The Clinical Effectiveness of Glucosamine and Chondroitin

Health Technology Assessment 2009; Vol. 13: No. 52

The clinical effectiveness of glucosamine and chondroitin supplements in slowing or arresting progression of osteoarthritis of the knee: a systematic review and economic evaluation

C Black, C Clar, R Henderson, C MacEachern, P McNamee, Z Quayyum, P Royle and S Thomas

November 2009 DOI: 10.3310/hta13520

Health Technology Assessment NIHR HTA programme www.hta.ac.uk HTA

How to obtain copies of this and other HTA programme reports An electronic version of this publication, in Adobe Acrobat format, is available for downloading free of charge for personal use from the HTA website (www.hta.ac.uk). A fully searchable CD-ROM is also available (see below). Printed copies of HTA monographs cost £20 each (post and packing free in the UK) to both public and private sector purchasers from our Despatch Agents. Non-UK purchasers will have to pay a small fee for post and packing. For European countries the cost is £2 per monograph and for the rest of the world £3 per monograph. You can order HTA monographs from our Despatch Agents: – fax (with credit card or official purchase order) – post (with credit card or official purchase order or cheque) – phone during office hours (credit card only). Additionally the HTA website allows you either to pay securely by credit card or to print out your order and then post or fax it.

Contact details are as follows: HTA Despatch Email: [email protected] Magellan Tel: 02392 492 000 Concept House, Bell Road Fax: 02392 478 555 Basingstoke, Hants RG24 8FB, UK Fax from outside the UK: +44 2392 478 555 NHS libraries can subscribe free of charge. Public libraries can subscribe at a very reduced cost of £100 for each volume (normally comprising 30–40 titles). The commercial subscription rate is £300 per volume. Please see our website for details. Subscriptions can be purchased only for the current or forthcoming volume.

Payment methods Paying by cheque If you pay by cheque, the cheque must be in pounds sterling, made payable to Direct Mail Works Ltd and drawn on a bank with a UK address. Paying by credit card The following cards are accepted by phone, fax, post or via the website ordering pages: Delta, Eurocard, Mastercard, Solo, Switch and Visa. We advise against sending credit card details in a plain email. Paying by official purchase order You can post or fax these, but they must be from public bodies (i.e. NHS or universities) within the UK. We cannot at present accept purchase orders from commercial companies or from outside the UK.

How do I get a copy of HTA on CD? Please use the form on the HTA website (www.hta.ac.uk/htacd.htm). Or contact Direct Mail Works (see contact details above) by email, post, fax or phone. HTA on CD is currently free of charge worldwide.

The website also provides information about the HTA programme and lists the membership of the various committees. AbstractAcknowledgements ...... iii 85

ContentsReferences ...... v 87 The clinical effectiveness of glucosamine ListAppendix of abbreviations 1 ...... vii Search strategies ...... 97 and chondroitin supplements in Executive summary ...... ix Appendix 2 slowing or arresting progression of ChapterTable 1 of excluded reviews ...... 99 Introduction ...... 1 osteoarthritis of the knee: a systematic AppendixAims and 3objectives ...... 1 Results of meta-analyses in the systematic review and economic evaluation Chapterreviews 2 ...... 101 Background ...... 3 AppendixOsteoarthritis 4 of the knee: an overview ...... 3 BurdenResults of OA of meta-analyses ...... – subgroup and 4 C Black,1* C Clar,1 R Henderson,2 Currentsensitivity standard analyses care/optimal ...... care ...... 7 105 3 4 Studying OA treatments in trials ...... 9 C MacEachern, P McNamee, AppendixInterventions 5 under investigation ...... 13 4 1 1 KeyResults questions of RCTsand rationale fulfilling ...... inclusion criteria 15 . 109 Z Quayyum, P Royle and S Thomas

1 ChapterAppendix 3 6 Section of Population Health, School of Medicine and Dentistry, University of ClinicalResults effectiveness of RCTs fulfilling of glucosamine inclusion and criteria chondroitin – Aberdeen, UK on thesubgroups progression ...... of OA of the knee ...... 17 121 2Public Health Medicine, NHS Grampian, Aberdeen, UK Methods ...... 17 3Department of Orthopaedics, Raigmore Hospital, Inverness, UK AppendixResults ...... 7 19 4Health Economics Research Unit, University of Aberdeen, UK SummaryHealth ...... economics methods – informing the58 model ...... 123 Chapter 4 *Corresponding author HealthCost-effectiveness Technology ofAssessment glucosamine reports and chondroitin published to in OA dateof the ...... knee ...... 61 125 Introduction ...... 61 Declared competing interests of the authors: none HealthMethods Technology ...... Assessment 61 Resultsprogramme ...... 66 145 Discussion ...... 68

Chapter 5 Biological plausibility ...... 71 Glucosamine, chondroitin and hyaline cartilage 71 Anabolic state ...... 73 Anticatabolic effects ...... 74 Published November 2009 Aggrecanase activity ...... 75 DOI: 10.3310/hta13520 Sulphate deficiency ...... 77 Summary ...... 77 This report should be referenced as follows: Chapter 6 Discussion and recommendations ...... 79 Black C, Clar C, Henderson R, MacEachern C, McNamee P, Quayyum Z, et al. The clinical Summary of findings ...... 79 effectiveness of glucosamine and chondroitin supplements in slowing or arresting progression Challenges for clinical effectiveness ...... 81 of osteoarthritis of the knee: a systematic review and economic evaluation. Health Technol Challenges for modelling ...... 82 Assess 2009;13(52). Research needs ...... 82 Conclusion ...... 83 Health Technology Assessment is indexed and abstracted in Index Medicus/MEDLINE, Excerpta Medica/EMBASE, Science Citation Expanded (SciSearch) and Current Contents/Clinical Medicine. NIHR Health Technology Assessment programme

he Health Technology Assessment (HTA) programme, part of the National Institute for Health TResearch (NIHR), was set up in 1993. It produces high-quality research information on the effectiveness, costs and broader impact of health technologies for those who use, manage and provide care in the NHS. ‘Health technologies’ are broadly defined as all interventions used to promote health, prevent and treat disease, and improve rehabilitation and long-term care. The research findings from the HTA programme directly influence decision-making bodies such as the National Institute for Health and Clinical Excellence (NICE) and the National Screening Committee (NSC). HTA findings also help to improve the quality of clinical practice in the NHS indirectly in that they form a key component of the ‘National Knowledge Service’. The HTA programme is needs led in that it fills gaps in the evidence needed by the NHS. There are three routes to the start of projects. First is the commissioned route. Suggestions for research are actively sought from people working in the NHS, from the public and consumer groups and from professional bodies such as royal colleges and NHS trusts. These suggestions are carefully prioritised by panels of independent experts (including NHS service users). The HTA programme then commissions the research by competitive tender. Second, the HTA programme provides grants for clinical trials for researchers who identify research questions. These are assessed for importance to patients and the NHS, and scientific rigour. Third, through its Technology Assessment Report (TAR) call-off contract, the HTA programme commissions bespoke reports, principally for NICE, but also for other policy-makers. TARs bring together evidence on the value of specific technologies. Some HTA research projects, including TARs, may take only months, others need several years. They can cost from as little as £40,000 to over £1 million, and may involve synthesising existing evidence, undertaking a trial, or other research collecting new data to answer a research problem. The final reports from HTA projects are peer reviewed by a number of independent expert referees before publication in the widely read journal series Health Technology Assessment. Criteria for inclusion in the HTA journal series Reports are published in the HTA journal series if (1) they have resulted from work for the HTA programme, and (2) they are of a sufficiently high scientific quality as assessed by the referees and editors. Reviews in Health Technology Assessment are termed ‘systematic’ when the account of the search, appraisal and synthesis methods (to minimise biases and random errors) would, in theory, permit the replication of the review by others.

The research reported in this issue of the journal was commissioned and funded by the HTA programme on behalf of NICE as project number 07/61/01. The protocol was agreed in February 2008. The assessment report began editorial review in December 2008 and was accepted for publication in March 2009. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The HTA editors and publisher have tried to ensure the accuracy of the authors’ report and would like to thank the referees for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this report. The views expressed in this publication are those of the authors and not necessarily those of the HTA programme or the Department of Health. Editor-in-Chief: Professor Tom Walley CBE Series Editors: Dr Aileen Clarke, Professor Chris Hyde, Dr John Powell,. Dr Rob Riemsma and Professor Ken Stein

ISSN 1366-5278 © 2009 Queen’s Printer and Controller of HMSO This monograph may be freely reproduced for the purposes of private research and study and may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NETSCC, Health Technology Assessment, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK. Published by Prepress Projects Ltd, Perth, Scotland (www.prepress-projects.co.uk), on behalf of NETSCC, HTA. Printed on acid-free paper in the UK by the Charlesworth Group. T DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Abstract The clinical effectiveness of glucosamine and chondroitin supplements in slowing or arresting progression of osteoarthritis of the knee: a systematic review and economic evaluation C Black,1* C Clar,1 R Henderson,2 C MacEachern,3 P McNamee,4 Z Quayyum,4 P Royle1 and S Thomas1

1Section of Population Health, School of Medicine and Dentistry, University of Aberdeen, UK 2Public Health Medicine, NHS Grampian, Aberdeen, UK 3Department of Orthopaedics, Raigmore Hospital, Inverness, UK 4Health Economics Research Unit, University of Aberdeen, UK *Corresponding author

Objective: To assess the clinical effectiveness and cost- > 12 months’ duration showed evidence of statistically effectiveness of glucosamine sulphate/hydrochloride and significant improvements in joint space loss, pain and chondroitin sulphate in modifying the progression of function for glucosamine sulphate, but the clinical osteoarthritis (OA) of the knee. importance of these differences was not clear. In two Data sources: Electronic databases were searched studies of glucosamine sulphate, the need for knee from 1950 to 2008 and included: MEDLINE and arthroplasty was reduced from 14.5% to 6.3% at 8 PubMed; EMBASE; Cochrane Library (including years’ follow-up. For other preparations of glucosamine, Cochrane Systematic Reviews Database, CENTRAL, chondroitin and combination therapy, there was less DARE, NHS EED and HTA databases); Allied and evidence to support a clinical effect. Cost-effectiveness Complementary Medicine (AMED); National Research modelling was restricted to glucosamine sulphate. Over Register (NRR); Web of Science Proceedings; Current a lifetime horizon the incremental cost per quality- Controlled Trials; and Clinical Trials.gov. Other sources adjusted life-year (QALY) gain for adding glucosamine included bibliographies of retrieved papers, registered sulphate to current care was estimated to be £21,335. but unpublished trials, internet searches and the Food Deterministic sensitivity analysis suggested that the Standards Agency website. cost-effectiveness of glucosamine sulphate therapy was Review methods: A search was conducted for particularly dependent on the magnitude of the quality systematic reviews of randomised controlled trials of life (QoL) gain, the change in knee arthroplasty (RCTs), which were used to identify RCTs of at least 12 probability with therapy and the discount rate. At a cost months’ duration and updated with searches for primary per QALY gained threshold of £20,000, the likelihood studies. A cost-effectiveness model was constructed that glucosamine sulphate is more cost-effective than using cohort simulation and drawing on available current care is 0.43, while at a threshold of £30,000, the evidence. Sensitivity analysis was undertaken and value probability rises to 0.73. Probabilistic sensitivity analysis of information analysis conducted. A review of studies showed that estimates were imprecise and subject to of mechanism of action was carried out to explore the a degree of decision uncertainty. Value of information biological plausibility of the preparations. analysis demonstrated the need for further research. Results: Five systematic reviews and one clinical Several biologically plausible mechanisms of action for guideline met the inclusion criteria. They reported glucosamine sulphate and chondroitin were proposed. inconsistent conclusions with only modest effects Conclusions: There was evidence that glucosamine on reported pain and function. A reduction in joint sulphate shows some clinical effectiveness in the space narrowing was more consistently observed, but treatment of OA of the knee. No trial data came from the effect size was small and the clinical significance the UK and caution should be exercised in generalising uncertain. A separate review of eight primary trials of the findings to the UK health-care setting. Cost- iii

effectiveness was not conclusively demonstrated. There mechanism of glucosamine sulphate and chondroitin was evidence to support the potential clinical impact of remains uncertain and, in particular, the proposal glucosamine sulphate. The value of information analysis that the active substance may be sulphate should be identified three research priorities: QoL, structural explored further. outcomes and knee arthroplasty. The biological

iv DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Contents

List of abbreviations ...... vii 6 Discussion and recommendations ...... 79 Summary of findings ...... 79 Executive summary ...... ix Challenges for clinical effectiveness ...... 81 Challenges for modelling ...... 82 1 Introduction ...... 1 Research needs ...... 82 Aims and objectives ...... 1 Conclusion ...... 83

2 Background ...... 3 Acknowledgements ...... 85 Osteoarthritis of the knee: an overview ...... 3 Burden of OA ...... 4 References ...... 87 Current standard care/optimal care ...... 7 Studying OA treatments in trials ...... 9 Appendix 1 Search strategies ...... 97 Interventions under investigation ...... 13 Key questions and rationale ...... 15 Appendix 2 Table of excluded reviews ...... 99

3 Clinical effectiveness of glucosamine and Appendix 3 Results of meta-analyses in chondroitin on the progression of OA of the systematic reviews ...... 101 the knee ...... 17 Methods ...... 17 Appendix 4 Results of meta-analyses – Results ...... 19 subgroup and sensitivity analysis ...... 105 Summary ...... 58 Appendix 5 Results of RCTs fulfilling 4 Cost-effectiveness of glucosamine and inclusion criteria ...... 109 chondroitin in OA of the knee ...... 61 Introduction ...... 61 Appendix 6 Results of RCTs fulfilling Methods ...... 61 inclusion criteria – subgroups ...... 121 Results ...... 66 Discussion ...... 68 Appendix 7 Health economics methods – informing the model ...... 123 5 Biological plausibility ...... 71 Glucosamine, chondroitin and hyaline . Health Technology Assessment reports cartilage ...... 71 published to date ...... 125 Anabolic state ...... 73 Anticatabolic effects ...... 74 Health Technology Assessment Aggrecanase activity ...... 75 programme ...... 145 Sulphate deficiency ...... 77 Summary ...... 77

DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

List of abbreviations

ACR American College of mRNA messenger ribonucleic acid Rheumatology NCCCC National Collaborative Centre BMI body mass index for Chronic Conditions

BNF British National Formulary NICE National Institute for Health and Clinical Excellence CI confidence interval NO nitric oxide COX cyclo-oxygenase NSAID non-steroidal anti-inflammatory CRD Centre for Reviews and drug Dissemination OA osteoarthritis EVPI expected value of perfect information OARSI Osteoarthritis Research Society International FDA Food and Drug Administration OMERACT outcome measurement in clinical GAIT Glucosamine/chondroitin trials Arthritis Intervention Trial

PGE2 prostaglandin E2 ICER incremental cost-effectiveness ratio PPP purchasing power parity

IL interleukin QALY quality-adjusted life-year

iNOS inducible nitric oxide synthase QoL quality of life

INR international normalised ratio QUOROM quality of reporting of meta- analyses JOA Japan Orthopaedic Association RCT randomised controlled trial MCID minimal clinically important difference RR relative risk

MHRA Medicines and Healthcare SD standard deviation Products Regulatory Agency SE standard error MMP matrix metalloproteinase SF-36 36-item Short Form General MRI magnetic resonance imaging Health Survey continued

vii

TKR total knee replacement WHO World Health Organization

TNF tumour necrosis factor WOMAC Western Ontario and McMaster Universities osteoarthritis index VAS visual analogue scale

All abbreviations that have been used in this report are listed here unless the abbreviation is well known (e.g. NHS), or it has been used only once, or it is a non-standard abbreviation used only in figures/tables/appendices, in which case the abbreviation is defined in the figure legend or in the notes at the end of the table.

viii DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Executive summary

Background review as well as from other relevant sources, a model to assess cost-effectiveness was constructed. Osteoarthritis (OA) of the knee is a major source Sensitivity analysis was undertaken and value of of disability in the UK, resulting in pain, loss information analysis conducted. of function and, for some, the need for knee arthroplasty. Two components of cartilage Furthermore, a review of studies of mechanism of structure, glucosamine and chondroitin, are action was carried out to explore the biological available as food supplements and/or licensed plausibility of the preparations under study. medicines. Reviews of short-term effectiveness in preventing disease progression and symptom control have been disappointing. Results

Five systematic reviews and one clinical guideline Objective met the inclusion criteria. They reported inconsistent conclusions with, at best, modest The aim of this systematic review and economic effects on reported pain and function. A reduction analysis was to assess the clinical effectiveness in joint space narrowing was more consistently and cost-effectiveness of glucosamine sulphate observed; however, the effect size was small and the or hydrochloride and chondroitin sulphate in clinical significance was reported to be uncertain. modifying the progression of OA of the knee. Data were not presented separately for long-term studies of > 12 months; therefore, we went on to review separately RCTs of > 12 months’ duration. Methods Eight primary trials were included with a duration To assess clinical effectiveness, we first conducted of at least 12 months. There was evidence of a search for systematic reviews of randomised statistically significant improvements in joint controlled trials (RCTs). Electronic databases space loss, pain and function for glucosamine were searched from 1950 to 2008 and included: sulphate; however, the clinical importance of MEDLINE and PubMed; EMBASE; Cochrane these differences was less clear. In two studies Library (including Cochrane Systematic Reviews of glucosamine sulphate, both funded by the Database, CENTRAL, DARE, NHS EED and HTA manufacturer (Rotta, Italy) of an oral powder databases); Allied and Complementary Medicine product, the need for knee arthroplasty was (AMED); National Research Register (NRR); Web reduced from 14.5% to 6.3% at 8 years’ follow-up. of Science Proceedings; Current Controlled Trials; For other preparations of glucosamine, chondroitin and Clinical Trials.gov. Other sources included and combination therapy, there was less evidence bibliographies of retrieved papers, registered but to support a clinical effect. unpublished trials, internet searches and the Food Standards Agency website. We used these reviews Cost-effectiveness modelling was restricted to to identify RCTs of at least 12 months’ duration glucosamine sulphate. Over a lifetime horizon and updated our findings with searches for primary the incremental cost per quality-adjusted life-year studies up to October 2007, with monthly alerts (QALY) gain for adding glucosamine sulphate being checked through to November 2008. Data to current care was estimated to be £21,335. were extracted from the reviews and RCTs and Deterministic sensitivity analysis suggested that the quality was checked. Where appropriate, meta- cost-effectiveness of glucosamine sulphate therapy analysis was undertaken. was particularly dependent on the magnitude of the quality of life (QoL) gain. At a cost per No cost-effectiveness studies were identified in the QALY gained threshold of £20,000, the likelihood published literature. Using cohort simulation, and that glucosamine sulphate is more cost-effective drawing on evidence from the clinical effectiveness than current care is 0.43, while at a threshold of ix

£30,000, the probability rises to 0.73. Probabilistic versus placebo over long-term treatment. sensitivity analysis showed that estimates were Any future trial should also inform our somewhat imprecise and subject to some degree of understanding of the relation between QoL decision uncertainty. Value of information analysis and costs of collecting resource use and cost indicated that further research to reduce decision data to allow estimation of the resource impact uncertainty would be beneficial, with priority being of any changes in QoL. given to determining the magnitude and duration 2. Structural outcomes – further long-term trial of QoL gains that arise following treatment. data are required to clarify the impact on the ultimate need for knee arthroplasty, including Several biologically plausible mechanisms of action the ability to delay the need for surgery. As for glucosamine sulphate and chondroitin were yet, surrogate marks continue to be proposed proposed. Importantly, bioavailability in the joint but, in the absence of long-term follow-up to space synovial fluid was demonstrated. surgery, the implications of change in surrogate end points remain uncertain. 3. Knee arthroplasty – a nationally representative Conclusions cohort study is required to understand what proportion of patients with OA (diagnosed in There was evidence that glucosamine sulphate primary care and referred to secondary care) shows some clinical effectiveness in the treatment require knee arthroplasty. of OA of the knee. No trial data came from the UK, and in the absence of good UK data about the Trials of interventions should focus on glucosamine current referral practice, management and surgical sulphate, and the Rotta product is the only one to rate, caution should be exercised in generalising date that has demonstrated effectiveness. While these data to the UK health-care setting. Cost- uncertainty about other preparations remains, effectiveness was not conclusively demonstrated, there was insufficient evidence of effectiveness and with substantial uncertainty related to the it was not possible to develop an economic case for magnitude and duration of QoL gain following further study at this time. Any trial should: treatment. There was evidence from biological studies to support the potential clinical impact of • include collection of information about co- glucosamine sulphate. For other preparations, the prescribing, the use of other interventions and evidence base was less consistent (chondroitin) or adverse events absent (glucosamine hydrochloride). • recruit obese and overweight participants and people across stages of OA severity Based on sensitivity analysis and value of • use the opportunity to gather a number of information analysis three research priorities were measures of joint structure and damage identified: • be of at least 3 years’ follow-up, with a mechanism to follow the cohort long term (e.g. 1. QoL – further clarification of the potential through record linkage to hospital data). QoL gains [using a generic preference-. based QoL measure (such as the Health The biological mechanism of glucosamine sulphate Utilities Index 3, Short Form-6D, EuroQol- and chondroitin remains uncertain and, in 5D) that can readily be used to estimate utility] particular, the proposal that the active substance from treatment with glucosamine sulphate may be sulphate should be explored further.

x DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Chapter 1 Introduction

steoarthritis (OA) of the knee is a leading This review sought evidence of the clinical Ocause of pain and disability in the UK. effectiveness of glucosamine and chondroitin in Treatments to date have largely helped only to modifying disease progression by focusing on manage symptoms and have not modified the trials of at least 12 months’ duration. Disease disease process. While the clinical course of OA is progression was defined in terms of pain, variable, for some the condition progresses to the function, joint structural change and arthroplasty. point where knee arthroplasty is the only option Economic analysis examined the cost-effectiveness for the management of pain and restoration of of the addition of these treatments to current function. care and, using value of information analysis, sought to identify where additional research Two components of the cartilage structure, might most effectively be targeted. A review of glucosamine and chondroitin, are available as biological studies has been included to explore the food supplements. Proponents of these treatments biological plausibility of a mechanism of action for suggest that oral supplements increase the glucosamine and chondroitin in joint preservation concentration of these components of cartilage in and repair. the joint and may help to preserve, or even repair, the damaged joint in OA. Aims and objectives As food supplements, glucosamine and chondroitin have not been subject to the same rigorous control The aim of this systematic review and economic and licensing processes as medicines. However, analysis was to assess the clinical effectiveness one product, a preparation of glucosamine and cost-effectiveness of glucosamine sulphate hydrochloride, has received a licence as a or hydrochloride and chondroitin sulphate in prescription-only medicine in the UK. A further modifying the progression of OA of the knee. glucosamine sulphate product, featured in many of the clinical trials, is licensed in Europe, In particular, we sought to address the following but not in the UK. A number of randomised questions: controlled trials (RCTs) have been conducted to assess the effectiveness of these treatments in • Do glucosamine and/or chondroitin prevent or the management of OA and several reviews have slow progression of OA of the knee? sought to summarise the clinical effectiveness. • Is treatment with glucosamine and/or While there has been a degree of heterogeneity chondroitin safe? in the conclusions drawn, in general, the impact • Is the addition of glucosamine and/or on short-term pain and function outcomes has chondroitin to current clinical practice cost- been disappointing. However, if the mechanism effective? of action is one of long-term disease modification, • Where could research most effectively be through joint preservation and repair, as has been focused to address remaining uncertainties? suggested, short-term studies may not have been able to demonstrate benefit.

DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Chapter 2 Background

Osteoarthritis of the process’ for synovial joints. Clinically, a dynamic knee: an overview process is in keeping with the range in severity of symptoms and signs experienced by individuals. Osteoarthritis is the most prevalent joint disorder In the long term, OA may lead to pain, deformity, and cause of disability in the UK.1 Although impairment of function and, in some patients, a the incidence increases with age, its onset is not need for knee arthroplasty. For a proportion of inevitable. Most commonly, OA affects the knees, patients, symptoms are reported to stabilise or even hips and small joints of the hands. The focus of this improve.1 review is on OA of the knee and the information presented below, while often applicable to OA at Diagnosis other anatomical sites, will concentrate on the knee joint. The diagnosis of OA is based on clinical examination of the joint and first-line investigation Although there are different systems with which by means of X-ray assessment. Classically, OA is classified, OA can be broadly separated into radiographic evidence of OA includes: primary and secondary OA. Primary, idiopathic, OA is the most common. Secondary OA results • decreased joint space from previous trauma, infection or congenital • osteophytosis abnormality.2 • formation of subchondral cysts • subchondral sclerosis. Clinically, OA leads to asymmetric joint swelling, joint crepitus, decreased range of movement The widely used American College of and occasionally joint locking. Symptomatically, Rheumatology (ACR) definition of OA is based patients may complain of joint pain with associated on a set of clinical criteria along with evidence, short-lived early morning stiffness; however, the from blood tests, of an absence of a primary degree of severity of symptoms is hugely variable inflammatory process (Box 1).4 and does not necessarily represent radiological change. The recently published NICE guideline for OA1 simplifies the definition of OA to: Definition of osteoarthritis • persistent joint pain that is worse with use Osteoarthritis has been defined as a chronic • age 45 years and over disorder characterised by:3 • morning stiffness lasting no more than half an hour • focal erosive lesions • cartilage destruction as the basis for a clinical working diagnosis. In the • subchondral sclerosis majority of patients, further investigation was felt • cyst formation not to be warranted in routine practice. • large osteophytes at the margins of the joint. Aetiology A member of the non-inflammatory arthritides, OA has been described as a degenerative process.2 The aetiology of OA remains unclear, but it is However, there is evidence of an inflammatory certainly multifactorial. There is a marked genetic response with upregulation of various pro- predisposition to the disease with high levels inflammatory mediators. Recent National of concordance noted in twin studies. Other Institute for Health and Clinical Excellence factors which come into play include joint shape, (NICE) guidelines for the management of OA occupation and leisure activities. At the cellular highlighted the dynamic nature of the joint level, there is an imbalance of the continuous remodelling process, describing it as a ‘repair 3

BOX 1 American College of Rheumatology definition of OA

:A diagnosis of OA of the knee is defned as presenting with pain, and meeting at least five of the following criteria

• patient older than 50 years of age • less than 30 minutes of morning stiffness • crepitus (noisy, grating sound) on active motion • bony tenderness • bony enlargement • no palpable warmth of synovium • erythrocyte sedimentation rate (ESR) < 40 mm/hour • rheumatoid factor < 1:40 • non-inflammatory synovial fluid

catabolic and anabolic metabolism, favouring It appears that there may be different risk factors catabolic degeneration of articular cartilage. for disease progression and development.6 For example, high bone density has been identified Osteoarthritis is the clinical and pathological as a risk factor for development, whereas low outcome of a range of disorders that results bone density has been identified as a risk factor in structural and functional failure of synovial for progression in knee OA.6 Other risk factors joints. Osteoarthritis occurs when the dynamic associated with the progression of knee OA include: equilibrium between the breakdown and repair obesity, low intake of vitamins C and D, and varus/ of joint tissues is overwhelmed.5 valgus malalignment.6–9 Deprivation and manual occupations have also been associated with higher In early OA, fibrillation is seen on the surface prevalence of knee pain and the need for knee layer of articular cartilage. This fibrillation tends arthroplasty.10,11 to correspond to areas of weight bearing and, on microscopy, has the appearance of superficial clefts. As the disease process worsens, these Burden of OA clefts become deeper and this is associated with Epidemiology progressive cartilage loss. Ultimately, the weight- bearing surface of cartilage is destroyed completely, The epidemiology of OA has been described as leaving an area of particularly dense subchondral ‘difficult to determine’.1 This reflects differences bone. These progressive changes lead to gradual between studies with respect to the method used reduction in joint space and ultimately abolition for diagnosing OA – radiological or clinical.12 of joint space with ‘bone on bone’ contact which is However, even comparing the results of studies visible on standard X-rays.5 using the same broad methodology can be difficult.13 For example, a variety of radiological Risk factors and diagnostic criteria are available, and the results predictors of disease may be dependent on which criteria are used.13 Furthermore, OA (whether radiologically or A variety of risk factors for developing OA have clinically diagnosed) will present with varying been identified. These have been classified as:1 degrees of severity. Whether or not patients with all degrees of severity are included in prevalence • genetic – it has been estimated that genetic estimates, or only those people with the most factors account for 40–60% of hand, knee and severe arthritis, will affect the results.13 In addition, hip OA radiological and clinical diagnoses do not always • constitutional (e.g. ageing, female gender, correlate.7,13 For example, patients with radiological obesity, high bone density) OA of the knee may not experience pain.14 • biomechanical (e.g. joint injury, occupational Similarly, patients experiencing knee pain may not and recreational issues, reduced muscle have any radiographic signs.9,12 strength).

4 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

There also appear to be differences between are genuine or an artefact resulting from the prevalence rates reported by studies undertaken method of case ascertainment.15 within different populations.15 However, due to the methodological issues described above, further Table 1 summarises estimates of the prevalence studies are required to clarify if these differences of knee OA drawn from several studies. While

TABLE 1 Prevalence of knee osteoarthritis contained within the published literature

Reference Population Criteria Prevalence National Collaborating UK Radiological 25% Centre for Chronic Adults aged over 50 years Conditions 20081 Garstang and Stitik 200612 Various countries Radiological At least 33% (and up to 68% Adults aged over 55 years in some studies) D’Ambrosia 200513 USA; Framingham study Radiological – severe (grade Women: 21.1% Adults aged 80 years and III or IV) Men: 17.4% over Arden and Nevitt 20067 USA Radiological 30% Adults aged 75 years and over Corti and Rigon 200315 Sweden Radiological Women: 45% Adults aged 75 years and Men: 33% over Netherlands Radiological Women: 35% Adults aged 65–75 years Men: 21% Italy Clinical Women: 26% Adults aged 65 years Men: 12% Bedson et al. 200716 UK Clinical 25–37% Adults over 50 years Dillon et al. 200610 USA; National Health and Radiological – severe (grade Women: 12.9% Nutrition Examination III or IV) Men: 6.5% Survey (1991–4) Adults aged 60 years and over White 200617 Population surveys Knee pain lasting more than ~ 25% conducted in the UK among 4 weeks adults aged over 55 years Adults aged over 60 years Clinical 10–15% Williams and Spector 200618 USA; National Health and Radiological – severe (grade Women: 6.6% Nutrition Examination III or IV) Men: 2.0% Survey (1971–5) Adults aged 65–74 years UK Clinical 18.1% Adults aged over 55 years Brooks 200219 Spain Clinical (self-completed 10.2% Adults (no age range postal questionnaire) specified)

The prevalence estimates are contained within review articles which cite primary studies.

not intended to be a comprehensive list, it Dieppe and colleagues20 found that there was no demonstrates the range of estimates reported correlation between radiographic change and within the published literature. clinical outcomes. They suggested that X-ray appearances may poorly reflect morphological Despite the methodological issues alluded to, some changes within joints, or alternatively that clinical broad conclusions can be drawn. Osteoarthritis of and radiographic changes are out of step with the knee becomes more common with increasing each other. Johnson and colleagues22 explored the age. In addition, prevalence rates among elderly association between self-assessed symptoms and women exceed those of elderly men.13 functional and radiological measures, reporting a poor correlation after 7 years. Improvements in Natural history of OA of the knee self-assessed symptoms (in 25% of participants) were not reflected in measures of function or on The natural history of OA is not fully understood.18 imaging. The authors raise the question of the role However, in contrast to the widely held view that of adaptation by patients as they develop strategies OA is an invariably progressive, deteriorating to avoid pain. Assessments focusing on pain or condition, it would appear that the condition can ability to complete certain tasks will not detect the remain stable for some time.7 While radiographic more subtle changes in social and physical activity recovery is uncommon, radiographic signs may that may be occurring as a coping mechanism.22 stabilise and clinical outcomes may even improve. This is reflected in the relatively low proportion of Consequences for health patients coming to surgery and has been reported even among patients referred to specialist clinics.20 The pain and disability associated with OA can have a marked effect on patients’ quality of life A recently published guideline on the management (QoL) and their ability to live independently.17 of OA states that, over a period of several years, People with the condition can experience approximately one-third of patients with knee OA significant limitations with respect to undertaking improve, one-third remain stable and one-third activities of daily living.14 OA of the large, weight- deteriorate.1 bearing joints (hips and knees) is particularly debilitating,7,15 and it has been claimed that Several authors17,20,21 have reported that the clinical knee OA is the most frequently reported cause of progression of OA is variable. Some patients may disability in older adults.23 Disability in relation experience rapidly progressive disease, while in to walking and using stairs that is attributable to others, the condition can remain stable for many knee OA equals that associated with cardiovascular years. White17 described a study which found that disease.15 A number of chronic conditions (e.g. up to 40% of patients with significant radiographic hypertension, diabetes mellitus) may contribute knee OA showed no deterioration when repeat to disability. If present, such conditions may add radiographs were performed 20 years later. It has to the difficulties experienced by patients as a been claimed that when followed up for over a consequence of their arthritis.15 decade, radiographic signs of joint degeneration did not progress in one-third to two-thirds of Hawker and colleagues24 conducted a study to patients with OA of the knee or hip.21 In one explore the nature of the pain experienced by small study that followed up 63 patients with knee people with knee and hip OA. Ninety-one of the OA over an 11-year period, the radiographic 143 participants recruited had knee OA. Two appearance improved in 10% of patients.21 Dieppe main types of pain were described: a constant dull, and colleagues20 followed up over a 3-year period aching pain and an intermittent, unpredictable, approximately 400 patients with OA who had been intense pain. The latter was associated with the referred to a specialist rheumatology clinic. Two- avoidance of social and recreational activities, thirds of the patients recruited had OA of the knee. as well having an adverse effect on the patients’ Fifty-eight per cent of patients reported an overall mood. worsening of their condition. Twenty-one per cent reported no change and 21% reported an overall Other studies have reported a high prevalence of improvement. With respect to pain, there was psychological morbidity among patients with knee considerable individual patient variation with many OA.23,25 As well as affecting patients’ QoL, this patients reporting markedly different levels of pain may also impact on their use of health services. at follow-up compared with baseline. All measures Rosemann and colleagues26 report that depression of function showed an overall deterioration. was positively associated with patients consulting 6 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

their GP and being referred to an orthopaedic details of current practice, a guideline produced specialist because of knee pain, although other by NICE in 2008,29 and based on work by the studies have reported an inverse relationship.16 National Collaborating Centre for Chronic However, both situations are problematic. It may Conditions (NCCCC),1 provides recommendations be the case that some patients referred to an for the care pathway for people with OA. orthopaedic specialist are undergoing surgery Figure 1 summarises the recommended care. when, in the first instance, treating their depression The three core treatments in the centre circle would have been more appropriate. It is also were recommended for all patients. Paracetamol possible that depressed patients who do not engage and topical NSAIDs were recommended as the with health services are missing out on care from first line in analgesia; treatments in the outer which they would benefit. circle then being recommended for only selected individuals. The guideline authors did, however, Several studies have reported that patients with note a number of areas of uncertainty including: knee pain may not consult health professionals.16,27 the most effective combinations of treatments; how A variety of contributory factors have been to improve adherence to the non-pharmacological proposed. It has been suggested that patients view interventions; and how most effectively to treat very knee OA pain as being an inevitable consequence elderly patients. of ageing and something which has to be tolerated.27 It has been suggested that the attitude Current care of health professionals can contribute to this (e.g. Non-pharmacological interventions by discussing knee pain and OA using terms such as ‘wear and tear’).27 Studies also suggest that some There are no national data about the levels patients have a degree of therapeutic nihilism of current practice in relation to the non- and believe that little can be done to alleviate pharmacological interventions outlined in the their pain.27 Other studies suggest that some centre circle. If guideline uptake was complete patients fear becoming addicted to analgesics or then all patients with OA would receive these core experiencing medication side effects.27 treatments. The guideline does not estimate the cost or service implications of implementing these changes. Chevalier and collegues30 conducted Current standard a questionnaire survey among French general care/optimal care practitioners to investigate the management and treatment of knee OA. They compared the survey Treatment results with recommendations produced by the A range of treatments can help to manage the European League Against Rheumatism in 2000 symptoms of OA of the knee. Non-pharmacological and found that non-pharmaceutical treatments, interventions include weight loss, exercise and a including recommencing exercise and weight range of physiotherapy interventions. Pain control loss as first-line approaches, appeared to be is achieved through the use of exercise, simple underutilised. They reported a high request rate analgesia, non-steroidal anti-inflammatory drugs for radiological examinations even if not required; (NSAIDs) and, if necessary, opiates. Transcutaneous the underutilisation of non-pharmacological electrical nerve stimulation (TENS) machines, heat treatments as a first-line approach; the therapy and topical NSAIDs may also be used. If inappropriate use of bed rest; and high use of oral non-invasive interventions fail to control symptoms, NSAIDs as a first-line analgesic. then intra-articular injections, arthroscopic lavage and debridement for joint locking, and ultimately Pharmacological interventions arthroplasty have been utilised with evidence of The main changes to practice identified in varying degrees of effectiveness. The primary aim the NICE guideline29 are around the use of of these treatments is management of symptoms, in paracetamol and topical NSAIDs as first-line particular pain and loss of function. Most of these care. This change in practice, if implemented, treatments do not modify disease progression.2 was identified as critical in reducing the adverse events experienced in relation to the use of Management of OA of the knee NSAIDs. Approximately 167,000 people with OA in England (based on a sample of primary care In the UK, information about current health- practices) were estimated to be treated currently care practice in the management of OA of the with topical NSAIDs, increasing to more than knee is scarce.28 While we may not have clear 2.5 million if the guideline was adopted (a total 7

Oral NSAIDs including COX-2 inhibitors Opioids Capsaicin Intra-articular corticosteroid injections Paracetamol Supports and braces Topical Education, advice, NSAIDs information access Local heat and cold Strengthening exercise, aerobic fitness training Shock- absorbing shoes Weight loss if or insoles overweight/obese Assistive devices

TENS

Joint Manual therapy arthroplasty (manipulation and stretching)

FIGURE 1 Targeting treatment: a summary of the treatments recommended. COX-2, cyclo-oxygenase-2; NSAIDs, non-steroidal anti- inflammatory drugs; TENS, transcutaneous electrical nerve stimulation. (Reproduced from: National Collaborating Centre for Chronic Conditions. Osteoarthritis: national clinical guideline for care and management in adults. London: Royal College of Physicians, 2008. Copyright © 2008 Royal College of Physicians. Reproduced by permission.)

cost of approximately £17million per year). Data Hospital Episode Statistics (HES) data in about oral NSAID and cyclo-oxygenase-2 (COX- 2005–6 reported that 19,686 people over the 2) inhibitor utilisation were limited, and based age of 45 years received arthroscopic lavage on clinical judgement, with an estimated 50% of and debridement in England. One of the major people with OA receiving these treatments. The conclusions of the NICE guidelines for OA was to other main change in practice would come from recommend that this procedure only be used where the dual prescription of proton pump inhibitors to patients have a clear history of mechanical locking. all patients receiving NSAIDs or COX-2 inhibitors, The guideline predicts that following this guidance but again estimates about current practice were would result in one-third fewer patients receiving not based on collected data. These data applied to arthroscopies. OA as a whole, i.e. were not restricted to OA of the knee. Audit data from England and Wales, complete for more than 85% of National Health Service (NHS) Intra-articular injections with hyaluronic acid were hospitals, reported that 57,597 knee arthroplasties not recommended for use. No data were presented were conducted in 2006; of these 97% were for OA. on the extent of use in current practice. The average age of patients undergoing surgery was 70 years and 57% were female.31 Invasive procedures Two main invasive procedures are used in the Economic burden of OA management of OA of the knee: It has been estimated that OA in general costs the • arthroscopic lavage and debridement US economy $60–65 billion annually.12,21 In the • knee arthroplasty. UK during the period 1999–2000, an estimated 8 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

36 million working days were lost as a result of required in the 12 months prior to the trial and 52 OA, with a cost of £3.2 billion in associated lost outpatient clinic appointments were necessary for productivity.1 management of their OA.32,33

Osteoarthritis has major impacts on health services. It is estimated that two million adults attend Studying OA their GP each year as a consequence of OA.1 With treatments in trials respect to knee pain, the incidence of new GP consultations among those aged 50 years and older Studying the impact of treatments on OA within is estimated at 10% per annum.1 As noted above, trials has been challenging. Progression of disease in the order of 58,000 primary knee arthroplasty occurs over years and decades. The number operations are carried out annually in England and of people deteriorating to need arthroplasty is Wales for OA of the knee. relatively small in relation to the total population with knee OA. Thus, studying interventions that A UK population survey conducted by Yong and aim to modify the course of the condition requires colleagues11 attempted to estimate the proportion either very long follow-up or a surrogate marker of the population aged 65 years and older that for progression. may benefit from knee arthroplasty based on Lequesne index scores. It was reported that 7.9% Surrogate end points for trials are often criticised. of respondents were assessed as possibly requiring In knee OA, two main types of end points have knee arthroplasty. This dropped to 5.1% once been adopted in trials of interventions: composite patients with specified contraindications to surgery scoring systems for pain and function specific to were taken into account. Of these patients, 5.3% OA; and measures of joint space. appeared to be candidates for bilateral knee arthroplasty. Measures of health state in OA

A trial of patient education for management of Standard generic measures of QoL such as the 36- OA of the knee conducted in the UK in 1996–7, item Short Form General Health Survey (SF-36) and including an economic evaluation, provided have not been widely adopted in trials of therapy a detailed and valuable source of costs of care at for knee OA. Several instruments are available that time.32 The estimated direct cost of knee OA- for assessing pain and function specifically in related health care over the 2-year study period OA. Lequesne34 developed a scoring system (excluding the trial intervention) was £212 per for severity of knee OA based on pain, walking patient for the NHS. Within this total, medication distance and activities of daily living, and using and primary care costs were estimated to be, on this, subdivided patients into levels of severity of average, £50 per patient. The mean hospital- disability. The Western Ontario and McMaster based costs per patient were £52 (including Universities OA index (WOMAC)35 is another radiology) with a further £20 for allied medical index for the assessment of OA of the knee or hip professions and complementary therapists on the and is based on assessment of pain, stiffness and NHS. Patients paid £78 per patient for private physical function. The WOMAC index is probably consultations with allied health-care professionals the most widely used within trials.35–37 It has been (largely physiotherapy), complementary therapy, rated as the leading tool in OA of the knee and travel expenses, prescription charges and over-the- hip; however, problems have been identified where counter medicines. The cost for society as a whole function and pain domains are divergent in terms was £291 per patient.32 of a patient’s experience.38 The most commonly used tools are summarised in Table 2. With a mean study population age of 65 years, the trial participants may have been a little younger Trials generally present a mean treatment group than the general population of people with OA difference from baseline to the end of the trial. of the knee. Seventy-one per cent were female In order to interpret the difference there is a and 55% of participants had at least 3 years of need to understand the clinical significance of OA history prior to recruitment. Approximately the change, sometimes known as the ‘minimal 15% of patients required a hospital inpatient stay clinically important difference’ (MCID). The during the 1-year follow-up period, accruing a total MCID is a threshold and how such cut-off points of 5.9 days as inpatients in orthopaedics. For the are defined is not standardised and is vulnerable 170 patients in the study, 276 visits to the GP were to the approach used. They do, however, give an 9

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Background Validated Validated in OA Yes Yes Yes Yes Knee Specific Yes No No Yes OA OA Specific Yes No Yes Yes Mental Health No Yes No No Stiffness No No Yes Yes Yes Yes Yes Yes Walk Walk distance Yes Yes Yes Yes Function Yes Yes Yes Yes Pain 13 Each question scores 1–5. 60 is worst score 0 is bad, 100 good. Scores are broken down into eight subcategories Scores are broken down into 0–20 (pain), 0–8 (stiffness), 0–68 (physical function) Combined severity score, 1–4 5–7 moderate; 8–10 mild OA; severe; 11–13 very > extremely severe Scale details mm scale 5–60 0–100 Uses VAS 100 or Likert 0–5 1–24 Scale 12 36 24 10 No. of questions Summary of composite health status scales commonly used in trials treatment for OA Oxford Knee Score SF-36 WOMAC Lequesne visual analogue scale. VAS, Scoring system

10 2 TABLE DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

idea of the magnitude of effect needed for patients • Grade 2 Mild: definite osteophytes and to express an improvement in their clinical state. possible narrowing of joint Tubach and colleagues39 estimated the MCID space using WOMAC and a 0–100 scale in patients with • Grade 3 Moderate: moderate multiple knee OA, stratifying by baseline symptom severity. osteophytes, definite narrowing The MCID for patients with low baseline scores of joint space and some sclerosis was estimated to be a 5-point improvement [95% and possible deformity of bone CI (confidence interval) 4 to 7]. For those who ends reported intermediate baseline scores, a 12-point • Grade 4 Severe: large osteophytes, (95% CI 10 to 15) improvement was considered marked narrowing of joint clinically important. For those with high baseline space, severe sclerosis and scores, the MCID was 20 (95% CI 18 to 23).39 definite deformity of bone ends.

None of these patient QoL scoring systems are in The use of scales is subjective and, in the context use in clinical practice outwith the context of a trial, of clinical trials and assessing disease progression and they have not been designed for individual over time, may lack reproducibility. With only patient clinical decision-making. A working group five grades, such tools are relatively crude and from OMERACT (Outcome Measurement in subtle changes cannot be represented. Joint space Clinical Trials) and OARSI (Osteoarthritis Research has the advantage that it can be measured using Society International) has been working on new standard X-rays. The relationship between joint definitions of severity of knee and hip OA, with space, progressive joint damage and progressive a view to considering when knee arthroplasty is symptoms is discussed below. necessary.40 Definitions of joint Measuring joint space width space in the knee Healthy knee Osteoarthritis has typically been considered a slowly progressive condition; however, in OA of the In a healthy knee joint, articular cartilage, not knee, the outcome is variable. While radiographic normally visible on standard X-rays, fills the joint recovery is uncommon, it can occur, as mentioned creating an apparent space on X-rays between the above.21 Clinical outcomes, measured using the visible femoral and tibial bones. The knee joint scales described above, may stabilise or improve. space can be summarised by a number of measures So, while clinical care of OA is based on clinical on standard X-rays; the commonly used measures symptoms, there has been substantial research are summarised below: interest in assessing if interventions are modifying the long-term disease process, reducing joint • mean medial joint space damage and, ultimately, reducing the progression • minimum joint space of symptoms and the need for knee arthroplasty or • mean lateral joint space other surgical interventions. • minimum lateral joint space.

Markers for those at risk of progressive disease Joint space loss is a characteristic radiological have been sought. Radiological markers of joint feature of OA, as well as rheumatoid arthritis damage, in particular joint space narrowing, have and other inflammatory arthropathies. In these been widely accepted as surrogate markers for conditions, joint space loss represents loss of structural damage and have been recommended as articular cartilage.42 What constitutes a ‘normal’ a primary end point for clinical trials by scientific healthy joint space on standard X-ray is not well organisations and regulatory agencies in Europe studied. X-ray studies among people attending and the USA.41 Accident and Emergency (A&E) for investigation of knee pain or following minor trauma, and Some studies use tools such as the Kellgren– considered ‘normal’ by a radiologist, have reported Lawrence scale based on radiological appearances: on joint space. In this population, there was little difference between weight bearing and non- • Grade 0 Normal weight bearing imaging; a finding believed to • Grade 1 Questionable: doubtful reflect the relative lack of ligament laxity in the narrowing of joint space and normal knee.42 Joint space in the healthy knee is possible osteophytic lipping consistently reported to be less in women than in 11

men. Dacre and colleagues42 reported that medial Change in joint space width and lateral joint spaces decreased with age, but as a predictor of symptoms were not associated with body mass index (BMI), Joint space narrowing over 2–3 years has been height or weight. Table 3 summarises measures of reported by some authors, to correlate with normal joint space.42 baseline pain scores and changes in measures of pain,47,51 but has not been a consistent finding by Osteoarthritic knee all.20 In OA of the knee, joint space volume varies among individuals. A number of approaches to Bingham and colleagues51 reported that a 60% systematically describing the changes seen on X-ray joint space narrowing in 2 years correlated with in OA of the knee have been reported. Several a 10 mm increase in pain, function and total grading systems have been used and these can be WOMAC scores using a 100 mm visual analogue split into two types: scale (VAS).

1. Semi-quantitative – defining the grade of joint Change in joint space width damage based on joint space loss, osteophyte as a predictor of joint formation, subchondral sclerosis, angular damage progression deformity and cyst formation. Various scoring systems have been reported since the original A correlation between joint space narrowing scale by Kellgren and Lawrence and accepted and knee arthroplasty has been reported, by the World Health Organization (WHO).43,44 with a minimum medial joint space narrowing 2. Quantitative – with formal measuring > 0.5 mm over 3 years being a predictor of knee (chondrometry) joint features. Again, various arthroplasty [relative risk (RR) 3.5, 95% CI 1.23 methods have been applied to how and where to 9.97] in the subsequent 5 years, but was based to measure.44 on small numbers of patients progressing to knee arthroplasty.52 Minimum medial joint space Single baseline measure as narrowing of > 0.6 mm in 3 years gave an RR of a predictor of severity of 5.16 (95% CI 1.76 to 15.12) of surgery.53 symptoms or progression Cicuttini and colleagues54 reported that joint space Joint space width at baseline was reported as a loss, as measured by magnetic resonance imaging predictor of complete joint space loss (score = 3 (MRI), correlated with knee arthroplasty. Of the 37 on a scale of 0–3) in a long follow-up study in the people with a rate of loss of < 3% (based on MRI at USA,45 with 50% of patients with a baseline joint year 0 and year 2), three required knee arthroplasty space width score of 1 progressing to complete by 4 years. In comparison, for the 36 with a tibial joint space loss in 12.03 years, and in 7.44 years if cartilage loss of > 8% per year, eight needed knee baseline joint space width score = 2. arthroplasty; an RR of 7.1 (95% CI 1.4 to 36.5) after adjustment for factors such as age, gender, A number of studies have reported little correlation BMI, etc.54 between baseline (one point in time) measures of joint space width (mean or minimum) and Issues with measuring symptoms of OA (pain, function or composite knee joint spaces measures such as WOMAC score).46–49 Comparisons Reproducibility of procedure for X-ray of joint space width with direct arthroscopy visualisation of the joint space also report poor In the diseased knee, the positioning during sensitivity and specificity.50 imaging is important, with substantial differences

TABLE 3 Summary of measures of normal joint space

Dacre 199142 Male Female Mean medial joint space (mm) 7.03 (6.91–7.15) 5.73 (5.58–5.88) Mean lateral joint space (mm) 7.44 (7.30–7.58) 6.44 (6.30–6.58)

12 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

in joint space measurements depending on weight Relation to clinical practice bearing and position. The OARSI issued a protocol While patients are often assessed using X-rays in for imaging the knee in guidelines published in clinical practice, pain is the major feature that 1996, recommending: determines when further intervention is required. The correlation between radiological changes and symptoms, as has been noted, is poor. X-rays are A standing (weight bearing) antero-posterior used widely in trials to assess disease progression, fully extended knee view.55 but this does not translate into clinical decision- making about when surgery is required. Since then various modifications have been tried. Variations in method do affect the absolute measures and require the position to be reproduced Interventions under in order to minimise within-study comparisons investigation or assessment of change over time.56–58 Protocols must also set out the imaging technical details if Glucosamine and chondroitin reproducibility over time is to be achieved, i.e. Both glucosamine and chondroitin are natural focus to film distance, radiological parameters and constituents of hyaline cartilage. With the control for joint rotation.59 exception of one preparation of glucosamine, in the UK they are considered, for regulatory Reproducibility of measurements purposes, to be food supplements. It has been Intraobserver reproducibility for joint space claimed that glucosamine and chondroitin reduce measurements has been reported to be high the pain of OA, but recent systematic reviews have [k > 0.88;49 k = 0.77 (0.72–0.82)].43 Inter-observer cast doubt on this.61,62 However, it has also been reproducibility was lower (lateral k = 0.6; medial suggested that glucosamine might have some effect k = 0.72).43,49 Grading joint space loss based on preserving cartilage in early OA, and hence on a scale and using comparative images as might slow down progression. examples improved intra- and inter-observer reproducibility.60 The use of automated systems to As a food supplement, glucosamine and measure joint space width from digital images is chondroitin can be bought by anyone, either over likely to reduce this source of error and has been the counter or by mail order, without the need for widely adopted in trials. a prescription. They come in different forms and doses, and may be taken on their own or mixed Impact of pain on positioning of the knee with food and drink. For example, glucosamine Knee pain has a major impact on a person’s may be added to a range of drinks, from iced tea to ability to fully straighten (extend) their knee joint. fruit smoothies. The degree to which extension is achieved may, therefore, be modified by treatments that modify Glucosamine pain. As a result, the joint space width would be Glucosamine is a naturally occurring amino artificially altered, reflecting not the cartilage sugar that is a building block for the complex volume, but rather the degree of extension proteins called glycosaminoglycans which are part achieved. This is a potential confounder in trials.59 of the structure of cartilage. Hyaline cartilage is composed of about 50% collagen, and the Assessment of knee OA remainder is mainly made up of proteoglycan progression in clinical trials molecules which serve to give the cartilage a viscoelastic resilience and thus to act as a cushion. Western Ontario and McMaster Universities The proteoglycan molecules (also called aggrecans) osteoarthritis index scores and measures of joint consist of numerous long-chain glycosaminoglycans space change have been adopted by OMERACT linked to a core protein. The glycosaminoglycans and regulatory bodies as the key outcome measures are repeating disaccharide units consisting of a for clinical trials in knee OA.40,59 Despite the hexuronic acid and a hexosamine (amino sugar). various caveats outlined above, these measures have Glucosamine (C6H13NO5) is the hexosamine been widely adopted by trialists. Few studies have constituent of keratan sulphate, which is found in achieved long-term follow-up of trial participants hyaline cartilage alongside the glycosaminoglycans to measure other hard clinical end points, such as chondroitin 4-sulphate and chondroitin 6-sulphate knee arthroplasty. (which are therefore much larger molecules than glucosamine).63 13

Two forms of glucosamine exist for use as The only other ‘prescription only medicine’ oral supplements: glucosamine sulphate and glucosamine product (glucosamine sulphate), glucosamine hydrochloride. Glucosamine is usually produced by Rotta, Italy, with a medicinal licence in derived from shellfish (chitin); however, it may some European countries is not currently available also be obtained from the chitin present in the cell in the UK. It is used at a dose of 1500 mg per day. walls of many fungi (which may be of advantage for people with shellfish allergy).64 Oral (either Food supplements in tablet form or as a powder dissolved in water), Glucosamine, chondroitin and combination intramuscular and even intra-articular preparations products are also available in the UK as non- exist. Doses vary considerably between and within licensed food supplements from health food shops, preparations. Commonly used dosages are between pharmacies and the internet. Food supplements are 1250 mg and 1500 mg daily. not subject to the same rigorous testing and control measures as licensed medicines and the products Chondroitin may vary in relation to combinations, compounds, Chondroitin is a large gel-forming molecule, which strengths and purities. forms part of cartilage and confers resistance to compression. Despite being a large molecule, it is Adebowale and colleagues66 studied 14 products partially absorbed from the diet or supplements. containing glucosamine sulphate or hydrochloride and 11 products containing chondroitin sulphate Chondroitin is used orally as chondroitin 4- and available on the US market. Deviations from label 6-sulphate. It is made from extracts of cartilaginous claims for glucosamine content ranged from cow and pig tissues (tracheal cartilage), or from as low as 25% to over 115% (although only two fish or bird cartilage.2 Chondroitin is usually products appeared to contain less than 90% of the taken orally, either in tablet form or as a powder label claim). For chondroitin sulphate, deviations dissolved in water, but intramuscular applications from the label claim ranged from less than 10% to have also been used. Commonly used dosages are 110%, with 4 of 11 products containing less than between 800 mg and 1200 mg daily, but again can 40% of the claimed chondroitin content. A further vary substantially between and within products. testing of 32 chondroitin sulphate products, using a phototrode titration method, found that over As regards the possible mechanism of action by half of the products contained less than 40% of which glucosamine and chondroitin might work the label claim (with all of the products costing in OA, the hypothesis is that oral supplements less than US$1 containing less than 10%). Barnhill of these compounds taken on their own or in and colleagues67 tested 20 chondroitin sulphate combination may help strengthen or repair products available in the US and found that none cartilage, thus having disease-modifying potential complied with the Food and Drug Administration in OA. We address in detail the biological (FDA) good manufacturing practices for plausibility of glucosamine and chondroitin as a pharmaceuticals. Table 4 provides an illustration of disease-modifying agent in OA in Chapter 5. the costs associated with different preparations.

Glucosamine and chondroitin Prescribing glucosamine and chondroitin availability in the UK Alateris (glucosamine hydrochloride) appears Licensed preparations in the BNF as the only product licensed as a ‘prescription only medicine’ available for In 2007, Alateris® (Randsom, UK) (glucosamine prescription in the UK. In June 2008, the Scottish hydrochloride) was licensed in the European Medicines Consortium (SMC) reviewed the Union and made available as a ‘prescription suitability of Alateris for use in NHS Scotland in only medicine’ in the UK. There were no clinical terms of clinical effectiveness and cost-effectiveness trials for this product to support the licensing and concluded that: application. The tablets are 625 mg and the recommended dosing is two tablets (1250 mg) per Alateris is not recommended for use within day. A 1-month supply was priced at £18.40 in the NHS Scotland for relief of symptoms in mild British National Formulary (BNF) in May 2008. As a to moderate osteoarthritis of the knee. No new product, Alateris is monitored under the UK direct clinical trial evidence of the efficacy Department of Health, Medicines and Healthcare and safety of this specific product is available. Products Regulatory Agency (MHRA) ‘black Randomised controlled trials of other triangle’ intensive surveillance scheme.65 formulations of glucosamine hydrochloride 14 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

TABLE 4 Example prices of preparations in the UK in 2008 (subject to special offers and deals between manufacturers and retailers)

Preparation Composition Price Valupak 30 tablets (400 mg Gl + 100 mg Ch) £0.99 Valupak Joint Care 30 tablets (500 mg Gl + 400 mg Ch) £4.99 Newmark 30 tablets (400 mg Gl + 100 mg Ch) £1.80 Newmark Glucosamine 30 tablets (500 mg Gl) £1.89 Natures Aid Glucosamine 90 tablets (100 mg Gl) £12.00 Alateris 60 tablets (625 mg Gl) £18.40 (BNF 2008)

BNF, British National Formulary; Ch, chondroitin; Gl, glucosamine.

indicate little or no benefit over placebo • Joint space is the main objective measure, but in improving symptoms in patients with there can be some subjectivity in quantifying. osteoarthritis of the knee. The link to long-term outcomes is not well established. In addition, the manufacturer did not present a sufficiently robust economic analysis to gain In undertaking this review and economic acceptance by SMC.68 evaluation, we were aware that a number of reviews had already been published. In general, these had Food supplements, as unlicensed products, are included studies of short duration and had not either ‘blacklisted’ as not for prescription or what focused on long-term outcomes. In this report, we is called ‘pay and report’, i.e. if a prescription were interested in the progression of knee OA and is issued, the pharmacist is reimbursed for the whether glucosamine and/or chondroitin could script but the prescription is reported to the local modify the disease process. We were, therefore, health board to follow up.69 The scale of use of only interested in trials of sufficient duration to glucosamine and chondroitin in the UK is not assess long-term outcomes. We have restricted well documented. One small survey in two GP our review to studies of a minimum of 12 months’ practices in the UK70 identified that 16% of people follow-up. with OA had used glucosamine and 5% had used chondroitin. In Scotland, in 2007, only 376 items The questions we planned to address in this report for glucosamine hydrochloride were dispensed are summarised below: on prescription (Mr D Pflegger, Robert Gordon University, 2009, personal communication). 1. Does glucosamine prevent or slow progression of OA of the knee? 2. Does chondroitin prevent or slow progression Key questions and rationale of OA of the knee? 3. If so, does the combination of both offer any Some of the findings in this chapter have greater effect? implications for the interpretation of the evidence, 4. If they are useful, what dosages should be including the following: used? 5. What are the costs, in terms of both the • OA of the knee is diagnosed mostly on clinical product costs and any side effects? grounds. 6. Are there any savings to the NHS from their • Progression is variable; symptoms improve in a use, for example by reduced consumption of significant minority. prescribed NSAIDs, or from avoidance of knee • Radiological changes do not usually regress, arthroplasty procedures? but often do not progress. 7. Is there evidence of cost-effectiveness from well • Clinical symptoms may not correlate with constructed economic evaluations? radiological changes. • Progression is slow, over many years. From the evidence review we sought to be able to • There are tried and tested instruments for come to one of three possible conclusions: assessing the impact of OA, such as WOMAC.

1. The evidence base is sufficient to conclude that term preservation of cartilage (which does they are not effective. imply longer-term reductions in symptoms, 2. The evidence base is sufficient to conclude that advanced OA and perhaps of the need for knee these products are clinically effective, in which arthroplasty), what should the comparator in case further research into clinical effectiveness trials be? is not required. 3. If the gain is in long-term outcomes, are there 3. The evidence is inconclusive, but suggestive short-term indicators of benefit which could be enough of benefit to justify further research. used, rather than waiting 20 years for advanced OA to become manifest? If further research was indicated, then we planned 4. Which variables or assumptions (e.g. costs, to address several further questions, where the data QoL, mortality, extrapolation of effectiveness permitted: over time) are most important in generating uncertainty in the cost-effectiveness estimates, 1. Are the alleged benefits of glucosamine and and which would provide the greatest return in chondroitin on knee cartilage biologically terms of reducing uncertainty at a reasonable plausible? cost. These variables and assumptions will 2. If, as suggested by the commissioning brief, be identified using value of information the evidence suggests that there is no gain in analysis, and will form the main basis in short-term symptom relief, but only in long- recommendations for further research.

16 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Chapter 3 Clinical effectiveness of glucosamine and chondroitin on the progression of OA of the knee

n this chapter we cover the evidence of clinical Where data permitted, glucosamine sulphate Ieffectiveness of glucosamine and chondroitin on and hydrochloride studies were to be analysed the progression of OA over at least 12 months. A separately. Trials of the Rotta product of the number of reviews have been published in recent sulphate form were also to be analysed separately. years. Many of these have included studies of short duration which, while adequate for addressing Recent NICE guidelines reiterate that none of short-term effects, including the effect on pain and the existing treatments have been demonstrated function in the short term, are not adequate for to modify OA disease progression. The main studying whether these treatments have an effect comparator was therefore placebo, but comparisons on the natural history and clinical progression of with analgesics and non-pharmacological the disease. interventions were not excluded. Different comparators would have been explored by sensitivity analysis if appropriate data had been Methods found. Inclusion criteria Types of outcome measures Types of studies Because this review was prompted by the In the first instance, we reviewed systematic reviews suggestion that glucosamine and chondroitin might of RCTs where the review included at least one have structural effect, systematic reviews were only trial which met our duration inclusion criteria. considered for inclusion if they assessed both pain/ The acceptable minimum duration of follow-up function and a measure of joint structure, such as was 12 months, i.e. reviews had to include at least joint space loss. Randomised controlled trials were one primary study with a study duration and/or considered if they assessed function or structure. duration of follow-up of 12 months. Data on the following outcome measures were We then used these reviews as a source to identify extracted: RCTs of at least 12 months’ duration. We also considered any additional RCTs published after • pain the last search of any relevant included review, or • function any RCTs not included in any of the reviews, but • composite measures of health status (WOMAC fulfilling our inclusion criteria. score, Lequesne index, SF-36 or similar) • joint space loss (as defined in the trials) Types of participants • knee arthroplasty People with OA of the knee were included. No • adverse events. restrictions were made regarding stage of OA, but information on OA stage was noted as indicated by Search strategy the reviews. The search strategy comprised the following main Types of interventions elements: Treatments with glucosamine or chondroitin taken in pharmacological form (capsule/tablet or powder) • searching of electronic databases were included (any salts and any doses, with • scrutiny of bibliographies of retrieved papers glucosamine and chondroitin being used alone or • checking for details of registered, but in combination). Topical applications or injections unpublished trials were excluded. Glucosamine or chondroitin taken • internet searches as additives to drinks were not included. 17

• the Food Standards Agency website for details 10. was the statistical analysis appropriate? on safety. Studies were considered to be systematic reviews Electronic databases included: MEDLINE and if they described inclusion criteria, gave details of PubMed; EMBASE; the Cochrane Library their literature searches and fulfilled at least half (including the Cochrane Systematic Reviews of criteria 3 to 10. In addition, the reviews had Database, CENTRAL, DARE, NHS EED and HTA to fulfil the remaining inclusion criteria for our databases); Allied and Complementary Medicine review, i.e. they (i) summarised RCTs, (ii) included (AMED); National Research Register (NRR); Web at least one RCT of at least 1 year’s duration, of Science Proceedings; Current Controlled Trials; (iii) included at least one RCT including patients and Clinical Trials.gov. with OA of the knee, (iv) considered the effects of glucosamine sulphate or hydrochloride or To identify additional recent RCTs that were chondroitin sulphate, alone or in combination, published after the publication of high quality compared with placebo, and (v) reported data on reviews, searches for RCTs were carried out back to both symptoms/pain and structure/joint space. the beginning of 2004 for glucosamine and back to 2005 for chondroitin. A separate search was done Studies were classified as being high, moderate or to identify studies specifically investigating adverse poor quality, according to the following criteria: events of glucosamine and/or chondroitin. See Appendix 1 for details. • high quality: all criteria fulfilled, or not more than one of ‘description of study selection’ or We considered only papers published in English, ‘description of study flow’ not fulfilled German, French, Italian and/or Spanish. Reports • moderate quality: more than one of the quality published as meeting abstracts only (with criteria not fulfilled, but most important insufficient methodological details reported to methodological aspects described [inclusion allow critical appraisal of study quality) were criteria, literature search, quality assessment, excluded, except as a guide to trials not yet details on individual studies (including quality), published in full. appropriate statistical analysis] • poor quality: any others.

Study selection The quality of RCTs, identified by any of the Titles and abstracts were examined for inclusion by methods outlined above, was assessed according one reviewer. Full copies were obtained of papers to the following criteria [modified from the that appeared to fulfil the inclusion criteria (or criteria suggested by the Centre for Reviews and where there was doubt) and these were checked by Dissemination (CRD) report on systematic review two reviewers independently for final inclusion. methodology]:72 There were no disagreements. • description and method of randomisation • allocation concealment Quality assessment • blinding of participants to active agent or Potential systematic reviews were assessed according placebo to the following criteria, which are based on the • blinding of outcomes assessors criteria of the Quality Of Reporting Of Meta- • numbers of participants randomised, excluded analyses (QUOROM) statement:71 (with reasons) and lost to follow-up • whether intention-to-treat analysis was 1. inclusion criteria described (study design, performed participants, interventions, outcomes) • methods for handling missing data 2. details of literature search given (databases, • appropriateness of statistical analysis. dates, keywords, restrictions) 3. study selection described 4. data extraction described Data extraction 5. study quality assessment described Data were extracted independently by one reviewer 6. study flow shown using a standardised data extraction table and 7. study characteristics of individual studies checked by another. Only a small number of described discrepancies occurred, and these were resolved by 8. quality of individual studies given discussion and by checking the original article. 18 9. results of individual studies shown DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Data analysis Review of systematic reviews The systematic reviews included were compared Search results according to the following criteria: The searches of MEDLINE, EMBASE and the Cochrane Library combined (as shown in • conclusions reached with respect to the Appendix 1) retrieved 677 articles. The duplicates effectiveness of glucosamine and/or were removed, and the remaining abstracts checked chondroitin (in relation to relevant outcome against the inclusion criteria. This resulted in 23 measures) articles, for which the full papers were obtained for • any differences in conclusions reached and the further checking. The QUOROM flow diagram of reasons for this; for example, inclusion criteria search results is shown in Figure 2. for studies, number and nature of studies included, outcome measures considered, Of the 23 potential relevant systematic reviews, importance placed on study quality or study five were included in the analysis. Twelve of the duration, different forms of glucosamine remaining reviews were excluded because they were considered separately (glucosamine sulphate not systematic reviews (i.e. they did not include and glucosamine hydrochloride), etc. a description of their methodology and did not • usefulness in addressing our research question systematically summarise the data). The remaining about the clinical effectiveness of glucosamine six were systematic reviews, but one was a review and/or chondroitin in modifying disease of nutritional supplements in OA specifically progression over at least 12 months. excluding glucosamine and chondroitin, one only reported short-term data (up to a maximum of 12 Data from systematic reviews were tabulated and weeks) and the remaining four did not consider discussed in a narrative review. Study characteristics structural outcomes (the bibliographies of these of the primary studies included in the reviews four reviews were searched to ensure no individual were compared against the inclusion criteria of the studies met our inclusion criteria for trials). present review. Characteristics and quality of RCTs Details of reasons for study exclusion are shown in fulfilling the inclusion criteria of this review (both Appendix 2. those included in previous reviews and studies newly identified) were also tabulated. Of the remaining five reviews, two dealt with glucosamine only,62,73 two dealt with chondroitin For the RCTs, where data permitted (i.e. more only 61,74 and one dealt with both glucosamine and than one study available for a given parameter chondroitin.75 All five reviews included a meta- and parameters measured in a comparable way) analysis. Additionally, a recent NCCCC guideline data from primary studies were summarised for the management of OA was identified, which in a meta-analysis using the Cochrane Review included a systematic assessment of glucosamine Manager Software 5.0.4. Continuous data were and chondroitin.1 This guideline was used to summarised as weighted mean differences using inform the NICE guideline for the management both fixed- and random-effects models. Where the of OA29 and was the only included review that point estimate was the same, the random-effects considered combination therapy with glucosamine model was quoted because the confidence intervals and chondroitin. Table 5 shows details of the five better reflect the uncertainty around the point reviews and one guideline included. estimates. No dichotomous results were analysed. Heterogeneity was assessed using the chi-squared Quality of included reviews test. For outcomes where no meta-analysis was Four of the five included reviews were classified as possible, results were summarised narratively. being high quality reviews, with three61,62,75 fulfilling all the quality criteria and one73 failing to give details of study selection. One review74 was rated Results as moderate quality, as neither study selection nor data extraction were described and details on The results have been presented in two sections. statistical analysis were lacking. The guideline1 First, we present the findings from the review of described certain methodological aspects rather systematic reviews. Secondly, we report the results briefly, but referred to methods manuals which of the review of the primary RCTs that met the appeared to assure a high quality systematic criteria for this review. reviewing process.

Search of MEDLINE, AMED and Remove duplicates and studies EMBASE Oct 2007 (main search and not relevant to topic specific searches) n = 677

RCTs: Systematic reviews: From reviews n = 46 selected from titles and From searches n = 8 Excluded: n = 47 abstracts n = 23 additional trials

Excluded: n = 18

Included systematic reviews: Included RCTs: Autoalerts (Oct From original search n = 5 From original search n = 6 2008): n = 1 From autoalerts n = 1 From searches n = 1 included trial guideline From autoalerts n = 1

FIGURE 2 QUOROM flow diagram of search results.

Inclusion criteria in the review by Towheed and colleagues62 for Key inclusion criteria and review characteristics glucosamine and Reichenbach and colleagues61 are shown in Table 6. The inclusion criteria varied for chondroitin, and these two reviews also tended substantially between reviews. One review73 had to provide the most detail. Therefore, most of the narrow inclusion criteria, looking at double-blind detail presented on the primary studies derives RCTs lasting at least 12 months and studying the from these two reviews. effect of oral glucosamine in primary knee OA. All the studies included in this review, therefore, The glucosamine trials included between 20 and met with our inclusion criteria of 12 months’ 1583 participants. With the exception of two trials, duration/follow-up. The other four reviews and one the proportion of women included was generally guideline included at least one study of a minimum greater than the proportion of men (up to 91% of 12 months’ duration (so were included here), women, with two trials including 48% and 5% but had wider inclusion criteria and, therefore, also women). The mean age range of participants included a number of studies of shorter duration. was between 51 and 75 years. Study duration was between 3 weeks and 3 years, with 15 trials having Trials included in systematic reviews a study duration of no longer than 3 months The five reviews and one guideline described and two having a duration of at least 1 year. The a total of 45 controlled trials – 21 studying chondroitin trials included between 17 and 622 glucosamine, 20 studying chondroitin – and participants. As in the glucosamine trials, there four studying a combination of chondroitin were generally more women than men in the and glucosamine (one of which compared trials (33–94%), and mean age was between 50 combination therapy with glucosamine or and 67 years. Trial duration was between 6 weeks chondroitin monotherapy). Table 7 summarises and 2 years, with the trials generally having a the characteristics of the 45 primary studies. longer duration than the glucosamine trials (one The NCCCC OA guideline1 included two had a duration of less than 3 months; five had a systematic reviews61,62 and six additional RCTs. duration of at least 1 year). One trial, studying a For glucosamine, only two RCTs (the long- combination of chondroitin and glucosamine, had term studies by Reginster and colleagues76 and 1583 participants with a mean age of 58 years and Pavelká and colleagues77) were included in all the 27% women, and a trial duration of 24 weeks. relevant reviews. For chondroitin, only six studies were included in all the relevant reviews. The Only one of the single therapy and two of the largest number of primary studies was included combination therapy trials gave glucosamine 20 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52 continued Quality Inclusion criteria described: yes Details of literature search given: yes Study selection described: no Data extraction described: yes Study quality assessment described: yes Study flow shown: yes Study characteristics of individual studies described: yes Quality of individual studies given: yes of individual studies Results shown: yes Statistical analysis appropriate: yes Overall quality: high Included studies Number of included trials: 2 Number of participants: 414 (range 202–212) quality: Jadad quality scores 4 Trial and 5 Study characteristics (comment): both trials fulfil the inclusion criteria of this review; see details the respective trials below Inclusion criteria and methodology Inclusion criteria study duration at least 1 year Study design: double-blind RCTs; patients with primary knee OA Participants: Interventions: oral glucosamine Outcomes: primary outcome – joint space narrowing; additional outcomes symptom modification, adverse events Methodology Search strategy: MEDLINE, EMBASE, BIOSIS, Evidence-Based Medicine Cochrane Library; search from inception of database; (EBM) Reviews, scanning of reference lists; search terms indicated; date last August 2004 Study selection: no details on study selection given Quality assessment: the methodological quality of each study was assessed using the scale developed by Jadad Data extraction: data were extracted independently by two authors; disagreements were resolved by consensus Data analysis: meta-analysis; statistical methodology appropriate Subgroups/sensitivity analyses: none 73 Characteristics of included systematic reviews and guideline Glucosamine 2005 Poolsup Study Thailand Focus: structural and symptomatic efficacy and safety of glucosamine in knee OA

TABLE 5 TABLE 21

22 5 TABLE DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52 continued Quality Inclusion criteria described: yes Details of literature search given: yes (but limited) Study selection described: no Data extraction described: no Study quality assessment described: yes Study flow shown: yes Study characteristics of individual studies described: yes Quality of individual studies given: yes of individual studies Results shown: yes Statistical analysis appropriate: unclear Overall quality: moderate Included studies Number of included trials: 7 Number of participants: 909 quality: demographic baseline Trial data well matched; quality scores: 6 7 for two studies, 9 for one study, 11 for two studies, 13 one study, one study Study characteristics (comment): three studies lasted a year or longer; authors noted that pain evaluation was biased owing to concomitant consumption of analgesics Inclusion criteria and methodology Inclusion criteria Study design: randomised, double-blind, placebo-controlled trials; study duration at least 4 weeks of the knee or hip patients with OA Participants: Interventions: oral chondroitin sulphate investigator’sOutcomes: symptoms – Lequesne index, WOMAC, or patient’s pain, walking time, NSAID or analgesic consumption; global assessment, VAS structure – mean joint space width, surface area, minimum narrowing; adverse events Methodology Register; Search strategy: MEDLINE/PubMed, Cochrane Controlled Trials published studies to limited indicated; terms search lists; reference of scanning search limited to studies published in peer-reviewed in English or French; journals between 1980 and 2005; date of last search 2005 Study selection: no details on study selection given Quality assessment: 14-item quality instrument used by two reviewers; differences resolved by consensus Data extraction: no details on data extraction given Data analysis: meta-analysis; statistical methodology appropriate; however, in two cases the authors included parts of a single study their meta- analyses as though they were separate studies – bias? Subgroups/sensitivity analyses: none 74 Chondroitin Bana 2006 Study France Focus: symptomatic and structural efficacy and safety of chondroitin sulphate 23

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Clinical effectiveness of glucosamine and chondroitin on the progression of OA of the knee Quality Inclusion criteria described: yes Details of literature search given: yes Study selection described: yes Data extraction described: yes Study quality assessment described: yes Study flow shown: yes Study characteristics of individual studies described: yes Quality of individual studies given: yes of individual studies Results shown: yes Statistical analysis appropriate: yes Overall quality: high Included studies Number of included trials: 22 (20 included in meta-analysis) Number of participants: 4056 (range 17–631) quality: recently performed trials Trial tended to be larger and of higher quality; generation of allocation sequence and allocation concealment unclear for 19 trials; only three trials used intention-to-treat analysis; most trials did not describe approaches for handling missing data; withdrawal rates reported in 13 trials (between 0 and 0.7, most between 0.02 0.27); all but three reported to be double blind Study characteristics (comment): most studies had a short study duration; some studies of non- oral chondroitin and non-knee applications included; noted that unequal consumption of analgesics between comparison groups was an unlikely reason for small effects found mg/day administered orally) 400 Inclusion criteria and methodology Inclusion criteria Study design: randomised or quasi-RCTs of the knee or hip patients with OA Participants: Interventions: chondroitin vs placebo or no treatment; trial groups given low doses excluded (< Outcomes: primary – pain [at end of trial or at a maximum 3 months after termination of chondroitin therapy (whichever came first)]; when study provided data on more than one pain scale, previously described hierarchy of pain-related outcomes referred to; secondary – changes in mean and minimum joint space width; adverse events (numbers/withdrawals because of/serious) Methodology Search strategy: Cochrane CENTRAL (1970–2006), MEDLINE (1966–2006), EMBASE (1980–2006), CINAHL (1982–2006); detailed search strategy shown; no language restrictions; extensive supplemental searches; date of last search 30 November 2006 Study selection: two reviewers independently evaluated reports for eligibility; disagreements resolved by discussion Quality assessment: two reviewers independently assessed concealment of treatment allocation, blinding and adequacy of analyses (handling missing data, intention-to-treat analysis); disagreements resolved by discussion with a third reviewer and consequent consensus; table presented listing study design, allocation concealment, reported to be double blind, adequate blinding of patients, adequate therapists, described as placebo controlled, withdrawal rate in chondroitin group, control group, intention to treat analysis performed, method handle missing data Data extraction: in duplicate, using standardised form; only first halves of crossover trials used; authors contacted for additional information where necessary; disagreements resolved by discussion with a third reviewer and consequent consensus Data analysis: meta-analysis; statistical methodology appropriate Subgroups/sensitivity analyses: analyses stratified by – concealment of allocation, use of a placebo control, patient blinding, adequacy analyses in accordance with the intention-to-treat principle, trial size (cut-off 200 patients), funding, route of administration, length follow-up (cut-off 26 weeks), difference in use of co-interventions in trial groups; included in univariable meta-regression – chondroitin dosage (in trials with oral administration), treatment duration, length of follow-up 61 Characteristics of included systematic reviews and guideline (continued) Reichenbach 2007 Reichenbach Study Switzerland/Germany/UK Focus: effects of chondroitin on pain in patients with OA

24 5 TABLE DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52 continued Criteria described: yes Details of literature search given: partially Study selection described: yes Data extraction described: partially Study quality assessment described: partially Study flow shown: no Study characteristics of individual studies described: partially Quality of individual studies given: yes of individual studies Results shown: yes Statistical analysis appropriate: yes Overall quality: high/moderate Quality /chondroitin) Trial quality: all glucosamine Trial trials described as high quality/ methodologically sound; all chondroitin and trials of combined glucosamine/chondroitin trials described as methodologically sound Study characteristics (comment): 2005 and Reichenbach Towheed 2007 summarised above; of the included, all have additional RCTs been considered by the present review and were either included or not considered eligible Included studies Number of included trials: two 2005, systematic reviews (Towheed 2007), six additional Reichenbach RCTs Number of participants: 10,342 (4504 glucosamine trials, 3999 chondroitin trials, 1839 combination glucosamine 40 excluded Inclusion criteria and methodology Inclusion criteria (observational studies); trials with Study design: systematic reviews, RCTs, sample size < adults with OA Participants: Interventions: glucosamine or chondroitin alone in compound form vs placebo Outcomes: symptoms, function, QoL, structural changes Methodology Search strategy: MEDLINE (1966 to April 2007), EMBASE (1980 to April 2007), CINAHL (1982 to April Cochrane Library (April AMED (1985 to April 2007); search terms not indicated; papers published/accepted journals included; conference paper abstracts and non- by peer-reviewed English language papers excluded Study selection: exclusion lists generated with reasons for exclusions; full papers obtained where relevant Quality assessment: critical appraisal checklists compiled for each paper by one researcher; compliant with NICE technical manual; evidence considered by guideline development group for accuracy and completeness Data extraction: data extracted by one researcher; evidence considered guideline development group for accuracy and completeness Data analysis: evidence tables and synthesis; statements graded Subgroups/sensitivity analyses: none 1 Glucosamine/chondroitin NCCCC guideline 2008 Study UK Focus: management of OA 25

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Clinical effectiveness of glucosamine and chondroitin on the progression of OA of the knee Quality Criteria described: yes Details of literature search given: yes Study selection described: yes Data extraction described: yes Study quality assessment described: yes Study flow shown: yes Study characteristics of individual studies described: yes Quality of individual studies given: yes of individual studies Results shown: yes Statistical analysis appropriate: yes Overall quality: high Included studies Number of included trials: 15 (seven glucosamine trials, eight chondroitin trials) Number of participants: 1775 (1020 glucosamine trials, 775 chondroitin trials) (range 24–319) quality: glucosamine trials Trial – three scored 4/5 and four 5/5; chondroitin trials – five scored 3/5, one 4/5, 5/5, had insufficient detail; individual demographic baseline data well matched, demographic baseline data for glucosamine and chondroitin trials also well matched Study characteristics (comment): many short-term studies; all studies of knee OA Inclusion criteria and methodology Inclusion criteria Study design: randomised, double-blind, parallel, placebo-controlled trials performed between January 1980 and March 2002; treatment period at least 4 weeks of the knee or hip patients with OA Participants: Interventions: oral glucosamine or chondroitin Outcomes: at least one of the following outcomes: joint space narrowing, for mobility assessment, pain, VAS VAS Lequesne index, WOMAC, responders to treatment, adverse effects Methodology HealthSTAR, BIOSIS Previews, Search strategy: MEDLINE, PreMEDLINE, Current Contents, EBM Register, EMBASE, Cochrane Controlled Trials search for studies published between January 1980 and March 2002; Reviews; search terms indicated; no language restrictions; extensive supplemental searches; date of last search March 2002 Study selection: potentially relevant publications identified were reviewed by two independent authors for eligibility; in the case of disagreement, a third author was consulted; authors’ names and sources were blinded for the selection process Quality assessment: trials fulfilling the inclusion criteria were blindly scored by two reviewers using the Jadad scale; differences were resolved by consensus; authors’ names and sources were blinded for quality assessment Data extraction: data for predefined outcomes were extracted blindly by two authors using a standardised form; third reviewer was consulted in case of disagreement Data analysis: meta-analysis; statistical methodology appropriate; data for glucosamine and chondroitin were analysed together after checking for all outcomes that there were no significant differences between the two compounds Subgroups/sensitivity analyses: none 75 Characteristics of included systematic reviews and guideline (continued) Study Richy 2003 Belgium/France Focus: structural and symptomatic efficacy of glucosamine sulphate and chondroitin sulphate in knee OA VAS, visual analogue scale. randomised controlled trial; NSAID, non-steroidal anti-inflammatory drug; QoL, quality of life; RCT,

26 5 TABLE DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

TABLE 6 Summary of review characteristics

NCCCC Reichenbach Poolsup 200573 Towheed 200562 Bana 200674 guideline 20081 200761 Richy 200375 Inclusion Inclusion Inclusion Inclusion Inclusion Inclusion RCTs, double RCTs, single or RCTs, double Systematic Randomised or RCTs, double blind, duration at double blind blind, duration at reviews, RCTs quasi-randomised blind, duration at least 1 year Primary or least 4 weeks OA in adults controlled trials least 4 weeks Primary knee OA secondary OA OA of the knee Glucosamine or OA of the knee OA of the knee Glucosamine vs Glucosamine vs or hip chondroitin alone or hip or hip placebo placebo or active Oral chondroitin or in compound Chondroitin vs Oral glucosamine Joint space intervention sulphate vs form vs placebo placebo or no or chondroitin vs narrowing, Pain, range of placebo Symptoms, treatment placebo symptoms, motion, functional Symptoms, pain, function, QoL, Pain, joint space Joint space width, adverse events assessment, structure, adverse structural changes width, adverse function, pain, Quality: high structural benefits, events Quality: moderate events adverse events adverse events Included trials: 2 Quality: moderate Included trials: Quality: high Quality: high (414 participants) Quality: high Included trials: 7 2 systematic Included trials: 22 Included Included trials: 20 (909 participants) reviews, 6 (4056 participants) trials: 15 [7 (2596 participants) additional glucosamine (1020 RCTs (10,342 participants), 8 participants) chondroitin (775 participants)]

QoL, quality of life; RCTs, randomised controlled trials.

hydrochloride rather than glucosamine sulphate. therapy trials, three had a duration of less than One single therapy trial switched from glucosamine a year, one studied a topical application of sulphate to glucosamine hydrochloride halfway glucosamine and chondroitin, and two examined through the trial owing to supply problems for no structural outcomes. These combination therapy the glucosamine/placebo product initially used. studies were included in the NCCCC OA guideline Towheed and colleagues62 reported data for nine 2008.1 trials using the Rotta preparation of glucosamine and for eight trials using a non-Rotta preparation. Appendix 3 summarises the results of the meta- None of the studies included Alateris, the only analyses of the included reviews and Appendix licensed glucosamine product in the UK. 4 shows the results of any relevant subgroup or sensitivity analyses. Fundamental to our research Findings question, none of the reviews presented data The findings of the reviews need to be compared separately for trials of longer duration for all with caution; some reviews were more inclusive outcomes. For structural changes, most of the than others. Of the glucosamine trials, three trial data have been limited to longer trials, so did not clearly include only knee OA, four gave conclusions from three of the reviews were based glucosamine by injection and five had active rather solely on trials of more than 1-year follow-up. The than placebo control groups (three compared other two reviews, and the guideline, based their with ibuprofen, two compared with piperazine conclusions about structural changes on a mixture plus chlorbutanol). These glucosamine trials of shorter- and long-term studies were included in the review by Towheed and colleagues,62 but in none of the others. Of the As the NCCCC OA guideline1 focused mainly on chondroitin trials, one considered hip rather than two systematic reviews already included here and knee OA and two gave chondroitin by injection. the only additional RCT eligible for the present These chondroitin trials were included in the review is already described below, the results of that review by Reichenbach and colleagues,61 but in guideline are not shown in further detail. Table 8 none of the others. Reichenbach and colleagues,61 summarises the conclusions and recommendations also included several studies published in abstract of the reviews. Trials and reviews are hereafter form only, although additional information was referred to by the name of the first author. obtained where possible. Of the combination 27

TABLE 7 Distribution of primary studies in the included reviews

Glucosamine only Chondroitin only Both

Poolsup Towheed Bana Reichenbach NCCCC Richy 200565 200573 200674 200761 20081 200375

Glucosamine Cibere 200478   Crolle 198079  D’Ambrosio 198180   Drovanti 198081  Herrero-Beaumont 200782  Houpt 199983   Hughes 200284   McAlindon 200485   Müller-Fassbender 199486   Noack 199487    Pavelká 200277    Pujalte 198088   Qiu 199889   Reginster 200176    Reichelt 199490  Rindone 200091    Rovati 199792   Usha 200493   Vajaradul 198194  Vaz 198295   Zenk 200296 

Chondroitin Alekseeva 199997   Bourgeois 199898     Bucsi 199899     Conrozier 1992100   Fleisch 1997101   Kahan 2007102   Kerzberg 1987103   L’Hirondel 1992104    Mazieres 1992105     Mazieres 2001106     Mazieres 2007107   Michel 2005108    Morreale 1996109   Nasonova 2001110   Pavelká 1999111    Rovetta 1991112  

28 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

TABLE 7 Distribution of primary studies in the included reviews (continued)

Glucosamine only Chondroitin only Both

Poolsup Towheed Bana Reichenbach NCCCC Richy 200565 200573 200674 200761 20081 200375 Soroka 2002113   Uebelhart 1998114     Uebelhart 1999115   Uebelhart 2004116    

Glucosamine/chondroitin Clegg 2006117   Cohen 2003118  Das 2000119  Rai 2004120 

Glucosamine no evidence of an adverse metabolic effect of Structure For glucosamine, only two trials were glucosamine. included in the analysis of structural outcomes in all the three relevant systematic reviews (including Chondroitin the Richy review where only glucosamine trials were Structure Both the review by Bana74 and the review included in the analysis of structural outcomes), by Reichenbach61 came to similar conclusions with namely the trials by Reginster76 and Pavelká.77 Both respect to the structural benefits of chondroitin the Poolsup73 and Towheed62 reviews concluded sulphate. Both suggested a small effect on that glucosamine may delay the structural joint space width in favour of chondroitin, with progression of OA of the knee. uncertain clinical significance (and possibly in the absence of a functional benefit). They considered Symptoms/pain While Poolsup73 concluded that the same two trials of more than one year follow- their results showed evidence in support of an up (with the addition of one further shorter study improvement of symptoms with glucosamine included in the Reichenbach review61). sulphate, Towheed62 was more cautious. They stated that overall, glucosamine was shown to have Symptoms/pain Both Bana74 and Reichenbach61 a moderate clinically significant effect on pain were cautious in their interpretation of the data compared with placebo. With respect to function, with respect to benefits of chondroitin sulphate results were mixed, with significant results for the on joint pain and function. Bana74 concluded that Lequesne index, but not for the WOMAC total there was a true, but modest, effect of pain relief score or any of the WOMAC subscores; in their and improvement of function but also concluded subgroup analyses, results were only significant for that the results were probably biased owing to the pain and the WOMAC total score for studies using poor quality of the data. They pointed out that the Rotta preparation of glucosamine, while they pain evaluation was complicated in most studies as were non-significant for trials using a non-Rotta concomitant analgesic or NSAID use was allowed, preparation. Poolsup73 only considered long-term but actual consumption was often not reported studies (whereas Towheed62 included mainly short- in detail. They also observed a latency period of term studies) and commented that an effect on about a month before an advantage of chondroitin physical function may only be achieved with long- with respect to pain relief and function was seen, term use of glucosamine. as well as an effect lasting beyond the cessation of treatment, which they thought might justify cycles Adverse events Both reviews concluded that of treatment to be used. glucosamine was safe. Towheed62 did, however, mention the concern and uncertainty with respect Reichenbach61 stressed the high degree of to a possible effect of glucosamine on glucose heterogeneity between studies and the poor quality metabolism; while not including this particular of many studies and concluded that large-scale question in their review, they did quote two studies methodologically sound trials indicated that the specifically addressing this question that found symptomatic effect of chondroitin sulphate was 29

TABLE 8 Conclusions and recommendations of reviews

Recommendations for Study Authors’ conclusions research Additional comments

Glucosamine Poolsup 200573 Pain/function: The results showed High quality, placebo- An effect on physical evidence in support of an improvement of controlled long-term trials function may only be symptoms by glucosamine sulphate of various forms (different achieved during long-term Structure (based on two studies of > 1-year salts, different forms of use follow-up): The results showed evidence glucosamine sulphate) of in support of a delay of structural glucosamine are needed progression by glucosamine sulphate Adverse effects: The evidence suggests that glucosamine sulphate is safe in long-term use Towheed 200562 Pain/function: Overall, glucosamine showed Questions remaining to be There is uncertainty a moderate clinically significant effect answered about the effect of on pain compared with placebo. Pooled Are different glucosamine glucosamine on glucose results from studies using a non-Rotta preparations sold by metabolism with long- preparation of glucosamine or adequate different manufacturers term use – no evidence allocation concealment failed to show equally effective and safe in of an adverse effect was benefit in pain and WOMAC function, the therapy of OA? found by two studies while those studies evaluating the Rotta preparation showed that glucosamine was Is glucosamine sulphate superior to placebo in the treatment of equally effective as pain and functional impairment resulting glucosamine hydrochloride? from symptomatic osteoarthritis. Is there any further benefit WOMAC outcomes of pain, stiffness and obtained by using mixed function did not show a superiority of glucosamine preparations glucosamine over placebo for both Rotta that contain additional and non-Rotta preparations of glucosamine therapeutic products, such Structure (based on three studies of variable as chondroitin sulphate? follow-up): Glucosamine may modify Is glucosamine helpful for all the natural radiological progression stages of OA severity? of osteoarthritis of the knee (delaying What is the optimal dose progression) for maximising efficacy and Adverse effects: Glucosamine was as safe as minimising toxicity? placebo What are the patient specific predictors of favourable effects on radiological progression of OA?

Chondroitin Bana 200674 Pain/function: The results analysed Further large studies lasting Pain evaluation is biased suggested a true, but modest effect of at least 2 years are needed as in most studies chondroitin sulphate on relief of pain and to confirm any effects of concomitant analgesic/ improvement of joint function in patients chondroitin sulphate on NSAID use is allowed – treated. However, the authors concluded preservation of structure/ but often not reported in that the evaluation could be biased, as only joint space width detail few methodologically sound papers were Studies needed to ascertain A latency period of about available for which OA patients 1 month was observed Structure (based on three studies > 1 year): these drugs are useful before an advantage of Structure (mean and minimum joint space (e.g. patients with mild to chondroitin for pain relief/ width) appears to be preserved with moderate disease) function could be seen chondroitin sulphate (versus deterioration The activity of chondroitin in the placebo groups), but in the main appears to last for up to study showing a structural advantage, no 3 months after the drug effect on the WOMAC score was seen is stopped, so cycles of Adverse effects: The tolerance of treatment/no treatment chondroitin sulphate can be considered as might be justified excellent 30 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

TABLE 8 Conclusions and recommendations of reviews (continued)

Recommendations for Study Authors’ conclusions research Additional comments Reichenbach 200761 Pain/function: There was a high degree Future trials should A significantly larger effect of heterogeneity between trials, making adhere to methodological of chondroitin was seen interpretation difficult; most studies were standards reducing possible in trials in which use of poor quality; large-scale, methodologically biases; reporting should rescue analgesics was sound trials indicate that the symptomatic adhere to standards such as reported to be higher in benefit of chondroitin is minimal or non- the CONSORT criteria the control groups existent A rigorously designed, Concomitant use of Structure (based on five studies of varying adequately powered medication (analgesics, duration): A small effect on joint space randomised placebo- NSAIDs) was poorly width in favour of chondroitin was seen, controlled trial restricted to reported – but small/ with uncertain clinical significance patients with low-grade OA lack of effect seen was Adverse effects: No evidence was found would be needed to assess unlikely to be related to that chondroitin is unsafe if chondroitin is more concomitant analgesic/ effective in patients with NSAID use Practice recommendation: The use of mild disease than in patients Recently performed chondroitin in routine clinical practice with more severe disease should be discouraged. In patients with methodologically sound advanced OA an effect is unlikely; in trials with effect sizes patients with low-grade OA use of near zero tended to have chondroitin should be restricted to RCTs lower proportions of patients with low-grade OA – but the effect of chondroitin may depend on OA stage

Glucosamine/chondroitin NCCCC guideline General/pain/function/adverse events: The Study of short- and long- 20081 evidence from the trials is often difficult term effects in the very to compare owing to differences in the elderly products used (and their bioavailability), Benefits of individual and between-study populations, patient BMI, combination OA therapies and the use of analgesia at the time of in people with multiple joint pain and function assessment in the trials. region pain Overall, trials using glucosamine sulphate as a single dose of 1500 mg rather than Identifying subsets of hydrochloride 500 mg three times a people with OA in whom day showed a small benefit compared existing treatments are with placebo for treatment of knee OA. more beneficial and cost- However, at the time the guideline was effective prepared, the hydrochloride preparation has been granted a European Medicines Evaluatory Agency licence, while the sulphate has not. The evidence for efficacy of chondroitin was less convincing Evidence to support the efficacy of glucosamine hydrochloride as a symptom modifier is poor. For the non-licensed product (glucosamine sulphate), the evidence is not strong enough to warrant recommending that it should be prescribed on the NHS. Notwithstanding some evidence of benefit and very little evidence of harm in clinical practice, and despite the extra scrutiny these agents have received, the economic cost–consequence table shows that only glucosamine sulphate is potentially cost-effective out of the interventions considered in this section.

continued

TABLE 8 Conclusions and recommendations of reviews (continued)

Recommendations for Study Authors’ conclusions research Additional comments A wide range of ICERs were reported and the poorest estimates of efficacy would take it beyond the threshold of affordability in the NHS. Because only one glucosamine hydrochloride product is licensed, it would not be cost-effective to prescribe glucosamine on the NHS Many people with OA take over-the- counter nutraceutical products and may benefit from clear, evidence-based information. In particular, the guideline development group felt that it would be beneficial to advise people who wanted to trial over-the-counter glucosamine that the only potential benefits identified in early research are related purely to a reduction of pain (to some people, and to only mild or modest degree) with glucosamine sulphate 1500 mg daily. They could also benefit from advice on how to perform their own trial of therapy, that is, to evaluate their pain before starting glucosamine and ensure they review the benefits of glucosamine after 3 months

Structure: In assessing the outcomes given in the evidence base, the guideline development group regarded measurement of joint space narrowing as of questionable value in assessing any potential beneficial structural modification, and convincing evidence of improvement in patient-centred outcomes consequent on any structural modification is still lacking. There is therefore no positive recommendation regarding structure modification Overall recommendation: The use of glucosamine or chondroitin products is not recommended for the treatment of OA Richy 200375 Pain/function: Comparable symptomatic Long-term studies are It is important to note efficacies (Lequesne index, VAS pain and needed to confirm and that rescue medications mobility) of chondroitin and glucosamine evaluate the structural were allowed in all trials; were shown efficacy of chondroitin however, cumulative For glucosamine, further doses were low and studies on the relationship it might be unlikely between structural and that rescue medication symptomatic changes, affected the pain-relieving controlling for baseline effect of glucosamine and characteristics including OA chondroitin stage, are needed, as well as studies on the possible use of glucosamine in prevention

BMI, body mass index; CONSORT, Consolidated Standards of Reporting Trials; ICERs, incremental cost-effectiveness ratios; NSAIDs, non-steroidal anti-inflammatory drugs; RCTs, randomised controlled trials; VAS, visual analogue scale.

32 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

minimal or non-existent. They suggested that if non-licensed product (glucosamine sulphate), there was an effect, chondroitin sulphate was more the evidence was not strong enough to warrant likely to benefit patients with low-grade OA than recommending that it should be prescribed on the patients with advanced OA. They also mentioned NHS. Notwithstanding some evidence of benefit the problem of concomitant use of rescue and very little evidence of harm in clinical practice, analgesia, but observed that in trials where use of and despite the extra scrutiny these agents had rescue analgesia was reported to be greater in the received, their economic cost–consequence table placebo group, the effect of chondroitin sulphate showed that only glucosamine sulphate was was actually reported to be larger than in trials potentially cost-effective out of the interventions reporting equal use in both comparison groups. considered in this section; however, a wide range of incremental cost-effectiveness ratios (ICERs) Adverse events Both reviews concluded that was reported and the poorest estimates of efficacy chondroitin sulphate was safe. would take it beyond the threshold of affordability in the NHS. Because only a glucosamine Glucosamine and/or chondroitin hydrochloride product was licensed, it would not be Analysing trials of glucosamine or chondroitin cost-effective to prescribe glucosamine on the NHS. versus placebo together in one meta-analysis, Richy75 found glucosamine/chondroitin to The guideline, however, also stated that many be superior to placebo for all the outcomes people with OA take over-the-counter nutraceutical investigated. The authors concluded that products and may benefit from clear, evidence- glucosamine sulphate or chondroitin sulphate based information. In particular, the guideline had a comparable benefit with respect to development group felt that it would be beneficial symptomatic outcomes (Lequesne index, VAS pain to advise people who wanted to trial over-the- and mobility). Authors reported that adequate counter glucosamine that the only potential structural data were only available for glucosamine, benefits identified in early research were related and a statistically significant benefit in the low purely to a reduction of pain (to some people, and to medium range was reported. Exclusion of to only a mild or modest degree) with glucosamine structural data from certain chondroitin studies sulphate 1500 mg daily. They could also benefit was a result of inadequately detailed reporting. from advice on how to perform their own trial The authors mentioned the concomitant use of of therapy, that is, to evaluate their pain before rescue medication, but concluded that cumulative starting glucosamine and ensure they review the doses were low and that rescue medication was benefits of glucosamine after 3 months. unlikely to have affected the pain-relieving effect of glucosamine and chondroitin. With respect to structural outcomes, the guideline development group regarded measurement of The NCCCC OA guideline1 did not recommend joint space narrowing as of questionable value the use of glucosamine or chondroitin products in assessing any potential beneficial structural for use in OA. They concluded that the evidence modification, and convincing evidence of from the trials was often difficult to compare owing improvement in patient-centred outcomes to differences in the products used (and their consequent on any structural modification was bioavailability), between-study populations, patient still lacking. There was therefore no positive BMI, and the use of analgesia at the time of pain recommendation regarding structure modification. and function assessment in the trials. They found that, overall, trials using glucosamine sulphate as a Research recommendations and single dose of 1500 mg rather than hydrochloride conclusions from the systematic reviews at 500 mg three times a day showed a small benefit Table 8 summarises the researchers’ conclusions and compared with placebo for treatment of knee OA. recommendations from the reviews and guideline. However, at the time the guideline was prepared, The conclusions about clinical effectiveness of the hydrochloride preparation had been granted glucosamine and chondroitin were not consistent. a European Medicines Evaluatory Agency licence, Authors noted that, at best, a modest effect on pain while the sulphate had not. The evidence for and function was observed. The effect on joint efficacy of chondroitin was less convincing. space was more consistent, but again the effect size was considered to be small and the long-term Their results suggested that the evidence to significance was hampered by the lack of long- support the efficacy of glucosamine hydrochloride term follow-up studies. The variability in quality as a symptom modifier was poor, and for the of some of the trials, in particular for chondroitin, 33

was noted, as was the variability in the preparations However, only one trial fulfilled our inclusion used. criteria,122 reporting only on symptoms, but not on structure. A fifth trial was identified through For the purposes of our research question about a PubMed auto-alert in June 2008. Kawasaki123 the effectiveness in modifying long-term disease reported both on symptoms and on structure. For progression, the reviews we identified, with the details of reasons for trial exclusion, see Table 9. exception of Poolsup,73 included only a very small proportion of trials of adequate follow-up. While One study excluded from our review is worthy of conclusions regarding structural changes were, noting here because of its high profile: the GAIT in three of the six reviews and guidelines, based (Glucosamine/chondroitin Arthritis Intervention on studies of at least 1 year, none presented their Trial) study, funded by the National Institute of data for long-term QoL outcomes separately. In Health (NIH) and conducted in multiple centres addition, the reporting of data extracted from across the USA.117 The original study was an RCT primary studies varied in the reviews with some of 24 weeks’ duration and was thus excluded from inconsistencies. We therefore went on to review our review.117 Recently (October 2008) a further RCTs of at least 12 months’ duration or follow-up report was published with 24 months of follow- and the results are reported below. up.124 This report is of a subset of participants in the original trial who met certain criteria and who Review of randomised accepted an invitation to participate in the longer controlled trials follow-up. Inclusion in the follow-up study required Search results that patients had appropriate baseline imaging and continued on treatment and participating in follow- The range of trials included in the reviews up at least until repeat imaging was done at 12 presented above varied widely and it proved months. The authors describe the subsequent study difficult to apply the findings to assessing the as a prospective observational study.124 Substantial effectiveness of glucosamine and chondroitin on numbers of patients who were eligible for this long-term outcomes such as disease progression. In extension study were not included in the analysis; addition, the reporting of the reviews on primary the most common reason was withdrawal prior to studies was not always consistent as not all reviews completing 12 months of follow-up. For example, described all the details about the primary studies. in the glucosamine arm of the study, 134 patients We have, therefore, reviewed separately all the were eligible, but only 77 were analysed, with 33 primary studies fulfilling our inclusion criteria withdrawing from the study. The withdrawals were (both the ones included in the reviews and new not described. The resulting study is at high risk ones identified in our searches and not included in of biases around participation, where patients the reviews). who dropped out were likely to differ, in a non- random way, from those who continued. The Within the reviews included above, there were baseline characteristics of patients participating in only two glucosamine RCTs fulfilling the inclusion the extension study differed between groups with criteria of our review,76,77 three chondroitin regard to sex, symptoms, radiological grade and trials108,114,116 (an earlier publication, Conrozier BMI. 1998,121 is included in the Reichenbach review61), and one combination therapy trial.120 The eight studies fulfilling our inclusion criteria are highlighted in grey in Table 9. Table 9 compares the characteristics of the primary studies included in the systematic reviews against The eight trials included did not allow a detailed the inclusion criteria of our review. The most comparison of different glucosamine preparations frequent reason for not fulfilling the inclusion (sulphate versus hydrochloride or Rotta versus non- criteria of the present review was short study Rotta preparations). Similarly, none of the trials duration (18 glucosamine, 15 chondroitin and one included a comparison between glucosamine and combined glucosamine/chondroitin trial). chondroitin, or a comparison between glucosamine and chondroitin combination therapy with In addition, Table 9 includes the details of studies either glucosamine or chondroitin monotherapy. identified in our additional searches. Through Likewise, none of the trials compared different examination of the excluded reviews and through doses of the treatment. supplemental searches, we identified four further RCTs studying glucosamine, and four studying Table 10 summarises the characteristics of the 34 a combination of glucosamine and chondroitin. included trials. The results for glucosamine and DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

TABLE 9 Randomised controlled trials (RCTs) included in reviews and identified in additional searches against inclusion criteria for this review (highlighted trials met all of the inclusion criteria for this review)

Structural Knee Duration Oral Placebo outcomes Full text RCT OA ≥ 1 year medication control included publication

Glucosamine From systematic reviews Cibere 200478  No   No  Clegg 2006117  No   No  Crolle 198079 ? No No No No  D’Ambrosio 198180 Various No No No No  sites Drovanti 198081 Various No   No  sites Herrero-Beaumont 200782  No   No  Houpt 199983  No   No  Hughes 200284  No   No  McAlindon 200485  No   No  Müller-Fassbender 199486  No  No No Noack 199487  No   No  Pavelká 200277       Pujalte 198088  No     Qiu 199889  No  No No  Reginster 200176       Reichelt 199490  No No No  Rindone 200091  No   No  Rovati 199792  No   No ? Usha 200493  No   No  Vajaradul 198194  No No  No  Vaz 198295  No  No No  Zenk 200296  No   No 

Additional searches Braham 2003125 ? No   No  Frestedt 2008126  No   No  Mehta 2007127  No  No No  Zhang 2007128  No  No No 

Chondroitin From systematic reviews Alekseeva 199997 No   No  Bourgeois 199898  No   No  Bucsi 199899  No   No  Conrozier 1992100 No No   No  Fleisch 1997101     No No

continued

Structural Knee Duration Oral Placebo outcomes Full text RCT OA ≥ 1 year medication control included publication Kahan 2007102      No Kerzberg 1987103  No No  No  L’Hirondel 1992104  No   No  Mazieres 1992105  No   No  Mazieres 2001106  No   No  Mazieres 2007107  No   No  Michel 2005108       Morreale 1996109  No   No  Nasonova 2001110  No   No  Pavelká 1999111  No   No  Rovetta 1991112  No No No  Soroka 2002113  No   No No Uebelhart 1998114/Conrozier       1998121 Uebelhart 1999115  No   No No Uebelhart 2004116       Glucosamine/chondroitin From systematic reviews Clegg 2006117  No   No  Cohen 2003118  No No No  Das 2000119  No     Rai 2004120       Additional searches Messier 2007122     No  Kawasaki 2008123       Debi 2000129  No No  No  Leffler 1999130  ? ?, not only  No  Gl/Ch Nguyen 2001131 No No   No 

Ch, chondroitin; Gl, glucose.

chondroitin are presented separately. Appendix 5 after cessation of treatment52). Both were funded provides all of the data in a single summary table by the Rotta Research Group. The third trial, by for direct comparison between treatments. Kawasaki123 was a Japanese open-label RCT with a duration of 18 months. Findings Glucosamine A total of 556 people participated in these trials. Trial characteristics The mean age of participants was between 61 and Three trials compared glucosamine with placebo. 70 years and more women than men were included The trials by Pavelká77 and Reginster76 were both (between 75% and 100% women). Most had OA of randomised, double-blind, placebo-controlled trials Kellgren–Lawrence grade 2 at baseline (between with a duration of 3 years. Follow-up data for knee 53% and 54% for Pavelká and between 70% and arthroplasty for both trials taken together were 71% for Reginster); the rest had OA of Kellgren– 36 available for a mean follow-up of 8 years (5 years Lawrence grade 3 (no details given in the trial by DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52 continued Method of randomisation: method not described Allocation concealed? Unclear blinded? No Participants Outcome assessors blinded? No; assessment of joint space width blinded of participants Proportion excluded/lost to follow- up: glucosamine: 35% withdrawals; C: 29% withdrawals; no significant difference in rates or reasons Handling of missing data: last observation carried forward Intention-to-treat analysis performed? Yes Statistical analysis appropriate? Yes Groups comparable at baseline? Yes Quality Primary: unclear Primary: pain WOMAC Pain: pain scale subscale, VAS reported WOMAC: Japan Function: Orthopaedic Association score (pain during (JOA) walking, pain when using stairs, range of motion, swelling, 100 points = full score); WOMAC Joint space loss: tibiofemoral joint space width (determined at the centre point of medial femoral condyle) Adverse events: no Other: NSAID consumption, urinary NTX (uNTX, bone absorption marker) Assessment: regular assessment every 3 months Outcome measures Glucosamine: 1500 mg/day glucosamine hydrochloride Risedronate: 2.5 mg/day risedronate (not considered here) Control: no supplement All groups: instructed by physical therapists to perform home exercise; instruction also given in a brochure; exercises as follows: 1. isometric muscle exercises of the lower limbs, 2. range motion exercise after the knee was warmed (bath or shower) Co-interventions/rescue analgesia: sodium loxoprofen prescribed in advance for rescue and any use of the medicine recorded analgesia use: no Rescue significant difference between groups Adherence assessment: not reported Interventions , SD 2.5; 2 mm or and , SD 3.0 2 49; 31 completed 51; 33 completed trial = = 42; 30 completed trial = mm (equivalent to K–L grade II or 6 OA stage: not reported OA duration: not reported OA Comorbidities: not reported Subgroups: effects of age, BMI, level of joint progression, urine NTX, concomitant use of glucosamine or risedronate with home exercise at baseline (but not for intervention groups separately) control: 24.0 kg/m Inclusion criteria: postmenopausal of medial women with untreated OA femorotibial compartment of knee and no other inflammatory diseases; selected according to criteria of ACR; joint space width ≥ 1 (not considered here) Control: n < III) Exclusion criteria: standard exclusion criteria (references given) Age: glucosamine: 68.5 years, SD 7.3; control: 69.5 years, SD 7.1 Gender: all female BMI: glucosamine: 23.9 kg/m trial Risedronate: n Participants number: 142 Total Glucosamine: n Design Focus: effects of glucosamine, risedronate or no treatment in patients with mild to moderate concurrently knee OA performing therapeutic exercise Single centre Duration: 18 months Follow-up: no post- intervention follow-up Setting: outpatient not reported Funding: 123 Characteristics of randomised controlled trials fulfilling the inclusion criteria this review Study Japan Glucosamine 2008 Kawasaki

TABLE 10 TABLE 37

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Clinical effectiveness of glucosamine and chondroitin on the progression of OA of the knee Method of randomisation: described, adequate Allocation concealed? Yes blinded? Yes Participants Outcome assessors blinded? Yes of participants Proportion excluded/lost to follow- up: intervention: 35% withdrawals; control: 46% withdrawals; no significant difference in rates or reasons Handling of missing data: worst case analysis and random sampling analysis Intention-to-treat analysis performed? Yes Statistical analysis appropriate? Yes Groups comparable at baseline? Yes Quality mm, other 5 Primary: changes of Primary: minimum joint space width joint pain WOMAC Pain: subscale reported WOMAC: WOMAC Function: function and stiffness subscales; Lequesne index Joint space loss: minimum joint space width, joint space narrowing > structural modifications Adverse events: withdrawal rates and reason for drop-out, adverse events, routine laboratory tests Other: paracetamol consumption Assessment: quarterly assessment of symptoms; radiographs and safety laboratory tests at end of each year Note: in patients with bilateral disease, the target knee was the one more severely affected Outcome measures mg mg/day Intervention: 1500 glucosamine sulphate (powder for oral solution) (Rotta) Control: placebo packets identical in external appearance Co-interventions/rescue analgesia: paracetamol 500 tablets, use recorded in patient diaries; no other pharmacologic allowed; treatments for OA exercise or only hydrotherapy, ultrasound allowed as physical therapies (no significant difference between groups) analgesia use: no Rescue significant difference between groups Adherence assessment: at least 86% of patients reported more than 90% drug intake Interventions , SD 2.1; 2 , SD 1.8 2 101; 66 completed = 101; 55 completed trial , clinically significant = 2 kg/m 27 alterations in haematological variables and renal, hepatic metabolic functions; systemic or intra-articular corticosteroid therapy in previous 3 months Age: intervention: 61.2 years, SD 7.3; control: 63.5 years, SD 6.9 Gender: intervention: 79% female; control: 76% female BMI: intervention: 25.7 kg/m OA stage: intervention: 54% K–L grade OA 2, 46% grade 3; control: 53% K–L grade 2, 47% 3 duration: intervention: 10.1 years, OA SD 8.1; control: 11.0 years, 6.8 Comorbidities: not reported Subgroups: none control: 25.7 kg/m Participants number: 202 Total Intervention: n trial Control: n Inclusion criteria: 45–70 years, based on clinical and primary knee OA Lequesne radiological criteria of ACR; index at least 4 points, but not > 12 Exclusion criteria: history of other clinically significant articular or rheumatic diseases, inflammatory rheumatic disease, secondary OA, history of trauma or lesions knee joint, severe articular inflammation, evidence of rapidly progressing OA (obtained before the trial); BMI > Design Focus: long- term effects of glucosamine sulphate on progression of joint structure and symptom changes in knee OA Single centre Duration: 3 years Follow-up: no post- intervention follow-up Setting: outpatient Rotta Funding: Group (Italy) Research 77 Characteristics of randomised controlled trials fulfilling the inclusion criteria this review (continued) Study 2002 Pavelká Czech Republic

38 10 TABLE DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52 continued Method of randomisation: described, adequate Allocation concealed? Yes blinded? Yes Participants Outcome assessors blinded? Yes of participants Proportion excluded/lost to follow- up: intervention: 33% withdrawals; control: 36% withdrawals; no significant difference in rates or reasons Handling of missing data: 3 approaches – (i) worst scenario analysis, (ii) last observation carried forward, (iii) random sampling approach Intention-to-treat analysis performed? Yes Statistical analysis appropriate? Yes Groups comparable at baseline? Yes Quality Primary: mean joint Primary: space width of medial compartment of tibiofemoral joint, total score WOMAC joint pain WOMAC Pain: subscale reported WOMAC: WOMAC Function: function and stiffness subscales Joint space loss: mean, minimum joint space width Adverse events: adverse events, routine laboratory tests, fasting glucose Other: use of rescue medication Assessment: radiographs taken at baseline, 1 and 3 years Outcome measures mg tablets, mg tablets); mg/day mg tablets) or mg capsules or Intervention: 1500 glucosamine sulphate (powder for oral solution) (Rotta) Control: placebo (no details) Co-interventions/rescue analgesia: paracetamol (500 NSAIDs (diclofenac 50 piroxicam 20 proglumetacin 150 washout before symptom assessment (at least five half-lives of selected medication allowed) analgesia use: no Rescue significant difference in medication rescue of consumption between groups Adherence assessment: over 70% drug intake in 81–91% without significant difference between groups Interventions , SD 2.6; 2 50 years, primary , SD 2.7 2 106; 68 completed = 106; 71 completed trial ; substantial abnormalities = 2 ) kg/m 213 30 in haematological, hepatic, renal or metabolic functions; intra-articular or systematic corticosteroids within 3 months preceding enrolment Age: intervention: 66.0 years, SD 8.1; control: 65.5 years, SD 7.5 Gender: intervention: 75% female; control: 78% female BMI: intervention: 27.3 kg/m OA stage: intervention: 71% K–L grade OA 2, 29% grade 3; control: 70% K–L grade 2, 30% 3 duration: intervention: 8.0 years, OA SD 7.5; control: 7.6 years, 7.5 Comorbidities: not reported Subgroups: highest vs lowest quartile for joint space width (see Bruyere 2003 control: 27.4 kg/m Participants number: 212 Total Intervention: n trial Control: n Inclusion criteria: > of medial femorotibial knee OA compartment, diagnosed according to clinical and radiological criteria of ACR Exclusion criteria: history or active presence of rheumatic diseases that could be cause of secondary knee OA; severe articular inflammation; traumatic knee lesions; BMI > Design Focus: effect of glucosamine sulphate on long-term progression of OA joint structure changes and symptoms Single centre Duration: 3 years Follow-up: no post- intervention follow-up Setting: outpatient Rotta Funding: Group (Italy) Research 76

Study 2001 Reginster Belgium 39

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Clinical effectiveness of glucosamine and chondroitin on the progression of OA of the knee Method of randomisation: described, adequate Allocation concealed? Yes blinded? Yes Participants Outcome assessors blinded? Yes of participants Proportion excluded/lost to follow- up: 27% in each study group; no significant difference in reasons for withdrawal Handling of missing data: data at last visit used; radiographic evaluation done at time of drop-out of patients who dropped out (except for 16 patients who dropped out within 1 month of study entry) Intention-to-treat analysis performed? Yes Statistical analysis appropriate? No measures of variability reported Groups comparable at baseline? Yes Quality mm; minimum and Primary: minimum and Primary: mean joint space width of more severely affected compartment of target knee joint pain WOMAC Pain: subscale reported WOMAC: WOMAC Function: function subscale Joint space loss: joint space narrowing in mean joint space width measured using digitised radiographs and an image analysis system Adverse events: reported Other: none Assessment: follow-up assessments conducted by mail every 3 months over 2-year study period and treatment (WOMAC diaries); radiographs at baseline and 24 months; clinical examination and routine laboratory tests at baseline, 12 and 24 months Outcome measures g/day allowed mg tablet mg tablets at (IBSA, Hungary) ® Intervention: 800 Condrosulf (chondroitins 4- and 6-sulphate) daily Control: identical placebo tablet Co-interventions/rescue analgesia: paracetamol 500 maximum dose of 3 as rescue analgesia; for secondary rescue, NSAIDs were allowed up to a maximum period of 5 consecutive days if primary rescue analgesia with paracetamol was insufficient; physical therapy limited to application of warmth and strengthening exercises when deemed necessary by patient; no other interventions allowed, including steroid injections analgesia use: similar Rescue amounts taken, no details Adherence assessment: pill counts and 12- 24-month visits; 69% in chondroitin group and 72% placebo group took more than 70% of tablets during trial (no significant difference between groups Interventions , SD 5.2; 2 , SD 5.5 2 150; 110 completed = 150; 109 completed trial = OA stage/severity: not reported (stated OA that K–L scores were not significantly different between groups) control: 28.1 kg/m Participants number: 300 Total Intervention: n trial Control: n Inclusion criteria: 40–85 years, (knee clinically symptomatic knee OA pain while standing, walking, and/ or motion for at least 25 of the 30 no required days prior to study entry, minimum level of pain on day entry); clinical diagnosed according to ACR and radiographic criteria for knee OA; patients with K–L grade 1, 2, 3; target knee defined as the most symptomatic knee on study entrance Exclusion criteria: presence of any causes of secondary knee OA, traumatic knee lesions, severe comorbidity (especially severe heart, renal, lung or neurological disease), intra-articular previous joint surgery, medications including corticosteroids in past month, foreseeable prospect of major surgery during 2-year study period; washout period of 3 months required for potentially longer-acting substances, e.g. glucosamine and chondroitin sulphate Age (mean): intervention: 62.5 years, SD 9.1; control: 63.1 years, 10.7 Gender: intervention: 51% female; control: 52% female BMI: intervention: 27.7 kg/m Design Focus: effect of chondroitin sulphate on cartilage loss in knee OA Single centre Duration: 2 years Follow-up: no post- intervention follow-up Setting: outpatient clinic unclear Funding: 108 Characteristics of randomised controlled trials fulfilling the inclusion criteria this review (continued) Study Chondroitin Michel 2005 Switzerland

40 10 TABLE DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52 continued Quality Method of randomisation: method not described Allocation concealed? Unclear blinded? Yes Participants Outcome assessors for joint blinded? Yes space measurements; other measurements unclear of participants Proportion excluded/lost to follow- up: 8.7% in both comparison groups; chondroitin group: 1 death, 1 left the country; placebo group: 1 left the 1 not satisfied country, Handling of missing data: not described Intention-to-treat analysis performed? Unclear (probably not) Statistical analysis appropriate? Yes Groups comparable at baseline? Yes Outcome measures spontaneous Primary: joint pain (Huskisson overall mobility VAS), capacity (VAS) Huskisson VAS Pain: no WOMAC: mobility Function: capacity (VAS) Joint space loss: joint space measurement Adverse events: tolerability Other: bone joint metabolism Assessment: at entry and 3, 6, 12 months; at entry and 12 X-rays months Note: only the most damaged knee was taken into account for inclusion and follow-up mg, dissolved in half a glass of Interventions Intervention: 800 mg/day chondroitin 4- and 6-sulphate two sachets of (Condrosulf, 400 water) Control: same daily amount of vehicle alone Co-interventions/rescue analgesia: free access to paracetamol as rescue medication analgesia use: analgesic Rescue consumption not reported 3 months mm at entry 5°, steroids or 23; 21 completed = 300) versus patients with 23; 21 completed trial = = 225) = minimum joint space ≥ 1 Participants duration: not reported OA Comorbidities: not reported Subgroups: for joint space changes, all patients ( n ( n number: 46 Total Intervention: n trial Control: n Inclusion criteria: 35–78 years, mono- or bilateral clinically symptomatic with a minimum remaining knee OA femorotibial joint space of 25% on standard X-ray Exclusion criteria: patients with inflammatory or systemic diseases other conditions involving joints, lower limbs axial deviation > bone-oriented treatments < before study began Age: intervention: 60 years, SD 13; control: 57 years, SD 11 Gender: intervention: 48% female; control: 56% female intervention: 72 kg, SD 11; Weight: control: 76 kg, SD 14 stage: intervention: 43% K–L grade OA 1, 48% grade 2, 9% 3; control: 48% K–L grade 1, 43% 2, 9% grade 3 duration: not reported OA Comorbidities: not reported Subgroups: none Design Focus: clinical, radiological, biological efficacy and tolerability of chondroitin 4- and 6-sulphate in knee OA Single centre Duration: 1 year Follow-up: no post- intervention follow-up Setting: in- or outpatient IBSA Funding: (producer) 114

Study Uebelhart 1998 Switzerland 41

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Clinical effectiveness of glucosamine and chondroitin on the progression of OA of the knee Method of randomisation: described, adequate Allocation concealed? Yes blinded? Yes Participants Outcome assessors blinded? Yes of participants Proportion excluded/lost to follow- up: 10 patients (of 120) not included in intention- to-treat population (did not turn up for second visit, did not take any medication); 11 drop- outs in intervention and 15 in placebo group; no significant difference Handling of missing data: last observation carried forward Intention-to-treat analysis performed? Yes Statistical analysis appropriate? Yes Groups comparable at baseline? Yes Quality m walking Primary: Lequesne index Primary: spontaneous joint Pain: pain (Huskisson VAS) no WOMAC: 20 Function: time Joint space loss: joint space surface area, mean and minimum joint space width Adverse events: adverse events, clinical laboratory evaluation Other: global patient and physician assessment; paracetamol consumption Note: in patients with the bilateral knee OA, most symptomatic knee was selected as the study target knee Assessment: every 3 months; radiographs at entry and 12 months Outcome measures mg/day g/day allowed; chondroitin 4- and 6-sulphate (Condrosulf sachets taken with a glass of water); treatment administered intermittently from entry to month 3, and between months 6 and 9 Control: identical sachets with vehicle only Co-interventions/rescue analgesia: medications as listed in exclusion criteria not allowed; paracetamol up to 4 paracetamol treatment stopped 24 hours prior to each visit analgesia use: after first Rescue month, paracetamol consumption significantly greater in placebo group (at 12 months 55.5 tablets, SD 68.1 versus 25.8, 37.0 in placebo group) Adherence assessment: adherence 93–98%, no significant difference between groups Interventions Intervention: 800 54; 43 completed trial = 56; 41 completed trial = Participants number: 110 Total Intervention: n Control: n mono- over, or 40 age criteria: Inclusion or bilateral clinically symptomatic according to ACR idiopathic knee OA criteria; patients with K–L score 1–3, with a minimum 25% remaining medial femorotibial joint space upon entry Exclusion criteria: other inflammatory joint diseases or other systemic conditions affecting or involving joints; neoplasias, bone metabolic diseases and/or other metabolic and systemic diseases; treatments such as intra-articular steroids, NSAIDs, symptom-modifying agents or bone- oriented therapies (e.g. fluorides, bisphosphonates, calcitonin, hormonal substitution within 3 months before beginning of study Age: intervention: 63.2 years, SD 9.1; control: 63.7 years, SD 8.1 Gender: intervention: 80% female; control: 82% female intervention: 76.8 kg, SD 15.8; Weight: control: 76.4 kg, SD 13.8 stage: intervention: 13% K–L grade OA 1, 59% grade 2, 28% 3; control: 11% K–L grade 1, 59% 2, 30% grade 3 duration: intervention: 50.1 OA months; control: 52.4 months Comorbidities: not reported Subgroups: none Design Focus: efficacy of a 3-month, twice a year intermittent treatment with chondroitin sulphate in knee OA Multicentre Duration: 1 year Follow-up: no post- intervention follow-up Setting: not reported IBSA Funding: (producer) 116 Characteristics of randomised controlled trials fulfilling the inclusion criteria this review (continued) Study Uebelhart 2004 Switzerland/Belgium/ France/Italy

42 10 TABLE DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52 continued Quality Method of randomisation: not explicitly described, computer allocation implied Allocation concealed? Yes blinded? Yes Participants Outcome assessors blinded? Yes of participants Proportion excluded/lost to follow- up: intervention: 18% withdrawals; control: 20% withdrawals; not reported if significant differences between groups Handling of missing data: last observation carried forward Intention-to-treat analysis performed? Yes Statistical analysis appropriate? Yes Groups comparable at baseline? No, GH/CS were significantly younger, more overweight, and had a higher income than placebo group Outcome measures Primary: WOMAC WOMAC Primary: function pain WOMAC Pain: subscale yes WOMAC: WOMAC Function: function subscale; mobility (distance walked in 6 minutes) Joint space loss: not assessed Adverse events: not systematically reported Other: knee concentric extension and flexion strength (Kin-Com 125E isokinetic dynamometer); balance (AMTI force platform) Assessment: screening visit, baseline, 6 and 12 months mg g/day Interventions Intervention: 1500/1200 glucosamine hydrochloride plus chondroitin sulphate (GH/CS); choice of 1 × or 3 a day regimen; 6 healthy lifestyle classes; after 6-month exercise programme (1 hour twice a week) added for last 6 months (combined facility and home-based exercise programme, two 15-minute walking sessions, separated by 20 minutes of strength training) Control: placebo of identical size, shape, regimen; lifestyle colour, classes and exercise programme after 6 months as in GH/CS group Co-interventions/rescue analgesia: paracetamol maximum dosage 4 analgesia use: decrease in Rescue paracetamol use by 12 months of 37% in intervention and 11% in placebo group (not stated if significant) period: 2 weeks Run-in/washout before start of intervention; discontinuation of all medications; rescue medications/required medication unrelated to OA permitted; unblinded placebo compliance supply (3 pills/day, assessed) Adherence assessment: 91–97% 0.03 = m in 6 45; 37 completed trial = 44; 35 completed trial = Participants number: 89 Total Intervention: n Control: n Inclusion criteria: ≥ 50 years, radiographic evidence of mild to (K–L grades 2–3) moderate knee OA classification criteria who met ACR Exclusion criteria: currently participating in another trial, dementia, active cancer other than skin cancer, anaemia, severe renal insufficiency, hepatic diseases, excess alcohol use, knee-joint replacement, inability to walk unassisted at least 128 minutes, completion of more than 20 minutes of formal exercise per week during past 3 months, planned absence for longer than 3 weeks during required study visits, residing further than 1 hour drive from research exposure to facilities, shellfish allergy, glucosamine/chondroitin in 6 months prior to randomisation, unwilling discontinue current arthritis medication during 2-week washout and 12-month intervention period (other than rescue medication), less than 80% compliance during run-in period, failure to complete graded exercise test, required human assistance with activities related to knee pain, recent knee surgeries or injections, e.g. cortisone and hyaluronic acid Age: intervention: 70.0 years, SE 1.28; control: 74.1 years, SE 1.32; p Gender: intervention: 75.6% female; control: 65.9% female Design Focus: effect of glucosamine hydrochloride plus chondroitin sulphate alone and combined with exercise on physical function, pain, strength, balance and mobility in older adults with knee OA Single centre Duration: 1 year Follow-up: no post- intervention follow-up Setting: outpatient clinical research centre Rexall Funding: Sundown Inc. 122

Study Glucosamine/chondroitin Messier 2007 USA 43

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Clinical effectiveness of glucosamine and chondroitin on the progression of OA of the knee Method of randomisation: not reported Allocation concealed? Unclear blinded? Yes, Participants described as double blind Outcome assessors described as blinded? Yes, double blind of participants Proportion excluded/lost to follow- up: not reported Handling of missing data: not reported Intention-to-treat analysis performed? Not reported Statistical analysis appropriate? Yes Groups comparable at baseline? Yes Quality Primary: joint space Primary: width changes in narrowest medial compartment of tibiofemoral joint no separate Pain: assessment no WOMAC: Lequesne index Function: Joint space loss: minimum joint space width Adverse events: not reported Other: none Assessment: every 4 weeks; standardised radiographs at enrolment and after 1 year Outcome measures , mg ® mg capsule) (Panacea Biotec Ltd, capsule) (Panacea India) – dosing not stated, presumably one capsule a contains 250 day; Kondro glucosamine sulphate and 200 chondroitin sulphate Control: matching placebo capsules Co-interventions/rescue analgesia: patients advised to take paracetamol for pain analgesia use: differences Rescue between groups not reported Interventions Intervention: oral combination of glucosamine sulphate and chondroitin sulphate (Kondro 0.005 = , SE 0.93; 2 50 years, primary , SE 0.71; p 2 50 = 50 = OA stage/severity/duration: not OA reported Comorbidities: listed in detail, no significant difference between groups control: 27.3 kg/m Control: n Inclusion criteria: > mainly of medial femorotibial knee OA compartment diagnosed according to clinical and radiological criteria of ACR Exclusion criteria: history or active presence of other rheumatic diseases severe responsible for secondary OA, articular inflammation as confirmed by physical examination, K–L grade 4, Lequesne index higher than 12, history of intra-articular injection or systemic steroid intake in 3 months preceding enrolment Age: intervention: 54.7 years; control: 53.9 years Gender: not reported intervention: 69.1 kg; control: Weight: 68.9 kg stage/severity: no details, stated OA was comparable at baseline that OA and mild to moderate duration: not reported OA Comorbidities: not reported Subgroups: none Participants number: 100 Total Intervention: n BMI: intervention: 30.7 kg/m Design Focus: effect of chondroitin plus glucosamine sulphate on joint structure and symptoms in knee OA Single centre Duration: 1 year Follow-up: no post- intervention follow-up Setting: orthopaedic outpatient centre unclear Funding: Characteristics of randomised controlled trials fulfilling the inclusion criteria this review (continued) 120 Rai 2004 Rai ACR, American College of Rheumatology; BMI, body mass index; K–L, Kellgren and Lawrence scores; NSAID(s), non-steroidal anti-inflammatory drug(s); NTX, terminal type I collagen American College of Rheumatology; BMI, body mass index; K–L, Kellgren ACR, visual analogue scale. telopeptides; VAS, Study India

44 10 TABLE DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Kawasaki which just stated that both Kellgren– Trial quality Lawrence grades 2 and 3 were included). In all Details for trial quality are shown in Table 10. Trial trials, baseline characteristics were well balanced in quality for the trials by Pavelká and Reginster was the comparison groups. good. The trial by Kawasaki was of poor quality. Randomisation and allocation concealment were Interventions not described and there was no blinding (with the In the Pavelká and Reginster trials, 1500 mg per exception of assessment of joint space width). day glucosamine sulphate (Rotta) was given as a powder oral solution. Kawasaki used 1500 mg per Results day glucosamine hydrochloride. All trials reported Structure Table 11 summarises the results for trials that there was no significant difference in rescue of glucosamine on joint structure. analgesia use between comparison groups and that medication adherence was good. The trials by Pavelká and Reginster showed a statistically significant effect in favour of Outcome measures and further trial details are glucosamine for minimum joint space width, summarised in Table 10. with a weighted mean difference from baseline to

TABLE 11 Results for trials of glucosamine on joint structure

Change from baseline/ p-value No. of difference between (between Study Outcome participants Baseline groups groups) Kawasaki 2008123 Joint space width I: n = 49 I: 2.6 mm, Change from baseline: NS (Hydrochloride) C: n = 42 SD 1.2 I: 0.0 mm (95% CI –2.2 to C: 3.1 mm, 1.7) SD 1.0 C: –0.3 mm (95% CI –1.8 to 1.6) Difference: 0.2 mm Pavelká 200277 Minimum joint I: n = 101 I: 3.89 mm, Change from baseline: 0.001 (Sulphate) space width C: n = 101 SD 1.48 I: +0.04 mm (95% CI –0.06 C: 3.63 mm, to 0.14) SD 1.57 C: –0.19 mm (95% CI –0.29 to –0.09) Difference: 0.23 mm (95% CI 0.09 to 0.37) Severe joint End of study: 0.05 space narrowing I: n = 5 (> 0.5 mm) C: n = 14 NNT = 11 Reginster 200176 Mean joint space I: n = 106 Total joint Change from baseline: 0.043 (Sulphate) width C: n = 106 space width: I: –0.06 mm (95% CI –0.22 I: 5.23 mm, to 0.09) SD 1.36 C: –0.31 mm (95% CI –0.48 C: 5.39 mm, to –0.13) SD 1.29 Difference: 0.24 mm (95% CI 0.01 to 0.48) Minimum joint I: n = 106 I: 3.82, Change from baseline: 0.003 space width C: n = 106 SD 1.32 mm I: –0.07 mm (95% CI –0.22 C: 3.95, to 0.07) SD 1.24 mm C: –0.40 mm (95% CI –0.56 to –0.24) Difference: 0.33 mm (95% CI 0.12 to 0.54)

C, control; I, intervention; NNT, number needed to treat; NS, not significant. 45

Glucosamine Placebo Study or Mean difference IV Mean difference IV subgroup Mean SD Total Mean SD Total Weight (random, 95% CI) (random, 95% CI)

Pavelká 200277 0.04 0.363 101 −0.19 0.363 101 67.3% 0.23 (0.13 to 0.33) Reginster 200176 −0.07 0.539 106 −0.40 0.594 106 32.7% 0.33 (0.18 to 0.48)

Total (95% CI) 207 207 100.0% 0.26 (0.17 to 0.35) Heterogeneity: τ2 = 0.00; χ2 = 1.15, df = 1 (p = 0.28); I2 = 13% Test for overall effect: z = 5.60 (p = 0.00001) –0.5 –0.25 0 0.25 0.5 Favours Favours placebo glucosamine

FIGURE 3 Glucosamine sulphate: a meta-analysis of minimum joint space width – difference from baseline (fixed-effects model produces same point estimate; random-effects model presented as confidence interval better reflects the uncertainty around that point estimate). IV, inverse difference.

end of study of 0.26 mm (95% CI 0.17 to 0.35), improvements in the joint space width were also indicating less joint space loss with glucosamine observed in the control group. (Figure 3). Both trials reported a point estimate in favour of glucosamine sulphate, with minimal Symptoms/pain For function and pain assessment, evidence of heterogeneity in the size of the effect the trials did not report data consistently (for being observed. Pavelká found significantly more example, Pavelká and Kawasaki reported WOMAC patients with severe joint space narrowing (more scores as points, whereas Reginster used per cent than 0.5 mm) in the placebo group than in the change on a VAS), only Pavelká used the Lequesne intervention group (n = 14 versus n = 5, p < 0.05). index, and Kawasaki used the Japan Orthopaedic Association (JOA) score and a visual analogue pain Mean joint space width was reported by Reginster scale, so data could not be summarised in a meta- only, who observed a significant difference of analysis (Table 13). 0.24 mm (95% CI 0.01 to 0.48) in mean joint space narrowing, indicating a preservation of joint space Pain Pavelká found a statistically significant effect with glucosamine. Kawasaki reported results for in favour of glucosamine compared with placebo joint space width (determined at the centre point at the end of the 3-year trial for the WOMAC of the medial femoral condyle) and found no pain subscale [difference between groups 0.7 significant effect of glucosamine hydrochloride points (95% CI 0.06 to 1.3)]. Similarly, Reginster versus control. found a statistically significant effect in favour of glucosamine for the WOMAC pain subscale A subgroup analysis of the Reginster trial (Table 12) [change from baseline approximately –36.7, compared patients in the lowest quartile for mean SE 8.3 mm (VAS) for glucosamine and –7.5 SE, joint space width (joint space width < 4.5 mm) 10.6 mm for the control group (estimated from with patients in the highest quartile for mean graph)]. Kawasaki did not find any statistically joint space width (joint space width > 6.2 mm). significant difference in the WOMAC pain subscale In the patients in the lowest quartile (n = 29 or the VAS pain scale between the glucosamine intervention, n = 23 control), mean joint space hydrochloride and control groups. Meta-analysis width increased by 0.22 mm (SD 0.66 mm) in the (Figure 4) illustrates the small, but statistically intervention group and 0.13 mm (SD 0.81 mm) significant favourable effect for glucosamine. in the control group (no statistically significant difference). In the patients in the highest quartile Little detail was reported in the three trials about (n = 27 intervention, n = 26 control), mean joint the use of rescue analgesia with the exception space width decreased by 0.45 mm (SD 1.04 mm) of stating in all three trials that there was no in the intervention group and by 1.05 mm (SD statistically significant difference in the volume of 1.28 mm) in the control group (difference not rescue analgesia used. statistically significant, p = 0.1). There was a statistically significant difference in joint space Function Pavelká found a statistically significant change between the two subgroups (p < 0.01). The effect in favour of glucosamine compared with authors speculated that the results may indicate placebo for both the WOMAC function and that patients with less severe disease may benefit stiffness subscales (difference between groups more from structure-modifying drugs, although 2.1 points (95% CI 0.28 to 3.9) for function and 46 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

TABLE 12 Subgroup analysis: summary from Reginster 2001 trial76

Change from baseline/ Study Subgroup Outcome difference between groups p-value Reginster 200176 Mean joint space width lowest Change in mean Joint space width < 4.5 mm: < 0.01 for joint quartile (< 4.5 mm, n = 29 joint space space change intervention, n = 23 control) width I: +6.2%, SD 17.5; +0.22 mm, between the SD 0.66 two quartiles vs C: +3.8%, SD 23.8; +0.13 mm, Highest quartile (> 6.2 mm, SD 0.81 n = 27 intervention, n = 26 control) Joint space width > 6.2 mm:

I: –6.0%, SD 15.1; –0.45 mm, SD 1.04

C: –14.9%, SD 17.9; –1.05 mm, SD 1.28

C, control; I, intervention.

TABLE 13 Summary of glucosamine trial results for pain, function and composite health state scores

No. of Change from baseline/difference Study participants Outcome Baseline between groups p-value Kawasaki I: n = 49 JOA total score I: 74.2, SD 13.3 I: +13.7 (95% CI –18.5 to 43) NS 123 2008 C: n = 42 (range 0–100) C: 69.6, SD 14.4 C: +16.3 (95% CI –4.8 to 39.8) JOA gait (range I: 23.3, SD 5.5 I: +4.4 (95% CI 2.8 to 11) NS 0–30) C: 21.5 SD 6.6 C: +6.3 (95% CI –3.6 to 20) JOA stairs (range I: 14.7, SD 5.6 I: +5.5 (95% CI –8.2 to 20.3) NS 0–25) C: 11.9, SD 6.1 C: +6.3 (95% CI –1.1 to 17.5) JOA range of I: 29.4, SD 5.0 I: +1.4 (95% CI –5.5 to 5.5) NS motion (range C: 29.5, SD 4.8 C: +1.9 (95% CI –6.1 to 10) 0–35) JOA swelling I: 6.8, SD 2.7 I: +2.6 (95% CI –0.3 to 5) NS (range 0–10) C: 6.7, SD 3.3 C: +1.9 (95% CI –5.0 to 10) VAS pain (range I: 5.0, SD 2.0 I: –1.6 (95% CI –5.0 to 2.2) NS 0–10) C: 4.9, SD 1.8 C: –1.7 (95% CI –5.5 to 2.5) WOMAC total I: 31.5, SD 15.6 I: –14.9 (95% CI –45.5 to 6) NS (range 0–96) C: 28.7, SD 16.0 C: –11.5 (95% CI –38.8 to 7.5) WOMAC pain I: 7.2, SD 3.9 I: –3.6 (95% CI –14.1 to 3) NS (range 0–20) C: 6.5, SD 3.4 C: –2.9 (95% CI –9.5 to 0.5) WOMAC stiffness I: 3.4, SD 1.6 I: –1.5 (95% CI –5.1 to 1.0) NS (range 0–8) C: 2.7, SD 1.6 C: –0.7 (95% CI –3 to 1.5) WOMAC function I: 20.8, SD 11.0 I: –9.8 (95% CI –31.2 to 5.2) NS (range 0–68) C: 19.6, SD 11.5 C: –7.5 (95% CI –27.2 to 7)

continued

TABLE 13 Summary of glucosamine trial results for pain, function and composite health state scores (continued)

No. of Change from baseline/difference Study participants Outcome Baseline between groups p-value Pavelká I: n = 101 Lequesne index I: 8.95, SD 2.30 Change from baseline: 0.002 77 2002 C: n = 101 (points) C: 8.94, SD 2.27 I: –1.7 (95% CI –2.2 to –1.2) C: –0.82 (95% CI –1.1 to –0.51) Difference: 0.91 (95% CI 0.34 to 1.5) WOMAC total I: 30.70, SD 14.40 Change from baseline: 0.01 (points) C: 30.48, SD 14.43 I: –8.0 (95% CI –9.8 to –6.3) C: –4.9 (95% CI –6.5 to 3.2) Difference: 3.1 (95% CI 0.77 to 5.5) WOMAC pain I: 6.61, SD 3.45 Change from baseline: 0.03 (points) C: 6.33, SD 3.13 I: –2.0 (95% CI –2.4 to –1.5) C: –1.3 (95% CI –1.7 to 0.88) Difference: 0.7 (95% CI 0.06 to 1.3) WOMAC function I: 21.84, SD 10.67 Change from baseline: 0.02 (points) C: 22.00, SD 11.03 I: –5.8 (95% CI –7.1 to –4.4) C: –3.7 (95% CI –4.9 to –2.5) Difference: 2.1 (95% CI 0.28 to 3.9) WOMAC stiffness I: 2.25, SD 1.47 Change from baseline: 0.01 (points) C: 2.15, SD 1.44 I: –0.31 (95% CI –0.55 to 0.07) C: +0.11 (95% CI –0.12 to 0.34) Difference: 0.42 (95% CI 0.09 to 0.75) Reginster I: n = 106 WOMAC total I: 1030.2 mm, Change from baseline: 0.020 76 2001 C: n = 106 (out of 2400mm) SD 473.8 I: –11.7% (95% CI –20.3 to –3.2) C: 939.7 mm, C: +9.8% (95% CI –6.2 to 25.8) SD 484.8 Difference: 21.6% (95% CI 3.5 to 39.6) WOMAC pain I: 194.1 mm, Change from baseline: 0.047 SD 101.9 I: ~–36.7 mm, SE 8.3 C: 172.2 mm, C: ~–7.5 mm, SE 10.6 SD 104.5 (estimated from graph) WOMAC function I: 740.1 mm, Change from baseline: 0.020 SD 364.2 I: ~–160 mm, SE 24.2 C: 670.8 mm, C: ~–65.5 mm, SE 31.5 SD 367.8 (estimated from graph) WOMAC stiffness I: 96.0 mm, NS SD 54.8 C: 96.7 mm, SD 54.6

C, control; I, intervention, NS, not significant.

0.42 points (95% CI 0.09 to 0.75) for stiffness). difference in the JOA total score or the gait, stairs, Reginster found statistically significant effects in range of motion or swelling subscores, or in the favour of glucosamine for the WOMAC function WOMAC stiffness and function subscores between subscale [change from baseline about –160 glucosamine hydrochloride and the control SE, 24.2 mm (VAS) for glucosamine and –65.5, group. Meta-analysis (Figure 5) illustrates the SE 31.5 mm for the control group (estimated small, but statistically significant favourable effect from graph)], but not for the stiffness subscale. for glucosamine. While the CIs overlap, visual Kawasaki did not find any statistically significant inspection identifies substantial heterogeneity, 48 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Glucosamine Placebo Standardised mean Study or difference IV Standardised mean difference IV subgroup Mean SD Total Mean SD Total Weight (random, 95% CI) (random, 95% CI)

Kawasaki 2008123 −3.6 20.74 49 −2.9 12.13 42 20.8% −0.04 (−0.45 to 0.37) Pavelká 200277 −2.0 1.63 101 −1.3 4.68 101 39.2% −0.20 (−0.48 to 0.08) Reginster 200176 −36.7 60.42 106 −7.5 77.17 106 40.0% −0.42 (−0.69 to −0.15)

Total (95% CI) 256 249 100.0% −0.25 (−0.46 to −0.05) Heterogeneity: τ2 = 0.01; χ2 = 2.60, df = 2 (p = 0.27); I2 = 23% Test for overall effect: z = 2.45 (p = 0.01) –1 –0.5 0 0.5 1 Favours glucosamine Favours placebo

FIGURE 4 Glucosamine sulphate and hydrochloride: a meta-analysis of WOMAC pain scores at end of study (fixed-effects model produces same point estimate; random-effects model presented as confidence interval better reflects the uncertainty around that point estimate). IV, inverse difference. with the glucosamine hydrochloride product were found between the patients followed up demonstrating little evidence of an effect. and those lost to follow-up. After a mean post- intervention follow-up of 5 years (8 years from trial Function/pain composite scores Pavelká found baseline), significantly fewer patients in the former a statistically significant effect in favour of glucosamine groups had received knee arthroplasty glucosamine compared with placebo for both than patients in the former placebo groups (9/144 the Lequesne index and the WOMAC total score in the glucosamine groups and 19/131 in the (difference between groups 0.91 points (95% placebo groups, RR 0.43 (95% CI 0.20 to 0.92, CI 0.34 to 1.5) for the Lequesne index and 3.1 p = 0.024). (95% CI 0.77 to 5.5) for the WOMAC total score). Reginster found statistically significant effects in Adverse events Neither Pavelká nor Reginster found favour of glucosamine for the WOMAC total score significant differences in reported adverse effects [per cent change from baseline, difference between between the glucosamine groups and the placebo groups 21.6% (95% CI 3.5 to 39.6)]. Kawasaki did groups. No significant differences in routine not find any significant difference in the WOMAC laboratory test results were seen, and Reginster total score between glucosamine hydrochloride and did not note any change in glucose homeostasis. the control group. Adverse events were not reported systematically in the study by Kawasaki. One patient in the Knee arthroplasty In a follow-up analysis52 of the glucosamine group and two in the control group trials by Reginster and Pavelká taken together, withdrew from the trial because of adverse events. patients with at least 12 months of treatment were followed up for their joint replacement status. Of Chondroitin 340 eligible patients, 275 (81%) could be followed Trial characteristics up. Of these, 144 had been in the glucosamine Of the three chondroitin trials, all were described groups of the trials and 131 had been in the as double-blind, placebo-controlled RCTs. The trial placebo groups. No significant baseline differences by Michel108 had a duration of 2 years, whereas the

Glucosamine Placebo Standardised mean Standardised mean Study or difference IV difference IV subgroup Mean SD Total Mean SD Total Weight (random, 95% CI) (random, 95% CI)

Kawasaki 2008123 −9.8 44.16 49 −7.5 41.49 42 22.3% −0.05 (−0.47 to 0.36) Pavelká 200277 −5.8 4.89 101 −3.7 4.35 101 38.3% −0.45 (−0.73 to −0.17) Reginster 200176 −160.0 176.18 106 −65.5 229.32 106 39.4% −0.46 (−0.73 to −0.19)

Total (95% CI) 256 249 100.0% −0.37 (−0.59 to −0.15) Heterogeneity: τ2 = 0.01; χ2 = 3.01, df = 2 (p = 0.22); I2 = 33% Test for overall effect: z = 3.25 (p = 0.001) –1 –0.5 0 0.5 1 Favours Favours glucosamine placebo

FIGURE 5 Glucosamine sulphate and hydrochloride: a meta-analysis of WOMAC function scores at end of study (fixed-effects model produces same point estimate; random-effects model presented as confidence interval better reflects the uncertainty around that point estimate). IV, inverse difference. 49

two trials by Uebelhart114,116 had a duration of 1 end of the trials (p = 0.001, random-effects model), year. The two trials by Uebelhart114,116 were industry no statistically significant heterogeneity, but the funded. No funding information was available for estimate was driven by the larger trial by Michel the trial by Michel.108 (Figure 6). Similarly, a significant effect in favour of chondroitin was shown for mean joint space width, A total of 456 people participated in the three with the weighted mean difference changing from trials. The trials included participants of a mean –0.01 mm (95% CI –0.30 to 0.28) at baseline to age of between 57 and 64 years, with between 48% 0.20 mm (95% CI 0.07 to 0.32) at the end of the and 82% being female. Osteoarthritis severity trials (p = 0.003, random-effects model) (Figure was not reported in the trial by Michel; in the 7). These results indicate less joint space loss with trials by Uebelhart, between 13% and 48% had chondroitin sulphate. Uebelhart (1998) also found OA of Kellgren–Lawrence grade 1 at baseline, a significant positive effect of chondroitin sulphate between 43% and 59% had grade 2, and between compared with placebo on some indicators of bone 9% and 30% had grade 3 [with more patients in metabolism. the Uebelhart (2004) study116 having more severe disease]. In all trials, baseline characteristics were Michel did a subgroup analysis of the patients well balanced in the comparison groups. with a minimum joint space width of 1 or more millimetres at entry (225/300 patients) and found Interventions no significant difference to their main results In the trial by Michel, 800 mg/day chondroitins (Table 15). They also state that mean joint space 4- and 6-sulphate [Condrosulf® (IBSA, Hungary)] width at baseline had no influence on radiographic were given as a tablet. In the trials by Uebelhart progression. (1998)114 and Uebelhart (2004),116 800 mg/day chondroitins 4- and 6-sulphate (Condrosulf) were Symptoms/pain Results for function and pain were given in powder form. In the Michel and Uebelhart very mixed and significant changes in joint space (1998) trials the treatment was given continuously, were not always mirrored by significant changes in whereas in the trial by Uebelhart (2004) treatment pain/function (Table 16). was administered intermittently for 3-month periods (from study entry to the third month, Pain Michel did not find any significant differences and from the sixth to the ninth months). Michel between chondroitin sulphate and placebo reported that analgesia use was similar between the for the WOMAC pain subscore. The trials by comparison groups; no information was given in Uebelhart (1998 and 2004), on the other hand, the trial by Uebelhart (1998), whereas in the trial found a statistically significant effect in favour by Uebelhart (2004) paracetamol consumption of chondroitin sulphate for the Huskisson visual was significantly greater in the placebo group at 12 analogue pain scale [a decrease of 3.66 cm in the months than in the intervention group. intervention group and 1.64 cm in the placebo group (p < 0.001) in Uebelhart (1998) and a Outcome measures and further details of the trials decrease of 24.5 mm in the intervention group are summarised in Table 10. and 15.3 mm in the placebo group (p < 0.05) in Uebelhart (2004). Trial quality The Michel and Uebelhart (2004) trials were Uebelhart (1998) did not report on the use of of good quality. The trial by Uebelhart (1998) rescue analgesia but Uebelhart (2004) noted was of poor quality, reporting little detail of statistically significantly less use of paracetamol the randomisation procedure and allocation in the chondroitin group compared with the concealment. placebo group at 12 months (25.8 tablets versus 55 tablets per month; p < 0.05). Michel reported no Results statistically significant difference. Structure Table 14 summarises the results for chondroitin on structural outcomes. Function Michel did not find any significant differences between chondroitin sulphate and The three chondroitin trials showed a significant placebo for the WOMAC function or stiffness effect in favour of chondroitin for minimum joint subscores. The trial by Uebelhart (1998) found no space width, with the weighted mean difference significant effect for overall mobility capacity, but changing from 0.02 mm (95% CI –0.20 to 0.24) at the trial by Uebelhart (2004) found statistically baseline to 0.18 mm (95% CI 0.07 to 0.30) at the significant results in favour of chondroitin sulphate 50 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52 continued 0.05 p -value (between groups) NS 0.04 < 0.05 NS 0.001 = mm, SD 0.57 (95% mm, SD 0.52 (95% mm, SD 0.48 (95% CI mm, SD 0.61 (95% CI –0.24 mm, SD 0.56 (95% CI –0.16 mm, SD 0.53 (95% CI –0.08 to Change from baseline/difference between groups Significant effect on bone metabolism found: osteocalcin, pyridinoline/creatinine and deoxypyridinoline/creatinine and keratin sulphate significantly lower than in significantly higher, placebo group, p CI 0.01 to 0.27) to –0.04) Difference: 0.14 Change from baseline: I: 0.0 0.09) C: –0.14 CI 0.00 to 0.24) to 0.02) Difference: 0.12 Change from baseline: I: +0.045 –0.03 to 0.12) C: –0.07 , SD 0.36 2 , SD 0.33 2 cm cm, SD 0.14 cm, SD 0.13 cm cm, SD 0.10 cm, SD 0.10 12) 1.17 12) 0.46 12) 0.36 14) 1.09 14) 0.44 14) 0.35 ======End of study I: ( n C: ( n I: ( n C: ( n I: ( n C: ( n , SD 2 , SD 2 cm cm, SD cm, SD cm cm, SD cm, SD mm, SD 0.15 mm, SD 0.14 12) 1.29 12) 0.51 12) 0.40 mm, SD 0.14 mm, SD 0.14 14) 1.08 14) 0.44 14) 0.34 ======Baseline 0.32 C: ( n 0.28 I: ( n 0.10 I: 3.04 C: 3.00 I: ( n 0.11 C: ( n 0.10 I: 2.41 C: 2.35 I: ( n 0.10 C: ( n Outcome Bone metabolism Medial femorotibial joint surface area Mean joint space width Medial femorotibial joint mean width Minimum joint space width Medial femorotibial joint minimum width 114 108 Summary of the effect chondroitin on joint structure Study Michel 2005 Uebelhart 1998

TABLE 14 TABLE 51

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Clinical effectiveness of glucosamine and chondroitin on the progression of OA of the knee 0.039 0.047 NS p -value (between groups) (95% CI –0.19 2 mm mm (95% CI 0.01 mm (95% CI 0.004 (95% CI –8.61 to 2 (95% CI –3.56 to 3.17) 2 mm (95% CI –0.20 to 0.18) mm mm (95% CI –0.53 to mm (95% CI –0.57 to mm (95% CI –0.23 to 0.14) mm C: –4.55 –0.49) Difference: 4.36 to 0.55) C: –0.29 –0.04) Difference: 0.28 Change from baseline: I: –0.006 Change from baseline: I: –0.04 to 0.55) to 8.91) C: –0.32 –0.07) Difference: 0.27 Change from baseline/difference between groups Change from baseline: I: –0.19 , SD 20.9 2 , SD 26.9 2 mm, SD 1.28 mm, SD 1.27 mm mm, SD 1.58 mm, SD 1.51 mm I: 4.20 C: 3.74 I: 3.61 C: 3.23 End of study I: 67.8 C: 58.7 , SD 24.4 2 , SD 27.2 2 mm, SD 1.47 mm, SD 1.39 mm mm, SD 1.51 mm, SD 1.46 mm I: 4.20 C: 4.03 I: 3.65 C: 3.54 Baseline C: 63.3 I: 68.0 Mean joint space width Minimum joint space width Outcome Joint surface area 116 Summary of the effect chondroitin on joint structure (continued) C, control; I, intervention; NS, not significant. Study Uebelhart 2004

52 14 TABLE DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Chondroitin sulphate Placebo Study or Mean difference IV Mean difference IV subgroup Mean SD Total Mean SD Total Weight (random, 95% CI) (random, 95% CI)

Michel 2005108 2.46 0.48 150 2.28 0.56 150 91.8% 0.18 (0.06 to 0.29) Uebelhart 1998114 3.50 1.00 14 3.6 1.30 12 1.6% −0.10 (−1.00 to 0.80) Uebelhart 2004116 3.61 1.51 77 3.23 1.27 76 6.6% 0.38 (−0.06 to 0.82)

Total (95% CI) 241 238 100.0% 0.18 (0.07 to 0.30) Heterogeneity: τ2 = 0.00; χ2 = 1.16, df = 2 (p = 0.56); I2 = 0% Test for overall effect: z = 3.319 (p = 0.001) –1 –0.5 0 0.5 1 Favours Favours control experimental

FIGURE 6 Chondroitin sulphate: a meta analysis of minimum joint space width at end of study (fixed-effects model produces same point estimate; random-effects model presented as confidence interval better reflects the uncertainty around that point estimate). IV, inverse difference

Chondroitin sulphate Placebo Study or Mean difference IV Mean difference IV subgroup Mean SD Total Mean SD Total Weight (random, 95% CI) (random, 95% CI)

Michel 2005108 3.04 0.53 150 2.86 0.61 150 90.3% 0.18 (0.05 to 0.31) Uebelhart 1998114 4.40 1.00 14 4.60 1.40 12 1.8% −0.20 (−1.15 to 0.75) Uebelhart 2004116 4.20 1.58 77 3.74 1.28 76 7.9% 0.46 (0.00 to 0.92)

Total (95% CI) 241 238 100.0% 0.20 (0.07 to 0.32) Heterogeneity: τ2 = 0.00; χ2 = 2.02, df = 2 (p = 0.36); I2 = 1% Test for overall effect: z = 2.98 (p = 0.003) –1 –0.5 0 0.5 1 Favours Favours control experimental

FIGURE 7 Chondroitin sulphate: a meta-analysis of minimum joint space width at end of study (fixed-effects model produces same point estimate; random-effects model presented as confidence interval better reflects the uncertainty around that point estimate). IV, inverse difference

TABLE 15 Subgroup analysis: summary from Michel 2005108

Change from baseline/difference between Study Subgroup Outcome groups p-value Michel 2005108 Patients with minimum Minimum joint Change from baseline: 0.01 joint space width ≥ 1 mm space width at entry (I: n = 114, C: I: +0.05 mm, SD 0.53 (95% CI –0.05 to 0.14) n = 111) C: –0.14 mm, SD 0.57 (95% CI –0.25 to –0.03)

Difference: 0.19 mm, SD 0.55 (95% CI 0.04 to 0.33) Mean joint Change from baseline: 0.006 space width I: +0.01 mm, SD 0.54 (95% CI –0.09 to 0.11)

C: –0.20 mm, SD 0.58 (95% CI –0.31 to –0.09)

Difference: 0.21 mm, SD 0.56 (95% CI 0.06 to 0.36)

C, control; I, intervention. 53

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Clinical effectiveness of glucosamine and chondroitin on the progression of OA of the knee 0.001, NS, < = 0.05 0.001 < < 0.01 0.05 0.05 Between treatments p between times p Between treatments p between times p NS < p -value NS < NS < NS Change from baseline: I: –3.9% C: +2.1% I: –36% C: –23% Change from baseline/difference between groups Change from baseline: I: –11.0% C: –6.2% I: –42% C: –25% Change from baseline: I: –7.8% C: –4.6% I: –18% C: –0.5% Change from baseline: I: –0.8% C: +5.9% mm, SD 27.6 mm, SD 27.4 cm, SD 2.5 cm, SD 2.3 cm, SD 2.1 cm, SD 1.4 I: 5.8, SD 3.6 C: 7.0, SD 3.9 I: 2.1 C: 4.8 I: 8.6 C: 6.8 End of study I: 34.3 C: 45.8 I: 20.1 seconds, SD 6.8 C: 22.7 seconds, SD 7.7 cm, SD 1.10 mm, SD 19.0 cm, SD 1.61 mm, SD 15.5 cm, SD 1.6 cm, SD 1.5 I: 2.3, SD 1.6 C: 2.6, SD 1.7 I: 5.76 C: 6.44 I: 9.0, SD 2.8 C: 9.1, SD 3.2 I: 5.1 C: 5.7 Baseline I: 2.5, SD 1.6 C: 2.7, SD 1.8 C: 61.1 I: 58.8 I: 2.1, SD 1.6 C: 2.5, SD 1.8 I: 24.5 seconds, SD 22.7 C: 22.8 seconds, SD 7.5 I: 3.0, SD 2.3 C: 3.5, SD 2.5 mm) WOMAC total WOMAC (0–10 Huskisson VAS cm) Lequesne index Overall mobility capacity (VAS) Outcome WOMAC pain WOMAC Huskisson VAS (0– Huskisson VAS 100 WOMAC function WOMAC Walking time Walking WOMAC stiffness WOMAC 114 116 108 Summary of the effect chondroitin on pain, function and composite measures health status Uebelhart 2004 Study Michel 2005 C, control; I, intervention; NS, not significant; VAS, visual analogue scale. C, control; I, intervention; NS, not significant; Uebelhart 1998

54 16 TABLE DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

for walking time [a decrease of 4.4 seconds in the The exercise programme was home and facility intervention group and 0.1 seconds in the placebo based and consisted of two 15-minute walking group (p < 0.05)]. sessions, separated by 20 minutes of strength training. Exercise sessions were held for an Function/pain composite scores Michel did not find hour each twice a week. In the trial by Messier, any significant differences between chondroitin paracetamol was allowed as a rescue medication. sulphate and placebo for the WOMAC total In the trial by Rai, patients were advised to take score. However, the trial by Uebelhart (2004) paracetamol for pain. Messier did not report found a statistically significant result in favour of whether the difference in analgesic consumption chondroitin sulphate for the Lequesne index [a between the groups was significant (a decrease of decrease of 3.2 points in the intervention group 37% by 12 months in the intervention and 11% and 2.1 points in the placebo group (p < 0.01)]. in the placebo group), and Rai did not report analgesic consumption. Adverse events None of the chondroitin trials found significant differences in reported adverse effects Outcome measures and further trial details are between the chondroitin groups and the placebo summarised in Table 10. groups. No significant differences in routine laboratory test results (where reported) were seen. Trial quality The trial by Messier was of moderate quality, Combination therapy with whereas the trial by Rai was of poor quality. Messier glucosamine and chondroitin did not explicitly describe randomisation, but Trial characteristics their description implies that randomisation was Both trials of glucosamine and chondroitin done centrally by the sponsor. Rai did not report combination therapy were double-blind, placebo- the method of randomisation, it was unclear controlled single centre trials with a duration of 1 whether allocation was concealed and the study was year. The trial by Messier122 was industry funded; described as double blind, but no further detail was no funding information was available for Rai.120 A given, withdrawals or drop-outs were not reported, total of 189 people participated in these trials. The and it was not reported whether analysis was by mean age was between 70 and 74 years for Messier intention to treat. and around 54 years for Rai. The trial by Messier included more women than men (between 66% and Results 76% women); no gender information was given by Structure Rai found a statistically significant result Rai. Neither of the trials gave details of OA severity for the preservation of minimum joint space or duration. Rai reported baseline characteristics with a combination of glucosamine sulphate to be well balanced in the comparison groups, and chondroitin sulphate compared with whereas in the trial by Messier, participants in the placebo [change from baseline –0.04 mm in the intervention group were significantly younger, intervention group and –0.13 mm in the placebo more overweight, and had a higher income than group (p ≤ 0.01)]. Table 17 summarises the trial those in the placebo group. findings for the effects on structure.

Interventions Symptoms/pain Table 18 summarises the findings for In the trial by Rai, it was only stated that Kondro, symptoms and pain. an oral combination of glucosamine sulphate and chondroitin sulphate, was given. No details Pain Messier did not report any statistically on dosing were given, but the Kondro capsules significant difference in the pain component of contain 250 mg glucosamine sulphate and WOMAC at 6 or 12 months. Rescue analgesia use 200 mg chondroitin sulphate, and presumably was decreased in both the treatment and placebo one capsule a day was given. In the trial by groups, but there was no statistically significant Messier, a combination of 1500 mg glucosamine difference between the groups. Rai did not report hydrochloride plus 1200 mg chondroitin on pain as a separate score nor did they report on sulphate was given each day, either in one dose analgesia use. or distributed over three doses a day. During the first 6 months, this was supplemented in Function Messier did not find any statistically both comparison groups by six healthy lifestyle significant differences either at 6 months (i.e. classes, and in the second 6 months, an exercise before the exercise intervention) or at 12 months programme was added that lasted for 6 months. for a combination of glucosamine hydrochloride 55

TABLE 17 Summary of the effect of combined glucosamine and chondroitin versus placebo on structure

Change from baseline/ Study Outcome Baseline End of study difference between groups p-value Rai Minimum joint I: 3.66 mm I: 3.62 mm, p = NS ≤ 0.01 2004120 space width vs baseline C: 3.65 mm C: 3.52 mm, p ≤ 0.01 vs baseline

C, control; I, intervention; NS, not significant.

plus chondroitin sulphate plus exercise (versus were seen with glucosamine administration at placebo plus exercise) for any of the following therapeutic doses. In 13 studies including over outcomes: WOMAC function subscale, the distance 800 participants, no laboratory or cardiovascular walked in 6 minutes or knee concentric extension abnormalities were seen with glucosamine strength. The intervention group had a statistically administration at therapeutic doses for a weighted significant benefit for knee concentric flexion average of around 40 weeks. Studies reporting strength and balance. common adverse events included 988 participants followed for a weighted average of 37 weeks. Pain/function composite scores Messier did not find Adverse events were reported less frequently any statistically significant differences either at with glucosamine than with placebo [rate ratio 6 months or at 12 months for a combination of 0.76 (95% CI 0.61 to 0.92, p < 0.05)]. Five studies glucosamine hydrochloride plus chondroitin reported adverse events with glucosamine in sulphate plus exercise versus placebo plus exercise comparison with adverse events with ibuprofen, for results of the WOMAC total score. In the trial and the prevalence of adverse events was 10.0% by Rai, a statistically significant benefit of the with glucosamine and 32.5% with ibuprofen. combination treatment compared with placebo was seen on the Lequesne index. However, compared Stumpf and colleagues132 reviewed data on the with other studies reporting results of the Lequesne effect of glucosamine on glucose control. The index, the deterioration seen in this trial in the review was not a systematic review, but a search placebo group seems rather extreme (4.6. to 3.7 in strategy was described. Of 16 clinical studies the intervention group versus 4.9 to 11.48 in the of oral glucosamine, only five included plasma placebo group). glucose concentrations that could be pooled. After 3–12 weeks of treatment, glucose values in the Adverse events Neither of the trials of glucosamine/ glucosamine groups were not significantly different chondroitin combination therapy reported from those in the placebo groups. Two 3-year systematically on adverse effects. studies (Pavelká and colleagues77 and Reginster and colleagues76) also found no significant effect Adverse effects of glucosamine or on glucose levels. However, the authors concluded chondroitin – additional data that data were limited and longer-term effects were No long-term observational studies were identified uncertain. that investigated any adverse events association with glucosamine (any form) or chondroitin One RCT by Scroggie and colleagues133 studied 34 sulphate or a combination of the two. patients with type 2 diabetes and a mean age of 69– 71 years. Patients in the intervention group (n = 22) Anderson and colleagues64 conducted a review were given a combination of 1500 mg glucosamine of adverse effects of glucosamine which did not hydrochloride plus 1200 mg chondroitin sulphate have the characteristics of a systematic review, but per day. Twelve patients were in the placebo group. seemed to aim at some thoroughness (literature After 90 days of treatment, glycaemic control search described, obviously some formal extraction (HbA1c levels) were not significantly different of data). For their safety assessments in humans, between the comparison groups. they summarised data from 33 studies in 3063 patients. In 16 studies of a total of 854 participants Knudsen and Sokol134 presented a case report of a followed for a weighted average of 37 weeks, potential interaction of glucosamine/chondroitin no significant changes in blood glucose values with warfarin in a single patient, resulting in an 56 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

TABLE 18 Summary of the effect of combined glucosamine and chondroitin versus placebo on measures of pain, function and composite health status scores

Study Outcome Baseline End of study p-value Messier WOMAC pain I: 7.1, SE 0.5 6 months: NS 122 2007 C: 5.9, SE 0.5 I: 6.2, SE0.4 C: 6.2, SE0.4 12 months: I: 6.0, SE0.5 C: 5.18, SE0.5 WOMAC function I: 25.9, SE 1.7 6 months: NS C: 21.1, SE 1.5 I: 21.9, SE 1.1 C: 22.9, SE 1.1 12 months: I: 19.4, SE 1.2 C: 20.6, SE 1.2 6-minute walk I: 384.7 m, SE17.6 6 months: NS C: 398.7 m, SE17.3 I: 393.6 m, SE 8.0 C: 396.5 m, SE 7.9 12 months: I: 409.2 m, SE 8.7 C: 410.5 m, SE 8.6 Knee concentric extension strength I: 209.4 N, SE 31.2 6 months: NS C: 163.9 N, SE 20.6 I: 176.9 N, SE 16.3 C: 202.7 N, SE 17.5 12 months: I: 207.6 N, SE 14.6 C: 209.7 N, SE 15.0 Knee concentric flexion strength I: 106.0 N, SE 16.1 6 months: NS at 6 months, C: 83.0 N, SE 10.9 I: 106.1 N, SE 7.3 p = 0.05 at 12 months C: 106.7 N, SE 7.8 12 months: I: 102.9 N, SE 7.7 C: 124.8 N, SE 8.3 Balance (foot length) I: 0.52, SE 0.04 6 months: p = 0.01 at 6 C: 0.53, SE 0.03 I: 0.523, SE 0.014 months, p = 0.05 at 12 months C: 0.583, SE 0.017 12 months: I: 0.538, SE 0.017 C: 0.591, SE 0.020 Rai 2004120 Lequesne index I: 4.6 I: 3.7 ≤ 0.01 C: 4.9 C: 11.48

C, control; I, intervention; N, Newton; NS, not significant.

increased international normalised ratio (INR) and the International Pharmaceutical Abstracts (i.e. a higher risk of bleeding). The authors did database. They identified 43 unique cases of a search of the US FDA MedWatch database and an increased INR in people using warfarin and the WHO Adverse Drug Reactions database, as glucosamine or a glucosamine/chondroitin sulphate well as a literature search of MEDLINE, EMBASE combination product (20 cases from MedWatch, 21 57

cases from the WHO database, and one published and we included eight trials that met our inclusion case report). The reports showed a temporal criteria of at least 12 months’ follow-up. relationship between glucosamine or combination product use and warfarin, with a number of the Evidence of clinical effectiveness reports stating that patients had a stable INR when Glucosamine they started taking glucosamine, and that the event resolved again in most cases when glucosamine was Two RCTs76,77 investigated glucosamine sulphate, stopped. The Therapeutic Goods Administration both using the Rotta preparation (including a of the Australian Government has received 12 total of 414 participants), and one123 investigated reports (nine included in the WHO data) of glucosamine hydrochloride (including a total of interactions between warfarin and glucosamine. In 142 participants). Owing to the small number 10 of these case reports, clinically significant rises of trials eligible, we could not compare different in INR were observed 4–20 days after initiation of forms of glucosamine preparations, different doses glucosamine.135 or monotherapy with combination therapy.

A case report in the UK136 reported the death Glucosamine sulphate (Rotta product) of a man who developed acute liver failure. Our meta-analysis showed a statistically significant The otherwise healthy 64-year-old man had effect of glucosamine sulphate in comparison recently started treatment with glucosamine. A with placebo on joint space width. The clinical fatal accident enquiry concluded that there was significance, while not meeting the pre-stated insufficient evidence to attribute the death to meaningful change used by trialists when glucosamine. calculating study power (change of 0.33–0.50 mm reported), was supported by the 8-year follow- up data that showed significantly fewer knee Summary arthroplasties in the glucosamine group than in the placebo group (6.3% versus 14.5%; RR 0.43). We summarise below the evidence of clinical Both trials also found statistically significant results effectiveness of glucosamine and chondroitin, in favour of glucosamine for almost all pain and highlighting where data were suitable for inclusion function parameters examined (except WOMAC in economic modelling and where there were stiffness in one trial). The clinical significance important gaps or uncertainty in the evidence. was uncertain. No a priori clinically significant change was specified by the trialists. Changes Systematic reviews in function were less than suggested by Tubach and colleagues39 to be of clinical importance. It This review summarised five systematic reviews is important to note that both these trials used and one clinical guideline – two on glucosamine, the Rotta preparation, a powdered oral form of two on chondroitin and two on glucosamine glucosamine. and/or chondroitin for OA of the knee. Taken together, the reviews summarised 45 trials – 21 of Participants in the two relevant trials were slightly glucosamine, 20 of chondroitin sulphate and four younger than might be expected in a clinical of a combination of glucosamine and chondroitin. population of people with OA of the knee (mean There were considerable differences between the 61–66 years) and included a high proportion of reviews in terms of the studies included – with women (75–79%). Participants had experienced OA respect to parameters such as study duration, for variable durations with a mean of 8–10 years, comparison interventions and inclusion of but with substantial heterogeneity. One of the treatment for disorders other than OA of the knee. trials77 required participants to have a Lequesne In particular, the lack of presentation of analysis for index of at least 4 and not exceeding 12, and so all outcomes and conclusions separately for long- excluded people with mild symptoms (1–4) and term studies made it difficult to draw conclusions those with very severe symptoms. The recruited about the effectiveness of glucosamine and population had a mean score of approximately chondroitin on long-term OA disease progression. 9 at baseline reflecting ‘severe’ symptoms. Obese This was less of an issue for structural change people were excluded from the trials. As indicated outcomes as few studies presented this finding, and in Chapter 2, the decision to undertake knee where they did, the duration was generally longer. arthroplasty is based on clinical assessment but is No major safety concerns were highlighted in the influenced by many factors, including patient and reviews. We, therefore, focused our assessment of health service characteristics. The comorbidities 58 clinical effectiveness on analysis of primary RCTs of participants were poorly reported. The clinical DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

setting for the two trials in the Czech Republic and groups. While statistically significant, the clinical Belgium may differ substantially from the UK. In significance of a 1-point difference on this scale the UK, there are few data about the proportion of is uncertain. The Lequesne index is scored out of people with OA receiving knee arthroplasty, so it is 24, but anything over 13 is considered to represent difficult to determine if the surgical rate of 14.5% extremely severe symptoms. in 8 years from ‘referral’ (plus a mean of 8 years of OA prior to referral) reflects practice that is Combined glucosamine and chondroitin generalisable to the UK. Two poor-to-moderate quality studies investigated a combination of glucosamine (one In addition, the studies with 8-year follow-up of glucosamine sulphate, one of glucosamine ended as RCTs after 3 years. During the ensuing hydrochloride) and chondroitin (including a follow-up, participants were not restricted to total of 189 participants).120,122 Only one of the treatment choices and substantial treatment group trials120 measured joint space parameters and switching may have occurred. The study design, found a statistically significant result in favour of utilising routine health-care data to identify the treatment with respect to loss of minimum knee arthroplasties, precluded the gathering of joint space width. One trial122 did not report any information about medicine use after the RCT significant difference between the combination ended or further information about QoL. This therapy and placebo on any of the WOMAC switching would, if anything, lessen the effect parameters measured. The other trial120 reported size observed when comparing the treatment a clinically and statistically significant effect on and control groups based on their original trial the Lequesne index, with an improvement in the allocation. treated group and a worsening of the mean score from moderate to very severe. However, the study Glucosamine hydrochloride was of poor quality with unclear randomisation, Only one poor quality trial of glucosamine allocation concealment and analysis. hydrochloride was identified.123 No significant differences between glucosamine hydrochloride The observational GAIT ancillary study was not and the placebo control were seen for any of the included in this review, but reported differences in structural, pain or functional outcomes assessed. joint space loss that were not statistically significant We therefore have very limited evidence about the when comparing glucosamine, chondroitin or a effectiveness of hydrochloride preparations, and combination of glucosamine and chondroitin with nothing to support its clinical effectiveness. placebo. The smallest joint space loss was observed in the glucosamine group, but the study suffered Chondroitin from being underpowered, with fewer participants Three studies investigated chondroitin sulphate meeting the inclusion criteria than anticipated. (including a total of 456 participants).108,114,116 Our meta-analysis showed small, but statistically All of the reviews or trials examined suggested that significant effects on both mean and minimum there were no safety concerns with glucosamine or joint space width in favour of chondroitin sulphate. chondroitin sulphate. Examination of additional As noted previously, while this surrogate end literature suggested that glucosamine (and point has been widely used within trials, the probably chondroitin) does not have any significant clinical significance of joint space narrowing effect on glucose homeostasis, despite clinical in relation to long-term disease progression concerns that there may be an impact, or on the and knee arthroplasty is uncertain. The lack of incidence of any other adverse events or laboratory correlation between symptoms and joint space abnormalities. However, a possible interaction in loss makes it difficult to translate this outcome people taking warfarin was suggested. In general, into a measure of health utility that can be utilised the studies investigating relevant adverse events in any health economics modelling. There were parameters were small and data on long-term no studies following participants for sufficient follow-up were lacking. The association with acute duration to assess the effect on knee arthroplasty. liver toxicity is uncertain. Results for pain and function were not consistent between studies, with studies using different Table 19 gives a summary of the findings from the tools to measure effect. Of the two larger, good clinical effectiveness review focusing on effect after quality studies, one108 reported no significant at least 12 months’ follow-up. Statistical significance effect on WOMAC scores while the other116 found is reported, but may not necessarily reflect clinically improvements on the Lequesne index for both important differences. the placebo (2 points) and treated (3 points) 59

In conclusion, we found evidence of long-term importance, in both the treatment and control effectiveness for the Rotta product of glucosamine groups. There was evidence of no reduction in sulphate from two good quality trials, although the use of analgesia and very limited long-term both were funded by the manufacturers. The safety data but no safety issues were identified in evidence of modification to structural change short-term studies. For all other preparations, was supported by a difference in rate of knee there was insufficient evidence of effectiveness, arthroplasty after 8 years of follow-up. WOMAC heterogeneity in the trial data or no evidence to scores were measured for 3 years and demonstrated enable conclusions to be drawn. minor improvements, of borderline clinical

TABLE 19 Summary of findings from the clinical effectiveness review

Effect Evidence summary Glucosamine sulphate prevents or slows progression (compared with placebo): Joint space loss Evidence of effectiveness Pain/function Evidence of effectiveness Knee arthroplasty Evidence of effectiveness Reduction in need for analgesia Evidence of no effectiveness Glucosamine hydrochloride prevents or slows progression (compared with placebo): Joint space loss Evidence of no effectiveness (poor quality) Pain/function Evidence of no effectiveness (poor quality) Knee arthroplasty No evidence Reduction in need for analgesia Evidence of no effectiveness (poor quality) Chondroitin prevents or slows progression (compared with placebo): Joint space loss Evidence of effectiveness Pain/function Heterogeneity of results Knee arthroplasty No evidence Reduction in need for analgesia Mixed evidence Glucosamine/chondroitin combination prevents or slows progression (compared with placebo): Joint space loss Evidence of no effectiveness (poor/moderate quality) Pain/function Heterogeneity of results Knee arthroplasty No evidence Reduction in need for analgesia Evidence of effectiveness (poor/moderate quality) Combination offers greater gains over single medicine No evidence treatment Identification of optimal dose or product No evidence Assurance of safety Short-term/trial setting Evidence of safety Long-term/clinical setting Very limited evidence

60 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Chapter 4 Cost-effectiveness of glucosamine and chondroitin in OA of the knee

Introduction prescribed medicines (includes antacids, hypnotics and anxiolytics, antidepressants, analgesics, The review of clinical effectiveness shows corticosteroids, drugs for nutritional, blood, that of the therapies that were reviewed, only musculo-skeletal and joint diseases, and local glucosamine sulphate was demonstrated to have a anaesthetics)’. statistically significant effect on QoL and/or disease progression. Thus, cost-effectiveness modelling was limited to a comparison of glucosamine sulphate Methods versus current care. The model draws upon several data sources, including the paper by Bruyere The costs and effectiveness of glucosamine sulphate and colleagues,52 which uses pooled data from were modelled using cohort simulation (Figure 8). two previous RCTs of the clinical effectiveness of Cohort simulation involves the placement of a glucosamine sulphate.76,77 However, there was a lack number of individuals with pre-defined baseline of data on costs and QoL from the pooled study. clinical characteristics into particular states Thus, the model also draws upon other studies that of health, with associated costs and QoL, who report this information. Further detail is outlined may then move or not move to other states of in the Methods section and Appendix 7. One health over time. Owing to a lack of well-defined important issue to note relates to the definition discrete states in knee OA, the approach taken of current care. In the model presented below, here involved the placement of the cohort into an estimates of the costs of current care are based on initial baseline level of health status, followed by two UK sources,32,137 while estimates of comparative change in health status over time. Thus, rather effectiveness are based on the definition of than using discrete states, health was modelled current care used by Bruyere and colleagues,52 along a continuum. In addition to the continuous that is: ‘…any care received from health-care health status measure, two additional states were providers, with or without physical therapies and used. First, to reflect the greater costs and QoL investigations (hydrotherapy, exercise, ultrasound, implications associated with knee replacement radiological procedures), and treatment with arthroplasty, a separate state was included for this

Asymptomatic

Glucosamine Death

TKR progression

Asymptomatic

Current practice Death

TKR progression

FIGURE 8 Outline schematic of model structure. TKR, total knee replacement. 61

outcome. Knee arthroplasty can include a number Predicted HUI3 utility score = 0.5274776 + of different surgical procedures. For the purpose (0.0079676 × pain) + (0.0065111 × stiffness) of the model, total knee replacement (TKR) – (0.0059571 × function) + (0.0019928 × pain surgery was assumed. Further, although therapy × stiffness) + (0.0010734 × pain × function) + had no effect on survival, as the time horizon was (0.0001018 × stiffness × function) – (0.0030813 remaining lifetime, the state of death was also pain2) – (0.0016583 × stiffness2) – (0.000243 included. An outline schematic of model structure × function2) + (0.0113565 × age in years) – is shown in Figure 8. (0.0000961 × age in years2) – (0.0172294 × female) – (0.0057865 × years since onset of OA The cohort was pre-defined according to the in the study knee) + (0.0001609 × years since baseline clinical characteristics reported within onset of OA in the study knee2). Pavelká and colleagues.77 We used this source as this was the only published randomised The authors show with the aid of a worked example placebo-controlled trial of therapy that reported that the model predicts that a 56-year-old woman baseline and follow-up QoL data, using an with WOMAC pain, stiffness, and function scores instrument (WOMAC) that permitted conversion of 10, 5, and 24, respectively, who has had OA for to a preference-based utility scale (HUI3). The 2.5 years, would have a HUI3 score of 0.68. For our conversion from Likert WOMAC score to HUI3 was cohort, we obtained a similar value (0.69) to this performed using Grootendorst and colleagues.138 with different parameters. We used less disabled Grootendorst and colleagues138 conducted a WOMAC pain, stiffness and function scores of multicentre randomised trial of hylan G-F 20 6, 2 and 22 (the higher the score, the worse the against appropriate care among 255 Canadian disability), and higher values for OA duration patients with symptomatic knee OA of mild to and age, reported from the mean intention- moderate severity, who had previously been treated to-treat values in the placebo group in Pavelká with analgesics, in which patients completed and colleagues.77 We were only able to use the WOMAC and HUI3 questionnaires at 2-monthly Pavelká and colleagues77 estimate as Reginster intervals over 12 months. It is important to use and colleagues76 reported VAS WOMAC scores preference-based scales such as the HUI3, in order rather than the Likert version. Further, for ease to permit comparisons of cost-effectiveness with of illustration we approximated the Pavelká and other therapies using quality-adjusted life-year colleagues77 demographic estimates by using OA (QALY) measures. A recent paper,139 published duration of 10 years and a starting age of 60 years after model development, has produced EQ-5D (the mean values in the placebo group were 11 estimates derived from WOMAC for knee pain years and 63 years respectively). patients. However, as the model does not account for disease duration, and was not derived from Once the cohort was placed in the starting level a population of knee OA patients, we chose to of health, thereafter individuals had a particular retain the Grootendorst and colleagues’ mapping probability of entering the TKR state, or avoiding algorithm. 138 TKR and progressing to a different state of health, or dying. Within the glucosamine group, Based on Pavelká and colleagues77 and individuals were assumed to remain on therapy Grootendorst and colleagues,138 the cohort had until death. Probability of entering the state of TKR an initial QoL score of 0.69 (on a 0–1 scale, was derived from Bruyere and colleagues;52 the where 0 = dead and 1 = perfect health). It is TKR risk was annualised using a standard formula, useful to note that this value is identical to the p* = 1 – (1 – p)1/t, where p = event probability over EQ-5D based value from Barton and colleagues,139 study duration and t = time in years. Probability of used by the NCCCC,1 which modelled the cost- death was estimated from age-specific all-cause life effectiveness of oral NSAID or COX inhibitor tables.140 As explained above, the QoL associated therapy for OA. The value of 0.69 was computed with health change over time among those neither using the best performing model (model 3); this dying nor entering the TKR state was taken from model had the lowest mean absolute error (MAE) Grootendorst and colleagues,138 who estimated and the lowest root mean square error statistics regression models that permit conversion of HUI3 (RMSE) (MAE = 0.1628, RMSE = 0.2065), relative scores using WOMAC subscale scores, age, and to the closest alternative model that included duration of OA in the study knee. Owing to a lack Kellgren radiographic grade (MAE = 0.1638, of data reporting QoL prior to TKR within Bruyere RMSE = 0.2083). Model 3 is shown as follows: and colleagues,52 QoL prior to TKR was estimated

62 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

from another source, using baseline (48 hours prior and follow-up, with and without therapy, were to surgery) WOMAC subscale scores published by converted to HUI3 using the regression model of Nunez and colleagues,141 giving a value of 0.51. Grootendorst and colleagues.138 The calculation is Within the model, individuals who entered the shown in Table 20. ‘prior to TKR’ health state did so at the beginning of each annual cycle and remained in that state Health-care costs associated with knee OA were for the complete cycle length, i.e. 12 months, then estimated from a UK study conducted by Lord and rejoined the larger non-progressive cohort in future colleagues.32 This study was an RCT of primary cycles. Added to this, for both progressors and care-based education for knee OA. In both control non-progressors, account was taken of the natural and intervention groups, costs were related to the history of knee OA, using data from Johnson and resources consumed for GP visits, medications, colleagues,22 who reported 7-year WOMAC subscale outpatient visits, inpatient care, professions score data for 43 Canadian patients (mean age at allied to medicine consultations, complementary baseline 69 years, 77% female) with moderate to therapists and X-ray procedures for a population severe hip or knee OA. Converting to HUI3 using of patients with radiographic evidence of knee OA. the Grootendorst and colleagues model, 138 the Data were collected from case notes, supplemented difference in QoL after 7 years of 0.04 gives an by patient interview for data on professions allied annual QoL decrement value of 0.006. to medicine consultations. We used the reported NHS direct cost element, uprated to 2007–8 The effectiveness of therapy on health states was prices, for the control group who did not receive obtained from two main sources. First, we used the education programme (£216 per patient per the estimate from Bruyere and colleagues52 of the year). This estimate was very similar to the values effectiveness of therapy in reducing the risk of reported in a more recent study by Hurley and undergoing TKR surgery (RR = 0.43, 95% CI 0.20 colleagues,137 who calculated 6-month health-care to 0.92). The Bruyere estimate was based on pooled and social care costs of £103 for usual care among data from 275 patients who had participated in two chronic knee pain patients in the UK at 2003–4 previous randomised placebo-controlled trials of prices. The cost of TKR surgery (£6126) was glucosamine sulphate,76,77 with follow-up data for a calculated by uprating to 2007–8 prices the 2005–6 mean period of 5 years after the end of both 3-year NHS Reference Costs estimate for elective primary trial periods. knee replacement (code HRG H04, £5747).142 We used the lower and upper interquartile range Second, the effectiveness of therapy in improving estimates, uprated to 2007–8 prices, in probabilistic QoL for those who did not progress to TKR was analysis, applying a uniform distribution. No UK estimated using Pavelká and colleagues,77 who published prices were available for glucosamine demonstrated a statistically significant change sulphate, as the product has no UK licence. UK in WOMAC total scale and subscale scores. To prices were, however, available for the licensed generate the annual utility improvement with hydrochloride preparation. Listed prices for therapy, the WOMAC subscales scores at baseline glucosamine sulphate in the USA were US$34.95

TABLE 20 Calculation of Health Utilities Index (HUI) from WOMAC scores

WOMAC HUI

Annualised Baseline Follow-up Baseline Follow-up Difference differencea QoL gain Placebo Pain 6.33 Pain 5.23 0.687186 0.717122 0.029936 0.010080 Function Function 22.00 18.30 Stiffness 2.15 Stiffness 2.26 Therapy Pain 6.61 Pain 4.61 0.689013 0.733497 0.044484 0.015050 0.004970 Function Function 21.84 16.04 Stiffness 2.25 Stiffness 1.94

a Calculated using a standard formula of 1 – (1 – v)1/t , where v = HUI difference (column 6) and t = 3. 63

for 30 days’ supply,143 giving a total annual cost of Second, to take account of parameter uncertainty, US$429.40 with international shipping, or £242.60 probabilistic sensitivity analysis was employed. at an exchange rate of £1 = US$1.77. Purchasing This involved taking 5000 random draws from power parity (PPP) conversion from US$ gives specific distributions of all stochastic parameters a cost of £280.40. The price to the NHS is likely (current care TKR probabilities, relative risk to be cheaper, however, given that the NHS is reduction of TKR associated with therapy, current a monopoly buyer, and glucosamine sulphate care QoL score, QoL gain with therapy, current would be likely to be priced at a similar level to care health-care costs, TKR costs). More details on competitor products, such as the hydrochloride the specific distributions used are outlined in the formulation. For the purposes of assessing the cost- footnotes to Table 23. The distributions selected effectiveness of a likely guide price in the future, we for each parameter reflect common candidate therefore used current UK published market prices distributions.146 of glucosamine hydrochloride, giving an annual cost of £221 per patient per year. In addition Value of information analysis was conducted to however, based on the PPP conversion price of investigate the worth of commissioning further £280.40, we used upper and lower bounds of 27% research on the cost-effectiveness of therapy. in deterministic sensitivity analysis to assess the Population expected value of perfect information impact of alternative prices on cost-effectiveness. (EVPI) was computed, assuming a constant Both costs and health outcomes were discounted population size of 500,000 knee OA patients at 3.5%. Assumptions for model parameters are over a discounted 10-year time horizon. To assess outlined in Table 21. which parameter, or set of parameters, should be the focus for any future research, analysis of Data analytic procedures covariance techniques were used to explore the proportionate contribution of each parameter to The sensitivity of the results to different model variation in incremental cost and QALYs. This was assumptions was explored in two ways. First, both implemented using the ‘Covariance’ command stochastic and non-stochastic parameters were in excel; this allows determination of whether varied individually using deterministic one-way large values of one variable tend to be associated sensitivity analysis. These parameters included with large values of another (positive covariance), glucosamine sulphate acquisition costs, discount whether small values of one variable tend to be rate, proportion of patients requiring TKR associated with large values of another (negative surgery, health-care costs, QoL scores and TKR covariance), or whether values of both variables probabilities. Table 22 provides information on the tend to be unrelated (covariance near zero). We assumptions used to generate the lower and upper used the covariance between log incremental bounds for each parameter. costs, log TKR costs, log health-care costs and

TABLE 21 Model assumptions and parameter

Parameter Value Data source QoL at baseline – current care 0.687186 Pavelká 2002,77 Grootendorst 2007138 Annual probability of progression to TKR – current care 0.019397 Bruyere 200852 Annual probability of progression to TKR – glucosamine sulphate 0.008035 Bruyere 200852 Annual probability of not progressing to TKR – current care 0.980603 Bruyere 200852 Annual probability of not progressing to TKR – glucosamine sulphate 0.991965 Bruyere 200852 QoL prior to TKR 0.511682 Nunez 2007,141 Grootendorst 2007138 Annual reduction in QoL due to natural history 0.005986 Johnson 200722 Annual QoL change due to glucosamine sulphate 0.004970 Pavelká 200277 Health-care costs of knee OA per patient per year £216 Lord 199932 Health-care TKR costs £6126 Department of Health 2006142 Glucosamine sulphate acquisition costs per patient per year £221 British National Formulary 2008144 Discount rate 3.5% The Green Book 2007145

QoL, quality of life; TKR, total knee replacement. 64 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

TABLE 22 Deterministic sensitivity analysis – parameter values and assumptions

Parameter Value Data source and notes Annual probability of TKR – current care 0.009698–0.029095 Bruyere 200852 and authors’ judgement. A 50% decrease and increase in probability of TKR described in Bruyere 200852 is applied TKR relative risk reduction 0.20–0.92 Bruyere 2008.52 Range is reported 95% CIs of TKR relative risk reduction Annual QoL change due to glucosamine 0.001242–0.008871 Pavelká 2002.77 Range is based on lower and upper sulphate 95% CIs of total WOMAC change score (0.77 to 5.5), expressed as percentage changes from mean difference, and applied to mean QoL change of 0.004970 calculated from Pavelká 200277 Glucosamine sulphate acquisition costs £161–£280 Upper value is US price of Rotta glucosamine sulphate per patient per year product converted to GBP using PPP. Lower value is computed using equivalent percentage difference of 27% Health-care costs of knee osteoarthritis £0–£1053 Hurley 2007.137 Range is lower and upper 95% CI of per patient per year means costs calculated from published standard deviation of £185 Health-care TKR costs £5171–£6938 Department of Health 2005 (Appendix NSRC1)147 Discount rate 0–5% Authors’ assumptions

QoL, quality of life; PPP, purchasing power parity; TKR, total knee replacement.

TABLE 23 Probabilistic sensitivity analysis – parameter distributions

Parameter Data source Distribution Annual probability of progression to TKR – Bruyere 200852 Beta binomiala (2, 98) current care Relative risk of TKR with therapy Bruyere 200852 Normalb (–0.84, 0.39) Health-care costs of knee OA per patient Lord 199932, Hurley 2007137 Gammac (42.14, 4.82) per year TKR costs Department of Health 2005 Uniformd (5171, 6938) (Appendix NSRC1)147 QoL score – current care Pavelká 2002,77 Grootendorst 2007138 Beta method of momentse (339, 523) QoL gain with therapy Pavelká 2002,77 Grootendorst 2007138 Normal (0.004970, 0.001950) QoL prior to TKR Nunez 2007,141 Grootendorst 2007138 Beta method of moments (438, 1570)

QoL, quality of life; TKR, total knee replacement. a Beta binomial (α, β) refers to a beta distribution with α = number of events and β = sample size minus number of events. b Uniform (α, β) refers to a uniform distribution with α = lower interquartile range and β = upper interquartile range. c Gamma (α, β) refers to a gamma distribution with α = (sample mean/SE)2 and β = SE2/sample mean. d Normal (α, β) refers to a normal distribution with α = sample mean and β = SE. e Beta method of moments (α, β) refers to beta distribution where α = sample mean (sample mean (1 – sample mean)/ SE2) – 1 and β = sample mean * (1 – sample mean)/(SE2) – 1 – α.

TKR probability with glucosamine sulphate, and across parameters. In addition, we conducted EVPI the covariance between incremental QALYs, for parameters, in which repeated 1000-model QoL score with glucosamine sulphate, QoL prior simulations were run with each parameter held to TKR and TKR probability with glucosamine constant in turn, while allowing other stochastic sulphate, and then computed the proportionate parameters to vary. contribution of each parameter to the overall sum

Results Figure 11 shows the cost and QALY differences between therapy and current care within the Over a lifetime horizon (mean life expectancy of cost-effectiveness acceptability curve framework. 22.61 years) the mean discounted costs per patient This depicts the change in the probability for associated with current care were calculated as differing values of willingness to pay for a QALY £4634. Care costs with the addition of glucosamine (or ‘ceiling ratio’). This shows that glucosamine sulphate were estimated to be £7039 over the same was not expected to be cost-effective if we were time period. Thus, the additional discounted costs only prepared to pay up to £10,000 per QALY associated with glucosamine were £2405. Mean (probability of cost-effectiveness = 0.08), but discounted QALYs were 8.17 for current care and for increasing amounts of willingness to pay per 8.28 for glucosamine, giving a QALY gain of 0.11. QALY, the probability of cost-effectiveness rises Avoiding rounding error, the incremental cost per (e.g. if we were prepared to pay up to £20,000 QALY gain was £21,335. per QALY, the probability of cost-effectiveness = 0.43). At a willingness-to-pay ceiling ratio of above Deterministic sensitivity analysis revealed that approximately £22,000, the probability of cost- estimates were reasonably robust to variations effectiveness becomes greater than 0.5. in assumptions (Figure 9). Estimates were most affected by changes in the QoL gain associated with The EVPI provides an upper bound of the value of therapy. Changes to the costs of current care or undertaking further research to reduce uncertainty TKR costs had negligible effects. over whether therapy is more cost-effective than current care. Figure 12 shows the relationship The results of the probabilistic sensitivity analysis between these values and different ceiling ratios are presented in Figures 10 and 11. Figure 10 shows of willingness to pay per QALY. This reveals the scatterplot of differences in costs and QALYs that the value of further research was high for a from the 5000 replications. It was apparent that range of commonly applied ceiling ratios (e.g. therapy was unlikely to be cost saving, as there were £10,000–£50,000), and far exceeded the likely only 41 occurrences of costs within the glucosamine costs of research. The EVPI became lower as the group being lower than those in the current ceiling ratio approached £100,000, as uncertainty care group. However, there was a wide scatter surrounding cost-effectiveness became less for across the x-axis (depicting QALY differences). higher levels of willingness to pay per QALY.

Incremental cost per QALY gain (£000) Incremental cost per QALY 0 50% less TKR 50% more TKR Lower QoL gain Lower TKR RRR Higher QoL gain Higher TKR RRR Lower care costs Lower TKR costs Lower drug price Higher care costs Higher TKR costs Higher drug price 5% discounted costs 0% discounted costs 5% discounted QALYs 0% discounted QALYs 5% discounted costs and QALYs 0% discounted costs and QALYs

FIGURE 9 Deterministic sensitivity analysis. QALY(s), quality-adjusted life-year(s); QoL, quality of life; RRR, relative risk reduction; TKR, total knee replacement. 66 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

7 6 5 4 3 2

1 0 –0.10 –0.05 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 Incremental cost (£000) –1 –2 –3 QALY gain

FIGURE 10 Scatterplot of incremental cost and quality-adjusted life-year (QALY) gained.

1.2

1 Current practice 0.8 Glucosamine

0.6

0.4

Probability cost-effective Probability 0.2

0 10 20 30 40 50 60 70 80 90 100 Value of ceiling ratio (£000)

FIGURE 11 Cost-effectiveness acceptability curves.

1400

1200

1000

800

600

400 Value of EVPI (£million) Value 200

0 0 10 20 30 40 50 60 70 80 90 100 Value of ceiling ratio (£000)

FIGURE 12 Expected value of perfect information (EVPI). 67

Figures 13 and 14 show the analysis of covariance Discussion results. Figure 13 reveals that the greatest uncertainty in costs was related to variation in TKR The addition of glucosamine sulphate therapy to costs, followed by variation in the probability of current care for knee OA was estimated to lead to TKR surgery with therapy, followed by variation in additional costs and gains in health-related QoL. costs of current care. Figure 14 shows that the most With a cost per QALY estimate of approximately important parameter that contributed to variation £21,335, the magnitude of potential gains appears in QALYs was the QoL change arising from the moderate in relation to the additional costs. addition of glucosamine to current care. Quality Deterministic sensitivity analysis suggested that the of life prior to TKR surgery, and the probability cost-effectiveness of glucosamine sulphate therapy of progression to TKR surgery, make smaller, was particularly dependent on the magnitude of although similar, contributions in comparison. the QoL gains produced by therapy. Probabilistic Figure 15 shows the partial EVPI for parameters. sensitivity analysis showed that these estimates of This reveals quite clearly that the most important cost-effectiveness were imprecise, suggesting some parameter is the QoL gain associated with therapy. degree of decision uncertainty. This is followed by the TKR probability with therapy. Current care costs, the quality of life Our estimate of cost-effectiveness was higher associated with TKR and TKR costs have much in comparison with other available studies; the smaller values in comparison. NCCCC1 produced two cost per QALY estimates

TKR surgery costs

Current care costs Parameter

TKR probability

0 10 20 30 40 50 60 70 Proportion of sum squares (%)

FIGURE 13 Individual parameter contribution to variation in costs. TKR, total knee replacement.

TKR QoL

QoL change Parameter

TKR probability

0 20 40 60 80 100 Proportion of sum squares (%)

FIGURE 14 Individual parameter contribution to variation in quality-adjusted life-years. QoL, quality of life; TKR, total knee replacement. 68 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

14,000

12,000

10,000

8000

6000

4000 Partial EVPI (£million) Partial 2000

0 TKR QoL TKR QoL Current TKR surgery probability change care costs costs Parameter

FIGURE 15 Expected value of perfect information (EVPI) for parameters. QoL, quality of life; TKR, total knee replacement. for glucosamine sulphate. Using only 3-year data using a generic preference-based QoL instrument, from Pavelká and colleagues,77 the mean cost such as the HUI3, SF-6D or EQ-5D. per QALY was £10,880 (range £3581–£21,219). Using data only from Reginster and colleagues,76 It is important to emphasise that the effectiveness the mean cost per QALY was £2427 (range data within economic model are built mainly £799–£6519). Without further information on the around data from one study that pooled data from sources of data or assumptions used, it is difficult to two trials,51 while the QoL gain with therapy is explain the differences. One possible explanation estimated from one trial.58 Further evidence is also is the difference in the timeframe adopted, required on the relationship between QoL and although this seems unlikely, as the longer follow- costs. Owing to lack of data, other than through up period adopted in the model presented here progression to TKR surgery, it was not possible allows the incorporation of cost savings and to build into the model estimates of cost changes QoL gains associated with the avoidance of knee that arise from changes in QoL. This may be replacement surgery, which would lead to lower, an important limitation of the current model. rather than higher, cost per QALY estimates. A We would anticipate that incorporation of these more plausible explanation is the use of alternative effects would have made current cost estimates estimates of QoL. As the same cost difference is less precise, as there was large variation around used to generate the two estimates of cost per the magnitude of QoL improvements, but would QALY reported by the NCCCC,1 the cost per have made little change to mean cost per QALY QALY difference is driven by differences in the estimates. This suggests that a future trial should QALY gain. Such differences may be due to the also collect resource use and cost data, to allow use of alternative ways to value health changes. For estimation of the resource impact of changes in example, the lower cost per QALY gain figures of QoL. Reginster and colleagues76 is generated from use of the VAS version of the WOMAC rather than the Estimates were not particularly sensitive to QoL Likert scale version. prior to TKR surgery, nor to costs of surgery. Further, cost-effectiveness was moderately Value of information analysis indicated that further related to the volume of TKR surgery in research to reduce decision uncertainty would be deterministic analysis. To assess the importance beneficial. It would appear that a high priority of this parameter, we were only able to apply should be given to obtaining further evidence of arbitrary 50% changes in volume in deterministic the QoL gains produced by glucosamine sulphate sensitivity analysis, resulting in small changes relative to placebo. The current uncertainty of between 1% and 2% in either direction in surrounding the magnitude of such gains may the likelihood of requiring surgery. A nationally be related to sample size and the requirement to representative cohort survey to determine the map between health status instruments to obtain current level of TKR surgery would help to ensure suitable QoL values in order to calculate QALYs. that the modelled estimates presented here are Thus, any future trial should aim to collect data generalisable to the UK health-care system. 69

DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Chapter 5 Biological plausibility

t the start of this review, we considered the of long-chain glycosaminoglycans linked to a core A possibility that the findings of the review of protein. Loss of aggrecan from the extracellular clinical effectiveness would be inconclusive – not matrix in cartilage leads to a change in the only uncertain about whether the compounds were biomechanical properties of the tissue, and in so clinically effective, but also that there might not be doing accelerates the loss of articular cartilage. sufficient basis to justify a large HTA trial. In that situation, evidence or not of biological plausibility The glycosaminoglycans consist of a hexuronic acid might help to decide whether it was worth doing and a hexosamine (amino sugar). a trial. For example, if (as has been alleged by one commentator) none of the glucosamine taken Glucosamine is the hexosamine constituent orally reaches the bloodstream, then that would of keratan sulphate, which is found in hyaline seem a good reason for not doing a trial. cartilage, alongside the glycosaminoglycans chondroitin 4-sulphate and chondroitin 6-sulphate The initial, rather simplistic presumption was that (which are therefore much larger molecules than exogenous administration of the precursors of glucosamine).63 Chondroitin is a large gel-forming cartilage would in some way either aid cartilage molecule which forms part of cartilage and confers regrowth or prevent destruction. For many years, resistance to compression. Despite being a large therefore, glucosamine and chondroitin have molecule, it is partially absorbed from the diet or been considered possible candidates as disease- supplements. modifying drugs. Improved understanding of the natural progression of OA has demonstrated Oral bioavailability that the aetiology and pathophysiology of OA is far more complicated than this. Hyaline cartilage Both glucosamine and chondroitin are taken and subchondral bone are in a constant state of orally, although the dosage advised varies among degradation and regeneration. Osteoarthritis is manufacturers. Clearly, compounds taken by mouth accepted to be caused by an imbalance of these will have no effect if they are not absorbed. There anabolic and catabolic processes. An effective has been some debate about the oral bioavailability disease-modifying drug would need to alter of both glucosamine and chondroitin, as such large this balance in favour of regeneration, by either macromolecules with charge density are not readily promoting anabolic effects or inhibiting catabolic absorbed from the gastrointestinal tract.149 effects. There are many potential therapeutic avenues by which this balance could be modified. So, before discussing the plausible mechanisms of These are summarised in Box 2, which has been action of both glucosamine and chondroitin, it is modified from that of Fioravanti and colleagues.148 first imperative to determine the pharmacokinetics and bioavailability of both products. In this chapter, we will briefly examine some of the ways in which glucosamine and chondroitin might Glucosamine affect some of these processes with reference to the Laverty and colleagues150 studied the effects of recent literature. glucosamine hydrochloride delivered by the nasogastric route on the serum and synovial glucosamine concentrations in eight female horses. Glucosamine, chondroitin The dose given was equivalent to 1500 mg for a and hyaline cartilage 75 kg man. Serum and synovial concentrations were measured before and up to 12 hours post dosing. As discussed in Chapter 1, glucosamine and Pre-dose serum and synovial concentrations chondroitin are contained within hyaline cartilage. were all below measurable levels. After dosing, The non-collagenous 50% of hyaline cartilage is synovial concentrations ranged from 0.3 µmol/l to mainly made up of proteoglycan molecules. These 0.7 µmol/l. Serum levels of glucosamine returned molecules (also called aggrecans) are composed to baseline by 6 hours, but synovial concentrations 71

BOX 2 Possible mechanisms of action for a disease-modifying drug (adapted from Fioravanti 2006148)

Anabolic effects

• Chondrocyte proliferation without cellular dedifferentiation • Proteoglycans synthesis • Collagen type 2 synthesis • Hyaluronic acid synthesis • Influence of local growth factors: transforming growth factor (TGF)-β, epidermal growth factor (EGF), fibroblast growth factor (FGF), platelet-derived growth factor (PDGF) • Effects of sulphate supplementation

Reduction of catabolic effects by:

• Inhibition of the release or the activity of chondrocitic matrix metalloproteases (MMPs) or negating the downregulation of tissue inhibitors of metalloproteinase (TIMPs) • Inhibition of the release or the activity of aggrecanases • Inhibition of cytokines such as interleukin (IL)-1β, tumour necrosis factor (TNF)-α, etc. • Inhibition of prostaglandin E2 (PGE2) • Inhibition of the release or the activity of free radicals of oxygen • Inhibition of the release or the activity of nitric oxide (NO) • Inhibition of lysosomal enzymes • Anti-apoptotic activity • Anti-inflammatory effect

Mixed effects

• Sulphate supplementation • Acting as biological response modifiers

remained elevated for up to 12 hours. Adebowale due to the hepatic first-pass effect. Biggee and and colleagues151 demonstrated the bioavailability colleagues157 gave 11 patients with known OA, who of oral glucosamine hydrochloride and chondroitin had not been taking glucosamine supplements, sulphate in beagle dogs. a single 1500 mg dose of commercial crystalline glucosamine sulphate after an overnight fast. Setnikar and colleagues152–156 have extensively Serum glucosamine levels were measured at studied the absorption, distribution, metabolism baseline and then regularly up to 8 hours post and excretion of glucosamine sulphate. Following ingestion. Baseline serum glucosamine was intravenous administration of radiolabelled undetectable in all subjects. Following ingestion glucosamine sulphate in rats, early uptake was seen glucosamine was detectable in 17 of the 18 principally in the liver and kidneys, but there was patients. After ingestion, peak levels ranged also early detection in the femoral head articular from 0 to 6.4 µmol/l (mean 3.6 µmol/l).157 Basak cartilage, peaking at 0.17 hours post dosing. and colleagues158 compared serum values after This uptake by articular cartilage was also shown supplementation with either a novel timed-release following oral dosing of radiolabelled glucosamine capsule or a standard powder-filled one, in a sulphate in both rat and dog studies. 12-patient multidose randomised crossover study using glucosamine sulphate. Serum glucosamine Using radiolabelled glucosamine sulphate, levels were measured at dosing and up to 24 Setnikar and colleagues156 studied oral hours post dosing. Glucosamine was not detected administration in healthy human volunteers, pre-dose, but the peak concentration afterwards and found gastrointestinal absorption of 90% was 543.12 ng/ml (± 151.16) and 520.98 ng/ml 72 with bioavailability of 44%. The difference was (± 152.55) for the powder-filled and timed-release DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

preparations respectively. They concluded that camera. As well as the bloodstream showing very there was a 21% bioavailability. high levels following injection, it was also noted that very high intensity was detected in the area Persiani and colleagues159 measured both plasma surrounding the knee joints.164 and synovial concentrations of glucosamine after 14 days of oral glucosamine sulphate (1500 mg) in Ebube and colleagues165 analysed the different 12 OA patients. Glucosamine levels were measured physiochemical and mechanical properties of at baseline and at 3 hours following their last dose. glucosamine hydrochloride and chondroitin After the 2 weeks, a 20.5-fold increase was found sulphate obtained from various sources. Particle in plasma, and a statistically significant 21.5-fold morphology, particle size and other physical increase in synovial concentrations was observed. properties were found to be greatly variable between sources. This will lead to different The optimum dose of oral crystalline glucosamine bioavailability profiles seen across the product sulphate was examined by Persiani and range for both supplements. colleagues.160 Twelve healthy volunteers received three consecutive daily doses of 750 mg, 1500 mg There is, therefore, a significant body of evidence or 3000 mg in an open randomised, crossover supporting the oral bioavailability of both fashion. Plasma levels increased in the dose range glucosamine and chondroitin products in human in 750–1500 mg, but became non-linear at 3000 mg. vivo studies. The blood levels remained elevated for 24 hours after each dose. Anabolic state Chondroitin The normal synovial concentration of chondroitin Many authors have studied the role of glucosamine sulphate has been studied in a healthy population on both chondrocytes themselves and also the group. Nakayama and colleagues161 found that the chondroprogenitor cells, such as the human normal level of synovial chondroitin, in 82 healthy mesenchymal stem cells (hMSCs). Dodge and volunteers aged between 20 and 79 years of age, Jimenez166 studied the effects of glucosamine fell with age. The mean synovial concentration sulphate on freshly isolated chondrocytes obtained was 103.1 nmol/ml for chondroitin 6-sulphate from patients with knee OA. They observed that and 18.4 nmol/ml for chondroitin 4-sulphate. The 40% of chondrocytes from OA patients failed concentration of chondroitin 6-sulphate in the to respond to glucosamine sulphate. Of the 70–79 years age group was approximately half remaining 60%, however, it was observed that the that of the 20–29 years age group. The ratio of presence of glucosamine sulphate upregulated chondroitin 6-sulphate to chondroitin 4-sulphate messenger ribonucleic acid (mRNA) levels of also steadily decreased with advancing age. the aggrecan gene in a dose-dependent manner, Interestingly, it was also noted that females had leading to a corresponding increase in observed lower concentrations of synovial chondroitin across levels of aggrecan. They also observed that the all age groups. presence of glucosamine sulphate decreased matrix metalloproteinase (MMP-3) protein Volpi162 gave 4 g oral chondroitin sulphate levels as well as enzymatic activity, and increased (bovine origin) to 20 healthy volunteers and glycosaminoglycan content.166 measured plasma levels of chondroitin at baseline and frequently until 48 hours post ingestion. Noyszewski and colleagues167 have also shown that After ingestion, chondroitin sulphate increased the addition of radiolabelled glucosamine sulphate significantly, peaking at 2 hours, with plasma to cultured chondrocytes facilitates the production levels increased by 200% at 2–4 hours. However, of proteoglycan components with direct uptake of a second, similar, study by Volpi,163 using shark the added radiolabelled glucosamine sulphate. chondroitin sulphate, found similar peak levels but at 8.7 hours. The author proposed that the In another in vitro study, Derfoul and colleagues168 difference may be due to the different molecular found that addition of 100 µM glucosamine weights and charge densities of the two molecules hydrochloride led to enhanced expression of studied. collagen II and aggregan, and an increased content of sulphated glycosaminoglycan. It was Technetium-labelled chondroitin sulphate was also observed that the addition of glucosamine injected in healthy human volunteers and its whole partially blocked the effects of interleukin (IL)- body distribution was monitored with a gamma 1β on downregulating collagen II and aggrecan 73

expression. Following addition of glucosamine, the Further work by Kim and colleagues172 explored expression of the matrix-degrading enzyme, MMP- the anabolic effects of physiological doses of 13, was also downregulated in both chondrocyte glucosamine sulphate on human osteoblast-like and hMSC cultures.168 MG-63 cells. The addition of glucosamine sulphate increased alkaline phosphatase (ALP) activity, The effects of glucosamine have also been collagen synthesis, osteocalcin secretion and attributed to the upregulation of chondrocyte mineralization of the cultured osteoblastic cells proliferation, matrix synthesis and gene expression. in vitro in a dose-dependent manner from 10– Varghese and colleagues169 found that the addition 1000 µg/ml, generally increasing anabolic activity. of glucosamine hydrochloride upregulated matrix production in 2D and 3D bovine chondrocyte Not all studies support an increase in anabolic layers. Up to a dose of 2 mM, this effect was seen activity with glucosamine sulphate, however. principally on aggrecan and type II collagen. This Using pre-cultured human osteoarthritic explants, was not seen at higher doses. It was also noted Uitterlinden and colleagues173 studied the effects that transforming growth factor (TGF)-β1 was of 0.5 mM or 5 mM addition of either glucosamine upregulated in a dose-dependent manner in the hydrochloride or glucosamine sulphate. A presence of glucosamine hydrochloride.169 5 mM concentration of both glucosamine subtypes was seen to significantly downregulate Nishomoto and colleagues170 observed aggrecan and collagen type II gene expression upregulation of the mRNA of collagen II and an suggesting inhibited anabolic activity. At the same increase in aggrecan mRNA expression in porcine doses, however, there was also significant anti- chondrocytes after the addition of chondroitin catabolic activity with downregulation of MMP- sulphate. However, the effect was not seen with all 3, and aggrecanase-1. The addition of 0.5 mM subtypes of chondroitin sulphate.170 glucosamine was not seen to alter either anabolic or catabolic activity at a significant level; however, Mroz and colleagues171 studied whether similar trends were noted when compared with the supplementation of glucosamine to cultured 5 mM group.173 human chondrocytes leads to increased production of chondroitin sulphate. Figure 16 shows possible pathways. Anticatabolic effects

They concluded that the addition of exogenous Piperno and colleagues174 studied the effects of H-glucosamine did not stimulate an increased glucosamine sulphate on proteoglycan synthesis, production of chondroitin sulphate.171 caseinase, collagenase, and phospholipase

Glucose Glucosamine N-acetylglucosamine

Glutamine Glucose Fructose Glucosamine N-acetylglucosamine 6-phosphate 6-phosphate 6-phosphate 6-phosphate

NH3

UDP-N-acetyl- UDP-N-acetyl- N-acetylglucosamine Chondroitin galactosamine glucosamine 1-phosphate

FIGURE 16 Biosynthetic scheme for incorporation of galactosamine into chondroitin from glucose or glucosamine precursors.171 (Reproduced from: Mroz PJ, Silbert JE. Use of 3H-glucosamine and 35S-sulfate with cultured human chondrocytes to determine the effect of glucosamine concentration on formation of chondroitin sulfate. Arthritis Rheum 2004;50(11):3574–9. Copyright 2004 American College of Rheumatology. Reprinted with permission of John Wiley & Sons, Inc.)

74 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

A2 (PLA2), and on production of nitric oxide glucosamine sulphate. However, this study failed to (NO), cyclic adenosine monophosphate (cAMP), demonstrate any significant difference between the and protein kinase C, in human osteoarthritic two groups when measuring glucosamine sulphate chondrocyte cell cultures. Glucosamine sulphate degradation products in serum and urine.177 was found to inhibit PLA2 and collagenase activity in a dose-dependent manner. A dose- dependent increase in protein synthesis (minimum Aggrecanase activity concentration 50 µM) and protein kinase C production was seen. The aggrecan degradation simulated by IL- 1, tumour necrosis factor (TNF)-α, or retinoic Monfort and colleagues175 studied the in vitro acid is thought largely to be due to the action of effects of chondroitin sulphate on MMP-3 aggrecanase. Ilic and colleagues178 studied the (stromelysin-1) synthesis in human osteoarthritic effects of long-term glucosamine on aggrecan chondrocytes. MMP-3 is a cartilage proteolytic degradation in cultures of bovine cartilage. enzyme which induces cartilage breakdown They concluded that a 5 mM concentration of and also acts as an inflammatory mediator. glucosamine indirectly inhibited degradation of Chondrocytes were obtained from human aggrecan under simulated catabolic activity. When osteoarthritic cartilage from patients of both sexes, glucosamine was then removed from the cultures, over the age of 40 years, who were undergoing aggrecanase activity increased. Uitterlinden arthroplasty. The study noted a marked inhibitory and colleagues173 also demonstrated that the effect of chondroitin sulphate on MMP-3 synthesis addition of 5 mM either glucosamine sulphate or at all concentrations tested. No significant glucosamine hydrochloride to cultured human differences were found between the various osteoarthritic explants demonstrated a significant concentrations.175 downregulation of aggrecanase-1. Suppression of aggrecanase-2 was also detected in the glucosamine The majority of the studies analysing the in vitro sulphate group. effects of glucosamine and chondroitin sulphate have focused purely on chondrocyte cultures. The Furthermore, work by Sandy and colleagues179,180 onset of OA is increasingly understood to involve concluded that aggrecanase activity is significantly changes within the subchondral bone, including decreased in the presence of glucosamine subchondral bone resorption. The balance of and that this may be via suppression of bone remodelling factors osteoprotegerin (OPG), glycosylphosphatidylinositol-linked proteins. and receptor activator of nuclear factor κB ligand (RANKL), are important in this process. Inhibition of oxidative stress Tat and colleagues176 studied the effects of both glucosamine sulphate and chondroitin sulphate As OA is generally (although not exclusively) and found that the chondroitin sulphate reduced perceived as an ageing phenomenon, it is production of OPG and RANKL, whereas postulated that degeneration of articular glucosamine sulphate significantly reduced cartilage and collagen may be a result of resorptive activity. Both chondroitin sulphate and oxidative stress. Tiku and colleagues181 explored glucosamine sulphate may, therefore, play an the hypothesis that glucosamine sulphate important role in reducing the catabolic effects of and glucosamine hydrochloride may protect subchondral bone in OA. collagen from fragmentation secondary to oxidation. Rabbit chondrocytes were cultured Cibere and colleagues177 studied the effects and activated with calcium ionophore which of glucosamine sulphate on type II collagen induces oxidative metabolism, with the addition breakdown in a randomised double-blind placebo- of either glucosamine sulphate or glucosamine controlled study. One hundred and thirty-seven hydrochloride. The study confirmed that the patients with proven knee OA who had gained addition of 25 mM of either glucosamine salts at least moderate benefit from glucosamine significantly decreased collagen degradation therapy were randomised to either placebo compared with controls. The author concluded (discontinuation) or glucosamine (continuation) that glucosamine was likely to inhibit advanced treatment over a 24-week period. The expectation lipoxidation reactions and therefore prevent was that the placebo group would show an increase oxidation and degradation of collagen matrix. in type II collagen markers after discontinuation of

Anti-inflammatory effects Glucosamine hydrochloride alone significantly Several authors have postulated that both reduced iNOS expression at 6 and 24 hours, but glucosamine and chondroitin may function as not at 48 hours, whereas in contrast, chondroitin anti-inflammatory mediators in relation to OA. sulphate reduced iNOS expression significantly The anti-inflammatory effects of glucosamine at 48 hours. Synthesis of NO was reduced by both sulphate and chondroitin sulphate on stimulated chondroitin sulphate alone and the combination chondrocyte cell cultures with IL-1β has been of chondroitin sulphate and glucosamine explored. Jomphe and colleagues182 found that hydrochloride at 24 hours. Furthermore Chan the addition of physiological levels of chondroitin and colleagues187 showed the downregulation of

sulphate reduced nuclear factor κ-light-chain- inflammatory mediators PGE2, NO, iNOS, MMP- enhancer of activated B cells (NF-κB) nuclear 3 and aggrecanase-2 in IL-β1-stimulated bovine translocation, an important transcription cartilage explants in the presence of glucosamine factor involved in the initiation of various pro- hydrochloride and chondroitin sulphate.187 The inflammatory genes such as COX-2, MMP-3 and upregulation of COX-2 with the addition of IL-1β TNF-α. They also concluded that chondroitin was abrogated with the addition of chondroitin sulphate diminishes IL-1β-induced increase in sulphate and glucosamine hydrochloride. They other proflammatory mediators (p38 mitogen also found that upregulation of tissue inhibitor of activated protein kinase and extracellular signal- metalloproteinase-3 transcripts was seen with the regulated kinase ½). Largo and colleagues183 also addition of both glucosamine hydrochloride and found that physiological doses of glucosamine chondroitin sulphate. sulphate inhibit NF-κB activation in IL-1β-activated 188 human chondrocytes as well as PGE2 synthesis. Huser and Davies demonstrated that a lipophilic Furthermore, they found that glucosamine sulphate derivative of glucosamine, Glu5, is able to prevent inhibited the expression and synthesis of COX-2, impact-induced chondrocyte death, possibly by although this result remains controversial within reducing mitochondrial depolarisation following the literature.152 a single impact load in vitro. Jomphe and colleagues182 studied the effects of the addition Kim and colleagues172 found that physiological of chondroitin sulphate to rabbit chondrocytes doses of glucosamine sulphate exhibited anti- stimulated with sodium nitroprusside, a compound inflammatory effects on the production of TNF-α, known to induce apoptosis, and found that the

IL-1β and PGE2 in macrophage cell lines in vitro. addition of chondroitin sulphate protected the chondrocytes from apoptosis to the same level Inhibition of nitric oxide as the group that was not exposed to sodium nitroprusside. Caraglia and colleagues189 also and prostaglandin E2 and anti-apoptotic activity showed that osteoarthritic mice models who were treated with chondroitin sulphate prior to Nitric oxide (NO) is thought to play a role in being sacrificed, showed a significant reduction the pathogenesis of OA. Production of NO is in chondrocyte apoptosis. Interestingly, this increased in chondrocytes from osteoarthritic effect was increased when they were treated with bone in comparison with that of normal bone.184 both chondroitin sulphate and mud therapy with Nitric oxide production is also thought to be an sulphur mineral water. important factor in the regulation of chondrocyte apoptosis, which in turn is thought to play an Biological response modifiers important role in the aetiology of OA.185 It has also been suggested that both glucosamine Chan and colleagues186 studied the effects and chondroitin may act as biological response of physiologically relevant concentrations of modifiers (BRMs) for chondrocytes under glucosamine hydrochloride and chondroitin simulated conditions of stress. A BRM is an agent sulphate on the regulation of inflammatory which promotes the defence of the host against 190 mediators NO and PGE2. They found that multiple stresses. Lippiello studied the effects glucosamine hydrochloride, chondroitin sulphate of the addition of glucosamine and chondroitin or the combination of both, at clinically relevant to bovine cartilage explants which were subject doses, reduced mRNA expression of inducible NO to various stressors, including mechanical stress, synthase (iNOS) at 24 and 48 hours post culture. heat, enzyme-induced matrix depletion and The combination of both reduced iNOS expression cytokine stress. The study concluded that both to control levels at 24 hours and 48 hours. glucosamine hydrochloride and chondroitin 76 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

sulphate modulate the protective metabolic the glucosamine sulphate. In a separate group, response in favour of repair and regeneration at 15 patients with known OA, who were to undergo doses in keeping with the bioavailability of these diagnostic needle aspiration in the outpatient agents. Interestingly, they also concluded that setting, consented to the measurement of cartilage explants from older animals were more both synovial and serum sulphate. This was to likely to undergo metabolic change with stress, investigate the relationship between serum and but also have a greater response to the addition synovial sulphate levels which were found to be of glucosamine hydrochloride and chondroitin virtually identical. sulphate. Nerucci and colleagues191 also found that human chondrocyte cultures exposed to a Of the seven healthy patients, the mean baseline simulated pressurisation cycle with the addition serum sulphate level was 331 µmol/l which of IL-1β showed a significant decrease in increased to 375 µmol/l 3 hours following ingestion proteoglycan concentration compared with controls of glucosamine sulphate. In the glucosamine (without IL-1β) at day 10 of culture. This decrease sulphate and paracetamol group, the baseline was significantly reduced with the presence of level decreased from 325 µmol/l to 290 µmol/l chondroitin sulphate at 10 µg/ml and 100 µg/ml. at 3 hours. The study therefore concluded that glucosamine sulphate increases serum inorganic sulphate concentration in normal individuals and Sulphate deficiency that this increase is reversed by coingestion of paracetamol.194 A further plausible hypothesis is based on the role of the sulphated amino sugars found in Cordoba and Nimni195 investigated the relationship glucosamine hydrochloride and chondroitin between dietary levels of protein and urinary sulphate. Cartilage contains tissue-specific excretion of sulphate. This study assessed glycosaminoglycans that require a source of whether consumption of chondroitin sulphate inorganic sulphate for its synthesis. Although the and glucosamine sulphate increased urinary recommended dietary allowance has previously excretion of sulphate. Healthy human volunteers been derived in young healthy individuals, this were instructed to record their dietary intake. does not take into account the increased turnover Patients were separated into five groups depending of glycosaminoglycans during the osteoarthritic on their routine dietary protein intake. In the process. Furthermore, many patients with OA take two lowest groups for dietary protein intake, paracetamol as a simple analgesic to manage their consuming 15 mM or less of sulphur over 24 hours, symptoms. Paracetamol is largely metabolised supplementation with chondroitin sulphate did not by sulphation and indeed 40% of paracetamol is lead to the expected increase in urinary secretion, excreted in urine conjugated with sulphate, which suggesting retention of sulphate, presumably due may contribute to sulphate deficiency. It has been to dietary insufficiency. In the groups consuming previously shown that sulphate depletion inhibits 25 mM or more of sulphur, supplementation glycosaminoglycans synthesis in vitro in rat and of either glucosamine sulphate or chondroitin human articular cartilage. Indeed, human cartilage sulphate led to an increase in urinary excretion is thought to be particularly sensitive to sulphate of sulphate over the measured daily dietary deficiency.192,193 consumption. The effect of glucosamine sulphate for patients with low dietary sulphur was not Hoffer and colleagues194 have also investigated the described in the paper. It is plausible, therefore, role of sulphate as the proposed mechanism of that chondroitin sulphate and glucosamine action of glucosamine sulphate. They first studied sulphate supplementation may be most effective whether oral glucosamine sulphate increases serum in patient populations with low dietary sulphur, sulphate concentrations. They then investigated such as vegetarians or those who consume little red whether this increase was also found within synovial meat. fluid. Finally, the group studied whether the use of paracetamol affected the serum sulphate response to glucosamine sulphate. Seven healthy adults Summary had venous blood measured for sulphate levels before, and then 3 hours following, ingestion of While much of the evidence presented above 1 g glucosamine sulphate (1.65 mmol sulphate). is derived from in vitro experiments which The study was then repeated in a similar fashion may not reflect the osteoarthritic human knee, with the addition of 1 g paracetamol along with there are a number of plausible mechanisms by 77

which glucosamine, chondroitin or the sulphate • Sulphate effect: there was evidence to support component of the treatments may act: the intriguing hypothesis that it may be the sulphate rather than the glucosamine which • Bioavailability: there was significant evidence has the effect. that oral supplements of both chondroitin and glucosamine led to an increase in both serum The exact mechanism(s) of either drug is as yet and synovial fluid concentrations. unknown. Furthermore, many but not all of the • Anabolic effect: in vitro studies supported some in vitro studies utilised concentrations in excess of anabolic function for glucosamine. those found from the bioavailability studies, and • Catabolic effect: there was a stronger evidence therefore may not be clinically relevant. base for inhibition of catabolism.

78 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Chapter 6 Discussion and recommendations

Summary of findings –– As there was currently no way to determine what, if any, significance joint space width In this review we set out to address a series of has on QoL, health utility or long-term questions with regard to the clinical effectiveness outcomes, it was not possible to draw any and cost-effectiveness of glucosamine and conclusions about the clinical effectiveness chondroitin. of chondroitin.

• Does glucosamine prevent or slow progression of OA • Does the combination of both offer any greater effect? of the knee? –– While a few studies did use combination –– There was evidence that the Rotta therapy, the comparison was placebo powdered oral preparation of glucosamine not single therapy. It was not possible, sulphate reduced joint space loss and therefore, to address whether there was may even improve joint space width after a relatively greater gain with such a 3 years of therapy when compared with combination. placebo. The clinical significance of joint space narrowing and changes in width over • If they are useful, what dosages should be used? time were less certain. Absolute joint space –– With only two trials for the one preparation width is a poor predictor of outcomes for which we had concluded that there was in terms of either QoL or need for knee evidence of effectiveness, and with both arthroplasty. There is, however, some trials using identical dosing regimens, it evidence that changes in joint space width was not possible to address the question do relate to clinical outcomes such as the of what was the optimal dose. It is worth need for future knee arthroplasty. noting that both trials used the Rotta –– There was evidence that glucosamine had glucosamine product and were sponsored a beneficial impact on long-term outcome by the pharmaceutical company. of knee arthroplasty (6.3% in the treated group versus 14.5%); an effect seen despite • What are the costs, in terms of both cost of the small numbers and the opportunity for products and any side effects? substantial treatment group switching –– The reported adverse event profile for during follow-up which would, if anything, glucosamine sulphate was unremarkable. lessen the observed difference between the The only potential issue identified two study groups. from data beyond clinical trials was an –– A small statistically significant interaction with warfarin. The one death improvement was observed in QoL associated with acute liver failure, however, measured using WOMAC in the treated highlighted the need for long-term group versus the placebo group. monitoring of the use of glucosamine and that, to date, experience from formal • Does chondroitin prevent or slow progression of OA study of the treatment draws on a relatively of the knee? small patient pool. There is no licensed –– We found evidence to support the preparation of glucosamine sulphate conclusion that chondroitin had a small available in the UK. Over-the-counter effect on structural changes, but this time preparations vary substantially in cost there was no corroborating evidence of a and content. For modelling, the cost of translation of this finding into long-term the licensed glucosamine hydrochloride clinically important outcomes. The impact product was taken as a marker for the likely on QoL appeared to be influenced by the price to the NHS of a sulphate licensed tool used and was inconsistent between product should one become available (BNF studies. price of £18.40).

• Are there any savings to the NHS from their use, for While a specific mechanism of action example by reduced consumption of prescribed non- has not been established, evidence for steroidal anti-inflammatory drugs, or from avoidance a number of potential actions on the of knee arthroplasty? joint has been observed for glucosamine, –– There was little evidence from the trials chondroitin and sulphate (in vitro, in that glucosamine sulphate reduced the animals and in healthy volunteers). need for oral analgesia, but as noted above, the long-term follow-up data did • If, as suggested by the commissioning brief, the support a reduction in the need for knee evidence suggests that there is no gain in short-term arthroplasty. symptom relief, but only in long-term preservation of cartilage (which does imply longer-term reductions • Is there evidence of cost-effectiveness from well in symptoms, advanced OA and perhaps the need for constructed economic evaluations? knee arthroplasty), what should the comparator in –– We did not identify any economic trials be? evaluations focusing on the long-term –– Because OA is now recognised to be a cost-effectiveness of glucosamine sulphate. condition that is not necessarily relentlessly Our own analysis estimated the cost per progressive and, in fact, the symptoms QALY to be £21,335. Gains in cost to experienced by individuals can range the health service were accompanied by widely, a comparator group is essential for modest health-related gains. However, the interpretation of trial data. NCCCC sensitivity analysis suggested that the cost- identified, in their review of treatments for effectiveness of glucosamine sulphate was OA of the knee, that none of the existing largely dependent on the magnitude of therapies have been shown to alter the QoL gains from therapy. The estimates clinical course of the condition. The of cost-effectiveness were imprecise comparator, therefore, has to be against and suggest some degree of decision placebo. uncertainty. • If the gain is in long-term outcomes, are there –– From our review we conclude that the short-term indicators of benefit which could be used, evidence is inconclusive, but suggestive rather than waiting 20 years for advanced OA to enough of benefit to justify further become manifest? Joint space narrowing on plain research for glucosamine sulphate. Our film has been used as one outcome. Could more conclusions are in keeping with those complex forms of imaging, such as MRI, positron of the recent NCCCC guidelines1 that emission tomography, computerised tomography, included an assessment of glucosamine and other radionuclide techniques or ultrasound, provide chondroitin. While they did not focus on more detailed assessment of the quality and thickness long-term outcomes and used a different of remaining cartilage, and hence give earlier methodology for determining utility confirmation of benefit? scores in the economics analysis, they did –– The two most significant challenges around report that a small effect was observed for imaging the knee using X-ray have been glucosamine sulphate on QoL measures the difficulty in relating changes in the and joint space. Like us, they concluded image to long-term outcomes and the from their economic analysis that evidence need to use quantitative measures to assess of cost-effectiveness was not conclusively change. established. –– There have been growing numbers of reports in the literature about the use of Similarly authors of the systematic reviews included other imaging techniques in assessing in our clinical effectiveness review reported joint space changes in OA. While there is evidence of some clinical effect associated, in evidence of a change in joint space being particular, with preparations of glucosamine associated with long-term outcomes such sulphate, and to some extent, chondroitin. as knee arthroplasty, regardless of the technique used to image the joint, the • Are the alleged benefits of glucosamine on knee difficulties remain about the reliability cartilage biologically plausible? of such measurements. MRI and other –– Bioavailability of the glucosamine in the techniques afford the additional benefit synovial fluid has been demonstrated. of providing more information about 80 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

structural changes with in the joint itself. outcomes tended to be longer term; therefore, the An evidence-driven consensus conference conclusions of previous reviews on this outcome in 2006 lead by OMERACT and OARSI are of relevance. Poolsup and colleagues73 focused attempted to summarise opinion about on long-term outcomes and identified two of the the most potentially useful features three trials we included. Their conclusions were derivable from MRI. They concluded that similar to ours, supporting evidence of clinical articular cartilage loss, osteophytes, bone effectiveness for both symptoms and structural marrow abnormalities, synovitis, meniscal outcomes. Towheed and colleagues62 included two abnormalities and synovial effusions held of the three trials we included in their assessment the greatest promise. Despite enthusiasm of structural outcomes, plus an additional shorter- for these new measures, the need for duration study that we excluded. They were long-term cohorts and follow-up data was cautious in their interpretation of the significance recognised as a priority to determine the of the surrogate end point of joint space change clinical significance. and at that time did not have access to the longer follow-up information about knee arthroplasty. • Which variables or assumptions (e.g. costs, QoL, mortality, extrapolation of effectiveness over time) Surrogate end points in trials are widely used are most important in generating uncertainty in the for many conditions and often lead to concerns cost-effectiveness estimates, and which would provide regarding their translation into long-term effects. the greatest return in terms of reducing uncertainty As a surrogate end point, joint space narrowing at reasonable cost? These variables and assumptions has been approved as an acceptable trial outcome will be identified using value of information analysis, measure for OA by OMERACT and regulatory and will form the main basis in recommendations for bodies41 but its insensitivity in diagnosis, measuring further research. disease activity and assessing progression has been –– Value of information analysis indicated noted. that further research to reduce uncertainty would be beneficial. Greatest uncertainty To address the challenges of surrogate end points around costs is generated by the in technology assessment, the HTA are currently uncertainty around the probability and cost establishing a working group to explore the of knee arthroplasty. The most important challenges of surrogate markers in clinical trials driver in QALY variability is uncertainty with a view to preparing guidance on their use around the QoL gain associated with in health technology assessment. Critical to our treatment. Overall, variability in cost- interpretation of the potential effectiveness of effectiveness is most heavily determined by glucosamine sulphate has been the evidence of a uncertainty regarding the magnitude and translation from the surrogate end point of joint duration of QoL gains following treatment. space narrowing to the clinical end point of knee arthroplasty. For chondroitin, biologically plausible mechanisms have been reported, but the clinical effectiveness Where evidence of effectiveness was observed evidence was less convincing with greater (for glucosamine sulphate) the conclusions are heterogeneity in conclusions. Trials of chondroitin, based on only three trials. While two of the three however, have tended to be of poorer quality, trials were well designed, they followed similar so uncertainties about the clinical effects of treatment protocols and therefore did not allow any chondroitin remain. assessment of the relative merits of different doses, products or regimens.

Challenges for clinical Other challenges included interpretation of QoL effectiveness measures. A number of different tools were used to assess pain, function and disability. Authors rarely Despite numerous reviews of these two treatments, stated a priori expectations of what might represent most conclusions have previously been driven by a clinically important change. There was little short-term findings, particularly for measures of information about the long-term natural history of symptoms and function. The failure to undertake such measures or how such measures relate to other subgroup analysis by duration of follow-up made outcomes, including knee arthroplasty or other it impossible to determine the effectiveness for service utilisation. long-term outcomes. Studies including structural 81

Finally, none of the studies were undertaken in the aspect of decision-making for surgeons, and UK and, given the importance of access to services treatments that delay the need for arthroplasty and complexity of decision-making around the may also reduce the need for subsequent timing of surgical intervention, generalisability to revision surgery. As yet, surrogate marks the UK health-care setting is difficult. continue to be proposed, but in the absence of long-term follow-up to surgery, the implications of change in surrogate end points remain Challenges for modelling uncertain. • Knee arthroplasty – a nationally representative Three main challenges for modelling are worth cohort study is required to understand what noting. First, the lack of robust epidemiological proportion of patients with OA (diagnosed in and service utilisation data for the UK meant primary care and referred to secondary care) that there was substantial uncertainty about the require knee arthroplasty. generalisability of trial data and the cost of ‘current care’. Second, there was no direct measure of a Trials of interventions should focus on glucosamine generic preference-based QoL measure (such as sulphate and the Rotta product is the only product HUI3, SF-6D, EQ-5D) and as a result mapping to date that has demonstrated effectiveness. While formulae had to be used to map WOMAC to a uncertainty about other preparations remains, utility measure. Finally, the economic model relied there was insufficient evidence of effectiveness and on the results of two studies for effect of treatment it was not possible to develop an economic case for on WOMAC and knee arthroplasty. The number further study at this time. Any trial should: of participants was relatively small, and only one treatment protocol and glucosamine preparation 1. include collection of information about co- was used. There remain, therefore, a number of prescribing, the use of other interventions and decision uncertainties that could not be addressed adverse events in the economic modelling. 2. recruit obese and overweight participants and people across stages of OA severity 3. use the opportunity to gather a number of Research needs measures of joint structure and damage 4. be of at least 3 years’ follow-up with a –– Based on sensitivity analysis and value mechanism to follow the cohort long term (e.g. of information analysis, three research through record linkage to hospital data). priorities were identified: In preparing the background and undertaking • Quality of life – further clarification of the the clinical effectiveness review and economics potential QoL gains from treatment with modelling, it was apparent that there was a lack of glucosamine sulphate versus placebo over information about the epidemiology, management long-term treatment would reduce uncertainty and natural history of OA of the knee in the UK. in the cost-effectiveness estimates. There was Greater understanding of the incidence and consistent evidence of some effect on various prevalence in the community, as presenting to composite scores, but no study utilised a primary care and as referred to specialist services, generic preference-based QoL measure (such would benefit any future technology assessment as HUI3, SF-6D, EQ-5D) that can readily in this clinical area. Alongside this, information be used to estimate utility. Any future trial about current care should be obtained. Some should also inform our understanding of the of this information could be obtained using relationship between QoL and costs collecting routine primary care databases, but would require resource use and cost data to allow estimation supplementation to estimate the impact in the of the resource impact of any changes in QoL. community and in secondary care. • Structural outcomes – further long-term trial data are required to clarify the impact on the A number of studies noted that expectations, ultimate need for knee arthroplasty, including coping and tolerance of symptoms varied between the ability to delay the need for surgery. As individuals and had substantial impacts on joint replacements have a finite lifespan, their utilisation of services. Further research to optimising the timing of surgery is a critical understand the impact of such patient factors

82 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

would help in the planning of services and target Conclusion need (versus demand). There was evidence that glucosamine sulphate The biological mechanism of glucosamine shows some clinical effectiveness in the treatment sulphate and chondroitin remains uncertain. In of OA of the knee. Cost-effectiveness was not particular, the proposal that the active substance conclusively demonstrated, but substantial may be sulphate should be explored further. Other uncertainty was observed for some key sulphate-rich supplements might provide a useful determinants in the model. There was evidence comparator, e.g. glutathione and l-methionine. from biological studies to support the potential Paracetamol, metabolised by sulphation may clinical impact of glucosamine sulphate. For potentially lead to sulphate deficiency, and other preparations, the evidence base was less therefore a better understanding of this potential consistent (chondroitin) or absent (glucosamine mechanism has important clinical implications. hydrochloride). Most work has focused on the effect of compounds on the chondrocyte and cartilage. Further work Further research to reduce uncertainty and confirm is needed to explore the impact on subchondral the impact on surgical outcomes in a UK clinical bone, which is increasingly seen as the primary setting would be beneficial. disease initiator.

DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Acknowledgements

e would like to acknowledge the help of Mr Rob Henderson (Consultant in Public Health WDavid Pflegger, Consultant in Public Health Medicine) drafted sections of the background. Pharmacy, NHS Grampian and Robert Gordon Campbell MacEachern (Specialist Registrar) University; and Professor David Reid, Professor drafted sections of the background and the chapter of Rheumatology, Head of Division of Applied on biological plausibility. Paul McNamee (Senior Medicine, University of Aberdeen. Research Fellow) and Zahidul Quayyum (Research Fellow) undertook the cost-effectiveness analysis Contribution of authors and drafted this section. Pam Royle (Senior Research Fellow) developed the search strategy and Corri Black (Senior Clinical Lecturer) led the undertook all searching. Sian Thomas (Systematic review and drafted the final report. Christine Reviewer) data extracted and contributed to the Clar (Systematic Reviewer) led the review of draft clinical effectiveness chapter. All authors read clinical effectiveness and drafted this chapter. and commented on the final draft.

DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

References

1. National Collaborating Centre for Chronic 13. D’Ambrosia RD. Epidemiology of osteoarthritis. Conditions. Osteoarthritis: national clinical guideline Orthopedics 2005;28:s201–s205. for care and management in adults. London: Royal College of Physicians; 2008. 14. Fautrel B, Hilliquin P, Rozenberg S, Allaert FA, Coste P, Leclerc A, et al. Impact of osteoarthritis: 2. Agency for Healthcare Research and Quality. results of a nationwide survey of 10,000 patients Treatment of primary and secondary osteoarthritis of the consulting for OA. Joint Bone Spine 2005;72:235–40. knee. URL: www.ahrq.gov/downloads/pub/evidence/ pdf/oaknee/oaknee.pdf (accessed 15 September 15. Corti MC, Rigon C. Epidemiology of osteoarthritis: 2008). prevalence, risk factors and functional impact. Aging Clin Experiment Res 2003;15:359–63. 3. Mankin HJ. Clinical features of osteoarthritis. In Kelley WN, Harris ED, Ruddy S, Sledge CB, editors. 16. Bedson J, Mottram S, Thomas E, Peat G. Knee Textbook of rheumatology. Philadelphia: W.B. Saunders pain and osteoarthritis in the general population: Company; 1981. what influences patients to consult? Fam Pract 2007;24:443–53. 4. Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K, et al. Development of criteria for 17. White A. Osteoarthritis of the knee – an the classification and reporting of osteoarthritis. introduction. Acupunct Med 2006;24:S1–S6. Classification of osteoarthritis of the knee. Diagnostic and Therapeutic Criteria Committee 18. Williams FMK, Spector TD. Osteoarthritis. Medicine of the American Rheumatism Association. Arthritis 2006;34:364–8. Rheum 1986;29:1039–49. 19. Brooks PM. Impact of osteoarthritis on individuals 5. Nuki G. Osteoarthritis: a problem of joint failure. Z and society: How much disability? Social Rheumatol 1999;58:142–7. consequences and health economic implications. Curr Opin Rheumatol 2002;14:573–7. 6. Doherty M. Risk factors for progression of knee osteoarthritis. Lancet 2001;358:775–6. 20. Dieppe PA, Cushnaghan J, Shepstone L. The Bristol ‘OA500’ study: progression of osteoarthritis 7. Arden N, Nevitt MC. Osteoarthritis: epidemiology. (OA) over 3 years and the relationship between Best Pract Res Clin Rheumatol 2006;20:3–25. clinical and radiographic changes at the knee joint. Osteoarthritis Cartilage 1997;5:87–97. 8. Grotle M, Hagen KB, Natvig B, Dahl FA, Kvien TK. Prevalence and burden of osteoarthritis: results 21. Buckwalter JA. The impact of osteoarthritis: from a population survey in Norway. J Rheumatol implications for research. Clin Orthop Relat Res 2004; 2008;35:677–84. 427(Suppl.)S6–S15.

9. Sowers M. Epidemiology of risk factors for 22. Johnson SR, Archibald A, Davis AM, Badley osteoarthritis: systemic factors. Curr Opin Rheumatol E, Wright JG, Hawker GA. Is self-reported 2001;13:447–51. improvement in osteoarthritis pain and disability reflected in objective measures? J Rheumatol 10. Dillon CF, Rasch EK, Gu Q, Hirsch R. Prevalence 2007;34:159–64. of knee osteoarthritis in the United States: arthritis data from the Third National Health and 23. Bookwala J, Harralson TL, Parmelee PA. Effects Nutrition Examination Survey 1991–94. J Rheumatol of pain on functioning and well-being in older 2006;33:2271–9. adults with osteoarthritis of the knee. Psychol Aging 2003;18:844–50. 11. Yong PF, Milner PC, Payne JN, Lewis PA, Jennison C. Inequalities in access to knee joint replacements 24. Hawker GA, Stewart L, French MR, Cibere J, for people in need. Ann Rheum Dis 2004;63:1483–9. Jordan JM, March L, et al. Understanding the pain experience in hip and knee osteoarthritis – an 12. Garstang SV, Stitik TP. Osteoarthritis: epidemiology, OARSI/OMERACT initiative. Osteoarthritis Cartilage risk factors, and pathophysiology. Am J Phys Med 2008;16:415–22. Rehabil 2006;85:S2–S12. 87

25. Viinamaki H. Sleep disturbance is associated 37. Bellamy N, Kirwan J, Boers M, Brooks P, Strand V, with mental disorder in patients with hip or knee Tugwell P, et al. Recommendations for a core set of osteoarthritis. Eur J Psychiatry 2002;16:9–13. outcome measures for future phase III clinical trials in knee, hip, and hand osteoarthritis. Consensus 26. Rosemann T, Joos S, Szecsenyi J, Laux G, Wensing development at OMERACT III. J Rheumatol M. Health service utilization patterns of primary 1997;24:799–802. care patients with osteoarthritis. BMC Health Serv Res 2007;7:169. 38. Stratford PW, Kennedy DM. Does parallel item content on WOMAC’s pain and function subscales 27. Jinks C, Ong BN, Richardson J. A mixed methods limit its ability to detect change in functional status? study to investigate needs assessment for knee BMC Musculoskelet Disord 2004;5:17. pain and disability: population and individual perspectives. BMC Musculoskeletal Disord 2007;8:59. 39. Tubach F, Wells GA, Ravaud P, Dougados M. Minimal clinically important difference, low disease 28. National Institute of Health and Clinical activity state, and patient acceptable symptom state: Excellence. Osteoarthritis: costing report. Implementing methodological issues. J Rheumatol 2005;32:2025–9. NICE guidelines. London: NICE; 2008. 40. Gossec L, Hawker G, Davis AM, Maillefert JF, 29. National Institute of Health and Clinical Lohmander LS, Altman R, et al. OMERACT/OARSI Excellence. Osteoarthritis: the care and management of initiative to define states of severity and indication osteoarthritis in adults. NICE clinical guideline 59. for joint replacement in hip and knee osteoarthritis. London: NICE; 2008. J Rheumatol 2007;34:1432–5.

30. Chevalier X, Marre JP, de Butler J, Hercek 41. Bruyere O, Richy F, Reginster JY. Three year joint A. Questionnaire survey of management and space narrowing predicts long term incidence of prescription of general practitioners in knee knee surgery in patients with osteoarthritis: an eight osteoarthritis: a comparison with 2000 EULAR year prospective follow up study. Ann Rheum Dis recommendations. Clin Experiment Rheumatol 2005;64:1727–30. 2004;22:205–12. 42. Dacre JE, Scott DL, Da Silva JA, Welsh G, Huskisson 31. The National Joint Registry. National Joint Registry EC. Joint space in radiologically normal knees. Br J for England and Wales 4th Annual Report. Hemel Rheumatol 1991;30:426–8. Hempstead: The National Joint Registry; 2007. 43. Cooper C, Cushnaghan J, Kirwan JR, Dieppe 32. Lord J, Victor C, Littlejohns P, Ross FM, Axford PA, Rogers J, McAlindon T, et al. Radiographic JS. Economic evaluation of a primary care- assessment of the knee joint in osteoarthritis. Ann based education programme for patients with Rheum Dis 1992;51:80–2. osteoarthritis of the knee. Health Technol Assess 1999;3(23). 44. Hochberg MC. Quantitative radiography in osteoarthritis: analysis. Baillieres Clin Rheumatol 33. Victor CR, Triggs E, Ross F, Lord J, Axford JS. 1996;10:421–8. Lack of benefit of a primary care-based nurse-led education programme for people with osteoarthritis 45. Wolfe F, Lane NE. The longterm outcome of of the knee. Clin Rheumatol 2005;24:358–64. osteoarthritis: rates and predictors of joint space narrowing in symptomatic patients with knee 34. Lequesne M. Indices of severity and disease osteoarthritis. J Rheumatol 2002;29:139–46. activity for osteoarthritis. Semin Arthritis Rheum 1991;20(Suppl. 2):48–54. 46. Barker K, Lamb SE, Toye F, Jackson S, Barrington S. Association between radiographic joint space 35. Bellamy N, Buchanan WW, Goldsmith CH, narrowing, function, pain and muscle power in Campbell J, Stitt LW. Validation study of WOMAC: severe osteoarthritis of the knee. Clin Rehabil a health status instrument for measuring 2004;18:793–800. clinically important patient relevant outcomes to antirheumatic drug therapy in patients with 47. Bruyere O, Honore A, Rovati LC, Giacovelli osteoarthritis of the hip or knee. J Rheumatol G, Henrotin YE, Seidel L, et al. Radiologic 1988;15:1833–40. features poorly predict clinical outcomes in knee osteoarthritis. Scand J Rheumatol 2002;31:13–6. 36. Bellamy N. Osteoarthritis – an evaluative index for clinical trials. Hamilton, Ontario: McMaster 48. Chang CB, Han I, Kim SJ, Seong SC, Kim University, 1982. TK. Association between radiological findings

88 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

and symptoms at the patellofemoral joint in 59. Pavelká K, Bruyere O, Rovati LC, Olejarova M, advanced knee osteoarthritis. J Bone Joint Surg Br Giacovelli G, Reginster JY. Relief in mild-to- 2007;89:1324–8. moderate pain is not a confounder in joint space narrowing assessment of full extension knee 49. Cicuttini FM, Baker J, Hart DJ, Spector TD. radiographs in recent osteoarthritis structure- Association of pain with radiological changes in modifying drug trials. Osteoarthritis Cartilage different compartments and views of the knee joint. 2003;11:730–7. Osteoarthritis Cartilage 1996;4:143–7. 60. Davies-Tuck ML, Wluka AE, Wang Y, Teichtahl 50. Brandt KD, Fife RS, Braunstein EM, Katz B. AJ, Jones G, Ding C, et al. The natural history of Radiographic grading of the severity of knee cartilage defects in people with knee osteoarthritis. osteoarthritis: relation of the Kellgren and Lawrence Osteoarthritis Cartilage 2008;16:337–42. grade to a grade based on joint space narrowing, and correlation with arthroscopic evidence of 61. Reichenbach S, Sterchi R, Scherer M, Trelle S, Burgi articular cartilage degeneration. Arthritis Rheum E, Burgi U, et al. Meta-analysis: chondroitin for 1991;34:1381–6. osteoarthritis of the knee or hip. Ann Intern Med 2007;146:580–90. 51. Bingham C, Beary J, Adami S, Clauw D, Cohen S, Dougados M, et al. Significant joint space 62. Towheed TE, Maxwell L, Anastassiades TP, Shea B, narrowing (JSN) in medial knee osteoarthritis (OA) Houpt J, Robinson V, et al. Glucosamine therapy for is associated with increased joint symptoms: results treating osteoarthritis. Cochrane Database Syst Rev of a 2-year randomized placebo-controlled study of 2005;CD002946. risedronate. Osteoarthritis Cartilage 2004;12:S142. 63. Owens S, Wagner P, Vangsness CT, Jr. Recent 52. Bruyere O, Pavelká K, Rovati LC, Gatterova advances in glucosamine and chondroitin J, Giacovelli G, Olejarova M, et al. Total joint supplementation. J Knee Surg 2004;17:185–93. replacement after glucosamine sulphate treatment in knee osteoarthritis: results of a mean 8-year 64. Anderson JW, Nicolosi RJ, Borzelleca JF. observation of patients from two previous 3-year, Glucosamine effects in humans: a review of randomised, placebo-controlled trials. Osteoarthritis effects on glucose metabolism, side effects, safety Cartilage 2008;16:254–60. considerations and efficacy. Food Chem Toxicol 2005;43:187–201. 53. Bruyere O, Richy F, Reginster JY. Minimal joint space narrowing is a better predictor of the need for 65. MHRA. New drugs under intensive surveillance. surgery than mean joint space narrowing: an 8-year URL: www.mhra.gov.uk/home/groups/pl-p/ prospective follow-up study in knee osteoarthritis. documents/websiteresources/con014660.pdf Osteoarthritis Cartilage 2004;12:S54. (accessed 15 September 2008).

54. Cicuttini FM, Jones G, Forbes A, Wluka AE. Rate of 66. Adebowale A, Cox DS, Liang Z, Eddington ND. cartilage loss at two years predicts subsequent total Analysis of glucosamine and chondroitin sulfate knee arthroplasty: a prospective study. Ann Rheum content in marketed products and the Caco-2 Dis 2004;63:1124–7. permeability of chondroitin sulfate raw materials. J Am Nutraceut Assoc 2000;3:37–44. 55. Altman R, Brandt K, Hochberg M, Moskowitz R. Design and conduct of clinical trials in patients with 67. Barnhill JG, Fye CL, Williams DW, Reda DJ, Harris osteoarthritis. Osteoarthritis Cartilage 1996;4:217–43. CL, Clegg DO. Chondroitin product selection for the glucosamine/chondroitin arthritis intervention 56. Botha-Scheepers S, Kloppenburg M, Kroon HM, trial. J Am Pharm Assoc (2003) 2006;46:14–24. Hellio Le Graverand MP, Breedveld FC, Ravaud P, et al. Fixed-flexion knee radiography: the sensitivity to 68. Scottish Medicines Consortium. Glucosamine detect knee joint space narrowing in osteoarthritis. (Alateris). URL: www.scottishmedicines.org.uk/ Osteoarthritis Cartilage 2007;15:350–3. smc/5777.html (accessed 15 September 2008).

57. Chaisson CE, Gale DR, Gale E, Kazis L, Skinner 69. NHS Business Services Authority. Prescription K, Felson DT. Detecting radiographic knee Pricing Division. URL: www.nhsbsa.nhs.uk/ osteoarthritis: what combination of views is optimal? prescriptions (accessed 13 July 200). Rheumatology 2000;39:1218–21. 70. Jordan KM, Sawyer S, Coakley P, Smith HE, 58. Davies AP, Calder DA, Marshall T, Glasgow MM. Cooper C, Arden NK. The use of conventional and Plain radiography in the degenerate knee. A case complementary treatments for knee osteoarthritis in for change. J Bone Joint Surg Br 1999;81:632–5. the community. Rheumatology (Oxford) 2004;43:381– 4. 89

71. Moher D, Cook DJ, Eastwood S, Olkin I, Rennie al. Glucosamine sulfate in the treatment of D, Stroup DF. Improving the quality of reports of knee osteoarthritis symptoms: a randomized, meta-analyses of randomised controlled trials: the double-blind, placebo-controlled study using QUOROM statement. Quality of reporting of meta- acetaminophen as a side comparator. Arthritis Rheum analyses. Lancet 1999;354:1896–900. 2007;56:555–67.

72. Khan, KS, ter Riet, G, Glanville, J, Sowden, A, 83. Houpt JB, McMillan R, Wein C, Paget-Dellio Kleijnen J. Undertaking systematic reviews of research on SD. Effect of glucosamine hydrochloride in the effectiveness: CRD’s guidance for those carrying out and treatment of pain of osteoarthritis of the knee. J commissioning reviews. CRD Report 4, 2nd edn. 2001. Rheumatol 1999;26:2423–30.

73. Poolsup N, Suthisisang C, Channark P, Kittikulsuth 84. Hughes R, Carr A. A randomized, double-blind, W. Glucosamine long-term treatment and the placebo-controlled trial of glucosamine sulphate progression of knee osteoarthritis: systematic review as an analgesic in osteoarthritis of the knee. of randomized controlled trials. Ann Pharmacother Rheumatology 2002;41:279–84. 2005;39:1080–7. 85. McAlindon T, Formica M, LaValley M, Lehmer 74. Bana G, Jamard B, Verrouil E, Mazieres B. M, Kabbara K. Effectiveness of glucosamine for Chondroitin sulfate in the management of hip and symptoms of knee osteoarthritis: results from an knee osteoarthritis: an overview. Adv Pharmacol internet-based randomized double-blind controlled 2006;53:507–22. trial. Am J Med 2004;117:643–9.

75. Richy F, Bruyere O, Ethgen O, Cucherat 86. Müller-Fassbender H, Bach GL, Haase W, Rovati M, Henrotin Y, Reginster JY. Structural and LC, Setnikar I. Glucosamine sulfate compared to symptomatic efficacy of glucosamine and ibuprofen in osteoarthritis of the knee. Osteoarthritis chondroitin in knee osteoarthritis: a comprehensive Cartilage 1994;2:61–9. meta-analysis. Arch Intern Med 2003;163:1514–22. 87. Noack W, Fischer M, Forster KK, Rovati LC, 76. Reginster JY, Deroisy R, Rovati LC, Lee RL, Setnikar I. Glucosamine sulfate in osteoarthritis of Lejeune E, Bruyere O, et al. Long-term effects of the knee. Osteoarthritis Cartilage 1994;2:51–9. glucosamine sulphate on osteoarthritis progression: a randomised, placebo-controlled clinical trial. 88. Pujalte JM, Llavore EP, Ylescupidez FR. Double- Lancet 2001;357:251–6. blind clinical evaluation of oral glucosamine sulphate in the basic treatment of osteoarthrosis. 77. Pavelká K, Gatterova J, Olejarova M, Machacek S, Curr Med Res Opin 1980;7:110–14. Giacovelli G, Rovati LC. Glucosamine sulfate use and delay of progression of knee osteoarthritis: a 89. Qiu GX, Gao SN, Giacovelli G, Rovati L, Setnikar 3-year, randomized, placebo-controlled, double- I. Efficacy and safety of glucosamine sulfate versus blind study. Arch Intern Med 2002;162:2113–23. ibuprofen in patients with knee osteoarthritis. Arzneimittelforschung 1998;48:469–74. 78. Cibere J, Kopec JA, Thorne A, Singer J, Canvin J, Robinson DB, et al. Randomized, double-blind, 90. Reichelt A, Forster KK, Fischer M, Rovati LC, placebo-controlled glucosamine discontinuation Setnikar I. Efficacy and safety of intramuscular trial in knee osteoarthritis. Arthritis Rheum glucosamine sulfate in osteoarthritis of the knee 2004;51:738–45. – a randomized, placebo-controlled, double-blind study. Arzneimittelforschung/Drug Res 1994;44:75–80. 79. Crolle G, D’Este E. Glucosamine sulphate for the management of arthrosis: a controlled clinical 91. Rindone JP, Hiller D, Collacott E, Nordhaugen investigation. Curr Med Res Opin 1980;7:104–9. N, Arriola G. Randomized, controlled trial of glucosamine for treating osteoarthritis of the knee. 80. D’Ambrosio E, Casa B, Bompani R, Scali G, Scali West J Med 2000;172:91–4. M. Glucosamine sulphate: a controlled clinical investigation in arthrosis. Pharmatherapeutica 92. Rovati L. The clinical profile of glucosamine 1981;2:504–8. sulfate as a selective symptom-modifying drug in osteoarthritis: current data and perspectives. 81. Drovanti A, Bignamini AA, Rovati AL. Therapeutic Osteoarthritis Cartilage 1997;5:72. activity of oral glucosamine sulfate in osteoarthrosis: a placebo-controlled double-blind investigation. 93. Usha PR. Randomised, double-blind, parallel, Clin Ther 1980;3:260–72. placebo-controlled study of oral glucosamine, methylsulfonylmethane and their combination in 82. Herrero-Beaumont G, Ivorra JA, Del Carmen osteoarthritis. Clin Drug Investig 2004;24:353–63. 90 TM, Blanco FJ, Benito P, Martin-Mola E, et DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

94. Vajaradul Y. Double-blind clinical evaluation of in the treatment of gonarthrosis and coxarthrosis. intra-articular glucosamine in outpatients with 5-months result of a multicenter double-blind gonarthrosis. Clin Ther 1981;3:336–43. controlled prospective study using placebo. Rev Rhum Mal Osteoartic 1992;59:466–72. 95. Vaz LA. Double-blind clinical evaluation of the relative efficacy of ibuprofen and glucosamine 106. Mazieres B, Combe B, Phan VA, Tondut J, Grynfeltt sulphate in the management of osteoarthrosis M. Chondroitin sulfate in osteoarthritis of the knee: of the knee in out-patients. Curr Med Res Opin a prospective, double blind, placebo controlled 1982;8:145–9. multicenter clinical study. J Rheumatol 2001;28:173– 81. 96. Zenk JL. The effects of milk protein concentrate on the symptoms of osteoarthritis in adults: 107. Mazieres B, Hucher M, Zaim M, Garnero P. Effect an exploratory, randomized, double-blind, of chondroitin sulphate in symptomatic knee placebo-controlled trial. Curr Ther Res – Clin Exp osteoarthritis: a multicentre, randomised, double- 2002;63:430–42. blind, placebo-controlled study. Ann Rheum Dis 2007;66:639–45. 97. Alekseeva LI, Benevolenskaia LI, Nasonov EL, Chichasova NV, Kariakin AN. Structum (chondroitin 108. Michel BA, Stucki G, Frey D, De VF, Vignon E, sulfate) – a new agent for the treatment of Bruehlmann P, et al. Chondroitins 4 and 6 sulfate in osteoarthrosis. Ter Arkh 1999;71:51–3. osteoarthritis of the knee: a randomized, controlled trial. Arthritis Rheum 2005;52:779–86. 98. Bourgeois P, Chales G, Dehais J, Delcambre B, Kuntz JL, Rozenberg S. Efficacy and tolerability 109. Morreale P, Manopulo R, Galati M, Boccanera of chondroitin sulfate 1200 mg/day vs chondroitin L, Saponati G, Bocchi L. Comparison of the sulfate 3 400 mg/day vs placebo. Osteoarthritis antiinflammatory efficacy of chondroitin sulfate Cartilage 1998;6(Suppl. A):25–30. and diclofenac sodium in patients with knee osteoarthritis. J Rheumatol 1996;23:1385–91. 99. Bucsi L, Poor G. Efficacy and tolerability of oral chondroitin sulfate as a symptomatic slow-acting 110. Nasonova VA, Alekseeva LI, Arkhangel’skaia GS, drug for osteoarthritis (SYSADOA) in the treatment Davydova AF, Karmil’tseva EA, Kogan KM, et al. of knee osteoarthritis. Osteoarthritis Cartilage Results of the multicenter clinical trial of structum 1998;6(Suppl. 6). preparation in Russia. Ter Arkh 2001;73:84–7.

100. Conrozier T, Vignon E. Die Wirkung von 111. Pavelká K, Manopulo R, Bucsi L. Double-blind, Chondroitinsulfat bei der Behandlung der dose-effect study of oral chondroitin 4 & 6 sulfate Hüftgelenksarthrose: eine Doppelblindstudie gegen 1200 mg, 800 mg, 200 mg, and placebo in the Placebo. Litera Rheumatol 1992;14:69–75. treatment of knee osteoarthritis. Litera Rheumatol 1999;24:21–30. 101. Fleisch A, Melin C, Imhoff A, Hodler J, Kissling R. A one-year randomized, double-blind, placebo- 112. Rovetta G. Galactosaminoglycuronoglycan sulfate controlled study with oral chondroitin sulfate in (matrix) in therapy of tibiofibular osteoarthritis of patients with knee osteoarthritis. Osteoarthritis the knee. Drugs Exp Clin Res 1991;17:53–7. Cartilage 1997;5:S70. 113. Soroka NF, Chyzh KA. Clinical efficiency and 102. Kahan A, Vignon E, Uebelhart D, Reginster JY. A pharmacoeconomical evaluation of treatment randomized controlled trial with chondroitin sulfate by chondroitinsulphate (Structumc) in patients involving over 600 patients: STOPP (STudy on with primary osteoarthritis (POA). Ann Rheum Dis Osteoarthritis Progression Prevention). Osteoporosis 2002;61:AB0289. Int 2007;18:S14–S15. 114. Uebelhart D, Thonar EJ, Delmas PD, Chantraine 103. Kerzberg EM, Roldan EJ, Castelli G, Huberman A, Vignon E. Effects of oral chondroitin sulfate on ED. Combination of glycosaminoglycans and the progression of knee osteoarthritis: a pilot study. acetylsalicylic acid in knee osteoarthrosis. Scand J Osteoarthritis Cartilage 1998;6(Suppl. A):39–46. Rheumatol 1987;16:377–80. 115. Uebelhart D, Knussel O, Theiler R. Effects of oral 104. L’Hirondel J. Doppelblind-Studie mit oral chondroitin sulfate on the progression of knee verabreichtem Chondroitinsulfat gegen Placebo osteoarthritis: a pilot study. Schweiz Med Wochenschr bei der tibiofemoralen Gonarthrose (125 patients). 1999;129:1174. Litera Rheumatol 1992;14:77–84. 116. Uebelhart D, Malaise M, Marcolongo R, DeVathaire 105. Mazieres B, Loyau G, Menkes CJ, Valat JP, F, Piperno M, Mailleux E, et al. Intermittent Dreiser RL, Charlot J, et al. Chondroitin sulfate treatment of knee osteoarthritis with oral 91

chondroitin sulfate: a one-year, randomized, 127. Mehta K, Gala J, Bhasale S, Naik S, Modak M, double-blind, multicenter study versus placebo. Thakur H, et al. Comparison of glucosamine sulfate Osteoarthritis Cartilage 2004;12:269–76. and a polyherbal supplement for the relief of osteoarthritis of the knee: a randomized controlled 117. Clegg DO, Reda DJ, Harris CL, Klein MA, O’Dell trial [ISRCTN25438351]. BMC Complement Altern JR, Hooper MM et al. Glucosamine, chondroitin Med 2007;7:34. sulfate, and the two in combination for painful knee osteoarthritis. New Engl J Med 2006;354:795–808. 128. Zhang WB, Zhuang CY, Li JM, Yang ZP, Chen XL. Efficacy and safety evaluation of glucosamine 118. Cohen M, Wolfe R, Mai T, Lewis D. A randomized, hydrochloride in the treatment of osteoarthritis. double blind, placebo controlled trial of a topical Chung-Hua Wai Ko Tsa Chih [Chinese Journal of cream containing glucosamine sulfate, chondroitin Surgery] 2007;45:998–1001. sulfate, and camphor for osteoarthritis of the knee. [erratum appears in J Rheumatol 2003 30:2512]. J 129. Debi R, Robinson D, Agar G, Halperin N. GAG Rheumatol 2003;30:523–8. for osteoarthritis of the knee – a prospective study. Harefuah 2000;138:451–3, 518. 119. Das A, Jr., Hammad TA. Efficacy of a combination of FCHG49 glucosamine hydrochloride, TRH122 130. Leffler CT, Philippi AF, Leffler SG, Mosure JC, Kim low molecular weight sodium chondroitin sulfate PD. Glucosamine, chondroitin, and manganese and manganese ascorbate in the management ascorbate for degenerative joint disease of the knee of knee osteoarthritis. Osteoarthritis Cartilage or low back: a randomized, double-blind, placebo- 2000;8:343–50. controlled pilot study. Military Med 1999;164:85–91.

120. Rai J. Efficacy of chondroitin sulfate and 131. Nguyen P, Mohamed SE, Gardiner D, Salinas T. glucosamine sulfate in the progression of A randomized double-blind clinical trial of the symptomatic knee osteoarthritis: a randomized, effect of chondroitin sulfate and glucosamine placebo-controlled, double blind study. Bull Postgrad hydrochloride on temporomandibular joint Inst Med Educ Res, Chandigarh 2004;38:18–22. disorders: a pilot study. Cranio 2001;19:130–9.

121. Conrozier T. Anti-arthrosis treatments: efficacy and 132. Stumpf JL, Lin SW. Effect of glucosamine on tolerance of chondroitin sulfates (CS 4&6). Presse glucose control. Ann Pharmacother 2006;40:694–8. Med 1998;27:1862–5. 133. Scroggie DA. The effect of glucosamine-chondroitin 122. Messier SP, Mihalko S, Loeser RF, Legault C, Jolla supplementation on glycosylated hemoglobin levels J, Pfruender J, et al. Glucosamine/chondroitin in patients with type 2 diabetes mellitus: A placebo- combined with exercise for the treatment of knee controlled, double-blinded, randomized clinical osteoarthritis: a preliminary study. Osteoarthritis trial. Arch Intern Med 2003;163:1587–90. Cartilage 2007;15:1256–66. 134. Knudsen JF, Sokol GH. Potential glucosamine– 123. Kawasaki T, Kurosawa H, Ikeda H, Kim SG, Osawa warfarin interaction resulting in increased A, Takazawa Y, et al. Additive effects of glucosamine international normalized ratio: case report and or risedronate for the treatment of osteoarthritis review of the literature and MedWatch database. of the knee combined with home exercise: a Pharmacotherapy 2008;28:540–8. prospective randomized 18-month trial. J Bone Miner Metab 2008;26:279–87. 135. Adverse Drug Reactions Advisory Committee (ADRAC), Adverse Drug Reactions Unit (ADRU. 124. Sawitzke AD, Shi H, Finco MF, Dunlop DD, Interaction between glucosamine and warfarin. Bingham CO, III, Harris CL, et al. The effect of Austr Adverse Drug React Bull 2008;27:3. glucosamine and/or chondroitin sulfate on the progression of knee osteoarthritis: a report from the 136. BBC News. Arthritis remedy theory for death. 2007. glucosamine/chondroitin arthritis intervention trial. URL: http://news.bbc.co.uk/1/hi/scotland/tayside_ Arthritis Rheum 2008;58:3183–91. and_central/7131026.stm (accessed 27 November 2008). 125. Braham R, Dawson B, Goodman C. The effect of glucosamine supplementation on people 137. Hurley MV, Walsh NE, Mitchell HL, Pimm TJ, experiencing regular knee pain. Br J Sports Med Williamson E, Jones RH, et al. Economic evaluation 2003;37:45–9. of a rehabilitation program integrating exercise, self-management, and active coping strategies for 126. Frestedt JL, Walsh M, Kuskowski MA, Zenk JL. A chronic knee pain. Arthritis Rheum 2007;57:1220–9. natural mineral supplement provides relief from knee osteoarthritis symptoms: a randomized 138. Grootendorst P, Marshall D, Pericak D, Bellamy N, 92 controlled pilot trial. Nutr J 2008;7:9. Feeny D, Torrance GW. A model to estimate health DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

utilities index mark 3 utility scores from WOMAC weight chondroitin sulfate after single and multiple index scores in patients with osteoarthritis of the doses to beagle dogs. Biopharm Drug Dispos knee. J Rheumatol 2007;34:534–42. 2002;23:217–25.

139. Barton G, Sach T, Jenkinson C, Avery A, Docherty 152. Setnikar I. Absorption, distribution, metabolism M, Muir K. Do estimates of cost–utility based on the and excretion of glucosamine sulfate: a review. EQ-5D differ from those based on the mapping of Arzneimittelforschung/Drug Res 2001;51:699–725. utility scores? Health Qual Life Outcomes 2008;6:51. 153. Setnikar I, Giachetti C, Zanolo G. Absorption, 140. Office for National Statistics. Health Statistics distribution and excretion of radioactivity after Quarterly. No 33, Spring 2007. Newport: Office for a single intravenous or oral administration of National Statistics; 2007. [14C] glucosamine to the rat. Pharmatherapeutica 1984;3:538–50. 141. Nunez M, Nunez E, del Val JL, Ortega R, Segur JM, Hernandez MV, et al. Health-related quality 154. Setnikar I. Pharmacokinetics of glucosamine in of life in patients with osteoarthritis after total the dog and in man. Arzneimittelforschung/Drug Res knee replacement: factors influencing outcomes 1986;36:729–35. at 36 months of follow-up. Osteoarthritis Cartilage 2007;15:1001–7. 155. Setnikar I. Absorption, distribution and excretion of radioactivity after single intravenous or oral 142. Department of Health. National Tarriff 2006–7. administration of D-[U-14C] glucose to the rat. Acta London: Department of Health; 2006. Ther 1986;12:353–71.

143. Rotta Pharmaceuticals. DONA: glucosamine sulfate. 156. Setnikar I, Palumbo R, Canali S, Zanolo G. 2008. URL: www.donausa.com/order.php (accessed Pharmacokinetics of glucosamine in man. 13 July 2009). Arzneimittelforschung/Drug Res 1993;43:1109–13.

144. BMJ Group and RPS Publishing. British national 157. Biggee BA, Blinn CM, McAlindon TE, Nuite M, formulary 55. 2008. URL: http://bnf.org/bnf/bnf/ Silbert JE. Low levels of human serum glucosamine current/index.htm (accessed 15 January 2008). after ingestion of glucosamine sulphate relative to capability for peripheral effectiveness. Ann Rheum 145. HM Treasury. The green book. 2007. URL: www.hm- Dis 2006;65:222–6. treasury.gov.uk/data_data_greenbook_index.htm (accessed 13 July 2009). 158. Basak M, Joseph S, Joshi S, Sawant S. Comparative bioavailability of a novel timed release and 146. Briggs A, Claxton K, Sculpher M. Decision modelling powder-filled glucosamine sulfate formulation – a for health economic evaluation. Oxford: Oxford multi-dose, randomized, crossover study. Int J Clin University Press, 2006. Pharmacol Ther 2004;42:597–601.

147. Department of Health. NHS reference costs 159. Persiani S, Rotini R, Trisolino G, Rovati LC, 2004. 2005. URL: http://www.dh.gov.uk/ Locatelli M, Paganini D, et al. Synovial and plasma en/Publicationsandstatistics/Publications/ glucosamine concentrations in osteoarthritic PublicationsPolicyAndGuidance/DH_4105545 patients following oral crystalline glucosamine (accessed 27 November 2008). sulphate at therapeutic dose. Osteoarthritis Cartilage 2007;15:764–72. 148. Fioravanti A, Collodel G. In vitro effects of chondroitin sulfate. Adv Pharmacol 2006;53:449–65. 160. Persiani S, Roda E, Rovati LC, Locatelli M, Giacovelli G, Roda A. Glucosamine oral 149. Volpi N. Chondroitin sulphate for the treatment of bioavailability and plasma pharmacokinetics after osteoarthritis. Curr Med Chem 2005;4:221–34. increasing doses of crystalline glucosamine sulfate in man. Osteoarthritis Cartilage 2005;13:1041–9. 150. Laverty S, Sandy JD, Celeste C, Vachon P, Marier JF, Plaas AH. Synovial fluid levels and serum 161. Nakayama Y, Narita T, Mori A, Uesaka S, Miyazaki pharmacokinetics in a large animal model following K, Ito H. The effects of age and sex on chondroitin treatment with oral glucosamine at clinically sulfates in normal synovial fluid. Arthritis Rheum relevant doses. Arthritis Rheum 2005;52:181–91. 2002;46:2105–8.

151. Adebowale A, Du J, Liang Z, Leslie JL, Eddington 162. Volpi N. Oral bioavailability of chondroitin sulfate ND. The bioavailability and pharmacokinetics of (Condrosulf) and its constituents in healthy male glucosamine hydrochloride and low molecular volunteers. Osteoarthritis Cartilage 2002;10:768–77.

163. Volpi N. Oral absorption and bioavailability of 174. Piperno M, Reboul P, Hellio Le Graverand MP, ichthyic origin chondroitin sulfate in healthy male Peschard MJ, Annefeld M, Richard M, et al. volunteers. Osteoarthritis Cartilage 2003;11:433–41. Glucosamine sulfate modulates dysregulated activities of human osteoarthritic chondrocytes in 164. Ronca F, Palmieri L, Panicucci P, Ronca G. Anti- vitro. Osteoarthritis Cartilage 2000;8:207–12. inflammatory activity of chondroitin sulfate. Osteoarthritis Cartilage 1998;6(Suppl. A):14–21. 175. Monfort J, Nacher M, Montell E, Vila J, Verges J, Benito P. Chondroitin sulfate and hyaluronic acid 165. Ebube NK, Mark W, Hahm H. Preformulation (500–730 kDa) inhibit stromelysin-1 synthesis in studies and characterization of proposed human osteoarthritic chondrocytes. Drugs Exp Clin chondroprotective agents: glucosamine HCl and Res 2005;31:71–6. chondroitin sulfate. Pharm Dev Technol 2002;7:457– 69. 176. Tat SK, Pelletier JP, Verges J, Lajeunesse D, Montell E, Fahmi H, et al. Chondroitin and glucosamine 166. Dodge GR, Jimenez SA. Glucosamine sulfate sulfate in combination decrease the pro-resorptive modulates the levels of aggrecan and matrix properties of human osteoarthritis subchondral metalloproteinase-3 synthesized by cultured human bone osteoblasts: a basic science study. Arthritis Res osteoarthritis articular chondrocytes. Osteoarthritis Ther 2007;9:R117. Cartilage 2003;11:424–32. 177. Cibere J, Thorne A, Kopec JA, Singer J, Canvin 167. Noyszewski EA, Wroblewski K, Dodge GR, J, Robinson DB, et al. Glucosamine sulfate Kudchodkar S, Beers J, Sarma AV, et al. and cartilage type II collagen degradation in Preferential incorporation of glucosamine into patients with knee osteoarthritis: randomized the galactosamine moieties of chondroitin sulfates discontinuation trial results employing biomarkers. in articular cartilage explants. Arthritis Rheum J Rheumatol 2005;32:896–902. 2001;44:1089–95. 178. Ilic MZ, Martinac B, Handley CJ. Effects of long- 168. Derfoul A, Miyoshi AD, Freeman DE, Tuan RS. term exposure to glucosamine and mannosamine Glucosamine promotes chondrogenic phenotype on aggrecan degradation in articular cartilage. in both chondrocytes and mesenchymal stem Osteoarthritis Cartilage 2003;11:613–22. cells and inhibits MMP-13 expression and matrix degradation. Osteoarthritis Cartilage 2007;15:646–55. 179. Sandy JD, Gamett D, Thompson V, Verscharen C. Chondrocyte-mediated catabolism of aggrecan: 169. Varghese S, Theprungsirikul P, Sahani S, Hwang N, aggrecanase-dependent cleavage induced by Yarema KJ, Elisseeff JH. Glucosamine modulates interleukin-1 or retinoic acid can be inhibited by chondrocyte proliferation, matrix synthesis, and glucosamine. Biochem J 1998;335 (Pt 1):59–66. gene expression. Osteoarthritis Cartilage 2007;15:59– 68. 180. Sandy JD, Thompson V, Verscharen C, Gamett D. Chondrocyte-mediated catabolism of aggrecan: 170. Nishimoto S, Takagi M, Wakitani S, Nihira T, evidence for a glycosylphosphatidylinositol-linked Yoshida T. Effect of chondroitin sulfate and protein in the aggrecanase response to interleukin-1 hyaluronic acid on gene expression in a three- or retinoic acid. Arch Biochem Biophys 1999;367:258– dimensional culture of chondrocytes. J Biosci Bioeng 64. 2005;100:123–6. 181. Tiku ML, Narla H, Jain M, Yalamanchili P. 171. Mroz PJ, Silbert JE. Use of 3H-glucosamine and Glucosamine prevents in vitro collagen degradation 35S-sulfate with cultured human chondrocytes to in chondrocytes by inhibiting advanced lipoxidation determine the effect of glucosamine concentration reactions and protein oxidation. Arthritis Res Ther on formation of chondroitin sulfate. Arthritis Rheum 2007;9:R76. 2004;50:3574–9. 182. Jomphe C, Gabriac M, Hale TM, Heroux L, 172. Kim MM, Mendis E, Rajapakse N, Kim SK. Trudeau LE, Deblois D, et al. Chondroitin sulfate Glucosamine sulfate promotes osteoblastic inhibits the nuclear translocation of nuclear differentiation of MG-63 cells via anti-inflammatory factor-kappaB in interleukin-1beta-stimulated effect. Bioorg Med Chem Lett 2007;17:1938–42. chondrocytes. Basic Clin Pharmacol Toxicol 2008;102:59–65. 173. Uitterlinden EJ, Jahr H, Koevoet JL, Jenniskens YM, Bierma-Zeinstra SM, DeGroot J, et al. 183. Largo R, Alvarez-Soria MA, Diez-Ortego I, Calvo Glucosamine decreases expression of anabolic and E, Sanchez-Pernaute O, Egido J, et al. Glucosamine catabolic genes in human osteoarthritic cartilage inhibits IL-1beta-induced NFkappaB activation in explants. Osteoarthritis Cartilage 2006;14:250–7. human osteoarthritic chondrocytes. Osteoarthritis 94 Cartilage 2003;11:290–8. DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

184. Pelletier JP, Mineau F, Ranger P, Tardif G, 195. Cordoba F, Nimni ME. Chondroitin sulfate and Martel-Pelletier J. The increased synthesis of other sulfate containing chondroprotective agents inducible nitric oxide inhibits IL-1ra synthesis by may exhibit their effects by overcoming a deficiency human articular chondrocytes: possible role in of sulfur amino acids. Osteoarthritis Cartilage osteoarthritic cartilage degradation. Osteoarthritis 2003;11:228–30. Cartilage 1996;4:77–84. 196. Ameye LG, Chee WS. Osteoarthritis and nutrition. 185. Blanco FJ, Guitian R, Vazquez-Martul E, de Toro From nutraceuticals to functional foods: a systematic FJ, Galdo F. Osteoarthritis chondrocytes die by review of the scientific evidence. Arthritis Res Ther apoptosis. A possible pathway for osteoarthritis 2006;8:R127. pathology. Arthritis Rheum 1998;41:284–9. 197. Biggee BA, McAlindon T. Glucosamine for 186. Chan PS, Caron JP, Rosa GJ, Orth MW. osteoarthritis: part I, review of the clinical evidence. Glucosamine and chondroitin sulfate regulate Med Health, Rhode Island 2004;87:176–9. gene expression and synthesis of nitric oxide and prostaglandin E(2) in articular cartilage explants. 198. Bjordal JM, Klovning A, Ljunggren AE, Slordal Osteoarthritis Cartilage 2005;13:387–94. L. Short-term efficacy of pharmacotherapeutic interventions in osteoarthritic knee pain: a meta- 187. Chan PS, Caron JP, Orth MW. Short-term gene analysis of randomised placebo-controlled trials. expression changes in cartilage explants stimulated Eur J Pain 2007;11:125–38. with interleukin beta plus glucosamine and chondroitin sulfate. J Rheumatol 2006;33:1329–40. 199. Clayton JJ. Nutraceuticals in the management of osteoarthritis. Orthopedics 2007;30:624–9. 188. Huser CA, Davies ME. Effect of a glucosamine derivative on impact-induced chondrocyte apoptosis 200. Distler J, Anguelouch A. Evidence-based practice: in vitro. A preliminary report. Osteoarthritis Cartilage review of clinical evidence on the efficacy of 2008;16:125–8. glucosamine and chondroitin in the treatment of osteoarthritis. J Am Acad Nurse Pract 2006;18:487– 189. Caraglia M, Beninati S, Giuberti G, D’Alessandro 93. AM, Lentini A, Abbruzzese A, et al. Alternative therapy of earth elements increases the 201. Felson DT, McAlindon TE. Glucosamine and chondroprotective effects of chondroitin sulfate in chondroitin for osteoarthritis: to recommend or not mice. Exp Mol Med 2005;37:476–81. to recommend? Arthritis Care Res 2000;13:179–82.

190. Lippiello L. Glucosamine and chondroitin sulfate: 202. Leeb BF, Schweitzer H, Montag K, Smolen JS. A biological response modifiers of chondrocytes under metaanalysis of chondroitin sulfate in the treatment simulated conditions of joint stress. Osteoarthritis of osteoarthritis. J Rheumatol 2000;27:205–11. Cartilage 2003;11:335–42. 203. Lozada CJ. Glucosamine in osteoarthritis: questions 191. Nerucci F, Fioravanti A, Cicero MR, Collodel remain. Cleve Clin J Med 2007;74:65–71. G, Marcolongo R. Effects of chondroitin sulfate and interleukin-1beta on human chondrocyte 204. Matheson AJ, Perry CM. Glucosamine: a review of cultures exposed to pressurization: a biochemical its use in the management of osteoarthritis. Drugs and morphological study. Osteoarthritis Cartilage Aging 2003;20:1041–60. 2000;8:279–87. 205. McAlindon TE. Glucosamine and chondroitin 192. Nimni ME. Chondroitin sulfate and sulfur for treatment of osteoarthritis: a systematic containing chondroprotective agents: is there a basis quality assessment and meta-analysis. JAMA for their pharmacological action? Curr Rheumatol 2000;283:1469–75. Rev 2006;2:137–49. 206. McAlindon T. Glucosamine and chondroitin for 193. van der Kraan PM. High susceptibility of human osteoarthritis? Bull Rheum Dis 2001;50:d-4. articular cartilage glycosaminoglycan synthesis to changes in inorganic sulfate availability. J Orthop Res 207. McAlindon TE. Nutritional factors and 1990;8:565–71. osteoarthritis: recent developments. Curr Opin Rheumatol 2005;17:647–52. 194. Hoffer LJ, Kaplan LN, Hamadeh MJ, Grigoriu AC, Baron M. Sulfate could mediate the therapeutic 208. McAlindon TE. Is treatment with chondroitin effect of glucosamine sulfate. Metabolism efficacious for osteoarthritis? Nat Clin Pract 2001;50:767–70. Rheumatol 2007;3:606–7.

209. Reginster JY, Bruyere O, Neuprez A. Current role 212. Zerkak D, Dougados M. The use of glucosamine of glucosamine in the treatment of osteoarthritis. therapy in osteoarthritis. Curr Pain Headache Rep Rheumatology 2007;46:731–5. 2004;8:507–11.

210. Towheed TE, Anastassiades T. Glucosamine 213. Bruyere O, Honore A, Ethgen O, Rovati LC, therapy for osteoarthritis: an update. J Rheumatol Giacovelli G, Henrotin YE, et al. Correlation 2007;34:1787–90. between radiographic severity of knee osteoarthritis and future disease progression. Results from 211. Vlad SC, LaValley MP, McAlindon TE, Felson DT. a 3-year prospective, placebo-controlled study Glucosamine for pain in osteoarthritis: why do trial evaluating the effect of glucosamine sulfate. results differ? Arthritis Rheum 2007;56:2267–77. Osteoarthritis Cartilage 2003;11:1–5.

96 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Appendix 1 Search strategies

Main search 2 osteoarthritis.mp. EMBASE 1996 to 2007 week 43 3 1 and 2 Search terms No. of hits 4 randomized controlled trial.pt. 1 exp Chondroitin/ 467 5 random$.tw. 2 chondroitin.tw. 3604 6 meta-analysis.pt. 3 1 or 2 3820 7 (meta-analysis or systematic review).tw. 4 exp Glucosamine/ 1792 8 4 or 5 or 6 or 7 5 glucosamine.tw. 2799 9 3 and 8 6 4 or 5 3468 Total number of hits = 101. 7 3 or 6 6899 8 exp Osteoarthritis, Knee/ 4582 EMBASE 1996 to 2007 week 43 9 (osteoarthritis adj3 knee$).tw. 2463 1 (glucosamine or chondroitin).mp. 10 8 or 9 5021 2 osteoarthritis.mp. 11 7 and 10 250 3 1 and 2 12 limit 11 to English language 213 4 random$.tw. 5 (meta-analysis or systematic review).tw. Ovid MEDLINE 1996 to 6 4 or 5 October week 3 2007 7 3 and 6 Total number of hits = 108. Search terms No. of hits 1 exp Chondroitin/ 2620 Cochrane Library issue 3 2007 2 chondroitin.tw. 3863 1 (osteoarthritis and knee):ti,ab,kw 3 1 or 2 4765 2 (chondroitin):ti,ab,kw or (glucosamine):ti,ab,kw 4 exp Glucosamine/ 2655 3 1 and 2 5 glucosamine.tw. 2920 Total number of hits = 77. 6 4 or 5 4373 7 3 or 6 8800 Additional searches 8 exp Osteoarthritis, Knee/ 4159 SCI meeting abstracts only (14/11/2007) 9 (osteoarthritis adj3 knee$).tw. 2406 10 8 or 9 4818 TS=((glucosamine or chondroitin) and 11 7 and 10 150 (osteoarthritis or knee)) 12 limit 11 to English language 125 DocType=Meeting Abstract OR Meeting Summary OR Meeting-Abstract; Language=All AMED (Allied and Complementary languages; Database=SCI-EXPANDED; Medicine) 1985 to October 2007 Timespan=1970–2007 Search terms No. of hits Websites 1 (glucosamine or chondroitin). 53 ACNFP – Advisory Committee on Novel Foods mp. [mp=abstract, heading and Processes: www.acnfp.gov.uk/assess/fullapplics/ words, title] glucosamine Food Standards Agency: www.food.gov.uk/news/ Repeat search strategies in MEDLINE and newsarchive/2007/sep/glucosamine EMBASE for high specificity for systematic reviews US Food and Drug Administration – Center for or RCTs on chondroitin or glucosamine for Food Safety and Applied Nutrition (CFSAN): www. osteoarthritis. cfsan.fda.gov/~dms/qhcosteo.html European Food Safety Authority (EFSA): www. Ovid MEDLINE 1996 to efsa.europa.eu/EFSA/efsa_locale-1178620753812_ October week 4 2007 home.htm 1 (glucosamine or chondroitin).mp. No documents found. 97

Searches for adverse effects of 5 3 and 4 87 glucosamine and chondroitin 6 exp Chondroitin/ae [Adverse 21 Ovid MEDLINE 1950 to Drug Reaction] March week 4 2008 7 exp Glucosamine/ae [Adverse 118 Search terms No. of hits Drug Reaction] 1 exp Chondroitin/ae [Adverse 129 8 6 or 7 123 Effects] 9 5 or 8 199 2 exp Glucosamine/ae [Adverse 144 AMED (Allied and Complementary Effects] Medicine) 1985 to March 2008 3 exp Chondroitin/ 7783 4 exp Glucosamine/ 11,072 (chondroitin or glucosamine) and (adverse or 5 3 or 4 18,390 safety or harm* or pharmacovigilance or risk* or 6 (risk or safety or adverse 332,967 interaction* or side-effect*) = 26 retrieved. or harm or interaction$or pharmacovigilance).m_titl. ISI Web of Science and ISI 7 5 and 6 429 Proceedings 1980 to March 2008 8 1 or 2 or 7 672 ((chondroitin or glucosamine) and (adverse or EMBASE 1980 to 2008 week 13 safety or harm* or pharmacovigilance or risk* or interaction*)) = 125 retrieved Search terms No. of hits 1 exp Chondroitin/ 651 Additional websites searched 2 exp Glucosamine/ 2649 • UKMI (www.ukmi.nhs.uk/) 3 1 or 2 3089 • FDA (www.fda.gov/) 4 (risk or safety or adverse 275,151 • MHRA (www.mhra.gov.uk/index.htm). or harm or interaction$or pharmacovigilance).m_titl.

98 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Appendix 2 Table of excluded reviews

Review Reason for exclusion AHRQ 20072 No outcomes regarding joint structure assessed; high quality review, but mainly a review of systematic reviews Ameye 2006196 Review of functional nutritional ingredients targeting osteoarthritis; glucosamine and chondroitin sulphate excluded Biggee 2004197 Not a systematic review Bjordal 2007198 Only short-term data reported; main data were reported for up to 4 weeks; the longest follow-up reported for 12 weeks Clayton 2007199 Not a systematic review Distler 2006200 Not a systematic review Felson 2000201 Not a systematic review Leeb 2000202 No outcomes regarding joint structure assessed Lozada 2007203 Not a systematic review Matheson 2003204 Not a systematic review (some information on search strategy, but no other methodology described) McAlindon 2000205 No outcomes regarding joint structure assessed McAlindon 2001206 Not a systematic review (summary of McAlindon 2000205) McAlindon 2005207 Not a systematic review McAlindon 2007208 Not a systematic review (comment on Reichenbach 200761) Reginster 2007209 Not a systematic review Towheed 2007210 Not a systematic review Vlad 2007211 No outcomes regarding joint structure assessed Zerkak 2004212 Not a systematic review

AHRQ, Agency for Healthcare Research and Quality.

DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Appendix 3 Results of meta-analyses in the systematic reviews

101

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Appendix 3 0.0003) 0.03) 0.0006) NS) NS) NS) NS) = = = NS) = = = = 0.001), equivalent to 0.02) 0.0001) 0.0001) 0.001) 0.0005) 0.0011) mm (95% CI 0.13 to = < = < < < = = 0.41 (95% CI 0.21 to 0.60, p minimal JSN difference of 0.27 0.41) 0.07 (95% CI –0.20 to 0.34, p 0.24 (95% CI 0.04 to 0.43, p 0.41 (95% CI 0.21 to 0.60, p –0.61 (95% CI –0.95 to –0.28, p –0.04 (95% CI –0.17 to 0.09, p 0.46 (95% CI 0.27 to 0.66, p 0.30 (95% CI 0.11 to 0.49, p –0.51 (95% CI –0.96 to –0.05, p 1.52 (95% CI 1.20 to 1.91, p –0.06 (95% CI –0.23 to 0.11, p –0.07 (95% CI –0.21 to 0.08, p –0.15 (95% CI –0.30 to 0.00, p 0.46 (95% CI 0.28 to 0.73, p –0.12 (95% CI –0.05 to –0.19, p 9 (95% CI 6 to 20) Results of meta-analyses Results Standardised mean difference (fixed-effects model) Standardised mean difference (fixed-effects model) risk (fixed-effects model) Relative Standardised mean difference (fixed-effects model) Standardised mean difference (fixed-effects model) Standardised mean difference (fixed-effects model) Global effect size Standardised mean difference (fixed-effects model) Effect size (standardised mean difference) Standardised mean difference (random-effects model) Effect size Global effect size Standardised mean difference (random-effects model) Relative risk Relative Risk difference NNT Effect measure a n 2 7 2 5 6 5 2 1 2 15 2 2 4 2 mm) 0.5 Outcome Minimum joint space width, glucosamine vs placebo pain subscale, glucosamine vs WOMAC placebo stiffness subscale, glucosamine vs WOMAC placebo function subscale, glucosamine vs WOMAC placebo total, glucosamine vs placebo WOMAC Disease progression (joint space narrowing > Minimum joint space narrowing (JSN) (only glucosamine studies included in this category) Mean joint space width, glucosamine vs placebo pain subscale WOMAC glucosamine vs placebo (VAS), Pain function subscale WOMAC total WOMAC Lequesne index, glucosamine vs placebo 62 62 62 73 73 73 75 75 Study Towheed 2005 Towheed Pain 2005 Poolsup 2005 Towheed Function 2005 Poolsup Richy 2003 2005 Towheed Richy 2003 102 Glucosamine Structure 2005 Poolsup DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52 continued cm VAS scale cm VAS NS) cm on a 10 = 0.34 mm (95% CI 0.09 to 0.37) mm (95% CI 0.08 to 0.24) ± 0.23 1.02 (95% CI 0.93 to 1.11, p =NS) of joint space suggested, but significance Preservation unclear 0.16 –0.75 (95% CI –0.99 to –0.50), equivalent a difference in pain scores of 1.6 0.60 (95% CI 0.26 to 0.94) 0.57 (95% CI 0.26 to 0.88) 1.83 0.99 (95% CI 0.76 to 1.31) 0.98 (95% CI 0.64 to 1.52) 1.52 (95% CI 0.12 to 19.97) All studies concluded on a good to excellent tolerability 0.97 (95% CI 0.88, 1.08, p =NS) 0.82 (95% CI 0.56 to 1.21, p Results of meta-analyses Results Effect size (Cohen) Relative risk Relative risk Relative risk Relative Relative risk Relative no Data summarised narratively, meta-analysis Effect size (Cohen) Effect size (based on differences of means and pooled SDs) Effect size (Cohen) Effect size (Cohen) risk Relative Relative risk (fixed-effects model), Relative three studies with no events risk (fixed-effects model), Relative eight studies with no events Effect measure a n 5 4 9 2 2 3 5 20 6 5 7 12 11 (14) 9 (17) Outcome Mean joint space width Being a responder withdrawing because of adverse Patients events Serious adverse events Any adverse events Mean/minimum joint space width Minimum joint space width (global, on activity), WOMAC Joint pain [VAS pain subscale] pain VAS Lequesne index Tolerance Any adverse events Number of patients reporting adverse events, glucosamine vs placebo Withdrawals due to adverse events, glucosamine vs placebo 61 61 61

62 73 74 74 74 74 Bana 2006 Study Pain 2007 Reichenbach Adverse events Bana 2006 Towheed 2005 Towheed 2007 Reichenbach Function Bana 2006 Reichenbach 2007 Reichenbach Chondroitin Structure Bana 2006 Adverse events 2005 Poolsup 103

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Appendix 3 0.001) 0.001) 0.001) 0.001) 0.15) < < < < = 0.45 (95% CI 0.33 to 0.57, p 0.43 (95% CI 0.32 to 0.54, p 0.59 (95% CI 0.25 to 0.92, p 1.59 (95% CI 1.39 to 1.83, p 20% (95% CI 15% to 26%) 4.9 0.80 (95% CI 0.59 to 1.08, p Results of meta-analyses Results Global effect size Global effect size Global effect size risk Relative NNT risk Relative Absolute risk difference Effect measure No data reported a n 12 10 3 9 11 Outcome VAS pain VAS Lequesne index Mobility Being a responder (vs placebo, based on dichotomisation by investigator or global assessment) Any adverse events 75 75 75 75 75 n refers to number of studies included in the meta-analysis. Study Function Richy 2003 Richy 2003 Richy 2003 Adverse events Richy 2003 VAS, visual analogue scale. number needed to treat; NS, not significant; NNT, a 104 Glucosamine/chondroitin Structure Pain Richy 2003 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Appendix 4 Results of meta-analyses – subgroup and sensitivity analyses

105

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Appendix 4 0.0001); non-Rotta –0.15 (95% CI –0.35 0.0001); non-Rotta 0.02); non-Rotta –0.02 (95% CI –0.27 to 0.02); non-Rotta NS); non-Rotta 0.01 (95% CI –0.16 to NS); non-Rotta NS); non-Rotta 0.04 (95% CI –0.18 to NS); non-Rotta NS); non-Rotta 0.03 (95% CI –0.18 to NS); non-Rotta = = = = = 0.05) 0.03) NS) = = = NS) NS) NS) NS) 0.02) = = = = = NS) = NS) NS) NS) NS) = = = = 0.01) 0.0005) = = Results Results Standardised mean difference (fixed-effects model): 0.24 (95% CI 0.04 to 0.43, p Standardised mean difference (random-effects model): –0.19 (95% CI –0.50 to 0.11, p Standardised mean difference (random-effects model): Rotta –1.31 Standardised mean difference (random-effects model): Rotta (95% CI –1.99 to –0.64, p to 0.05, p Standardised mean difference (fixed-effects model): –0.04 (95% CI –0.17 to 0.09, p Standardised mean difference (fixed-effects model): Rotta –0.10 Standardised mean difference (fixed-effects model): (95% CI –0.29 to 0.09, p 0.17, p Standardised mean difference (random-effects model): –0.61 (95% CI –1.21 to –0.01, p Relative risk (fixed-effects model): 1.43 (95% CI 1.08 to 1.91, Relative p Standardised mean difference (random-effects model): –0.51 (95% CI –0.96 to –0.05, p risk (fixed-effects model): 1.52 (95% CI 1.20 to 1.91, Relative p Standardised mean difference (fixed-effects model): –0.06 (95% CI –0.23 to 0.11, p Standardised mean difference (fixed-effects model): Rotta –0.22 Standardised mean difference (fixed-effects model): (95% CI –0.50 to 0.06, p 0.25, p Standardised mean difference (fixed-effects model): –0.07 (95% CI –0.21 to 0.08, p Rotta –0.14 Standardised mean difference (fixed-effects model): (95% CI –0.34 to 0.05, p 0.25, p Standardised mean difference (fixed-effects model): –0.15 (95% CI –0.30 to 0.00, p Standardised mean difference (fixed-effects model): Rotta –0.23 Standardised mean difference (fixed-effects model): (95% CI –0.42 to –0.03, p 0.22, p n 2 8 7 Rotta, 8 non-Rotta 7 Rotta, 7 2 Rotta, 5 non-Rotta 2 Rotta, 3 1 4 Rotta 2 Rotta 5 1 Rotta, 4 non-Rotta 1 Rotta, 6 4 non-Rotta 2 Rotta, 5 2 Rotta, 3 non-Rotta 2 Rotta, Factor Allocation concealment adequate, glucosamine vs placebo Allocation concealment adequate, glucosamine vs placebo Glucosamine preparation Allocation concealment adequate, glucosamine vs placebo Glucosamine preparation Allocation concealment adequate, glucosamine vs placebo Allocation concealment adequate, glucosamine vs placebo Glucosamine preparation Allocation concealment adequate, glucosamine vs placebo Glucosamine preparation Allocation concealment adequate, glucosamine vs placebo Glucosamine preparation Allocation concealment adequate, glucosamine vs placebo Glucosamine preparation Outcome Minimum joint space width Pain (VAS) Pain Pain (VAS) Pain WOMAC pain WOMAC subscale WOMAC pain WOMAC subscale Lequesne index Lequesne index stiffness WOMAC subscale WOMAC stiffness WOMAC subscale WOMAC function WOMAC subscale function WOMAC subscale WOMAC total WOMAC WOMAC total WOMAC 62 Study 106 Glucosamine 2005 Towheed DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52 NS) NS) = = p p 200; –0.93 (95% CI 0.001 = No significant effect Results Results 1.08, to 0.88 CI (95% 0.97 model): (fixed-effects risk Relative Benefit of chondroitin decreased by an effect size 0.08 per year (95% CI 0.04 to 0.12), i.e. newer publications had smaller effects, p No significant difference in effect size No significant difference in effect size No significant difference in effect size No significant difference in effect size No significant difference in effect size Effect size: –0.01 (95% CI –0.12 to 0.10) for adequate allocation concealment; –0.84 (95% CI –1.08 to –0.59) for inadequate allocation concealment; p for interaction 0.05 –0.88 (95% CI Effect size: –0.03 (95% CI –0.13 to 0.07) for ITT; –1.13 to –0.64) for no ITT or unclear; p interaction 0.017 Effect size: –0.30 (95% CI –0.62 to 0.02) for similar analgesic cointervention; –0.92 (95% CI –1.26 to –0.59) for less analgesic in experimental group or unclear; p for interaction 0.043 Effect size: –0.26 (95% CI –0.56 to 0.04) for > –1.22 to –0.65) for ≤ 200; p interaction 0.022 Relative risk (fixed-effects model): 0.78 (95% CI 0.52 to 1.18, to 0.52 CI (95% 0.78 model): (fixed-effects risk Relative 6 months, 9 ≤ 200, 15 ≤ 200 n 7 20 17 yes, 3 no 12 adequate, 8 unclear or no 11 > months 1 yes, 19 unclear or no Oral 18, intramuscular 2 2 adequate, 18 unclear 3 yes, 17 no or unclear 15 less in 5 similar, experimental group or unclear 5 > 7 (9) 6 months vs ≤ 200 patients randomised vs ≤ Factor Allocation concealment adequate, glucosamine vs placebo Year of publication Year Placebo control (yes/no) blinding Patient Duration of follow-up (> 6 months) Funding by non-profit organisation Funding of administration (oral, Route intramuscular) Concealment of allocation Intention-to-treat (ITT) analysis Analgesic cointervention > patients randomised Univariable meta-regression: follow-up duration, maximum treatment dosage of chondroitin Allocation concealment adequate, glucosamine vs placebo Outcome Number of patients reporting adverse events Pain (combined Pain measures, see above) Pain Pain Pain Pain Pain Pain Pain Pain Pain Pain Number of withdrawals due to adverse events 61 NS, not significant; VAS, visual analogue scale. NS, not significant; Study Chondroitin 2007 Reichenbach 107

DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Appendix 5 Results of RCTs fulfilling inclusion criteria

Intention-to-treat values are reported where possible; all results are from end of trial unless stated otherwise.

109

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Appendix 5 p -value (between groups) 0.003 NS 0.001 0.05 0.043 NS NS NS Timing of outcome measure 3 years 18 months 3 years 3 years 3 years 18 months 18 months 18 months mm (95% CI mm (95% CI mm (95% CI mm (95% CI –0.56 to mm (95% CI –0.22 to mm (95% CI –0.22 to Change from baseline/ difference between groups 0.12 to 0.54) Change from baseline: I: –0.07 0.07) C: –0.40 –0.24) Difference: 0.33 I: +4.4 (95% CI 2.8 to 11) C: +6.3 (95% CI –3.6 to 20) I: +5.5 (95% CI –8.2 to 20.3) C: +6.3 (95% CI –1.1 to 17.5) I: 0.0 (95% CI –2.2 to 1.7) C: –0.3 (95% CI –1.8 to 1.6) I: +0.04 (95% CI –0.06 to 0.14) C: –0.19 (95% CI –0.29 to –0.09) Difference: 0.23 0.09 to 0.37) Change from baseline: I: –0.06 0.09) C: –0.31mm (95% CI –0.48 to –0.13) Difference: 0.24 0.01 to 0.48) I: +13.7 (95% CI –18.5 to 43) C: +16.3 (95% CI –4.8 to 39.8) 11 14 5 = = = End of study I: n C: n NNT mm, SD 1.24 mm, SD 1.57 mm, SD 1.29 mm, SD 1.32 mm, SD 1.48 mm, SD 1.36 mm, SD 1.0 mm, SD 1.2 Baseline I: 3.82 C: 3.95 I: 23.3, SD 5.5 C: 21.5, SD 6.6 I: 14.7, SD 5.6 C: 11.9, SD 6.1 I: 2.6 C: 3.1 I: 3.89 C: 3.63 joint space width: Total I: 5.23 C: 5.39 I: 74.2, SD 13.3 C: 69.6, SD 14.4 mm) 0.5 Outcome Severe joint space narrowing (> Minimum joint space width gait JOA (range 0–30) stairs JOA (range 0–25) Joint space width Minimum joint space width Mean joint space width total score JOA (range 0–100) 77 123 76 123 Glucosamine Structure Kawasaki 2008 Study Pain/function Kawasaki 2008 110 2002 Pavelká Reginster 2001 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52 continued p -value (between groups) NS NS NS NS NS NS NS 0.002 0.01 0.03 Timing of outcome measure 18 months 18 months 18 months 18 months 18 months 18 months 18 months 3 years 3 years 3 years Change from baseline/ difference between groups I: +1.4 (95% CI –5.5 to 5.5) C: +1.9 (95% CI –6.1 to 10) I: +2.6 (95% CI –0.3 to 5) C: +1.9 (95% CI –5.0 to 10) I: –1.6 (95% CI –5.0 to 2.2) C: –1.7 (95% CI –5.5 to 2.5) I: –14.9 (95% CI –45.5 to 6) C: –11.5 (95% CI –38.8 to 7.5) I: –3.6 (95% CI –14.1 to 3) C: –2.9 (95% CI –9.5 to 0.5) I: –1.5 (95% CI –5.1 to 1.0) C: –0.7 (95% CI –3 to 1.5) I: –9.8 (95% CI –31.2 to 5.2) C: –7.5 (95% CI –27.2 to 7) Change from baseline: I: –8.0 (95% CI –9.8 to –6.3) C: –4.9 (95% CI –6.5 to 3.2) Difference: 3.1 (95% CI 0.77 to 5.5) Change from baseline: I: –2.0 (95% CI –2.4 to –1.5) C: –1.3 (95% CI –1.7 to 0.88) Difference: 0.7 (95% CI 0.06 to 1.3) Change from baseline: I: –1.7 (95% CI –2.2 to –1.2) C: –0.82 (95% CI –1.1 to –0.51) Difference: 0.91 (95% CI 0.34 to 1.5) End of study I: 29.4, SD 5.0 C: 29.5, SD 4.8 I: 6.8, SD 2.7 C: 6.7, SD 3.3 I: 5.0, SD 2.0 C: 4.9, SD 1.8 I: 31.5, SD 15.6 C: 28.7, SD 16.0 I: 7.2, SD 3.9 C: 6.5, SD 3.4 I: 3.4. SD 1.6 C: 2.7, SD 1.6 I: 20.8, SD 11.0 C: 19.6, SD 11.5 Baseline I: 30.70, SD 14.40 C: 30.48, SD 14.43 I: 6.61, SD 3.45 C: 6.33, SD 3.13 I: 8.95, SD 2.30 C: 8.94, SD 2.27 JOA range of motion JOA (range 0–35) swelling JOA (range 0–10) pain VAS (range 0–10) total WOMAC (range 0–96) pain WOMAC (range 0–20) stiffness WOMAC (range 0–8) function WOMAC (range 0–68) Outcome WOMAC total WOMAC (points) pain WOMAC (points) Lequesne index (points) 77 Pavelká 2002 Pavelká Study 111

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Appendix 5 p -value (between groups) 0.02 0.01 0.020 No significant difference between groups (no data shown) 0.047 0.020 Timing of outcome measure 3 years 3 years Mean 5 years after trial termination (8 years from beginning of treatment) 3 years 3 years 3 years 3 years mm mm, SE 31.5 mm, SE 8.3 mm, SE 10.6 mm, SE 24.2 Change from baseline/ difference between groups Change from baseline: I: –5.8 (95% CI –7.1 to –4.4) C: –3.7 (95% CI –4.9 to –2.5) Difference: 2.1 (95% CI 0.28 to 3.9) Change from baseline: I: –0.31 (95% CI –0.55 to 0.07) C: +0.11 (95% CI –0.12 to 0.34) Difference: 0.42 (95% CI 0.09 to 0.75) Change from baseline: I: ~–160 C: ~–65.5 (estimated from graph) Change from baseline: I: ~–36.7 C: ~–7.5 (estimated from graph) Change from baseline: I: –11.7% (95% CI –20.3 to –3.2) C: +9.8% (95% CI –6.2 to 25.8) Difference: 21.6% (95% CI 3.5 to 39.6) 0.024) = End of study I: 9/144 (6.3%) C: 19/131 (14.5%) RR 0.43 (95% CI 0.20 to 0.92, p mm, SD 473.8 mm, SD 367.8 mm, SD 104.5 mm, SD 484.8 mm, SD 364.2 mm, SD 101.9 mm, SD 54.6 mm, SD 54.8 I: 21.84, SD 10.67 C: 22.00, SD 11.03 I: 2.25, SD 1.47 C: 2.15, SD 1.44 Baseline I: 96.0 C: 96.7 I: 740.1 C: 670.8 I: 194.1 C: 172.2 I: 1030.2 C: 939.7 131 144 = = WOMAC function WOMAC (points) stiffness WOMAC (points) Outcome WOMAC stiffness WOMAC WOMAC function WOMAC WOMAC pain WOMAC placebo) glucosamine, n WOMAC total WOMAC Joint replacement [in 275 patients (of 340 patients with at least 12 months’ treatment), n 52 76 Reginster Reginster 2001 Study 112 Joint replacement Reginster 2001/Pavelká 2002 (pooled study) DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52 continued p -value (between groups) No significant difference between groups; no significant difference in routine laboratory tests between groups Timing of outcome measure 18 months 3 years Change from baseline/ difference between groups 2 1 = = End of study I: n C: n I: 8 drop-outs due to adverse effects; at least one adverse event 66%; gastrointestinal tract and liver 25%; musculoskeletal 30%; cardiovascular 23%; skin and appendages 10%; respiratory tract 17%; urinary tract 12%; metabolic and nutritional 7%; other 14% C: 10 drop-outs due to adverse effects; at least one adverse event 64%; gastrointestinal tract and liver 28%; musculoskeletal 22%; cardiovascular 20%; skin and appendages 15%; respiratory tract 7%; urinary tract 11%; metabolic and nutritional 6%; other 14% Baseline Outcome Withdrawals due to adverse events 77 123 Adverse events Kawasaki 2008 Study Pavelká 2002 Pavelká 113

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Appendix 5 p -value (between groups) 0.05 No significant difference between groups; no great abnormalities in laboratory tests in both groups; no change glycaemic homeostasis Timing of outcome measure 2 years 3 years Change from baseline/ difference between groups Change from baseline: –0.03 CI (95% 0.48 SD +0.045, I: to 0.12) C: –0.07, SD 0.56 (95% CI –0.16 to 0.02) Difference between groups: 0.12, SD 0.52 (95% CI 0.00 to 0.24) End of study I: 18 drop-outs due to adverse effects; at least one adverse event 94%; abdominal pain 12%; dyspepsia 4%; diarrhoea 9%; increased BP 14%; decreased BP 2%; cardiac failure 4%; fatigue 9%; headache 6%; vertigo 7%; neuritis 4%; depressive mood 6%; allergic episode 4% C: 21 drop-outs due to adverse effects; at least one adverse event 93%; abdominal pain 17%; dyspepsia 8%; diarrhoea 10%; increased BP 14%; decreased BP 8%; cardiac failure 7%; fatigue 7%; headache 4%; vertigo 3%; neuritis 6%; depressive mood 4%; allergic episode 7% Baseline I: 2.41, SD 0.14 C: 2.35, SD 0.14 Outcome Minimum joint space width 108 76 Reginster Reginster 2001 Study 114 Chondroitin Structure Michel 2005 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52 0.001 continued = 0.05 p -value (between groups) 0.04 NS NS Significant effect on bone metabolism found: osteocalcin, pyridinoline/creatinine, and deoxypyridinoline/creatinine and keratin significantly higher, sulphate significantly lower than in placebo group, p < NS Timing of outcome measure 2 years 1 year 1 year 1 year 1 year 1 year (95% CI 2 mm (95% CI –8.61 to 2 (95% CI –3.56 to 2 mm mm Change from baseline/ difference between groups Change from baseline: I: 0.0, SD 0.53 (95% CI –0.08 to 0.09) C: –0.14, SD 0.61 (95% CI –0.24 to –0.04) Difference between groups: 0.14, SD 0.57 (95% CI 0.01 to 0.27) 3.17) C: –4.55 –0.49) Difference: 4.36 –0.19 to 8.91) Change from baseline: I: –0.19 , SD 2 , SD 2 cm cm, SD cm, SD cm cm, SD cm, SD , SD 20.9 2 , SD 26.9 2 mm 12) 1.17 12) 0.46 12) 0.36 mm 14) 1.09 14) 0.44 14) 0.35 ======I: ( n C: 58.7 0.33 C: ( n 0.36 End of study 0.14 0.13 I: ( n 0.10 C: ( n I: ( n 0.10 C: ( n I: 67.8 , SD 2 , SD 2 cm cm, SD cm, SD cm cm, SD cm, SD , SD 24.4 2 , SD 27.2 2 mm 12) 1.29 12) 0.51 12) 0.40 mm 14) 1.08 14) 0.44 14) 0.34 ======I: 3.04, SD 0.14 C: 3.00, SD 0.15 C: 63.3 0.32 C: ( n 0.28 Baseline I: 68.0 I: ( n I: ( n 0.11 C: ( n 0.10 I: ( n 0.10 C: ( n 0.10 Mean joint space width Bone metabolism Outcome Joint surface area Medial femorotibial joint surface area Medial femorotibial joint mean width Medial femorotibial joint minimum width 114 116 Uebelhart 1998 Uebelhart 2004 Study 115

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Appendix 5 0.001, NS, < = 0.05 0.001 < < 0.01 0.039 0.047 p -value (between groups) Between treatments p between times p Between treatments p between times p < NS NS NS NS 1 year 1 year Timing of outcome measure 1 year 1 year 1 year 2 years 2 years 2 years 2 years mm (95% CI mm (95% CI mm (95% CI –0.20 to mm (95% CI –0.53 to mm (95% CI –0.57 to mm (95% CI –0.23 to 0.01 to 0.55) Change from baseline: I: –0.006 0.18) C: –0.29 –0.04) Difference: 0.28 0.004 to 0.55) Change from baseline: I: –0.04 0.14) C: –0.32 –0.07) Difference: 0.27 Change from baseline/ difference between groups I: –36% C: –23% I: –11.0% C: –6.2% I: –7.8% C: –4.6% I: –0.8% C: +5.9% I: –3.9% C: +2.1% mm, SD 1.28 mm, SD 1.27 mm, SD 1.58 mm, SD 1.51 cm, SD 2.3 cm, SD 2.5 cm, SD 1.4 cm, SD 2.1 I: 4.20 C: 3.74 I: 3.61 C: 3.23 I: 8.6 C: 6.8 End of study I: 2.1 C: 4.8 I: 5.8, SD 3.6 C: 7.0, SD 3.9 mm, SD 1.47 mm, SD 1.39 cm, SD 1.10 mm, SD 1.51 mm, SD 1.46 cm, SD 1.61 cm, SD 1.6 cm, SD 1.5 I: 4.20 C: 4.03 I: 3.65 C: 3.54 I: 5.1 C: 5.7 Baseline I: 2.5, SD 1.6 C: 2.7, SD 1.8 I: 2.1, SD 1.6 C: 2.5, SD 1.8 I: 3.0, SD 2.3 C: 3.5, SD 2.5 I: 2.3, SD 1.6 C: 2.6, SD 1.7 I: 5.76 C: 6.44 I: 9.0, SD 2.8 C: 9.1, SD 3.2 Mean joint space width Minimum joint space width Overall mobility capacity (VAS) Outcome WOMAC pain WOMAC function WOMAC stiffness WOMAC WOMAC total WOMAC Huskisson VAS (0–10 cm) Lequesne index 108 114 116 Pain/function Michel 2005 Uebelhart 2004 Study 116 Uebelhart 1998 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52 continued 0.05 0.05 < < p -value (between groups) No details, stated that there was no significant difference in clinical side effects between the two groups No details, stated that there was no significant difference between the two groups All differences NS No significant differences between groups, laboratory abnormalities only observed in placebo group 1 year 1 year Timing of outcome measure 1 year 1 year 2 years 1 year I: –18% C: –0.5% I: –42% C: –25% Change from baseline/ difference between groups mm, SD 27.6 mm, SD 27.4 I: 20.1 seconds, SD 6.8 C: 22.7 seconds, SD 7.7 I: 34.3 C: 45.8 End of study I: 9 drop-outs due to adverse effects; upper respiratory tract infections 29%; headache 7%; abdominal pain 4%; allergic episode 6%; cardiac problem 6%; urinary tract infection 5% C: 9 drop-outs due to adverse effects; upper respiratory tract infections 31%; headache 9%; abdominal pain 11%; allergic episode 6%; cardiac problem 5%; urinary tract infection 5% I: 2 drop-outs due to adverse events C: 1 drop-out due to adverse events mm, SD 19.0 mm, SD 15.5 I: 24.5 seconds, SD 22.7 C: 22.8 seconds, SD 7.5 I: 58.8 C: 61.1 Baseline mm) Walking time Walking Huskisson VAS Huskisson VAS (0–100 Outcome Biological tolerability (blood, renal, liver biological parameters) Clinical tolerance 108 114 116 Adverse events Michel 2005 Study Uebelhart 2004 Uebelhart 1998 117

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Appendix 5 p -value (between groups) NS NS ≤ 0.01 NS NS Timing of outcome measure 6 months and 1 year 6 months and 1 year 1 year 6 months and 1 year 6 months and 1 year Change from baseline/ difference between groups NS versus = m, SE 7.9 m, SE 8.6 m, SE 8.0 m, SE 8.7 mm, p ≤ 0.01 mm, p 6 months: I: 176.9 N, SE 16.3 C: 202.7 N, SE 17.5 12 months: I: 207.6 N, SE 14.6 C: 209.7 N, SE 15.0 End of study baseline C: 3.52 versus baseline 6 months: I: 21.9, SE 1.1 C: 22.9, SE 1.1 12 months: I: 19.4, SE 1.2 C: 20.6, SE 1.2 6 months: I: 393.6 C: 396.5 12 months: I: 409.2 C: 410.5 I: 3.62 6 months: I: 6.2, SE 0.4 C: 6.2, SE 0.4 12 months: I: 6.0, SE 0.5 C: 5.18, SE 0.5 m, SE 17.3 m, SE 17.6 mm mm I: 209.4 N, SE 31.2 C: 163.9 N, SE 20.6 Baseline I: 25.9, SE 1.7 C: 21.1, SE 1.5 I: 384.7 C: 398.7 I: 3.66 C: 3.65 I: 7.1, SE 0.5 C: 5.9, SE 0.5 Knee concentric extension strength Outcome WOMAC function WOMAC 6-minute walk Minimum joint space width pain WOMAC 120 122 Glucosamine/chondroitin Structure 2004 Rai Study 118 Pain/function Messier 2007 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52 0.05 = 0.05 at 12 = 0.01 at 6 months, p = NS at 6 months, p months p p -value (between groups) at 12 months ≤ 0.01 6 months and 1 year 6 months and 1 year Timing of outcome measure 1 year 1 year Change from baseline/ difference between groups 6 months: I: 0.523, SE 0.014 C: 0.583, SE 0.017 12 months: I: 0.538, SE 0.017 C: 0.591, SE 0.020 6 months: I: 106.1 N, SE 7.3 C: 106.7 N, SE 7.8 12 months: I: 102.9 N, SE 7.7 C: 124.8 N, SE 8.3 End of study I: 3.7 C: 11.48 No details given but reported that the only adverse event was hair loss in one participant the intervention group I: 0.52, SE 0.04 C: 0.53, SE 0.03 I: 106.0 N, SE 16.1 C: 83.0 N, SE 10.9 Baseline I: 4.6 C: 4.9 Balance (foot length) Knee concentric flexion strength Outcome Lequesne index Adverse events 120 122 Rai 2004 Rai Study Adverse events Messier 2007 BP, blood pressure; C, control; N, Newton; NNT, number needed to treat; I, intervention; JOA, Japan Orthopaedic Association; NS, not significant; RR, relative risk; VAS, visual Japan Orthopaedic Association; NS, not significant; RR, relative risk; number needed to treat; I, intervention; JOA, blood pressure; C, control; N, Newton; NNT, BP, analogue scale. 119

DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Appendix 6 Results of RCTs fulfilling inclusion criteria – subgroups

121

© 2009 Queen’s Printer and Controller of HMSO. All rights reserved. Appendix 6 0.01 for joint p -value 0.01 0.006 < space change between the two quartiles mm, mm, mm, SD mm, SD mm: mm: 4.5 6.2 Change from baseline/difference between groups Change from baseline: I: +0.05, SD 0.53 (95% CI –0.05 to 0.14) C: –0.14, SD 0.57 (95% CI –0.25 to –0.03) Difference: 0.19, SD 0.55 (95% CI 0.04 to 0.33) Change from baseline: I: +0.01, SD 0.54 (95% CI –0.09 to 0.11) C: –0.20, SD 0.58 (95% CI –0.31to –0.09) Difference: 0.21, SD 0.56 (95% CI 0.06 to 0.36) Joint space width < I: +6.2%, SD 17.5; +0.22 0.66 C: +3.8%, SD 23.8; +0.13 SD 0.81 Joint space width > I: –6.0%, SD 15.1; –0.45 1.04 C: –14.9%, SD 17.9; –1.05 SD 1.28 End of study Baseline Minimum joint space width Mean joint space width Change in mean joint space width Outcome 27 114, = = 29 mm, n = 6.2 23 control) versus 26 control) = = mm, n 4.5 mm at entry (I: n 111) = C: n Patients with minimum joint space Patients width ≥ 1 intervention, n highest quartile (> intervention, n Mean joint space width lowest quartile (< Subgroup ) 76 213 108 Glucosamine Structure 2001 Reginster C, control; I, intervention. Chondroitin Structure Michel 2005 122 Study (Bruyere 2003 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Appendix 7 Health economics methods – informing the model

or developing the model of cost-effectiveness and/or ‘chondroitin’, combined with ‘economics’, Fof glucosamine, information from the clinical ‘costs or economics’, ‘quality of life’, ‘quality review was used as the basis for evidence of the adjusted life years’, ‘satisfaction’. This search clinical effectiveness and QoL gains. A separate retrieved 217 articles from MEDLINE(R), and structured search was carried out to identify the we also ran those key words in EMBASE for the evidence relating to the costs and health outcome period 1980 to week 6 of 2008 and retrieved 180 measures that are used for economic evaluations of papers. From the articles retrieved, the health health-care interventions for knee OA. economics research fellow read each abstract to identify whether the study was relevant for A systematic review of literature was conducted inclusion. A study was relevant if a comparison to search for evidence on cost-effectiveness and of costs and consequences (effectiveness, QoL cost of illness studies relating to OA and use of or QALY) between placebo (or usual care) and glucosamine and chondroitin. The MEDLINE glucosamine sulphate, glucosamine hydrochloride and EMBASE medical literature databases were or chondroitin (or some combination of the three) searched from 1950 to week 1 of February 2008 was undertaken and reported in the paper. No to identify relevant studies. We only considered relevant studies matched these criteria; thus, it was studies in English. not possible to undertake a critical appraisal of published studies. The following search terms were used to find relevant cost-effectiveness studies: ‘glucosamine’

123

DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Health Technology Assessment reports published to date

Volume 1, 1997 No. 11 No. 6 Newborn screening for inborn errors of Effectiveness of hip prostheses in No. 1 metabolism: a systematic review. primary total hip replacement: a critical Home parenteral nutrition: a systematic By Seymour CA, Thomason MJ, review of evidence and an economic review. Chalmers RA, Addison GM, Bain MD, model. By Richards DM, Deeks JJ, Sheldon Cockburn F, et al. By Faulkner A, Kennedy LG, Baxter TA, Shaffer JL. K, Donovan J, Wilkinson M, Bevan G. No. 12 No. 2 Routine preoperative testing: a No. 7 Diagnosis, management and screening systematic review of the evidence. Antimicrobial prophylaxis in colorectal of early localised prostate cancer. By Munro J, Booth A, Nicholl J. surgery: a systematic review of A review by Selley S, Donovan J, randomised controlled trials. Faulkner A, Coast J, Gillatt D. No. 13 By Song F, Glenny AM. Systematic review of the effectiveness of No. 3 No. 8 laxatives in the elderly. The diagnosis, management, treatment Bone marrow and peripheral By Petticrew M, Watt I, Sheldon T. and costs of prostate cancer in England blood stem cell transplantation for and Wales. malignancy. No. 14 A review by Chamberlain J, Melia J, A review by Johnson PWM, When and how to assess fast-changing Moss S, Brown J. Simnett SJ, Sweetenham JW, Morgan GJ, technologies: a comparative study of Stewart LA. No. 4 medical applications of four generic Screening for fragile X syndrome. technologies. No. 9 A review by Murray J, Cuckle H, A review by Mowatt G, Bower DJ, Screening for speech and language Taylor G, Hewison J. Brebner JA, Cairns JA, Grant AM, delay: a systematic review of the McKee L. literature. No. 5 By Law J, Boyle J, Harris F, A review of near patient testing in Harkness A, Nye C. primary care. Volume 2, 1998 By Hobbs FDR, Delaney BC, No. 10 Fitzmaurice DA, Wilson S, Hyde CJ, No. 1 Resource allocation for chronic Thorpe GH, et al. Antenatal screening for Down’s stable angina: a systematic review of effectiveness, costs and No. 6 syndrome. cost-effectiveness of alternative Systematic review of outpatient services A review by Wald NJ, Kennard A, for chronic pain control. Hackshaw A, McGuire A. interventions. By McQuay HJ, Moore RA, Eccleston By Sculpher MJ, Petticrew M, C, Morley S, de C Williams AC. No. 2 Kelland JL, Elliott RA, Holdright DR, Screening for ovarian cancer: a Buxton MJ. No. 7 systematic review. Neonatal screening for inborn errors of By Bell R, Petticrew M, Luengo S, No. 11 metabolism: cost, yield and outcome. Sheldon TA. Detection, adherence and control of A review by Pollitt RJ, Green A, hypertension for the prevention of McCabe CJ, Booth A, Cooper NJ, No. 3 stroke: a systematic review. Leonard JV, et al. Consensus development methods, By Ebrahim S. and their use in clinical guideline No. 12 No. 8 development. Postoperative analgesia and vomiting, Preschool vision screening. A review by Murphy MK, Black NA, with special reference to day-case A review by Snowdon SK, Lamping DL, McKee CM, Sanderson Stewart-Brown SL. surgery: a systematic review. CFB, Askham J, et al. By McQuay HJ, Moore RA. No. 9 No. 4 Implications of socio-cultural contexts No. 13 A cost–utility analysis of interferon beta for the ethics of clinical trials. Choosing between randomised and for multiple sclerosis. A review by Ashcroft RE, Chadwick nonrandomised studies: a systematic DW, Clark SRL, Edwards RHT, Frith L, By Parkin D, McNamee P, Jacoby A, review. Hutton JL. Miller P, Thomas S, Bates D. By Britton A, McKee M, Black N, McPherson K, Sanderson C, Bain C. No. 10 No. 5 A critical review of the role of neonatal Effectiveness and efficiency of methods No. 14 hearing screening in the detection of of dialysis therapy for end-stage renal Evaluating patient-based outcome congenital hearing impairment. disease: systematic reviews. measures for use in clinical trials. By Davis A, Bamford J, Wilson I, By MacLeod A, Grant A, Donaldson A review by Fitzpatrick R, Davey C, Ramkalawan T, Forshaw M, Wright S. C, Khan I, Campbell M, Daly C, et al. Buxton MJ, Jones DR. 125

No. 15 No. 4 No. 15 Ethical issues in the design and conduct A randomised controlled trial of Near patient testing in diabetes clinics: of randomised controlled trials. different approaches to universal appraising the costs and outcomes. antenatal HIV testing: uptake and A review by Edwards SJL, Lilford RJ, By Grieve R, Beech R, Vincent J, acceptability. Annex: Antenatal HIV Mazurkiewicz J. Braunholtz DA, Jackson JC, Hewison J, testing – assessment of a routine Thornton J. voluntary approach. By Simpson WM, Johnstone FD, No. 16 No. 16 Boyd FM, Goldberg DJ, Hart GJ, Positron emission tomography: Qualitative research methods in health Gormley SM, et al. establishing priorities for health technology assessment: a review of the technology assessment. literature. No. 5 A review by Robert G, Milne R. Methods for evaluating area-wide and By Murphy E, Dingwall R, organisation-based interventions in No. 17 (Pt 1) Greatbatch D, Parker S, Watson P. health and health care: a systematic The debridement of chronic wounds: a review. systematic review. No. 17 By Ukoumunne OC, Gulliford MC, By Bradley M, Cullum N, Sheldon T. The costs and benefits of paramedic Chinn S, Sterne JAC, Burney PGJ. skills in pre-hospital trauma care. No. 17 (Pt 2) By Nicholl J, Hughes S, Dixon S, No. 6 Assessing the costs of healthcare Systematic reviews of wound care Turner J, Yates D. technologies in clinical trials. management: (2) Dressings and topical A review by Johnston K, Buxton MJ, agents used in the healing of chronic No. 18 Jones DR, Fitzpatrick R. wounds. Systematic review of endoscopic By Bradley M, Cullum N, Nelson EA, ultrasound in gastro-oesophageal No. 7 Petticrew M, Sheldon T, Torgerson D. cancer. Cooperatives and their primary care emergency centres: organisation and By Harris KM, Kelly S, Berry E, No. 18 impact. Hutton J, Roderick P, Cullingworth J, By Hallam L, Henthorne K. A systematic literature review of et al. spiral and electron beam computed No. 8 tomography: with particular reference No. 19 Screening for cystic fibrosis. to clinical applications in hepatic Systematic reviews of trials and other A review by Murray J, Cuckle H, lesions, pulmonary embolus and studies. Taylor G, Littlewood J, Hewison J. coronary artery disease. By Sutton AJ, Abrams KR, Jones DR, By Berry E, Kelly S, Hutton J, No. 9 Sheldon TA, Song F. Harris KM, Roderick P, Boyce JC, et al. A review of the use of health status measures in economic evaluation. No. 20 By Brazier J, Deverill M, Green C, No. 19 Primary total hip replacement surgery: Harper R, Booth A. What role for statins? A review and a systematic review of outcomes economic model. and modelling of cost-effectiveness No. 10 By Ebrahim S, Davey Smith associated with different prostheses. Methods for the analysis of quality- G, McCabe C, Payne N, Pickin M, A review by Fitzpatrick R, Shortall of-life and survival data in health Sheldon TA, et al. technology assessment. E, Sculpher M, Murray D, Morris R, A review by Billingham LJ, No. 20 Lodge M, et al. Abrams KR, Jones DR. Factors that limit the quality, number No. 11 and progress of randomised controlled Volume 3, 1999 Antenatal and neonatal trials. haemoglobinopathy screening in the A review by Prescott RJ, Counsell CE, No. 1 UK: review and economic analysis. Gillespie WJ, Grant AM, Russell IT, By Zeuner D, Ades AE, Karnon J, Informed decision making: an Kiauka S, et al. Brown J, Dezateux C, Anionwu EN. annotated bibliography and systematic No. 21 review. No. 12 Antimicrobial prophylaxis in total hip By Bekker H, Thornton JG, Assessing the quality of reports of replacement: a systematic review. Airey CM, Connelly JB, Hewison J, randomised trials: implications for the By Glenny AM, Song F. Robinson MB, et al. conduct of meta-analyses. A review by Moher D, Cook DJ, No. 22 No. 2 Jadad AR, Tugwell P, Moher M, Jones A, et al. Health promoting schools and health Handling uncertainty when performing promotion in schools: two systematic economic evaluation of healthcare No. 13 reviews. interventions. ‘Early warning systems’ for identifying By Lister-Sharp D, Chapman S, A review by Briggs AH, Gray AM. new healthcare technologies. Stewart-Brown S, Sowden A. By Robert G, Stevens A, Gabbay J. No. 3 No. 23 No. 14 The role of expectancies in the placebo Economic evaluation of a primary A systematic review of the role of effect and their use in the delivery of human papillomavirus testing within a care-based education programme for health care: a systematic review. cervical screening programme. patients with osteoarthritis of the knee. By Crow R, Gage H, Hampson S, By Cuzick J, Sasieni P, Davies P, A review by Lord J, Victor C, 126 Hart J, Kimber A, Thomas H. Adams J, Normand C, Frater A, et al. Littlejohns P, Ross FM, Axford JS. DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Volume 4, 2000 No. 11 No. 21 Cost and outcome implications of the Systematic reviews of wound care No. 1 organisation of vascular services. management: (3) antimicrobial agents The estimation of marginal time By Michaels J, Brazier J, for chronic wounds; (4) diabetic foot preference in a UK-wide sample Palfreyman S, Shackley P, Slack R. ulceration. (TEMPUS) project. By O’Meara S, Cullum N, Majid M, A review by Cairns JA, No. 12 Sheldon T. van der Pol MM. Monitoring blood glucose control in diabetes mellitus: a systematic review. No. 22 No. 2 By Coster S, Gulliford MC, Seed PT, Using routine data to complement Geriatric rehabilitation following Powrie JK, Swaminathan R. and enhance the results of randomised fractures in older people: a systematic controlled trials. By Lewsey JD, Leyland AH, Murray review. No. 13 GD, Boddy FA. By Cameron I, Crotty M, Currie C, The effectiveness of domiciliary Finnegan T, Gillespie L, Gillespie W, health visiting: a systematic review of international studies and a selective No. 23 et al. review of the British literature. Coronary artery stents in the treatment of ischaemic heart disease: a rapid and No. 3 By Elkan R, Kendrick D, Hewitt M, Robinson JJA, Tolley K, Blair M, et al. systematic review. Screening for sickle cell disease and By Meads C, Cummins C, Jolly K, thalassaemia: a systematic review with No. 14 Stevens A, Burls A, Hyde C. supplementary research. The determinants of screening uptake By Davies SC, Cronin E, Gill M, and interventions for increasing No. 24 Greengross P, Hickman M, Normand C. uptake: a systematic review. Outcome measures for adult critical By Jepson R, Clegg A, Forbes C, care: a systematic review. No. 4 Lewis R, Sowden A, Kleijnen J. By Hayes JA, Black NA, Jenkinson C, Community provision of hearing aids Young JD, Rowan KM, Daly K, et al. and related audiology services. No. 15 A review by Reeves DJ, Alborz A, The effectiveness and cost-effectiveness No. 25 Hickson FS, Bamford JM. of prophylactic removal of wisdom A systematic review to evaluate the teeth. effectiveness of interventions to No. 5 A rapid review by Song F, O’Meara S, promote the initiation of breastfeeding. By Fairbank L, O’Meara S, False-negative results in screening Wilson P, Golder S, Kleijnen J. programmes: systematic review of Renfrew MJ, Woolridge M, Sowden AJ, impact and implications. No. 16 Lister-Sharp D. By Petticrew MP, Sowden AJ, Ultrasound screening in pregnancy: No. 26 Lister-Sharp D, Wright K. a systematic review of the clinical Implantable cardioverter defibrillators: effectiveness, cost-effectiveness and arrhythmias. A rapid and systematic No. 6 women’s views. review. By Bricker L, Garcia J, Henderson J, Costs and benefits of community By Parkes J, Bryant J, Milne R. postnatal support workers: a Mugford M, Neilson J, Roberts T, et al. randomised controlled trial. No. 27 No. 17 By Morrell CJ, Spiby H, Stewart P, Treatments for fatigue in multiple Walters S, Morgan A. A rapid and systematic review of the sclerosis: a rapid and systematic review. effectiveness and cost-effectiveness of By Brañas P, Jordan R, Fry-Smith A, No. 7 the taxanes used in the treatment of Burls A, Hyde C. Implantable contraceptives (subdermal advanced breast and ovarian cancer. implants and hormonally impregnated By Lister-Sharp D, McDonagh MS, No. 28 intrauterine systems) versus other Khan KS, Kleijnen J. Early asthma prophylaxis, natural forms of reversible contraceptives: two history, skeletal development and No. 18 systematic reviews to assess relative economy (EASE): a pilot randomised Liquid-based cytology in cervical effectiveness, acceptability, tolerability controlled trial. screening: a rapid and systematic and cost-effectiveness. By Baxter-Jones ADG, Helms PJ, review. By French RS, Cowan FM, Russell G, Grant A, Ross S, Cairns JA, By Payne N, Chilcott J, McGoogan E. Mansour DJA, Morris S, Procter T, et al. Hughes D, et al. No. 19 No. 29 Randomised controlled trial of non- No. 8 Screening for hypercholesterolaemia directive counselling, cognitive– versus case finding for familial An introduction to statistical methods behaviour therapy and usual general hypercholesterolaemia: a systematic for health technology assessment. practitioner care in the management of review and cost-effectiveness analysis. A review by White SJ, Ashby D, depression as well as mixed anxiety and By Marks D, Wonderling Brown PJ. depression in primary care. D, Thorogood M, Lambert H, By King M, Sibbald B, Ward E, Humphries SE, Neil HAW. No. 9 Bower P, Lloyd M, Gabbay M, et al. Disease-modifying drugs for multiple No. 30 sclerosis: a rapid and systematic review. No. 20 A rapid and systematic review of By Clegg A, Bryant J, Milne R. Routine referral for radiography of the clinical effectiveness and cost- patients presenting with low back pain: effectiveness of glycoprotein IIb/IIIa No. 10 is patients’ outcome influenced by GPs’ antagonists in the medical management Publication and related biases. referral for plain radiography? of unstable angina. A review by Song F, Eastwood AJ, By Kerry S, Hilton S, Patel S, By McDonagh MS, Bachmann LM, Gilbody S, Duley L, Sutton AJ. Dundas D, Rink E, Lord J. Golder S, Kleijnen J, ter Riet G. 127

No. 31 Volume 5, 2001 No. 11 A randomised controlled trial Effectiveness of autologous chondrocyte of prehospital intravenous fluid No. 1 transplantation for hyaline cartilage replacement therapy in serious trauma. Clinical and cost-effectiveness defects in knees: a rapid and systematic By Turner J, Nicholl J, Webber L, of donepezil, rivastigmine and review. Cox H, Dixon S, Yates D. galantamine for Alzheimer’s disease: a By Jobanputra P, Parry D, Fry-Smith rapid and systematic review. A, Burls A. No. 32 By Clegg A, Bryant J, Nicholson T, No. 12 Intrathecal pumps for giving opioids in McIntyre L, De Broe S, Gerard K, et al. Statistical assessment of the learning chronic pain: a systematic review. curves of health technologies. By Williams JE, Louw G, No. 2 The clinical effectiveness and cost- By Ramsay CR, Grant AM, Wallace Towlerton G. SA, Garthwaite PH, Monk AF, Russell IT. effectiveness of riluzole for motor neurone disease: a rapid and systematic No. 33 No. 13 Combination therapy (interferon review. The effectiveness and cost-effectiveness alfa and ribavirin) in the treatment By Stewart A, Sandercock J, Bryan S, of temozolomide for the treatment of of chronic hepatitis C: a rapid and Hyde C, Barton PM, Fry-Smith A, et al. recurrent malignant glioma: a rapid systematic review. and systematic review. By Shepherd J, Waugh N, No. 3 By Dinnes J, Cave C, Huang S, Hewitson P. Equity and the economic evaluation of Major K, Milne R. healthcare. No. 34 By Sassi F, Archard L, Le Grand J. No. 14 A systematic review of comparisons of A rapid and systematic review of effect sizes derived from randomised No. 4 the clinical effectiveness and cost- and non-randomised studies. Quality-of-life measures in chronic effectiveness of debriding agents in treating surgical wounds healing by By MacLehose RR, Reeves BC, diseases of childhood. By Eiser C, Morse R. secondary intention. Harvey IM, Sheldon TA, Russell IT, By Lewis R, Whiting P, ter Riet G, Black AMS. No. 5 O’Meara S, Glanville J. Eliciting public preferences for No. 35 healthcare: a systematic review of No. 15 Intravascular ultrasound-guided techniques. Home treatment for mental health interventions in coronary artery By Ryan M, Scott DA, Reeves C, Bate problems: a systematic review. disease: a systematic literature review, A, van Teijlingen ER, Russell EM, et al. By Burns T, Knapp M, Catty J, with decision-analytic modelling, of Healey A, Henderson J, Watt H, et al. outcomes and cost-effectiveness. No. 6 No. 16 By Berry E, Kelly S, Hutton J, General health status measures for How to develop cost-conscious Lindsay HSJ, Blaxill JM, Evans JA, et al. people with cognitive impairment: guidelines. learning disability and acquired brain No. 36 By Eccles M, Mason J. injury. A randomised controlled trial to By Riemsma RP, Forbes CA, No. 17 evaluate the effectiveness and cost- Glanville JM, Eastwood AJ, Kleijnen J. effectiveness of counselling patients The role of specialist nurses in multiple sclerosis: a rapid and systematic review. with chronic depression. No. 7 By De Broe S, Christopher F, By Simpson S, Corney R, An assessment of screening strategies Waugh N. Fitzgerald P, Beecham J. for fragile X syndrome in the UK. By Pembrey ME, Barnicoat AJ, No. 18 No. 37 Carmichael B, Bobrow M, Turner G. A rapid and systematic review Systematic review of treatments for of the clinical effectiveness and atopic eczema. No. 8 cost-effectiveness of orlistat in the By Hoare C, Li Wan Po A, Issues in methodological research: management of obesity. Williams H. perspectives from researchers and By O’Meara S, Riemsma R, commissioners. Shirran L, Mather L, ter Riet G. No. 38 By Lilford RJ, Richardson A, Stevens Bayesian methods in health technology A, Fitzpatrick R, Edwards S, Rock F, et al. No. 19 assessment: a review. The clinical effectiveness and cost- By Spiegelhalter DJ, Myles JP, No. 9 effectiveness of pioglitazone for Jones DR, Abrams KR. Systematic reviews of wound type 2 diabetes mellitus: a rapid and care management: (5) beds; systematic review. By Chilcott J, Wight J, Lloyd Jones No. 39 (6) compression; (7) laser therapy, M, Tappenden P. The management of dyspepsia: a therapeutic ultrasound, electrotherapy systematic review. and electromagnetic therapy. No. 20 By Delaney B, Moayyedi P, Deeks J, By Cullum N, Nelson EA, Extended scope of nursing practice: Innes M, Soo S, Barton P, et al. Flemming K, Sheldon T. a multicentre randomised controlled trial of appropriately trained nurses No. 40 No. 10 and preregistration house officers in A systematic review of treatments for Effects of educational and psychosocial preoperative assessment in elective severe psoriasis. interventions for adolescents with general surgery. By Griffiths CEM, Clark CM, diabetes mellitus: a systematic review. By Kinley H, Czoski-Murray C, Chalmers RJG, Li Wan Po A, By Hampson SE, Skinner TC, Hart J, George S, McCabe C, Primrose J, 128 Williams HC. Storey L, Gage H, Foxcroft D, et al. Reilly C, et al. DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

No. 21 No. 31 No. 5 Systematic reviews of the effectiveness Design and use of questionnaires: a The clinical effectiveness and cost- of day care for people with severe review of best practice applicable to effectiveness of inhaler devices used mental disorders: (1) Acute day hospital surveys of health service staff and in the routine management of chronic versus admission; (2) Vocational patients. asthma in older children: a systematic rehabilitation; (3) Day hospital versus By McColl E, Jacoby A, Thomas L, review and economic evaluation. outpatient care. Soutter J, Bamford C, Steen N, et al. By Peters J, Stevenson M, Beverley C, By Marshall M, Crowther R, Lim J, Smith S. Almaraz- Serrano A, Creed F, Sledge W, No. 32 Kluiter H, et al. A rapid and systematic review of No. 6 the clinical effectiveness and cost- The clinical effectiveness and cost- No. 22 effectiveness of paclitaxel, docetaxel, effectiveness of sibutramine in the The measurement and monitoring of gemcitabine and vinorelbine in non- management of obesity: a technology surgical adverse events. small-cell lung cancer. assessment. By Bruce J, Russell EM, Mollison J, By Clegg A, Scott DA, Sidhu M, By O’Meara S, Riemsma R, Shirran Krukowski ZH. Hewitson P, Waugh N. L, Mather L, ter Riet G.

No. 23 No. 33 No. 7 Action research: a systematic review and Subgroup analyses in randomised The cost-effectiveness of magnetic guidance for assessment. controlled trials: quantifying the risks resonance angiography for carotid By Waterman H, Tillen D, Dickson R, of false-positives and false-negatives. artery stenosis and peripheral vascular de Koning K. By Brookes ST, Whitley E, Peters TJ, disease: a systematic review. Mulheran PA, Egger M, Davey Smith G. By Berry E, Kelly S, Westwood ME, No. 24 Davies LM, Gough MJ, Bamford JM, A rapid and systematic review of No. 34 et al. the clinical effectiveness and cost- Depot antipsychotic medication effectiveness of gemcitabine for the in the treatment of patients with No. 8 treatment of pancreatic cancer. schizophrenia: (1) Meta-review; (2) Promoting physical activity in South By Ward S, Morris E, Bansback N, Patient and nurse attitudes. Asian Muslim women through ‘exercise Calvert N, Crellin A, Forman D, et al. By David AS, Adams C. on prescription’. By Carroll B, Ali N, Azam N. No. 25 No. 35 No. 9 A rapid and systematic review of the A systematic review of controlled Zanamivir for the treatment of evidence for the clinical effectiveness trials of the effectiveness and cost- influenza in adults: a systematic review and cost-effectiveness of irinotecan, effectiveness of brief psychological and economic evaluation. oxaliplatin and raltitrexed for the treatments for depression. By Burls A, Clark W, Stewart T, treatment of advanced colorectal By Churchill R, Hunot V, Corney R, Preston C, Bryan S, Jefferson T, et al. cancer. Knapp M, McGuire H, Tylee A, et al. By Lloyd Jones M, Hummel S, No. 10 Bansback N, Orr B, Seymour M. No. 36 Cost analysis of child health A review of the natural history and surveillance. epidemiology of multiple sclerosis: No. 26 implications for resource allocation and Comparison of the effectiveness of By Sanderson D, Wright D, Acton C, Duree D. health economic models. inhaler devices in asthma and chronic By Richards RG, Sampson FC, obstructive airways disease: a systematic Beard SM, Tappenden P. review of the literature. By Brocklebank D, Ram F, Wright J, Volume 6, 2002 No. 11 Barry P, Cates C, Davies L, et al. Screening for gestational diabetes: No. 1 a systematic review and economic No. 27 A study of the methods used to select evaluation. The cost-effectiveness of magnetic review criteria for clinical audit. By Scott DA, Loveman E, McIntyre resonance imaging for investigation of By Hearnshaw H, Harker R, L, Waugh N. the knee joint. Cheater F, Baker R, Grimshaw G. By Bryan S, Weatherburn G, Bungay No. 12 H, Hatrick C, Salas C, Parry D, et al. No. 2 The clinical effectiveness and cost- Fludarabine as second-line therapy for effectiveness of surgery for people with No. 28 B cell chronic lymphocytic leukaemia: a morbid obesity: a systematic review and A rapid and systematic review of technology assessment. economic evaluation. the clinical effectiveness and cost- By Hyde C, Wake B, Bryan S, Barton By Clegg AJ, Colquitt J, Sidhu MK, effectiveness of topotecan for ovarian P, Fry-Smith A, Davenport C, et al. Royle P, Loveman E, Walker A. cancer. By Forbes C, Shirran L, Bagnall A-M, No. 3 No. 13 Duffy S, ter Riet G. Rituximab as third-line treatment for The clinical effectiveness of refractory or recurrent Stage III or IV trastuzumab for breast cancer: a No. 29 follicular non-Hodgkin’s lymphoma: systematic review. Superseded by a report published in a a systematic review and economic By Lewis R, Bagnall A-M, Forbes C, later volume. evaluation. Shirran E, Duffy S, Kleijnen J, et al. By Wake B, Hyde C, Bryan S, Barton No. 30 P, Song F, Fry-Smith A, et al. No. 14 The role of radiography in primary The clinical effectiveness and cost- care patients with low back pain of at No. 4 effectiveness of vinorelbine for breast least 6 weeks duration: a randomised A systematic review of discharge cancer: a systematic review and (unblinded) controlled trial. arrangements for older people. economic evaluation. By Kendrick D, Fielding K, Bentley By Parker SG, Peet SM, McPherson By Lewis R, Bagnall A-M, King S, E, Miller P, Kerslake R, Pringle M. A, Cannaby AM, Baker R, Wilson A, et al. Woolacott N, Forbes C, Shirran L, et al. 129

No. 15 No. 24 No. 33 A systematic review of the effectiveness A systematic review of the effectiveness The effectiveness and cost-effectiveness and cost-effectiveness of metal-on- of interventions based on a stages-of- of imatinib in chronic myeloid metal hip resurfacing arthroplasty for change approach to promote individual leukaemia: a systematic review. treatment of hip disease. behaviour change. By Garside R, Round A, Dalziel K, By Vale L, Wyness L, McCormack K, By Riemsma RP, Pattenden J, Bridle Stein K, Royle R. McKenzie L, Brazzelli M, Stearns SC. C, Sowden AJ, Mather L, Watt IS, et al. No. 34 No. 16 No. 25 A comparative study of hypertonic The clinical effectiveness and cost- A systematic review update of the saline, daily and alternate-day rhDNase effectiveness of bupropion and nicotine clinical effectiveness and cost- in children with cystic fibrosis. replacement therapy for smoking effectiveness of glycoprotein IIb/IIIa By Suri R, Wallis C, Bush A, cessation: a systematic review and antagonists. Thompson S, Normand C, Flather M, economic evaluation. By Robinson M, Ginnelly L, Sculpher et al. By Woolacott NF, Jones L, Forbes CA, M, Jones L, Riemsma R, Palmer S, et al. Mather LC, Sowden AJ, Song FJ, et al. No. 35 No. 26 A systematic review of the costs and No. 17 A systematic review of the effectiveness, effectiveness of different models of A systematic review of effectiveness cost-effectiveness and barriers to paediatric home care. and economic evaluation of new drug implementation of thrombolytic and By Parker G, Bhakta P, Lovett CA, treatments for juvenile idiopathic neuroprotective therapy for acute Paisley S, Olsen R, Turner D, et al. arthritis: etanercept. ischaemic stroke in the NHS. By Cummins C, Connock M, By Sandercock P, Berge E, Dennis M, Fry-Smith A, Burls A. Forbes J, Hand P, Kwan J, et al. Volume 7, 2003

No. 18 No. 27 No. 1 Clinical effectiveness and cost- A randomised controlled crossover trial How important are comprehensive effectiveness of growth hormone in of nurse practitioner versus doctor- literature searches and the assessment children: a systematic review and led outpatient care in a bronchiectasis of trial quality in systematic reviews? economic evaluation. clinic. Empirical study. By Bryant J, Cave C, Mihaylova B, By Caine N, Sharples LD, By Egger M, Juni P, Bartlett C, Hollingworth W, French J, Keogan M, Chase D, McIntyre L, Gerard K, et al. Holenstein F, Sterne J. Exley A, et al. No. 19 No. 2 No. 28 Clinical effectiveness and cost- Systematic review of the effectiveness Clinical effectiveness and cost – effectiveness of growth hormone and cost-effectiveness, and economic consequences of selective serotonin in adults in relation to impact on evaluation, of home versus hospital or reuptake inhibitors in the treatment of quality of life: a systematic review and satellite unit haemodialysis for people sex offenders. economic evaluation. with end-stage renal failure. By Adi Y, Ashcroft D, Browne K, By Bryant J, Loveman E, Chase D, By Mowatt G, Vale L, Perez J, Wyness Beech A, Fry-Smith A, Hyde C. Mihaylova B, Cave C, Gerard K, et al. L, Fraser C, MacLeod A, et al. No. 29 No. 3 No. 20 Treatment of established osteoporosis: Systematic review and economic Clinical medication review by a a systematic review and cost–utility evaluation of the effectiveness of pharmacist of patients on repeat analysis. infliximab for the treatment of Crohn’s prescriptions in general practice: a By Kanis JA, Brazier JE, Stevenson disease. randomised controlled trial. M, Calvert NW, Lloyd Jones M. By Clark W, Raftery J, Barton P, By Zermansky AG, Petty DR, Raynor Song F, Fry-Smith A, Burls A. DK, Lowe CJ, Freementle N, Vail A. No. 30 Which anaesthetic agents are cost- No. 4 No. 21 effective in day surgery? Literature A review of the clinical effectiveness The effectiveness of infliximab and review, national survey of practice and and cost-effectiveness of routine anti-D etanercept for the treatment of randomised controlled trial. prophylaxis for pregnant women who rheumatoid arthritis: a systematic By Elliott RA Payne K, Moore JK, are rhesus negative. review and economic evaluation. Davies LM, Harper NJN, St Leger AS, By Chilcott J, Lloyd Jones M, Wight By Jobanputra P, Barton P, Bryan S, et al. J, Forman K, Wray J, Beverley C, et al. Burls A. No. 31 No. 5 No. 22 Screening for hepatitis C among Systematic review and evaluation of the A systematic review and economic injecting drug users and in use of tumour markers in paediatric evaluation of computerised cognitive genitourinary medicine clinics: oncology: Ewing’s sarcoma and behaviour therapy for depression and systematic reviews of effectiveness, neuroblastoma. anxiety. modelling study and national survey of By Riley RD, Burchill SA, By Kaltenthaler E, Shackley P, current practice. Abrams KR, Heney D, Lambert PC, Stevens K, Beverley C, Parry G, By Stein K, Dalziel K, Walker A, Jones DR, et al. Chilcott J. McIntyre L, Jenkins B, Horne J, et al. No. 6 No. 23 No. 32 The cost-effectiveness of screening for A systematic review and economic The measurement of satisfaction with Helicobacter pylori to reduce mortality evaluation of pegylated liposomal healthcare: implications for practice and morbidity from gastric cancer and doxorubicin hydrochloride for ovarian from a systematic review of the peptic ulcer disease: a discrete-event cancer. literature. simulation model. By Forbes C, Wilby J, Richardson G, By Crow R, Gage H, Hampson S, By Roderick P, Davies R, Raftery J, 130 Sculpher M, Mather L, Reimsma R. Hart J, Kimber A, Storey L, et al. Crabbe D, Pearce R, Bhandari P, et al. DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

No. 7 No. 16 No. 26 The clinical effectiveness and cost- Screening for fragile X syndrome: a Can randomised trials rely on existing effectiveness of routine dental checks: literature review and modelling. electronic data? A feasibility study to a systematic review and economic By Song FJ, Barton P, Sleightholme explore the value of routine data in evaluation. V, Yao GL, Fry-Smith A. health technology assessment. By Davenport C, Elley K, Salas By Williams JG, Cheung WY, C, Taylor-Weetman CL, Fry-Smith A, No. 17 Cohen DR, Hutchings HA, Longo MF, Bryan S, et al. Systematic review of endoscopic sinus Russell IT. surgery for nasal polyps. No. 8 No. 27 By Dalziel K, Stein K, Round A, A multicentre randomised controlled Evaluating non-randomised Garside R, Royle P. trial assessing the costs and benefits intervention studies. of using structured information and By Deeks JJ, Dinnes J, D’Amico R, No. 18 analysis of women’s preferences in the Sowden AJ, Sakarovitch C, Song F, et al. management of menorrhagia. Towards efficient guidelines: how to By Kennedy ADM, Sculpher MJ, monitor guideline use in primary care. No. 28 Coulter A, Dwyer N, Rees M, Horsley S, By Hutchinson A, McIntosh A, A randomised controlled trial to assess et al. Cox S, Gilbert C. the impact of a package comprising a patient-orientated, evidence-based self- No. 9 No. 19 help guidebook and patient-centred Clinical effectiveness and cost–utility Effectiveness and cost-effectiveness consultations on disease management of photodynamic therapy for wet of acute hospital-based spinal cord and satisfaction in inflammatory bowel age-related macular degeneration: injuries services: systematic review. disease. a systematic review and economic By Bagnall A-M, Jones L, Richardson By Kennedy A, Nelson E, Reeves D, evaluation. G, Duffy S, Riemsma R. Richardson G, Roberts C, Robinson A, By Meads C, Salas C, Roberts T, et al. Moore D, Fry-Smith A, Hyde C. No. 20 Prioritisation of health technology No. 29 No. 10 assessment. The PATHS model: The effectiveness of diagnostic tests for Evaluation of molecular tests for the assessment of shoulder pain due prenatal diagnosis of chromosome methods and case studies. to soft tissue disorders: a systematic abnormalities. By Townsend J, Buxton M, review. By Grimshaw GM, Szczepura A, Harper G. By Dinnes J, Loveman E, McIntyre L, Hultén M, MacDonald F, Nevin NC, Waugh N. Sutton F, et al. No. 21 Systematic review of the clinical No. 30 No. 11 effectiveness and cost-effectiveness of The value of digital imaging in diabetic First and second trimester antenatal tension-free vaginal tape for treatment retinopathy. screening for Down’s syndrome: of urinary stress incontinence. By Sharp PF, Olson J, Strachan F, the results of the Serum, Urine and By Cody J, Wyness L, Wallace S, Hipwell J, Ludbrook A, O’Donnell M, Ultrasound Screening Study (SURUSS). Glazener C, Kilonzo M, Stearns S, et al. By Wald NJ, Rodeck C, Hackshaw et al. AK, Walters J, Chitty L, Mackinson AM. No. 22 No. 31 The clinical and cost-effectiveness of No. 12 Lowering blood pressure to prevent patient education models for diabetes: The effectiveness and cost-effectiveness myocardial infarction and stroke: a new a systematic review and economic of ultrasound locating devices for preventive strategy. central venous access: a systematic evaluation. By Law M, Wald N, Morris J. review and economic evaluation. By Loveman E, Cave C, Green C, By Calvert N, Hind D, McWilliams Royle P, Dunn N, Waugh N. No. 32 RG, Thomas SM, Beverley C, Clinical and cost-effectiveness of Davidson A. No. 23 capecitabine and tegafur with uracil for The role of modelling in prioritising the treatment of metastatic colorectal No. 13 and planning clinical trials. cancer: systematic review and economic A systematic review of atypical By Chilcott J, Brennan A, Booth A, evaluation. antipsychotics in schizophrenia. Karnon J, Tappenden P. By Ward S, Kaltenthaler E, Cowan J, By Bagnall A-M, Jones L, Lewis R, Brewer N. Ginnelly L, Glanville J, Torgerson D, No. 24 et al. Cost–benefit evaluation of routine No. 33 Clinical and cost-effectiveness of new influenza immunisation in people and emerging technologies for early No. 14 65–74 years of age. localised prostate cancer: a systematic Prostate Testing for Cancer and By Allsup S, Gosney M, Haycox A, Treatment (ProtecT) feasibility study. review. Regan M. By Donovan J, Hamdy F, Neal D, By Hummel S, Paisley S, Morgan A, Peters T, Oliver S, Brindle L, et al. Currie E, Brewer N. No. 25 The clinical and cost-effectiveness of No. 15 No. 34 Early thrombolysis for the treatment pulsatile machine perfusion versus cold Literature searching for clinical and of acute myocardial infarction: a storage of kidneys for transplantation cost-effectiveness studies used in health systematic review and economic retrieved from heart-beating and non- technology assessment reports carried evaluation. heart-beating donors. out for the National Institute for By Boland A, Dundar Y, Bagust A, By Wight J, Chilcott J, Holmes M, Clinical Excellence appraisal system. Haycox A, Hill R, Mujica Mota R, et al. Brewer N. By Royle P, Waugh N. 131

No. 35 No. 2 No. 11 Systematic review and economic Systematic review and modelling of the The use of modelling to evaluate decision modelling for the prevention investigation of acute and chronic chest new drugs for patients with a chronic and treatment of influenza A and B. pain presenting in primary care. condition: the case of antibodies By Turner D, Wailoo A, Nicholson K, By Mant J, McManus RJ, Oakes RAL, against tumour necrosis factor in Cooper N, Sutton A, Abrams K. Delaney BC, Barton PM, Deeks JJ, et al. rheumatoid arthritis. By Barton P, Jobanputra P, Wilson J, No. 36 No. 3 Bryan S, Burls A. A randomised controlled trial The effectiveness and cost-effectiveness No. 12 to evaluate the clinical and cost- of microwave and thermal balloon Clinical effectiveness and cost- effectiveness of Hickman line insertions endometrial ablation for heavy effectiveness of neonatal screening in adult cancer patients by nurses. menstrual bleeding: a systematic review for inborn errors of metabolism using By Boland A, Haycox A, Bagust A, and economic modelling. tandem mass spectrometry: a systematic Fitzsimmons L. By Garside R, Stein K, Wyatt K, review. Round A, Price A. By Pandor A, Eastham J, Beverley C, No. 37 Chilcott J, Paisley S. Redesigning postnatal care: a No. 4 randomised controlled trial of protocol- A systematic review of the role of No. 13 based midwifery-led care focused bisphosphonates in metastatic disease. Clinical effectiveness and cost- on individual women’s physical and By Ross JR, Saunders Y, effectiveness of pioglitazone and psychological health needs. Edmonds PM, Patel S, Wonderling D, rosiglitazone in the treatment of type By MacArthur C, Winter HR, Normand C, et al. 2 diabetes: a systematic review and Bick DE, Lilford RJ, Lancashire RJ, economic evaluation. Knowles H, et al. No. 5 By Czoski-Murray C, Warren E, Systematic review of the clinical Chilcott J, Beverley C, Psyllaki MA, No. 38 effectiveness and cost-effectiveness Cowan J. Estimating implied rates of discount in of capecitabine (Xeloda®) for locally No. 14 healthcare decision-making. advanced and/or metastatic breast Routine examination of the newborn: By West RR, McNabb R, Thompson cancer. the EMREN study. Evaluation of an AGH, Sheldon TA, Grimley Evans J. By Jones L, Hawkins N, Westwood M, extension of the midwife role including Wright K, Richardson G, Riemsma R. a randomised controlled trial of No. 39 appropriately trained midwives and Systematic review of isolation policies No. 6 paediatric senior house officers. in the hospital management of Effectiveness and efficiency of guideline By Townsend J, Wolke D, Hayes J, methicillin-resistant Staphylococcus dissemination and implementation Davé S, Rogers C, Bloomfield L, et al. aureus: a review of the literature strategies. with epidemiological and economic By Grimshaw JM, Thomas RE, No. 15 modelling. MacLennan G, Fraser C, Ramsay CR, Involving consumers in research and By Cooper BS, Stone SP, Kibbler CC, Vale L, et al. development agenda setting for the Cookson BD, Roberts JA, Medley GF, NHS: developing an evidence-based et al. No. 7 approach. Clinical effectiveness and costs of the By Oliver S, Clarke-Jones L, Rees R, No. 40 Sugarbaker procedure for the treatment Milne R, Buchanan P, Gabbay J, et al. Treatments for spasticity and pain in of pseudomyxoma peritonei. No. 16 multiple sclerosis: a systematic review. By Bryant J, Clegg AJ, Sidhu MK, A multi-centre randomised controlled By Beard S, Hunn A, Wight J. Brodin H, Royle P, Davidson P. trial of minimally invasive direct coronary bypass grafting versus No. 41 No. 8 percutaneous transluminal coronary The inclusion of reports of randomised Psychological treatment for insomnia angioplasty with stenting for proximal trials published in languages other than in the regulation of long-term hypnotic stenosis of the left anterior descending English in systematic reviews. drug use. coronary artery. By Moher D, Pham B, Lawson ML, By Morgan K, Dixon S, Mathers N, By Reeves BC, Angelini GD, Bryan Klassen TP. Thompson J, Tomeny M. AJ, Taylor FC, Cripps T, Spyt TJ, et al.

No. 42 No. 9 No. 17 The impact of screening on future Improving the evaluation of Does early magnetic resonance imaging health-promoting behaviours and therapeutic interventions in multiple influence management or improve health beliefs: a systematic review. sclerosis: development of a patient- outcome in patients referred to By Bankhead CR, Brett J, Bukach C, based measure of outcome. secondary care with low back pain? A Webster P, Stewart-Brown S, Munafo M, By Hobart JC, Riazi A, Lamping DL, pragmatic randomised controlled trial. et al. Fitzpatrick R, Thompson AJ. By Gilbert FJ, Grant AM, Gillan MGC, Vale L, Scott NW, Campbell MK, et al. No. 10 A systematic review and economic Volume 8, 2004 No. 18 evaluation of magnetic resonance The clinical and cost-effectiveness No. 1 cholangiopancreatography compared of anakinra for the treatment of What is the best imaging strategy for with diagnostic endoscopic retrograde rheumatoid arthritis in adults: a acute stroke? cholangiopancreatography. systematic review and economic By Wardlaw JM, Keir SL, Seymour J, By Kaltenthaler E, Bravo Vergel Y, analysis. Lewis S, Sandercock PAG, Dennis MS, Chilcott J, Thomas S, Blakeborough T, By Clark W, Jobanputra P, Barton P, 132 et al. Walters SJ, et al. Burls A. DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

No. 19 No. 28 No. 37 A rapid and systematic review and Effectiveness and cost-effectiveness Rituximab (MabThera®) for aggressive economic evaluation of the clinical of imatinib for first-line treatment non-Hodgkin’s lymphoma: systematic and cost-effectiveness of newer drugs of chronic myeloid leukaemia in review and economic evaluation. for treatment of mania associated with chronic phase: a systematic review and By Knight C, Hind D, Brewer N, bipolar affective disorder. economic analysis. Abbott V. By Bridle C, Palmer S, Bagnall A-M, By Dalziel K, Round A, Stein K, Darba J, Duffy S, Sculpher M, et al. Garside R, Price A. No. 38 Clinical effectiveness and cost- No. 29 No. 20 effectiveness of clopidogrel and Liquid-based cytology in cervical VenUS I: a randomised controlled trial modified-release dipyridamole in the screening: an updated rapid and of two types of bandage for treating secondary prevention of occlusive systematic review and economic venous leg ulcers. vascular events: a systematic review and analysis. By Iglesias C, Nelson EA, Cullum By Karnon J, Peters J, Platt J, NA, Torgerson DJ, on behalf of the economic evaluation. Chilcott J, McGoogan E, Brewer N. VenUS Team. By Jones L, Griffin S, Palmer S, Main C, Orton V, Sculpher M, et al. No. 21 No. 30 Systematic review of the long-term Systematic review of the effectiveness No. 39 effects and economic consequences of and cost-effectiveness, and economic Pegylated interferon α-2a and -2b treatments for obesity and implications evaluation, of myocardial perfusion in combination with ribavirin in the for health improvement. scintigraphy for the diagnosis and treatment of chronic hepatitis C: By Avenell A, Broom J, Brown TJ, management of angina and myocardial a systematic review and economic Poobalan A, Aucott L, Stearns SC, et al. infarction. evaluation. By Mowatt G, Vale L, Brazzelli M, By Shepherd J, Brodin H, Cave C, Hernandez R, Murray A, Scott N, et al. No. 22 Waugh N, Price A, Gabbay J. Autoantibody testing in children No. 31 with newly diagnosed type 1 diabetes No. 40 A pilot study on the use of decision mellitus. Clopidogrel used in combination with By Dretzke J, Cummins C, theory and value of information analysis as part of the NHS Health aspirin compared with aspirin alone Sandercock J, Fry-Smith A, Barrett T, Technology Assessment programme. in the treatment of non-ST-segment- Burls A. By Claxton K, Ginnelly L, Sculpher elevation acute coronary syndromes: a systematic review and economic No. 23 M, Philips Z, Palmer S. evaluation. Clinical effectiveness and cost- No. 32 By Main C, Palmer S, Griffin S, Jones effectiveness of prehospital intravenous The Social Support and Family Health fluids in trauma patients. L, Orton V, Sculpher M, et al. Study: a randomised controlled trial By Dretzke J, Sandercock J, Bayliss and economic evaluation of two S, Burls A. No. 41 alternative forms of postnatal support Provision, uptake and cost of cardiac for mothers living in disadvantaged No. 24 rehabilitation programmes: improving inner-city areas. Newer hypnotic drugs for the short- services to under-represented groups. By Wiggins M, Oakley A, Roberts I, term management of insomnia: a By Beswick AD, Rees K, Griebsch I, Turner H, Rajan L, Austerberry H, et al. systematic review and economic Taylor FC, Burke M, West RR, et al. evaluation. No. 33 No. 42 By Dündar Y, Boland A, Strobl J, Psychosocial aspects of genetic Dodd S, Haycox A, Bagust A, et al. screening of pregnant women and Involving South Asian patients in newborns: a systematic review. clinical trials. No. 25 By Green JM, Hewison J, Bekker HL, By Hussain-Gambles M, Leese B, Development and validation of Bryant, Cuckle HS. Atkin K, Brown J, Mason S, Tovey P. methods for assessing the quality of diagnostic accuracy studies. No. 34 No. 43 By Whiting P, Rutjes AWS, Dinnes J, Evaluation of abnormal uterine Clinical and cost-effectiveness of Reitsma JB, Bossuyt PMM, Kleijnen J. bleeding: comparison of three continuous subcutaneous insulin outpatient procedures within cohorts infusion for diabetes. No. 26 defined by age and menopausal status. By Colquitt JL, Green C, Sidhu MK, EVALUATE hysterectomy trial: By Critchley HOD, Warner P, Lee AJ, Hartwell D, Waugh N. a multicentre randomised trial Brechin S, Guise J, Graham B. comparing abdominal, vaginal and No. 44 laparoscopic methods of hysterectomy. No. 35 Identification and assessment of By Garry R, Fountain J, Brown J, Coronary artery stents: a rapid ongoing trials in health technology Manca A, Mason S, Sculpher M, et al. systematic review and economic assessment reviews. evaluation. By Song FJ, Fry-Smith A, Davenport No. 27 By Hill R, Bagust A, Bakhai A, Methods for expected value of Dickson R, Dundar Y, Haycox A, et al. C, Bayliss S, Adi Y, Wilson JS, et al. information analysis in complex health economic models: developments on No. 36 No. 45 the health economics of interferon-β Review of guidelines for good practice Systematic review and economic and glatiramer acetate for multiple in decision-analytic modelling in health evaluation of a long-acting insulin sclerosis. technology assessment. analogue, insulin glargine By Tappenden P, Chilcott JB, By Philips Z, Ginnelly L, Sculpher M, By Warren E, Weatherley-Jones E, Eggington S, Oakley J, McCabe C. Claxton K, Golder S, Riemsma R, et al. Chilcott J, Beverley C. 133

No. 46 No. 4 No. 13 Supplementation of a home-based Randomised evaluation of alternative Cervical screening programmes: can exercise programme with a class- electrosurgical modalities to treat automation help? Evidence from based programme for people bladder outflow obstruction in men systematic reviews, an economic with osteoarthritis of the knees: a with benign prostatic hyperplasia. analysis and a simulation modelling randomised controlled trial and health By Fowler C, McAllister W, Plail R, exercise applied to the UK. By Willis BH, Barton P, Pearmain P, economic analysis. Karim O, Yang Q. Bryan S, Hyde C. By McCarthy CJ, Mills PM, Pullen R, No. 5 Richardson G, Hawkins N, Roberts CR, No. 14 A pragmatic randomised controlled et al. Laparoscopic surgery for inguinal trial of the cost-effectiveness of hernia repair: systematic review of No. 47 palliative therapies for patients with effectiveness and economic evaluation. Clinical and cost-effectiveness of once- inoperable oesophageal cancer. By McCormack K, Wake B, Perez J, daily versus more frequent use of same By Shenfine J, McNamee P, Steen N, Fraser C, Cook J, McIntosh E, et al. potency topical corticosteroids for Bond J, Griffin SM. atopic eczema: a systematic review and No. 15 economic evaluation. No. 6 Clinical effectiveness, tolerability and By Green C, Colquitt JL, Kirby J, Impact of computer-aided detection cost-effectiveness of newer drugs for Davidson P, Payne E. prompts on the sensitivity and epilepsy in adults: a systematic review specificity of screening mammography. and economic evaluation. No. 48 By Taylor P, Champness J, Given- By Wilby J, Kainth A, Hawkins N, Epstein D, McIntosh H, McDaid C, et al. Acupuncture of chronic headache Wilson R, Johnston K, Potts H. disorders in primary care: randomised No. 7 No. 16 controlled trial and economic analysis. A randomised controlled trial to By Vickers AJ, Rees RW, Zollman CE, Issues in data monitoring and interim analysis of trials. compare the cost-effectiveness of McCarney R, Smith CM, Ellis N, et al. tricyclic antidepressants, selective By Grant AM, Altman DG, Babiker serotonin reuptake inhibitors and AB, Campbell MK, Clemens FJ, No. 49 lofepramine. Generalisability in economic evaluation Darbyshire JH, et al. By Peveler R, Kendrick T, Buxton M, studies in healthcare: a review and case Longworth L, Baldwin D, Moore M, et al. studies. No. 8 By Sculpher MJ, Pang FS, Manca A, Lay public’s understanding of equipoise No. 17 and randomisation in randomised Drummond MF, Golder S, Urdahl H, Clinical effectiveness and cost- controlled trials. et al. effectiveness of immediate angioplasty By Robinson EJ, Kerr CEP, for acute myocardial infarction: No. 50 Stevens AJ, Lilford RJ, Braunholtz DA, systematic review and economic evaluation. Virtual outreach: a randomised Edwards SJ, et al. By Hartwell D, Colquitt J, Loveman controlled trial and economic No. 9 E, Clegg AJ, Brodin H, Waugh N, et al. evaluation of joint teleconferenced Clinical and cost-effectiveness of medical consultations. electroconvulsive therapy for depressive No. 18 By Wallace P, Barber J, Clayton W, illness, schizophrenia, catatonia A randomised controlled comparison of Currell R, Fleming K, Garner P, et al. and mania: systematic reviews and alternative strategies in stroke care. economic modelling studies. By Kalra L, Evans A, Perez I, Knapp M, Swift C, Donaldson N. By Greenhalgh J, Knight C, Hind D, Volume 9, 2005 Beverley C, Walters S. No. 19 The investigation and analysis of No. 1 No. 10 Randomised controlled multiple critical incidents and adverse events in Measurement of health-related quality healthcare. treatment comparison to provide a cost- of life for people with dementia: By Woloshynowych M, Rogers S, effectiveness rationale for the selection development of a new instrument Taylor-Adams S, Vincent C. of antimicrobial therapy in acne. (DEMQOL) and an evaluation of By Ozolins M, Eady EA, Avery A, current methodology. No. 20 Cunliffe WJ, O’Neill C, Simpson NB, By Smith SC, Lamping DL, Banerjee Potential use of routine databases in et al. S, Harwood R, Foley B, Smith P, et al. health technology assessment. By Raftery J, Roderick P, Stevens A. No. 2 No. 11 Do the findings of case series studies Clinical effectiveness and cost- No. 21 vary significantly according to effectiveness of drotrecogin alfa Clinical and cost-effectiveness of newer methodological characteristics? (activated) (Xigris®) for the treatment immunosuppressive regimens in renal By Dalziel K, Round A, Stein K, of severe sepsis in adults: a systematic transplantation: a systematic review and modelling study. Garside R, Castelnuovo E, Payne L. review and economic evaluation. By Woodroffe R, Yao GL, Meads C, By Green C, Dinnes J, Takeda A, Bayliss S, Ready A, Raftery J, et al. No. 3 Shepherd J, Hartwell D, Cave C, et al. Improving the referral process No. 22 for familial breast cancer genetic No. 12 A systematic review and economic counselling: findings of three A methodological review of how evaluation of alendronate, etidronate, randomised controlled trials of two heterogeneity has been examined in risedronate, raloxifene and teriparatide interventions. systematic reviews of diagnostic test for the prevention and treatment of By Wilson BJ, Torrance N, accuracy. postmenopausal osteoporosis. Mollison J, Wordsworth S, Gray JR, By Dinnes J, Deeks J, Kirby J, By Stevenson M, Lloyd Jones M, De 134 Haites NE, et al. Roderick P. Nigris E, Brewer N, Davis S, Oakley J. DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

No. 23 No. 32 No. 40 A systematic review to examine Longer term clinical and economic A randomised controlled trial and the impact of psycho-educational benefits of offering acupuncture care to cost-effectiveness study of systematic interventions on health outcomes patients with chronic low back pain. screening (targeted and total and costs in adults and children with By Thomas KJ, MacPherson population screening) versus routine difficult asthma. H, Ratcliffe J, Thorpe L, Brazier J, practice for the detection of atrial By Smith JR, Mugford M, Holland Campbell M, et al. fibrillation in people aged 65 and over. R, Candy B, Noble MJ, Harrison BDW, The SAFE study. et al. No. 33 By Hobbs FDR, Fitzmaurice DA, Cost-effectiveness and safety of Mant J, Murray E, Jowett S, Bryan S, No. 24 epidural steroids in the management et al. An evaluation of the costs, effectiveness of sciatica. and quality of renal replacement No. 41 By Price C, Arden N, Coglan L, therapy provision in renal satellite units Displaced intracapsular hip fractures Rogers P. in England and Wales. in fit, older people: a randomised By Roderick P, Nicholson T, Armitage comparison of reduction and fixation, No. 34 A, Mehta R, Mullee M, Gerard K, et al. bipolar hemiarthroplasty and total hip The British Rheumatoid Outcome arthroplasty. No. 25 Study Group (BROSG) randomised By Keating JF, Grant A, Masson M, Imatinib for the treatment of patients controlled trial to compare the Scott NW, Forbes JF. with unresectable and/or metastatic effectiveness and cost-effectiveness of gastrointestinal stromal tumours: aggressive versus symptomatic therapy No. 42 systematic review and economic in established rheumatoid arthritis. Long-term outcome of cognitive evaluation. By Symmons D, Tricker K, Roberts behaviour therapy clinical trials in By Wilson J, Connock M, Song F, C, Davies L, Dawes P, Scott DL. central Scotland. Yao G, Fry-Smith A, Raftery J, et al. By Durham RC, Chambers JA, No. 35 Power KG, Sharp DM, Macdonald RR, No. 26 Conceptual framework and systematic Major KA, et al. Indirect comparisons of competing review of the effects of participants’ interventions. and professionals’ preferences in No. 43 By Glenny AM, Altman DG, Song F, randomised controlled trials. The effectiveness and cost-effectiveness Sakarovitch C, Deeks JJ, D’Amico R, By King M, Nazareth I, Lampe F, of dual-chamber pacemakers compared et al. Bower P, Chandler M, Morou M, et al. with single-chamber pacemakers for bradycardia due to atrioventricular No. 27 No. 36 block or sick sinus syndrome: systematic Cost-effectiveness of alternative review and economic evaluation. The clinical and cost-effectiveness of strategies for the initial medical By Castelnuovo E, Stein K, Pitt M, implantable cardioverter defibrillators: management of non-ST elevation acute Garside R, Payne E. a systematic review. coronary syndrome: systematic review and decision-analytical modelling. By Bryant J, Brodin H, Loveman E, No. 44 By Robinson M, Palmer S, Sculpher Payne E, Clegg A. Newborn screening for congenital heart M, Philips Z, Ginnelly L, Bowens A, et al. defects: a systematic review and cost- No. 37 effectiveness analysis. No. 28 A trial of problem-solving by By Knowles R, Griebsch I, Outcomes of electrically stimulated community mental health nurses for Dezateux C, Brown J, Bull C, Wren C. gracilis neosphincter surgery. anxiety, depression and life difficulties By Tillin T, Chambers M, Feldman R. among general practice patients. The No. 45 CPN-GP study. The clinical and cost-effectiveness of No. 29 By Kendrick T, Simons L, left ventricular assist devices for end- The effectiveness and cost-effectiveness Mynors-Wallis L, Gray A, Lathlean J, stage heart failure: a systematic review of pimecrolimus and tacrolimus for Pickering R, et al. and economic evaluation. atopic eczema: a systematic review and By Clegg AJ, Scott DA, Loveman E, economic evaluation. No. 38 Colquitt J, Hutchinson J, Royle P, et al. By Garside R, Stein K, Castelnuovo The causes and effects of socio- E, Pitt M, Ashcroft D, Dimmock P, et al. demographic exclusions from clinical No. 46 trials. The effectiveness of the Heidelberg No. 30 Retina Tomograph and laser diagnostic By Bartlett C, Doyal L, Ebrahim S, Systematic review on urine albumin glaucoma scanning system (GDx) in Davey P, Bachmann M, Egger M, et al. testing for early detection of diabetic detecting and monitoring glaucoma. complications. By Kwartz AJ, Henson DB, Harper No. 39 By Newman DJ, Mattock MB, RA, Spencer AF, McLeod D. Is hydrotherapy cost-effective? Dawnay ABS, Kerry S, McGuire A, A randomised controlled trial of Yaqoob M, et al. No. 47 combined hydrotherapy programmes Clinical and cost-effectiveness of No. 31 compared with physiotherapy land autologous chondrocyte implantation Randomised controlled trial of the cost- techniques in children with juvenile for cartilage defects in knee joints: effectiveness of water-based therapy for idiopathic arthritis. systematic review and economic lower limb osteoarthritis. By Epps H, Ginnelly L, Utley M, evaluation. By Cochrane T, Davey RC, Southwood T, Gallivan S, Sculpher M, By Clar C, Cummins E, McIntyre L, Matthes Edwards SM. et al. Thomas S, Lamb J, Bain L, et al. 135

No. 48 No. 6 No. 15 Systematic review of effectiveness of Systematic review and evaluation Measurement of the clinical and cost- different treatments for childhood of methods of assessing urinary effectiveness of non-invasive diagnostic retinoblastoma. incontinence. testing strategies for deep vein By McDaid C, Hartley S, Bagnall By Martin JL, Williams KS, Abrams thrombosis. A-M, Ritchie G, Light K, Riemsma R. KR, Turner DA, Sutton AJ, Chapple C, By Goodacre S, Sampson F, et al. Stevenson M, Wailoo A, Sutton A, No. 49 Thomas S, et al. Towards evidence-based guidelines No. 7 The clinical effectiveness and cost- for the prevention of venous No. 16 thromboembolism: systematic effectiveness of newer drugs for Systematic review of the effectiveness reviews of mechanical methods, oral children with epilepsy. A systematic and cost-effectiveness of HealOzone® anticoagulation, dextran and regional review. for the treatment of occlusal pit/fissure anaesthesia as thromboprophylaxis. By Connock M, Frew E, Evans B-W, By Roderick P, Ferris G, Wilson K, Bryan S, Cummins C, Fry-Smith A, et al. caries and root caries. Halls H, Jackson D, Collins R, et al. By Brazzelli M, McKenzie L, Fielding No. 8 S, Fraser C, Clarkson J, Kilonzo M, et al. No. 50 Surveillance of Barrett’s oesophagus: exploring the uncertainty through The effectiveness and cost-effectiveness No. 17 systematic review, expert workshop and of parent training/education Randomised controlled trials of economic modelling. programmes for the treatment conventional antipsychotic versus By Garside R, Pitt M, Somerville M, of conduct disorder, including new atypical drugs, and new atypical Stein K, Price A, Gilbert N. oppositional defiant disorder, in drugs versus clozapine, in people with children. No. 9 schizophrenia responding poorly to, or By Dretzke J, Frew E, Davenport C, Topotecan, pegylated liposomal intolerant of, current drug treatment. Barlow J, Stewart-Brown S, Sandercock J, doxorubicin hydrochloride and By Lewis SW, Davies L, Jones PB, et al. paclitaxel for second-line or subsequent Barnes TRE, Murray RM, Kerwin R, treatment of advanced ovarian cancer: et al. a systematic review and economic Volume 10, 2006 evaluation. No. 18 By Main C, Bojke L, Griffin S, Diagnostic tests and algorithms used No. 1 Norman G, Barbieri M, Mather L, et al. in the investigation of haematuria: The clinical and cost-effectiveness of systematic reviews and economic donepezil, rivastigmine, galantamine No. 10 evaluation. Evaluation of molecular techniques and memantine for Alzheimer’s By Rodgers M, Nixon J, Hempel S, disease. in prediction and diagnosis Aho T, Kelly J, Neal D, et al. By Loveman E, Green C, Kirby J, of cytomegalovirus disease in Takeda A, Picot J, Payne E, et al. immunocompromised patients. By Szczepura A, Westmoreland D, No. 19 Cognitive behavioural therapy in No. 2 Vinogradova Y, Fox J, Clark M. FOOD: a multicentre randomised trial addition to antispasmodic therapy for evaluating feeding policies in patients No. 11 irritable bowel syndrome in primary admitted to hospital with a recent Screening for thrombophilia in high- care: randomised controlled trial. stroke. risk situations: systematic review By Kennedy TM, Chalder T, By Dennis M, Lewis S, Cranswick G, and cost-effectiveness analysis. The McCrone P, Darnley S, Knapp M, Forbes J. Thrombosis: Risk and Economic Jones RH, et al. Assessment of Thrombophilia No. 3 Screening (TREATS) study. No. 20 The clinical effectiveness and cost- By Wu O, Robertson L, Twaddle S, A systematic review of the Lowe GDO, Clark P, Greaves M, et al. effectiveness of computed tomography clinical effectiveness and cost- screening for lung cancer: systematic effectiveness of enzyme replacement No. 12 reviews. A series of systematic reviews to inform therapies for Fabry’s disease and By Black C, Bagust A, Boland A, a decision analysis for sampling and mucopolysaccharidosis type 1. Walker S, McLeod C, De Verteuil R, et al. treating infected diabetic foot ulcers. By Connock M, Juarez-Garcia A, By Nelson EA, O’Meara S, Craig D, Frew E, Mans A, Dretzke J, Fry-Smith A, No. 4 Iglesias C, Golder S, Dalton J, et al. et al. A systematic review of the effectiveness and cost-effectiveness of neuroimaging No. 13 No. 21 assessments used to visualise the seizure Randomised clinical trial, observational Health benefits of antiviral therapy for focus in people with refractory epilepsy study and assessment of cost- mild chronic hepatitis C: randomised being considered for surgery. effectiveness of the treatment of controlled trial and economic By Whiting P, Gupta R, Burch J, varicose veins (REACTIV trial). evaluation. Mujica Mota RE, Wright K, Marson A, By Michaels JA, Campbell WB, By Wright M, Grieve R, Roberts J, et al. Brazier JE, MacIntyre JB, Palfreyman SJ, Main J, Thomas HC, on behalf of the Ratcliffe J, et al. No. 5 UK Mild Hepatitis C Trial Investigators. Comparison of conference abstracts No. 14 and presentations with full-text articles The cost-effectiveness of screening for No. 22 in the health technology assessments of oral cancer in primary care. Pressure relieving support surfaces: a rapidly evolving technologies. By Speight PM, Palmer S, Moles DR, randomised evaluation. By Dundar Y, Dodd S, Dickson R, Downer MC, Smith DH, Henriksson M, By Nixon J, Nelson EA, Cranny G, 136 Walley T, Haycox A, Williamson PR. et al. Iglesias CP, Hawkins K, Cullum NA, et al. DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

No. 23 No. 31 No. 40 A systematic review and economic Etanercept and infliximab for the What are the clinical outcome and cost- model of the effectiveness and cost- treatment of psoriatic arthritis: a effectiveness of endoscopy undertaken effectiveness of methylphenidate, systematic review and economic by nurses when compared with doctors? dexamfetamine and atomoxetine evaluation. A Multi-Institution Nurse Endoscopy for the treatment of attention deficit By Woolacott N, Bravo Vergel Y, Trial (MINuET). Hawkins N, Kainth A, Khadjesari Z, hyperactivity disorder in children and By Williams J, Russell I, Durai D, Misso K, et al. adolescents. Cheung W-Y, Farrin A, Bloor K, et al. By King S, Griffin S, Hodges Z, No. 32 Weatherly H, Asseburg C, Richardson G, No. 41 et al. The cost-effectiveness of testing for hepatitis C in former injecting drug The clinical and cost-effectiveness of oxaliplatin and capecitabine for the No. 24 users. adjuvant treatment of colon cancer: The clinical effectiveness and cost- By Castelnuovo E, Thompson-Coon systematic review and economic effectiveness of enzyme replacement J, Pitt M, Cramp M, Siebert U, Price A, et al. therapy for Gaucher’s disease: a evaluation. systematic review. By Pandor A, Eggington S, Paisley S, No. 33 By Connock M, Burls A, Frew E, Tappenden P, Sutcliffe P. Computerised cognitive behaviour Fry-Smith A, Juarez-Garcia A, McCabe C, therapy for depression and anxiety et al. No. 42 update: a systematic review and A systematic review of the effectiveness economic evaluation. of adalimumab, etanercept and No. 25 By Kaltenthaler E, Brazier J, infliximab for the treatment of Effectiveness and cost-effectiveness De Nigris E, Tumur I, Ferriter M, of salicylic acid and cryotherapy for Beverley C, et al. rheumatoid arthritis in adults and cutaneous warts. An economic decision an economic evaluation of their cost- model. No. 34 effectiveness. By Thomas KS, Keogh-Brown MR, Cost-effectiveness of using prognostic By Chen Y-F, Jobanputra P, Barton P, Chalmers JR, Fordham RJ, Holland RC, information to select women with breast Jowett S, Bryan S, Clark W, et al. Armstrong SJ, et al. cancer for adjuvant systemic therapy. By Williams C, Brunskill S, Altman D, No. 43 No. 26 Briggs A, Campbell H, Clarke M, et al. Telemedicine in dermatology: a A systematic literature review of the randomised controlled trial. effectiveness of non-pharmacological No. 35 By Bowns IR, Collins K, Walters SJ, interventions to prevent wandering in Psychological therapies including McDonagh AJG. dementia and evaluation of the ethical dialectical behaviour therapy for implications and acceptability of their borderline personality disorder: a No. 44 use. systematic review and preliminary Cost-effectiveness of cell salvage and By Robinson L, Hutchings D, Corner economic evaluation. alternative methods of minimising L, Beyer F, Dickinson H, Vanoli A, et al. By Brazier J, Tumur I, Holmes M, Ferriter M, Parry G, Dent-Brown K, et al. perioperative allogeneic blood No. 27 transfusion: a systematic review and A review of the evidence on the effects No. 36 economic model. and costs of implantable cardioverter Clinical effectiveness and cost- By Davies L, Brown TJ, Haynes S, defibrillator therapy in different effectiveness of tests for the diagnosis Payne K, Elliott RA, McCollum C. patient groups, and modelling of cost- and investigation of urinary tract effectiveness and cost–utility for these infection in children: a systematic No. 45 groups in a UK context. review and economic model. Clinical effectiveness and cost- By Buxton M, Caine N, Chase D, By Whiting P, Westwood M, Bojke L, effectiveness of laparoscopic surgery Connelly D, Grace A, Jackson C, et al. Palmer S, Richardson G, Cooper J, et al. for colorectal cancer: systematic reviews and economic evaluation. No. 37 No. 28 By Murray A, Lourenco T, de Verteuil Cognitive behavioural therapy Adefovir dipivoxil and pegylated R, Hernandez R, Fraser C, McKinley A, in chronic fatigue syndrome: a interferon alfa-2a for the treatment of randomised controlled trial of an et al. chronic hepatitis B: a systematic review outpatient group programme. and economic evaluation. By O’Dowd H, Gladwell P, Rogers No. 46 By Shepherd J, Jones J, Takeda A, CA, Hollinghurst S, Gregory A. Etanercept and efalizumab for the Davidson P, Price A. treatment of psoriasis: a systematic No. 38 review. No. 29 A comparison of the cost-effectiveness By Woolacott N, Hawkins N, An evaluation of the clinical and cost- of five strategies for the prevention Mason A, Kainth A, Khadjesari Z, Bravo effectiveness of pulmonary artery of nonsteroidal anti-inflammatory Vergel Y, et al. catheters in patient management in drug-induced gastrointestinal toxicity: intensive care: a systematic review and a a systematic review with economic No. 47 randomised controlled trial. modelling. By Harvey S, Stevens K, Harrison D, Systematic reviews of clinical decision By Brown TJ, Hooper L, Elliott RA, tools for acute abdominal pain. Young D, Brampton W, McCabe C, et al. Payne K, Webb R, Roberts C, et al. By Liu JLY, Wyatt JC, Deeks JJ, No. 30 No. 39 Clamp S, Keen J, Verde P, et al. Accurate, practical and cost-effective The effectiveness and cost-effectiveness assessment of carotid stenosis in the of computed tomography screening No. 48 UK. for coronary artery disease: systematic Evaluation of the ventricular assist By Wardlaw JM, Chappell FM, review. device programme in the UK. Stevenson M, De Nigris E, Thomas S, By Waugh N, Black C, Walker S, By Sharples L, Buxton M, Caine N, Gillard J, et al. McIntyre L, Cummins E, Hillis G. Cafferty F, Demiris N, Dyer M, et al. 137

No. 49 No. 7 No. 16 A systematic review and economic Glucocorticoid-induced osteoporosis: Additional therapy for young model of the clinical and cost- a systematic review and cost–utility children with spastic cerebral palsy: a analysis. effectiveness of immunosuppressive randomised controlled trial. By Kanis JA, Stevenson M, therapy for renal transplantation in McCloskey EV, Davis S, Lloyd-Jones M. By Weindling AM, Cunningham CC, children. Glenn SM, Edwards RT, Reeves DJ. By Yao G, Albon E, Adi Y, Milford D, No. 8 Bayliss S, Ready A, et al. Epidemiological, social, diagnostic and No. 17 economic evaluation of population Screening for type 2 diabetes: literature No. 50 screening for genital chlamydial review and economic modelling. Amniocentesis results: investigation of infection. By Waugh N, Scotland G, McNamee By Low N, McCarthy A, Macleod J, anxiety. The ARIA trial. P, Gillett M, Brennan A, Goyder E, et al. By Hewison J, Nixon J, Fountain J, Salisbury C, Campbell R, Roberts TE, et al. Cocks K, Jones C, Mason G, et al. No. 18 No. 9 The effectiveness and cost-effectiveness Methadone and buprenorphine for the of cinacalcet for secondary Volume 11, 2007 management of opioid dependence: hyperparathyroidism in end-stage renal a systematic review and economic disease patients on dialysis: a systematic No. 1 evaluation. review and economic evaluation. Pemetrexed disodium for the treatment By Connock M, Juarez-Garcia A, By Garside R, Pitt M, Anderson R, of malignant pleural mesothelioma: Jowett S, Frew E, Liu Z, Taylor RJ, et al. Mealing S, Roome C, Snaith A, et al. a systematic review and economic evaluation. No. 10 Exercise Evaluation Randomised No. 19 By Dundar Y, Bagust A, Dickson R, Trial (EXERT): a randomised trial The clinical effectiveness and cost- Dodd S, Green J, Haycox A, et al. comparing GP referral for leisure effectiveness of gemcitabine for centre-based exercise, community-based metastatic breast cancer: a systematic No. 2 walking and advice only. review and economic evaluation. A systematic review and economic By Isaacs AJ, Critchley JA, See Tai model of the clinical effectiveness S, Buckingham K, Westley D, Harridge By Takeda AL, Jones J, Loveman E, and cost-effectiveness of docetaxel SDR, et al. Tan SC, Clegg AJ. in combination with prednisone or prednisolone for the treatment of No. 11 No. 20 hormone-refractory metastatic prostate Interferon alfa (pegylated and non- A systematic review of duplex pegylated) and ribavirin for the cancer. ultrasound, magnetic resonance treatment of mild chronic hepatitis angiography and computed By Collins R, Fenwick E, Trowman R, C: a systematic review and economic Perard R, Norman G, Light K, et al. evaluation. tomography angiography for By Shepherd J, Jones J, Hartwell D, the diagnosis and assessment of No. 3 Davidson P, Price A, Waugh N. symptomatic, lower limb peripheral A systematic review of rapid diagnostic arterial disease. tests for the detection of tuberculosis No. 12 By Collins R, Cranny G, Burch J, infection. Systematic review and economic Aguiar-Ibáñez R, Craig D, Wright K, evaluation of bevacizumab and By Dinnes J, Deeks J, Kunst H, et al. Gibson A, Cummins E, Waugh N, et al. cetuximab for the treatment of metastatic colorectal cancer. By Tappenden P, Jones R, Paisley S, No. 21 No. 4 Carroll C. The clinical effectiveness and cost- The clinical effectiveness and cost- effectiveness of treatments for children effectiveness of strontium ranelate for No. 13 with idiopathic steroid-resistant the prevention of osteoporotic fragility A systematic review and economic nephrotic syndrome: a systematic fractures in postmenopausal women. evaluation of epoetin alfa, epoetin review. By Stevenson M, Davis S, Lloyd-Jones beta and darbepoetin alfa in anaemia By Colquitt JL, Kirby J, Green C, M, Beverley C. associated with cancer, especially that attributable to cancer treatment. Cooper K, Trompeter RS. No. 5 By Wilson J, Yao GL, Raftery J, No. 22 A systematic review of quantitative and Bohlius J, Brunskill S, Sandercock J, et al. A systematic review of the routine qualitative research on the role and monitoring of growth in children of effectiveness of written information No. 14 primary school age to identify growth- available to patients about individual A systematic review and economic related conditions. medicines. evaluation of statins for the prevention By Raynor DK, Blenkinsopp of coronary events. By Fayter D, Nixon J, Hartley S, A, Knapp P, Grime J, Nicolson DJ, By Ward S, Lloyd Jones M, Pandor A, Rithalia A, Butler G, Rudolf M, et al. Pollock K, et al. Holmes M, Ara R, Ryan A, et al. No. 23 No. 6 No. 15 Systematic review of the effectiveness of A systematic review of the effectiveness Oral naltrexone as a treatment for preventing and treating Staphylococcus and cost-effectiveness of different aureus carriage in reducing peritoneal relapse prevention in formerly opioid- models of community-based respite dependent drug users: a systematic care for frail older people and their catheter-related infections. review and economic evaluation. carers. By McCormack K, Rabindranath K, By Adi Y, Juarez-Garcia A, Wang D, By Mason A, Weatherly H, Spilsbury Kilonzo M, Vale L, Fraser C, McIntyre L, 138 Jowett S, Frew E, Day E, et al. K, Arksey H, Golder S, Adamson J, et al. et al. DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

No. 24 No. 32 No. 40 The clinical effectiveness and cost Current practice, accuracy, effectiveness Taxanes for the adjuvant treatment of of repetitive transcranial magnetic and cost-effectiveness of the school early breast cancer: systematic review stimulation versus electroconvulsive entry hearing screen. and economic evaluation. therapy in severe depression: a By Bamford J, Fortnum H, Bristow By Ward S, Simpson E, Davis S, Hind multicentre pragmatic randomised K, Smith J, Vamvakas G, Davies L, et al. D, Rees A, Wilkinson A. controlled trial and economic analysis. By McLoughlin DM, Mogg A, Eranti No. 33 No. 41 S, Pluck G, Purvis R, Edwards D, et al. The clinical effectiveness and cost- The clinical effectiveness and cost- effectiveness of inhaled insulin in effectiveness of screening for open No. 25 diabetes mellitus: a systematic review angle glaucoma: a systematic review A randomised controlled trial and and economic evaluation. and economic evaluation. economic evaluation of direct versus By Black C, Cummins E, Royle P, By Burr JM, Mowatt G, Hernández indirect and individual versus group Philip S, Waugh N. R, Siddiqui MAR, Cook J, Lourenco T, modes of speech and language therapy et al. for children with primary language No. 34 impairment. Surveillance of cirrhosis for No. 42 By Boyle J, McCartney E, Forbes J, hepatocellular carcinoma: systematic Acceptability, benefit and costs of early O’Hare A. review and economic analysis. screening for hearing disability: a study By Thompson Coon J, Rogers G, of potential screening tests and models. No. 26 Hewson P, Wright D, Anderson R, By Davis A, Smith P, Ferguson M, Hormonal therapies for early breast Cramp M, et al. Stephens D, Gianopoulos I. cancer: systematic review and economic evaluation. No. 35 No. 43 By Hind D, Ward S, De Nigris E, The Birmingham Rehabilitation Contamination in trials of educational Uptake Maximisation Study (BRUM). Simpson E, Carroll C, Wyld L. interventions. Homebased compared with hospital- By Keogh-Brown MR, Bachmann No. 27 based cardiac rehabilitation in a multi- MO, Shepstone L, Hewitt C, Howe A, Cardioprotection against the toxic ethnic population: cost-effectiveness Ramsay CR, et al. effects of anthracyclines given to and patient adherence. children with cancer: a systematic By Jolly K, Taylor R, Lip GYH, No. 44 review. Greenfield S, Raftery J, Mant J, et al. Overview of the clinical effectiveness of By Bryant J, Picot J, Levitt G, positron emission tomography imaging Sullivan I, Baxter L, Clegg A. No. 36 in selected cancers. A systematic review of the clinical, By Facey K, Bradbury I, Laking G, No. 28 public health and cost-effectiveness of Payne E. Adalimumab, etanercept and infliximab rapid diagnostic tests for the detection for the treatment of ankylosing and identification of bacterial intestinal No. 45 spondylitis: a systematic review and pathogens in faeces and food. The effectiveness and cost-effectiveness economic evaluation. By Abubakar I, Irvine L, Aldus CF, of carmustine implants and By McLeod C, Bagust A, Boland A, Wyatt GM, Fordham R, Schelenz S, et al. temozolomide for the treatment of Dagenais P, Dickson R, Dundar Y, et al. No. 37 newly diagnosed high-grade glioma: No. 29 A randomised controlled trial a systematic review and economic Prenatal screening and treatment examining the longer-term outcomes evaluation. strategies to prevent group B of standard versus new antiepileptic By Garside R, Pitt M, Anderson R, streptococcal and other bacterial drugs. The SANAD trial. Rogers G, Dyer M, Mealing S, et al. infections in early infancy: cost- By Marson AG, Appleton R, Baker effectiveness and expected value of GA, Chadwick DW, Doughty J, Eaton B, No. 46 information analyses. et al. Drug-eluting stents: a systematic review By Colbourn T, Asseburg C, Bojke L, and economic evaluation. Philips Z, Claxton K, Ades AE, et al. No. 38 By Hill RA, Boland A, Dickson R, Clinical effectiveness and cost- Dundar Y, Haycox A, McLeod C, et al. No. 30 effectiveness of different models Clinical effectiveness and cost- of managing long-term oral anti- No. 47 effectiveness of bone morphogenetic coagulation therapy: a systematic The clinical effectiveness and proteins in the non-healing of fractures review and economic modelling. cost-effectiveness of cardiac and spinal fusion: a systematic review. By Connock M, Stevens C, Fry-Smith resynchronisation (biventricular pacing) By Garrison KR, Donell S, Ryder J, A, Jowett S, Fitzmaurice D, Moore D, for heart failure: systematic review and Shemilt I, Mugford M, Harvey I, et al. et al. economic model. By Fox M, Mealing S, Anderson R, No. 31 No. 39 Dean J, Stein K, Price A, et al. A randomised controlled trial of A systematic review and economic postoperative radiotherapy following model of the clinical effectiveness No. 48 breast-conserving surgery in a and cost-effectiveness of interventions Recruitment to randomised trials: minimum-risk older population. The for preventing relapse in people with strategies for trial enrolment and PRIME trial. bipolar disorder. participation study. The STEPS study. By Prescott RJ, Kunkler IH, Williams By Soares-Weiser K, Bravo Vergel Y, By Campbell MK, Snowdon C, LJ, King CC, Jack W, van der Pol M, Beynon S, Dunn G, Barbieri M, Duffy S, Francis D, Elbourne D, McDonald AM, et al. et al. Knight R, et al. 139

No. 49 No. 4 No. 12 Cost-effectiveness of functional Does befriending by trained lay workers The clinical effectiveness and cost- cardiac testing in the diagnosis and improve psychological well-being and effectiveness of central venous catheters management of coronary artery quality of life for carers of people treated with anti-infective agents in disease: a randomised controlled trial. with dementia, and at what cost? A preventing bloodstream infections: The CECaT trial. randomised controlled trial. a systematic review and economic By Sharples L, Hughes V, Crean A, By Charlesworth G, Shepstone L, evaluation. Dyer M, Buxton M, Goldsmith K, et al. Wilson E, Thalanany M, Mugford M, By Hockenhull JC, Dwan K, Boland Poland F. A, Smith G, Bagust A, Dundar Y, et al. No. 50 No. 13 Evaluation of diagnostic tests when No. 5 Stepped treatment of older adults on there is no gold standard. A review of A multi-centre retrospective cohort laxatives. The STOOL trial. methods. study comparing the efficacy, safety By Mihaylov S, Stark C, McColl E, By Rutjes AWS, Reitsma and cost-effectiveness of hysterectomy Steen N, Vanoli A, Rubin G, et al. JB, Coomarasamy A, Khan KS, and uterine artery embolisation for Bossuyt PMM. the treatment of symptomatic uterine No. 14 fibroids. The HOPEFUL study. A randomised controlled trial of No. 51 By Hirst A, Dutton S, Wu O, Briggs cognitive behaviour therapy in Systematic reviews of the clinical A, Edwards C, Waldenmaier L, et al. adolescents with major depression effectiveness and cost-effectiveness of treated by selective serotonin reuptake proton pump inhibitors in acute upper No. 6 inhibitors. The ADAPT trial. gastrointestinal bleeding. Methods of prediction and prevention By Goodyer IM, Dubicka B, By Leontiadis GI, Sreedharan of pre-eclampsia: systematic reviews of Wilkinson P, Kelvin R, Roberts C, A, Dorward S, Barton P, Delaney B, accuracy and effectiveness literature Byford S, et al. Howden CW, et al. with economic modelling. By Meads CA, Cnossen JS, Meher S, No. 15 No. 52 Juarez-Garcia A, ter Riet G, Duley L, The use of irinotecan, oxaliplatin A review and critique of modelling in et al. and raltitrexed for the treatment of prioritising and designing screening advanced colorectal cancer: systematic programmes. No. 7 review and economic evaluation. By Karnon J, Goyder E, Tappenden The use of economic evaluations in By Hind D, Tappenden P, Tumur I, P, McPhie S, Towers I, Brazier J, et al. NHS decision-making: a review and Eggington E, Sutcliffe P, Ryan A. empirical investigation. No. 16 No. 53 By Williams I, McIver S, Moore D, Ranibizumab and pegaptanib for An assessment of the impact of the Bryan S. the treatment of age-related macular NHS Health Technology Assessment degeneration: a systematic review and Programme. No. 8 economic evaluation. By Hanney S, Buxton M, Green C, Stapled haemorrhoidectomy By Colquitt JL, Jones J, Tan SC, Coulson D, Raftery J. (haemorrhoidopexy) for the treatment Takeda A, Clegg AJ, Price A. of haemorrhoids: a systematic review and economic evaluation. No. 17 Volume 12, 2008 By Burch J, Epstein D, Baba-Akbari Systematic review of the clinical A, Weatherly H, Fox D, Golder S, et al. effectiveness and cost-effectiveness No. 1 of 64-slice or higher computed A systematic review and economic No. 9 tomography angiography as an model of switching from The clinical effectiveness of diabetes alternative to invasive coronary nonglycopeptide to glycopeptide education models for Type 2 diabetes: a angiography in the investigation of antibiotic prophylaxis for surgery. systematic review. coronary artery disease. By Cranny G, Elliott R, Weatherly H, By Loveman E, Frampton GK, By Mowatt G, Cummins E, Waugh N, Chambers D, Hawkins N, Myers L, et al. Clegg AJ. Walker S, Cook J, Jia X, et al.

No. 2 No. 10 No. 18 ‘Cut down to quit’ with nicotine Payment to healthcare professionals for Structural neuroimaging in psychosis: replacement therapies in smoking patient recruitment to trials: systematic a systematic review and economic cessation: a systematic review of review and qualitative study. evaluation. By Albon E, Tsourapas A, Frew E, effectiveness and economic analysis. By Raftery J, Bryant J, Powell J, Davenport C, Oyebode F, Bayliss S, et al. By Wang D, Connock M, Barton P, Kerr C, Hawker S. Fry-Smith A, Aveyard P, Moore D. No. 19 No. 11 Systematic review and economic No. 3 Cyclooxygenase-2 selective non- analysis of the comparative A systematic review of the effectiveness steroidal anti-inflammatory drugs effectiveness of different inhaled of strategies for reducing fracture risk (etodolac, meloxicam, celecoxib, corticosteroids and their usage with in children with juvenile idiopathic rofecoxib, etoricoxib, valdecoxib and long-acting beta2 agonists for the arthritis with additional data on long- lumiracoxib) for osteoarthritis and treatment of chronic asthma in adults term risk of fracture and cost of disease rheumatoid arthritis: a systematic and children aged 12 years and over. management. review and economic evaluation. By Shepherd J, Rogers G, Anderson By Thornton J, Ashcroft D, O’Neill T, By Chen Y-F, Jobanputra P, Barton P, R, Main C, Thompson-Coon J, 140 Elliott R, Adams J, Roberts C, et al. Bryan S, Fry-Smith A, Harris G, et al. Hartwell D, et al. DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

No. 20 No. 28 Volume 13, 2009 Systematic review and economic Intravenous magnesium sulphate analysis of the comparative and sotalol for prevention of atrial No. 1 effectiveness of different inhaled fibrillation after coronary artery Deferasirox for the treatment of iron corticosteroids and their usage with bypass surgery: a systematic review and overload associated with regular blood transfusions (transfusional long-acting beta agonists for the economic evaluation. 2 haemosiderosis) in patients suffering treatment of chronic asthma in children By Shepherd J, Jones J, Frampton GK, Tanajewski L, Turner D, Price A. with chronic anaemia: a systematic under the age of 12 years. review and economic evaluation. By Main C, Shepherd J, Anderson R, No. 29 By McLeod C, Fleeman N, Kirkham Rogers G, Thompson-Coon J, Liu Z, J, Bagust A, Boland A, Chu P, et al. Absorbent products for urinary/faecal et al. incontinence: a comparative evaluation No. 2 of key product categories. No. 21 Thrombophilia testing in people with By Fader M, Cottenden A, Getliffe K, Ezetimibe for the treatment of venous thromboembolism: systematic Gage H, Clarke-O’Neill S, Jamieson K, hypercholesterolaemia: a systematic review and cost-effectiveness analysis. et al. By Simpson EL, Stevenson MD, review and economic evaluation. Rawdin A, Papaioannou D. By Ara R, Tumur I, Pandor A, No. 30 Duenas A, Williams R, Wilkinson A, et al. A systematic review of repetitive No. 3 functional task practice with modelling Surgical procedures and non-surgical No. 22 of resource use, costs and effectiveness. devices for the management of non- Topical or oral ibuprofen for chronic By French B, Leathley M, Sutton C, apnoeic snoring: a systematic review of knee pain in older people. The TOIB McAdam J, Thomas L, Forster A, et al. clinical effects and associated treatment study. costs. By Underwood M, Ashby D, Carnes No. 31 By Main C, Liu Z, Welch K, Weiner D, Castelnuovo E, Cross P, Harding G, The effectiveness and cost-effectivness G, Quentin Jones S, Stein K. et al. of minimal access surgery amongst people with gastro-oesophageal reflux No. 4 Continuous positive airway pressure No. 23 disease – a UK collaborative study. The devices for the treatment of obstructive reflux trial. A prospective randomised comparison sleep apnoea–hypopnoea syndrome: a By Grant A, Wileman S, Ramsay C, of minor surgery in primary and systematic review and economic analysis. Bojke L, Epstein D, Sculpher M, et al. secondary care. The MiSTIC trial. By McDaid C, Griffin S, Weatherly H, By George S, Pockney P, Primrose J, Durée K, van der Burgt M, van Hout S, No. 32 Smith H, Little P, Kinley H, et al. Akers J, et al. Time to full publication of studies of anti-cancer medicines for breast cancer No. 24 No. 5 and the potential for publication bias: a A review and critical appraisal Use of classical and novel biomarkers short systematic review. of measures of therapist–patient as prognostic risk factors for localised By Takeda A, Loveman E, Harris P, prostate cancer: a systematic review. interactions in mental health settings. Hartwell D, Welch K. By Sutcliffe P, Hummel S, Simpson E, By Cahill J, Barkham M, Hardy G, Young T, Rees A, Wilkinson A, et al. Gilbody S, Richards D, Bower P, et al. No. 33 Performance of screening tests for No. 6 No. 25 child physical abuse in accident and The harmful health effects of The clinical effectiveness and cost- emergency departments. recreational ecstasy: a systematic review effectiveness of screening programmes By Woodman J, Pitt M, Wentz R, of observational evidence. for amblyopia and strabismus in Taylor B, Hodes D, Gilbert RE. By Rogers G, Elston J, Garside R, children up to the age of 4–5 years: Roome C, Taylor R, Younger P, et al. a systematic review and economic No. 34 No. 7 evaluation. Curative catheter ablation in atrial Systematic review of the clinical fibrillation and typical atrial flutter: By Carlton J, Karnon J, Czoski- effectiveness and cost-effectiveness systematic review and economic Murray C, Smith KJ, Marr J. of oesophageal Doppler monitoring evaluation. in critically ill and high-risk surgical No. 26 By Rodgers M, McKenna C, Palmer patients. A systematic review of the clinical S, Chambers D, Van Hout S, Golder S, By Mowatt G, Houston G, Hernández effectiveness and cost-effectiveness et al. R, de Verteuil R, Fraser C, Cuthbertson and economic modelling of minimal B, et al. No. 35 incision total hip replacement Systematic review and economic approaches in the management of No. 8 modelling of effectiveness and cost The use of surrogate outcomes in arthritic disease of the hip. utility of surgical treatments for men model-based cost-effectiveness analyses: By de Verteuil R, Imamura M, Zhu S, with benign prostatic enlargement. a survey of UK Health Technology Glazener C, Fraser C, Munro N, et al. By Lourenco T, Armstrong N, N’Dow Assessment reports. J, Nabi G, Deverill M, Pickard R, et al. By Taylor RS, Elston J. No. 27 A preliminary model-based assessment No. 36 No. 9 of the cost–utility of a screening Immunoprophylaxis against respiratory Controlling Hypertension and programme for early age-related syncytial virus (RSV) with palivizumab Hypotension Immediately Post Stroke macular degeneration. in children: a systematic review and (CHHIPS) – a randomised controlled By Karnon J, Czoski-Murray C, economic evaluation. trial. By Potter J, Mistri A, Brodie F, Smith K, Brand C, Chakravarthy U, By Wang D, Cummins C, Bayliss S, Chernova J, Wilson E, Jagger C, et al. Davis S, et al. Sandercock J, Burls A. 141

No. 10 No. 19 No. 27 Routine antenatal anti-D prophylaxis Dipsticks and diagnostic algorithms in Paracetamol and ibuprofen for the for RhD-negative women: a systematic urinary tract infection: development treatment of fever in children: the review and economic evaluation. and validation, randomised trial, PITCH randomised controlled trial. By Pilgrim H, Lloyd-Jones M, Rees A. economic analysis, observational cohort By Hay AD, Redmond NM, Costelloe and qualitative study. C, Montgomery AA, Fletcher M, No. 11 By Little P, Turner S, Rumsby K, Hollinghurst S, et al. Amantadine, oseltamivir and zanamivir Warner G, Moore M, Lowes JA, et al. for the prophylaxis of influenza No. 28 (including a review of existing guidance No. 20 A randomised controlled trial to no. 67): a systematic review and Systematic review of respite care in the compare minimally invasive glucose economic evaluation. frail elderly. monitoring devices with conventional By Tappenden P, Jackson R, Cooper By Shaw C, McNamara R, Abrams monitoring in the management of K, Rees A, Simpson E, Read R, et al. K, Cannings-John R, Hood K, Longo insulin-treated diabetes mellitus M, et al. (MITRE). No. 12 By Newman SP, Cooke D, Casbard A, Improving the evaluation of No. 21 Walker S, Meredith S, Nunn A, et al. therapeutic interventions in multiple Neuroleptics in the treatment of sclerosis: the role of new psychometric aggressive challenging behaviour for No. 29 methods. people with intellectual disabilities: Sensitivity analysis in economic By Hobart J, Cano S. a randomised controlled trial evaluation: an audit of NICE current (NACHBID). practice and a review of its use and No. 13 By Tyrer P, Oliver-Africano P, Romeo value in decision-making. By Andronis L, Barton P, Bryan S. Treatment of severe ankle sprain: a R, Knapp M, Dickens S, Bouras N, et al. pragmatic randomised controlled trial Suppl. 1 comparing the clinical effectiveness No. 22 Trastuzumab for the treatment of and cost-effectiveness of three types of Randomised controlled trial to primary breast cancer in HER2-positive mechanical ankle support with tubular determine the clinical effectiveness women: a single technology appraisal. bandage. The CAST trial. and cost-effectiveness of selective By Ward S, Pilgrim H, Hind D. By Cooke MW, Marsh JL, Clark M, serotonin reuptake inhibitors plus Nakash R, Jarvis RM, Hutton JL, et al., supportive care, versus supportive care Docetaxel for the adjuvant treatment on behalf of the CAST trial group. alone, for mild to moderate depression of early node-positive breast cancer: a with somatic symptoms in primary single technology appraisal. No. 14 care: the THREAD (THREshold for By Chilcott J, Lloyd Jones M, Non-occupational postexposure AntiDepressant response) study. Wilkinson A. prophylaxis for HIV: a systematic By Kendrick T, Chatwin J, Dowrick C, review. Tylee A, Morriss R, Peveler R, et al. The use of paclitaxel in the By Bryant J, Baxter L, Hird S. management of early stage breast No. 23 cancer. No. 15 Diagnostic strategies using DNA testing By Griffin S, Dunn G, Palmer S, Blood glucose self-monitoring in type 2 for hereditary haemochromatosis in Macfarlane K, Brent S, Dyker A, et al. diabetes: a randomised controlled trial. at-risk populations: a systematic review Rituximab for the first-line treatment By Farmer AJ, Wade AN, French DP, and economic evaluation. of stage III/IV follicular non-Hodgkin’s Simon J, Yudkin P, Gray A, et al. By Bryant J, Cooper K, Picot J, Clegg lymphoma. A, Roderick P, Rosenberg W, et al. By Dundar Y, Bagust A, Hounsome J, No. 16 McLeod C, Boland A, Davis H, et al. How far does screening women for No. 24 domestic (partner) violence in different Enhanced external counterpulsation Bortezomib for the treatment of health-care settings meet criteria for for the treatment of stable angina and multiple myeloma patients. a screening programme? Systematic heart failure: a systematic review and By Green C, Bryant J, Takeda A, reviews of nine UK National Screening economic analysis. Cooper K, Clegg A, Smith A, et al. Committee criteria. By McKenna C, McDaid C, Fludarabine phosphate for the first- By Feder G, Ramsay J, Dunne D, Suekarran S, Hawkins N, Claxton K, line treatment of chronic lymphocytic Rose M, Arsene C, Norman R, et al. Light K, et al. leukaemia. By Walker S, Palmer S, Erhorn S, No. 17 No. 25 Brent S, Dyker A, Ferrie L, et al. Spinal cord stimulation for chronic Development of a decision support pain of neuropathic or ischaemic tool for primary care management of Erlotinib for the treatment of relapsed origin: systematic review and economic patients with abnormal liver function non-small cell lung cancer. evaluation. tests without clinically apparent liver By McLeod C, Bagust A, Boland A, By Simpson, EL, Duenas A, Holmes disease: a record-linkage population Hockenhull J, Dundar Y, Proudlove C, MW, Papaioannou D, Chilcott J. cohort study and decision analysis et al. (ALFIE). Cetuximab plus radiotherapy for the No. 18 By Donnan PT, McLernon D, Dillon treatment of locally advanced squamous The role of magnetic resonance JF, Ryder S, Roderick P, Sullivan F, et al. cell carcinoma of the head and neck. imaging in the identification of By Griffin S, Walker S, Sculpher M, suspected acoustic neuroma: a No. 26 White S, Erhorn S, Brent S, et al. systematic review of clinical and cost- A systematic review of presumed effectiveness and natural history. consent systems for deceased organ Infliximab for the treatment of adults By Fortnum H, O’Neill C, Taylor R, donation. with psoriasis. Lenthall R, Nikolopoulos T, Lightfoot By Rithalia A, McDaid C, Suekarran By Loveman E, Turner D, Hartwell 142 G, et al. S, Norman G, Myers L, Sowden A. D, Cooper K, Clegg A. DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

No. 30 No. 38 Rituximab for the treatment of Psychological interventions for The effectiveness and cost-effectiveness rheumatoid arthritis. postnatal depression: cluster of methods of storing donated kidneys By Bagust A, Boland A, Hockenhull randomised trial and economic from deceased donors: a systematic J, Fleeman N, Greenhalgh J, Dundar Y, evaluation. The PoNDER trial. review and economic model. et al. By Morrell CJ, Warner R, Slade P, By Bond M, Pitt M, Akoh J, Moxham Omalizumab for the treatment of Dixon S, Walters S, Paley G, et al. T, Hoyle M, Anderson R. severe persistent allergic asthma. By Jones J, Shepherd J, Hartwell D, No. 31 No. 39 Harris P, Cooper K, Takeda A, et al. The effect of different treatment Rehabilitation of older patients: day durations of clopidogrel in patients hospital compared with rehabilitation Rituximab for the treatment of relapsed with non-ST-segment elevation acute at home. A randomised controlled trial. or refractory stage III or IV follicular coronary syndromes: a systematic By Parker SG, Oliver P, Pennington non-Hodgkin’s lymphoma. review and value of information M, Bond J, Jagger C, Enderby PM, et al. analysis. By Boland A, Bagust A, Hockenhull By Rogowski R, Burch J, Palmer S, J, Davis H, Chu P, Dickson R. No. 40 Craigs C, Golder S, Woolacott N. Breastfeeding promotion for infants in Adalimumab for the treatment of psoriasis. No. 32 neonatal units: a systematic review and economic analysis By Turner D, Picot J, Cooper K, Systematic review and individual Loveman E. patient data meta-analysis of diagnosis By Renfrew MJ, Craig D, Dyson L, of heart failure, with modelling of McCormick F, Rice S, King SE, et al. Dabigatran etexilate for the prevention implications of different diagnostic of venous thromboembolism in patients strategies in primary care. No. 41 undergoing elective hip and knee By Mant J, Doust J, Roalfe A, Barton The clinical effectiveness and cost- surgery: a single technology appraisal. P, Cowie MR, Glasziou P, et al. effectiveness of bariatric (weight loss) By Holmes M, C Carroll C, surgery for obesity: a systematic review and Papaioannou D. No. 33 economic evaluation. A multicentre randomised controlled By Picot J, Jones J, Colquitt JL, Romiplostim for the treatment trial of the use of continuous positive Gospodarevskaya E, Loveman E, Baxter of chronic immune or idiopathic airway pressure and non-invasive L, et al. thrombocytopenic purpura: a single positive pressure ventilation in the early technology appraisal. treatment of patients presenting to the No. 42 By Mowatt G, Boachie C, Crowther emergency department with severe Rapid testing for group B streptococcus M, Fraser C, Hernández R, Jia X, et al. acute cardiogenic pulmonary oedema: during labour: a test accuracy study Sunitinib for the treatment of the 3CPO trial. with evaluation of acceptability and gastrointestinal stromal tumours: a By Gray AJ, Goodacre S, Newby cost-effectiveness. DE, Masson MA, Sampson F, Dixon critique of the submission from Pfizer. By Daniels J, Gray J, Pattison H, By Bond M, Hoyle M, Moxham T, S, et al., on behalf of the 3CPO study Roberts T, Edwards E, Milner P, et al. investigators. Napier M, Anderson R. No. 43 No. 45 No. 34 Screening to prevent spontaneous Early high-dose lipid-lowering therapy Vitamin K to prevent fractures in preterm birth: systematic reviews of to avoid cardiac events: a systematic older women: systematic review and accuracy and effectiveness literature review and economic evaluation. economic evaluation. with economic modelling. By Ara R, Pandor A, Stevens J, Rees By Stevenson M, Lloyd-Jones M, A, Rafia R. By Honest H, Forbes CA, Durée KH, Papaioannou D. Norman G, Duffy SB, Tsourapas A, et al. No. 35 No. 46 Adefovir dipivoxil and pegylated No. 44 The effects of biofeedback for the interferon alpha for the treatment The effectiveness and cost-effectiveness treatment of essential hypertension: a of chronic hepatitis B: an updated of cochlear implants for severe to systematic review. systematic review and economic profound deafness in children and By Greenhalgh J, Dickson R, evaluation. adults: a systematic review and Dundar Y. By Jones J, Shepherd J, Baxter L, economic model. Gospodarevskaya E, Hartwell D, Harris By Bond M, Mealing S, Anderson R, No. 47 P, et al. Elston J, Weiner G, Taylor RS, et al. A randomised controlled trial of the use of aciclovir and/or prednisolone for No. 36 Suppl. 2 the early treatment of Bell’s palsy: the Methods to identify postnatal Gemcitabine for the treatment of BELLS study. depression in primary care: an metastatic breast cancer. By Sullivan FM, Swan IRC, Donnan integrated evidence synthesis and value By Jones J, Takeda A, Tan SC, PT, Morrison JM, Smith BH, McKinstry of information analysis. Cooper K, Loveman E, Clegg A. B, et al. By Hewitt CE, Gilbody SM, Brealey S, Paulden M, Palmer S, Mann R, et al. Varenicline in the management of Suppl. 3 smoking cessation: a single technology Lapatinib for the treatment of HER2- No. 37 appraisal. overexpressing breast cancer. A double-blind randomised placebo- By Hind D, Tappenden P, Peters J, By Jones J, Takeda A, Picot J, von controlled trial of topical intranasal Kenjegalieva K. Keyserlingk C, Clegg A. corticosteroids in 4- to 11-year-old children with persistent bilateral otitis Alteplase for the treatment of acute Infliximab for the treatment of media with effusion in primary care. ischaemic stroke: a single technology ulcerative colitis. By Williamson I, Benge S, Barton S, appraisal. By Hyde C, Bryan S, Juarez-Garcia A, Petrou S, Letley L, Fasey N, et al. By Lloyd Jones M, Holmes M. Andronis L, Fry-Smith A. 143

Rimonabant for the treatment of Mifamurtide for the treatment of No. 50 overweight and obese people. osteosarcoma: a single technology Cessation of attention deficit By Burch J, McKenna C, Palmer S, appraisal. hyperactivity disorder drugs Norman G, Glanville J, Sculpher M, et By Pandor A, Fitzgerald P, Stevenson in the young (CADDY) – a al. M, Papaioannou D. pharmacoepidemiological and qualitative study. Telbivudine for the treatment of Ustekinumab for the treatment of By Wong ICK, Asherson P, Bilbow A, chronic hepatitis B infection. moderate to severe psoriasis. Clifford S, Coghill D, R DeSoysa R, et al. By Hartwell D, Jones J, Harris P, By Gospodarevskaya E, Picot J, Cooper K. Cooper K, Loveman E, Takeda A. No. 51 ARTISTIC: a randomised trial of Entecavir for the treatment of chronic No. 48 hepatitis B infection. human papillomavirus (HPV) testing in Endovascular stents for abdominal primary cervical screening. By Shepherd J, Gospodarevskaya E, aortic aneurysms: a systematic review Frampton G, Cooper, K. By Kitchener HC, Almonte M, and economic model. Gilham C, Dowie R, Stoykova B, Sargent Febuxostat for the treatment of By Chambers D, Epstein D, Walker S, A, et al. hyperuricaemia in people with gout: a Fayter D, Paton F, Wright K, et al. single technology appraisal. By Stevenson M, Pandor A. No. 49 Clinical and cost-effectiveness of Rivaroxaban for the prevention of epoprostenol, iloprost, bosentan, venous thromboembolism: a single sitaxentan and sildenafil for pulmonary technology appraisal. arterial hypertension within their By Stevenson M, Scope A, Holmes M, licensed indications: a systematic review Rees A, Kaltenthaler E. and economic evaluation. By Chen Y-F, Jowett S, Barton P, Cetuximab for the treatment of Malottki K, Hyde C, Gibbs JSR, et al. recurrent and/or metastatic squamous cell carcinoma of the head and neck. By Greenhalgh J, Bagust A, Boland A, Fleeman N, McLeod C, Dundar Y, et al.

144 DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Health Technology Assessment programme

Director, Deputy Director, Professor Tom Walley, Professor Jon Nicholl, Director, NIHR HTA Director, Medical Care Research programme, Professor of Unit, University of Sheffield Clinical Pharmacology, University of Liverpool

Prioritisation Strategy Group Members

Chair, Dr Andrew Cook, Professor Paul Glasziou, Ms Lynn Kerridge, Professor Tom Walley, Consultant Advisor, NETSCC, Professor of Evidence-Based Chief Executive Officer, Director, NIHR HTA HTA Medicine, University of Oxford NETSCC and NETSCC, HTA programme, Professor of Clinical Pharmacology, Dr Peter Davidson, Dr Nick Hicks, Dr Ruairidh Milne, University of Liverpool Director of Science Support, Director of NHS Support, Director of Strategy and NETSCC, HTA NETSCC, HTA Development, NETSCC Deputy Chair, Professor Jon Nicholl, Professor Robin E Ferner, Dr Edmund Jessop, Ms Kay Pattison, Director, Medical Care Research Consultant Physician and Medical Adviser, National Section Head, NHS R&D Unit, University of Sheffield Director, West Midlands Centre Specialist, National Programme, Department of for Adverse Drug Reactions, Commissioning Group (NCG), Health Dr Bob Coates, City Hospital NHS Trust, Department of Health, London Consultant Advisor, NETSCC, Birmingham Ms Pamela Young, HTA Specialist Programme Manager, NETSCC, HTA

HTA Commissioning Board Members

Programme Director, Professor Deborah Ashby, Professor Freddie Hamdy, Professor Ian Roberts,. Professor Tom Walley, Professor of Medical Statistics, Professor of Urology, Professor of Epidemiology & Director, NIHR HTA Queen Mary, University of University of Sheffield Public Health, London School programme, Professor of London of Hygiene and Tropical Clinical Pharmacology, Professor Allan House, Medicine University of Liverpool Professor John Cairns, Professor of Liaison Psychiatry, Professor of Health Economics, University of Leeds Professor Mark Sculpher, Chair, London School of Hygiene and Professor of Health Economics, Professor Jon Nicholl, Tropical Medicine Dr Martin J Landray, University of York Director, Medical Care Research Reader in Epidemiology, Unit, University of Sheffield Professor Peter Croft, Honorary Consultant Physician, Professor Helen Smith, Director of Primary Care Clinical Trial Service Unit, Professor of Primary Care, Deputy Chair, Sciences Research Centre, Keele University of Oxford University of Brighton Dr Andrew Farmer, University Senior Lecturer in General Professor Stuart Logan, Professor Kate Thomas, Practice, Department of Professor Nicky Cullum, Director of Health & Social Professor of Complementary & Primary Health Care, Director of Centre for Evidence- Care Research, The Peninsula Alternative Medicine Research, University of Oxford Based Nursing, University of Medical School, Universities of University of Leeds York Exeter and Plymouth Professor Ann Ashburn, Professor David John Professor of Rehabilitation Professor Jenny Donovan, Dr Rafael Perera, Torgerson, and Head of Research, Professor of Social Medicine, Lecturer in Medical Statisitics, Director of York Trials Unit, Southampton General Hospital University of Bristol Department of Primary Health University of York Care, Univeristy of Oxford Professor Steve Halligan, Professor Hywel Williams, Professor of Gastrointestinal Professor of Dermato- Radiology, University College Epidemiology, University of Hospital, London Nottingham

Observers

Ms Kay Pattison, Dr Morven Roberts, Section Head, NHS R&D Clinical Trials Manager, Programme, Department of Medical Research Council Health 145

Diagnostic Technologies & Screening Panel Members

Chair, Dr Stephanie Dancer, Dr Anne Mackie, Dr W Stuart A Smellie, Professor Paul Glasziou, Consultant Microbiologist, Director of Programmes, UK Consultant in Chemical Professor of Evidence-Based Hairmyres Hospital, East National Screening Committee Pathology, Bishop Auckland Medicine, University of Oxford Kilbride General Hospital Dr Michael Millar, Deputy Chair, Professor Glyn Elwyn, Consultant Senior Lecturer in Dr Nicholas Summerton, Dr David Elliman, Primary Medical Care Research Microbiology, Barts and The Consultant Clinical and Public Consultant Paediatrician and Group, Swansea Clinical School, London NHS Trust, Royal Health Advisor, NICE Honorary Senior Lecturer, University of Wales London Hospital Great Ormond Street Hospital, Ms Dawn Talbot, London Dr Ron Gray, Mr Stephen Pilling, Service User Representative Consultant Clinical Director, Centre for Outcomes, Professor Judith E Adams, Epidemiologist, Department Research & Effectiveness, Dr Graham Taylor, Consultant Radiologist, of Public Health, University of Joint Director, National Scientific Advisor, Regional Manchester Royal Infirmary, Oxford Collaborating Centre for DNA Laboratory, St James’s Central Manchester & Mental Health, University University Hospital, Leeds Manchester Children’s Professor Paul D Griffiths, College London Professor of Radiology, Professor Lindsay Wilson University Hospitals NHS Trust, Turnbull, and Professor of Diagnostic University of Sheffield Mrs Una Rennard, Service User Representative Scientific Director of the Radiology, Imaging Science Dr Jennifer J Kurinczuk, Centre for Magnetic Resonance and Biomedical Engineering, Consultant Clinical Dr Phil Shackley, Investigations and YCR Cancer & Imaging Sciences, Epidemiologist, National Senior Lecturer in Health Professor of Radiology, Hull University of Manchester Perinatal Epidemiology Unit, Economics, School of Royal Infirmary Ms Jane Bates, Oxford Population and Health Consultant Ultrasound Sciences, University of Dr Susanne M Ludgate, Newcastle upon Tyne Practitioner, Ultrasound Medical Director, Medicines & Department, Leeds Teaching Healthcare Products Regulatory Hospital NHS Trust Agency, London

Observers

Dr Tim Elliott, Dr Catherine Moody, Dr Ursula Wells, Team Leader, Cancer Programme Manager, Principal Research Officer, Screening, Department of Neuroscience and Mental Department of Health Health Health Board

Pharmaceuticals Panel Members

Chair, Dr Peter Elton, Professor Jonathan Ledermann, Dr Martin Shelly, Professor Robin Ferner, Director of Public Health, Professor of Medical Oncology General Practitioner, Leeds, Consultant Physician and Bury Primary Care Trust and Director of the Cancer and Associate Director, NHS Director, West Midlands Centre Research UK and University Clinical Governance Support for Adverse Drug Reactions, Dr Ben Goldacre, College London Cancer Trials Team, Leicester City Hospital NHS Trust, Research Fellow, Division of Centre Birmingham Psychological Medicine and Dr Gillian Shepherd, Psychiatry, King’s College Dr Yoon K Loke, Director, Health and Clinical Deputy Chair, London Senior Lecturer in Clinical Excellence, Merck Serono Ltd Professor Imti Choonara, Pharmacology, University of Professor in Child Health, Mrs Barbara Greggains, East Anglia Mrs Katrina Simister, University of Nottingham Service User Representative Assistant Director New Professor Femi Oyebode, Medicines, National Prescribing Mrs Nicola Carey, Dr Bill Gutteridge, Consultant Psychiatrist Centre, Liverpool Senior Research Fellow, . Medical Adviser, London and Head of Department, School of Health and Social Strategic Health Authority University of Birmingham Mr David Symes, Service User Representative Care, The University of Dr Dyfrig Hughes, Reading Dr Andrew Prentice, Reader in Pharmacoeconomics Senior Lecturer and Consultant Dr Lesley Wise, Mr John Chapman, and Deputy Director, Centre Obstetrician and Gynaecologist, Unit Manager, Service User Representative for Economics and Policy in The Rosie Hospital, University Pharmacoepidemiology Health, IMSCaR, Bangor of Cambridge Research Unit, VRMM, University Medicines & Healthcare Products Regulatory Agency Observers Mr Simon Reeve, Dr Heike Weber, Dr Ursula Wells, Ms Kay Pattison, Head of Clinical and Cost- Programme Manager, Principal Research Officer, Section Head, NHS R&D Effectiveness, Medicines, Medical Research Council Department of Health Programme, Department of Pharmacy and Industry Group, Health 146 Department of Health

Current and past membership details of all HTA programme ‘committees’ are available from the HTA website (www.hta.ac.uk) DOI: 10.3310/hta13520 Health Technology Assessment 2009; Vol. 13: No. 52

Therapeutic Procedures Panel Members

Chair, Mrs Val Carlill, Mr Paul Hilton, Dr Kate Radford, Dr John C Pounsford, Service User Representative Consultant Gynaecologist Senior Lecturer (Research), Consultant Physician, North and Urogynaecologist, Royal Clinical Practice Research Bristol NHS Trust Mrs Anthea De Barton-Watson, Victoria Infirmary, Newcastle Unit, University of Central Service User Representative upon Tyne Lancashire, Preston Deputy Chair, Professor Scott Weich, Mr Mark Emberton, Professor Nicholas James, Mr Jim Reece Professor of Psychiatry, Division Senior Lecturer in Oncological Professor of Clinical Oncology, Service User Representative of Health in the Community, Urology, Institute of Urology, University of Birmingham, University of Warwick, University College Hospital, and Consultant in Clinical Dr Karen Roberts, Coventry London Oncology, Queen Elizabeth Nurse Consultant, Dunston Hill Hospital Hospital Cottages Professor Jane Barlow, Professor Steve Goodacre, Professor of Public Health in Professor of Emergency Dr Peter Martin, the Early Years, Health Sciences Medicine, University of Consultant Neurologist, Research Institute, Warwick Sheffield Addenbrooke’s Hospital, Medical School, Coventry Professor Christopher Griffiths, Cambridge Ms Maree Barnett, Professor of Primary Care, Barts Acting Branch Head of Vascular and The London School of Programme, Department of Medicine and Dentistry Health

Observers

Dr Phillip Leech, Dr Morven Roberts, Professor Tom Walley, Dr Ursula Wells, Principal Medical Officer for Clinical Trials Manager, Director, NIHR HTA Principal Research Officer, Primary Care, Department of Medical Research Council programme, Professor of Department of Health Health Clinical Pharmacology, University of Liverpool Ms Kay Pattison, Section Head, NHS R&D Programme, Department of Health Disease Prevention Panel Members Chair, Dr John Jackson, Dr Julie Mytton, Dr Kieran Sweeney, Dr Edmund Jessop, General Practitioner, Parkway Locum Consultant in Public Honorary Clinical Senior Medical Adviser, National Medical Centre, Newcastle Health Medicine, Bristol Lecturer, Peninsula College Specialist, National upon Tyne Primary Care Trust of Medicine and Dentistry, Commissioning Group (NCG), Universities of Exeter and London Professor Mike Kelly, Miss Nicky Mullany, Plymouth Director, Centre for Public Service User Representative Deputy Chair, Health Excellence, NICE, Professor Carol Tannahill, Dr David Pencheon, London Professor Ian Roberts, Glasgow Centre for Population Director, NHS Sustainable Professor of Epidemiology and Health Development Unit, Cambridge Dr Chris McCall, Public Health, London School General Practitioner, The of Hygiene & Tropical Medicine Professor Margaret Thorogood, Dr Elizabeth Fellow-Smith, Hadleigh Practice, Corfe Professor of Epidemiology, Medical Director, West London Mullen, Dorset Professor Ken Stein, University of Warwick Medical Mental Health Trust, Middlesex Senior Clinical Lecturer in School, Coventry Ms Jeanett Martin, Public Health, University of Director of Nursing, BarnDoc Exeter Limited, Lewisham Primary Care Trust

Observers

Ms Christine McGuire, Dr Caroline Stone, Research & Development, Programme Manager, Medical Department of Health Research Council

147

Expert Advisory Network Members

Professor Douglas Altman, Mr Jonothan Earnshaw, Professor Alan Horwich, Professor Miranda Mugford, Professor of Statistics in Consultant Vascular Surgeon, Dean and Section Chairman, Professor of Health Economics Medicine, Centre for Statistics Gloucestershire Royal Hospital, The Institute of Cancer and Group Co-ordinator, in Medicine, University of Gloucester Research, London University of East Anglia Oxford Professor Martin Eccles, Professor Allen Hutchinson, Professor Jim Neilson, Professor John Bond, Professor of Clinical Director of Public Health and Head of School of Reproductive Professor of Social Gerontology Effectiveness, Centre for Health Deputy Dean of ScHARR, & Developmental Medicine & Health Services Research, Services Research, University of University of Sheffield and Professor of Obstetrics University of Newcastle upon Newcastle upon Tyne and Gynaecology, University of Tyne Professor Peter Jones, Liverpool Professor Pam Enderby, Professor of Psychiatry, Professor Andrew Bradbury, Dean of Faculty of Medicine, University of Cambridge, Mrs Julietta Patnick, Professor of Vascular Surgery, Institute of General Practice Cambridge National Co-ordinator, NHS Solihull Hospital, Birmingham and Primary Care, University of Cancer Screening Programmes, Sheffield Professor Stan Kaye, Sheffield Mr Shaun Brogan, Cancer Research UK Professor Chief Executive, Ridgeway Professor Gene Feder, of Medical Oncology, Royal Professor Robert Peveler, Primary Care Group, Aylesbury Professor of Primary Care Marsden Hospital and Institute Professor of Liaison Psychiatry, Research & Development, of Cancer Research, Surrey Royal South Hants Hospital, Mrs Stella Burnside OBE, Centre for Health Sciences, Southampton Chief Executive, Regulation Barts and The London School Dr Duncan Keeley, and Improvement Authority, of Medicine and Dentistry General Practitioner (Dr Burch Professor Chris Price, Belfast & Ptnrs), The Health Centre, Director of Clinical Research, Mr Leonard R Fenwick, Thame Bayer Diagnostics Europe, Ms Tracy Bury, Chief Executive, Freeman Stoke Poges Project Manager, World Hospital, Newcastle upon Tyne Dr Donna Lamping, Confederation for Physical Research Degrees Programme Professor William Rosenberg, Therapy, London Mrs Gillian Fletcher, Director and Reader in Professor of Hepatology Antenatal Teacher and Tutor Psychology, Health Services and Consultant Physician, Professor Iain T Cameron, and President, National Research Unit, London School University of Southampton Professor of Obstetrics and Childbirth Trust, Henfield of Hygiene and Tropical Gynaecology and Head of the Medicine, London Professor Peter Sandercock, School of Medicine, University Professor Jayne Franklyn, Professor of Medical Neurology, of Southampton Professor of Medicine, Mr George Levvy, Department of Clinical University of Birmingham Chief Executive, Motor Neurosciences, University of Dr Christine Clark, Neurone Disease Association, Edinburgh Medical Writer and Consultant Mr Tam Fry, Northampton Pharmacist, Rossendale Honorary Chairman, Child Dr Susan Schonfield, Growth Foundation, London Professor James Lindesay, Consultant in Public Health, Professor Collette Clifford, Professor of Psychiatry for the Hillingdon Primary Care Trust, Professor of Nursing and Professor Fiona Gilbert, Elderly, University of Leicester Middlesex Head of Research, The Consultant Radiologist and Medical School, University of NCRN Member, University of Professor Julian Little, Dr Eamonn Sheridan, Birmingham Aberdeen Professor of Human Genome Consultant in Clinical Genetics, Epidemiology, University of St James’s University Hospital, Professor Barry Cookson, Professor Paul Gregg, Ottawa Leeds Director, Laboratory of Hospital Professor of Orthopaedic Infection, Public Health Surgical Science, South Tees Professor Alistaire McGuire, Dr Margaret Somerville, Laboratory Service, London Hospital NHS Trust Professor of Health Economics, Director of Public Health London School of Economics Learning, Peninsula Medical Dr Carl Counsell, Bec Hanley, School, University of Plymouth Clinical Senior Lecturer in Co-director, TwoCan Associates, Professor Rajan Madhok, Neurology, University of West Sussex Medical Director and Director Professor Sarah Stewart-Brown, Aberdeen of Public Health, Directorate Professor of Public Health, Dr Maryann L Hardy, of Clinical Strategy & Public Division of Health in the Professor Howard Cuckle, Senior Lecturer, University of Health, North & East Yorkshire Community, University of Professor of Reproductive Bradford & Northern Lincolnshire Warwick, Coventry Epidemiology, Department Health Authority, York of Paediatrics, Obstetrics & Mrs Sharon Hart, Professor Ala Szczepura, Gynaecology, University of Healthcare Management Professor Alexander Markham, Professor of Health Service Leeds Consultant, Reading Director, Molecular Medicine Research, Centre for Health Services Studies, University of Professor Robert E Hawkins, Unit, St James’s University Dr Katherine Darton, Hospital, Leeds Warwick, Coventry Information Unit, MIND – The CRC Professor and Director Mental Health Charity, London of Medical Oncology, Christie Dr Peter Moore, Mrs Joan Webster, CRC Research Centre, Freelance Science Writer, Consumer Member, Southern Professor Carol Dezateux, Christie Hospital NHS Trust, Ashtead Derbyshire Community Health Professor of Paediatric Manchester Council Epidemiology, Institute of Child Dr Andrew Mortimore, Health, London Professor Richard Hobbs, Public Health Director, Professor Martin Whittle, Head of Department of Primary Southampton City Primary Clinical Co-director, National Mr John Dunning, Care & General Practice, Care Trust Co-ordinating Centre for Consultant Cardiothoracic University of Birmingham Women’s and Children’s Surgeon, Papworth Hospital Dr Sue Moss, Health, Lymington NHS Trust, Cambridge Associate Director, Cancer Screening Evaluation Unit, Institute of Cancer Research, Sutton 148

Current and past membership details of all HTA programme ‘committees’ are available from the HTA website (www.hta.ac.uk)

Feedback The HTA programme and the authors would like to know your views about this report. The Correspondence Page on the HTA website (www.hta.ac.uk) is a convenient way to publish your comments. If you prefer, you can send your comments to the address below, telling us whether you would like us to transfer them to the website. We look forward to hearing from you.

NETSCC, Health Technology Assessment Alpha House University of Southampton Science Park Southampton SO16 7NS, UK Email: [email protected] www.hta.ac.uk ISSN 1366-5278