Online Appendices
Appendix 1. Further details on methods.
Appendix 1. Table 1. Methodological quality assessment of the clinical trials using a tool developed by the Cochrane Collaboration Back Review Group and the additional six criteria.
Appendix1. Table A2. Modified Newcastle-Ottawa Quality Assessment Scale (NOS): Scoring of the cohort studies.
Appendix 2. Search strategies of electronic databases.
Appendix 3. Study Protocol.
Appendix 4. Study Eligibility Form.
Appendix 5. PRISMA Checklist. Appendix 1. Further details on methods
Our study follows the PRISMA guidelines for the conduct and reporting of systematic reviews (study protocol, Appendix5). Bellow we report details on methods to supplement information in the main manuscript.
Methods
Systematic search
We systematically searched electronic databases, Medline and EMBASE (from inception through June 2015), the Cochrane Library, the Cochrane Controlled Clinical Trials Register, and proceedings from annual meetings of the American Society of Hematology (URL: www.hematology.org) and the American Society of Bone and Mineral Research (URL: www.asbmr.org). Separate strategies for Medline and EMBASE databases, developed with an experienced librarian, included the following MeSH terms and key words: heparin/sodium heparin, low molecular weight heparin, bone, bone density, bone loss, fracture/s, vertebral fracture/s, osteoporosis (Appendix 2). The bibliographies of relevant hematological, osteoporosis, and general medicine journals were also manually searched over the past 10 years (i.e., Osteoporosis Int, J Bone Miner Res, Calified Tissue Int, Bone, J Thromb Haemost, J Thromb Thrombolysis, Thromb Haemostasis, Thromb Res, Semin.Thromb Hemost, Circulation, Blood, Am J Hematol, Brit J Haematolol, New Engl J Med, Lancet, JAMA, BMJ).
Data extraction
Two reviewers (OGV and CP) extracted data on the number of all reported fractures and absolute or relative changes in BMD (bone mineral density) as measured by dual-energy X-ray absorptiometry (DXA). In most studies, these outcomes were extracted from the reports on adverse events. The corresponding author of one publication was contacted for clarification of the data reported for fractures.1 We also abstracted information about study design, study duration, age, sex and co-morbidities of study population, daily dose, duration and purpose of long-term LMWH use.
Methodological quality assessment
At least two reviewers (OGV, CP and PSS) independently assessed methodological quality of the included studies using previously validated quality assessment checklists for clinical trials and observational studies.2, 3
Clinical trials were appraised by a 11-item tool developed by the Cochrane Collaboration Back Review Group for the assessment of randomized controlled clinical trials (RCTs), controlled clinical trials or cross-over trials.2 This checklist assesses the presence of selection bias (randomization, treatment allocation – 3 items), performance bias (blinding, co-interventions, compliance – 4 items), detection bias (outcome assessment – 2 items) and attrition bias (loss to follow up, intention-to-treat analysis – 2 items). Acceptable quality is defined as six of the maximum 11 points. Additionally, we appraised the included trials with six a priori defined quality criteria specific to osteoporosis research (Appendix 1. Table 1).
2 The quality of observational studies was evaluated using the Newcastle Ottawa Scale (NOS).3 It is a 9-item tool that consists of four criteria related to the selection of study participants (4 stars), one criterion related to comparability of study groups (2 stars), and three criteria related to the ascertainment of outcome in cohort studies (3 stars). Increasing number of stars correlates with higher study quality. The scale can be tailored a priori to fit objectives of the review. Because this review was related to the adverse effect of LMWH on bone, we a priori modified several NOS items to evaluate fractures and BMD (Appendix 1. Table 2). We also created two new quality criteria specific to osteoporosis research, one related to study power regarding detection of the fracture outcome and another related to ascertainment of the fracture or BMD outcome.
References
1. FRISC II Investigators. Long-term low-molecular-mass heparin in unstable coronary- artery disease: FRISC II prospective randomised multicentre study. FRagmin and Fast Revascularisation during InStability in Coronary artery disease. Investigators. Lancet. 1999;354:701-7. 2. van Tulder M, Furlan A, Bombardier C, Bouter L. Updated method guidelines for systematic reviews in the Cochrane collaboration back review group. Spine. 2003;28:1290-9.
3. Wells GA, Shea B, O'Connell D, Peterson J, Welch V, Losos M. The Newcastle - Ottawa Scale (NOS) for assessing the quality if nonrandomized studies in meta- analyses. At : http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed 16 Dec 2015.
3 Appendix 2. Search strategies in electronic databases
Database: Ovid MEDLINE: 1946- June 11, 2015
Search Strategy: ------1 (randomized controlled trial or controlled clinical trial).pt. 2 randomized controlled trials/ or random allocation/ or double-blind method/ or single-blind method/ 3 1 or 2 4 animal/ not human/ 5 3 not 4 6 clinical trial.pt. 7 exp clinical trials/ or placebos/ or research design/ 8 (clinic$ adj25 trial$).mp. 9 ((singl$ or doubl$ or trebl$ or tripl$) adj (mask$ or blind$)).mp. 10 (placebo$ or random$).mp. 11 (latin adj square).mp. 12 or/6-11 13 12 not 4 14 13 not 5 15 comparative study/ or exp evaluation studies/ or follow-up studies/ or prospective studies/ or cross-over studies/ 16 (control$ or prospective$ or volunteer$).mp. 17 15 or 16 18 17 not 4 19 18 not (14 or 5) 20 5 or 13 or 18 21 (random: or (doubl: adj2 dummy) or ((Singl: or double: or trebl:) adj25 (blind: or mask:)) or RCT or RCTs or (control: adj25 trial:) or multicent: or placebo: or metaanalysis: or (meta adj5 analus:) or sham or effectiveness or efficacy or compar:).ti,ab. 22 randomized controlled trial.pt. 23 clinical trial, phase i.pt. 24 clinical trial, phase ii.pt. 25 clinical trial, phase iv.pt. 26 clinical trial, phase iii.pt. 27 controlled clinical trial.pt. 28 meta analysis.pt. 29 multicenter study.pt. 30 or/22-29 31 or/21-29 32 20 or 31 33 exp cohort studies/ 34 exp case-control studies/ 35 exp prognosis/ 36 disease-free survival.mp. 37 medical: futil:.mp. 38 treatment outcome:.mp. 39 treatment failure:.mp. 40 exp disease progression/ 41 (disease adj1 progress:).mp. 42 exp morbidity/ 43 exp mortality/ 44 fatal outcome:.mp. 45 hospital mortality:.mp. 46 exp survival analysis/
4 47 natural histor:.mp. 48 or/33-47 49 32 or 48 50 heparin/ or heparin, low-molecular-weight/ 51 exp Heparin/ 52 Heparin/ae, tu, th, to [Adverse Effects, Therapeutic Use, Therapy, Toxicity] 53 50 or 51 or 52 54 osteoporosis/ or osteoporosis, postmenopausal/ 55 bone demineralization, pathologic/ or osteoporosis/ 56 colles' fracture/ or femoral fractures/ or femoral neck fractures/ or fractures, bone/ or humeral fractures/ or fractures, spontaneous/ or shoulder fractures/ or radius fractures/ or rib fractures/ or spinal fractures/ or tibial fractures/ or ulna fractures/ or monteggia's fracture/ or fractures, closed/ or fractures, open/ 57 bone density/ 58 exp absorptiometry, photon/ 59 bone resorption/ or osteolysis/ 60 (bone: adj25 marker:).mp. 61 exp heparin/ or heparin, low-molecular-weight/ or heparinoids/ or dermatan sulfate/ or (heparin: or Clarin or Contusol or Disebrin or Eleparon or Elheparin or Elheparon or Epiheparin or "Gag 98" or Gag98 or Hepalean or Hepcon or "Hep Lock" or HepLock or Hepsal or "Lipo Hepin" or LipoHepin or Liquaemin or Liquemin or Menaven or Monoparin or Mucoitin or Multiparin or Noparin or Panheparin or Panhepin or Panheprin or Praecivenin or Pularin or Thrombareduct or Thromboliquin: or Thrombophlogat or Thrombophob or Thrombosamine or "Thrombo Vetren" or ThromboVetren or Vetren or Vister).mp. or ("Bm 2123" or Bm2123 or choay or "Ebpm 1" or Ebpm1 or "Ebmp 2" or Ebpm2 or "Ebmp 3" or Embp3).mp. or ("FF 1034" or Ff1034 or "fr 860" or fr860 or "gag 869" or gag869 or "Pk 007" or pk007 or "sandoz 5100" or Sandoz5100 or "Sandoz 6700" or sandoz6700 or Traxyparine).mp. 62 bone demineralization, pathologic/ or decalcification, pathologic/ or osteoporosis/ or osteoporosis, postmenopausal/ or Bone Diseases, Metabolic/ or Bone Density/ or densitometry/ or absorptiometry, photon/ or exp Fractures, Bone/ or exp "Bone and Bones"/ or osteoporo:.mp. 63 53 or 61 64 54 or 55 or 56 or 57 or 58 or 59 or 60 or 62 65 49 and 63 and 64
5 Database: Embase Classic+Embase: 1946- June 11, 2015
Search Strategy: ------1 ct.fs. 2 randomized controlled trial/ or random allocation/ or double-blind method/ or single-blind method/ or controlled clinical trial/ 3 (rct or rcts or sham).mp. 4 or/1-3 5 Animal/ not human/ 6 4 not 5 7 exp clinical trials/ or placebo/ or latin square design/ 8 (clinic$ adj25 trial$).mp. 9 ((singl$ or doubl$ or trebl$ or tripl$) adj (mask$ or blind$)).mp. 10 (placebo$ or random$).mp. 11 (latin adj square).mp. 12 or/7-11 13 12 not 5 14 13 not 6 15 comparative study/ or exp evaluation studies/ or follow-up studies/ or prospective studies/ or cross-over studies/ 16 (control$ or prospective$ or volunteer$).mp. 17 or/15-16 18 17 not 5 19 18 not (6 or 13) 20 6 or 13 or 18 21 heparin derivative/ or heparin/ or heparin calcium/ or heparinoid/ or exp low molecular weight heparin/ or (heparin: or Clarin or Contusol or Disebrin or Eleparon or Elheparin or Elheparon or Epiheparin or "Gag 98" or Gag98 or Hepalean or Hepcon or "Hep Lock" or HepLock or Hepsal or "Lipo Hepin" or LipoHepin or Liquaemin or Liquemin or Menaven or Monoparin or Mucoitin or Multiparin or Noparin or Panheparin or Panhepin or Panheprin or Praecivenin or Pularin or Thrombareduct or Thromboliquin: or Thrombophlogat or Thrombophob or Thrombosamine or "Thrombo Vetren" or ThromboVetren or Vetren or Vister).mp. 22 (37187-54-5 or 8057-48-5 or 8065-01-8 or 9005-48-5).rn. 23 ("Bm 2123" or Bm2123 or choay or "Ebpm 1" or Ebpm1 or "Ebmp 2" or Ebpm2 or "Ebmp 3" or Embp3).mp. 24 ("FF 1034" or Ff1034 or "fr 860" or fr860 or "gag 869" or gag869 or "Pk 007" or pk007 or "sandoz 5100" or Sandoz5100 or "Sandoz 6700" or sandoz6700 or Traxyparine).mp. 25 21 or 22 or 23 or 24 26 osteopenia/ or bone demineralization/ or osteoporosis/ or postmenopause osteoporosis/ or primary osteoporosis/ or secondary osteoporosis/ or senile osteoporosis/ or bone density/ or bone examination/ or bone biopsy/ or bone densitometry/ or fracture/ or foot fracture/ or fracture dislocation/ or fragility fracture/ or limb fracture/ or open fracture/ or pathologic fracture/ or stress fracture/ or exp arm fracture/ or exp face fracture/ or exp leg fracture/ or exp pelvis fracture/ or exp rib fracture/ or exp skull fracture/ or exp spine fracture/ or Osteolysis/ or exp bone/ or osteoporo:.mp. 27 25 and 26 28 6 and 27 29 14 and 27 30 28 or 29 31 19 and 27 32 28 or 29 or 31 33 29 not 28 34 31 not 30 35 6 or 12 or 17 36 25 and 26 and 35 37 20 and 25 and 26
6 Database: Cochrane Central Register of Controlled Trials: until June 2015
Search Strategy: ------1 Heparin.mp. [mp=title, original title, abstract, mesh headings, heading words, keyword] 2 bone.mp. [mp=title, original title, abstract, mesh headings, heading words, keyword] 3 bone density.mp. [mp=title, original title, abstract, mesh headings, heading words, keyword] 4 osteoporosis.mp. [mp=title, original title, abstract, mesh headings, heading words, keyword] 5 bone loss.mp. [mp=title, original title, abstract, mesh headings, heading words, keyword] 6 bone loss.mp. [mp=title, original title, abstract, mesh headings, heading words, keyword] 7 3 or 4 or 5 or 6 8 1 and 7
7 Appendix 1. Table 1: Methodological quality assessment of the clinical trials using a tool developed by the Cochrane Collaboration Back Review Group and the additional six criteria
Study quality criteria Instructions for scoring Yes (i.e., 1 point) if the authors described a random component Adequate randomization in the sequence generation Yes if assignment generated by an independent person not Concealed allocation responsible for determining the eligibility of the participants Prognostic factor balance at baseline for the Yes if prognostic factor balance at baseline shown for the primary disease primary disease (e.g., venous thromboembolism) Blinding of patients Yes if blinding of the patients described Blinding of healthcare providers Yes if blinding of the healthcare providers described Yes if blinding of the outcome assessor for the primary Blinding of outcome assessor outcome was described Co-intervention similar Yes if co-intervention were described and were similar Yes if assessment of compliance was described, the data on Acceptable compliance in both groups compliance were provided (e.g., % of heparin syringes returned) and the compliance was >90% Yes if the drop-out was described and the calculated rate was Drop-out rate described and acceptable (<10%) <10% Comparable timing of the primary outcome Yes if the primary outcome assessed in both groups at same assessment in both groups follow-up times Yes if all randomized patients were analyzed in the group they were initially allocated to by randomization Intention to treat analysis Or Yes if this was modified intention to treat analysis Van Tulder internal validity score Van Tulder internal validity score: maximum 11 points Additional criterion 1: Prognostic factor Yes if prognostic factors at baseline for osteoporosis were balance at baseline for osteoporosis assessed and balanced Additional criterion 2: Contamination, use of Yes if contamination occurred during the follow-up and heparin in both intervention groups patients from the control group were allowed to use of heparin Additional criterion 3: Comparable timing of Yes if the timing of the assessment of bone outcomes was same bone outcome assessment in both groups for both groups Additional criterion 4: If fracture was assessed Yes if fracture was assessed among the adverse events and the among adverse events, was duration of follow- duration of follow-up was >=1 year up to detect a fracture adequate (>=1 year) Additional criterion 5: Sample large enough to Yes if sample included over 1000 patients (no sample size detect differences between the groups regarding calculations were required) fracture outcome Additional criterion 6: Bone outcomes assessed Yes if bone outcomes were assessed among the adverse events as secondary
8 Appendix 1. Table 2: Modified Newcastle-Ottawa Quality Assessment Scale (NOS): Scoring of the cohort studies
Study quality criteria Instructions for scoring Selection Selection – maximum 4 stars 1 star if truly or somewhat representative of the average Representativeness of the exposed cohort patient indicated for long-term heparin 1 star if drawn from the same community as the exposed Selection of the non exposed cohort cohort 1 star if heparin use was confirmed through the secure record Ascertainment of exposure (database) or the structured interview Demonstration that outcome of interest (i.e., 1 star if osteoporosis and fractures were assessed and not bone outcome) was not present at start of demonstrated at start of the study study Comparability Comparability – maximum 2 stars 2 stars if the adjustment was made at two levels: 1) study design: cohorts matched for important factors at start of study; Comparability of cohorts on the basis of the and, 2) statistical analysis: multivariable analysis design or analysis 1 star if comparability of cohorts was assured on the basis of either the design or analysis Outcome (i.e., bone outcome) Outcome – maximum 3 stars 1 star if bone density measured or fractures evaluated systematically and independently in heparin-exposed and heparin non-exposed patients Ascertainment of outcome or 1 star if bone density or fractures were systematically recorded in the registry in both cohorts 1 star if bone density assessed and the follow-up was at least 3 months Follow-up long enough for outcomes to or occur 1 star if fractures assessed and the follow-up (and exposure) was at least 1 year 1 star if statement was provided about potential reasons for loss to follow-up and if this was less than 10% Adequacy of follow up of cohorts or 1 star if all patients were followed up completely till the end of study NOS Score NOS Score: maximum 9 Additional criterion 1: Large sample that provides sufficient statistical power to detect Yes if sample included over 1000 patients (no sample size differences between the groups regarding calculations were required) fracture outcome Additional criterion 2: Bone outcome Yes if bone outcomes were assessed among the adverse events assessed as secondary
9 Appendix 3. Study Protocol
BACKGROUND
Description of the condition
Osteoporosis is a systemic skeletal disease characterized by low bone mass and micro-architectural deterioration of bone tissue, resulting in bone fragility and fractures 1. It is a major public health problem in Canada, affecting one in six Canadian women aged ≥ 50 years 2. Also, one in two Canadian women aged ≥ 50 years are diagnosed with ostepenia (low bone mass), a condition that precedes osteoporosis 2.
Osteoporotic (fragility) fractures, the major clinical expression of osteoporosis, occur in both women with osteoporosis and women with osteopenia 3. They decrease quality of life, increase morbidity and mortality (10-40% hip-fracture excess mortality) and are associated with considerable healthcare costs 4-6. The primary exemplars of osteoporotic fractures are vertebral, hip and wrist fractures 7-9. Prevalent osteoporotic fractures increase the risk of next vertebral or non-vertebral fractures 2-5 times, independently of BMD (bone mineral density) and age 10-12.
One of the secondary causes of osteoporosis is long-term treatment with heparin 13-17, used in the management of venous thromboembolism (VTE). VTE is also a major health problem in North America, with an overall age- and sex-adjusted annual incidence of over 1 per 1000 18. It is associated with significant mortality (up to 6%), morbidity and substantial costs 19-22.
Description of the intervention
Heparin (unfractionated heparin (UFH) or low-molecular weight heparin (LMWH)) is commonly used short-term (5-10 days), as an initial therapeutic or prophylactic treatment for VTE 15;23-25. It is often concurrent with or followed by oral anticoagulants that are given long-term for at least 3 months 15;23;24.
UFH or LMWH may cause three adverse effects: thrombocytopenia, bleeding and osteoporosis (if given long-term 3 months) 13-15. In current clinical practice, LMWH has largely replaced UFH, as an alternative equally effective and more convenient for administration (it has 2-4 times longer half-life and better bioavailability to UFH, and a predictable anticoagulant response) 15;23;26. Data also suggest that compared to UFH, LMWH may be safer as to bleeding and thrombocytopenia 14;27-31.
Long-term anticoagulation with heparin ( 3 months) is used in pregnant patients with hereditary thrombophilias, a history of VTE or past miscarriages (heparin causes no teratogenic effects as it does not cross placenta) 32-35. Long-term thromboprophylaxis with heparin is also recommended in non-pregnant patients with a poor response or an allergy to oral anticoagulants (e.g., cancer patients) 23;24;36-38. However, numerous clinical studies in general non-pregnant patient populations have also examined long-term treatment with LMWH against oral anticoagulants, suggesting an interest in extending the use of heparin for general long-term primary or secondary thromboprophylaxis. (e.g., 39-52)
How the intervention might work
Studies in rats have shown that heparin treatment causes significant bone loss. Matzsch et al. showed that 33 days of UFH treatment was enough to induce a significant decrease of bone volume 53. A number of animal studies have demonstrated that UFH induced a significantly greater decrease of BMD compared to LMWH 54-57. Differences between UFH and LMWH in
10 the effects on bone metabolism were also shown: UFH accelerates bone resorption and suppresses bone formation, while LMWH suppresses bone formation only 54;58. However, a study by Matzsch et al. suggested that these differences became negligent once the dose of LMWH was high. Thus, they found that after 34 days of treatment, high-dose LMWH induced a similar decrease in the femoral neck BMD as that seen with high-dose UFH 57. Studies in rat models examined changes in bone mechanical properties (extrinsic stiffness, ultimate and breaking load, deformation caused by the load) and found that LMWHs (enoxaparin and nadroparin) caused a comparable weakening of the whole femur as compared to UFH 59. An in- vitro analysis of osteoclast formation showed differences among the species in the sensitivity to heparins 60. Thus, in the rat culture, LMWH at much higher doses than UFH was needed to induce osteclast formation, whereas in the mouse culture, no LMWH dose increase was required to see an effect on osteoclast formation. The authors speculated that accordingly, in humans, long-term treatment with LMWH may harm bones more than expected 60.
Observational and experimental studies in pregnant and non-pregnant patients examined the effect of long-term heparin treatment on within and between-subject changes in trabecular (lumbar spine or distal radius) and cortical (femoral neck) bone densities and incidence of fractures (e.g., 61-65). Some data suggest that both UFH and LMWH may cause clinically significant within-subject decreases in BMD of 4-5% at the lumbar spine after 12 or more weeks of treatment in pregnant patients 66;67and after 12 months of treatment in non-pregnant patients 68. Depending on the type, dose and duration of heparin, the incidence of spine and all clinical fractures ranged from 2% to 5% in pregnant patient populations 63;69 and from 2% to 15% in non-pregnant patient populations 70.
Several systematic reviews compared the side effects of UFH and LMWH and provided the evidence of an equal or safer effect of LMWH concerning thrombocytopenia and bleeding 14;27-31;71;72. However, they either failed to systematically evaluate heparin-induced osteoporosis or drawn their conclusions regarding a less deleterious effect of LMWH on bone from small and selected studies in pregnant and non-pregnant patients.
Why it is important to do this review (Rationale)
Due to lack of systematically evaluated evidence, numerous clinical trials comparing long-term LMWH ( 3 months) to other thromboprophylactic options failed to assess bone outcomes in elderly and non- pregnant patients 39-52.
In addition, the current guidelines for the prevention and treatment of osteoporosis 16;17;73 provide a general recommendation regarding heparin-induced osteoporosis with no attempt to differentiate fracture risks and the magnitude of bone loss by the type or dose of heparin.
Since an increase in the use of long-term heparin (LMWH) may be expected in the near future, there is a need to systematically examine all available evidence to provide balanced and unbiased conclusions regarding the effects of heparins (UFH and LMWH) on bone health.
OBJECTIVES
Overall Research Question(s)
11 Our overall research question(s) is: “In non-pregnant patients undergoing thromboprophylaxis or thrombotherapy, is long-term heparin treatment (i.e., 3 months) associated with worse bone outcomes?”
Primary Research Objectives
Our primary objectives will determine:
The effect of long-term heparin treatment on incidence of fragility fractures (i.e., all clinical fractures, clinical and/or morphometric vertebral fractures, hip fractures, non-vertebral fractures) in non-pregnant patient populations undergoing long-term thromboprophylaxis or thrombotherapy; and,
Secondary Research Objectives
We will examine studies in non-pregnant patient populations to determine the effect of long-term heparin on mean changes in:
BMD at the femoral neck, total hip, lumbar spine or radius;
bone quality parameters;
bone strength parameters; and,
bone remodeling parameters (bone formation and bone resorption markers).
METHODS
Criteria for considering studies for this review
Types of studies
We will include all published original studies (no language restrictions) that reported the primary or any of the secondary outcomes in non-pregnant study populations exposed to long-term heparin. As to study design, we will include randomized controlled trials – RCTs (unit of randomization are individuals), quasi-randomized controlled trials, before-after studies and observational studies (i.e., retrospective and prospective cohort studies and case-control studies). As to the type of publication, in addition to published original studies, we will consider relevant abstracts, dissertations, letters and commentaries that contain some original data (and will contact the primary authors to obtain or update the study information).
12 We will exclude descriptive studies (i.e., case series and case reports), and reviews, letters and commentaries with no original data.
Types of participants
We will include studies done in non-pregnant patient populations who were on heparin for at least 3 months for the purpose of thromboprophylaxis or thrombotherapy (no restrictions as to heparin dose or duration (if longer than 3 months) will be made).
Types of interventions
We will compare the effect of:
UFH versus LMWH (e.g., enoxaparin, nadroparin, fragmin, tinzaparin), placebo, control (oral anticoagulants (e.g., warfarin) or low-dose aspirin), or no treatment; or
LMWH versus placebo, control, or no treatment (if there is no exposure to UFH).
Types of outcome measures
Primary outcome Our primary outcome is fragility fractures including all clinical fractures (clinically recognized (symptomatic) vertebral fractures, hip fractures, and non-vertebral fractures) and morphometric (asymptomatic) vertebral fractures (as detected by lateral spine radiography). Fragility fractures are defined as fractures that result from non-identifiable or minimal trauma (i.e., a fall from a standing height or less) 17.
Secondary Outcomes Our secondary outcomes are: Mean absolute (areal BMD: g/cm2) and relative (%) within and between-group changes in BMD (as measured by central or peripheral Dual Energy X-ray Absorptiometry-DXA) at the femoral neck, lumbar spine (L1-L4, L2-L4), total hip and radius (proximal, distal and ultradistal radius); The mean percentage change in bone densities will be recalculated from the mean absolute difference (if missing). A clinically significant decrease in BMD will be defined at 3% 17;74. Mean changes in bone quality parameters: cortical thickness (mm), trabecular thickness (mm), trabecular number and trabecular spacing (mm–1), cortical and trabecular volumetric BMD (g/cm3) (as measured by peripheral or central quantitative computer tomography); Mean changes in cross-sectional bending stiffness of a long bone (Nm2) - a parameter that determines mechanical properties of the bone (as measured by mechanical response tissue analyzer-MRTA); Mean changes in bone formation (procollagen type 1 N-terminal propeptide (P1NP), osteocalcin (OC) and bone specific alkaline phosphatase (BAP)) and bone resorption
13 markers (hydroxyproline, pyridinoline (PYD), deoxypyridinoline (DPD), and type 1 collagen crosslinked N-and C- telopeptides (NTX and CTX)).
Search methods for identification of studies
To identify relevant articles, we will systematically search electronic databases: Medline(1950–), EMBASE(1980-) and The Cochrane Controlled Clinical Trials Register and other electronic sources such as conference proceedings databases (the American Society of Hematology at URL: www.hematology.org and the American Society of Bone and Mineral Research at URL: www.asbmr.org) ), and various dissertations’ databases (e.g., Foreign Doctoral Dissertations at URL: http://www.crl.edu/; ProQuest Dissertations and Theses at URL: http://proquest.umi.com/; Thesis Canada Portal at URL: http://www.collectionscanada.gc.ca/thesescanada/).
We will also hand-search reference lists (reviews and original studies) and relevant hematological, osteoporosis, and general medicine journals (i.e., Osteoporosis Int, J Bone Miner Res, Calified Tissue Int, Bone, J Thromb Haemost, J Thromb Thrombolysis, Thromb Haemostasis, Thromb Res, Semin.Thromb.Hemost.,Circulation, Blood, Am J Hematol, Brit J Haematolol, New Engl J Med, Lancet, JAMA, BMJ).
For electronic database search (Medline and EMBASE), we developed a systematic search strategy together with a librarian (Ms. Elizabeth Uleryk) that includes the most sensitive Robinson-Dickersin’s strategies 75 for identification clinical trials and observational studies in Medline and EMBASE (separate strategies). We combined these with the following search terms (as both MeSH terms and key words): heparin, low molecular weight heparin, bone, bone density, bone loss/osteopenia, osteoporosis, fracture/s to identify experimental and non- experimental studies relevant to heparin-induced osteoporosis (please see Appendix 1 for the complete description of the search strategies).
As suggested in the literature, we will use a QUOROM flow diagram to report a yield of our systematic search 76.
DATA COLLECTION AND ANALYSIS
Selection of studies
Title stage: One reviewer (Olga Gajic-Veljanoski-OGV) will assess titles of all electronically retrieved articles to exclude articles irrelevant to the topic. 14 Abstract stage: Two reviewers (OGV, Prakesh S Shah-PSS) will independently assess abstracts of potentially relevant studies. The study will be included and reviewed in full if it satisfies all of the following criteria: one of the primary or secondary outcomes were reported heparin was identified as an exposure or an intervention it is an observational or experimental study (and not a case report) We will record all inclusions and reasons for exclusions in Study Eligibility Forms (Appendix 1) and will keep track of all excluded studies using the reference manger software.
Data extraction and management Two reviewers will independently assess the methodological quality of studies and will also independently extract the data from the relevant studies using Data Abstraction Forms.
Data Collection The Data Abstraction Form will collect information on study design and quality score (potential biases) and will include information related to heparin exposure, study population characteristics and potential confounders. For example, we will collect information on mean age and gender of study participants, major comorbidities and reasons for heparin treatment, calcium and vitamin D intake (food and/or supplements), mobility or amount of time spent in bed (if available), history of fractures or osteoporosis at baseline. As to heparin exposure, we will collect type of intervention and control treatment, dose and duration of intervention. As to the outcome measures, we will collect type and number of fractures and mean BMD (and standard deviations-SDs) at baseline and follow-up visits at different skeletal sites (femoral neck, lumbar spine (L1-L4, L2-L4), trochanter, Ward’s triangle, total hip, proximal, distal or ultradistal radius).
Pilot testing of Data Abstraction Tool: We will use a random sample of three observational and three experimental studies to test this abstraction tool for completeness, preciseness and feasibility. We will modify it accordingly.
Inadequate or missing information and discrepancy: If we are not able to extract the relevant data, the primary authors will be contacted. Discrepancies will be resolved by consensus.
Data management: We will use tables in Review Manager 5 to manage the data and prepare them for further subgroup analyses.
Assessment of risk of bias in included studies
We will independently perform the quality assessments of observational and experimental studies using two previously validated checklists 77;78.
The quality of observational (non-randomized) studies will be evaluated using the Newcastle Ottawa Scale (NOS) 78. It is a 9-item tool that consists of four criteria related to the selection of study participants (4 stars), one criterion related to comparability of study groups (2 stars), and three criteria related to the ascertainment of outcome in cohort studies (3 stars).
The internal validity of experimental studies will be appraised utilizing an 11-item tool developed by the Cochrane Collaboration Back Review Group for the quality assessment of RCTs, controlled clinical trials and crossover studies 77. This checklist includes criteria that examine selection bias (randomization, treatment allocation–3 items), performance bias (blinding, co-interventions,
15 compliance–4 items), detection bias (outcome assessment–2 items) and attrition bias (loss to follow up, intention-to-treat analysis–2 items). It takes 10 minutes to complete and the maximum score is 11. Acceptable quality will be defined at a cut-off of 6/11, as suggested by van Tulder et al. 77.
In addition to these instruments, we will add topic-specific study quality items related to the timing of BMD measurement, baseline measurement of morphometric vertebral fractures (that are asymptomatic and usually clinically unrecognized) and potential bias due to regression to the mean80.
Agreement - Study Quality: Any discrepancies will be resolved by consensus. If consensus cannot be reached, we will seek an opinion/assessment from the third reviewer (Dr Angela M. Cheung -AMC).
Measures of treatment effect
For our primary binary outcome (fragility fractures) we will use relative risk (RR) and risk difference (RD) to measure treatment effect in prospective studies. A RR of 1.0 or less or RD of 0 or less indicates a lower risk of fractures among patients given heparin long-term. If observational case- control studies are included, the measure of treatment effect will be odds ratio (OR). As fractures are rare outcome, we could assume that RR and OR are approximately the same.
All our secondary outcomes are continuous variables, and thus we will use weighted mean difference (WMD) to measure effect size.
Unit of analysis issues
We expect unit of randomization to be at patient level. BMD measurements and related SDs could be reported differently among the studies. Primary authors will be contacted to resolve any unit of analysis issues.
Dealing with missing data
In case of missing or inaccurate original data, we will contact the primary authors to try to obtain appropriate information. If we are unable to obtain this information, and if the amount of missing data is not larger than 5%, we may decide to impute data (using appropriate statistical methods)81;82. Sensitivity analysis will be performed to test how much our conclusions change with and without data imputations.
Assessment of heterogeneity
We will use Q test (Cochran’s Chi-Square test of homogeneity) to assess statistical heterogeneity 83. As this test has low power to detect heterogeneity, we will use a higher significance threshold level of 0.1. I2 statistic will be used to quantify degree of statistical heterogeneity beyond chance (%)83. I2 statistic > 50% indicates substantial amount of heterogeneity between the studies and a need for random-effects meta- analysis.
16 Assessment of reporting biases
We will be using funnel plot to explore publication bias83.
DATA SYNTHESIS
We will synthesize the data separately for experimental and observational studies. If we decide to perform meta-analysis, we will use the DerSimonian and Laird random-effects method regardless of the Q-test results. The DerSimonian and Laird random-effects method adjust for between- study variability using the inverse variance method83. If there is no heterogeneity, the estimates from the fixed-effects and random-effects methods should be the same. For the case of fractures which are rare outcomes, we will also perform the Mantel-Haenszel method which is more robust than inverse variance method 83. These results will be compared in sensitivity analysis. All tests of significance will be 2-sided and statistical significance will be defined as p < 0.05.
Subgroup analysis and investigation of heterogeneity
We will explore clinical (between-study) heterogeneity using a priori defined subgroup analyses. If data allow, we will perform subgroup analyses on the type of fractures (e.g., spine fractures, all clinical fractures), BMD skeletal site, type of control treatment, heparin dose, heparin duration, and comorbidity status (cancer vs. other).
Sensitivity analysis
Sensitivity analyses will be performed to explore changes in our conclusions in the presence of bias (low quality studies vs. high quality studies) and missing data (estimates from models with imputed data vs. models with missing data (studies included) vs. models with deleted missing data (studies excluded)). We will also examine whether and how much use of different meta-analytic methods will change our pooled estimates and study conclusions (fixed-effects vs. random-effects method, Mantel-Haenszel method vs. inverse variance method).
17 REFERENCES [STUDY PROTOCOL]
1. Anonymous. Consensus Development Conference: Diagnosis, Prophylaxis, and Treatment of Osteoporosis. Am J Med 1993; 94:646-50.
2. Tenenhouse A, Joseph L, Kreiger N, et al. Estimation of the prevalence of low bone density in Canadian women and men using a population-specific DXA reference standard: the Canadian Multicentre Osteoporosis Study (CaMos). Osteoporos Int 2000; 11:897-04.
3. Kanis JA, Johnell O, Oden A, Dawson A, De Laet C, Jonsson B. Ten year probabilities of osteoporotic fractures according to BMD and diagnostic thresholds. Osteoporos Int 2001; 12:989-95.
4. Wiktorowicz ME, Goeree R, Papaioannou A, Adachi JD, Papadimitropoulos E. Economic implications of hip fracture: health service use, institutional care and cost in Canada. Osteoporos Int 2001; 12:271-8.
5. Kanis JA, Oden A, Johnell O, De Laet C, Jonsson B, Oglesby AK. The components of excess mortality after hip fracture. Bone 2003; 32:468-3.
6. Cooper C. The crippling consequences of fractures and their impact on quality of life. Am J Med 1997; 103:12S-7S.
7. Cummings SR, Melton LJ. Epidemiology and outcomes of osteoporotic fractures. Lancet 2002; 359:1761-7.
8. Kanis JA, Oden A, Johnell O, Jonsson B, de Laet C, Dawson A. The burden of osteoporotic fractures: a method for setting intervention thresholds. Osteoporos Int 2001; 12:417-27.
9. Seeley DG, Browner WS, Nevitt MC, Genant HK, Scott JC, Cummings SR. Which fractures are associated with low appendicular bone mass in elderly women? The Study of Osteoporotic Fractures Research Group. Ann Intern Med 1991; 115:837-42.
10. Klotzbuecher CM, Ross PD, Landsman PB, Abbott TA, III, Berger M. Patients with prior fractures have an increased risk of future fractures: a summary of the literature and statistical synthesis. J Bone Miner Res 2000; 15:721-39.
11. Kanis JA, Johnell O, De Laet C, et al. A meta-analysis of previous fracture and subsequent fracture risk. Bone 2004; 35:375-82.
12. Cauley JA, Hochberg MC, Lui LY, et al. Long-term risk of incident vertebral fractures. JAMA 2007; 298:2761-7. 18 13. Weitz JI. Low-molecular-weight heparins. N Engl J Med 1997; 337:688-98.
14. Hirsh J, Warkentin TE, Raschke R, Granger C, Ohman EM, Dalen JE. Heparin and low- molecular-weight heparin: mechanisms of action, pharmacokinetics, dosing considerations, monitoring, efficacy, and safety. Chest 1998; 114:489S-10S.
15. Hirsh J, Warkentin TE, Shaughnessy SG, et al. Heparin and low-molecular-weight heparin: mechanisms of action, pharmacokinetics, dosing, monitoring, efficacy, and safety. Chest 2001; 119:64S-94S.
16. Cheung AM, Feig DS, Kapral M, Diaz-Granados N, Dodin S. Prevention of osteoporosis and osteoporotic fractures in postmenopausal women: recommendation statement from the Canadian Task Force on Preventive Health Care. CMAJ 2004; 170:1665-7.
17. Brown JP, Josse RG. 2002 clinical practice guidelines for the diagnosis and management of osteoporosis in Canada. CMAJ 2002; 167:S1-34.
18. Silverstein MD, Heit JA, Mohr DN, Petterson TM, O'Fallon WM, Melton LJ, III. Trends in the incidence of deep vein thrombosis and pulmonary embolism: a 25-year population- based study. Arch Intern Med 1998; 158:585-93.
19. Gathof BS, Picker SM, Rojo J. Epidemiology, etiology and diagnosis of venous thrombosis. Eur J Med Res 2004; 9:95-103.
20. Hull RD, Pineo GF. Heparin and low-molecular-weight heparin therapy for venous thromboembolism: will unfractionated heparin survive? Semin Thromb Hemost 2004; 30:11-23.
21. Menon J, Salman MM, Md GH. Venous Thrombolysis: Current Perspectives. Curr Treat Options Cardiovasc Med 2004; 6:159-68.
22. Ridker PM. Inherited risk factors for venous thromboembolism: implications for clinical practice. Clin Cornerstone 2002; 4:18-30.
23. Hyers TM, Agnelli G, Hull RD, et al. Antithrombotic therapy for venous thromboembolic disease. Chest 2001; 119:176S-93S.
24. Geerts WH, Heit JA, Clagett GP, et al. Prevention of venous thromboembolism. Chest 2001; 119:132S-75S.
25. Fareed J, Hoppensteadt DA, Fareed D, et al. Survival of heparins, oral anticoagulants, and aspirin after the year 2010. Semin Thromb Hemost 2008; 34:58-73.
19 26. Warkentin TE, Levine MN, Hirsh J, et al. Heparin-induced thrombocytopenia in patients treated with low-molecular- weight heparin or unfractionated heparin. N Engl J Med 1995; 332:1330-5.
27. Boneu B. Low molecular weight heparins: are they superior to unfractionated heparins to prevent and to treat deep vein thrombosis? Thromb Res 2000; 100:113-20.
28. Eikelboom JW, Quinlan DJ, Douketis JD. Extended-duration prophylaxis against venous thromboembolism after total hip or knee replacement: a meta-analysis of the randomised trials. Lancet 2001; 358:9-15.
29. Pineo GF, Hull RD. Low-molecular-weight heparin for the treatment of venous thromboembolism in the elderly. Clin Appl Thromb Hemost 2005; 11:15-23.
30. Sonawane S, Kasbekar N, Berns JS. The safety of heparins in end-stage renal disease. Semin Dial 2006; 19:305-10.
31. Turpie AG. Extended duration of thromboprophylaxis in acutely ill medical patients: optimizing therapy? J Thromb Haemost 2007; 5:5-11.
32. Atrash HK, Rowley D, Hogue CJ. Maternal and perinatal mortality. Curr Opin Obstet Gynecol 1992; 4:61-71.
33. Melissari E, Parker CJ, Wilson NV, et al. Use of low molecular weight heparin in pregnancy. Thromb Haemost 1992; 68:652-6.
34. Shaul WL, Hall JG. Multiple congenital anomalies associated with oral anticoagulants. Am J Obstet Gynecol 1977; 127:191-8.
35. Stevenson RE, Burton OM, Ferlauto GJ, Taylor HA. Hazards of oral anticoagulants during pregnancy. JAMA 1980; 243:1549-51.
36. Lee AY, Levine MN. Venous thromboembolism and cancer: risks and outcomes. Circulation 2003; 107:I17-I21.
37. Lee AY. The role of low-molecular-weight heparins in the prevention and treatment of venous thromboembolism in cancer patients. Curr Opin Pulm Med 2003; 9:351-5.
38. Bates SM, Greer IA, Hirsh J, Ginsberg JS. Use of antithrombotic agents during pregnancy: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004; 126:627S-44S.
39. Sideras K, Schaefer PL, Okuno SH, et al. Low-molecular-weight heparin in patients with advanced cancer: a phase 3 clinical trial. Mayo Clin Proc 2006; 81:758-67.
20 40. Santamaria A, Juarez S, Reche A, Gomez-Outes A, Martinez-Gonzalez J, Fontcuberta J. Low-molecular-weight heparin, bemiparin, in the outpatient treatment and secondary prophylaxis of venous thromboembolism in standard clinical practice: the ESFERA Study. Int J Clin Pract 2006; 60:518-25.
41. Pini M, Aiello S, Manotti C, et al. Low molecular weight heparin versus warfarin in the prevention of recurrences after deep vein thrombosis. Thromb Haemost 1994; 72:191-7.
42. Meyer G, Marjanovic Z, Valcke J, et al. Comparison of low-molecular-weight heparin and warfarin for the secondary prevention of venous thromboembolism in patients with cancer: a randomized controlled study. Arch Intern Med 2002; 162:1729-35.
43. Luk C, Wells PS, Anderson D, Kovacs MJ. Extended outpatient therapy with low molecular weight heparin for the treatment of recurrent venous thromboembolism despite warfarin therapy. Am J Med 2001; 111:270-3.
44. Lopez-Beret P, Orgaz A, Fontcuberta J, et al. Low molecular weight heparin versus oral anticoagulants in the long-term treatment of deep venous thrombosis. Journal of Vascular Surgery 2001; 33:77-90.
45. lopaciuk S, Bielska-Falda H, Noszczyk W, et al. Low molecular weight heparin versus acenocoumarol in the secondary prophylaxis of deep vein thrombosis. Thromb Haemost 1999; 81:26-31.
46. Lee AY, Levine MN, Baker RI, et al. Low-molecular-weight heparin versus a coumarin for the prevention of recurrent venous thromboembolism in patients with cancer. N Engl J Med 2003; 349:146-53.
47. Lablanche JM, McFadden EP, Meneveau N, et al. Effect of nadroparin, a low-molecular- weight heparin, on clinical and angiographic restenosis after coronary balloon angioplasty: the FACT study. Fraxiparine Angioplastie Coronaire Transluminale. Circulation 1997; 96:3396-02.
48. Kher A, Samama MM. Primary and secondary prophylaxis of venous thromboembolism with low-molecular-weight heparins: prolonged thromboprophylaxis, an alternative to vitamin K antagonists. J Thromb Haemost 2005; 3:473-81.
49. Iorio A, Guercini F, Pini M. Low-molecular-weight heparin for the long-term treatment of symptomatic venous thromboembolism: meta-analysis of the randomized comparisons with oral anticoagulants. Journal of Thrombosis & Haemostasis 2003;1:1906-13.
50. Husted SE, Ziegler BK, Kher A. Long-term anticoagulant therapy in patients with coronary artery disease. Eur Heart J 2006; 27:913-9.
21 51. Deitcher SR, Kessler CM, Merli G, Rigas JR, Lyons RM, Fareed J. Secondary prevention of venous thromboembolic events in patients with active cancer: enoxaparin alone versus initial enoxaparin followed by warfarin for a 180-day period. Clin Appl Thromb Hemost 2006; 12:389-96.
52. Bergqvist D, Agnelli G, Cohen AT, et al. Duration of prophylaxis against venous thromboembolism with enoxaparin after surgery for cancer. N Engl J Med 2002; 346:975-80.
53. Matzsch T, Bergqvist D, Hedner U, Nilsson B, Ostergaard P. Heparin-induced osteoporosis in rats. Thromb Haemost 1986; 56:293-4.
54. Nishiyama M, Itoh F, Ujiie A. Low-molecular-weight heparin (dalteparin) demonstrated a weaker effect on rat bone metabolism compared with heparin. Jpn J Pharmacol 1997; 74:59-68.
55. Muir JM, Andrew M, Hirsh J, et al. Histomorphometric analysis of the effects of standard heparin on trabecular bone in vivo. Blood 1996; 88:1314-20.
56. Monreal M, Vinas L, Monreal L, Lavin S, Lafoz E, Angles AM. Heparin-related osteoporosis in rats. A comparative study between unfractioned heparin and a low- molecular-weight heparin. Haemostasis 1990; 20:204-7.
57. Matzsch T, Bergqvist D, Hedner U, Nilsson B, Ostergaard P. Effects of low molecular weight heparin and unfragmented heparin on induction of osteoporosis in rats. Thromb Haemost 1990; 63:505-9.
58. Muir JM, Hirsh J, Weitz JI, Andrew M, Young E, Shaughnessy SG. A histomorphometric comparison of the effects of heparin and low- molecular-weight heparin on cancellous bone in rats. Blood 1997; 89:3236-42.
59. Folwarczna J, Janiec W, Sliwinski L. Effects of heparin and low-molecular-weight heparins on bone mechanical properties in rats. Thromb Haemost 2004; 92:940-6.
60. Folwarczna J, Sliwinski L, Janiec W, Pikul M. Effects of standard heparin and low- molecular-weight heparins on the formation of murine osteoclasts in vitro. Pharmacol Rep 2005; 57:635-45.
61. Casele HL, Laifer SA. Prospective evaluation of bone density in pregnant women receiving the low molecular weight heparin enoxaparin sodium. J Matern Fetal Med 2000; 9:122-5.
22 62. Dahlman T, Lindvall N, Hellgren M. Osteopenia in pregnancy during long-term heparin treatment: a radiological study post partum. Br J Obstet Gynaecol 1990; 97:221-8.
63. Dahlman TC. Osteoporotic fractures and the recurrence of thromboembolism during pregnancy and the puerperium in 184 women undergoing thromboprophylaxis with heparin. Am J Obstet Gynecol 1993; 168:1265-70.
64. Dahlman TC, Sjoberg HE, Ringertz H. Bone mineral density during long-term prophylaxis with heparin in pregnancy. Am J Obstet Gynecol 1994; 170:1315-20.
65. Monreal M, Olive A, Lafoz E, del Rio L. Heparins, coumarin, and bone density. Lancet 1991; 338:706.
66. Backos M, Rai R, Thomas E, Murphy M, Dore C, Regan L. Bone density changes in pregnant women treated with heparin: a prospective, longitudinal study. Hum Reprod 1999; 14:2876-80.
67. Shefras J, Farquharson RG. Bone density studies in pregnant women receiving heparin. Eur J Obstet Gynecol Reprod Biol 1996; 65:171-4.
68. Wawrzynska L, Tomkowski WZ, Przedlacki J, Hajduk B, Torbicki A. Changes in Bone Density during Long-Term Administration of Low-Molecular-Weight Heparins or Acenocoumarol for Secondary Prophylaxis of Venous Thromboembolism. Pathophysiology of Haemostasis & Thrombosis 2003; 33:64-7.
69. Howell R, Fidler J, Letsky E, de Swiet M. The risks of antenatal subcutaneous heparin prophylaxis: a controlled trial. Br J Obstet Gynaecol 1983; 90:1124-8.
70. Monreal M, Lafoz E, Olive A, del Rio L, Vedia C. Comparison of subcutaneous unfractionated heparin with a low molecular weight heparin (Fragmin) in patients with venous thromboembolism and contraindications to coumarin. Thromb Haemost 1994; 71:7-11.
71. Greer IA, Nelson-Piercy C. Low-molecular-weight heparins for thromboprophylaxis and treatment of venous thromboembolism in pregnancy: a systematic review of safety and efficacy. Blood 2005; 106:401-7.
72. Eldor A. The use of low-molecular-weight heparin for the management of venous thromboembolism in pregnancy. Eur J Obstet Gynecol Reprod Biol 2002; 104:3-13.
73. National Osteoporosis Foundation. Physician's guide to prevention and treatment of osteoporosis. Belle Mead(NJ): Excerpta Medica; 1998.
23 74. Blake GM, Wahner HW, Fogelman I. The Evaluation of Osteoporosis: Dual Energy X- Ray Absorptiometry and Ultrasound in Clinical Practice. 2nd ed. London: Martin Dunitz; 1999.
75. Robinson KA, Dickersin K. Development of a highly sensitive search strategy for the retrieval of reports of controlled trials using PubMed. Int J Epidemiol 2002; 31(1):150- 153.
76. Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF. Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement. Quality of Reporting of Meta-analyses. Lancet 1999; 354:1896-00.
77. van Tulder M, Furlan A, Bombardier C, Bouter L. Updated method guidelines for systematic reviews in the Cochrane collaboration back review group. Spine 2003; 28:1290-9.
78. Wells G, Shea B, O'Connell D, Peterson J, Welch V, Losos M. The Newcastle - Ottawa Scale (NOS) for assessing the quality if nonrandomized studies in meta-analyses. Available at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed 16 Dec 2015.
79. Deeks J, Dinnes J, D'Amico R, et al. Evaluating non-randomized intervention studies. Health Technology Assessment 2003; 7(27). Available at: http://www.journalslibrary.nihr.ac.uk/__data/assets/pdf_file/0007/64933/FullReport- hta7270.pdf. Accessed 16 Dec 2015.
80. Cummings SR, Bates D, Black DM. Clinical use of bone densitometry: scientific review. JAMA 2002;288:1889-97.
81. Harrell FE Jr. Regression Modeling Strategies: with applications to linear models, logistical regression, and survival analysis. New York: Springer-Verlag New York, Inc.; 2001.
82. Fairclough D. Design and Analysis of Quality of Life Studies in Clinical Trials. Boca Raton, Florida: Chapman & Hall/CRC; 2002.
83. Egger M, Smith GD, Altman DG. Systematic Reviews in Health Care: Meta-Analysis in Context. 2nd ed. London: BMJ Publishing Group; 2001.
24 Appendix 4. Study Eligibility Form
Level of review: =Yes Title Abstract Article
Article Information: Author Last name and first initial: Journal Year of publication Volume Issue Pages Language English Other:______ Sent for translation (if included)
Inclusion Criteria: Publications: =Yes Original study Letter/Commentaries with original data Yes, Authors were contacted for data Dissertations Yes, Authors were contacted to resolve duplicate publications Study designs: Experimental study designs: Observational study designs: RCT Retrospective cohort study Quasi-randomized trial Prospective cohort study Crossover trial Nested case-cohort study Before-after trial Nested case-cohort study Case-control study (retrospective) Study Population: Pregnant population EXCLUDE Non-pregnant patients Exposure to Heparin > 90 days: Yes Cannot tell contact the author No, short-term EXCLUDE Measured (reported) one or more of the following outcomes: YES BMD measured by: DXA (Dual X-Ray Absorptiometry) SPA (Single Photon Absorptiometry)
Fracture(s) Bone markers Bone quality parameters Bone strength parameters NO EXCLUDE
Exclusion Criteria: Descriptive Studies: case report Descriptive Studies: cross-sectional study Letters/Commentary/Abstracts with no original data Reviews Other: ______
25 Appendix 5. PRISMA Checklist
Section/topic Checklist item Reported on page #
TITLE
Title Identify the report as a systematic review, meta-analysis, or both. 1 ABSTRACT
Structured summary Provide a structured summary including, as applicable: background; objectives; data sources; study eligibility criteria, participants, and interventions; 2 study appraisal and synthesis methods; results; limitations; conclusions and implications of key findings; systematic review registration number.
INTRODUCTION
Rationale Describe the rationale for the review in the context of what is already known. 4
Objectives Provide an explicit statement of questions being addressed with reference to participants, interventions, comparisons, outcomes, and study design 5 (PICOS). METHODS
Protocol and registration Indicate if a review protocol exists, if and where it can be accessed (e.g., Web address), and, if available, provide registration information including Appendix 1, 3 registration number. Eligibility criteria Specify study characteristics (e.g., PICOS, length of follow-up) and report characteristics (e.g., years considered, language, publication status) used as 5 criteria for eligibility, giving rationale.
Information sources Describe all information sources (e.g., databases with dates of coverage, contact with study authors to identify additional studies) in the search and 5, Appendix 1 &2 date last searched.
Search Present full electronic search strategy for at least one database, including any limits used, such that it could be repeated. Appendix 2
Study selection State the process for selecting studies (i.e., screening, eligibility, included in systematic review, and, if applicable, included in the meta-analysis). 5, Appendix 1
Data collection process Describe method of data extraction from reports (e.g., piloted forms, independently, in duplicate) and any processes for obtaining and confirming data 5, Appendices 1&4 from investigators.
Data items List and define all variables for which data were sought (e.g., PICOS, funding sources) and any assumptions and simplifications made. 5, Appendix 1
Risk of bias in individual studies Describe methods used for assessing risk of bias of individual studies (including specification of whether this was done at the study or outcome level), 5, Appendix 1, and how this information is to be used in any data synthesis. Appendix1 Tables 1 &2 Summary measures State the principal summary measures (e.g., risk ratio, difference in means). 6
26 Synthesis of results Describe the methods of handling data and combining results of studies, if done, including measures of consistency (e.g., I2) for each meta-analysis. 6
Reported on Section/topic # Checklist item page #
Risk of bias across studies 15 Specify any assessment of risk of bias that may affect the cumulative evidence (e.g., publication bias, selective reporting within studies). 5-6, Appendix 1
Additional analyses 16 Describe methods of additional analyses (e.g., sensitivity or subgroup analyses, meta-regression), if done, indicating which were pre-specified. 6,
RESULTS
Study selection 17 Give numbers of studies screened, assessed for eligibility, and included in the review, with reasons for exclusions at each stage, ideally with a flow diagram. 6-7, Fig1
Study characteristics 18 For each study, present characteristics for which data were extracted (e.g., study size, PICOS, follow-up period) and provide the citations. 7-8, Table 1
Risk of bias within studies 19 Present data on risk of bias of each study and, if available, any outcome level assessment (see item 12). 8, Tables 2&3
Results of individual studies 20 For all outcomes considered (benefits or harms), present, for each study: (a) simple summary data for each intervention group (b) effect estimates and 8-11, Fig2 confidence intervals, ideally with a forest plot.
Synthesis of results 21 Present results of each meta-analysis done, including confidence intervals and measures of consistency. 9, Fig2
Risk of bias across studies 22 Present results of any assessment of risk of bias across studies (see Item 15). 9-10,
Additional analysis 23 Give results of additional analyses, if done (e.g., sensitivity or subgroup analyses, meta-regression [see Item 16]). 9-10, Figs2b&2c
DISCUSSION
Summary of evidence 24 Summarize the main findings including the strength of evidence for each main outcome; consider their relevance to key groups (e.g., healthcare providers, 11-12 users, and policy makers).
Limitations 25 Discuss limitations at study and outcome level (e.g., risk of bias), and at review-level (e.g., incomplete retrieval of identified research, reporting bias). 12-15
Conclusions 26 Provide a general interpretation of the results in the context of other evidence, and implications for future research. 16
FUNDING
Funding 27 Describe sources of funding for the systematic review and other support (e.g., supply of data); role of funders for the systematic review. 17
27 From: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med 6(6): e1000097. doi:10.1371/journal.pmed1000097 for more information, visit: www.prisma-statement.org.
28