BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
Intravesical administration of combined hyaluronic acid (HA) and chondroitin sulphate (CS) for the treatment of female recurrent urinary tract infections: a European multicenter nested case-control study For peer review only
Journal: BMJ Open
Manuscript ID: bmjopen-2015-009669
Article Type: Research
Date Submitted by the Author: 10-Aug-2015
Complete List of Authors: Ciani, Oriana; University of Exeter , University of Exeter Medical School; CeRGAS Bocconi, Arendsen, Erik; Diaconessenhuis, Dept. of Urology Romancik, Martin; St. Cyril and Method University Hospital, Dept. of Urology Lunik, Richard; Fakultná nemocnica s poliklinikou, Dept. of Urology Costantini, Elisabetta; University of Perugia, Dept. of Surgical and Biomedical Science Di Biase, Manuel; University of Perugia, Dept. of Surgical and Biomedical Science
Morgia, Giuseppe; University of Catania, Dept. of Urology http://bmjopen.bmj.com/ Fragala', Eugenia; University of Catania, Dept. of Urology Tomaskin, Roman; Jessenius School of Medicine, Dept. of Urology Bernat, Marian; Fakultná nemocnica s poliklinikou, Dept. of Urology Guazzoni, Giorgio; Humanitas Clinical and Research Center, Dept. of Urology Tarricone, Rosanna; Bocconi University, Dept. of Policy Analysis and Public Management Lazzeri, Massimo; Humanitas Clinical and Research Center, Dept. of
Urology on September 29, 2021 by guest. Protected copyright.
Primary Subject Urology Heading:
Secondary Subject Heading: Infectious diseases
Urinary tract infections < UROLOGY, Antimicrobial Resistance, Hyaluronic Keywords: acid, chondroitin sulphate
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 25 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 4 5 Intravesical administration of combined hyaluronic 6 7 8 acid (HA) and chondroitin sulphate (CS) for the 9 10 treatment of female recurrent urinary tract infections: 11 12 a European multicenter nested case-control study 13 14 15 For1 peer2 review 3 only 4 16 Oriana Ciani , Erik Arendsen , Martin Romancik , Richard Lunik , Elisabetta 17 Costantini5, Manuel Di Biase 5, Giuseppe Morgia6, Eugenia Fragalà6, 18 7 8 9 19 Tomaskin Roman , Marian Bernat , Giorgio Guazzoni , Rosanna 20 1,10 9 21 Tarricone , Massimo Lazzeri 22 23 1 24 Centre For Research on Health and Social Care Management, Università 25 26 Bocconi, via Rontgen, 1, 20136, Milan, Italy 27 2 Diaconessenhuis, Dept. of Urology, Leiden, The Netherlands 28 3 29 St. Cyril and Method University Hospital, Dept. of Urology, Bratislava, 30 31 Slovakia 32 4 Fakultná nemocnica s poliklinikou, Dept. of Urology, Prešov, Slovakia 33 5 http://bmjopen.bmj.com/ 34 University of Perugia, Dept. of Surgical and Biomedical Science, Urology 35 36 and Andrology Clinic, Perugia, Italy 37 6 University of Catania, Dept. of Urology, Catania, Italy 38 7 39 Jessenius School of Medicine, University Hospital, Dept. of Urology, Martin, 40 41 Slovakia 42 8 Fakultná nemocnica s poliklinikou, Dept of Urology, Nové Zámky, Slovakia on September 29, 2021 by guest. Protected copyright. 43 9 44 Humanitas Clinical and Research Center, Humanitas University, Department 45 46 of Urology, Milan, Italy 47 10 Università Bocconi, Department of Policy Analysis and Public Management, 48 49 Via Roentgen, 1, 20136 Milano, Italy 50 51 Correspondence to: [email protected] 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 1 BMJ Open Page 2 of 25 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Abstract 4 5 Objectives: To compare the clinical effectiveness of the intravesical 6 administration of combined hyaluronic acid and chondroitin sulphate (HA + 7 8 CS) vs. current standard management in adult women with recurrent urinary 9 10 tract infections (RUTIs). 11 Setting: A EU based, multicenter, retrospective nested case control study. 12 13 Participants: 276 adult women treated for RUTIs starting from 2009 to 2013. 14 15 Interventions:For Patients peer treated withreview either intravesical only administration of HA + 16 CS or standard of care (antimicrobial/immunoactive 17 18 prophylaxis/probiotics/cranberry). 19 20 Primary and secondary outcome measures: The primary outcome was 21 occurrence of bacteriologically confirmed recurrence within 12 months. 22 23 Secondary outcomes were time to recurrence, total number of recurrences, 24 25 health related quality of life and healthcare resource consumption. Crude and 26 adjusted results for unbalanced characteristics are presented. 27 28 Results: 181 patients treated with HA + CS and 95 patients treated with 29 30 standard of care from 7 centers were included. The crude and adjusted OR 31 32 (95% CI) for the primary endpoint were 0.77 (0.46 to 1.28) and 0.51 (0.27 to 33
0.96), respectively. The benefit of intravesical HA + CS therapy improves http://bmjopen.bmj.com/ 34 35 when the number of instillations is ≥ 5. 36 37 Conclusions: In real world setting, bladder instillations of combined HA + CS 38 reduces by 49% the risk of bacteriologically confirmed recurrences compared 39 40 to the current standard management of RUTIs. Total incidence rates and 41 42 hazard rates were instead non significantly different between the two groups on September 29, 2021 by guest. Protected copyright. 43 after adjusting for unbalanced factors. In contrast to what happens with 44 45 antibiotic prophylaxis, the effectiveness of HA + CS therapy improves over 46 47 time. In an era of worryingly increased antimicrobial resistance, these findings 48 confirm the promising role of the HA + CS reinstatement therapy for 49 50 prevention and treatment of RUTIs. 51 52 Trial registration: ClinicalTrials.gov NCT02016118 53 54 55 Article Summary 56 57 'Strengths and limitations of this study' 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 2 Page 3 of 25 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 • Real world data show bladder instillations of combined hyaluronic acid 4 5 and chondroitin sulphate reduce the risk of bacteriologically confirmed 6 urinary tract infections vs current standard management 7 8 • Number of instillations is an important marker of success for this non 9 10 antimicrobial therapy 11 12 • Due to the retrospective observational design, these findings need 13 confirmation from prospective and preferably randomised studies 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 3 BMJ Open Page 4 of 25 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 4 Background 5 6 Urinary tract infection (UTI) is a major healthcare concern in women with an 7 annual incidence of 30 per 1000 (1). Nearly 33% of women will have had at 8 9 least one UTI episode, with characteristics of acute cystitis, requiring 10 11 antimicrobial therapy by the age of 24 years and as many as 60% of women 12 report having had a UTI in their lifetime (2, 3). 13 14 UTIs have a propensity to recur (4, 5); evidence shows that between 24% and 15 For peer review only 16 50% of initial episodes are followed by a second infection within six months 17 (6 9). A widely accepted definition of RUTIs is two or more UTI episodes over 18 19 six months, or three or more episodes over 12 months (10). On a population 20 21 scale, high incidence and prevalence of RUTIs results in considerable 22 healthcare costs, at the individual level the impact of this condition on health 23 24 related quality of life (HRQoL) is not negligible (11 13). 25 26 The pathogenesis of RUTI involves colonisation of the vagina with 27 uropathogenic bacteria and subsequent migration per urethra to the bladder. 28 29 About 68 77% of recurrences caused by Escherichia coli involve strains 30 31 genetically indistinguishable from those that caused previous infections (14). 32 The diagnosis is often made on clinical presentation with local genitourinary 33 http://bmjopen.bmj.com/ 34 symptoms of dysuria, frequency, and urgency or hesitancy appearing 35 36 suddenly (14). However, urine culture is useful in women presenting with 37 RUTI to confirm the diagnosis, direct antimicrobial therapy and exclude 38 39 infection from overactive bladder or interstitial cystitis (15). Evidence based 40 41 clinical practice guidelines recommend empiric initial therapy for acute on September 29, 2021 by guest. Protected copyright. 42 management or continuous antimicrobial therapy or self initiated therapy and 43 44 prophylaxis, either antimicrobial or non antimicrobial based (16, 17). 45 46 The choice of specific strategy for care depends on the number of 47 recurrences experienced per year, the patient’s preferences and careful 48 49 review of modifiable risk factors (14, 18). As the second most common reason 50 51 for prescribing antibiotics (following otitis media), there is currently increasing 52 concern about empiric use of these agents due to increased antimicrobial 53 54 resistance. Antibiotic use selects for resistant pathogens: a major risk factor 55 56 for an antibiotic resistant UTI is prior antibiotic use (5). In an international 57 survey investigating the prevalence and susceptibility of pathogens causing 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 4 Page 5 of 25 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 cystitis, 10.3% of Escherichia coli isolates were resistant to at least three 4 5 different classes of antimicrobial agents, including ampicillin (48.3%), 6 trimethoprim/sulfamethoxazole (29.4%) and nalidixic acid (18.6%) (19). 7 8 Nonantimicrobial prevention strategies have become popular in the age of 9 10 increasing antimicrobial use and resistance. However, no probiotic agent has 11 been approved for therapeutic use and the potential benefit of cranberry in 12 13 terms of product type (solid vs liquid), dosing, and optimal patient population 14 15 remainsFor to be elucidated peer (14, 18). review A new therapy based only on the reinstatement 16 of the glycosaminoglycan (GAG) bladder epithelium has recently been 17 18 proposed for the treatment of RUTIs (20). This GAG layer consists of non 19 20 sulphated, e.g. hyaluronic acid (HA), and sulphated, e.g. heparan sulphate 21 and heparin, chondroitin sulphate (CS), dermatan sulphate and keratan 22 23 sulphate, GAGs. Limited evidence has shown the preventive activity of 24 25 intravesical GAG substation therapy (with HA alone or with HA + CS) on 26 recurrence of infections in patients with recurrent bacterial cystitis (21). 27 28 However, large scale studies are needed to underline the benefit of this 29 30 therapy (22). 31 Therefore, we decided to perform a European retrospective multicenter study 32 33 to compare the clinical effectiveness of the intravesical administration of http://bmjopen.bmj.com/ 34 35 combined HA + CS (Ialuril®, IBSA Institut Biochimique SA) vs. current 36 standard management of RUTIs in adult women. 37 38 39 40 41 Methods 42 on September 29, 2021 by guest. Protected copyright. 43 Study design 44 45 This was a EU based, multicenter, retrospective nested case control 46 comparison of prospectively collected data based on individual patient 47 48 electronic medical records and/or administrative databases available at the 49 50 participating institutions. Centers using the intravesical administration of 51 combined HA + CS, in the countries where Ialuril® was already registered and 52 53 on the market (Ialuril® received a CE mark for this indication in 2009), were 54 55 identified and invited to take part in the study. 56 57 58 Study population 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 5 BMJ Open Page 6 of 25 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 All patients treated with either HA + CS or standard of care at the participating 4 5 centers starting from 2009 were included if they were women, aged 18 75 6 years, diagnosed with RUTIs, defined as at least three episodes of 7 8 uncomplicated UTIs accompanied by clinical symptoms and documented by 9 3 10 urine culture with the isolation of >10 CFU/ml of an identified pathogen in the 11 last 12 months. Uncomplicated UTI is defined as an infection in a person with 12 13 normal urinary tract and function (17). Women with complicated UTIs (i.e. 14 15 individualsFor with functional peer or structural review abnormalities ofonly the genitourinary tract) 16 were excluded. Based on a previous cohort study (11), we estimated that 208 17 18 patients were needed to observe a 50% difference in the proportions of 19 20 patients recurring between the two groups within 12 months with 90% power 21 and alpha level of 0.05. 22 23 24 25 Groups and interventions 26 Patients treated with intravesical administration of combined HA 1.6% and CS 27 28 2.0%. The recommended scheme is one instillation per week for the first 29 30 month, followed by one instillation every two weeks for the second month and 31 one instillation per month afterward until stable remission of the symptoms, 32 33 however different patterns are seen in clinical practice. These patients were http://bmjopen.bmj.com/ 34 35 compared to patients treated with antimicrobial prophylaxis (continuous or 36 postcoital), or immunoactive prophylaxis or prophylaxis with probiotics or 37 38 prophylaxis with cranberry, or combination of these (17), as recommended by 39 40 the European Association of Urology. 41
on September 29, 2021 by guest. Protected copyright. 42 43 Study outcomes 44 45 The primary outcome for this study was the occurrence of objective urinary 46 tract infection recurrence, defined as the occurrence of at least one 47 48 bacteriologically confirmed urinary tract infection within 12 months after 49 50 treatment initiation for RUTIs. Information about clinically confirmed 51 recurrences were also sought although they are not reported in this 52 53 manuscript as they are assumed to be less objective than the bacteriologically 54 55 confirmed ones. The secondary outcome measures were the time to 56 recurrence (defined as the time from the start of the treatment until the 57 58 occurrence of the first objective recurrence); the total number of recurrences; 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 6 Page 7 of 25 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 HRQoL as assessed through the SF 36 (23) or EQ 5D (24) questionnaires. 4 5 Dutch (25), Italian (26) and UK (27) tariffs were used to estimate utility values 6 from the EQ 5D questionnaires in the Netherlands, Italy and Slovakia, 7 8 respectively. Information about healthcare resource consumption were also 9 10 collected. A cost analysis was planned and will be the subject of a future 11 publication. 12 13 14 15 Data collectionFor peer review only 16 General patient demographic characteristics, diagnosis and treatment 17 18 information were collected based on a predefined form designed on the input 19 20 obtained from collaborating centers during a workshop held in July 2013. An 21 intuitive electronic system was implemented (Advice Pharma ltd) to record 22 23 and store data on a secure remote server provider. 24 25 26 Statistical analyses 27 28 Continuous baseline characteristics are presented as median and interquartile 29 30 range (IQR) or mean and standard deviation (SD), as appropriate. For 31 proportions, absolute and relative frequencies are reported. Wilcoxon–Mann– 32 33 Whitney test or t test were used for continuous and ordinal variables baseline http://bmjopen.bmj.com/ 34 35 differences, whereas the Chi Square test was used for proportions. In our 36 primary analyses, we applied univariate and multivariate logistic, poisson and 37 38 Cox model for objective recurrence, number of recurrences and time to 39 40 recurrence outcomes, respectively. Results were presented as crude and 41
adjusted odds ratio (OR), incidence rate ratio (IRR) and hazard ratio (HR) on September 29, 2021 by guest. Protected copyright. 42 43 respectively, with their 95% confidence intervals (CI). A prespecified 44 45 sensitivity analysis was conducted to investigate the impact of adherence to 46 HA + CS treatment on clinical outcomes considering patients who had ≥ 5 47 48 instillations. Pairwise deletion was used to deal with missing data. All 49 50 significance tests were two tailed at the 0.05 significance level. All the 51 analyses were conducted using Stata SE StataCorp LP 11. 52 53 54 55 Ethical approval and funding 56 The study protocol was reviewed and approved at the coordinating center by 57 58 an Independent Ethics Committee at the Department of Urology, University of 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 7 BMJ Open Page 8 of 25 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Perugia. The study was funded by the Study Group for Urogenital Diseases in 4 5 Italy, an association on practioners and clinical urologists. The study is 6 registered at clinicaltrials.gov (NCT02016118). Extra data is available by 7 8 emailing the corresponding author. 9 10 11 12 Results 13 14 Overall, 276 patients treated for RUTIs at seven European centers from 15 For peer review only 16 January 2009 up to December 2013 were included in the analyses. Of these, 17 181 women were treated with HA + CS intravesical administration and 95 18 19 women received standard management of RUTIs. Baseline socio 20 21 demographic and clinical characteristics of patients are reported in Table 1. 22 Given the non experimental nature of the study, the distribution of several 23 24 characteristics was not homogeneous between the two groups, in particular 25 26 women treated with HA + CS were older, with a higher body mass index and 27 probability of dyspareunia. 28 29 30 31 Primary analyses 32 Primary analyses showed that the adjusted OR (95% CI) for developing a 33 http://bmjopen.bmj.com/ 34 bacteriologically confirmed recurrence within 12 months was 0.51 (0.27 to 35 36 0.96), meaning that, other characteristics being equal, there is a 49% reduced 37 risk of developing a recurrence in patients treated with HA + CS compared to 38 39 standard care (Table 2). When the number of recurrences is considered, we 40 41 observed an adjusted IRR (95% CI) of 0.99 (0.69 to 1.43), showing non on September 29, 2021 by guest. Protected copyright. 42 significant differences in the incidence rates between the group treated with 43 44 HA + CS and the control. Similar results were obtained from the univariate 45 46 and multivariate Cox regression models used to estimate the HR (95% CI) for 47 the time to first bacteriologically confirmed recurrence, with an unadjusted 48 49 estimate of 0.99 (0.61 to 1.61) (Table 2). Although during the 12 months 50 51 follow up the median time to first recurrence was higher in the standard care 52 group (169.5 days (IQR, 72.5 to 341.5) vs 320 days (IQR, 179 to 365); p 53 54 value < 0.001), we observed the distribution of recurrences at separate follow 55 56 up times, and noted that the incident proportion of patients who developed the 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 8 Page 9 of 25 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 recurrence vs those still at risk was lower in the HA + CS group in the latest 4 5 part of follow up, after 8 months (Table 3). 6 7 8 Health related quality of life and resources consumption 9 10 In a subset of patients, a measure of the HRQoL as measured through the 11 SF 36 or EQ 5D 3L questionnaires was available at baseline and after 12 12 13 months follow up. There was no evidence of better improvement in HRQoL in 14 15 the HA +For CS group peercompared to controlreview with SF 36 results, only whereas when 16 EQ 5D data were considered the HA + CS group seemed to have received a 17 18 higher benefit in terms of HRQoL than the control (Table 4). 19 20 There is a general reduction in physical units of health resources (i.e. medical 21 visits, laboratory and imaging tests) consumed by the two groups before the 22 23 treatment and during the follow up (Supplementary table 1) without significant 24 25 differences between the two groups. 26 27 28 Sensitivity analysis 29 30 We repeated all primary analyses and considered different exposure intensity 31 (i.e. number of intravesical administrations received) in the HA + CS group. All 32 33 findings consistently show additional benefit gained to patients when the http://bmjopen.bmj.com/ 34 35 number of instillations increases, revealing the importance of adherence to 36 this medical device therapy for the treatment of RUTIs (Table 5). 37 38 39 40 41 Discussion on September 29, 2021 by guest. Protected copyright. 42 In this European multicenter retrospective observational study we compared 43 44 bacteriologically confirmed recurrence rates at 12 months follow up after the 45 46 initiation of intravesical administration of HA + CS vs standard of care for the 47 treatment of RUTIs. After adjusting for unbalanced confounding factors 48 49 between the two groups, we observed that the HA + CS patients had a 49% 50 51 reduction (OR 0.51, 95% CI: 0.27 to 0.96) in the risk of a bacteriologically 52 confirmed recurrence, whereas there was no statistical evidence for a 53 54 difference in the incidence rates and hazard rates of such recurrences 55 56 between the two groups. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 9 BMJ Open Page 10 of 25 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Four clinical studies (28 31) have been performed to investigate the efficacy 4 5 and tolerability of intravesically administered GAG for RUTI prophylaxis, all 6 showing that HA alone or HA + CS instillations reduce the number of UTI per 7 8 patient per year at no increased risk of severe adverse events and prolong the 9 10 time interval between RUTI episodes, with a high rate of patients being free of 11 recurrence at the end of the study period. In particular, two RCTs studies 12 13 compared HA + CS administration to either placebo (29) or long term 14 15 antibioticFor prophylaxis peer using sulfamethoxazole review 200 mg onlyand trimethoprim 40 mg 16 (30). Damiano and colleagues report a decrease in UTI rate per patient of 17 18 77% (95% CI, 72.3 to 80.8) in the experimental vs placebo group, whereas De 19 20 Vita et al. report the mean ± SD number of recurrent cystitis per patient per 21 year as 1 ± 1.2 vs 2.3 ± 1.4 in HA + CS and antibiotic treated patients, 22 23 respectively. 24 25 Despite the prospective and randomized design, these trials were limited by 26 the small sample size (i.e. they included 57 and 28 patients, respectively) and 27 28 the single center setting that considerably reduces the generalizability of the 29 30 findings. In this respect, our study involved seven centers across three 31 European countries and 276 patients, thus providing important additional 32 33 evidence about currently available treatment options for RUTIs. Furthermore, http://bmjopen.bmj.com/ 34 35 the non experimental observational design allows for a closer representation 36 of the routine clinical practice of the use of the HA + CS reinstatement therapy 37 38 as compared to standard of care in place at high volume university hospitals, 39 40 that is on purpose defined as very broad given the variety of recommended 41
strategies (17) and general scarce adherence to clinical guidelines (32). on September 29, 2021 by guest. Protected copyright. 42 43 On the other hand, the retrospective design limited the availability of data to 44 45 that previously collected at the centers. Contacting patients ex post to gather 46 additional data was not applicable (e.g. as in the case of HRQoL assessment) 47 48 or not helpful given the potential significant recall bias introduced by delayed 49 50 reporting. In this regard, the issue of missing data was dealt with by assuming 51 they were missing at random and applying pairwise deletion. 52 53 Health related quality of life assessment in routine practice is still uncommon 54 55 as indicated by the significant proportion of missing information (up to 73% in 56 the control group) on this outcome. However our findings are in line with 57 58 previous reports showing that the GAG replacement treatment in women with 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 10 Page 11 of 25 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 RUTIs had a positive impact on patients’ quality of life, reducing the 4 5 symptoms and improving the maximum cystometric capacity (29, 30). 6 Recurrent UTIs in women are a common condition, associated with significant 7 8 morbidity and burden for the whole society. In a study of 684 women aged 18 9 10 to 70 years with UTI, participants reported an average of 3.83 symptom days, 11 2.89 restricted activity days, and 3.13 days during which they were unwell 12 13 (33). In another study patients reported 1.2 days in which they were not able 14 15 to attendFor classes orpeer work, and 0.4 review days in bed (34). New only effective prevention 16 strategies are needed, in particular non antimicrobial approaches would be 17 18 desirable for several reasons. First of all, a prolonged antibiotic use as a 19 20 prophylactic approach to RUTI increases the risk of side effects, including 21 vaginal and oral candidiasis, and gastrointestinal symptoms (10). This in turn, 22 23 lowers patients’ compliance and therefore the effectiveness of the treatment 24 25 (35). However, the most worrying effect of the antibiotic use (and misuse) is 26 the exacerbation of the antimicrobial resistance (AMR) (36). Recently, a UK 27 28 commissioned report on health and macroeconomic consequences of AMR 29 30 estimated 10 million extra deaths a year and costs up to €90tn for the global 31 economy by 2050 if this problem is not properly tackled (37). Although this 32 33 first report might not be as scientifically rigorous or informed by evidence as http://bmjopen.bmj.com/ 34 35 possible (38), it brought renewed interest in the worldwide AMR crisis. The 36 war against the spread of drug resistant microbes is attracting considerable 37 38 attention as well as investment from all major governments and research 39 40 organizations in EU and beyond (39,40). The way forward outlined by all of 41
these major research initiatives includes establishing appropriate funding and on September 29, 2021 by guest. Protected copyright. 42 43 rewards to subsidize access to and development of new antibiotic agents, 44 45 preservation of existing drugs antimicrobial activity through prescription 46 tailored to diagnosis, prioritization and controlled access, and identification of 47 48 novel approaches and therapies for microbial diseases. 49 50 The case of GAG reinstatement therapy is a good example of an innovative 51 approach to prevent and manage bacterial urinary infections, a medical 52 53 device intervention as opposed to a drug. In contrast to antibiotic therapy, 54 55 which aims at eradicating pathogens, treatment with HA + CS targets 56 bacterial adherence to the bladder mucosa by physically recovering a 57 58 damaged GAG layer that facilitates bacterial adherence and, therefore, 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 11 BMJ Open Page 12 of 25 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 RUTIs. As regards the economic profile of the two alternative approaches, it 4 5 has been reported that the cost of the HA + CS could be even five times 6 higher than the cost for a 6 month antibiotic prophylaxis. However, this 7 8 consideration corresponds to a very restrictive, if not naïve, cost analysis 9 10 as it is well known that all direct healthcare costs and consequences, 11 including those for the wider society, as would be containment of drug 12 13 resistance spreading, should be taken into account when assessing the cost 14 15 effectivenessFor profiles peer of health technologies. review As part ofonly this study, an 16 evaluation of direct healthcare resource consumption showed similar profiles 17 18 between the two groups. 19 20 21 22 Conclusions 23 24 In order to treat and prevent RUTI, there is a need of effective and safe 25 26 alternative strategies to antimicrobial therapy. Our study showed that in real 27 world setting bladder instillation of combined HA + CS reduces by 49% the 28 29 risk of bacteriologically confirmed recurrences compared to the current 30 31 standard management in this study population. Total incidence rates and 32 hazard rates were instead non significantly different between the two groups. 33 http://bmjopen.bmj.com/ 34 The number of HA + CS instillations seems to be an important marker of 35 36 success for the intravesical administration therapy. Furthermore, in contrast to 37 what happens with antibiotic prophylaxis, due to side effects and development 38 39 of resistance, the effectiveness of GAG reinstatement therapy improves over 40 41 time, with an even better expected comparative effectiveness profile in the on September 29, 2021 by guest. Protected copyright. 42 long run. 43 44 Although firm conclusions are difficult due to the retrospective observational 45 46 design, these findings highlight the relevance of additional prospective and 47 randomized studies in this area and the promising role of the HA + CS 48 49 reinstatement therapy for prevention and treatment of RUTI in an era of 50 51 worryingly increased antimicrobial resistance. 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 12 Page 13 of 25 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 4 References 5 6 7 1. Laupland KB, Ross T, Pitout JD, Church DL, Gregson DB. Community 8 onset urinary tract infections: a population based assessment. Infection. 9 10 2007;35(3):150 3. 11 2. Foxman B. Epidemiology of urinary tract infections: incidence, 12 morbidity, and economic costs. The American journal of medicine. 2002;113 13 Suppl 1A:5s 13s. 14 3. Foxman B, Barlow R, D'Arcy H, Gillespie B, Sobel JD. Urinary tract 15 infection:For self reported peer incidence andreview associated costs. only Annals of 16 epidemiology. 2000;10(8):509 15. 17 4. Hooton TM, Scholes D, Hughes JP, Winter C, Roberts PL, Stapleton 18 AE, et al. A prospective study of risk factors for symptomatic urinary tract 19 20 infection in young women. The New England journal of medicine. 21 1996;335(7):468 74. 22 5. Foxman B. The epidemiology of urinary tract infection. Nature reviews 23 Urology. 2010;7(12):653 60. 24 6. Sen A. Recurrent cystitis in non pregnant women. BMJ clinical 25 evidence. 2008;2008. 26 7. Ikaheimo R, Siitonen A, Heiskanen T, Karkkainen U, Kuosmanen P, 27 Lipponen P, et al. Recurrence of urinary tract infection in a primary care 28 setting: analysis of a 1 year follow up of 179 women. Clinical infectious 29 30 diseases : an official publication of the Infectious Diseases Society of 31 America. 1996;22(1):91 9. 32 8. Foxman B. Recurring urinary tract infection: incidence and risk factors. 33 American journal of public health. 1990;80(3):331 3. http://bmjopen.bmj.com/ 34 9. Foxman B, Gillespie B, Koopman J, Zhang L, Palin K, Tallman P, et al. 35 Risk factors for second urinary tract infection among college women. 36 American journal of epidemiology. 2000;151(12):1194 205. 37 10. Albert X, Huertas I, Pereiro, II, Sanfelix J, Gosalbes V, Perrota C. 38 Antibiotics for preventing recurrent urinary tract infection in non pregnant 39 40 women. The Cochrane database of systematic reviews. 2004(3):Cd001209. 41 11. Ciani O, Grassi D, Tarricone R. An economic perspective on urinary 42 tract infection: the "costs of resignation". Clinical drug investigation. on September 29, 2021 by guest. Protected copyright. 43 2013;33(4):255 61. 44 12. Renard J, Ballarini S, Mascarenhas T, Zahran M, Quimper E, Choucair 45 J, et al. Recurrent Lower Urinary Tract Infections Have a Detrimental Effect on 46 Patient Quality of Life: a Prospective, Observational Study. Infectious 47 diseases and therapy. 2014. 48 13. Bermingham SL, Ashe JF. Systematic review of the impact of urinary 49 50 tract infections on health related quality of life. BJU international. 2012;110(11 51 Pt C):E830 6. 52 14. Gupta K, Trautner BW. Diagnosis and management of recurrent 53 urinary tract infections in non pregnant women. BMJ. 2013;346:f3140. 54 15. Arya LA, Northington GM, Asfaw T, Harvie H, Malykhina A. Evidence of 55 bladder oversensitivity in the absence of an infection in premenopausal 56 women with a history of recurrent urinary tract infections. BJU international. 57 2012;110(2):247 51. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 13 BMJ Open Page 14 of 25 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 16. Gupta K, Hooton TM, Naber KG, Wullt B, Colgan R, Miller LG, et al. 4 International clinical practice guidelines for the treatment of acute 5 uncomplicated cystitis and pyelonephritis in women: A 2010 update by the 6 Infectious Diseases Society of America and the European Society for 7 Microbiology and Infectious Diseases. Clinical infectious diseases : an official 8 9 publication of the Infectious Diseases Society of America. 2011;52(5):e103 10 20. 11 17. Grabe M, Bartoletti R, Bjerklund Johansen T, Cai T, Çek M, Köves B, 12 et al. Guidelines on Urological Infections. European Association of Urology, 13 2015. 14 18. Shepherd AK, Pottinger PS. Management of urinary tract infections in 15 the era ofFor increasing peer antimicrobial review resistance. The Medical only clinics of North 16 America. 2013;97(4):737 57, xii. 17 19. Schito GC, Naber KG, Botto H, Palou J, Mazzei T, Gualco L, et al. The 18 19 ARESC study: an international survey on the antimicrobial resistance of 20 pathogens involved in uncomplicated urinary tract infections. International 21 journal of antimicrobial agents. 2009;34(5):407 13. 22 20. Lazzeri M, Montorsi F. The therapeutic challenge of "chronic cystitis": 23 search well, work together, and gain results. European urology. 24 2011;60(1):78 80. 25 21. De Vita D, Antell H, Giordano S. Effectiveness of intravesical 26 hyaluronic acid with or without chondroitin sulfate for recurrent bacterial 27 cystitis in adult women: a meta analysis. International urogynecology journal. 28 29 2013;24(4):545 52. 30 22. Madersbacher H, van Ophoven A, van Kerrebroeck PE. GAG layer 31 replenishment therapy for chronic forms of cystitis with intravesical 32 glycosaminoglycans a review. Neurourology and urodynamics. 2013;32(1):9 33 18. http://bmjopen.bmj.com/ 34 23. Brazier JE, Harper R, Jones NM, O'Cathain A, Thomas KJ, Usherwood 35 T, et al. Validating the SF 36 health survey questionnaire: new outcome 36 measure for primary care. Bmj. 1992;305(6846):160 4. 37 24. EuroQol a new facility for the measurement of health related quality of 38 39 life. Health policy (Amsterdam, Netherlands). 1990;16(3):199 208. 40 25. Lamers LM, McDonnell J, Stalmeier PF, Krabbe PF, Busschbach JJ. 41 The Dutch tariff: results and arguments for an effective design for national 42 EQ 5D valuation studies. Health Econ. 2006;15(10):1121 32. on September 29, 2021 by guest. Protected copyright. 43 26. Scalone L, Cortesi PA, Ciampichini R, Belisari A, D'Angiolella LS, 44 Cesana G, et al. Italian population based values of EQ 5D health states. 45 Value in health : the journal of the International Society for 46 Pharmacoeconomics and Outcomes Research. 2013;16(5):814 22. 47 27. Dolan P. Modeling valuations for EuroQol health states. Medical care. 48 49 1997;35(11):1095 108. 50 28. Constantinides C, Manousakas T, Nikolopoulos P, Stanitsas A, 51 Haritopoulos K, Giannopoulos A. Prevention of recurrent bacterial cystitis by 52 intravesical administration of hyaluronic acid: a pilot study. BJU international. 53 2004;93(9):1262 6. 54 29. Damiano R, Quarto G, Bava I, Ucciero G, De Domenico R, Palumbo 55 MI, et al. Prevention of recurrent urinary tract infections by intravesical 56 administration of hyaluronic acid and chondroitin sulphate: a placebo 57 controlled randomised trial. European urology. 2011;59(4):645 51. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 14 Page 15 of 25 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 30. De Vita D, Giordano S. Effectiveness of intravesical hyaluronic 4 acid/chondroitin sulfate in recurrent bacterial cystitis: a randomized study. 5 International urogynecology journal. 2012;23(12):1707 13. 6 31. Lipovac M, Kurz C, Reithmayr F, Verhoeven HC, Huber JC, Imhof M. 7 Prevention of recurrent bacterial urinary tract infections by intravesical 8 9 instillation of hyaluronic acid. International journal of gynaecology and 10 obstetrics: the official organ of the International Federation of Gynaecology 11 and Obstetrics. 2007;96(3):192 5. 12 32. Taur Y, Smith MA. Adherence to the Infectious Diseases Society of 13 America guidelines in the treatment of uncomplicated urinary tract infection. 14 Clinical infectious diseases : an official publication of the Infectious Diseases 15 Society Forof America. peer 2007;44(6):769 74. review only 16 33. Little P, Merriman R, Turner S, Rumsby K, Warner G, Lowes JA, et al. 17 Presentation, pattern, and natural course of severe symptoms, and role of 18 19 antibiotics and antibiotic resistance among patients presenting with suspected 20 uncomplicated urinary tract infection in primary care: observational study. 21 Bmj. 2010;340:b5633. 22 34. Ronald A. The etiology of urinary tract infection: traditional and 23 emerging pathogens. The American journal of medicine. 2002;113 Suppl 24 1A:14S 9S. 25 35. Hooton TM, Levy SB. Antimicrobial resistance: a plan of action for 26 community practice. American family physician. 2001;63(6):1087 98. 27 36. Fauci AS, Marston l D. The perpetual challenge of antimicrobial 28 29 resistance. Jama. 2014;311(18):1853 4. 30 37. Review on Antimicrobial Resistance. Antimicrobial Resistance: 31 Tackling a crisis for the health and wealth of nations. 2014. 32 38. Antimicrobial resistance: in terms politicians understand. Lancet. 33 2014;384(9961):2173. http://bmjopen.bmj.com/ 34 39. Woolhouse M, Farrar J. Policy: An intergovernmental panel on 35 antimicrobial resistance. Nature. 2014;509(7502):555 7. 36 40. Geoghegan Quinn M. Funding for antimicrobial resistance research in 37 Europe. Lancet. 2014;384(9949):1186. 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 15 BMJ Open Page 16 of 25 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Contributorship: OC, RT, ML designed the study. EA, MR, RL, EC, MDB, 4 5 GM, EF, TR, MB, GG contributed to the data collection. OC analyzed the data 6 and drafted the manuscript. All authors commented and approved the final 7 8 version of the manuscript. 9 10 Ethics: The study was approved by the Comitato Etico delle Aziende 11 Sanitarie della Regione Umbria, protocol N. 2230/14. Given the retrospective 12 13 nature of the study, patient consent was not obtained but the presented data 14 15 are anonymisedFor and peer risk of identification review is low. only 16 Funding: This study was funded by an unrestricted grant from the TETI 17 18 Association Study group for urogenital diseases (www.tetiassociation.com). 19 20 The authors declare no competing conflict of interest. 21 Data sharing: The dataset is available by emailing the corresponding author. 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 16 Page 17 of 25 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 4 Table 1 Baseline characteristics 5 Characteristics HA + CS Standard care P - value 6 7 (N = 181) (N = 95) 8 9 Age – mean (SD) 55.24 (17.33) 48.84 (21.16) 0.017 10 Employed – n (%) 83 (46) 62 (65) 0.003 11 12 Partner – n (%) 148 (82) 72 (76) 0.172 13 14 Sexually active – n (%) 114 (63) 60 (63) 0.931 15 Menopause – n For(%) peer69 (28) review49 (51) only 0.038 16 17 BMI – mean (SD) 26.20 (6.37) 24.56 (5.03) 0.019 18 19 Postcoital infections – n (%) 31 (17) 29 (31) 0.012 20 21 Dyspareunia – n (%) 40 (22) 5 (5) < 0.001 22 Severity RUTI – median (IQR)* 4 (2 – 5) 4 (3 – 4) 0.316 23 24 FSFI – mean (SD) 5.35 (8.71) 3.13 (5.43) 0.018 25 a b 26 EQ 5D– median (IQR) 0.69 (0.24 0.73) 0.69 (0.28 0.81) 0.696 27 SF 36 PCS – median (IQR) 60 (54.5 70)c 60 (53 – 67.5)d 0.500 28 29 SF 36 MCS – median (IQR) 60 (54 70)e 60 (53 – 67.5)d 0.551 30 31 BMI = body mass index; EQ 5D = Euro QoL 5D 3 level; FSFI = female sexual 32 33 function index; IQR = interquartile range; SD = standard deviation; RUTI = http://bmjopen.bmj.com/ 34 recurrent urinary tract infection; SF 36 PCS = Short From 36 physical 35 36 component score; SF 36 MCS = Short From 36 mental component score 37 38 *According to the European Association of Urology Guidelines on Urological 39 Infections where 1 is low severity cystitis and 6 is extreme severity including 40 41 organ failure (21) on September 29, 2021 by guest. Protected copyright. 42 a b c d e 43 N = 90 patients; N = 29 patients; N = 72 patients; N = 60 patients; N = 73 44 patients 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 17 BMJ Open Page 18 of 25 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Table 2 Bacteriologically confirmed recurrence, total number of 4 5 recurrences and time to first recurrence between HA + CS vs standard 6 of care treated patients 7 8 9 Outcome OR (95% CI) Adjusted* OR (95% CI) 10 Bacteriologically 11 0.77 (0.46 to 1.28) 0.51 (0.27 to 0.96) 12 confirmed recurrence 13 14 IRR (95% CI) Adjusted* IRR (95% CI) 15 For peer review only 16 Total number of 17 bacteriologically 1.73 (1.27 to 2.37) 0.99 (0.69 to 1.43) 18 19 confirmed recurrence 20 21 HR (95% CI) Adjusted* HR (95% CI) 22 Time to first 23 24 bacteriologically 1.66 (1.09 to 2.54)§ 0.99 (0.61 to 1.61) 25 26 confirmed recurrence 27 OR = odds ratio, IRR = incidence rate ratio, HR = hazard ratio 28 29 *Adjusted for age, BMI, employment and menopause status, postcoital 30 31 infections, dyspareunia, FSFI and severity of RUTI 32 §Log rank test p value 0.018 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 18 Page 19 of 25 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Table 3 Incidence of bacteriologically confirmed recurrences during 12 4 5 months follow-up 6 7 Incidence of HA + CS Standard care P value 8 9 bacteriologically 10 11 confirmed 12 recurrences 13 14 0 – 90 days 15.3% 12.7% 0.610 15 For peer review only 16 91 – 180 days 15.6% 12.9% 0.628 17 181 – 240 days 10.7% 3.7% 0.132 18 19 241 – 365 days 16.3% 30.8% 0.043 20 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 19 BMJ Open Page 20 of 25 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Table 4 Health related quality of life assessment before treatment 4 initiation and after 12 months for the HA + CS and standard care group 5 6 Baseline 12 months Baseline 12 months P- 7 8 value 9 HA + CS (N = 79) Standard care (N = 26) 10 11 EQ 5D 3L 0.53 (0.36) 0.76 (0.25) 0.57 (0.31) 0.53 (0.40) 0.020 12 Mean (SD) 13 14 HA + CS (N = 83) Standard care (N = 27) 15 EQ 5D For59.53 peer(18.34) 75.88 review (16.90) 62.5 (11.67) only 66.84 (14.17) <0.001 16 17 VAS 18 Mean (SD) 19 20 HA + CS (N = 67) Standard care (N = 59) 21 22 SF 36 PCS 61.88 (10.14) 76.99 (11.69) 60.68 (10.02) 74.66 (13.32) 0.118 23 Mean (SD) 24 25 SF 36 MCS 61.74 (10.26) 77.35 (11.33) 60.68 (10.02) 74.66 (13.32) 0.105 26 Mean (SD) 27 28 MCS = Mental component score; PCS = Physical component score; SD = 29 standard deviation; VAS = visual analogue scale 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 20 Page 21 of 25 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Table 5 Sensitivity analysis – impact of number of intravesical 4 5 administration of HA + CS on clinical outcomes 6 7 Bacteriologically confirmed OR (95% CI) Adjusted* OR (95% CI) 8 9 recurrence 10 11 ≥ 5 instillations vs standard of care 0.81 (0.48 to 1.36) 0.52 (0.27 to 0.99) 12 ≥ 6 instillations vs standard of care 0.69 (0.40 to 1.18) 0.47 (0.25 to 0.91) 13 14 ≥ 7 instillations vs standard of care 0.63 (0.34 to 1.14) 0.43 (0.21 to 0.88) 15 For peer review only 16 Total number of IRR (95% CI) Adjusted* IRR (95% CI) 17 bacteriologically confirmed 18 19 recurrence 20 21 ≥ 5 instillations vs standard of care 1.82 (1.33 to 2.49) 1.05 (0.73 to 1.52) 22 ≥ 6 instillations vs standard of care 1.64 (1.18 to 2.28) 0.97 (0.66 to 1.40) 23 24 ≥ 7 instillations vs standard of care 1.46 (1 to 2.13) 0.90 (0.60 to 1.36) 25 26 Time to first bacteriologically HR (95% CI) Adjusted* HR (95% CI) 27 28 confirmed recurrence 29 ≥ 5 instillations vs standard of care 1.72 (1.12 to 2.65) 1.04 (0.64 to 1.71) 30 31 ≥ 6 instillations vs standard of care 1.55 (0.99 to 2.41) 0.96 (0.58 to 1.57) 32 33 ≥ 7 instillations vs standard of care 1.33 (0.79 to 2.22) 0.85 (0.49 to 1.47) http://bmjopen.bmj.com/ 34 OR = odds ratio, IRR = incidence rate ratio, HR = hazard ratio 35 36 *Adjusted for age, BMI, employment and menopause status, postcoital 37 38 infections, dyspareunia, FSFI and severity of RUTI 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 21 BMJ Open Page 22 of 25 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 STROBE Statement—Checklist of items that should be included in reports of case-control studies 3 Item Page 4 No Recommendation 5 Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the 1 6 7 abstract 8 (b) Provide in the abstract an informative and balanced summary of what 2 9 was done and what was found 10 11 Introduction 12 Background/rationale 2 Explain the scientific background and rationale for the investigation being 4-5 13 reported 14 Objectives 3 State specific objectives, including any prespecified hypotheses 5 15 For peer review only 16 Methods 17 Study design 4 Present key elements of study design early in the paper 5 18 Setting 5 Describe the setting, locations, and relevant dates, including periods of 5-6 19 recruitment, exposure, follow-up, and data collection 20 21 Participants 6 (a) Give the eligibility criteria, and the sources and methods of case 6 22 ascertainment and control selection. Give the rationale for the choice of 23 cases and controls 24 (b) For matched studies, give matching criteria and the number of controls NA 25 26 per case 27 Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, 6-7 28 and effect modifiers. Give diagnostic criteria, if applicable 29 Data sources/ 8* For each variable of interest, give sources of data and details of methods of 6-7 30 31 measurement assessment (measurement). Describe comparability of assessment methods 32 if there is more than one group 33 Bias 9 Describe any efforts to address potential sources of bias 7 http://bmjopen.bmj.com/ 34 Study size 10 Explain how the study size was arrived at 6 35 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If 7 36 37 applicable, describe which groupings were chosen and why 38 Statistical methods 12 (a) Describe all statistical methods, including those used to control for 7 39 confounding 40 (b) Describe any methods used to examine subgroups and interactions 7 41
c on September 29, 2021 by guest. Protected copyright. 42 ( ) Explain how missing data were addressed 7 43 (d) If applicable, explain how matching of cases and controls was addressed NA 44 (e) Describe any sensitivity analyses 7 45 46 Results 47 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers 8 48 potentially eligible, examined for eligibility, confirmed eligible, included in 49 the study, completing follow-up, and analysed 50 51 (b) Give reasons for non-participation at each stage 8 52 (c) Consider use of a flow diagram NA 53 Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, 8 54 social) and information on exposures and potential confounders 55 (b) Indicate number of participants with missing data for each variable of Tables 56 57 interest 58 Outcome data 15* Report numbers in each exposure category, or summary measures of 8 59 exposure 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml1 Page 23 of 25 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted Tables 3 estimates and their precision (eg, 95% confidence interval). Make clear 4 which confounders were adjusted for and why they were included 5 b 6 ( ) Report category boundaries when continuous variables were categorized Tables 7 (c) If relevant, consider translating estimates of relative risk into absolute NA 8 risk for a meaningful time period 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml2 BMJ Open Page 24 of 25 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity 9 4 analyses 5 6 7 Discussion 8 Key results 18 Summarise key results with reference to study objectives 9 9 Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or 10- 10 11 imprecision. Discuss both direction and magnitude of any potential bias 11 12 Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, 12 13 multiplicity of analyses, results from similar studies, and other relevant evidence 14 Generalisability 21 Discuss the generalisability (external validity) of the study results 10 15 For peer review only 16 Other information 17 Funding 22 Give the source of funding and the role of the funders for the present study and, if 8 18 applicable, for the original study on which the present article is based 19 20 21 *Give information separately for cases and controls. 22 23 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and 24 published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely 25 26 available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at 27 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is 28 available at http://www.strobe-statement.org. 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml3 Page 25 of 25 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Supplementary Table 1 Resource consumption before treatment and 4 after 12 months follow-up. Median (IQR) 5 6 Resources HA + CS Standard Care P Value* 7 8 (N=181) (N=95) 9 N. GP visits Before treatment 5 (3-6) 3 (3-4) 10 11 Follow-up 2 (1.5-3) 0 (0-1) 0.964 12 13 N. Specialist visits 6 (2-10) 6 (4-8) 14 2 (1-5) 4 (4-4) 0.644 15 For peer review only 16 N. Urine microscopic analysis 5 (3-7) 6 (6-8) 17 18 2 (0-3) 4 (1-4) 0.219 19 20 N. Urine culture 4.5 (3-7) 6 (5-8)
21 2 (1-4) 1 (0-3) <0.001 22 23 N. Urine cytology 2 (1-3) 3 (1-3) 24 25 1 (1-1) 0 (0-1) 0.003 26 N. Vaginal swab test 3 (1-3) 3 (1-3) 27 28 0 (0-0) 0 (0-0) 0.809 29 30 N. Transabdominal ultrasound 2 (1-2) 2 (1-2) 31 1 (0-1) 0 (0-1) 0.091 32 33
N. Transvaginal ultrasound 1 (1-2) 2 (1-2) http://bmjopen.bmj.com/ 34 35 0 (0-0) 0 (0-0) 0.938 36 N. Uroflowmetry 1 (0-1) 0 (0-1) 37 38 0 (0-1) 0 (0-1) 0.933 39 40 N. Invasive urodynamic test 1 (1-1) 0 (0-1) 41
1 (1-1) on September 29, 2021 by guest. Protected copyright. 42 0 (0-0) 0.777 43 N. Cystoscopy with bladder byopsy 0 (0-1) 0 (0-0) 44 45 0 (0-0) 0 (0-0) 0.234 46 47 N. Cystoscopy 1 (1-3) 1 (0-1) 48 1 (0-1) 0 (0-1) 0.047 49 50 IQR = Interquartile range 51 52 * Wilcoxon-Mann-Withney test on the differences in the before 53 treatment/follow-up resource consumption between the two groups 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
Intravesical administration of combined hyaluronic acid (HA) and chondroitin sulphate (CS) for the treatment of female recurrent urinary tract infections: a European multicenter nested case-control study
For peer review only Journal: BMJ Open
Manuscript ID bmjopen-2015-009669.R1
Article Type: Research
Date Submitted by the Author: 20-Nov-2015
Complete List of Authors: Ciani, Oriana; University of Exeter , University of Exeter Medical School; CeRGAS Bocconi, Arendsen, Erik; Diaconessenhuis, Dept. of Urology Romancik, Martin; St. Cyril and Method University Hospital, Dept. of Urology Lunik, Richard; Fakultná nemocnica s poliklinikou, Dept. of Urology Costantini, Elisabetta; University of Perugia, Dept. of Surgical and Biomedical Science Di Biase, Manuel; University of Perugia, Dept. of Surgical and Biomedical Science Morgia, Giuseppe; University of Catania, Dept. of Urology
Fragala', Eugenia; University of Catania, Dept. of Urology http://bmjopen.bmj.com/ Tomaskin, Roman; Jessenius School of Medicine, Dept. of Urology Bernat, Marian; Fakultná nemocnica s poliklinikou, Dept. of Urology Guazzoni, Giorgio; Humanitas Clinical and Research Center, Dept. of Urology Tarricone, Rosanna; Bocconi University, Dept. of Policy Analysis and Public Management Lazzeri, Massimo; Humanitas Clinical and Research Center, Dept. of Urology on September 29, 2021 by guest. Protected copyright. Primary Subject Urology Heading:
Secondary Subject Heading: Infectious diseases, Health services research
Urinary tract infections < UROLOGY, Antimicrobial Resistance, Hyaluronic Keywords: acid, chondroitin sulphate
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 4 5 Intravesical administration of combined hyaluronic 6 7 8 acid (HA) and chondroitin sulphate (CS) for the 9 10 treatment of female recurrent urinary tract infections: 11 12 a European multicenter nested case-control study 13 14 15 For1 peer2 review 3 only 4 16 Oriana Ciani , Erik Arendsen , Martin Romancik , Richard Lunik , Elisabetta 17 Costantini5, Manuel Di Biase 5, Giuseppe Morgia6, Eugenia Fragalà6, 18 7 8 9 19 Tomaskin Roman , Marian Bernat , Giorgio Guazzoni , Rosanna 20 1,10 9 21 Tarricone , Massimo Lazzeri 22 23 1 24 Centre For Research on Health and Social Care Management, Università 25 26 Bocconi, via Rontgen, 1, 20136, Milan, Italy 27 2 Diaconessenhuis, Maatschap, Urologie, Leiden, The Netherlands 28 3 29 St. Cyril and Method University Hospital, Dept. of Urology, Bratislava, 30 31 Slovakia 32 4 Fakultná nemocnica s poliklinikou, Dept. of Urology, Prešov, Slovakia 33 5 http://bmjopen.bmj.com/ 34 University of Perugia, Dept. of Surgical and Biomedical Science, Urology 35 36 and Andrology Clinic, Perugia, Italy 37 6 University of Catania, Dept. of Urology, Catania, Italy 38 7 39 Jessenius School of Medicine, University Hospital, Dept. of Urology, Martin, 40 41 Slovakia 42 8 FNsP, Urologicka Klinika, Nové Zámky, Slovakia on September 29, 2021 by guest. Protected copyright. 43 9 44 Department of Urology, Humanitas Clinical and Research Center, 45 46 Humanitas University, Milan, Italy 47 10 Department of Policy Analysis and Public Management, Università Bocconi, 48 49 Via Roentgen, 1, 20136 Milano, Italy 50 51 Correspondence to: [email protected] 52 Trial registration: ClinicalTrials.gov NCT02016118 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 1 BMJ Open Page 2 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Word count: 3,552 (excluding references) 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 2 Page 3 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 4 Abstract 5 6 Objectives: To compare the clinical effectiveness of the intravesical 7 administration of combined hyaluronic acid and chondroitin sulphate (HA + 8 9 CS) vs. current standard management in adult women with recurrent urinary 10 11 tract infections (RUTIs). 12 Setting: A EU based, multicenter, retrospective nested case control study. 13 14 Participants: 276 adult women treated for RUTIs starting from 2009 to 2013. 15 For peer review only 16 Interventions: Patients treated with either intravesical administration of HA + 17 CS or standard of care (antimicrobial/immunoactive prophylaxis/ probiotics/ 18 19 cranberry). 20 21 Primary and secondary outcome measures: The primary outcome was 22 occurrence of bacteriologically confirmed recurrence within 12 months. 23 24 Secondary outcomes were time to recurrence, total number of recurrences, 25 26 health related quality of life and healthcare resource consumption. Crude and 27 adjusted results for unbalanced characteristics are presented. 28 29 Results: 181 patients treated with HA + CS and 95 patients treated with 30 31 standard of care from 7 centers were included. The crude and adjusted OR 32 (95% CI) for the primary endpoint were 0.77 (0.46 to 1.28) and 0.51 (0.27 to 33 http://bmjopen.bmj.com/ 34 0.96), respectively. However no evidence of improvement in terms of total 35 36 number or recurrences (incidence rate ratio (95%CI), 0.99 (0.69 to 1.43)) or 37 time to first recurrence was seen (hazard ratio (95%CI), 0.99 (0.61 to 1.61)). 38 39 The benefit of intravesical HA + CS therapy improves when the number of 40 41 instillations is ≥ 5. on September 29, 2021 by guest. Protected copyright. 42 Conclusions: Our results show that bladder instillations of combined HA + 43 44 CS reduces the risk of bacteriologically confirmed recurrences compared to 45 46 the current standard management of RUTIs. Total incidence rates and hazard 47 rates were instead non significantly different between the two groups after 48 49 adjusting for unbalanced factors. In contrast to what happens with antibiotic 50 51 prophylaxis, the effectiveness of the HA + CS reinstatement therapy improves 52 over time. 53 54 55 56 Trial registration: ClinicalTrials.gov NCT02016118 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 3 BMJ Open Page 4 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Article Summary 4 5 'Strengths and limitations of this study' 6 7 • This real world data shows bladder instillations of combined hyaluronic 8 acid and chondroitin sulphate reduce the risk of bacteriologically 9 10 confirmed urinary tract infections vs current standard management 11 12 • However if the recurrence occurs, there is no evidence of benefit in 13 terms of total number or time to first recurrence 14 15 • NumberFor of instillationspeer is anreview important marker ofonly success for this non 16 17 antimicrobial therapy 18 • Due to the retrospective observational design, these findings need 19 20 confirmation from prospective and preferably randomised studies 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 4 Page 5 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 4 Background 5 6 Urinary tract infection (UTI) is a major healthcare concern in women with an 7 annual incidence of 30 per 1000 (1). Nearly 33% of women will have had at 8 9 least one UTI episode, with characteristics of acute cystitis, requiring 10 11 antimicrobial therapy by the age of 24 years and as many as 60% of women 12 report having had a UTI in their lifetime (2, 3). 13 14 UTIs have a propensity to recur (4, 5); evidence shows that between 24% and 15 For peer review only 16 50% of initial episodes are followed by a second infection within six months 17 (6 9). A widely accepted definition of RUTIs is two or more UTI episodes over 18 19 six months, or three or more episodes over 12 months (10). On a population 20 21 scale, high incidence and prevalence of RUTIs results in considerable 22 healthcare costs, at the individual level the impact of this condition on health 23 24 related quality of life (HRQoL) is not negligible (11 13). 25 26 The pathogenesis of RUTI involves colonisation of the vagina with 27 uropathogenic bacteria and subsequent migration per urethra to the bladder. 28 29 About 68 77% of recurrences caused by Escherichia coli involve strains 30 31 genetically indistinguishable from those that caused previous infections (14). 32 The diagnosis is often made on clinical presentation with local genitourinary 33 http://bmjopen.bmj.com/ 34 symptoms of dysuria, frequency, and urgency or hesitancy appearing 35 36 suddenly (14). However, urine culture is useful in women presenting with 37 RUTI to confirm the diagnosis, direct antimicrobial therapy and exclude 38 39 infection from overactive bladder or interstitial cystitis (15). Evidence based 40 41 clinical practice guidelines recommend empiric initial therapy for acute on September 29, 2021 by guest. Protected copyright. 42 management or continuous antimicrobial therapy or self initiated therapy and 43 44 prophylaxis, either antimicrobial or non antimicrobial based (16, 17). 45 46 The choice of specific strategy for care depends on the number of 47 recurrences experienced per year, the patient’s preferences and careful 48 49 review of modifiable risk factors (14, 18). As the second most common reason 50 51 for prescribing antibiotics (following otitis media), there is currently increasing 52 concern about empiric use of these agents due to increased antimicrobial 53 54 resistance. Antibiotic use selects for resistant pathogens: a major risk factor 55 56 for an antibiotic resistant UTI is prior antibiotic use (5). In an international 57 survey investigating the prevalence and susceptibility of pathogens causing 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 5 BMJ Open Page 6 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 cystitis, 10.3% of Escherichia coli isolates were resistant to at least three 4 5 different classes of antimicrobial agents, including ampicillin (48.3%), 6 trimethoprim/sulfamethoxazole (29.4%) and nalidixic acid (18.6%) (19). 7 8 Nonantimicrobial prevention strategies have become popular in the age of 9 10 increasing antimicrobial use and resistance. However, no probiotic agent has 11 been approved for therapeutic use and the potential benefit of cranberry in 12 13 terms of product type (solid vs liquid), dosing, and optimal patient population 14 15 remainsFor to be elucidated peer (14, 18). review A new therapy based only on the reinstatement 16 of the glycosaminoglycan (GAG) bladder epithelium has recently been 17 18 proposed for the treatment of RUTIs (20). This GAG layer consists of non 19 20 sulphated, e.g. hyaluronic acid (HA), and sulphated, e.g. heparan sulphate 21 and heparin, chondroitin sulphate (CS), dermatan sulphate and keratan 22 23 sulphate, GAGs. Limited evidence has shown the preventive activity of 24 25 intravesical GAG substation therapy (with HA alone or with HA + CS) on 26 recurrence of infections in patients with recurrent bacterial cystitis (21). 27 28 However, large scale studies are needed to underline the benefit of this 29 30 therapy (22). 31 Therefore, we decided to perform a European retrospective multicenter study 32 33 to compare the clinical effectiveness of the intravesical administration of http://bmjopen.bmj.com/ 34 35 combined HA + CS (Ialuril®, IBSA Institut Biochimique SA) vs. current 36 standard management of RUTIs in adult women. 37 38 39 40 41 Methods 42 on September 29, 2021 by guest. Protected copyright. 43 Study design 44 45 This was a EU based, multicenter, retrospective nested case control 46 comparison of individual patient data collected from electronic medical 47 48 records and/or administrative databases available at the participating 49 50 institutions. Centers using the intravesical administration of combined HA + 51 CS, in the countries where Ialuril® was already registered and on the market 52 53 (Ialuril® received a CE mark for this indication in 2009), were identified and 54 55 invited to take part in the study. 56 57 58 Study population 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 6 Page 7 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 All patients treated with either HA + CS or standard of care at the participating 4 5 centers, high volume organizations with specific expertise in the treatment of 6 UTIs, starting from 2009 were included if they were women, aged 18 75 7 8 years, diagnosed with RUTIs, defined as at least three episodes of 9 10 uncomplicated UTIs accompanied by clinical symptoms and documented by 11 urine culture with the isolation of >103 CFU/ml of an identified pathogen in the 12 13 last 12 months. Uncomplicated UTI is defined as an infection in a person with 14 15 normal urinaryFor tract peer and function (17).review Women with complicated only UTIs (i.e. 16 individuals with functional or structural abnormalities of the genitourinary tract) 17 18 were excluded. Within Europe, patients at first diagnosis of RUTIs are offered 19 20 an approach based on behavioral changes, antimicrobial prophylaxis or 21 aspecific non antimicrobial prevention. However, several women refuse to 22 23 take antimicrobials over an extended period of time, hence intravesical 24 25 administration of HA + CS is intended for women refractory or not satisfied 26 with first line management of RUTIs. Based on a previous cohort study (11), 27 28 we estimated that 208 patients were needed to observe a 50% difference in 29 30 the proportions of patients recurring between the two groups within 12 months 31 with 90% power and alpha level of 0.05. 32 33 http://bmjopen.bmj.com/ 34 35 Groups and interventions 36 Patients treated with intravesical administration of combined HA 1.6% and CS 37 38 2.0%. The recommended scheme is one instillation per week for the first 39 40 month, followed by one instillation every two weeks for the second month and 41
one instillation per month afterward until stable remission of the symptoms, on September 29, 2021 by guest. Protected copyright. 42 43 however different patterns are seen in clinical practice. These patients were 44 45 compared to patients treated with antimicrobial prophylaxis (continuous or 46 postcoital), or immunoactive prophylaxis or prophylaxis with probiotics or 47 48 prophylaxis with cranberry, or combination of these (17), as recommended by 49 50 the European Association of Urology. 51 52 53 Study outcomes 54 55 The primary outcome for this study was the occurrence of objective urinary 56 tract infection recurrence, defined as the occurrence of at least one 57 58 bacteriologically confirmed urinary tract infection within 12 months after 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 7 BMJ Open Page 8 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 treatment initiation for RUTIs. According to current clinical guidelines, in non 4 5 pregnant women, urine culture is recommended in symptomatic patients only. 6 Information about clinically confirmed recurrences were also sought although 7 8 they are not reported in this manuscript as they are assumed to be less 9 10 objective than the bacteriologically confirmed ones. Patients who developed a 11 UTI whilst on the HA + CS instillation protocol were treated according to 12 13 clinical guidelines with antibiotics but could continue the instillations afterward. 14 15 After theFor first bacteriologically peer confirmed review recurrence, notonly only the time to first 16 recurrence was recorded, but also the number of additional UTIs. The 17 18 secondary outcome measures were the time to recurrence (defined as the 19 20 time from the start of the treatment until the occurrence of the first objective 21 recurrence); the total number of recurrences; HRQoL as assessed through 22 23 the SF 36 (23) or EQ 5D (24) questionnaires. Dutch (25), Italian (26) and UK 24 25 (27) tariffs were used to estimate utility values from the EQ 5D questionnaires 26 in the Netherlands, Italy and Slovakia, respectively. Information about 27 28 healthcare resource consumption were also collected. A cost analysis was 29 30 planned and will be the subject of a future publication. 31 32 33 Data collection http://bmjopen.bmj.com/ 34 35 General patient demographic characteristics, diagnosis and treatment 36 information were collected based on a predefined form designed on the input 37 38 obtained from collaborating centers during a workshop held in July 2013. An 39 40 intuitive electronic system was implemented (Advice Pharma ltd) to record 41
and store data on a secure remote server provider. on September 29, 2021 by guest. Protected copyright. 42 43 44 45 Statistical analyses 46 Continuous baseline characteristics are presented as median and interquartile 47 48 range (IQR) or mean and standard deviation (SD), as appropriate. For 49 50 proportions, absolute and relative frequencies are reported. Wilcoxon–Mann– 51 Whitney test or t test were used for continuous and ordinal variables baseline 52 53 differences, whereas the Chi Square test was used for proportions. In our 54 55 primary analyses, we applied univariate and multivariate logistic, poisson and 56 Cox model for objective recurrence, number of recurrences and time to 57 58 recurrence outcomes, respectively. Results were presented as crude and 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 8 Page 9 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 adjusted odds ratio (OR), incidence rate ratio (IRR) and hazard ratio (HR) 4 5 respectively, with their 95% confidence intervals (CI). Adjusting variables were 6 age, body mass index (BMI), employment and menopause status, postcoital 7 8 infections, dyspareunia, Female sexual function index (FSFI) and severity of 9 10 RUTI. A prespecified sensitivity analysis was conducted to investigate the 11 impact of adherence to HA + CS treatment on clinical outcomes considering 12 13 patients who had ≥ 5 instillations. Pairwise deletion was used to deal with 14 15 missing Fordata. All significance peer tests review were two tailed at theonly 0.05 significance 16 level. All the analyses were conducted using Stata SE StataCorp LP 11. 17 18 19 20 Ethical approval and funding 21 The study protocol was reviewed and approved at the coordinating center by 22 23 an Independent Ethics Committee at the Department of Urology, University of 24 25 Perugia. The study was funded by the Study Group for Urogenital Diseases in 26 Italy, an association on practioners and clinical urologists. The study is 27 28 registered at clinicaltrials.gov (NCT02016118). Extra data is available by 29 30 emailing the corresponding author. 31 32 33 http://bmjopen.bmj.com/ 34 Results 35 36 Overall, 276 patients treated for RUTIs at seven European centers from 37 January 2009 up to December 2013 were included in the analyses. Of these, 38 39 181 women were treated with HA + CS intravesical administration and 95 40 41 women received standard management of RUTIs. The numerical imbalance on September 29, 2021 by guest. Protected copyright. 42 was probably due to the participating organizations being tertiary referral 43 44 centers for patients who are not satisfied with standard management of 45 46 RUTIs. A flow diagram reporting number of patients at each stage of the study 47 is shown in Figure 1. Baseline socio demographic and clinical characteristics 48 49 of patients are reported in Table 1. Given the non experimental nature of the 50 51 study, the distribution of several characteristics was not homogeneous 52 between the two groups, in particular women treated with HA + CS were 53 54 older, with a higher BMI and probability of dyspareunia. 55 56 57 Primary analyses 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 9 BMJ Open Page 10 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 In the HA + CS group, 55.7% of patients showed bacteriologically confirmed 4 5 recurrences, whereas 62.1% had such recurrence in the standard of care 6 group (p = 0.313). However, the adjusted OR (95% CI) for developing a 7 8 bacteriologically confirmed recurrence within 12 months was 0.51 (0.27 to 9 10 0.96), meaning that, other characteristics being equal, there is a 49% reduced 11 risk of developing a recurrence in patients treated with HA + CS compared to 12 13 standard care (Table 2). When the number of recurrences is considered, in 14 15 the HA +For CS group peerthere were 121 review bacteriologically confirmedonly recurrences in 16 61.5 person year whereas in the standard treatment group there were 59 17 18 bacteriologically confirmed recurrences in 51.1 person year (p = 0.001). 19 20 However we observed an adjusted IRR (95% CI) of 0.99 (0.69 to 1.43), 21 showing non significant differences in the incidence rates between the group 22 23 treated with HA + CS and the control. Similar results were obtained from the 24 25 univariate and multivariate Cox regression models used to estimate the HR 26 (95% CI) for the time to first bacteriologically confirmed recurrence, with an 27 28 unadjusted estimate of 0.99 (0.61 to 1.61) (Table 2). Although during the 12 29 30 months follow up the median time to first recurrence was higher in the 31 standard care group (169.5 days (IQR, 72.5 to 341.5) vs 320 days (IQR, 179 32 33 to 365); p value < 0.001), we observed the distribution of recurrences at http://bmjopen.bmj.com/ 34 35 separate follow up times, and noted that the incident proportion of patients 36 who developed the recurrence vs those still at risk was lower in the HA + CS 37 38 group in the latest part of follow up, after 8 months (Table 3). All patients were 39 40 alive at 12 months follow up. There were 14 all cause hospitalizations in the 41
HA + CS and 1 in the control group. on September 29, 2021 by guest. Protected copyright. 42 43 44 45 Health related quality of life and resources consumption 46 In a subset of patients, a measure of the HRQoL as measured through the 47 48 SF 36 or EQ 5D 3L questionnaires was available at baseline and after 12 49 50 months follow up. There was no evidence of better improvement in HRQoL in 51 the HA + CS group compared to control with SF 36 results, whereas when 52 53 EQ 5D data were considered the HA + CS group seemed to have received a 54 55 higher benefit in terms of HRQoL than the control (Supplementary table 1). 56 There is a general reduction in physical units of health resources (i.e. medical 57 58 visits, laboratory and imaging tests) consumed by the two groups before the 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 10 Page 11 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 treatment and during the follow up (Supplementary table 2) without significant 4 5 differences between the two groups. 6 7 8 Sensitivity analyses 9 10 We repeated all primary analyses and considered different exposure intensity 11 (i.e. number of intravesical administrations received) in the HA + CS group. All 12 13 findings consistently show additional benefit gained to patients when the 14 15 number Forof instillations peer increases, revealingreview the importance only of adherence to 16 this medical device therapy for the treatment of RUTIs (Table 4). As a post 17 18 hoc subgroup analysis, we repeated primary analyses in non sexually active 19 20 patients only and obtained similar patterns of results as in the whole sample, 21 although with loss of statistical significance. 22 23 24 25 26 Discussion 27 In this European multicenter retrospective observational study we compared 28 29 bacteriologically confirmed recurrence rates at 12 months follow up after the 30 31 initiation of intravesical administration of HA + CS vs standard of care for the 32 treatment of RUTIs. After adjusting for unbalanced confounding factors 33 http://bmjopen.bmj.com/ 34 between the two groups, we observed that the HA + CS patients had a 49% 35 36 reduction (OR 0.51, 95% CI: 0.27 to 0.96) in the risk of a bacteriologically 37 confirmed recurrence, whereas there was no statistical evidence for a 38 39 difference in the incidence rates and hazard rates of such recurrences 40 41 between the two groups. on September 29, 2021 by guest. Protected copyright. 42 Four clinical studies (28 31) have been performed to investigate the efficacy 43 44 and tolerability of intravesically administered GAG for RUTI prophylaxis, all 45 46 showing that HA alone or HA + CS instillations reduce the number of UTI per 47 patient per year at no increased risk of severe adverse events and prolong the 48 49 time interval between RUTI episodes, with a high rate of patients being free of 50 51 recurrence at the end of the study period. In particular, two RCTs studies 52 compared HA + CS administration to either placebo (29) or long term 53 54 antibiotic prophylaxis using sulfamethoxazole 200 mg and trimethoprim 40 mg 55 56 (30). Damiano and colleagues report a decrease in UTI rate per patient of 57 77% (95% CI, 72.3 to 80.8) in the experimental vs placebo group, whereas De 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 11 BMJ Open Page 12 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Vita et al. report the mean ± SD number of recurrent cystitis per patient per 4 5 year as 1 ± 1.2 vs 2.3 ± 1.4 in HA + CS and antibiotic treated patients, 6 respectively. Despite the prospective and randomized design, these trials 7 8 were limited by the small sample size (i.e. they included 57 and 28 patients, 9 10 respectively) and the single center setting that considerably reduces the 11 generalizability of the findings. Whilst waiting for definitive RCT evidence 12 13 clarifying the comparative effectiveness profile of this therapy in support of its 14 15 adoption,For our observational peer study reviewdesign provides useful only information around 16 its effectiveness in real world practice. 17 18 In this respect, our study involved seven centers across three European 19 20 countries and 276 patients, thus providing important additional evidence with 21 respect of current treatment options for RUTIs. Furthermore, the non 22 23 experimental observational design allows for a closer representation of the 24 25 routine clinical practice of the use of the HA + CS reinstatement therapy as 26 compared to standard of care in place at high volume university hospitals, that 27 28 is on purpose defined as very broad given the variety of recommended 29 30 strategies (17) and general scarce adherence to clinical guidelines (32). 31 On the other hand, the retrospective design limited the availability of data to 32 33 that previously collected at the centers. Contacting patients ex post to gather http://bmjopen.bmj.com/ 34 35 additional data was not applicable (e.g. as in the case of HRQoL assessment) 36 or not helpful given the potential significant recall bias introduced by delayed 37 38 reporting. The issue of missing data was dealt with by assuming they were 39 40 missing at random (i.e. given the observed data, data are missing 41
independently of unobserved data, that is missing data does not depend on on September 29, 2021 by guest. Protected copyright. 42 43 the level of their outcome) and applying pairwise deletion. In this regard, we 44 45 performed two additional analyses, first by restricting the primary analyses to 46 all complete cases (i.e. no missing in both outcomes or adjusting factors). 47 48 That provided similar results to those presented here (data not shown). 49 50 Second, for all outcomes and adjusting variables we tested through Fisher’s 51 exact test whether proportions of missing was different between HA + CS and 52 53 Standard of Care groups. No significant difference was observed with the 54 55 exception of the resource consumption where the number of missing was 56 higher in the HA + CS group. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 12 Page 13 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Data on uropathogens and antimicrobial resistance within the groups was 4 5 unfortunately not available from this database, although we know most 6 commonly prescribed antibiotics were Ciprofloxacin (13.2% of all 7 8 prescriptions), Cefuroxime (6.9%), Fosfomycin (6.9%), Nitrofurantoin (6.4%) 9 10 and E. Coli bacterial extract (OM 89, 4.8%). 11 Health related quality of life assessment in routine practice is still uncommon 12 13 as indicated by the significant proportion of missing information (up to 73% in 14 15 the controlFor group) onpeer this outcome. review However our findings only are in line with 16 previous reports showing that the GAG replacement treatment in women with 17 18 RUTIs had a positive impact on patients’ quality of life, reducing the 19 20 symptoms and improving the maximum cystometric capacity (29, 30). 21 Recurrent UTIs in women are a common condition, associated with significant 22 23 morbidity and burden for the whole society. In a study of 684 women aged 18 24 25 to 70 years with UTI, participants reported an average of 3.83 symptom days, 26 2.89 restricted activity days, and 3.13 days during which they were unwell 27 28 (33). In another study patients reported 1.2 days in which they were not able 29 30 to attend classes or work, and 0.4 days in bed (34). New effective prevention 31 strategies are needed, in particular non antimicrobial approaches would be 32 33 desirable for several reasons. First of all, a prolonged antibiotic use as a http://bmjopen.bmj.com/ 34 35 prophylactic approach to RUTI increases the risk of side effects, including 36 vaginal and oral candidiasis, and gastrointestinal symptoms (10). This in turn, 37 38 lowers patients’ compliance and therefore the effectiveness of the treatment 39 40 (35). However, the most worrying effect of the antibiotic use (and misuse) is 41
the exacerbation of the antimicrobial resistance (AMR) (36). Recently, a UK on September 29, 2021 by guest. Protected copyright. 42 43 commissioned report on health and macroeconomic consequences of AMR 44 45 estimated 10 million extra deaths a year and costs up to €90tn for the global 46 economy by 2050 if this problem is not properly tackled (37). Although this 47 48 first report might not be as scientifically rigorous or informed by evidence as 49 50 possible (38), it brought renewed interest in the worldwide AMR crisis. The 51 war against the spread of drug resistant microbes is attracting considerable 52 53 attention as well as investment from all major governments and research 54 55 organizations in EU and beyond (39,40). The way forward outlined by all of 56 these major research initiatives includes establishing appropriate funding and 57 58 rewards to subsidize access to and development of new antibiotic agents, 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 13 BMJ Open Page 14 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 preservation of existing drugs antimicrobial activity through prescription 4 5 tailored to diagnosis, prioritization and controlled access, and identification of 6 novel approaches and therapies for microbial diseases. 7 8 The case of GAG reinstatement therapy is a good example of an innovative 9 10 approach to prevent and manage bacterial urinary infections, a medical 11 device intervention as opposed to a drug. In contrast to antibiotic therapy, 12 13 which aims at eradicating pathogens, treatment with HA + CS targets 14 15 bacterialFor adherence peer to the bladder review mucosa by physically only recovering a 16 damaged GAG layer that facilitates bacterial adherence and, therefore, 17 18 RUTIs. The different mechanism of action could explain the apparent 19 20 reduction in the incidence of UTIs in the group treated with HA + CS 21 instillations compared with standard care when considering later time intervals 22 23 (Table 3). Whilst antibiotics are immediately effective, although subject and 24 25 conducive to resistance, GAG layer administration is progressively restoring 26 the epithelium that will protect women from future uropathogen infections. On 27 28 the other hand, catheterization induced UTIs might represent an unintended 29 30 consequence of this procedure. Previous reports (29, 30) have highlighted 31 good tolerability and safety of the intervention that must be performed in 32 33 under sterile condition by nurses trained in the procedure. As regards the http://bmjopen.bmj.com/ 34 35 economic profile of the two alternative approaches, it has been reported that 36 the cost of the HA + CS could be even five times higher than the cost for a 6 37 38 month antibiotic prophylaxis. However, this consideration corresponds to a 39 40 very restrictive, if not naïve, cost analysis 41
as it is well known that all direct healthcare costs and consequences, on September 29, 2021 by guest. Protected copyright. 42 43 including those for the wider society, as would be containment of drug 44 45 resistance spreading, should be taken into account when assessing the cost 46 effectiveness profiles of health technologies. Future methodologically sound 47 48 economic evaluation studies are recommended to compare the societal or 49 50 payer value of the two treatment strategies. 51 52 53 54 Conclusions 55 56 In order to treat and prevent RUTI, there is a need of effective and safe 57 alternative strategies to antimicrobial therapy. Our study showed that in real 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 14 Page 15 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 world setting bladder instillation of combined HA + CS reduces the risk of 4 5 bacteriologically confirmed recurrences compared to the current standard 6 management in this study population. Total incidence rates and hazard rates 7 8 were instead non significantly different between the two groups. The number 9 10 of HA + CS instillations seems to be an important marker of success for the 11 intravesical administration therapy. Furthermore, in contrast to what happens 12 13 with antibiotic prophylaxis, due to side effects and development of resistance, 14 15 the effectivenessFor ofpeer GAG reinstatement review therapy improves only over time, with an 16 even better expected comparative effectiveness profile in the long run. 17 18 Although firm conclusions are difficult due to the retrospective observational 19 20 design, these findings highlight the relevance of additional prospective and 21 randomized studies in this area and the promising role of the HA + CS 22 23 reinstatement therapy for prevention and treatment of RUTI in an era of 24 25 worryingly increased antimicrobial resistance. 26 27 28 Contributorship: OC, RT, ML designed the study. EA, MR, RL, EC, MDB, 29 30 GM, EF, TR, MB, GG contributed to the data collection. OC analyzed the data 31 and drafted the manuscript. All authors commented and approved the final 32 33 version of the manuscript. http://bmjopen.bmj.com/ 34 35 Competing interest: No competing interest**. 36 Funding: This study was funded by an unrestricted grant from the TETI 37 38 Association Study group for urogenital diseases (www.tetiassociation.com). 39 40 This is a practitioners and clinical urologists association in Italy. Members of 41
the association were involved in the data collection and revised the on September 29, 2021 by guest. Protected copyright. 42 43 manuscript. 44 45 Data sharing: The dataset is available by emailing the corresponding author 46 Oriana Ciani ([email protected]) 47 48 . 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 15 BMJ Open Page 16 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 4 References 5 6 7 1. Laupland KB, Ross T, Pitout JD, Church DL, Gregson DB. Community 8 onset urinary tract infections: a population based assessment. Infection. 9 10 2007;35(3):150 3. 11 2. Foxman B. Epidemiology of urinary tract infections: incidence, 12 morbidity, and economic costs. The American journal of medicine. 2002;113 13 Suppl 1A:5s 13s. 14 3. Foxman B, Barlow R, D'Arcy H, Gillespie B, Sobel JD. Urinary tract 15 infection:For self reported peer incidence andreview associated costs. only Annals of 16 epidemiology. 2000;10(8):509 15. 17 4. Hooton TM, Scholes D, Hughes JP, Winter C, Roberts PL, Stapleton 18 AE, et al. A prospective study of risk factors for symptomatic urinary tract 19 20 infection in young women. The New England journal of medicine. 21 1996;335(7):468 74. 22 5. Foxman B. The epidemiology of urinary tract infection. Nature reviews 23 Urology. 2010;7(12):653 60. 24 6. Sen A. Recurrent cystitis in non pregnant women. BMJ clinical 25 evidence. 2008;2008. 26 7. Ikaheimo R, Siitonen A, Heiskanen T, Karkkainen U, Kuosmanen P, 27 Lipponen P, et al. Recurrence of urinary tract infection in a primary care 28 setting: analysis of a 1 year follow up of 179 women. Clinical infectious 29 30 diseases : an official publication of the Infectious Diseases Society of 31 America. 1996;22(1):91 9. 32 8. Foxman B. Recurring urinary tract infection: incidence and risk factors. 33 American journal of public health. 1990;80(3):331 3. http://bmjopen.bmj.com/ 34 9. Foxman B, Gillespie B, Koopman J, Zhang L, Palin K, Tallman P, et al. 35 Risk factors for second urinary tract infection among college women. 36 American journal of epidemiology. 2000;151(12):1194 205. 37 10. Albert X, Huertas I, Pereiro, II, Sanfelix J, Gosalbes V, Perrota C. 38 Antibiotics for preventing recurrent urinary tract infection in non pregnant 39 40 women. The Cochrane database of systematic reviews. 2004(3):Cd001209. 41 11. Ciani O, Grassi D, Tarricone R. An economic perspective on urinary 42 tract infection: the "costs of resignation". Clinical drug investigation. on September 29, 2021 by guest. Protected copyright. 43 2013;33(4):255 61. 44 12. Renard J, Ballarini S, Mascarenhas T, Zahran M, Quimper E, Choucair 45 J, et al. Recurrent Lower Urinary Tract Infections Have a Detrimental Effect on 46 Patient Quality of Life: a Prospective, Observational Study. Infectious 47 diseases and therapy. 2014. 48 13. Bermingham SL, Ashe JF. Systematic review of the impact of urinary 49 50 tract infections on health related quality of life. BJU international. 2012;110(11 51 Pt C):E830 6. 52 14. Gupta K, Trautner BW. Diagnosis and management of recurrent 53 urinary tract infections in non pregnant women. BMJ. 2013;346:f3140. 54 15. Arya LA, Northington GM, Asfaw T, Harvie H, Malykhina A. Evidence of 55 bladder oversensitivity in the absence of an infection in premenopausal 56 women with a history of recurrent urinary tract infections. BJU international. 57 2012;110(2):247 51. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 16 Page 17 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 16. Gupta K, Hooton TM, Naber KG, Wullt B, Colgan R, Miller LG, et al. 4 International clinical practice guidelines for the treatment of acute 5 uncomplicated cystitis and pyelonephritis in women: A 2010 update by the 6 Infectious Diseases Society of America and the European Society for 7 Microbiology and Infectious Diseases. Clinical infectious diseases : an official 8 9 publication of the Infectious Diseases Society of America. 2011;52(5):e103 10 20. 11 17. Grabe M, Bartoletti R, Bjerklund Johansen T, Cai T, Çek M, Köves B, 12 et al. Guidelines on Urological Infections. European Association of Urology, 13 2015. 14 18. Shepherd AK, Pottinger PS. Management of urinary tract infections in 15 the era ofFor increasing peer antimicrobial review resistance. The Medical only clinics of North 16 America. 2013;97(4):737 57, xii. 17 19. Schito GC, Naber KG, Botto H, Palou J, Mazzei T, Gualco L, et al. The 18 19 ARESC study: an international survey on the antimicrobial resistance of 20 pathogens involved in uncomplicated urinary tract infections. International 21 journal of antimicrobial agents. 2009;34(5):407 13. 22 20. Lazzeri M, Montorsi F. The therapeutic challenge of "chronic cystitis": 23 search well, work together, and gain results. European urology. 24 2011;60(1):78 80. 25 21. De Vita D, Antell H, Giordano S. Effectiveness of intravesical 26 hyaluronic acid with or without chondroitin sulfate for recurrent bacterial 27 cystitis in adult women: a meta analysis. International urogynecology journal. 28 29 2013;24(4):545 52. 30 22. Madersbacher H, van Ophoven A, van Kerrebroeck PE. GAG layer 31 replenishment therapy for chronic forms of cystitis with intravesical 32 glycosaminoglycans a review. Neurourology and urodynamics. 2013;32(1):9 33 18. http://bmjopen.bmj.com/ 34 23. Brazier JE, Harper R, Jones NM, O'Cathain A, Thomas KJ, Usherwood 35 T, et al. Validating the SF 36 health survey questionnaire: new outcome 36 measure for primary care. Bmj. 1992;305(6846):160 4. 37 24. EuroQol a new facility for the measurement of health related quality of 38 39 life. Health policy (Amsterdam, Netherlands). 1990;16(3):199 208. 40 25. Lamers LM, McDonnell J, Stalmeier PF, Krabbe PF, Busschbach JJ. 41 The Dutch tariff: results and arguments for an effective design for national 42 EQ 5D valuation studies. Health Econ. 2006;15(10):1121 32. on September 29, 2021 by guest. Protected copyright. 43 26. Scalone L, Cortesi PA, Ciampichini R, Belisari A, D'Angiolella LS, 44 Cesana G, et al. Italian population based values of EQ 5D health states. 45 Value in health : the journal of the International Society for 46 Pharmacoeconomics and Outcomes Research. 2013;16(5):814 22. 47 27. Dolan P. Modeling valuations for EuroQol health states. Medical care. 48 49 1997;35(11):1095 108. 50 28. Constantinides C, Manousakas T, Nikolopoulos P, Stanitsas A, 51 Haritopoulos K, Giannopoulos A. Prevention of recurrent bacterial cystitis by 52 intravesical administration of hyaluronic acid: a pilot study. BJU international. 53 2004;93(9):1262 6. 54 29. Damiano R, Quarto G, Bava I, Ucciero G, De Domenico R, Palumbo 55 MI, et al. Prevention of recurrent urinary tract infections by intravesical 56 administration of hyaluronic acid and chondroitin sulphate: a placebo 57 controlled randomised trial. European urology. 2011;59(4):645 51. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 17 BMJ Open Page 18 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 30. De Vita D, Giordano S. Effectiveness of intravesical hyaluronic 4 acid/chondroitin sulfate in recurrent bacterial cystitis: a randomized study. 5 International urogynecology journal. 2012;23(12):1707 13. 6 31. Lipovac M, Kurz C, Reithmayr F, Verhoeven HC, Huber JC, Imhof M. 7 Prevention of recurrent bacterial urinary tract infections by intravesical 8 9 instillation of hyaluronic acid. International journal of gynaecology and 10 obstetrics: the official organ of the International Federation of Gynaecology 11 and Obstetrics. 2007;96(3):192 5. 12 32. Taur Y, Smith MA. Adherence to the Infectious Diseases Society of 13 America guidelines in the treatment of uncomplicated urinary tract infection. 14 Clinical infectious diseases : an official publication of the Infectious Diseases 15 Society Forof America. peer 2007;44(6):769 74. review only 16 33. Little P, Merriman R, Turner S, Rumsby K, Warner G, Lowes JA, et al. 17 Presentation, pattern, and natural course of severe symptoms, and role of 18 19 antibiotics and antibiotic resistance among patients presenting with suspected 20 uncomplicated urinary tract infection in primary care: observational study. 21 Bmj. 2010;340:b5633. 22 34. Ronald A. The etiology of urinary tract infection: traditional and 23 emerging pathogens. The American journal of medicine. 2002;113 Suppl 24 1A:14S 9S. 25 35. Hooton TM, Levy SB. Antimicrobial resistance: a plan of action for 26 community practice. American family physician. 2001;63(6):1087 98. 27 36. Fauci AS, Marston l D. The perpetual challenge of antimicrobial 28 29 resistance. Jama. 2014;311(18):1853 4. 30 37. Review on Antimicrobial Resistance. Antimicrobial Resistance: 31 Tackling a crisis for the health and wealth of nations. 2014. 32 38. Antimicrobial resistance: in terms politicians understand. Lancet. 33 2014;384(9961):2173. http://bmjopen.bmj.com/ 34 39. Woolhouse M, Farrar J. Policy: An intergovernmental panel on 35 antimicrobial resistance. Nature. 2014;509(7502):555 7. 36 40. Geoghegan Quinn M. Funding for antimicrobial resistance research in 37 Europe. Lancet. 2014;384(9949):1186. 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 18 Page 19 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Table 1 Baseline characteristics 4 5 6 Characteristics HA + CS Standard care P - value 7 (N = 181) (N = 95) 8 9 Age – mean (SD) 55.24 (17.33) 48.84 (21.16) 0.017 10 11 Employed – n (%) 83 (46) 62 (65) 0.003 12 Partner – n (%) 148 (82) 72 (76) 0.172 13 14 Sexually active – n (%) 114 (63) 60 (63) 0.931 15 For peer review only 16 Menopause – n (%) 69 (28) 49 (51) 0.038 17 BMI – mean (SD) 26.20 (6.37) 24.56 (5.03) 0.019 18 19 Postcoital infections – n (%) 31 (17) 29 (31) 0.012 20 21 Dyspareunia – n (%) 40 (22) 5 (5) < 0.001 22 23 Severity RUTI – median (IQR)* 4 (2 – 5) 4 (3 – 4) 0.316 24 Prophylaxis n (%) 25 26 Antimicrobial prophylaxis 27 na 42 (44.21) 28 continuous 29 Antimicrobial prophylaxis 30 na 19 (20) 31 postcoital 32 33 Intermittent immunoactive na 28 (29.47) http://bmjopen.bmj.com/ 34 prophylaxis 35 36 On demand immunoactive 37 na 1 (1.05) 38 therapy 39 Others na 5 (5.26) 40 41 FSFI – mean (SD) 5.35 (8.71) 3.13 (5.43) 0.018 42 on September 29, 2021 by guest. Protected copyright. a b 43 EQ 5D– median (IQR) 0.69 (0.24 0.73) 0.69 (0.28 0.81) 0.696 44 c d 45 SF 36 PCS – median (IQR) 60 (54.5 70) 60 (53 – 67.5) 0.500 46 SF 36 MCS – median (IQR) 60 (54 70)e 60 (53 – 67.5)d 0.551 47 48 BMI = body mass index; EQ 5D = Euro QoL 5D 3 level; FSFI = female 49 50 sexual function index; IQR = interquartile range; na = not applicable; SD = 51 standard deviation; RUTI = recurrent urinary tract infection; SF 36 PCS = 52 53 Short From 36 physical component score; SF 36 MCS = Short From 36 54 55 mental component score 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml19 BMJ Open Page 20 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 *According to the European Association of Urology Guidelines on 4 5 Urological Infections where 1 is low severity cystitis and 6 is extreme 6 severity including organ failure (21) 7 8 aN = 90 patients; bN = 29 patients; cN = 72 patients; dN = 60 patients; eN = 9 10 73 patients 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml20 Page 21 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Table 2 Bacteriologically confirmed recurrence, total number of 4 5 recurrences and time to first recurrence between HA + CS vs standard 6 of care treated patients 7 8 9 Outcome OR (95% CI) Adjusted* OR (95% CI) 10 Bacteriologically 11 0.77 (0.46 to 1.28) 0.51 (0.27 to 0.96) 12 confirmed recurrence 13 14 IRR (95% CI) Adjusted* IRR (95% CI) 15 For peer review only 16 Total number of 17 bacteriologically 1.73 (1.27 to 2.37) 0.99 (0.69 to 1.43) 18 19 confirmed recurrence 20 21 HR (95% CI) Adjusted* HR (95% CI) 22 Time to first 23 24 bacteriologically 1.66 (1.09 to 2.54)§ 0.99 (0.61 to 1.61) 25 26 confirmed recurrence 27 OR = odds ratio, IRR = incidence rate ratio, HR = hazard ratio 28 29 *Adjusted for age, BMI, employment and menopause status, postcoital 30 31 infections, dyspareunia, FSFI and severity of RUTI 32 §Log rank test p value 0.018 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 21 BMJ Open Page 22 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Table 3 Incidence of bacteriologically confirmed recurrences during 12 4 5 months follow-up 6 7 Incidence of HA + CS Standard care P value 8 9 bacteriologically 10 11 confirmed 12 recurrences 13 14 0 – 90 days 15.3% 12.7% 0.610 15 For peer review only 16 91 – 180 days 15.6% 12.9% 0.628 17 181 – 240 days 10.7% 3.7% 0.132 18 19 241 – 365 days 16.3% 30.8% 0.043 20 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 22 Page 23 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Table 4 Sensitivity analysis – impact of number of intravesical 4 5 administration of HA + CS on clinical outcomes 6 7 Bacteriologically confirmed recurrence OR (95% CI) Adjusted* OR (95% CI) 8 9 ≥ 5 instillations (n = 156) vs standard of care 0.81 (0.48 to 1.36) 0.52 (0.27 to 0.99) 10 11 ≥ 6 instillations (n = 134) vs standard of care 0.69 (0.40 to 1.18) 0.47 (0.25 to 0.91) 12 ≥ 7 instillations (n = 82) vs standard of care 0.63 (0.34 to 1.14) 0.43 (0.21 to 0.88) 13 14 Total number of bacteriologically IRR (95% CI) Adjusted* IRR (95% CI) 15 For peer review only 16 confirmed recurrence 17 ≥ 5 instillations vs standard of care 1.82 (1.33 to 2.49) 1.05 (0.73 to 1.52) 18 19 ≥ 6 instillations vs standard of care 1.64 (1.18 to 2.28) 0.97 (0.66 to 1.40) 20 21 ≥ 7 instillations vs standard of care 1.46 (1 to 2.13) 0.90 (0.60 to 1.36) 22 23 Time to first bacteriologically confirmed HR (95% CI) Adjusted* HR (95% CI) 24 recurrence 25 26 ≥ 5 instillations vs standard of care 1.72 (1.12 to 2.65) 1.04 (0.64 to 1.71) 27 28 ≥ 6 instillations vs standard of care 1.55 (0.99 to 2.41) 0.96 (0.58 to 1.57) 29 ≥ 7 instillations vs standard of care 1.33 (0.79 to 2.22) 0.85 (0.49 to 1.47) 30 31 OR = odds ratio, IRR = incidence rate ratio, HR = hazard ratio 32 33 *Adjusted for age, BMI, employment and menopause status, postcoital http://bmjopen.bmj.com/ 34 infections, dyspareunia, FSFI and severity of RUTI 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 23 BMJ Open Page 24 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Flow diagram describing numbers of individuals at each stage of study 33 254x190mm (72 x 72 DPI) http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Supplementary Table 1 Health related quality of life assessment before 4 treatment initiation and after 12 months for the HA + CS and standard 5 care group 6 7 Baseline 12 months Baseline 12 months P- 8 value 9 10 HA + CS (N = 79) Standard care (N = 26) 11 12 EQ-5D 3L 0.53 (0.36) 0.76 (0.25) 0.57 (0.31) 0.53 (0.40) 0.020 13 Mean (SD) 14 15 For peerHA + CS (N = 83)review Standard only care (N = 27) 16 EQ-5D 59.53 (18.34) 75.88 (16.90) 62.5 (11.67) 66.84 (14.17) <0.001 17 18 VAS 19 Mean (SD) 20 21 HA + CS (N = 67) Standard care (N = 59) 22 23 SF-36 PCS 61.88 (10.14) 76.99 (11.69) 60.68 (10.02) 74.66 (13.32) 0.118 24 Mean (SD) 25 26 SF-36 MCS 61.74 (10.26) 77.35 (11.33) 60.68 (10.02) 74.66 (13.32) 0.105 27 Mean (SD) 28 29 MCS = Mental component score; PCS = Physical component score; SD = 30 31 standard deviation; VAS = visual analogue scale 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Supplementary Table 2 Resource consumption before treatment and 4 after 12 months follow-up. Median (IQR) 5 6 Resources HA + CS Standard Care P Value* 7 8 (N=181) (N=95) 9 N. GP visits Before treatment 5 (3-6) 3 (3-4) 10 11 Follow-up 2 (1.5-3) 0 (0-1) 0.964 12 13 N. Specialist visits 6 (2-10) 6 (4-8) 14 2 (1-5) 4 (4-4) 0.644 15 For peer review only 16 N. Urine microscopic analysis 5 (3-7) 6 (6-8) 17 18 2 (0-3) 4 (1-4) 0.219 19 20 N. Urine culture 4.5 (3-7) 6 (5-8)
21 2 (1-4) 1 (0-3) <0.001 22 23 N. Urine cytology 2 (1-3) 3 (1-3) 24 25 1 (1-1) 0 (0-1) 0.003 26 N. Vaginal swab test 3 (1-3) 3 (1-3) 27 28 0 (0-0) 0 (0-0) 0.809 29 30 N. Transabdominal ultrasound 2 (1-2) 2 (1-2) 31 1 (0-1) 0 (0-1) 0.091 32 33
N. Transvaginal ultrasound 1 (1-2) 2 (1-2) http://bmjopen.bmj.com/ 34 35 0 (0-0) 0 (0-0) 0.938 36 N. Uroflowmetry 1 (0-1) 0 (0-1) 37 38 0 (0-1) 0 (0-1) 0.933 39 40 N. Invasive urodynamic test 1 (1-1) 0 (0-1) 41
1 (1-1) on September 29, 2021 by guest. Protected copyright. 42 0 (0-0) 0.777 43 N. Cystoscopy with bladder byopsy 0 (0-1) 0 (0-0) 44 45 0 (0-0) 0 (0-0) 0.234 46 47 N. Cystoscopy 1 (1-3) 1 (0-1) 48 1 (0-1) 0 (0-1) 0.047 49 50 N. Antibiotic prescriptions - - 51 52 0 (0-1)^ 1 (0-1) 0.001 53 IQR = Interquartile range 54 55 * Wilcoxon-Mann-Withney test on the differences in the before treatment/ 56 follow-up resource consumption between the two groups 57 ^ mean (SD), 0.99 (1.85) in HA + CS vs 1.31 (2.47) in Standard care 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 STROBE Statement—Checklist of items that should be included in reports of case-control studies 3 Item Page 4 No Recommendation 5 Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or 1 6 7 the abstract 8 (b) Provide in the abstract an informative and balanced summary of what 2 9 was done and what was found 10 11 Introduction 12 Background/rationale 2 Explain the scientific background and rationale for the investigation 4-5 13 being reported 14 Objectives 3 State specific objectives, including any prespecified hypotheses 5 15 For peer review only 16 Methods 17 Study design 4 Present key elements of study design early in the paper 5 18 Setting 5 Describe the setting, locations, and relevant dates, including periods of 5-6 19 recruitment, exposure, follow-up, and data collection 20 21 Participants 6 (a) Give the eligibility criteria, and the sources and methods of case 6 22 ascertainment and control selection. Give the rationale for the choice of 23 cases and controls 24 (b) For matched studies, give matching criteria and the number of NA 25 26 controls per case 27 Variables 7 Clearly define all outcomes, exposures, predictors, potential 6-7 28 confounders, and effect modifiers. Give diagnostic criteria, if applicable 29 Data sources/ 8* For each variable of interest, give sources of data and details of methods 6-7 30 31 measurement of assessment (measurement). Describe comparability of assessment 32 methods if there is more than one group 33 Bias 9 Describe any efforts to address potential sources of bias 7 http://bmjopen.bmj.com/ 34 Study size 10 Explain how the study size was arrived at 6 35 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If 7 36 37 applicable, describe which groupings were chosen and why 38 Statistical methods 12 (a) Describe all statistical methods, including those used to control for 7 39 confounding 40 (b) Describe any methods used to examine subgroups and interactions 7 41
c on September 29, 2021 by guest. Protected copyright. 42 ( ) Explain how missing data were addressed 7 43 (d) If applicable, explain how matching of cases and controls was NA 44 addressed 45 (e) Describe any sensitivity analyses 7 46 47 Results 48 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers 8 49 potentially eligible, examined for eligibility, confirmed eligible, included 50 in the study, completing follow-up, and analysed 51 52 (b) Give reasons for non-participation at each stage 8 53 (c) Consider use of a flow diagram Figure 1 54 Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, 8 55 social) and information on exposures and potential confounders 56 57 (b) Indicate number of participants with missing data for each variable of Table 1 58 interest 4 Suppl 59 Outcome data 15* Report numbers in each exposure category, or summary measures of 8 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml1 BMJ Open Page 28 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 exposure 3 Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted Table 2 4 estimates and their precision (eg, 95% confidence interval). Make clear 5 6 which confounders were adjusted for and why they were included 7 (b) Report category boundaries when continuous variables were Tables 8 categorized 9 (c) If relevant, consider translating estimates of relative risk into absolute NA 10 risk for a meaningful time period 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml2 Page 29 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity 9 4 analyses 5 6 7 Discussion 8 Key results 18 Summarise key results with reference to study objectives 9 9 Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or 10- 10 11 imprecision. Discuss both direction and magnitude of any potential bias 11 12 Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, 12 13 multiplicity of analyses, results from similar studies, and other relevant evidence 14 Generalisability 21 Discuss the generalisability (external validity) of the study results 10 15 For peer review only 16 Other information 17 Funding 22 Give the source of funding and the role of the funders for the present study and, if 8 18 applicable, for the original study on which the present article is based 19 20 21 *Give information separately for cases and controls. 22 23 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and 24 published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely 25 26 available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at 27 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is 28 available at http://www.strobe-statement.org. 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml3 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
Intravesical administration of combined hyaluronic acid (HA) and chondroitin sulphate (CS) for the treatment of female recurrent urinary tract infections: a European multicenter nested case-control study
For peer review only
Journal: BMJ Open
Manuscript ID bmjopen-2015-009669.R2
Article Type: Research
Date Submitted by the Author: 20-Jan-2016
Complete List of Authors: Ciani, Oriana; University of Exeter , University of Exeter Medical School; CeRGAS Bocconi, Arendsen, Erik; Diaconessenhuis, Dept. of Urology Romancik, Martin; St. Cyril and Method University Hospital, Dept. of Urology Lunik, Richard; Fakultná nemocnica s poliklinikou, Dept. of Urology Costantini, Elisabetta; University of Perugia, Dept. of Surgical and Biomedical Science Di Biase, Manuel; University of Perugia, Dept. of Surgical and Biomedical Science Morgia, Giuseppe; University of Catania, Dept. of Urology
Fragala', Eugenia; University of Catania, Dept. of Urology http://bmjopen.bmj.com/ Tomaskin, Roman; Jessenius School of Medicine, Dept. of Urology Bernat, Marian; Fakultná nemocnica s poliklinikou, Dept. of Urology Guazzoni, Giorgio; Humanitas Clinical and Research Center, Dept. of Urology Tarricone, Rosanna; Bocconi University, Dept. of Policy Analysis and Public Management Lazzeri, Massimo; Humanitas Clinical and Research Center, Dept. of Urology on September 29, 2021 by guest. Protected copyright. Primary Subject Urology Heading:
Secondary Subject Heading: Infectious diseases, Health services research
Urinary tract infections < UROLOGY, Antimicrobial Resistance, Hyaluronic Keywords: acid, chondroitin sulphate
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 4 5 Intravesical administration of combined hyaluronic 6 7 8 acid (HA) and chondroitin sulphate (CS) for the 9 10 treatment of female recurrent urinary tract infections: 11 12 a European multicenter nested case-control study 13 14 15 For1 peer2 review 3 only 4 16 Oriana Ciani , Erik Arendsen , Martin Romancik , Richard Lunik , Elisabetta 17 Costantini5, Manuel Di Biase 5, Giuseppe Morgia6, Eugenia Fragalà6, 18 7 8 9 19 Tomaskin Roman , Marian Bernat , Giorgio Guazzoni , Rosanna 20 1,10 9 21 Tarricone , Massimo Lazzeri 22 23 1 24 Centre For Research on Health and Social Care Management, Università 25 26 Bocconi, via Rontgen, 1, 20136, Milan, Italy 27 2 Diaconessenhuis, Maatschap, Urologie, Leiden, The Netherlands 28 3 29 St. Cyril and Method University Hospital, Dept. of Urology, Bratislava, 30 31 Slovakia 32 4 Fakultná nemocnica s poliklinikou, Dept. of Urology, Prešov, Slovakia 33 5 http://bmjopen.bmj.com/ 34 University of Perugia, Dept. of Surgical and Biomedical Science, Urology 35 36 and Andrology Clinic, Perugia, Italy 37 6 University of Catania, Dept. of Urology, Catania, Italy 38 7 39 Jessenius School of Medicine, University Hospital, Dept. of Urology, Martin, 40 41 Slovakia 42 8 FNsP, Urologicka Klinika, Nové Zámky, Slovakia on September 29, 2021 by guest. Protected copyright. 43 9 44 Department of Urology, Humanitas Clinical and Research Center, 45 46 Humanitas University, Milan, Italy 47 10 Department of Policy Analysis and Public Management, Università Bocconi, 48 49 Via Roentgen, 1, 20136 Milano, Italy 50 51 Correspondence to: [email protected] 52 Trial registration: ClinicalTrials.gov NCT02016118 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 1 BMJ Open Page 2 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Word count: 3,579 (excluding references) 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 2 Page 3 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 4 Abstract 5 6 Objectives: To compare the clinical effectiveness of the intravesical 7 administration of combined hyaluronic acid and chondroitin sulphate (HA + 8 9 CS) vs. current standard management in adult women with recurrent urinary 10 11 tract infections (RUTIs). 12 Setting: A EU based, multicenter, retrospective nested case control study. 13 14 Participants: 276 adult women treated for RUTIs starting from 2009 to 2013. 15 For peer review only 16 Interventions: Patients treated with either intravesical administration of HA + 17 CS or standard of care (antimicrobial/immunoactive prophylaxis/ probiotics/ 18 19 cranberry). 20 21 Primary and secondary outcome measures: The primary outcome was 22 occurrence of bacteriologically confirmed recurrence within 12 months. 23 24 Secondary outcomes were time to recurrence, total number of recurrences, 25 26 health related quality of life and healthcare resource consumption. Crude and 27 adjusted results for unbalanced characteristics are presented. 28 29 Results: 181 patients treated with HA + CS and 95 patients treated with 30 31 standard of care from 7 centers were included. The crude and adjusted OR 32 (95% CI) for the primary endpoint were 0.77 (0.46 to 1.28) and 0.51 (0.27 to 33 http://bmjopen.bmj.com/ 34 0.96), respectively. However no evidence of improvement in terms of total 35 36 number or recurrences (incidence rate ratio (95%CI), 0.99 (0.69 to 1.43)) or 37 time to first recurrence was seen (hazard ratio (95%CI), 0.99 (0.61 to 1.61)). 38 39 The benefit of intravesical HA + CS therapy improves when the number of 40 41 instillations is ≥ 5. on September 29, 2021 by guest. Protected copyright. 42 Conclusions: Our results show that bladder instillations of combined HA + 43 44 CS reduces the risk of bacteriologically confirmed recurrences compared to 45 46 the current standard management of RUTIs. Total incidence rates and hazard 47 rates were instead non significantly different between the two groups after 48 49 adjusting for unbalanced factors. In contrast to what happens with antibiotic 50 51 prophylaxis, the effectiveness of the HA + CS reinstatement therapy improves 52 over time. 53 54 55 56 Trial registration: ClinicalTrials.gov NCT02016118 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 3 BMJ Open Page 4 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Article Summary 4 5 'Strengths and limitations of this study' 6 7 • This real world data shows bladder instillations of combined hyaluronic 8 acid and chondroitin sulphate may reduce the risk of bacteriologically 9 10 confirmed urinary tract infections vs current standard management 11 12 • However if the recurrence occurs, there is no evidence of benefit in 13 terms of total number or time to first recurrence 14 15 • NumberFor of instillationspeer seems review an important marker only of success for this 16 17 non antimicrobial therapy 18 • Due to the retrospective observational design, these findings need 19 20 confirmation from prospective and preferably randomised studies 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 4 Page 5 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 4 Background 5 6 Urinary tract infection (UTI) is a major healthcare concern in women with an 7 annual incidence of 30 per 1000 (1). Nearly 33% of women will have had at 8 9 least one UTI episode, with characteristics of acute cystitis, requiring 10 11 antimicrobial therapy by the age of 24 years and as many as 60% of women 12 report having had a UTI in their lifetime (2, 3). 13 14 UTIs have a propensity to recur (4, 5); evidence shows that between 24% and 15 For peer review only 16 50% of initial episodes are followed by a second infection within six months 17 (6 9). A widely accepted definition of RUTIs is two or more UTI episodes over 18 19 six months, or three or more episodes over 12 months (10). On a population 20 21 scale, high incidence and prevalence of RUTIs results in considerable 22 healthcare costs, at the individual level the impact of this condition on health 23 24 related quality of life (HRQoL) is not negligible (11 13). 25 26 The pathogenesis of RUTI involves colonisation of the vagina with 27 uropathogenic bacteria and subsequent migration per urethra to the bladder. 28 29 About 68 77% of recurrences caused by Escherichia coli involve strains 30 31 genetically indistinguishable from those that caused previous infections (14). 32 The diagnosis is often made on clinical presentation with local genitourinary 33 http://bmjopen.bmj.com/ 34 symptoms of dysuria, frequency, and urgency or hesitancy appearing 35 36 suddenly (14). However, urine culture is useful in women presenting with 37 RUTI to confirm the diagnosis, direct antimicrobial therapy and exclude 38 39 infection from overactive bladder or interstitial cystitis (15). Evidence based 40 41 clinical practice guidelines recommend empiric initial therapy for acute on September 29, 2021 by guest. Protected copyright. 42 management or continuous antimicrobial therapy or self initiated therapy and 43 44 prophylaxis, either antimicrobial or non antimicrobial based (16, 17). 45 46 The choice of specific strategy for care depends on the number of 47 recurrences experienced per year, the patient’s preferences and careful 48 49 review of modifiable risk factors (14, 18). As the second most common reason 50 51 for prescribing antibiotics (following otitis media), there is currently increasing 52 concern about empiric use of these agents due to increased antimicrobial 53 54 resistance. Antibiotic use selects for resistant pathogens: a major risk factor 55 56 for an antibiotic resistant UTI is prior antibiotic use (5). In an international 57 survey investigating the prevalence and susceptibility of pathogens causing 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 5 BMJ Open Page 6 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 cystitis, 10.3% of Escherichia coli isolates were resistant to at least three 4 5 different classes of antimicrobial agents, including ampicillin (48.3%), 6 trimethoprim/sulfamethoxazole (29.4%) and nalidixic acid (18.6%) (19). 7 8 Nonantimicrobial prevention strategies have become popular in the age of 9 10 increasing antimicrobial use and resistance. However, no probiotic agent has 11 been approved for therapeutic use and the potential benefit of cranberry in 12 13 terms of product type (solid vs liquid), dosing, and optimal patient population 14 15 remainsFor to be elucidated peer (14, 18). review A new therapy based only on the reinstatement 16 of the glycosaminoglycan (GAG) bladder epithelium has recently been 17 18 proposed for the treatment of RUTIs (20). This GAG layer consists of non 19 20 sulphated, e.g. hyaluronic acid (HA), and sulphated, e.g. heparan sulphate 21 and heparin, chondroitin sulphate (CS), dermatan sulphate and keratan 22 23 sulphate, GAGs. Limited evidence has shown the preventive activity of 24 25 intravesical GAG substation therapy (with HA alone or with HA + CS) on 26 recurrence of infections in patients with recurrent bacterial cystitis (21). 27 28 However, large scale studies are needed to underline the benefit of this 29 30 therapy (22). 31 Therefore, we decided to perform a European retrospective multicenter study 32 33 to compare the clinical effectiveness of the intravesical administration of http://bmjopen.bmj.com/ 34 35 combined HA + CS (Ialuril®, IBSA Institut Biochimique SA) vs. current 36 standard management of RUTIs in adult women. 37 38 39 40 41 Methods 42 on September 29, 2021 by guest. Protected copyright. 43 Study design 44 45 This was a EU based, multicenter, retrospective nested case control 46 comparison of individual patient data collected from electronic medical 47 48 records and/or administrative databases available at the participating 49 50 institutions. Centers using the intravesical administration of combined HA + 51 CS, in the countries where Ialuril® was already registered and on the market 52 53 (Ialuril® received a CE mark for this indication in 2009), were identified and 54 55 invited to take part in the study. 56 57 58 Study population 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 6 Page 7 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 All patients treated with either HA + CS or standard of care at the participating 4 5 centers, high volume organizations with specific expertise in the treatment of 6 UTIs, starting from 2009 were included if they were women, aged 18 75 7 8 years, diagnosed with RUTIs, defined as at least three episodes of 9 10 uncomplicated UTIs accompanied by clinical symptoms and documented by 11 urine culture with the isolation of >103 CFU/ml of an identified pathogen in the 12 13 last 12 months. Uncomplicated UTI is defined as an infection in a person with 14 15 normal urinaryFor tract peer and function (17).review Women with complicated only UTIs (i.e. 16 individuals with functional or structural abnormalities of the genitourinary tract) 17 18 were excluded. Within Europe, patients at first diagnosis of RUTIs are offered 19 20 an approach based on behavioral changes, antimicrobial prophylaxis or 21 aspecific non antimicrobial prevention. However, several women refuse to 22 23 take antimicrobials over an extended period of time, hence intravesical 24 25 administration of HA + CS is intended for women refractory or not satisfied 26 with first line management of RUTIs. Based on a previous cohort study (11), 27 28 we estimated that 208 patients were needed to observe a 50% difference in 29 30 the proportions of patients recurring between the two groups within 12 months 31 with 90% power and alpha level of 0.05. 32 33 http://bmjopen.bmj.com/ 34 35 Groups and interventions 36 Patients treated with intravesical administration of combined HA 1.6% and CS 37 38 2.0%. The recommended scheme is one instillation per week for the first 39 40 month, followed by one instillation every two weeks for the second month and 41
one instillation per month afterward until stable remission of the symptoms, on September 29, 2021 by guest. Protected copyright. 42 43 however different patterns are seen in clinical practice. These patients were 44 45 compared to patients treated with antimicrobial prophylaxis (continuous or 46 postcoital), or immunoactive prophylaxis or prophylaxis with probiotics or 47 48 prophylaxis with cranberry, or combination of these (17), as recommended by 49 50 the European Association of Urology. 51 52 53 Study outcomes 54 55 The primary outcome for this study was the occurrence of objective urinary 56 tract infection recurrence, defined as the occurrence of at least one 57 58 bacteriologically confirmed urinary tract infection within 12 months after 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 7 BMJ Open Page 8 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 treatment initiation for RUTIs. According to current clinical guidelines, in non 4 5 pregnant women, urine culture is recommended in symptomatic patients only. 6 Information about clinically confirmed recurrences were also sought although 7 8 they are not reported in this manuscript as they are assumed to be less 9 10 objective than the bacteriologically confirmed ones. Patients who developed a 11 UTI whilst on the HA + CS instillation protocol were treated according to 12 13 clinical guidelines with antibiotics but could continue the instillations afterward. 14 15 After theFor first bacteriologically peer confirmed review recurrence, notonly only the time to first 16 recurrence was recorded, but also the number of additional UTIs. The 17 18 secondary outcome measures were the time to recurrence (defined as the 19 20 time from the start of the treatment until the occurrence of the first objective 21 recurrence); the total number of recurrences; HRQoL as assessed through 22 23 the SF 36 (23) or EQ 5D (24) questionnaires. Dutch (25), Italian (26) and UK 24 25 (27) tariffs were used to estimate utility values from the EQ 5D questionnaires 26 in the Netherlands, Italy and Slovakia, respectively. Information about 27 28 healthcare resource consumption were also collected. A cost analysis was 29 30 planned and will be the subject of a future publication. 31 32 33 Data collection http://bmjopen.bmj.com/ 34 35 General patient demographic characteristics, diagnosis and treatment 36 information were collected based on a predefined form designed on the input 37 38 obtained from collaborating centers during a workshop held in July 2013. An 39 40 intuitive electronic system was implemented (Advice Pharma ltd) to record 41
and store data on a secure remote server provider. on September 29, 2021 by guest. Protected copyright. 42 43 44 45 Statistical analyses 46 Continuous baseline characteristics are presented as median and interquartile 47 48 range (IQR) or mean and standard deviation (SD), as appropriate. For 49 50 proportions, absolute and relative frequencies are reported. Wilcoxon–Mann– 51 Whitney test or t test were used for continuous and ordinal variables baseline 52 53 differences, whereas the Chi Square test was used for proportions. In our 54 55 primary analyses, we applied univariate and multivariate logistic, poisson and 56 Cox model for objective recurrence, number of recurrences and time to 57 58 recurrence outcomes, respectively. Results were presented as crude and 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 8 Page 9 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 adjusted odds ratio (OR), incidence rate ratio (IRR) and hazard ratio (HR) 4 5 respectively, with their 95% confidence intervals (CI). Adjusting variables were 6 age, body mass index (BMI), employment and menopause status, postcoital 7 8 infections, dyspareunia, Female sexual function index (FSFI) and severity of 9 10 RUTI. A prespecified sensitivity analysis was conducted to investigate the 11 impact of adherence to HA + CS treatment on clinical outcomes considering 12 13 patients who had ≥ 5 instillations. Pairwise deletion was used to deal with 14 15 missing Fordata. All significance peer tests review were two tailed at theonly 0.05 significance 16 level. All the analyses were conducted using Stata SE StataCorp LP 11. 17 18 19 20 Ethical approval and funding 21 The study protocol was reviewed and approved at the coordinating center by 22 23 an Independent Ethics Committee at the Department of Urology, University of 24 25 Perugia. The study was funded by the Study Group for Urogenital Diseases in 26 Italy, an association on practioners and clinical urologists. The study is 27 28 registered at clinicaltrials.gov (NCT02016118). Extra data is available by 29 30 emailing the corresponding author. 31 32 33 http://bmjopen.bmj.com/ 34 Results 35 36 Overall, 276 patients treated for RUTIs at seven European centers from 37 January 2009 up to December 2013 were included in the analyses. Of these, 38 39 181 women were treated with HA + CS intravesical administration and 95 40 41 women received standard management of RUTIs. The numerical imbalance on September 29, 2021 by guest. Protected copyright. 42 was probably due to the participating organizations being tertiary referral 43 44 centers for patients who are not satisfied with standard management of 45 46 RUTIs. A flow diagram reporting number of patients at each stage of the study 47 is shown in Figure 1. Baseline socio demographic and clinical characteristics 48 49 of patients are reported in Table 1. Given the non experimental nature of the 50 51 study, the distribution of several characteristics was not homogeneous 52 between the two groups, in particular women treated with HA + CS were 53 54 older, with a higher BMI and probability of dyspareunia. 55 56 57 Primary analyses 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 9 BMJ Open Page 10 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 In the HA + CS group, 55.7% of patients showed bacteriologically confirmed 4 5 recurrences, whereas 62.1% had such recurrence in the standard of care 6 group (p = 0.313). However, the adjusted OR (95% CI) for developing a 7 8 bacteriologically confirmed recurrence within 12 months was 0.51 (0.27 to 9 10 0.96), meaning that, other characteristics being equal, there is a 49% reduced 11 risk of developing a recurrence in patients treated with HA + CS compared to 12 13 standard care (Table 2). When the number of recurrences is considered, in 14 15 the HA +For CS group peerthere were 121 review bacteriologically confirmedonly recurrences in 16 61.5 person year whereas in the standard treatment group there were 59 17 18 bacteriologically confirmed recurrences in 51.1 person year (p = 0.001). 19 20 However we observed an adjusted IRR (95% CI) of 0.99 (0.69 to 1.43), 21 showing non significant differences in the incidence rates between the group 22 23 treated with HA + CS and the control. Similar results were obtained from the 24 25 univariate and multivariate Cox regression models used to estimate the HR 26 (95% CI) for the time to first bacteriologically confirmed recurrence, with an 27 28 unadjusted estimate of 0.99 (0.61 to 1.61) (Table 2). Although during the 12 29 30 months follow up the median time to first recurrence was higher in the 31 standard care group (169.5 days (IQR, 72.5 to 341.5) vs 320 days (IQR, 179 32 33 to 365); p value < 0.001), we observed the distribution of recurrences at http://bmjopen.bmj.com/ 34 35 separate follow up times, and noted that the incident proportion of patients 36 who developed the recurrence vs those still at risk was lower in the HA + CS 37 38 group in the latest part of follow up, after 8 months (Table 3). All patients were 39 40 alive at 12 months follow up. There were 14 all cause hospitalizations in the 41
HA + CS and 1 in the control group. on September 29, 2021 by guest. Protected copyright. 42 43 44 45 Health related quality of life and resources consumption 46 In a subset of patients, a measure of the HRQoL as measured through the 47 48 SF 36 or EQ 5D 3L questionnaires was available at baseline and after 12 49 50 months follow up. There was no evidence of better improvement in HRQoL in 51 the HA + CS group compared to control with SF 36 results, whereas when 52 53 EQ 5D data were considered the HA + CS group seemed to have received a 54 55 higher benefit in terms of HRQoL than the control (Supplementary table 1). 56 There is a general reduction in physical units of health resources (i.e. medical 57 58 visits, laboratory and imaging tests) consumed by the two groups before the 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 10 Page 11 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 treatment and during the follow up (Supplementary table 2) without significant 4 5 differences between the two groups. 6 7 8 Sensitivity analyses 9 10 We repeated all primary analyses and considered different exposure intensity 11 (i.e. number of intravesical administrations received) in the HA + CS group. All 12 13 findings consistently show additional benefit gained to patients when the 14 15 number Forof instillations peer increases, possiblyreview revealing the only importance of 16 adherence to this medical device therapy for the treatment of RUTIs (Table 4). 17 18 As a post hoc subgroup analysis, we repeated primary analyses in non 19 20 sexually active patients only and obtained similar patterns of results as in the 21 whole sample, although with loss of statistical significance. 22 23 24 25 26 Discussion 27 In this European multicenter retrospective observational study we compared 28 29 bacteriologically confirmed recurrence rates at 12 months follow up after the 30 31 initiation of intravesical administration of HA + CS vs standard of care for the 32 treatment of RUTIs. After adjusting for unbalanced confounding factors 33 http://bmjopen.bmj.com/ 34 between the two groups, we observed that the HA + CS patients had a 49% 35 36 reduction (OR 0.51, 95% CI: 0.27 to 0.96) in the risk of a bacteriologically 37 confirmed recurrence, whereas there was no statistical evidence for a 38 39 difference in the incidence rates and hazard rates of such recurrences 40 41 between the two groups. on September 29, 2021 by guest. Protected copyright. 42 Four clinical studies (28 31) have been performed to investigate the efficacy 43 44 and tolerability of intravesically administered GAG for RUTI prophylaxis, all 45 46 showing that HA alone or HA + CS instillations reduce the number of UTI per 47 patient per year at no increased risk of severe adverse events and prolong the 48 49 time interval between RUTI episodes, with a high rate of patients being free of 50 51 recurrence at the end of the study period. In particular, two RCTs studies 52 compared HA + CS administration to either placebo (29) or long term 53 54 antibiotic prophylaxis using sulfamethoxazole 200 mg and trimethoprim 40 mg 55 56 (30). Damiano and colleagues report a decrease in UTI rate per patient of 57 77% (95% CI, 72.3 to 80.8) in the experimental vs placebo group, whereas De 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 11 BMJ Open Page 12 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Vita et al. report the mean ± SD number of recurrent cystitis per patient per 4 5 year as 1 ± 1.2 vs 2.3 ± 1.4 in HA + CS and antibiotic treated patients, 6 respectively. Despite the prospective and randomized design, these trials 7 8 were limited by the small sample size (i.e. they included 57 and 28 patients, 9 10 respectively) and the single center setting that considerably reduces the 11 generalizability of the findings. Whilst waiting for definitive RCT evidence 12 13 clarifying the comparative effectiveness profile of this therapy in support of its 14 15 adoption,For our observational peer study reviewdesign provides useful only information around 16 its effectiveness in real world practice. 17 18 In this respect, our study involved seven centers across three European 19 20 countries and 276 patients, thus providing important additional evidence with 21 respect of current treatment options for RUTIs. Furthermore, the non 22 23 experimental observational design allows for a closer representation of the 24 25 routine clinical practice of the use of the HA + CS reinstatement therapy as 26 compared to standard of care in place at high volume university hospitals, that 27 28 is on purpose defined as very broad given the variety of recommended 29 30 strategies (17) and general scarce adherence to clinical guidelines (32). 31 On the other hand, the retrospective design limited the availability of data to 32 33 that previously collected at the centers. Contacting patients ex post to gather http://bmjopen.bmj.com/ 34 35 additional data was not applicable (e.g. as in the case of HRQoL assessment) 36 or not helpful given the potential significant recall bias introduced by delayed 37 38 reporting. The issue of missing data was dealt with by assuming they were 39 40 missing at random (i.e. given the observed data, data are missing 41
independently of unobserved data, that is missing data does not depend on on September 29, 2021 by guest. Protected copyright. 42 43 the level of their outcome) and applying pairwise deletion. In this regard, we 44 45 performed two additional analyses, first by restricting the primary analyses to 46 all complete cases (i.e. no missing in both outcomes or adjusting factors) and 47 48 providing similar results to those presented here (data not shown). Second, 49 50 for all outcomes and adjusting variables we tested through Fisher’s exact test 51 whether proportions of missing was different between HA + CS and Standard 52 53 of Care groups. No significant difference was observed with the exception of 54 55 the resource consumption where the number of missing was higher in the HA 56 + CS group. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 12 Page 13 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Data on uropathogens and antimicrobial resistance within the groups was 4 5 unfortunately not available from this database, although we know most 6 commonly prescribed antibiotics were Ciprofloxacin (13.2% of all 7 8 prescriptions), Cefuroxime (6.9%), Fosfomycin (6.9%), Nitrofurantoin (6.4%) 9 10 and E. Coli bacterial extract (OM 89, 4.8%). 11 Health related quality of life assessment in routine practice is still uncommon 12 13 as indicated by the significant proportion of missing information (up to 73% in 14 15 the controlFor group) onpeer this outcome. review However our findings only are in line with 16 previous reports showing that the GAG replacement treatment in women with 17 18 RUTIs had a positive impact on patients’ quality of life, reducing the 19 20 symptoms and improving the maximum cystometric capacity (29, 30). 21 Recurrent UTIs in women are a common condition, associated with significant 22 23 morbidity and burden for the whole society. In a study of 684 women aged 18 24 25 to 70 years with UTI, participants reported an average of 3.83 symptom days, 26 2.89 restricted activity days, and 3.13 days during which they were unwell 27 28 (33). In another study patients reported 1.2 days in which they were not able 29 30 to attend classes or work, and 0.4 days in bed (34). New effective prevention 31 strategies are needed, in particular non antimicrobial approaches would be 32 33 desirable for several reasons. First of all, a prolonged antibiotic use as a http://bmjopen.bmj.com/ 34 35 prophylactic approach to RUTI increases the risk of side effects, including 36 vaginal and oral candidiasis, and gastrointestinal symptoms (10). This in turn, 37 38 lowers patients’ compliance and therefore the effectiveness of the treatment 39 40 (35). However, the most worrying effect of the antibiotic use (and misuse) is 41
the exacerbation of the antimicrobial resistance (AMR) (36). Recently, a UK on September 29, 2021 by guest. Protected copyright. 42 43 commissioned report on health and macroeconomic consequences of AMR 44 45 estimated 10 million extra deaths a year and costs up to €90tn for the global 46 economy by 2050 if this problem is not properly tackled (37). Although this 47 48 first report might not be as scientifically rigorous or informed by evidence as 49 50 possible (38), it brought renewed interest in the worldwide AMR crisis. The 51 war against the spread of drug resistant microbes is attracting considerable 52 53 attention as well as investment from all major governments and research 54 55 organizations in EU and beyond (39,40). The way forward outlined by all of 56 these major research initiatives includes establishing appropriate funding and 57 58 rewards to subsidize access to and development of new antibiotic agents, 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 13 BMJ Open Page 14 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 preservation of existing drugs antimicrobial activity through prescription 4 5 tailored to diagnosis, prioritization and controlled access, and identification of 6 novel approaches and therapies for microbial diseases. 7 8 The case of GAG reinstatement therapy is a good example of an innovative 9 10 approach to prevent and manage bacterial urinary infections, a medical 11 device intervention as opposed to a drug. In contrast to antibiotic therapy, 12 13 which aims at eradicating pathogens, treatment with HA + CS targets 14 15 bacterialFor adherence peer to the bladder review mucosa by physically only recovering a 16 damaged GAG layer that facilitates bacterial adherence and, therefore, 17 18 RUTIs. Although patients who benefit from the treatment in first place might 19 20 decide to undertake higher number of instillations compared to patients who 21 do not benefit immediately, the different mechanism of action could explain 22 23 the apparent reduction in the incidence of UTIs in the group treated with HA + 24 25 CS instillations compared with standard care when considering later time 26 intervals (Table 3). Whilst antibiotics are immediately effective, although 27 28 subject and conducive to resistance, GAG layer administration is 29 30 progressively restoring the epithelium that will protect women from future 31 uropathogen infections. On the other hand, catheterization induced UTIs 32 33 might represent an unintended consequence of this procedure. Previous http://bmjopen.bmj.com/ 34 35 reports (29, 30) have highlighted good tolerability and safety of the 36 intervention that must be performed under sterile conditions by nurses trained 37 38 in the procedure. As regards the economic profile of the two alternative 39 40 approaches, it has been reported that the cost of the HA + CS could be even 41
five times higher than the cost for a 6 month antibiotic prophylaxis. However, on September 29, 2021 by guest. Protected copyright. 42 43 this consideration corresponds to a very restrictive, if not naïve, cost analysis 44 45 as it is well known that all direct healthcare costs and consequences, 46 including those for the wider society, as would be containment of drug 47 48 resistance spreading, should be taken into account when assessing the cost 49 50 effectiveness profiles of health technologies. Future methodologically sound 51 economic evaluation studies are recommended to compare the societal or 52 53 payer value of the two treatment strategies. 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 14 Page 15 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Conclusions 4 5 In order to treat and prevent RUTI, there is a need of effective and safe 6 alternative strategies to antimicrobial therapy. Our study showed that in real 7 8 world setting bladder instillation of combined HA + CS may reduce the risk of 9 10 bacteriologically confirmed recurrences compared to the current standard 11 management in this study population. Total incidence rates and hazard rates 12 13 were instead non significantly different between the two groups. The number 14 15 of HA + ForCS instillations peer seems to reviewbe an important marker only of success for the 16 intravesical administration therapy. Furthermore, in contrast to what happens 17 18 with antibiotic prophylaxis, due to side effects and development of resistance, 19 20 the effectiveness of GAG reinstatement therapy improves over time, with an 21 even better expected comparative effectiveness profile in the long run. 22 23 Although firm conclusions are difficult due to the retrospective observational 24 25 design, these findings highlight the relevance of additional prospective and 26 randomized studies in this area and the promising role of the HA + CS 27 28 reinstatement therapy for prevention and treatment of RUTI in an era of 29 30 worryingly increased antimicrobial resistance. 31 32 33
Contributorship: OC, RT, ML designed the study. EA, MR, RL, EC, MDB, http://bmjopen.bmj.com/ 34 35 GM, EF, TR, MB, GG contributed to the data collection. OC analyzed the data 36 37 and drafted the manuscript. All authors commented and approved the final 38 version of the manuscript. 39 40 Competing interest: No competing interest**. 41 42 Funding: This study was funded by an unrestricted grant from the TETI on September 29, 2021 by guest. Protected copyright. 43 Association Study group for urogenital diseases (www.tetiassociation.com). 44 45 This is a practitioners and clinical urologists association in Italy. Members of 46 47 the association were involved in the data collection and revised the 48 manuscript. 49 50 Data sharing: The dataset is available by emailing the corresponding author. 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 15 BMJ Open Page 16 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 4 References 5 6 7 1. Laupland KB, Ross T, Pitout JD, Church DL, Gregson DB. Community 8 onset urinary tract infections: a population based assessment. Infection. 9 10 2007;35(3):150 3. 11 2. Foxman B. Epidemiology of urinary tract infections: incidence, 12 morbidity, and economic costs. The American journal of medicine. 2002;113 13 Suppl 1A:5s 13s. 14 3. Foxman B, Barlow R, D'Arcy H, Gillespie B, Sobel JD. Urinary tract 15 infection:For self reported peer incidence andreview associated costs. only Annals of 16 epidemiology. 2000;10(8):509 15. 17 4. Hooton TM, Scholes D, Hughes JP, Winter C, Roberts PL, Stapleton 18 AE, et al. A prospective study of risk factors for symptomatic urinary tract 19 20 infection in young women. The New England journal of medicine. 21 1996;335(7):468 74. 22 5. Foxman B. The epidemiology of urinary tract infection. Nature reviews 23 Urology. 2010;7(12):653 60. 24 6. Sen A. Recurrent cystitis in non pregnant women. BMJ clinical 25 evidence. 2008;2008. 26 7. Ikaheimo R, Siitonen A, Heiskanen T, Karkkainen U, Kuosmanen P, 27 Lipponen P, et al. Recurrence of urinary tract infection in a primary care 28 setting: analysis of a 1 year follow up of 179 women. Clinical infectious 29 30 diseases : an official publication of the Infectious Diseases Society of 31 America. 1996;22(1):91 9. 32 8. Foxman B. Recurring urinary tract infection: incidence and risk factors. 33 American journal of public health. 1990;80(3):331 3. http://bmjopen.bmj.com/ 34 9. Foxman B, Gillespie B, Koopman J, Zhang L, Palin K, Tallman P, et al. 35 Risk factors for second urinary tract infection among college women. 36 American journal of epidemiology. 2000;151(12):1194 205. 37 10. Albert X, Huertas I, Pereiro, II, Sanfelix J, Gosalbes V, Perrota C. 38 Antibiotics for preventing recurrent urinary tract infection in non pregnant 39 40 women. The Cochrane database of systematic reviews. 2004(3):Cd001209. 41 11. Ciani O, Grassi D, Tarricone R. An economic perspective on urinary 42 tract infection: the "costs of resignation". Clinical drug investigation. on September 29, 2021 by guest. Protected copyright. 43 2013;33(4):255 61. 44 12. Renard J, Ballarini S, Mascarenhas T, Zahran M, Quimper E, Choucair 45 J, et al. Recurrent Lower Urinary Tract Infections Have a Detrimental Effect on 46 Patient Quality of Life: a Prospective, Observational Study. Infectious 47 diseases and therapy. 2014. 48 13. Bermingham SL, Ashe JF. Systematic review of the impact of urinary 49 50 tract infections on health related quality of life. BJU international. 2012;110(11 51 Pt C):E830 6. 52 14. Gupta K, Trautner BW. Diagnosis and management of recurrent 53 urinary tract infections in non pregnant women. BMJ. 2013;346:f3140. 54 15. Arya LA, Northington GM, Asfaw T, Harvie H, Malykhina A. Evidence of 55 bladder oversensitivity in the absence of an infection in premenopausal 56 women with a history of recurrent urinary tract infections. BJU international. 57 2012;110(2):247 51. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 16 Page 17 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 16. Gupta K, Hooton TM, Naber KG, Wullt B, Colgan R, Miller LG, et al. 4 International clinical practice guidelines for the treatment of acute 5 uncomplicated cystitis and pyelonephritis in women: A 2010 update by the 6 Infectious Diseases Society of America and the European Society for 7 Microbiology and Infectious Diseases. Clinical infectious diseases : an official 8 9 publication of the Infectious Diseases Society of America. 2011;52(5):e103 10 20. 11 17. Grabe M, Bartoletti R, Bjerklund Johansen T, Cai T, Çek M, Köves B, 12 et al. Guidelines on Urological Infections. European Association of Urology, 13 2015. 14 18. Shepherd AK, Pottinger PS. Management of urinary tract infections in 15 the era ofFor increasing peer antimicrobial review resistance. The Medical only clinics of North 16 America. 2013;97(4):737 57, xii. 17 19. Schito GC, Naber KG, Botto H, Palou J, Mazzei T, Gualco L, et al. The 18 19 ARESC study: an international survey on the antimicrobial resistance of 20 pathogens involved in uncomplicated urinary tract infections. International 21 journal of antimicrobial agents. 2009;34(5):407 13. 22 20. Lazzeri M, Montorsi F. The therapeutic challenge of "chronic cystitis": 23 search well, work together, and gain results. European urology. 24 2011;60(1):78 80. 25 21. De Vita D, Antell H, Giordano S. Effectiveness of intravesical 26 hyaluronic acid with or without chondroitin sulfate for recurrent bacterial 27 cystitis in adult women: a meta analysis. International urogynecology journal. 28 29 2013;24(4):545 52. 30 22. Madersbacher H, van Ophoven A, van Kerrebroeck PE. GAG layer 31 replenishment therapy for chronic forms of cystitis with intravesical 32 glycosaminoglycans a review. Neurourology and urodynamics. 2013;32(1):9 33 18. http://bmjopen.bmj.com/ 34 23. Brazier JE, Harper R, Jones NM, O'Cathain A, Thomas KJ, Usherwood 35 T, et al. Validating the SF 36 health survey questionnaire: new outcome 36 measure for primary care. Bmj. 1992;305(6846):160 4. 37 24. EuroQol a new facility for the measurement of health related quality of 38 39 life. Health policy (Amsterdam, Netherlands). 1990;16(3):199 208. 40 25. Lamers LM, McDonnell J, Stalmeier PF, Krabbe PF, Busschbach JJ. 41 The Dutch tariff: results and arguments for an effective design for national 42 EQ 5D valuation studies. Health Econ. 2006;15(10):1121 32. on September 29, 2021 by guest. Protected copyright. 43 26. Scalone L, Cortesi PA, Ciampichini R, Belisari A, D'Angiolella LS, 44 Cesana G, et al. Italian population based values of EQ 5D health states. 45 Value in health : the journal of the International Society for 46 Pharmacoeconomics and Outcomes Research. 2013;16(5):814 22. 47 27. Dolan P. Modeling valuations for EuroQol health states. Medical care. 48 49 1997;35(11):1095 108. 50 28. Constantinides C, Manousakas T, Nikolopoulos P, Stanitsas A, 51 Haritopoulos K, Giannopoulos A. Prevention of recurrent bacterial cystitis by 52 intravesical administration of hyaluronic acid: a pilot study. BJU international. 53 2004;93(9):1262 6. 54 29. Damiano R, Quarto G, Bava I, Ucciero G, De Domenico R, Palumbo 55 MI, et al. Prevention of recurrent urinary tract infections by intravesical 56 administration of hyaluronic acid and chondroitin sulphate: a placebo 57 controlled randomised trial. European urology. 2011;59(4):645 51. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 17 BMJ Open Page 18 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 30. De Vita D, Giordano S. Effectiveness of intravesical hyaluronic 4 acid/chondroitin sulfate in recurrent bacterial cystitis: a randomized study. 5 International urogynecology journal. 2012;23(12):1707 13. 6 31. Lipovac M, Kurz C, Reithmayr F, Verhoeven HC, Huber JC, Imhof M. 7 Prevention of recurrent bacterial urinary tract infections by intravesical 8 9 instillation of hyaluronic acid. International journal of gynaecology and 10 obstetrics: the official organ of the International Federation of Gynaecology 11 and Obstetrics. 2007;96(3):192 5. 12 32. Taur Y, Smith MA. Adherence to the Infectious Diseases Society of 13 America guidelines in the treatment of uncomplicated urinary tract infection. 14 Clinical infectious diseases : an official publication of the Infectious Diseases 15 Society Forof America. peer 2007;44(6):769 74. review only 16 33. Little P, Merriman R, Turner S, Rumsby K, Warner G, Lowes JA, et al. 17 Presentation, pattern, and natural course of severe symptoms, and role of 18 19 antibiotics and antibiotic resistance among patients presenting with suspected 20 uncomplicated urinary tract infection in primary care: observational study. 21 Bmj. 2010;340:b5633. 22 34. Ronald A. The etiology of urinary tract infection: traditional and 23 emerging pathogens. The American journal of medicine. 2002;113 Suppl 24 1A:14S 9S. 25 35. Hooton TM, Levy SB. Antimicrobial resistance: a plan of action for 26 community practice. American family physician. 2001;63(6):1087 98. 27 36. Fauci AS, Marston l D. The perpetual challenge of antimicrobial 28 29 resistance. Jama. 2014;311(18):1853 4. 30 37. Review on Antimicrobial Resistance. Antimicrobial Resistance: 31 Tackling a crisis for the health and wealth of nations. 2014. 32 38. Antimicrobial resistance: in terms politicians understand. Lancet. 33 2014;384(9961):2173. http://bmjopen.bmj.com/ 34 39. Woolhouse M, Farrar J. Policy: An intergovernmental panel on 35 antimicrobial resistance. Nature. 2014;509(7502):555 7. 36 40. Geoghegan Quinn M. Funding for antimicrobial resistance research in 37 Europe. Lancet. 2014;384(9949):1186. 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 18 Page 19 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Table 1 Baseline characteristics 4 5 6 Characteristics HA + CS Standard care P - value 7 (N = 181) (N = 95) 8 9 Age – mean (SD) 55.24 (17.33) 48.84 (21.16) 0.017 10 11 Employed – n (%) 83 (46) 62 (65) 0.003 12 Partner – n (%) 148 (82) 72 (76) 0.172 13 14 Sexually active – n (%) 114 (63) 60 (63) 0.931 15 For peer review only 16 Menopause – n (%) 69 (28) 49 (51) 0.038 17 BMI – mean (SD) 26.20 (6.37) 24.56 (5.03) 0.019 18 19 Postcoital infections – n (%) 31 (17) 29 (31) 0.012 20 21 Dyspareunia – n (%) 40 (22) 5 (5) < 0.001 22 23 Severity RUTI – median (IQR)* 4 (2 – 5) 4 (3 – 4) 0.316 24 Prophylaxis n (%) 25 26 Antimicrobial prophylaxis 27 na 42 (44.21) 28 continuous 29 Antimicrobial prophylaxis 30 na 19 (20) 31 postcoital 32 33 Intermittent immunoactive na 28 (29.47) http://bmjopen.bmj.com/ 34 prophylaxis 35 36 On demand immunoactive 37 na 1 (1.05) 38 therapy 39 Others na 5 (5.26) 40 41 FSFI – mean (SD) 5.35 (8.71) 3.13 (5.43) 0.018 42 on September 29, 2021 by guest. Protected copyright. a b 43 EQ 5D– median (IQR) 0.69 (0.24 0.73) 0.69 (0.28 0.81) 0.696 44 c d 45 SF 36 PCS – median (IQR) 60 (54.5 70) 60 (53 – 67.5) 0.500 46 SF 36 MCS – median (IQR) 60 (54 70)e 60 (53 – 67.5)d 0.551 47 48 BMI = body mass index; EQ 5D = Euro QoL 5D 3 level; FSFI = female 49 50 sexual function index; IQR = interquartile range; na = not applicable; SD = 51 standard deviation; RUTI = recurrent urinary tract infection; SF 36 PCS = 52 53 Short From 36 physical component score; SF 36 MCS = Short From 36 54 55 mental component score 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml19 BMJ Open Page 20 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 *According to the European Association of Urology Guidelines on 4 5 Urological Infections where 1 is low severity cystitis and 6 is extreme 6 severity including organ failure (21) 7 8 aN = 90 patients; bN = 29 patients; cN = 72 patients; dN = 60 patients; eN = 9 10 73 patients 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml20 Page 21 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Table 2 Bacteriologically confirmed recurrence, total number of 4 5 recurrences and time to first recurrence between HA + CS vs standard 6 of care treated patients 7 8 9 Outcome OR (95% CI) Adjusted* OR (95% CI) 10 Bacteriologically 11 0.77 (0.46 to 1.28) 0.51 (0.27 to 0.96) 12 confirmed recurrence 13 14 IRR (95% CI) Adjusted* IRR (95% CI) 15 For peer review only 16 Total number of 17 bacteriologically 1.73 (1.27 to 2.37) 0.99 (0.69 to 1.43) 18 19 confirmed recurrence 20 21 HR (95% CI) Adjusted* HR (95% CI) 22 Time to first 23 24 bacteriologically 1.66 (1.09 to 2.54)§ 0.99 (0.61 to 1.61) 25 26 confirmed recurrence 27 OR = odds ratio, IRR = incidence rate ratio, HR = hazard ratio 28 29 *Adjusted for age, BMI, employment and menopause status, postcoital 30 31 infections, dyspareunia, FSFI and severity of RUTI 32 §Log rank test p value 0.018 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 21 BMJ Open Page 22 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Table 3 Incidence of bacteriologically confirmed recurrences during 12 4 5 months follow-up 6 7 Incidence of HA + CS Standard care P value 8 9 bacteriologically 10 11 confirmed 12 recurrences 13 14 0 – 90 days 15.3% 12.7% 0.610 15 For peer review only 16 91 – 180 days 15.6% 12.9% 0.628 17 181 – 240 days 10.7% 3.7% 0.132 18 19 241 – 365 days 16.3% 30.8% 0.043 20 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 22 Page 23 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Table 4 Sensitivity analysis – impact of number of intravesical 4 5 administration of HA + CS on clinical outcomes 6 7 Bacteriologically confirmed recurrence OR (95% CI) Adjusted* OR (95% CI) 8 9 ≥ 5 instillations (n = 156) vs standard of care 0.81 (0.48 to 1.36) 0.52 (0.27 to 0.99) 10 11 ≥ 6 instillations (n = 134) vs standard of care 0.69 (0.40 to 1.18) 0.47 (0.25 to 0.91) 12 ≥ 7 instillations (n = 82) vs standard of care 0.63 (0.34 to 1.14) 0.43 (0.21 to 0.88) 13 14 Total number of bacteriologically IRR (95% CI) Adjusted* IRR (95% CI) 15 For peer review only 16 confirmed recurrence 17 ≥ 5 instillations vs standard of care 1.82 (1.33 to 2.49) 1.05 (0.73 to 1.52) 18 19 ≥ 6 instillations vs standard of care 1.64 (1.18 to 2.28) 0.97 (0.66 to 1.40) 20 21 ≥ 7 instillations vs standard of care 1.46 (1 to 2.13) 0.90 (0.60 to 1.36) 22 23 Time to first bacteriologically confirmed HR (95% CI) Adjusted* HR (95% CI) 24 recurrence 25 26 ≥ 5 instillations vs standard of care 1.72 (1.12 to 2.65) 1.04 (0.64 to 1.71) 27 28 ≥ 6 instillations vs standard of care 1.55 (0.99 to 2.41) 0.96 (0.58 to 1.57) 29 ≥ 7 instillations vs standard of care 1.33 (0.79 to 2.22) 0.85 (0.49 to 1.47) 30 31 OR = odds ratio, IRR = incidence rate ratio, HR = hazard ratio 32 33 *Adjusted for age, BMI, employment and menopause status, postcoital http://bmjopen.bmj.com/ 34 infections, dyspareunia, FSFI and severity of RUTI 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 23 BMJ Open Page 24 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Flow diagram describing numbers of individuals at each stage of study 33 150x112mm (300 x 300 DPI) http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 29 BMJ Open
1 2 3
Supplementary Table 1 Health related quality of life assessment before BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from 4 5 treatment initiation and after 12 months for the HA + CS and standard 6 care group 7 8 Baseline 12 months Baseline 12 months P- 9 value 10 11 HA + CS (N = 79) Standard care (N = 26) 12 13 EQ-5D 3L 0.53 (0.36) 0.76 (0.25) 0.57 (0.31) 0.53 (0.40) 0.020 14 Mean (SD) 15 16 HA + CS (N = 83) Standard care (N = 27) 17 For peer review only 18 EQ-5D 59.53 (18.34) 75.88 (16.90) 62.5 (11.67) 66.84 (14.17) <0.001 19 VAS 20 21 Mean (SD) 22 23 HA + CS (N = 67) Standard care (N = 59) 24 SF-36 PCS 61.88 (10.14) 76.99 (11.69) 60.68 (10.02) 74.66 (13.32) 0.118 25 26 Mean (SD) 27 28 SF-36 MCS 61.74 (10.26) 77.35 (11.33) 60.68 (10.02) 74.66 (13.32) 0.105 29 Mean (SD) 30 31 MCS = Mental component score; PCS = Physical component score; SD = 32 33 standard deviation; VAS = visual analogue scale 34 35 36 37 38 http://bmjopen.bmj.com/ 39 40 41 42 43 44 45
46 on September 29, 2021 by guest. Protected copyright. 47 48 49 50 51 52 53 54 55 56 57 58 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 29
1 2 3
Supplementary Table 2 Resource consumption before treatment and BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from 4 5 after 12 months follow-up. Median (IQR) 6 7 Resources HA + CS Standard Care P Value* 8 (N=181) (N=95) 9 10 N. GP visits Before treatment 5 (3-6) 3 (3-4) 11 12 Follow-up 2 (1.5-3) 0 (0-1) 0.964 13 14 N. Specialist visits 6 (2-10) 6 (4-8) 15 16 For peer review2 (1-5) only4 (4-4) 0.644 17 N. Urine microscopic analysis 5 (3-7) 6 (6-8) 18
19 2 (0-3) 4 (1-4) 0.219 20 21 N. Urine culture 4.5 (3-7) 6 (5-8) 22 23 2 (1-4) 1 (0-3) <0.001 24 25 N. Urine cytology 2 (1-3) 3 (1-3)
26 1 (1-1) 0 (0-1) 0.003 27 28 N. Vaginal swab test 3 (1-3) 3 (1-3) 29 30 0 (0-0) 0 (0-0) 0.809 31 32 N. Transabdominal ultrasound 2 (1-2) 2 (1-2) 33 34 1 (0-1) 0 (0-1) 0.091 35 N. Transvaginal ultrasound 1 (1-2) 2 (1-2) 36 37 0 (0-0) 0 (0-0) 0.938 38 http://bmjopen.bmj.com/ 39 N. Uroflowmetry 1 (0-1) 0 (0-1) 40 41 0 (0-1) 0 (0-1) 0.933 42 43 N. Invasive urodynamic test 1 (1-1) 0 (0-1) 44 1 (1-1) 0 (0-0) 0.777 45 46 N. Cystoscopy with bladder byopsy 0 (0-1) 0 (0-0) on September 29, 2021 by guest. Protected copyright. 47 48 0 (0-0) 0 (0-0) 0.234 49 50 N. Cystoscopy 1 (1-3) 1 (0-1) 51 52 1 (0-1) 0 (0-1) 0.047 53 N. Antibiotic prescriptions - - 54 55 0 (0-1)^ 1 (0-1) 0.001 56 57 IQR = Interquartile range 58 59 * Wilcoxon-Mann-Withney test on the differences in the before treatment/ 60 follow-up resource consumption between the two groups ^ mean (SD), 0.99 (1.85) in HA + CS vs 1.31 (2.47) in Standard care
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 STROBE Statement—Checklist of items that should be included in reports of case-control studies 3 Item Page 4 No Recommendation 5 Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or 1 6 7 the abstract 8 (b) Provide in the abstract an informative and balanced summary of what 2 9 was done and what was found 10 11 Introduction 12 Background/rationale 2 Explain the scientific background and rationale for the investigation 4-5 13 being reported 14 Objectives 3 State specific objectives, including any prespecified hypotheses 5 15 For peer review only 16 Methods 17 Study design 4 Present key elements of study design early in the paper 5 18 Setting 5 Describe the setting, locations, and relevant dates, including periods of 5-6 19 recruitment, exposure, follow-up, and data collection 20 21 Participants 6 (a) Give the eligibility criteria, and the sources and methods of case 6 22 ascertainment and control selection. Give the rationale for the choice of 23 cases and controls 24 (b) For matched studies, give matching criteria and the number of NA 25 26 controls per case 27 Variables 7 Clearly define all outcomes, exposures, predictors, potential 6-7 28 confounders, and effect modifiers. Give diagnostic criteria, if applicable 29 Data sources/ 8* For each variable of interest, give sources of data and details of methods 6-7 30 31 measurement of assessment (measurement). Describe comparability of assessment 32 methods if there is more than one group 33 Bias 9 Describe any efforts to address potential sources of bias 7 http://bmjopen.bmj.com/ 34 Study size 10 Explain how the study size was arrived at 6 35 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If 7 36 37 applicable, describe which groupings were chosen and why 38 Statistical methods 12 (a) Describe all statistical methods, including those used to control for 7 39 confounding 40 (b) Describe any methods used to examine subgroups and interactions 7 41
c on September 29, 2021 by guest. Protected copyright. 42 ( ) Explain how missing data were addressed 7 43 (d) If applicable, explain how matching of cases and controls was NA 44 addressed 45 (e) Describe any sensitivity analyses 7 46 47 Results 48 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers 8 49 potentially eligible, examined for eligibility, confirmed eligible, included 50 in the study, completing follow-up, and analysed 51 52 (b) Give reasons for non-participation at each stage 8 53 (c) Consider use of a flow diagram Figure 1 54 Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, 8 55 social) and information on exposures and potential confounders 56 57 (b) Indicate number of participants with missing data for each variable of Table 1 58 interest 4 Suppl 59 Outcome data 15* Report numbers in each exposure category, or summary measures of 8 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml1 BMJ Open Page 28 of 29 BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 exposure 3 Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted Table 2 4 estimates and their precision (eg, 95% confidence interval). Make clear 5 6 which confounders were adjusted for and why they were included 7 (b) Report category boundaries when continuous variables were Tables 8 categorized 9 (c) If relevant, consider translating estimates of relative risk into absolute NA 10 risk for a meaningful time period 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml2 Page 29 of 29 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2015-009669 on 31 March 2016. Downloaded from
1 2 3 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity 9 4 analyses 5 6 7 Discussion 8 Key results 18 Summarise key results with reference to study objectives 9 9 Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or 10- 10 11 imprecision. Discuss both direction and magnitude of any potential bias 11 12 Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, 12 13 multiplicity of analyses, results from similar studies, and other relevant evidence 14 Generalisability 21 Discuss the generalisability (external validity) of the study results 10 15 For peer review only 16 Other information 17 Funding 22 Give the source of funding and the role of the funders for the present study and, if 8 18 applicable, for the original study on which the present article is based 19 20 21 *Give information separately for cases and controls. 22 23 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and 24 published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely 25 26 available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at 27 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is 28 available at http://www.strobe-statement.org. 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 29, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml3