BACKGROUND: OBJECTIVE: DESIGN: Multicentre, pragmatic, open-label, randomised, non-inferiority trial of 12 months' treatment with the allocated intervention, including an early, embedded qualitative study and a 6-month post-treatment observation phase. The predefined non-inferiority margin was one urinary tract infection per person-year. SETTING: Eight UK NHS secondary care sites. PARTICIPANTS: A total of 240 adult women with recurrent urinary tract infection requiring preventative treatment participated in the trial. INTERVENTIONS: MAIN OUTCOME MEASURES: The primary clinical outcome was incidence of symptomatic antibiotic-treated urinary tract infection during the 12-month treatment period. Cost-effectiveness was assessed by incremental cost per quality-adjusted life-year gained, extrapolated over the patient's expected lifetime using a Markov cohort model. Secondary outcomes included post-treatment urinary tract infections, total antibiotic use, microbiologically proven urinary tract infections, antimicrobial resistance, bacteriuria, hospitalisations and treatment satisfaction. RESULTS: Primary modified intention-to-treat analysis comprised 205 (85%) randomised participants [102/120 (85%) participants in the antibiotics arm and 103/120 (86%) participants in the methenamine hippurate arm] with at least 6 months' data available. During treatment, the incidence rate of symptomatic, antibiotic-treated urinary tract infections decreased substantially in both arms to 1.38 episodes per person-year (95% confidence interval 1.05 to 1.72 episodes per person-year) for methenamine hippurate and 0.89 episodes per person year (95% confidence interval 0.65 to 1.12 episodes per person-year) for antibiotics (absolute difference 0.49; 90% confidence interval 0.15 to 0.84). This absolute difference did not exceed the predefined, strict, non-inferiority limit of one urinary tract infection per person-year. On average, methenamine hippurate was less costly and more effective than antibiotics in terms of quality-adjusted life-years gained; however, this finding was not consistent over the longer term. The urinary tract infection incidence rate 6 months after treatment completion was 1.72 episodes per year in the methenamine hippurate arm and 1.19 in the antibiotics arm. During treatment, 52% of urine samples taken during symptomatic urinary tract infections were microbiologically confirmed and higher proportions of participants taking daily antibiotics (46/64; 72%) demonstrated antibiotic resistance in Escherichia coli cultured from perineal swabs than participants in the methenamine hippurate arm (39/70; 56%) (p-value = 0.05). Urine cultures revealed that during treatment higher proportions of participants and samples from the antibiotic arm grew E. coli resistant to trimethoprim/ co-trimoxazole and cephalosporins, respectively. Conversely, post treatment, higher proportions of participants in the methenamine hippurate arm (9/45; 20%) demonstrated multidrug resistance in E. coli isolated from perineal swabs than participants in the antibiotic arm (2/39; 5%) (p = 0.06). All other secondary outcomes and adverse events were similar in both arms. LIMITATIONS: This trial could not define whether or not one particular antibiotic was more beneficial, and progressive data loss hampered economic evaluation. CONCLUSIONS: This large, randomised, pragmatic trial in a routine NHS setting has clearly shown that methenamine hippurate is not inferior to current standard care (daily low-dose antibiotics) in preventing recurrent urinary tract infections in women. The results suggest that antimicrobial resistance is proportionally higher in women taking prophylactic antibiotics. RECOMMENDATIONS FOR RESEARCH: Future research should include evaluation of other non- antibiotic preventative treatments in well-defined homogeneous patient groups, preferably with the comparator of daily antibiotics. TRIAL REGISTRATION: This trial is registered as ISRCTN70219762 and EudraCT 2015-003487-36. FUNDING: This project was funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 26, No. 23. See the NIHR Journals Library website for further project information.
|