Interventions for treating urinary incontinence in older women: a network meta-analysis

BACKGROUND: Urinary incontinence is highly prevalent among women 60 years and over, impacting their quality of life. The condition is often overlooked and untreated. Various treatments are available, but their benefits and harms in older women remain uncertain. OBJECTIVES: To compare the benefits and harms of conservative, pharmacological, and surgical treatments for urinary incontinence in terms of 'cure', 'cure or improvement', and serious adverse events (SAEs) in women 60 years and over using network meta-analyses (NMA), and to rank interventions within a single treatment network. SEARCH METHODS: We searched the Cochrane Incontinence Specialized Register, comprising trials from CENTRAL, MEDLINE, MEDLINE In-Process, MEDLINE Epub Ahead of Print, MEDLINE Daily, and two major international clinical trial registries, on 23 March 2025. We handsearched journals, conference proceedings, and reference lists of relevant articles. We placed no limitations on the searches. SELECTION CRITERIA: We included randomized controlled trials (RCTs) that examined the benefits and harms of conservative, pharmacological, and/or surgical treatments in women 60 years and over with urinary incontinence. Our primary outcomes were 'cure' and 'cure or improvement' of urinary incontinence symptoms. Secondary outcomes included the number of women with SAEs. DATA COLLECTION AND ANALYSIS: At least two review authors independently assessed trials for eligibility and risk of bias using Cochrane's risk of bias 2 (RoB 2) tool. A third author resolved any disagreements. We followed the guidance on undertaking NMA in Chapter 11 of the Cochrane Handbook for Systematic Reviews of Interventions. MAIN RESULTS: We included 43 RCTs involving 8506 participants, a mean of 198 per study (range 14 to 1438). Conservative treatments predominated (20/43, 46.5%) in the studies, followed by pharmacological (17/43, 39.5%), surgical (4/43, 9.3%), and mixed (2/43, 4.7%) treatments. The RCTs had variable risks of bias, often presenting 'some concerns' or 'high risk,' with poor reporting on randomization, blinding, and protocol details. Conservative and pharmacological treatments were often at a high risk of bias for all outcomes (cure, cure or improvement, and SAEs). For the 'cure' outcome, we excluded three studies to address network disconnections; hence, comparisons focused on conservative and pharmacological treatments. Results indicated that all treatments might be better than control, with physical therapies - mainly pelvic floor muscle training with or without complementary therapies or education - showing the best performance for 'cure': physical therapies combined with complementary therapies (odds ratio (OR) 17.79, 95% confidence interval (CI) 2.97 to 106.46; 1 study, 71 participants), physical therapies (OR 7.20, 95% CI 2.59 to 20.03; 4 studies, 310 participants), and physical therapies with education (OR 3.25, 95% CI 1.19 to 8.84; 4 studies, 364 participants), with the evidence for all three results being of very low certainty, followed by complementary therapies (OR 4.65, 95% CI 0.74 to 29.37; 1 study, 37 participants; very low-certainty evidence) and education (OR 2.68, 95% CI 0.61 to 11.73; 2 studies, 180 participants; low-certainty evidence). The mean ranks for best treatment, P scores, and surface under the cumulative ranking curve (SUCRA) values demonstrate the superiority of physical therapies, suggesting that the addition of complementary therapies may be the optimal treatment for 'cure' (SUCRA value ranged from 57% to 85% across the three interventions that included physical therapies). However, due to imprecision in effect estimates and sparse data, uncertainty regarding optimal treatment remains (low- to very low-certainty evidence). For 'cure or improvement of urinary incontinence', after adjusting for disconnected networks by excluding three studies, the analysis showed that physical therapies, with or without education, performed best compared with their controls, with very low-certainty evidence for all the following results (physical therapies: OR 3.98, 95% CI 2.02 to 7.82; 3 studies, 197 participants; physical therapies combined with education: OR 3.20, 95% CI 1.45 to 7.02; 3 studies, 236 participants; β3-adrenergic agonists: OR 2.44, 95% CI 1.28 to 4.62; 1 study, 360 participants) followed by education (OR 2.09, 95% CI 1.05 to 4.17; 2 studies, 213 participants) and antimuscarinic drugs (OR 1.90, 95% CI 1.19 to 3.03; 2 studies, 1469 participants). Both physical therapies, with or without an educational intervention, and β3-adrenergic agonists performed the best compared with their controls (physical therapies: SUCRA = 90%; physical therapies combined with education: SUCRA = 77%; β3-adrenergic agonists: SUCRA = 63%). However, the evidence was of very low certainty, suggesting the need for more trials. Notably, there were no SAEs in conservative treatments, and pharmacological treatments reported some. However, no treatment showed significantly less chance of SAEs, with very low-certainty evidence for all the following results (serotonin-noradrenaline uptake inhibitors: OR 0.40, 95% CI 0.10 to 1.59; 1 study, 264 participants; β3-adrenergic agonists: OR 0.61, 95% CI 0.04 to 10.19; 1 study, 404 participants; complementary therapies: OR 0.53, 95% CI 0.00 to 71.04; no direct evidence, 18 participants; antimuscarinic drugs: OR 0.81, 95% CI 0.46 to 1.42; 4 studies, 2731 participants; physical therapies combined with education: OR 0.99, 95% CI 0.10 to 9.80; 3 studies, 130 participants). AUTHORS' CONCLUSIONS: Due to the limited number of trials and generally small sample sizes, the precision of estimates regarding treatment benefits and harms was low. We successfully conducted an NMA, but there was insufficient evidence to support a robust overall analysis. To establish a connected treatment network, we excluded surgical intervention studies. As a result, the analysis focused on conservative and pharmacological treatment comparisons. For the outcome of 'cure', evidence rated as very low to low certainty suggests that physical therapies combined with complementary therapies may be the most effective option, followed by physical therapies alone or combined with education. For 'cure or improvement', both physical therapies (with or without education) and β3-adrenergic agonists showed potential benefits. SAEs were absent for conservative interventions, whereas most of the pharmacological treatment studies reported some. However, there was insufficient evidence to determine whether any treatment reduced the likelihood of SAEs. Overall, the number and quality of studies were insufficient to draw firm conclusions about the most effective treatment for urinary incontinence in older women. To strengthen the evidence base, larger, high-quality trials with clearly defined interventions and consistently reported outcomes are needed.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.