Systematic voiding programme in adults with urinary incontinence following acute stroke : the ICONS-II RCT

Watkins, Caroline and Tishkovskaya, Svetlana and Brown, Chris and Sutton, Chris and Garcia, Yvonne Sylvestre and Forshaw, Denise and Prescott, Gordon and Thomas, Lois and Roffe, Christine and Booth, Joanne and Bennett, Kina and Roe, Brenda and Hollingsworth, Bruce and Mateus, Ceu and Britt, David and Panton, Cliff (2022) Systematic voiding programme in adults with urinary incontinence following acute stroke : the ICONS-II RCT. Health Technology Assessment, 26 (31). pp. 1-88. ISSN 1366-5278

Abstract

Background: Urinary incontinence affects around half of stroke survivors in the acute phase, and it often presents as a new problem after stroke or, if pre-existing, worsens significantly, adding to the disability and helplessness caused by neurological deficits. New management programmes after stroke are needed to address urinary incontinence early and effectively. Objective: The Identifying Continence OptioNs after Stroke (ICONS)-II trial aimed to evaluate the clinical effectiveness and cost-effectiveness of a systematic voiding programme for urinary incontinence after stroke in hospital. Design: This was a pragmatic, multicentre, individual-patient-randomised (1: 1), parallel-group trial with an internal pilot. Setting: Eighteen NHS stroke services with stroke units took part. Participants: Participants were adult men and women with acute stroke and urinary incontinence, including those with cognitive impairment. Intervention: Participants were randomised to the intervention, a systematic voiding programme, or to usual care. The systematic voiding programme comprised assessment, behavioural interventions (bladder training or prompted voiding) and review. The assessment included evaluation of the need for and possible removal of an indwelling urinary catheter. The intervention began within 24 hours of recruitment and continued until discharge from the stroke unit. Main outcome measures: The primary outcome measure was severity of urinary incontinence (measured using the International Consultation on Incontinence Questionnaire) at 3 months post randomisation. Secondary outcome measures were taken at 3 and 6 months after randomisation and on discharge from the stroke unit. They included severity of urinary incontinence (at discharge and at 6 months), urinary symptoms, number of urinary tract infections, number of days indwelling urinary catheter was in situ, functional independence, quality of life, falls, mortality rate and costs. The trial statistician remained blinded until clinical effectiveness analysis was complete. Results: The planned sample size was 1024 participants, with 512 allocated to each of the intervention and the usual-care groups. The internal pilot did not meet the target for recruitment and was extended to March 2020, with changes made to address low recruitment. The trial was paused in March 2020 because of COVID-19, and was later stopped, at which point 157 participants had been randomised (intervention, n = 79; usual care, n = 78). There were major issues with attrition, with 45% of the primary outcome data missing: 56% of the intervention group data and 35% of the usual-care group data. In terms of the primary outcome, patients allocated to the intervention group had a lower score for severity of urinary incontinence (higher scores indicate greater severity in urinary incontinence) than those allocated to the usual-care group, with means (standard deviations) of 8.1 (7.4) and 9.1 (7.8), respectively. Limitations: The trial was unable to recruit sufficient participants and had very high attrition, which resulted in seriously underpowered results. Conclusions: The internal pilot did not meet its target for recruitment and, despite recruitment subsequently being more promising, it was concluded that the trial was not feasible owing to the combined problems of poor recruitment, poor retention and COVID-19. The intervention group had a slightly lower score for severity of urinary incontinence at 3 months post randomisation, but this result should be interpreted with caution.