This is the first study to test if a physiotherapist-supervised individual telehealth‑delivered pelvic floor muscle training (PFMT) program incorporating a home-based intra-vaginal biofeedback device reduces urinary incontinence (UI) more than usual care after gynaecological cancer, following the intervention and at 52 weeks. Our primary aim was to investigate if a 16-week PFMT program with biofeedback is effective in this population with UI.

This was a pragmatic, two‑arm, stratified superiority randomised controlled trial. Recruitment sites included gynaecology‑oncology outpatient clinics and community foundations/social media/care groups. Participants had completed primary cancer treatment ≥6 months prior or adjuvant therapy ≥3 months prior, for Stage I, II or III uterine, cervical, fallopian tube, primary peritoneal or ovarian cancer or borderline ovarian tumour, and experienced UI at least weekly. Participants randomised to usual care received bladder/bowel advice handouts and one audio-telehealth consultation with a researcher to answer queries about the handouts. Participants randomised to the intervention additionally received eight video-telehealth physiotherapist consultations of PFMT with a biofeedback device (Femfit®), delivered individually, alongside a home‑based program over 16 weeks, and were encouraged to continue a self-directed PFMT maintenance program. The primary outcome was change in participant‑reported UI, measured using the International Consultation on Incontinence Questionnaire-Urinary Incontinence Short Form (ICIQ-UI SF) [1], from baseline to immediately post‑intervention (primary endpoint) and 52 weeks (secondary). Secondary outcomes are listed in Table 1. Adherence and change in intra‑vaginal squeeze pressure from the biofeedback device were also investigated in the intervention group.
72 women (36 per group) were required to detect a minimum difference of 2.5-units in mean change in UI (2) between groups with 80% power and a 5% two-tailed significance level, assuming a standard deviation (SD) of 3.2 units, pre-post correlation of 0.4, intra-cluster correlation of 0.05 with 3 physiotherapists treating approximately 10 patients each, and 15% attrition. The primary outcome was analysed using a constrained longitudinal data analysis model (3), including the stratification variable (radiotherapy) and random effects for physiotherapist (intervention arm only). Missing data were handled using multiple imputation.

Overall, 78 women (mean age 59 [SD=11] years) recruited between July 2022 and September 2024 were randomised (n=40 PFMT, n=38 usual care) (Figure 1). Uterine cancer (54%) and stage 1 diagnosis (54%) were most common among participants. 35% of patients received post-operative radiotherapy.
A clinically meaningful difference in mean change in ICIQ-UI SF was observed between groups (difference: -3.2, 95% confidence interval: -4.5, -1.8), indicating greater improvement for PFMT participants. However, benefits were not sustained at 52 weeks. Between-group differences favoured PFMT immediately post-intervention for most secondary outcomes; these persisted at 52 weeks for quality‑of‑life, bother of pelvic floor symptoms, leakage episodes and global impression of change (Table 1). Self-reported PFMT was higher in the intervention group immediately post-intervention (74% vs 31%) but rates were similar between groups at 52 weeks (33% vs 28%). In the intervention group, adherence to the program was good during the physiotherapy-supervised phase. Pelvic floor muscle strength and endurance increased over 100%. Three participants reported transient discomfort with insertion/use of the intra-vaginal device.

The 16-week PFMT telehealth program incorporating intra-vaginal biofeedback reduced UI in women by a clinically relevant magnitude following treatment for gynaecological cancer immediately post-intervention, although this benefit was not sustained at 52 weeks. Several secondary outcomes improved, including some at 52 weeks. However, once supervision of the PFMT program ceased, self-reported participant exercise levels declined. This may explain why improvements in the ICIQ-UI SF observed immediately post-intervention were not sustained at 52 weeks.

This model of service delivery can be recommended in clinical practice. Periodic physiotherapy contact may be required post-supervised PFMT to support patient continuation of exercise and optimise results.

Avery, K., J. Donovan, T. J. Peters, C. Shaw, M. Gotoh and P. Abrams (2004). "ICIQ: A brief and robust measure for evaluating the symptoms and impact of urinary incontinence." Neurourology and Urodynamics 23(4): 322–330.
Nystrom, E., M. Sjostrom, H. Stenlund and E. Samuelsson (2015). "ICIQ symptom and quality of life instruments measure clinically relevant improvements in women with stress urinary incontinence." Neurourology and Urodynamics 34(8): 747–751.
Liang K-Y, Zeger SL. (2000) "Longitudinal data analysis of continuous and discrete responses for pre-post designs". Sankhya: Indian J Stat Ser B (1960–2002), 62(1):134–48.