More than 30% of competitive rhythmic gymnasts experience urinary incontinence (UI), of which 70% report that UI negatively influence sports performance (1). Strength training of the pelvic floor muscles (PFM) has 1A level of evidence/recommendation to treat UI in the general female population; however, there is little knowledge of this among young female athletes and no randomized controlled trials have assessed the effects in rhythmic gymnasts (2). The aim of the present study was to assess the effect of pelvic floor muscle training (PFMT) compared with no intervention on bother and prevalence of UI among competitive rhythmic gymnasts.

This was an assessor-blinded cluster-randomized controlled trial with concealed allocation. Twenty-three rhythmic gymnastics clubs were randomized to an intervention group (12 clubs, 119 gymnasts) and a control group (11 clubs, 86 gymnasts). Included gymnasts (a mixed group with and without UI) had to be ≥12 years of age and training ≥3 days per week. All gymnasts, or parents of gymnasts younger than 16 years of age, gave written consent to participate. 
Before commencing the PFMT program, the intervention group gymnasts had an individual session with a physiotherapist where a portable 2D ultrasound machine (GE Healthcare –Logiq e 17 R7, GE>12L-RS – 5-13 MHz Wideband Linear Probe) was used to teach and assess the ability to perform a correct PFM contraction (probe placed suprapubically). The intervention group performed 1 set of 8–12 repetitions of close to maximum PFM contractions during each training/warm-up for 8 months (November 2022-June 2023). The degree of coach supervised PFMT varied between the clubs in the intervention group. The control group continued rhythmic gymnastics training as normal without PFMT during training/warm-up. Adherence to the training program was reported at a monthly basis through an online survey sent to the gymnasts’ mobile phones.
The International Consultation on Incontinence Questionnaire - Urinary Incontinence Short Form (ICIQ-UI-SF) was used to measure the primary outcome, bother of UI (change in the ICIQ-UI-SF total score), and the secondary outcome, prevalence of UI. Gymnasts in both groups answered the ICIQ-UI-SF at baseline and after the intervention period (October 2022 and June 2023, respectively). A statistician conducted an a-priori power calculation for two outcomes: change in the ICIQ-UI-SF total score (primary outcome here) and prevalence of overuse injuries (not part of this study). With 80% power, 5% significance level and a decrease in the primary outcome measure from 4.7 to 2.5 (SD 2.1) in the intervention group and no change in the control group, at least 30 gymnasts were required. To account for randomization at the cluster level, the sample size was multiplied by a factor of 1.4 (assuming ICC=0.05 and 9 gymnasts per club), resulting in a total sample size of 42 (21 in each group). The sample size for the outcome prevalence of overuse injuries was much greater (n=220) and this was thereby the trial's target sample size. The ICIQ-UI-SF total score after the intervention period was analyzed with linear regression, with adjustment for the baseline ICIQ-UI-SF total score (ANCOVA). The standard errors were adjusted for the clustering of gymnasts in clubs. Prevalence of UI was analyzed with mixed effect logistic regression, adjusted for the baseline prevalence, with a random intercept at the club level.

One hundred and seventy-nine (88%) of the rhythmic gymnasts completed the study (mean age 13.8 years (SD 2.0), mean BMI 19.0 (SD 2.8), mean weekly training load: 15.2 hours (SD 6.2)). Among the gymnasts in the intervention group, 68% adhered to the prescribed training protocol. The between-group difference (intervention – control) in the total score of the ICIQ-UI-SF was -0.07 (95% CI -0.96 to 0.82), p = 0.88. The prevalence of UI in the intervention group changed from 46% at baseline to 41% after the intervention period, while the prevalence of UI in the control group changed from 32% at baseline to 34% after the intervention period; odds ratio for intervention vs control = 1.1 (95% CI 0.45 to 2.7); p = 0.83. No adverse effects were reported.

There were no differences in bother of UI and prevalence of UI after eight months of targeted PFMT compared with no targeted training in competitive rhythmic gymnasts. The high weekly training load may require a higher dose of PFMT for effect. In addition, more targeted individual supervision during PFMT may be important to achieve significant effect in gymnasts, a group of female athletes regularly exposed to heavy load on the pelvic floor. PFMT did not seem to cause any adverse effects for the rhythmic gymnasts, thus it might be safe to implement PFMT after all, because symptoms of UI at a young age is a known risk factor of developing UI later in life, e.g. during pregnancy and after childbirth (3). The lack of effect at a group level might also stress the importance of individual treatment of gymnasts diagnosed with UI.

The results suggest that 1 set of 8–12 repetitions of close to maximum PFM contractions during each training/warm-up is not enough to reduce the bother or prevalence of UI among competitive rhythmic gymnasts at a group level. Individually adapted measures might be of more importance to competitive rhythmic gymnasts, but more research is needed to conclude.

Gram, M.C.D. and K. Bø, High level rhythmic gymnasts and urinary incontinence: Prevalence, risk factors, and influence on performance. Scandinavian journal of medicine & science in sports, 2020. 30(1): p. 159-165
Dumoulin, C., L.P. Cacciari, and E.J.C. Hay-Smith, Pelvic floor muscle training versus no treatment, or inactive control treatments, for urinary incontinence in women. Cochrane database of systematic reviews, 2018(10).
Bø, K. and J. Sundgot-Borgen, Are former female elite athletes more likely to experience urinary incontinence later in life than non-athletes? Scandinavian journal of medicine & science in sports, 2010. 20(1): p. 100-104.

Continence 12S (2024) 101359
DOI: 10.1016/j.cont.2024.101359