More than one in five women experience the embarrassment of urine leakage while exercising, and this is a substantial barrier to exercise [1]. As many as one in three women with urinary incontinence report that they limit their physical activity due to their incontinence; this includes not exercising at all (11.6%), exercising less (11.3%) or changing the type of exercise they perform (12.4%) [2]. Urine leakage during exercise is experienced predominantly as stress urinary incontinence (SUI), which is characterized by urine leakage which occurs when bladder pressure exceeds urethral closure pressure in the absence of detrusor contraction. As with SUI in general, neuromuscular damage and structural damage to the urethra and the levator ani muscles (PFMs) as well as their associated connective tissues are likely implicated, yet repetitive loading of the pelvic floor may, over a bout of exercise, lead to muscle fatigue or tissue strain as contributing factors. The main goal of this study is to understand the mechanistic factors associated with SUI when it is experienced predominantly during exercise. More specifically, we aim to investigate differences in and compare the effects of an acute bout of running on pelvic floor morphology and function between female runners who do and do not regularly experience urinary incontinence while they run.

This paper reports a preliminary analysis of data from an ongoing cross-sectional, observational case-control study which received approval from the local institutional research ethics board. Women aged 18 years and over with no known risk factors related to physical activity (PAR-Q+), who run at least 5 km in under 50 minutes, twice per week and more than 10km a week, and who have done so for at least one year, are being recruited into two cohorts: females who regularly (≥ 1 per month) experience urine leakage while exercising, and those who do not. An a priori power calculation estimated that a minimum of 26 women per group would be necessary to detect within-group differences. Women are being recruited from local running groups and physiotherapy clinics, and are excluded if they have a history of urogenital surgery, symptoms of the female athlete triad, dyspareunia, a neurologic disorder, are pregnant or have delivered a baby within the previous year. All eligible females provide demographic information and are asked to complete the International Consultation on Incontinence - Female Lower Urinary Tract Symptoms (ICIQ-FLUTS) and the International Physical Activity Questionnaire (IPAQ). 

All participants undergo an ultrasound imaging evaluation of their pelvis. 3D volume images of the levator ani are acquired using a transperineal probe while participants remain in quiet standing. Next, participants are instrumented with an intravaginal dynamometer in standing and are asked to perform three repetitions of a maximal voluntary contraction (MVC) using a dynamometric anteroposterior diameter of 35mm. In supine, passive forces are recorded while the anteroposterior diameter of the dynamometer arms moved from 15mm to 40mm at a constant speed of 50mm/s. Pelvic floor muscle elongation is held for 7s at the maximum (40mm) diameter before the arms return to their initial position. Three repetitions of passive tissue elongation are performed. Participants then run on a treadmill for 37 minutes using a standardized protocol: 2 minutes at 7km/h, 2 minutes at 10 km/h, 2 minutes at 15 km/h, 30 minutes at a self-selected pace, followed by 30 seconds at 10 then 7 km/h.  Following the run, the ultrasound imaging and dynamometer protocols are repeated. Outcomes include levator hiatal area at rest, and maximum relative peak forces and rate of force development during MVC and during tissue elongation. Stress relaxation is measured using the force data acquired by the dynamometer arms once they are opened to 40mm and held for 7s, quantified using the coefficient of the exponential decay function. Based on data acquired to date, all outcomes were tested for normality using the Kolmogorov-Smirnov test. A mixed-model ANOVA was used to compare dynamometry and ultrasound outcomes between group (continence status) and across time (pre-run vs post-run). Effect sizes (Cohen’s f) were calculated using partial eta-squared values, and statistical power was determined for each variable.

To date, 23 experienced female runners have participated out of the targeted n=52. Demographic information is presented in Table 1. The runners with and without leakage do not differ on any variables except the ICIQ-FLUTS score, which is significantly higher in the runners with running-induced SUI. 

All outcomes are normally distributed. Table 2 describes the results for all outcomes. No significant two-way interactions between factors (continence status) and within-participants (pre-run vs post-run) were found for any outcome (p >0.05). To date, no difference in force generating capacity of the PFMs is evident between participants with and without running-induced SUI. While the rate of force development during MVC does not appear to change after the acute bout of running, a significant small increase in the relative peak force achieved during MVC is observed after the run compared to before the run in both groups. To date, women with running-induced SUI demonstrate a slower decay in passive force during the stress relaxation response than those without SUI. After the run, there is a significant difference in relative peak forces and rate of force development measured during tissue elongation—forces were lower after the run for both groups. The effect sizes computed for the interaction between continence status and time suggest that there may be moderate effect sizes for changes in relative peak force measured during tissue elongation and for changes in levator hiatal area at rest (effect size=0.35 and 0.26 respectively), while the effects sizes may be small for the other outcomes (<0.13). There appears to be a change in the shape of the levator hiatus after the run, and we are currently developing a method to quantify this change for further investigation.

Although this study is currently underpowered, this interim power analysis suggests that this study will be adequately powered to detect differences in most outcomes in terms of the morphological and mechanical changes observed after an acute bout of running (power ranging from 82 to 93% for 4 of 6 outcomes). Based on these preliminary findings, PFM strength and fatigue appear to play a limited role, if any, in running-induced SUI in experienced female runners. Although no significant interaction effect between continence status and time (before/after the run) is evident, the moderate effect size for the interaction between relative peak force at 40mm dynamometer diameter and time suggests that changes in the mechanical properties of the supportive connective tissues may be associated with running-induced SUI in females.

These preliminary results do not indicate significant differences in PFM strength between female runners with and without running-induced SUI, nor any PFM fatigue induced by an acute bout of running. The effect sizes suggest that differences between runners with and without SUI may exist in the passive properties of PFMs and surrounding connective tissues. These differences might have implications in terms of the support offered to the pelvic organs and urethra in response to repetitive loading during running. These are preliminary results only, recruitment is ongoing.

S. Almousa and A. Bandin van Loon, “The prevalence of urinary incontinence in nulliparous adolescent and middle-aged women and the associated risk factors: A systematic review.,” Maturitas, vol. 107, pp. 78–83, 2018.
I. Nygaard, T. Girts, N. H. Fultz, K. Kinchen, G. Pohl, and B. Sternfeld, “Is urinary incontinence a barrier to exercise in women?,” Obstet. Gynecol., vol. 106, no. 2, pp. 307–314, 2005.