Both runners with and without RI-SUI experienced significant transient changes, particularly in the morphology of their pelvic floor as measured by USI after they completed a 37-min treadmill running protocol. The changes induced by running appear to predominantly affect the passive support system, as evidenced by the larger area and AP diameter of the levator hiatus, the more caudal position of the BN, and the elongated LP observed in quiet standing after running compared with baseline. Although PFM function, as measured by intravaginal dynamometry, was not significantly affected by the run, the directionality of the observed effects was consistent with the morphological changes. Specifically, the relative peak force and stiffness measured during passive tissue elongation tended to be lower after the run than before the run. And, although differences were not statistically significant, the runners who experienced RI-SUI tended to generate less power after running than before the run, whereas their strength (relative peak force) did not change. Conversely, the runners without RI-SUI tended to generate greater force and more power after the run than before the run. These findings suggest that an acute bout of running might result in strain of the passive tissues within the pelvis, which was observed in all female runners, regardless of whether they experience urine leakage during running. Such changes are likely related to transient changes in the viscoelastic characteristics of the tissues after an episode of repetitive loading—the tissues likely return to their prestressed state assuming that there is no underlying damage; however, the cumulative effects of repeated running bouts should be investigated.

Running caused significant transient morphological changes in pelvic morphology, as measured by 2D and 3D USI, mainly reflecting elongation of the passive supporting structures. Indeed, the passive support structures appeared to be equally affected by the activity in both runners with and those without RI-SUI. Based on these findings, it is plausible that the initial morphological features of the levator hiatus, BN, and LP may serve as a predisposing factor for the development of RI-SUI over the course of a run. It is possible that some threshold exists for the levator hiatus area, LP length, BN height, or a combination of these factors beyond which the passive support to the urethra is no longer sufficient to support and compress it between the anterior vaginal wall and the pubic symphysis to prevent urine leakage. Consistent with this theory, a recent study investigating the likelihood of SUI symptoms being cured by a PFM training intervention found that a resting BN position lower than 14.3 mm in standing, as measured on 2D USI, was predictive of treatment failure [28]. In the current study, continent runners had a resting BN height that was above this threshold both before and after the run (17.98 mm before vs 14.66 mm after); however, the BN of runners with RI-SUI dropped below this threshold after the run (16.22 mm before vs 13.92 mm after). Although the BN may descend over the course of a run in all female runners, leakage may occur when the BN position ultimately reaches some failure point [28]. This theory is consistent with findings that athletes most often leak urine toward the middle and end of a training session [10].

The significant increase in the extent of shortening of the LP observed during the MVC after the run, and the significant decrease in the extent of lengthening of the LP during the MVM after the run are likely due to a transient increase in the length of the LP. With a longer LP length at rest, the LP had to shorten more during the MVC to achieve the same end-MVC length observed before the run, and similarly the LP did not have as much room to lengthen during MVM before it reached its ultimate endpoint. As with the LP, the significant reduction in both groups in the extent of BN descent observed during the MVM after the run can be attributed to the lower resting position of the BN after the run, which was already close to its maximal descent.

Although not significant, the trends observed on dynamometry are worth discussing, as the effect sizes were moderate to large; yet these results should be interpreted with caution. Runners with RI-SUI demonstrated a 9% reduction in the contractile power of their levator ani muscles after running, whereas their strength remained unchanged. In contrast, the runners without RI-SUI tended to have greater strength (14% increase) and power (8% increase) after the run than before the run. These results align somewhat with previous findings among nulliparous females with mild SUI, where a mean reduction in maximal vaginal closure pressure induced by MVC of the levator ani muscles was substantive (-20%) but not significant following a 90-min exercise session that included 20 min of running [29]. Also, in line with the current findings, there were changes neither in vaginal resting pressure nor in the time over which an MVC could be sustained (i.e., static endurance) following the exercise session [29]. The current results are also consistent with findings that, following a 20-min as many repetitions as possible (AMRAP) workout, strenuous exercisers (27.7% with symptoms of urine leakage) demonstrated significantly more vaginal descent and significantly lower vaginal resting pressure [30], without concurrent changes in maximal vaginal closure pressure during MVC after the exercise [30]. A control group of nonstrenuous exercisers (8.57% with symptoms of urine leakage) completed a 20-min walk, and no differences were observed between the strenuous exercisers and the control group in terms of effect on pelvic organ support and vaginal closure pressure despite the different intensity of the exercise [30]. Both the current study and these previous results reinforce the idea that an acute bout of exercise may have a greater impact on passive pelvic organ support rather than PFM contractile function.

Although not significant, stiffness and peak passive force measured through dynamometry tended to be compromised in the runners with RI-SUI relative to their continent counterparts, which align with the observed morphological changes in all runners. Although both groups demonstrated similar reductions in stiffness and peak passive forces after completing the running protocol, the group with RI-SUI had lower absolute stiffness and lower peak passive forces after the running protocol. Further, the runners with RI-SUI whose leakage was provoked during the running protocol appeared to experience greater strain than those with RI-SUI whose leakage was not provoked. More research with a larger sample size is needed to establish the contribution of the loss of passive tissue stiffness in RI-SUI.

This study focused on experienced female runners and thus the findings may not be generalizable to infrequent runners. The sample included middle-aged nulliparous and parous, mostly white, healthy females. Although the outcomes of this investigation might be transferable to other athletes who engage in running as a component of other physical activities, the results may not be applicable to all athletes. Whereas the runners without RI-SUI selected a faster treadmill speed than their incontinent counterparts (10.8 km/h versus 9.7 km/h), step counts during the running protocol were not different between the groups, suggesting that, even though loading may have been higher in the continent group, the pelvic tissues were exposed to a similar number of loading cycles in both groups with comparable levels of exertion (15–16 on the Borg Scale). Indeed, had the runners with RI-SUI run at the same intensity as their continent counterparts, the observed changes may have been greater.

Vaginal birth is a known risk factor for SUI, and it impacts the morphology and functioning of the PFMs. Although efforts were made to balance parity as a dichotomous variable between the recruited groups, there was a higher proportion of female runners with multiple childbirths in the RI-SUI category in contrast to those without. Nonetheless, an equal number of participants in both groups (n = 4) exhibited levator avulsion (partial or complete). Additional alterations in the pelvic floor linked to parity such as nerve damage might have played a role in affecting the findings of this investigation.

Neural, metabolic, and vascular components of the continence system could impact differences in the force generation capacity of the PFMs after the run; however, these were not assessed in the present study. Indeed, although PFM surface electromyography (EMG) was pilot tested as an outcome to be measured concurrently with intravaginal dynamometry in this study, it was deemed not to be feasible, as the arms of the dynamometer caused significant motion artifact in the EMG signals, intravaginal electrodes were dislodged by the dynamometer arms, and adhesive electrodes lifted, particularly after the run, owing to perspiration, urine, and/or vaginal secretions.

Not all runners with RI-SUI leaked urine during the protocol. A longer or more strenuous protocol may have induced greater changes in morphology and function than was observed here. Although the sample size was small, this hypothesis is supported by the exploratory post-hoc analysis presented in Table 4. Among the runners with RI-SUI, those whose leakage was provoked during the running protocol, findings point to greater reductions in pelvic organ support compared with those with RI-SUI who did not experience leakage during the protocol. Although participants were evaluated quickly after the running protocol (within 15 min), it is possible that the passive PFM tissue properties may have recovered somewhat by the time the passive tissue elongation was performed. As such, the extent of tissue strain induced by the running protocol may be larger than that reported here.

A final limitation of this study is the extent of missing data that resulted from hardware failure in the intravaginal dynamometer and poor image quality during USI, which resulted in a smaller sample size for some outcomes. The interpretation of the results is also subject to the limitations of the instruments themselves. For instance, although intravaginal dynamometry is considered a gold standard for the measurement of PFM active and passive tissue properties, it remains a coarse measurement as it is influenced by all closure forces acting on the dynamometer arms [31, 32]. USI measurements can be influenced by changes in the angle of the transducer during acquisition and the incorrect identification of landmarks [31].

That said, the innovative approach used in this study to investigate the effects of an acute bout of running on active and passive pelvic floor support structures contributes to our understanding of RI-SUI. The findings point to practical approaches that focus on enhancing pelvic floor support such as PFM training and the use of pessaries to effectively mitigate or prevent symptoms of athletic incontinence in females. If pelvic floor support is enhanced prior to an acute bout of running, it might prevent the pelvic structures from descending below a critical threshold that leads to the failure of the continence system over the course of a run.