It has been suggested that excessive or repetitive increases in abdominal pressure associated with exercise, specifically high impact exercise (HIE), may promote the development of pelvic floor dysfunction (1). Research assessing intra-abdominal pressure (IAP) generation during different exercises report wide variations amongst individuals and there remains no established threshold of maximal IAP to guide activity restrictions (2). The contribution of the pelvic floor muscles (PFM)s to a women’s ability to tolerate increases in IAP during exercise also needs to be considered. Assessment of the PFMs using traditional perineometry tools is limited by their inability to differentiate IAP from that produced by the PFMs(3); rendering it invalid as a tool for measuring the PFMs during dynamic activities. The femfit® is a device capable of measuring a pressure profile along the length of the vagina. The device is thin, soft and flexible with eight pressure sensing areas, and is able to be used during dynamic and static exercise. Simultaneous measurements are made of IAP (sensor 8) and within the PFM zone covered by sensors numbered 1 to 6. The aim of this study is to assess the ability of the vaginal pressure profile recorded by the femfit® to describe the PFMs, during distinct tasks, by comparing it with simultaneously measured changes in the pelvic floor anatomy using transperineal ultrasound (TPUS). The hypothesis is that the pressure profile could be used to describe the changes in the geometry of the PFMs during dynamic activity.

This was a prospective observational study. Nulliparous volunteers who had no, or minimal non-bothersome pelvic floor dysfunction were recruited from university campuses, or by advertisement. Exclusion criteria included women younger than 18 years of age, previous or current pregnancy, current vaginal infections, pelvic organ prolapse, incontinence, or discomfort using an intra-vaginal device. Participants self-inserted the femfit®, in private, with the instruction ‘insert as you would a tampon’. Participants were then asked to perform a series of exercises that included: a PFM contraction and straining maneuverer in lying and standing, posterior pelvic tilt, bridge, abdominal curl and standing holding a 13.5kg weight. Each task was held between five and ten seconds. During each task, concurrent TPUS measurements were performed using a GE Voluson/4D ultrasound machine (GE health care, Oslo, Norwary) with 4-8Mhz curved array 3D/4D ultrasound transducer (RAB 4-8l/obstetric). Femfit® data was processed using excel, the resting pressure was calculated from the time interval before the exercise started and subtracted from the raw pressure measurements, then the average of each sensor for the middle two seconds of the task was used to provide the pressure values for each exercise (see figure 1). The sensor within the PFM zone that demonstrated the greatest positive change for the PFM contraction was selected as the PFM sensor, and sensor eight was selected as the IAP sensor. For comparison with ultrasound measurements, the PFM pressure was calculated by subtracting the IAP from the PFM sensor, and designated as PFMt (PFM true). Offline post-processing of ultrasound images were performed using proprietary software 4D view. Changes in PFM anatomy were measured using the anterior-posterior diameter of the levator hiatus (LHap) (figure 1). The LHap was measured at rest and midway through the task. The proportional change in LHap length was then calculated ((Measurement Rest - Measurement maneuver)/Measurement Rest) X 100%. 
Paired T-tests were used to determine if there were significant differences in PFMt between a PFM contraction and a straining manoeuvre in both lying and standing. To assess the ability of the femfit® to describe changes in the AP length of the PFMs, measurements for both tools for each individual task were grouped as either positive (>0mmhg/>0%LHap shortening) or negative (<0mmHg/<0% LHap Lengthening) and presented as a two by two contingency table. This was used to calculate the proportion of agreement between the two measures and proportion of disagreement between the two measures for all the tasks observed.

18 participants volunteered for the study. Mean age was 30.9 years (±8.5), mean BMI was 23.2kg/m2 (±2.2). 5 participants reported very occasional UI and one had undergone a coccygectomy, none had prolapse symptoms. Ultrasound images were poor for four tasks, and for two tasks the femfit® data was unable to be analysed, resulting in missing data in six instances.  There was a statistically significant difference in PFMt between a PFM contraction and a straining maneuver in both lying (mean difference 16.2 (95% CI, 7.7 to 24.7) mmHg, P =.001; and standing (mean difference 13.1 (95% CI, 7.3 to 18.9) mmHg, P < .0005. The femfit ® was 100% accurate at predicting a shortening of the LHap for a PFM contraction in both lying and standing. For the remaining tasks agreement was between 75% - 83% (Table 1)

The femfit® was able to distinguish between IAP and the pressure produced by constriction of the PFMs. Incorrect performance of a PFM contraction is common. Hence, the femfit® offers an exciting addition to the tools available for clinicians to assess and teach the correct performance of a PFM contraction in both lying and standing. The femfit® had the best agreement with the PFM contraction tasks and  poorest agreement with the straining tasks. Closer analysis of the data suggested that a PFMt pressure of ≥ 4mmHg corresponded with a shortening of the LHap by ≥ 10% across all the tasks assessed. A PFMt pressure < 4mmHg corresponded to a less than 10% shortening or lengthening of the LHap for all tasks. Commonly seen for the straining task was a negative PFMt pressure despite co-contraction and shortening of the LHap; this was frequently associated with visible bladder neck descent on the ultrasound images. Further studies that include vaginal wall descent measures as a second variable alongside PFM changes would likely add value in understanding changes of the pressure profile in these situations.

The femfit® was able to distinguish between pressures developed in the region of the PFM and IAP, therefore providing a more valid assessment of LHap changes during a series of tasks. It is the first PFM measurement device to offer independent PFM and IAP biofeedback during dynamic exercise activity.

Bø, K., & Sherburn, M. (2005). Evaluation of female pelvic-floor muscle function and strength. Physical Therapy, 85(3), 269–282.
Shaw, J. M., & Nygaard, I. E. (2017). Role of chronic exercise on pelvic floor support and function. Current Opinion in Urology, 27(3), 257–261.
Tian, T., Budgett, S., Smalldridge, J., Hayward, L., Stinear, J., & Kruger, J. (2018). Assessing exercises recommended for women at risk of pelvic floor disorders using multivariate statistical techniques. International Urogynecology Journal, 29, 1447–1454.