This cross-sectional, observational study received approval from the local institutional research ethics board and all participants provided written informed consent prior to engaging in any study activities. Healthy females, naïve to HEs, were recruited from the local community. Participants were excluded if they were <18 years old, were in menopause, experienced chronic or recurrent pelvic pain, had a history of pelvic surgery, or were pregnant or < six months postpartum. Participants attended an initial training session where they learned from a certified Hypopressive trainer how to perform HEs in supine and standing in two HPs (Demeter and Athenas, respectively), a follow-up training session one week later to ensure the exercises were being performed correctly, and a data collection session one week after that.
The primary outcomes were transient changes observed in IAP, EMG amplitude of the LAMs and EAS, and pelvic morphology [levator plate length (LPL), bladder neck height (BNH) and levator plate angle (LPA)] observed on 2D transperineal ultrasound imaging (USI), acquired during the HEs.
At the data collection session, LAMs strength and stiffness were recorded using a custom intravaginal dynamometer. Electrodes were then located intravaginally over the LAMs (differential suction electrode) and on the skin surface overlying the EAS, interfaced with Delsys D.E. 2.1 preamplifiers and a Bagnoli-8 EMG amplifier system (Delsys Inc, Boston, USA). An IAP sensor [2] was inserted into the posterior fornix of the vagina. Participants first performed three maximum voluntary contractions (MVCs) of their PFMs (maximal effort squeeze and lift). Next, in random order, they performed three repetitions of the HE maneuver with and without the HP in supine and standing. All outcomes [(EMG, IAP and transperineal USI videos (GE Voluson S6; RAB6-D 4D convex curvilinear probe, GE, Toronto, Canada)] were acquired synchronously throughout these tasks.
EMG data had any offset removed, were full-wave rectified, and were smoothed using a 4th order, dual-pass low-pass Butterworth filter (cut-off 6 Hz). The peak of the EMG signal during each HE task was normalized to the highest peak achieved during the PFM MVCs.
Outcome data were tested for normality (Shapiro-Wilk test). Student t-tests were used to determine whether there were changes in IAP, EMG amplitude or pelvic morphology during the HEs performed in supine or standing in conjunction with the corresponding HP. Separate two-way repeated-measures analysis of variance (RM-ANOVA) models and non-parametric Friedman’s ANOVAs were used to determine if there were differences in outcomes between positions (supine vs standing) or postures (HP vs no HP) or interactions between position and posture. An adjusted α=0.05/10 was set for all hypothesis testing to account for multiple outcomes. Despite some non-normal variables, there were no differences in the outcomes of hypothesis testing between parametric and nonparametric analyses, and the histograms and Q-Q plots suggested pseudonormality; all outcomes are therefore presented using parametric analyses for consistency.
The sample size was determined apriori based on pilot (n=7) IAP and EMG data acquired to address objective 1. IAP was lower (supine: 13.47±14.82 cmH20; standing 18.90±14.90 cmH20) and LAM EMG was higher (supine: 9.6±6.2 uV; standing 5.5 ±3.7 uV) during the HE compared to rest. To achieve power=0.80 (α=0.05), the required minimum sample size was n=9 for a reduction in IAP and n=12 for activation of the LAMs. A moderate effect size (d=0.50) was assumed for the effect of the HP, requiring a sample size of n=30 to reach statistical significance.