Study design, materials and methods
This study was based on building a model of the human pelvic floor. The size of the genital hiatus was re-created using data describing the smallest and largest levator hiatal measurements as identified by 3D trans-perineal ultrasound [3]. A 1 inch thick board had small and large levator hiatus cut outs with the dimensions, 3.4cm by 3.7cm and 8.2cm by 6.4cm respectively. A neonatal blood pressure cuff (Phillips) was placed within the hiatus around the circumference to represent the pelvic floor. Inflation of the blood pressure cuff was to set pressure readings of 10, 20, 40, 60, 80, 100mmHg represented different strengths of a maximal pelvic floor muscle contraction.
Digital examination of the model levator hiatus (mLH) was carried out using the Oxford Scale of muscle strength assessment score to assess the strength of the stimulated pelvic floor contraction (sPFC). The positioning of the model was to mimic a patient in the supine position. Examinations were carried out by 3 experienced healthcare practitioners who were blinded to pressures in the pelvic floor cuff. The maximum and minimum contraction of each mLH was first palpated by each practitioner. Digital examination of the mLH was then carried out at set inflation pressure points of 10, 20, 40, 60, 80, 100mmHg in random order. The observer was blinded to the strength of the simulated pelvic floor muscle contraction as well as being blinded to the findings obtained by other observers. Each pressure point was assessed twice at different times of the day. Between stimulated contractions the cuff was completely deflated (stage 0), digital examination was carried out from this baseline point to the set pressure point of the sPFC.
Examination of the small and large model hiatus was repeated using a perineometer (Peritron, (NEEN Healthcare, Oldham, UK) ). A baseline reading of 0mmHg was set to assess all simulated pelvic floor contraction pressures for both the large and small model hiatus sizes. The test was carried out twice for each hiatus size and pressures to enable a test retest analysis. The tests were repeated with increasing perineometer baseline settings of 10, 20, 30, 40, 50, 60, 70 mmHg by inflating the perineometer with air.
Results
Correlation between increasing cuff pressure was positively associated with a higher Oxford score following digital examination for both the small and large levator ani hiatuses, Spearmans rank correlation coefficient (rs 0.87, p<0.05) and (rs 0.95, p<0.05) respectively.
The intra-observer correlation coefficients were 0.75, 0.74, 0.71 for digital assessment of the small model hiatus and 0.87, 0.89, 0.88 for the large model hiatus. The inter-observer correlation coefficients were 0.86 and 0.94 for the small and large model hiatus’ respectively. Correlations above 0.61–0.80 are good and 0.81–1.00 excellent.
A Mann-Whitney U test was carried to assess differences in Oxford scores between the two models. Oxford scores for the small levator ani hiatus (33) were not significantly different to the large levator ani hiatus (45), U = 12, z = -0.96, p > 0.05 for each sPFC strength.
Intra-observer correlation coefficient for perineometer readings varied between the size of the mLH and the perineometer baseline pressure. For the small mLH the ICC ranged from 0.96 – 0.99. For the large mLH the ICC varied from 0.70, 0.26, 0.70, 0.65, 0.76, 0.95, 0.92, 0.77 for each 10mmHg increment starting at a baseline of 0mmHg to 70mmHg.
An independent-samples t-test was run to demonstrate a significant difference between the recorded sPFC pressure rise from baseline to maximum sPFC (cuff pressure of 0mmHg to 100mmHg). The average pressure rise in the small levator ani hiatus was 39mmHg ± 9.9 and 4.4mmHg ± 1.6 for the large levator ani hiatus. A statistically significant difference was observed t = 3.3, P< 0.01.
There were significant differences between the perineometer readings with the pelvic floor pressures below 40mmHg between the small and large levator ani hiatuses (t-test, p<0.03), however above this pressure there was no significant difference in readings.
Interpretation of results
Through a positive correlation, we demonstrate the functionality of the model created for this study. There was good or excellent inter and intra-observation correlation between examination findings of sPFC between both levator ani hiatus models. There was no difference in the digital assessment scores given to each model for each sPFC strength.
Perineometer measurements showed a larger incremental rise from baseline in the small levator ani hiatus compared with the large levator ani hiatus with equal sPFC strength. However this is not associated with an actual higher strength PFC as confirmed by concordant Oxford scores from digital examination for the differing levator ani hiatus sizes.