Non-invasive methods of assessing pelvic floor muscle (PFM) contractions are desirable for patients with inflammatory bowel disease (IBD), especially when intra-anal examination or biofeedback may be painful or distressing. Transperineal ultrasound (TPUS) assessment of the PFM, during contraction and simulated defecation, correlates well with digital, manometric and defecographic assessments, is minimally invasive and can be used as biofeedback [1,2]. 
The aim of this study was to assess PFM activity in patients with quiescent IBD and persistent bowel symptoms using TPUS as a basis for determining whether any obtained measures can be correlated to pelvic floor dysfunction.

TPUS assessments were performed on patients with quiescent IBD and symptoms of fecal urgency, fecal incontinence or evacuation difficulty as part of an intervention study including PFM training, modification of toileting behaviors, lifestyle advice and practical management strategies.

Examinations were performed by one assessor, blinded to symptoms. An Aixplorer V11 ultrasound with a 3.5 - 4.5 MHz curved transducer recorded cine-loops in B-mode.  The transducer was placed on the perineum in the mid-sagittal plane providing a two-dimensional image with a 70° field of view. 

Patients emptied their bladder before reclining with hips and knees flexed and back supported 45° from vertical. A brief practice of PFM contraction preceded formal assessment.  The ultrasound screen could not be viewed by patients, thus preventing altered performance from visual feedback during the tests:
- Maximum pelvic floor voluntary muscle contraction, held for three seconds
- Maximum simulated defecatory effort, pushing for six seconds 
each repeated three times at one-minute intervals
The cine-loops were analyzed by a different examiner, blinded to patient information. Still images best depicting maximum contraction and simulated defecation were extracted for measurement.
  
The most anterior point of the anorectal angle (ARA), being the junction between the posterior rectal and anal walls, was designated the inflection point (IP). The most posterior point of the pubic symphysis (PS) border was identified for use (Figure 1A).

The methods for image extraction and measurement were tested for reliability before computer measurements using ImageJ software:
- Distance between the IP and PS (puborectalis length) at rest, on contraction and on simulated defecation
- IP displacement, from maximum contraction and maximum simulated defecation, measured in two ways from the IP resting position: 
(a) the shortest straight line (direct) displacement and
(b) the anatomical co-ordinate directions anterior (ventral) and superior (cranial) on contraction, and inferior (caudal) and posterior (dorsal) on simulated defecation (Figure1B). 
- Change in ARA relative to the resting value on contraction and simulated defecation

Thirty-three (94%) of 34 patients enrolled in the intervention study had TPUS assessment at baseline Twenty-two (67%) of these 33 patients (15 females) had TPUS assessments at the completion of treatment. Three had poor quality images, three did not have an ultrasound at study completion, and five withdrew from the study for medical reasons. 

The PS was identifiable in the images of 21 subjects. At baseline the mean (SD) puborectalis lengths (distances between the IP and PS) at rest and on maximum contraction, were 50.2 (8.5) mm and 44.4 (6.5) mm respectively, with a mean change of 5.7 (4.4) mm. Continent patients had significantly shorter puborectalis lengths on maximum contraction at baseline than patients with faecal incontinence [41.9 (1.5) mm versus 49.1(2.8) mm; p = 0.02].

With simulated defecation the mean (SD) distance between the IP and PS was 55.7 (9.9) mm, with a mean change of 3.4 (6.5) mm. The puborectalis lengths on PFM contraction and simulated defecation are similar to measurements reported in healthy female controls [3]. 

Measurements are reported in Table 1. All 22 patients produced an anterior displacement of the IP on PFM contraction at both time points. The anterior displacement increased after treatment. Although not statistically significant, the effect size was moderate. There was significant individual variation, with displacement on contraction ranging from 0.8 to 22.8mm.

Correct push (posteroinferior displacement of the IP) was observed in 11 (50%) and 16 (73%) of the 22 patients, before and after treatment respectively. The inferior IP displacement with simulated defecation was reduced following treatment suggesting less downward straining force.

There was a strong correlation between the change in IP to PS length and anterior displacement of the IP with maximum contraction (Pearson’s r 0.86, p < 0.001), indicating that IP anterior displacement is a reliable substitute for the IP to PS measurement when the PS is not identifiable.
The changes in the ARA during contraction and simulated defecation were less than the 15 to 20 degrees considered normal and did not change significantly following treatment. 
There were no significant correlations between any of the ultrasound measurements and symptoms following treatment.

These results suggest that TPUS assessment is accepted by most patients with IBD and suitable images can be obtained in the majority of cases. Though the mean puborectalis lengths measured approximated those reported for healthy women, continent patients were able to reduce puborectalis length significantly more during contraction than incontinent patients. While all patients were able to contract correctly, half the patients had difficulty coordinating simulated defecation at baseline. 

Anterior displacement of the IP of the ARA, during voluntary maximum PFM contraction, correlated strongly with the measurement using the pubic symphysis as a reference point.  This method could be useful when the pubic symphysis is not identifiable in recorded images.

Limitations include a small cohort and the lack of simultaneous electromyographic recordings of PFM activity and movements observed could not be assumed to be PFM contraction or relaxation alone. Simulated defecation was not practiced in a normal toileting position and patients, despite reassurance, may have been concerned about fecal or flatal leakage.

TPUS assessment was accepted by more than 90% of patients with IBD. Baseline puborectalis lengths on PFM contraction were significantly shorter in continent patients but no associations were found following treatment. Powered studies are recommended to examine PFM function in patients with IBD and the relationship between TPUS measurements and clinical symptoms. 
Anterior displacement of the IP, during voluntary maximum PFM contraction, appears to be a reliable alternative measure when the pubic symphysis is not identifiable as a reference point. 
TPUS is recommended as a non-invasive means of PFM assessment and biofeedback in patients with IBD.

Beer-Gabel M, Carter D. Comparison of dynamic transperineal ultrasound and defecography for the evaluation of pelvic floor disorders. Int J Colorectal Dis 2015;30:835-841.
Dietz HP, Jarvis SK, Vancaillie TG. The assessment of levator muscle strength: A validation of three ultrasound techniques. Int Urogynecol J Pelvic Floor Dysfunct 2002;13:156-159.
Thibault-Gagnon S, McLean L, Goldfinger C,et al. Differences in the biometry of the levator hiatus at rest, during contraction, and during valsalva maneuver between women with and without provoked vestibulodynia assessed by transperineal ultrasound imaging. J Sex Med 2016;13:243-252.