Study design, materials and methods
Two authors (AH, DS) independently reviewed the abstracts and screened them for their relevance and content. All articles comparing transperineal, translabial or introital ultrasound with defaecation proctography were included. The examination of the anal sphincters was not included, but rather the abnormalities associated with obstructed defaecation (rectocoele, enterocoele, sigmoidocoele, peritoneocoele, intussusception, perineal descent, and dyssynergy). The literature search encompassed all publications to the end of February 2022. Review articles and those which were not published in English were not included.
Due to the heterogeneous nature of the studies reviewed, it was not possible to perform statistical analysis. For studies which provided the number of true and false positives and negatives but did not calculate accuracy, this was calculated using Stats Direct software. Agreement using Cohen’s Kappa coefficient was poor (0 - 0.20), fair (0.21 - 0.40), moderate (0.41 - 0.60), good (0.61 - 0.80) and excellent (> 0.81).
There were 570 articles identified during the initial literature search of which 28 were duplicates, and 304 were excluded as irrelevant after review of the abstract. There were 238 articles where the full text was reviewed. There were 14 which compared findings on transperineal (translabial/ introital) ultrasound with defaecation proctography. One study compared transperineal ultrasound with defaecation MRI.
Results
Transperineal or translabial ultrasound was performed with either a curved or linear array probe rested externally. One study performed introital ultrasound using an endocavity transducer placed near the hymeneal ring, on the posterior wall of the vulva. All performed ultrasound scanning during squeezing and maximal straining. There was variation in the use of gel, contrast and patient preparation. Some authors used rectal gel and encouraged evacuation of the gel toward the end of examination and performed ultrasound during rectal evacuation where possible.
Agreement for rectocoele detection between transperineal ultrasound and defaecation proctography was poor to excellent, specificity was 67 to 100% and sensitivity 29 to 88.9%. There was variation in the method for rectocoele measurement and the cut off for a pathological rectocoele on ultrasound. Grasso et al. found 100% of large rectocoeles (>4cm on proctography), 12% of moderate (2-4cm on proctography) but no small rectocoeles (<2cm on proctography) were visible on introital ultrasound. Steensma et al. showed that 87% of rectocoeles 2cm or greater on proctography but only 25% of those less than 2cm, were seen on ultrasound. Likewise, Beer Gabel et al. showed higher concordance for rectocoeles over 4 cm than those 2 to 4 cm.
A Cul de Sac hernia refers to either an enterocoele, sigmoidocoele or peritineocoele. Agreement for enterocoele (or Cul de Sac hernia) detection was good to excellent, specificity 92 to 96% and sensitivity 64 to 80% (table 3). Steensma et al. showed concordance between the two modalities was higher for grade 2 – 3 enterocoeles only than for all enterocoeles (kappa 0.77 versus kappa 0.65; 14/16 grade 2 – 3 enterocoeles but only 2/9 grade 1 enterocoeles were seen on ultrasound).
The description and grading for intussusception varied. Some calculated accuracy and agreement for internal (i.e. intussusception) and external rectal prolapse as one entity. Concordance for the detection of intussusception/rectal prolapse was poor to excellent, specificity 84 to 100% and sensitivity 22 to 100%.
Interpretation of results
Transperineal ultrasound is cheap, safe and portable. It is non-invasive and therefore avoids the soft tissue distortion associated with endoluminal probes and is well tolerated by patients. It allows the dynamic assessment of the entire pelvic floor including those functional and anatomical elements associated with obstructive defaecation namely rectocoele, enterocoele, sigmoidocoele, peritoneocoele, intussusception, perineal descent and dyssynergy (anismus).
Transperineal ultrasound has a high positive predictive value and specificity for rectocoele, enterocoele and intussusception. Negative predictive value and sensitivity are generally lower. Accuracy for detection of rectal prolapse is less clear. Concordance for all pathologies is variable. Transperineal ultrasound may be a suitable screening tool for obstructive defaecation and may avoid the need for defaecatory imaging in some patients.