IS TYPE OF RECTOVAGINAL SEPTUMDEFECT ASSOCIATED WITH THE SEVERITY OF POSTERIOR COMPARTMENTPROLAPSE? A 2D AND 3D ENDOVAGINAL ULTRASOUND ASSESSMENT
A. HEGDE1, N.CHANDRASEKARAN 2, V. AGUILAR 3, G. DAVILA 2;
1CENTER FOR UROGYNECOLOGY AND PELVIC HEALTH (C.U.P),NEW DELHI, India, 2Cleveland Clinic Florida, Weston, FL,3UTSW at Austin, Austin, TX.
Introduction: Though recentimaging  and histochemical  studies have supported the presenceof the rectovaginal septum (RVS), its length and thickness is amatter of debate. Also a matter of controversy is the issue ofwhether or not the RVS, if present, is a crucial structure involvedin the development of posterior vaginal wall prolapse . Recentlydefects in the RVS including gaps and diffuse blurring were found tobe associated with the development of posterior compartment prolapseon 2D and 3D endovaginal ultrasound (EVUS).
Objective:To use 2D and 3D EVUS to correlate the type of RVS defect with theseverity of posterior compartment prolapse.
Methods: Thisis a retrospective cohort study of 90 women attending our center fromAugust 2011 and June 2013. 50 patients with stage 2 rectocele onPOP-Q assessment constituted Group A and 40 patients with rectocelestages 3 and 4 on POP-Q assessment constituted Group B. Exclusioncriteria in the two groups was history of any prior pelvic surgery orlaser treatment of the vagina. All patients underwent 2D and 3D EVUSof the posterior pelvic compartment with the 8848 transducer (BKMedical Ultrafocus, Peabody, MA) by a fellow who was blinded to theresults of the POP-Q assessment. The 30 seconds 3D volumes obtainedat rest were assessed to determine the presence of the RVS. Thelength of the RVS was determined in two halves: from the anorectalangle caudad until the perineal body and from the anorectal anglecranially until the cul-de-sac. The thickness of the RVS wasdetermined at the level of the anorectal angle. The presence of anydefects in the RVS including gaps in the RVS, thinning or diffuseblurring was noted. 2D assessment was performed during cough andValsalva maneuvers to determine the in vivo behavior of the RVSduring stress.
Results: The two groups matched with respectto their demographic data of age, BMI, parity, smoking history andmenopausal status (p > 0.05). The RVS was visualized as a distincthyperechogenic layer between the vaginal and anorectal muscularis inall patients (figure 1). The characteristics of the rectovaginalseptum including presence of defects in the two groups are as givenin table 1. The two groups were similar with respect to the length ofthe RVS and thickness of the RVS at the anorectal junction (p >0.05). The number of patients with a gap in the RVS that was distalto the anorectal junction was similar between the two groups (p =0.3992). However, number of patients with diffuse apical (i.e. at theattachment of the RVS to the cul-de-sac) blurring of the RVS or anapical gap was significantly more in group B when compared with groupA (p < 0.0001). The number of patients with folding or crumblingof the RVS on Valsalva maneuver was similar between the two groups (p> 0.05). Concomitant enterocele was noted in all the 24 patientsin Group A and 35 (94.5%) patients in Group B with apicalblurring/gap of the RVS.
Conclusions: RVS is a definedentity and exists as a hyperechogenic entity between the vaginal andanorectal muscularis. Patients with higher grade posteriorcompartment prolapse have a higher prevalence of apical blurring/gapof the RVS.
References: 1. Ultrasound in Obstetrics andGynecology 2011; 37 (3): 348-352. 2. Dis Colon Rec. 2009;52(9):1564-1571.3. IUJ 2016 Suppl (1): s121 - 122.
Table1: Characteristics of rectovaginal septum including presence of adefect
GroupA (n = 50) Patients with stage 2 rectocele
GroupB (n = 40) Patients with stage 3 and 4 rectocele
Lengthof RVS above the anorectal junction*
Lengthof RVS below the anorectal junction*
RVSthickness at the anorectal junction*
Gapin RVS distal to the anorectal junction n (%)^
Diffuseapical blurring or apical gap of RVS n (%)^
Foldingof RVS on Valsalva n (%) ^
Crumblingof RVS on Valsalva n (%)^
*Median(Interquartile range); p value: Mann Whitney U test. ^Fisher Exacttest