Reid 2014.

Study characteristics
Patient sampling	Primary objectives: to evaluate the use of a newly modified sonovaginography (SVG) technique, outpatient 'office gel SVG', for prediction of posterior compartment deep infiltrating endometriosis (DIE) Study population: women who presented to pelvic pain clinic with symptoms suggestive of endometriosis Selection criteria: Inclusion criteria: reproductive age, history of chronic pelvic pain ± history of endometriosis, laparoscopy within 6 months of gel SVG examination. Exclusion criteria: malignancy, menopause, pregnancy Study design: multi‐centre, prospective, observational study; consecutive enrolment
Patient characteristics and setting	Clinical presentation: chronic pelvic pain, dysmenorrhoea, dyspareunia, dyschezia; mean duration of pain 39.7 ± 47.5 months; history of infertility 44/220; history of endometriosis 92/220; history of bowel DIE in the past 10/220 Age: mean 32.2 ± 7.5 years Number enrolled: 220 women Number available for analysis: 189 women Setting: 4 university teaching hospitals, tertiary referral centres: Nepean Hospital, Royal Hospital for Women, Royal Prince Alfred Hospital, Liverpool Hospital; 5 private hospitals: Norwest Private Hospital, Hurstville Private Hospital, St. Luke’s Private Hospital, Prince of Wales Private Hospital, St. George Private Hospital Place of study: NSW, Australia Period of study: January 2009 to February 2013 Language: English
Index tests	Index test:Sonovaginography (SVG) Description of positive case definition by index test as reported: RVS DIE predicted during SVG when hyperechoic RVS layer was interrupted or was no longer visible. USL nodules identified as defined hypoechoic lesions located laterally alongside the cervix. Vaginal DIE identified as solid nodule in vaginal wall. Diagnosis of rectal/rectosigmoid DIE made when the normally appearing linear hypoechoic longitudinal muscle was thickened by a solid hypoechoic mass, which was continuous within the longitudinal muscle of the bowel and could be visualised in both sagittal and transverse planes. The 'sliding sign' was performed during gel SVG to determine whether adhesions existed between anterior rectum/rectosigmoid and posterior vaginal wall/cervix/uterus (i.e. POD obliteration). Description is supported by pictures of images for each compartment. Technique is described in detail Examiners: all SVG examinations performed by 2 operators (1 was an expert gynaecological sonologist with experience in diagnosis of DIE; the other was a gynaecological ultrasound fellow supervised by an experienced operator). Same person who performed SVG performed the gynaecological examination and TVS. Operators were not blinded to clinical history Interobserver variability: not provided
Target condition and reference standard(s)	Target condition: posterior DIE ‐ overall and separate anatomical sites (USL, RVS, vagina, bowel including anterior rectum and rectosigmoid) Prevalence of target condition in the sample: pelvic endometriosis 146/189 (77%), posterior DIE 57/189 (30%), separate compartments of endometriosis: bowel 43/189 (23%), vaginal 11/189 (6%), RVS 11/189 (6%), USL 10/189 (5%), POD obliteration 47/189 (25%) Reference standard: laparoscopy 189/189 (100%) + histopathology Description of positive case definition by reference test as reported: surgical diagnosis of endometriosis made if any of the following was satisfied: 1 ‐ histological confirmation of endometriosis in ≥ 1 resected nodule; 2 ‐ visualisation and palpation of subperitoneal nodule without biopsy and another histologically proven location of endometriosis; 3 ‐ visualisation of complete obliteration of cul‐de‐sac. Surgical findings verified through primary author's review of detailed operation reports and diagrams made at the time of surgery; surgical procedure described Examiners: surgery performed by a total of 13 laparoscopic surgeons: 9 advanced laparoscopic surgeons and 4 general gynaecological surgeons. Surgeons not blinded to patient data, including results of the index test
Flow and timing	Time interval between index test and reference standard: within 6 months Withdrawals: 31 women chose not to undergo surgery after consultation with their surgeon; no additional details provided
Comparative
Key conclusions by the authors	Office gel SVG appears an effective imaging technique for detection of bowel DIE, with higher accuracy for prediction of rectosigmoid vs anterior rectum DIE. SVG found to have high specificity and NPV for all forms of DIE, indicating that negative SVG examination correlates highly with absence of DIE at laparoscopy. This new technique may not only aid in triaging of women for referral to an advanced laparoscopic surgeon ± colorectal input, but may act as a useful learning tool for visualisation of posterior pelvic compartment in women with suspected DIE
Conflict of interests	Not reported
Notes	Reported accuracy estimates for SVG for diagnosis of posterior DIE, overall and specific sites (rectosigmoid/anterior rectum, USL, RVS/vagina, POD obliteration), confirmed as correct
Methodological quality
Item	Authors' judgement	Risk of bias	Applicability concerns
DOMAIN 1: Patient Selection
Was a consecutive or random sample of patients enrolled?	Yes
Did the study avoid inappropriate exclusions?	Yes
Was a 'two‐gate' design avoided?	Yes
		Low	Low
DOMAIN 2: Index Test Any test
Were the index test results interpreted without knowledge of the results of the reference standard?	Yes
Did the study provide a clear pre‐specified definition of what was considered to be a “positive” result of index test?	Yes
Was the index test performed by a single operator or interpreted by consensus in a joint session?	Yes
Were the same clinical data available when the index test results were interpreted as would be available when the test is used in practice?	No
		High	Low
DOMAIN 3: Reference Standard
Is the reference standards likely to correctly classify the target condition?	Yes
Were the reference standard results interpreted without knowledge of the results of the index tests?	No
		High	Low
DOMAIN 4: Flow and Timing
Was there an appropriate interval between index test and reference standard?	Yes
Did all patients receive the same reference standard?	Yes
Were all patients included in the analysis?	Yes
		Low