Background
Detailed pre‐operative description of endometriotic lesions by non‐invasive methods is an important tool for accurate diagnosis and effective treatment of the disease. Transvaginal ultrasound (TVUS) is a sensitive method for diagnosis of deep infiltrating endometriosis (DIE); however, it is highly operator‐dependent and consistent results require adequately trained and experienced clinicians.

Aims
The aim of the study is to assess the accuracy of TVUS in predicting DIE by comparing it with laparoscopic findings. We also compared US done in the community by general radiologists with examinations done by specialist gynaecologists.

Materials and Methods
A retrospective cohort study of patients who underwent laparoscopy for excision of possible endometriosis between July 2014 to February 2019 who had a TVUS prior to laparoscopy.

Results
A total of 119 patients were included. TVUS was shown to be useful in detecting all but bladder DIE. Community TVUS was no better than chance at identifying most DIE (area under the curve (AUC) of 0.48–0.60) except in the detection of ovarian endometriomas and adhesions (AUC = 0.84). Specialist TVUS correctly identified most DIE with greatest utility for DIE in rectosigmoid (AUC = 0.85, P < 0.000), followed by pouch of Douglas/pouch of Douglas adhesions (AUC = 0.82, P < 0.000), ovarian endometriomas/ovarian adhesions (AUC = 0.79, P < 0.000), uterosacral ligaments (AUC = 0.75, P < 0.000) and rectovaginal septum (AUC = 0.69, P < 0.05). Conclusion Specialist TVUS is informative in examining the presence of DIE particularly in posterior compartments which may increase surgical complexity. Community TVUS is significantly less beneficial; however, it is more accessible to the general public. This adds to the argument that increasing access to DIE‐TVUS appears favourable.