MRI Has Selective Role After Expert Ultrasound in Deep Endometriosis

Adding magnetic resonance imaging (MRI) after expert transvaginal ultrasound (TVS) improved detection of bowel and uterosacral ligament deep endometriosis (DE) in selected patients but lowered specificity at several anatomical sites, according to a prospective cohort study in Acta Obstetricia et Gynecologica Scandinavica.¹ Adhesions and fibrosis make surgical excision of DE technically difficult.² The authors concluded that the findings support a selective rather than routine role for second-line MRI.¹

Researchers Compared Ultrasound Alone With Ultrasound Plus MRI

Investigators at Stockholm South General Hospital, a tertiary referral center, enrolled a nonconsecutive series of 221 patients referred for suspected endometriosis between January 2019 and May 2022. Of these, 189 underwent surgery, including 112 who had been referred for second-line MRI because of suspected advanced disease, inconclusive TVS, or symptoms suggestive of advanced disease despite negative TVS findings. The paired analysis included 107 patients with available MRI scans.

TVS followed the International Deep Endometriosis Analysis protocol. Surgical visualization served as the primary reference standard, with histopathology as a secondary standard. The combined approach was defined as positive when either TVS or MRI was positive, a rule that can raise sensitivity but cannot improve specificity.

Bowel Detection Improved

Sensitivity for bowel DE rose from 81.0% with TVS alone to 91.4% with TVS plus MRI, a gain of 10.3 percentage points (95% CI, 3.9 to 21.2), equal to 6 additional true positives and 3 additional false positives. Specificity declined by 6.1 points. Among 58 patients with surgically confirmed bowel DE, TVS missed 11 lesions, 10 of them in the mid or upper rectum. MRI detected 6 of those 11, all in the upper rectum or rectosigmoid.

Sensitivity for uterosacral ligament involvement increased from 58.3% to 87.5%, a gain of 29.2 points (95% CI, 17.0 to 44.1), with 14 additional true positives and 12 additional false positives. Specificity fell from 59.3% to 39.0%.

Limited Value at Other Sites

MRI added little at other sites. Specificity for endometriomas dropped 19.0 points, and specificity for pouch of Douglas obliteration fell 13.3 points, while sensitivity rose only 2.3 and 1.6 points, respectively. Urinary tract sensitivity stayed low for both approaches, at 42.9% with TVS and 47.6% with the combined method, while specificity fell 9.3 points. Among the 12 patients who received urologic assistance that TVS had not flagged as necessary, MRI identified no additional true-positive urinary tract lesions.

In an exploratory analysis, combined sensitivity for bowel DE was 100% (35 of 35) among patients with previous abdominal surgery vs 77.3% (17 of 22) among those without (P = .006). The authors cautioned that this finding is hypothesis-generating.

In an exploratory review, 9 patients received colorectal surgical assistance that was not considered necessary after TVS. Seven had confirmed bowel DE, and in 3 of them MRI correctly identified involvement that TVS missed. The surgical plan was revised to include colorectal surgeons in all 3.

Limitations and Implications

Imaging was not interpreted independently or under blinded conditions, and MRI was read with knowledge of the preceding TVS findings. The paired cohort was a high-prevalence tertiary population, which may introduce spectrum bias and overestimate predictive values. The study was not designed to determine whether adding MRI improves surgical or patient-reported outcomes, and its 2019 to 2022 period predates recent consensus addenda for lateral pelvic compartment assessment.

“Whether this additional diagnostic information justifies the organizational effort and healthcare resources required for MRI cannot be determined from the present study,” the authors wrote. “In settings with flexible surgical resources, the added value of MRI may be limited, whereas in centers requiring predefined multidisciplinary planning, MRI may aid decision-making despite false positives.”

Study author Elisabeth Epstein reported a pending patent on AI-driven diagnostic support for ultrasound assessment of ovarian tumors, as well as stock in and a board role at Intelligyn. The other authors declared no conflicts of interest.

References

1. Rehal M, Brunes M, Suzuki C, Epstein E. Second-line MRI after expert transvaginal ultrasound in selected patients with suspected complex deep endometriosis: a prospective cohort study. Acta Obstet Gynecol Scand. 2026;00:1-13. doi:10.1111/aogs.70372
2. Becker CM, Bokor A, Heikinheimo O, et al. ESHRE guideline: endometriosis. Hum Reprod Open. 2022; 2022(2):hoac009. doi:10.1093/hropen/hoac009