A systematic review of the accuracy of ultrasound in the diagnosis of endometriosis.

Abstract

To evaluate transvaginal and transabdominal ultrasound scanning, with or without Doppler, as a diagnostic test for the accurate diagnosis of pelvic endometriosis. The MEDLINE (1966-2001) and EMBASE (1980-2001) databases were searched for relevant studies, published in English. Only studies fulfilling predefined criteria were selected. An assessment of quality was made for each study, and data were then reanalyzed using likelihood ratios to determine the usefulness of the test. In total, 67 papers were identified using the search strategy, of which 17 described relevant studies. Of these, seven fulfilled the inclusion criteria. All seven related to the use of transvaginal gray-scale imaging in the diagnosis of ovarian endometriomata specifically, rather than endometriosis. The positive likelihood ratios ranged from 7.6 to 29.8, and the negative likelihood ratios ranged from 0.1 to 0.4. Confidence intervals were wide. One paper addressed the use of conventional color Doppler with ultrasound: the positive likelihood ratio was 1.2, with a negative likelihood ratio of 0.4. One paper assessed the use of color Doppler energy imaging, and showed a positive likelihood ratio of 33.5 and a negative likelihood ratio of 0.1. Transvaginal ultrasound appears to be a useful test both to make and to exclude the diagnosis of an ovarian endometrioma.