Endovaginal sonography for the diagnosis of upper genital tract infection

To determine the clinical utility of transvaginal sonography for the diagnosis of upper genital tract infection. Fifty-five women who either met the Centers for Disease Control and Prevention's minimal criteria for acute pelvic inflammatory disease or were being seen for nonclassic signs of upper genital tract infection were evaluated. During abdominal and endovaginal ultrasound testing, fluid in the cul-de-sac, discrete tubes with or without tubal fluid, multicystic ovaries, and adnexal masses were noted. Upper genital tract infection was confirmed by laparoscopic visualization or histologic or microbiologic evidence of salpingitis or endometritis. The specificity of identifying fallopian tubes with or without intraluminal fluid on ultrasound was 97% (35 of 36); the sensitivity, however, was only 32% (six of 19). Calculated using Bayes theorem and based on a prevalence rate of 50%, the positive predictive value of visualizing fallopian tubes was 91%. The sensitivities associated with the visualization of a multicystic ovary or tubo-ovarian abscess were 42% (eight of 19) and 32% (six of 19), with specificities of 86% (31 of 36) and 97% (35 of 36), and positive predictive values of 75% and 91%, respectively. Cul-de-sac fluid was associated with low sensitivity (37%; seven of 19), low specificity (58%; 21 of 36), and the lowest positive predictive value (47%). Endovaginal sonography has limited clinical utility in the diagnosis of upper genital tract infection due to its low sensitivity.