Using transvaginal ultrasounds for the identification of follicle-like structures ≥10 mm in a multi-center randomized contraceptive clinical trial of three 84-day oral ulipristal acetate regimens.

PURPOSE: We examine the use of transvaginal sonography in imaging ovarian follicles among non-pregnant reproductive-aged women enrolled in a contraceptive clinical trial. METHODS: Ten sites conducted a clinical trial comparing three oral ulipristal acetate regimens for ovulation inhibition. Enrollees underwent twice weekly transvaginal sonography and hormonal blood testing throughout treatment and until the second menses post-treatment. The primary outcome of the present analysis was success in identifying follicles ≥10 mm in a subgroup of sonograms performed at times of likely follicular activity (estradiol <100 pg/ml and progesterone >3 ng/ml). Secondarily, an overall analysis assessed all sonograms regardless of timing. RESULTS: At times of likely follicular activity, 135 women underwent 969 sonograms. The proportion of sonograms without follicles ≥10 mm ranged from 8.3% to 46.3% when stratified by site (p = .01). The odds of no follicle among higher-body mass index (BMI) women were 1.91 times those among lower-BMI women after controlling for site and treatment arm (95% CI: 1.13, 3.22). Results were similar for the overall population. CONCLUSION: Sonographers were less likely to identify follicles for higher-BMI women and results varied by site. Machine quality slightly impacted sonogram results, but did not sufficiently explain site differences.