A noninferiority randomized controlled trial to compare transabdominal and transvaginal sonography for eligibility assessment prior to medical abortion

Abstract

Objectives: To compare transabdominal sonography (TAS) to transvaginal sonography (TVS) in medical abortion eligibility assessment, specifically to measure how often clinicians chose to order additional testing for eligibility assessment following TAS and TVS, and to look for differences by patient and clinician characteristics. Also, to compare patient acceptability between the two modalities.

Study design: This pragmatic multisite randomized noninferiority trial compared TAS to TVS at 10 New York City and New Jersey health centers that provide medical abortion. Women seeking medical abortion were randomized 1:1 to receive TAS or TVS. Following the study ultrasound examination, clinicians determined whether participants were eligible for medical abortion based on these results or warranted further testing. All participants completed an acceptability questionnaire. We compared additional testing and acceptability between TAS and TVS.

Results: Of those randomized to TAS, 63/317 (19.9%) received additional testing compared to 15/312 (4.8%) randomized to TVS. After TAS, most additional testing consisted of a same-day TVS. Other tests included β-hCG testing, scheduled repeat sonography or return visit. After TAS, 13.4% seen by physicians and 27.6% seen by advanced practice nurses (APNs) received additional testing (p<.01). Additional testing was more common in early gestational ages for both groups. We enrolled too few women with a body mass index (BMI) >35 kg/m² to make comparisons. Participants found TAS more acceptable than TVS, and two thirds preferred TAS for future care.

Conclusions: TAS provided sufficient information for clinicians to assess medical abortion eligibility without additional tests for most patients. However, the frequency of additional testing was exceedingly close to our predefined noninferiority boundary. Why APNs ordered substantially more additional testing than physicians is unclear. TAS was more acceptable to patients than TVS.

Implications: TVS use requires high-level disinfection, which is resource-intensive and thus can be a barrier to care. Instead, TAS can be first-line for most women, reducing resources needed to provide medical abortion. Further research could help to establish gestational age and BMI thresholds beyond which TVS would be a more informative first test. We also need to evaluate whether additional training in using TAS would decrease additional testing.

Trial registration: ClinicalTrials.gov NCT03047551.

Keywords: Medical abortion; Medication abortion; Mifepristone; Transabdominal ultrasound; Transvaginal ultrasound.