Effect of Clinical Setting on Prevalence of Pregnancy of Unknown Location: A Retrospective Cohort Study [ID: 1377630

Abstract

INTRODUCTION: Pregnancy of unknown location (PUL) is the diagnosis of a positive pregnancy test without ultrasonographic confirmation of pregnancy location. Using varying diagnostic criteria, studies from the 1980s to early 2000s estimated the prevalence of PUL diagnoses at 8–31%. With advances in ultrasound technology and increasing use of early ultrasound, we aimed to determine current prevalence of PUL and differences by clinical setting. METHODS: With IRB approval, we conducted a retrospective cohort study of all initial pregnancy ultrasounds less than 14 weeks performed at a single academic medical center from January 1, 2018, to December 31, 2019. Ultrasounds occurred in the outpatient perinatal ultrasound suite or inpatient radiology suite via the emergency department. We compared diagnoses by setting via chi-square tests and logistic regression using Stata SE. RESULTS: Of 1,775 ultrasounds, only 81 (4.6%) were PULs, while 308 (17.4%) were intrauterine pregnancies of uncertain viability (IPUV). Both types of uncertain pregnancy diagnosis were more common in the emergency department than the outpatient obstetrics and gynecology setting (PUL: 8.3% versus 2.0%, P<.001; IPUV: 23.8% versus 12.9%, P<.001). Patients with symptoms of cramping or vaginal bleeding were more likely to have uncertain diagnoses (33.1% versus 16.8%, P<.001); however, presence of symptoms was not independently associated with uncertain diagnosis after controlling for setting. CONCLUSION: Diagnoses of PUL have declined, largely replaced by diagnoses of IPUV. Differences in diagnosis prevalence unexplained by patient symptoms support existing evidence of disparate early pregnancy experiences between the outpatient and emergency settings. As over one-fifth of patients will have an uncertain pregnancy diagnosis on initial ultrasound, safe, efficient, and patient-centered follow-up protocols are crucial.