Work

Evaluation and performance of the β-hCG discriminatory zone and targeted ultrasound findings for early pregnancy diagnosis in a high-risk population: a retrospective cohort study at a US safety-net hospital

Year: 2026

Type: article

Abstract: Objectives To evaluate the diagnostic performance of the American College of Obstetricians and Gynecologists (ACOG)-recommended β-hCG discriminatory zone (DZ) of 3500 mIU/mL for detecting hallmarks of early intrauterine pregnancy (IUP) and for identifying ectopic pregnancy. To assess how ultrasound findings in the absence of an IUP modify the likelihood of ectopic pregnancy. Design This study is a retrospective review of same-day quantitative beta-human chorionic gonadotropin (β-hCG) and pelvic US (February 2018–March 2025) was performed. Targeted US markers (adnexal lesions, peritoneal free fluid) were recorded. Diagnostic accuracy metrics and receiver operating characteristic analyses were calculated. Setting All patients were evaluated at Hurley Medical Center, a safety-net hospital, serving a predominantly low-resource, high-risk population. Participants Patients with a uterus and a positive qualitative serum β-hCG result with a same-day transvaginal pelvic ultrasound from February 2018 through March 2025 were included. Patients with no available information on same-day pelvic ultrasound and quantitative β-hCG levels were excluded. Primary and secondary outcome measures Outcomes included visualisation of gestational sac (GS), yolk sac (YS), fetal pole (FP), fetal heartbeat (FHB) and EP (ectopic pregnancy). Results Among 1517 patients, visualisation of GS, YS, FP and FHB was significantly more likely above the DZ (all p<0.001), whereas EP was below the DZ (p<0.001). β-hCG showed excellent discrimination for early IUP markers [area under the curve (AUC) (95% CI): GS 0.925 (0.910 to 0.939), YS 0.921 (0.907 to 0.935), FP 0.907 (0.892 to 0.922), FHB 0.906 (0.891 to 0.921) but poor discrimination for EP (AUC (95% CI): 0.236 (0.201 to 0.270) for all pregnancies; 0.537 (0.484 to 0.590) after excluding IUPs]. Notably, 86% of EPs occurred with β-hCG<3500 mIU/mL. Across all pregnancies, EP risk increased stepwise with targeted US markers: 1.6% with neither adnexal lesion nor free fluid, 7.5% with one, and 26.0% with both (p<0.001). In the absence of an IUP, corresponding risks were 4.5%, 13.4% and 41.3% (p<0.001). Conclusion In this high-risk, low-resource cohort, the ACOG β-hCG DZ reliably predicts early IUP US features but is unreliable for diagnosing EP. Integrating targeted US markers, particularly adnexal lesions and peritoneal free fluid, markedly improves EP risk stratification, supporting a tiered approach.

Source: BMJ Open

DOI: 10.1136/bmjopen-2025-116039