We conducted a prospective diagnostic-accuracy study across two obstetric units evaluating first-trimester ultrasound for early detection of gynecological conditions during pregnancy. Consecutive patients ≤14 weeks’ gestation underwent protocolized transvaginal scanning (with complementary transabdominal views and targeted Doppler under ALARA), structured reporting of adnexal/cervical findings (IOTA/O-RADS descriptors, cervical-length technique), and blinded verification against a composite reference standard (surgery/pathology when available or standardized 30-day clinical follow-up). Analyses (IBM SPSS v27) included accuracy metrics, ROC curves, and inter-reader reliability. Among 320 participants (prevalence of target conditions 25%), overall performance was high: sensitivity 90.0%, specificity 92.1%, accuracy 91.6%, AUC 0.94, with strong rule-in/rule-out value (LR+ 11.4; LR− 0.11). Condition-specific results were excellent for ectopic pregnancy (sensitivity 94%, specificity 97%, AUC 0.97) and strong for adnexal torsion (86%, 95%, AUC 0.93), clinically significant adnexal masses requiring surgical referral (82%, 90%, AUC 0.90), and actionable cervical pathology (80%, 92%, AUC 0.89). Reliability was substantial-to-excellent (Cohen’s κ 0.82 for categorical calls; ICC(2,1) 0.90 for cervical length), and no scan-related adverse events were recorded. These data support a protocolized first-trimester ultrasound pathway as a front-door test that delivers accurate, reproducible triage for high-impact conditions during pregnancy and is ready for health-system scale-up with standardized training, documentation, and quality audit.
Nadaf et al. (2025) studied this question.