Accuracy of point-of-care ultrasound in diagnosing retained products of conception

Introduction: Retained products of conception (RPOC) refers to residual intrauterine tissue in female patients after a recent birth, pregnancy termination, or miscarriage. Ultrasound is the primary diagnostic modality for RPOC, but the accuracy of point-of-care ultrasound (POCUS) has not been evaluated for this diagnosis. Our aim was to determine the test characteristics of POCUS in detecting RPOC, and to evaluate the management of RPOC in the Emergency Department.

Methods: This was a retrospective cohort study of all patients presenting to an Emergency Department with over 110,000 annual visits between January 1, 2017, and December 31, 2023, who had an Emergency Department performed pelvic POCUS. All Emergency Department patients ≥18 years old who had a pelvic POCUS performed during the study period and had no identifiable intrauterine pregnancy were included. Patients were excluded if their sex was male, age was greater than 55, if there was inadequate chart data, or if there were inadequate or no POCUS images saved. Chart review was conducted by three nonblinded investigators, and POCUS review was conducted by two ultrasound fellows, blinded to the chart review. RPOC on POCUS was defined as heterogenous or hyperechoic products within the endometrium measuring greater than 10 mm. The gold standard for the diagnosis of RPOC on chart review was radiology ultrasound or obstetrics and gynecology (OBGYN) diagnosis for patients who did not receive a radiology ultrasound.

Results: There were 703 patients included in the study while 58 met exclusion criteria, leaving 645 patients for review. Radiology ultrasound was performed in 512 patients (79.4 %), and identified 42 cases of RPOC. In the 133 patients who did not receive a radiology ultrasound, 20 were confirmed to have RPOC based on OBGYN diagnosis, for a total of 62 patients with RPOC (a 9.6 % prevalence). There were 70 total POCUS examinations identified as RPOC, with 17 indeterminant examinations. In the 265 patients with a history concerning for RPOC based on chart review, the sensitivity and specificity for RPOC on POCUS were 79.0 % (95 % CI; 66.1 %-88.6 %) and 93.8 % (95 % CI; 90.0 %-96.6 %) respectively when compared to the gold standard diagnosis. If all eight indeterminant POCUS examinations from this group were considered positive, the sensitivity and specificity were 80.7 % (95 % CI; 68.1 %-90.0 %) and 90.4 % (95 % CI; 85.5 %-94.0 %) and 80.7 % (95 % CI; 68.1 %-90.0 %). Of the 62 total patients with RPOC based on chart review, 21 (33.9 %) were admitted, 26 (41.9 %) had a surgical procedure, 26 (41.9 %) were managed medically, and 10 (16.1 %) were expectantly managed.

Conclusion: POCUS demonstrated high specificity and low sensitivity for diagnosing RPOC in patients with a history concerning for RPOC. POCUS can be used to diagnose RPOC, but caution should be exercised when making the diagnosis in early pregnancy given the variable appearance of the endometrium on transabdominal POCUS during that timeframe.