Development and validation of a rapid 15-minute gastric ultrasound protocol for suspected gastroparesis: a...

Gastric emptying scintigraphy (GES) is the reference standard for gastroparesis but is limited by time, cost, and accessibility. We aimed to validate a practical 15-minute gastric ultrasonography (GUS) threshold for early assessment of suspected gastroparesis.

We conducted a prospective diagnostic accuracy study comparing GUS with GES. Participants were divided into training and validation cohorts by enrollment date. Using 4-hour GES as reference, the optimal GUS cut-off was derived by ROC analysis in the training cohort and tested in the validation cohort. Reproducibility and associations with symptom severity were assessed.

Of 123 participants screened, 113 completed both GUS and GES (training, n = 67; validation, n = 46). GUS_15min correlated with GES-derived emptying rates (GES_120min: r = 0.51; GES_240min: r = 0.50; both p < 0.0001). The optimal training cut-off was 0.195 (AUC 0.91, 95%CI 0.81-1.00). In validation, a threshold of 0.200 yielded an AUC of 0.90 (95%CI 0.80-0.99), sensitivity of 93.2%, and specificity of 88.3%, with substantial agreement with GES_240min (κ = 0.758). Reproducibility was excellent (ICC = 0.976; κ = 0.904). GUS had a higher completion rate than GES (99.2% vs. 93.4%, p = 0.039). GUS_15min was more strongly associated with total GCSI score than GES_240min (r = -0.52 vs. -0.29; p = 0.0001).

A GUS_15min threshold < 20.0% showed good validation performance, excellent reproducibility, and higher completion than GES. GUS may provide a useful bedside screening approach for suspected gastroparesis and help identify patients who should proceed to GES.