Correlation between simultaneous scintigraphic and ultrasonographic measurement of gastric emptying in patients with type 1 diabetes mellitus.

Abstract

To compare scintigraphic measurements of total stomach emptying of a semisolid meal with ultrasonographic measurements of changes in antral area as estimates of antral emptying in type 1 diabetic patients. Eleven patients with insulin-dependent diabetes mellitus were studied with simultaneous measurements of gastric emptying by scintigraphy and ultrasonography. Patients were imaged immediately after ingestion (time 0) and every 15 minutes over 120 minutes. The gastric emptying rate was expressed as percent reduction in antral cross-sectional area from 15 to 90 minutes after meal ingestion. Ultrasonographic measurements showed a postprandial maximal antral area at 15 minutes, continuously decreasing with time, and reaching a plateau 45 to 90 minutes after the end of meal ingestion, whereas the scintigraphic counts attained their maximum immediately after the meal and began to fall thereafter. Between 15 and 90 minutes, the residual radioactivity and antral ultrasonographically measured distension curves were concordant. The curves then showed a tendency toward deviation for the last 15 minutes (median, 51% versus 59% at 105 minutes and 40% versus 57% at 120 minutes, respectively). A strong significant correlation could be seen between the ultrasonographic gastric emptying rate and scintigraphic half-time values (r = -0.94; P < .001). Comparing scintigraphic and ultrasonographic half-time values showed a systematic measurement error of 9.9 minutes and a random measurement error of 18.6 minutes. The use of standardized real-time ultrasonography to determine the gastric emptying rate of semisolid meals in diabetic patients, with the use of the change in gastric antral cross-sectional area in a single section of the stomach 15 and 90 minutes postprandially, offers a valid method for clinical practice.