The Importance of Region-of-Interest Placement in Gastric Emptying Studies with Small Bowel Activity at Time 0

Gastric emptying is assessed in nuclear medicine by having patients consume a standardized radioactive meal, followed by imaging after meal ingestion (time 0) and at each hour up to 4 h. Usually, all of the meal is in the stomach at time 0. However, in some patients, a portion is in the small bowel at time 0. This study evaluates the effects of region-of-interest (ROI) placement on gastric retention percentages when meal is present in small bowel at time 0. Methods: In 624 patients who underwent gastric emptying scintigraphy, images were examined retrospectively for small bowel activity at time 0 (small bowel sign). We analyzed gastric retention by comparing time 0 ROIs drawn around the stomach and those drawn around the stomach plus small bowel. ROIs were drawn around the stomach at 1-4 h after meal ingestion. An equal number of age- and sex-matched patients without the small bowel sign were evaluated. Qualitative assessment was used to determine whether most counts were in the gastric fundus or antrum at time 0. Results: The small bowel sign was present in 8.3% of patients (52/624). Over 1-4 h, the mean initial ROIs without bowel analysis were significantly higher than those with bowel analysis (38% ± 36% vs. 30% ± 30%, P < 0.0001). A normal gastric retention value of no more than 10% at 4 h was found in 94.2% of patients (49/52) using the initial ROI without bowel analysis and in 100% of patients (52/52) using the initial ROI with bowel analysis. Rapid gastric transit (retention of ≤30% at 1 h) was seen in 3.8% of cases (2/52) using the initial ROI without bowel analysis and in 9.6% of cases (5/52) using the initial ROI with bowel analysis (P not significant). At time 0, most counts were in the antrum in 33% of patients with the small bowel sign versus 8% in patients without the small bowel sign (P = 0.001). Conclusion: In patients with a portion of the meal in the small bowel on time 0 images, the exclusion of small bowel counts on ROI analysis can lead to spuriously high 1- and 4-h gastric retention values. This can lead to underdiagnosis of rapid gastric transit at 1 h and overdiagnosis of delayed gastric emptying at 4 h.