Single photon emission computed tomography for gastric volume assessment: A method with observer-defined regions of interest

Abstract

Single photon emission computed tomography (SPECT) imaging allows non-invasive measurement of gastric volume. In previous studies, the processing of the SPECT data involved global threshold algorithms that do not take into account the non-uniform distribution of radioactivity in the gastric wall. The aim of this study was to develop a simple alternative method based on observer-defined regions of interest. A phantom study was performed to standardize volume calculations from SPECT derived cross-sectional areas. In 12 healthy volunteers, the principle was then used to determine gastric volume before and after a 600 ml liquid meal. Furthermore, gastric emptying of the meal was followed with planar scintigraphy. The median volume of the stomach was 86 ml (range 62-130 ml) at baseline, 642 ml (536-748 ml) immediately after the meal, and 370 ml (221-481 ml) 1 h after the meal. The coefficient of variation for the calculations was 9%, 2% and 4%, respectively. The median increase in gastric volume was 562 ml (501-628 ml) immediately after the meal and 294 ml (159-370 ml) after 1 h. Gastric retention of the meal was 68% (50-73%) after 0.5 h and 51% (39-57%) after 1 h. The present manual technique may be a reliable alternative to the automated SPECT methods for assessing gastric volume. The liquid meal that was used in our study did not seem to cause an increase in gastric volume that differed from the volume of the meal.