Diagnostic performance of ultrasound in the assessment of gastric contents: a meta-analysis and systematic review

Abstract

Objective To systematically analyze the accuracy of ultrasonic techniques in assessing the nature of gastric contents and their volume. Methods English-language articles that used ultrasonic techniques to assess the nature of gastric contents and their volume in patients were selected. In eligible studies, data were recalculated and analyzed for forest plots and subject summary curves of operating characteristics (SROC). Study quality was assessed using the diagnostic accuracy study quality assessment tool QUADAS-2. Publication bias was tested using funnel plots. Results Nine articles with a total of 523 study subjects were identified for this review. All studies were feasibility studies. The sensitivity of ultrasound assessment of gastric contents ranged from 53 to 100% and the specificity from 48 to 99%. The combined analysis yielded an area under the working characteristic curve for subjects of 97% (95% confidence interval (CI), 95–98%), a sensitivity of 95% (95% CI, 84–99%), and a specificity of 88% (95% CI, 72–95%). There was a high degree of heterogeneity among the studies due to inter-operator differences and small sample sizes. Conclusion Ultrasound techniques have high diagnostic accuracy in assessing the nature of gastric contents and their volume in patients. However, most of the studies were feasibility studies with small sample sizes, lacked standardization, and had high risk of bias. More studies are needed in the future to investigate the diagnostic performance of gastric ultrasound assessment techniques. Critical relevance statement Ultrasonography can be used to assess gastric contents, but standardized data integration and reporting are needed to account for the diagnostic capabilities of this technology. Key points • Ultrasound is a safe and feasible tool for assessing gastric contents. • Ultrasound has good diagnostic performance for gastric contents. • There is still a certain heterogeneity within our analysis process; more research is needed in the future to improve our results. Graphical Abstract