Ultrasound evaluation of gastric emptying and comparison with patient-reported risk factors in elective surgical patients

Abstract

Background/Objectives: Despite standard preoperative fasting guidelines, residual gastric content can persist in some patients, increasing the risk of aspiration pneumonitis. Multiple patient-specific factors may predict gastric content retention, but their predictive accuracy is limited. We hypothesized that ultrasound would more reliably identify residual gastric content compared to a comprehensive questionnaire and aimed to determine the most practical approach for risk assessment in elective surgical patients. Methods: We conducted a prospective, observational study in adult patients scheduled for elective surgery at a single center. All participants adhered to an 8 h fasting period. The primary outcome was the incidence of a “full stomach” on ultrasound. Secondary outcomes included the Perlas risk classification, comparisons of established volume estimation formulas, correlations with self-reported fasting duration, agreement between questionnaire-based predictions and ultrasound findings, and the time efficiency of each method. Multivariable logistic regression and Cohen’s kappa were used for analyses. Results: Data from 404 patients were analyzed. Despite prolonged fasting, 16.3% had a full stomach by ultrasound, suggesting incomplete gastric emptying. Early satiety and cholelithiasis significantly predicted a full stomach; prolonged fasting duration and female sex were protective. Questionnaire-based predictions demonstrated fair agreement with ultrasound (kappa = 0.327). The Michiko formula often yielded negative volumes, highlighting limitations in volume estimations. Ultrasound examination (3 min median) was faster than questionnaire completion (5 min). Conclusions: Ultrasound accurately detects residual gastric content, outperforming questionnaire-based assessments. Integrating it into routine preoperative evaluation may improve patient safety, although research is needed to refine volume estimation formulas and expand feasibility for patients with positioning limitations.