Interrater reliability of qualitative ultrasound assessment of gastric content in the third trimester of pregnancy.

BACKGROUND Pulmonary aspiration of gastric contents in pregnant women undergoing general anaesthesia is one of the most feared complications in obstetric anaesthesia. Bedside gastric ultrasonography is a feasible imaging tool to assess the gastric content. The purpose of this study was to investigate the reliability of qualitative bedside assessment of the gastric content performed by anaesthesiologists on third trimester pregnant women. METHODS Pregnant women (≥32 weeks gestational age) were randomized to undergo ultrasound (US) assessments of their stomach in a fasting state (>8 h), or after ingestion of clear fluids only, or solid food. Three anaesthesiologists trained in gastric ultrasonography performed the assessments using a low-frequency curved-array US transducer (5-2 MHz). Primary outcome of the study was the consistency of raters in diagnosing the correct status of the gastric content, which was used to determine the interrater reliability among the three anaesthesiologists. Secondary outcomes were overall proportion of correct and incorrect diagnoses and the specific proportions of correct diagnosis across the three gastric content groups. RESULTS We analysed 32 pregnant women. The interrater reliability displayed a kappa statistic of 0.74 (bias corrected 95% CI: 0.68-0.84). The overall proportion of correct diagnosis was 87.5% (84 of 96). The odds of correct diagnosis for 'solid contents' were 16.7 times the odds for 'empty', and 14.3 times for 'clear fluid'. CONCLUSIONS Our results show the consistency of the qualitative US assessment of gastric contents of pregnant women in the third trimester by anaesthesiologists. A kappa of 0.74 suggests substantial agreement in terms of interrater reliability for this diagnostic measurement. CLINICAL TRIAL REGISTRATION ClinicalTrials.gov identifier: NCT01564030.