Capnography in diabetic ketoacidosis (DKA) diagnosis: a systematic review and meta-analysis

The study investigates the role of capnography as an auxiliary diagnostic tool for diagnosing diabetic ketoacidosis (DKA) in order to reduce treatment delays and associated morbidities.

A registered protocol followed PRISMA guidelines and included all diabetic patients undergoing capnography, excluding those with metabolic disturbances unrelated to DKA. A comprehensive review of electronic medical resources was performed, with quality assessment conducted using the Newcastle-Ottawa Scale (NOS). Heterogeneity was expressed using Cochran's Q statistic, along with I² and tau-squared (τ²) statistics. Potential publication bias was evaluated through the Egger test and funnel plots, utilizing Comprehensive Meta-Analysis (CMA) and Stata/MP 17 for the analyses.

Seven studies, including two cohort and five cross-sectional, were analyzed. Among the participants, 322 (28.67%) were diagnosed with DKA, while 801 (71.33%) were not. The main outcome assessed was DKA diagnosis using capnography. The analysis of studies has revealed a statistically significant difference in EtCO₂ levels (standardized mean difference: -1.80, 95% CI: -2.33 to -1.28, p < 0.001), the pooled EtCO₂ odds ratio (OR: 0.03, 95% CI: 0.01-0.09, p < 0.001), and a pooled standardized mean difference (SMD) (-1.80, 95% CI: -2.33 to -1.28, p < 0.001) between DKA and non-DKA patients. The pooled sensitivity of EtCO₂ for DKA was 0.91 (95%CI: 0.87, 0.94), and the pooled specificity was 0.90 (95%CI: 0.81, 0.94).

The measurement of EtCO₂ by capnography, a non-invasive and readily available tool, may facilitate the expedited diagnosis of DKA in diabetic patients, pending further validation in RCTs and diverse settings.