Capnography as a tool for triaging and diagnosis of diabetic ketoacidosis in the emergency department: A prospective observational study

Objectives: The cornerstone of management of acidosis in a patient with diabetic ketoacidosis (DKA) has traditionally been carried out by blood gas analysis, which is expensive and associated with significant risk. It is against this background that the correlation between end-tidal carbon dioxide (EtCO₂), blood pH, and EtCO₂ bicarbonate levels was analyzed. The predictive value of EtCO₂ was also analyzed in the diagnosis of DKA. Finally, we aimed to determine the value of EtCO₂ as a screening test for the exclusion of DKA.

Materials and methods: This was a prospective cohort study carried out in the emergency department of a tertiary care teaching hospital from September 2020 to September 2021. Patients with suspected DKA underwent simultaneous blood gas collection and EtCO₂ analysis.

Results: A total of 123 patients with blood sugar levels >250 mg/dl and moderate-to-large (≥2+) urine ketones were studied. A cut-off value of EtCO₂ ≤24 was determined to rule in DKA with a sensitivity of 93.02% and specificity of 91.9%. EtCO₂ >26 could effectively rule out the diagnosis of DKA with sensitivity of 98.8% and specificity of 75.7%. A significant linear correlation between pH and EtCO₂ (P < 0.0001, r = 0.82) and HCO3 and EtCO₂ (r = 0.896, P < 0.0001) was found.

Conclusions: EtCO₂ values ≤24 can accurately identify patients with DKA in the presence of elevated blood sugar and urinary ketones and must be considered a valuable addition to the diagnostic criteria. EtCO₂ values >26 can be an effective triaging tool for ruling our DKA. A significant linear correlation between pH and EtCO₂ and pH and HCO3 was observed. EtCO₂ can be considered a surrogate marker for the degree of response to the treatment in DKA.