Résumé
Acute alcohol intoxication is a frequent cause of emergency department (ED) visits. Evaluating a patient’s alcohol intoxication is commonly based on both a physical examination and determination of blood alcohol concentration (BAC).
To demonstrate the feasibility and usefulness of using a last-generation infrared breath analyzer as a non-invasive and rapid screening tool for alcohol intoxication in the ED.
Adult patients suspected of acute alcohol intoxication were prospectively enrolled over 10 days. Breath alcohol concentrations (BrAC) were measured using a handheld infrared breath analyzer. BAC was determined simultaneously by automated enzymatic analysis of a venous blood sample. The relationship between BAC and BrAC values was examined by both linear regression and Bland-Altman analysis.
The study included 54 patients (mean age 40±14 years, sex ratio M/F of 3/1). Breath and blood alcohol concentrations ranged from 0 to 1.44mg/L and from 0 to 4.40g/L (0–440mg/dL), respectively. The mean individual BAC/BrAC ratio was 2615±387, 95% confidence interval 2509–2714, which is 30% higher than the legal ratio in France (2000). The correlation between both measurements was excellent: r=0.95 (0.92–0.97). Linear regression revealed BAC=0.026+1.29 (BrAC×2000) and BAC=0.026+0.99 (BrAC×2615). Mean BAC-BrAC differences and limits of agreement were 0.49 g/L [−0.35, 1.34] (or 49 mg/dL [−35, 134] and 0.01 g/L [−0.68, 0.71] (or 1 mg/dL [−68, 71]), for the 2000 and 2615 ratios, respectively.
The calculated conversion coefficient provided a satisfactory determination of blood alcohol concentration. Breath alcohol testing, using appropriate BAC/BrAC conversion, different from the legal BAC/BrAC, could be a reliable alternative for routine screening and management of alcohol intoxication in the ED.