Measuring ethanol in saliva with the QED enzymatic test device: comparison of results with blood- and breath-alcohol concentrations.

An enzymatic test device, QED (quantitative ethanol detector), intended for on-the-spot analysis of ethanol not equal to saliva was evaluated in laboratory experiments. Tests were made in vitro with known-strength aqueous ethanol standards and also in vivo after healthy volunteers drank a moderate dose of alcohol (0.80 g/kg), either on an empty stomach or after breakfast. Acetone, methanol, 2-butanone, and ethylene glycol did not react with the QED when tested in vitro. However, n-propanol (100 mg/dL) gave an apparent ethanol response of 60 mg/dL; the secondary alcohol, 2-propanol (100 mg/dL), gave a response of 20 mg/dL, but a longer time was needed to reach an endpoint. When ethanol solutions of 50, 100, and 140 mg/dL were analyzed, the QED gave readings of 50 +/- 2 (+/- standard deviation [SD]), 102 +/- 3, and 136 +/- 5, corresponding to recoveries of 100, 102, and 97%, respectively. The analytical precision, expressed as coefficients of variation, ranged from 3.2-4.0%. The mean difference between the QED test result and venous blood ethanol was 4.4 +/- 0.91 mg/dL (+/- standard error [SE]), and the 95% limits of agreement (LOA) were -18 and 26 mg/dL. The QED saliva test also compared well with breath-alcohol readings; the mean difference was 3.3 +/- 0.71 mg/dL, and the 95% LOA were now slightly narrower: -12 and 19 mg/dL. The QED device proved useful for quickly analyzing alcohol in saliva, and the results agreed well with the concentrations of ethanol in venous blood and end-expired breath.