Observations on Interpretation of Blood Alcohol Levels Derived from Analysis of Urine

Abstract

The fact that gas chromatography when combined with an internal standard and an integrator will measure the alcohol content in minute samples of body fluids rapidly and accurately has been confirmed. The analysis of blood and urine samples from 35 motorists suspected of driving while under the influence of drink gave a mean urine to blood ratio of 1.44:1, substantially higher than the recommended conversion factor of 1.33:1. Moreover the range of values found—from 1.10 to 2.44:1—suggests that the uncritical acceptance of this urine to blood ratio of 1.33:1 will lead to miscarriages of justice when it becomes an offence to drive or to be in charge of a motor vehicle with more than a stipulated amount of alcohol in the blood. The failure to obtain good agreement between capillary blood samples and the corresponding venous samples when analysed for alcohol indicates that it would be unwise to depend on capillary blood analysis for accurate results at the present moment. Probably the discrepancies are due to sample- handling difficulties, which may be overcome with more experience. The analysis of five samples of human blood showed that when alcohol is present the plasma contains over 40% more than whole blood (plasma to R.B.C. ratio, 1.43:1 ; plasma to whole blood ratio, 1.15:1). Thus the alcohol content of a given blood sample will be critically dependent on the separation or otherwise of plasma from red blood cells during handling. © 1967, British Medical Journal Publishing Group. All rights reserved.