Breath Alcohol Concentration and Breath Temperature

Breath alcohol concentration, breath temperature, and body temperature are used to extrapolate the breath alcohol concentration to the concentration of alcohol in the deep lung air, which is in equilibrium with the pulmonary blood. Introduction The variability that is seen in the breath alcohol test (note: “alcohol” means ethyl alcohol) has always been more or less ignored as inevitable “biological variability”. It has been known from the beginning of the use of breath test for numerical evidence of drunk driving that breath temperature plays a part in the test result. With the intention of reducing breath test variability, authorities in Germany and Alabama have recently begun extrapolation of the test result from the measured breath temperature to a standard breath temperature, 34C. The procedure is described by Schoknect and Stock (1) and is based on Henry’s Law. Extrapolation to a standard temperature does not take into account pulmonary blood temperature, which varies from person to person and with time of day and with it, breath alcohol concentration. Therefore, extrapolation to body temperature, the procedure described below, should further reduce the variability of the breath test as well as to make the BrAC more equitable with the BAC. The procedure assumes equilibrium between pulmonary blood and breath in the deep lung region; but in the upper lung region only ‘near equilibrium’ is assumed between the breath and the water in the mucous surface that lines the airways. Recent Work In recent years, several workers have investigated the process leading to a breath alcohol measurement. Lubkin et al (2) have developed a model of the breath alcohol profile, but it does not have a temperature component. Curiously, Lefranc and Montamat (3) concluded that the breath test is not influenced by breath temperature, although they did not measure it. They noticed the effect of hyperventilation on breath alcohol but ascribed it to “a temporary thinning of the arterial blood in the alveoli”. Tsu et al (4) have developed a detailed model of the interaction of alcohol with the lung surfaces that has Brass tube Breath thermistor Control thermistor Breath hose Mouth pc been tested by George et al (5). It predicts alcohol profiles, but is too complex for police use. Methods The inlet end of the breath hose of a National Patent Analytical Systems, Inc. Datamaster infrared type police evidential breath alcohol tester was fitted with a thermostated (at 34C) fast response gas thermistor temperature probe assembly. The thermistor was calibrated by passing air saturated at 37C through a 10 foot long 3/8 inch diameter copper tube coiled into a water bath thermostated at temperatures from 30C to 36C. Breath alcohol concentration calibration data was obtained immediately before and after each human subject test run using a Repco Marketing simulator operating at 34C to provide reference alcohol-in-air samples. The flow thermistor was calibrated by use of a Warren Collins, Inc. spirometer connected to the Datamaster sample vents. Compressed air saturated at 34C was used for this calibration. Oral temperatures to the nearest tenth of a degree Fahrenheit using a digital oral thermometer were obtained from each of 30 drinking human subjects just before being tested with the apparatus. All temperature sensors used were checked with a NIST traceable digital thermometer. The output of the breath temperature thermistor, and the alcohol and flow outputs of the Datamaster, were fed to a data logger, and then to a computer for spreadsheet analysis. The breath temperature probe is shown in Figure 1. Four types of breath sample were obtained from each subject in the following order: • A normal police type sample where the subject takes a deep breath before blowing fully into the Datamaster. • A hyperventilated sample where the subject inhales deeply then fully exhales. This is repeated ten times before finally inhaling and blowing into the Datamaster as above. The purpose of this maneuver is to aggressively cool the airway surfaces of the lung. • A hypoventilated sample where the subject takes a deep breath and holds it for 20 seconds before blowing into the Datamaster. The purpose of this maneuver is to allow the airway surfaces to warm. • A re-breathed sample where the subject inhales deeply, then blows his breath fully into an unheated 4 liter plastic bag, then re-inhales the same air from the bag, all the while keeping his nostrils pinched shut to prevent fresh air from entering his lungs. After the 7 inhale, the subject blows into the Datamaster. The purpose of this maneuver is to stabilize the interaction of alcohol vapor in the breath with the airway surfaces. Results The relationship of BrAC (g alcohol/210L air) to breath temperature of a breath sample Figure 1. Breath temperature probe can be seen in Figure 2a and Figure 2b and are typical of the results obtained. Breath flow rate data where used to calculate breath volume. The profiles are from a single subject and were obtained within a period of less than 10 minutes. Comparison of the alcohol and the temperature profiles clearly shows the effect of cooling the lung airway surfaces on the breath alcohol concentration. The sharp rise seen in the alcohol profiles over the first 150 ml or so corresponds the displacement of air in the Datamaster plus air from the conducting airway volume (from the mouth to the bronchi of the subject). The slower increase that follows corresponds to air from the lungs. The order of the tests had a small influence in the test result, indicating that it takes a few minutes for the lung surfaces to stabilize after the previous blow. It is surprising that the temperature of hypo samples does not rise to deep lung temperature, assumed to be equal to oral temperature. This may reflect the disturbance of heat balance at the mucous water/air interface due to O2/CO2/water condensation/evaporation during breath holding. Re-breathed samples showed similar behavior. Re-breathed samples were obtained specifically for use as a surrogate for the pulmonary blood alcohol concentration, pBAC (g/100ml blood). Following Harger’s work (6) to determine a more stable relationship between BrAC and (venous) BAC, an earlier study in this laboratory (7) involving 30 subjects who had passed into the post-absorption period (at least 2 hours after drinking had stopped) yielded the following empirical relationship between re-breathed BrAC and BAC (at this stage in the post absorption period, BAC=pBAC): Eq. 1 pBAC = 0.9573BrAC(re-breathed) 0.004 for which R = 0.9844, and the standard error for predicted pBAC = 0.004. Since one cannot be certain whether or not the subject is fully post-absorptive unless several hours have elapsed after drinking had stopped, which was not the case in the present study, re-breathed BrAC was used instead of BAC in the calculations below. Fingertip blood, which is equivalent to arterial blood and hence to pulmonary blood, could have been used but this requires specialized equipment and techniques that were not available to this study. With the assumption that Henry’s law holds at the mucous water/air interface in the upper lung to a close enough approximation, an extrapolated deep lung alcohol concentration 0.000 0.020 0.040 0.060 0.080 0.100