**Gas Detection Instruments and Sensors, Drägerwerk AG, Abt. ESG, Moislinger Allee 53155, Lübeck 23542, Germany
The technical requirements for future evidential breath testing in Germany have been published recently in a draft of the german standard DIN/VDE 0405. The requirements are based on a study of G. Schoknecht, Federal Health Office Berlin 1992, and were afterwards formed into a standard under consideration of the existing OIML specifications for evidential breath tester. In addition to the OIML specifications several requirements had to be added to achieve legal and technical acceptance. The most important are: two consecutive breath tests for mouth alcohol detection; measurement of the endexpiratorial breath temperature to compensate for changes in breath alcohol concentration due to the subject's body temperature and breathing technique; a sex and age dependent minimum blowing volume; and verification of the calibration by including two independent measuring systems with mutual control.
Up to now no available instruments fulfil these requirements. We present a detailed description of a breathtesting instrument comprising two independent analytical systems (infrared and electrochemical) and means for breath temperature measurement. Furthermore, results from field and laboratory tests are given. Our data show a very good correlation between blood and breath alcohol concentration (corr. coef. 0.99) and show that breath temperature measurement considerably improves the performance of an evidential breath tester.
The technical requirements for future evidential breath testing in Germany have been published recently as a draft German standard DIN/VDE 0405 (1994). The requirements are based on an expertise of the Federal Health Office Berlin (Schoknecht, 1992a), and the preliminary draft of OIML recommendations (1992) for Evidential Breath Testers (EBT) as far as they concern the basic reliability of the instruments. But with respect to the assessment of physiological influences on breath alcohol analysis the expertise revealed that several requirements had to be added to the OIML recommendations to achieve legal and technical acceptance of the German authorities. Investigations have shown that breath temperature (Schoknecht, 1992b) and the expired volume (Schoknecht, et al; 1990) have a markable influence on the results. Breath alcohol concentration (BrAC), breath temperature and flow of volume are the most relevant parameters , which continuously vary during the dynamic process of the breath alcohol analysis. Measurement of the endexpiratorial breath temperature allows to compensate for changes in breath alcohol concentration due to the subject's body temperature and breathing techniques like hyper- and hypoventilation (Schoknecht, 1992b). This is achieved by referring the measured BrAC to a reference temperature of 34°C. The coefficient of temperature for this calculation amounts to 6.59%/°C according to our own experiments and complies very well with the data found by Dubowski (1980).
For a valid breath alcohol analysis a minimum expired volume is necessary to ensure that the measured BrAC corresponds to the alcohol concentration of deep lung air (alveolar air). Since the vital capacity of a person depends on sex and age a fixed minimum exhaled volume appears to be not sufficient as a validation criterion. From an epidemiological field study (including 4000 persons) a list of sex and age dependent minimum blowing volumina has been derived to assure a deviation of less than 10% rel. between the BrAC taken at the minimum volume and the endexpiratorial BrAC (Table 1).
Required Minimum Blowing Volumina for Male and Female Subjects
According to the OIML recommendations for EBTs two expirations per operation cycle are required. The results of both measurements have to be within prescribed limits. The requirement has been extended in a way that the two breath tests must be analysed by two independent measuring systems with mutual control of the results. Each system comprises means for gas analysis, breath temperature measurement and volume control and has to comply with the basic OIML specifications. Furthermore the two systems have to be of different analytical specificity, i.e. only for ethanol in exhaled air the results are equal; if interfering substances are present this equality is disturbed and an error is indicated. A waiting time of at least 2 minutes between the two expirations ensures a reliable detection of mouth alcohol and other substances with short time influences like e.g. mouthfreshners. By this procedure two independent and valid results are obtained and the average of both is given as the valid BrAC in terms of mg/L.
The first realisation of a breath testing device according to the aforementioned specifications was based on a regular infrared unit (IR) Dräger Alcotest 7110 to which an electro-chemical (EC) sensor has been added as the second independent system. From a look-up table the minimum blowing volume is set according to Table 1 after the subject's personal data have been entered via the keyboard. The combination of an IR unit and an EC sensor offers the advantage that the breath sample is analysed by two substantially different methods of ethanol detection. This approach is very similar to blood alcohol analysis in Germany and increases public and legal acceptance. Furthermore, the different analytical specificities of the two measuring systems ensure excellent selectivity for ethanol while guarding against the presence of potentially interfering endogenous substances. Many volatile compounds have negligible small influence on the response of the EC sensor but cause a strong absorption of infrared light like for instance toluene and ethylacetat, solvents often used in paints and glue. Figure 1 shows a block diagram of the combined systems. The EC sensor together with a small sampling pump is piggybacked to the IR cuvette, a multi-reflection cuvette operating at a wavelength of 9.5µm. A double hot filament sensor, located close to the inlet orifice of the IR chamber, measures the incoming sample flow from which the exhaled volume is calculated via flow integration.
Block Diagram of the Two Measuring Systems
The sensor for breath temperature measurement, as shown in Figure 2, consists of a massive piece of brass including means for taking the mouthpiece at one end. The other end is connected to the sample hose of the instrument. Two small thermistors are mounted in the breath passageway close to the outlet orifice of the mouthpiece . To avoid condensation inside and to warm up the mouthpiece the whole sensor is heated to approx. 37°C. Warming up the mouthpiece before the test turned out to be advantageous to keep its influence on the breath temperature as small as possible. Our experiments revealed an uncertainty of the breath temperature measurement of less than 0.5°C, including the influence of the mouthpiece.
Sensor for Breath Temperature Measurement
Six instruments were passed to the police authorities at various locations in Germany for a field test starting in Sept. 93 and lasting for 15 months. Before the test the instruments were adjusted to 0.48 mg/L at a liquid standard with an ethanol concentration of 1.21g/L H2O held at 34°C. Calibration checks were performed regularly every six month. The instruments were mainly operated at police stations but two of them were also tested in mobile use for several weeks. More than 700 tests have been successfully performed during the testing period where for approx. 300 tests additional blood samples were taken. The comparison between the blood alcohol concentration (BAC) and the corresponding BrAC is shown in Figure 3. No corrections have been made for the time delay (a few minutes up to two hours) between breath and blood alcohol analysis because in general it is not known whether the subject is in the absorption or desorption phase. Despite this fact, the data show a very good agreement with a correlation coefficient of 0.98 and nearly all of the data points are within a ±10% (rel.) distribution centered around the also plotted regression line. No outliers are observed. The regression line intercepts the BrAC axis at 0.02 mg/l which we attribute to an average time delay of 15 min. between BAC and BrAC analysis. From our data we calculate a BAC versus BrAC ratio of 2090. Where the ratio is only 1960 if breath temperature correction is not taken into account.
BrAC vs. their Corresponding BAC Values. The Linear Regression and the ±10% Range are Indicated by the Solid Lines
The distribution of breath temperatures, obtained during the course of the test for stationary (triangles) use and mobile (circles) operation, is illustrated in Figure 4. As seen from this Figure, most of the subjects had breath temperatures within a span of ±1°C relative to the mean breath temperature of 35°C. Considerable lower breath temperatures down to 30.9°C were found for roadside testing during winter time. The difference of 5.8°C between the lowest (30.9°C) and highest value (36.7°C) corresponds to a relative change in BrAC of nearly 40%. Even for purely stationary use we found confirmatory evidence for a correlation between the breath temperature and the ambient temperature. For clarification Figure 4 also shows the averaged breath temperature (solid upper line) together with the averaged ambient minimum night temperature (lower curve, right scale). During cold days in winter with outside temperatures below 0°C the average breath temperature decreases down to 34.5°C while during warm periods in summer it climbs up to 35.5°C.
Distribution of Breath Temperatures Obtained for Stationary (triangles) and Mobile (circles) Use. The Upper Solid Line Indicates the Averaged Breath Temperature, the Lower Curve the Averaged Minimum Night Temperatures
To give an impression for the reliability of the instruments Figure 5 shows the difference between the results of the first breath sample, analysed by the IR system, and of the second sample, analysed by the EC system. A further data analysis revealed that approx. 50% of the total difference is due to physiological variations between the two breath samples. The rest can be attributed to the statistical and systematic errors of the measuring systems. The solid lines indicate the prescribed limits (±10% in relative value or 0.040mg/L, whichever is larger, accord. to DIN/VDE 0405;1994), which must not be exceeded during the measuring procedure. As illustrated by this Figure the majority of the results is clearly within the permitted range. Only a few of them come close to the limits. During the testing period only one of the 700 tests failed, because the difference of the results between the two breath samples was too large.
Difference Between the Results of the First (IR) and Second Breath (EC) Sample Analysis. The Permitted Range is Indicted by the Solid Lines
Our results have shown that breath testing instruments according to the specifications listed in DIN/VDE 0405 (1994) provide a precise and reliable breath alcohol analysis. The instruments also satisfy the requirements for handling and stability in daily use. The main reason for inadequate attempts was that approx. 15% of subjects were not able or not willing to provide the required blowing volume. We believe that this percentage will decrease considerably after evidential breath testing has been introduced by law. The subjects are then faced with the fact to provide a blood sample in case of a non-performed breath test (Due to the present legal situation in Germany the instruments were only used as pretest devices during the field test). In addition our data clearly point out in accordance with (Schoknecht, 1992a) that breath temperature measurement improves the performance of breath alcohol analysis with respect to an equal treatment of the subjects. The data strongly support the OIML recommendation that breath temperature has to be taken into account for evidential roadside testing. But perhaps as the most important result the field trial revealed that the problem of BAC versus BrAC outliers has been overcome. Because of these outliers, which have been reported occasionally from pretest devices (Wilske, 1992), breath alcohol analysis in general has been often blamed to be unreliable and not to meet the necessary requirements of evidential purposes for law enforcement in Germany.
DIN/VDE 0405: Bestimmung der Atemalkoholkonzentration, Entwurf Feb. 1994.
Dubowski, K.M.: Breath-alcohol testing: disposable breath tester, Part 1, Dec. 1980, Nat. Technical Information Service, Springfield, Virginia 22, 162 (1980).
OIML: Third preliminary draft of an international recommendation relating to evidential breath analysers (1992).
Schoknecht, G: Beweissicherheit der Atemalkoholanalyse; Gutachten des Bundesgesundheitsamtes. Unfall- und Sicherheitsforschung Straßenverkehr, Heft 86 (1992a).
Schoknecht, G: The influence of temperature on breath-alcohol analysis, Proceedings of the ICADTS T-92, 1992b.
Schoknecht, K. Fleck, B. Kophamel : Einfluß des Atemvolumes auf die Atemalkoholanalyse. Blutalkohol 26, 83-94 (1990).
Wilske, J: Problems with outliers in breath alcohol testing, Proceedings of the ICADTS T- 92, 1992.
Return to Contents
Return to T'95 Home Page