|
|
 |
 |
 |
DWI
August 2009, Page 44
DWI
By Wayne A. Morris
A New Look at Breath Alcohol Testing
Instrumental breath alcohol concentration (BrAC) testing has been in wide use in the United States since 1954 with the introduction of the Borkenstein Breathalyzer.1 Rolla N. Harger conducted the first-ever “short course” on chemical tests for intoxication in 1937 and in 1938 introduced the Drunkometer, the first practical instrument for testing breath alcohol. It was the Breathalyzer invented by Dr. Robert F. Borkenstein, however, that was the most popular of the early instruments for testing breath alcohol.2 As an alternate method to blood alcohol concentration (BAC) testing, it is currently used to prove that a person was driving with a BrAC above the legal limit. Originally, BrAC testing instruments reported the results in terms of grams of ethyl alcohol per 100 milliliters of blood, i.e., the units of blood alcohol concentration. The current state of the art breath alcohol testing of human subjects uses infrared spectroscopy to quantify the contents of a flow cell that opens to the ambient air through an exit portal.
The Blood-Breath Ratio
This use of breath testing to measure BAC was based upon what was then known of the relationship between the weight of ethyl alcohol in the blood to the weight of ethyl alcohol in the breath. It should be noted for those not familiar with the blood-breath ratio (BBR) that this ratio means that at 2100:1, for example, one milliliter of blood has the same weight of ethanol as 2100 milliliters of breath. Each country has adopted by fiat its own value for this ratio, such as 2000:1 in Austria,3 2300:1 in the United Kingdom and the Netherlands,4 and 2100:1 in the United States.5 The use of grams per 210 liters as the units of BrAC currently used in the United States arises from the fact that BAC is measured in grams per 100 milliliters of blood. Since the volume of breath containing the same weight of ethyl alcohol as contained in100 milliliters of blood is 2100 times greater, the breath volume at equality would be 210,000 milliliters or 210 liters. This change of units was predicated by the arguments raised by defense attorneys as to the range of the blood-breath ratio seen in the general population and was based upon a suggestion by Kurt Dubowski that the legal limit be expressed as both a BAC and BrAC level.6 The fact that both legal limits have the same numerical value is based upon the tacit assumption that the BAC and BrAC values are equal at the 2100:1 ratio.
Over the years, defense attorneys and scientific experts have taken issue with perceived shortcomings of breath alcohol concentration testing (BrACT), including the lack of scientific basis for the rules of breath testing, the misuse of instrumental techniques, and the misapplication or misinterpretation of scientific principles, theories, and observations.7 This article is an effort to re-examine the scientific principles and theories upon which BrACT is based. While several issues are addressed, this is not meant to be a complete review of all the issues, but mainly addresses the blood-breath ratio and the simultaneous blood and breath alcohol correlation.
The measurement of the ethyl alcohol vapor concentrations using simulator samples has shown that the instrumental measurements are accurate and precise. Accuracy refers to how close an observed measurement is to the actual value of the property being measured. For breath alcohol testing, accuracy is shown by the use of an alcohol reference solution (ARS) or dry gas standards having a known alcohol concentration.
The alcohol concentration of the vapor produced in a simulator at a set temperature of 34ºC +/- 1ºC is set by the alcohol concentration of the water solution or by the addition of a known amount of alcohol vapor mixed with air. In either instance, the amount of alcohol added to the liquid or gas will produce the correct standard value of the breath alcohol concentration reported in terms of grams per 2100 milliliters of breath at the temperature of 34ºC within the accuracy parameters of the measurements used in the creation of the standard and within the accuracy parameters of the instrument being used to make the measurements. If the breath alcohol result is reported to three decimal places, the actual concentration of that bottle of ARS used or the vapor concentration of the dry gas standard must be known to three decimal places. If the concentration is reported only to two decimal places, the required accuracy cannot be determined or shown. If the actual concentration of the standard is known to the same number of decimal places as is being reported, then it can be shown that the instrument is giving an accurate representation of the standard concentration within the instrumental accuracy of the instrument. In other words, the instrument is accurately measuring the concentration of the sample that has been introduced into the sample cell of the instrument. Since the sample chamber is connected to the simulator or the dry gas standard bottle at both the inlet and outlet when the measurement of the vapor concentrations arise from the ARS standards or dry gas standards, this concentration is fixed, known and in equilibrium with the liquid after the initial filling of the sample cell with the standard. Thus, the observed breath alcohol concentrations reported are accurate representations of the vapor concentrations produced by the simulator or the dry gas standard that fill the sample cell. As these concentrations will not change, the multiple testing of these standards within the sample cell will also show that the instruments are precise.8
The Accepted Model of Ethanol Expiration
The 1950s model of the pulmonary role in the expiration of ethanol in the breath was based upon a model in which the capillary blood transferred ethanol into the alveolar sacs according to Henry’s Law. Then the incoming breath exchanged with the alcohol-containing alveolar air from which simple diffusion allowed the ethanol vapors to move through the bronchial tubes.9 Forced exhalation brought this alveolar vapor into a mouthpiece and tubing system leading to a sample cell or to a collection device for introduction into a sample cell.10 This model has been supported by the blood-breath correlation studies that have shown a “nearness” of BrAC and BAC values for near simultaneous collected samples. The blood-breath correlation studies have also been used to calculate the apparent blood-breath ratio values for each of the subjects taking part in these studies and the average value of the ratio for the study subjects.
While there is always a wide range for these individual values, the average values calculated for each study group are close to the theoretical values derived from equilibrium studies involving blood-air and water-air systems. This “nearness” of observed values to those obtained in the in vitro equilibrium experiments has been used to confirm this model.11 (While Camps and Robinson12 reported higher breath alcohol concentrations than for the blood alcohol concentrations, the design of the experiment may have actually shown the effect of breath sampling during the absorption phase.) All these studies have been interpreted in light of the then-accepted model of breath testing measuring either alveolar alcohol concentrations or diffused alveolar air concentrations. The observed scattering of values that has been reported in each study has been largely imbued with little significance other than a demonstration of the physiological variations present in the human population, although there have been some expressions of concern.13
The blood-breath correlation study performed by the Florida Department of Law Enforcement in 199214 gave rise to the graph shown in Figure 1. This graph is representative of the blood-breath correlation reported in other studies. The many values obtained showing values of the BrAC lower than those obtained from the corresponding BACT samples have been given a high significance in showing that the selected value of the blood-breath ratio favors the majority of subjects while those BrAC higher than those obtained from the corresponding BrACT samples carry little significance in the interpretation of these results. The lack of a universally accepted average value for the in vivo blood-breath ratio prompted Dubowski to state that if such an average value indeed exists, it should lie between 1900:1 and 2250:1.15 Harger established the in vitro equilibrated blood-air ratio as being approximately 2020:1 at 34ºC and cites several other studies with results near 2000:1.16 Jones17 reported the in vitro equilibrated average blood-air ratio for blood obtained from 20 men to be 2167 +/- 9.6 and for blood obtained from 15 women to be 2195 +/- 10.9 at 34ºC. In this same article, Jones also adjusted the Harger result by 5.4 percent to allow for the “salting out” effect of NaF, resulting in an average value of 2143 at 34ºC.
The significance of the lack of universal agreement upon a single value for the average of the blood-breath ratio and the wide variation observed for individuals in the general population has largely been discounted. When there has been some concern as to the observed scattering and lack of agreement in values, the concerns have generally resulted in slight modifications of the theory (using “substantially alveolar deep lung air” instead of alveolar air), interpretation of the results (changing the units of BrAC from gram/100 mL of blood to grams per 210 L of breath), and the legal use of BrAC testing (changing the statutes to refer to either a BAC or a BrAC value) rather than reconsidering the theoretical basis of breath testing.18
Studies using rebreathed air samples have been conducted in an effort to obtain a better sampling of alveolar breath. While the theoretical in vitro values have been approached using rebreathed air samples,19 the results have been interpreted with the expectation that alveolar or alveolar-like deep lung air is the source of the breath alcohol. BrACT of single end-expired breath samples gives a lower BrAC result than that obtained from rebreathed samples. The explanation for these observations using the above discussed theory in which the source of the breath alcohol is the alveolar alcohol content requires that the mucous lining removes the ethanol from the breath on the single pass and that rebreathing allows a steady-state condition in which the ethanol reaches a higher constant concentration in the breath. This adjustment can be considered equivalent to an unknown loss of sample being replaced by an unknown amount of “contaminant.” However, the overall conclusion of this adjustment must be a BrAC lower than the alveolar content and again, the lack of agreement with the theoretical blood-breath ratio. Alteration of the breathing pattern prior to the giving of a breath sample also affects the BrACT results.20 Gullberg21 has reported the presence of aberrations in breath profile curves and differences between BrAC results for normal breathing and breath holding prior to the breath sampling.
Hlastala’s Model of Ethanol Expiration
The above discussion of the blood-breath ratio based upon the diffusion or measurement of alveolar-like breath does not comport with the current knowledge or understanding of the exhalation of water-soluble gases. The role of the alveolar air in breath testing needs to be re-examined in light of Michael Hlastala’s new paradigm of the model for the exhalation of ethanol in the breath.22 These studies show a more vibrant model of lung respiration.23 The introduction of breath testing instruments that measure the volume of the breath sample introduced into the instrument, such as the DataMaster and the Intoxilyzer 8000, and spirometry measurements of the longer sampling times used in breath testing as compared to spirometry, have raised concerns as to the nature of the breath sample.
From its inception, BrAC studies and testing procedures referred to the breath sample as alveolar air or “deep lung air meaning alveolar air.”24 CMI, the manufacturer of the Intoxilyzer line of breath testing instruments, used the currently accepted terminology “deep lung air” rather than “alveolar air” in its 1991 operator manual. Although it is now routinely accepted that deep lung air is not alveolar air, the blood-breath ratio is still the accepted criteria for relating the BrAC result to the BAC.25 The belief that the longer one gives the breath sample, the closer the BrAC gets to the alveolar air concentration, is still considered valid. Studies of rebreathed air appear to confirm this belief with results that approach the in vitro equilibrium values for the blood-breath ratio.26
The diffusion model, shown in Figure 2, represents the pulmonary pathway as a cylinder with the highest BrAC located at the bottom (the alveolar sac) and lower concentration gradients occurring as the distance from the base of the cylinder (the alveoli) increases. These concentration gradients are produced by simple diffusion of the vapors moving up the pulmonary pathway. The portion labeled as “deep lung air” represents the segment at which a manufacturer’s requirements for a valid or reliable breath alcohol concentration result are first met. The longer a subject takes to deliver the breath sample, the lower down in the cylinder will the final sample be taken and the closer the sample will be to the alveolar alcohol concentration (i.e., the closer the sample will approach the blood-breath ratio). Much of the scattering seen in the blood-breath correlation studies was explained by the height of the sample in the “cylinder,” with those samples nearer to the alveolar air giving rise to the most accurate values for the BrAC.
Advocates of breath testing acknowledge that biological and physiological factors produce the observed scattering in the BBR correlation studies. They also acknowledge that a constant BBR is not feasible when simultaneous blood and breath samples are taken due to these same biological and physiological factors. However, while agreeing with the lack of a reliable BBR, which lowers the reliability of the breath alcohol testing, they counter this loss of reliability by the practical utility of breath alcohol testing including the ease of administering the test and the ability of getting the results at the time of testing. They also want to believe that the loss of reliability does not affect the meaningfulness of the numerical result. Analytical chemistry principles dictate that a lowering of reliability to test results also lowers the meaningfulness of those numerical values. In other words, if two people both have a result of 0.20 grams per 210 liters, there is no assurance that their blood alcohol concentrations and their state of impairment will be equal or nearly equal. Any such inference needs to be based upon accurate and repeatable results, which cannot be obtained thru the current usage of BrACT based upon a BBR that varies with individuals.
The directions given to the breath operator in the CMI manuals prior to 1989 were to have the subject give the breath sample until the tone produced by the pressure sensor stopped. The appearance of a zero in the units place was a visual cue to the operator that the minimum adequate sample had been obtained (that the plateau region of the breath curve had been reached). While this earlier view of a level plateau has changed, it is interesting to note that at least one state agency still utilizes this concept. Figure 3 demonstrates the typical breath curve showing the initial rise of the volume and the level plateau as described in these earlier manuals.27 Figure 4 shows the breath curve obtained from a male subject. Note the small rise for the plateau region. It is similar to those presented in the later CMI manuals that inform the operator that the tone does not stop until the subject stops giving the breath sample.
Consider this brief description of respiration according to lung physiologists.28 The tidal volume is the inspired or expired volume in normal breathing. The expiratory reserve volume is also called the forced expiration volume; it refers to the maximum volume of air that can be exhaled. The residual volume is the volume of air remaining in the lungs after a forced expiration. The alveolar air makes up part of the residual volume.29 The lung volume, with the lung pathway represented using the same cylindrical model used for the diffusion model, is shown in Figure 5. As the subject blows longer into the mouthpiece and the breath volume increases, the expiratory reserve volume is still being measured. It is clear from a comparison of the two models that the deep lung air is the expiratory reserve volume. Alveolar air, in the residual volume, is never measured.
Spirometry is a medical test that measures the bronchial function of the lung. The bronchial air samples correspond to the forced expiratory air portion of the breath profile (the expiratory reserve volume) and alveolar air to the reserve volume. One of the graphs generated by spirometry is that of volume per time of sampling. Figures 6, 7, and 8 are spirograms obtained during testing supervised by the author and conducted by trained operators in a medical office. Spirometry testing utilizes a much larger mouthpiece having a very low resistance to the breath than does breath alcohol testing.
The breath sample in spirometry is quickly provided with a stronger force and lasts for a shorter time than that given in breath testing. As such, the breath profile for breath alcohol samples differs from those for spirometry. Spirograms labeled A in Figure 6 were taken from a healthy male subject giving normal spirometry samples. The initial rise to the plateau region takes approximately 1.1 seconds. Spirogram B in Figure 6 was taken from the same subject who gave a slow and steady breath sample similar to that given as if one blew into the mouthpiece of a breath alcohol tester. The initial rise to the plateau occurs after three seconds with the rise during the plateau region being about 100 or 200 milliliters depending upon the length of the sampling time. This rise seems to be independent of lung volume and gender.
The spirograms in Figure 7 were taken from the same healthy female subject. Spirogram A in Figure 7 is the normal spirometry sample while Spirogram B is that obtained from the long and steady blow similar to the breath testing sample. The graphs are similar to that presented in Figure 4, with both showing the slight rise in the plateau region. While the slope of the rising portion obtained from the BrAC-type sampling has a different slope than that obtained with actual breath curves seen in BrACT situations, this difference is most likely explained by the sampling not being given through the mouthpiece itself. The minimum adequate sample is obtained in about 1.5 seconds in an actual BrACT. Even with this difference in slope, the final breath volumes for each individual subject have the same approximate volumes. Figure 8 shows three successive spirograms obtained from a female subject diagnosed with asthma. The middle spirogram was for the third sample. The different final volumes obtained agree with the observations that asthmatics give unreliable breath alcohol samples. Spirometry testing confirms that the deep lung air referred to in discussions of BrAC testing is, in fact, the forced expiration portion of the breath profile and is limited to the sampling of the bronchial region of the lungs. These findings are consistent with the paradigm proposed by Hlastala limiting the breath alcohol sample to the bronchi.
Hlastala has proposed that the mucous on the surface of the bronchi is the major influence to the breath alcohol concentration arising from the absorption and desorption of pulmonary alcohol.30 It is widely accepted that the ingested ethanol is quickly distributed to all the aqueous portions of the body. Bronchial mucous is approximately 95 percent water31 and would thus be involved in the distribution of the ingested ethanol as would the mouth saliva, which is approximately 99 percent water.32 The observations reported by Gullberg33 support some type of exchange in the bronchial and/or mouth air.
Ramifications of Hlastala’s Model
While there would be the exchange of ethyl alcohol from the capillaries to the alveolar air, Hlastala has proposed that this alveolar alcohol is absorbed by the mucous lining in the bronchi at the start of exhalation and then desorbed back into the later portions of the warmer exhaled air. Since there is already an alcohol concentration in the mucous lining from the normal distribution of alcohol into the aqueous tissues and fluids, it is unlikely that the total alveolar alcohol content enters the mucous lining. It is just as unlikely that the total mucous alcohol would be transferred to the later exhaled breath. At any time, t, the total expired breath alcohol concentration, [BrAC]exp, can be expressed as the summation of the individual components that may affect it by the general equation:
[BrAC]exp,t=a[BrAC]avl,t+ b[BrAC]muc,t+g[BrAC]sal,t+ d[BrAC]oth,t .
The co-efficients for each component represent the relative contribution that component makes to the total expired BrAC. The total concentration for the expired air [BrAC], arises from the possible contributions from the alveolar air [BrAC]avl, the positive and negative influence from the mucous lining of the bronchi [BrAC]muc, the saliva [BrAC]sal, and from other possible sources [BrAC]oth. These other sources include, but are not limited to, mouth alcohol, trapped alcohol,34 and stomach alcohol vapors contaminating the breath sample when collected during the absorption phase,35 or regurgitation. In the case of a nondrinking subject conducting a test on the mouth alcohol detector, all the terms would be zero except for [BrAC]oth arising from the swishing of the alcohol in the mouth. In this scenario, the value of the coefficient d, would be one and would be zero for each of the other coefficients. According to Hlastala’s new paradigm, b should be one and a, g, and d should be zero. Based upon the consideration of the spirometry data and distribution of alcohol into the aqueous portions of the body, and depending upon the factors present, d may or may not be zero and b and g should both have nonzero values.
At the beginning of the exhalation, the breath alcohol content would be expected to be mostly alveolar in origin, but with a much lower alcohol content than that found in the alveoli. This lower concentration would arise from the dilution occurring from the diffusion of the alcohol from the alveoli and by the absorption of the ethyl alcohol by the mucous. These effects would agree with the proposals presented in the literature that were used to explain the lower BrAC found in the blood-breath correlation studies mentioned earlier. For this breath sample, b could be considered a negative value with a and g then having nonzero values. At the latter portion of the exhalation, there would still be a contribution from the alveoli. In other words, a would have some nonzero value, but the prominent contribution would arise from desorption of the alcohol from the mucous; b would be relatively large, with g, the saliva contribution, still being nonzero.
If one assumes that the diffusion of the alveolar alcohol is constant over the course of the breath sampling, and the short time interval for the sampling would suggest that this assumption is valid, the contribution of the alveolar breath alcohol would be constant during the entire breath sampling time. Likewise, the contribution from the saliva would then also be expected to be constant over the course of the breath sampling time. The initial rise of the alcohol content in the sample chamber of the breath testing instrument is due to the initial filling of the sample chamber with breath containing alcohol. The remainder of the rise would be from the increasing alcohol content of the breath mixing with the existing content of the sample chamber. Since the alcohol content of the later portions of the exhalation would be due to contributions from all the components, the reported BrAC would then be significantly affected by the contribution of the desorbed alcohol from the mucous. Since the contributions from each component are unknown, the reported BrAC cannot be simply related to the BAC (as is now done by assuming the blood-breath ratio) or to the mucous alcohol content (as Hlastala’s new paradigm would suggest).
Another component of [BrAC]oth is the rise in the plateau portion of the breath curve when the collected sample is given beyond what would be considered the minimum acceptable sample. The contribution to the total breath volume for this portion of the breath is less than 20 percent of the total volume. However, in unpublished studies by this author, an increase of no greater than 0.022 grams per 210 liters has been observed depending upon the length of the sampling time beyond that needed for the minimum acceptable sample. This value seems to be independent of the instrument used, as both the Intoxilyzer 5000 and 8000 give this result and similar increases in the lung volume.
One possible explanation for the observed rise up to 0.022 grams per 210 liters, which would be independent of lung volume and instrument, is based upon the construction of the breath chamber or cell in which the sample is collected and measured during the breath sampling. In any infrared gas cell, only that portion of the cell volume through which the infrared beam passes provides the measured portion of the sample. The rest of the cell volume is not measured. If a sample were introduced into a breath cell that is then closed to the outer air, simple diffusion would allow the sample to equilibrate with the cell volume, with all portions of the cell volume having the same concentration. Thus, the infrared beam would be absorbed to the same extent for each similar sample volume introduced. This closed system occurs in simulator samples with the exit port connected to the simulator by tubing. Once the simulator vapors fill the sample cell and the flow continues for a brief time period, an equilibrium state exists with the concentration in the sample cell being constant. Each successive measurement of the simulator should give the same value as the certified simulator sample within the accuracy and precision of the breath testing instrument. If one would observe the LED display as the initial increments of simulator vapors fill the cell, one would observe this quick rise in concentration to the certified concentration of the simulator solution vapor concentration with only slight changes due to the accuracy and precision of the instrument.
In the case of the breath alcohol testing of a living subject, the sample cell acts as a flow cell open to the outer air at the exit port. As the breath sample enters the breath chamber or cell, a small portion of the contents of the cell nearest the exit port will be forced out into the surrounding air. The first increment of the incoming breath sample will start to commingle with the contents remaining in the breath cell by simple diffusion possibly modified by a small eddy effect arising from the flow into the breath cell.
The next increment of incoming breath again will force a small portion of the contents out into the surrounding air and again start to commingle with the remaining contents in the cell. Each following increment affects the contents of the breath cell in the same manner, resulting in a concentration gradient through the length of the sample cell in the direction of the infrared beam. This concentration gradient will be highest near the incoming sample port and lowest near the exit port. The infrared beam passes through this concentration with the majority of the absorption of the light occurring nearer the entrance port and a decreasing absorption occurring as the beam nears the detector.
The detector will only observe the total loss of intensity. However, as more sample passes into the detector, more molecules of ethanol enter the breath chamber and more molecules of ethanol will enter into the portion of the volume measured by the infrared light beam. The majority of the bronchial ethanol vapor will be in the breath cell within a few seconds and the incoming increments of the breath sample after the plateau is reached will not significantly affect the higher concentration gradients in the breath cell. However, as these final increments enter the breath cell, a portion of the cell volume with lowest concentration will be exiting the exit port. This is shown in Figure 9. The end result is that the sampling of the breath beyond the minimum acceptable sample will result in a smaller difference between the high and low concentration gradients, and produce an increased value for the observed [BrAC]exp that will be independent of lung volume or instrument. Since the volume of the breath cell is constant and the slight rise in plateau is fairly constant, the rise in breath test results observed should be fairly constant and independent of subject, which is what has been observed. This increase is not due to the longer sampling time producing a sample with more alveolar character, but rather can be considered a result of the better mixing of the breath cell contents and corresponding decrease of the concentration gradient in the breath cell.
Conclusion
Forensically, the question now arises as to the use of BrAC testing as a scientifically sound method. As this discussion indicates, it is clear that there can be no simple relationship between BrAC and BAC. This alone would explain the observed scattering in the blood-breath correlation studies and the lack of universal agreement on the blood-breath ratio. While the concept of a blood-breath ratio is useful in simulator testing, the concept of a blood-breath ratio for human subject testing should be discarded. As such, the idea that the BrAC numerical values as now reported have the same significance as the BAC in describing impairment has no theoretical basis. As the brain alcohol content has a direct relationship to the BAC, the BAC can be related to impairment due to the ingestion of alcohol. As the BrAC does not have a direct relationship with the BAC, the brain alcohol content cannot have a direct relationship to the BrAC. This casts into question those studies relating BrACT to impairment. The conclusion that is reached based upon the above discussion is that breath alcohol testing may be, at best, more suited to a role of a presumptive test requiring a BACT to determine if the BAC is above the legal limit. This would be similar to the forensic use of enzymatic testing of urine samples for drugs and would necessitate the establishment of cut-off levels above which BACT would be done. Studies on saliva and sweat alcohol concentration testing may ultimately provide high enough scientific reliability for use of these testing methods for forensic and legal purposes.36
Thanks to Dr. Michael Hlastala (for suggestions and guidance in his review of the early drafts of this paper); Dr. Lawrence Siegel and his staff (for assistance in the spirometry testing, explanations of the testing, and interpretation of the results); and Dionne Foley (for assistance in preparing the drawings).
© Wayne A. Morris, 2009. All rights reserved.
Notes
1. Y.H. Caplan, D. Yohman & J.A. Schaefer, An In Vitro Study of the Accuracy and Precision of Breathalyzer Models 900, 900A, and 1000, 30 J. Forensic Sci. 1058 (1985).
2. A.W. Jones, Enforcement of Drunk-Driving Laws by Use of ‘Per Se’ Legal Alcohol Limits: Blood and/or Breath Concentrations as Evidence of Impairment, 4 Alc. Drugs and Driving 99 (1998).
3. D. Labianca & G. Simpson, Medicolegal Alcohol Determination: Variability of the Blood- to Breath-Alcohol Ratio and Its Effect on Reported Breath Alcohol Concentrations, 33 Eur. J. Clin. Chem. Clin. Biochem. 919 (1995).
4. Id.
5. National Safety Council, Special Report on the Conversion Factor 1976, in Committee Handbook, Committee on Alcohol and Other Drugs 139 (1996).
6. M.F. Mason & K.M. Dubowski, Alcohol, Traffic, and Chemical Testing in the United States: A Résumé and Some Remaining Problems, 20 Clin. Chem. 126 (1974).
7. W. Giguiere & G. Simpson, Medicolegal Alcohol Determination: In Vivo Blood/Breath Ratios as a Function of Time, in Proceedings of the 27th Meeting of the International Association of Forensic Toxicologists, Perth (Australia) 494 (1990); comment in 12 Drinking/Driving L.
Meeting of the International Association of Forensic Toxicologists, Perth
(Australia) 494 (1990); comment in 12 Drinking/Driving L. Letter 141 (1993); K. Smith, Science, The Intoxilyzer and Breath Alcohol Testing, The Champion, May 1987 at 8; L.W. Masten, Does Breath Alcohol Accurately Predict the BAC? Sometimes … Fact Is Stranger Than Fiction, Florida Defender 54 (Fall 2003); A.WFLORIDA DEFENDER 54 (Fall 2003); A.W. Jones, Top TenDefense Challenges Among Drinking Drivers in Sweden, 31 MED. SCI. L. 229 (1991);G. Simpson, Accuracy and Precision of Breath-Alcohol Measurements for a RandomSubject in the Post-Absorptive State, 33 CLIN. CHEM. 261 (1987); G. Simpson, Accuracy and Precision of Breath-Alcohol Measurements for a RandomSubject in the Absorptive State, 33 CLIN. CHEM. 753 (1987); M.P. Hlastala, Physiological Errors Associated With Alcohol
Breath Testing,THE CHAMPION, July 1985 at 16; A. Flores, L.K. Eliason & Y.C. Wu, Breath Alcohol Sampling Simulator (BASS) for Qualification Testing of Breath Alcohol Measurement Devices, U.S. Department of Commerce, National Bureau of Standards, special publication 480-41 (1981).
8.Caplan et al., supra note 1;D.J. Smith & R. Laslett, Evaluation of the Drager Alcotest Model 7110 Infrared Breath Alcohol Analyzing Instrument, 30 J. FORENSIC SCI. SOC. 349 (1990);M.P. Hlastala, The Alcohol Breath Test, 93 J. APPL. PHYSIOL. 405 (2002).
9. G. Simpson, Accuracy and Precision of Breath-Alcohol Measurements for a Random Subject in the Absorptive State, 33 CLIN. CHEM. 753 (1987); Hlastala, supra note
8; A.W. Jones, Role of Rebreathing in Determination of the Blood-Breath Ratio of Expired Ethanol, 55 J. APPL. PHYSIOL. 1237 (1983).
10.W. Fowler, Lung Function Studies, II: The Respiratory Dead Space, 154 AM. J. PHYSIOL. 405 (1948); Understanding the Intoxilyzer 5000 [manufacturer instructional packet]. CMI/MPH, 6: 13-16 (1991); H. Karsten invented a breath alcohol detection system with identity verification (U.S. patent 6,967,581) (Nov. 22, 2005); A.J. Crockett,D.A.Schembri,D.J.Smith,R.Laslett & J.H. Alpers,MinimumRespiratory Function for Breath Alcohol Testing in South Australia,
32 J.FORENSIC SCI.SOC. 333 (1992);A.Norberg, J. Gabrielsson, A.W. Jones & R.G. Hahn, Within- and Between-Subject Variations in
Pharmacokinetic Parameters of Ethanol by nalysis of Breath, Venous Blood and Urine, 49 BR. J. CLIN. PHARMACOL. 399 (2000); R.F. Borkenstein, The Evolution of Modern Instruments for Breath Alcohol Analysis, 5 J. FORENSIC SCI. 395 (1960). K.M. Dubowski, Biological Aspects of Breath-Alcohol
Analysis, 20 CLIN. CHEM. (1974).
11. Dubowski, supra note 10; A.W. Jones, Variability of the Blood-Breath Ratio In Vivo, 39 J. STUD. ALCOHOL 1931 (1978); P. Harding & P.H. Field, Breathalyzer Accuracy in Actual Law Enforcement Practice: A Comparison of Blood- and Breath-Alcohol Results in Wisconsin Drivers, J. FORENSIC SCI.
1235 (1987); F.E. Camps & A.E. Robinson, Experiments Designed to Establish the Amount of Alcohol in the Blood Under Social Drinking Conditions, 8MED. SCI. L.153 (1968).
12. Camps et al., supra note 11.
13. G. Simpson, Medicolegal Alcohol Determination: Comparison and Consequences of Breath and Blood Analysis, 13 J. ANAL. TOXICOL. 361 (1989); A.G. Skåle, L. Slørdal,G.Wethe & J.Mørland, Blood/Breath Ratio at Low Alcohol Levels: A Controlled Study (abstract only), 14 ANN. TOXICOL. ANAL.
280 (2002); L.A. Greenberg, Physiological Factors Affecting Breath Samples, 5 J. FORENSIC SCI. 411 (1960); G.R. Fox & J.S. Hayward, Effect of Hyperthermia on Breath- Alcohol Analysis, 34 J. FORENSIC SCI. 836 (1989).
14. Florida Department of Law Enforcement Implied Consent Program, Breath Test Instrument Report (1993).
15. M.F. Mason & K.M. Dubowski, Breath-Alcohol Analysis: Uses, Methods, and Some Forensic Problems — Review and Opinion, 9 J. FORENSIC SCI. 9 (1976).
16. R.N.Harger, B.B. Raney, E.G. Bridwell & M.F. Kitchel, The Partition Ratio of Alcohol Between Air, Water, Urine and Blood: Estimation and Identification of Alcohol in These Liquids From Analysis of Air Equilibrated With Them, 183 J. BIOLOGICAL CHEM. 197 (1950).
17. A.W. Jones, Determination of Liquid/Air Partition Coefficients for Dilute Solutions of Ethanol in Water, Whole Blood, and Plasma, 7 J. ANAL.TOXICOL. 193 (1983).
18. Jones, supra note 2; Masten, supra note 7;M.F.Mason & K.Dubowski, Breath as a Specimen for Analysis for Ethanol and Other Low Molecular Weight Alcohols, in MEDICAL ASPECTS OF ALCOHOL 172 (James C. Garriott ed., 3d ed.) 1966; D.R.Wilkinson, P. Haines, R. Moargner, D. Sockrider, C.L. Wilkinson & M. Spartz, The 2100/1 Ratio Used in Alcohol Programs Is Once Again Under Attack, in ALCOHOL, DRUGS AND TRAFFIC SAFETY-T86 391 (P.C. Noordzij & R. Roszbach eds.) (1987).
19. Jones, supra note 9; J.Ohlsson, D.D. Ralph, M.A. Mandelkorn, A.L. Babb & M.P. Hlastala, Accurate Measurement of Blood Alcohol Concentration With Isothermal Rebreathing, 51 J. STU . ALCOHOL 6 (1990); A.W. Jones, Role of Rebreathing in Determination of the Blood-Breath Ratio of Expired Ethanol, 55 J. APPL. PHYSIOL. 1237 (1983).
20. Mason, supra note 18; A.W. Jones, How Breathing Techniques Can Influence the Results of Breath-Alcohol Analysis, 22 MED SCI. L. 275 (1982).
21. R.G. Gullberg, Duplicate Breath Alcohol Analysis: Some Further Parameters for Evaluation, 31 MED. SCI. L. 239 (1991).
22. M.P. Hlastala, The Alcohol Breath Test — A Review, 84 J. APPL. PHYSIOL. 401 (1998);M.P. Hlastala,W.J.E. Lamm & J. Nesci, The Slope Detector Does Not Always Detect the Presence ofMouth Alcohol, THE CHAMPION, March 2006 at 57; M.P. Hlastala, A New J.: L.& SCI. 1 (1998);M.P.Hlastala,Why Breath
Tests of Blood-Alcohol Don’tWork, in DWI ON TRIAL — THE BIG APPLE SEMINARY (May 2001); M.P. Hlastala, Breathing-Related Limitations to the Alcohol Breath Test, 17 DWI J.: L. & SCI. 1 (2002); M.P. Hlastala, Scientific Aspects of the Alcohol Breath Test, in DUI DEFENSE SKILLS CERTIFICATION PROGRAM 1, Washington Law School Continuing Legal Education (1995); M.P. Hlastala & E.R. Swenson, Airway Gas Exchange, in THE BRONCHIAL CIRCULATION 417
(John Butler ed.) (1992).
23. See note 22, supra; Hlastala, A New Paradigm for the Alcohol Breath Test; Hlastala, Why Breath Tests of Blood-Alcohol Don’t Work; Hlastala, Breathing-Related Limitations to the Alcohol Breath Test; Hlastala, Scientific Aspects of the Alcohol Breath Test; Hlastala & Swenson, Airway Gas
Exchange.
24. Jones, supra note 2; Mason, supra note 6; Understanding the Intoxilyzer 5000 [manufacturer instructional packet]. CMI/MPH, 6: 13-16 (1991); Borkenstein, supra note 10; Dubowski, supra note 10; Florida Department of Law Enforcement Implied Consent Program, Breath Test Instrument Report (1993); Mason & Dubowski, supra note 15; Federal Signal Corporation, Features/Functions/Benefits, Product: Intoxilyzer 5000 (manufacturer
packet); H.W. Smith, Methods for Determining Alcohol, in METHODS OF FORENSIC SCIENCE, IV 53 (A.S. Curry ed.) (1965).
25. NATIONAL SAFETY COUNCIL, Special Repo t on the Conversion Factor 1976, in Committee Handbook, Committee on Alcohol and Other Drugs 139 (1996).
26. Jones, supra note 9; Ohlsson et al., supra note 19.
27. Florida Department of Law Enforcement, Criminal Justice Standards and Training Commission, Breath Test Operator Course—A 24-Hour Course (2007
manual) (public record site http://www.fdle.stae.fl.us/atp/PublicRecor ds.htm).
28. S.I. FOX, HUMAN PHYSIOLOGY 452 (3d ed.) (1990).
29. PULMONARY PHYSIOLOGY 57 (Michael G. Levitzky ed., 5th ed.) (1999).
30. See note 22, supra; Hlastala, The Alcohol Breath Test — A Review; Hlastala, Lamm & Nesci, The Slope Detector Does Not Always Detect the Presence of Mouth Alcohol; Hlastala, A New Paradigm for the Alcohol Breath Test; Hlastala, Scientific Aspects of the Alcohol Breath Test;Hlastala & Swenson, Airway Gas Exchange.
31. C. Basbaum & M.J. Welsh, Mucus Secretion and Ion Transport in Airways, in TEXTBOOK OF RESPIRATORY MEDICINE 327 (J.F. Murray, J.A. Nadel & R.J Mason eds., 3d ed., vol. 1) (2000).
32. P.D.V. de Almeida, A.M.T. Gregio, M.A.N. Machado, A.A.S. de Lima & L.R. Azevedo, Saliva Composition and Functions: A Comprehensive Review,9 J.CONTEMP.DENTAL PRAC. 1 (2008).
33. R.G.Gullberg, supra note 21.
34. W.A. Morris, Mouth Alcohol and Trapped Alcohol, 20 DWI J.: L. & SCI. 4 (2005).
35. G. Simpson, Accuracy and Precision of Breath-Alcohol Measurements for a RandomSubject in the Post-Absorptive State, 33 CLIN. CHEM. 261 (1987); E.M.P. Widmark, PRINCIPLES AND APPLICATIONS OF MEDICOLEGAL ALCOHOL DETERMINATION 40 (English trans.) (1981).
36. E.M.P. Widmark, PRINCIPLES AND APPLICATIONS OF MEDICOLEGAL ALCOHOL DETERMINATION 40 (English trans.) (1981); K M. Dubowski, Absorption, Distribution and Elimination of Alcohol: Highway Safety Aspects, 10 J. STUD. ALCOHOL 98 (1985); R.C. Basalt, Disposition of Alcohol in Man, in MEDICAL ASPECTS OF ALCOHOL 73 (James C. Garriott, ed., 3d ed.) (1966); M.J. Buono, Sweat Alcohol Concentrations Are Highly Correlated With Co-Existing Blood Values in Humans, 84 EXP. PHYSIOL. 401 (1999); J.C. Anderson & M.P. Hlastala, The Kinetics of Transdermal Ethanol Exchange, 100 J. APPL. PHYSIOL. 649 (2006); R.P. Roine, R. Yulikahri & M. Salaspuro, Urinary Dilichol — A New Marker of Alcoholism, 11 ALCOHOLISM:CLINICAL AND EXPERIMENTAL RES. 525 (1987).
|
 |
National Association of Criminal Defense Lawyers (NACDL)
1660 L St., NW, 12th Floor, Washington, DC 20036
(202) 872-8600 • Fax (202) 872-8690 •
assist@nacdl.org
|
|