Why Breath Tests of Blood-Alcohol Don't Work

by Michael P. Hlastala, Ph.D.
Division of Pulmonary and Critical Care Medicine
Box 356522
University of Washington
Seattle, WA 98195-6522

The Lungs: Effects of Moisture and Temperature

The lungs have a relatively simple, but non-uniform, anatomical structure. The airways are a branching, tree-like arrangement of tubes. Inspired air moves through progressively shorter, narrower and more numerous airways. These airways are lined with mucus at a temperature varying between approximately 34^oC at the mouth and 37^oC in the very smallest airways. However, this temperature range varies depending on the breathing pattern. The membranes separating the air in the alveoli and the blood in the capillaries are so thin that inert gases such as alcohol equilibrate between blood and air very rapidly. With exhalation, air within the alveoli is conducted along the airways to the mouth.

During inspiration, air is heated and humidified as it passes through the upper airways. Some water within the mucous layer or watery sub-mucous layer will vaporize and heat stored in the airways will be picked up by the inspired gas and taken to the alveoli. During exhalation, the process reverses; fully humidified air at core body temperature is cooled by the cooler airway mucosa and water vapor condenses on the mucosa. This water and heat exchange process is vital because it conditions the inspired air to avoid damaging the delicate alveolar cells while conserving water and heat from major loss in the exhaled air. Under normal environmental conditions, exhaled gas has less heat and less water vapor than does alveolar air.

The dynamics of soluble gas exchange are similar to the dynamics of heat and water exchange. These processes are analyzed using analogous equations. The fact that respired air exchanges heat and water with the airways implies similar soluble gas exchange processes. This interaction of soluble gases with airway mucosa is well documented. The degree of interaction is directly related to the solubility of the gas in the airway mucosa and mucous lining. The very high solubility of alcohol in water guarantees its strong interaction with airway tissue. Because this interaction depends on temperature and airflow characteristics, variations in tidal volume and frequency can have a substantial effect on the alcohol concentration in the breath sample. This variation is affected by the difference in temperature between the outside air and the alveolar air.

The exchanges of heat and of gas with the airways are complex and interactive processes. The relative significance of this exchange depends on the effective solubility of the gas in the mucosa. For the respiratory gases, oxygen and carbon dioxide, airway tissue solubility is small. For both water and alcohol airway solubility is quite large. Moreover, the exchange processes are interactive. During inspiration, heat, water and alcohol are transported from the mucosa to the air. The exchange of heat cools the mucosa causing an increase in its alcohol solubility and, hence, a decrease in the partial pressure of alcohol in the mucosa and a reduction in alcohol flux into the airway lumen. These various processes have been integrated into a mathematical model developed by Tsu et al and further refined by George et al which shows that during normal breathing, the inspired air is equilibrated with alcohol, picking it up from the airways, before reaching the seventeenth generation airways (start of the alveolar region). Upon reaching the alveoli a small amount of additional alcohol is picked up because the solubility of alcohol in blood is lower than the solubility of alcohol in water. The equilibrium partial pressure of alcohol in vapor above blood is greater than in vapor above water at the same temperature. With exhalation, the excess alcohol picked up in the alveoli is rapidly lost to the airways within the sixteenth or fifteenth generation. Along the airway, more alcohol is lost to the airways. The alcohol that arrives at the mouth comes essentially from the airways and not from the alveoli. This is also the case for water vapor. The humidification of inspired air is performed by the airways.

Figure 4. Airway alcohol exchange.

The flux of alcohol from the mucus surface into the air (positive values) during inspiration and the flux of alcohol from the air to the mucus surface (negative values) during expiration is demonstrated in Figure 4. This figure was calculated using a mathematical model of the human airway structure. During inspiration, alcohol is taken up into the inspired air immediately at the mouth. The greatest alcohol uptake occurs in the trachea and generations 6 through 13. During expiration, the redeposition of alcohol occurs primarily at these same airway generations. The important conclusion from this work is that all of the alcohol that comes out of the mouth in the breath comes from the airway surfaces rather than from the alveolar regions.

The early basic assumption of the alcohol breath test was that the breath alcohol concentration was the same irrespective of the exhaled volume as long as the dead space volume is exhaled (as shown in Figure 3). However, others have shown that the breath alcohol concentration depends on exhaled volume. The breath testing instrument takes a sample of air from the end of the breath whenever the subject stops but the volume of breath exhaled is neither controlled nor measured. Therefore, the apparent breath alcohol concentration depends on the volume of air delivered to the breath testing instrument. The last part of the breath can be well above the average single breath alcohol level because the alveolar plateau has a positive slope that is dependent on air temperature.

A sloping alveolar plateau for various low solubility gases has been explained by several factors including stratified inhomogeneity (gas phase diffusion limitation), convection-diffusion interaction, sequential exhalation from regions with differing V_A /Q and continuing gas exchange. None of these factors contribute substantially to the slope of the exhaled alcohol profile. Continuing gas exchange will contribute to the slope of the exhaled profile for respiratory gases (CO2 and O2), but not inert gases.

Further variation in BrAC will result from changing the breathing pattern immediately before delivering the sample breath. Hyperventilation for 20 seconds prior to delivering a sample breath to the breath tester causes an 11% reduction in BrAC. Three deep breaths prior to the sample breath reduces BrAC by 4%. After breath-holding for 15 seconds prior to exhalation, the BrAC increases by 12% (for a minimum exhalation) and 6% (for a maximum exhalation). A 30 second breathhold prior to exhalation increases BrAC by 16%. These effects are caused by the respective cooling or warming of the airways and the data further support the airway surface interaction of alcohol as the mechanism causing the changing alcohol concentration during exhalation.