In Reply:-We appreciate Dr. Fletcher's careful review of our article and considerate comments. We considered the ratio PaCO(2):minute ventilation to be a simple calculation of ventilatory efficiency in the restrictive setting of cardiopulmonary and metabolic stability with "…relatively constant PaCO(2) and carbon dioxide production…"  We did not and would not advise use of the calculation in a setting of variable CO2production, alveolar ventilation, or PaCO(2). As a separate measure, we calculated the P(a-et)CO(2) and confirmed that intermittent CPAP results in more efficient ventilation than traditional controlled mechanical ventilation (CMV). Fletcher also referred to an increase in alveolar ventilation, which was precluded by study design. We produced equivalent levels of alveolar ventilation during CMV and intermittent CPAP, as evidenced by similar PaCO(2) values, but with significantly less total ventilation during the latter. Using the calculation P(a-et)CO(2)/PaCO(2) as an indication of the fraction of unperfused, but ventilated alveoli  we found a mean alveolar deadspace of 4% during intermittent CPAP and 17% during CMV.
It is well known that an inspiratory pause applied during CMV may result in a smaller calculated physiologic deadspace.  It is less apparent that "conducting airways," which define the "anatomic" deadspace, are reduced in volume by an end-inspiratory pause. It also is not clear that CPAP will have a pulmonary effect similar to that observed with conventional positive pressure ventilation with an inspiratory plateau. Our data clearly support a significant decrease in alveolar deadspace during intermittent CPAP. To test Fletcher's hypothesis that a decrease in anatomic deadspace might account for the observed increase in efficiency of ventilation during intermittent CPAP, we calculated the difference in mean anatomic and alveolar deadspace (VD) volumes during CMV and intermittent CPAP, using data in the manuscript  and standard equations.  We assumed a normal anatomic deadspace of 194 ml during CMV. During intermittent CPAP alveolar V (D) was 15 ml, anatomic VD139 ml, and total physiologic VD153 ml. During CMV alveolar VDwas 74 ml, anatomic VD194 ml, as assumed, and total physiologic VD268 ml. We observed that intermittent CPAP caused modest decrease in calculated anatomic VDcompared with CMV, of 28%. More important however, was the fivefold increase in alveolar deadspace generated by controlled ventilation. Overall physiologic VDwas 74% higher during CMV than during intermittent CPAP, emphasizing our conclusion that intermittent CPAP provides more efficient ventilation.
John B. Downs, M.D.
Department of Anesthesiology, MDC 59; University of South Florida; College of Medicine; 12901 Bruce B. Downs Boulevard; Tampa, Florida;email@example.com
(Accepted for publication December 17, 1998.)