Relationship between invasive and noninvasive measurements of gas exchange in anesthetized infants and children

Abstract

Minute ventilation (V̇(E), tidal volume (V(T)), carbon dioxide elimination (V̇(CO2)), and end-tidal (P(ET(CO2)) and arterial CO2 tensions (Pa(CO2)) were measured in 39 anesthetized infants and children with body weights ranging from 3.1 to 31 kg. Eighteen children had normal cardiopulmonary function, seven had acyanotic congenital heart disease, and 11 had cyanotic congenital heart disease. One child had left heart failure and pulmonary congestion, and two had severe parenchymal lung disease. To evaluate differences between pulmonary gas exchange calculated from Pa(CO2)) versus P(ET(CO2)), dead space volume (V(D)) and alveolar ventilation (V̇(A)) based on a Pa(CO2) (V(D)a, V̇(A)a) as well as on P(ET(CO2)) (V(D)(ET), V̇(A)(ET)) were performed, and correlations between Pa(CO2) - P(ET(CO2)), V(D)a/V(T) - V(D)(ET)/V(T), and V̇(A)a - V̇(A)(ET) were carried out. It was demonstrated that in normal children, as well as in those with acyanotic congenital heart disease, P(ET(CO2)) correlated closely with Pa(CO2) (r = 0.94, 0.98, respectively). In children with cyanotic congenital heart disease, however, correlation between P(ET(CO2)) and Pa(CO2) was relatively poor (r = 0.61). Mean values for Pa(CO2) were significantly higher than P(ET(CO2)) in the cyanotic children (P < 0.01), resulting in significant underestimation of physiologic dead space (P < 0.05) and significant overestimation of alveolar ventilation (P < 0.01). In three patients with pulmonary disease, large differences between Pa(CO2) and P(ET(CO2)) were comparable with those observed in the children with cyanotic congenital heart disease. It is concluded that in normal children and in children with acyanotic heart disease during anesthesia, noninvasive measurement of P(ET(CO2)) can be used as a reliable estimate of Pa(CO2) and for calculations of physiologic dead space and alveolar ventilation. In children with cyanotic congenital heart disease and severe pulmonary disease, P(ET(CO2)) does not provide a precise estimate of Pa(CO2), and calculations of physiologic dead space and alveolar ventilation should be based on direct measurements of Pa(CO2).