Transcutaneous oxygen and carbon dioxide monitoring. Use in adult surgical patients in an intensive care unit

E. N. Schachter, T. D. Rafferty, C. Knight, R. Yocher, R. Mentelos, L. Giambalvo, L. Firestone and P. G. Barash

We evaluated a combined transcutaneous oxygen and carbon dioxide (tcPO2 and tcPO2) monitor in 33 adult surgical patients in an intensive care unit. Surgical procedures included cardiothoracic, general, vascular, and orthopedic operations. Ninety-three paired, transcutaneous values were compared with simultaneously determined arterial blood gas measurements. The correlation coefficient for PaO2 was .75 (P less than .01) whereas for PaCO2 it was .55 (P less than .01). Although statistically significant, individual transcutaneous values differed by as much as 50 mm Hg from levels determined from arterial samples. Such differences make it inappropriate to use this monitor to predict actual PaO2 or PaCO2. By contrast, continuous monitoring allowed observation of acute changes associated with respiratory-care procedures and cardiovascular decompensation. Such acute changes changes were appreciated on the tcPO2 tracing but not the tcPO2 tracing. We conclude that the tcPO2 monitor is a valuable trend indicator of cardiopulmonary dysfunction, but that the tcPCO2 monitor is not.