End-tidal carbon dioxide as an indicator of arterial carbon dioxide in neurointensive care patients.

The relationship between the arterial partial pressure of carbon dioxide (Paco2) and the end-tidal carbon dioxide partial pressure (PEtco2) was evaluated in 11 critically ill adult neurointensive care patients during mechanical ventilation. It was hypothesized that the Paco2 to PEtco2 gradient, or P(a-Et)co2, was maintained and that PEtco2 can be used to determine Paco2 accurately in these patients. After approval by the Clinical Investigations Committee, when clinically indicated arterial blood gases (with Paco2) were measured, the PEtco2 was determined from the capnograph (Hewlett Packard 78520A infrared capnometer). The P(a-Et)co2 was evaluated for possible effects from changes in the other monitored hemodynamic and respiratory parameters. Linear regression analysis was used to determine the significance of the relationship between Paco2 and PEtco2 and other assessed variables. Student's t tests were used where applicable. A p value </=0.05 determined significance. One hundred thirty-five comparisons, 12.3 +/- 5.8 per patient, of Paco2 and PEtco2 were made. (All values are means +/- SD.) The P(a-Et)co2 was 6.9 +/- 4.4 mm Hg (-11-21 mm Hg), with Paco2 = 34 +/- 6 mm Hg and PEtco2 = 27 +/- 6 mm Hg. There was a significant correlation between Paco2 and PEtco2 values for the total study population (r = 0.72, p = 0.001). However, when the relationship between Paco2 and PEtco2 values for individual patients was analyzed, only seven of 11 patients (64%) had significant correlations. The direction of Paco2 change was inaccurately predicted by PEtco2 changes in 31.9% of measurements. PEtco2 does not provide a stable reflection of Paco2 in all neurointensive care patients. Arterial blood gases cannot be eliminated when monitoring respiratory acid-base balance in mechanically ventilated neurointensive care patients.