Blood gas measurements during exercise: errors due to temperature correction.

This study assessed the degree to which correcting blood gas measurements to rectal temperature (Tre) rather than to the temperatures at which gas exchange occurs [pulmonary arterial (Tpa) or intramuscular (Tm)] introduces errors into blood gas analysis of exercising mammals. Horses and steers weighing 450 kg were run on a treadmill at speeds up to those eliciting maximal rates of O2 consumption (VO2max), and temperatures were measured in various body compartments. In both species Tpa rose faster than Tre during the run, the degree of dissociation being a function of exercise intensity and duration. Tm was measured only in horses, and it rose faster than Tpa during the run and decreased more slowly postrun. Correcting blood gas values measured at an analyzer temperature of 37 degrees C to Tre without accounting for transient increases during the run of Tpa and Tm that were never reflected in Tre significantly biased estimates of blood gases. The biased estimates erroneously indicated that both species experienced more severe hypoxemia than they actually did at VO2max and masked the hypercapnia experienced by the horses at VO2max.