RESUMO
Circulation time (Ct) between lung and periphery may be a surrogate for cardiac output, estimated here, for the most part, as the time between taking a breath of nitrogen and peripheral detection of a desaturation pulse. Use of pulse oximetry involves an internal, instrument delay; however, using the ear, we found shortening with exercise (12.1 +/- 0.37 sec, at rest; 9.1 +/- 0.25 sec at 100 watts), lengthening after beta-blockade, and lengthening in patients with echocardiographic and clinical left heart failure (8 patients 16.2 +/- 1.1 sec; 6 controls 12.0 +/- 0.5 sec). Pulse oximetry failed, however, to discriminate heart failure from normal in several patients. In patients referred to a department of nuclear medicine for assessment of chest pain, pulse oximetry (finger and ear) showed unacceptable variability. Nuclide delays between lung and carotid artery correlated significantly with the reciprocal of gated SPECT estimated cardiac output (Q(gs)); not so, however, for lung to finger. In normal subjects, an old Waters fast response oximeter gave short, reproducible Ct estimates and a significant correlation with the reciprocal of (indirect Fick) cardiac output (Q(if)). The relationship for normal subjects was: Ct = 0.28 x 60/Q(if) + 2.8 sec (Q(if) in L min.; P slope < .001).