Assessment of cardiac output from systemic arterial pressure in humans

OBJECTIVE: To evaluate the reliability, by comparison with established techniques, of a new method to assess cardiac output, called pressure recording analytical method (PRAM), deriving from the analysis of the arterial pressure profile in the time domain the arterial-pressure-blood flow relationship. DESIGN: Criterion standard. SETTING: Hemodynamics laboratory at an university medical center. PATIENTS: Twenty-two hemodynamically stable cardiac patients scheduled for diagnostic right and left heart catheterization. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Cardiac index was simultaneously estimated by direct-oxygen Fick method, thermodilution, and PRAM applied to pressure signals recorded either invasively from an aortic catheter (PRAMa) or noninvasively at the finger (PRAMf) by photoplethysmography. Cardiac index values obtained by established techniques were significantly correlated with those estimated by PRAM: Fick method vs. PRAMa, r(2) =.88, vs. PRAMf, r(2) =.94; thermodilution vs. PRAMa, r(2) =.77, vs. PRAMf, r(2) =.77. The Bland-Altman analysis showed agreement between the Fick method and PRAM, with all data points comprised within the limits of agreement (+/-2SD) (mean difference +/- SD: -0.012 +/- 0.187 L x min(-1) x m(-2) for PRAMa; 0.024 +/- 0.167 L x min(-1) x m(-2) for PRAMf). Agreement was also found between thermodilution and PRAM, with all but one data point lying within the limits of agreement (mean difference +/- SD: -0.154 +/- 0.348 L x min(-1) x m(-2) for PRAMa; -0.108 +/- 0.348 L x min(-1) x m(-2) for PRAMf). CONCLUSIONS: In the range evaluated (cardiac index from 1.65 to 3.91 L x min(-1) x m(-2) by the Fick method), PRAM provides reliable invasive and noninvasive estimates of cardiac output in hemodynamically stable cardiac patients. PRAM may prove clinically useful for the beat-to-beat monitoring of cardiac output.