Cardiac Index Validation Using the Pressure Recording Analytic Method in Unstable Patients

Objective: The authors investigated the accuracy and precision of the pressure recording analytic method (PRAM) in cardiac index measurement compared with thermodilution in unstable patients, a setting in which minimally invasive monitoring devices often fail. Design: Criterion standard. Setting: Intensive care unit. Patients: Thirty-two consecutive patients with low cardiac output syndrome treated with an intra-aortic balloon pump and/or high doses of inotropic drugs but without atrial fibrillation were studied after cardiac surgery. Interventions: None. Pulmonary and radial artery catheters were already in situ for clinical reasons. Measurements and Main Results: Four patients (12.5%) were excluded from the study because of artifacts caused by under- or overdamping of the arterial pressure monitoring system. The authors performed 3 injections of the thermal indicator in 5 minutes through the pulmonary artery catheter. Mean cardiac index values of 12 consecutive beats were considered for the PRAM. A significant correlation was found between the cardiac index assessed by thermodilution and PRAM (r = 0.72, p <0.001). The mean bias between the 2 techniques was 0.072 ± 0.41 L/min/m² with lower and upper 95% limits of confidence of -0.089 and 0.233 L/min/m², respectively. The percentage error was 30%. Sufficient agreement between the two techniques was evidenced by the Bland-Altman plot with only two points above the limits of agreement. Conclusions: This study showed that PRAM, a minimally invasive method for cardiac index assessment, is clinically useful even in unstable patients such as those receiving intra-aortic balloon pump and/or ongoing high doses of a inotropic drugs because of a low cardiac output syndrome but without atrial fibrillation.