The reliability of pulse contour-derived cardiac output during hemorrhage and after vasopressor administration.

Abstract

Reliable measurement of cardiac output (CO) is important in the critically ill. Pulse contour-derived CO (PCCO) has been evaluated during stable hemodynamics, but is sensitive to changes in vascular tone and has not been validated under conditions of changing hemodynamics. Furthermore, PCCO requires calibration for the individual vascular impedance by transpulmonary thermodilution CO (TPCO), and the required frequency of recalibration to maintain accurate measurements, especially during changing conditions, has not been confirmed. We compared PCCO measurements of CO with TPCO and continuous and bolus pulmonary artery CO (CCO and BCO, respectively) during conditions of uncontrolled hemorrhage and resuscitation with norepinephrine. Thirteen pigs were anesthetized and instrumented for determination of CO by BCO and CCO, respectively, as well as bolus TPCO and PCCO. Uncontrolled hemorrhage was accomplished by liver incision. When mean arterial blood pressure was <25 mm Hg, or heart rate declined progressively to <20% of its peak value, vasopressor therapy was started. TPCO and BCO were performed after induction of anesthesia and 15 min after start of therapy, and PCCO and CCO were obtained repeatedly. CO measurements were compared using Bland-Altman analysis. Mean arterial blood pressure, CO and systemic vascular resistance decreased after hemorrhage (P < 0.001 and <0.01, respectively). Bias and limits of agreement between CCO and PCCO (0.54 L/min; 1.46 L/min) increased after hemorrhage (-3.49; 6.12) and further deteriorated after norepinephrine administration (-8.01; 9.9). After recalibration, bias and limits of agreement returned to -0.51 and 1.28. PCCO needs frequent recalibration during hemorrhage and after vasopressor administration.