ABSTRACT
BACKGROUND: Liver transplantation (LT) implies hemodynamic instability, making invasive monitoring of cardiac output (CO) mandatory. Intermittent thermodilution with pulmonary artery catheter (PAC) remains the clinical gold standard to measure CO. The agreement between PAC and new monitoring methods in LT needs to be further investigated. Our aim is to clarify whether cardiac index (CI) measurements with transpulmonary intermittent thermodilution, and continuous pulmonary thermodilution methods agree sufficiently with those performed intermittently with PAC to be considered interchangeable during LT. METHODS: We studied prospectively hemodynamic parameters of 72 consecutive patients undergoing LT. Each CI was obtained simultaneously with three different techniques: intermittent (PACi) and continuous (CCI) pulmonary artery thermodilution with PAC, and intermittent transpulmonary thermodilution (TPTD) with PiCCO2 in 8 time points of the procedure, obtaining 1350 paired measurements. Exclusion criteria was retransplantation. The statistical Bland Altman method for repeated measures was used to assess agreement, and polar plot methodology to evaluate trending ability. RESULTS: Analysis of agreement between PACi and TPTD measurements (N.=474 paired measurements) showed a bias of -0.42 L/min/m2, 95% limits of agreement (95%LoA) of ±1.5 L/min/m2 and percentage error of 45%. PACi-CCI comparisons (N.=431) showed bias of -0.02 L/min/m2, 95%LoA of ±1.96 L/min/m2, and percentage error of 64%. These results demonstrated questionable clinical agreement between PACi and TPTD, and no agreement between PACi and CCI. TPTD and CCI showed poor CO trending ability. CONCLUSION: Continuous pulmonary thermodilution with PAC is not an alternative monitoring method of CO. Transpulmonary thermodilution CO monitoring with PiCCO2 shows too questionable agreement with the clinical gold standard (PACi) being in the limit of acceptance to be considered interchangeable during liver transplantation.
