Comparison of thermodilution, lithium dilution, and pulse contour analysis for the measurement of cardiac output in 3 different hemodynamic states in dogs.

OBJECTIVE: To (1) evaluate lithium dilution (LiDCO) and transpulmonary thermodilution (PiCCOTD ) in relation to traditional thermodilution (PAC-TD) for determining cardiac output (CO) in 3 different hemodynamic states in dogs and to (2) compare the continuous CO values obtained using power analysis (PulseCO) with continuous PiCCO (PiCCOc).

DESIGN: Prospective randomized study.

SETTING: University research laboratory.

ANIMALS: Fourteen healthy Beagles.

INTERVENTIONS: CO was measured using PAC-TD, LiDCO, and PiCCOTD in 3 different hemodynamic states induced in random order and defined on the basis of the mean arterial pressure (MAP). Normodynamic state was defined as the baseline MAP and 1 MAC sevoflurane. The hypodynamic state was induced with a deep level of sevoflurane anesthesia. The hyperdynamic state was induced with noradrenaline. After these measurements were obtained in each hemodynamic state, CO was monitored continuously for 30 min using PulseCO and PiCCOc. Agreement was assessed using Bland-Altman analysis and intraclass correlation coefficients, and a trend score was determined for the continuous CO measurements.

MEASUREMENTS AND MAIN RESULTS: There was good agreement among the 3 modalities of CO measurement in each hemodynamic state. The mean CIPAC-TD /CIPICCOTD bias was -0.04 ± 1.19 L/min/m(2) (limits of agreement, -2.37/1.93 L/min/m(2) ), and the mean CIPAC-TD /CILiDCO bias was -0.11 ± 1.55 L/min/m(2) (limits of agreement, -3.04/2.93 L/min/m(2) ). The mean CIPulseCO -CIPiCCOc bias was -0.04 ± 1.91 L/min/m(2) (limits of agreement, -1.95/1.87 L/min/m(2) ), which suggested good agreement. The CIPulseCO -CIPiCCOc trend score, calculated from 252 paired comparisons, was 93.3% positive after zone exclusion (∆CI < 15%).

CONCLUSIONS: Both LiDCO and PiCCOTD agreed well with PAC-TD for the measurement of CO under different hemodynamic conditions. Moreover, PiCCOc appears to be an accurate method for monitoring continuous CO in dogs as its performance for measurement was similar to that of PulseCO.