Use of lithium dilution and pulse contour analysis cardiac output determination in anaesthetized horses: a clinical evaluation

OBJECTIVE: To assess the suitability of a human algorithm for calculation of continuous cardiac output from the arterial pulse waveform, in anaesthetized horses. STUDY DESIGN: Prospective clinical study. ANIMALS: Twenty-four clinical cases undergoing anaesthesia for various conditions. MATERIALS AND METHODS: Cardiac output (Qt), measured by lithium dilution (QtLiDCO), was compared with a preceding, calibrated Qt measured from the pulse waveform (QtPulse). These comparisons were repeated every 20-30 minutes. Positive inotropes or vasopressors were administered when clinically indicated. Cardiac indices from 30.7 to 114.9 mL kg(-1) minute(-1) were recorded. Unusually shaped QtLiDCO curves were rejected and the measurement was repeated immediately. RESULTS: Eighty-nine comparisons were made between QtLiDCO and QtPulse. The bias between the mean (+/-SD) of the two methods (QtLiDCO - QtPulse) was -0.07 L minute(-1)(+/-3.08) (0.24 +/- 6.48 mL kg(-1) minute(-1)). The limits of agreement were -12.72 and 13.2 mL kg(-1) minute(-1) (Bland & Altman 1986; Mantha et al. 2000). Linear regression analysis demonstrated a correlation coefficient (r2) of 0.89. Cardiac output in individual patients varied from 49.1 to 183% of the initial measurement at the time of calibration. Linear regression of log-transformed Qt variation for each method found a mean difference of 9% with limits of agreement of -4.1 to 22.1%. CONCLUSIONS AND CLINICAL RELEVANCE: This method of pulse contour analysis is a relatively noninvasive and reliable way of monitoring continuous Qt in the horse under anaesthesia. The ability to easily monitor Qt might decrease morbidity and mortality in the anaesthetized horse.