Ventilation, thermal noise, and errors in cardiac output measurements after cardiopulmonary bypass

The authors observed transient increases in the amplitude of respiratory variations in pulmonary artery blood temperature in many patients after cardiopulmonary bypass (CPB). This increased "thermal noise" may significantly influence measurements of thermodilution cardiac outputs (TDCO) performed during this time.

The authors recorded the peak-to-peak amplitude of respiratory variations in pulmonary artery blood temperature in 15 patients during the first 35 min after CPB. Possible relationships between the amplitude of these variations and the magnitude of temperature differences between commonly monitored body temperature sites (nasopharyngeal, rectal, bladder, and pulmonary artery) were also examined. In ten additional patients, the authors investigated the influence of these increased respiratory variations on TDCO measurements by correlating the maximum variation in three successive TDCO measurements with the peak-to-peak amplitude of the respiratory variations in pulmonary artery blood temperature. Potential error in TDCO measurements caused by these increased respiratory variations in pulmonary artery blood temperature were also examined using model calculations of the effects of respiratory variations in pulmonary artery blood temperature on measured TDCO thermal areas.

In the first 15 patients, the mean amplitude of respiratory variations in pulmonary artery blood temperature after CPB (mean +/- SEM) were: (1) within 5 min after CPB, 0.037 +/- 0.004 degrees C; (2) 10 min after #1, 0.025 +/- 0.003 degrees C; (3) 20 min after #1, 0.019 +/- 0.003 degrees C; and (4) 30 min after #1, 0.012 +/- 0.002 degrees C. There were no significant correlations between the magnitude of the respiratory variation in pulmonary artery blood temperature and the observed temperature differences between body sites. Four patients had pulmonary artery blood temperature variations in excess of the maximum amplitude previously reported in man (0.05 degrees C). In the next ten patients, the maximum variation between three successive TDCO measurements taken at specified times in the respiratory cycle (end inspiration, end exhalation, and 3 s after end exhalation) was significantly correlated with the amplitude of respiratory variations in pulmonary artery blood temperature (r = 0.83, P < 0.001). Four patients with increased respiratory variations in pulmonary artery blood temperature had variations in TDCO measurements exceeding 2 l/min. Subsequent model calculations demonstrated that the magnitude of potential error in TDCO measurements is dependent on both the amplitude of the respiratory variations in pulmonary artery blood temperature and the baseline cardiac output. On the basis of these thermal area calculations, potential errors of 15-50% could be caused by respiratory variations in pulmonary artery blood temperature > 0.05 degrees C.

The authors concluded that respiratory variations in pulmonary artery blood temperature are transiently increased in many patients after CPB, and that this increased "thermal noise" may cause significant errors in TDCO measurements.