Background: To compare tracheal temperature (T_ET) with nasopharyngeal temperature (T_NP) in patients undergoing cardiac surgery using cardiopulmonary bypass (CPB). 

Methods: T_ET was measured using a thermistor in the cuff of an endotracheal tube and T_NP was monitored using an esophageal stethoscope. Depending on the management of the CPB strategy, the operation was divided into four periods (pre-CPB, cooling, rewarming, and post-CPB). A Bland–Altman analysis was carried out to compare T_ET with T_NP in each period. The concordance correlation coefficient for repeated measures analysis and various time lags was used to validate the time lag maximizing the concordance correlation coefficient between the two methods. 

Results: A total of 7191 pairs of temperature recordings acquired from 24 patients were included in the analysis. During steady normothermia, 81.7% (95% confidence interval [CI], 79.8–83.5%) of the pairs had a difference within ±0.5 °C, with a mean bias of −0.28 °C and limits of agreement (LOA) of −0.74 °C to 0.18 °C. The LOA during the cooling and rewarming phase of CPB were −1.13 °C to 0.51 °C and −0.91 °C to 1.29 °C, respectively. The mean bias and LOA throughout the entire operation were −0.10 °C and −0.98 °C to 0.77 °C, respectively. Throughout the entire operation, −2 min of time lag on T_ET maximized the concordance correlation coefficient (0.94 [95% CI, 0.92–0.96] to 0.95 [95% CI, 0.93–0.96]), indicating an earlier response of T_ET than T_NP. 

Conclusions: T_ET could be an alternative to T_NP during cardiac surgery using CPB.