Comparison of tracheal temperature and core temperature measurement in living donor liver transplant recipients: a clinical comparative study

Abstract

Background: Body temperature is a vital sign, and temperature monitoring during liver transplantation is important. Tracheal temperature can be measured via an endotracheal tube with a temperature sensor on the cuff of the tube. This study aimed to investigate the accuracy and trending ability of tracheal temperature measurement compared to those of the core temperature measured at the esophagus and pulmonary artery (PA) in living donor liver transplant recipients. Methods: Twenty-two patients who underwent living donor liver transplantation (LDLT) were enrolled. Patients were intubated using an endotracheal tube with a temperature sensor placed on the inner surface of the tube cuff. Tracheal, esophageal, and PA temperatures were recorded at five time points corresponding to the different phases of liver transplantation. The tracheal and esophageal, tracheal and PA, and esophageal and PA temperatures were compared using Bland–Altman analysis, four-quadrant plot/concordance analysis, and polar plot analysis. Results: Bland–Altman analysis showed an overall mean bias (95% limits of agreement) between tracheal and esophageal temperatures of -0.10 °C (-0.37 °C to 0.18 °C), with a percentage error of 0.27%; between tracheal and PA temperatures, -0.05 °C (-0.91 °C to 0.20 °C), with a percentage error of -0.15%; and between esophageal and PA temperatures, 0.04 °C (-0.27 °C to 0.35 °C), with a percentage error of 0.12%. The concordance rates between tracheal and esophageal temperatures, tracheal and PA temperatures, and esophageal and PA temperatures were 96.2%, 96.2%, and 94.94%, respectively. The polar plot analysis showed a mean angular bias (radial limits of agreement) of 4° (26°), -3° (13°), and 2° (21°). Conclusions: Monitoring core temperature at the inner surface of the endotracheal tube cuff is accurate in all phases of LDLT with good trending ability; thus, it can be an excellent alternative for monitoring during LDLTs.