Intraoperative zero-heat-flux thermometry overestimates nasopharyngeal temperature by 0.39 °C: an observational study in patients undergoing congenital heart surgery

Abstract

During surgery for congenital heart disease (CHD) temperature management is crucial. Vesical (T_ves) and nasopharyngeal (T_NPH) temperature are usually measured. Whereas T_ves slowly responds to temperature changes, T_NPH carries the risk of bleeding. The zero-heat-flux (ZHF) temperature monitoring systems SpotOn™ (T_SpotOn), and Tcore™ (T_core) measure temperature non-invasively. We evaluated accuracy and precision of the non-invasive devices, and of T_ves compared to T_NPH for estimating temperature. In this prospective observational study in pediatric and adult patients accuracy and precision of T_SpotOn, T_core, and T_ves were analyzed using the Bland-Altman method. Proportion of differences (PoD) and Lin´s concordance correlation coefficient (LCC) were calculated. Data of 47 patients resulted in sets of matched measurements: 1073 for T_SpotOn vs. T_NPH, 874 for T_core vs. T_NPH, and 1102 for T_ves vs. T_NPH. Accuracy was - 0.39 °C for T_SpotOn, -0.09 °C for T_core, and 0.07 °C for T_ves. Precisison was between - 1.12 and 0.35 °C for T_SpotOn, -0.88 to 0.71 °C for T_core, and - 1.90 to 2.05 °C for T_ves. PoD ≤ 0.5 °C were 71% for T_SpotOn, 71% for T_core, and 60% for T_ves. LCC was 0.9455 for T_SpotOn, 0.9510 for T_core, and 0.9322 for T_ves. Temperatures below 25.2 °C (T_SpotOn) or 27.1 (T_core) could not be recorded non-invasively, but only with T_ves. Trial registration German Clinical Trials Register, DRKS00010720.

Keywords: Congenital heart disease; Core temperature; Hypothermia; Non-invasive monitoring; Zero-heat-flux thermometer.

Fig. 1

Flowchart of enrollment of patients

Fig. 2

(a) Temperatures measured with SpotOn™ (T_SpotOn), Tcore™ (T_core), vesical (T_ves) and nasopharyngeal (T_NPH) temperature probes in a 2-day-old patient. During cardiopulmonary bypass temperature was measured in the arterial branch of the oxygenator (T_CPB). (b) Temperatures measured with SpotOn™ (T_SpotOn), Tcore™ (T_core), vesical (T_ves) and nasopharyngeal (T_NPH) temperature probes in a 15-year-old patient. During cardiopulmonary bypass temperature was measured in the arterial branch of the oxygenator (T_CPB)

Fig. 3

(a) Bland-Altman plot with multiple temperature measurements (47 patients with 1073 measurement pairs) of the SpotOn^TM-Sensor (SOT) and a nasopharyngeal probe (Nasal). Blue line indicates mean bias with 95% confidence interval (CI): -0.39 °C (-0.48; -0.30); red lines indicate 95% limits of agreement (LoA) with 95% confidence interval (CI). LoA: ±0.74 °C, lower LoA: -1.12 °C (-1.22; -1.04), upper LoA: 0.35 °C (0.26; 0.44). (b) Bland-Altman plot with multiple temperature measurements (39 patients with 874 measurement pairs) of the Tcore^TM-Sensor (T_core) and a nasopharyngeal probe (T_NPH). Blue line indicates mean bias with 95% confidence interval (CI): -0.09 °C (-0.23; 0.05); red lines indicate 95% limits of agreement (LoA) with 95% confidence interval (CI). LoA: ±0.80 °C, lower LoA: -0.88 °C (-1.04; -0.75), upper LoA: 0.71 °C (0.58; 0.87). (c) Bland-Altman plot with multiple temperature measurements (47 patients with 1102 measurement pairs) of a vesical (T_ves) and a nasopharyngeal probe (T_NPH). Blue line indicates mean bias with 95% confidence interval (CI): 0.07 °C (0.09; 0.24); red lines indicate 95% limits of agreement (LoA) with 95% confidence interval (CI). LoA: ±1.98 °C, lower LoA: -1.90 °C (-2.06; -1.74), upper LoA: 2.05 °C (1.89; 2.21)

Fig. 4

(a) Bland-Altman plot with multiple temperature measurements of the SpotOn^TM-Sensor (SOT) and a nasopharyngeal probe (Nasal). Blue line indicates mean bias with 95% confidence interval (CI): -0.45 °C (-0.60; -0.30); red lines indicate 95% limits of agreement (LoA) with 95% confidence interval (CI). LoA: ±0.78 °C, lower LoA: -1.23 °C (-1.39; -1.09), upper LoA: 0.32 °C (0.18; 0.49). (b) Bland-Altman plot with multiple temperature measurements of the Tcore^TM-Sensor (T_core) and a nasopharyngeal probe (T_NPH). Blue line indicates mean bias with 95% confidence interval (CI): -0.13 °C (-0.38; 0.11); red lines indicate 95% limits of agreement (LoA) with 95% confidence interval (CI). LoA: ±0.89 °C, lower LoA: -1.02 °C (-1.33; -0.79), upper LoA: 0.75 °C (1.06; 0.52). (c) Bland-Altman plot with multiple temperature measurements of a vesical (T_ves) and a nasopharyngeal probe (T_NPH). Blue line indicates mean bias with 95% confidence interval (CI): 0.05 °C (-017; 0.26); red lines indicate 95% limits of agreement (LoA) with 95% confidence interval (CI). LoA: ±2.00 °C, lower LoA: -1.96 °C (-2.18; -1.76), upper LoA: 2.05 °C (1.85; 2.28)

Fig. 5

(a) Bland-Altman plot with multiple temperature measurements of the SpotOn^TM-Sensor (SOT) and a nasopharyngeal probe (Nasal). Blue line indicates mean bias with 95% confidence interval (CI): -0.34 °C (-0.44; -0.23); red lines indicate 95% limits of agreement (LoA) with 95% confidence interval (CI). LoA: ±0.70 °C, lower LoA: -1.03 °C (-1.15; -0.93), upper LoA: 0.36 °C (0.26; 0.47). (b) Bland-Altman plot with multiple temperature measurements of the Tcore^TM-Sensor (T_core) and a nasopharyngeal probe (T_NPH). Blue line indicates mean bias with 95% confidence interval (CI): -0.05 °C (-0.21; 0.11); red lines indicate 95% limits of agreement (LoA) with 95% confidence interval (CI). LoA: ±0.73 °C, lower LoA: -0.77 °C (-0.95; -0.62), upper LoA: 0.68 °C (0.53; 0.86). (c) Bland-Altman plot with multiple temperature measurements of a vesical (T_ves) and a nasopharyngeal probe (T_NPH). Blue line indicates mean bias with 95% confidence interval (CI): 0.10 °C (-0.14; 0.34); red lines indicate 95% limits of agreement (LoA) with 95% confidence interval (CI). LoA: ±1.94 °C, lower LoA: -1.84 °C (-2.08; -1.61), upper LoA: 2.04 °C (1.82; 2.29)