Jay Ollie, Molgat-Seon Yannick, Chou Shirley, Murto Kimmo
Faculty of Health Sciences, Thermal Ergonomics Laboratory, University of Ottawa, Ottawa, ON, Canada; Children's Hospital of Eastern Ontario Research Institute, Ottawa, ON, Canada.
Paediatr Anaesth. 2013 Dec;23(12):1109-16. doi: 10.1111/pan.12262. Epub 2013 Sep 25.
The accurate measurement of core temperature is an essential aspect of intraoperative management in children. Invasive measurement sites are accurate but carry some health risks and cannot be used in certain patients. An accurate form of noninvasive thermometry is therefore needed. Our aim was to develop, and subsequently validate, separate models for estimating core temperature using different skin temperatures with an individualized correction factor.
Forty-eight pediatric patients (0-36 months) undergoing elective surgery were separated into a modeling group (MG, n = 28) and validation group (VG, n = 20). Skin temperature was measured over the carotid artery (Tsk_carotid ), upper abdomen (Tsk_abd ), and axilla (Tsk_axilla ), while nasopharyngeal temperature (Tnaso ) was measured as a reference.
In the MG, derived models for estimating Tnaso were: Tsk_carotid + 0.52; Tsk_abd + (0.076[body mass] + 0.02); and Tsk_axilla + (0.081[body mass]-0.66). After adjusting raw Tsk_carotid, Tsk_abd , and Tsk_axilla values in the independent VG using these models, the mean bias (Predicted Tnaso - Actual Tnaso [with 95% confidence intervals]) was +0.03[+0.53, -0.50]°C, -0.05[+1.02, -1.07]°C, and -0.06[+1.21, -1.28°C], respectively. The percentage of values within ±0.5°C of Tnaso was 93.2%, 75.4%, and 66.1% for Tsk_carotid, Tsk_abd , and Tsk_axilla , respectively. Sensitivity and specificity for detecting hypothermia (Tnaso < 36.0°C) was 0.88 and 0.91 for Tsk_carotid , 0.61 and 0.76 for Tsk_abd , and 0.91 and 0.73 for Tsk_axilla . Goodness-of-fit (R(2) ) relative to the line-of-identity was 0.74 (Tsk_carotid ), 0.34 (Tsk_abd ), and 0.15 (Tsk_axilla ).
Skin temperature over the carotid artery, with a simple correction factor of +0.52°C, provides a viable noninvasive estimate of Tnaso in young children during elective surgery with a general anesthetic.
准确测量核心体温是儿童术中管理的重要环节。有创测量部位虽准确，但存在一定健康风险，且某些患者无法使用。因此，需要一种准确的无创测温方法。我们的目标是开发并随后验证使用不同皮肤温度及个体化校正因子来估计核心体温的独立模型。
48例接受择期手术的儿科患者（0 - 36个月）被分为建模组（MG，n = 28）和验证组（VG，n = 20）。测量颈动脉处（Tsk_carotid）、上腹部（Tsk_abd）和腋窝（Tsk_axilla）的皮肤温度，同时测量鼻咽温度（Tnaso）作为参考。
在建模组中，估计Tnaso的推导模型为：Tsk_carotid + 0.52；Tsk_abd +（0.076[体重] + 0.02）；以及Tsk_axilla +（0.081[体重] - 0.66）。使用这些模型对独立验证组中的原始Tsk_carotid、Tsk_abd和Tsk_axilla值进行调整后，平均偏差（预测Tnaso - 实际Tnaso[95%置信区间]）分别为+0.03[+0.53, -0.50]°C、-0.05[+1.02, -1.07]°C和-0.06[+1.21, -1.28]°C。Tsk_carotid、Tsk_abd和Tsk_axilla在Tnaso±0.5°C范围内的值的百分比分别为93.2%、75.4%和66.1%。检测体温过低（Tnaso < 36.0°C）时，Tsk_carotid的敏感性和特异性分别为0.88和0.91，Tsk_abd为0.61和0.76，Tsk_axilla为0.91和0.73。相对于恒等线的拟合优度（R²）分别为0.74（Tsk_carotid）、0.34（Tsk_abd）和0.15（Tsk_axilla）。
在全身麻醉下进行择期手术的幼儿中，颈动脉处的皮肤温度加上简单的校正因子+0.52°C，可提供对Tnaso可行的无创估计。
