Inadvertent perioperative hypothermia is relatively common in patients undergoing general anesthesia and may significantly impact patient outcomes adversely [1]. The risk of perioperative hypothermia is higher in pediatric patients than in adult patients, owing to their proportionately larger body surface area and impaired physiological compensatory mechanisms [2,3]. Adequate core temperature measurement is a prerequisite for detecting and controlling core temperature in pediatric patients during surgery.

Core temperature can be measured at various sites using different techniques. The gold standard for thermoregulation in pediatric patients remains rectal temperature (T_R) measurement [4]. However, this technique can cause complications such as rectal perforation, increased blood pressure, and decreased partial pressure of oxygen [5]. As an alternative, esophageal temperature (T_E), which allows continuous monitoring with minimal adverse effects, is commonly employed in pediatric patients.

The supraglottic airway device (SAD) is widely used in patients undergoing general anesthesia. However, its use makes measuring T_E challenging because it fills the oral cavity. Most second-generation SADs, including AuraGain, i-Gel, Supreme, and ProSeal laryngeal mask airways, which are commonly used nowadays, have a gastric channel that facilitates access to the esophagus and stomach [6]. The gastric channel provides a passage for T_E probe insertion, enabling gastric decompression through removal of air in the stomach resulting from mask ventilation.

In this study, we aimed to assess the feasibility and accuracy of core temperature measurement using a T_E probe inserted through the gastric lumen of a second-generation SAD in pediatric patients. Furthermore, we aimed to examine correlations between T_E and tympanic membrane (T_TM) and forehead (T_ZHF) cutaneous temperatures in pediatric patients, using T_R as the reference. Additionally, we assessed the gastric decompression effect obtained by performing suction through the T_E probe inserted through the SAD.

A total of 67 pediatric patients, aged < 6 years and weighing 10–30 kg, who were scheduled to undergo general anesthesia, were assessed for eligibility. Among them, one who had a congenital anomaly of the rectum and 24 whose guardians declined participation were excluded. Subsequently, the remaining 42 patients were enrolled. However, one patient was excluded from the final analysis due to an error in the thermometer device. The STROBE flow diagram of patient selection and dropout is presented in Fig. 2. The patients’ demographic characteristics and perioperative variables are presented in Table 1. Table 2 shows the range of body temperature data according to the measurement sites during anesthesia.

In all patients, the esophageal probe was successfully inserted through the gastric channel of the SAD. The results of the equivalence test are presented in Table 3, and the Bland–Altman plots illustrating the relationships of T_E, T_TM, and T_ZHF with T_R are shown in Fig. 3. When considering the initial temperature measurements, the smallest difference was observed with T_E, while the largest difference was noted with T_TM, using T_R as a reference. Overall, the minimal difference between temperatures was observed between measurements in the esophagus and rectum during the 0–40-min period, and between the tympanic membrane and rectum from 50–60-min period. When agreement was examined with T_R as a reference, T_E demonstrated equivalent results to T_R (P < 0.001 at 0, 10, 20, 30, and 40-min intervals and P = 0.018 at 50-min intervals), except at the 60-min interval (P = 0.697). Compared with T_R, T_TM was not equivalent until 30 min after surgery (P > 0.999 at 0, 10, and 20-min intervals and P = 0.084 at the 30-min interval) and then showed equivalent results until the end of the surgery (P = 0.006, P = 0.025, P = 0.013 at the 40, 50, 60-min intervals). When T_ZHF was compared with T_R as a reference, there was no difference from 10 min to 40 min after the surgery (P = 0.216, P < 0.001, P = 0.003, P < 0.001, P = 0.048, P > 0.999, P > 0.999 at 10-min intervals from 0 to 60 min).

Analysis of the intraclass correlation coefficients (ICCs) at each time point revealed that T_E exhibited good reliability with T_R as the reference throughout the intraoperative period (ICC > 0.75, Table 4). Both T_ZHF and T_TM exhibited reliability exceeding the moderate level intraoperatively; however, T_E exhibited the best reliability among the three measurement methods.

In 33 of the 41 included patients, gastric suction through the T_E probe effectively reduced gastric air volume (80.5%). Gastric decompression was not clearly observed on sonography in the remaining eight patients.

We observed T_E measurement using a second-generation SAD to be highly accurate and reliable in pediatric patients undergoing general anesthesia, with T_R as a reference, and it was feasible in all patients. Moreover, effective reduction of gastric air volume was achieved through suction via the gastric channel of the second-generation SAD.

Accurate intraoperative temperature monitoring is important, especially in children. Several guidelines exist for perioperative temperature monitoring, but none of them specify the best device or site for temperature measurement [15,16]. To ensure patient safety, an ideal temperature monitoring technique should accurately reflect the core temperature while being non-invasive [17]. The device or site used for measuring body temperature is selected based on the type of surgery, accessibility of monitoring sites, and physician’s preference. To the best of our knowledge, this is the first study investigating the most appropriate temperature measurement device and site for accurately determining the core temperature with the utilization of SAD in pediatric patients undergoing surgery.

Our study demonstrated that T_E measurements obtained using a SAD exhibited higher agreement and reliability over an extended period than T_TM or T_ZHF measurements when T_R was used as a reference. This could be attributed to the fact that the distal esophagus receives sufficient blood perfusion from the core, resulting in more accurate and consistent temperature measurements than those at other sites. However, when T_R serves as the reference point in an equivalent test, accuracy of T_E decreases towards the end of the surgery. Conversely, T_TM may be more accurate toward the end of surgery, as its accuracy is maintained throughout the later stages. The tympanic membrane, which is located near the hypothalamus, can indeed reflect core temperature [15]. However, accurately inserting the temperature probe can be challenging, particularly in pediatric patients with a small aural canal. Furthermore, concerns exist regarding the feasibility of continuous measurement, reliance on the operator’s skill for determining accuracy, and variability of measurements [18]. These limitations prevent precise core temperature measurement using this method [19]. Other monitoring devices, such as zero-heat-flux cutaneous thermometers, have an appealing theoretical basis; however, inconsistent accuracy has been noted, both in our study and in previous studies [19,20].

SADs have gained popularity as advantageous airway management tools in pediatric patients [21]. However, the absence of space for temperature probe insertion in previous SADs necessitated its insertion through the nasopharynx [6]. Consequently, accurate body temperature measurement was challenging due to the cooling effect of anesthetic gases and the potential for procedural trauma during probe insertion into the nasopharynx [22,23]. In a previous study, body temperature measurements in patients with SAD insertion primarily focused on pharyngeal measurements [24]; however, the incorporation of a gastric channel in second-generation SADs for improved esophageal access enables accurate temperature measurement in the distal esophagus and smooth insertion of temperature probes.

In contrast to tracheal intubation, utilization of SADs may lead to increased gastric insufflation in a pressure-dependent manner [25,26]. Moreover, even minor malpositioning of the SAD can result in air leakage from its seal, causing air to flow into the esophagus [27]. Gastric insufflation may increase the risk of pulmonary aspiration of gastric contents and postoperative nausea and vomiting, particularly in high-risk pediatric patients, such as those with an impaired ability to protect their airways [28,29]. The gastric channel of second-generation SADs offers the additional advantage of effectively allowing the suctioning of air that might be introduced into the stomach during mask ventilation [30].

The first limitation of our study pertains to the accuracy of the T_E probe placement in routine clinical practice. The sensor is intended to be placed in the distal esophagus close to the pulmonary artery, and this is confirmed using sonography [31]. A previous study has demonstrated that the accuracy of T_E measurement increases when the probe is appropriately positioned [32]. However, performing sonography in routine clinical practice is challenging, which might reduce the accuracy compared with that in our study. The second limitation is that our findings are applicable only within a limited range of temperature variations. While events such as tourniquet deflation and cardiopulmonary bypass application could induce rapid and extreme changes in body temperature, such events were not examined in this study.

In conclusion, we demonstrated that T_E measurement through a second-generation SAD is feasible and accurate, making it suitable for routine use in pediatric patients with SAD insertion. Moreover, utilization of suction through the gastric channel of a second-generation SAD effectively reduces gastric air volume.