Abstract
Background
There are several sites for measuring body temperature. Correct reading of core temperature is imperative for patients undergoing major operations under anesthesia. In certain situations, the sites of measurement may be close to the surgical area, and thus the measurement is easily prejudiced by the influence environment. We hypothesized that the body temperature, if monitored in the esophagus, would be lower than obtained from the tympanic membrane during thoracotomy for lung pathology under general anesthesia.
Materials and methods
The study involved 32 patients, of American Society of Anesthesiologists (ASA) physical status I or II, who were to undergo elective thoracotomy for lung disorders. General anesthesia was induced with fentanyl, propofol, and rocuronium and maintained with sevoflurane in oxygen. The tympanic membrane probe was placed prior to when general anesthesia was administered, and the esophageal probe was inserted after administration of general anesthesia. Both the individualized temperatures were recorded at 5-minute intervals, and were compared at each change of surgical situation.
Results
The tympanic membrane temperature was higher than esophageal temperature after initiation of one-lung ventilation (OLV) with statistical significance. The magnitude of decrease in temperature between two individualized temperatures, as compared from start of OLV, was greater in tympanic membrane temperature, especially at 30 minutes after OLV (p < 0.02, difference = –0.09 ± 0.22) and at the time point of the lowest temperature (p = 0.002, difference = −0.14 ± 0.24). There was no clinical difference of situation found (difference > 0.5°C) in the measuring sequences.
Conclusion
The accuracy of esophageal temperature seemed not to be affected during thoracotomy for lung lesion, in comparison with that of tympanic temperature. From clinical viewpoints, the monitoring of esophageal temperature could be more reliable in such surgical situation.
Keywords
esophagus; one-lung ventilation; temperature; thoracotomy;
1. Introduction
Accurate monitoring of core body temperature is important in patients under general anesthesia, because of the need for early detection of malignant hyperthermia or hypothermia due to loss of body heat. Temperature disturbances could cause numerous complications, including wound infection, coagulopathy, delayed wound healing, and prolonged postanesthetic recovery.1, 2, 3
The tympanic membrane, which is almost regarded as an ideal site for measurement,4 is located in close proximity to the brain and considered to rationally reflect the brain temperature more accurately. However, core temperature also can be measured at various sites, such as the lower esophagus, nasopharynx, pulmonary artery, rectum, and tympanic membrane.5
Temperature measurement taken at the distal esophagus is reliable because it is unaffected by even extreme airway cooling.6 However, during liver transplantation,7 esophageal temperature monitoring can give a seemingly correct but really incorrect core temperature reading because of the close proximity of the temperature probe to the diaphragm, which is subject to extremely cooling.
During thoracotomy, core temperature monitoring is important, and the measurement is usually taken from the tympanic membrane or the esophagus. Both sites are regarded as core temperature compartments; however, the esophagus is close to the site of thoracotomy for lung lesion. We hypothesized that esophageal temperature would be lower than the tympanic temperature during thoracotomy and the difference of the two temperatures, if exceeding 0.5°C, could be regarded as having clinical importance. In this prospective study, we investigated the validity of esophageal temperature by comparing the core temperature readings taken at the distal esophagus and at the tympanic membrane during thoracotomy in the lateral position.
2. Materials and methods
This study was approved by the Institutional Review Board of the China Medical University Hospital, Taiwan. The study group comprised 32 patients of American Society of Anesthesiologists (ASA) physical status I or II, undergoing thoracic surgery for lung disorders in the lateral decubitus position.
Exclusion criteria were hypothermia (<36°C), hyperthermia (>38°C), significant esophageal pathology, and nasogastric tube insertion. The measurement of tympanic membrane temperatures might be less accurate in the accumulation of cerumen or in the presence of otologic disease; therefore, under direct otoscopy we removed the cerumen prior to inserting the probe for measuring tympanic membrane temperature. Each patient was kept warm with a circulating-water mattress at 40°C until he or she left the operating room. The temperature of the operating room was maintained at 22 ± 1°C. Anesthetic gases were not intentionally warmed and the intravenous fluids were infused at room temperature. Crystalloid fluid was maintained using the 4:2:1 rule (4 mL/kg for the first 10 kg, 2 mL/kg for the next 10 kg, and 1 mL/kg for each additional 10 kg), and adjusted by hemodynamic status. Tympanic membrane temperature was measured prior to the administration of general anesthesia with placement of the aural probe while the patient was awake, and the probe was fixed firmly when the patient felt that the probe had touched the tympanic membrane. Esophageal temperature was measured with the adult esophageal/rectal reusable temperature probe, which was en route through a nostril to the esophagus after general anesthesia; the insertion length of the probe was calculated according to the following formula: length (cm) = 0.228 × standing height (cm) − 0.194.8
All patients received fentanyl (1–1.5 μg/kg), propofol (1.5–2.5 mg/kg), and rocuronium (0.5–0.6 mg/kg) for induction of general anesthesia. General anesthesia was maintained with sevoflurane in 50–100% oxygen, and the fresh gas flow was maintained at a constant rate of 2 L/min throughout the surgery. Patients were mechanically ventilated at a tidal volume of 8–10 mL/kg to maintain normocapnia (end-tidal carbon dioxide = 30–35 mm Hg). All core temperatures were recorded at 5-minute intervals throughout the surgery.
The paired Student t test was used to assess temperature differences and variations at two locations (tympanic membrane and esophagus) at each time point and time events. Furthermore, the mixed model was used to evaluate the effect of locations at which temperature was measured and the times of temperature readings after adjusting for age, sex, weight, height, and the side of the surgical site (left or right). The dependent variable was core temperature. The within-factors were locations and time points (or time events). The between-factors and covariates were age, sex, weight, height, and the side (left or right) on which the surgery was performed. Adjusted means were reported from mixed model. Simple effects analysis combined with Bonferroni-type multiple comparisons was performed to assess the effects of the time points (or time events) and locations (tympanic membrane and esophagus). SAS 9.1 (SAS Institute, Cary, NC) software was used for all statistical analysis.
3. Results
The demographics (age, sex, weight, height, fluid infusion, chest surgical-site, and duration of surgery) are shown in Table 1. The tympanic membrane temperature was overall higher than esophageal temperature after initiation of one-lung ventilation (OLV), inclusive of the lowest temperatures (p < 0.001), the one taken at the end of OLV (p < 0.001), and the one taken at the end of surgery (p = 0.004; Table 2). Additionally, similar results were found in the following time points after OLV, including the temperatures at 30 minutes (p < 0.001), 60 minutes (p < 0.001), 90 minutes (p < 0.001), and 120 minutes (p = 0.007) after OLV (Table 3). Taking the temperature at the start of OLV as a standard, similar results (higher temperature in tympanic membrane) were also found in time events and time points by mixed model to adjust parameters (Figs. 1 and 2). Treating the tympanic membrane as the standard, the difference of decreasing temperature variance, as compared with that at the start of OLV, was smaller in esophageal temperature in the event of the lowest temperature (p = 0.002, difference = –0.14 ± 0.24) and 30 minutes after OLV (p = 0.02, difference = – 0.09 ± 0.22; Tables 2 and 3). There was no temperature difference > 0.5°C in the measurements in time sequences.
Download full-size image
Download full-size image
4. Discussion
We found that the esophageal temperature was significantly lower than the tympanic temperature from 30 minutes after starting OLV until the end of surgery, inclusive of serial time points (30 minutes, 60 minutes, and 90 minutes after starting OLV) and sequent measurement points (the lowest temperature reading, the end of OLV, and the end of surgery). Therefore the temperature readings were lower in the esophagus than in the tympanic membrane in thoracotomy.
There were no differences in tympanic membrane and esophageal temperatures prior to opening the chest, even though they could not be measured simultaneously prior to anesthesia. Insertion of a temperature probe into the esophagus is uncomfortable while the patient is awake; therefore, we usually insert the probe after administration of general anesthesia. The core thermal compartments are defined as well-perfused tissue spaces in which the temperature is higher than other tissue parts of the body,9 and would change faster than that of the peripheral thermal compartments. Therefore, the tympanic membrane, central blood, nasopharyngeal space, urinary bladder, and esophageal temperatures are regarded as good sites for monitoring core temperature.10 Indeed, we found that the decrease in core temperature was larger in the compartment of the tympanic membrane. Unlike the esophagus, the tympanic membrane is located away from the surgical site in thoracotomy.
In general anesthesia, the esophageal temperature is a practical choice for core temperature during most thoracotomized procedures. However, the esophageal temperature readings can be affected by malposition of the temperature probe or by placing the probe too close to the surgical site.7, 9, 11Although there were some reports that showed a close correlation between tympanic and esophageal temperatures,12, 13, 14 in other studies the esophageal temperature readings were lower than or parallel to the tympanic temperature readings.15, 16, 17 In our study, we provided extreme conditions to determine whether the so-called core temperature would be affected, and would the difference be large enough (defined as 0.5°C) to affect our clinical practice.
The esophageal temperature probe, which stayed outward in OLV in our patients undergoing thoracotomy, caused the core thermal compartments to be exposed outside. Therefore, in such a situation, we assumed that the temperatures taken in the esophagus might be affected by ambient temperature, which is much lower than that of innards or by surgical manipulations and inclusive of water irrigation and contact with cool surgical instruments. During the period of thoracotomy, the temperatures of all core thermal compartments would change simultaneously and have similar variations. In addition to comparison of different patients in the same condition, we showed two different results in the event points and time sequence points to standardize the personal conditions.
OLV is necessary for manipulations in traditional transthoracic esophagectomy, and provides direct contact to the esophagus.18 Therefore, the OLV imposes the exposure of the esophagus to the ambient environment, and the decreasing level of temperature reading is greater in the esophagus in comparison with that of the tympanic membrane. It has been reported that the reading of esophageal temperature may be faulty during thoracotomy because the thoracic cavity is opened and exposed to ambient air.5 Russell and Freeman7 pointed out that the temperature measured in the lower third of the esophagus might be affected by the cold liver proximal to the tip of the esophageal temperature probe. However, there was no comparison between different sites of core thermal compartments, and the degree of decreased esophageal temperature was not mentioned in their study.
During thoracotomy, the esophageal temperature probe is more proximal to the colder ambient air than to the upper abdominal compartment, and other core thermal compartments such as the tympanic membrane are not affected by this alteration. Nevertheless, in our study, the esophageal temperature was lower than the tympanic temperature and there was a minor statistical but not clinical change after OLV was started (>0.5°C).
Our results indicated that the reading of core temperature would not be affected, even if the position of the temperature probe was near the site of thoracotomy. In addition, temperature of the tympanic membrane and the esophagus reliably indicated core temperature during thoracotomy.
In conclusion the placement of the temperature probe was not influenced by extrinsic factors in thoracotomy for pulmonary surgery. However, subclinical differences between esophageal and tympanic temperatures were found after the start of OLV.
Acknowledgments
The authors would like to thank W.M. Liang, PhD, Public Health, China Medical University, Taichung, Taiwan, for her help with statistical analysis. This project was supported by a grant from China Medical University Hospital (DMR-98-082).