INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

CORE TEMPERATURE is one of the most important indexes for clinical practice and thermoregulatory research in humans. It is necessary that the gold standard for measuring core temperature responds to rapid changes in the core temperature. Esophageal (T_es) and tympanic (T_ty) temperatures have been used to estimate core temperature in clinical practice and thermoregulatory research. Benzinger (1, 2) proposed that T_ty reflected the core temperature because of the proximity of the tympanic membrane to the carotid artery, where it flows into the hypothalamus. Although some researchers agree with his opinion (6, 7, 18, 24), others do not accept that T_ty is a true measurement of core temperature because T_ty is affected by a decrease in skin temperature resulting from facial cooling or fanning (5, 14, 15, 20). In a study directly measuring brain temperature, Shiraki et al. (21) indicated that T_ty did not reflect the brain temperature during hyperthermia. On the other hand, Mariak et al. (12) suggested that the changes in brain temperature were better approximated by T_ty than by T_es. This discrepancy may result from the imperfections in the method used for measuring T_ty. Brinnel and Cabanac (6) and Sato et al. (18) reported that when a probe was successfully placed on the tympanic membrane, and the external canal was covered with cotton or a thermal insulating material, then T_ty was not affected by ambient temperature or by a decrease in skin temperature on the face or head.

Generally, the traditional methods for measuring T_es and T_ty require that the thermometry keeps in contact with the esophagus or tympanic membrane, respectively. However, these methods have some risks (spread infection, injure the esophagus or tympanic membrane, and cause subjects discomfort or pain). Infrared tympanic thermometry has been developed to solve these problems, as a method of measuring T_ty more safely and conveniently (19). Terndrup noted that this method has been mentioned in many publications since the late 1980s (see Ref. 23 for review). However, the accuracy and reliability of existing devices are not high enough to allow their use in clinical medicine and thermoregulatory research (3, 10, 13, 17, 18, 22, 23). The reasons for such inadequate performance in measuring T_ty are attributed to the characteristics of the infrared sensor and its size. An infrared sensor measures the average temperature in its "field of vision." Existing devices may measure the infrared radiation from the auditory canal as well as the tympanic membrane. Furthermore, existing devices are not suitable for continuous monitoring, although continuous information on the core temperature during experiments is valuable. We have developed an infrared tympanic thermometry with an optical fiber (optical fiber thermometry) to detect the limited amount of infrared radiation from the tympanic membrane. A chalcogenide glass fiber was used for the optical fiber, and this made it possible to detect the temperature of the tympanic membrane selectively.

In this study, we investigated whether the optical fiber thermometry accurately registered the temperature of the tympanic membrane during different experimental maneuvers. Head cooling and facial fanning tests were used to determine whether the temperature registered by the optical fiber thermometry was affected by a decrease in skin temperature. The passive-heat-stress and progressive-exercise tests were used to compare T_ty measured by the optical fiber thermometry (infrared-T_ty), T_ty measured with a thermistor (contact-T_ty), and T_es measured by a thermocouple. We also used the relationship between the rate of sweating and the two core temperatures in these experiments as an additional test of the utility of the optical fiber thermometry.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Subjects. Three healthy untrained male subjects participated in the head cooling and facial fanning tests, and these three and five other male subjects took part in the passive-heat-stress and progressive-exercise tests. Their mean age, height, weight, and body surface area were 22.8 ± 0.9 (SD) yr, 1.70 ± 0.05 m, 64.5 ± 9.8 kg, and 1.74 ± 0.14 m², respectively. Each subject was informed of the nature of the procedures and the risks associated with the experiments, and each signed an informed consent agreement when he first visited our laboratory.

Procedures. For the first group of tests, each subject sat on a chair in an environmental chamber (SR-3000, Nagano Science, Osaka, Japan) maintained at an ambient temperature of 25°C and 50% relative humidity. After a 5-min baseline period, the subject's temple was cooled by application of an ice pack or his forehead was cooled by air blown from an electrical fan. Cooling lasted for 15-min followed by a 10-min recovery period. The two tests were performed on different days.

Measurement of T_ty and T_es. In this study, two types of tympanic thermometry were used to measure T_ty. The first was an optical fiber thermometry, consisting of an infrared sensor and an optical fiber. The infrared sensor used was a pyroelectric infrared detector (DLATGS, Shimadzu). This infrared sensor can detect infrared energy independent of the wavelength of the radiation emitted by an object (11), but the incident infrared radiation must be interrupted at regular intervals. Therefore, this system is equipped with a mechanical chopper, inserted between the infrared sensor and the optical fiber (Fig. 1), which has two blades that rotate at 120 Hz.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Schematic illustration of left external auditory canal and infrared tympanic thermometry and optical fiber. Infrared rays from tympanic membrane were transmitted to a pyroelectric infrared sensor along a chalcogenide glass fiber. DC, direct current; A/D, analog to digital.

Measurement of skin temperatures, sweating rate, and heart rate. During the tests of head cooling and facial fanning, the skin temperature of the forehead and right temple was measured with the copper-constantan thermocouple. These temperatures were sampled every second. During the passive-heat-stress and progressive-exercise tests, the local sweating rate and heart rate were measured. Sweating rate was recorded continuously at two sites, the left side of the chest and the forearm, by the ventilated capsule method. Dry nitrogen gas was pumped through the capsules (chest: 8.54 cm², forearm: 5.31 cm²) at a rate of 1.5 l/min. The humidity of the nitrogen gas flowing out of the capsules was measured with a capacitance hygrometer (HMP 133Y, Vaisala, Finland). The skin temperatures and hygrometer output signals were stored in the same system used for measuring T_es and were calculated every minute. Heart rate was measured by using lead V₅ of an electrocardiogram.

Statistical analysis. A one-way analysis of variance was performed to assess the difference between calculated and measured values for the thermal parameters, e.g., the control values, the final values, T_core, and the slope of sweating rate vs. core temperature. Repeated two-way analysis of variance with repeated measures was used to assess the change in thermal parameters with time. Values of infrared-T_ty, T_es, and contact-T_ty sampled every 5 min were used for the statistical analysis. In addition, Sheffé's test was used to test treatment mean values among the parameters when there were significant differences. Statistical significance was set at P < 0.05 for all statistical tests.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

The results of the stability and linearity tests of the optical fiber thermometry, carried out with the blackbody, are shown in Fig. 2. The stability tests that used the blackbody calibrated at 37°C were carried out at an ambient temperature of 25°C for 60 min. There is no long-term drift in this system, and the maximum difference between the temperature the optical fiber thermometry measured and the temperature of the blackbody was ±0.2°C (Fig. 2A). The temperature measured by the optical fiber thermometry shows a good linear relationship with that of the blackbody (Fig. 2B).

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Results of blackbody stability test at constant temperature (A) and linear response between infrared tympanic thermometry using an optical fiber (optical fiber thermometry) and blackbody temperature (B).

The head cooling and facial fanning tests were used to investigate whether the optical fiber thermometry accurately measured only the limited amount of infrared radiation from the tympanic membrane. The skin temperatures on the right temple and forehead decreased remarkably when the ice pack was pressed against the right temporal region (Fig. 3A) or the electrical fan blew air in the subject's face (Fig. 3B). Infrared-T_ty and T_es, however, are not influenced by the decrease in these temperatures. The differences in infrared-T_ty between the control and final values in the cooling and fanning tests are 0.0026 ± 0.0480 (SD) and 0.0411 ± 0.0731°C, respectively.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   A: time course of tympanic temperature measured by an infrared tympanic thermometry with an optical fiber (infrared-Tty), esophageal temperature (Tes), and right-side temporal skin temperature (Ttemple) during head cooling test. Ice pack was positioned against right temporal area for 15-min after a 5-min baseline period. Values of infrared-Tty and Tes were 36.88 ± 0.03 (SD) and 36.72 ± 0.06°C for 3-5 min before cooling and were 36.93 ± 0.05 and 36.71 ± 0.07°C for 18-20 min during cooling. B: time course of infrared-Tty, Tes, and forehead skin temperature (Tforehead) during facial fanning test. Electrical fan with a 300-mm-diameter blade was placed 150 mm in front of the face. Facial skin was cooled for 15 min after a 5-min baseline period. Values of infrared-Tty and Tes were 36.75 ± 0.03 (SD) and 36.67 ± 0.05°C for 3-5 min before cooling and 36.75 ± 0.05 and 36.66 ± 0.06°C for 18-20 min during fanning.

No significant differences between infrared-T_ty, contact-T_ty, and T_es are observed for the control and final values and for T_core in the passive-heat-stress and progressive-exercise experiments (Table 1). During both experiments, there are no significant differences in the changes in infrared-T_ty, contact-T_ty, and T_es over time (Fig. 4). In the progressive-exercise experiment, there was a significant delay in the rise in infrared-T_ty and contact-T_ty compared with T_es. There was no significant difference in the onset of the elevation in infrared-T_ty, contact-T_ty, and T_es in the passive-heat-stress experiment (Table 1). The correlation between infrared-T_ty and contact-T_ty in both experiments is statistically significant [correlation coefficient: 0.82 ± 0.10 (passive-heat-stress test) and 0.87 ± 0.07 (progressive-exercise test)].

View larger version (32K): [in this window] [in a new window]   Fig. 4.   Comparison of core temperatures during passive-heat-stress (A) and progressive-exercise (B) tests. Time course of infrared-Tty, tympanic temperature measured with a thermistor (contact-Tty), and Tes during both experiments. One subject became exhausted 27 min after the start of exercise, so data plotted for all the subjects' data are for 27 min.

There are highly positive correlations, which range from 0.864 to 0.950, between sweating rate on the chest and forearm and the core temperatures. There are no significant differences in the slopes of sweating rate vs. infrared-T_ty and sweating rate vs. contact-T_ty at each site in both experiments. No significant differences in the threshold values of infrared-T_ty, contact-T_ty, and T_es for sweating were observed for either site.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Although many thermometries have been developed and used to measure core temperature, use of these devices involves great risks, technical difficulties, and controversies (4). Recently, a method of infrared tympanic thermometry that did not involve contact with the tympanic membrane was developed to measure T_ty safely and easily. The accuracy of predecessors has been questioned (3, 10, 13, 17, 22). The insufficient accuracy is caused by the fact that these devices may measure the temperature of the auditory canal. To prevent this, we used a chalcogenide glass fiber to detect infrared radiation solely from the tympanic membrane (Fig. 1). Because the optical fiber thermometry accurately measured the regulated temperature of a blackbody (Fig. 2), it is believed that the optical fiber thermometry measurements are stable and have a linear response, making the optical fiber suitable for thermometry.

Because the tympanic membrane is near the point where the carotid artery enters the hypothalamus, T_ty is believed to reflect the brain and core temperatures (2). However, there were several problems with methods of measuring T_ty. Some authors have reported that T_ty did not reflect the core temperature, because it was influenced by changes in the ambient temperature and by cooling of the head or face (4, 14, 15). Sato et al. (18) reported that T_ty fell by as much as 0.6°C during head cooling by an ice pack when the sensor was not in contact with the tympanic membrane. When the sensor was carefully positioned and insulated, T_ty was not affected by the decrease in head and face skin temperatures and was recommended as the core temperature (6, 7, 18). Therefore, the head cooling and facial fanning tests in normothermic subjects were used to determine whether the optical fiber thermometry accurately measured T_ty. Figure 3 shows that infrared-T_ty was not affected by the decreased skin temperatures during the head cooling or facial fanning tests, when the same methods as those of Sato et al. were used. The results of the head cooling and facial fanning suggested that the tip of the optical fiber was close to the tympanic membrane, although this system cannot determine the critical distance between the tympanic membrane and the tip of the optical fiber. Calculating from the diameter of the tympanic membrane (~1 mm) and the visual fields of the optical fiber (~45°), the distance might be <2.5 mm. In the passive-heat-stress experiment, infrared-T_ty, contact-T_ty and T_es showed almost identical responses (Fig. 4): no significant differences among infrared-T_ty, contact-T_ty, and T_es are observed for the control and final values, T_core, and the onset of the rise in core temperatures (Table 1). In the progressive-exercise experiment, the only difference among infrared-T_ty, contact-T_ty, and T_es was in the onset of the rise in core temperature. However, there were no significant differences in the change in infrared-T_ty, contact-T_ty, and T_es over time in both experiments (Fig. 4). Earlier studies (8, 9, 15) also reported T_es increased before T_ty during exercise. In both experiments, infrared-T_ty and contact-T_ty showed almost identical responses, and the responses between T_ty (infrared-T_ty and contact-T_ty) and T_es were similar to earlier studies (8, 9, 15, 18). These results suggest that the temperature measured by the optical fiber thermometry is the T_ty and that the optical fiber thermometry can be recommended as a device for measuring T_ty.

Because our system uses the infrared sensor, it was necessary to determine whether the polyethylene sponge, attached to the optical fiber and used to insulate the auditory canal, held the fiber close to the tympanic membrane throughout exercise. Two subjects had a low correlation between infrared-T_ty and contact-T_ty during exercise. These subjects reported that they felt that the fiber was slipping out of their auditory canals. It is necessary to improve the method of attaching the fiber, perhaps by fashioning an attachment similar to a hearing aid. On the other hand, this system accurately measured T_ty, even in three subjects whose heart rate reached 200 beats/min at the end of the exercise, suggesting that the optical fiber thermometry could accurately measure the T_ty during intense exercise when the fiber was positioned in external canal.

Thermoregulatory sweating responses are generally assessed by two parameters, the core temperature threshold for the onset of sweating and the slope of the relationship between core temperature and the rate of sweating (i.e., sensitivity). We used these two parameters to investigate whether infrared-T_ty could be used to evaluate the thermoregulatory sweating responses as well as contact-T_ty and T_es. In this study, there was a remarkable positive relationship between the core temperatures measured by the three methods and sweating rate on the chest and forearm. In both experiments, there were no significant differences in the sensitivity of measurement determined by sweating rate vs. infrared-T_ty and sweating rate vs. contact-T_ty at each site. Furthermore, there were no significant differences in infrared-T_ty, contact-T_ty, and T_es thresholds for sweating in both experiments (Table 1), although the increase in infrared-T_ty and contact-T_ty lagged behind the rise in T_es during exercise (Fig. 4). Consequently, these results also suggest that infrared-T_ty accurately reflects the temperature that determines the thermoregulatory sweating response and can be used in thermoregulatory research and clinical medicine.

In conclusion, the optical fiber thermometry performs extremely well, and it accurately measures the T_ty. Head cooling and facial fanning tests proved that the optical fiber thermometry accurately measured only the limited amount of infrared radiation from the tympanic membrane. The changes in infrared-T_ty during the passive-heat-stress and progressive-exercise experiments were the same as those in contact-T_ty. Furthermore, the changes in infrared-T_ty and contact-T_ty were similar to those in T_es during both experiments. There were also no significant differences between infrared-T_ty and contact-T_ty and the changes in the two parameters for monitoring sweating responses (threshold temperature and sensitivity). Therefore, we can recommend the optical fiber thermometry as a device for continuously measuring T_ty.