Investigation of gender difference in human response to temperature step changes

Physiology & Behavior 151 (2015) 426–440

Contents lists available at ScienceDirect

Physiology & Behavior journal homepage: www.elsevier.com/locate/phb

Investigation of gender difference in human response to temperature step changes Jing Xiong a, Zhiwei Lian a,⁎, Xin Zhou a, Jianxiong You b, Yanbing Lin b a b

Department of Architecture, School of Naval Architecture, Ocean & Civil Engineering, Shanghai Jiao Tong University, Shanghai, PR China The Third People's Hospital Affiliated to School of Medicine, Shanghai Jiao Tong University, Shanghai, PR China

H I G H L I G H T S • • • •

Women lowered oral temperatures more intensively than men after up-step of S15. Men witnessed a more remarkable decrease in RMSSD after up-step of S15 than women. Men regulated skin temperatures more robustly and swiftly than women. Women felt lower temperature cooler and higher temperature warmer than men.

a r t i c l e

i n f o

Article history: Received 1 July 2015 Received in revised form 22 July 2015 Accepted 29 July 2015 Available online 8 August 2015 Keywords: Temperature steps Gender difference Subjective perception Biomarkers Physiological parameters

a b s t r a c t The purpose of this study was to examine gender difference in human response to temperature step changes. A total of three step-change conditions (S5: 32 °C–37 °C–32 °C, S11: 26 °C–37 °C–26 °C, and S15: 22 °C–37 °C–22 °C) were designed and a laboratory experiment with 12 males and 12 females was performed. Results of this study support our hypothesis that females differ from males in human response to sudden temperature changes from the perspectives of psychology, physiology and biomarkers. Females are more prone to show thermal dissatisfaction to cool environments while males are more likely to feel thermal discomfort in warm environments. It is logical that men have a stronger thermoregulation ability than women as male skin temperature change amplitude is smaller while the time to be stable for skin temperature is shorter than that of females after both up-steps and down-steps. In S15, males witnessed a more intensive decrease in RMSSD while females underwent a remarkable instant reduce in oral temperatures after the up-step. Marginal significance was observed in male IL-6 before and after the up-step in S15 while female IL-6 prominently increased after the down-step in S15. © 2015 Elsevier Inc. All rights reserved.

1. Introduction Human beings are often likely to expose themselves to sudden temperature changes in daily life. For example, people will encounter temperature steps when entering or exiting air-conditioned buildings and getting on or off planes. Many studies have been carried out to examine human subjective and objective responses to transient thermal environments. However, most of them did not consider gender difference as some studies only enrolled males in their experiments [1–6], and some even did not give subjects' gender information [7,8]. Moreover, although some dynamic thermal environment research did recruit both males and females as subjects [9–12], only Chen discussed gender

⁎ Corresponding author at: Room 405, Mulan Chu Chao Building, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, PR China. E-mail address: [email protected] (Z. Lian).

http://dx.doi.org/10.1016/j.physbeh.2015.07.037 0031-9384/© 2015 Elsevier Inc. All rights reserved.

difference from the aspect of thermal sensation and skin physiology [12], followed by Zhai presenting a simple gender comparison in thermal sensation and dissatisfaction [13]. No circumstantial analysis on gender difference in human response to temperature step changes has been published. As far as thermal perceptions are concerned, Karjalainen conducted a detailed review comprising chamber studies and field surveys, and gave a synopsis of thermal comfort experienced by men and women, which concluded that females were more likely than males to express thermal dissatisfaction; however, no significant gender difference was found in neutral temperatures in most studies; females are more sensitive than males to a deviation from an optimal temperature and express more dissatisfaction, especially in cooler conditions [14]. Nevertheless, as the review shows, most studies focus on human thermal comfort under steady environments rather than transient environments and could not reflect human dynamic response. In addition, objective measures like physiological parameters and biomarkers are applied more

J. Xiong et al. / Physiology & Behavior 151 (2015) 426–440

Nomenclature S5, S11, S15 experiment conditions (S5: 32 °C–37 °C–32 °C, S11: 26 °C–37 °C–26 °C, and S15: 22 °C–37 °C–22 °C) S5-M, S11-M, S15-M male response to S5, S11 and S15, respectively S5-F, S11-F, S15-F female response to S5, S11 and S15, respectively TS thermal sensation TC thermal comfort TA thermal acceptability ANOVA analysis of variance IL-6 Interleukin-6, ng/L HSP70 heat stress protein 70, ng/L mean skin temperature, °C Tskin oral temperature, °C Toral blood oxygen saturation, % SPO2 RR respiratory rate, bpm ECG electrocardiograph HRV heart rate variability HR heart rate, bpm mRR the average of the beat intervals, ms SDRR standard deviation of the beat intervals, ms RMSSD the square root of the mean of the sum of the squares of differences between adjacent beat intervals, ms TP (5 min total power) the variance of NN intervals over temporal segment, ms2 LFnorm LF power (0.04–0.15 Hz) in normalized units; LF / (TP − VLF) ∗ 100, n.u. HFnorm HF power (0.15–0.4 Hz) in normalized units; HF / (TP − VLF) ∗ 100, n.u. LF/HF ratio LF/HF Δ instant change which was defined as the first value after temperature steps minus the last value before temperature steps (e.g. ΔTS, ΔTC, ΔHR, ...)

and more widely in thermal environment studies [15–18], but few gender difference analyses were delivered and even much less studies on the bodily reaction to temperature steps between different gender groups were performed. The purpose of this study was to investigate gender difference in human response to the alteration in thermal environment. We hypothesized that females are different from males in their responses to different temperature steps from the perspectives of psychology, physiology and biomarkers. 2. Material and methods

427

those who might suffer cardiovascular disease, respiratory disease and skin disease which might interfere with the physiological and biochemical measures used in this experiment. Table 1 is the summary of their anthropometric information. Every subject's body mass index rates in the normal range [19]. Skin surface area was calculated based on the Chinese formula [20]. Table 1 also shows profound gender differences in height, weight and skin surface area. Participants were required to wear short-sleeved T-shirts, short trousers and slippers. All of them were asked to avoid caffeine, alcohol, and intense physical activity at least 12 h prior to each experiment. All protocols were approved by the university's ethics committee. Verbal and written informed consents were obtained from each subject prior to participation. 2.2. Measurements 2.2.1. Thermal environment measurement The experiment was conducted in a climate chamber which contained two adjacent rooms (Room A: 3.8 m ∗ 3.6 m ∗ 2.65 m, Room B: 3.8 m ∗ 3.8 m ∗ 2.65 m) connected by an internal door. The step changes in air temperature of heat exposure were supposed to range from 5 °C (small) to 11 °C (medium) to 15 °C (large). Room A was set at 37 °C to represent the outdoor temperature while Room B was set at 22/26/32 °C to represent the typical temperature levels found in air conditioned and naturally ventilated environments in summer. By this means, three temperature step conditions, namely, S5: 32 °C–37 °C– 32 °C, S11: 26 °C–37 °C–26 °C, and S15: 22 °C–37 °C–22 °C, were developed. The measurement site was placed at the center of each room. The air temperature and relative humidity were recorded every 10 s at 0.1 m and 1.1 m height. The air velocity was also monitored. The mean radiant temperature was computed from the globe temperature. All apparatus used in the experiment are listed in Table 3. The measured physical parameters describing the indoor environment are summarized in Table 2. The relative humidity in all rooms was controlled in the range of 30% to 70%. The air speed was kept under 0.1 m/s. The mean radiant temperature was close to air temperature during the experiment. 2.2.2. Subjective measurement In this study, psychological measurements consist of health selfreported symptoms, fatigue, thermal perceptions and endurance. Subjects were asked to answer whether or not they were suffering from health symptoms like perspiration, dizziness, accelerated respiration, eyestrain and accelerated heart rate at present time. Fatigue was assessed using the Japanese subjective fatigue symptoms (2002 version) [21]. The fatigue check-list contains 25 items and is divided into 5 subtypes. The participants answered each item using a five-point discrete scale from + 1 (none) to +5 (extremely severe). In addition, subjects gave their subjective thermal feelings on ASHRAE continuous voting scales and endurance status on a four-point scale (Fig. 1).

2.1. Subjects Twenty four healthy undergraduate students (half males and half females) with an average age of 22 ± 1 years were recruited in the study. Preliminary evaluation of applicants was conducted to exclude Table 1 Anthropometric information of subjects participating in the study. Gender

Number

Age (years)

Height (cm)

Weight (kg)

BMI (kg/m2)

As (m2)

Male Female P

12 12 –

22 ± 1 22 ± 1 0.122

176.8 ± 4.9 164.1 ± 5.7 b0.001⁎⁎⁎

66.5 ± 6.3 55.1 ± 4.0 b0.001⁎⁎⁎

21.3 ± 2.3 20.5 ± 1.3 0.270

1.90 ± 0.08 1.69 ± 0.08 b0.001⁎⁎⁎

Note: BMI, body mass index and As, body surface area calculated by formula for Chinese adults [20]. ⁎⁎⁎ P b 0.001.

2.2.3. Biochemical and physiological measurements When encountering a temperature step, many biochemical and physiological variations would occur for humans to adapt themselves to the stress. In this study, parameters that can reflect temperature step effects on the human immune system (IL-6 and HSP70), heat metabolism system (Tskin and Toral), respiratory system (SPO2 and RR) and cardiovascular system (ECG) were recorded. Devices used for measurements are summarized in Table 3. Venous blood was sampled four times during the experiment for the test of IL-6 and HSP70. Each time, each subject provided about 2 ml of blood. The samples were then centrifuged for 15 min at the speed of 2000 RPM. After separation from blood cells, serum was stored in a freezer at − 20 °C before being sent for analysis. The analysis was performed by an external specialized laboratory using ELISA (enzymelinked immunosorbent assay) kits.

428

J. Xiong et al. / Physiology & Behavior 151 (2015) 426–440

Table 2 Experimental conditions. Conditions (nominal temperature)

S5 S11 S15

32 °C–37 °C–32 °C 26 °C–37 °C–26 °C 22 °C–37 °C–22 °C

Phase 1 (Room A)

Phase 2 (Room B)

Phase 3 (Room A)

T

RH

T

RH

T

RH

32.2 ± 0.5 26.0 ± 0.3 22.0 ± 0.4

65.0 ± 8.8 59.2 ± 7.4 46.6 ± 8.1

37.2 ± 0.3 37.2 ± 0.4 37.4 ± 0.4

44.8 ± 4.5 40.0 ± 2.5 31.0 ± 4.3

32.1 ± 0.6 26.5 ± 0.2 22.2 ± 0.7

62.2 ± 6.0 65.0 ± 2.0 50.0 ± 5.4

Note: T and RH represent air temperature (°C) and relative humidity (%).

Skin temperature was measured at an interval of 10 s and on seven body parts. The mean skin temperature was calculated according to Eq. (1) [22]. Tskin ¼ 0:07Tforehead þ 0:35Tchest þ 0:14Tlower arm þ 0:05Thand back þ 0:19Tthigh þ 0:13Tlower leg þ 0:07Tinstep

ð1Þ

Electrocardiograph (ECG) is a non-invasive technique to evaluate the balance and capacity of the autonomic nervous system [23]. For short-term recordings, mRR, SDRR and RMSSD are recommended indexes for time domain measures while LFnrom, HFnorm and LF/HF are the main frequency domain parameters [24]. In this study, HR and HRV were computed on the basis of ECG. 2.3. Experiment procedure The experiment was conducted in the summer of 2014 and lasted for 18 days. Considering the rhythm of biomarker and physiological parameters, the experiment was performed in the afternoon of each day. After arrival, subjects first stayed in Room A to wear the portable physiological recorder and to have the Pyrobuttons attached to the skin in 15 min. Then the experiment started. Each test lasted for 135 min (Fig. 2). First, subjects stayed in Room A for 30 min (phase A).Then, they moved to Room B for 60 min (phase B). Finally, subjects returned to Room A remaining for 45 min (phase 3). The questionnaires of health symptoms and thermal perceptions were filled out at the 15, 31, 33, 36, 53, 63, 85, 91, 93, 96, 110, 126, and 135 min while subjects' fatigue status was asked at the middle of each phase. Tskin, RR, SPO2 and ECG were continuously monitored while Toral was measured 15 min before and after temperature steps. In addition, blood was sampled 10 min before and after temperature steps. Subjects were allowed to have a read during the experiment.

2.4. Statistical analysis ECG was divided into 5 min segments according to the task force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology [24]. For consistency, SPO2 and RR were also computed at an interval of 5 min. In order to reveal the influence of temperature step magnitude, we also calculated the instant change (Δ) under a temperature step which was defined as the first vote after the step minus the last vote before the step. The distributions of all data were tested for normality using the Shapiro–Wilk test. Within-subject ANOVA was performed using time as a within-subject factor and gender as a between-subject factor. The interaction of time and gender was also included in the model. A chisquare test, the Mann–Whitney U test and Student's t-test were applied to test gender difference in binary data, non-normal distribution data and normal distribution data, separately. The statistical analysis was performed by the software SPSS 19.0, and the significance level was set to be 0.05 (P b 0.05). 3. Results 3.1. Subjective perceptions 3.1.1. Health symptoms The change of self-reported symptoms over time in response to temperature steps for males and females is displayed in Fig. 3. In S5, when returning to 32 °C from 37 °C, fewer males reported the perspiration symptom and the percentage of males ended at about 10% while that of females was 50%. The chi-square test demonstrates significant gender difference at the end of phase 3 (P b 0.05). In general, the percentage of self-reported eyestrain for females is higher than that for males. Particularly, when subjects moved into a cool environment from 37 °C, their

Fig. 1. Voting scales.

J. Xiong et al. / Physiology & Behavior 151 (2015) 426–440

429

Table 3 Measurement devices of this experiment. Name

Type

Specification

Measurement

Physical measurements Thermometer

TR-72, Japan

Air temperature

Psychrometer

TR-72, Japan

Anemometer

TESTO 425, German

Black globe thermometer

Shanghai Huo er Co. China

Accuracy: ±0.3 °C Range: 0 °C–50 °C Accuracy: ±5% Range:10%–95% Accuracy: ±(0.03 m/s + 5% measured value) Range: 0 m/s–20 m/s Diameter: 150 mm Accuracy: ±0.1 °C Range: −2 °C–50 °C

Biochemical and physiological measurements Thermometer

Omron MC-246

Oral temperature

Thermometer

Pyrobutton-L,USA

Accuracy: ±0.1 °C Range: 32 °C–42 °C Accuracy: ±0.2 °C Range: −40 °C–85 °C

Portable multi-channel physiological recorder ELISA kits

Somté, Australia –

The lower limits of detection: 0.05 ng/L for IL-6 and 1.0 ng/L for HSP70

eyestrain symptoms would be eased; however, significantly more females still perceived eyestrain than males in S15 (P b 0.05) and S11 (P b 0.1) right after entering phase 3. 3.1.2. Thermal perceptions and endurance Fig. 4 shows the variation of subjects' thermal perceptions and endurance over time. It can be seen that males and females have almost the same thermal sensation under 26 °C. In contrast, female thermal sensation is lower in the cool environment (22 °C) and higher in the warm environment (32 °C and 37 °C) than that of males. Comparison between thermal perceptions just before and after temperature steps was conducted as partitioned by gender and temperature step conditions, and significances were observed for male and female thermal sensations under all up and down temperature steps. As for thermal comfort, when subjects encountered sudden heating, their thermal comfort worsened and reached steady state at the end of phase 2. In contrast, when meeting with temperature down-steps, subjects regained thermal comfort. It is noticeable that when moving to 37 °C from 22 °C, the thermal comfort of women, other than decreased, instead slightly increased at first. Significances are obtained between thermal comfort votes before and after up-steps in S5 and S11, and downsteps in S5, S11 and S15 for both males and females. Thermal acceptability shares a similar changing pattern with thermal comfort. Subjects' endurance statuses were also investigated. Endurance statuses worsened

Relative humidity Air velocity Global temperature

Skin temperature SPO2/RR/ECG IL-6, HSP70

significantly after subjects entered 37 °C in S5 and S11, and were alleviated when people returned to the cool room with significance detected only in the male group. Gender differences in thermal perceptions over an adaptive process after temperature up-steps and down-steps were displayed in Table 4. During the up-step adaptive process (phase 2), female thermal sensation is statistically larger than that of males in S5, and variation patterns of thermal sensation in S11 are significantly different for gender groups (gender × time: P b 0.05). Contrarily, during the down-step adaptive process (phase 3), the variation trends of thermal comfort in S11 and thermal acceptability in S15 were significantly different between genders (gender × time: P b 0.05); besides, female thermal acceptability is statistically higher than that of males in S11. Fig. 5 shows instant changes in thermal sensation, comfort, acceptability and endurance immediately after temperature steps for males and females. The changes of female thermal perceptions are similar to but more intensive than those of males, and there are more significances observed for females. Specifically, the instant change of thermal comfort after the up-step in S15 for women significantly differs from that for men. Fig. 6 illustrates the correlation of thermal comfort and thermal sensation for males and females. It can be seen that females are more prone to show thermal dissatisfaction to cool environments while males are more likely to feel thermal discomfort in warm environments.

Fig. 2. Experimental procedure.

430

J. Xiong et al. / Physiology & Behavior 151 (2015) 426–440

Fig. 3. Changes of self-reported symptoms over time in response to temperature steps.

Fig. 4. Changes of subjects' thermal perceptions and endurance over time in response to temperature steps.

J. Xiong et al. / Physiology & Behavior 151 (2015) 426–440

431

Table 4 Gender difference in multiple subjective perceptions in different conditions. Parameters

Thermal sensation

Thermal comfort

Thermal acceptability

Endurance

Phase 2

S5 S11 S15 S5 S11 S15 S5 S11 S15 S5 S11 S15

Phase 3

Time

Gender

Time ∗ gender

Time

Gender

Time ∗ gender

0.439 0.009⁎⁎ b0.001⁎⁎⁎ 0.168 b0.001⁎⁎⁎ b0.001⁎⁎⁎

0.019⁎ 0.557 0.257 0.865 0.939 0.142 0.872 0.821 0.248 0.552 0.626 0.093

0.081 0.024⁎ 0.701 0.109 0.346 0.081 0.740 0.052 0.665 0.680 1.000 0.258

0.543 0.295 0.013⁎ 0.044⁎ b0.001⁎⁎⁎ b0.001⁎⁎⁎ 0.027⁎ 0.004⁎⁎ 0.012⁎

0.247 0.899 0.095 0.342 0.281 0.145 0.328 0.049⁎ 0.402 0.187 0.855 0.154

0.960 0.872 0.220 0.589 0.009⁎⁎ 0.093 0.648 0.223 0.020⁎ 0.251 0.330 0.357

0.069 b0.001⁎⁎⁎ b0.001⁎⁎⁎ 0.045⁎ 0.017⁎ b0.001⁎⁎⁎

0.054 0.330 0.040⁎

⁎ P b 0.05. ⁎⁎ P b 0.01. ⁎⁎⁎ P b 0.001.

3.1.3. Fatigue ratings As Fig. 7 shows, fatigue ratings are categorized into five subtypes. Repeated ANOVA demonstrates that the change patterns of fatigue subtypes including tiredness and insecurity significantly differ between females and males during the experiment period (time × gender: P b 0.05). Fatigue status aggravated after temperature upsteps and eased after down-steps for both men and women. Females' tiredness fatigue scores tend to be higher than those of males in

phase 2 of all three conditions though no statistically significant gender difference was obtained. 3.2. Biochemical responses Marginal significance existed in the male serum level of IL-6 after sudden heating while female IL-6 significantly increased after sudden cooling in S15 (Fig. 8). At the same time, the instant changes of female

Fig. 5. Instant changes of thermal perceptions and endurance after temperature steps.

432

J. Xiong et al. / Physiology & Behavior 151 (2015) 426–440

Fig. 6. Thermal sensation and corresponding thermal comfort.

IL-6 after down-steps in S11 and S15 are significantly more intensive than that in S5 (Fig. 9). When entering into 37 °C from 22 °C, the female IL-6 level is markedly lower than that of males. No significance was observed for the serum level of HSP70 under any conditions. 3.3. Physiological responses Female oral temperatures are generally higher than that of males. In detail, noticeable gender differences exist at the 15th minute of S15, and the 75th minute and 105th minute of S5 (Fig. 10). After sudden heating in S15, oral temperatures of both males and females significantly

reduced, indicating that subjects lowered core temperatures to protect core parts of their bodies. Fig. 11 and Table 5 show the temporal change of mean skin temperatures. In the first 30 min (phase 1), male mean skin temperatures are higher than female skin temperatures at 22 °C. After up-steps, the difference between males and females tends to narrow down, and the changing patterns of males and females are significantly different (time × gender: P b 0.05) in S11. Next, subjects returned to the cool room. During the down-step adaptive process, there is a more rapid decrease of female skin temperatures in S15 than that of males. In contrast, in S5 which is made up of relatively high temperatures of 32 °C and 37 °C, male mean skin temperatures are constantly lower than female mean skin temperatures during the whole experiment period. By comparing the 5 min mean skin temperature just before and after temperature steps, it is clear that skin temperatures are significantly increased and decreased after sudden heating and cooling accordingly for both men and women. In order to compare the time for skin temperature to reach steady state for different gender groups, skin temperatures were managed as follows: mean skin temperatures were considered to reach stability in this study if the variation range of them after a certain time point was constantly within the range of ±0.1 °C and then nonlinear fitting was performed with Eq. (2) to get the stable time. tmsk ¼ k1 exp

t


τ

þ k2

ð2Þ

where tmsk (°C) is the predicted mean skin temperature and t (min) is the exposure time after the step change. The coefficients k1 (°C) and k2 (°C) represent the amplitude of the change in the mean skin temperature and the steady value. τ (min) is a time coefficient. All these unknowns were computed by means of the Levenberg–Marquardt algorithm. As ± 0.1 °C was considered as the standard of stability, stable

Fig. 7. Changes of fatigue scores over time in response to temperature steps.

J. Xiong et al. / Physiology & Behavior 151 (2015) 426–440

433

Fig. 8. Changes of IL-6 and HSP70 over time in response to temperature steps.

time was computed using Eq. (2). For example, it is obvious in Fig. 11 that male skin temperature reached stability after sudden heating in S5 and the nonlinear regression shows that k1 and τ are equal to −0.8 °C and 5.3 min. When t equals 2.1τ, namely 11.0 min, the absolute value of k1 expðt =τ Þ is 0.1 °C and it will continue to decrease with the time. Hence, the time for male skin temperatures to reach steady state after the up-step in S5 is 11.0 min. It should be noted that subjects'

skin temperatures failed to reach stability after the down-step in S15 in this experiment; their stable time is undoubtedly longer than 45 min. Table 6 displays the results of regression. For both males and females, more time is needed for skin temperatures to get stability after more intensive temperature steps. Moreover, it takes less time for both men and women reach stable status after down-steps than after up-steps. Gender difference is also presented in Table 6 which

Fig. 9. Instant changes of IL-6 and HSP70 in response to temperature steps.

434

J. Xiong et al. / Physiology & Behavior 151 (2015) 426–440

Fig. 10. Changes of oral temperature over time in response to temperature steps.

shows that the steady time of skin temperature for males is shorter than that for females; meanwhile, change amplitudes in the mean skin temperatures (k1 in Eq. (2)) for males are smaller than the female counterparts for both up-steps and down-steps. Fig. 12 shows the change of SPO2 and RR in response to temperature steps. Female blood oxygen saturation is clearly larger than that of males as a whole. Specific analysis observed significance for females between the values before and after the down-step in S15. Besides, a profound gender difference exists during phase 2 in S11 and S15 while during the adaptive process of the down-step (phase 3) females still own a significantly higher level of SPO2 than males in S5 with noticeable interaction between gender and time in S11 (Table 5). As for respiration rate, the general trends for males and females bear some similarities. Analysis shows that male respiration rates increase significantly after both the up-step and down-step in S11; meanwhile, female respiration rates change vigorously after the up-step in S15 and after down-steps in S5 and S11. Fig. 13 and Fig. 14 illustrate the change of cardiovascular indexes over time. As a whole, female HR is faster than male HR while their mRR, SDRR and RMSSD are lower than that of males. Based on analysis

performed for different gender groups, it can be also seen in Fig. 13 and Fig. 14 that males significantly quicken their heart rates after the upstep in S11 while males and females remarkably slow down their heart rates after down-steps in S11 and S15 separately; female mRR shows significant changes after up-steps in all three conditions and after down-steps in S11 and S15, while there is significance between male mRR values before and after up-steps in S11 and S15; SDRR for both males and females profoundly increases after step changes in S11 and S15 except for males after the up-step in S15; for RMSSD, after

Table 5 Gender difference in multiple physiological parameters in different conditions. Parameters

Tskin

SPO2

RR

HR

mRR

SDRR

RMSSD

LFnorm

HFnorm

LF/HF

Fig. 11. Changes of mean skin temperature over time in response to temperature steps.

S5 S11 S15 S5 S11 S15 S5 S11 S15 S5 S11 S15 S5 S11 S15 S5 S11 S15 S5 S11 S15 S5 S11 S15 S5 S11 S15 S5 S11 S15

⁎ P b 0.05. ⁎⁎ P b 0.01. ⁎⁎⁎ P b 0.001.

Phase 2

Phase 3

Time

Gender

b0.001⁎⁎⁎ b0.001⁎⁎⁎ b0.001⁎⁎⁎

0.049⁎

0.552 0.224 0.005⁎⁎ 0.088 0.197 b0.001⁎⁎⁎ 0.070 b0.001⁎⁎⁎ b0.001⁎⁎⁎ 0.156 b0.001⁎⁎⁎ b0.001⁎⁎⁎ 0.010⁎ b0.001⁎⁎ b0.001⁎⁎⁎ 0.015⁎ 0.088 b0.001⁎⁎⁎ 0.014⁎ 0.012⁎ 0.002⁎⁎ 0.006⁎⁎ 0.012⁎ 0.002⁎⁎ 0.005⁎⁎ 0.068 0.005⁎⁎

0.798 0.703 0.062 0.032⁎ 0.003⁎⁎ 0.819 0.423 0.709 0.016⁎ 0.073 0.421 0.019⁎ 0.011⁎ 0.453 0.148 0.043⁎ 0.852 0.092 0.035⁎ 0.680 0.790 0.725 0.343 0.766 0.324 0.371 0.641 0.424 0.381

T×G

Time

Gender

T×G

0.077 0.018⁎ 0.168 0.408 0.293 0.092 0.582 0.909 0.468 0.325 0.398 0.963 0.994 0.788 0.592 0.536 0.164 0.882 0.781 0.303 0.816 0.517 0.706 0.188 0.464 0.946 0.189 0.426 0.262 0.135

b0.001⁎⁎⁎ b0.001⁎⁎⁎ b0.001⁎⁎⁎

0.035⁎

0.528 b0.001⁎⁎⁎ 0.156 0.553 0.034⁎

0.677 0.058 0.067 b0.001⁎⁎⁎ b0.001⁎⁎⁎ b0.001⁎⁎⁎ 0.045⁎ 0.007⁎⁎ 0.001⁎⁎ 0.038⁎ b0.001⁎⁎⁎ b0.001⁎⁎⁎ 0.156 b0.001⁎⁎⁎ b0.001⁎⁎⁎ 0.598 0.368 0.027⁎ 0.567 0.065 0.040⁎ 0.500 0.128 0.061 0.196 0.561 0.007⁎⁎

0.665 0.049⁎ 0.050⁎ 0.381 0.593 0.617 0.735 0.934 0.009⁎⁎ 0.102 0.669 0.009⁎⁎ 0.009⁎⁎ 0.257 0.175 0.008⁎⁎ 0.946 0.069 0.014⁎ 0.307 0.414 0.979 0.544 0.386 0.466 0.555 0.191 0.682 0.469

0.371 0.891 0.608 0.125 0.359 0.705 0.152 0.657 0.260 0.249 0.642 0.596 0.605 0.212 0.336 0.740 0.099 0.882 0.850 0.148 0.614 0.830 0.447 0.733 0.444

J. Xiong et al. / Physiology & Behavior 151 (2015) 426–440

435

Table 6 The results of non-linear regression model applied to the mean skin temperature. Condition Up-steps

S5: 32–37 °C S11: 26–37 °C S15: 22–37 °C Down-steps S5: 37–32 °C S11: 37–26 °C S15: 37–22 °C

Gender k1 (°C) Male Female Male Female Male Female Male Female Male Female Male Female

−0.8 −1.0 −1.9 −2.4 −3.3 −4.0 0.6 0.8 1.6 2.0

τ Tstable (min) (min) 5.3 5.8 4.7 6.4 9.0 8.7 4.1 4.3 3.6 6.6

2.1τ 2.3τ 2.9τ 3.2τ 3.5τ 3.7τ 1.8τ 2.1τ 2.8τ 3.0τ

11.0 13.4 13.8 20.3 31.5 32.1 7.3 8.9 10.0 19.8

k2 (°C)

R2

Se

35.3 35.9 35.7 35.9 35.6 35.6 34.4 35.0 33.6 33.5

0.910 0.970 0.947 0.990 0.986 0.987 0.787 0.979 0.982 0.968

0.002 0.001 0.006 0.002 0.007 0.010 0.003 0.000 0.001 0.006

Note: R2, coefficient of determination and Se, residual sum of squares.

the up-step, significance appears only for men in S15 while after the down-step, there is significance for males in all three intensities and for females in S11 and S15; the last value of phase 1 and the first value of phase 2 differ significantly for males in S11 in LFnorm and HFnorm and for females in S15 just in LFnorm; for LF/HF, significance is available only between the last value of phase 2 and the first value of phase 3 for females in S11 and for males in S5 and S15. Table 5 contains the gender difference in multiple cardiovascular parameters. During both the up-step and down-step adaptive processes, female hearts beat significantly faster than male hearts in S5 while mRR in S5 and S11 and SDRR and RMSSD in S11 of females are significantly smaller than their male counterparts. Gender differences in multiple physiological parameters at typical time, that is the end of phase 1 (27.5 min), the start of phase 2 (32.5 min), the end of phase 2 (87.5 min), the start of phase 3 (32.5 min) and the end of phase 2 (87.5 min), were discussed using GLM (general linear model) with condition as a within-subject factor and gender as a between-subject factor (Table 7). At the end of phase 1, there is no difference among either different conditions or genders for respiratory indexes like SPO2 and RR. In contrast, condition

Fig. 13. Change of HR over time in response to temperature steps.

difference is significant for cardiovascular indexes (HR and HRV) with noticeable gender differences in HR and mRR, which is reflected in female hearts beating significantly faster than males in S11 (26 °C) and their significantly shorter mRR in S11 (26 °C) and S5 (32 °C). For mean skin temperatures, the interaction between condition and gender is notable and Student's t test confirms the significant gender difference in S15 (22 °C). At the start of phase 2, male skin temperatures remain significantly lower than that of females in S15 and gender difference is observed for mRR in S5 and S11. At the end of phase 2, after nearly an hour of adaption to the hot environment of 37 °C, there is no significant condition difference in any of these physiological parameters, indicating the erasure of temperature step change effects. Statistical gender differences are found in SPO2, HR and three time domain HRV parameters. At

Fig. 12. Changes of SPO2 and RR over time in response to temperature steps.

436

J. Xiong et al. / Physiology & Behavior 151 (2015) 426–440

Fig. 14. Changes of HRV indexes over time in response to temperature steps.

the start of phase 3, when returning to the relative cool room, significant interactions between gender and condition are there for skin temperature and heart rate. Specifically, there are remarkable gender differences for skin temperature in S11 and heart rate in S5. Besides, male mRR values are significantly higher than females in S5 and S11. After the down-step adaptive process of about 45 min, gender differences were observed for skin temperature in S15, for HR in S5 and for mRR, SDRR and RMSSD in S5 and S11. Effects of temperature step amplitude on various physiological parameters are also examined in this research by the paired t test performed between instant changes after the up-step or down-step for males and females, separately (Fig. 15 and Fig. 16). Male instant change of oral temperature after the up-step in S11 is significantly different

from that in S15 while ΔToral of females in S15 is more intensive than that in S5 and S11. There are statistical significances between any two of the three conditions for female ΔTskin except for up-steps between S11 and S15, and it is the same case with male skin temperatures apart from the fact that no noticeable discrepancy is detected after down-steps between S11 and S15. As for ΔmRR, there are significances between S5 and S11, and between S5 and S15 for males after up-steps, and between S5 and S15 for females after down-steps. As for ΔSDRR, male values in S11 are remarkably higher than that in S5 and S15 after up-steps while male values in S5 are lower than that in S11 and S15 after down-steps; female values after both the up-step and down-step in S5 are significantly smaller than that in S11 and S15. As for ΔRMSSD, there appear significances for pairwise comparison

J. Xiong et al. / Physiology & Behavior 151 (2015) 426–440

437

Table 7 Gender difference in multiple physiological parameters at different time. Time (min)

Condition

Tskin

SPO2

RR

HR

mRR

SDRR

RMSSD

LF

HF

LF/HF

27.5

ANOVA S5 S11 S15 ANOVA S5 S11 S15 ANOVA S5 S11 S15 ANOVA S5 S11 S15 ANOVA S5 S11 S15

▼ 0.159 0.056 0.007⁎⁎ ▼ 0.317 0.098 0.048⁎

n.s. 0.303 0.245 0.301 ▼ 0.243 0.832 0.523

n.s. 0.989 0.394 0.715 n.s. 0.904 0.696 0.784 n.s. 0.665 0.147 0.891 n.s. 0.354 0.283 0.795 n.s. 0.868 0.632 0.794

▼ 0.077 0.027⁎ 0.676 ▼ 0.216 0.117 0.528

▼ 0.030⁎ 0.007⁎⁎ 0.479 ▼ 0.026⁎ 0.010⁎

▼ 0.211 0.169 0.282 ▼ 0.260 0.161 0.720

▼ 0.437 0.066 0.215 ▼ 0.178 0.123 0.911

0.003⁎⁎ 0.036⁎ 0.369 ▼ 0.005⁎⁎

0.018⁎ 0.008⁎⁎ 0.275

0.173 0.013⁎ 0.976 ▼ 0.443 0.067 0.851 ▼ 0.010⁎ 0.002⁎⁎ 0.673

0.053 0.022⁎ 0.444 ▼ 0.138 0.081 0.387 ▼ 0.030⁎ 0.006⁎⁎ 0.213

▼ 0.519 0.462 0.210 n.s. 0.700 0.663 0.082 n.s. 0.521 0.909 0.771 n.s. 0.135 0.594 0.793 ▼ 0.111 0.618 0.838

▼ 0.678 0.219 0.163 n.s. 0.651 0.406 0.139 n.s. 0.559 0.406 0.798 n.s. 0.165 0.753 0.945 ▼ 0.097 0.723 0.733

▼ 0.374 0.454 0.194 n.s. 0.408 0.594 0.132 n.s. 0.438 0.487 0.719 n.s. 0.363 0.664 0.259 ▼ 0.232 0.627 0.779

32.5

87.5

92.5

132.5

n.s. 0.134 0.294 0.831 ▼ 0.088 0.012⁎ 0.459 ▼ 0.136 0.521 0.008⁎⁎

0.807 0.052 0.028⁎ n.s. 0.069 0.297 0.703 n.s. 0.643 0.209 0.768

0.327 0.609 ▼ 0.010⁎⁎ 0.226 0.092

Note: ▼ denotes significant condition difference; denotes significant gender difference; that no significance was detected among condition, gender and condition × gender. ⁎ P b 0.05. ⁎⁎ P b 0.01.

among the three conditions after up-steps for males while instant change of males or females in S15 is profoundly different from S5 and S11 after down-steps. Besides, significant gender differences exist in ΔToral after the up-step in S15, in ΔTskin after the up-step in S5 and the down-step in S11 and in ΔRMSSD after the up-step in S15 (Fig. 15 and Fig. 16).

0.644

0.014⁎ 0.019⁎ 0.511 ▼ 0.016⁎ 0.006⁎⁎ 0.127

denotes significant condition and gender difference (condition × gender); and n.s. denotes

4. Discussion 4.1. Subjective perceptions Our results show that significantly more females reported perspiration at the end of the down-step adaptive process of S5. One possible

Fig. 15. Instant changes of oral temperature, skin temperature, SPO2 and RR in response to temperature steps.

438

J. Xiong et al. / Physiology & Behavior 151 (2015) 426–440

Fig. 16. Instant changes of HRV indexes in response to temperature steps.

reason is that heat convection between males and the environment is more intensive than that of females because male skin temperature is lower than its female counterpart in S5. So, males may gain thermal balance just through heat convection while females may have to exert heat dissipation through convection and sweat evaporation. Women being more sensitive to the thermal environment may account for the relatively higher percentage of self-reported eyestrain. Gender differences in thermal perceptions are illustrated in Fig. 4. Males and females have similar thermal sensations under 26 °C while males feel less thermal comfort than females. Some studies concluded that there was no significant gender difference on neutral temperature [13,14,25] but females rated 26 °C more pleasant than males [26]. Lan [25] also found that female comfortable temperature (26.2 °C) was higher than male comfortable temperature (25.3 °C), which can explain the result of male thermal satisfaction being lower than that of females in 26 °C. It is clear in this study that females feel lower temperature (22 °C) cooler and higher temperature (32 °C and 37 °C) warmer than males. This gender difference could be demonstrated from the perspective of skin temperature. Fig. 11 shows that female skin temperatures are lower in cool environments and higher in warm environments than corresponding male values. As a result, the higher sensitivity of skin temperature to the environmental temperature causes the relatively more intensive change for females in thermal sensation. The correlation of thermal comfort and thermal sensation (Fig. 6) proves that females are prone to be more dissatisfied in cool temperatures whereas males are prone to be more dissatisfied in warm temperatures. The larger deviation from the comfort temperature results in the severer thermal discomfort. Since male comfort temperature is lower than that of females [25,27], male thermal comfort is worse than female thermal comfort in warm environments and better than that of females in cool environments. It is worth mentioning that gender difference would decrease in extremely high temperature, for example, both males and females reported the similar level of thermal discomfort when voting “hot” (+3) on the ASHRAE thermal sensation scale (Fig. 6). One plausible explanation is that the deviation from optimal

temperature is so large in an uncompensable hot environment like 37 °C for both males and females that all of them show dissatisfaction to nearly the same extent. This is the same case for thermal acceptability and endurance. It is obvious that after the up-step of S15, ΔTC, ΔTA and ΔEndurance of women are smaller than that of men in spite of the larger ΔTS. This can be also explained by the reason that females prefer a slightly warm environment while males prefer a slightly cool environment. 4.2. Biomarkers Two kinds of biomarkers, namely IL-6 and HSP70 were measured in this study. IL-6 is a kind of cytokines, which is reported to be related to heat stress and can be treated as an indicator to reflect immunityspecific homeostasis. As for HSP70, nearly all cells respond to intensive heat stress by producing heat-shock proteins or stress proteins which contain HSP70. Marginal significance was observed in male serum levels of IL-6 before and after the up-step in S15 whereas female IL-6 increased prominently after the down-step in S15 (Fig. 8). This discrepancy can be explained by the stress theory. Because males are more prone to show discomfort in warm environments, it is reasonable to assume that males regard 37 °C as a more unfavorable environment than females. Hence, when encountering sudden heating, males think of it as a worse and unfavorable condition, and their stress reactions become more intensive than females, which resulted in a high level of IL-6. Likewise, because females are more prone to show dissatisfaction in cool environments, when coming up with sudden cooling in S15, females take it as a more serious threat and their stress reactions would be activated and cause a significant increase in IL-6. No obvious difference was shown in the serum level of HSP70 which could account for by the blood sampling time and temperature conditions [6,28]. 4.3. Physiological parameters Gender difference in skin temperature is illustrated in Fig. 11. Female skin temperature is lower in a cool environment and higher in a warm

J. Xiong et al. / Physiology & Behavior 151 (2015) 426–440

environment than that of males. Under heat stress, women generally present lower sweating response [29–31] and greater body heat storage [32], and these would lead to a higher skin temperature in females. In the cold, women generally have less capability for maximum heat production by either exercising or shivering and greater surface heat losses, which would result in a lower skin temperature in females [32]. In summary, it is rational to suppose that men have a stronger thermoregulation ability than women. The shorter stable time and smaller change amplitude of mean skin temperature shown in Table 6 also prove the more robust and swift thermoregulation ability for men. Interestingly, we found that after the up-step adaptive process of about 60 min, male skin temperatures in S5 tend to be lower than other conditions although no significance is observed, which implies that previous temperature has an impact on the present status. One possible explanation is that male sweat glands are more intensely activated in a warm environment like 32 °C, and when moving to a hotter environment (37 °C), the sweat rate augments and more latent heat is dissipated by means of evaporation, leading to a lower skin temperature. More detailed skin parameters that can reflect sweating like skin moisture and sweat loss are suggested to be applied in future work to explore the cause of such discrepancy. Additionally, Lan [25] exposed subjects to four temperature levels (21 °C, 24 °C, 26 °C, 29 °C) and found that female skin temperature was constantly lower than that of males. We can logically determine from the results of this study and Lan's research that the turning point for skin temperature below which male skin temperature is higher and above which male skin temperature is lower than its female counterpart lies in the range between 29 °C and 32 °C. When passing to 22 °C from 37 °C, oral temperatures of males and females are significantly lowered to protect the core part of the body. Meanwhile, female instant change of oral temperature after the upstep in S15 is profoundly intensive than that of male, inferring that female internal environments are more susceptible to external influence and males can be more successfully heat acclimated. In the same way, when encountering sudden cooling in S15, change of male oral temperature is prone to be smaller than that of females, though no significance is observed. Gender difference in oral temperature in S15 also supports the speculation that men have a stronger thermoregulation ability than women. Female SPO2 values are lower versus males. The difference could be related to women on average having a lower diffusing capacity smaller airway diameter and lung volumes than men [33]. Another possibility is progesterone-related ventilatory stimulation in women [34]. Several cardiovascular indexes were analyzed in this study. Conforming to some studies [35], female heart rate is significantly higher than males, and the female mean heart beat interval is shorter than the latter accordingly. RMSSD is a parasympathetic index. After the up-step in S15, RMSSD decreases for men and women, and in particular, male ΔRMSSD is significantly more intensive than female ΔRMSSD. When meeting with intensive up-steps, males could exert more vigorous inhibition on the vagus nerve to adapt themselves more quickly to the cool-to-hot temperature shock than females. Besides, in the present study, SDRR that estimates the cyclic component responsible for variability is generally higher in males, indicating the stronger regulative capability of the autonomic nerve system on the heart. No gender difference was observed in HRV frequency domain measures (LFnorm, HFnorm and LF/HF) though RMSSD significantly correlated with the other parasympathetic index HFnorm. It is interesting that the variation patterns of instant change in various parameters under down-steps are very different, especially for cardiovascular indexes, between men and women. The reason why such discrepancy exists needs further investigation. There are some limitations in the paper. For example, only young students were included in this study. Further research work is still needed to validate whether these results are also suitable for other age groups. And temperature levels investigated in this study mainly simulated the summer condition; as to the condition in winter, more studies are required. As far as practical applications are concerned, the gender

439

differences in human response to temperature steps are found, and these differences can be considered in environmental designs for practical buildings and other similar environments. 5. Conclusions Gender difference in human response to the alteration in thermal environment was investigated from the aspect of psychology, physiology and biomarkers. It can be concluded that: 1. When encountering the up-step in S15, both males and females significantly lowered their oral temperatures to protect the core part of the body. However, female instant change of oral temperature after the up-step in S15 is profoundly more intensive than that of males. 2. It is logical that males have a stronger thermoregulation ability than females. After both up-steps and down-steps, male skin temperature change amplitude is smaller while the stable time for skin temperature is shorter than that of females. Additionally, the turning point for difference in skin temperature between males and females is supposed to lie between 29 °C and 32 °C. 3. Males showed a more intensive decrease in RMSSD after the up-step in S15 than females. No significant gender difference is detected in frequency domain indexes of heart rate variability. 4. Marginal significance was observed in male IL-6 values before and after the up-step in S15, while female IL-6 prominently increased after the down-step in S15. This discrepancy can be explained by the diversity of judgment of thermal stress property between males and females. 5. Females feel lower temperature cooler and higher temperature warmer than males. Furthermore, females are more prone to show thermal dissatisfaction to cool environments while males are more likely to feel thermal discomfort in warm environments. 6. Significantly more females reported perspiration at the end of the down-step adaptive process of S5. The percentage of eyestrain symptom of females is generally higher than that of males during the up-step or down-step adaptive process. Acknowledgments This work is financially supported by Key Program of the National Natural Science Foundation of China (51238005). References [1] Gagge AP, Stolwijk J, Hardy J. Comfort and thermal sensations and associated physiological responses at various ambient temperatures. Environ. Res. 1967;1:1–20. [2] R. Dear, J. Ring, P. Fanger, Thermal sensations resulting from sudden ambient temperature changes, Indoor Air 3 (1993) 181–192. [3] K. Nagano, A. Takaki, M. Hirakawa, Y. Tochihara, Effects of ambient temperature steps on thermal comfort requirements, Int. J. Biometeorol. 50 (2005) 33–39. [4] X. Du, B. Li, H. Liu, D. Yang, W. Yu, J. Liao, et al., The response of human thermal sensation and its prediction to temperature step-change (cool–neutral–cool), PLoS ONE 9 (2014) e104320. [5] H. Liu, J. Liao, D. Yang, X. Du, P. Hu, Y. Yang, et al., The response of human thermal perception and skin temperature to step-change transient thermal environments, Build. Environ. 73 (2014) 232–238. [6] J. Y., Q. O., Y. Z., H. S., G. C., W. Cu., A comparison of the thermal adaptability of people accustomed to air-conditioned environments and naturally ventilated environments, Indoor Air 22 (2011) 110–118. [7] C. H., H. Z., E. A., D. W., Skin and core temperature response to partial- and wholebody heating and cooling, J. Therm. Biol. 29 (2004) 549–558. [8] C. Chun, A. Tamura, Thermal comfort in urban transitional spaces, Build. Environ. 40 (2005) 633–639. [9] Y. Zhang, J. Zhang, H. Chen, X. Du, Q. Meng, Effects of step changes of temperature and humidity on human responses of people in hot-humid area of China, Build. Environ. 80 (2014) 174–183. [10] Z. Rongyi, X. Yizai, L. Jun, New conditioning strategies for improving the thermal environment, Proceedings of International Symposium on Building and Urban Environmental Engineering, Tianjin, 1997. [11] C.Y. Chun, A. Tamura, Thermal environment and human responses in underground shopping malls vs department stores in Japan, Build. Environ. 33 (1998) 151–158. [12] C.-P. Chen, R.-L. Hwang, S.-Y. Chang, Y.-T. Lu, Effects of temperature steps on human skin physiology and thermal sensation response, Build. Environ. 46 (2011) 2387–2397.

440

J. Xiong et al. / Physiology & Behavior 151 (2015) 426–440

[13] Y. Zhai, Y. Zhang, Q. Meng, H. Chen, J. Wang, Gender Differences in Thermal Comfort in a Hot–Humid Climate, 2014. [14] S. Karjalainen, Thermal comfort and gender: a literature review, Indoor Air 22 (2012) 96–109. [15] S. Qian, Q. Jiang, K. Liu, B. Li, M. Li, L. Li, et al., Effects of short-term environmental hyperthermia on patterns of cerebral blood flow, Physiol. Behav. 128 (2014) 99–107. [16] Y. Liu, Human behavior in different TDRAs, Physiol. Behav. 119 (2013) 25–29. [17] Y. Y., Z. L., W. L., Q. S., Experimental study on physiological responses and thermal comfort under various ambient temperatures, Physiol. Behav. 93 (2008) 310–321. [18] Z.J. Schlader, S.E. Simmons, S.R. Stannard, T. Mündel, The independent roles of temperature and thermal perception in the control of human thermoregulatory behavior, Physiol. Behav. 103 (2011) 217–224. [19] WHO, Obesity: preventing and managing the global epidemic. Report of a WHO consultation, World Health Organization Technical Report Series, 2000. [20] Y. Hu, X. Wu, Z. Hu, Study of formula for calculating body surface areas of the Chinese adults, Acta Phys. Sin. 51 (1999) 45–48. [21] H. He, G. Dong, Application study on Japanese “Subjective Fatigue Symptoms” (2002 version) in a Chinese manufacturer, Chinese Journal of Ergonomics 15 (2009) 26–28. [22] J.D. Hardy, E.F.D. Bois, G.F. Soderstrom, The technic of measuring radiation and convection: one figure, J. Nutr. (1938). [23] W. Liu, Z. Lian, Y. Liu, Heart rate variability at different thermal comfort levels, Eur. J. Appl. Physiol. 103 (2008) 361–366. [24] Cardiology TFotESo, Cardiology TFotESo. The North American Society of Pacing and Electrophysiology. Heart rate variability: standards of measurement, physiological interpretation and clinical use, Circulation 93 (1996) 1043–1065.

[25] L. Lan, Z. Lian, W. Liu, Y. Liu, Investigation of gender difference in thermal comfort for Chinese people, Eur. J. Appl. Physiol. 102 (2008) 471–480. [26] D.K. Tiller, L.M. Wang, A. Musser, M.J. Radik, Combined effects of noise and temperature on human comfort and performance, ASHRAE Trans. 522-40 (2010). [27] M.Y. Beshir, J.D. Ramsey, Comparison between male and female subjective estimates of thermal effects and sensations, Appl. Ergon. 12 (1981) 29–33. [28] Y. Xuejun, Human Physiological Reactions to Heat Stress in Space Flight, Fourth Military Medical University, 2000. [29] W.L. Kenney, A review of comparative responses of men and women to heat stress, Environ. Res. 37 (1985) 1–11. [30] P. Mehnert, P. Brode, B. Griefahn, Gender-related difference in sweat loss and its impact on exposure limits to heat stress, Int. J. Ind. Ergon. 29 (2002) 343–351. [31] G.S. Anderson, R. Ward, I.B. M., Gender differences in physiological reactions to thermal stress, Eur. J. Appl. Physiol. Occup. Physiol. 71 (1995) 95–101. [32] R.L. B., Sex differences in human thermoregulatory response to heat and cold stress, Human Factors: the Journal of the Human Factors & Ergonomics Society 21 (1979) 687–699. [33] J. Waugh, D. Gardner, D. Vines, Differences in capnography and pulse oximetry measurements related to gender, Chest 142 (2012) 939A. [34] M.D. Witting, S.M. Scharf, Diagnostic room-air pulse oximetry: effects of smoking, race, and sex, Am. J. Emerg. Med. 26 (2008) 131–136. [35] J. Zhang, Effect of age and sex on heart rate variability in healthy subjects, J. Manip. Physiol. Ther. 30 (2007) 374–379.