Core temperature responses to compensable versus uncompensable heat stress in young adults (PSU HEAT Project)

Abstract

INTRODUCTION

The frequency, duration, and severity of heat waves have increased globally over the past several decades (1). The continued warming of the climate and associated heat waves are projected to cause an increase in heat-related morbidity and mortality. For example, extreme heat events are associated with an increased risk in emergency department visits due to heat-related illness (2). Furthermore, in Europe, future climate projections based on Representative Concentration Pathway (RCP) 4.5 and 8.5 have predicted 30,867–45,930 excess heat-related deaths from 2036–2064 and 46,690–117,333 excess attributable deaths from 2071–2099, respectively (3). In addition, projections incorporating socioeconomic factors into climates represented by RCP4.5 and 8.5 simulations project 5,032–7,935 and 5,130–8,079 excess heat-related deaths, respectively, in Houston, Texas from 2061–2080 (4). However, many of these studies detailing future heat-related health risks are overly simplistic as it can be hard to incorporate population-wide adaptative measures to extreme heat or the complex physiological responses that humans experience during these events (5). It is imperative to identify the specific combinations of temperature and humidity that increase the risk of heat-related mortality and illness among various populations. Identifying those critical environmental thresholds can lead to the development of _targeted policy and safety interventions to mitigate future heat-related morbidity and mortality. That is the overarching goal of the PSU HEAT (Pennsylvania State University—Human Environmental Age Thresholds) project. To date, the PSU HEAT project has primarily focused on a heterogenous sample of young, healthy adults to establish a best-case baseline for comparative purposes with future data in vulnerable populations (6).

The specific combinations of temperature and humidity above which heat balance cannot be maintained, resulting in a continuous rise in body core temperature (T_c), are termed critical environmental limits (7–10). Environmental conditions just below the critical environmental limit are compensable and increases in T_c are minimal for a given metabolic rate; combinations of temperature and humidity above the critical environmental limit are uncompensable and T_c increases at a higher rate. As part of the PSU HEAT project, we previously published critical environmental limits in young, healthy adults at two low metabolic rates—those associated with activities of daily living (MinAct; ∼0.45 L/min V̇o₂) and during light ambulation (LightAmb; ∼0.85 L/min V̇o₂)—across a wide range of thermal environments (6). However, heat storage (S) and the rate of change in T_c above and below those critical limits, i.e., during compensable and uncompensable heat stress can provide important information regarding human thermoregulation, health, and safety during prolonged heat exposures in a variety of environments.

The primary aim of this investigation was to compare and contrast calculated S and the change in core (gastrointestinal) temperature (T_gi) over time (i.e., the slope of the T_gi response) during MinAct and LightAmb above and just below critical environmental limits in warm-humid (WH) versus hot-dry (HD) environments. In addition, we examined whether individual characteristics (i.e., sex, aerobic fitness, body size) or time of year of testing (seasonality) were associated with the rate of increase in T_gi in WH or HD environmental conditions at each metabolic rate. By calculating the slopes of the T_gi response over time during low-intensity physical activity in minimally compensable and uncompensable conditions, the time to meet a generally accepted T_c criterion associated with heat-related illness can be calculated from any theoretical T_c.

METHODS

Testing Procedures

Experimental trials were conducted on separate days with at least 72 h between visits. Before each experimental session, subjects were instructed to abstain from alcoholic beverages and vigorous exercise for 24 h and from caffeine for 12 h. On arrival, participants provided a urine sample to ensure euhydration, defined as urine specific gravity ≤ 1.020 (USG; PAL-S, Atago, Bellevue, WA) (8). Subjects performed light physical activity in an environmental chamber at two low metabolic intensities reflecting the metabolic demand of activities of daily living (MinAct) or slow walking (LightAmb) (11, 12). Subjects cycled on a cycle ergometer (Lode Excalibur, Groningen, The Netherlands) against zero resistance at a cadence of 40–50 rpm for MinAct trials and walked on a motor-driven treadmill at a speed of 2.2 mi/h and grade of 3% for LightAmb trials.

Critical environmental limits were identified as previously described (6, 13) using a controllable environmental chamber at five constant dry-bulb temperatures (T_db) of 34°C, 36°C, 38°C, 40°C, and 42°C and two constant water-vapor pressures (P_a) of 12 and 16 mmHg. Following a 30-min equilibration period, either the P_a (during constant T_db trials, P_crit) or T_db (during constant P_a trials, T_crit) in the environmental chamber was increased in a stepwise fashion (1 mmHg or 1°C) every 5 min (Fig. 1). We have previously reported excellent reliability and validity of this protocol to identify critical environmental limits (14). In this paper, warm-humid conditions (WH; ∼35°C and 59%–72% RH) include data combined from 34°C and 36°C P_crit trials, and hot-dry conditions (HD; 43–48°C and 18%–22% RH) include data combined from 12 and 16 mmHg T_crit trials. During each experiment, chamber data (i.e., T_db, P_a, and RH), T_gi, and HR were continuously monitored and subjects free-pedaled or walked continuously until a clear rise in T_gi was observed. Each subject completed experimental trials in multiple environmental conditions at both metabolic rates. Experimental trials lasted ∼90–120 min.

Open in a separate window

Figure 1.

Representative tracing of the time course of gastrointestinal temperature (T_gi; B), dry-bulb temperature (T_db), and ambient water vapor pressure (P_a) for a MinAct trial with increasing P_a (A). The dashed vertical line in A and B denotes the P_a at which T_gi begins to continuously rise (i.e., heat stress becomes uncompensable). In this case, the T_gi inflection point (i.e., critical water vapor pressure, P_crit) occurs at P_a = 26 mmHg. B: the compensable and uncompensable slope of T_gi and heat storage components. MinAct, minimal activity.

Determination of Critical Environments

The critical environmental limit was determined from the raw data by a continuous rise in T_gi following a steady state. A line was drawn between the data points starting at the point at which T_gi began to equilibrate (typically 30–60 min after beginning physical activity), and a second line was drawn at the end of the T_gi steady state until the completion of the study (Fig. 3). The average T_db and P_a for the 2 min immediately preceding the inflection point was defined as the critical environmental limit. Following the inflection point, subjects continued exercising in the environmental chamber for ∼15–30 min to ensure a clear and continuous rise in T_gi occurred. Inflection points were determined by visual inspection, as previously reported (9, 10, 15, 16). We have previously demonstrated excellent interrater reliability for the determination of the T_gi inflection point (intraclass correlation coefficient = 0.913) (10). Mean critical environmental loci from each environmental condition were plotted on a standard psychrometric chart with lower bounds of the 95% CI for MinAct and LightAmb trials to delineate relatively safe (compensable) from potentially unsafe (uncompensable) environmental conditions (Fig. 2, updated from Ref. 6 with additional subject data).

Open in a separate window

Figure 2.

A standard psychrometric chart showing empirically derived mean critical environmental limits (symbols and solid lines) for minimal activity (MinAct; A) and light ambulatory activity (LightAmb; B) trials. Dashed lines denote the lower bounds of the 95% confidence interval for each condition. Updated from Wolf et al. (6).

Open in a separate window

Figure 3.

Representative tracing of the time course of gastrointestinal temperature (T_gi) during a progressive heat stress protocol, as described in methods. Dashed lines are drawn through the data points to represent the slope of T_gi used to calculate the time (h) to 40.2°C from a theoretical 37°C T_gi baseline.

RESULTS

DISCUSSION

The PSU HEAT Project is designed to determine critical environmental limits for heat balance, i.e., those combinations of ambient temperature and humidity above which thermal balance (defined as no or minimal rise in T_c during prolonged exposure) is not possible. Those psychrometric limits have been published (6) and are updated here. To our knowledge, this is the first study to quantify heat storage and the rate of change in T_c above and just below these critical environmental thresholds in young adults. Under compensable heat stress in HD environments, heat storage was higher than in WH environments (P < 0.05) resulting in a greater but more variable ΔT_c (P ≥ 0.06) despite the same metabolic heat production. Increased heat storage in HD was due to increased dry heat gain, and therefore increased skin temperature, despite greater evaporative heat losses. On the other hand, once critical environmental limits were exceeded (uncompensable heat stress), there was no effect of environment or metabolic rate.

As expected, sweat rate and V̇O₂ were greater during LightAmb versus MinAct trials and sweat rate was higher in HD than WH. There was no effect of sex, V̇o_2max, or seasonality of testing on S or ΔT_c in any condition. However, a larger body surface area-to-mass ratio (i.e., smaller body size) was associated with a lower β₁ during LightAmb in WH environments. Thus, larger individuals have a higher metabolic heat production when walking up even low grades on a treadmill, resulting in a greater ΔT_c in WH but not HD environments. These data are consistent with the results being driven by biophysical heat exchange principles between humans and the environment at a given metabolic heat production.

Under HD ambient conditions in which heat balance is limited by sweating capacity rather than evaporative capacity, higher maximal sweating rates (commonly observed in men compared with women) should be advantageous during prolonged heat stress to slow the rise of T_gi. However, there were no sex differences in β₁ at either metabolic rate or environmental condition. The absence of a sex difference in β₁ is likely due to the fact that, although lower sweat rates in women may translate to lower critical environmental limits when the sexes are matched for metabolic heat production (10, 16), differences in sweat rate likely have minimal influence on the slopes of T_c over time (0.06–0.15°C/h in the present study) when heat stress is compensable (i.e., the requirements for heat balance are met).

Because body surface area-to-mass ratio influences heat dissipation capacity and, therefore, the maintenance of heat balance (21, 22), we examined associations between body surface area-to-mass ratio and the rate of change in T_gi during compensable and uncompensable heat stress. Our study showed a correlation between body surface area-to-mass ratio and β₁ during LightAmb in WH conditions such that the slope of T_gi was lower for those with a larger body surface area-to-mass ratio. However, this correlation did not exist during LightAmb in HD environments or during MinAct in either environment. Although speculative, the lack of correlation during LightAmb in HD may be due to increased dry heat gain for those with a larger body surface area despite similar sweating rates, potentially off-setting any differences in metabolic heat production. Conversely, in WH conditions a larger body surface area confers an advantage for both evaporative and dry heat losses. The lack of correlation between body surface area-to-mass ratio and β₁ during MinAct is likely explained by similar metabolic heat production across body masses given the nonweight bearing nature of cycle ergometry, thus negating the thermoregulatory disadvantage imparted by increased body mass.

Associations between heat acclimatization, aerobic fitness, and thermoregulatory function are well documented (23–26). Using the same progressive heat stress protocol detailed herein, we have previously reported that in heat acclimated subjects T_c typically equilibrates as a relatively horizontal line, whereas unacclimated subjects often display a gradual rise that precedes the upward inflection at the critical environmental limit (9, 27). Although we did not account for acclimatization status during subject recruitment, any effect of acclimatization on the rise in T_gi during prolonged heat stress should be captured by randomizing season of testing. Similarly, because aerobic fitness is associated with improved thermoregulatory function (24, 28), we hypothesized that subjects with a greater V̇o_2max would have an attenuated rise in T_gi during compensable heat stress. However, in this sample, there was no correlation between seasonality or aerobic fitness with the rate of change in T_gi. The lack of an effect of fitness may be explained by the very low metabolic rates at which the subjects were working during this study. Further, although seasonality was used as a proxy for acclimatization, it is likely that these subjects did not spend adequate time outdoors in the heat to become acclimatized, explaining the lack of a seasonality effect.

The change in T_c during prolonged heat stress is driven by a change in body heat storage during thermal imbalance. As such, we calculated S and the change in T_c during compensable and uncompensable heat stress in varying environmental conditions. In the present study, S was greater in HD compared with WH environments at both metabolic rates; however, there was not a significant difference in β₁. However, the mean increase in β₁ from HD to WH was 100% in MinAct and 67% in LightAmb. This statistical discrepancy is driven by an increase in mean skin temperature during HD conditions in which T_db and convective heat gain were greatest, eliciting a greater heat storage, although increased heat losses via sweating in HD offset the increased skin temperature. Importantly, quantifying S and ΔT_c during heat stress is critical for developing safety guidelines, evidence-based alert communication, policy decisions, triage for impending heat events, and implementation of other safety interventions during extreme heat events.

In the present study, we calculated the rate of change in T_gi during minimally compensable (just below the upper limit for the maintenance of heat balance) and uncompensable heat stress in WH versus HD environments. Together, these findings provide important information regarding thermoregulation and health that can be used for future policy decisions and safety interventions during prolonged heat stress. From the rate of change in T_gi calculated in this study, projections can be made to and from any hypothetical T_gi temperature for an infinite number of real-world scenarios. For example, during light ambulation in a compensable WH environment with a baseline T_gi of 37°C, it would take an average of 36 h without any intervention to reach a T_gi of 40.2°C (see Fig. 3). However, during minimal activity in an uncompensable HD environment with a baseline T_gi of 37.5°C, it would only take 3–5 h without any intervention to reach a T_gi of 40.2°C.

These findings should be taken with caution for a variety of reasons. In independent living situations, it would be rare to continue steady-state activity once subjective and objective manifestations of heat strain appear. Similarly, in our experimental paradigm, no mitigating intervention strategies were employed (slowing down, replacing lost fluid, fans, movement to air conditioning, etc.) throughout the trial. The thresholds calculated using these slopes of T_gi should not be considered as a demarcation between healthy and unhealthy or as a threshold for survival, but rather limits where mitigation efforts become vital. In addition, a longer period of activity in uncompensable heat stress may lead to a more precise slope.

These results support the notion that, given prolonged exposure to combinations of temperature and humidity associated with uncompensable heat stress and a continued rise in T_c, heat-related morbidity and mortality will become of greater risk. With increases in the frequency, severity, and intensity of heat waves, along with continued warm-season temperature increases due to climate change, we are experiencing the environmental thresholds identified in the present study more frequently (29). When environmental conditions are above the thresholds for heat balance, it is crucial to encourage people to limit physical activity, replace lost fluid, and/or seek cooler temperatures to prevent rises in T_c to temperatures associated with increased risk of heat-related illness. Importantly, the data in this study were collected in young subjects in which thermoregulatory capacity is preserved; thus, further investigation is critical to understand how more vulnerable populations, such as the aged, respond to compensable and uncompensable heat stress.

GRANTS

This research was supported by the National Institute on Aging Grant T32 {"type":"entrez-nucleotide","attrs":{"text":"AG049676","term_id":"16586568","term_text":"AG049676"}}AG049676 to the Pennsylvania State University (to R.M.C. and D.J.V.) and National Institutes of Health Grant R01 AG067471 (to W.L.K.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

R.M.C. and W.L.K. conceived and designed research; R.M.C., Z.S.L., D.J.V., and S.T.W. performed experiments; R.M.C. and Z.S.L. analyzed data; R.M.C., S.T.W., and W.L.K. interpreted results of experiments; R.M.C., D.J.V., and S.T.W. prepared figures; R.M.C. drafted manuscript; R.M.C., Z.S.L., D.J.V., S.T.W., and W.L.K. edited and revised manuscript; R.M.C., Z.S.L., D.J.V., S.T.W., and W.L.K. approved final version of manuscript.

REFERENCES