Participants

Participant characteristics are presented in Table 1. We conducted a power analysis using 80% power and a significance level of 0.05 to calculate the minimum sample size required based on reported age-related differences in whole body heat loss in young, middle-aged, and older males during exercise in the heat at a moderate heat load (30). Assuming an effect of 30% and standard deviation of 25%, the calculated minimum sample size per group was n = 6. Thus 40 males volunteered for the study and were divided into 4 groups of 10 young (21 ± 1 yr), 10 middle-aged trained (49 ± 5 yr), 10 middle-aged untrained (48 ± 5 yr), and 10 older (65 ± 3 yr) males. Participants were matched for height (P = 0.436), body mass (P = 0.895), and body surface area (P = 0.762). We carefully matched our participants for physical characteristics by first recruiting an individual for one of the groups (i.e., young) and then selected a match for each of the other groups (i.e., middle-aged trained/untrained and older). However, there was a significant difference among groups for percent body fat (P < 0.001) and o_2peak (P < 0.001). Specifically, the young males had a lower percent body fat compared with the middle-aged untrained and older males (P < 0.05). Moreover, the young and middle-aged trained males had a similar o_2peak (P > 0.05), which was greater compared with the middle-aged untrained and older males (P < 0.05). All participants were healthy and nonsmokers. The Kohl physical activity questionnaire (27) was used to assess the participant's physical activity level (i.e., duration, frequency, and mode) for a period of 3 mo before the start of the experimental test session. The results from the questionnaire revealed that young, middle-aged untrained, and older adults were performing a minimum of 30 min of exercise/day (i.e., walking, running, cycling) for 1–3 day/wk (habitually active), while middle-aged trained males were highly endurance trained (≥30 min of exercise per/day of running, cycling, or cross country skiing for 4–7 days/wk).

Table 1.

Participant characteristics

	Age, yr	Height, cm	Mass, kg	BSA, m2	o2peak, ml·kg−1·min−1	Body fat, %
Y	21 ± 1	179 ± 6	79 ± 9	1.98 ± 0.12	50.0 ± 3.7	13.4 ± 3.3
MAT	49 ± 5*	181 ± 5	82 ± 8	2.02 ± 0.11	51.0 ± 6.8	19.1 ± 4.3
MUT	48 ± 5*	179 ± 8	82 ± 12	2.00 ± 0.18	37.3 ± 3.5*†	23.1 ± 6.3*
O	65 ± 3*	176 ± 5	80 ± 7	1.96 ± 0.06	37.9 ± 7.8*†	21.2 ± 5.6*

Open in a separate window

Values are means ± SD. Y, young; MAT, middle-aged trained; MUT, middle-aged untrained; O, older; BSA, body surface area; o_2peak, peak oxygen uptake.

^*Significant difference from young males.

^†Significant difference from middle-age trained males.

Experimental Design

Each participant completed one preliminary and one experimental session. During the preliminary session, body height, mass, and density, as well as o_2peak were determined. Body height was determined using a stadiometer (model 2391; Detecto, Webb City, MO), while body mass was measured using a digital high-performance weighing terminal (model CBU150X; Mettler Toledo, Mississauga, ON, Canada). Body surface area was subsequently calculated from the measurements of body height and mass (8). Body density was measured using the hydrostatic weighing technique and used to calculate body fat percentage (36). o_2peak was measured on a recumbent cycle ergometer (Corival; Lode, Groningen, The Netherlands) during a progressive incremental exercise protocol that consisted of a 2-min warm-up at 40 W followed by 20-W increments every min until the participant could no longer maintain a pedaling cadence of at least 60 rpm. For the older males, a 12-lead ECG was monitored continuously throughout the maximal exercise test by a qualified technician to detect any abnormalities in heart activity. If detected, participants were excluded from the study and referred to their physician. No participants were excluded from the study based on this criterion.

The experimental protocol was performed in a whole body direct air calorimeter outside of summer months (September to May only) to ensure that there was no potential confounding effect of acclimatization. An equal number of participants within each group performed the experimental protocol in the morning and in the afternoon. Participants consumed a standardized light meal or snack of dry toast and orange juice before their arrival (a minimum of 2 h before testing) and avoided major thermal stimuli on their way to the laboratory (i.e., running, cycling, etc.). Strenuous activity and alcohol were avoided for 24 h and caffeine for 12 h before testing sessions. To ensure euhydration, participants drank ~250 ml of water before bed in the morning of the experimental trial and within 2 h of the start of the trial. No fluid was ingested for the duration of the experimental protocol.

The calorimeter was regulated to an ambient temperature of 40°C and 15% RH. Since some participants may not have been able to withstand high levels of exercise intensity, and therefore high exercise-induced heat loads (i.e., >600 W), the high ambient temperature (i.e., 40°C) was used to create an additional heat load (i.e., environmental heat load). All participants wore a light pair of athletic shorts and sandals. Following instrumentation, participants rested for a 30-min habituation period on an upright seated cycle ergometer in the calorimeter. Habituation was followed by three bouts of 30-min cycling exercise (Ex) at increasingly greater rates of metabolic heat production (Ex1: 300 W; Ex2: 400 W; and Ex3: 500 W). Each exercise bout was followed by a 15-min recovery (Rec) period (Rec1, Rec2, and Rec3) in the calorimeter. The rates of metabolic heat production employed were equivalent to ~25 and ~33% o_2peak for Ex1, ~35 and ~47% o_2peak for Ex2 and ~44 and ~60% o_2peak for Ex3 for young/middle-aged trained and older/middle-aged untrained, respectively.

Measurements

Whole body evaporative loss and dry heat exchange as well as change in body heat content were quantified using the modified Snellen direct air calorimeter. A full technical description of the fundamental principles and performance characteristics of the Snellen calorimeter is available elsewhere (33). Data from the direct calorimeter were collected continuously at 8-s intervals during the experimental sessions. Real-time data were displayed and recorded on a personal computer with Lab-VIEW software (version 7.0; National Instruments, Austin, TX). The rate of evaporative heat loss was calculated from the calorimetry data using the following equation: [mass flow × (humidity_out − humidity_in] × 2,426)/60, where mass flow is the rate of air mass (kg air/s); (humidity_out − humidity_in) is the difference in absolute humidity (g water/kg air) between the in and out flows of the calorimeter; and 2,426 is the latent heat of vaporization of sweat (J/g sweat). The rate of dry heat loss, from radiation, convection, and conduction was calculated from calorimetry data using the following equation: [mass flow × (temperature_out − temperature_in) × 1,005]/60, where mass flow is the rate of air mass (kg air/s); (temperature_out − temperature_in) is the difference in inflow-outflow air temperature (°C) of the calorimeter; and 1,005 is the specific heat of air [J·(kg air·°C)⁻¹]. Direct calorimetry measures whole body evaporative loss and dry heat exchange (radiation, conduction, and convection), yielding an accuracy of ± 2.3 W for the measurement of total heat loss. Indirect calorimetry was used to measure metabolic heat production. Electrochemical gas analyzers located outside of the calorimeter (AMETEK model S-3A/1 and CD 3A; Applied Electrochemistry, Pittsburgh, PA) were used to continuously determine the concentration of expired O₂ and CO₂ during testing sessions. To account for respiratory heat exchange, expired air was recycled back into the calorimeter. The change in body heat content was subsequently calculated by subtracting the total amount of heat production and heat loss over the experimental protocol. The amount of evaporation required to achieve heat balance (E_req) was defined as the combination of metabolic heat production and dry heat exchange.

Local sweat production was measured using the ventilated capsule technique. A 3.8-cm² plastic capsule was attached to three skin sites (upper back, chest, and forearm) with an adhesive ring and topical skin glue (Collodion HV; Mavidon Medical Products, Lake Worth, FL). Compressed dry air was passed through the capsule at a rate of 1.0 l/min. Water content of the effluent air was measured using high precision dew point mirrors (model 473; RH Systems, Albuquerque, NM). Local sweat rate was calculated using the difference in water content between effluent and influent air multiplied by the flow rate and normalized for the skin surface area under the capsule. Laser-Doppler velocimetry was employed to measure skin blood flow (PeriFlux System 5000, main control unit; PF5010 LDPM). A laser-Doppler probe (Perimed Integrating Probe 413, Järfälla, Sweden) was affixed to the forearm's surface in an area that did not seem overly vascular upon visual inspection and provided stable readings at rest. To measure maximal skin blood flow, the heater housing the laser-Doppler probe was heated to 44°C at the end of Rec3 until maximal skin vasodilation was achieved (~30–40 min) (42).

Esophageal temperature was measured using a thermocouple temperature probe (Mon-a-therm General Purpose Temperature Probe; Mallinckrodt Medical, St. Louis, MO). The esophageal probe was inserted 40 cm past the nostril entrance while the participants sipped water (100–300 ml) through a straw. Mean skin temperature was calculated as the weighted average of four skin temperature measurements: biceps 30%, chest 30%, quadriceps 20%, and front calf 20% (32). Data were collected in 15-s intervals and were displayed and recorded in spreadsheet format using a HP Agilent data acquisition module (model 3497A; Agilent Technologies Canada, Mississauga, ON, Canada) and a personal computer with LabVIEW software (Version 7.0; National Instruments).

Heart rate was monitored, recorded continuously, and stored using a Polar coded WearLink and transmitter, Polar RS400 interface, and Polar ProTrainer 5 software (Polar Electro, Oy, Finland). Preexperimental session urine specific gravity was determined in duplicate using a handheld total solids refractometer (model TS400; Reichter, Depew, NY).

Whole Body Direct Calorimetry

Baseline and exercise.

The rate of metabolic heat production at baseline was similar for young (117 ± 19 W), middle-aged trained (123 ± 9 W), middle-aged untrained (112 ± 11 W), and older (112 ± 13 W) males (P = 0.212). By experimental design, the rate of metabolic heat production was similar among groups (P = 0.758) during each of the exercise periods (average across groups: Ex1 = 305 ± 18; Ex2: 404 ± 18; and Ex3: 504 ± 17 W). The rates of evaporative heat loss and E_req during exercise are shown in Fig. 1. The rate of evaporative heat loss at baseline was similar for young (146 ± 42 W), middle-aged trained (150 ± 46 W), middle-aged untrained (132 ± 40 W), and older (125 ± 34 W) males (P = 0.516). However, a main effect of group on the rate of evaporative heat loss (P = 0.002) was detected during exercise, whereby the rate of evaporative heat loss was greater for the young and middle-aged trained compared with the middle-aged untrained and older males at the end of Ex2 (P < 0.05) and Ex3 (P < 0.05). Rates of dry heat exchange during baseline were not different among the young (−77 ± 16 W), middle-aged trained (−83 ± 12 W), middle-aged untrained (−79 ± 11 W), and older (−78 ± 12 W) males (P = 0.696). Likewise, there were no differences in dry heat exchange among groups at the end of each exercise bout, an indication of the amount of heat gained by the body from the environmental heat load (average among groups: Ex1 = −105 ± 14; Ex2: −114 ± 17; and Ex3: −117 ± 20 W; P = 0.384). As a consequence, the total heat load for each exercise (rate of metabolic heat production and dry heat gain or E_req) was similar among groups (Ex1 = 410 ± 28; Ex2: 518 ± 25; and Ex3: 621 ± 25 W; P = 0.443). Whole body sweat rate was significantly different among groups during exercise (P = 0.001), whereby whole body sweat rate was significantly greater in the young and middle-aged trained groups compared with the older and middle-aged untrained groups at the end of Ex2 (P < 0.05) and Ex3 (P < 0.05; see Table 3).

Open in a separate window

Fig. 1.

Means ± SE values for evaporative heat loss (solid bars) and the required amount of evaporation for heat balance (E_req, striped bars) measured over three 30-min exercise bouts (Ex 1, Ex 2, and Ex 3) and three 15-min recovery periods (Rec 1, Rec 2, and Rec 3) in a hot, dry [40°C, 15% relative humidity (RH)] environment. The black bars/stripes represent the young (Y) group, the dark grey bars/stripes represent the middle-aged trained (MAT) group, the light grey bars/stripes represent the middle-aged untrained (MUT) group, and the white bars/grey and black stripes represent the older (O) group. There were no differences in the required amount of evaporation for heat balance between groups. Significance level was set at P ≤ 0.05. *Significant difference in evaporative heat loss from young. †Significant difference in evaporative heat loss from middle-aged trained.

Table 3.

Local heat loss, whole body sweat rate, and heart rate responses during each Ex/Rec cycle

	Ex1	Rec1	Ex2	Rec2	Ex3	Rec3
LSR–chest, mg·min−1·cm−2
Y	0.37 ± 0.19	0.18 ± 0.11	0.50 ± 0.20	0.18 ± 0.14	0.66 ± 0.26	0.20 ± 0.16
MAT	0.44 ± 0.15	0.29 ± 0.13	0.60 ± 0.26	0.26 ± 0.09	0.79 ± 0.38	0.34 ± 0.15
MUT	0.37 ± 0.16	0.23 ± 0.12	0.55 ± 0.27	0.29 ± 0.14	0.68 ± 0.31	0.40 ± 0.21
O	0.38 ± 0.19	0.22 ± 0.11	0.51 ± 0.17	0.26 ± 0.12	0.58 ± 0.22	0.39 ± 0.25
LSR–back, mg·min−1·cm−2
Y	0.44 ± 0.24	0.23 ± 0.11	0.54 ± 0.22	0.22 ± 0.10	0.67 ± 0.22	0.27 ± 0.12
MAT	0.44 ± 0.18	0.28 ± 0.14	0.59 ± 0.22	0.26 ± 0.11	0.80 ± 0.25	0.31 ± 0.11
MUT	0.30 ± 0.14	0.23 ± 0.10	0.48 ± 0.20	0.30 ± 0.15	0.60 ± 0.25	0.36 ± 0.17
O	0.39 ± 0.16	0.28 ± 0.13	0.56 ± 0.17	0.35 ± 0.11	0.67 ± 0.15	0.39 ± 0.13
LSR–arm, mg·min−1·cm−2
Y	0.36 ± 0.13	0.16 ± 0.07	0.53 ± 0.16	0.20 ± 0.10	0.82 ± 0.14	0.24 ± 0.13
MAT	0.40 ± 0.18	0.23 ± 0.07	0.52 ± 0.08	0.25 ± 0.10	0.77 ± 0.23	0.25 ± 0.12
MUT	0.34 ± 0.14	0.29 ± 0.11	0.53 ± 0.22	0.38 ± 0.12	0.77 ± 0.21	0.41 ± 0.15*†
O	0.43 ± 0.17	0.26 ± 0.10	0.55 ± 0.14	0.30 ± 0.10	0.68 ± 0.16	0.43 ± 0.14*†
WBSR, g/min
Y	9.1 ± 1.1	5.9 ± 1.1	11.7 ± 1.0	6.2 ± 1.2	13.8 ± 1.3	5.7 ± 0.7
MAT	9.1 ± 1.1	5.7 ± 1.1	11.7 ± 0.5	7.0 ± 1.4	13.7 ± 0.6	6.3 ± 0.6
MUT	8.0 ± 0.9	5.5 ± 1.0	10.5 ± 0.8*†	6.3 ± 1.0	12.3 ± 1.3*†	6.1 ± 2.3
O	8.3 ± 0.9	5.7 ± 0.9	10.5 ± 0.9*†	6.8 ± 0.8	12.0 ± 1.1*†	7.0 ± 1.0
SkBF, %max
Y	43.2 ± 12.9	33.9 ± 9.5	44.3 ± 12.8	32.8 ± 9.3	49.1 ± 12.7	30.9 ± 9.0
MAT	49.4 ± 9.9	39.7 ± 16.2	54.4 ± 14.0	44.3 ± 16.9	62.2 ± 15.7	41.0 ± 9.8
MUT	44.6 ± 15.7	32.3 ± 14.5	52.5 ± 16.0	31.8 ± 14.6	54.7 ± 14.9	36.9 ± 13.5
O	51.9 ± 12.0	42.4 ± 9.5	55.1 ± 14.1	41.9 ± 8.9	60.2 ± 16.1	40.2 ± 8.5
HR, beats/min
Y	95 ± 8	81 ± 10	111 ± 11	84 ± 12	131 ± 14	90 ± 7
MAT	85 ± 14	70 ± 15	99 ± 13	74 ± 14	117 ± 18	79 ± 17
MUT	99 ± 11	85 ± 11	119 ± 18†	92 ± 14	143 ± 23†	101 ± 17
O	93 ± 11	76 ± 8	109 ± 17	84 ± 10	135 ± 14	94 ± 14

Open in a separate window

Values are means ± SD. LSR, local sweat rate. WBSR, whole body sweat rate. SkBF, skin blood flow. HR, heart rate.

^*Significant difference from young males.

^†Significant difference from middle-age trained males.

The changes in body heat content are presented in Fig. 2. There was a main effect of group on change in body heat content during exercise (P < 0.001). The young males had a lower change in body heat content for all three exercise periods (Ex1–3) compared with the middle-aged untrained (percent difference from young = Ex1: 48%; Ex2: 55%; and Ex3: 57%) and older males (percent difference from young = Ex1: 47%; Ex2: 57%; and Ex3: 54%; P < 0.05). Additionally, the middle-aged trained males had a lower change in body heat content for Ex3 (P < 0.05) compared with the middle-aged untrained (percent difference from middle-aged trained for Ex3 = 44%) and older (percent difference from middle-aged trained for Ex3 = 41%) males.

Open in a separate window

Fig. 2.

Means ± SE values for changes in body heat content during each exercise/recovery cycle as well as the total change in body heat content over the exercise protocols in a hot, dry (40°C, 15% RH) environment. The solid bars represent changes in body heat content during exercise and the striped bars represent the changes in body heat content during recovery. The black bars/stripes represent the young (Y) group, the dark grey bars/stripes represent the middle-aged trained (MAT) group, the light grey bars/stripes represent the middle-aged untrained (MUT) group, and the white bars/grey and black stripes represent the older (O) group. Significance level was set at P ≤ 0.05. *Significantly different from young. †Significantly different from middle-aged trained.

Onset Thresholds and Thermosensitivities, Time Constant, and Amplitude Values for Whole Body Evaporative Heat Loss

Exercise.

Mean body temperature onset thresholds, thermosensitivities, time constant, and amplitude values for whole body evaporative heat loss during each exercise bout are presented in Table 2. The mean body temperature onset threshold for whole body evaporative heat loss did not differ among groups (P = 0.294). However, a main effect of group on thermosensitivity was observed (P = 0.001). For Ex1, the young males had a significantly greater thermosensitivity for the whole body evaporative heat loss response compared with the other three groups (P < 0.05). Additionally, during the third exercise bout, the young and middle-aged trained males had a significantly greater thermosensitivity compared with both the middle-aged untrained and older males (P < 0.05). Furthermore, there was a significant difference among groups in time constant values for the rate of evaporative heat loss during exercise (P = 0.045). This was demonstrated by the young having a shorter time to reach 63.2% of the evaporative heat loss response during the third exercise bout compared with the other three groups. Likewise, the amplitude for evaporative heat loss was different among groups (P = 0.036) such that the young and middle-aged trained males had a significantly greater amplitude for Ex3 compared with both the middle-aged untrained and older males.

Table 2.

Time constants, amplitudes, onset thresholds, and thermosensitivities of evaporative heat loss for each exercise/recovery bout

	Ex1	Rec1	Ex2	Rec2	Ex3	Rec3
τ, min
Y	11.0 ± 10.4	2.99 ± 1.67	6.5 ± 3.1	3.09 ± 0.74	4.8 ± 1.4	2.87 ± 1.03
MAT	16.4 ± 5.4	5.69 ± 4.70	8.0 ± 5.2	5.39 ± 1.22*	7.9 ± 3.1*	3.93 ± 1.08
MUT	21.9 ± 14.1	3.98 ± 1.85	6.9 ± 2.9	5.84 ± 0.91*	7.0 ± 2.6*	5.51 ± 2.10*
O	14.0 ± 6.1	5.09 ± 2.90	8.7 ± 3.2	7.33 ± 2.60*	8.5 ± 2.3*	7.09 ± 4.66*
Amplitude, W
Y	172 ± 43	119 ± 71	198 ± 52	223 ± 54	268 ± 51	255 ± 53
MAT	175 ± 46	145 ± 52	207 ± 51	185 ± 51	265 ± 69	272 ± 29
MUT	148 ± 56	110 ± 50	179 ± 46	177 ± 28	215 ± 61*†	201 ± 66*†
O	158 ± 28	129 ± 44	180 ± 31	161 ± 48	189 ± 50*†	189 ± 69*†
Onset threshold of evaporative heat loss, °C
Y	36.63 ± 0.40		36.82 ± 0.37		36.94 ± 0.33
MAT	36.57 ± 0.21		36.72 ± 0.20		36.83 ± 0.26
MUT	36.91 ± 0.70		36.88 ± 0.14		37.08 ± 0.21
O	36.64 ± 0.28		36.88 ± 0.31		36.97 ± 0.31
Thermosensitivity of evaporative heat loss, W/°C
Y	1,575 ± 1,160		1,217 ± 559		1,295 ± 629
MAT	1,022 ± 270*		1,141 ± 536		1,031 ± 352
MUT	879 ± 415*		833 ± 604		513 ± 238*†
O	721 ± 318*		694 ± 424		411 ± 263*†

Open in a separate window

Values are means ± SD. Ex, exercise; Rec, recovery; τ, time constant.

^*Significant difference from young males.

^†Significant difference from middle-age trained group. Mean body temperature was used to calculated the onset threshold and thermosensitivity.

Esophageal and Mean Skin Temperatures

Baseline and exercise.

Esophageal and mean skin temperatures are presented in Fig. 3. Esophageal temperatures during baseline were not different among the young (36.77 ± 0.40°C), middle-aged trained (36.72 ± 0.22°C), middle-aged untrained (36.83 ± 0.24°C), and older (36.83 ± 0.26°C) males (P = 0.787). However, there was a main effect of group on esophageal temperature during exercise (P = 0.013). This was evidenced by the young and middle-aged trained males having lower esophageal temperatures at the end of Ex3 compared with middle-aged untrained and older males (P < 0.05). Baseline values for mean skin temperature were not different among the young (35.33 ± 0.66°C), middle-aged trained (35.26 ± 0.44°C), middle-aged untrained (35.16 ± 0.37°C), and older (35.26 ± 0.48°C) males (P = 0.882). Likewise, there was no difference among groups for mean skin temperature (P = 0.765) during exercise. Similarly, when values were examined from a change from baseline, a main effect of group on esophageal temperature during exercise (P = 0.016) was observed. The young and middle-aged trained males demonstrated smaller changes in esophageal temperatures at the end of Ex3 (young: 0.65°C; middle aged trained: 0.57°C) compared with middle-aged untrained (1.10°C) and older (0.95°C) males (P < 0.05). However, there was no difference in the change in mean skin temperature (P = 0.357) during exercise among groups.

Open in a separate window

Fig. 3.

Means ± SE values for esophageal and mean skin temperatures during each exercise/recovery cycle in a hot, dry (40°C, 15% RH) environment. The black circles represent the young (Y) group, the dark grey circles represent the middle-aged trained (MAT) group, the light grey circles represent the middle-aged untrained (MUT) group, and the white circles represent the older (O) group. Significance level was set at P ≤ 0.05. Analysis was performed at the end of each exercise/recovery period. *Significantly different from young. †Significantly different from middle-aged trained.

Effects of Age on the Physiological Capacity to Dissipate Heat

In the current study, we employed a unique incremental exercise model involving progressive increases in heat load (and therefore E_req) to determine the threshold at which age-related differences in the body's physiological capacity to dissipate heat occurs. Of note, when we factor in the heat gained from the environment (i.e., dry heat gain), the net heat loads, or E_req, were ~400 W for Ex1, ~500 W for Ex 2, and ~600 W for Ex 3 (Fig. 1). In the present study, we showed that whole body evaporative heat loss was reduced in both middle-aged untrained and older males compared with young males at the end of the two highest heat loads employed (Fig. 1). Similar findings were reported by previous studies employing an exercise-induced heat load similar to that of Ex2 in our study (i.e., 400 W), albeit lower environmental stress (35°C and 20% RH), and therefore minimal dry heat gain (28, 30). As such, heat loads greater than ~400 W, derived from a combination of metabolic and environmental heat, are sufficient to exceed the physiological capacity to dissipate heat of middle-aged as well as older adults matched for aerobic fitness. In line with our study hypothesis, as the heat load increased, the degree of impairment in whole body evaporative loss was greater in both the middle-aged untrained and older adults. While the young adults were able to increase their rate of whole body evaporative heat loss from Ex1 to Ex 2 by ~28% and Ex2 to Ex3 by ~18%, the older adults had an increase in whole body evaporative heat loss from Ex1 to Ex2 of ~26% and from Ex2 to Ex3 of only ~14%. This translated into 69 and 95 kJ of greater body heat storage at the end of Ex2 and Ex3, respectively, in the older compared with the young adults. Although, a true measure of maximal heat loss capacity would be observed if no increase in the rate of whole body heat loss occurred from Ex2 to Ex3, our observations provide evidence to indicate that both the middle-aged untrained and older adults are approaching near maximal levels of heat dissipation.

Insight into whether impairment in heat loss occurs centrally (neural activity/integration) or peripherally (end organ) can be gleaned from examining the mean body temperature at which heat loss is activated (central) or from the rate of rise in the heat loss response relative to increasing mean body temperature (peripheral) (12, 31). While the mean body temperature onset threshold of evaporative heat loss was not different between each group (Table 2), the thermosensitivity of the evaporative heat loss response was reduced in the middle-aged untrained and older males. These results are consistent with previous studies that have reported that older adults have a reduced rate of heat loss (i.e., sweating) for a given increase in mean body temperature (1, 17, 30), which is indicative of a peripheral modification (31). In our study, we observed the greatest level of impairment during Ex3 where the time constant of the evaporative heat loss response was lower in young (4.8 ± 1.4 min) compared with both middle-aged untrained (7.0 ± 2.6 min) and older (8.5 ± 2.3 min) males (Table 2). The lower thermosensitivities and greater time constants were coupled with reduced amplitude of evaporative heat loss during the third exercise bout. It has been suggested that the attenuated heat loss response in older adults during heat stress is due to a decrease in cholinergic sensitivity and lower sweat gland output, rather than fewer heat-activated sweat glands ultimately resulting in a reduced capacity for heat dissipation in older adults (1, 22). We showed that this may be true for middle-aged untrained males as well. As a consequence of the reduced thermosensitivity and magnitude of evaporative heat loss, the change in body heat content was greater in middle-aged untrained and older males relative to their younger counterparts, respectively, by 48 and 47% for Ex1, 55 and 57% for Ex2, and 57 and 54% for Ex3.

Effects of Aerobic Fitness on Age-Related Impairments in Heat Loss

We observed that the middle-aged trained group had a significantly greater capacity for whole body heat dissipation at the moderate (400 W) and high (500 W) heat load conditions compared with the age-matched untrained group (Fig. 1). The differences in heat loss between the two middle-aged groups were solely due to the level of aerobic fitness, since the middle-aged trained males participated in regular endurance-type exercise training (i.e., running, cycling, cross country skiing), while the untrained males did not. Furthermore, participants were matched for physical characteristics such as body mass, body surface area, and body composition, which are known to influence heat dissipation and therefore body heat storage (13, 34). The increased capacity for heat dissipation in middle-aged trained males was the result of an increased thermosensitivity of the evaporative heat loss response combined with greater amplitude of increase in the rate of evaporative heat loss (Table 2). As such, evaporative heat loss reached significantly greater values for a given mean body temperature in the trained compared with untrained middle-aged counterparts in addition to the higher rate of evaporative heat loss at the last two exercise bouts. This finding is consistent with previous studies showing that endurance-trained adults demonstrate greater thermosensitivity of the sweating and/or skin blood flow response compared with their untrained counterparts (14, 41). It has been proposed by some that this response is in part attributed to partial acclimation caused by regular vigorous exercise training (3, 5, 11, 13, 34). In the present study, the greater rate of heat dissipation measured in the middle-aged trained adults translated into a 20 and 44% reduction in the amount of heat stored in the moderate and high heat load exercise bouts (i.e., Ex2 and Ex3), respectively, relative to the middle-aged untrained group. Future studies should be conducted to assess differences between nontrained adults with a low vs. high o_2peak to discern if the level of o_2peak per se (likely genetically determined) may influence this response.

Improvements in thermoregulatory capacity due to high levels of aerobic fitness are not always observed. Jay et al. (21) examined two groups of young (19–24 yr) males matched for body mass and surface area during exercise for 60 min at a fixed rate of metabolic heat production (540 W) in a thermoneutral environment. Jay et al. reported no differences in rectal temperature, whole body sweat rate, or local sweat rate (upper back) between the two groups despite the large differences in o_2peak (~40 vs. 60 ml·kg⁻¹·min⁻¹). Thus, while aerobic fitness may not be a factor affecting heat loss in young males when exercising at moderate levels of heat production, we showed that maintaining a higher level of aerobic fitness in middle-aged males promotes a greater capacity to dissipate heat. Furthermore, consistent with previous studies, when older untrained are compared with older trained adults during exercise (35–67% o_2peak), heat loss responses of sweating and skin blood flow are reported to be lower in the older untrained adults (18, 41). However, it is unknown if the results would have been the same if both groups had exercised at the same rate of heat production, thereby providing a similar thermal drive (9). Therefore, further research is required to examine the effect of aerobic fitness with increasing age (i.e., >60 yr).

Whole Body Calorimetry vs. Core Temperature and Local Heat Loss Responses

An interesting finding of the current study is that we only observed a significantly greater increase in esophageal temperature at the end of the third exercise bout (Fig. 3), despite the greater change in body heat content measured at all heat loads in the older and middle-aged untrained males compared with their younger counterparts. Ultimately, direct calorimetry allows us to precisely measure the amount of heat that is stored in the body irrespective of differences in tissue heat distribution that may occur as a result of age-related changes in tissue blood distribution during and following exercise. While changes in esophageal temperature provide us with an indication of where some of the heat is stored, they cannot be used as a reliable measure to quantify whole body heat storage (25). By the same logic, the reverse is correct. The calorimeter can only tell us how much heat is stored in the body but not where it is stored. Of note, the measurement of esophageal temperature is affected by the powerful vascular changes that occur in this region. This effect is clearly observed in our study findings such that, as noted above, we only observed a significantly greater increase in esophageal temperature by the end of the third exercise bout between the middle-aged untrained and older group compared with both the young and middle-aged trained group. In contrast, we observed significant differences in heat storage by Ex1 for the middle-aged untrained and older males relative to the young and middle-aged trained males. From a practical standpoint, measurement of body core temperature using esophageal temperature may underestimate the level of thermal strain experienced by a middle-age untrained and older adult exercising in the heat especially when exercise is of short duration and/or low intensity.

In the present study, age-related differences in whole body evaporative heat loss (and therefore sweat production) were not paralleled by similar differences in local sweat rate at the three skin sites. Previous studies have reported within-group regional differences in sweating during passive heating and/or exercise within groups (10, 19, 20, 37, 38). However, the measurement of whole body evaporative heat loss by direct calorimetry allows us to precisely measure the net consequences of aging per se on whole body sweat production without the confounding effects of regional differences in sweating. While age-related differences in local skin blood flow responses have been observed (15, 16, 23), the pattern of response can differ between skin sites (20, 37, 38). In the present study, we only measured skin blood flow at one site, thus it is plausible that differences might have been observed at other sites. Taken together, our findings demonstrate that caution should be used when employing surrogate measures of body heat storage (i.e., core temperature) and local heat loss responses for the purpose of assessing age-related differences in the body's physiological capacity to dissipate heat, especially at low to moderate heat loads. These thermoregulatory measures can underestimate the extent to which aging may alter the body's physiological capacity to dissipate heat.

Postexercise Response

We observed a rapid reduction in whole body evaporative and local heat loss in all groups at the cessation of exercise despite sustained elevations in esophageal temperature, which is the typical pattern of response postexercise (25, 26). The rapid reductions in heat loss have been attributed to nonthermal factors overriding the thermal control of heat loss during postexercise recovery in young adults (25, 26), thus altering the body's ability to dissipate heat during recovery. In the present study, we observed a lower amplitude of change in evaporative heat loss for the third recovery period in the middle-aged untrained and older males compared with their young and middle-aged trained counterparts (Table 2). Furthermore, it took a longer amount of time to achieve ~63% of the evaporative heat loss response for Rec3 in older males. These findings may imply that the level of influence of nonthermal factors on heat loss postexercise may be reduced in middle-aged untrained and older adults as a result of the greater thermal drive, associated with the progressively greater heat storage during exercise. Moreover, we see that the rates of whole body heat loss are slightly elevated in the middle-aged untrained and older males compared with the young and middle-aged trained males during the third recovery period. It is possible that the greater rates of heat loss during recovery in the middle-aged untrained and older groups are due to some compensatory mechanism to try to offset the greater amounts of heat stored during exercise. Therefore, it brings into question if these differences would become more pronounced had we extended the recovery time period. Despite the few previous studies examining heat loss during the postexercise recovery period in older adults (29, 30, 40), future research is required to examine the potential influence of thermal and nonthermal factors on heat loss responses postexercise as a function of age and aerobic fitness.

Limitations

Due to technical limitations associated with performing measurements of whole body heat loss in a calorimeter (i.e., subject must remain isolated in the calorimeter and access in and out of temperature chamber housing the calorimeter is restricted during the trial), we were unable to measure changes in hydration status during the experimental trial. As a consequence, we were unable to determine the potential effects of age-related differences in fluid distribution on whole body heat dissipation as a function of increases in exercise-induced heat loads. However, by design we employed an incremental exercise model that required that the young, middle-aged, and older adults exercise at the same fixed absolute heat load and therefore requirement for heat loss. This experimental model was used to ensure that any differences in sweating (and therefore fluid loss) would be the result of age-related differences and not due to differences in the requirement for heat loss. Additional research is required to determine how changes in hydration status may influence this pattern of response as a function of increasing heat loads.

Perspectives

Advanced aging is associated with adverse changes to the thermoregulatory system, but when do these alterations matter? The results from this study provide important insight into the level of heat load in which differences in the capacity to dissipate heat occur as a function of age and aerobic fitness. We showed for the first time that the impairment in thermoregulatory function in older as well as middle-aged untrained males is heat load dependent. Thus our results can be used by health care providers, exercise specialists, health and safety managers, sporting event organizers (i.e., marathons, soccer tournaments, etc.), and others to better understand the level of heat stress in which dangers associated with exercise in hot conditions occur and decide when it is safe to perform physical activity in the heat. As such, it is critical that the previously mentioned groups reevaluate heat exposure thresholds and adjust these limits to include information based on the age as well as level of aerobic fitness of the individual to reduce the risk of heat-related illness/injury for all adults. However, during a work shift, it is possible that older adults self-pace differently relative to their younger counterparts as a way to mitigate the level of physical/thermal strain they experience. Thus, while older adults, and/or those who are less physically active, might have a reduced capacity to dissipate heat during work in the heat, they may experience similar levels of thermal strain as their younger and/or endurance-trained counterparts when allowed to self-pace. Furthermore, the participants in the present study were relatively healthy. Therefore, older adults with health conditions such as cardiovascular disease, diabetes etc. may be at an even greater risk of heat-related illness and/or injury during work in the heat.

We thank all the participants who volunteered for the present study. The authors thank Michael Sabino of Can-Trol Environmental Systems Limited (Markham, ON, Canada) for their support. We also thank all the members of the Human and Environmental Physiology Research Unit who assisted with data collection.