Sex differences in thermophysiological responses of elderly to low-intensity exercise during uncompensable heat strain

A significant rise in mortality, particularly among the elderly population, has been observed during heatwaves (Robine et al. 2008; Semenza et al. 1996). Although most studies indicate that females face a heightened mortality risk during heatwaves (Ballester et al. 2023; Folkerts et al. 2022; van Steen et al. 2019), some have demonstrated higher mortality risk in males (Khatana et al. 2022; Vaidyanathan et al. 2020). Females exhibit lower sweat gland output and less evaporative heat loss than males during heavy exercise (Gagnon 2012) resulting in less effective cooling in particular in dry heat. Gagnon and Kenny (2012) imposed a relative humidity of 12% which made it easy to evaporate the produced sweat. Additionally, females show an earlier onset of body core temperature rises when dehydrated compared to males (Wickham et al. 2021). However, the lower muscle mass and smaller body size of women lead to less heat generation and, therefore, require less sweat output in high ambient temperatures leading to no inherent disadvantage (Yanovich et al. 2020). On the other hand, body surface area of females is lower than that of males, thus reducing the area over which sweat can be effectively evaporated. Consequently, the unresolved nature of sex differences in vulnerability to heat exposure persists.

Humidity ramp protocols are commonly employed to assess human vulnerability to heat (e.g., Kenney and Zeman 2002; Ravanelli et al. 2018). In this protocol, humidity is incrementally increased to determine the environmental conditions (temperature and humidity) at which the core body temperature exhibits a significant rise. This shift is interpreted as the transition from compensable heat strain to uncompensable heat strain (Wolf et al. 2021a; Cottle et al. 2023; Cottle et al. 2022). The threshold from compensable to uncompensable heat strain is higher in unacclimatized females than males in wet humid climates exercising at 30% VO₂max, but not in hot dry climates (Kenney and Zeman 2002; Wolf et al. 2022). The threshold for males was about 31 °C WBGT ((indoor) wet-bulb globe temperature) and close to 33 °C for females in a wet humid climate (Kenney and Zeman 2002). The reason for this difference is likely that the reported metabolism of females is much lower (about 215 W) than that of males (about 400 W) at 30% VO₂max; and, therefore, also require a lower amount of sweat to evaporate to compensate for the heat production. In hot dry climates, the limit of evaporative heat loss is related to the amount of sweat that is produced, therefore, females have a relative disadvantage in hot dry conditions because on average they cannot sweat as much as males. While the thresholds from compensable to uncompensable heat strain are described, also for elderly recently (Wolf et al. 2023), the thermophysiological responses to heat in elderly males and females during uncompensable heat strain have not been investigated.

Therefore, we conducted a study to investigate differences in thermophysiological responses to low-intensity exercise in the heat between healthy elderly (60 years and older) males and females. Considering that elderly people have a considerable earlier transition from compensable to uncompensable heat strain that young people in increasing heat (Wolf et al. 2023), we expected that the elderly would be exposed to uncompensable heat strain with ambient temperatures of 39 °C and relative humidities increasing stepwise from 25% onwards. Non-weight bearing exercise (cycling) was chosen to avoid differences in metabolic rate due to differences in body weight (Yanovich et al. 2020) and enable a fair comparison of heat loss mechanisms between sexes when exposed to heat. We hypothesized that no sex differences would exist in core, skin and mean body temperature during uncompensabe heat strain, and no sex differences in sweat rate for low humidities, but higher sweat rates in males for high humidities in line with observations from Kenney and Zeman (2002) in young adults.

The estimated VO₂max using the YMCA test was approximately 35 ml/kg/min for females and 36 ml/kg/min for males (p = 0.69). The experiment lasted 74.5 ± 12.7 min for females and 86.9 ± 18.4 min for males on average (p = 0.04). Each participant was in the climatic chamber for at least 60 min. No participant reached Tgi values exceeding 39.5 °C so the termination was based on a Tgi rise of > 0.1 °C every 5 min for 15 consecutive minutes for all participants. Heart rate data were missing due to technical failure for one male and one female participant, metabolism for one female participant and Tgi for one male participant; all other data were acquired successfully. Tgi of all subjects increased after 20–30 min in the climatic chamber (Fig. 1). The individual data were shown to illustrate that the participants were experiencing uncompensable heat strain.

Figure 2 shows (the increase in) Tgi, Tsk, Tbody, and LSR against time for males and females. The first panel shows relative humidity as a reference. No sex differences were observed in increase in Tgi, Tsk, Tbody. However, LSR increases considerable more for males than females after minute 32, in particular for relative humidities exceeding 30%.

Table 2 shows the main results for males and females regarding metabolism, heart rate, temperatures and thermal sensation/thermal comfort for the 60 min period. Metabolism (in W/m²) did not differ between males and females (p = 0.10), and since the external power was fixed at 10 W, also absolute heat production did not differ between sexes (p = 0.75). Tgi was not significantly different between females and males (0.012 ± 0.004 °C/min vs. 0.011 ± 0.005 °C/min; p = 0.64). The increase in Tsk also did not differ between females and males (0.044 ± 0.007 °C/min vs. 0.038 ± 0.011 °C/min; p = 0.07). Females exhibited lower whole body sweat rates than males (0.29 ± 0.06 vs. 0.45 ± 0.14 mg/m²/min; p < 0.001) in particular at relative humidities exceeding 30%. WBSR and LSR were well correlated (r = 0.54, P < 0.002). Thermal sensation and thermal comfort did not differ between sexes (p = 0.30 for both).

To elucidate if elderly females or males are more vulnerable to heat stress, 15 females and 16 males performed low-intensity exercise in a high temperature setting with increasing humidities. The WBGT value corresponding to the temperature/humidity in the first 30 min was 28.9 °C which corresponds to a water vapour pressure of 1.7 kPa or 13 mmHg. Following the 30 min mark, a consistent increase in core temperature (Tgi) was observed, indicating uncompensable heat strain in the elderly participants. The increases in core, skin and mean body temperature during the period of increasing heat stress did not differ between sexes, thereby confirming the hypothesis.

As humidity levels increased, so did the observed difference in evaporated sweat between sexes, in line with our hypothesis. Local evaporative sweat rate was about 22% higher in males than females after 60 min of heat exposure coinciding with a relative humidity nearing 40% (Fig. 2). Despite the lower evaporative sweat rate in females under relatively humid conditions, they managed to constrain the rise in their mean body temperature to values similar to those in males. Excessive sweating beyond the necessary level implies rapid dehydration, potentially leading to an increase in core temperature due to the reduction in circulating blood volume (Sawka et al. 1985). The key challenge is to sweat an amount sufficient for cooling purposes, and females appear to be more adept at achieving this balance. Body core temperatures exceeding values of 42 °C are considered life-threatening (Kosaka et al. 2004). Since no discernible differences in core temperature were observed between sexes after 60 min of heat exposure, we cannot assert the existence of health-related disparities between the sexes in response to heat exposure.

A recent study investigated heat strain in elderly also using a humidity ramp protocol, but did not specifically look at sex differences (Wolf et al. 2023). In their research, uncompensable heat strain typically commenced at relative humidities averaging around 35%, a level higher than observed in our study. Participant characteristics like age (71 ± 6 years versus 69 ± 5 years in our study) and BMI (26 ± 5 kg/m² versus 24 ± 3 kg/m²) were comparable. In our study the elderly cycled at 10 W resistance while no resistance was imposed in the US study, but the metabolism (84 ± 15 W/m² at 40 °C versus 84 ± 19 W/m² in our study) was comparable. Gross efficiency in our study was less than 7%. This is common at low workloads—most energy goes to the movement itself, not to the bike (Reger et al. 2013). One possibility is that the US subjects may have undergone some level of heat acclimation, whereas in our study, all experiments were conducted during the winter months.

The amount of evaporated sweat was measured both for the entire body (WBSR) and locally (LSR). These two parameters were highly correlated and of the same magnitude, which suggests that the evaporated sweat measured at the lower arm can be considered as indicative for total evaporated sweat. In dry conditions, sweat production equals sweat evaporation, but in humid circumstances sweat production exceeds sweat evaporation. Females have been shown to produce less sweat than males during heavy exercise Gagnon and Kenny (2012). The observed sex differences in sweat production are likely not related to differences in the number of sweat glands per cm² between sexes (Schwiening et al. 2011) but to activation: females have a lower slope of the sweating response than males (Andérson et al. 1995). Although not significantly different (p = 0.07), the increase in Tsk tended to be higher in females than in males (0.044 ± 0.007 °C/min vs. 0.038 ± 0.011 °C/min). This may indicate that females partly compensate for the reduced amount of evaporated sweat by reducing dry heat gain to a larger extent than males.

A recent review on sex differences in thermal sensitivity (Greenfield et al. 2023) reported that most studies, but not all, reported greater warm sensitivity in young women than young men. Elderly females reported to feel a little hot to very hot in 28% of the cases, with males reporting this sensation in only 22% of the cases, showing a similar trend in elderly (Farias et al. 2022). In our study, non-parametric testing showed higher thermal sensation ratings in females than males for the first hour of heat exposure but not in thermal comfort (Table 2). Taken together, females tend to feel warmer than males in similar hot conditions but similarly comfortable.

A limitation of this study is that the findings may not be applicable to the general elderly population in the Netherlands since individuals with underlying health conditions were excluded. Additionally, most participants were recruited from sports facilities, potentially leading to a group that was relatively well-trained. Research indicates that the YMCA test tends to overestimate actual VO₂max in the elderly by about 7.5 ml/kg/min (Takeshima et al. 1992), resulting in actual values of approximately 28 ml/kg/min, considered average for males and above average for females (Letnes et al. 2023). In the older population, individuals are generally less physically fit as they tend to become more sedentary with age (Schutzer and Graves 2004). Furthermore, the general older population often experiences more age-related pathologies that can impact thermoregulation (Meade et al. 2020).

It is undisputed that core temperatures exceeding 42 °C are life-threatening (Kosaka et al. 2004), but is likely that heat related mortality is not directly linked to those extreme core temperature values (Kovats and Hajat 2008). Several heat stress indices are, therefore, available to warn people when critical values in thermal stress and/or strain are reached (Ioannou et al. 2022). One of the most common indicators is the WBGT value. The ISO WBGT-standard threshold for working 8 h at low metabolic rate (125–235 W: 150 W in this study) is set at WBGT 29 °C for non-acclimated adults (ISO_7243 2017). Although these values are specific for workers, they are often also applied to the general public. We observed uncompensable heat strain above 29.5 °C WBGT within a 1 h exposure. Standards are typically established to ensure that only 5% of the population is excluded; however, for the elderly, the average WBGT already approaches the limit values for a fit sample of the aged population. Therefore, it is advisable to consider age-dependent standards to account for this variation.

In conclusion, elderly females exhibit significantly lower sweat rates than their male counterparts during low-intensity exercise at ambient temperatures of 39 °C when humidity exceeds 30%. However, both elderly males and females demonstrate a comparable rise in core temperature, skin temperature, and mean body temperature, indicating similar health-related risks associated with heat exposure.