Short-term endurance training in the cold with layered clothing mitigates perceptual strain during exercise in the heat

Abstract

Purpose

We examined whether short-term endurance training in the cold combined with layered clothing enhances cardiovascular, thermoregulatory, and performance adaptations.

Methods

Sixteen healthy males completed five consecutive days of endurance cycling in the cold (10 °C, 50% relative humidity [rH]) while wearing either thermal layered clothing (WEAR) or light clothing (CON). Before and after training, participants performed a graded exercise test, a time-to-exhaustion test, and a heat stress test (30 min at 35 °C, 50% rH).

Results

During training, heart rate, skin temperature, and sweat rate were higher in WEAR than CON (P < 0.05). Maximal power output increased similarly in both groups (WEAR: + 7.2 ± 9.5%; CON: + 8.4 ± 7.1%, P < 0.05), while maximal oxygen uptake improved only in WEAR (52.4 ± 7.3 to 55.2 ± 7.1 mL/kg/min, P < 0.05). During the heat stress test, ratings of perceived exertion and thermal sensation decreased after training in WEAR only (P < 0.05), with no changes in sweat rate, skin temperature, or rectal temperature in either group.

Conclusion

Short-term endurance training in the cold combined with layered clothing attenuated perceptual strain during subsequent exercise in the heat, implying a practical alternative approach to traditional heat acclimation strategies.

Introduction

Exercising in hot environments impairs endurance performance and increases the risk of heat illness (Periard et al. 2021). To mitigate such adverse effects, athletes commonly undergo repeated heat exposure (i.e., heat acclimation) before competition in the heat. Traditional heat acclimation protocols typically involve 60–90 min of daily exercise in hot conditions over ~ 10–14 days (Periard et al. 2015). These interventions improve endurance capacity and heat tolerance through cardiovascular and thermoregulatory adaptations such as plasma volume expansion, increased sweat rate, and reductions in heart rate and core temperature (Periard et al. 2015, 2016).

Short-term (≤ 7 days) heat acclimation has gained increasing attention as a more practical alternative across many sports due to congested competition schedules and limited preparation time (Garrett et al. 2011; Chalmers et al. 2014). However, its implementation can be challenging in cold climates or during winter, particularly when environmental chambers are unavailable. To address this limitation, alternative approaches aimed at increasing thermal strain have been proposed, including sauna exposure, hot water immersion, and the use of thermal clothing (Heathcote et al. 2018; Stevens 2018; Jenkins et al. 2025). Among these methods, thermal clothing is particularly attractive because it requires no specialized facilities. For example, wearing an upper-body sauna suit increased peak rectal temperature (T_rec) and sweat rate during 30 min of cycling under both temperate (22 °C) and hot (45 °C) conditions (Willmott et al. 2018). Moderate increases in core temperature and sweat rate were also observed during 80 min of outdoor cycling (~ 17 °C) when wearing thermal clothing compared with light clothing (Stevens et al. 2017). Likewise, treadmill running for 60 min in a cool (15 °C) environment while wearing multiple clothing layers elevated T_rec above 38.5 ℃ (Ely et al. 2018), a threshold considered sufficient to induce heat acclimation adaptations (Fox et al. 1963).

In terms of training adaptations, an early review article highlighted that exercise training with thermal clothing is likely to improve heat tolerance, while evidence regarding effects beyond those induced by training itself is limited (Dawson 1994). Recently, Lundby et al. (2021) found that 10 days of cycling in a cool (18.4 °C) environment while wearing layered clothing significantly improved performance during a 30-min all-out cycling test in the heat, with effects comparable to training at 35℃ without additional clothing. Furthermore, 5 days of treadmill running in a temperate condition (20 °C) with overdressing induced greater reduction in resting and peak core temperature, skin temperature, and heart rate during 60-min running in the heat (40 °C) compared with the same training with light clothing (Greenfield et al. 2025). However, because these assessments were conducted under temperate conditions, it remains unclear whether comparable adaptations occur when training is undertaken in the cold (e.g., 10 °C). In cold conditions, a larger skin-environment thermal gradient enhances dry heat loss via convection and radiation, limiting core temperature rise (Blokker et al. 2022; Wallace et al. 2024). Wearing layered clothing restricts this heat dissipation, allowing metabolic heat production to elevate internal thermal strain. This combination may induce heat acclimation-like cardiovascular and thermoregulatory adaptations.

Therefore, this study tested the hypothesis that short-term endurance training in the cold (10 °C) while wearing layered clothing enhances cardiovascular, thermoregulatory, and performance adaptations compared with identical training performed in light clothing.

Methods

Participants

Sixteen recreationally active males (age: 20.8 ± 0.9 years, height: 172.7 ± 4.6 cm, body weight: 64.3 ± 7.2 kg) participated in the study. All participants were accustomed to low- to moderate-intensity cycling and had not performed any heat-based training for at least six months before the experiment. Participants received detailed information about the experimental procedures and provided written informed consent. The study was approved by the Research Ethics Committee of Ritsumeikan University, Japan (BKC-IRB-2020-048), and conducted in accordance with the Declaration of Helsinki.

Experimental overview

The study consisted of three performance tests – a graded exercise test (GXT), time-to-exhaustion (TTE) test, and heat stress test (HST) – conducted before and after five consecutive days of endurance training in the cold (Fig. 1). A five-day training regimen was chosen based on previous short-term heat acclimation programs as a practically applicable strategy for athletes (Garrett et al. 2011; Chalmers et al. 2014). Participants were randomly assigned to either a layered clothing group (WEAR, n = 8) or a light clothing group (CON, n = 8) for all training sessions. WEAR trained while wearing layered thermal clothing including caps, neck warmers, long-sleeved shirts, insulated jackets, long tights, insulated trousers, socks, and shoes. In contrast, CON wore standard (light clothing) attire including T-shirts, shorts, socks, and shoes.

Fig. 1.

Experimental overview of the short-term endurance training intervention performed with layered clothing (WEAR) or light clothing (CON), including pre- and post-training testing sequences. GXT graded exercise test, TTE time to exhaustion, HST heat stress test

Pre- and post-training tests were performed in a fixed order (GXT, TTE, HST) over three consecutive days, with two rest days following the final training session. Participants were instructed to refrain from caffeine, alcohol, or supplements and to avoid additional exercise from 24 h before the pre-training tests until completion of the post-training tests. All tests and training sessions were conducted in an environmental chamber (FCC-500S; Fuji Medical Science Co. Ltd., Chiba, Japan), with ambient temperature and relative humidity (rH) continuously monitored.

Pre- and post-training tests

Graded exercise test

The GXT was conducted on an electromagnetically braked cycle ergometer (Aerobike 75XLⅢ; Konami Corp., Tokyo, Japan) in a thermoneutral environment (25 °C, 50% rH) to determine maximal aerobic capacity. The test began at 70 W and increased by 30 W every 2 min until volitional exhaustion at a cadence of 70 rpm. Maximal power output (W_max) was defined as the highest workload completed for ≥ 1 min. Expired gases were continuously collected and analyzed using an automatic gas analyzer (AE300S; Minato Medical Science Co. Ltd., Tokyo, Japan). Heart rate (HR), stroke volume (SV), and cardiac output (CO) were measured using an impedance cardiac output meter (PhysioFlow Q-Link; Manatec Biomedical, Paris, France). Six electrodes were placed on the neck, chest, and back. Cardiorespiratory data were averaged over 30-s intervals, and maximal values for oxygen uptake (VO_2max), HR (HR_max), SV (SV_max), and CO (CO_max) were determined.

Time-to-exhaustion test

In a thermoneutral environment (25 °C, 50% rH), participants performed a constant-load cycling test to volitional exhaustion (70 rpm) on a cycle ergometer (Aerobike 75XLⅢ; Konami Corp., Tokyo, Japan). Workload was fixed at 80% of VO_2max (198 ± 38 W). The test was terminated when cadence dropped below 65 rpm for > 5 s, and time to exhaustion (TTE) was recorded.

Heat stress test

The HST consisted of 30 min of continuous cycling at 65% of VO_2max (152 ± 31 W) in a hot environment (35 °C, 50% rH). HR was continuously monitored using a wireless HR monitor (RCX5; Polar Electro Oy., Kempele, Finland). T_rec was measured using a thermistor (ITP010-11; Nikkiso Therm Co. Ltd., Tokyo, Japan) inserted 10 cm past the anal sphincter. Skin temperature was measured using surface thermistors (ITP082-24; Nikkiso Therm Co. Ltd.) attached to the left chest, arm, thigh, and calf. Mean skin temperature (T_skin) was calculated as T_skin = 0.3 × (chest + arm) + 0.2 × (thigh + calf) (Ramanathan 1964). Data were averaged over 5-min intervals. Ratings of perceived exertion (RPE, 10-point scale) and thermal sensation (TS, 9-point scale: 1 = very cold to 9 = very hot) were assessed every 10 min. Nude body weight was measured before and after the HST to estimate sweat rate.

Body composition and fluid balance

Body composition and fluid balance (including blood sampling) were assessed immediately before pre- and post-training GXT, following an overnight fast. Participants arrived at the laboratory at 8.30 am and rested for 10 min before measurements. Body weight, lean body mass, and body fat percentage were measured using a multi-frequency bioelectrical impedance analyzer (Inbody770; InBody Co. Ltd., Seoul, South Korea). Venous blood samples were collected using a 2.5 mL syringe, and hemoglobin (Hb) and hematocrit (Hct) concentrations were analyzed by a certified clinical laboratory (SRL Inc., Tokyo, Japan). Changes in plasma volume (ΔPV) were calculated as ΔPV = 100 × ([Hb_pre/Hb_post] × [100 − Hct_post]/[100 − Hct_pre] − 1) (Dill and Costill 1974). Urine samples were collected to assess urine specific gravity using a refractometer (PAL-09S; Atago Co. Ltd., Tokyo, Japan).

Training protocol

Participants completed five consecutive days of endurance cycling in a cold environment (10 °C, 50% rH). The ambient temperature of 10 °C represents the typical daily maximum temperature in winter in major cities in Japan (e.g., Tokyo, Osaka, and Nagoya). Each training session comprised four 20-min blocks, each including 5 min of cycling at 50 W followed by 15 min at 62% of W_max, and ended with 20 min at 38% of W_max (total duration of 100 min; Fig. 1). Pilot testing confirmed that this protocol elicited sufficient core temperature elevation (> 38.5 °C) when layered clothing was worn. During each 50 W stage of blocks 2–4, participants consumed 200 mL of water (20 °C; total 600 mL in both WEAR and CON) to minimize dehydration. HR, T_rec, T_skin, sweat rate, RPE, and TS were monitored on training days 1, 3, and 5. The cumulative heat strain index (CHSI) was calculated from HR and T_rec during each training session as follows (Frank et al. 2001):

  

Where: hb = heart beats, f_c0 = baseline HR (bpm), T_rec0 = baseline T_rec ( °C), and t = time (min) from the onset of exercise.

Statistical analyses

Data are expressed as mean ± standard deviation (SD). Two-way repeated-measures ANOVA (group × period [pre- and post-training]) was used to compare body composition and fluid balance, performance and physiological variables during the GXT, TTE, and training sessions (SPSS; IBM Corp., Armonk, USA). Physiological and perceptual responses during the HST were analyzed using two-way repeated-measures ANOVA (period [pre- and post-training] × time [baseline to 30 min]). Effect sizes were calculated as partial eta squared (η_p²), and values of 0.01, 0.06, and > 0.14 were considered as small, medium, and large, respectively (Cohen 1988). When significant main effects or interaction were observed, Tukey post hoc tests were performed. Statistical significance was set at P < 0.05.

Results

Training data

Table 1,Summarizes physiological and perceptual responses during the training sessions. Peak and mean HR were significantly higher in WEAR than CON (main effect of group: P < 0.001 [η_p² = 0.643] for peak HR; P = 0.001 [η_p² = 0.567] for mean HR), with significant reductions from day 1 to day 5 observed only in WEAR (main effect of period: P < 0.001 [η_p² = 0.475] for peak HR; P < 0.001 [η_p² = 0.474] for mean HR). Peak and mean T_skin were also higher in WEAR than in CON throughout the training period (main effect of group: P < 0.001 [η_p² = 0.838] for peak T_skin; P < 0.001 [η_p² = 0.893] for mean T_skin). Peak T_rec was significantly higher in WEAR on day 1, but this difference diminished on subsequent days (main effect of group: P = 0.017 [η_p² = 0.363]). Within WEAR, both peak and mean T_rec decreased from day 1 to day 5, whereas no changes were observed in CON (main effect of period: P = 0.002 [η_p² = 0.370] for peak T_rec; P = 0.023 [η_p² = 0.253] for mean T_rec). CHSI was significantly higher in WEAR than in CON on day 1 and 3, but not on day 5 (main effect of group: P = 0.008 [η_p² = 0.407]). Sweat rate, RPE, and TS were consistently higher in WEAR than in CON (main effect of group: P = 0.007 [η_p² = 0.415], P = 0.003 [η_p² = 0.481], and P < 0.001 [0.665], respectively).

Table 1.

Physiological and perceptual responses during training sessions in the layered clothing (WEAR) and light clothing (CON) groups

	WEAR group			CON group
	Day 1	Day 3	Day 5	Day 1	Day 3	Day 5
HR (bpm)
Peak	181 ± 7†	174 ± 7*†	171 ± 9*†	161 ± 10	156 ± 9	155 ± 5
Mean	159 ± 9†	154 ± 7†	150 ± 8*†	140 ± 9	137 ± 9	136 ± 6
Trec (℃)
Peak	38.9 ± 0.7†	38.6 ± 0.4	38.4 ± 0.4*	38.2 ± 0.2	38.1 ± 0.2	38.0 ± 0.2
Mean	38.2 ± 0.5	38.2 ± 0.3	37.9 ± 0.3*	37.9 ± 0.2	37.8 ± 0.2	37.8 ± 0.2
Tskin (℃)
Peak	37.1 ± 0.9†	36.8 ± 0.6†	36.8 ± 0.4†	34.2 ± 1.4	34.1 ± 0.9	34.3 ± 0.9
Mean	36.1 ± 1.0†	35.9 ± 0.7†	35.9 ± 0.4†	31.9 ± 1.9	31.6 ± 1.6	31.7 ± 2.3
CHSI (units)	428 ± 60†	315 ± 34*†	296 ± 39*	252 ± 40	184 ± 49	215 ± 59
Sweat rate (L/h)	1.0 ± 0.4†	1.1 ± 0.5†	1.1 ± 0.3†	0.6 ± 0.2	0.6 ± 0.2	0.6 ± 0.2
RPE	7.5 ± 2.1†	7.0 ± 1.7†	6.9 ± 1.5†	4.6 ± 1.3	4.6 ± 1.6	4.3 ± 1.6
TS	7.5 ± 1.8†	7.5 ± 1.9†	7.5 ± 1.4†	3.1 ± 2.1	3.1 ± 2.0	3.1 ± 1.5

HR heart rate, T_rec rectal temperature, T_skin mean skin temperature, CHSI cumulative heat strain index, RPE rating of perceived exertion, TS thermal sensation, Values are mean ± SD

*P < 0.05 vs. Day 1

†P < 0.05 vs. CON group

Body composition and fluid balance

No significant pre- to post-training changes were noted in either group for body weight (64.3 ± 9.8 to 64.2 ± 9.8 kg in WEAR; 64.2 ± 4.1 to 63.9 ± 4.2 kg in CON), lean body mass (55.5 ± 8.1 to 55.1 ± 8.1 kg in WEAR; 56.4 ± 4.3 to 55.7 ± 4.0 kg in CON), body fat percentage (13.7 ± 3.8 to 14.0 ± 3.9% in WEAR; 12.1 ± 2.3 to 12.8 ± 2.3% in CON), urine specific gravity (1.022 ± 0.009 to 1.024 ± 0.009 in WEAR; 1.022 ± 0.010 to 1.018 ± 0.008 in CON), Hb (15.2 ± 0.6 to 15.3 ± 0.8 g/dL in WEAR; 15.0 ± 0.9 to 15.2 ± 0.8 g/dL in CON), Hct (46.0 ± 1.4 to 46.2 ± 2.6% in WEAR; 45.5 ± 2.4 to 46.0 ± 1.9% in CON), or ΔPV (−0.7 ± 9.0% in WEAR; −2.0 ± 4.8% in CON).

Graded exercise test

As shown in Table 2, W_max increased from pre- to post-training in both groups (+ 7.2 ± 9.5% and + 8.4 ± 7.1% in WEAR and CON, respectively; main effect of period: P = 0.002 [η_p² = 0.521]). Absolute (+ 5.3 ± 4.6%) and relative VO_2max (+ 5.6 ± 3.8%) increased in WEAR only (main effect of period: P = 0.009 [η_p² = 0.400] and P = 0.004 [η_p² = 0.464], respectively). HR_max (−1.4 ± 2.5% in WEAR and −1.8 ± 4.3% in CON), SV_max (+ 13.2 ± 21.2% in WEAR and −2.8 ± 12.9% in CON), and CO_max (+ 11.6 ± 19.7% in WEAR and −1.8 ± 15.5% in CON) did not differ significantly within or between groups.

Table 2.

Performance and cardiorespiratory responses during the graded exercise test in the layered clothing (WEAR) and light clothing (CON) groups

	WEAR group		CON group
	Pre	Post	Pre		Post
Wmax (W)	269 ± 36	288 ± 42*	258 ± 59		276 ± 49*
Absolute VO2max (mL/min)	3365 ± 510	3544 ± 568*	3253 ± 621		3344 ± 523
Relative VO2max (mL/kg/min)	52.4 ± 7.3	55.2 ± 7.1*	50.8 ± 7.8		52.0 ± 5.8
HRmax (bpm)	197 ± 9	194 ± 11	189 ± 10		187 ± 9
SVmax (mL)	106 ± 18	118 ± 21	122 ± 23		121 ± 27
COmax (L/min)	20.1 ± 2.8	22.1 ± 3.2	22.4 ± 4.1		21.7 ± 4.5

W_max maximal power output, VO_2max maximal oxygen uptake, HR_max maximal heart rate, SV_max maximal stroke volume, CO_max maximal cardiac output

Values are mean ± SD

^*P < 0.05 vs. Pre

Time-to-exhaustion test

The TTE improved significantly after training in both groups (from 21.5 ± 9.2 to 30.7 ± 14.1 min [+ 46.5 ± 65.6%] in WEAR and from 25.4 ± 8.0 to 37.8 ± 14.4 min [+ 51.4 ± 49.1%] in CON; main effect of period: P = 0.001 [η_p² = 0.573]), with no between-group difference.

Heat stress test

Peak HR decreased significantly after training in both groups (WEAR: −5.9 ± 4.8%, main effect of period: P = 0.032 [η_p² = 0.563]; CON: −4.3 ± 2.9%, main effect of period: P = 0.008 [η_p² = 0.662]). Peak T_rec (WEAR: −0.3 ± 0.7%; CON: −0.3 ± 0.9%) and T_skin (WEAR: + 1.4 ± 3.4%; CON: −0.4 ± 1.5%) did not change (Fig. 2). RPE and TS scores decreased significantly after training in WEAR only (main effect of period: P = 0.013 [η_p² = 0.613] and P = 0.007 [η_p² = 0.667], respectively; Fig. 3). Sweat rate was unchanged in both groups (+ 15.1 ± 27.4% and + 4.2 ± 24.1% in WEAR and CON, respectively; main effect of period: P = 0.234 [η_p² = 0.100]).

Fig. 2.

Heart rate (A), rectal temperature (B),  and mean skin temperature (C) during the heat stress test in the layered clothing (WEAR) and light clothing (CON) groups. Values are mean ± SD. †P < 0.05 vs. post

Fig. 3.

Rating of perceived exertion (A) and thermal sensation (B) during the heat stress test in the layered clothing (WEAR) and light clothing (CON) groups. Values are mean ± SD. †P < 0.05 vs. post

Discussion

We examined the effects of wearing layered clothing during five consecutive days of endurance training in a cold environment. The main findings were that this strategy improved VO_2max, did not affect TTE in a thermoneutral environment, and attenuated perceptual strain during a HST without concomitant changes in thermoregulatory responses.

Elevating core temperature above 38.5°C is considered a prerequisite for effective heat acclimation (Fox et al. 1963). During the first training session, peak T_rec in WEAR exceeded this threshold and remained elevated for a substantial duration despite the cold environment (10°C), reaching levels comparable to those observed during traditional heat acclimation interventions (Lorenzo et al. 2010; Moss et al. 2020) and exercise while wearing thermal clothing in temperate environments (Ely et al. 2018; Lundby et al. 2021). These findings indicate that the present training strategy successfully induced meaningful heat strain during early exposure. However, peak T_rec gradually decreased across sessions and no longer reached 38.5°C by the final training day. Although this reduction is consistent with the onset of acclimation, the reduced heat stimulus in later sessions may explain the absence of thermoregulatory adaptations (T_rec, T_skin, sweat rate) during the HST in WEAR. Stevens et al. (2018) examined performance and physiological adaptations following 2 weeks of field-based training (six running/cycling sessions per week) in a temperate environment (~ 18 °C, ~ 67% rH) while wearing light clothing or thermal clothing. However, they found no benefit of wearing thermal clothing on 40-min time trial performance or thermoregulatory adaptation in the heat (32.5°C, 55% rH). Although absolute core temperature data during training sessions were not reported, they suggested that thermophysiological strain was insufficient to elicit heat acclimation, as mean HR during training was similar between groups (139 bpm). In contrast, Greenfield et al. (2025) conducted 5 days of treadmill-based training in a temperate condition (20°C) while participants wore either thermal or light clothing, and found higher peak T_rec, T_skin, HR, and sweat rate with thermal clothing than light clothing during both the initial and final training sessions. Notably, peak T_rec remained elevated to 38.5°C with thermal clothing even during the final session (Greenfield et al. 2025). Furthermore, the 5-day training with thermal clothing induced significant thermoregulatory and cardiorespiratory adaptations (i.e., reductions of resting and peak core temperature, skin temperature, and heart rate) during 60 min of treadmill running in the heat (40°C) (Greenfield et al. 2025). Our results showed markedly higher HR, T_skin, and CHSI in WEAR than CON over the training period; however, the difference in T_rec diminished during the latter training sessions. Maintaining sufficient heat stress throughout training –via higher exercise intensity or greater insulation –may therefore be required to elicit robust thermoregulatory adaptations.

In contrast to the unchanged thermoregulatory responses, perceptual strain (RPE and TS) was reduced after five consecutive days of training. Although T_skin is closely associated with TS and thermal comfort (Song et al. 2025), the perceptual improvements observed here were not accompanied by changes in T_rec or T_skin. This dissociation aligns with previous findings showing continued reductions in TS and improvements in thermal comfort when heat acclimation was extended from 5 to 10 days, despite plateaued T_rec and T_skin after five days (Moss et al. 2020), implying that perceptual adaptations may occur independently of thermoregulatory adjustments. Lower RPE and TS at a given workload decrease subjective strain, whereas elevated effort and thermal perception impair self-paced performance in the heat (Flouris and Schlader 2015). Moreover, exacerbated thermal sensation has been shown to impair cognitive performance during exercise in the heat (Qin et al. 2025). Although exercise capacity during the HST was not directly assessed, short-term cold training with layered clothing may nonetheless enhance endurance performance in hot environments by attenuating perceptual strain compared to training with light clothing.

Another notable finding is that VO_2max increased significantly after training in WEAR (+ 5.6 ± 3.8%) but not in CON (+ 3.0 ± 6.1%). Traditional 10-day heat acclimation protocols improve VO_2max and CO_max, and these effects are largely attributed to plasma volume expansion (Lorenzo et al. 2010), although improvements in VO_2max remains inconsistent across studies (Karlsen et al. 2015; Keiser et al. 2015). In our study, plasma volume did not increase, likely because the heat strain induced by layered clothing was insufficient to evoke this response. Nevertheless, the magnitude of increases in VO_2max (+ 5.6%) and CO_max (+ 2.0 L/min) in WEAR was comparable to values reported following traditional heat acclimation (Lorenzo et al. 2010). The contribution of plasma volume expansion in mediating VO_2max improvements remains debated, as increases in plasma volume without concurrent increases in hemoglobin mass may cause hemodilution without enhancing oxygen delivery (Warburton et al. 2000). Alternative mechanisms may therefore explain the observed VO_2max increase, including augmented muscle vasodilatory capacity (Richey et al. 2022), optimized blood flow redistribution (Chen et al. 2013), and enhanced muscle oxidative capacity (Patton et al. 2018). These adaptations could facilitate oxygen delivery and utilization during maximal exercise, partially accounting for the improved aerobic capacity. Higher HR through training sessions in WEAR represents augmented physiological strain (internal training load), which may support cardiovascular adjustments, as the time spent under high internal load is considered crucial to improve VO_2max (Buchheit and Laursen 2013). Consistent with this interpretation, Lundby et al. (2021) also reported improved aerobic capacity after endurance training with thermal clothing in the absence of hematological and thermoregulatory adaptations. Further research is needed to elucidate the mechanisms underpinning these responses.

Both W_max during the GXT and TTE at 80% VO_2max significantly improved after training in WEAR and CON, indicating that the training stimulus itself was sufficient to enhance performance, particularly given that participants were not accustomed to high-intensity cycling. These general training adaptations may have masked any additional effects of layered clothing, and the observed improvement in VO_2max may be partially attributable to exercise training itself. Therefore, caution is needed when interpreting these results. Importantly, exercise capacity (i.e., W_max and TTE) was assessed only under thermoneutral conditions. Meta-analytic evidence indicates that the performance benefits of heat acclimation are small when aerobic capacity is measured in thermoneutral conditions, but moderate to large when measured in the heat (Waldron et al. 2021). Thus, training with layered clothing may confer greater improvements in exercise capacity measured under hot conditions compared with light clothing.

The present study has several limitations. First, only male participants were recruited. A recent review highlighted that the magnitude of physiological adaptations (e.g., HR and core temperature) following heat acclimation is comparable between males and females (Wickham et al. 2021), and females can adapt to heat regardless of menstrual cycle phase (Lundby et al. 2023; Richard et al. 2025). On the other hand, females may require a longer heat acclimation period to achieve physiological adaptations comparable to those observed in males (Wickham et al. 2021), suggesting that a 5-day training program may be insufficient in females. In addition, the menstrual cycle influences both resting and exercising core temperatures in the heat (Christison et al. 2025), which could complicate the interpretations of training-induced changes in core temperature. For these reasons, only male participants were included in the present study. However, the next step should be to examine the effects of short-term endurance training with layered clothing in females. Second, the sample size was relatively small (n = 8 per group) because all experiments were completed within a single season to minimize confounding effects of natural heat acclimation. Third, the training duration was short (5 days), which may be insufficient compared to traditional heat-acclimation protocols (~ 10–14 days). Fourth, post-tests were performed after a two-day rest period following completion of training to minimize the influence of fatigue resulting from relatively high exercise intensity compared with typical heat acclimation protocols, as well as differences in internal load between groups. However, several physiological adaptations may be more pronounced immediately after training and may therefore have been underestimated in the present study (Daanen et al. 2018). Finally, exercise capacity was assessed only under thermoneutral conditions; future studies should evaluate exercise capacity in differing hot environments (i.e., hot-dry and hot-humid) to enhance practical relevance.

In conclusion, short-term endurance training with layered clothing improved VO_2max and attenuated perceptual strain during exercise in the heat, without inducing measurable cardiovascular or thermoregulatory adaptations. This strategy may offer a practical and time-efficient preparatory strategy for athletes training for competition in hot conditions.

Abbreviations

ANOVA

Analysis of variance

CO

Cardiac output

CO_max

Maximal cardiac output

CON

Light clothing group

CHSI

Cumulative heat strain index

GXT

Graded exercise test

HR

Heart rate

HR_max

Maximal heart rate

HST

Heat stress test

Hb

Hemoglobin

Hct

Hematocrit

RPE

Ratings of perceived exertion

SD

Standard deviation

SV

Stroke volume

SV_max

Maximal stroke volume

TS

Thermal sensation

TTE

Time to exhaustion

T_rec

Rectal temperature

T_skin

Mean skin temperature

VO_2max

Maximal oxygen uptake

WEAR

Layered clothing group

W_max

Maximal power output

rH

Relative humidity

ΔPV

Plasma volume change

η_p²

Partial eta squared

Author contributions

KY, AM, RY, and KG designed the research. KY, SO, and AM conducted experiments and analyses. KY wrote the manuscript, and OG and KG revised it. All authors approved the final manuscript.

Funding

Open Access funding provided by Ritsumeikan University. The present study was funded by Panasonic Corporation.

Declarations

Conflict of interest

RY is an employee of Panasonic Corporation. All other authors declare no conflicts of interest.