Combined torso and foot heat exposure reduces acute ad libitum energy intake in young adults Authors: Ge Li, Daiki Watanabe, Yumiko Inoue, Qinlin Jin, Chiho Ezoe, Nan An, Jason Kai Wei Lee, Motohiko Miyachi Published: 2026 ABSTRACT ABSTRACTLowering body temperature with cooling vests has been shown to increase energy intake (EI), but it is unclear whether a heating strategy would have the opposite effect and modify EI. We examined the acute effects of combined torso and foot heat exposure via a hot vest and a warm foot bath on EI. Eight young females and eight young males performed two randomized and crossover trials, 28 d apart. Trials on female were conducted during the early follicular phases. In the control trial (CON), participants remained seated from 09:30 to 11:30. In the hot trial (HOT), they remained seated from 09:30 to 10:30, after which they wore a hot vest (~45°C) and took a foot bath (~43°C) from 10:30 to 11:30. From 11:40, participants were given up to an hour to consume a buffet meal. EI was calculated from the buffet meal. Tympanic and skin temperatures, appetite, and thermal sensation were assessed. Tympanic temperature, skin temperature and thermal sensations were higher in HOT than in CON during 10:30–11:30 (all p < 0.05). EI, food amount, hot foods, and hot beverage consumption were lower in HOT compared with CON (all p < 0.05). Subjective appetite was similar between the two trials. Linear mixed model revealed that lower EI trended to be linked to higher tympanic temperature (β = −6.060, p = 0.069). Combined torso and foot heat exposure decreased EI in young healthy adults, with tympanic temperature predicting this reduction and no changes in subjective appetite. Introduction The modern environment allows people to access and consume excess food by upregulating their appetite through abundant food cues [1–3]. Excessive energy from overeating, rather than alterations in energy expenditure (EE) alone, can gradually induce overweight or obesity [4,5]. It is well known that obesity remains a major global issue and is associated with several adverse health outcomes, including cardiovascular diseases, diabetes, and various metabolic disorders [6]. Numerous strategies have been developed to combat obesity [7–9]. A typical method of combating obesity is to decrease energy intake (EI). Therefore, strategies for decreasing the EI require further investigation. Emerging studies have shown that EI is modulated by ambient temperature, with reduction in hot environments and increases during cold exposure, either at rest or after exercise [10–18]. These findings suggest that exposure to different temperatures can alter subsequent EI. Shorten et al. [12] first reported a decreased relative EI (defined as total EI minus EE) after exercising at 36°C. Thereafter, lower EI was found after resting in a 30°C chamber for 5.5 h [10], or 7-h office work in a 26–27°C room, compared to the thermoneutral environments [15]. However, other studies have reported contradictory results, showing similar EI after exercise [18–20] or during rest [21–24] in heat compared to thermoneutral conditions. The heterogeneity of the methods may lead to discrepancies in EI. Specifically, the differences in these studies with respect to temperature setting, buffet meal design, exposure duration, and heating methods may have caused these inconsistent results, as highlighted by a recent systematic review and meta-analysis [25]. There has been little discussion on the relationship between body temperature and EI under different thermal conditions. Westerterp-Plantenga et al. [21] reported an association between rectal temperature and EI during passive cold exposure. Indeed, our previous study found that EI increased after wearing a cooling vest following high-intensity exercise or rest, and it was associated with lower rectal temperature in young males [26]. Together, these findings suggest that variations in core body temperature may contribute to the change in EI after exposure to different ambient temperatures. Furthermore, data examining the relationship regarding EI and body temperature following passive heat exposure remain limited. Here, we examined the acute effects of combined torso and foot heat exposure, induced by a hot vest and a warm-water foot bath, on subsequent appetite and EI in healthy young adults. We hypothesized that this heat exposure would suppress appetite and decrease EI in young adults by increasing tympanic temperature (Ttym) at rest. Methods Study design This randomized crossover study included two trials (control trial (CON) and hot trial (HOT)) conducted in a randomized order, separated by at least 28 d, to minimize potential carry-over effects. For female participants, the trials were conducted during the first week of the follicular phase, given the effects of sex hormones on EI during different menstrual phases [27]. Figure 1 shows a schematic illustration of the study protocol. Figure 1.The schematic representation of the study protocol.The schematic representation of the study protocol shows two trials: a hot trial (HOT) and a control trial (CON). In CON, participants remained seated from 0930 to 1130. In HOT, participants remained seated from 0930 to 1030, followed by wearing a hot vest (~45°C) and taking a foot bath (~43°C) from 1030 to 1130. From 1140, participants were allowed up to 1 hour to consume an ad libitum buffet meal, from which energy intake (EI) was calculated. Subjective feelings are recorded at multiple intervals. Tympanic and skin temperatures are measured continuously, while body mass are assessed at specific points. The protocol illustrates the sequence and timing of these activities and measurements.Schematic of study protocol with HOT and CON trials, showing activities and measurements over time. The schematic representation of the study protocol. The Ethics Committee on Human Research of Waseda University (approval number: 2023–368) reviewed and approved the study protocol, which complied with the Declaration of Helsinki. Prior to providing consent, participants were informed a brief explanation of the general research background and experimental procedures but were not informed of the specific hypotheses. All participants provided written informed consent before data collection. Participants The approximate sample size was estimated based on the EI difference (153 kcal) from our previous study [26]. Assuming a significance level of 0.05 and a statistical power of 80% for a within-subject crossover design, 16 participants were required. Sixteen young adults were recruited through advertisement flyers from Waseda University. The inclusion criteria were as follows: 1) aged 18–39 years; 2) body mass index (BMI) of 18.5–24.9 kg/m2; 3) nonsmokers; 4) no eating disorders; 5) no plans to gain or lose weight within 3 months prior to and during the study; 6) no food allergies; 7) for females, regular menstrual cycles (a consistent cycle length of 25–38 d). Screening and anthropometric assessment All the participants completed a screening form detailing medical history. Habitual physical activity was assessed using the International Physical Activity Questionnaire-Short Form. All food items used in the ad libitum buffet meal were checked and familiarized with all participants with their acceptance. Participants were familiar with the ad libitum buffet-style meal format. Height was measured using a stadiometer (213; Seca Corporation, Japan). Body mass and fat composition were measured using a body composition analyzer (MC-780 MA-N, Tanita Corporation, Japan). Waist circumference was measured using a flexible plastic tape at the umbilicus level. Standardization before the main trial On the day before the first trial, participants were instructed to record their dietary intake before dinner and repeat a similar diet pattern before the second trial. For dinner, the participants consumed a bento box (Plump Rice and Demi-Glace Hamburger; Nissui Corporation, Japan) from 19:30 to 20:30, followed by an overnight fast. Water intake was standardized: 1500 mL between 09:30 and 19:30, and 500 mL between 19:30 and 09:30 on the trial day. Planned exercise and alcohol were restricted on the day before each trial. Instructions were sent by e-mail to the participants and compliance was confirmed verbally before each trial with this standardization immediately before each trial. Main trials In each trial, the participants reported to the laboratory at approximately 08:45 after an overnight fast without breakfast. Each participant was instructed to change into a long-sleeved cotton T-shirt and sports shorts. Skin temperature (Tsk) probes were attached to the skin at predefined sites. The room temperature and relative humidity were targeted around 24°C and 45%. In CON, the participants remained seated from 09:30 until 11:30 without intervention. In HOT, the participants remained seated from 09:30 until 11:30. From 10:30, the participants were instructed to wear a hot vest and undergo a foot bath until 11:30. At 11:30, the hot vest and foot bath were removed, sweat was immediately wiped off, and the top was changed to a long-sleeved cotton T-shirt. Body mass was assessed. At 11:40, the participants consumed the ad libitum buffet meal until they felt “comfortably full and satisfied,” with a 1-hour time limit. After the buffet meal, the participants rested untill 12:40. Thereafter, body mass was assessed. Hot vest and foot bath setting In HOT, a hot vest and a warm-water foot bath were used as practical heat-exposure methods, offering advantages in terms of portability, low cost, and ease of implementation compared with traditional environmental chambers or saunas. This hot vest (heat vest-001, Excitech Corporation, Japan) was used to heat the upper body and neck. It is a non-sleeved garment with a warming effect. It contains 17 heating zones covering the abdomen, back, shoulders, and neck and is powered by a portable charger for heat generation. The vest temperature was ~45°C. An electric foot bath with a temperature-control function (B-HA23001, GOQRAQ Patent, Japan) was used to heat the lower legs and feet at ~43°C. The water level was set ~2 cm below the Tsk probe placed at the left calf. Before the main experiment, a pilot test was conducted to evaluate the effects of the hot vest and the warm foot bath used separately on Ttym. The results indicated that neither device alone produced a substantial increase in Ttym. Thus, these two interventions were combined in the main trial to achieve sufficient thermal load. Body temperatures and thermal sensation The Ttym was continuously measured using an ear canal thermometer (CE thermo, NIPRO Corporation, Japan) every 5 min throughout each trial. Ttym was selected as a noninvasive alternative of core body temperature to minimize participant discomfort and to avoid potential interference with ad libitum EI during the prolonged resting protocol. The Ttym probe was gently inserted approximately 0.5 cm into the external auditory canal and positioned to face the tympanic membrane. The device estimates Ttym based on infrared radiation reflected from the tympanic membrane and surrounding tissues. The Tsk was continuously assessed using button-type loggers (DS1921G-F5, Thermochron iButton, Analog Devices Corporation, USA) every 5 min throughout each trial. Data were subsequently retrieved using the associated software (OneWireViewer; Analog Devices Corporation, USA). Ten loggers were strategically positioned on each participant’s forehead, neck, left chest (Tchest), left upper arm (Tupper arm), left calf (Tcalf), left thigh (Tthigh), medial abdomen, left scapula, and lower medial back. Mean Tsk was calculated through Ramanathan’s formula: mean Tsk = 0.3(Tchest +Tupper arm) + 0.2(Tcalf +Tthigh) [28]. Additional skin temperature sensors were placed at multiple body sites to enable exploratory analyses of regional skin temperatures in relation to EI. Subjective thermal sensation was evaluated at the same intervals as subjective appetite using a 17-point thermal sensation scale [29], with scores ranging from 0 (unbearably cold) to 8 (unbearably hot) in 0.5-unit increments. Ad libitum buffet meal design and energy intake The ad libitum buffet meal contained bananas, apples, fruit jelly, cakes, milk chocolate, chips, oolong tea, orange juice, milk, Japanese dressing, mayonnaise, cabbage salads, chicken salads, smoked salmon, rice balls, fried cabbage, fried salmon, fried chicken, and rice (Supplemental Table S1). Rice, fried chicken, fried salmon, and fried cabbage were classified as hot foods and reheated using the same microwave oven protocols at proper temperatures before being served to the participants. Rice balls, smoked salmon, chicken salads, cabbage salads, mayonnaise, and Japanese dressings were classified as cold foods. Hot beverages included hot milk and oolong tea, which were heated and stored in thermal jugs. Cold beverages included cold milk, cold oolong tea, and orange juice. The temperatures of foods are listed in Supplemental Table S1. The buffet was designed to include both hot and cold foods as well as beverages to allow participants to freely select foods across a range of temperatures and textures. This design was consistent with our previous work and with a previous study examining the effects of food temperature on food choice and EI [22,26]. In each trial, all food items were provided in excess quantities. Participants were directed to consume the buffet meal until they felt “comfortably full and satisfied” within 60 min, with the option to take additional servings if desired. Total food intake (g) was determined by subtracting the remaining food (including any leftovers) from the amount initially served using a kitchen scale (Tanita, Japan). Energy and macronutrient intake (carbohydrates, proteins, and fats) were calculated based on The Standard Tables of Food Composition in Japan 2020 8th Revised Editio) using computer software (Excel Eiyoukun Ver 9.0, Kenpakusha, Japan) and/or macronutrient labels provided for food packaging. Water intake was prohibited during the experimental sessions prior to the meal because expired gas measurements were continuously performed using a face mask. Subjective feelings of appetite Subjective appetite was measured in the fasting state (at 09:30), at 10:00, 10:30 (immediately after setting the hot vest and foot bath), 11:00, 11:30 (immediately after removal), immediately after the meal, and at 12:40, using a visual analogue scale. This scale included measures of hunger, fullness, satiety, prospective food consumption, and desire to eat [30]. Each was scored on a 100 mm line anchored with “not at all” (0 mm) and “extremely” (100 mm). Resting energy expenditure (REE) and respiratory quotient (RQ) In each trial, expired air samples were collected from 09:30 to 11:30 using a breath-by-breath gases analyzer (AE310s, MINATO Medical Science Corporation, Japan) to assess REE and substrate utilization in relation to subsequent EI. REE was evaluated using Weir’s equation [31]. RQ was calculated as VCO2/VO2. Statistical analysis Data were analyzed using IBM SPSS Statistics for Windows, version 29.0 (IBM Corporation, USA). Normal distribution was checked using the Shapiro–Wilk test. Area under the curve (AUC) for the appetite perception over the 0–2 h period (09:30–11:30), and for body temperature and thermal sensation over the 0–1 h period (10:30–11:30), were calculated using the trapezoidal method. No baseline subtraction was applied. Paired t-tests were used to examine the differences between trials in energy and nutrient intake, food intake, AUCs of appetite perceptions and body mass, REE, RQ, room temperature, and relative humidity. Results are shown as mean ± standard deviation, with 95% confidence intervals. A linear mixed model was used to detect differences in Ttym and mean Tsk and thermal sensations over time, assessing trial and time effects. In addition, EI was analyzed using a linear mixed model with trial as a within-subject factor, and trial order as a between-subject factor, to test the trial order effect. Absolute standardized effect sizes were estimated, and effect sizes were interpreted as small (0.2), moderate (0.5), or large (0.8). Linear mixed models were used to examine the effects of heat exposure on EI while accounting for repeated measurements within participants. Participant was included as a random intercept. Trial (CON vs. HOT) was specified as a fixed effect in all models. To explore potential factors modulating EI, additional models included body mass, body mass change (09:30–11:30), BMI, AUCs of Ttym, Tsk, forehead temperature, neck temperature, Tchest, Tupper arm, Tcalf, Tthigh, medial abdomen temperature, left scapula temperature, lower medial back temperature and thermal sensations (10:30–11:30) as covariates in separate models, as well as their interactions with Trial. In addition, an exploratory multiple regression analysis, using a stepwise method, was performed to analyze the relationship among EI and the above factors. Statistical significance was set at p < 0.05. Results Participants Sixteen healthy, recreationally active young adults (50% of females) were recruited. The physical characteristics are presented in Table 1.Table 1.Participant characteristics.CharacteristicTotal n = 16Female n = 8Male n = 8Age (years)24 ± 324 ± 324 ± 2Height (cm)167 ± 8161 ± 5173 ± 5Body mass (kg)59.9 ± 9.053.8 ± 5.965.9 ± 7.5Body mass index (kg/m2)21.5 ± 2.020.8 ± 1.922.2 ± 2.0Body fat (%)22 ± 829 ± 515 ± 4Lean body mass (kg)46.6 ± 9.038.8 ± 3.254.4 ± 5.2Waist circumference (cm)74.1 ± 6.671.7 ± 4.676.4 ± 7.7Walking (MET-min/week)a925 ± 1383986 ± 4481950 ± 1373Moderate physical activities (MET-min/week)a353 ± 590435 ± 619270 ± 504Vigorous physical activities (MET-min/week)a899 ± 1346120 ± 3391308 ± 1765Values are mean ± standard deviation for 8 young females and 8 young males.aData were assessed using the International Physical Activity Questionnaire (short form).The relatively large standard deviation in walking METs among males was influenced by one participant reporting a high volume of walking activity.MET, metabolic equivalent of task. Table 1.Participant characteristics.CharacteristicTotal n = 16Female n = 8Male n = 8Age (years)24 ± 324 ± 324 ± 2Height (cm)167 ± 8161 ± 5173 ± 5Body mass (kg)59.9 ± 9.053.8 ± 5.965.9 ± 7.5Body mass index (kg/m2)21.5 ± 2.020.8 ± 1.922.2 ± 2.0Body fat (%)22 ± 829 ± 515 ± 4Lean body mass (kg)46.6 ± 9.038.8 ± 3.254.4 ± 5.2Waist circumference (cm)74.1 ± 6.671.7 ± 4.676.4 ± 7.7Walking (MET-min/week)a925 ± 1383986 ± 4481950 ± 1373Moderate physical activities (MET-min/week)a353 ± 590435 ± 619270 ± 504Vigorous physical activities (MET-min/week)a899 ± 1346120 ± 3391308 ± 1765Values are mean ± standard deviation for 8 young females and 8 young males.aData were assessed using the International Physical Activity Questionnaire (short form).The relatively large standard deviation in walking METs among males was influenced by one participant reporting a high volume of walking activity.MET, metabolic equivalent of task. Participant characteristics. Values are mean ± standard deviation for 8 young females and 8 young males. aData were assessed using the International Physical Activity Questionnaire (short form). The relatively large standard deviation in walking METs among males was influenced by one participant reporting a high volume of walking activity. MET, metabolic equivalent of task. Room temperature and relative humidity Ambient temperatures (CON: 24.8 ± 0.6°C and HOT: 24.8 ± 0.7°C, p = 0.918) and relative humidity (CON: 48 ± 9% and HOT: 46 ± 9%, p = 0.671) were similar between trials. Body temperature and thermal sensations The body temperatures and thermal sensations across the trial are presented in Figure 2. Ttym became significantly higher from 10:50 (20 minutes after initiation of heat exposure) until 11:45 (5 minutes after initiation of the buffet meal) in HOT than CON (main effect of trial: p < 0.001; main effect of time: p < 0.001; trial × time interaction: p < 0.001). Mean Tsk became significantly higher from 10:35 until 11:40 in HOT than CON (all p < 0.001 for trial, time, and interaction effects). Thermal sensations were also significantly higher from 10:30 until immediately after the meal in HOT than CON (all p < 0.001 for trial, time, and interaction effects). Figure 2.Time-course changes in (a) tympanic temperature, (b) mean skin temperature, and (c) thermal sensations during the control (CON) and hot (hot) trials. A linear mixed model revealed significant main effects of trial, time, and a trial × time interaction for all measures (all p < 0.001). Data are presented as mean ± standard deviation. * post-hoc analyses indicated significant differences between trials at specific time points.Image A displays tympanic temperature (°C) over time (09:30-12:40). The y-axis ranges from 35.5 to 37.5°C. Two curves show control (CON) and hot (HOT) trials, with HOT showing a marked temperature rise during the heat phase, indicated by an asterisk. Tympanic temperature became significantly higher in HOT than CON from 10:50 (20 minutes after initiation of heat exposure) until 11:45 (5 minutes after initiation of the buffet meal). Image B illustrates mean skin temperature (°C) over the same time frame, with the y-axis spanning 32.0 to 37.0°C. Again, the HOT trial exhibits a significant temperature increase during the heat phase, marked by an asterisk. Mean skin temperature became significantly higher in HOT than CON from 10:35 until 11:40. Image C presents thermal sensation over time, with the y-axis ranging from 0.0 to 8.0. The HOT trial shows a notable rise in thermal sensations during the heat phase until after the meal, marked by an asterisk, compared to the CON trial.Graphs of tympanic, skin temperatures and thermal sensations during control and hot trials. Time-course changes in (a) tympanic temperature, (b) mean skin temperature, and (c) thermal sensations during the control (CON) and hot (hot) trials. A linear mixed model revealed significant main effects of trial, time, and a trial × time interaction for all measures (all p < 0.001). Data are presented as mean ± standard deviation. * post-hoc analyses indicated significant differences between trials at specific time points. Energy and nutrient intake, and body mass Energy and nutrient intakes are presented in Table 2, and individual profiles of EI and hot food consumption are presented in Figure 3. EI was decreased in HOT than CON (p = 0.027), with no effect of trial order (p = 0.356). Absolute and relative fat intakes were lower in HOT than CON (all p < 0.05, Table 2). Furthermore, food amount, hot food consumption, and hot beverage consumption were also lower in HOT than CON (all p < 0.05, Table 2). Figure 3.(a) Total ad libitum energy intake (EI) and (b) hot food consumption in the control (CON) and the hot (hot) trials. Bars represent the standard deviation of the mean, which is represented by error bars. Data points and lines represent individual responses. In individuals, 12 participants showed lower EI, and 12 participant reported decreased hot food consumption after the heat exposure. * the heat exposure significantly reduced both total EI (p = 0.027) and hot food intake (p = 0.011) compared to the CON trial.Image A showing a bar graph labeled 'EI (kcal)' with the x-axis labeled 'CON' and 'HOT' and the y-axis labeled '0 to 2000'. The graph displays individual data points connected by lines, indicating energy intake in control and hot trials. The mean energy intake is lower in the hot trial, marked by an asterisk. Image B showing a bar graph labeled 'Hot foods (g)' with the x-axis labeled 'CON' and 'HOT' and the y-axis labeled '0 to 800'. This graph also shows individual data points connected by lines, representing hot food consumption in control and hot trials. The mean hot food consumption is lower in the hot trial, marked by an asterisk.Two bar graphs showing energy intake and hot food consumption in control and hot trials.Table 2.Energy and macronutrient intake with and without a hot vest and foot bath measured from an ad libitum buffet meal. Total n = 16ComparisonControlHotMean difference(95% CI)Effect sizeEI (kcal)1025 ± 343849 ± 317−176 (−329, −23)*−0.615Carbohydrate (kcal)554 ± 223470 ± 185−85 (−171, 2)−0.521Carbohydrate (%)53 ± 955 ± 102 (−2, 7)0.277Fat (kcal)320 ± 128241 ± 142−79 (−153, −4)*−0.565Fat (%)31 ± 727 ± 9−4 (−8, 0)*−0.557Protein (kcal)164 ± 48149 ± 51−16 (−33, 2)−0.471Protein (%)17 ± 518 ± 52 (−1, 4)0.349Total amount (g)1013 ± 345888 ± 320−125 (−278, 28)−0.436Food amount (g)638 ± 232566 ± 172−72 (−143, −1)*−0.540Beverage amount (g)375 ± 245322 ± 161−53 (−183, 77)−0.219Energy density(kcal/g)1.0 ± 0.21.0 ± 0.2−0.1 (−0.1, 0)−0.346Hot foods (g)a316 ± 157236 ± 174−81 (−148, −14)*−0.640Hot beverages (g)b179 ± 17747 ± 90−133 (−202, −63)*−1.015Cold foods (g)c159 ± 125160 ± 901 (−59, 61)0.008Cold beverages (g)d196 ± 276275 ± 14180 (−76, 235)0.273Values are presented as the mean ± standard deviation. Data were analyzed using a paired t-test. Significant differences are presented as mean differences (95% CI) with asterisks *. Absolute standardized effect sizes were estimated, and thresholds of 0.2, 0.5, and 0.8 describe small, moderate, and large effects, respectively.aHot foods include rice, fried chicken, fried salmon, and fried cabbage.bHot beverages included hot milk and hot oolong tea.cCold foods included rice balls, salad chicken, smoked salmon, salad cabbage, mayonnaise, and Japanese dressing.dCold beverages included cold milk, orange juice, and cold oolong tea.EI, energy intake; CI, confidence interval.Table 3.AUC for appetite perception over 0–2 h with and without the hot vest and foot bath before the ad libitum buffet meal. n = 16ComparisonControlHotMean difference (95% CI)Hunger (mm·h)139 ± 36133 ± 29−8 (−27, 11)Satiety (mm·h)55 ± 2854 ± 27−1 (−18, 17)Fullness (mm·h)42 ± 3053 ± 3011 (−10, 32)PFC (mm·h)136 ± 29137 ± 32−4 (−21, 14)Desire to eat (mm·h)129 ± 31131 ± 35−1 (−18, 17)Values are presented as the mean ± standard deviation. AUC (mm·h), where mm refers to the VAS score (0–100 mm). Data were analyzed using a paired t-test. Significant differences are presented as mean differences (95% CI) with asterisks *.AUC, area under the curve; PFC, prospective food consumption; CI, confidence interval. (a) Total ad libitum energy intake (EI) and (b) hot food consumption in the control (CON) and the hot (hot) trials. Bars represent the standard deviation of the mean, which is represented by error bars. Data points and lines represent individual responses. In individuals, 12 participants showed lower EI, and 12 participant reported decreased hot food consumption after the heat exposure. * the heat exposure significantly reduced both total EI (p = 0.027) and hot food intake (p = 0.011) compared to the CON trial. Table 2.Energy and macronutrient intake with and without a hot vest and foot bath measured from an ad libitum buffet meal. Total n = 16ComparisonControlHotMean difference(95% CI)Effect sizeEI (kcal)1025 ± 343849 ± 317−176 (−329, −23)*−0.615Carbohydrate (kcal)554 ± 223470 ± 185−85 (−171, 2)−0.521Carbohydrate (%)53 ± 955 ± 102 (−2, 7)0.277Fat (kcal)320 ± 128241 ± 142−79 (−153, −4)*−0.565Fat (%)31 ± 727 ± 9−4 (−8, 0)*−0.557Protein (kcal)164 ± 48149 ± 51−16 (−33, 2)−0.471Protein (%)17 ± 518 ± 52 (−1, 4)0.349Total amount (g)1013 ± 345888 ± 320−125 (−278, 28)−0.436Food amount (g)638 ± 232566 ± 172−72 (−143, −1)*−0.540Beverage amount (g)375 ± 245322 ± 161−53 (−183, 77)−0.219Energy density(kcal/g)1.0 ± 0.21.0 ± 0.2−0.1 (−0.1, 0)−0.346Hot foods (g)a316 ± 157236 ± 174−81 (−148, −14)*−0.640Hot beverages (g)b179 ± 17747 ± 90−133 (−202, −63)*−1.015Cold foods (g)c159 ± 125160 ± 901 (−59, 61)0.008Cold beverages (g)d196 ± 276275 ± 14180 (−76, 235)0.273Values are presented as the mean ± standard deviation. Data were analyzed using a paired t-test. Significant differences are presented as mean differences (95% CI) with asterisks *. Absolute standardized effect sizes were estimated, and thresholds of 0.2, 0.5, and 0.8 describe small, moderate, and large effects, respectively.aHot foods include rice, fried chicken, fried salmon, and fried cabbage.bHot beverages included hot milk and hot oolong tea.cCold foods included rice balls, salad chicken, smoked salmon, salad cabbage, mayonnaise, and Japanese dressing.dCold beverages included cold milk, orange juice, and cold oolong tea.EI, energy intake; CI, confidence interval. Energy and macronutrient intake with and without a hot vest and foot bath measured from an ad libitum buffet meal. Values are presented as the mean ± standard deviation. Data were analyzed using a paired t-test. Significant differences are presented as mean differences (95% CI) with asterisks *. Absolute standardized effect sizes were estimated, and thresholds of 0.2, 0.5, and 0.8 describe small, moderate, and large effects, respectively. aHot foods include rice, fried chicken, fried salmon, and fried cabbage. bHot beverages included hot milk and hot oolong tea. cCold foods included rice balls, salad chicken, smoked salmon, salad cabbage, mayonnaise, and Japanese dressing. dCold beverages included cold milk, orange juice, and cold oolong tea. EI, energy intake; CI, confidence interval. Table 3.AUC for appetite perception over 0–2 h with and without the hot vest and foot bath before the ad libitum buffet meal. n = 16ComparisonControlHotMean difference (95% CI)Hunger (mm·h)139 ± 36133 ± 29−8 (−27, 11)Satiety (mm·h)55 ± 2854 ± 27−1 (−18, 17)Fullness (mm·h)42 ± 3053 ± 3011 (−10, 32)PFC (mm·h)136 ± 29137 ± 32−4 (−21, 14)Desire to eat (mm·h)129 ± 31131 ± 35−1 (−18, 17)Values are presented as the mean ± standard deviation. AUC (mm·h), where mm refers to the VAS score (0–100 mm). Data were analyzed using a paired t-test. Significant differences are presented as mean differences (95% CI) with asterisks *.AUC, area under the curve; PFC, prospective food consumption; CI, confidence interval. AUC for appetite perception over 0–2 h with and without the hot vest and foot bath before the ad libitum buffet meal. Values are presented as the mean ± standard deviation. AUC (mm·h), where mm refers to the VAS score (0–100 mm). Data were analyzed using a paired t-test. Significant differences are presented as mean differences (95% CI) with asterisks *. AUC, area under the curve; PFC, prospective food consumption; CI, confidence interval. For each food items, fried chicken, fried salmon, and hot oolong tea were consumed less, whereas cold oolong tea was consumed more in HOT than CON (all p < 0.05; Supplemental Table S2). Baseline body mass and the changes in body mass before and after the buffet meal were similar between HOT and CON (0.8 ± 0.3 kg vs. 0.8 ± 0.3 kg, respectively, p = 0.887). Subjective feelings of appetite AUCs of subject appetite before the buffet meal (0–2 hour) were listed in Table 3. Subjective feelings of hunger, satiety, fullness, prospective food consumption, and desire to eat were similar between HOT and CON (all p > 0.05). REE and RQ REE and RQ were listed in Supplemental Table S3. REE and RQ were similar from 09:30 to 11:30 in both trials (all p > 0.05). Modeling of energy intake Linear mixed model revealed that Ttym tended to be associated with EI (β = −6.060, standard error = 3.813, p = 0.069), indicating that participants exhibiting greater increases in Ttym tended to consume less energy. Sex was not significantly associated with EI (p = 0.073), although males tended to consume more energy than females. Stepwise regression modeling detected that a higher total AUC of Ttym (β = −5.681, standard error with β = 2.348, t = −2.420, p = 0.022, VIF = 1.000) was associated with a lower EI (directional coefficient = 13338.751; R = 0.404, R2 = 0.163, adjusted R = 0.135; F = 5.856, p = 0.022). Discussion The primary finding of this study was that combined torso and foot heat exposure for 1 h decreased subsequent EI in young adults. Furthermore, absolute and relative fat intake, food intake, hot foods, and hot beverages were also decreased. Notably, this reduction in EI occurred without concomitant changes in subjective appetite ratings before the meal. Additionally, higher Ttym tended to be associated with a lower EI. To our knowledge, this is the first application of using a hot vest and foot bath to induce heat exposure and examine its effects on EI in young females and males. Our previous study showed that cold exposure induced by a cooling vest for 1 h increased EI after exercise or at rest [26]. Consistent with other studies [10,12,15], this study further examined the efficacy and application of hot vests and foot baths, wearable and portable devices, in reducing EI among young adults. However, some earlier studies did not detect a decrease in EI after heat exposure during rest [20,22,24]. A previous study reported similar EI after a sauna or rest among obese and non-obese males [24]. It seems that the hot exposure conducted in the sauna study was not sufficient to downregulate EI because this study set up three 20-min sauna sessions (80°C) with interval rest sessions (22°C) to achieve sweat loss rather than increasing body temperature [24]. Another study also reported similar EI among 24-h hot exposure (32°C), cold exposure (16°C), and control (24°C) in young healthy males, despite significant differences in core temperature between cold (16°C) and both temperate and hot conditions, while core temperature did not differ significantly between the 24°C and 32°C conditions [22]. Our previous study explored the inverse relationship between rectal temperature and EI [26]. The present study found that higher Ttym tended to be associated with a lower EI. Therefore, it is likely that the variation in core body temperature may represent one of several factors modulating the subsequent EI. Instead, modulation of food intake itself may represent an adaptive response to thermal stress, potentially involving diet-induced thermogenesis (DIT). DIT, the postprandial increase in EE and heat production following food ingestion, contributes approximately 5–15% of daily EE and varies with meal composition and nutrient processing [32,33]. Core body temperature also modestly increases with feeding and may be related to diet-induced thermogenic responses [34]. Interestingly, it has been reported that overeating attenuated the decreased in rectal temperature under mild cold exposure and DIT was increased [35]. Our data has shown that the Ttym were higher in HOT than CON until 5 minutes until meal initiation, where the Ttym gradually became higher in CON. The results of thermal sensation also suggested the thermal load from the heat exposure. Interestingly, a previous study has shown that core temperature continued to decrease after cold-water immersion, where the subsequent EI was increased compared to the neutral conditions [17]. It could be speculated that in cold environments, increased EI and the associated diet-induced thermogenesis may support the maintenance of core body temperature, whereas in hot environments, a reduction in EI may help limit additional heat production following a meal. Although DIT was not directly assessed in the present study, this physiological adaptation may partly explain the reduced EI observed after heat exposure. Moreover, the present study has also assessed the REE and RQ during the heat exposure. Previous studies have discussed the potential for compensatory changes in EI after higher EE. However, the REE and RQ was similar between the HOT and CON conditions. Unlu et al. [16] also reported that 24-h EI was increased following cold exposure but was independent of 24-h EE. These findings suggest that the variation in EI is unlikely to be driven by acute changes in REE. Further research is warranted to investigate these potential mechanisms. The present study suggested that the heat exposure decreased absolute and relative fat intake, which contributed to the energy value of foods. These results were consistent with higher fat intake after or during cold exposure observed in previous studies [16,26]. It is possible that explicit liking and implicit wanting for high-fat foods was decreased by hot environment [22,36]. Furthermore, the reduced food intake, especially the decreased hot foods and beverages consumption, likely contributed to the decreased EI. Prior studies suggested that increased EI may result from increased consumption of hot foods after wearing a cooling vest, while cold dishes were consumed more, and the implicit desire for cold and low-fat foods increased in a hot environment [22,26]. However, it is worth noting that cold foods consumption and low-fat foods was not increased after the heat exposure in the present study, which is opposite to Coca et al. study [22]. This finding indicated that participants did not increase consumption of these items to offset the heat-induced reduction in EI. Moreover, another study observed a decrease preference for savory food at hot condition, suggesting the influence of ambient temperature on foods selection [37]. Therefore, it can be assumed that ambient temperature may influence EI through food selection behaviors, considering the macronutrients and temperatures of foods and beverages. Interestingly, intake of specific beverage types differed between trials: hot beverages (e.g. hot oolong tea) were less frequently consumed after the heat exposure, while intake of cold beverages such as orange juice may have been limited by sweetness. It may explain why total beverage intake tended to be lower during the heat trial, although this difference did not reach statistical significance. This observation may be partly influenced by beverage temperature preferences and cultural drinking habits. However, given the exploratory nature of this finding and the absence of plain water in the buffet, this result should be interpreted with caution. Although EI was decreased, subjective appetite did not differ before the meal in the present study. Indeed, most studies have reported inconsistent results between appetite and EI at different temperatures [13,18,19,21]. Similarly, a systematic review showed that subjective ratings of appetite are unlikely to predict EI [38]. Moreover, some previous studies have also examined the potential involvement of appetite-related hormones under different thermal exposures reporting small or variable changes in hormones such as ghrelin (appetite-stimulating hormone) [14,22,39] or PYY (appetite-suppressing hormone) [12]. However, our previous work suggested a lack of association between appetite-related hormones and the EI response to different external temperatures, as well as other studies [13,14,18,20,26,40]. Together, these findings suggest that temperature-related changes in EI may not be fully explained by subjective appetite or hormonal pathways. Further studies incorporating integrated physiological and behavioral assessments are required to elucidate the mechanism(s) behind these processes. The randomized crossover design, well-controlled water and dietary intake before the main trials, and inclusion of female participants enhanced the validity of our findings. Our meal design also strengthened this study, as previously reported [26]. Furthermore, compared to hot chambers, our results suggest the combined torso and foot heat exposure decrease EI over a shorter duration (1 h versus 5.5 h) [10]. However, limitations include the recruitment of healthy, normal-weight young adults, rather than overweight or obese groups who might benefit more from EI suppression. However, we recruited normal-weight participants to consider broader possible applications, such as rapid weight loss in athletes or models. Second, we did not include a thermoneutral control condition involving a thermoneutral vest and foot bath to limit participant burden and the total number of experimental sessions. However, the absence of a thermoneutral control condition limits the interpretation of heat-specific effects. Third, thermal comfort and appetite-related hormones were not evaluated, and rectal, esophageal or gastrointestinal temperatures are more accurate than Ttym in representing core temperature [41]. Nevertheless, the middle ear cavity can still be considered as a potential core temperature site for measurement [42]. Moreover, as EI was decreased based on a single buffet meal, it remains unclear whether heat exposure would influence total daily EI. The empirical findings of this study provide new insights of hot vests and foot baths for reducing EI. The reduction in EI (~176 kcal) suggested a potential strategy for preventing weight gain. However, these findings should not be extrapolated to long-term weight change or chronic interventions, which require dedicated longitudinal studies. Conclusion Overall, this study showed that EI was decreased after the combined torso and foot heat exposure for 1 h in young adults. This reduction was accompanied by reductions of intake of certain foods, but not the modifications of subjective appetite. Furthermore, higher Ttym was a predictor of reduction in EI. Compared to bariatric surgery or other medications, the use of hot vests and warm foot baths has advantages, such as low price, accessibility, transportation, and no invasion. However, further long-term studies on overweight individuals are required to know whether it could constitute a valid and cheap solution to manage body mass. Supplementary Material Li-EIandHotExposure-SupplementalTable1.docx Li-EIandHotExposure-SupplementalTable2.docx Li-EIandHotExposure-SupplementalTable3.docx