Abstract
Background
While aerobic training is well-established for improving body composition and metabolic health in normoxia, its effectiveness in hypoxia remains unclear.
Objective
This meta-analysis examines whether aerobic training in hypoxia is more effective than in normoxia for improving body composition and metabolic health in overweight and/or obese individuals, and identifies optimal exercise prescription variables.
Methods
A search of five databases was conducted through 10 November 2024. Random-effects meta-analyses evaluated body composition (e.g., body mass and fat mass) and metabolic health markers (e.g., triglycerides and glucose). Subgroup analyses were performed based on hypoxic severity, hypoxic dose, exercise duration, frequency, session length, and age.
Results
Aerobic training in hypoxia resulted in greater reductions in body mass (mean difference [MD] = -0.90, 95% confidence interval [CI]: -1.80 to -0.01), triglycerides (MD = -10.78, 95% CI: -20.68 to -0.88), low-density lipoprotein cholesterol (MD = -3.74, 95% CI: -6.92 to -0.56, p < 0.05), and insulin resistance (MD = -0.22, 95% CI: -0.33 to -0.11) (all p < 0.05), with a trend towards larger fat mass loss (MD = -1.22, 95% CI: -2.59 to 0.15, p = 0.08). These benefits were more prominent in moderate hypoxia (inspired oxygen fraction [FiO2] ≥ 15%), with hypoxic dose ≥ 55 km·h, in individuals < 40 years, and with protocols involving ≥ 4 days/week, ≥ 60-min sessions, and < 8 weeks of training.
Conclusion
Aerobic training in hypoxia is more effective than in normoxia for reducing body mass, fat mass, triglycerides, low-density lipoprotein cholesterol, and insulin resistance in overweight and/or obese individuals. These findings could help inform obesity management strategies using hypoxic training.
Supplementary Information
The online version contains supplementary material available at 10.1186/s40798-025-00918-6.
Keywords: Aerobic training, Hypoxia, Body composition, Metabolic health, Obesity
Key Points
• Aerobic exercise training in hypoxia led to larger reductions in body mass, fat body mass, triglycerides, and low-density lipoprotein cholesterol in overweight and/or obese individuals.
• Moderate hypoxia (FiO2 ≥ 15%) with hypoxic dose ≥ 55 km·h, sessions lasting ≥ 60 min, ≥ 4 days/week, and < 8 weeks of training were effective in managing obesity, particularly in individuals aged < 40 years.
Introduction
Aerobic exercise, defined as the rhythmic movement of the body’s large muscles for at least 10 min [1], is a well-established strategy for obesity management [2], improving metabolism and supporting healthy body composition [3, 4]. The World Health Organization recommends moderate-to-high intensity aerobic exercise as an effective intervention for weight loss in individuals with obesity [2, 5]. However, higher intensity exercises may be less enjoyable and harder do due to the greater body mass of overweight and obese individuals [6]. Additionally, such exercises can increase the risk of musculoskeletal injuries, potentially undermining weight loss efforts [7–9]. Consequently, it is imperative to develop safe, well-tolerated, and effective aerobic exercise strategies tailored to the specific needs of this population.
In recent years, exercise training in low oxygen conditions (hypoxia) has emerged as a promising intervention for obesity management [10–12]. For instance, studies have compared aerobic exercise training (50–60 min of running, 3 days/week for 4 weeks at 65% of maximal oxygen uptake [V̇O2max] or 60% of maximum heart rate [HRmax]) in normoxia and hypoxia (inspired oxygen fraction [FiO2] = 15%) [13, 14]. These studies found that hypoxic training leads to greater improvements in body composition (e.g., body mass index [BMI], fat body mass [FBM], and waist circumference) and metabolic markers (e.g., blood glucose, insulin, total cholesterol, and triglycerides [TG]) at lower absolute exercise intensities [13, 14]. These benefits are thought to result from hypoxia’s potential to boost glucose uptake and transport [15], fat oxidation [16], and appetite regulation through the activation of hypoxia-inducible factors (HIF) [17], ultimately improving health markers [16].
Several meta-analyses have compared the effects of exercise training on body composition and metabolism between hypoxia and normoxia [18–21], but the results remain inconsistent. Some meta-analyses reported larger reductions in TG [21], FBM and BMI [20] following hypoxic compared to normoxic training, while others found no significant differences between conditions [18, 19]. However, these meta-analyses present several limitations: (a) the inclusion of other exercise modalities, such as resistance training and high-intensity interval training [18–21], which are less effective for fat loss in obese populations compared to aerobic exercise [3, 4, 22], potentially reducing the effects of hypoxic interventions; (b) the lack of subgroup analyses for accounting for factors like hypoxic dose (e.g., severity), exercise intervention (e.g., duration, frequency), and population characteristics (e.g., age), leading to substantial heterogeneity (I² >50%) [19–21]. Since our previous meta-analysis on the effects of hypoxic exposure on fat loss and cardiometabolic markers [21], several new randomized controlled trials have been published [23–25], highlighting the need for an updated meta-analysis that incorporates more comprehensive data.
The purpose of this meta-analysis was to compare the effects of aerobic exercise training in hypoxia and normoxia on body composition and metabolic health in overweight and/or obese individuals. We hypothesized that aerobic training in hypoxia would be more effective in improving body composition, such as BM and FBM, as well as metabolic health markers, including TG and low-density lipoprotein cholesterol (LDL-C). Additionally, we examined potential moderating factors (e.g., hypoxia severity, training duration, and frequency) to inform practical recommendations to optimize hypoxic training interventions.
Methods
This meta-analysis was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines [26] and registered with the International Prospective Register of Systematic Reviews (PROSPERO) database (CRD42024577543).
Literature Search
We searched five databases (PubMed, Web of Science, EMBASE, Scopus, and CNKI [China National Knowledge Infrastructure]) from their inception to 10 November 2024. The search used specific terms and Boolean operators (‘AND,’ ‘OR,’ ‘NOT’): “hypoxia”, “altitude”, “hypoxic”, “exercise”, “aerobic”, “body composition”, “FBM”, “lean mass”, “muscle mass”, “fat-free mass”, “weight loss”, “fat loss”, “metabolism”, “metabolic” “overweight”, “obesity”, “obese”. Furthermore, we conducted manual searches using Google Scholar to ensure comprehensive coverage of relevant literature.
Selection Criteria
Each study included was required to meet the following the criteria based on the participants, interventions, comparators, outcomes, and study design (PICOS) framework: (1) participants were overweight (body mass index [BMI] ≥ 25 kg/m2) and/or obese (BMI ≥ 30 kg/m2), without physical activity restrictions or severe diseases that could impede exercise interventions; (2) the intervention involved aerobic exercise training in hypoxia (simulated altitude ≥ 2000 m or FiO2 ≤ 16.5%) at least 2 days/week with moderate-to-high intensity (≥ 45% V̇O2max or ≥ 50% heart rate reserve [HHR] or ≥ 60% HRmax) [21, 27]; (3) An equivalent aerobic exercise training protocol (relative intensity, training duration, frequency, and session length identical to the hypoxic group) conducted in normoxia was used as the comparator; (4) outcomes related to body composition (body mass [BM], BMI, lean body mass [LBM], FBM, waist circumference [WC], hip circumference [HC], waist-to-hip ratio [WHR]) and metabolic health (total cholesterol [TC], TG, LDL-C, high-density lipoprotein cholesterol [HDL-C], glucose, insulin, and homeostasis model assessment index of insulin resistance [HOMA-IR]) were reported; (5) the study design utilized a randomized controlled trial.
Studies were excluded based on the following criteria: (1) participants had severe diseases that could hinder exercise interventions (e.g., cardiovascular diseases or type 2 diabetes); (2) the intervention combined aerobic exercise with other types of training (e.g., interval training, resistance training, or whole-body vibration); (3) blood flow restriction was used as the hypoxic intervention; (4) full-text articles or data extraction were unavailable; (5) the study was a conference abstract, review, case report, or animal experiment; (6) the study was not published in English or Chinese.
Study Selection
The retrieved literature was imported into EndNote (Version X9, Clarivate Analytics, Philadelphia, USA) reference management software, and duplicates were removed. During the initial screening, two independent investigators (H.J. and D.L.) reviewed the titles and abstracts of all studies to identify potentially relevant ones. Full-texts of the relevant studies were then obtained and assessed for compliance with the inclusion and exclusion criteria. Any disagreements between the two investigators were resolved through consultation with a third investigator (C.Y.), who helped finalize the list of included studies.
Data Extraction
Two independent investigators (H.J. and D.L.) extracted relevant data from the eligible studies, including: (1) basic information (author [s], year of publication and title); (2) participant characteristics (sample size, age, sex, and BMI); (3) hypoxic exposure (hypoxia severity, duration, and pattern); (4) exercise protocol (duration, frequency, and session length); (5) outcome indicators (mean value, standard deviation [SD], and sample size). If studies did not report the mean values and SD of the differences before and after the intervention, these were calculated using the following formula [28]:
 |
 |
where Meanbefore and Meanafter represent the means of the pre-intervention and post-intervention, respectively. SDbefore and SDafter represent the SDs of the pre-intervention and post-intervention, respectively. Corr is the correlation coefficient between the hypoxic and normoxia groups, often assumed to be 0.5 when no correlation is reported [28].
Assessment of Methodological Quality
The methodological quality and risk of bias of all included studies were assessed following the Cochrane guidelines [29]. The Cochrane risk of bias assessment tool examines studies across several domains: (1) random sequence generation; (2) allocation concealment; (3) blinding of participants and personnel; (4) blinding of outcome assessment; (5) incomplete outcome data; (6) selective reporting; and (7) other potential sources of bias. Each domain was rated as ‘low risk’ (‘+’), ‘high risk’ (‘-‘), or ‘unclear risk’ (‘?‘). Two independent investigators (D.L. and H.J.) carried out the assessments using Review Manager software (Version 5.4, the Cochrane Collaboration, Oxford, UK). Any discrepancies in the risk of bias ratings were resolved through discussion and, when necessary, by involving a third reviewer (C.Y.).
Statistical Analysis
Data analysis was performed using Review Manager software (version 5.4, The Cochrane Collaboration, Oxford, UK) and Stata software (version 14, Stata Corp, TX, USA). After standardizing the extracted variables to consistent units, a random-effects model was applied to assess the differences in outcomes between hypoxic and normoxic interventions. Results were reported as mean differences (MD), 95% confidence intervals (CI), and p-values [4, 30]. Study heterogeneity was quantified using the I² statistic, classified as low (I² < 25%), moderate (25% ≤ I² ≤ 50%), or high (I² >50%) [31]. Publication bias was assessed via funnel plots, along with Begg’s and Egger’s tests. Statistical significance was set at p < 0.05.
In line with prior research [18–21, 32], subgroup analyses were conducted to examine the effect of moderating factors: (1) age (< 40 years vs. ≥ 40 years); (2) hypoxia severity (FiO2 < 15% vs. ≥ 15%). This threshold was selected because more severe hypoxia is associated with a greater physiological burden [18, 33], and approximately 137 million individuals (27% of plateau residents) living at moderate altitudes (2000–2500 m) are exposed to a FiO₂ of around 15% [34]; (3) hypoxic dose (< 50 km·h vs. ≥ 50 km·h), expressed as ‘kilometer hours’ (km·h), where ‘km’ represents exposure elevation in kilometers and ‘h’ denotes total exposure duration in hours [63]; (4) frequency (< 4 days/week vs. ≥ 4 days/week); (5) training duration (< 8 weeks vs. ≥ 8 weeks); and (5) session length (≤ 60 min vs. >60 min).
Results
Search Results
A total of 5020 studies were identified from the databases. After removing 2241 duplicates, 2779 studies were screened based on titles and abstracts. Fifty-one studies underwent full-text screening, and 13 studies met the inclusion criteria and were included in the analysis [13, 14, 23–25, 35–42] (Fig. 1).
Fig. 1.
Flow diagram of study selection
Study Characteristics
A total of 14 trials from 13 studies (one study was included twice due to two groups with different hypoxia severities) were analyzed quantitatively (Table 1). The total number of participants across the 14 trials was 346 (hypoxia: 176; normoxia: 170). Participants’ ages ranged from 18 to 57 years, and BMIs between 25.1 and 37.9 kg/m2. The aerobic training interventions lasted 3–32 weeks, with session of 20–90 min, 2–5 days/week, and intensities of 60–75% V̇O2max or HRmax. In hypoxia groups, FiO2 ranged from 13.0% to 16.5%, with hypoxic doses of 16.5–672 km·h.
Table 1.
Characteristics of the included studies
| Study | Participants | Intervention | Outcome | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Sample size (M/F) | Age (years) | BMI | Duration (week)/Frequency (days/week) | Exercise protocol | Type and severity of hypoxia | Pattern and duration of hypoxic exposure | Hypoxic dose (km·h) | |||||
| HYP | NOR | HYP | NOR | HYP | NOR | |||||||
| Netzer et al. [39] | 10 (2/8) | 10 (2/8) | 50.1 ± 13.0 | 45.5 ± 13.3 | 33.4 ± 3.1 | 32.8 ± 3.3 | 8/3 | 90 min of stepper, treadmill, bicycle ergometer at 60% HRmax | Normobaric, | Intermittent, 90 min/day | 90 | BM, TG, HDL-C |
| FiO2 = 15.0% (2500 m) | ||||||||||||
| Haufe et al. [37] | 10 (10/0) | 10 (10/0) | 29.0 ± 5.9 | 28.1 ± 5.2 | 25.1 ± 1.9 | 24.0 ± 1.6 | 4/3 | 60 min of running at 3 mmol/L lactate heart rate | Normobaric, | Intermittent, 60 min/day | 30 | BM, BMI, TG, LDL-C, HDL-C, HOMA-IR |
| FiO2 = 15.0% (2500 m) | ||||||||||||
| Wiesner et al. [13] | 24 (10/14) | 21 (8/13) | 42.2 ± 1.2 | 42.1 ± 1.7 | 33.1 ± 0.3 | 32.5 ± 0.8 | 4/3 | 60 min of running at 65% of V̇O2max heart rate | Normobaric, | Intermittent, 60 min/day | 30 | LBM, WC, LDL-C,, Insulin, HOMA-IR |
| FiO2 = 15.0% | ||||||||||||
| (2500 m) | ||||||||||||
| Yang et al., [40] | 10 (10/0) | 8 (8/0) | 22.5 ± 1.3 | 22.1 ± 2.2 | 30.7 ± 4.3 | 28.4 ± 2.8 | 4/5 | 60 min of cycling and running at 70–75% HRmax | Normobaric, | Intermittent, 60 min/day | 50 | BM, BMI, FBM, TC, TG, LDL-C, HDL-C |
| FiO2 = 15.4% | ||||||||||||
| (2500 m) | ||||||||||||
| Gatterer et al. [36] | 16 (4/12) | 16 (6/10) | 50.3 ± 10.3 | 52.4 ± 7.9 | 37.9 ± 8.1 | 36.3 ± 4.0 | 32/2 | 90 min of cycling, running and cross trainer at 65–70% HRmax and rested for additional 90 min in normobaric hypoxic chambers | Normobaric, | Intermittent, 180 min/day | 672 | BM, BMI, FBM, WC, HC, WHR, TC, TG, HDL-C, Glucose, |
| FiO2 = 14.0 ± 0.2% (3500 m) | ||||||||||||
| Zhao et al. [41] | 9 (9/0) | 9 (9/0) | 18.2 ± 2.2 | 18.1 ± 1.7 | 32. 9 | 31. 5 | 8/5 | 60 min of cycling at 65–75% V̇O2max | Normobaric, | Intermittent, 60 min/day | 109.6 | BM, FBM |
| FiO2 = 14.7% (2740 m) | ||||||||||||
| Park et al. [42] | 11 (0/11) | 12 (0/12) | 42.0 ± 4.4 | 47.2 ± 6.3 | > 30 | > 30 | 6/5 | 60 min of treadmill and bicycle at 75% HRmax | Normobaric, | Intermittent, 60 min/day | 60/90 | BM, FBM, TC, LDL-C, HDL-C |
| FiO2 = 16.5% (2000 m) and 14.5% (3000 m) | ||||||||||||
| Shin et al. [14] | 8 (8/0) | 9 (9/0) | 45.6 ± 20.9 | 46.0 ± 20.5 | 26.8 ± 2.3 | 27.0 ± 3.0 | 4/3 | 50 min of running in treadmill at 60% HRmax, (5 min warm-up; 40 min main set and 5 min cold down) | Normobaric, | Intermittent, 50 min/day | 24.5 | BM, BMI, FBM, WC, TC, TG, LDL-C, HDL-C, HOMA-IR |
| FiO2 = 15.4% (2500 m) | ||||||||||||
| Klug et al. [38] | 12 (12/0) | 11 (11/0) | 55.0 ± 7.3 | 57.6 ± 7.3 | 35.5 ± 4.8 | 34.1 ± 3.0 | 6/3 | 60 min of walking at 60% HRmax | Normobaric, | Intermittent, 60 min/day | 45 | BM, BMI, LBM, FBM, WC, HC, WHR, TG, LDL-C, HDL-C, Glucose, Insulin |
| FiO2 = 15.0% (2500 m) | ||||||||||||
| Fernández et al. [35] | 12 (2/10) | 11 (2/9) | 34.8 ± 4.7 | 32.2 ± 8.4 | 34.0 ± 2.6 | 32.9 ± 2.7 | 3/3 | 60 min of walking at six different speed | Normobaric, | Intermittent, 60 min/day | 27 | BM, BMI, LBM, FBM, TC, TG, LDL-C, HDL-C, Insulin, HOMA-IR |
| FiO2 = 14.5% (3000 m) | ||||||||||||
| Zhang et al. [23] | 20 (20/0) | 20 (20/0) | 22.3 ± 2.2 | 21.9 ± 2.3 | 32.8 ± 1.2 | 33.3 ± 2.1 | 4/5 | 60 min of cycling at 65% V̇O2max | Normobaric, | Intermittent, 60 min/day | 50 | BM, BMI, FBM, TC, TG, LDL-C, HDL-C |
| FiO2 = 15.0% (2500 m) | ||||||||||||
| Chacaroun et al., [24] | 12 (11/1) | 11 (8/3) | 52.0 ± 12.0 | 56.0 ± 11.0 | 31.2 ± 2.4 | 31.8 ± 3.2 | 8/3 | 45 min of cycling at 75 ± 2% HRmax | Normobaric, | Intermittent, 45 min/day | 55.6 | BMI, LBM, FBM, WC, HC, TC, TG, LDL-C, HDL-C, Insulin, HOMA-IR |
| FiO2 = 13.0% (3700 m) | ||||||||||||
| Namboonlue et al. [25] | 10 (10/0) | 10 (10/0) | 20.3 ± 0.9 | 19.8 ± 0.4 | 26.3 ± 3.4 | 27.0 ± 1.9 | 5/3 | 30 min of running at 60% HRR (5 min warm-up; 20 min main set and 5 min cold down) | Normobaric, | Intermittent, 30 min/day | 16.5 | BM, BMI, LBM |
| FiO2 = 15.8% (2200 m) | ||||||||||||
M: male; F: female; NA: no available; FiO2: inspired fraction of oxygen; HRmax: maximum heart rate; V̇O2max: maximal oxygen consumption; HRR: heart rate reserve; BM: body mass; BMI: body mass index; LBM: lean body mass; FBM: fat body mass; WC: waist circumference; HC: hip circumference; WHR: waist-to-hip ratio; TC: total cholesterol; TG: triglyceride; LDL-C: low-density lipoprotein cholesterol; HDL-C: high density lipoprotein cholesterol; HOMA-IR: homeostasis model assessment of insulin resistance; intermittent: hypoxic training consisting of hypoxia exposure lasting seconds to hours with a return to normoxia or lower levels of hypoxia and repetition over days to weeks
Quality of Study Methods
Risk of bias was evaluated for the 13 included studies. One study was rated as high risk due to incomplete outcome data, while the remaining 12 studies were considered moderate risk (Fig. 2). The funnel plot indicated slight asymmetry, suggesting possible publication bias (Fig. 3). Since funnel plots are subjective, further objective evaluation was performed using Begg’s and Egger’s tests. Most major outcomes showed p-values greater than 0.05, except for LDL-C (p = 0.04), indicating no significant publication bias for these outcomes (tests for WHR were not conducted due to fewer than three studies) (Table 2).
Fig. 2.
Risk of bias summary of included studies
Fig. 3.
Funnel plot showing the mean difference against the standard error for outcomes related to body composition and metabolic health
Table 2.
Summary of meta-analyses
| Outcomes | Indices | N | Mean difference (95% of CI) | p | I2 (p) | Bias |
|---|---|---|---|---|---|---|
| Body Composition | BM | 12 | -0.90 (-1.80, -0.01) | 0.05 | 0% (0.98) | Egger: p = 0.30; Begg: p = 0.63 |
| BMI | 9 | -0.57 (-1.35, 0.20) | 0.15 | 0% (0.63) | Egger: p = 0.88; Begg: p = 0.47 | |
| LBM | 4 | 0.71 (-3.09, 4.52) | 0.71 | 0% (0.98) | Egger: p = 0.49; Begg: p = 1.00 | |
| FBM | 10 | -1.22 (-2.59, 0.15) | 0.08 | 0% (0.97) | Egger: p = 0.09; Begg: p = 0.37 | |
| WC | 5 | -0.52 (-3.95, 2.91) | 0.77 | 0% (0.99) | Egger: p = 0.52; Begg: p = 0.81 | |
| HC | 3 | -0.22 (-4.41, 3.89) | 0.92 | 0% (0.95) | Egger: p = 0.59; Begg: p = 1.00 | |
| WHR | 2 | -0.00 (-0.05, 0.04) | 0.85 | 0% (0.82) | ||
| Metabolism | TC | 8 | -1.51 (-9.82, 6.80) | 0.72 | 0.5% (0.43) | Egger: p = 0.43; Begg: p = 0.90 |
| TG | 9 | -10.78 (-20.68, -0.88) | < 0.05 | 0% (0.53) | Egger: p = 0.98; Begg: p = 0.75 | |
| HDL-C | 11 | -1.72 (-6.17, 2.73) | 0.45 | 44% (0.06) | Egger: p = 0.84; Begg: p = 0.36 | |
| LDL-C | 10 | -3.74 (-6.92, -0.56) | < 0.05 | 0% (0.77) | Egger: p = 0.11; Begg: p = 0.22 | |
| Glucose | 3 | -1.74 (-16.24, 12.76) | 0.81 | 0% (0.97) | Egger: p = 0.31; Begg: p = 1.00 | |
| Insulin | 5 | 0.62 (-1.66, 2.89) | 0.60 | 0% (0.49) | Egger: p = 0.70; Begg: p = 1.00 | |
| HOMA-IR | 5 | -0.22 (-0.33, -0.11) | < 0.05 | 0% (0.50) | Egger: p = 0.40; Begg: p = 0.81 |
BM: body mass; BMI: body mass index; CI: confidence interval; LBM: lean body mass; FBM: fat body mass; WC: waist circumference; HC: hip circumference; WHR: waist-to-hip ratio; TC: total cholesterol; TG: triglyceride; LDL-C: low-density lipoprotein cholesterol; HDL-C: high-density lipoprotein cholesterol; HOMA-IR: homeostasis model assessment of insulin resistance. Bold text signifes statistically significant results
Meta-Analysis Results
Body Composition
The meta-analysis summary of body composition indicators is shown in Table 2. Compared to normoxia, aerobic exercise training in hypoxia resulted in a larger reduction in BM (MD = -0.90, 95% CI: -1.80 to -0.01, p < 0.05, I2 = 0%, Fig. 4) and a favorable trend towards larger FBM loss (MD = -1.22, 95% CI: -2.59 to 0.15, p = 0.08, I2 = 0%, Fig. 5) in overweight and/or obese populations. However, no significant differences were observed between hypoxia and normoxia for BMI, LBM, WC, HC, and WHR (all p > 0.05) (Table 2).
Fig. 4.
Total effects of the intervention on body mass in hypoxia compared to normoxia. “a” and “b” represent the number of trials of the same study. Filled green squares represent study-specific estimates, while the filled diamond represents pooled estimates of random-effects. CI: confidence interval, SD: standard deviation
Fig. 5.
Total effects of the intervention on fat body mass in hypoxia compared to normoxia. “a” and “b” represent the number of trials of the same study. Filled green squares represent study-specific estimates, while the filled diamond represents pooled estimates of random-effects. CI: confidence interval, SD: standard deviation
The summary of the subgroup analysis is shown in Table 3. Aerobic exercise training in hypoxia resulted in greater BM loss compared to normoxia for session lengths of ≥ 60 min (MD = -0.89, 95% CI: -1.80 to -0.01, p = 0.05, I² = 0%), but not for session lengths of < 60 min (MD = -1.25, 95% CI: -7.55 to 5.06, p = 0.70, I² = 0%). A favorable trend towards greater BM loss in hypoxia was also noted at FiO₂ ≥ 15% (MD = -0.85, 95% CI: -1.77 to 0.06, p = 0.07, I² = 0%) and hypoxic dose ≥ 50 km·h (MD = -1.26, 95% CI: -2.50 to -0.02, p = 0.05, I² = 0%), but not at FiO₂ < 15% (MD = -2.00, 95% CI: -6.29 to 2.30, p = 0.36, I² = 0%) or hypoxic dose < 50 km·h (MD = -0.52, 95% CI: -1.80 to -0.77, p = 0.43, I² = 0%). Furthermore, hypoxia led to larger reductions in FBM reductions with hypoxic dose ≥ 50 km·h (MD = -1.54, 95% CI: -3.09 to -0.02, p = 0.05, I² = 0%), training frequency of ≥ 4 days/week (MD = -1.78, 95% CI: -3.46 to -0.01, p < 0.05, I² = 0%) and in participants aged < 40 years (MD = -1.84, 95% CI: -3.66 to -0.02, p = 0.05, I² = 0%), but not with a frequency of < 4 days/week (MD = -0.09, 95% CI: -2.47 to 2.30, p = 0.95, I² = 0%) or hypoxic dose < 50 km·h (MD = -0.16, 95% CI: -3.03 to 2.72, p = 0.91, I² = 0%) or in participants aged ≥ 40 years (MD = -0.43, 95% CI: -2.50 to 1.64, p = 0.69, I² = 0%) (Table 3).
Table 3.
Summary of subgroup analyses
| Outcomes | N | Mean difference (95% of CI) | p | I2 (p) |
|---|---|---|---|---|
| BM | ||||
| Age: ≥ 40 years | 6 | -0.94 (-2.26, 0.38) | 0.16 | 0% (0.99) |
| Age: < 40 years | 6 | -0.87 (-2.09, 0.34) | 0.16 | 0% (0.64) |
| FiO2: ≥ 15% | 8 | -0.85 (-1.77, 0.06) | 0.07 | 0% (0.99) |
| FiO2: < 15% | 4 | -2.00 (-6.29, 2.30) | 0.36 | 0% (0.58) |
| Hypoxic dose ≥ 50 km·h | 7 | -1.26 (-2.50, -0.02) | 0.06 | 0% (0.84) |
| Hypoxic dose < 50 km·h | 5 | -0.52 (-1.80. 0.77) | 0.43 | 0% (0.99) |
| Frequency: ≥ 4 days/week | 5 | -2.27 (-4.98, 0.44) | 0.10 | 0% (0.72) |
| Frequency: < 4 days/week | 7 | -0.73 (-1.68, 0.21) | 0.13 | 0% (0.99) |
| Session length: ≥ 60 min | 10 | -0.89 (-1.80, 0.01) | 0.05 | 0% (0.94) |
| Session length: < 60 min | 2 | -1.25 (-7.55, 5.06) | 0.70 | 0% (0.89) |
| Duration: ≥ 8 weeks | 3 | -1.30 (-3.36, 0.75) | 0.21 | 4.2% (0.35) |
| Duration: < 8 weeks | 9 | -0.73 (-1.90, 0.44) | 0.22 | 0% (0.99) |
| BMI | ||||
| Age: ≥ 40 years | 4 | 0.08 (-1.40, 1.55) | 0.92 | 0% (0.91) |
| Age: < 40 years | 5 | -0.86 (-1.88, 0.15) | 0.10 | 0% (0.63) |
| FiO2: ≥ 15% | 6 | 0.20 (-1.33, 1.73) | 0.07 | 0% (0.86) |
| FiO2: < 15% | 3 | -0.85 (-1.75, 0.06) | 0.80 | 0% (0.48) |
| Hypoxic dose ≥ 50 km·h | 4 | -1.04 (-3.03, 0.95) | 0.31 | 44.6% (0.14) |
| Hypoxic dose < 50 km·h | 5 | -0 38 (-1 31, 0.55) | 0.42 | 0% (0.99) |
| Frequency: ≥ 4 days/week | 2 | 0.46 (-1.63, 2.55) | 0.67 | 0% (0.67) |
| Frequency: < 4 days/week | 7 | -0.24 (-1.09, 0.60) | 0.08 | 0% (0.63) |
| Session length: ≥ 60 min | 6 | -0.74 (-1.58, 0.10) | 0.09 | 0% (0.44) |
| Session length: < 60 min | 3 | -0.05 (-1.46, 1.37) | 0.95 | 0% (0.74) |
| Duration: ≥ 8 weeks | 2 | 0.46 (-1.63, 2.55) | 0.67 | 0% (0.67) |
| Duration: < 8 weeks | 7 | -0.74 (-1.58, 0.10) | 0.08 | 0% (0.56) |
| LBM | ||||
| Age: ≥ 40 years | 2 | 0.29 (-5.43, 6.02) | 0.92 | 0% (0.82) |
| Age: < 40 years | 2 | 1.05 (-4.03, 6.13) | 0.69 | 0% (0.98) |
| FiO2: ≥ 15% | 2 | 0.95 (-4.87, 6.77) | 0.75 | 0% (0.69) |
| FiO2: < 15% | 2 | 0.54 (-4.48, 5.56) | 0.83 | 0% (0.92) |
| FBM | ||||
| Age: ≥ 40 years | 6 | -0.43 (-2.50, 1.64) | 0.69 | 0% (0.99) |
| Age: < 40 years | 4 | -1.84 (-3.66, -0.02) | 0.05 | 0% (0.97) |
| FiO2: ≥ 15% | 5 | -1.54 (-3.55, 0.47) | 0.13 | 0% (0.94) |
| FiO2: < 15% | 5 | -0.95 (-2.82, 0.92) | 0.32 | 0% (0.74) |
| Hypoxic dose ≥ 50 km·h | 7 | -1.54 (-3.09, -0.02) | 0.05 | 0% (0.93) |
| Hypoxic dose < 50 km·h | 3 | -0.16 (-3.03, 2.72) | 0.91 | 0% (0.77) |
| Frequency: ≥ 4 days/week | 5 | -1.78 (-3.46, -0.01) | < 0.05 | 0% (0.89) |
| Frequency: < 4 days/week | 5 | -0.09 (-2.47, 2.30) | 0.95 | 0% (0.97) |
| Session length: ≥ 60 min | 8 | -1.24 (-2.70, 0.21) | 0.09 | 0% (0.90) |
| Session length: < 60 min | 2 | -1.06 (-5.08, 2.97) | 0.61 | 0% (0.74) |
| Duration: ≥ 8 weeks | 2 | -1.99 (-4.73, 0.75) | 0.16 | 0% (0.97) |
| Duration: < 8 weeks | 8 | -0.97 (-2.55, 0.61) | 0.23 | 0% (0.97) |
| WC | ||||
| Age: ≥ 40 years | 3 | -0.20 (-4.55, 4.15) | 0.93 | 0% (0.94) |
| Age: < 40 years | 2 | -1.05 (-6.63, 4.52) | 0.71 | 0% (0.70) |
| FiO2: ≥ 15% | 3 | -1.26 (-6.04, 3.51) | 0.60 | 0% (0.94) |
| FiO2: < 15% | 2 | 0.27 (-4.66, 5.91) | 0.92 | 0% (0.98) |
| Hypoxic dose ≥ 50 km·h | 2 | 0.27 (-4.66, 5.19) | 0.92 | 0% (0.98) |
| Hypoxic dose < 50 km·h | 3 | -1.26 (-6.04, 3.51) | 0.60 | 0% (0.93) |
| TC | ||||
| Age: ≥ 40 years | 5 | 9.75 (-2.95, 22.45) | 0.13 | 0% (0.92) |
| Age: < 40 years | 3 | -9.87 (-20.74,1.01) | 0.08 | 0% (0.66) |
| FiO2: ≥ 15% | 4 | -7.93 (-18.58, 2.73) | 0.15 | 0% (0.57) |
| FiO2: < 15% | 4 | 8.02 (-5.06, 21.09) | 0.23 | 0% (0.67) |
| Hypoxic dose ≥ 50 km·h | 6 | 0.45 (-10.98, 11.88) | 0.93 | 28.4% (0.22) |
| Hypoxic dose < 50 km·h | 2 | -2.12 (-28.07, 23.84) | 0.87 | 0% (0.82) |
| Frequency: ≥ 4days/week | 4 | -2.82 (-16.74,11.07) | 0.69 | 33.4% (0.21) |
| Frequency: < 4days/week | 4 | 5.61 (-9.94, 21.16) | 0.48 | 0% (0.71) |
| Session length: ≥ 60 min | 2 | 0.01 (-11.48, 11.50) | 0.99 | 28.5% (0.22) |
| Session length: < 60 min | 6 | 0.73 (-24.62, 26.09) | 0.95 | 0% (0.92) |
| Duration: ≥ 8 weeks | 5 | 9.93 (-0.48, 20.35) | 0.32 | 0% (0.37) |
| Duration: < 8 weeks | 6 | -4.11 (-13.23, 5.02) | 0.38 | 0% (0.47) |
| TG | ||||
| Age: ≥ 40 years | 5 | -8.56 (-37.83, 20.70) | 0.57 | 0% (0.43) |
| Age: < 40 years | 4 | -10.93 (-20.78, -1.08) | 0.08 | 8.1% (0.35) |
| FiO2: ≥ 15% | 6 | -13.28 (-23.49, -3.08) | < 0.05 | 0% (0.69) |
| FiO2: < 15% | 3 | 20.98 (-16.60, 58.55) | 0.27 | 0% (0.59) |
| Hypoxic dose ≥ 50 km·h | 5 | -11.99 (-23.26, -0.73) | < 0.05 | 0.1% (0.40) |
| Hypoxic dose < 50 km·h | 4 | -7.25 (-28.02, 13.53) | 0.49 | 0% (0.39) |
| Frequency: ≥ 4 days/week | 2 | -20.95 (-61.42, 19.52) | 0.31 | 27.1% (0.24) |
| Frequency: < 4 days/week | 7 | -6.00 (-23.31, 11.31) | 0.50 | 0% (0.51) |
| Session length: ≥ 60 min | 7 | -10.80 (-20.85, -0.70 | < 0.05 | 0% (0.46) |
| Session length: < 60 min | 2 | -22.68 (-94.91, 49.55) | 0.59 | 31.6% (0.23) |
| Training duration: ≥ 8 weeks | 3 | -2.38 (-36.04, 31.28) | 0.89 | 12.4% (0.32) |
| Training duration: < 8 weeks | 6 | -11.78 (-22.16, -1.40) | < 0.05 | 0% (0.47) |
| LDL-C | ||||
| Age: ≥ 40 years | 5 | -2.04 (-7.61, 3.51) | 0.47 | 0% (0.83) |
| Age: < 40 years | 5 | -0.46 (-14.05, 13.13) | 0.94 | 0% (0.86) |
| FiO2: ≥ 15% | 7 | -3.77 (-7.03, -0.52) | < 0.05 | 0% (0.79) |
| FiO2: < 15% | 3 | -2.94 (-17.82, 11.93) | 0.69 | 0% (0.68) |
| Hypoxic dose ≥ 50 km·h | 5 | -2.96 (-7.86, 1.94) | 0.34 | 10.8% (0.34) |
| Hypoxic dose < 50 km·h | 5 | 0.54 (-10.78, 11.86) | 0.96 | 0% (0.96) |
| Frequency: ≥ 4 days/week | 4 | -1.98 (-8.84, 4.87) | 0.57 | 32% (0.22) |
| Frequency: < 4 days/week | 6 | 0.47 (-10.10, 11.05) | 0.92 | 0% (0.51) |
| Session length: ≥ 60 min | 8 | -3.85 (-7.06, -0.64) | < 0.05 | 0% (0.46) |
| Session length: < 60 min | 2 | 2.94 (-20.98, 26.87) | 0.80 | 0% (0.47) |
| HDL-C | ||||
| Age: ≥ 40 years | 7 | 1.05 (-2.68, 4.78) | 0.58 | 0% (0.97) |
| Age: < 40 years | 4 | -6.67 (-16.00, 2.66) | 0.16 | 51.4% (0.10) |
| FiO2: ≥ 15% | 7 | -2.55 (-9.40, 4.29) | 0.46 | 54.4% (0.04) |
| FiO2: < 15% | 4 | 0.43 (-4.52, 5.39) | 0.86 | 5.4% (0.36) |
| Hypoxic dose ≥ 50 km·h | 7 | -2.42 (-9.12, 4.28) | 0.47 | 65.5% (0.01) |
| Hypoxic dose < 50 km·h | 4 | -0.80 (-6.97, 5.36) | 0.79 | 0% (0.94) |
| Frequency: ≥ 4 days/week | 4 | -4.10 (-14.09, 6.06) | 0.43 | 78.5% (0.01) |
| Frequency: < 4 days/week | 7 | -0.73 (-5.44, 3.97) | 0.76 | 0% (0.79) |
| Session length: ≥ 60 min | 9 | -1.24 (-6.18, 3.68) | 0.62 | 49.9% (0.04) |
| Session length: < 60 min | 2 | -5.63 (-17.72, 6.45) | 0.36 | 15.0% (0.27) |
| Training duration: ≥ 8 weeks | 3 | -0.53 (-10.19, 9.12) | 0.91 | 40.8% (0.04) |
| Training duration: < 8 weeks | 8 | -2.13 (-7.54, 3.26) | 0.43 | 51.1% (0.18) |
| Insulin | ||||
| Age: ≥ 40 years | 3 | 0.23 (-3.62, 4.08) | 0.91 | 9.4% (0.33) |
| Age: < 40 years | 2 | 1.03 (-2.28, 4.34) | 0.54 | 9.5% (0.29) |
| FiO2: ≥ 15% | 3 | 0.26 (-3.17, 3.69) | 0.88 | 9.1% (0.33) |
| FiO2: < 15% | 2 | 1.19 (-2.46, 4.85) | 0.52 | 4.7% (0.31) |
| Session length: ≥ 60 min | 3 | 1.37 (-1.47, 4.21) | 0.34 | 0% (0.39) |
| Session length: < 60 min | 2 | -0.74 (-4.55, 3.07) | 0.70 | 0% (0.38) |
| HOMA-IR | ||||
| Age: ≥ 40 years | 2 | -0.04 (-0.59, 0.50) | 0.88 | 0% (0.40) |
| Age: < 40 years | 3 | -0.19 (-0.41, 0.02) | 0.08 | 11.7% (0.32) |
| FiO2: ≥ 15% | 3 | -0.24 (-0.35, -0.13) | < 0.05 | 0% (0.79) |
| FiO2: < 15% | 2 | 0.14 (-0.33, 0.61) | 0.56 | 0% (0.50) |
| Session length: ≥ 60 min | 3 | -0.19 (-0.41, 0.02) | 0.08 | 11.7% (0.32) |
| Session length: < 60 min | 2 | -0.04 (-0.59, 0.50) | 0.88 | 0% (0.40) |
N: Number of studies; CI: confidence interval; FiO2: inspired fraction of oxygen; BM: body mass; BMI: body mass index; LBM: lean body mass; FBM: fat body mass; WC: waist circumference; HC: hip circumference; WHR: waist-to-hip ratio; TC: total cholesterol; TG: triglyceride; LDL-C: low-density lipoprotein cholesterol; HDL-C: high- density lipoprotein cholesterol; HOMA-IR: homeostasis model assessment of insulin resistance. Bold text signifes statistically significant results
Metabolic Health
The meta-analysis summary of metabolic health indicators is displayed in Table 2. Compared to normoxia, aerobic exercise training in hypoxia lead to larger reductions in TG (MD = -10.78, 95% CI: -20.68 to 0.88, p = 0.05, I2 = 0%, Fig. 6), LDL-C (MD = -3.74, 95% CI: -6.92 to -0.56, p < 0.05, I2 = 0%, Fig. 7), and HOMA-IR (MD = -0.22, 95% CI: -0.33 to -0.11, p < 0.01, I2 = 0%, Fig. 8). No significant differences were observed for TC, HDL-C, blood glucose and insulin between the two conditions (all p > 0.05) (Table 2).
Fig. 6.
Total effects of the intervention on triglyceride in hypoxia compared to normoxia. Filled green squares represent study-specific estimates, while the filled diamond represents pooled estimates of random-effects. CI: confidence interval, SD: standard deviation
Fig. 7.
Total effects of the intervention on low-density lipoprotein cholesterol in hypoxia compared to normoxia. “a” and “b” represent the number of trials of the same study. Filled green squares represent study-specific estimates, while the filled diamond represents pooled estimates of random-effects. CI: confidence interval, SD: standard deviation
Fig. 8.
Total effects of the intervention on homeostasis model assessment of insulin resistance in hypoxia compared to normoxia. Filled green squares represent study-specific estimates, while the filled diamond represents pooled estimates of random-effects. CI: confidence interval, SD: standard deviation
The results of the subgroup analysis are presented in Table 3. Larger reductions in TG were reported after aerobic exercise training in hypoxia compared to normoxia for subgroups with training duration of < 8 weeks (MD = -11.78, 95% CI: -22.16 to -1.40, p < 0.05, I² = 0%), session length of ≥ 60 min (MD = -10.80, 95% CI: -20.85 to 0.70, p = 0.05, I² = 0%), FiO₂ ≥ 15% (MD = -12.75, 95% CI: -23.02 to -2.49, p < 0.05, I² = 0%) and hypoxic dose ≥ 50 km·h (MD = -11.99, 95% CI: -23.26 to -0.73, p < 0.05, I² = 0%). However, no significant effects were found in other subgroups (all p > 0.05) (Table 3). For LDL-C, aerobic exercise training in hypoxia induced larger effects than normoxia in subgroups with a session length ≥ 60 min (MD = -3.85, 95% CI: -7.06 to -0.64, p < 0.05, I² = 0%), FiO₂ ≥ 15% (MD = -3.77, 95% CI: -7.03 to -0.52, p < 0.05, I² = 0%). No significant differences were reported in the other subgroups (Table 3).
Discussion
This meta-analysis is the first to compare the independent (not combined with other exercise forms) effects of aerobic exercise training interventions in hypoxia versus normoxia on body composition and metabolic health in overweight and/or obese individuals. Our results support our hypothesis, indicating that training in hypoxia is more effective than in normoxia for reducing BM, FBM, TG, LDL-C, and HOMA-IR. These findings highlight the benefits of aerobic exercise training in hypoxia, supporting its potential as an effective intervention for obesity management.
Body Composition Outcomes
Our results indicate that aerobic exercise training in hypoxia leads to a greater reduction in BM compared to normoxia in overweight and/or obese individuals (Fig. 4). Typically, BM loss results from reductions in FBM, LBM, or both [43]. Given the similar benefits observed on LBM (p = 0.71) and other body composition indicators (BMI, LBM, WC, HC, WHR, p = 0.15–0.92) between the two conditions (Table 2), it can be speculated that the larger BM reduction is primarily due to a trend towards larger FBM loss (p = 0.08). This is consistent with a meta-analysis by He et al. [20], which demonstrated that exercise training in hypoxia more effectively reduces FBM than in normoxia in middle-aged and older adults. Several physiological mechanisms may explain the enhanced fat loss observed with hypoxic aerobic exercise. Aerobic exercise training in hypoxia may elevate basal metabolic rate by enhancing mitochondrial quantity and efficiency [44–46], which could reduce fat accumulation. Improved mitochondrial function could also increase fat oxidation, further promoting FBM loss [47]. Supporting this, a previous study reported an increasing trend in fat oxidation during the post-exercise rest period after training in hypoxia, while a decreasing trend was observed after the same training in normoxia [47]. Additionally, hypoxic exposure may stimulate leptin secretion while lowering acylated ghrelin levels [48, 49], thereby suppressing appetite and reducing calorie intake. The reduction in calorie intake contributes to a larger energy deficit, which is critical for effective obesity management [50].
A previous meta-analysis that included various exercise modalities (e.g., resistance, high-intensity interval, and aerobic exercise) showed no greater reduction in FBM with hypoxic exercise compared to normoxia in overweight and/or obese individuals [19]. The discrepancy between their results and ours regarding FBM loss may stem from differences in the types of exercise modalities included (a combination of aerobic exercise, resistance exercise, and high-intensity interval training in [19] vs. aerobic exercise only in the present study). Recent studies suggest that aerobic exercise is more effective than resistance exercise and high-intensity interval training for reducing FBM in overweight and/or obese populations [3, 4, 22]. While high-intensity interval training and resistance exercise primarily promote hormone secretion (e.g., catecholamines and growth hormone) and increase LBM [51, 52], aerobic exercise mainly targets fat mass reduction [52]. However, there is currently no direct evidence comparing the effects of different exercise modalities on FBM reduction in hypoxia, which warrants further research.
Our subgroup analysis showed that, compared to normoxia, aerobic exercise training in hypoxia led to a greater reduction in FBM for individuals aged < 40 years, while no such effect was observed in those aged ≥ 40 years (Table 3). However, a meta-analysis by Ramos-Campo et al. [18], which included various exercise modalities, failed to report larger reduction in FBM with hypoxic exercise training over normoxia in overweight and/or obese individuals aged 14–57 years. In addition to the influence of mixed exercise types discussed earlier, these divergent findings may also stem from the confounding factor of age. In their meta-analysis, approximately half of the participants were over 40 years old [18]. Research has demonstrated that aging is associated with a decline in both daily physical activity and basal metabolic rate [20, 53], which may reduce energy expenditure and limit the effectiveness of hypoxic exercise interventions. Overall, compared to normoxia, aerobic exercise training in hypoxia results in larger reductions in FBM and BM in overweight and/or obese individuals, particularly those aged < 40 years, suggesting its potential for improving body composition.
Metabolic Health Outcomes
Aerobic exercise training in hypoxia resulted in a greater reduction in TG compared to normoxia in overweight and/or obese individuals (Figs. 6 and 7). This is consistent with a previous meta-analysis, which found that hypoxic exercise training led to a greater decrease in TG in this population [21]. We also observed a larger reduction in LDL-C following aerobic exercise training in hypoxia. These findings further support previous research suggesting that hypoxic aerobic exercise can alleviate dyslipidemia and maintain lipid metabolic health [13, 14]. Among all the studies included in our meta-analysis, the only difference between the normoxia and hypoxia groups was exposure to a hypoxic environment during exercise. Consequently, the observed improvements in lipid metabolism can be predominantly attributed to hypoxic exposure. The underlying mechanisms are likely related to the activation of HIF triggered by hypoxic exposure [54]. Specifically, HIF upregulates peroxisome proliferator-activated receptor gamma and peroxisome proliferator-activated receptor coactivator 1, both of which play key roles in improving lipid metabolism [45, 55]. Additionally, HIF stimulates erythropoietin and nitric oxide secretion [54], thereby increasing blood flow in adipose tissue [56] and promoting fatty acid oxidation [21].
Insulin resistance is a common complication in obese populations [57], and our meta-analysis is the first to examine the effects of hypoxic intervention on insulin resistance by including the HOMA-IR index. Our results reveal that aerobic exercise training in hypoxia leads to a greater reduction in HOMA-IR (Fig. 8). This may be due to hypoxic conditions alleviating insulin resistance by improving insulin sensitivity, which in turn helps maintain healthy glucose metabolism [15]. Additionally, we found no significant difference in insulin and blood glucose reduction between hypoxia and normoxia when aerobic exercise training was performed at similar relative intensities (with lower absolute intensity in hypoxia) (Table 2). This aligns with a previous meta-analysis [18], suggesting that hypoxic interventions may provide similar benefits for glucose metabolism as normoxia, yet at a lower absolute intensity [39, 58]. Furthermore, no significant differences were observed in HDL-C between hypoxia and normoxia following aerobic training (p = 0.45) (Table 2). This may be due to differences in exercise intensity. Previous studies have suggested that high exercise intensities (> 60% V̇O2max or > 60% HRR) are required to meaningfully improve HDL-C levels [1, 59]. Most studies included in our meta-analyses utilized moderate-intensity (45–65% V̇O2max or 50–65% HRR or 65–75% HRmax) exercise protocols, which may not have been sufficient to effectively increase HDL-C levels in either hypoxic or normoxic conditions.
Since the studies in our review employed aerobic exercise training at the same relative intensity (e.g., % V̇O 2max and %HRmax) across conditions, the absolute intensity (and thus energy expenditure) was lower in the hypoxic group than in the normoxic group. This suggests that hypoxic training at a lower absolute intensity may elicit greater improvements in body composition and metabolic health compared to normoxic training. Supporting this, a previous study showed that exercising at the same relative intensity resulted in lower workloads (1.39 vs. 1.67 W/kg) in hypoxia versus normoxia, yet led to greater improvements in FBM and HOMA-IR [37]. Therefore, the larger improvement in FBM, TG, LDL-C, and HOMA-IR observed in our meta-analysis was likely attributable to the effects of hypoxic exposure.
Subgroup Analysis
Hypoxic Severity and Dose
Our findings indicate that FiO₂ ≥ 15% more effectively reduces TG and LDL-C compared to FiO₂ < 15%, resulting in greater benefits for lipid metabolism (Table 3). Previous studies have indicated that total work and absolute intensity of exercise are key factors in improving lipid metabolism [18]. However, more severe hypoxic conditions may provide a stronger hypoxic stimulus [60], potentially reducing participants’ total work and limiting the physiological benefits of aerobic exercise training in hypoxia for overweight and/or obese populations [18, 33]. Furthermore, considering that more severe hypoxia is associated with a higher physiological burden [61], our study recommends a moderate hypoxic level (FiO₂ ≥ 15%) for obese individuals to achieve metabolic health benefits. Additionally, a greater reduction in BF, FBM, and TG were observed with a hypoxic dose of ≥ 50 km·h (Table 3). This indicates that the design of hypoxia protocols should not only consider hypoxic severity (FiO₂ ≥ 15%) but also ensure sufficient exposure time to achieve an adequate hypoxic dose, thereby maximizing health benefits. Nonetheless, due to limited evidence directly comparing different hypoxia severities and doses, these findings should be interpreted with caution.
Frequency, Session Length, and Training Duration
Compared to normoxia, aerobic exercise training in hypoxia with a frequency of ≥ 4 days/week and session length ≥ 60 min resulted in larger reductions in BM, FBM, TG and LDL-C, effects not seen in other subgroups (Table 3). Consistent with our results, another meta-analysis reported that hypoxic exercise sessions lasting at least 60 min are necessary to produce larger reductions in FBM and BM compared to normoxic exercise training [20]. Furthermore, our results indicated that intervention durations of < 8 weeks were more effective than those ≥ 8 weeks. This may be due to physiological adaptation or tolerance to prolonged hypoxic exposure, which could diminish the additional benefits over time. Supporting this, a previous study found that aerobic exercise training in hypoxia (90 min of cycling, running and cross-training at 65–70% HRmax) was more effective in improving body composition during the first five weeks, but the added benefits disappeared after three months [36]. This implies that hypoxic training may be particularly effective as a short-term intervention (< 8 weeks), while longer-term interventions (≥ 8 weeks) may require an appropriate wash-out period. However, it should be noted that fewer studies in our review used intervention durations of ≥ 8 weeks compared to < 8 weeks (4 vs. 9), which limits the robustness of this conclusion. Future research should aim to clarify the long-term effects of hypoxic exercise training on obesity.
Age
Several previous studies indicated that hypoxic exercise training (30–60 min of running, 4–5 days/week for 3 weeks at 60% HRR or at a heart rate corresponding to 3 mmol/L lactate) provides greater reductions in BF and TG compared to normoxic exercise training in young adults (19–29 years) [25, 37]. In contrast, other studies using comparable protocols (60 min of walking or bicycle, 6 days/week for 3–6 weeks at 60–75% HRmax) did not observe additional benefits in older adults (50–57 years) [38, 42]. This discrepancy may be due to the effects of hypoxic exercise training being moderated by age. Our subgroup analysis further supported this, revealing that aerobic exercise training in hypoxia was more effective in improving FBM, TG, and LDL-C in overweight and/or obese individuals aged < 40 years compared to those aged ≥ 40 years (Table 3). This could be attributed to younger individuals having higher levels of hormones (e.g., growth hormone and testosterone) which promote fat mobilization and utilization while inhibiting fat synthesis, thereby leading to greater benefits from hypoxia exposure [53]. However, due to the lack of direct evidence comparing the effects of aerobic exercise training in hypoxia across different age groups, potential age-related differences in its effectiveness warrant further investigation through original research.
Practical Implications
Our research suggests that aerobic exercise training in hypoxia is a more effective option for improving body composition and maintaining metabolic health in overweight and/or obese populations compared to normoxia. Practitioners should consider factors such as age, severity of hypoxia, training frequency, session duration, and overall training duration to optimize the benefits of the hypoxic intervention strategy.
Limitations and Future Directions
This study has several limitations. First, most of the included studies employed moderate-intensity aerobic exercise training in hypoxia, meaning our meta-analysis did not examine the effects of aerobic exercise at varying intensities. It remains unclear whether higher exercise intensities would provide additional physiological benefits. Second, due to the lack of studies focusing on females only, our meta-analysis did not explore the impact of sex on the effects of hypoxic interventions. A study by Sandoval et al. [62] specifically examined sex differences in substrate utilization under hypoxic conditions and found that women tend to rely more on fat during exercise, while men tend to rely more on carbohydrates. Lastly, some studies have reported that combined exercise modalities (e.g., aerobic exercise with resistance training) lead to larger improvements in body composition and metabolic health compared to aerobic exercise alone [3, 4, 22]. Future research should determine the most effective exercise modalities and consider the impact of potential factors (e.g., sex and exercise intensity). This would help develop more personalized, effective, and systematic intervention strategies tailored to individual characteristics.
Conclusion
Aerobic exercise training in hypoxia is more effective than in normoxia for reducing body mass, fat body mass, triglycerides, and low-density lipoprotein cholesterol in overweight and/or obese individuals. These findings could inform the development of effective hypoxic exercise interventions for managing obesity, as this approach gains popularity.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
We sincerely appreciate the participants for their valuable time and diligent effort in this study.
Abbreviations
- BM
Body mass
- BMI
Body mass index
- CI
Confidence intervals
- FBM
Fat body mass
- FiO2
Inspired oxygen fraction
- LBM
Lean body mass
- HC
Hip circumference
- HDL-C
High-density lipoprotein cholesterol
- HOMA-IR
Homeostasis model assessment index of insulin resistance
- LDL-C
Low-density lipoprotein cholesterol
- MD
Mean difference
- O2
Oxygen
- SD
Standard deviation
- SE
Standard error
- SEM
Standard error of mean
- TC
Total cholesterol
- TG
Triglycerides
- WC
Waist circumference
- WHR
Waist-to-hip ratio
- V̇O2max
Maximal oxygen uptake
Author Contributions
Y.C. and L.D. conceived and designed the study. L.D. and J.H. contributed to data collection. L.D., J.H. and B.C. performed data analysis. J.L., L.D., L.G., and Y.C. interpreted the experimental results. L.D. and J.H. prepared the figures. J.H. and L.D. the drafted the manuscript. L.D., J.H., J.L., Y.X., B.C., L.G., O.G. and Y.C. edited and revised the manuscript. All authors read and approved the final manuscript.
Funding
This study was supported by the Key Lab of Exercise and Health Sciences, Ministry of Education at Shanghai University of Sport (grant number: 2025KF0005) and the Shanghai Key Laboratory of Human Performance at Shanghai University of Sport (grant number: 11DZ2261100).
Data Availability
Data are available from the corresponding author upon reasonable request.
Declarations
Ethics Approval and Consent to Participate
Not applicable.
Consent for Publication
Not applicable.
Competing Interests
L.D., J.H., J.L., Y.X., B.C., L.G., O.G. and Y.C. declare that they have no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Li Ding, Jin Huang and Bin Chen contributed equally to this work.
References
- 1.ACSM. ACSM’s guidelines for exercise testing and prescription. 10th ed. Philadelphia: Wolters Kluwer Health; 2018. [Google Scholar]
- 2.Bull FC, Al-Ansari SS, Biddle S, Borodulin K, Buman MP, Cardon G, et al. World health organization 2020 guidelines on physical activity and sedentary behaviour. Br J Sports Med. 2020;54(24):1451–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chen X, He H, Xie K, Zhang L, Cao C. Effects of various exercise types on visceral adipose tissue in individuals with overweight and obesity: A systematic review and network meta-analysis of 84 randomized controlled trials. Obes Reviews: Official J Int Association Study Obes. 2024;25(3):e13666. [DOI] [PubMed] [Google Scholar]
- 4.O’Donoghue G, Blake C, Cunningham C, Lennon O, Perrotta C. What exercise prescription is optimal to improve body composition and cardiorespiratory fitness in adults living with obesity? A network meta-analysis. Obes Reviews: Official J Int Association Study Obes. 2021;22(2):e13137. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Elmaleh-Sachs A, Schwartz JL, Bramante CT, Nicklas JM, Gudzune KA, Jay M. Obesity management in adults: A review. JAMA. 2023;330(20):2000–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ekkekakis P, Hall EE, Petruzzello SJ. The relationship between exercise intensity and affective responses demystified: to crack the 40-year-old nut, replace the 40-year-old nutcracker! Annals Behav Medicine: Publication Soc Behav Med. 2008;35(2):136–49. [DOI] [PubMed] [Google Scholar]
- 7.Browning RC, Kram R. Effects of obesity on the biomechanics of walking at different speeds. Med Sci Sports Exerc. 2007;39(9):1632–41. [DOI] [PubMed] [Google Scholar]
- 8.Sheehan KJ, Gormley J. The influence of excess body mass on adult gait. Clin Biomech (Bristol Avon). 2013;28(3):337–43. [DOI] [PubMed] [Google Scholar]
- 9.Wearing SC, Hennig EM, Byrne NM, Steele JR, Hills AP. The biomechanics of restricted movement in adult obesity. Obes Reviews: Official J Int Association Study Obes. 2006;7(1):13–24. [DOI] [PubMed] [Google Scholar]
- 10.Dünnwald T, Gatterer H, Faulhaber M, Arvandi M, Schobersberger W. Body composition and body weight changes at different altitude levels: A systematic review and Meta-Analysis. Front Physiol. 2019;10:430. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hamad N, Travis SPL. Weight loss at high altitude: pathophysiology and practical implications. Eur J Gastroenterol Hepatol. 2006;18. [DOI] [PubMed]
- 12.Girard O, Malatesta D, Millet GP. Walking in hypoxia: an efficient treatment to lessen mechanical constraints and improve health in obese individuals?? Front Physiol. 2017;8:73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Wiesner S, Haufe S, Engeli S, Mutschler H, Haas U, Luft FC, et al. Influences of Normobaric hypoxia training on physical fitness and metabolic risk markers in overweight to obese subjects. Obes (Silver Spring Md). 2010;18(1):116–20. [DOI] [PubMed] [Google Scholar]
- 14.Sohee Shin WYS. Influences of Short-Term Normobaric hypoxic training on metabolic Syndrome-Related markers in overweight and Normal-Weight men. J Mens Health. 2018;14(1):e44–52. [Google Scholar]
- 15.Soo J, Raman A, Lawler NG, Goods PSR, Deldicque L, Girard O, et al. The role of exercise and hypoxia on glucose transport and regulation. Eur J Appl Physiol. 2023;123(6):1147–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Gaspar JM, Velloso LA. Hypoxia inducible factor as a central regulator of Metabolism - Implications for the development of obesity. Front NeuroSci. 2018;12:813. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Wenger RH. Cellular adaptation to hypoxia: O2-sensing protein hydroxylases, hypoxia-inducible transcription factors, and O2-regulated gene expression. Faseb J. 2002;16(10):1151–62. [DOI] [PubMed] [Google Scholar]
- 18.Chen S, Su H, Liu X, Li Q, Yao Y, Cai J, et al. Effects of exercise training in hypoxia versus normoxia on fat-reducing in overweight and/or obese adults: A systematic review and meta-analysis of randomized clinical trials. Front Physiol. 2022;13:940749. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Guo H, Cheng L, Duolikun D, Yao Q. Aerobic exercise training under Normobaric hypoxic conditions to improve glucose and lipid metabolism in overweight and obese individuals: A systematic review and Meta-Analysis. High Alt Med Biol. 2023;24(4):312–20. [DOI] [PubMed] [Google Scholar]
- 20.He Z, Qiang L, Liu Y, Gao W, Feng T, Li Y, et al. Effect of hypoxia conditioning on body composition in Middle-Aged and older adults: A systematic review and Meta-Analysis. Sports Med - open. 2023;9(1):89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ramos-Campo DJ, Girard O, Pérez A, Rubio-Arias J. Additive stress of Normobaric hypoxic conditioning to improve body mass loss and cardiometabolic markers in individuals with overweight or obesity: A systematic review and meta-analysis. Physiol Behav. 2019;207:28–40. [DOI] [PubMed] [Google Scholar]
- 22.Huang Z, Li J, Liu Y, Zhou Y. Effects of different exercise modalities and intensities on body composition in overweight and obese children and adolescents: a systematic review and network meta-analysis. Front Physiol. 2023;14:1193223. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Zhang N. Effects of hypoxia and moderate intensity training therapy on body composition and blood lipid metabolism in obese subjects. Genomics Appl Biology. 2019;38(12):5657–63. [Google Scholar]
- 24.Chacaroun S, Borowik A, Vega-Escamilla YGI, Doutreleau S, Wuyam B, Belaidi E, et al. Hypoxic exercise training to improve exercise capacity in obese individuals. Med Sci Sports Exerc. 2020;52(8):1641–9. [DOI] [PubMed] [Google Scholar]
- 25.Namboonlue C, Namboonlue S, Wanna S. The effects of five weeks Low-Intensity aerobic training under hypoxia on body composition and oxygen consumption in overweight/obesity men. J Exerc Physiol Online. 2021;08(01):24:33–44. [Google Scholar]
- 26.Moher D, Shamseer L, Clarke M, Ghersi D, Liberati A, Petticrew M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Reviews. 2015;4(1):1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Thompson PD, Arena R, Riebe D, Pescatello LS. ACSM’s new preparticipation health screening recommendations from ACSM’s guidelines for exercise testing and prescription, ninth edition. Current sports medicine reports. 2013;12(4):215–7. [DOI] [PubMed]
- 28.Higgins JPTTJ, Chandler J, Cumpston M, Li T, Page M. J. Cochrane handbook for systematic reviews of interventions. version 6.3 2022 [cited; Available from.
- 29.Cohen J. Statistical Power for the Behavioural Sciences: Statistical power analysis for the behavioral sciences /; 1988.
- 30.Hsu KJ, Liao CD, Tsai MW, Chen CN. Effects of exercise and nutritional intervention on body composition, metabolic health, and physical performance in adults with sarcopenic obesity: A Meta-Analysis. Nutrients. 2019;11(9). [DOI] [PMC free article] [PubMed]
- 31.Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med. 2002;21(11):1539–58. [DOI] [PubMed] [Google Scholar]
- 32.Montero D, Roberts CK, Vinet A. Effect of aerobic exercise training on arterial stiffness in obese populations. Sports Med. 2014;44(6):833–43. [DOI] [PubMed] [Google Scholar]
- 33.Kelly LP, Basset FA. Acute Normobaric hypoxia increases Post-exercise lipid oxidation in healthy males. Front Physiol. 2017;8:293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Tremblay JC, Ainslie PN. Global and country-level estimates of human population at high altitude. Proc Natl Acad Sci U S A. 2021;118(18). [DOI] [PMC free article] [PubMed]
- 35.Fernandez Menendez A, Saudan G, Sperisen L, Hans D, Saubade M, Millet GP, et al. Effects of Short-Term Normobaric hypoxic walking training on energetics and mechanics of gait in adults with obesity. Obes (Silver Spring). 2018;26(5):819–27. [DOI] [PubMed] [Google Scholar]
- 36.Gatterer H, Haacke S, Burtscher M, Faulhaber M, Melmer A, Ebenbichler C, et al. Normobaric intermittent hypoxia over 8 months does not reduce body weight and metabolic risk Factors–a randomized, single blind, Placebo-Controlled study in Normobaric hypoxia and Normobaric Sham hypoxia. Obes Facts. 2015;8(3):200–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Haufe S, Wiesner S, Engeli S, Luft FC, Jordan J. Influences of Normobaric hypoxia training on metabolic risk markers in human subjects. Med Sci Sports Exerc. 2008;40(11):1939–44. [DOI] [PubMed] [Google Scholar]
- 38.Klug L, Mahler A, Rakova N, Mai K, Schulz-Menger J, Rahn G, et al. Normobaric hypoxic conditioning in men with metabolic syndrome. Physiol Rep. 2018;6(24):e13949. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Netzer NC, Chytra R, Kupper T. Low intense physical exercise in Normobaric hypoxia leads to more weight loss in obese people than low intense physical exercise in Normobaric Sham hypoxia. Sleep Breath. 2008;12(2):129–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Yang X, He W, Shi D, He Y, Hu Y. The effects of hypoxic exercise on the energy intake, body composition and lipid metabolism in overweight and obese young men. Chin J Sports Med. 2014;33(7):638–45. [Google Scholar]
- 41.Zhao S, Shi H. Effects of hypoxic training on body composition and endurance capacity of obese students. Chin J School Health. 2016;37(11):1637–40. [Google Scholar]
- 42.Park HY, Lim K. The effects of aerobic exercise at hypoxic condition during 6 weeks on body composition, blood pressure, arterial stiffness, and blood lipid level in obese women. Int J Sports Sci. 2017;3(8):001–5. [Google Scholar]
- 43.Shi Q, Wang Y, Hao Q, Vandvik PO, Guyatt G, Li J, et al. Pharmacotherapy for adults with overweight and obesity: a systematic review and network meta-analysis of randomised controlled trials. Lancet (London England). 2024;403(10434):e21–31. [DOI] [PubMed] [Google Scholar]
- 44.Park HY, Kim J, Park MY, Chung N, Lim K. Exposure and exercise training in hypoxic conditions as a new obesity therapeutic modality: A mini review. J Obes Metabolic Syndrome. 2018;27(2):93–101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Roels B, Thomas C, Bentley DJ, Mercier J, Hayot M, Millet G. Effects of intermittent hypoxic training on amino and fatty acid oxidative combustion in human permeabilized muscle fibers. J Appl Physiol (Bethesda Md: 1985). 2007;102(1):79–86. [DOI] [PubMed] [Google Scholar]
- 46.Urdampilleta A, González-Muniesa P, Portillo MP, Martínez JA. Usefulness of combining intermittent hypoxia and physical exercise in the treatment of obesity. J Physiol Biochem. 2012;68(2):289–304. [DOI] [PubMed] [Google Scholar]
- 47.Camacho-Cardenosa A, Camacho-Cardenosa M, Burtscher M, Martínez-Guardado I, Timon R, Brazo-Sayavera J, et al. High-Intensity interval training in Normobaric hypoxia leads to greater body fat loss in overweight/obese women than High-Intensity interval training in normoxia. Front Physiol. 2018;9:60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Michèle G-M. Alexandra, grosfeld, tarik, issad. LeptinIs a Hypoxia-Inducible gene. Obes Res. 2012;10(8):856. [DOI] [PubMed] [Google Scholar]
- 49.Bailey DP, Smith LR, Chrismas BC, Taylor L, Stensel DJ, Deighton K, et al. Appetite and gut hormone responses to moderate-intensity continuous exercise versus high-intensity interval exercise, in normoxic and hypoxic conditions. Appetite. 2015;89:237–45. [DOI] [PubMed] [Google Scholar]
- 50.Debevec T. Hypoxia-Related hormonal appetite modulation in humans during rest and exercise: mini review. Front Physiol. 2017;8:366. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Cayanan EA, Keating SE, O’Connor HT, Sainsbury A, Johnson NA. Continuous exercise but not high intensity interval training improves fat distribution in overweight adults. J Sci Med Sport. 2014;2014(2):834865. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Willis LH, Slentz CA, Bateman LA, Shields AT, Piner LW, Bales CW, et al. Effects of aerobic and/or resistance training on body mass and fat mass in overweight or obese adults. J Appl Physiol. 2012;113(12):1831–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Veldhuis JD, Roemmich JN, Richmond EJ, Rogol AD, Lovejoy JC, Sheffield-Moore M, et al. Endocrine control of body composition in infancy, childhood, and puberty. Endocr Rev. 2005;26(1):114–46. [DOI] [PubMed] [Google Scholar]
- 54.Millet GP, Debevec T, Brocherie F, Malatesta D, Girard O. Therapeutic use of exercising in hypoxia: promises and limitations. Front Physiol. 2016;7:224. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Zoll J, Ponsot E, Dufour S, Doutreleau S, Ventura-Clapier R, Vogt M et al. Exercise training in normobaric hypoxia in endurance runners. III. Muscular adjustments of selected gene transcripts. Journal of applied physiology (Bethesda, Md: 1985). 2006;100(4):1258-66. [DOI] [PubMed]
- 56.Heinonen IH, Boushel R, Kalliokoski KK. The circulatory and metabolic responses to hypoxia in Humans - With special reference to adipose tissue physiology and obesity. Front Endocrinol (Lausanne). 2016;7:116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Michael PC. Insulin action and resistance in obesity and type 2 diabetes. Nat Med. 2017 0701. [DOI] [PMC free article] [PubMed]
- 58.Montero D, Lundby C. Effects of exercise training in hypoxia versus normoxia on vascular health. Sports Med. 2016;46(11):1725–36. [DOI] [PubMed] [Google Scholar]
- 59.Mann S, Beedie C, Jimenez A. Differential effects of aerobic exercise, resistance training and combined exercise modalities on cholesterol and the lipid profile: review, synthesis and recommendations. Sports Med. 2014;44(2):211–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60.Bozzini CE, Alippi RM, Barcelo AC. Mandibular growth retardation in growing rats chronically exposed to hypobaria. J Dent Res. 1987;66(1):65–6. [DOI] [PubMed] [Google Scholar]
- 61.Pandit A. The physiologic basis of high-altitude diseases. Ann Intern Med. 2005;142(7):591. author reply 2. [DOI] [PubMed] [Google Scholar]
- 62.Sandoval DA, Matt KS. Gender differences in the endocrine and metabolic responses to hypoxic exercise. J Appl Physiol (1985). 2002;92(2):504–12. [DOI] [PubMed] [Google Scholar]
- 63.Garvican-Lewis LA, Sharpe K, Gore CJ. Time for a new metric for hypoxic dose? J Appl Physiol (1985). 2016;121(1):352–5. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are available from the corresponding author upon reasonable request.