The Institutional Review Board (IRB) of the Brazilian National Health Council approved the study (# CAAE 02279512.1.0000.0012), under number 025794/2012, and based on the National Law 466/2012, and following the ethical principles of the Declaration of Helsinki. All participants signed the consent form after receiving verbal instruction and before any intervention. The volunteers were seen by a physician who released them to perform vigorous exercise before participating in the study. The volunteers were recruited to the study after checking into one of the public Basic Health Centers of the City of Porto Velho, Rondônia, Brazil.

The intervention was conducted at a public Basic Health Center. The environment was not climate-controlled; however, since the assessments took place in the morning, the conditions were adequate. Measurements obtained with a thermometer and thermo-hygrometer indicated average values of 26–28 °C and approximately 70% relative humidity every day. The region where the study was carried out is characterized by about six months of dry season and six months of rainfall, which contributes to variations in air quality. All procedures followed the guidelines of the Brazilian Unified Health System (SUS), emphasizing accessibility and low-cost interventions.

No volunteer participated in sports or physical training regularly. Metabolic syndrome (MetS) was determined based on the criteria defined by the World Health Organization (WHO) and the National Heart, Lung, and Blood Institute of the United States National Institutes of Health (NIH) (30). Only subjects exhibiting three out of the four criteria described in Table 1 and being 60 years or older were included in the study. Subjects presenting any orthopedic limitation to the exercise training were excluded from the study. Figure 1 shows the consort flow chart for subject's selection.

Forty subjects, 36 females and four males, were enrolled and randomly divided into the control group (no exercise; n = 20) or HIIT (n = 20). Before the study, participants had the opportunity to familiarize themselves with all the procedures on two occasions. First, they visited the laboratory after fasting for 12 h and refraining from caffeine, alcohol, and exercise for 24 h. Body mass, height, and blood pressure were assessed. Baseline blood samples were obtained to determine hematological and biochemical parameters. In the second visit (4 days later), subjects of the HIIT group were familiarized with the HIIT protocol. After the familiarization sessions, baseline measurements of maximal exercise capacity were assessed.

Participants were allocated 1:1 to HIIT or control using computer-generated random numbers and, after, the allocation was conducted in accordance with the result. Due to the nature of the intervention, participants and trainers were not blinded; however, outcome assessors and laboratory staff were blinded to group allocation whenever feasible.

All evaluations were conducted at a public Basic Health Center, between 7:00 and 9:00 a.m. They were instructed to fast for 8–10 h, avoid alcohol for 24 h, and refrain from vigorous physical activity for 24 h prior to each assessment. Subjects were followed up four and eight weeks after initiating the HIIT protocol.

Body weight, waist circumference, and body mass index (BMI), were assessed using the octopolar scale (InBody, model 170, South Corea) (32). Blood pressure was determined following the previously described American Heart Association (AHA) procedures (33). The measurement was always performed on the right arm after 5 min of resting on a chair and using a sphygmomanometer and stethoscope (PA.MED, Brazil). The blood pressure was assessed by a very experienced nurse from the Basic Health Center.

Blood samples were obtained by venipuncture and collected in EDTA-containing tubes for plasma and tubes with gel to separate the serum (BD Vacutainer, Franklin Lakes, NJ, USA). Hematocrit, hemoglobin, mean corpuscular volume, and mean corpuscular hemoglobin were analyzed using the Hemocauto Analyzer 2120® (Siemens, AG, Erlangen, Germany). Blood glucose, glycosylated hemoglobin, and high-density lipoprotein C were assessed with the automatic device Konelab 60i after calibration. The coefficient of variation was 1.2% to red globule count, 0.93% to hemoglobin, and 2.93% to plaquette count. All analysis was conducted in duplicates. All measurements were performed until 60 min after the blood collection.

To determine maximal exercise capacity, volunteers underwent the 6-minute walking test (6MWT). The test was performed outdoors, and participants were instructed to walk at their own pace and alone, covering as much distance as possible within 6 min. They were allowed to stop and rest, if necessary, before resuming (34).

All individuals were advised to wear light clothing and to refrain from exercise prior to the test. The following sex-specific equations were used to estimate cardiorespiratory fitness from the 6-min walk test (6MWT), expressed as estimated VO₂max (mL·kg⁻¹·min⁻¹):Men:estimatedVO2max=(7.57×height[cm])−(5.02×age[years])−(1.76×bodymass[kg])−309Women:estimatedVO2max=(2.11×height[cm])−(2.29×bodymass[kg])−(5.78×age[years])+667.6

The HIIT program involved bodyweight exercises, requiring no specialized equipment besides exercise mats and small step boxes. Exercise sessions lasted 40 min including 5 min warmup and cool-down. Participants were instructed on proper form and safety during a pre-study familiarization session led by certified trainers. Warm-up included dynamic stretching (arm swings, leg swings, trunk rotations) and light calisthenics (marching in place, low intensity jumping jacks). The protocol was standardized across sessions, ensuring consistency in intensity and execution. Each of the 10 exercises was performed once for 60 s with 120 s of passive recovery between exercises. Passive recovery consisted of sitting quietly or standing still.

Although each exercise was performed for 60 s, totaling 20 stimuli per session (five exercises repeated four times), not all exercises were performed in every daily session. Thus, the effective duration of high-intensity effort was estimated to be approximately 15–18 min, representing the sum of the active period and the time during which heart rate remained elevated throughout the recovery phase.

The cooldown included 5 min of breathing and stretching exercises. Participants performed the HIIT sessions three times a week for eight weeks. Adherence to the protocol was monitored weekly using attendance logs, and participants were instructed to maintain their usual diet and lifestyle during the intervention. Additionally, they were advised to refrain from other structured physical activities to isolate the effects of the HIIT intervention.

Subjects performed a standardized warm-up of stretching and calisthenics exercises for 5 min before HIIT sessions. All HIIT sessions were conducted in the following exercise order: a) ten meters back and forth running, b) air squats, c) push-ups, d) sit-ups, e) jumping jacks, f) forward lunges, g) abdominal planks, h) reverse lunges, i) abdominal plank, j) side steps, direction-changing footwork and going up and down steps (25 cm). A 5-min cooldown was carried out at the end of the session with easy stretching and breathing exercises.

Exercise intensity during the HIIT sessions was monitored using heart rate measurements obtained every 5 s with a ROSSMAX finger oximeter. Maximal heart rate (HRmax) was estimated for each participant using the age-predicted formula (HRmax = 220 − age). Target heart rate zones for the HIIT protocol were defined as percentages of HRmax, with high-intensity intervals executed at >90% of HRmax to ensure vigorous effort consistent with published HIIT guidelines. This proceeding ensured that participants exercised within predetermined high-intensity zones, providing objective control of cardiovascular stress throughout the intervention.

The triglyceride–glucose (TyG) index was calculated using the following validated formula:TyG=ln[(fastingtriglycerides(mg/dL)×fastingglucose(mg/dL))/2]This formula yields a dimensionless score, with higher values indicating greater insulin resistance. A cutoff value of TyG ≥4.5 has been proposed to identify insulin resistance in adults. All values reported in this study were expressed consistently using this definition (35, 36).

To determine the effects of HIT on MetS parameters, we assessed the metabolic syndrome severity score (MetS-Z) (37). The MetS-Z is based on data from the National Health and Nutrition Examination Survey (NHANES) collected between 1999 and 2010, which represents a broad cross-section of the U.S. population. MetS-Z functions as a standardized z-score, normally distributed with a mean of 0 and a standard deviation of 1. In the context of this study, it indicates how many standard deviations a participant's MetS-Z deviates from the NHANES population average.

Data are presented as mean ± standard deviation. Data normality was assessed using the Shapiro–Wilk test. Two-tail Student's T-test was used to compare means of control and HIIT groups before exercise (Baseline measurements; Table 2). Two-way ANOVA followed by Sìdak post-hoc test was used to determine differences between means within (over time) and between groups (Figures 1–3). Because an a priori sample size calculation was not performed, we report effect sizes and 95% confidence intervals for all primary outcomes. Statistical significance was set as p < 0.05 and assessed using GraphPad Prism 9.

Anthropometric, physiological, and biochemical parameters were comparable between groups at baseline (week 0) (body mass: control vs. HIIT mean difference = 1.13 kg, 95% CI: −6.31 to 8.58; p = 0.7636; BMI: control vs. HIIT mean difference = 2.37 kg·m⁻², 95% CI: −2.06 to 6.80; p = 0.2917; waist circumference: control vs. HIIT mean difference = 6.43 cm, 95% CI: 0.64–12.23; p = 0.0299). Detailed baseline data are presented in Table 2.

Eight weeks of HIIT did not decrease body weight or body mass index [no group × time interaction for body mass: F(2,114) = 0.15, p = 0.8616; and BMI: F(2,114) = 0.35, p = 0.7028] (Figures 2A,B). However, waist circumference was reduced at the end of the HIIT program [significant main group effect: F(1,114) = 4.84, p = 0.0299; mean difference = −6.43 cm, 95% CI: −12.23 to −0.64] (Figure 2C).

The HIIT intervention resulted in significant improvements in cardiovascular parameters and exercise capacity. Systolic blood pressure, resting heart rate, double-product, and VO2max were improved significantly in the HIIT group compared to sedentary subjects at four weeks of training (SBP mean difference = −8.81 mmHg, 95% CI: −19.48 to 1.86; p = 0.203; VO₂max mean difference = 4.70 mL·kg⁻¹·min⁻¹, 95% CI: 1.48–7.92; p = 0.0004) (Figure 3). After 8 weeks, the HIIT group showed reductions in systolic blood pressure by 9.4% (−14.90 mmHg, 95% CI: −25.57 to −4.23; p = 0.0009), diastolic blood pressure by 6.2% (−7.41 mmHg, 95% CI: −14.33 to −0.49; p = 0.0260), and mean arterial pressure had and tendency to decrease (−9.91 mmHg, 95% CI: −21.16 to 1.34; p = 0.134) (Figures 3A–C). While HIIT decreased resting heart rate (−11.80 bpm, 95% CI: −22.85 to −0.75; p = 0.0269) and the double product (−2,224 units, 95% CI: −3,251 to −1,197; p < 0.0001), it increased VO2max by 20% at the end of the intervention (+6.50 mL·kg⁻¹·min⁻¹, 95% CI: 3.28–9.72; p < 0.0001) (Figures 3D–F). In contrast, no changes were observed in the control group (Figure 3).

HIIT induced moderate alterations in hematological markers. Hematocrit decreased by 4.9% (mean difference = −2.20%, 95% CI: −3.28 to −1.12; p < 0.0001) while hemoglobin increased by 7.7% (mean difference = +3.04 g·dL⁻¹, 95% CI: 2.14–3.95; p < 0.0001) in the HIIT group after 8 weeks (Figures 4A,B). Mean corpuscular volume (mean difference = −3.72 fL, 95% CI: −5.13 to −2.32; p < 0.0001) and mean corpuscular hemoglobin (mean difference = −2.15 pg, 95% CI: −3.36 to −0.94; p < 0.0001) also declined significantly (Figures 4C,D). Platelet counts remained stable across both groups (HIIT vs. control at 8 weeks: mean difference = −28.32 × 10³ µL⁻¹, 95% CI: −97.10 to 40.46; p = 0.976) (Figure 4E).

Metabolic improvements were evident as early as 4 weeks in the HIIT group. Plasma glucose, HbA1c, and triglycerides were significantly lower in the HIIT group compared to sedentary subjects and baseline measurements (glucose at 4 weeks: mean difference = −30.4 mg·dL⁻¹ vs. control, 95% CI: −53.86 to −6.94; p = 0.0027; HbA1c at 4 weeks: mean difference = −3.10%, 95% CI: −5.16 to −1.04; p = 0.0002; triglycerides at 4 weeks: mean difference = −39.3 mg·dL⁻¹, 95% CI: −74.57 to −4.03; p = 0.0173) (Figures 5A,C). At 8 weeks, fasting glucose levels decreased by 32% (−49.8 mg·dL⁻¹, 95% CI: −73.26 to −26.34; p < 0.0001), HbA1c levels by 35% (−3.30%, 95% CI: −5.36 to −1.24; p < 0.0001), and triglycerides by 39% (−70.8 mg·dL⁻¹, 95% CI: −106.1 to −35.53; p < 0.0001) (Figures 5A,C). According to Guerrero-Romero coworkers (35, 36), all subjects enrolled in our study were insulin resistant since the TyG index was above the 4.5 cutoff level. Eight weeks of HIIT decreased the TyG index by 12%, reaching values close to the cutoff [mean difference = −0.61, 95% CI: −1.06 to −0.16; p = 0.0016; from (RED) 6.63 ± 0.7 to 6.02 ± 0.3] (Figure 5D). In contrast, glucose increased in the control group compared to baseline levels (+25.6 mg·dL⁻¹, 95% CI: 2.14–49.06; p = 0.0216) (Figure 5A). HbA1c, triglycerides, and TyG index remained the same in the control group during the study.

At baseline, both the control and HIIT groups presented comparable MetS-Z scores (mean ± SD: 1.63 ± 0.59 vs. 1.59 ± 0.50, respectively), indicating elevated cardiometabolic risk in both groups prior to intervention (baseline control vs. HIIT: mean difference = 0.04, 95% CI: −0.45 to 0.53; p > 0.9999). Over time, while the MetS-Z did not change in the control group, it progressively decreased in the HIIT after four and eight weeks (Figure 6). The MetS-Z went from 1.59 ± 0.50 to 1.00 ± 0.27 at 4 weeks (mean difference vs. baseline = −0.59, 95% CI: −1.08 to −0.10; p = 0.0071) and further to 0.26 ± 0.093 at 8 weeks (mean difference vs. baseline = −1.33, 95% CI: −1.82 to −0.84; p < 0.0001) (Figure 6), representing a 84% reduction. At both time points, MetS-Z values were significantly lower in the HIIT group compared to controls (4 weeks: mean difference = −0.69, 95% CI: −1.18 to −0.20; p = 0.0008; 8 weeks: mean difference = −1.77, 95% CI: −2.26 to −1.28; p < 0.0001).