Impacts of Concurrent Training on Heart Rate Variability, Hemodynamic Parameters, and Behavioral Aspects of Older Adults

Population aging is a global phenomenon, and in Brazil, older adults are estimated to account for 37.8% of the population by 2070. This demographic transition requires interventions aimed at mitigating the impact of comorbidities associated with senescence, such as cardiovascular diseases (CVDs), functional decline, and metabolic disorders.

In this context, heart rate variability (HRV) stands out as a non-invasive marker of autonomic health, with different indices and markers exhibiting distinct patterns in cardiac modulation and potentially being correlated with increased risk of CVD, all-cause mortality, and cognitive deterioration.

Concurrent training (CT), which integrates aerobic and resistance exercises, has demonstrated promising effects on autonomic nervous system (ANS) modulation, with increased parasympathetic activity (PA). However, evidence regarding its combined effects on hemodynamic parameters (such as systemic blood pressure, heart rate, and rate-pressure product), markers of psychological stress, and quality of life-including perceived stress and sleep quality-in older adults with different comorbidity profiles remains limited.

This study aimed to analyze and compare the responses to concurrent, endurance, and resistance training after 12 weeks regarding resting cardiac autonomic modulation, systolic and diastolic blood pressure, resting heart rate, rate-pressure product, perceived stress, and sleep quality in older adults. The study was characterized as a prospective, non-controlled clinical trial following the Transparent Reporting of Evaluations with Nonrandomized Designs (TREND) guidelines. The present study was registered on the ClinicalTrials.gov platform under protocol ID 88413825.2.0000.5168, ensuring transparency, traceability, and protocol reliability. The interventions were conducted at the Resistance Exercise and Health Laboratory (LERES) of the University of the State of Pará (UEPA), Campus III. The intervention period lasted 12 weeks, the duration specified in the study protocol.

Participants included older adults aged 60 years or older with hypertension, diabetes, osteopenia, heart disease, or metabolic syndrome, whether clinically stable or not, provided that participants who were clinically unstable were deemed fit to exercise at the time of the training session, as well as participants without any comorbidities. The exclusion criteria were participants with diseases or injuries that prevented locomotion or exercise; participants without medical clearance for physical activity; participants using beta-blocking medications, as these medications directly influence the HR/HRV relationship and may therefore constitute a physiological confounding factor; and participants who engaged in other physical activities or sports outside the study. Participant recruitment and outreach were conducted through social media advertisements, posters displayed on university campuses, and a waiting list of individuals awaiting vacancies in other projects at the same laboratory. The sample was allocated by convenience according to the availability of each group. The first 23 participants were allocated to the RT group because it was the first intervention group to be implemented. The second group to be implemented was the CT group; therefore, newly recruited participants were allocated to this group. The third and final group was the ET group, to which the remaining participants were allocated. The ET group was conducted at a different time from the other groups, which may have influenced its final sample size. However, HRV assessments were performed at the same time as those of the other groups (2:00 p.m. to 6:00 p.m.).

Sample size was estimated using G*Power software (version 3.1.9.7). The primary outcome used to support the calculation was HRV, considering comparisons among three intervention groups: resistance training, endurance training, and concurrent training. A significance level of 5% (α = 0.05), statistical power of 80% (1 - β = 0.80), effect size of 0.40, and three comparison groups were adopted. Based on these parameters, a minimum sample of 64 participants was estimated. To minimize the impact of potential sample losses and dropouts throughout follow-up, 20% was added to the initial sample, resulting in a final planned sample of 78 participants, distributed approximately as 26 individuals per group. The groups performed training sessions three times per week. The Endurance Training group performed an adapted functional circuit consisting exclusively of exercises predominantly focused on cardiorespiratory demands in a controlled environment. Effort was monitored using the modified Borg Rating of Perceived Exertion (CR10) scale. The workload was increased or modified whenever the participant indicated that the current workload was not producing the intended effort. The Borg scale score was standardized to 7-9, ranging from "very light" to "very hard,"

During the first two weeks, participants in the Endurance Training group exercised for 15-20 minutes. From that period onward, the duration increased to 25-30 minutes. The exercises used in the functional circuit were forward walking between cones, stationary marching, lateral movement between cones, and adapted jumping jacks, totaling four exercises. A 5-10-minute warm-up was performed before Endurance Training in all sessions.

Progression in this modality was achieved through the RPE scale, targeting a score of 7-9, combined with the duration of the Endurance exercise sets, which ranged from 20 to 40 seconds, and rest intervals between sets ranging from 15 to 30 seconds. This duration was considered sufficient to induce cardiometabolic changes during the intervention period. During the first two weeks, the exercise stations lasted 20 seconds, followed by 15 seconds of rest. From the third to the sixth week, stations lasted 25 seconds with 15 seconds of rest. From the seventh to the tenth week, stations lasted 30 seconds with 15 seconds of rest. Finally, during the last two weeks, stations lasted 40 seconds with 20 seconds of rest. The rest interval could be increased to 30 seconds if the RPE score exceeded 18 points.

The Resistance Training group performed three weekly training sessions. The sessions consisted of six exercises involving all major muscle groups, using resistance-training machines and free weights: free-weight squat, leg press, leg curl, vertical chest press, unilateral row, and abdominal exercises. Participants performed two sets of 8-12 repetitions using the submaximal repetition method combined with the RPE scale to control intensity, effort, and progression. Rest intervals were at least 60 seconds for healthy participants and 90 seconds for participants with comorbidities, particularly those with hypertension and coronary artery disease, as these conditions may require longer rest intervals to allow heart rate stabilization.

Progression was based on each participant's execution and, when necessary, the load was increased by 2% to 10%, assuming that the participant was performing one or more repetitions beyond the desired range. A 5-10-minute warm-up was performed before the main training session. All participants were instructed to perform repetitions in a moderate and controlled manner. Participants were supervised by physical education professionals and undergraduate students.

The protocol followed a structured progression of resistance exercise.

The Concurrent Training group performed three weekly training sessions, with Endurance Training performed immediately after Resistance Training. The Resistance Training sessions in the CT group followed the same protocol used in the Resistance Training group, consisting of six exercises using resistance-training machines and free weights: free-weight squat, leg press, leg curl, vertical chest press, unilateral row, and abdominal exercises.

Participants performed two sets of 8-12 repetitions using the submaximal repetition method combined with the RPE scale to control intensity, effort, and progression, with rest intervals of at least 60 seconds for healthy participants and 90 seconds for participants with hypertension and CVDs. Progression was based on each participant's execution and, when necessary, the load was increased by 2% to 10%, assuming that the participant was performing one or more repetitions beyond the desired range. A 5-10-minute warm-up was performed before the main training session.

The Endurance Training sessions in the CT group differed from those of the isolated Endurance Training group. Endurance Training was performed using ergometers (bicycle and elliptical trainer) and walking in a controlled environment. The Endurance Training sessions in the CT group lasted 15-25 minutes, and effort and intensity were controlled and adjusted using the RPE scale, maintaining an effort level between 7 and 9. Participants in the CT group received the full combined training dose.

An anamnesis (Appendix A) and the International Physical Activity Questionnaire (IPAQ), adapted for older adults (Appendix A), were administered to assess participants' physical activity levels and the presence of comorbidities. The Brazilian Portuguese version of the Pittsburgh Sleep Quality Index (PSQI) (Appendix B) was also administered to assess sleep quality, along with the Perceived Stress Scale-10 (PSS-10) to assess stress levels (Appendix C).

The PSS-10 score ranges from 0 to 40 points, combining positively and negatively worded items through reverse scoring. It measures the extent to which participants perceived situations in their lives as unpredictable, uncontrollable, and overwhelming during the previous month.

Participants underwent weight and height measurements to calculate body mass index (BMI). The equipment used for data collection included a CESCORF inelastic measuring tape, a stadiometer integrated into an anthropometric scale, and a G-Tech digital scale.

To assess cardiac autonomic modulation, a Polar H10 heart rate sensor was used to record resting heart rate variability for 15 minutes. This analysis was based on the recording of R-R intervals, in milliseconds (ms), which were transmitted, recorded, and stored on a digital device for subsequent processing.

Before the test, participants received the necessary instructions and recommendations: no caffeine consumption or moderate-to-vigorous physical exertion during the preceding 24 hours; avoidance of tobacco and alcohol during the pre-test period; control of sleep and medications containing components that could affect autonomic modulation were also recommended as test-performance criteria.

The test was conducted in a controlled environment at room temperature, between 2:00 p.m. and 6:00 p.m. Before the test, participants rested for at least 15 minutes. HRV data collection was standardized as follows: participants were instructed to undergo the test in the supine position, with spontaneous breathing, for 12 minutes.

The segment selected from these 12 minutes for analysis depended on visual inspection performed by the researcher using Kubios HRV software. On average, the initial 25% and final 25% of the recording were removed, and the remaining 50% was analyzed. The time of day used for reassessments followed the same standardized schedule, from 2:00 p.m. to 6:00 p.m.

In addition to this assessment, the Stroop Color Test was administered as a mental stress test involving the presentation of words in different orders and colors. The Stroop Color Test was performed immediately after the first HRV assessment, followed by a second HRV assessment, with the purpose of comparing the results and investigating the potential influence of mental stress on cardiac autonomic modulation.

Data corresponding to the R-R intervals obtained from the heart rate sensor were exported through the Elite HRV application, version 5.5.8, and stored in electronic files for subsequent analysis using Kubios HRV Scientific Lite software, version 4.1.2.1.

Initially, the recordings underwent visual inspection to identify inadequate markings and artifacts that could compromise signal quality. When necessary, the correction resources available in the software itself were used. In addition, for spectral analysis, the software applied cubic spline interpolation to generate equidistant time series from the R-R intervals, as well as a detrending procedure recommended to reduce the influence of very slow signal components and increase sensitivity to short-term autonomic modulation.

Following this processing, HRV time series were analyzed in the time and frequency domains. In the time domain, the following indices were used: mean normal-to-normal intervals (MeanNN), the standard deviation of successive normal R-R intervals (SDNN), and parasympathetic activity (RMSSD).

In the frequency domain, absolute spectral power was obtained in the previously established bands: low frequency (LF: 0.04-0.15 Hz) and high frequency (HF: 0.15-0.40 Hz). These frequency bands allow interpretation of cardiac autonomic behavior based on the distribution of spectral power.

Blood pressure and heart rate were monitored daily using the office measurement technique, following the recommendations of the Brazilian Society of Cardiology (SBC). Measurements were performed on the bare arm, supported at heart level, with the palm facing upward; in a quiet environment at room temperature; without prior smoking, eating, or caffeine intake; after 5 minutes of rest before measurement; with the participant seated, back supported against the chair, legs uncrossed, and feet supported on the floor; with an empty bladder and without the need to urinate; and without speaking.

The device used to measure blood pressure and heart rate was a Microlife MAM-PC BP3AC1PC automatic tabletop blood pressure monitor.

Statistical analyses were performed using IBM SPSS Statistics, version 20.0, adopting a significance level of 5%. Data normality was assessed using the Shapiro-Wilk test, and homogeneity of variances was assessed using Levene's test. Data were presented as mean ± standard deviation, median [Q1-Q3], or absolute and relative frequencies, as appropriate.

Baseline comparisons between groups were performed using one-way ANOVA, Welch's test, or the Kruskal-Wallis test, according to the assumptions. Tukey's or Games-Howell post hoc tests were used when necessary.

For post-intervention outcomes, analyses of covariance (ANCOVA) were performed, considering group as the factor and the respective baseline value as the covariate. Homogeneity of regression slopes was assessed through the group × baseline value interaction, and model adequacy was additionally examined through residual analysis.

Pairwise comparisons of adjusted means were performed with Bonferroni correction, and partial eta squared (ηp²) was reported as a measure of effect size.

For HRV, RMSSD and HF were transformed using the natural logarithm for inferential analyses. MeanNN, SDNN, lnRMSSD, and lnHF were analyzed using ANCOVA, considering the respective baseline value as the covariate. The p-values of the four HRV indices were adjusted using the Holm method.

For sleep quality, due to violation of the homogeneity of regression slopes assumption in ANCOVA, pre-post changes were compared between groups using the Kruskal-Wallis test, considering only participants with complete data. No imputation of missing data was performed. In addition to this test, the pre-post change (ΔPSQI) was calculated for each group and presented as median [Q1-Q3]; this was the variable effectively compared.

A sensitivity analysis was planned for SDNN to assess the influence of extreme observations.

This study was conducted in strict accordance with the ethical guidelines established by Resolution No. 466/12 of the Brazilian National Health Council, which regulates research involving human participants in Brazil. All procedures adopted respected the fundamental principles of bioethics: autonomy, beneficence, non-maleficence, and justice, ensuring the comprehensive protection of participants.

The project received approval from the Research Ethics Committee (REC) 5168 - University of the State of Pará - Campus XII - Tapajós - UEPA, under CAAE No. 88413825.2.0000.5168 (Appendix E).

Participation in the study was voluntary, and all individuals from the three groups received clear and complete information regarding the objectives, procedures, expected benefits, and potential risks, as established in the Informed Consent Form (ICF).

Participants were free to withdraw from the study at any time without any prejudice to the care or assistance they might receive. To ensure confidentiality and privacy, collected data were stored anonymously and identified only by codes, accessible exclusively to the researchers responsible for the study. In addition, all records were maintained in a secure environment, in accordance with the requirements of the Brazilian General Data Protection Law (LGPD, Law No. 13.709/18).