The study population consisted of healthy volunteers. Subjects were excluded if they had cancer, acute myocardial infarction in the previous 6 months, clinical overt heart failure, liver failure (AST and ALT >50 IU/mL and/or serum bilirubin >3 mg/dL), serum creatinine ≥ 2 mg/dL, secondary hypertension, or thyroid disease (hypo- or hyperthyroidism requiring hormonal treatment). Additional exclusion criteria included a history of large-vessel stroke and atrial fibrillation since their presence interferes with the accuracy of aorta stiffness measurements. After informed consent was provided, all patients underwent a medical history, a clinic visit, and measurements of body composition, heart rate variability (HRV), and arterial aging.

A self-administered questionnaire was also used to collect information on physical activity. The first question in the questionnaire exploring physical activity was “How many days a week do you perform a moderate to vigorous physical activity such as walking at a fast speed?”; the second question exploring physical activity was “On average, how many minutes per week do you dedicate to these activities?”. Every participant provided a clear response to question 1 (how many days per week). However, approximately one-third of the participants in the present study were not able to provide a clear and detailed answer to the second question. Therefore, participants were classified as follows: sedentary, if they reported no physical activity during the week; physically active, if they reported regular physical activity at least once a week.

The adopted definition differs from that adopted by the World Health Organization (WHO), which defines physical activity as “at least 150 min of moderate-intensity physical activity per week” (https://www.who.int/news-room/fact-sheets/detail/physical-activity). Implications are discussed in the limitations of the study.

Height, weight, and waist circumference were determined for all participants. Body mass index (BMI) was calculated as body weight (kg)/height (m²).

A bioelectrical impedance measure was performed using the Tanita MC-780^® (Tanita Europe BV, Amsterdam, Netherlands) multi-frequency segmental body composition analyzer under standardized conditions: ambient temperature between 23 °C and 25 °C, fasting for 3 h, and an empty bladder. The device acquired measurements of total fat mass, body fat percentage, total muscle mass, skeletal muscle mass, and visceral fat, along with total, intracellular, and extracellular water.

Blood pressure was measured in the non-dominant arm with a sphygmomanometer using an appropriately sized cuff after a 10-min rest in a dark, quiet room. The average of the second and third measurements for systolic blood pressure (SBP) and diastolic blood pressure (DBP) was used in the analysis. Pulse pressure was computed as PP= (SBP−DBP); mean BP was computed as mean blood pressure (MBP) = DBP +(PP/3).

Arterial aging was assessed non-invasively using the carotid–femoral PWV.

PWV was automatically determined as foot-to-foot velocity using the validated SphygmoCor^® device (AtCor Medical, Cardiex, Australia). The pulse transit time was determined as the average of 10 consecutive beats. The distance travelled by the pulse wave was measured over the body surface as the distance between the two recording sites (carotid and femoral pulse). The carotid location–sternal notch distance was subtracted from the carotid–femoral site distance. PWV was calculated as the ratio of distance to transit time.

To account for the known influence of distending blood pressure on PWV, PWV was also normalized for MBP (PWV/MBP).

HRV was measured using AtCor Medical HRV software to assess sympathetic/parasympathetic autonomic function. A standard 2-min electrocardiogram recording was performed with the patient in the supine position, maintaining a regular and calm breathing pattern (5-min resting period) in a quiet room.

Spectral analysis of a series of consecutive RR intervals allowed the quantification of sympathetic and vagal influences on the heart. The evaluated parameters were as follows:

– power spectral density at the high-frequency (HF) range (0.15–0.4 Hz);

Vagal activity is the major contributor to the HF component and time-domain parameters, while LF reflects both sympathetic and vagal activities (Ismail et al., 2012).

All analyses were performed using SAS University Edition.

Univariate regression analysis with Pearson’s r was calculated using PROC CORR.

ANCOVA was used to calculate age-adjusted least-square means of study variables. Therefore, data are presented as least-square means ± standard error (SEM), with post hoc Bonferroni group comparisons. PROC ANCOVA was also used to evaluate the interaction between sex and physical activity.

Multivariable models were constructed to identify factors associated with PWV. These models included the following parameters as covariates: age, blood pressure, heart rate, muscular and fat mass, visceral adiposity, LF, and HF.

An analysis of covariance was used to test whether the slope of the association between PWV and autonomic nervous system activity differed in younger (≤50 years) or older (>50 years) subjects. Interaction terms were constructed between autonomic nervous system activity and age group, sex, and physical activity. A significant triple interaction term indicated that the effect of sex and physical activity on arterial stiffness differed in young and older subjects.

A two-sided p-value <0.05 indicated statistical significance.

The study population consisted of 265 volunteers: 152 men with a mean age of 39.3 ± 17.8 years and 113 women with a mean age of 40 ± 16 years.

To better identify possible sex differences in the impact of physical activity on arterial aging and its possible determinants, participants were divided into four subgroups depending on sex and physical activity. Age significantly differed in the four groups; therefore, age-adjusted means were calculated and are illustrated in Table 1.

Physically active men presented a significantly lower PWV than the sex-matched sedentary group (8.2 ± 0.2 versus 9.0 ± 0.3 m/s, p < 0.01), with no differences in blood pressure. Waist circumference, visceral adiposity, and HR were also lower in physically active men than in sedentary men. When compared to sedentary women, physical activity was not associated with differences in PWV. A lower HR was observed in physically active women than in sedentary controls. Both physically active and sedentary men showed higher SBP, waist circumference, and visceral adiposity than paired women.

Univariate regression analyses showed that determinants of PWV differed in men and women. As illustrated in Table 2, blood pressure and adiposity were significantly associated with PWV in both men and women. Conversely, physical activity was negatively associated with PWV in men (the greater the physical activity, the lower the PWV) but did not appear as a significant determinant in women. HR, LF, and HF were significantly and positively associated with PWV in men, but not in women.

To investigate potential sex differences in the association between arterial aging and its possible determinants, multivariable regression models were constructed. They revealed no significant differences in the association between age, blood pressure, fat mass, or visceral adiposity and PWV between men and women, sedentary or physically active (Supplementary Figure S1).

LF (t = 2.67, p = 0.081) and HF (t = 2.51, p = 0.0125) were significantly associated with PWV in the multivariable regression models. However, their association with PWV differed in the four groups (p for interaction = 0.017 for LF and p = 0.027 for HF). As illustrated in Figure 1, the slopes of the association between PWV and autonomic activity, indexed by LF or HF, differed in sedentary or physically active men and women—after controlling for age, blood pressure, HR, and adiposity. In more detail, sedentary men showed a steeper increase in PWV with any increase in LF or HF than physically active men, which showed an association between PWV and LF or HF similar to that observable in women. No significant physical difference in the slope of the association between PWV and LF or HF was observed between active and sedentary women.

Similar findings were observed when waist or fat mass was—alternatively—added or substituted for visceral adiposity in the multivariable regression models.

We also sought to explore whether menopause could impact the abovementioned sex differences. Therefore, we first divided subjects into two age groups: ≤50 and >50 years. Then, an interaction term between age group, sex, physical activity, and the autonomic nervous system was introduced into the multivariable regression models. The interaction term did not reach statistical significance, suggesting that the abovementioned association—as illustrated in Figure 1—did not differ between men and women younger or older than 50 years.

The present study showed that more physical activity was associated with lower arterial aging, indexed as carotid–femoral PWV, for any given autonomic nervous system activity in physically active men compared to sedentary men. However, this association did not differ between sedentary and physically active women.

Future research should aim to understand the biological pathways underlying the effect of physical activity on the negative association between autonomic nervous system activity and arterial aging in men and women. This new knowledge may eventually contribute to a more personalized approach to the prevention of CV events and a better understanding of vascular aging.