Chronic aerobic exercise reduces the aortic age of an elderly cohort

Central arterial stiffness is a well-established morbidity and mortality marker, particularly in the context of cardiovascular disease (Zieman et al. 2005). Increased arterial stiffness is a physiological aspect of senescence, and is exacerbated by non-communicable diseases such as hypertension, atherosclerosis and diabetes mellitus. Structural alterations in large arterial vessel walls that lead to stiffness include overproduction of collagen, underproduction/breakdown of elastin, medial smooth muscle hypertrophy, increased intimal permeability, migration of monocytes/macrophages, and fibrosis of the adventitia (Zieman et al. 2005). Regular lifelong exercise (i.e. more than 25 years) is a primary preventive strategy known to curtail the rate of arterial stiffening, as described in a recent article in The Journal of Physiology by Shibata et al. (2018). A previous cross-sectional study by Shibata and Levine showed that senior (65+ years) masters athletes have more compliant arteries compared to their sedentary yet healthy counterparts in the control group (Shibata & Levine, 2011). However, the magnitude and duration of exercise that masters athletes participate in is not practical for most. To better understand arterial stiffness in the elderly cohort, Shibata et al. (2018) conducted a cross-sectional exercise study to elucidate whether lower levels of exercise still have similar effects on arterial compliance. Healthy seniors recruited by Shibata et al. (2018) were stratified into four groups based on self-reported exercise levels over at least 20 years: (1) a healthy sedentary group (n = 27) that participated in one exercise session or less per week, (2) a casual group (n = 25) that exercised for two to three sessions per week, (3) a committed group (n = 25) that performed four to five exercise sessions per week (meeting current physical activity recommendations), and (4) a masters group (n = 25) that performed near-daily exercise. Shibata et al. (2018) then measured and calculated multiple surrogate indices to assess both central and peripheral arterial stiffness for each participant. The authors adopted a previously validated mathematical algorithm (Modelflow aortic age) to determine biological aortic age, an index of central arterial stiffness that is independent of blood pressure, calculated from stroke volume (SV) measured using the non-invasive acetylene method, a central blood pressure waveform extrapolated from a finger waveform, and chronological age (Shibata & Levine, 2011). Both Doppler ultrasonography and applanation tonometry (SphygmoCor MX) were used to measure central (carotid–femoral) and upper limb (carotid–radial) pulse wave velocity (PWV), with Doppler ultrasonography additionally used to assess lower limb (femoral–dorsal) PWV. Furthermore, ascending and descending aortic β-stiffness and strain were calculated using aortic cross-sectional areas using thoracic magnetic resonance imaging and brachial cuff blood pressure measurements. Right common carotid β-stiffness index and distensibility coefficient measurements were assessed using carotid pressures derived primarily from brachial applanation tonometry and carotid cross-sectional areas measured with high-resolution ultrasound. Additionally, carotid augmentation index was mathematically derived from radial artery waveform via applanation tonometry, which was then normalized to a heart rate of 75 beats min−1. Finally, total arterial compliance index (TACi), effective arterial elastance index (Eai), and total peripheral resistance index (TPRi) were all calculated with stroke volume index (SVi), which is SV from the acetylene method corrected for participants’ individual body surface area. SVi showed an upward trend with increasing exercise intensity with a significant difference noted between the sedentary and physically active participants. SVi-related measurements showed analogous results: TACi was significantly lower in the committed and masters group and Eai was significantly lower in all exercise groups when compared to control. Additionally, the carotid β-stiffness index and distensibility coefficient were significantly higher and lower, respectively, in the sedentary group than all exercise groups with no significant difference among exercise groups, suggesting that even a casual commitment to exercise over a lifetime has similar cardioprotective effects to lifetime intense exercise. Doppler ultrasonography PWV measurements showed significant reductions in a trend-like fashion across all exercise groups compared to the sedentary group, and calculated aortic age showed a clear declining trend with higher exercise levels; committed and masters cohorts had mean aortic ages that were 11 and 24 years younger than that of the sedentary cohort, respectively (P = 0.066 and P < 0.001). A few subject-related weaknesses became evident in relation to participant recruitment and characteristics. Firstly, Shibata et al. (2018) defined exercise sessions loosely as at least 30 min of aerobic exercise, without strictly defining type, intensity or duration. Previous work by Guimaraes et al. (2010) found that 40 min exercise sessions alternating between 50% and 80% of participants’ reserve heart rate had favourable effects on arterial stiffness, while sessions at a continuous 60% of heart rate reserve had none (Guimaraes et al. 2010). Therefore, the uncontrolled characteristics of exercise in the present study may act as confounding variables. Secondly, researchers intentionally enriched their masters athletes group with top masters contestants (top 10th–15th percentile) of local and regional endurance events, further augmenting the already exceptionally active and healthy group. Thirdly, groups had varying male-to-female (M:F) ratios: the sedentary group had a M:F ratio of 1.25, while that of the committed group was 4. Moreover, out of 102 participants, 100 of them were Caucasian, which is not representative of the general population. Sex- and ethnicity-related differences in atherosclerosis and arterial stiffness are well documented in the literature; discrepant M:F ratios between groups and the overwhelmingly Caucasian cohort may have compromised comparability between groups, and generalizability to the public, respectively. Other limitations include the significant differences in body mass index and body fat percentages between groups (both decreasing with higher exercise levels). It seems reasonable to assume that more active people are leaner than their sedentary counterparts, but such variations are still considered to be confounding variables. Another undiscussed weakness of the Shibata et al. (2018) study was the non-significant central PWV results yielded by applanation tonometry using SphygmoCor MX, compared to the Doppler method. This is particularly alarming, as SphygmoCor MX tonometry is one of the most common and well-validated systems used in this endeavour (Rajzer et al. 2008). Nevertheless, Shibata et al. (2018) have shown that the results of Modelflow aortic age and central PWV are largely analogous and parallel. Comparable conclusions drawn from both methods for the effect of chronic exercise on central arterial stiffness pose as a major strength, intrinsically validating both methods. An additional technical strength pertains to the way by which Shibata et al. (2018) elucidated the Modelflow aortic age. The Modelflow technique requires inputs of actual subject SV, which may be measured either invasively using the Fick or thermodilution method, or non-invasively using various rebreathing methods. Shibata et al. (2018) opted for the use of the non-invasive acetylene rebreathing technique which, compared to the gold standard invasive Fick method, waives the need for arterial catheterization (Liu et al. 1997). Secondly, many studies, including that of Liu et al. (1997), found that measurements by both methods were almost identical, validating the acetylene rebreathing method as a reliable non-invasive alternative to measure SV (Liu et al. 1997). As such, Shibata et al. (2018) have successfully coupled the acetylene technique to the Modelflow aortic age algorithm, and successfully demonstrated that the resultant aortic age dimensions were consistent with those of applanation tonometry PWV. Central PWV has been described repeatedly as the gold standard for assessing arterial stiffness, with a large body of evidence supporting its reliability in predicting cardiovascular disease (CVD) incidence (Reference Values for Arterial Stiffness’ Collaboration, 2010). While Shibata et al. (2018) have shown the preliminary usefulness of the Modelflow algorithm, its clinical utility and ability to predict CVD incidence has not yet been definitively elucidated. As such, future research should explore correlation between Modelflow aortic age indices and incidence of CVD. To do so, future longitudinal studies could set out to calculate the Modelflow aortic age of participants using the acetylene rebreathing method and elucidate its correlation with surrogate end points such as fatal and non-fatal coronary events, and all-cause mortality, identifying its ability to predict mortality. If results show strong predictive power, then the Modelflow method may rise as a new alternative to PWV. A recent review paper identified numerous interventional studies presenting positive effects of anti-inflammatory and cholesterol-reducing drugs on aortic arterial stiffness (Maki-Petaja & Wilkinson, 2009). Therefore, additional future studies may set out to identify baseline Modelflow aortic ages of subjects with atherosclerosis and, after a short intervention with such therapies, calculate their aortic ages once again. If their aortic ages do in fact decrease, then the Modelflow method would gain validation as not only a mortality predictive tool, but also as a useful method to track the effects of pharmacological therapies. To conclude, Shibata et al. (2018) have further explored and validated a mathematically ingenious algorithm that may in the future show true clinical value in predicting CVD and monitoring therapeutic interventions for the management of atherosclerosis and arterial stiffness. None declared. M.A.I. and T.T. initiated the journal club, were involved in drafting the manuscript, and analysis and interpretation. Z.A. thoroughly critiqued the methods and results section, and assisted in drafting its strengths and weaknesses. All three authors were involved in reviewing and finalizing the manuscript. All authors approve of the final version of this manuscript and agree to be accountable for all aspects of the work. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed. None.