Purpose: It is unclear if higher daily step counts are associated with lower arterial stiffness (AS) in older adults. Less is known about the effects of aerobic stepping cadence (steps/minute) independent of daily step counts on AS in older adults. We examined the independent and combined associations of objectively-measured step counts and aerobic stepping cadence withAS among older adults.

Methods: This cross-sectional study included 409 older adults aged ≥65 years (mean age = 72 ± 6, 59% female). AS was derived from carotid-femoral pulse wave velocity (cfPWV) using the SphygmoCor device (SphygmoCor system, AtCor Medical, Sydney, Australia). High AS, an established risk factor of cardiovascular diseases, was defined as cfPWV >10 m/s.. Step counts and aerobic stepping cadence averaged over 7 days were measured with accelerometer-based pedometers (Omron Alvita Optimized pedometer HJ-321, Kyoto, Japan). Odds ratios (ORs) and 95% confidence intervals (CIs) for high AS were calculated among quintiles (fifths) of step counts and five groups of aerobic stepping cadence (slow walkers with 0 aerobic steps and quartiles [fourths] of aerobic stepping cadence for fast walkers). Slow walkers included those participants who accumulated no aerobic stepping cadence data (cadence of >60 steps per minute for more than 10 consecutive minutes). The lowest step count quintile and the slow walkers were considered the reference groups in the corresponding analysis. Logistic regression was used to investigate the independent associations between step counts or aerobic stepping cadence with high AS. Participants were dichotomized as fast/slow walkers (obtaining any steps at >60 steps/minute or not) or active/inactive (≥5,000 steps/day or not) for a joint analysis. Covariates were sex, age, body mass index, smoking, heavy drinking, diabetes, hyperlipidemia, hypertension medication, systolic blood pressure, and step counts or aerobic stepping cadence in respective analyses.

Results: Eighty-six (21%) older adults were identified as having high AS. Compared to the lowest quintile of step counts, ORs (95% CIs) of having high AS were 0.47 (0.23-0.99), 0.38 (0.18-0.82), 0.42 (0.19-0.93), 0.52 (0.22-1.11) for second, third, fourth and fifth quintile, respectively, after controlling for age and sex. After further adjustment for comorbidities, lifestyle factors, and aerobic stepping cadence, the second and third quintile of step counts still remained significant (OR, 0.38 [95% CI 0.16-0.91] and OR, 0.38 [95% CI 0.15-0.97]). Compared to those with 0 steps/min aerobic stepping cadence (slow walking), ORs (95% CIs) of having high AS were 0.50 (0.21-1.19), 0.47 (0.18-1.20), 0.46 (0.18-1.16), and 0.42 (0.16-1.09) for ascending groups of fast walking aerobic stepping cadence after adjustment for age, sex, comorbidities, lifestyle factors, and step counts. In the joint analysis, compared to Inactive & Slow walkers, ORs (95% CIs) of having high AS were 1.16 (0.31-4.34), 0.43 (0.20-0.95), and 0.48 (0.23-0.95) for Active & Slow walkers, Inactive & Fast walkers, and Active & Fast walkers, respectively.

Conclusion: Higher total daily step counts appeared to be associated with a lower prevalence of high AS status among older adults after adjusting for possible confounders including aerobic stepping cadence in logistic regression analyses. Although aerobic stepping cadence was not significant after adjusting for possible confounders including total daily step counts, this study also suggests a possible association between higher aerobic stepping cadence and lower prevalence of high AS with ORs of <1.00 in all fast walkers (≥1 aerobic steps) compared to slower walkers (0 aerobic steps).