A Vascular Aging Index as Independent Predictor of Cardiovascular Events and Total Mortality in an Elderly Urban Population

Abstract

The morphology and function of the arteries can be directly measured using different established methods. This prospective cohort study aimed to translate 2 of these, aortic pulse wave velocity (aPWV) and carotid intima–media thickness (cIMT), into a combined Vascular Aging Index (VAI) and then evaluate the predictive power of aPWV, cIMT, and VAI. Patients (n = 2718) were included from the cardiovascular arm of the Malmö Diet and Cancer Study (median age 71.9 years, 62.2% females). Total follow-up time was 16 448 person-years and a composite cardiovascular disease (CVD) end point was used. Cox regressions yielded adjusted hazard ratios (95% confidence interval) per 1 standard deviation increment of loge aPWV, loge cIMT, and loge VAI of 1.25 (1.08-1.45, P = .003), 1.27 (1.13-1.44, P < .001), and 1.45 (1.26-1.68, P < .001), respectively. The C-statistics increased from 0.714 to 0.734 when adding aPWV and cIMT to a model of conventional risk factors. Net Reclassification Index also showed a significant ( P < .001) improvement for the classification of event-free patients and no change for patients with events. A VAI based on aPWV and cIMT had a good predictive performance. Used together, aPWV and cIMT incrementally and significantly improve the prediction of CVD events by correctly down-adjusting the predicted risk for noncases.