Observational studies have shown that some of the classic CV risk factors, namely hypertension or hypercholesterolemia, become nebulous, or even act in the reverse direction, in the oldest people. We investigated whether in the elderly, increased aortic stiffness was associated with higher mortality risk, before and after adjustments on common geriatric confounders. In a cohort of 331 (86 men) subjects aged >70 years (mean age (± s.d.): 85 ± 7 years), aortic stiffness was assessed by carotid-femoral pulse wave velocity (PWV). Classical CV risk factors were determined simultaneously, in association with inflammation and denutrition parameters. One hundred and ten subjects died during a 2-year follow-up period. In crude analysis, a positive non-significant trend was observed between PWV and mortality risk. Multivariate Cox regression analysis showed that five parameters entered the prediction model: two were positively related to mortality risk, PWV (P = 0.008) and orosomucoide (P = 0.045), and three were related negatively, total cholesterol (P = 0.006), albumin (P = 0.026) and body weight (P = 0.035). Interaction analysis revealed that the effect of PWV on mortality was increased in the presence of renal dysfunction and increased inflammation. In conclusion, although marginally significant in crude analysis, PWV is a powerful determinant of prognosis in the oldest people taking into account inflammation and denutrition.