|| Checking for direct PDF access through Ovid
An individual participant data meta-analysis was conducted in the data of 14 673 Japanese participants without a history of cardiovascular disease (CVD) to examine the association of the brachial-ankle pulse wave velocity (baPWV) with the risk of development of CVD. During the average 6.4-year follow-up period, 687 participants died and 735 developed cardiovascular events. A higher baPWV was significantly associated with a higher risk of CVD, even after adjustments for conventional risk factors (P for trend <0.001). When the baPWV values were classified into quintiles, the multivariable-adjusted hazard ratio for CVD increased significantly as the baPWV quintile increased. The hazard ratio in the subjects with baPWV values in quintile 5 versus that in those with the values in quintile 1 was 3.50 (2.14–5.74; P<0.001). Every 1 SD increase of the baPWV was associated with a 1.19-fold (1.10–1.29; P<0.001) increase in the risk of CVD. Moreover, addition of baPWV to a model incorporating the Framingham risk score significantly increased the C statistics from 0.8026 to 0.8131 (P<0.001) and also improved the category-free net reclassification (0.247; P<0.001). The present meta-analysis clearly established baPWV as an independent predictor of the risk of development of CVD in Japanese subjects without preexisting CVD. Thus, measurement of the baPWV could enhance the efficacy of prediction of the risk of development of CVD over that of the Framingham risk score, which is based on the traditional cardiovascular risk factors.