Data on the association between serum nitrate/nitrite and cardiovascular disease (CVD) is controversial; the aim of this study was to examine the potential utility of serum nitrate/nitrite in prediction of CVD. A total of 2443 adult participants, free from CVD at baseline were included. The probability of developing CVD was estimated by incorporating traditional CVD risk factors into a logistic regression model in the presence of serum nitrate/nitrite. The probability of CVD was calculated using Ln-transformed CVD risk factors as a covariate in the presence of Ln-transformed nitrate/nitrite in the final model. The added value of nitrate/nitrite was estimated using receiver operating characteristic curves and the net reclassification index (NRI). During follow-up, 169 events occurred. The multivariate odds ratio (95% confidence interval) for each 1 unit of increase in the Ln-transformed traditional risk factors and Ln-transformed nitrate/nitrite, were 3.20 (2.67-3.84) and 1.35 (1.01-1.80) for incident CVD, respectively. The areas under the curves of discrimination for models with and without nitrate/nitrite were not significantly different; however, incorporating nitrate/nitrite to the traditional CVD risk model can help appropriately reclassify over 6% of individuals at risk. In conclusion, serum nitrate/nitrite levels were independently associated with incident CVD. Measurement of serum nitrate/nitrite provided information beyond individual data on risk factors and improved prediction of CVD.
Keywords: Biomarker; Cardiovascular disease; Nitrate/nitrite.
Copyright © 2016 Elsevier Inc. All rights reserved.