What predicts the occurrence of the metabolic syndrome in a population‐based cohort of adult healthy subjects?

Background Metabolic syndrome (MS), the concurrence of hyperglycaemia, dyslipidaemia, hypertension and visceral obesity, increases cardiovascular risk and mortality. Predictors of MS were previously evaluated in patients without the full syndrome, but with some of its traits. This might confound the resulting associations. Methods The relationship between baseline variables and MS development was evaluated in healthy middle‐aged subjects without any MS component at baseline, over a 4.5‐year follow‐up. Results From a population‐based cohort of 1658 subjects, 241 individuals showed no MS components and 201 (83.4%) of them participated in a follow‐up screening. At baseline, patients who developed the MS ( n = 28/201; 13.9%) showed significantly higher Homeostasis Model Assessment‐Insulin Resistance score (HOMA‐IR) and C‐reactive protein (CRP) values, and lower exercise level than subjects who did not. In a multiple logistic regression analysis, after multiple adjustments, the only baseline variable significantly ( p < 0.01) associated with the MS was CRP (OR = 4.05; 95% CI 2.23–7.38; p < 0.001). Results did not change after adjusting for weight gain. The area under the receiver‐operating curve was 0.83 for CRP after multiple adjustments. The optimal cut‐off point of baseline CRP values was 2.1 mg/L, with 86% (95% CI 81–90) sensitivity and 75% (69–81) specificity in predicting the MS. Baseline CRP resulted associated with after‐study glucose values in a multiple regression model (β = 0.14; 0.08–0.20; p < 0.001). Conclusions Higher baseline CRP values confer a significant increased risk of developing the MS in healthy subjects, independently of weight gain. Copyright © 2009 John Wiley & Sons, Ltd.