抄録
Background HbA1c is proposed as an optional component of metabolic syndrome (MS).
Therefore, we studied an optimal cut point of HbA1c as a component of MS. Methods Medical check-up data of 1354 men and 814 women were examined using Spearman's correlation coefficients among MS-related components, stepwise logistic regressions using MS and Japanese MS (JMS) as dependent variables, and receiver operating characteristic (ROC) curves for diagnosing MS and JMS. MS was defined by revised National Cholesterol Education Program (NCEP) criteria for Japanese and JMS was defined by the Examination Committee for Criteria of Metabolic Syndrome. The optimal cut point was defined as the point on the ROC curve from where the distance to the point where both sensitivity and specificity were one was minimal.
Results The correlation coefficients of HbA1c with other MS-related components and the area under ROC curve (AUC) of HbA1c for diagnosing MS and JMS were comparable to those of other MS-related cornponents. HbA1c was an independent marker for JMS, but not for MS. AUC of HbA1c for diagnosing MS and JMS were 0.731 and 0.717 in men and 0.737 and 0.802 in women, respectively. The optimal cut point with its sensitivity and specificity of HbA1c for diagnosing MS and JMS were 5.15% with 0.61 and 0.75 and 5.2% with 0.57 and 0.79, respectively in men and 5.2% with 0.56 and 0.77 and 5.2% with 0.68 and 0.75, respectively in women. Conclusion We recommend 5.2% as an optimal cut point of HbA1c for diagnosing MS among Japanese.
