Woodward Mark, Tunstall-Pedoe Hugh
Cardiovascular Epidemiology Unit, University of Dundee, Scotland.
Eur J Cardiovasc Prev Rehabil. 2009 Apr;16(2):210-4. doi: 10.1097/HJR.0b013e3283282f8d.
The metabolic syndrome (MS) is a popularly used risk marker for coronary heart disease (CHD), yet its utility is in doubt.
Cohort study based in Glasgow, Scotland, of 1471 men and women free of cardiovascular disease, followed up for a median of 13.7 years.
MS was defined according to current criteria, requiring at least three of five dichotomous risk factors to be positive. Cox models were used to obtain hazard ratios (HRs) and discrimination was quantified by areas under receiver operating characteristic curves (AUCs) using 500 bootstrap samples.
The HR (95% confidence interval) for CHD, MS versus no MS was 2.23 (1.67-2.97). However, the HR rose monotonically when plotted against the number of positive components, with no suggestion of a threshold effect at three positive components. Furthermore, the HR also changed monotonically as each of the five continuous variables defining the different components increased, again with no obvious threshold effects. The AUC for MS was low, at 0.5764, this being significantly (P<0.0001) lower than the AUCs for other risk prediction models, including the Framingham score, 0.7517.
Although MS is related to CHD, there is no epidemiological justification for using it, rather than other criteria, as a risk predictor for CHD.
代谢综合征（MS）是常用的冠心病（CHD）风险标志物，但其效用存疑。
基于苏格兰格拉斯哥的队列研究，纳入1471名无心血管疾病的男性和女性，中位随访13.7年。
根据当前标准定义MS，要求五个二分风险因素中至少三个为阳性。使用Cox模型获得风险比（HRs），并通过使用500个自抽样样本的受试者工作特征曲线下面积（AUCs）对辨别力进行量化。
MS组与非MS组相比，冠心病的HR（95%置信区间）为2.23（1.67 - 2.97）。然而，当按阳性成分数量绘制时，HR单调上升，在三个阳性成分时未显示阈值效应。此外，随着定义不同成分的五个连续变量中的每一个增加，HR也单调变化，同样没有明显的阈值效应。MS的AUC较低，为0.5764，显著低于其他风险预测模型的AUC，包括弗雷明汉评分的0.7517（P<0.0001）。
尽管MS与CHD相关，但没有流行病学依据使用它而非其他标准作为CHD的风险预测指标。
