Meijnikman Abraham S, De Block Christophe E M, Verrijken An, Mertens Ilse, Van Gaal Luc F
Department of Endocrinology, Diabetology & Metabolism, Faculty of Medicine, Antwerp University Hospital and University of Antwerp, Wilrijkstraat 10, 2650 Edegem, Belgium.
Diabetol Metab Syndr. 2018 Mar 1;10:12. doi: 10.1186/s13098-018-0310-0. eCollection 2018.
The aim of this study to compare the diagnostic accuracy of the metabolic syndrome (MetS) with the FINDRISC score to screen for type 2 diabetes mellitus T2DM in an overweight/obese population.
Subjects 18 years or older visiting the obesity clinic of the Antwerp University Hospital were consecutively recruited between 2012 and 2014. Every patient underwent a standard metabolic work-up including a clinical examination with anthropometry. Glucose status was tested using OGTT and Hba1c. FINDRISC questionnaire and MetS were examined.
Of 651 subjects, 50.4% were diagnosed with prediabetes, whereas 11.1% was diagnosed with T2DM. FINDRISC score increased with worsening of glucose status 11 ± 3, 13 ± 4 and 15 ± 5 in respectively, subjects without T2DM, prediabetes and T2DM. 312 subjects had the MetS. The aROC of the FINDRISC to identify subjects with T2DM was 0.76 (95% CI 0.72-0.82), sensitivity was 64% and specificity was 63% with 13 as cutoff point. Adding FPG or HbA1c to FINDRISC, the aROC increased significantly to 0.91(95% CI 0.88-0.95) and 0.93(95% CI 0.90-0.97), respectively (p < 0.001). The aROC of the MetS to identify subjects with diabetes was 0.72 (95% CI 0.65-0.78), sensitivity was 75% and specificity was 55%. The aROC of the FINDRISC + HbA1c was significantly higher than the MetS for predicting T2DM (p < 0.001).
Prediction of type 2 diabetes is important for timely intervention and to avoid chronic complications associated with the disease. Our findings suggest, that it may be of good clinical practice to use the FINDRISC score + HbA1c in a two-step screening model for diabetes rather than using the metabolic syndrome.
本研究旨在比较代谢综合征（MetS）与芬兰糖尿病风险评分（FINDRISC）在超重/肥胖人群中筛查2型糖尿病（T2DM）的诊断准确性。
2012年至2014年期间，连续招募了安特卫普大学医院肥胖门诊18岁及以上的受试者。每位患者均接受了包括人体测量临床检查在内的标准代谢检查。使用口服葡萄糖耐量试验（OGTT）和糖化血红蛋白（Hba1c）检测血糖状态。对FINDRISC问卷和MetS进行了检查。
在651名受试者中，50.4%被诊断为糖尿病前期，而11.1%被诊断为T2DM。在无T2DM、糖尿病前期和T2DM的受试者中，FINDRISC评分分别随着血糖状态的恶化而升高，分别为11±3、13±4和15±5。312名受试者患有MetS。FINDRISC用于识别T2DM受试者的曲线下面积（aROC）为0.76（95%置信区间0.72-0.82），以13为截断点时，敏感性为64%，特异性为63%。在FINDRISC中加入空腹血糖（FPG）或糖化血红蛋白（HbA1c）后，aROC分别显著提高到0.91（95%置信区间0.88-0.95）和0.93（95%置信区间0.90-0.97）（p<0.001）。MetS用于识别糖尿病受试者的aROC为0.72（95%置信区间0.65-0.78），敏感性为75%，特异性为55%。在预测T2DM方面，FINDRISC+HbA1c的aROC显著高于MetS（p<0.001）。
预测2型糖尿病对于及时干预和避免与该疾病相关的慢性并发症很重要。我们的研究结果表明，在糖尿病两步筛查模型中使用FINDRISC评分+HbA1c而非代谢综合征可能是良好的临床实践。
