Abstract: Objective The purpose of this study was to determine the prevalence of insulin resistanceamong nondiabetic patients with acute ischemic stroke and to assess the association between thetriglycerides-HDL-cholesterol ratio (TG:HDL-C) and impaired insulin sensitivity.Methods From August 2008 to December 2009, We enrolled 610 unselected subjects who wereadmitted to the neurological department in TianTan Hospital in this study prospectively. Eligiblesubjects without medical history of diabetes underwent an oral glucose tolerance test (OGTT) 14±3 days after their event. The homeostatic model assessment (HOMA) was used to estimate insulinresistance (HOMA-IR). Risk factors correlated with insulin resistance included blood pressure,body mass index(BMI), HDL cholesterol, triglycerides, triglycerides-HDL-cholesterol ratio (TG:HDL-C), high-sensitivity C-reactive protein (hsCRP), fibrinogen and hyperhomocysteinemia etc.Results In 375 out of 480 ischemic stroke patients without previous DM, glucometabolic statewas determined by oral glucose tolerance test (OGTT). A normal glucose tolerance (NGT) wasidentified in 123 patients, isolated-impaired glucose regulation (I-IFG) in 5 patients, isolatedimpairedglucose tolerance(I-IGT) in 108 patients, combined glucose intolerance (CGI) in 10patients.The propotion of insulin resistance was 8.1% in the NGT group, as compared with 20.0%,17.6%, 30.0% in the I-IFG group, I-IGT group and CGI group respectively (χ2 for trends, P <0.01).TG:HDL-C and hsCRP was associated with an increased risk of impaired insulin sensitivityindependently. The adjusted odds ratio of having insulin resistance was 6.09 [95% confidenceinterval (CI ) 1.43-25.87, P <0.05] for the top TG/HDL level quater and 4.11(95%CI 1.06-15.95,P <0.05) for middle quarters combined than for the bottom quater. The adjusted odds ratio of havinginsulin resistance was 1.19 (95%CI 0.38-3.74, P >0.05) for the top sCRP level quater and 3.98(95%CI1.14-13.89, P <0.05) for middle quarters combined than for the bottom quater.Conclusion Our data revealed that HOMA-estimated insulin resistance and subclinicalinflammation are present in nondiabetic patients with acute ischemic stroke. The TG:HDL-C was apredictor for impaired insulin sensitivity. These findings have important implications for diagnosisand therapy in case that treatment to improve insulin sensitivity is shown to reduce risk forsubsequent stroke and heart disease.

Key words: Brain infarction; Nondiabetic; Insulin resistance; Triglycerides; Cholesterol; HDL

摘要： 目的 在非糖尿病缺血性卒中患者中观察胰岛素抵抗的患病情况及甘油三酯（triglycerides，TG）/高密度脂蛋白胆固醇（high density lipoprotein- cholesterol，HDL-C）比值和胰岛素抵抗的关系。方法 在2008年8月～2009年12月期间天坛医院神经内科连续入选符合研究标准的610名缺血性卒中患者。无糖尿病既往史患者，在卒中发病14±3d行口服糖耐量试验（oral glucose tolerance test，OGTT）。应用胰岛素抵抗指数（homeostasis model assessment-insulin resistance，HOMA-IR）评定胰岛素抵抗情况。测定胰岛素抵抗可能相关的危险因素，包括血压、体重指数（body mass index，BMI）、HDL-C、TG/HDL-C比值、超敏C反应蛋白（high-sensitivity C-reactive protein，hsC-RP）、纤维蛋白原和高同型半胱氨酸血症等。结果 在入选缺血性卒中患者中，480名患者既往无糖尿病史。375名符合条件患者完成OGTT检查和HOMA测定。123名患者诊断为正常糖耐量（normal glucose tolerance，NGT），5名患者为单纯性空腹血糖受损（isolated-Impaired glucose regulation，I-IFG），108名患者为单纯性糖耐量受损（isolated-Impaired glucose tolerance，I-IGT），10名患者为复合性糖耐量受损（combined glucose intolerance，CGI）。胰岛素抵抗在NGT，I-IFG，I-IGT，CGI组中的患病率分别为8.1% 20.0%，17.6%和30.0%（χ2趋势检验，P <0.01）。多因素分析显示TG/HDL-C，hsC-RP和胰岛素抵抗独立相关。以TG/HDL-C的最低1/4组作参照，最高1/4组存在胰岛素抵抗的危险是参照的6.09倍[95%可信区间（95% confidence interval，95%CI）1.43～25.87，P <0.05]；中间组存在胰岛素抵抗的危险是参照的4.11倍（95%CI 1.06～15.95，P <0.05）。以h s C-RP的最低1/4组作参照，最高1/4组存在胰岛素抵抗的危险是参照的1.19倍（95%CI 0.38～3.74，P＞0.05）；中间组存在胰岛素抵抗的危险是参照的3.98倍（95%CI 1.14～13.89，P <0.05）。结论 在非糖尿病缺血性卒中患者存在胰岛素抵抗和亚急性炎性现象。TG/HDL-C是胰岛素抵抗的预测指标。若能进一步证明治疗胰岛素抵抗能减少卒中和心血管疾病，则上述发现对诊断和治疗具有重要的意义。

关键词: 脑梗死; 非糖尿病; 胰岛素抵抗; 甘油三酯类; 胆固醇; HDL