Juhan-Vague I, Morange P E, Frere C, Aillaud M F, Alessi M C, Hawe E, Boquist S, Tornvall P, Yudkin J S, Tremoli E, Margaglione M, Di Minno G, Hamsten A, Humphries S E
Laboratoire d'Hématologie, CHU Timone, Inserm EPI 99-36, Marseilles, France.
J Thromb Haemost. 2003 Nov;1(11):2322-9. doi: 10.1046/j.1538-7836.2003.00458.x.
Although the potential role of plasminogen activator inhibitor-1 (PAI-1) in the development of coronary artery disease is strongly supported by its biological characteristics, results of clinical studies remain controversial.
To investigate whether plasma PAI-1 concentrations and the -675 4G/5G polymorphism located in the PAI-1 gene could constitute risk markers for myocardial infarction (MI).
We used a European case-control study, the HIFMECH study, comparing 598 men with MI and 653 age-matched controls.
Insulin resistance explained a major part of the variation in PAI-1 (24%) whereas inflammation had only a minor contribution (0.01%). For both cases and controls plasma PAI-1 concentrations were significantly higher in the North than the South, and in both regions were higher in individuals with MI compared with control subjects [overall odds ratio (OR) for a 1 SD increase=1.54, 95% confidence interval (CI) 1.34, 1.77]. This difference was observed in all the centers studied. Overall, the difference between cases and control subjects remained significant after controlling for inflammation variables (OR=1.30, 95% CI 1.08, 1.57), but lost significance after controlling for insulin resistance variables (OR=1.17, 95% CI 0.98, 1.40). The 4G allele was associated with significantly higher PAI-1 levels in cases but not controls and, taken independently, did not modify the risk of MI (P=0.9). However, a significant interaction was observed with both insulin or proinsulin and the risk of MI (P=0.05 and 0.02, respectively), but not with triglycerides or body mass index (BMI). The insulin or proinsulin effect on risk was observed only in the carriers of the 4G/4G genotype. This interaction appeared not to be mediated by plasma PAI-1 antigen concentrations (P=0.01 and 0.02 after adjustment for PAI-1 plasma levels). The interaction with proinsulin but not insulin remained statistically significant after further adjustment for other factors associated with insulin resistance (triglycerides and BMI) and C-reactive protein (P=0.01).
This study suggests that PAI-1 has a role in risk of MI in the presence of underlying insulin resistance. A significant interaction between insulin or proinsulin and the -675 4G/5G polymorphism was observed in risk for MI. The mechanisms for these interactions remain to be determined.
尽管纤溶酶原激活物抑制剂-1(PAI-1)在冠状动脉疾病发展中的潜在作用得到其生物学特性的有力支持,但临床研究结果仍存在争议。
研究血浆PAI-1浓度及PAI-1基因中-675 4G/5G多态性是否可构成心肌梗死(MI)的风险标志物。
我们采用了一项欧洲病例对照研究,即HIFMECH研究,比较了598例MI男性患者和653例年龄匹配的对照者。
胰岛素抵抗解释了PAI-1变异的主要部分(24%),而炎症的贡献较小(0.01%)。对于病例组和对照组,血浆PAI-1浓度在北方均显著高于南方,且在两个地区,MI患者的血浆PAI-1浓度均高于对照者[1个标准差增加的总体比值比(OR)=1.54,95%置信区间(CI)1.34,1.77]。在所有研究中心均观察到这种差异。总体而言,在控制炎症变量后,病例组和对照组之间的差异仍具有统计学意义(OR=1.30,95%CI 1.08,1.57),但在控制胰岛素抵抗变量后失去统计学意义(OR=1.17,95%CI 0.98,1.40)。4G等位基因与病例组中显著更高的PAI-1水平相关,但与对照组无关,单独来看,它并未改变MI的风险(P=0.9)。然而,观察到胰岛素或胰岛素原与MI风险之间存在显著的相互作用(分别为P=0.05和0.02),但与甘油三酯或体重指数(BMI)无关。胰岛素或胰岛素原对风险的影响仅在4G/4G基因型携带者中观察到。这种相互作用似乎不是由血浆PAI-1抗原浓度介导的(在调整血浆PAI-1水平后P=0.01和0.02)。在进一步调整与胰岛素抵抗相关的其他因素(甘油三酯和BMI)及C反应蛋白后,与胰岛素原而非胰岛素的相互作用仍具有统计学意义(P=0.01)。
本研究表明,在存在潜在胰岛素抵抗的情况下,PAI-1在MI风险中起作用。观察到胰岛素或胰岛素原与-675 4G/5G多态性之间在MI风险方面存在显著的相互作用。这些相互作用的机制仍有待确定。