摘要
目的 探讨黔东南地区心脏瓣膜置换术后患者CYP2C9和VKORC1基因多态性对华法林稳定剂量及心肌损伤的影响。方法 收集2019年11月至2024年10月期间在贵州医科大学第二附属医院心胸外科行心脏瓣膜置换手术治疗黔东南地区患者110例的临床资料。按照患者CYP2C9*3基因型不同分为野生型(AA型)和突变型(AC型和CC型),VKORC1-1639G>A基因型分为野生型(GG型)和突变型(GA型和AA型)。不同基因型间华法林剂量、组间INR达标时间比较均先进行Levene方差齐性检验。将影响华法林稳定剂量的因素进行分层回归分析。采用BP神经网络模型进行风险预测。采用广义线性混合效应模型(GLMMs)分析CYP2C9、VKORC1基因状态与术后心肌损伤的关系。结果 所有基因分布符合Hardy-Weinberg遗传平衡(均P>0.05)。CYP2C9*3、VKORC1-1639G>A基因状态与房颤、心肌损伤、血栓和出血及合并药物(胺碘酮、利福平、苯巴比妥、伏立康唑)比较差异均具有统计学意义(均P<0.05)。CYP2C9*3基因型AA的华法林稳定剂量和华法林抗凝治疗的达标率(TTR)显著高于基因型AC(均P<0.01),INR达标时间显著低于基因型AC(均P<0.01)。VKORC1-1639G>A基因型GG+GA的华法林稳定剂量和INR达标时间均显著高于基因型AA(均P<0.01),TTR显著低于基因型AA(P<0.01)。分层回归分析结果显示,CYP2C9、VKORC1、年龄、胺碘酮、BSA、BMI、性别对华法林稳定剂量均产生显著的影响(均P<0.05)。BP神经网络模型构建显示,CYP2C9和VKORC1指标对模型分类的贡献度最高,其次是胺碘酮、BSA、BMI、性别、年龄。对模型验证结果显示,准确率为93.54%,灵敏度、特异度分别为85.62%、95.89%,阳性预测值为94.23%,阴性预测值为93.15%,预测模型拟合ROC-AUC为0.958。CYP2C9*3、VKORC1-1639G>A基因突变型患者的cTnI、CK-MB、hs-CRP水平均显著高于野生型患者(均P<0.05)。GLMMs模型分析显示,无论是否调整因素,CYP2C9、VKORC1基因突变型与术后心肌损伤的风险性仍存在统计学关联(均P<0.01)。结论 黔东南地区心脏瓣膜置换术后患者的CYP2C9*3和VKORC1-1639G>A基因多态性显著影响华法林稳定剂量及抗凝达标率,且突变型基因与术后心肌损伤指标升高及风险增加密切相关。
Objective To investigate the effects of CYP2C9 and VKORC1 gene polymorphism on warfarin stable dose and myocardial injury in patients after heart valve replacement in Qiandongnan region.Methods Clinical data of 110 patients in Qiandongnan who underwent heart valve surgery at Guizhou Medical University's Second Affiliated Hospital's Cardiothoracic Department(form November 2019 to October 2024)were collected.According to CYP2C9×3 genotype,patients were divided into wild type(AA type)and mutant type(AC type and CC type),and VKORC1-1639G>A genotype was divided into wild type(GG type)and mutant type(GA type and AA type).Levene variance homogeneity test was performed first to compare warfarin dose between different genotypes and INR compliance time between groups.The factors affecting the stable dose of warfarin were analyzed by hierarchical regression.BP neural network model is used for risk prediction.Generalized linear mixed effects model(GLMMs)was used to analyze the relationship between CYP2C9 and VKORC1 gene status and postoperative myocardial injury.Results The distribution of all genes was consistent with Hardy-Weinberg genetic balance(both P>0.05).CYP2C9*3 and VKORC1-1639G>A gene status were significantly different from atrial fibrillation,myocardial injury,thrombosis and hemorrhage,and combined drugs(amiodarone,rifampicin,phenobarbital,voriconazole)(all P<0.05).The compliance rate(TTR)and warfarin stable dose of warfarin anticoagulation therapy for CYP2C9*3 genotype AA were significantly higher than that of genotype AC(both P<0.01),and the INR compliance time was significantly lower than that of genotype AC(both P<0.01).The warfarin stable dose and INR reach time of VKORC1-1639G>A genotype GG+GA were significantly higher than genotype AA(both P<0.01),and TTR was significantly lower than genotype AA(both P<0.01).Stratified regression analysis showed that CYP2C9,VKORCl,age,amiodarone,BSA,BMI and gender had significant effects on warfarin stable dose(all P<0.O5).BP neural network model construction showed that CYP2C9 and VKORC1 had the highest contribution to model classification,followed by amiodarone,BSA,BMI,gender and age.The model verification results showed that the accuracy was 93.54%,the sensitivity and specificity were 85.62%and 95.89%,respectively.The positive predictive value was 94.23%,the negative predictive value was 93.15%,and the ROC AUC fitting of the prediction model was 0.958.The levels of cTnI,CK-MB and hs-CRP in CYP2C9*3 and VKORC1-1639G>A mutant patients were significantly higher than those in wild-type patients(P<0.05).GLMMs model analysis showed that CYP2C9 and VKORCl gene mutations were still statistically associated with the risk of postoperative myocardial injury regardless of adjustment factors(P<0.01).Conclusion The CYP2C9*3 and VKORC1-1639G>A gene polymorphisms in patients who underwent heart valve replacement surgery in Qiandongnan area significantly affect the stable dose of warfarin and the rate of achieving anticoagulation targets.Moreover,the mutant genes are closely related to the increase in postoperative myocardial injury indicators and the increased risk.
作者
杨榕
宁绍东
杨俊
童立钦
Yang Rong;Ning Shaodong;Yang Jun(Department of Cardiothoracic Surgery,the Second Af filiated Hospital of Guizhou Medical University,Kaili 556000,China)
出处
《华中科技大学学报(医学版)》
北大核心
2025年第4期552-560,共9页
Acta Medicinae Universitatis Scientiae et Technologiae Huazhong
基金
贵州省卫健委科学技术基金项目(No.gzwkj2022-350)
黔东南州科技计划项目(No.黔东南科合基础[2023]18号)。
