血清NLR、sST2对心房颤动射频消融术后晚期复发的影响及风险预测模型构建

Influence of neutrophil/lymphocyte ratio and soluble growth stimulation expressed gene 2 on late recurrence of atrial fibrillation after radiofrequency ablation and risk prediction model establishment

目的基于血清中性粒细胞/淋巴细胞比值(NLR)、可溶性生长刺激表达基因2(sST2)构建预测心房颤动(AF)患者射频消融术(RFA)后晚期复发的风险预测模型并进行验证。方法前瞻性分析2020年1月至2022年6月期间于黄冈市中心医院收治的接受射频消融术治疗的117例心房颤动患者,根据晚期复发情况将患者可分为未复发组(n=85)和复发组(n=32)。比较两组临床资料,采用二元Logistic回归分析法明确晚期复发的影响因素,将独立危险因素引入R软件构建风险列线图,采用Bootstrap法验证模型区分度,绘制Calibration曲线和受试者工作特征(ROC)曲线进行拟合度及预测效能评估。结果复发组患者左心房内径(LAD)值、心房颤动早期复发(ERAF)占比、白细胞计数(WBC)、超敏C反应蛋白(hs-CRP)、脑钠肽(BNP)及中性粒细胞/淋巴细胞比值(NLR)、可溶性生长刺激表达基因2(sST2)水平高于未复发组,差异有统计学意义(P<0.05)。二元Logistic回归模型显示,LAD值偏大、血清NLR偏高、sST2偏高是导致AF心房颤动患者RFA术后晚期复发的独立危险因素(P<0.05)。根据独立危险因素建立心房颤动患者射频消融术后晚期复发的预测模型方程,采用Bootstrap法对预测模型进行内部验证,结果显示模型区分度良好,Calibration curve显示模型拟合度好。基于血清NLR、sST2构建的预测模型的AUC为0.879(95%CI:0.816~0.938,P<0.05),预测效能优于NLR、sST2单独预测。结论血清NLR、sST2是导致心房颤动患者射频消融术后晚期复发的危险因素,基于此构建风险预测模型区分度、拟合度良好,对射频消融术后晚期复发具有较高的预测价值。

Objective To establish a risk predictive model and verify it based on neutrophil/lymphocyte ratio(NLR)and soluble growth stimulation expressed gene 2 on(sST2)in patients with late recurrence of atrial fibrillation(AF)after radiofrequency ablation(RFA).Methods AF patients(n=117)received RFA were prospectively analyzed in Central Hospital of Huanggang City from Jan.2020 to June 2022.The patients were divided into non-recurrence group(n=85)and recurrence group(n=32).The clinical materials were compared between 2 groups.The influence factors for late AF recurrence were analyzed by using binary Logistic regression analysis.The independent risk factors were introduced into R software to build a risk nomogram,and model discrimination was verified by using Bootstrap method.The degree of fitting and prediction efficiency were reviewed after drawing Calibration curve and ROC curve.Results The indexes of left atrial diameter(LAD),proportion of early recurrenceatrial fibrillation(ERAF),white blood cell count(WBC),high sensitivityC-reactive protein(hs-CRP),brain natriuretic peptide(BNP)and levels of NLR and sST2 were higher in recurrence group than those in non-recurrence group(P<0.05).The results of binary Logistic regression analysis showed that longer LAD and higher NLR and sST2 were independent risk factors for late recurrence on AF patients after RFA(P<0.05).The predictive model equation was established according to independent risk factors,and verified internally by using Bootstrap method.The results showed thatmodel discrimination was good,and Calibration curve showed that model fitting degree was good.The AUC of predictive model established based on NLR and sST2 was 0.879,95%CI:0.816~0.938,(P<0.05),and predictive efficacy of combination of NLR and sST2 was higher the that of single NLR or single sST2.Conclusion NLR and sST2 are risk factors for late recurrence in AF patients after RFA,and the risk predictive model established based on which has good discrimination and fitting degree,and has higher predictive value to late AF recurrenceafter RFA.