【摘要】 目的 通过视网膜分支静脉阻塞(BRVO)继发黄斑水肿(ME)患者治疗前相干光层析成像术(OCT)生物标志物基线特征,构建预测BRVO继发ME患者治疗后水肿复发的临床模型。方法 回顾性病例研究。选取2022年9月至2023年9月我院收治的90例(90只眼)BRVO继发ME患者,给予玻璃体腔注射雷珠单抗治疗,随访期间采用3+PRN方案。根据治疗后6个月ME是否复发分为复发组(n=31)和未复发组(n=59);采集患者一般资料及OCT影像学资料,对两组患者治疗前基线资料进行差异性比较,采用Logistic回归分析影响患者ME复发的预测因素,构建列线图模型并进行内部验证。结果 多因素Logistic回归分析结果显示,BRVO缺血型(β=2.090,OR=8.081,95%CI=2.194～29.763)、治疗前中央视网膜厚度(CRT)(β=0.004,OR=1.004,95%CI=1.001～1.006)、视网膜内层结构紊乱(DRIL)(β=1.469,OR=4.344,95%CI=1.138～16.574)、视网膜下液(SRF)(β=-2.151,OR=0.116,95%CI=0.033～0.413)是水肿复发的独立危险因素,性别、眼别、年龄、眼压、注射前视力、高血压、高血脂、糖尿病、吸烟、高反射灶(HF)数量、外界膜(ELM)和椭圆体带(EZ)、黄斑区视网膜下出血、玻璃体黄斑粘连、脉络膜厚度不是水肿复发的影响因素(均P> 0.05)。由4项危险因素构建的BRVO继发ME患者水肿复发的列线图模型拟合优度霍斯默-莱梅肖检验(χ~2=8.156,P=0.418);针对模型绘制预测BRVO继发ME患者水肿复发列线图模型的预测风险能力指数(C-index)为0.885,提示列线图预测能力较好,绘制决策分析曲线(DCA)评价模型的临床效应显示该模型具有明显的临床正向净收益。BRVO缺血型预测BRVO继发ME患者治疗后ME复发的ROC曲线下面积(AUC)为0.749,灵敏度为87.1%,特异度为62.7%,SRF预测BRVO继发ME患者治疗后ME复发的AUC为0.702,灵敏度为69.5%,特异度为71%,DRIL预测BRVO继发ME患者治疗后ME复发的AUC为0.616,灵敏度为77.4%,特异度为45.8%,治疗前CRT预测BRVO继发ME患者治疗后ME复发的AUC为0.658,灵敏度为93.5%,特异度为37.3%,四者联合预测的AUC为0.885,灵敏度为80.6%,特异度为81.4%,该模型对判断患者复发具有较好的临床诊断价值。结论 BRVO缺血型、基线CRT、DRIL、SRF是影响BRVO继发ME患者治疗后ME复发的独立危险因素,基于此构建的列线图模型可以预测BRVO继发ME患者术后黄斑水肿复发概率,为临床治疗和预后评估提供参考价值。更多还原

【Abstract】 Objective To establish a clinical model for predicting the recurrence of macular edema(ME) after treatment in patients with branch retinal vein occlusion(BRVO) secondary macular edema(ME) based on the baseline characteristics of optical coherence tomography(OCT) biomarkers before treatment.Methods retrospective case study.A total of 90 patients(90 eyes) with BRVO secondary ME treated in our hospital from September 2022 to September 2023were selected and given intravitreal injection of leizumab.During the follow-up period,the 3+PRN regimen was adopted.According to the recurrence of macular edema 6 months after treatment,the patients were divided into recurrence group(n=31) and non-recurrence group(n=59).General data and OCT imaging data of patients were collected,and the difference of baseline data before treatment was compared between the two groups.Logistic regression was used to analyze the predictors of ME recurrence in patients,and a nomogram model was constructed and internal verification was carried out.Results Multivariate Logistic regression analysis showed that BRVO lacking blood type(β=2.090,OR=8.081,95% CI=2.194～29.763),central retinal thickness(CRT)(β=0.004,OR=1.004,95% CI=1.001～1.006),disorganization of the inner retinal layers,(DRIL)(β=1.469,OR=4.344,95% CI=1.138～16.574) and sub retinal fluid(SRF)(β=-2.151,OR=0.116,95% CI=0.033～0.413) were independent risk factors for edema recurrence.Gender,eye type,age,IOP,pre-injection visual acuity,hypertension,hyperlipidemia,diabetes,smoking,HF number,ELM and EZ bands,subretinal hemorrhage in macular area,vitreous macular adhesion,and choroid thickness were not the influencing factors for edema recurrence(P> 0.05).Hosmer-lemeshaw test of goodness of fit for the nomogram model of edema recurrence in BRVO patients secondary to ME constructed by 4 risk factors(χ~2=8.156,P=0.418);The predictive risk ability index(C-index) of the model drawn to predict edema recurrence in BRVO secondary ME patients was 0.885,indicating that the predictive ability of the column graph was good.The clinical effect of the model evaluated by decision analysis curve(DC A) showed that the model had significant clinical positive net benefit.The area under curve AUC of BRVO deficiency predicted ME recurrence in patients with BRVO secondary ME after treatment was 0.749,the sensitivity was 87.1%,and the specificity was 62.7%.The AUC of SRF predicted ME recurrence in patients with BRVO secondary ME after treatment was 0.702.The sensitivity and specificity were 69.5% and 71% respectively.DRIL predicted the AUC of ME recurrence in patients with BRVO secondary ME after treatment was 0.616,the sensitivity was 77.4% and the specificity was 45.8%;CRT predicted the AUC of ME recurrence in patients with BRVO secondary ME after treatment was 0.658 and the sensitivity was 93.5%.The specificity was37.3%,the combined prediction AUC was 0.885,the sensitivity was 80.6%,and the specificity was 81.4%.This model has good clinical diagnostic value in judging patients ’ recurrence.Conclusions BRVO lack blood type,baseline CRT,DRIL and SRF are independent risk factors for ME recurrence in patients with BRVO secondary ME after treatment.Based on this,the column graph model constructed can predict the recurrence probability of postoperative macular edema in patients with BRVO secondary ME,providing reference value for clinical treatment and prognosis evaluation.更多还原