Other TitlesComparison of different measure methods of macular hole closure index for predicting the anatomical prognosis of idiopathic macular hole surgery
Retinal perforations/surgery
Tomography,optical coherence
Issue Date2017
Citation中华眼底病杂志. 2017, 33(4), 341-345.
Abstract目的 对比观察光相干断层扫描(OCT)仪内置测量工具及ImageJ图像处理软件两种不同方法测得的黄斑裂孔(MH)愈合指数(MHCI)对特发性MH手术后MH闭合的预测作用.方法 前瞻性预探索性临床研究.接受玻璃体切割联合内界膜剥除、气体填充手术治疗的特发性MH患者63例63只眼纳入研究.所有患眼手术前后均行OCT检查.分别采用OCT仪内置测量工具、ImageJ图像处理软件测量OCT仪显示屏上及OCT仪自动调整纵横比例后图像的MH两侧外界膜断开处到视网膜光感受器与视网膜色素上皮层脱离起点间的距离(M、N)及MH底径(BASE).MHCI=(M+N) /BASE.其中,OCT仪内置测量工具测得的MHCI以MHCI1表示;ImageJ图像处理软件测得的MHCI以MHCI2表示.同时应用OCT仪内置测量工具测量MH最小径(MHD).根据OCT图像特征将MH闭合程度分为桥状闭合(A级)、完全闭合(B级)、不良闭合(C级)3个等级.A、B级闭合合并为良好闭合;C级闭合为不良闭合.手术后3、6、12个月,采用与手术前相同的设备与方法行相关检查.以末次随访时OCT检查显示的MH闭合等级为最终结局指标,对比分析MHCI1、MHCI2与MH闭合等级之间的相关性.应用受试者工作特征(ROC)曲线分析测量MHCI1、MHCI2与MHD预测MH手术后良好闭合与不良闭合、A级闭合与B级闭合的曲线下面积(AUC),确定预测MH闭合等级的界值及对应的灵敏度和特异度.以AUC>0.9为该指标预测效能良好.结果 患眼MHCI1为0.30~1.35,平均MHCI1为0.68±0.21;MHCI2为0.41~1.55,平均MHCI2为0.95士0.26;MHD为127~956 μm,平均MHD为(476.244210.18)μm.MHCI1、MHCI2与手术后MH闭合等级均呈负相关(r=-0.665、-0.691,P<0.001).ROC曲线分析结果显示,判断良好闭合及不良闭合时,MHCI1、MHCI2的AUC分别为0.928、0.957,下界值分别为0.505、0.670,灵敏度分别为96.2%、92.3%,特异度分别为81.8%、72.7%;MHD的AUC为0.916,下界值为559 μm;灵敏度为90.9%,特异度为76.9%.判断A、B级闭合时,MHCI1、MHCI2的AUC分别为0.840、0.847,上界值分别为0.705、0.965,灵敏度分别为80.0%、82.9%,特异度分别75.0%、85.7%;MHD的AUC为0.653,上界值为364 μm,灵敏度为63.4%,特异度为65.9%.结论 OCT仪内置测量工具与ImageJ图像处理软件测得的MHCI对特发性MH手术后MH闭合等级均有良好预测作用.
Objective To compare the predicted efficiency ofmacular hole closure index (MHCI) calculated by 2 different methods for postoperative anatomical outcomes after idiopathic macular hole (MH) surgery.Methods This is a prospective exploratory clinical study.A total of 63 patients (63 eyes) with idiopathic MH,who received vitrectomy,inner limiting membrane peeling and gas tamponade,were enrolled in this study.All the patients received optical coherence tomography (OCT) examination at each visit to measure the MHCI using the formula MHCI=(M+N) /BASE,M and N is the distance from outer limiting membrane break points to the beginning points of detached photoreceptor from retinal pigment epithelium of both side of the hole,respectively.BASE is the length of MH base.MHCI1 was measured by built-in caliper of OCT software,MHCI2 was measured by ImageJ software.The minimum macular diameter (MHD) was measured by built-in caliper of OCT software.Based on the OCT images,the anatomical outcomes were classified grade A (bridge-like shape closure),grade B (complete closure) and grade C (poor closure).Grade A and B are considered as good closure,grade C as poor closure.Patients were followed up at 3,6 and 12 months after surgery.The closure grades at last visit were the final outcome.The relationship between MHCI 1,MHCI2 and closure grades was analyzed.And the predicted efficiency of MHD,MHCI1 and MHCI2 for anatomical outcomes after the surgery was studied.Results The mean MHCI1 was 0.68±0.21 (0.30-1.35),MHCI2 was 0.95±0.26 (0.41-1.55),and MHD was (476.24±210.18) μm (127-956 μm).MHCI1 and MHCI2 were both negative correlated with the closure grades (r=-0.665,-0.691;P<0.001).The receiver operating characteristic (ROC) curve analysis ofMHCI1,MHCI2 and MHD for the prediction of good or poor closure showed that area under the curve (AUC) was 0.928,0.957 and 0.916 respectively,and 0.505,0.67 and 559 μm were set as the lower cut-offvalue.The sensitivity was 96.2%,92.3% and 90.9% respectively,and specificity was 81.8%,72.7% and 76.9% respectively.Accordingly,the ROC curve analysis for the prediction of grade A or B closure showed that AUC was 0.840,0.847 and 0.653 respectively,and 0.705,0.965 and 364 μm were set as the upper cut-off value.The sensitivity was 80.0%,82.9%,63.4% respectively and specificity was 75.0%,85.7%,65.9%.Conclusion MHCI1 and MHCI2,measured by built-in caliper of OCT software or ImageJ software,both have good predictive efficiency for the anatomical outcomes of MH surgery.
Appears in Collections:人民医院

Files in This Work
There are no files associated with this item.

Web of Science®

Checked on Last Week


Checked on Current Time


Checked on Current Time

Google Scholar™

License: See PKU IR operational policies.