倍频视野计对原发性青光眼诊断效能的荟萃分析

Meta-analysis of diagnostic capability of frequency-doubling technology (FDT) for primary glaucoma

目的以标准自动视野计的异常检测结果和(或)眼底改变为金标准,对研究倍频视野计(FDT)诊断原发性青光眼的文献进行荟萃(Meta)分析,评价FDT的临床诊断效能。方法电子检索MEDLINE、EMBASE、CochraneLibrary、BIOSIS、Previews、HMIC、IPA、OVID循证医学数据库、中国期刊网专题全文数据库、中国生物医学文献数据库、维普中文科技期刊数据库、中国医学学术会议论文、中国优秀硕博士论文数据库、超星数字图书馆、书生之家数字图书馆;手工检索国内外相关教科书、杂志、会议论文等及其参考文献。根据Cochrane协作网推荐的诊断试验纳入标准筛选文献,并对纳入文献进行质量评估,提取纳入研究的特征信息。数据分析采用Metatestversion0.6软件,检验异质性,并根据检验结果选择相应的效应模型进行Meta分析。对所有研究予以加权定量合并,计算敏感性、特异性及95%可信区间。绘制汇总受试者工作特征曲线(SROC),并计算曲线下面积。最后进行敏感性分析。结果共检索到相关文献206篇,15篇英文文献(21个研究)符合纳入标准,研究对象共3172例。FDT诊断原发性青光眼的敏感性范围为0.51～1.00,特异性范围为0.58～1.00。纳入研究存在异质性。按照随机效应模型进行计算汇总,其敏感性、特异性分别为0.86(0.80～0.90)、0.87(0.81～0.91)。SROC曲线下面积为93.01%。纳入文献稳定性好。结论纳入的文献具有较高质量,汇总Meta分析显示FDT诊断原发性青光眼的效能较高,但还需要高质量的前瞻性研究作进一步分析。

Objective To evaluate the overall diagnostic capabilities of frequency-doubling technology （FDT） in patients of primary glaucoma, with standard automated perimetry （SAP） and/or optic disc appearance as the gold standard. Methods A comprehensive electric search in MEDLINE, EMBASE, Cochrane Library, BIOSIS, Previews, HMIC, IPA, OVID, CNKI, CBMdisc, VIP information, CMCC, CCPD, SSreader and 21dmedia and a manual search in related textbooks, journals, congress articles and their references were performed to identify relevant English and Chinese language articles. Criteria for adoptability were established according to validity criteria for diagnostic research published by the Cochrane Methods Group on Screening and Diagnostic Tests. Quality of the included articles was assessed and relevant materials were extracted for studying. Statistical analysis was performed with Meta Test version 0.6 software. Heterogeneity of the included articles was tested, which was used to select appropriate effect model to calculate pooled weighted sensitivity and specificity. Summary Receiver Operating Characteristic （SROC） curve was established and the area under the curve （AUC） was calculated. Finally, sensitivity analysis was performed. Results Fifteen English articles （21 studies） of 206 retrieved articles were included in the present study, with a total of 3172 patients. The reported sensitivity of FDT ranged from 0.51 to 1.00, and specificity from 0.58 to 1.00. The pooled weighted sensitivity and specificity for FDT with 95% confidence intervals （95% CI） after correction for standard error were 0. 86 （0.80-0.90）, 0. 87 （0. 81-0. 91 ）, respectively. The AUC of SROC was 93.01%. Sensitivity analysis demonstrated no disproportionate influences of individual study. Conclusions The included articles are of good quality and FDT can be a highly efficient diagnostic test for primary glaucoma based on Meta-analysis. However, a high quality perspective study is still required for further analysis. （Chin J Opbthalmol, 2006,42：403-408）