摘要目的 对已发表的使用荧光原位杂交(FISH)技术对膀胱癌患者的诊断性试验进行系统评价,比较其是否在诊断效能上优于传统的尿液细胞学检查.方法 检索相关文献,根据纳入标准筛选文献,利用Meta-DiSc软件分别计算FISH检查组和细胞学检查组灵敏度、特异度、似然比和诊断性比数比.绘制总受试者工作特征曲线,计算曲线下面积,估计总诊断精确度.结果 有14篇文献被纳入,FISH检测膀胱癌总灵敏度为0.69(95%CI 0.66～0.73),特异度为0.84(95%CI 0.81～0.86),总受试者工作特征曲线的曲线下面积为0.87;细胞学检查检测膀胱癌总灵敏度为0.46(95%CI 0.42～0.51),特异度为0.88(95%CI 0.84～0.90),总受试者工作特征曲线的曲线下面积为0.81.FISH检查在总体诊断效能上与细胞学检查差异无统计学意义(P>0.05).结论 FISH检查对膀胱癌的诊断灵敏度高于细胞学检查,但尚不能证实FISH技术检查膀胱癌对初发患者和随访患者混合构成的人群具有较高价值.

abstractsObjective To compare the diagnostic value of fluorescence in-situ hybridization (FISH) and traditional cytology in detection of bladder cancer, and to evaluate if FISH test is superior to cytologic examination in this respect. Methods The electronic database of PubMed was searched. Criteria for inclusion and exclusion into the meta-analysis were established. The studies fulfilling the inclusion criteria were selected. Data of these studies were extracted for calculation of pooled odds ratio, likelihood ratio, sensitivity and specificity. Summary receiver operating curve and area under curve (AUC) were produced with Meta-DiSc software. Results Fourteen studies were included, and data related to cytologic examination were available in nine of them. The pooled sensitivity and specificity of FISH test were 0.69 (95% CI 0. 66 to 0. 73) and 0. 84 (95% CI 0. 81 to 0.86), respectively. The pooled sensitivity and specificity of cytology were 0. 46 (95% CI 0. 42 to 0. 51) and 0. 88 (95% CI 0. 84 to 0. 90), respectively.The AUC for FISH and cytology were 0. 87 and 0. 81, respectively. The difference in AUC between the two tests was not statistically significant (P > 0. 05). Conclusions Comparing with traditional cytological examination, FISH test is more sensitive. However, FISH test is still not proven to have better diagnostic value in detection of bladder cancer in both new and follow-up cases. More studies in this area are required for further evaluation.