中国日本血吸虫病人体感染检测的免疫诊断效果：一项荟萃分析。

Immunodiagnostic efficacy of detection of Schistosoma japonicum human infections in China: a meta analysis.

机构信息

Jiangsu Institute of Parasitic Diseases, 117 Yangxiang, Meiyuan, Wuxi, Jiangsu Province, People's Republic of China.

出版信息

Asian Pac J Trop Med. 2012 Jan;5(1):15-23. doi: 10.1016/S1995-7645(11)60238-1.

DOI:10.1016/S1995-7645(11)60238-1

PMID:22182637

Abstract

OBJECTIVE

To assess the diagnostic efficacy of the currently most widely used indirect hemagglutination assay (IHA) and enzyme-linked immunosorbent assay (ELISA) for detection of Schistosoma japonicum human infections.

METHODS

A comprehensive search was undertaken from China National Knowledge Infrastructure, Wanfang Database, VIP Database, PubMed, Cochrane Library, Science Citation Index Expanded, Proquest, and the inclusion and exclusion criteria were strictly settled. The funnel plot was used to assess the publication bias, Cochran's Q test was employed to measure the homogeneity between studies, a summary receiver operating characteristic (SROC) curve was used to compare the diagnostic accuracy between the IHA and ELISA qualitatively by means of the Weighted Least Square method, the Ordinary Least Square method and the Robust regression method, and the diagnostic odds ratio (DOR) was drawn to compare the accuracy quantitatively.

RESULTS

Out of 785 publications, 19 papers were eventually selected for analysis. Literature quality assessment indicated that minor publication bias existed in studies pertaining IHA test, but no bias was found in literatures regarding ELISA test. The heterogeneity test showed a heterogeneity between studies was present (χ(2)=466.07 and 34.67, both P values<0.0001). The areas under the SROC curves of IHA were all higher than that of ELISA test using the three methods (Weighted Least Square method: 0.766 vs. 0.695, Ordinary Least Square method: 0.826 vs. 0.741, Robust regression: 0.815 vs. 0.715). The TPR* values for IHA and ELISA were 0.710, 0.759, 0.749, and 0.650, 0.686 and 0.666, respectively, and OR values were 5.997, 9.937, 8.893, and 3.432, 4.784 and 3.959, respectively. The DOR of IHA was 9.41 (95% CI: 4.88-18.18), and 4.78 (95% CI: 3.21-7.13) for ELISA.

CONCLUSIONS

All above results revealed that the diagnostic performance of IHA is better than that of ELISA. However, taking into account their unsatisfactory diagnostic value in areas with low infection intensity, a search for a better diagnostic test that can be applied in field situations in China should be given high priority.

摘要

目的

评估目前应用最广泛的间接血凝试验（IHA）和酶联免疫吸附试验（ELISA）检测日本血吸虫人体感染的诊断效果。

方法

在中国知网、万方数据库、维普数据库、PubMed、Cochrane 图书馆、科学引文索引扩展版、ProQuest 等数据库中进行全面检索，严格设定纳入和排除标准。采用漏斗图评估发表偏倚，采用 Cochran's Q 检验测量研究间的同质性，采用加权最小二乘法、普通最小二乘法和稳健回归法对 IHA 和 ELISA 进行定性比较，绘制综合受试者工作特征（SROC）曲线，采用诊断优势比（DOR）进行定量比较。

结果

共检出 785 篇文献，最终纳入 19 篇文献进行分析。文献质量评价显示，IHA 试验研究存在轻度发表偏倚，而 ELISA 试验研究则不存在发表偏倚。异质性检验显示，IHA 试验研究间存在异质性（χ²=466.07，P<0.0001；χ²=34.67，P<0.0001）。SROC 曲线下面积显示，IHA 的曲线下面积均高于 ELISA 试验（加权最小二乘法：0.766 比 0.695；普通最小二乘法：0.826 比 0.741；稳健回归：0.815 比 0.715）。IHA 和 ELISA 的 TPR*值分别为 0.710、0.759、0.749 和 0.650、0.686 和 0.666，OR 值分别为 5.997、9.937、8.893 和 3.432、4.784 和 3.959，DOR 值分别为 9.41（95%CI：4.88-18.18）和 4.78（95%CI：3.21-7.13）。