CMR对急性心肌炎诊断价值的Meta分析

Meta analysis of the diagnostic performance of cardiac magnetic resonance in acute myocarditis

目的通过Meta分析评价心脏磁共振(CMR)各参数对急性心肌炎(AM)的诊断效能。方法检索Embase、PubMed、Cochrane、中国知网、万方、维普数据库的CMR诊断AM的中英文文献,按照纳入排除标准筛选CMR对AM的诊断价值的研究,检索时间从建库至2020年7月31日。根据QUADAS-2条目评价纳入文献的质量,采用RevMan5.4评价偏倚风险及临床适用性,采用Meta-disc 1.4软件对纳入研究的合并敏感度、合并特异度、合并阳性似然比、合并阴性似然比和合并诊断比值比进行分析。绘制森林图和综合受试者操作特征(SROC)曲线,并计算相应曲线下面积(AUC)。采用不一致指数(I^(2))和Chochrane Q指数评价异质性,采用阈值效应、Meta回归、敏感性分析探讨异质性来源。采用Stata 15.1绘制漏斗图,评估是否存在发表偏倚,并进行Begg和Egger检验。结果最终纳入21篇中英文文献,包括1678例病人。路易斯湖标准(LLC)和native T_(1) mapping的合并敏感度、合并特异度分别为77%、88%和85%、85%,两参数的合并敏感度、合并特异度相当;native T_(1) mapping的AUC(0.9419)大于LLC的(0.8681)。T_(2)r、EGE、LGE、LLC、native T_(1) mapping、T_(2) mapping和ECV参数均存在中度以上异质性,阈值效应是EGE异质性的重要来源,Meta回归分析显示设备、场强、研究方法和确诊标准均不是异质性来源(均P>0.05)。Begg和Egger检验结果显示纳入研究的以上参数均不存在明显的发表偏倚(均P>0.05)。结论native T_(1) mapping具有良好的诊断效能,且只需评估单一参数,可优化检查流程。

Objective To evaluate the efficacy of cardiac magnetic resonance(CMR)in the diagnosis of acute myocarditis(AM)by Meta analysis.Method We searched the Embase,PubMed,Cochrane,CNKI,Wanfang,and VIP databases for the English and Chinese literatures about diagnosis of AM using CMR,and selected the value of CMR in AM according to the inclusion criteria.The retrieval time was from construction to July 31,2020.The quality of included literatures was evaluated according to QUADAS-2,RevMan 5.4 was used to evaluate the risk of bias and clinical applicability,Meta-disc 1.4 was used for consolidation analysis of combined sensitivity,combined specificity,combined positive likelihood ratio,combined negative likelihood ratio,and diagnostic odds ratio.Forest plots and summary receiver operating characteristics(SROC)curves were plotted and the area under the SROC curve(AUC)was calculated.The heterogeneity between studies was evaluated with I2 and Chochrane Q.Threshold effect,meta regression,and sensitivity analysis were performed to explore the sources of heterogeneity.Stata 15.1 was used to draw funnel plots to evaluate whether there was publication bias,and Begg and Egger test were performed.Result Twenty-one studies were included,and a total of 1678 patients were analyzed.The combined sensitivity,combined specificity of Lake Louise Criteria(LLC)and native T_(1) mapping were 77%and 88%,and 85%and 85%,respectively.The combination sensitivity and specificity of the two parameters were similar.The AUC of native T_(1) mapping(0.9419)was greater than LLC(0.8681).T_(2)r,EGE,LGE,LLC,native T_(1) mapping,T_(2) mapping,and ECV all showed over moderate heterogeneity.The threshold effect was an important source of heterogeneity of EGE,and meta regression showed that equipment,field intensity,study method,and diagnostic criteria were not the sources of heterogeneity(all P>0.05).Begg and Egger test indicated that there was no obvious publication bias for the above parameters in the included study(all P>0.05).Conclusion Native T_(1) mapping has good diagnostic efficacy,and can be used as a single parameter to evaluate AM,thus the examination process can be optimized.