背景:越来越多的系统综述(SR)评估了下一代测序(NGS)在感染性疾病(ID)中的诊断价值。
目的:这项总括分析旨在评估现有SR的潜在偏倚风险,并总结已发表的NGS在不同ID中的诊断价值。
方法:我们搜索了PubMed,Embase,和Cochrane图书馆,直到2023年9月,用于评估NGS对ID的诊断有效性的SRs。两名调查人员独立确定审查资格,提取的数据,并评估报告质量,偏见的风险,方法学质量,以及所包括SR中的证据确定性。
结果:分析了11个SR。大多数SR表现出中等水平的报告质量,而所有SRs均存在严重偏倚风险.NGS在检测肺囊虫肺炎和假体周围/假体关节感染方面的诊断性能非常强大,表现出优异的敏感性(肺囊虫肺炎:0.96,95%CI0.90-0.99,非常低的确定性;假体周围/假体关节感染:0.93,95%CI0.83-0.97,非常低的确定性)和特异性(肺囊虫肺炎:0.96,95%CI0.92-0.98,非常低的确定性;假体周围/假体关节感染:0.95,95%CI0.92-0.97,非常低的确定性)。NGS对中枢神经系统感染表现出高度特异性,细菌性脑膜脑炎,和结核性脑膜炎.对这些传染病的敏感性是中等的。NGS对多种感染和肺部感染表现出中等的敏感性和特异性。
结论:这项伞式分析表明,NGS是诊断肺孢子虫肺炎和假体周围/假体关节感染的一种有前途的技术,具有出色的敏感性和特异性。需要更多高质量的原始研究和SRs来验证当前的发现。
BACKGROUND: An increasing number of systematic reviews (SRs) have evaluated the diagnostic values of next-generation sequencing (NGS) in infectious diseases (IDs).
OBJECTIVE: This umbrella analysis aimed to assess the potential risk of bias in existing SRs and to summarize the published diagnostic values of NGS in different IDs.
METHODS: We searched PubMed, Embase, and the Cochrane Library until September 2023 for SRs assessing the diagnostic validity of NGS for IDs. Two investigators independently determined review eligibility, extracted data, and evaluated reporting quality, risk of bias, methodological quality, and evidence certainty in the included SRs.
RESULTS: Eleven SRs were analyzed. Most SRs exhibited a moderate level of reporting quality, while a serious risk of bias was observed in all SRs. The diagnostic performance of NGS in detecting pneumocystis pneumonia and periprosthetic/prosthetic joint infection was notably robust, showing excellent sensitivity (pneumocystis pneumonia: 0.96, 95% CI 0.90-0.99, very low certainty; periprosthetic/prosthetic joint infection: 0.93, 95% CI 0.83-0.97, very low certainty) and specificity (pneumocystis pneumonia: 0.96, 95% CI 0.92-0.98, very low certainty; periprosthetic/prosthetic joint infection: 0.95, 95% CI 0.92-0.97, very low certainty). NGS exhibited high specificity for central nervous system infection, bacterial meningoencephalitis, and tuberculous meningitis. The sensitivity to these infectious diseases was moderate. NGS demonstrated moderate sensitivity and specificity for multiple infections and pulmonary infections.
CONCLUSIONS: This umbrella analysis indicates that NGS is a promising technique for diagnosing pneumocystis pneumonia and periprosthetic/prosthetic joint infection with excellent sensitivity and specificity. More high-quality original research and SRs are needed to verify the current findings.