The diagnostic value of metagenomic next-generation sequencing in pulmonary infectious diseases

Abstract

Objective To compare the application value of metagenomic next generation sequencing (mNGS) with traditional culture in diagnosis of pulmonary infection pathogens.

Methods The clinical documents of 310 patients with suspected pulmonary infection admitted to the General Hospital of Center Theater Command from February 2021 to September 2022 were retrospectively analyzed. The results of mNGS and traditional culture were analyzed, followed by comparison on the positive rate, sensitivity, specificity, accuracy (ACC), positive predictive value (PPV) and negative predictive value (NPV) between the two methods.

Results The study revealed that mNGS can simultaneously detect multiple pathogens, with the highest efficiency of detection for bacteria and the lowest for fungi. And the sequencing numbers of bacteria, fungi and viruses shown by mNGS were significantly different ( H=70.361, P<0.001). In comparison, mNGS displayed a higher positive detection rate (88.40%) than traditional culture (29.70%) (χ ²=162.373, P<0.001), but the consistency between the two methods was not significant ( Kappa = −0.003, P=0.902). The sensitivity, specificity, ACC, PPV and NPV of mNGS were 91.29%, 28.26%, 81.94%, 87.96%, and 36.11% respectively, compared to corresponding 30.30%, 73.91%, 36.77%, 86.96% and 15.60% of traditional culture respectively. Through analysis, it is confirmed that the sensitivity and specificity between the two methods were statistically significant (91.29% vs 30.30%, χ ²=148.120, P<0.001 and 28.26% vs 73.91, χ ²=13.793, P<0.001).

Conclusions mNGS can significantly improve the detection rate of pathogens in pulmonary infections and provide a complementary tool besides to traditional culture method for accurate anti-infection therapy. Furthermore, both traditional culture and mNGS pathogen detection methods are highly dependent on sample quality and detection quality control. mNGS requires the correct interpretation of comprehensive, non-destructive pathogenic genetic information to accurately identify pathogens.

Abstract

摘要： 目的 比较宏基因组二代测序 (metagenomic next-generation sequencing, mNGS) 与实验室传统培养在肺部 感染病原中诊断的应用价值。 方法 回顾性分析 2021 年 2 月 —2022 年 9 月中部战区总医院收治的 310 例疑诊肺部感 染患者临床资料, 分析 mNGS 与实验室传统培养检测结果, 对比 2 种检测方法检测肺部感染病原的阳性率、灵敏性、特 异性、准确性 (accuracy, ACC)、阳性预计值 (positive predictive value, PPV)、阴性预计值 (negative predictive value, NPV)。 结果 研究发现, mNGS 能够同时检出多种病原, 对细菌的检出效率最高, 对真菌的检出阳性率最低, 且 mNGS 检出的 细菌、真菌与病毒的序列数之间有显著差异 ( H=70.361, P<0.001)。在比较分析 mNGS 与实验室传统培养方法中发现, mNGS检出阳性率(88.40%)高于实验室传统培养 (29.70%), 差异有统计学意义(χ ²=162.373, P<0.001), 检测结果一致性 比 较 Kappa=−0.003, P=0.902; mNGS 的 灵 敏 性、特 异 性、ACC、PPV、NPV 分 别 为 91.29%、28.26%、81.94%、87.96%、36.11%, 传统培养分别为 30.30%、73.91%、36.77%、86.96%、15.60%; 其中, 两种方法间的灵敏性和特异性差异有统计学 意义 (91.29% vs 30.30%, χ ²=148.120, P<0.001; 28.26% vs 73.91%, χ ²=13.793, P<0.001)。 结论 mNGS 能够显著提高肺 部感染病原的检出率, 是传统实验室培养方法的必要补充, 为指导临床抗感染治疗提供有力依据。同时, 传统培养与 mNGS 病原检测方法都高度依赖于标本质量和检测质控, mNGS 在提供无损的多种病原遗传信息的同时也需正确解读, 才能正确甄别致病原。