靶向高通量测序技术在疑似肺炎患者中的诊断价值

Diagnostic Value of Targeted Next-Generation Sequencing in Patients with Suspected Pneumonia

摘要: 目的：探讨靶向高通量测序(tNGS)在疑似肺炎患者中的诊断价值。方法：回顾性分析335例疑似肺炎患者的病例资料，所有患者均采集下呼吸道标本(首选肺泡灌洗液)并进行了tNGS检测和传统病原学检测(CMTs)。以综合诊断为标准分析tNGS和CMTs在疑似肺炎患者中的诊断效能，比较两者所检出的致病病原体分布以及在不同类型病原体中的诊断效能。结果：335名患者中，34名患者(10.1%)被诊断为非感染性病因，301名(89.9%)被诊断为感染性肺炎，其中270名患者被诊断为微生物学确诊的肺炎。以综合诊断为标准，tNGS的总体诊断敏感性(81.1% vs. 38.9%, P < 0.05)、准确度(80.0% vs. 44.2%, P < 0.05)、阴性预测值(33.8% vs. 14.8%, P < 0.05)显著高于CMTs；CMTs的总体诊断特异度则优于tNGS (91.2% vs. 70.6%, P < 0.05)。以综合诊断为标准，tNGS共鉴定出细菌166株、真菌48株、病毒28株、非典型病原体34株、分枝杆菌47株；CMTs共鉴定出细菌46株、真菌33株、病毒8株、非典型病原体10株、分枝杆菌33株。tNGS对不同病原菌的诊断效果不同，tNGS诊断总体细菌、真菌、病毒、非典型病原体、分枝杆菌的敏感性、阴性预测值和准确度均高于CMTs，但在诊断曲霉菌、结核分枝杆菌方面，两种方法诊断效能无明显统计学差异。结论：tNGS在疑似肺炎患者诊断中具有更广泛的病原体鉴定和更高的灵敏度和准确度。tNGS可以与传统病原学检测互为补充，为临床诊治提供有效指导。

Abstract: Objective: To explore the diagnostic value of targeted next-generation sequencing (tNGS) in patients with suspected pneumonia. Method: A retrospective analysis was conducted on the medical records of 335 patients with suspected pneumonia. All patients underwent collection of lower respiratory tract specimens (preferably bronchoalveolar lavage fluid), followed by testing with targeted next-generation sequencing (tNGS) and conventional microbiological tests (CMTs). Using a comprehensive diagnostic standard, the diagnostic performance of tNGS and CMTs in patients with suspected pneumonia was evaluated. The distribution of pathogenic microorganisms detected by both methods and their diagnostic efficacy for different types of pathogens were compared. Result: Among the 335 patients, 34 (10.1%) were diagnosed with non-infectious causes, and 301 (89.9%) were diagnosed with infectious pneumonia, with 270 cases confirmed microbiologically. Based on the comprehensive diagnostic standard, tNGS exhibited significantly higher overall diagnostic sensitivity (81.1% vs. 38.9%, P < 0.05), accuracy (80.0% vs. 44.2%, P < 0.05), and negative predictive value (33.8% vs. 14.8%, P < 0.05) than CMTs. However, CMTs demonstrated superior overall diagnostic specificity compared to tNGS (91.2% vs. 70.6%, P < 0.05). According to the comprehensive diagnostic standard, tNGS identified a total of 166 bacterial strains, 48 fungal strains, 28 viral strains, 34 atypical pathogen strains, and 47 mycobacterial strains, while CMTs identified 46 bacterial strains, 33 fungal strains, 8 viral strains, 10 atypical pathogen strains, and 33 mycobacterial strains. The diagnostic performance of tNGS varied for different pathogens. Overall, tNGS showed higher sensitivity, negative predictive value, and accuracy in diagnosing bacteria, fungi, viruses, atypical pathogens, and mycobacteria compared to CMTs. However, there was no statistically significant difference in diagnostic performance between the two methods for Aspergillus and Mycobacterium tuberculosis. Conclusion: tNGS demonstrates a broader range of pathogen identification and higher sensitivity and accuracy in the diagnosis of suspected pneumonia patients. tNGS can complement traditional microbiological testing and provide effective guidance for clinical diagnosis and treatment.