Etiology of lower respiratory tract in pneumonia based on metagenomic next-generation sequencing: a retrospective study

Abstract The purpose of this study was to explore the ability of metagenomic next-generation sequencing (mNGS) to detect pathogens in lower respiratory tract of patients with pulmonary infection. A retrospective study was conducted on 112 patients with pulmonary infection admitted to the First Affiliated Hospital of Zhengzhou University from April 30th 2018 to June 30th 2020. Sputum and pharyngeal swabs were collected from all patients, and bronchoalveolar lavage fluid (BALF) was collected during bronchoalveolar lavage. Respiratory tract pathogenic microorganisms were detected by routine methods (bacterial smear, PCR, culture, etc.), and BALF pathogenic microorganisms were detected by mNGS. The average age of the patients was 53.0 years old and 94.6% (106/112) of the patients obtained the results of pathogenic microorganisms. The overall detection rate of pathogenic microorganisms by mNGS was significantly higher than that of conventional methods (93.7% vs. 33.1%, P <0.05). mNGS detected bacteria in 75.0% (84/112) of patients, while conventional methods detected bacteria in only 28.6% (32/112) of all patients. The most commonly detected bacteria were Acinetobacter baumannii (19.6%), Klebsiella pneumoniae (17.9%), Pseudomonas aeruginosa (14.3%), Staphylococcus faecium (12.5%), Enterococcus faecium (12.5%) and Haemophilus parainfluenzae (11.6%). Fungi were detected by mNGS in 29.5% (33/112) of patients, including 23 cases of Candida albicans (20.5%), 18 cases of Pneumocystis carinii (16.1%), and 10 cases of Aspergillus (8.9%). Conventional methods detected fungi only in 7.1% (8/112) of patients. The detection rate of virus by mNGS was significantly higher than that by conventional methods (43.8% vs.0.9%, P <0.05). The most commonly detected viruses were EBV (15.2%), CMV (13.4%), circovirus (8.9%), human coronavirus (4.5%) and rhinovirus (4.5%). The results of Kappa analysis showed that the consistency of the two methods was poor ( P = 0.276), only 32.1% (36/112) of patients were positive for both methods, and 5.4% (6/112) of patients were negative for both methods. mNGS shows great advantages in the detection of pathogenic microorganisms in patients with pulmonary infection. For those with poor therapeutic effect, mNGS detection can obtain etiological basis as far as possible, which is helpful to achieve accurate anti-infective treatment.