Comparison of next-generation sequencing with traditional methods for pathogen detection in cases of lower respiratory tract infection at a community hospital in Eastern China

Abstract

Lower respiratory tract infection (LRTI) is still a threat to human health. Metagenomics next-generation sequencing (NGS) provides an efficient and unbiased way to identify LRTI pathogens, and has been shown to have several advantages over traditional methods. However, its application is currently limited in low-resource settings. Our aim was to collect and analyze data on LRTI cases at a county-level community hospital in Eastern China over one year, in order to compare the efficiency of NGS and traditional methods including culture, nucleic acid amplification and antibody techniques. We performed NGS of bronchoalveolar lavage fluid (BALF) for pathogen identification in 71 patients with LRTI. We compared the detection rates, identified pathogens, and turnaround time of NGS with traditional methods. Pathogens were detected using traditional methods in 19 cases, and the results were compared with those obtained with the NGS technique in 60 cases. The pathogen detection rate of NGS (84.5%) was much higher than that of the traditional methods (26.8%). Moreover, with the traditional methods considered the gold standard, the consistency rate between NGS and traditional methods was 68.4%. For the 19 cases in which the traditional method was used, the main pathogens included invasive Aspergillus (5 cases), Pseudomonas aeruginosa (3 cases), Candida albicans (3 cases), and Staphylococcus aureus (2 cases). Among the 60 cases detected by NGS, the main pathogens included Mycobacterium (12 cases), Streptococcus pneumoniae (5 cases), Klebsiella pneumoniae (3 cases), P. aeruginosa (3 cases), Haemophilus influenzae (3 cases), and S. aureus (3 cases), Aspergillus (9 cases), Pneumocystis jiroveci (5 cases), C. albicans (3 cases), Human Papilloma Virus (9 cases), Epstein-Barr virus (8 cases), and parvovirus (6 cases). In addition, 2 cases of chlamydia and 1 case of mycoplasma infection were detected by NGS. The time taken to perform the NGS tests was significantly shorter than that taken with the traditional method. NGS analysis of bronchoalveolar lavage fluid, in combination with traditional pathogen detection methods, can improve the efficiency of pathogen detection. More attention should be paid to the regional epidemic characteristics of infectious pathogens in LRTI.