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

Abstract

Introduction: Pneumonia are the leading cause of death worldwide, and antibiotic treatment remains fundamental. However, conventional sputum smears or cultures are still inefficient for obtaining pathogenic microorganisms.Metagenomic next-generation sequencing (mNGS) has shown great value in nucleic acid detection, however, the NGS results for lower respiratory tract microorganisms are still poorly studied.

Methods: This study dealt with investigating the efficacy of mNGS in detecting pathogens in the lower respiratory tract of patients with pulmonary infections. A total of 112 patients admitted at the First Affiliated Hospital of Zhengzhou University between April 30, 2018, and June 30, 2020, were enrolled in this retrospective study. The bronchoalveolar lavage fluid (BALF) was obtained from lower respiratory tract from each patient. Routine methods (bacterial smear and culture) and mNGS were employed for the identification of pathogenic microorganisms in BALF.

Results: The average patient age was 53.0 years, with 94.6% (106/112) obtaining pathogenic microorganism results. The total mNGS detection rate of pathogenic microorganisms significantly surpassed conventional methods (93.7% vs. 32.1%, P < 0.05). Notably, 75% of patients (84/112) were found to have bacteria by mNGS, but only 28.6% (32/112) were found to have bacteria by conventional approaches. The most commonly detected bacteria included 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%). In 29.5% (33/112) of patients, fungi were identified using mNGS, including 23 cases of Candida albicans (20.5%), 18 of Pneumocystis carinii (16.1%), and 10 of Aspergillus (8.9%). However, only 7.1 % (8/112) of individuals were found to have fungi when conventional procedures were used. The mNGS detection rate of viruses was significantly higher than the conventional method rate (43.8% vs. 0.9%, P < 0.05). The most commonly detected viruses included Epstein-Barr virus (15.2%), cytomegalovirus (13.4%), circovirus (8.9%), human coronavirus (4.5%), and rhinovirus (4.5%). Only 29.4% (33/112) of patients were positive, whereas 5.4% (6/112) of patients were negative for both detection methods as shown by Kappa analysis, indicating poor consistency between the two methods (P = 0.340; Kappa analysis).

Conclusion: Significant benefits of mNGS have been shown in the detection of pathogenic microorganisms in patients with pulmonary infection. For those with suboptimal therapeutic responses, mNGS can provide an etiological basis, aiding in precise anti-infective treatment.

Keywords: antibiotics; etiology; lower respiratory tract; metagenomic next-generation sequencing; pulmonary infection.