Metagenomic Next-generation Sequencing For Mixed Pulmonary Infection Diagnosis

Objective: Metagenomic next-generation sequencing(m NGS)is emerging as a promising technique for pathogens detection.However,reports on the application of m NGS in mixed pulmonary infection remain scarce.In this study,we evaluated the performance of this approach in the diagnosis of mixed pulmonary infection and detection of pathogens.Methods: 55 cases in Tianjin Medical University General Hospital were enrolled in this retrospective analysis.Pulmonary biopsy and bronchoalveolar lavage fluid of patients with suspected pulmonary infection were collected by bronchoscopy.The lung biopsies were sent to histopathology laboratories within 2 h of collection.A portion of bronchoalveolar lavage fluid was used for culture of aerobic bacteria,anaerobic bacteria,fungi,viruses and mycobacteria.Another part of the bronchoalveolar lavage fluid was used for Xpert MTB,galactomannan test and smear.The remaining specimens were stored at-80°C for m NGS.Results: The sensitivity of m NGS in mixed pulmonary infection diagnosis was much higher than that of conventional test(97.2% vs 13.9%;P < 0.01),but the specificity was the opposite(63.2% vs 94.7%;P = 0.07).A total of 5(9.1%)cases were identified as mixed pulmonary infection by both conventional tests and m NGS.The pathogens detected by m NGS in 3(5.5%)cases were consistent with those by conventional test.According to our data,m NGS had a broader spectrum for pathogen detection than conventional tests.In particular,application of m NGS improved the diagnosis of pulmonary fungal infections.Of 28 cases which had negative results by conventional test,m NGS analysis produced negative results in 4 cases,produced monomicrobial detection in 5 cases and polymicrobial detection in 19 cases.Of 21 cases which were identified as single infection by conventional test,m NGS results were consistent with conventional tests in 2 cases.Another 13 results were partially matched,among which 6 cases were positive for Pneumocystis jirovecii,Staphylococcus epidermidis,Acinetobacter baumannii,Pseudomonas aeruginosa and Klebsiella pneumoniae by culturing.However,m NGS detected more pathogens.Among the other 6 patients,results of conventional tests were paradoxical with that ofm NGS.Of 6 cases which were identified as mixed infection by conventional test,m NGS results were consistent with conventional tests in 1 case.In 4 out of the 6cases,results of m NGS and conventional tests were partially matched.In conclusion,even in the specimen where conventional tests identified mixed pulmonary infection,m NGS still played an important role,because it has the ability to identify both common and rare pathogens without any prior hypothesis.Within the 55 cases,5species of pathogens(Mycobacterium abscessus,Rhizopus,Haemophilus parainflfluenzae,Rhizomucor pusillus and Streptococcus pneumoniae)were identified by m NGS,but not by conventional tests.m NGS detected and identified fungi in 31(56.4%)cases,of which only 10(18.2%)cases were positive for the same fungi by conventional test.Followed by Pneumocystis jirovecii,Pseudomonas aeruginosa,Acinetobacter baumannii,Klebsiella pneumoniae and Aspergillus fumigatus.Due to the high detection rate of Pneumocystis jirovecii in this study,we continued to study the performance of metagenomics next-generation sequencing in the diagnosis of Pneumocystis pneumonia.The sensitivity and specificity of m NGS for P.jirovecii identification were 70.0% and 97.2% or BALF samples.The sensitivity and specificity m NGS were 75.0% and 99.1% for lung biopsy samples.Sensitivity of m NGS(70.0% VS.35.0%;P=0.039)for BALF and for lung biopsy(75.0% VS.35.0%;P=0.021)were both significantly higher than that of GMS staining.Cases with m NGS unique reads ≤100 had a longer duration from disease onset to the sample collection than those with unique reads >100(10 VS.5 days;P=0.038).Conclusions: m NGS is a promising technique to detect co-pathogens in mixed pulmonary infection,with potential benefits in speed and sensitivity.