Utilizing metagenomic next-generation sequencing for diagnosis and lung microbiome probing of pediatric pneumonia through bronchoalveolar lavage fluid in pediatric intensive care unit: results from a large real-world cohort

Front Cell Infect Microbiol. 2023 Aug 15:13:1200806. doi: 10.3389/fcimb.2023.1200806. eCollection 2023.

Abstract

Background: Metagenomic next-generation sequencing (mNGS) is a powerful method for pathogen detection in various infections. In this study, we assessed the value of mNGS in the pathogen diagnosis and microbiome analysis of pneumonia in pediatric intensive care units (PICU) using bronchoalveolar lavage fluid (BALF) samples.

Methods: A total of 104 pediatric patients with pneumonia who were admitted into PICU between June 2018 and February 2020 were retrospectively enrolled. Among them, 101 subjects who had intact clinical information were subject to parallel comparison of mNGS and conventional microbiological tests (CMTs) for pathogen detection. The performance was also evaluated and compared between BALF-mNGS and BALF-culture methods. Moreover, the diversity and structure of all 104 patients' lung BALF microbiomes were explored using the mNGS data.

Results: Combining the findings of mNGS and CMTs, 94.06% (95/101) pneumonia cases showed evidence of causative pathogenic infections, including 79.21% (80/101) mixed and 14.85% (15/101) single infections. Regarding the pathogenesis of pneumonia in the PICU, the fungal detection rates were significantly higher in patients with immunodeficiency (55.56% vs. 25.30%, P =0.025) and comorbidities (40.30% vs. 11.76%, P=0.007). There were no significant differences in the α-diversity either between patients with CAP and HAP or between patients with and without immunodeficiency. Regarding the diagnostic performance, the detection rate of DNA-based BALF-mNGS was slightly higher than that of the BALF-culture although statistically insignificant (81.82% vs.77.92%, P=0.677) and was comparable to CMTs (81.82% vs. 89.61%, P=0.211). The overall sensitivity of DNA-based mNGS was 85.14% (95% confidence interval [CI]: 74.96%-92.34%). The detection rate of RNA-based BALF-mNGS was the same with CMTs (80.00% vs 80.00%, P>0.999) and higher than BALF-culture (80.00% vs 52.00%, P=0.045), with a sensitivity of 90.91% (95%CI: 70.84%-98.88%).

Conclusions: mNGS is valuable in the etiological diagnosis of pneumonia, especially in fungal infections, and can reveal pulmonary microecological characteristics. For pneumonia patients in PICU, the mNGS should be implemented early and complementary to CMTs.

Keywords: bronchoalveolar lavage fluid; diagnosis; lung microbiome; metagenomic next-generation sequencing; pediatric intensive care unit; pneumonia.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Bronchoalveolar Lavage Fluid
  • Child
  • High-Throughput Nucleotide Sequencing
  • Humans
  • Intensive Care Units, Pediatric
  • Lung
  • Microbiota* / genetics
  • Pneumonia* / diagnosis
  • Retrospective Studies

Grants and funding

This work was supported by the National Key Research and Development Program of China (2021YFC2701800, 2021YFC2701805), the Scientific Research Project of Shanghai Municipal Health Commission (No. 202140442), National Natural Science Foundation of China (82202374 to CZ), Shanghai Committee of Science and Technology (22ZR1408500 to CZ).