Etiology of Severe Pneumonia in Children in Alveolar Lavage Fluid Using a High-Throughput Gene Targeted Amplicon Sequencing Assay

Fei Li, Yin Wang, Yuhan Zhang, Peng Shi, Linfeng Cao, Liyun Su, Qiguo Zhu, Libo Wang, Roujian Lu, Wenjie Tan, Jun ShenFrontiers in Pediatrics 2021 · DOI

TL;DR

This study introduces a high-throughput gene targeted amplicon sequencing (TAS) assay for detecting a broad range of pathogens in alveolar lavage fluid (ALF) from children with severe pneumonia. The TAS assay demonstrated comparable performance to conventional methods and multiplex PCR, offering a promising avenue for improved etiological diagnosis in pediatric respiratory infections.

What Problem Does This Paper Address?

Diagnosing the cause of severe community-acquired pneumonia (SCAP) in children is crucial for effective treatment, but current methods have limitations. Traditional culture-dependent diagnostic tests (CDTs) often have low positive rates. While molecular methods like PCR and metagenomic sequencing (mNGS) offer higher sensitivity, mNGS can be complex and prone to contamination issues. This paper aims to evaluate a high-throughput gene targeted amplicon sequencing (TAS) assay as a more accessible and comprehensive alternative for identifying both bacterial and viral pathogens in the lower respiratory tract of children with SCAP, specifically using alveolar lavage fluid (ALF).

Methods

The researchers conducted a retrospective study using 48 frozen ALF samples from 47 children diagnosed with SCAP. They employed three detection methods:

  1. Conventional Tests (CT): This included standard methods like bacterial culture, direct fluorescent antibody (DFA) for specific viruses, and singleplex PCR.
  2. Multiplex PCR (Multi-PCR): A commercial kit designed to detect 22 common respiratory pathogens.
  3. High-Throughput Gene Targeted Amplicon Sequencing (TAS): This novel assay targets a broad panel of 412 pathogenic microorganisms (bacteria, fungi, viruses, parasites).

The study compared the results of the TAS assay against both CT and Multi-PCR, using CT as a "Golden Standard" in one comparison and Multi-PCR as the "Golden Standard" in another. They also explored an "ID Reading" approach where infectious disease doctors interpreted the raw TAS data in conjunction with clinical features.

Datasets

The study utilized a dataset of 48 alveolar lavage fluid (ALF) samples collected from 47 pediatric patients with severe community-acquired pneumonia (SCAP) at the Children's Hospital of Fudan University between January 1, 2019, and March 31, 2019. No new benchmarks were introduced; the study focused on evaluating the TAS assay against existing diagnostic methods.

Key Insights and Ideas

  • TAS as a Broad-Spectrum Diagnostic Tool: The TAS assay demonstrated its ability to detect a wide array of pathogens, including bacteria and viruses, in a single test. This is a significant advantage over singleplex PCR or DFA, which target specific organisms.
  • Comparable Performance to Established Methods: While not always outperforming, the TAS assay showed comparable diagnostic value to conventional tests (CT) and multiplex PCR, suggesting it can reliably identify pathogens in ALF.
  • Potential for Uncovering Missed Diagnoses: In some instances, TAS identified pathogens that were missed by CT, highlighting its potential to improve diagnostic yield.
  • High Co-infection Rates: The study confirmed that mixed infections (co-detection of multiple pathogens) are common in severe pediatric pneumonia, a finding consistent across all three tested methods.
  • Need for Standardized Interpretation: The authors noted that interpreting raw TAS data can be challenging, especially distinguishing between infection, colonization, and contamination. They suggest that further research is needed to establish standardized interpretation criteria for TAS reads and analysis platforms.

Key Results

  • CT Detection Rate: Conventional tests (CT) detected pathogens in 81.2% (39/48) of samples.
  • TAS vs. CT: When CT was used as the "Golden Standard," TAS detected pathogens in 70.8% (34/48) of samples. The difference in detection rates was not statistically significant (p = 0.232). TAS showed a sensitivity of 87.1% and a specificity of 100.0% compared to CT.
  • TAS vs. Multi-PCR: When Multi-PCR was used as the "Golden Standard," TAS detected pathogens in 83.3% (40/48) of samples. This showed a statistically significant difference compared to Multi-PCR (p = 0.003), with TAS having a lower detection rate in this comparison.
  • Dominant Pathogens: Mycoplasma pneumoniae (MP) and Adenovirus (ADV) were consistently the most frequently detected pathogens across all three methods.
  • Co-infection: Mixed infections were identified in a substantial proportion of samples, with rates varying by method (e.g., 35.4% by TAS based on CT or Multi-PCR).

Limitations and Future Work

  • Sample Size and Study Design: The study had a relatively small sample size (48 samples) and was a single-center, retrospective study. This limits the generalizability of the findings.
  • Distinguishing Infection from Colonization/Contamination: A key limitation of nucleic acid-based methods like TAS is the difficulty in differentiating between active infection, colonization, or contamination, especially with low read counts. The authors acknowledge that some detected microorganisms might be part of the normal flora or represent contamination.
  • Interpretation Criteria: The lack of standardized interpretation criteria for TAS reads is a significant hurdle for widespread clinical adoption. The authors emphasize the need for further research to develop robust interpretation guidelines.
  • Future Directions: The authors anticipate that high-throughput gene TAS assays will become standard in clinical microbiology laboratories in the coming decade. They suggest that future research should focus on establishing clear interpretation criteria and potentially expanding the pathogen targets within the TAS panel.

Summary updated

Abstract

<b>Objective:</b> To evaluate the diagnostic value of a high-throughput gene targeted amplicon sequencing (TAS) assay for detecting pathogenic microorganisms in alveolar lavage fluid (ALF) from children with severe community-acquired pneumonia (SCAP). <b>Methods:</b> A retrospective study was performed on 48 frozen ALF samples from 47 severe pneumonia cases admitted to Children's Hospital of Fudan University from January 1, 2019, to March 31, 2019. All samples were tested by a multiplex PCR (Multi-PCR) assay and a TAS assay. The results of the TAS panels were parallel compared with Multi-PCR and Conventional Tests (CT) including culture, direct fluorescent antibody method (DFA), and singleplex polymerase chain reaction (PCR). <b>Results:</b> The proportion of pathogens detection by CT was 81.2% (39/48). The 8 common respiratory viruses including respiratory syncytial virus (RSV), adenovirus (ADV), influenza A virus (FLUA), influenza B virus (FLUB), parainfluenza virus 1-3 (PIV1-3), and human Metapneumovirus (hMPV) were found in 31.2% (15/48) of the 48 samples by DFA. With the criteria of CT results used as "Golden Standard" for determing of TAS results, the proportion of pathogens detection by TAS was 70.8% (34/48). The difference of proportion of pathogens detection between TAS and CT was not statistically significant (<i>p</i> = 0.232). The sensitivity and specificity of TAS for pathogens detection based on CT were 87.1% (95% CI, 71.77-95.18%) and 100.0% (95% CI, 62.88-100%), the positive predictive value (PPV) and negative predictive value (NPV) were 100.0% (95% CI, 87.35-100%) and 64.2% (95% CI, 35.62-86.02%), respectively. While Multi-PCR results were used as "Golden Standard," the total pathogens detection rate of TAS was 83.3% (40/48), which had a significant difference with that of Multi-PCR (<i>p</i> = 0.003). The sensitivity and PPV of TAS compared with Multi-PCR were 83.3% (95% CI, 69.23-92.03%) and 100.0% (95% CI, 89.08-100%), respectively. High rates of co-infection were proved by CT, Multi-PCR, and TAS. Mycoplasma pneumoniae (MP) and ADV were the two most frequently detected pathogens in all three assays. <b>Conclusion:</b> Compared with the CT and Multi-PCR methods, this TAS assay had a good performance in detecting bacteriological and viral pathogens from ALF. More research is needed to establish interpretation criteria based on TAS reads or analysis platforms.
BibTeX
@inproceedings{li2021etiology,
  author = {Fei Li and Yin Wang and Yuhan Zhang and Peng Shi and Linfeng Cao and Liyun Su and Qiguo Zhu and Libo Wang and Roujian Lu and Wenjie Tan and Jun Shen},
  title = {{Etiology of Severe Pneumonia in Children in Alveolar Lavage Fluid Using a High-Throughput Gene Targeted Amplicon Sequencing Assay}},
  booktitle = {Frontiers in Pediatrics},
  year = {2021},
  doi = {10.3389/fped.2021.659164},
}
Figure