Clinical Features and Value of Tracheal Aspirate Metagenomic Next-Generation Sequencing for Severe Pneumonia in Children in Pediatric Intensive Care Unit

Pneumonia is a significant cause of morbidity and mortality in children (Liu et al. 2015). Severe pneumonia accounts for 7–13% of these cases (Rudan et al. 2008; Black et al. 2010; Rudan et al. 2013). Due to its rapid onset, severe symptoms, rapid progression, and high incidence of complications, accurate diagnosis and treatment of severe pneumonia are crucial for reducing mortality rates and improving cure rates. However, identifying the pathogen that causes pneumonia can be challenging due to the diverse clinical symptoms that vary depending on children’s age, gender, immune status, and underlying diseases (Shi et al. 2020; Anteneh et al. 2023).

The distribution of pathogens in severe pneumonia cases among children of different age groups was studied. Respiratory syncytial virus (RSV) was detected more frequently in younger children, while parainfluenza, adenovirus, and influenza were identified more commonly in older children (Ning et al. 2017; Chen et al. 2018; Shin et al. 2018; Zhao et al. 2019; Chen et al. 2023). Older children were more likely to be infected with multiple pathogens, with mycoplasma and viruses commonly found in their pathogenic spectrum (Chen et al. 2023). Nonetheless, there remains a lack of research on the changes in respiratory system microbial components among children of different ages.

Pathogens, the lung microbiome, and the host response are the three core elements of respiratory tract infections (Zhang et al. 2021). Altering the respiratory microflora early in life may lead to the development of respiratory diseases. Streptococcus, Moraxella, or Haemophilus were the dominant taxa for most children aged 6 months to 5 years, and Streptococcus, Prevotella, or Pasteurellaceae were the dominant taxa for children aged 5 to 18 years (Pettigrew et al. 2016). Additionally, specific taxa in the lung microbiota are linked to the severity of community-acquired pneumonia in children (Pettigrew et al. 2016). No further comparative analysis between the age groups was carried out in the previous study.

Tracheal aspirate (TA) sampling is less invasive than mini-bronchoalveolar lavage (mBAL). Nevertheless, TA was previously considered a suboptimal sample due to contamination by oral microbes. Previous studies showed that the lung microbiome composition is similar between TA and mBAL in patients with LRTI (Kalantar et al. 2019). TA samples can also better reflect the lung microbiome of children with LRTI (Mick et al. 2023). The Infectious Diseases Society of America (IDSA) and American Thoracic Society (ATS) currently recommend non-invasive TA sampling with semiquantitative cultures over other investigative techniques, including BAL, in the diagnosis of ventilator-associated pneumonia (VAP) in adults (Kalil et al. 2016). Katayama et al. (2010) demonstrated that gram staining of TAs is highly effective in guiding antibiotic therapy for VAP in extremely preterm neonates with lifelong hospital admission and mean onset of VAP at > 30 days. These results show that TA can be used for pathogen identification and lung microbiome research. The prospects of studying the lung microbiome include various aspects, such as investigating its potential as a biomarker for diagnosing respiratory diseases, exploring its spatial dynamics, and understanding the mechanisms underlying its interaction with the host (Yi et al. 2022). Metagenomic next-generation sequencing (mNGS) is an emerging method that has shown significant advantages in the unbiased detection of pathogens causing various infectious diseases (Gu et al. 2019; Li et al. 2022b, 2022a; Chen et al. 2024; Wu et al. 2023). Additionally, it can analyze the microbial landscape of the respiratory tract (Langelier et al. 2018; Zhang et al. 2023). mNGS has been widely used to study upper respiratory tract samples in children (Ogunbayo et al. 2023). However, its application in less invasive lower respiratory tract samples, such as tracheal aspirates, remains in the preliminary exploration stage. Systematic research on its clinical utility in diagnosing pediatric lung infections is still lacking. Thus, in-depth exploration of metagenomics in this sample type could elucidate the pathogen profile and microbiome composition in childhood pneumonia and provide new insights for optimizing clinical diagnosis and treatment strategies.

The goal of etiologic diagnosis and lung microbiome research is to achieve precision medicine. The purpose of this study was to investigate the relationship between clinical features, specific pathogen distribution, and microbiome characteristics in children of different ages with severe pneumonia. The findings have significant implications for the clinical management of infectious diseases and future research directions.

Respiratory microbes undergo dramatic changes in the early stages of childhood (Biesbroek et al. 2014). Several studies have suggested that respiratory microbes play a key role in the pathogenesis of respiratory infections (Hurst et al. 2022). From infancy, microbial diversity increases significantly with age (Hurst et al. 2022). The action of external factors has a long-term impact on establishing the microbiota. These effects may be due to the natural environment, diet, seasons, exposure history, drug use, and lifestyle habits (McCauley et al. 2022). With advancing age, the immune system shapes microbiota diversity through regulatory mechanisms such as immune tolerance, selective immune responses, and adaptation (Kloepfer and Kennedy 2023). Therefore, this study aimed to comprehensively evalate the association between these microorganisms and the clinical characteristics of patients with severe pneumonia in different age groups.

Our study found that mNGS demonstrated good consistency and a broad pathogen profile in children with severe pneumonia when CMT was used as the reference standard, with RSV, S. pneumoniae, and C. albicans being the most common pathogens. Pathogen distribution and microbiota composition vary with age. In our study, co-infection was common (80%), particularly bacterial-viral co-infection, consistent with previous studies (Li M et al. 2023; Shen et al. 2023). These co-infections correlated with disease severity, and RSV co-infection with H. influenzae, S. aureus, and rhinovirus significantly increased the incidence of severe pneumonia (Cebey-López et al. 2016; Meskill and O’Bryant 2020; Liu et al. 2023). In this study, one case of co-infection involving P. jirovecii and CMV was identified. Previous studies have indicated that co-infection-induced pneumonia can have fatal consequences, particularly in infants and young children, if not promptly diagnosed and treated (Williams et al. 2001; Lyu et al. 2022). Investigating pathogen co-infections provides valuable insights for clinicians, helping them make more accurate diagnoses and treatment decisions. In children with severe pneumonia, mNGS can help doctors identify complex co-infections in a timely manner, enabling them to select targeted antimicrobial drugs for combination therapy and reduce unnecessary drug use.

By analyzing the respiratory microbiota of children with severe pneumonia, we found that bacteria and viruses are more likely to coexist in older children. Over time, older children’s immune systems mature, and environmental exposure increases the diversity of microorganisms in the lower respiratory tract (Muhlebach et al. 2018; Zhu et al. 2021). Most children under one year of age have low bacterial densities in the lower respiratory tract due to the immaturity of their immune system (Renz et al. 2018). This increased diversity may also contribute to a higher risk of opportunistic infections by colonizing bacteria. The LEfSe analysis revealed differential species between the two age groups. RSV and S. mitis are microbiological markers of severe pneumonia in children under 12 months of age, while HBoV-1 and P. histicola are more commonly found in children aged 12 to 144 months. This study highlights RSV as the main causative species, consistent with previous research (Esposito et al. 2013; Khuri-Bulos et al. 2018). Additionally, the uneven distribution of RSV among hospitalized children of different ages supports the idea that younger age is closely associated with severe RSV infection, as mentioned by Do et al. (2020) and Gonapaladeniya et al. (2021). The difference is that more S. mitis is found in children under 12 months of age. S. mitis, a respiratory colonizing bacterium, has been reported to inhibit the growth of various respiratory pathogens (Men et al. 2025). Our study found that S. mitis may be involved in developing severe pneumonia as a potential pathogen in infants with incomplete immune function. HBoV-1 was detected only in children aged 12 to 144 months, which differs from previous studies. Several studies have found that severe HBoV pneumonia most commonly occurs in younger children under 2 years of age, particularly in infants, who are more likely to develop HBoV co-infection (Zhang et al. 2018; Li Z et al. 2025). The epidemiological characteristics of different regions may influence these differences. Rapid identification of early species-specific infections is essential for prompt treatment in infants and children. When RSV and S. mitis are detected in infants, and HBoV-1 and P. histicola are detected in children, clinicians should focus more on controlling disease exacerbations.

The microbiome and host interaction significantly influences human health (Vayssier-Taussat et al. 2014; Kumpitsch et al. 2019). Our study showed that the high abundance of microorganisms in TA samples from children was closely related to inflammatory markers, especially S. peroris, HBoV-1, Staphylococcus oralis, S. pneumoniae, and Streptococcus parasanguinis. Previous studies have also found that an increased abundance of colonizing microorganisms may induce more significant epithelial damage and the expression of inflammatory factors, leading to exacerbation of the disease (McCauley et al. 2019; Kloepfer and Kennedy 2023).

However, there are several limitations worth noting. Firstly, this study is retrospective, potentially introducing a certain degree of selection bias. Secondly, the sample size was relatively small, consisting of only 55 patients from a single center, which may limit the statistical power. Thirdly, although TA mNGS is superior to CMT in detecting infections in children with severe pneumonia, the actual clinical situation is that not all patients undergo CMT testing of TA samples.

Therefore, we used CMT results from different sampling sites for comparative analysis. Differences in sample types may impact the interpretation of the results. Lastly, not all viral events detected by mNGS underwent confirmatory PCR testing.

Additionally, our study is retrospective, and a single sampling site cannot thoroughly analyze the dynamics of microorganisms throughout the disease course. Previous studies have conducted longitudinal research on adult lung microbes to explore the relationship between microbes, disease, and disease progression after treatment. In future studies, longitudinal research on the microbial communities of children with severe pneumonia using mNGS detection technology can better clarify the disease’s potential mechanisms and therapeutic effects.

In conclusion, we performed a comprehensive analysis to elucidate the characteristics of pediatric patients suffering from severe pneumonia, specifically focusing on variations in clinical symptoms, characteristic indices, pathogens, and the microbiome among various age groups. RSV, S. pneumoniae, and C. albicans were the dominant pathogens, and co-infections were the primary cause of severe pneumonia. Older children showed increased microbial richness and diversity compared to infants, with representative microorganisms identified in both groups. These results provide a solid basis for enhancing our understanding of the relationship between severe pneumonia and pulmonary microorganisms in children.