Abstract
Objective
This study characterized the microbial spectrum and clinical manifestations in hospitalized children with acute respiratory infections (ARI) using targeted next-generation sequencing (tNGS) and identified predictors for severe disease requiring ICU admission.
Methods
10,153 hospitalized children with ARI were enrolled. Microbial detection was performed using tNGS on pediatric respiratory tract samples. Epidemiological trends, clinical features, laboratory and imaging findings, and outcomes were analyzed. Logistic regression identified factors associated with ICU admission.
Results
The microbial detection rate was 97.77%, with a co-detection rate of 82.87%. Distinct clinical profiles emerged: human respiratory syncytial virus (HRSV) and human bocavirus were associated with younger age and more severe disease; Mycoplasma pneumoniae (MP) with frequent respiratory complications yet lower overall severity, particularly in older children; and adenovirus and Haemophilus influenzae with milder illness. A bidirectional severity pattern was identified: monodetections of higher-virulence agents (e.g., HRSV, Acinetobacter baumannii, human bocavirus) correlated with more severe disease than their co-detections, whereas the opposite was true for lower-pathogenicity microbes such as MP, adenovirus, and Haemophilus influenzae. ICU patients had significantly higher detection rates of HRSV (29.01% vs. 17.53%), cytomegalovirus (25.66% vs. 19.75%), and Acinetobacter baumannii (19.25% vs. 12.51%) (all P < 0.001). Independent predictors for ICU admission included respiratory complications (OR = 8.289), fever (OR = 7.338), and other system complications (OR = 5.564).
Conclusion
This large-scale study uncovers complex co-detection patterns and microbe-dependent outcomes in ARI. A novel bidirectional severity pattern and distinct ICU microbial profiles with robust predictive factors have been identified, aiding early risk stratification and tailored management.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12967-026-08164-4.
Keywords: Respiratory infections, Microbial spectrum, Co-detection, Pediatric patients, tNGS
Introduction
Respiratory infections remain a leading global cause of morbidity and hospitalization in children under five, posing a substantial threat to child health and imposing significant medical and economic burdens [1, 2]. Their etiology involves diverse co-circulating pathogens with distinct seasonal and regional patterns [3]. Advances in molecular diagnostics have further revealed the high frequency of concurrent microbial detection (co-detection) [4], which, compared to single-pathogen detection, is often associated with more complex presentations, higher complication rates, and greater disease severity [5–7]. The developing immune systems of children heighten their susceptibility to these pathogens [8]. A detailed understanding of the respiratory microbial spectrum and its clinical correlates is therefore essential for developing effective prevention and treatment strategies.
Targeted next-generation sequencing (tNGS) integrates multiplex PCR with deep sequencing to rapidly and accurately detect dozens to hundreds of microbial targets in clinical samples. Compared with traditional culture methods, tNGS offers superior sensitivity and specificity, particularly in low-biomass samples [9]. A key advantage of tNGS is its ability to detect multiple microbes in a single assay, which is critical for identifying co-detection—a common scenario in pediatric respiratory illnesses where children frequently harbor diverse microbes simultaneously [10]. Furthermore, the rapid turnaround time of tNGS offers the potential to inform timely clinical decisions and may contribute to reducing unnecessary antibiotic use [11].
Although tNGS has been increasingly used to study pediatric respiratory infections [12–17], two key limitations persist in the existing literature: limited statistical power due to small sample sizes (typically a few hundred cases) and a narrow analytical scope often restricted to single-pathogen studies or fragmented assessments, rather than a systematic evaluation of polymicrobial detection. To address these gaps, we analyzed a tNGS-based clinical-microbiological dataset from 10,153 children hospitalized with acute respiratory illness, systematically characterizing microbial distributions, co-detection patterns, and their clinical correlates. Our specific aims were to: (1) identify prevalent microbes and their associated clinical profiles through integrated analysis of laboratory, imaging, and clinical data; (2) elucidate agent-specific manifestations under both mono- and co-detection conditions; and (3) identify high-risk subgroups and predictors of severe disease to inform risk-stratified clinical management. By providing a comprehensive, real-world epidemiological picture of respiratory agents and their clinical impact, this study aims to generate evidence that can ultimately support the translation of agent-informed diagnostics into more tailored therapeutic strategies for pediatric respiratory disease.
Materials and methods
Study population
This retrospective analysis included 10,153 pediatric patients with acute respiratory tract infection (ARI) hospitalized at the Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region from April 2021 to December 2024. ARI was characterized as an acute onset of respiratory symptoms lasting up to 21 days. Common symptoms encompassed cough, sputum production, shortness of breath, sore throat, and rhinorrhea. The condition mainly refers to acute upper respiratory tract infections, acute bronchitis, and community-acquired pneumonia (CAP).
Eligibility criteria
The study enrolled pediatric patients (≤ 14 years) diagnosed with ARI and with complete tNGS data. Exclusion criteria comprised: (1) absence of key clinical or laboratory records (e.g., demographics, tNGS reports, final diagnosis); (2) acute respiratory inflammation due to non-infectious etiologies like allergies, asthma, environmental irritants; and (3) a diagnosis of hospital-acquired (nosocomial) infection.
Targeted next-generation sequencing
10,153 specimens were analyzed via tNGS, comprising 4,578 bronchoalveolar lavage fluid samples, 4,648 throat swabs, and 927 sputum samples. Nucleic acids were extracted and purified using the MagPure Pathogen DNA/RNA Kit (R6672-01B, Magen, Guangzhou, China). Multiplex PCR was performed using the Respiration100™ library system (KingCreate, Guangzhou, China) to detect common respiratory pathogens. Throughout the extended duration of the study, iterative updates to the detection panel significantly expanded its coverage, increasing the number of detectable agents from an initial 65 to 198 (Supplementary Tables 1 and 2). Post-library preparation, products were bead-purified and amplified with barcoded primers. The quality and concentration of the final library were assessed respectively with a Qsep100 Biofragment Analyzer (Guangding Bio) and a Qubit 4.0 Fluorometer (Thermo Scientific). Sequencing was performed on the Illumina MiniSeq platform.
Following adapter trimming and removal of low-quality sequences with fastp v0.20.1 [18], reads longer than 50 bp with over 75% of bases achieving a Q30 score were retained for downstream analysis. These quality-filtered reads were aligned to a curated reference database (sourced from NCBI’s GenBank, RefSeq, and Nucleotide databases) using Bowtie2 v2.4.1 [19]. Interpretation of targeted sequencing results was based on amplicon coverage and normalized read count. A microorganism was considered detected if it met the following criteria: (i) For bacteria (excluding the Mycobacterium tuberculosis complex), fungi, and atypical agents: amplicon coverage ≥ 50% and normalized read count ≥ 10; (ii) For viruses: amplicon coverage ≥ 50% and normalized read count ≥ 3, or normalized read count ≥ 10; (iii) For the Mycobacterium tuberculosis complex: normalized read count ≥ 1.
Subsequently, two experienced clinicians independently evaluated the patients’ clinical data to assess the association between the acute respiratory illness and potential clinically relevant agents. This comprehensive assessment incorporated the patient’s medical history, symptoms, imaging, tNGS, and laboratory results. To establish a final determination, a structured set of integrated criteria was applied to differentiate genuine infection from colonization. These criteria were synthesized into a multidimensional diagnostic framework encompassing four key aspects: (1) host factors (e.g., immunocompromised states such as neutropenia, HIV infection, organ transplantation, or prematurity); (2) microbiological evidence (including high pathogen load via tNGS, low cycle threshold [Ct] values in PCR assays, microscopic observation of invasive forms such as hyphae or intracellular organisms, and sampling from sterile sites); (3) clinical and imaging correlation (such as new or progressive pulmonary infiltrates, ground-glass opacities, viral interstitial patterns, and elevated inflammatory markers including procalcitonin [PCT], C-reactive protein [CRP], galactomannan [GM], or virus-specific antibody responses); and (4) treatment response (e.g., clinical improvement within 72 h of initiating targeted antimicrobial therapy or confirmed virological clearance). Where the two clinicians’ initial assessments differed, a consensus was reached through further consultation with a senior physician. This integrated approach enhances diagnostic accuracy in distinguishing true infection from colonization.
For this study, only microorganisms that passed this comprehensive clinical adjudication process and were ultimately classified as ‘suspected pathogens’ were included in the subsequent epidemiological analyses, co-detection calculations, and subgroup comparisons (e.g., ICU vs. non-ICU). Microbial detections that were deemed by the clinicians to represent colonization, environmental contamination, or background airway microbiota were excluded from the final analytical dataset.
Statistical methods
Statistical analysis was performed using IBM SPSS 26.0. Quantitative data conforming to normal distribution and homogeneity of variance are presented as mean ± standard deviation (x̄±s) and compared via the independent samples t-test. Data with non-normal distributions are expressed as median (interquartile range) and analyzed using the Mann-Whitney U test. Qualitative data are presented as frequency (percentage, %) and compared via the chi-square test. Univariate and multivariate logistic regression analyses were conducted to identify factors independently associated with ICU admission. In this study, “poor prognosis” was defined as in-hospital mortality or discharge against medical advice due to poor treatment response. P < 0.05 was considered statistically significant.
Results
Baseline characteristics
Among the 10,153 hospitalized children with ARI, 6,071 (59.80%) were male and 4,082 (40.20%) were female (male-to-female ratio 1.49:1). The age of the cohort ranged from 1 day to 14 years, with a median of 24 months (IQR: 8–54). Age distribution was as follows: ≤1 year (3,753, 36.96%), 1–3 years (2,507, 24.69%), 3–5 years (1,895, 18.66%), and > 5 years (1,998, 19.68%).
Detection rate and profile of predominant microbes
Following clinical adjudication of the tNGS results using the integrated criteria, suspected pathogens were identified in 9,927 (97.77%) of the 10,153 children. The positive samples included 4,511 bronchoalveolar lavage fluid, 4,574 throat swabs, and 842 sputum specimens, with detection rates of 98.54%, 98.41%, and 90.83%, respectively.
The ten most frequently detected agents, in descending order of prevalence, were: Haemophilus influenzae (Hi; 30.4%, 3,088/10,153), Mycoplasma pneumoniae (MP; 30.4%, 3,086/10,153), rhinovirus (26.5%, 2,686/10,153), Streptococcus pneumoniae (21.2%, 2,153/10,153), cytomegalovirus (CMV; 20.2%, 2,048/10,153), human respiratory syncytial virus (HRSV; 18.3%, 1,862/10,153), adenovirus (13.8%, 1,396/10,153), Staphylococcus aureus (13.5%, 1,369/10,153), Acinetobacter baumannii (13.0%, 1,318/10,153), and Moraxella catarrhalis (10.8%, 1,095/10,153) (Fig. 1). The detection rates of these microbes also varied considerably across different age groups. For instance, MP was detected in up to 54.6% of children aged 5–14 years, Hi in 42.6% of those aged 3–5 years, and CMV in 37.3% of children aged ≤ 1 year (Fig. 1). Microbial profiles differed significantly by sample type. In BALF, MP (47.2%) was most common, followed by rhinovirus (26.7%) and Hi (24.2%). In throat swabs, Hi (37.7%) predominated, followed by rhinovirus (25.8%) and Acinetobacter baumannii (23.2%). In sputum, the leading microbes were rhinovirus (28.4%), Hi (24.8%), and Streptococcus pneumoniae (19.4%) (Supplementary Fig. 1).
Fig. 1.
Distribution of commonly detected agents in hospitalized children with respiratory tract illness. RV: rhinovirus, CMV: cytomegalovirus, HRSV: human respiratory syncytial virus, AdV: adenovirus, HPIV-3: human parainfluenza virus type 3, EBV: Epstein-Barr virus, HMPV: human metapneumovirus, HBoV: human bocavirus, EV: enterovirus, IAV: influenza A virus, HHV-7: human herpesvirus 7, HCoV: human coronavirus, IBV: influenza B virus, HPIV-1: human parainfluenza virus type 1, HHV-6: human herpesvirus 6, HHV-1: human herpesvirus 1, HPIV-4: human parainfluenza virus type 4, HPIV-2: human parainfluenza virus type 2, CV: coxsackievirus, WUPyV: WU polyomavirus, MP: Mycoplasma pneumoniae, CT: Chlamydia trachomatis, Cp: Chlamydia pneumoniae, Uu: Ureaplasma urealyticum, PJ: Pneumocystis jirovecii, Ca: Candida albicans, Af: Aspergillus fumigatus, Cpa: Candida parapsilosis, Ct: Candida tropicalis, Hi: Haemophilus influenzae, Spn: Streptococcus pneumoniae, SA: Staphylococcus aureus, AB: Acinetobacter baumannii, MC: Moraxella catarrhalis, KP: Klebsiella pneumoniae, Si: Streptococcus intermedius, Bp: Bordetella pertussis, Fn: Fusobacterium nucleatum, Tw: Tropheryma whipplei, Sag: Streptococcus anginosus, PA: Pseudomonas aeruginosa, E. coli: Escherichia coli, Hh: Haemophilus hemolyticus, Sm: Stenotrophomonas maltophilia, Sp: Streptococcus pyogenes, Ec: Enterobacter cloacae complex, Sc: Streptococcus constellatus, Ef: Enterococcus faecalis, Sa: Streptococcus agalactiae
Analysis of monthly detection trends (April 2021–December 2024) revealed distinct seasonal patterns for several key targets (Fig. 2). Viral agents such as rhinovirus, CMV, and HRSV each showed seasonal peaks. Among bacterial targets, Hi and Streptococcus pneumoniae were predominant, while Staphylococcus aureus, Acinetobacter baumannii, and Moraxella catarrhalis also exceeded 20% in specific months. Regarding atypical agents, Pneumocystis jirovecii was prominent until May 2022, after which MP became dominant.
Fig. 2.
Monthly detection frequency of common respiratory agents in hospitalized children with respiratory tract illness
Distribution of microbial detection patterns
Among the 10,153 cases, a single microbial agent was detected in 1,513 cases (14.90%), while multiple agents were co-detected in 8,414 cases (82.87%). The distribution by the number of agents detected per case was as follows: 2 agents (n = 2,394, 23.58%), 3 agents (n = 2,213, 21.79%), 4 agents (n = 1,760, 17.33%), and ≥ 5 agents (n = 2,047, 20.16%). Among the 25 major agents analyzed, MP was distinctive: it was detected as a single agent in 18.90% of all MP-positive cases. In contrast, the single-detection rate for each of the other 24 agents was below 8% (Fig. 3A). The comparison of microbial load between the single-detection and co-detection groups for the 10 most frequently single-detected agents revealed that the vast majority of microbes showed a significantly higher load in the single-detection group than in the co-detection group (P < 0.05) (Fig. 3B). The most frequent co-detection profiles were bacterial-viral combinations (n = 3,856, 37.98%) and bacterial-viral-atypical agent combinations (n = 1,076, 10.60%). Detailed patterns are provided in Supplementary Table 3.
Fig. 3.
Distribution Patterns and Microbial Load Between the Single-detection and Co-detection Groups. (A) Distribution Patterns of Single and Co-detection Among the Top 25 Most Frequently Detected Agents. (B) Comparison of Microbial Load Between the Single-detection and Co-detection Groups for the 10 Most Frequently Single-detected Agents. MP: Mycoplasma pneumoniae, HRSV: human respiratory syncytial virus, HMPV: human metapneumovirus, Bp: Bordetella pertussis, AdV: adenovirus, HPIV-3: human parainfluenza virus type 3, AB: Acinetobacter baumannii, RV: rhinovirus, HBoV: human bocavirus, IAV: influenza A virus, KP: Klebsiella pneumoniae, SA: Staphylococcus aureus, CMV: cytomegalovirus, EV: enterovirus, PJ: Pneumocystis jirovecii, HCoV: human coronavirus, Hi: Haemophilus influenzae, Si: Streptococcus intermedius, EBV: Epstein-Barr virus, Fn: Fusobacterium nucleatum, Tw: Tropheryma whipplei, MC: Moraxella catarrhalis, Spn: Streptococcus pneumoniae, Ca: Candida albicans, HHV-7: human herpesvirus 7
To assess potential bias arising from variable panel size over time, we performed a sensitivity analysis restricted to a core panel of 64 uniformly screened targets. This analysis yielded similar overall detection (96.89%) and co-detection (79.91%) rates, indicating that the addition of less common microbes to the panel did not significantly alter the primary epidemiological trends (see Supplementary Table 4).
Comparison of laboratory and clinical characteristics between single- and co-detected cases
As summarized in Table 1, children with single-agent detection were significantly older (median 36.00 vs. 24.00 months) with a higher proportion of females (45.59% vs. 39.35%) compared to those with co-detection (P < 0.001). The single-detection group exhibited higher levels of inflammatory markers, including serum amyloid A (38.92 vs. 27.28 mg/L, P = 0.021), procalcitonin (0.10 vs. 0.09 ng/mL, P = 0.003), and erythrocyte sedimentation rate (28.00 vs. 24.00 mm/h, P < 0.001). Clinically, patients with single detection experienced higher frequencies of fever (65.30% vs. 60.58%) and respiratory complications (49.83% vs. 41.97%), longer durations of fever (median 3.00 vs. 2.00 days) and hospitalization (median 7.00 vs. 6.00 days), greater need for respiratory support (24.85% vs. 22.36%), higher total treatment costs (median 1181.75 vs. 1147.77 USD), and higher rates of poor prognosis (7.05% vs. 4.71%) (all P < 0.05). Although other parameters (e.g., WBC, lymphocyte percentage, neutrophil percentage) also showed significant intergroup differences (all P < 0.05), median values remained within normal reference ranges.
Table 1.
Clinical characteristics associated with monomicrobial versus polymicrobial detection
| Clinical features | Monomicrobial detection (n = 1513) | Polymicrobial detection (n = 8414) | P value |
|---|---|---|---|
| Quantitative data [M (IQR)] | |||
| Age (months) | 36.00 (64.10) | 24.00 (41.33) | ***<0.001 |
| WBC (×109/L) | 8.50 (4.70) | 9.70 (5.50) | ***<0.001 |
| NEU (%) | 53.35 (30.23) | 48.10 (30.75) | ***<0.001 |
| LYM (%) | 35.65 (27.60) | 40.80 (29.10) | ***<0.001 |
| Mon (%) | 7.60 (4.00) | 7.60 (4.00) | 0.613 |
| Eos (%) | 1.30 (2.63) | 1.20 (2.50) | **0.008 |
| Baso (%) | 0.30 (0.20) | 0.30 (0.20) | 0.176 |
| RBC (×1012/L) | 4.50 (0.70) | 4.60 (0.70) | ***<0.001 |
| PLT (×109/L) | 337.00 (169.25) | 364.00 (167.00) | ***<0.001 |
| CRP (mg/L) | 5.00 (15.18) | 4.46 (11.87) | 0.075 |
| PCT (ng/mL) | 0.10 (0.16) | 0.09 (0.15) | **0.003 |
| SAA (mg/L) | 38.92 (101.66) | 27.28 (92.24) | *0.021 |
| ESR (mm/h) | 28.00 (25.50) | 24.00 (26.00) | ***<0.001 |
| D-dimer (ng/mL FEU) | 450.00 (430.00) | 383.00 (380.00) | ***<0.001 |
| LDH (U/L) | 299.00 (88.00) | 301.00 (87.00) | 0.474 |
| CK (U/L) | 94.00 (72.75) | 93.00 (69.00) | 0.085 |
| CK-MB (U/L) | 23.00 (11.00) | 24.00 (11.00) | ***<0.001 |
| AST (U/L) | 32.00 (13.00) | 34.00 (14.00) | ***<0.001 |
| ALT (U/L) | 14.00 (10.00) | 15.00 (12.00) | ***<0.001 |
| CYs-C (mg/L) | 0.88 (0.30) | 0.89 (0.27) | 0.605 |
| IgE (g/L) | 174.05 (466.70) | 190.55 (451.95) | 0.210 |
| IgG (g/L) | 8.73 (4.26) | 8.34 (4.06) | **0.003 |
| IgA (g/L) | 0.86 (1.07) | 0.64 (0.86) | ***<0.001 |
| IgM (g/L) | 1.18 (0.76) | 1.17 (0.75) | 0.676 |
| Length of hospitalization (days) | 7.00 (5.00) | 6.00 (4.00) | ***<0.001 |
| Treatment expense (USD) | 1181.75 (892.48) | 1147.77 (854.42) | ***<0.001 |
| Duration of fever (days) | 3.00 (3.00) | 2.00 (3.00) | ***<0.001 |
| Categorical data N (%) | |||
| Gender (female) | 688.00 (45.59%) | 3320.00 (39.35%) | ***<0.001 |
| Fever | 909.00 (65.30%) | 4700.00 (60.58%) | ***<0.001 |
| ICU | 118.00 (7.80%) | 566.00 (6.73%) | 0.130 |
| Respiratory complications | 754.00 (49.83%) | 3531.00 (41.97%) | ***<0.001 |
| Other system complications | 532.00 (35.16%) | 2952.00 (35.08%) | 0.954 |
| Respiratory support | 376.00 (24.85%) | 1881.00 (22.36%) | *0.033 |
| Poor prognosis | 98.00 (7.05%) | 365.00 (4.71%) | ***<0.001 |
*P < 0.05, **P < 0.01, and ***P < 0.001 represent statistically significant differences. WBC: Whiteblood cell; NEU: Neutrophil; LYM: Lymphocyte; Mon: Monocyte; Eos: Eosinophil; Baso: Basophil; RBC: Red Blood Cell; PLT: Platelet; CRP: C-reactive protein; PCT: Procalcitonin; SAA: Serum Amyloid A; ESR: Erythrocyte Sedimentation Rate; LDH: Lactic acid dehydrogenase; CK: Creatine kinase; CK-MB: Creatine kinase-MB; AST: Aspartate aminotransferase; ALT: Alanine Aminotransferase; CYs-C: Cystatin C; USD: United States Dollar; IgE: Immunoglobulin E; IgG: Immunoglobulin G; IgA: Immunoglobulin A; IgM: Immunoglobulin M
Comparison of clinical severity between single- and co-detection across major agents
Analysis of 12 common agents revealed that the clinical severity difference between single- and co-detection was agent-dependent (Table 2). For MP, adenovirus, and Hi, co-detection was associated with younger age and more severe manifestations—including higher costs, greater need for respiratory support, and increased complication rates (all P < 0.05).
Table 2.
Comparison of clinical characteristics by single or co-detection of common agents (N≥20)
| Agents | Numbers | Age (months) | Length of hospitalization (days) | Treatment expense (USD) | WBC (×109/L) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Single detection | Co-detection | Single detection | Co-detection | P value | Single detection | Co-detection | Pvalue | Single detection | Co-detection | P value | Single detection | Co-detection | P value | |
| Haemophilus influenzae | 57 | 3031 | 48.00 (52.00) | 36.00 (45.00) | 0.005 | 5.00 (2.00) | 6.00 (3.00) | <0.001 | 757.92 (695.19) | 1006.84 (703.78) | 0.024 | 9.90 (4.70) | 10.30 (5.70) | 0.601 |
| Mycoplasma pneumoniae | 583 | 2503 | 66.00 (49.25) | 42.00 (55.00) | <0.001 | 7.00 (3.00) | 7.00 (4.00) | 0.158 | 1178.71 (360.05) | 1232.08 (534.74) | 0.003 | 7.70 (3.50) | 8.80 (4.60) | <0.001 |
| Rhinovirus | 107 | 2579 | 25.00 (40.00) | 24.00 (39.00) | 0.444 | 6.00 (4.00) | 6.00 (4.00) | 0.845 | 1196.65 (1225.15) | 1133.65 (796.11) | 0.095 | 9.80 (5.50) | 10.70 (5.80) | 0.050 |
| Streptococcus pneumoniae | 22 | 2131 | 56.00 (51.75) | 36.00 (47.00) | 0.024 | 6.50 (8.00) | 6.00 (3.00) | 0.341 | 1308.27 (1420.12) | 1097.99 (677.79) | 0.092 | 9.05 (8.15) | 9.80 (5.50) | 0.919 |
| Cytomegalovirus | 53 | 1995 | 6.00 (8.00) | 9.00 (11.00) | 0.004 | 8.00 (11.00) | 7.00 (5.00) | 0.225 | 1384.28 (3484.16) | 1194.57 (1051.36) | 0.463 | 11.80 (5.95) | 10.70 (5.70) | 0.177 |
| Human respiratory syncytial virus | 147 | 1715 | 10.50 (21.48) | 12.00 (24.00) | 0.004 | 8.00 (8.00) | 7.00 (5.00) | 0.032 | 1369.99 (2412.39) | 1248.14 (1218.87) | 0.028 | 7.50 (4.20) | 9.20 (5.17) | <0.001 |
| Adenovirus | 64 | 1332 | 44.00 (36.25) | 36.00 (39.00) | 0.013 | 6.00 (3.75) | 6.00 (4.00) | 0.041 | 771.18 (1010.01) | 1416.67 (947.64) | 0.003 | 10.35 (7.75) | 10.45 (6.90) | 0.962 |
| Staphylococcus aureus | 37 | 1332 | 6.50 (44.63) | 20.00 (51.00) | 0.037 | 7.00 (14.50) | 6.00 (3.00) | 0.365 | 972.62 (4238.56) | 948.24 (760.68) | 0.268 | 10.00 (6.98) | 9.30 (4.90) | 0.417 |
| Acinetobacter baumannii | 52 | 1266 | 11.38 (56.98) | 13.00 (30.00) | 0.273 | 6.50 (13.75) | 7.00 (4.50) | 0.251 | 869.73 (5157.40) | 879.51 (1026.52) | 0.188 | 8.90 (5.63) | 10.00 (6.00) | 0.275 |
| Human parainfluenza virus type 3 | 45 | 957 | 11.00 (24.00) | 13.00 (22.00) | 0.097 | 8.00 (7.25) | 7.00 (4.00) | 0.170 | 1258.02 (1430.49) | 1206.34 (917.10) | 0.328 | 7.95 (5.08) | 10.00 (5.48) | 0.009 |
| Human metapneumovirus | 46 | 672 | 28.00 (47.55) | 25.50 (38.00) | 0.875 | 6.00 (4.50) | 6.00 (3.00) | 0.567 | 1096.38 (833.48) | 1140.36 (776.53) | 0.405 | 7.70 (4.20) | 8.65 (5.03) | 0.085 |
| Human bocavirus | 21 | 631 | 14.50 (24.32) | 16.00 (24.00) | 0.631 | 9.00 (15.75) | 7.00 (5.00) | 0.169 | 3075.41 (6259.43) | 1313.36 (954.34) | 0.023 | 9.35 (6.23) | 9.90 (5.13) | 0.899 |
| Agents | Numbers | NEU (%) | CRP (mg/L) | PCT (ng/mL) | SAA (mg/L) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Single detection | Co-detection | Single detection | Co-detection | P value | Single detection | Co-detection | P value | Single detection | Co-detection | P value | Single detection | Co-detection | P value | |
| Haemophilus influenzae | 57 | 3031 | 53.80 (23.35) | 52.60 (27.85) | 0.284 | 8.17 (24.29) | 6.50 (19.16) | 0.808 | 0.09 (0.12) | 0.11 (0.19) | 0.364 | 38.92 (225.85) | 44.42 (126.44) | 0.693 |
| Mycoplasma pneumoniae | 583 | 2503 | 60.50 (15.80) | 54.10 (24.80) | <0.001 | 9.20 (17.77) | 5.90 (14.54) | <0.001 | 0.10 (0.12) | 0.09 (0.11) | 0.124 | 60.04 (100.49) | 36.80 (96.99) | <0.001 |
| Rhinovirus | 107 | 2579 | 44.00 (38.00) | 48.20 (30.55) | 0.136 | 2.52 (7.14) | 4.16 (10.91) | 0.021 | 0.08 (0.08) | 0.08 (0.11) | 0.349 | 21.61 (80.33) | 25.03 (85.74) | 0.538 |
| Streptococcus pneumoniae | 22 | 2131 | 49.05 (36.08) | 51.20 (27.50) | 0.925 | 3.21 (29.09) | 5.32 (13.72) | 0.468 | 0.06 (0.24) | 0.09 (0.17) | 0.433 | 19.95 (259.33) | 34.83 (100.64) | 0.499 |
| Cytomegalovirus | 53 | 1995 | 28.20 (27.35) | 35.30 (27.55) | 0.009 | 0.79 (8.35) | 2.73 (8.19) | 0.172 | 0.09 (0.09) | 0.08 (0.11) | 0.840 | 5.06 (244.76) | 10.72 (52.43) | 0.429 |
| Human respiratory syncytial virus | 147 | 1715 | 37.00 (27.60) | 41.45 (29.60) | 0.016 | 1.51 (4.50) | 3.15 (7.65) | 0.002 | 0.10 (0.13) | 0.09 (0.12) | 0.668 | 9.28 (28.56) | 13.74 (55.57) | 0.059 |
| Adenovirus | 64 | 1332 | 67.65 (27.75) | 54.25 (29.33) | 0.022 | 16.72 (40.59) | 7.08 (21.30) | 0.007 | 0.42 (1.15) | 0.12 (0.25) | <0.001 | 166.73 (270.21) | 52.29 (157.30) | 0.003 |
| Staphylococcus aureus | 37 | 1332 | 41.35 (32.35) | 43.95 (32.58) | 0.479 | 0.63 (10.50) | 2.88 (8.77) | 0.086 | 0.09 (0.31) | 0.08 (0.11) | 0.604 | 5.11 (37.99) | 23.43 (78.97) | 0.045 |
| Acinetobacter baumannii | 52 | 1266 | 54.70 (31.85) | 43.10 (31.70) | 0.084 | 2.04 (12.69) | 3.36 (9.21) | 0.166 | 0.11 (0.35) | 0.09 (0.17) | 0.445 | 19.64 (69.00) | 18.01 (70.48) | 0.755 |
| Human parainfluenza virus type 3 | 45 | 957 | 37.30 (27.70) | 39.75 (30.35) | 0.725 | 0.90 (3.94) | 2.28 (7.12) | 0.054 | 0.08 (0.11) | 0.08 (0.09) | 0.927 | 8.57 (41.48) | 13.76 (65.72) | 0.375 |
| Human metapneumovirus | 46 | 672 | 43.40 (37.75) | 45.35 (31.55) | 0.936 | 3.03 (6.71) | 4.23 (9.51) | 0.121 | 0.10 (0.18) | 0.11 (0.24) | 0.639 | 15.28 (116.31) | 21.85 (66.30) | 0.505 |
| Human bocavirus | 21 | 631 | 54.10 (32.75) | 46.80 (29.38) | 0.631 | 2.27 (5.31) | 4.22 (10.58) | 0.120 | 0.11 (0.08) | 0.09 (0.13) | 0.222 | 27.91 (71.76) | 20.66 (90.75) | 0.835 |
| Agents | Numbers | ESR (mm/h) | Respiratory support, n (%) | Proportion of invasive respiratory support, n (%) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Single detection | Co-detection | Single detection | Co-detection | P value | Single detection | Co-detection | P value | Single detection | Co-detection | P value | ||||
| Haemophilus influenzae | 57 | 3031 | 30.50 (31.75) | 29.00 (25.00) | 0.274 | 4.00 (7.02%) | 507.00 (16.73%) | 0.076 | 1.00 (25.00%) | 129.00 (25.44%) | 1.000 | |||
| Mycoplasma pneumoniae | 583 | 2503 | 36.00 (19.00) | 31.00 (24.00) | <0.001 | 64.00 (10.98%) | 423.00 (16.90%) | <0.001 | 18.00 (28.13%) | 117.00 (27.66%) | 0.938 | |||
| Rhinovirus | 107 | 2579 | 18.00 (19.00) | 22.00 (23.00) | 0.056 | 38.00 (35.51%) | 554.00 (21.48%) | <0.001 | 15.00 (39.47%) | 182.00 (32.85%) | 0.402 | |||
| Streptococcus pneumoniae | 22 | 2131 | 28.00 (46.50) | 27.00 (25.00) | 0.938 | 4.00 (18.18%) | 345.00 (16.19%) | 1.000 | 0.00 (0.00%) | 97.00 (28.12%) | 0.579 | |||
| Cytomegalovirus | 53 | 1995 | 15.00 (38.00) | 20.00 (24.00) | 0.540 | 17.00 (32.08%) | 582.00 (29.17%) | 0.647 | 9.00 (52.94%) | 193.00 (33.16%) | 0.089 | |||
| Human respiratory syncytial virus | 147 | 1715 | 16.00 (16.75) | 21.00 (23.00) | 0.004 | 69.00 (46.94%) | 608.00 (35.45%) | 0.005 | 30.00 (43.47%) | 200.00 (32.89%) | 0.079 | |||
| Adenovirus | 64 | 1332 | 36.00 (27.00) | 30.00 (27.00) | 0.175 | 10.00 (15.63%) | 260.00 (19.52%) | 0.441 | 4.00 (40.00%) | 87.00 (33.46%) | 0.930 | |||
| Staphylococcus aureus | 37 | 1332 | 12.00 (13.00) | 20.00 (23.00) | 0.065 | 9.00 (24.32%) | 270.00 (20.27%) | 0.546 | 5.00 (55.56%) | 85.00 (31.48%) | 0.129 | |||
| Acinetobacter baumannii | 52 | 1266 | 19.00 (24.75) | 22.00 (24.00) | 0.701 | 20.00 (38.46%) | 381.00 (30.09%) | 0.199 | 12.00 (60.00%) | 132.00 (34.65%) | 0.021 | |||
| Human parainfluenza virus type 3 | 45 | 957 | 18.50 (23.25) | 21.00 (24.00) | 0.270 | 15.00 (33.33%) | 216.00 (22.57%) | 0.094 | 4.00 (26.67%) | 39.00 (18.06%) | 0.627 | |||
| Human metapneumovirus | 46 | 672 | 17.50 (12.00) | 23.00 (20.00) | 0.014 | 13.00 (28.26%) | 136.00 (20.24%) | 0.194 | 5.00 (38.46%) | 34.00 (25.00%) | 0.291 | |||
| Human bocavirus | 21 | 631 | 19.50 (18.50) | 22.00 (23.00) | 0.187 | 10.00 (47.62%) | 190 (30.11%) | 0.087 | 9.00 (90.00%) | 80.00 (42.11%) | 0.008 | |||
| Agents | Numbers | ICU admission rate, n (%) | Respiratory complications, n (%) | Other system complications, n (%) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Single detection | Co-detection | Single detection | Co-detection | P value | Single detection | Co-detection | P value | Single detection | Co-detection | P value | ||||
| Haemophilus influenzae | 57 | 3031 | 2.00 (3.51%) | 147.00 (4.85%) | 0.876 | 24.00 (42.11%) | 1168.00 (38.54%) | 0.583 | 13.00 (22.81%) | 893.00 (29.46%) | 0.274 | |||
| Mycoplasma pneumoniae | 583 | 2503 | 10.00 (1.72%) | 85.00 (3.40%) | 0.034 | 418.00 (71.70%) | 1483.00 (59.25%) | <0.001 | 112.00 (19.21%) | 709.00 (28.33%) | <0.001 | |||
| Rhinovirus | 107 | 2579 | 12.00 (11.21%) | 196.00 (7.60%) | 0.170 | 45.00 (42.06%) | 1013.00 (39.28%) | 0.565 | 37.00 (34.58%) | 862.00 (33.42%) | 0.804 | |||
| Streptococcus pneumoniae | 22 | 2131 | 1.00 (4.55%) | 110.00 (5.16%) | 1.000 | 8.00 (36.36%) | 885.00 (41.53%) | 0.625 | 7.00 (31.82%) | 645.00 (30.27%) | 0.875 | |||
| Cytomegalovirus | 53 | 1995 | 3.00 (5.66%) | 181.00 (9.07%) | 0.539 | 11.00 (20.75%) | 788.00 (39.50%) | 0.006 | 37.00 (69.81%) | 876.00 (43.91%) | <0.001 | |||
| Human respiratory syncytial virus | 147 | 1715 | 31.00 (21.09%) | 177.00 (10.32%) | <0.001 | 84.00 (57.14%) | 800.00 (46.65%) | 0.014 | 70.00 (47.62%) | 668.00 (38.95%) | 0.039 | |||
| Adenovirus | 64 | 1332 | 1.00 (1.56%) | 73.00 (5.48%) | 0.280 | 12.00 (18.75%) | 550.00 (41.29%) | <0.001 | 26.00 (40.63%) | 482.00 (36.19%) | 0.471 | |||
| Staphylococcus aureus | 37 | 1332 | 3.00 (8.11%) | 105.00 (7.88%) | 1.000 | 8.00 (21.62%) | 470.00 (35.29%) | 0.085 | 17.00 (45.95%) | 446.00 (33.48%) | 0.114 | |||
| Acinetobacter baumannii | 52 | 1266 | 10.00 (19.23%) | 128.00 (10.11%) | 0.035 | 16.00 (30.77%) | 416.00 (32.86%) | 0.753 | 29.00 (55.77%) | 502.00 (39.65%) | 0.020 | |||
| Human parainfluenza virus type 3 | 45 | 957 | 3.00 (6.67%) | 43.00 (4.49%) | 0.752 | 12.00 (26.67%) | 357.00 (37.30%) | 0.148 | 20.00 (44.44%) | 378.00 39.50%) | 0.508 | |||
| Human metapneumovirus | 46 | 672 | 2.00 (4.35%) | 33.00 (4.91%) | 1.000 | 23.00 (50.00%) | 282.00 (41.96%) | 0.286 | 14.00 (30.43%) | 190.00 (28.27%) | 0.753 | |||
| Human bocavirus | 21 | 631 | 7.00 (33.33%) | 91.00 (14.42%) | 0.017 | 12.00 (57.14%) | 332.00 (52.61%) | 0.683 | 10.00 (47.62%) | 248.00 (39.30%) | 0.443 | |||
USD: United States Dollar; WBC: White blood cell; NEU: Neutrophil; CRP: C-reactive protein; PCT: Procalcitonin; SAA: Serum Amyloid A; ESR: Erythrocyte Sedimentation Rate
Conversely, for HRSV, Acinetobacter baumannii, and bocavirus, single detection was associated with more severe disease than co-detection. Children infected solely with HRSV were younger and had poorer outcomes, including longer stays, higher costs, greater need for respiratory support, increased intensive care unit (ICU) admission, and more frequent complications (all P < 0.05). Similarly, the single detection of A. baumannii or human bocavirus—compared to their co-detection with other agents—was linked to higher rates of invasive respiratory support, ICU admission, extra-pulmonary complications, and treatment costs (all P < 0.05). A structured overview of these microbe-specific severity patterns is provided in Table 3.
Table 3.
Microbe severity patterns in single vs. co-detection
| Microbe | Pathogenicity group | Severity pattern | Primary endpoint(s) supporting the pattern |
|---|---|---|---|
| Acinetobacter baumannii | High | Mono-detection more severe | ICU admission rate, Invasive respiratory support, Other system complications |
| Human respiratory syncytial virus | High | Mono-detection more severe | ICU admission rate, Length of hospitalization, Treatment expense, Respiratory support, Respiratory complications, Other system complications |
| Human bocavirus | High | Mono-detection more severe | ICU admission rate, Invasive respiratory support, Treatment expense |
| Mycoplasma pneumoniae | Low | Co-detection more severe | ICU admission rate, Treatment expense, Respiratory support, Respiratory complications, Other system complications |
| Adenovirus | Low | Co-detection more severe | Length of hospitalization, Treatment expense, Respiratory complications |
| Haemophilus influenzae | Low | Co-detection more severe | Length of hospitalization, Treatment expense |
We performed stratified analyses by sample type. The trends observed across different sample types were consistent with those in the overall cohort (Supplementary Table 5).
Comparison of clinical profiles associated with single detection of specific agents
We compared children with single detection of 12 common targets (each n ≥ 20) against the overall single-detection cohort (n = 1234) to delineate clinical features associated with individual agents (Table 4). Distinct patterns were observed. Single detection of HRSV, Acinetobacter baumannii, and bocavirus was characterized by younger age and more severe disease. Compared to the overall single-detection baseline, these cases had significantly higher rates of respiratory support, ICU admission, extrapulmonary complications, and—especially for bocavirus—markedly higher treatment costs (all P < 0.01). In contrast, single MP detection involved older children. While showing the highest respiratory complication rate (71.70% vs. 53.54%; P < 0.001), they required less respiratory support (10.98% vs. 22.12%) and had lower ICU admission (1.72% vs. 6.89%) and fewer extrapulmonary complications (all P < 0.001). Single detection of adenovirus and Hi was generally milder, including shorter hospital stays, lower costs, and less respiratory support (all P < 0.05). Notably, cases with adenovirus detection showed significantly elevated levels of inflammatory markers (CRP, PCT, SAA; all P < 0.001).
Table 4.
Comparison of clinical features in single-agent detection
| Agents | Numbers | Proportion | Age (months) | P value | Length of hospitalization (days) | P value | Treatment expense (USD) | P value | WBC (×109/L) | P value | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total | 1234 | 40.00 (60.00) | 7.00 (4.00) | 1182.55 (702.36) | 8.30 (4.20) | |||||||
| Haemophilus influenzae | 57 | 4.62% | 48.00 (52.00) | 0.213 | 5.00 (2.00) | <0.001 | 760.10 (697.18) | <0.001 | 9.90 (4.70) | 0.002 | ||
| Mycoplasma pneumoniae | 583 | 47.24% | 66.00 (49.25) | <0.001 | 7.00 (3.00) | 0.390 | 1182.09 (361.08) | 0.870 | 7.70 (3.50) | <0.001 | ||
| Rhinovirus | 107 | 8.67% | 25.00 (40.00) | <0.001 | 6.00 (4.00) | 0.427 | 1200.08 (1228.67) | 0.669 | 9.80 (5.50) | <0.001 | ||
| Streptococcus pneumoniae | 22 | 1.78% | 56.00 (51.75) | 0.183 | 6.50 (8.00) | 0.992 | 1312.03 (1424.20) | 0.377 | 9.05 (8.15) | 0.158 | ||
| Cytomegalovirus | 53 | 4.29% | 6.00 (8.00) | <0.001 | 8.00 (11.00) | 0.083 | 1388.26 (3494.17) | 0.486 | 11.80 (5.95) | <0.001 | ||
| Human respiratory syncytial virus | 147 | 11.91% | 10.50 (21.48) | <0.001 | 8.00 (8.00) | <0.001 | 1373.92 (2419.32) | 0.001 | 7.50 (4.20) | 0.008 | ||
| Adenovirus | 64 | 5.19% | 44.00 (36.25) | 0.449 | 6.00 (4.00) | 0.021 | 773.40 (1012.91) | 0.002 | 10.35 (7.75) | <0.001 | ||
| Staphylococcus aureus | 37 | 3.00% | 6.50 (44.63) | 0.001 | 7.00 (15.00) | 0.966 | 975.41 (4250.73) | 0.963 | 10.00 (6.98) | 0.043 | ||
| Acinetobacter baumannii | 52 | 4.21% | 11.38 (56.98) | <0.001 | 6.50 (13.75) | 0.312 | 872.22 (5172.21) | 0.847 | 8.90 (5.62) | 0.154 | ||
| Human parainfluenza virus type 3 | 45 | 3.65% | 11.00 (24.00) | <0.001 | 8.00 (7.00) | 0.025 | 1258.92 (1431.51) | 0.394 | 7.95 (5.08) | 0.724 | ||
| Human metapneumovirus | 46 | 3.73% | 28.00 (47.55) | 0.038 | 6.00 (5.00) | 0.185 | 1099.53 (835.88) | 0.094 | 7.70 (4.20) | 0.177 | ||
| Human bocavirus | 21 | 1.70% | 14.50 (24.32) | 0.002 | 9.00 (16.00) | 0.073 | 3084.24 (6277.40) | 0.002 | 9.35 (6.23) | 0.075 |
| Agents | Numbers | Proportion | NEU (%) | P value | CRP (mg/L) | Pvalue | PCT (ng/mL) | P value | SAA (mg/L) | P value | ESR (mm/h) | P value |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total | 1234 | 54.60 (28.23) | 5.74 (17.28) | 0.10 (0.14) | 42.11 (104.83) | 30.00 (26.00) | ||||||
| Haemophilus influenzae | 57 | 4.62% | 53.80 (23.35) | 0.481 | 8.17 (24.29) | 0.548 | 0.09 (0.12) | 0.841 | 38.92 (225.85) | 0.434 | 30.50 (31.75) | 0.229 |
| Mycoplasma pneumoniae | 583 | 47.24% | 60.50 (15.80) | <0.001 | 9.20 (17.77) | <0.001 | 0.10 (0.12) | 0.822 | 60.04 (100.49) | 0.001 | 36.00 (19.00) | <0.001 |
| Rhinovirus | 107 | 8.67% | 44.00 (38.00) | 0.003 | 2.52 (7.14) | <0.001 | 0.08 (0.08) | 0.005 | 21.61 (80.33) | 0.088 | 18.00 (19.00) | <0.001 |
| Streptococcus pneumoniae | 22 | 1.78% | 49.05 (36.08) | 0.772 | 3.21 (29.09) | 0.480 | 0.06 (0.24) | 0.319 | 19.95 (259.34) | 0.536 | 28.00 (47.00) | 0.928 |
| Cytomegalovirus | 53 | 4.29% | 28.20 (27.35) | <0.001 | 0.79 (8.35) | 0.002 | 0.09 (0.09) | 0.207 | 5.06 (244.76) | 0.073 | 15.00 (38.00) | 0.017 |
| Human respiratory syncytial virus | 147 | 11.91% | 37.00 (27.60) | <0.001 | 1.51 (4.50) | <0.001 | 0.10 (0.13) | 0.939 | 9.28 (28.56) | <0.001 | 16.00 (17.00) | <0.001 |
| Adenovirus | 64 | 5.19% | 67.65 (27.75) | 0.002 | 16.72 (40.59) | <0.001 | 0.42 (1.15) | <0.001 | 166.73 (270.21) | <0.001 | 36.00 (27.00) | 0.081 |
| Staphylococcus aureus | 37 | 3.00% | 41.35 (32.35) | 0.006 | 0.63 (10.50) | 0.005 | 0.09 (0.31) | 0.864 | 5.11 (37.99) | 0.007 | 12.00 (13.00) | <0.001 |
| Acinetobacter baumannii | 52 | 4.21% | 54.70 (31.85) | 0.324 | 2.04 (12.69) | 0.008 | 0.11 (0.35) | 0.439 | 19.64 (69.00) | 0.163 | 19.00 (25.00) | 0.030 |
| Human parainfluenza virus type 3 | 45 | 3.65% | 37.30 (27.70) | <0.001 | 0.90 (3.94) | <0.001 | 0.08 (0.11) | 0.133 | 8.57 (41.48) | 0.018 | 18.50 (23.25) | 0.003 |
| Human metapneumovirus | 46 | 3.73% | 43.40 (37.75) | 0.069 | 3.03 (6.71) | 0.019 | 0.10(0.18) | 0.539 | 15.28 (116.31) | 0.242 | 17.50 (12.00) | <0.001 |
| Human bocavirus | 21 | 1.70% | 54.10 (32.75) | 0.629 | 2.67 (5.31) | 0.017 | 0.11 (0.08) | 0.506 | 27.91 (71.76) | 0.275 | 19.50 (18.50) | 0.007 |
| Agents | Numbers | Proportion | Respiratory support, n (%) | P value | Proportion of invasive respiratory support, n (%) | P value | ICU admission rate, n (%) | P value | Respiratory complications, n (%) | P value | Other system complications, n (%) | P value |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total | 1234 | 273.00 (22.12%) | 112.00 (41.03%) | 85.00 (6.89%) | 673.00 (53.54%) | 392.00 (31.77%) | ||||||
| Haemophilus influenzae | 57 | 4.62% | 4.00 (7.02%) | 0.011 | 1.00 (25.00%) | 0.893 | 2.00 (3.51%) | 0.469 | 24.00 (42.11%) | 0.066 | 13.00 (22.81%) | 0.154 |
| Mycoplasma pneumoniae | 583 | 47.24% | 64.00 (10.98%) | <0.001 | 18.00 (28.13%) | 0.056 | 10.00 (1.72%) | <0.001 | 418.00 (71.70%) | <0.001 | 112.00 (19.21%) | <0.001 |
| Rhinovirus | 107 | 8.67% | 38.00 (35.51%) | 0.002 | 15.00 (39.47%) | 0.855 | 12.00 (11.21%) | 0.097 | 45.00 (42.06%) | 0.013 | 37.00 (34.58%) | 0.550 |
| Streptococcus pneumoniae | 22 | 1.78% | 4.00 (18.18%) | 0.855 | 0.00 (0.00%) | 0.150 | 1.00 (4.55%) | 0.996 | 8.00 (36.36%) | 0.090 | 7.00 (31.82%) | 0.996 |
| Cytomegalovirus | 53 | 4.29% | 17.00 (32.08%) | 0.089 | 9.00 (52.94%) | 0.334 | 3.00 (5.66%) | 0.945 | 11.00 (20.75%) | <0.001 | 37.00 (69.81%) | <0.001 |
| Human respiratory syncytial virus | 147 | 11.91% | 69.00 (46.94%) | <0.001 | 30.00 (43.48%) | 0.712 | 31.00 (21.09%) | <0.001 | 84.00 (57.14%) | 0.549 | 70.00 (47.62%) | <0.001 |
| Adenovirus | 64 | 5.19% | 10.00 (15.63%) | 0.220 | 4.00 (40.00%) | 1.000 | 1.00 (1.56%) | 0.158 | 12.00 (18.75%) | <0.001 | 26.00 (40.63%) | 0.139 |
| Staphylococcus aureus | 37 | 3.00% | 9.00 (24.32%) | 0.751 | 5.00 (55.56%) | 0.384 | 3.00 (8.11%) | 1.000 | 8.00 (21.62%) | <0.001 | 17.00 (45.95%) | 0.069 |
| Acinetobacter baumannii | 52 | 4.21% | 20.00 (38.46%) | 0.006 | 12.00 (60.00%) | 0.089 | 10.00 (19.23%) | <0.001 | 16.00 (30.77%) | <0.001 | 29.00 (55.77%) | <0.001 |
| Human parainfluenza virus type 3 | 45 | 3.65% | 15.00 (33.33%) | 0.077 | 4.00 (26.67%) | 0.405 | 3.00 (6.67%) | 1.000 | 12.00 (26.67%) | <0.001 | 20.00 (44.44%) | 0.074 |
| Human metapneumovirus | 46 | 3.73% | 13.00 (28.26%) | 0.326 | 5.00 (38.46%) | 0.854 | 2.00 (4.35%) | 0.709 | 23.00 (50.00%) | 0.544 | 14.00 (30.43%) | 0.849 |
| Human bocavirus | 21 | 1.70% | 10.00 (47.62%) | 0.006 | 9.00 (90.00%) | 0.006 | 7.00 (33.33%) | <0.001 | 12.00 (57.14%) | 0.812 | 10.00 (47.62%) | 0.123 |
USD: United States Dollar; WBC: White blood cell; NEU: Neutrophil; CRP: C-reactive protein; PCT: Procalcitonin; SAA: Serum Amyloid A; ESR: Erythrocyte Sedimentation Rate
Imaging findings
Among the 10,153 cases, 4,580 underwent bronchoscopy. The most common findings were mucosal congestion and edema (91.64%, 4,197/4,580) and white viscous secretions (71.14%, 3,258/4,580). Other observations included follicular hyperplasia (12.86%), longitudinal folds (10.81%), stringy-flocculent lavage fluid (69.17%), mucus plugs (5.02%), plastic bronchitis (0.59%), and bronchiectasis (0.50%).
Of the 6,900 children who received CT scans, thickened lung markings/bronchial walls (89.33%, 6,164/6,900) and patchy opacities (85.80%, 5,920/6,900) were predominant. Additional findings comprised mucus plugging (17.55%), tree-in-bud opacities (10.51%), ground-glass opacities (8.23%), linear opacities (19.04%), consolidation (27.75% unilateral, 17.88% bilateral), lymphadenopathy (8.19%), pleural effusion (3.81%), atelectasis (4.58%), and necrotizing pneumonia (0.16%).
Respiratory support
Of the 10,153 cases, 2,374 (23.38%) received respiratory support. Non-invasive ventilation was employed in 1,460 cases (61.50% of supported cases), primarily via conventional oxygen therapy (nasal cannula/mask, n = 1,219, 83.49%), followed by continuous positive airway pressure (n = 119, 8.15%) and high-flow nasal cannula oxygen therapy (n = 4, 0.27%). The remaining 914 cases (38.50%) required invasive support, which predominantly involved invasive mechanical ventilation (n = 885, 96.83%), with smaller proportions receiving high-frequency oscillatory ventilation (n = 17, 1.86%) or extracorporeal membrane oxygenation (n = 12, 1.31%).
Complications and clinical outcomes
Among the 10,153 cases, 4,360 (42.94%) developed respiratory complications, primarily pulmonary consolidation (26.61%), hypoxemia (10.41%), and respiratory failure (8.0%). Other-system complications were recorded in 3,628 cases (35.73%), primarily anemia (7.50%), gastrointestinal dysfunction (5.65%), electrolyte disturbances (5.55%), myocardial injury (4.70%), and hepatic injury (3.76%) (Fig. 4). Regarding outcomes, the median hospital stay was 7 days (IQR: 5–9), and the median cost was 1,170.47 USD (IQR: 732.03–1,622.82). ICU admission was required for 717 children (7.06%). The proportion of ICU admissions varied significantly across different sample types: sputum samples had the highest ICU admission rate (15.1%, 140/927), followed by bronchoalveolar lavage fluid (BALF) (6.8%, 311/4578), and throat swabs (5.7%, 266/4648). Following treatment, most patients (9,653, 95.08%) were discharged in improved condition, while 475 (4.68%) opted for early discharge against medical advice, and 25 (0.25%) died.
Fig. 4.
Common respiratory and systemic complications in hospitalized children with respiratory tract infections
Factors associated with ICU admission
Logistic regression analysis of factors associated with ICU admission revealed significant intergroup differences in numerous variables by univariate analysis (Table 5). Subsequent multivariate analysis identified several independent predictors, among which respiratory complications emerged as the strongest factor (OR = 8.289, 95% CI: 6.551–10.489), followed by fever (OR = 7.338, 95% CI: 5.285–10.188) and other system complications (OR = 5.564, 95% CI: 4.601–6.730). Additional independent predictors included younger age, longer fever duration, lower RBC, lower monocyte percentage, elevated procalcitonin, lower ESR, higher creatine kinase-MB, and lower immunoglobulin M levels. In contrast, variables such as WBC count, neutrophil percentage, lymphocyte percentage, eosinophil percentage, CRP, SAA, and IgA lost significance in the multivariate model.
Table 5.
Clinical characteristics of ICU-patients and Non-ICU patients
| Clinical features | ICU-patients (n = 717) | Non-ICU patients (n = 9436) | Univariate analysis P value |
Multivariate analysis Odds Ratio (95% CI) |
Multivariate analysis P value |
|---|---|---|---|---|---|
| Quantitative data [M (IQR)] | |||||
| Age (months) | 12.00 (24.50) | 27.00 (47.07) | ***<0.001 | 0.971 (0.964–0.979) a | ***<0.001 |
| RBC (×1012/L) | 4.30 (1.00) | 4.59 (0.70) | ***<0.001 | 0.728 (0.589-0.900) a | **0.003 |
| WBC (×109/L) | 10.10 (6.63) | 9.50 (5.40) | ***<0.001 | 1.021 (0.998–1.045)a | 0.073 |
| NEU (%) | 57.85 (31.80) | 48.20 (30.50) | ***<0.001 | 0.546 (0.281–1.059)a | 0.074 |
| LYM (%) | 31.75 (27.90) | 40.60 (29.00) | ***<0.001 | 0.531 (0.273–1.032)a | 0.062 |
| Mon (%) | 7.20 (4.90) | 7.60 (3.90) | ***<0.001 | 0.508 (0.260–0.990) a | *0.047 |
| Eos (%) | 0.60 (2.20) | 1.30 (2.60) | ***<0.001 | 0.559 (0.284–1.097)a | 0.091 |
| Baso (%) | 0.20 (0.20) | 0.30 (0.20) | ***<0.001 | - | |
| PLT (×109/L) | 360.00 (168.00) | 360.00 (169.00) | 0.353 | ||
| CRP (mg/L) | 5.09 (13.65) | 4.41 (11.92) | **0.002 | 1.003 (0.998–1.009)a | 0.239 |
| PCT( ng/mL) | 0.16 (0.36) | 0.09 (0.14) | ***<0.001 | 1.041 (1.002–1.083) a | *0.042 |
| SAA (mg/L) | 19.02 (78.17) | 30.67 (98.10) | ***<0.001 | 1.000 (0.999–1.002)a | 0.900 |
| ESR (mm/h) | 19.00 (23.00) | 25.00 (26.00) | ***<0.001 | 0.968 (0.959–0.978) a | ***<0.001 |
| D-dimer (ng/mL FEU) | 520.00 (691.00) | 390.00 (380.00) | ***<0.001 | 1.000 (1.000–1.000)a | 0.750 |
| LDH (U/L) | 324.50 (133.00) | 299.00 (85.00) | ***<0.001 | 1.000 (0.999–1.001)a | 0.775 |
| AST (U/L) | 34.00 (17.00) | 33.00 (15.00) | *0.016 | 0.999 (0.996–1.003)a | 0.774 |
| ALT (U/L) | 17.00 (14.00) | 15.00 (11.00) | ***<0.001 | 1.002 (0.999–1.006)a | 0.230 |
| CYs-C (mg/L) | 0.94 (0.44) | 0.89 (0.28) | ***<0.001 | 0.764 (0.399–1.464)a | 0.418 |
| CK (U/L) | 95.00 (96.00) | 93.00 (68.00) | 0.400 | ||
| CK-MB (U/L) | 27.00 (15.00) | 24.00 (11.00) | ***<0.001 | 1.010 (1.003–1.016) a | **0.003 |
| IgG (g/L) | 7.48 (4.31) | 8.47 (4.05) | ***<0.001 | 0.095 (0.945–1.048)a | 0.848 |
| IgA (g/L) | 0.40 (0.65) | 0.70 (0.92) | ***<0.001 | 1.210 (0.891–1.642)a | 0.222 |
| IgM (g/L) | 0.94 (0.72) | 1.18 (0.75) | ***<0.001 | 0.690 (0.538–0.886) a | **0.004 |
| IgE (g/L) | 166.30 (560.93) | 190.10 (448.65) | 0.167 | ||
| Length of hospitalization (days) | 14.00 (13.00) | 6.00 (3.00) | ***<0.001 | - | |
| Treatment expense (USD) | 5690.81 (5521.52) | 1113.34 (757.02) | ***<0.001 | - | |
| Duration of fever (days) | 6.00 (7.00) | 2.00 (2.00) | ***<0.001 | 1.206 (1.166–1.247) a | ***<0.001 |
| Categorical data [N (%)] | |||||
| Fever | 650.00 (94.20%) | 5103.00 (58.84%) | ***<0.001 | 7.338 (5.285–10.188) b | ***<0.001 |
| Respiratory complications | 604.00 (84.24%) | 3756.00 (39.81%) | ***<0.001 | 8.289 (6.551–10.489) b | ***<0.001 |
| Other system complications | 532.00 (74.20%) | 3096.00 (32.81%) | ***<0.001 | 5.564 (4.601–6.730) b | ***<0.001 |
| Respiratory support | 643.00 (89.68%) | 1731.00 (18.34%) | ***<0.001 | - | |
| Poor prognosis | 100.00 (14.51%) | 400.00 (4.61%) | ***<0.001 | - |
*P < 0.05, **P < 0.01, and ***P < 0.001 represent statistically significant differences.WBC: White blood cell; NEU: Neutrophil; LYM: Lymphocyte; Mon: Monocyte; Eos: Eosinophil; Baso: Basophil; RBC: Red Blood Cell; PLT: Platelet; CRP: C-reactive protein; PCT: Procalcitonin; SAA: Serum Amyloid A; ESR: Erythrocyte Sedimentation Rate; LDH: Lactic acid dehydrogenase; CK: Creatine kinase; CK-MB: Creatine kinase-MB; AST: Aspartate aminotransferase; ALT: Alanine Aminotransferase; CYs-C: Cystatin C; USD: United States Dollar; IgE: Immunoglobulin E; IgG: Immunoglobulin G; IgA: Immunoglobulin A; IgM: Immunoglobulin M. a Multivariate regression analysis was performed on quantitative data with statistically significant differences between groups; b Multivariate regression analysis was performed on categorical data with statistically significant differences between groups. - Length of hospitalization, Poor prognosis and Treatment expense were not ICU treatment covariates and thus were not included in the multivariate regression analysis; Respiratory support was not included as a covariate in the logistic regression model as it was initiated after ICU admission in a subset of patients; Baso (%) was automatically excluded by the Logistic regression model because it had complete collinearity with the percentages of other white blood cell categories (Baso% ≈ 100% - NEU% - LYM% - Mon% - Eos%)
Comparison of agent distribution between ICU and non-ICU patients
Detected agents differed markedly between ICU and non-ICU patients. The ICU group (n = 717) showed significantly higher detection rates for HRSV (29.01% vs. 17.53%), CMV (25.66% vs. 19.75%), Acinetobacter baumannii (19.25% vs. 12.51%), bocavirus (13.67% vs. 5.87%), and Streptococcus intermedius (9.34% vs. 5.24%) (all P < 0.001). In contrast, the non‑ICU group (n = 9436) had higher prevalences of Haemophilus influenzae (31.15% vs. 20.78%), MP (31.70% vs. 13.25%), Streptococcus pneumoniae (21.64% vs. 15.48%), adenovirus (14.01% vs. 10.32%), and other viral and fungal agents (all P < 0.05) (Fig. 5). These findings highlight a distinct agent profile in critically ill children, characterized by a greater representation of agents associated with more severe disease in ICU patients.
Fig. 5.
Comparative analysis of predominant agent distribution between ICU and non-ICU groups of children with respiratory tract illness. *: P < 0.05, **: P < 0.01, ***: P < 0.001, represents the difference is statistically significant
Discussion
Pediatric respiratory infections involve a diverse spectrum of pathogens, and the predominant causative agents can vary significantly across different geographic regions, time periods, and age groups [20–22]. In our tNGS analysis of 10,153 hospitalized children, the most frequently detected agents were Haemophilus influenzae (30.41%), MP (30.40%), and rhinovirus (26.45%). A notable epidemiological feature was the high co‑detection rate of 82.87%, exceeding previously reported rates (23–63%) [23–25]. This observation likely stems from the combined effect of two factors: the high sensitivity and wide detection spectrum of the tNGS methodology, and the generally more severe clinical profile—and thus potentially richer microbial burden—of the hospitalized children in our cohort. It is important to note that not all detected microorganisms represent active infectious pathogens. Although a multidimensional diagnostic framework was employed to aid in differentiation, distinguishing among true infection, colonization, and latent infection remains challenging in clinical practice. For the clinical management of detected microorganisms, a comprehensive assessment incorporating host factors, clinical presentation, imaging findings, and laboratory results is essential to avoid overdiagnosis, overtreatment, and inappropriate use of antibiotics. Collectively, these findings highlight the common occurrence and complexity of microbial co-detection in hospitalized children with respiratory illnesses, while underscoring that the pathogenic significance of co-detected microorganisms and the interactions among different agents warrant further in-depth investigation.
Through etiological screening in a large cohort of hospitalized children, this study systematically compared the clinical characteristics associated with the detection of individual major respiratory agents. Given that co-detection is common in pediatric respiratory illnesses, prior research—often limited by incomplete exclusion of co-detected agents, modest sample sizes, and insufficient multidimensional clinical comparison—has left a gap in delineating the distinct clinical phenotypes linked to single agents; our analysis helps address this gap. The results reveal agent-specific profiles: detection of HRSV and HBoV occurred more frequently in younger children and was associated with greater need for respiratory support, higher ICU admission rates, and more severe illness; MP detection showed the highest frequency of respiratory complications yet required less respiratory support and lower ICU utilization, suggesting a more localized disease process; detection of Haemophilus influenzae was linked to a milder clinical course; while adenovirus detection correlated with marked systemic hyperinflammation. These findings highlight the potential clinical relevance of an agent-informed approach to patient assessment—for example, recognizing that HRSV or HBoV detection in infants may warrant closer monitoring, MP detection may prompt heightened vigilance for respiratory complications, and adenovirus detection may signal the need for early assessment of systemic inflammatory response. Overall, this study clarifies distinct clinical patterns associated with different detected agents, providing a basis for future mechanistic investigation and the development of individualized management strategies.
This study reveals a bidirectional and agent-specific pattern in disease severity associations: the sole detection of certain agents with higher virulence (such as HRSV, Acinetobacter baumannii, and HBoV) was linked to worse clinical outcomes, whereas for agents such as MP and adenovirus, co-detection was correlated with more severe illness. This finding challenges the conventional view that co-detection generally exacerbates disease. Niche competition may serve as an explanatory mechanism: during single-agent detection, highly virulent organisms may dominate available resources, leading to severe manifestations, while co-detection could restrain their excessive proliferation. Conversely, for less virulent microbes, co-detection with more virulent agents—or synergistic effects in certain combinations—may aggravate disease progression. Our observation that microbial loads were generally higher in mono-detection than in co-detection (Fig. 3B) is consistent with the competition hypothesis. Therefore, clinical severity appears to depend not simply on the number of agents detected, but rather on a combination of the inherent virulence of the agents, complex inter-microbial interactions (such as competition, suppression, or synergy), and individual host differences. Future research is needed to further elucidate these underlying mechanisms. A comprehensive consideration of both the type of agent(s) detected and the specific clinical context of the patient may ultimately help optimize risk stratification and guide more precise clinical decision-making.
In this cohort of 10,153 children hospitalized with ARI, the median length of stay was 7 days (IQR 5–9) and median treatment cost was 1,170.47 USD (IQR 732.03–1,622.82), with an overall ICU admission rate of 7%. Multivariate analysis identified several independent predictors of ICU admission, among which respiratory complications emerged as the strongest factor (OR = 8.289, 95% CI: 6.551–10.489), followed by fever on admission (OR = 7.338, 95% CI: 5.285–10.188) and other system complications (OR = 5.564, 95% CI: 4.601–6.730). ICU patients also exhibited a distinct microbial profile, with significantly higher detection rates of HRSV, CMV, Acinetobacter baumannii, and bocavirus. Notably, CMV was detected in 20.17% of the overall cohort, with the highest rate in infants ≤ 1 year (37.3%). It is important to distinguish between congenital and perinatal/postnatal CMV infection given their divergent clinical implications: congenital CMV may lead to severe sequelae such as sensorineural hearing loss and neurodevelopmental impairment, whereas perinatal/postnatal CMV is typically mild in term infants 26, 27. In our cohort, only 4 CMV-positive cases were aged ≤ 3 weeks (confirmed congenital), while 41 cases aged > 3 weeks to 1 month could not be definitively classified due to lack of early testing. Additionally, due to the absence of longitudinal serological data, we were unable to differentiate between primary infection and viral reactivation in this cohort. This finding, combined with the identified predictors highlights the multidimensional determinants of severe disease. Together, these results underscore the value of integrating early clinical and laboratory indicators with agent-specific data to enhance risk stratification and guide timely intervention in critically ill children.
The distribution of detected microbes varied across different sample types, and the choice of sampling method may be associated with disease severity. Generally, children who underwent bronchoscopy for BALF collection presented with more severe illness. However, in clinical practice, due to the high cost of bronchoscopy, some families of severely ill children may opt for throat swabs or sputum as alternative samples for economic reasons. Additionally, in the ICU, many critically ill children receiving invasive mechanical ventilation often undergo tNGS testing using endotracheal aspirates. This specimen type has a lower risk of contamination and is easily accessible, making it a more feasible and reliable testing option in the ICU setting—which partly explains why sputum samples accounted for the highest proportion among ICU children. Nevertheless, this sampling preference may introduce selection bias: some severely ill children who did not undergo bronchoscopy were categorized into the sputum or throat swab groups, potentially diluting the differences in disease severity across sample types and even underestimating the true diagnostic value of BALF in critically ill children. The selection of sample type is influenced by multiple factors, including the child’s clinical condition, family economic status, and clinical practice.
This study has several limitations. First, as a single-center retrospective analysis conducted in Guangxi, China, the findings are subject to potential selection bias (e.g., only hospitalized children with tNGS data were included) and information bias. Differences across testing batches or panel versions may also introduce variability. Second, the study period (2021–2024) coincided with the COVID-19 pandemic, during which respiratory pathogen epidemiology was likely influenced by non-pharmaceutical interventions; thus, the results are time- and region-specific. Third, although tNGS is highly sensitive, it cannot definitively distinguish colonization from active infection, and its finite panel coverage may miss uncommon or emerging pathogens. Fourth, data on prior or concurrent antibiotic use were not collected. Empirical antibiotic therapy, common in this population, may have reduced bacterial detection rates and confounded associations with clinical outcomes. Future multicenter prospective studies with standardized antibiotic documentation are needed to validate these findings.
In summary, this large-scale study of 10,153 cases systematically elucidates key epidemiological and clinical characteristics of acute respiratory illness in children, highlighting a notably high co-detection rate (82.87%) and distinct clinical profiles associated with specific agents. Disease severity exhibited a bidirectional pattern: detection of single agents associated with higher baseline virulence (e.g., HRSV and HBoV) correlated with more severe outcomes, whereas for agents with relatively lower virulence (e.g., MP and adenovirus), co-detection was associated with exacerbated illness. Respiratory complications emerged as the strongest predictor of ICU admission (OR = 8.289), and ICU patients demonstrated a distinct microbial profile characterized by higher detection rates of agents such as HRSV and HBoV. Collectively, these findings provide a comprehensive epidemiological and clinical characterization of pediatric respiratory infections, offering a foundation for future research aimed at integrating microbe-specific information into early risk stratification and clinical management strategies for critically ill children.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
Thank all the participants who participated in this study. Thank the Health Commission of Guangxi Zhuang Autonomous Region for the financial support.
Author contributions
Chunyun Fu and Jie Tan drafted the original manuscript. Junming Lu, Xuehua Hu, Qifei Li, and Xiaoying Chen critically revised the manuscript. All the authors contributed to the data collection and statistical analysis, and the approved final manuscript.
Funding
This study was supported by the Nanning Science and Technology Base Special Project (Grant No. 20233071), the Guangxi Clinical Key Specialty (Laboratory Medicine) Project, the Guangxi Medical and Health Appropriate Technology Development and Promotion Application Project (Grant No. S2024082), and the Guangxi Health Commission Self-Financing Projects (Grant No. Z-A20220257).
Data availability
Data and materials may be made available upon written request to the corresponding author.
Declarations
Human ethics and consent to participate
The study was approved by the Medical Ethics Committee of Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region. Informed consent from the patients was waived due to the retrospective nature of the study. The study was conducted following the guidelines of the Declaration of Helsinki.
Consent for publication
All authors have reviewed and approved the final version of the manuscript for publication.
Competing interests
No potential conflict of interest was reported by the author(s).
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Chunyun Fu, Junming Lu and Xuehua Hu contributed equally to this work.
Contributor Information
Xiaoying Chen, Email: cxy183763@163.com.
Qifei Li, Email: qxl368@med.miami.edu.
Jie Tan, Email: tanjie273@163.com.
References
- 1.Collaborators GL. Age-sex differences in the global burden of lower respiratory infections and risk factors, 1990–2019: results from the Global Burden of Disease Study 2019. Lancet Infect Dis. 2022;22(11):1626–47. 10.1016/S1473-3099(22)00510-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.McAllister DA, Liu L, Shi T, Chu Y, Reed C, Burrows J, et al. Global, regional, and national estimates of pneumonia morbidity and mortality in children younger than 5 years between 2000 and 2015: a systematic analysis. Lancet Glob Health. 2019;7(1):e47–57. 10.1016/S2214-109X(18)30408-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Collaborators GBDLRI. Estimates of the global, regional, and national morbidity, mortality, and aetiologies of lower respiratory infections in 195 countries, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Infect Dis. 2018;18(11):1191–210. 10.1016/S1473-30991830310-4. [DOI] [PMC free article] [PubMed]
- 4.Tan J, Huang Z, Tang W, Liu G, Li P, Wen J, et al. Human bocavirus-1 infection in hospitalized pediatric patients with acute respiratory tract infections. Microbiol Spectr. 2025;13(4):e0298524. 10.1128/spectrum.02985-24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Niu D, Gao Y, Zhang Y, Lv Q, Jiang Y, Jia Y, et al. Systematic Review and Meta-Analysis of the Association Between Clinical Severity and Co-Infection of Human Adenovirus With Other Respiratory Pathogens in Children. J Med Virol. 2025;97(5):e70370. 10.1002/jmv.70370. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ruuskanen O, Lahti E, Jennings LC, Murdoch DR. Viral pneumonia. Lancet. 2011;377(9773):1264–75. 10.1016/S0140-6736(10)61459-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lalbiaktluangi C, Yadav MK, Singh PK, Singh A, Iyer M, Vellingiri B, et al. A cooperativity between virus and bacteria during respiratory infections. Front Microbiol. 2023;14:1279159. 10.3389/fmicb.2023.1279159. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kollmann TR, Kampmann B, Mazmanian SK, Marchant A, Levy O. Protecting the Newborn and Young Infant from Infectious Diseases: Lessons from Immune Ontogeny. Immunity. 2017;46(3):350–63. 10.1016/j.immuni.2017.03.009. [DOI] [PubMed] [Google Scholar]
- 9.Blauwkamp TA, Thair S, Rosen MJ, Blair L, Lindner MS, Vilfan ID, et al. Analytical and clinical validation of a microbial cell-free DNA sequencing test for infectious disease. Nat Microbiol. 2019;4(4):663–74. 10.1038/s41564-018-0349-6. [DOI] [PubMed] [Google Scholar]
- 10.Gadsby NJ, Russell CD, McHugh MP, Mark H, Conway Morris A, Laurenson IF, et al. Comprehensive Molecular Testing for Respiratory Pathogens in Community-Acquired Pneumonia. Clin Infect Dis. 2016;62(7):817–23. 10.1093/cid/civ1214. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Fu C, Ruan J, Mo L, Zhao J, Lu J, Huang Y, et al. Epidemiological and clinical profiles of respiratory syncytial virus infections in hospitalized children: a retrospective cohort study utilizing targeted next-generation sequencing. Eur J Clin Microbiol Infect Dis. 2025. 10.1007/s10096-025-05112-w. [DOI] [PubMed] [Google Scholar]
- 12.Deng Z, Li C, Wang Y, Wu F, Liang C, Deng W, et al. Targeted next-generation sequencing for pulmonary infection diagnosis in patients unsuitable for bronchoalveolar lavage. Front Med. 2023;10:1321515. 10.3389/fmed.2023.1321515. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Sun W, Zheng L, Kang L, Chen C, Wang L, Lu L, et al. Comparative analysis of metagenomic and targeted next-generation sequencing for pathogens diagnosis in bronchoalveolar lavage fluid specimens. Front Cell Infect Microbiol. 2024;14:1451440. 10.3389/fcimb.2024.1451440. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ma R, Liu Z, Zhang L, Chen C, Yuan B, Luo Y, et al. Epidemiological characteristics of severe community-acquired pneumonia in children admitted to two tertiary hospitals in Shihezi, Xinjiang Region, China in 2023: a cross-sectional analysis. J Thorac disease. 2024;16(10):6969–82. 10.21037/jtd-24-1417. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Xiao Y, Dekyi, Wang X, Feng S, Yang Y, Zheng J, et al. Interpretation of pathogenicity and clinical features of multiple pathogens in pediatric lower respiratory tract infections by tNGS RPTM analysis. Eur J Clin Microbiol Infect Dis. 2025. 10.1007/s10096-025-05094-9. [DOI] [PubMed] [Google Scholar]
- 16.Kuang Y, Tan W, Hu C, Dai Z, Bai L, Wang J, et al. Diagnosis value of targeted and metagenomic sequencing in respiratory tract infection. Front Cell Infect Microbiol. 2024;14:1498512. 10.3389/fcimb.2024.1498512. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Wei M, Ai X, Gu D, Zhang S, Xu K, Li S, et al. Ultra-broad hybrid capture-based targeted next-generation sequencing for sensitive plasma pathogen cfDNA detection in bloodstream infections. J Transl Med. 2025;23(1):1203. 10.1186/s12967-025-07258-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Chen S, Zhou Y, Chen Y, Gu J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinf (Oxford England). 2018;34(17):i884–90. 10.1093/bioinformatics/bty560. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Langmead B, Trapnell C, Pop M, Salzberg SL. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 2009;10(3):R25. 10.1186/gb-2009-10-3-r25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Waites KB, Xiao L, Liu Y, Balish MF, Atkinson TP. Mycoplasma pneumoniae from the Respiratory Tract and Beyond. Clin Microbiol Rev. 2017;30(3):747–809. 10.1128/CMR.00114-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Li ZJ, Zhang HY, Ren LL, Lu QB, Ren X, Zhang CH, et al. Etiological and epidemiological features of acute respiratory infections in China. Nat Commun. 2021;12(1):5026. 10.1038/s41467-021-25120-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kaya S, Telci Caklili O, Oguz A. Community-Acquired Pneumonia Requiring Hospitalization. N Engl J Med. 2015;373(24):2381. 10.1056/NEJMc1511751. [DOI] [PubMed] [Google Scholar]
- 23.Jain S, Williams DJ, Arnold SR, Ampofo K, Bramley AM, Reed C, et al. Community-acquired pneumonia requiring hospitalization among U.S. children. N Engl J Med. 2015;372(9):835–45. 10.1056/NEJMoa1405870. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Wang Q, Wu D, Zhang Y, Zeng Q, Wang J, Lv X. tNGS-based detection of respiratory pathogens in a single center: associations with age, gender, season, and co-infections. Front Cell Infect Microbiol. 2025;15:1663234. 10.3389/fcimb.2025.1663234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Trifonova I, Korsun N, Levterova V, Pavlova D, Simeonovski I, Ivanova M, et al. Respiratory infections in the post-COVID-19 era: impact, prevalence, and clinical characteristics of bacterial and viral co-infections. Front Med (Lausanne). 2025;12:1597782. 10.3389/fmed.2025.1597782. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Swanson EC, Schleiss MR. Congenital cytomegalovirus infection: new prospects for prevention and therapy. Pediatr Clin North Am. 2013;60(2):335–49. 10.1016/j.pcl.2012.12.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Coskun E, Kakkar F, Riley LE, Ciaranello AL, Prabhu M. Evaluation and Management of Congenital Cytomegalovirus Infection. Obstet Gynecol. 2025;145(3):297–306. 10.1097/aog.0000000000005840. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Citations
- Collaborators GBDLRI. Estimates of the global, regional, and national morbidity, mortality, and aetiologies of lower respiratory infections in 195 countries, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Infect Dis. 2018;18(11):1191–210. 10.1016/S1473-30991830310-4. [DOI] [PMC free article] [PubMed]
Supplementary Materials
Data Availability Statement
Data and materials may be made available upon written request to the corresponding author.