The case numbers were collected according to the study aims (Supplementary Table S1). All cases (aged 1 month–94 years) were diagnosed with acute respiratory infections (ARIs) or pneumonia refer to the previous study (Li et al., 2021).To be specific, for comprehensive respiratory pathogen panel testing, we included 2484 patients presenting with respiratory tract infection (RTI) symptoms to the emergency department of Peking Union Medical College Hospital (PUMCH) (during September 2023–February 2024). Nasopharyngeal swab (NPS) specimens were collected from all participants and transferred to the clinical laboratory for comprehensive respiratory pathogen panel testing before initiating therapeutic interventions.

For cytokine profiling analysis, we retrospectively examined 209 RTI cases with available biospecimens collected during three distinct epidemiological periods: November 2018–December 2019 (Group 1,n=41), September 2023–February 2024 (Group 2,n=70), and September 2024–December 2024 (Group 3,n=98). All specimens were immediately processed within 24 hours of collection following standardized protocols. The study protocol received ethical approval from the PUMCH Ethics Committee (No. I-22PJ860).

The NPS were used for nucleic acid extraction by using the Nucleic Acid Extraction and Purification Kit (Xi ‘an Tianlong Technology Co., Ltd, Xi’an, China) on GeneRotex 96 Automatic Nucleic Acid Extraction instrument (Xi ‘an Tianlong Technology Co., Ltd, Xi’an, China) following producer’s protocols.

The detection of respiratory pathogens was performed using the Respiratory Tract Pathogen Nucleic Acid Detection Kit (Coyote Bioscience, Beijing, China), for the qualitative detection of nucleic acids from 29 respiratory pathogens, including 14 types of RNA viruses (Influenza A virus (IFVA), Influenza B virus (IFVB), Parainfluenza virus type 1 (HPIV1), Parainfluenza virus type 2 (HPIV2), Parainfluenza virus type 3(HPIV3), Parainfluenza virus type 4 (HPIV4), Coronavirus 229E(HCoV229E), Coronavirus OC43 (HCoVOC43), Coronavirus NL63(HCoVNL63), Coronavirus HKU1(HCoVHKU1), Respiratory syncytial virus (RSV), Human metapneumovirus (HMPV), Human Rhinovirus (HRV), Measles virus (MeV), 2 types of DNA viruses (Human Adenovirus (HAdV), Human bocavirus (HBoV)), 2 types of atypical pathogens (Mycoplasma pneumoniae, Chlamydia pneumoniae), and 11 types of bacterial species (Group A streptococcus (GAS), Streptococcus pneumoniae, Haemophilus influenzae, Legionella pneumophila, Klebsiella pneumoniae, Pseudomonas aeruginosa, Staphylococcus aureus, Moraxella catarrhalis, Escherichia coli, Acinetobacter baumannii, Bordetella pertussis from nasopharyngeal and oropharyngeal swab samples. The detection process utilizes PCR amplification combined with fluorescence probe technology. Specific primers and fluorescent probes (FAM, VIC, Cy5, Texas Red) targeted microbial genetic material in eight separate reaction wells. Real-time fluorescence signals were recorded during amplification on Tianlong Gentier 96E. For amplification, 7µL nucleic acid extraction was added into an 18 µL prepared PCR reaction mix. The PCR cycling conditions included an initial reverse transcription step at 42°C for 5 minutes, followed by denaturation at 95°C for 1 minute, and 45 cycles of amplification (95°C for 5 seconds and 60°C for 30 seconds, with fluorescence collection). Positive and negative controls were included in each run to ensure assay validity. Data were analyzed by monitoring the threshold cycle (Ct) values, with results interpreted according to predefined quality control parameters.

Cytokine levels were measured using multiple cytokines (12-items) Detection Kit (Flowcytometry Fluorescence Luminance Method, Joinstar Biomedical Technology Co,.Ltd., Hangzhou, China), which is based on double antibody sandwich flow cytometry and liquid suspension chip technology to detect a variety of cytokines. The 12 cytokines included interleukin (IL)-1β, IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-12p70, IL-17, tumor necrosis factor (TNF)-α,interferon (IFN)-α, and IFN-γ. 200µL NPS samples were measured using the kit following producer’s protocols. The fluorescent antibody signal is captured, decoded and clustered by iMatrix 100 system. According to the intensity of the fluorescent signal, the concentration of each cytokine in the sample is calculated through the calibration curve.

Descriptive statistics for continuous variables are presented as medians with interquartile ranges (IQR). The Mann–Whitney U-test was used for comparisons between two groups. Categorical variables were expressed as % (m/n) and examined using χ2/Fisher’s exact test. P value < 0.05 was considered statistically significant. Statistical analyses were performed and graphs were plotted using R (4.2.1).

This study encompassed 2484 cases with a relatively balanced sex distribution (48.11% male, 51.89% female). The cohort was predominantly pediatric, with 69.04% (n=1715) of cases in children (<18 years) and 30.96% (n=769) in adults, reflecting a clear majority of pediatric cases during the surveillance period. Cases were temporally distributed across six months (September 2023–February 2024), with peak incidence in October (36.07%) and December (24.19%). Overall, 85.02% of cases presented as ARI, while 14.98% were diagnosed with pneumonia (Table 1). Characteristics of patients involved in this study were presented in Supplementary Table S2.

Among 2484 patients screened for 29 respiratory pathogens, 70.73%(1757/2484) tested positive for at least one pathogen 2894 pathogens were detected in 2484 cases with viral, bacterial and atypical pathogens were identified in 40.42%(1004/2484), 51.45%(1278/2484)cases, respectively (Supplementary Table S3). A statistically significant age-related difference was observed, with pediatric cases demonstrating higher overall positivity rates (74.11%, 1271/1715) compared to adults (63.20%, 486/769) (Table 1, P<0.05) This disparity was evident across both viral (41.57% vs. 37.84%) and non-viral pathogens (57.38% vs. 38.23%) in children versus adults, respectively (P < 0.05, Supplementary Table S3).

The predominant viral pathogens during the observation period were IFVA (9.78%, 243/2484), IFVB(7.97%, 198/2484), and HAdV(6.32%, 157/2484). Among bacterial pathogens, Haemophilus influenzae (26.28%, 653/2484) and Moraxella catarrhalis (8.90%, 221/2484) demonstrated highest prevalence, while Mycoplasma pneumoniae (6.76%, 168/2484) predominated among atypical pathogens. Notably, none of the 2484 cases tested positive for the three specific pathogens, HCoVNL63, MeV, and L. pneumophila (Figure 1A).

Age-specific analysis showed peak positivity among school-aged children (7–12 years: 76.61%, 583/761) and lowest detection rates in older adults (>60 years: 52.74%, 77/146) (Table 1). Median age of infection for most pathogens clustered around 10 years, with notable exceptions being IFVB (median age 26 years, IQR 8-38) and HCoVOC43 (median age 23 years, IQR 17-45.5) (Figure 1B). Age-stratified analysis of 29 respiratory pathogens demonstrated significant variations in detection patterns: both IFVA and IFVB demonstrated consistently high detection rates across all age cohorts, with particularly pronounced prevalence among youths (13–18 years)and adults (19–35 and 36–60 years). Pediatric populations exhibited significantly higher infection rates for HPIV(primarily HPIV3 and HPIV4 subtypes) compared to adult groups (Figure 2A). HMPV demonstrated a bimodal distribution across age extremes, with elevated rates in both pediatric and elderly cohorts. Regarding bacterial and atypical pathogens, H. influenzae maintained high prevalence across all age groups except seniors (>60 years), while M. catarrhalis showed predilection for younger pediatric populations. Notably, M. pneumoniae maintained broad age distribution with significant prevalence among school-aged children (Figure 2B).

Pneumonia complications occurred in 372 cases (14.98%,372/2484), with significant age-related disparity: pediatric patients (<18 years) exhibited lower incidence (10.79%, 185/1715) compared to adults (24.32%, 187/769; p<0.05) (Table 1). Pathogen-specific pneumonia risk stratification (Figure 1C) identified M. pneumoniae as having strongest association (41.1%, 69/168), followed by A. baumannii (25.0%, 9/36), HRV (22.2%, 24/108), and HMPV (20.8%, 22/106). Within the pneumonia group, a statistically significant disparity in viral versus non-viral pathogen detection rates was identified in pediatric patients (P < 0.05)(Supplementary Table S3).

Bacterial and viral etiologies exhibited distinct age-specific distribution patterns in pneumonia patients (Supplementary Table S4). M. pneumoniae was identified as the predominant pathogen in pediatric pneumonia cases, accounting for 33.5% of identified pathogens, whereas H. influenzae demonstrated higher prevalence in adult populations, constituting 12.3% of cases. Respiratory virus detection revealed significant age-related variations: HRV showed greater frequency in pediatric cases (10.3%) compared to adult cases (2.7%). Conversely, IFVB exhibited an inverse distribution pattern, being detected in 10.2% of adult patients versus 2.2% of pediatric cases.

Among the cohort, 38.85% (965/2484) tested positive for a single pathogen, 31.88%(792/2484) for multiple pathogens(Supplementary Table S5, P<0.05). Pediatric populations demonstrated significantly higher co-infection prevalence compared to adults (37.26% vs. 19.90%, P < 0.05), with peak co-infection rates observed in school-aged children (7–18 years: 37.39%, 341/912) compared to the lowest rates in elderly adults (>60 years: 17.81%, 26/146) (Supplementary Table S5).

Network analysis revealed distinct age-specific co-infection patterns. H. influenzae occupied a central hub in the interaction network, with 66.0% (431/653)of its cases involving co-infections.(Figure 3A). Pediatric cases predominantly featured H. influenzae with HAdV, IFVA or M. pneumoniae, while adult cases showed higher prevalence of H. influenzae-IFVB, H. influenzae-IFVA and S. pneumoniae-IFVB combinations (Figures 3B, C). Several pathogen pairs demonstrated statistically significant correlations: M. pneumoniae significantly co-occurred with HRV (43.52% co-occurrence vs. 5.09% without HRV; P < 0.001), while K. pneumoniae-E. coli and S. pneumoniae-IFVB combinations also showed significant associations (P < 0.05, Figure 3D). High-frequency co-infection combinations (prevalence >1%) peaked in children around 10 years of age (Figure 3E).

While co-infection status did not elevate pneumonia risk overall, specific combinations significantly increased fever incidence. Co-infection with IFVB and S. pneumoniae (median age 12 years) resulted in markedly elevated fever ratio (81.5%) compared to single infections with IFVB alone (57.3%, median age 32 years) or S. pneumoniae alone (60.7%, median age 34 years; P < 0.05). Similar trends were observed for IFVA and M. catarrhalis co-infections.

Temporal trends revealed fluctuating testing positive rate(TPR) throughout the surveillance period, commencing at 54.79% (103/188) in September 2023. Positivity rates escalated to >70% during October 2023–January 2024, before decreasing to 63.8% (104/163) by February 2024 (Table 1).

Pathogen-specific temporal patterns demonstrated distinct epidemiological trajectories. IFVA, HAdV, RSV, and HMPV exhibited progressive increases from September baseline levels, while HRV maintained stable detection rates throughout the surveillance period. Notably, IFVA detection peaked in December 2023 (17.46%) before sharply declining to 2.3% by January 2024. Conversely, IFVB showed an inverse pattern, rising from December 2023 to peak at 30.17% in January 2024 (Figure 4A). Among bacterial pathogens, H. influenzae remained persistently elevated across all surveillance months. M. catarrhalis, K. pneumoniae, and E. coli showed transient elevations during October-November before declining to baseline levels from December onward. In contrast, S. pneumoniae detection rates increased substantially during December compared to preceding months (Figure 4B).

To characterize post-pandemic immune responses to respiratory pathogens, we analyzed cytokine profiles in nasopharyngeal swab (NPS) samples across three epidemiological cohorts. Multiplex cytokine analysis revealed significant intergroup variations specifically in IL-6, IL-8, and IL-1β levels. Group 2 demonstrated markedly reduced cytokine concentrations compared to both Group 1 and Group 3, particularly evident in IL-6 (Group 2: 3.82 [3.59-5.73] pg/mL vs Group 1: 6.12 [3.95-12.32] pg/mL vs Group 3: 4.60 [3.71-21.40] pg/mL) and IL-8 levels (Group 2: 18.35 [15.38-26.68] pg/mL vs Group 1: 37.98 [17.41-56.11] pg/mL vs Group 3: 34.7 [17.38-141.48] pg/mL) (Figure 5A). Subgroup analysis revealed more pronounced cytokine level alterations in pediatric patients compared to adults across all study groups (Figure 5B). Similarly, pneumonia patients exhibited enhanced cytokine responses relative to non-pneumonia cases, mirroring the age-related response patterns (Figure 5C).