Etiological characteristics of community-acquired bacterial pneumonia in children: A seven-year retrospective, hospital-based study

To the Editor: Community-acquired bacterial pneumonia (CABP) is a significant contributor to morbidity and mortality, accounting for approximately 64% of pneumonia-related deaths in children under 5 years old.[1] Since 2020, a series of non-pharmaceutical interventions have had a significant impact on the global seasonal patterns of respiratory pathogens.[2] Because of the “immunity debt” caused by decreased exposure to prevalent pathogens, a resurgence of certain infectious diseases has occurred in recent years. In 2023, a surge in Mycoplasma pneumoniae (M. pneumoniae) infections attracted global attention, notable not only for their increased incidence but also for their severity. These respiratory infection outbreaks have required us to refocus on the epidemic characteristics of bacterial community-acquired pneumonia (CAP) in children, especially those of younger age. Co-infections have been reported to exacerbate disease severity,[3] underscoring the importance of understanding the intricate interactions between diverse pathogens. The pathogen spectrum of CAP is influenced by multiple factors and shows dynamic variations across regions, seasons, and age groups,[4] yet there remains a significant gap in our understanding of bacterial co-infections, particularly in terms of epidemiological patterns and contribution to disease severity. It is necessary to perform updated etiological surveillance, particularly for bacterial pathogens, which may exhibit distinct patterns of transmission and resistance in the post-pandemic era. This study, conducted at Beijing Children’s Hospital from January 2017 to December 2023, aimed to explore the epidemiological characteristics of 18 bacterial pathogens in pediatric inpatients with CAP by age and year and to assess their correlation with disease severity. The detailed methods are presented in the Supplementary Files, https://links.lww.com/CM9/C636. The study was approved by the Ethics Committee of Beijing Children’s Hospital (No. 2024-Y-126-D), with informed consent waived. Of 15,556 eligible hospitalized children, 5135 (33.0%) were enrolled in the final analysis [Supplementary Figure 1, https://links.lww.com/CM9/C636], with 2395 (46.6%) stratified into the severe CAP (SCAP) cohort and 2740 (53.4%) in the non-SCAP cohort. The majority of patients were male (54.4%) and the median age was 5 years (interquartile range: 1 to 8 years). A gradual shift of increasing SCAP prevalence was observed (from 16.8% [43/256] to 66.1% [1419/2145]) from 2017 to 2023 (P <0.001). SCAP cases had significantly elevated intensive care unit (ICU) admission rates (59.9% [226/377] vs. 40.1% [151/2740]) [Supplementary Table 1, https://links.lww.com/CM9/C636]. The most commonly detected bacterial pathogen was M. pneumoniae (36.3% [1863/5135]), followed by Haemophilus influenzae (H. influenzae;12.2% [624/5135]), Streptococcus pneumoniae (S. pneumoniae; 11.6% [595/5135]), Staphylococcus aureus (4.6% [235/5135]), and Klebsiella pneumoniae (K. pneumoniae; 2.1% [107/5135]); Nocardia stellate, Chlamydia trachomatis, Pneumocystis jirovecii, and Chlamydia psittaci were not detected. M. pneumoniae was the predominant bacteria detected in children >5 years of age, with the highest proportion observed in those 5–9 years old (55.8% [1166/2091]). In total, 3102 (60.4%) children had at least one positive result for the 18 bacterial pathogens examined [Supplementary Table 2, https://links.lww.com/CM9/C636]; among them, 74.9% (2322/3102) had positive detection for a single species, with the rankings from most to least prevalent being M. pneumoniae (59.7% [1387/2322]), H. influenzae (11.5% [268/2322]), and S. pneumoniae (11.3% [263/2322]). Co-infections with two bacteria were detected in 625 (20.1% [625/3102]) children; multi-bacterial co-infections (>2 bacteria) were relatively rare, detected in only 155 cases (5.0% [155/3102]). The pathogen spectrum in children with positive bacterial results was further analyzed based on infection severity and patient age [Figure 1A]. In total, 1384 patients were diagnosed with SCAP (44.6%). The distribution of SCAP differed among the four groups analyzed (P <0.001); children <1 year old had fewer SCAP cases (9.2%, 128/1384) than older groups, especially those 5–9 years old (49.8%, 689/1384). The prevalence of M. pneumoniae was significantly higher in SCAP-positive patients compared with that in non-SCAP patients (58.7% [959/1634] vs. 41.9% [904/2156], respectively; P <0.001). In contrast, S. pneumoniae (12.8% [209/1634] vs. 17.9%, [386/2156], P <0.001), S. aureus (4.3% [70/1634] vs. 7.7% [165/2156], P <0.001) and H. influenzae (13.4% [219/1634] vs. 18.7% [405/2156], P <0.001) were found in significantly lower proportions in SCAP cases compared with the findings in non-SCAP cases, respectively. When the disease severity of 3102 patients was further analyzed within age subgroups, M. pneumoniae (43.5% [147/338] vs. 35.9% [142/395], P = 0.037), K. pneumoniae (3.3% [11/338] vs. 0.8% [3/395], P = 0.014), and Acinetobacter baumannii (7.1% [24/338] vs. 3.3% [13/395], P = 0.019) were found in significantly higher proportions in SCAP cases compared with the findings in non-SCAP cases among children aged 1–4 years, respectively. In contrast, S. pneumoniae (10.0% [69/689] vs. 18.4% [133/722], P <0.001), H. influenzae (12.0% [83/689] vs. 25.2% [182/722], P <0.001) and S. aureus (2.6% [18/689] vs. 6.2% [45/722], P = 0.001) were found in a significantly lower proportion in SCAP cases compared with the findings in non-SCAP cases among children aged 5–9 years, respectively [Supplementary Table 3, https://links.lww.com/CM9/C636].Figure 1: Bacterial epidemiology and risk factors of pediatric community-acquired pneumonia, 2017–2023. (A) Proportion of pathogens in SCAP and non-SCAP cases, calculated by positive cases over total positive cases. The lengths of the bars indicate the positive rate or proportion of each bacterial pathogen. (B) Seasonal positive case numbers of the top five pathogens causing CAP, stratified by disease severity. (C) Heatmap of co-infection rates in SCAP and non-SCAP patients across the three periods. The color of dots represents the CAP/SCAP co-infection rates; the color of grids color represents non-SCAP. Larger, darker circles indicate higher rates. (D) Risk factors of ICU admission for CAP. CAP: Community-acquired pneumonia; CI: Confidence interval; MRSA: Methicillin-resistant Staphylococcus aureus; NCAP: Non-community-acquired pneumonia; OR: Odds ratio; PICU: Pediatric intensive care unit; SCAP: Severe community-acquired bacterial pneumonia.From 2017 to 2023, M. pneumoniae consistently emerged as the most prevalent pathogen; after decreasing from 2018 (46.0%, 435/946) to the lowest level in 2021 (25.2%, 111/441), the rate of detection increased during 2022 and 2023 with yearly positive rates of 28.8% and 33.0%, respectively. The prevalence rates of H. influenzae and S. pneumoniae were second- and third-most abundant, exhibiting similarity during the examined years. However, the prevalence of S. pneumoniae, H. influenzae, S. aureus, and K. pneumoniae decreased over the examined years (P <0.001 and 0.016, 0.002, and 0.008, respectively) [Supplementary Figure 2, https://links.lww.com/CM9/C636]. From 2017 to 2019, there was no statistically significant difference in M. pneumoniae detection rates between patients with or without SCAP (odds ratio [OR]: 1.1, 95% confidence interval [CI]: 0.90–1.33, P = 0.351). However, starting from 2020, there was an increase in the detection rate of M. pneumoniae among severe cases, with a significant rise observed from 2020 to 2022 (OR: 2.5, 95% CI: 1.83–3.34, P <0.001) and 2023 (OR: 3.00, 95% CI: 2.27–3.91, P <0.001). Between 2020 to 2022, there was no discrepancy in the detection rates of H. influenzae between SCAP and non-SCAP (OR: 0.90, 95% CI: 0.53–1.67, P = 0.841). However, both before and following this time frame, its detection rate in SCAP consistently remained lower during 2017–2019 (OR: 0.60, 95% CI: 0.42–0.74, P <0.001) and 2023 (OR: 0.60, 95% CI: 0.42–0.78, P <0.002). The increase in the detection of S. aureus in SCAP was only observed from 2020 to 2022 (OR: 0.40, 95% CI: 0.17–0.96, P = 0.042) [Supplementary Table 4, https://links.lww.com/CM9/C636]. The detection rate of M. pneumoniae exhibited a clear seasonal peak during autumn and winter, with a significant outbreak in 2023 that accounted for 32.5% of the total M. pneumoniae-associated CAP during the 6-year period. H. influenzae and S. pneumoniae showed similar but relatively low seasonal trends. Detection rates for all bacterial pathogens were low in 2020. Before 2020, a majority of the M. pneumoniae-associated CAP were non-severe; a shift occurred in subsequent years, especially after the summer of 2023, leading to a dramatic increase in the reported cases of SCAP [Figure 1B]. Co-infections were identified in 780 (15.1% [780/5135]) children, with 3.9% (202/5135) being H. influenzae–M. pneumonia, 3.7% (188/5135) being S. pneumoniae–M. pneumoniae, and 2.2% (113/5135) being S. pneumoniae–H. influenzae. The co-infection rates among patients with SCAP significantly decreased to 4.7% (17/361) and 7.0% (100/1419) between 2020–2022 and 2023, respectively, compared with 26.3% (162/615) in previous years (P <0.001). The period from 2020–2022 exhibited a more intricate pattern of co-infections compared to before and following years [Figure 1C]. The risk factors identified for ICU admission include being male (OR: 1.35, 95% CI: 1.02–1.81); having disease onset during the cold season in Q4 or Q1 (OR: 1.79, 95% CI: 1.35–2.39); younger age, specifically children aged 1–4 years (OR: 3.34, 95% CI: 1.97–5.79) and <1 year (OR: 4.33, 95% CI: 2.50–7.49), and multi-bacterial co-infections with >2 bacteria (OR: 2.51, 95% CI: 1.55–4.05) [Figure 1D, Supplementary Table 5, https://links.lww.com/CM9/C636]. Our study identified M. pneumoniae as the predominant pathogen in bacterial pneumonia, particularly among children aged 5–9 years, consistent with previous research.[5] These findings underscore the necessity for age-specific strategies in diagnosing and managing pediatric CAP because susceptibilities vary across different age groups. A gradual increase in M. pneumoniae detection rates among younger children and a rise in SCAP, especially in 2023, highlights the evolving epidemiological landscape of M. pneumoniae pneumonia in pediatrics. Significant differences in the infection and co-infection patterns were observed across age groups, years, and seasons. The coronavirus disease-2019 pandemic and its associated non-pharmaceutical interventions caused a temporary dip in detection rates. However, with the easing of these measures and the return to regular social activities, there was a resurgence of cases in 2023, potentially because of the “immunity debt” accumulated during the pandemic period. The changes in circulating epidemic strains, the increasing resistance to commonly used antibiotics, and vaccine coverage targeting specific bacteria may have contributed to the changed epidemiological patterns. In this study, co-infections particularly involving combinations of bacterial pathogens including M. pneumoniae, S. pneumoniae, and H. influenzae exacerbated CAP disease severity. Bacterial and viral co-infection also plays a significant role in childhood CAP. The mechanisms of co-infection may include enhancing colonization, inducing synergistic inflammatory responses, altering quorum sensing to increase virulence, and facilitating the exchange of antibiotic-resistance genes, ultimately leading to higher ICU admission in pediatric CAP. Given these risk factors associated with ICU admission, younger children, especially those who are infected in cold seasons, are a key population for continuous surveillance. Early detection and timely treatment of children infected with multiple pathogens are crucial for mitigating the rising risk of developing severe CAP and improving patient outcomes. The bacterial etiology of pediatric CAP has evolved significantly from 2017 to 2023, with a marked rise in cases of SCAP. M. pneumoniae remains the most predominant pathogen, especially in severe cases. These trends emphasize the importance of ongoing surveillance and the development of targeted strategies to effectively manage bacterial CAP in children. Funding This work was supported by grants from the Beijing Municipal Natural Science Foundation (No. L246011), the National Natural Science Foundation of China (No.82073573), the Training Plan for High-Level Public Health Technical Talents of Beijing Municipal Health Commission (No.2022-02-04) and Funding for Reform and Development of Beijing Municipal Health Commission (No. EYGF-HX-05). Conflicts of interest None.