Abstract
Paediatric brain abscess (BA) is a rare but life–threatening condition, and large-scale epidemiological studies are limited. We aimed to characterize the temporal trend of pediatric BA in China during 2016–2023. We conducted a nationwide study using the FuTang Update Medical Records (FUTURE) database, which includes face-sheets of discharge medical records from 33 tertiary children’s hospitals across China. Case-identification strategies were validated, and the optimal strategy was applied to identify BA cases. Demographic information, predisposing factors, complications, and clinical outcomes of included cases were collected and analyzed. The main analytic cohort was constructed at the index qualifying episode level. A total of 799 children were identified, with 6.4–8.4 patients per 100,000 hospitalized children annually (P = 0.377). After 2020, the proportion of patients aged 1–3 years decreased, whereas the proportions of patients aged 10–12 and 13–17 years increased (all P < 0.05). Among predisposing factors, contiguous-site infections were more common in 2020–2023 than in 2016–2019 (OR, 1.997; 95% CI, 1.287–3.100; P < 0.001), whereas congenital heart disease remained stable. Neurosurgical intervention, complications, length of stay (LOS), and hospitalization costs were higher in 2020–2023 after adjustment for age, sex, and predisposing factors (all P < 0.05).
Conclusion: Although the proportion of BA among hospitalized children in China remains stable, the distribution of its predisposing factors has changed, and increases in LOS and hospitalization costs were observed.
|
What is Known: • Bacterial meningitis was reported as the most common predisposing factor for paediatric brain abscess (BA) in a single-centre study in China during 2006–2016. • Large-scale epidemiological data on paediatric BA are limited. | |
|
What is New: • This nationwide study in China showed changes in the predisposing factors for paediatric BA from 2016 to 2023. • Contiguous-site infections increased over time and were more common in 2020–2023 than in 2016–2019. |
Supplementary Information
The online version contains supplementary material available at 10.1007/s00431-026-07186-w.
Keywords: Brain Abscess, Child, Sinusitis, Otitis Media, Neurosurgery, COVID-19
Introduction
Brain abscess (BA) is a rare but life-threatening central nervous system (CNS) infection in children, with reported mortality rates ranging from 4 to 21% [1–3]. Epidemiological studies on this condition, particularly with respect to high–risk factors, are important for guiding diagnosis and empirical antibiotic treatment. Nationwide studies from Denmark have provided valuable insights into the epidemiology of BA across all age groups [4–7]. However, the paediatric subgroup in these studies was relatively small, comprising 155 children under 20 years of age. Most other studies on paediatric BA are limited to small case series [1, 8–11].
In China, a single-centre study conducted from 2006 to 2016 revealed bacterial meningitis and congenital heart disease (CHD) as the two most common predisposing factors for paediatric BA [3]. However, several changes have occurred in China over the past few years. The coronavirus disease 2019 (COVID-19) pandemic occurred between 2020 and 2022. During this period, the epidemiology of respiratory viral infections [12] and bacterial meningitis [13] was reported. In addition, although CHD was reported to be the most common birth defect in 2005, the prevalence of CHD as well as CHD-related hospitalizations in children in China has decreased in recent years [14, 15]. The impact of these epidemiological shifts on paediatric BA in China remains unknown.
Therefore, we conducted a nationwide retrospective study to characterize the temporal trend of paediatric BA from 2016 to 2023.
Materials and methods
Participants and setting
Data for this study were gathered from the FuTang Update Medical Records (FUTURE) database, which stores face-sheets of discharge medical records (FSMRs) from 33 tertiary children's hospitals [16]. Children with BA aged 0–17 years who were admitted to 33 hospitals between January 1, 2016, and December 31, 2023, were included from the FUTURE database.
The overall workflow of the study is illustrated in Fig. 1. First, we evaluated the positive predictive values (PPVs) and sensitivity of different case-identification strategies in the FUTURE database. The detailed strategies are presented in Online Resource 1 Supplementary Table S1. Second, on the basis of the results of the first step, we selected the optimal strategy to include the analysis cohort. Finally, we collected and analysed the data.
Fig. 1.
Study flowchart. Confirmed BA was defined as definite or probable BA. BA, brain abscess; BCH, Beijing Children’s Hospital; FUTURE, FuTang Update Medical Records database; MRS, medical record system; PPV, positive predictive value
Evaluation of the PPVs and sensitivity
The calculation of the PPVs and sensitivity for each case-identification strategy is shown in Fig. 1. The PPV is defined as the proportion of children with confirmed BA among all children with BA identified by the strategy. The sensitivity of each strategy is defined as the proportion of confirmed BA cases identified in the FUTURE database compared to all confirmed BA cases identified in the MRS of BCH [17]. For children identified in the FUTURE database who were admitted to Beijing Children's Hospital (BCH), detailed medical notes could be retrieved from the Medical Record System (MRS) of BCH. On the basis of the detailed medical notes from the MRS of BCH, the children identified in the FUTURE database were classified into four categories: definite BA, probable BA, possible BA, and not BA [17]. The first two categories were defined as confirmed BA. The specific criteria for these categories are provided in Online Resource 1 Supplementary Table S2. Two experienced infectious disease specialists independently reviewed the clinical information and classified the children with BA. Any disagreements in classification were resolved through adjudication by a third specialist.
Data cleaning and case identification
The workflow of data cleaning and case identification is shown in Fig. 2. Briefly, all hospitalization records with BA-related ICD–10 codes in any diagnostic field were retrieved as candidate records. Repeated records from the same child were grouped into a cluster, and consecutive admissions were merged into a single hospitalization episode when the discharge and readmission dates were the same day or the next day and when the principal ICD–10 diagnosis codes were identical. For merged episodes, clinical events were combined across linked records, and the LOS and hospitalization costs were summed. The validated optimal strategy was then applied to all the candidate episodes. For children with more than one qualifying BA episode, only the index qualifying episode was retained. Therefore, the final analytic cohort included one index qualifying BA episode per child.
Fig. 2.
Workflow of data cleaning and case identification. Hospitalization records with BA-related ICD-10 codes in any diagnostic field were retrieved as candidate records. Repeated records from the same child were grouped into clusters. Consecutive admissions were merged into a single hospitalization episode when the discharge and readmission dates were the same day or the next day and when the principal ICD-10 diagnosis codes were identical. For children with more than one qualifying BA episode, only the first qualifying BA episode was retained as the index qualifying BA episode. The final analytic cohort included one index qualifying BA episode per child. BA, brain abscess
Variables and statistical analysis
The variable definitions are shown in Online Resource 1 Supplementary Table S3. Discharge outcomes were categorized according to the discharge status recorded on the discharge face-sheet: discharge by doctor’s orders, leave against medical advice, and in-hospital death. In-hospital mortality was defined only as a discharge status of death. Children who left against medical advice while alive were analysed as a separate category and were not considered in-hospital deaths. No missing values were identified for the variables included in the main analyses.
Categorical variables are presented as counts and percentages. Continuous variables were assessed for normality using the Shapiro–Wilk test, histograms, and Q–Q plots (Online Resource 3); non-normally distributed variables are presented as medians and interquartile ranges (IQRs). Period differences between 2016–2019 and 2020–2023 and annual temporal trends from 2016 to 2023 were assessed for age distribution, predisposing factors, complications, neurosurgical intervention, length of stay (LOS), and hospitalization costs. Binary variables were analysed using logistic regression models, and the LOS and hospitalization costs were analysed using generalized linear models with a gamma distribution and log link. For complications, neurosurgical intervention, LOS, and hospitalization costs, both period comparisons and temporal trend analyses were further adjusted for sex, age, and predisposing factors. Effect estimates are reported as odds ratios for logistic models and ratios for generalized linear models, with 95% confidence intervals. Statistical analysis was performed using SPSS 21.0 (SPSS Inc., Chicago, IL), and a P value of < 0.05 was considered to indicate statistical significance.
To account for potential hospital-level effects, we further fitted mixed-effects models with hospital included as a random intercept. To assess the influence of high-contribution hospitals, we also performed a sensitivity analysis in which hospitals contributing more than 5% of all included BA cases were excluded one at a time. The two-period comparisons (2016–2019 vs. 2020–2023) were then repeated for predisposing factors, neurosurgical intervention, complications, LOS, and hospitalization costs.
Results
The positive predictive value and sensitivity of the FUTURE database
Through an initial search in the MRS of BCH for children admitted to BCH during the years 2016–2023, 164 children were diagnosed with BA (primary, secondary, or otherwise). A subsequent review of their detailed clinical information led to the following classification: 36 as not BA, 11 as possible BA, 16 as probable BA, and 101 as definite BA. Therefore, 117 children were identified as having confirmed BA in the MRS of BCH. Detailed information on these children with confirmed BA is shown in Online Resource 1 Supplementary Table S4.
The detailed results for the PPV and sensitivity for each strategy are presented in Online Resource 1 Supplementary Table S1. The strategy used to identify the analytic cohort across the entire FUTURE database had a PPV of 81.8% and a sensitivity of 95.7%.
Demographics
A total of 799 children diagnosed with brain abscess between January 2016 and December 2023 were included. The annual number of cases ranged from 84 to 125, accounting for 6.4–8.4 cases per 100,000 hospitalized children (Online Resource 1 Supplementary Table S5), with no statistically significant changes across years (P = 0.377). The detailed clinical characteristics are shown in Table 1.
Table 1.
Age distribution and clinical characteristics of the analytic cohort
| Total, n (%) N = 799 |
2016–2019, n (%) N = 435 |
2020–2023, n (%) N = 364 |
|
|---|---|---|---|
| Age groups | |||
| < 1 y | 225 (28.2) | 135 (31.0) | 90 (24.7) |
| 1–3 y | 159 (19.9) | 100 (23.0) | 59 (16.2) |
| 4–6 y | 132 (16.5) | 73 (16.8) | 59 (16.2) |
| 7–9 y | 127 (15.9) | 60 (13.8) | 67 (18.4) |
| 10–12 y | 109 (13.6) | 49 (11.3) | 60 (16.5) |
| 13–17 y | 47 (5.9) | 18 (4.1) | 29 (8.0) |
| Predisposing factors | |||
| Meningitis | 128 (16.0) | 82 (18.9) | 46 (12.6) |
| CHD | 117 (14.6) | 59 (13.6) | 58 (15.9) |
| Contiguous-site infections | 94 (11.8) | 37 (8.5) | 57 (15.7) |
| Cranial or skull-base defects | 27 (3.4) | 14 (3.2) | 13 (3.6) |
| Immunocompromising conditions | 47 (5.9) | 23 (5.3) | 24 (6.6) |
| Bloodstream infection | 66 (8.3) | 35 (8.0) | 31 (8.5) |
| Dental infections | 4 (0.5) | 0 (0.0) | 4 (1.1) |
| Previous neurosurgery | 4 (0.5) | 2 (0.5) | 2 (0.5) |
| Pulmonary artery fistula | 4 (0.5) | 1 (0.2) | 3 (0.8) |
| Other underlying diseases | 13 (1.6) | 4 (0.9) | 9 (2.5) |
| Unknown origin | 295 (36.9) | 178 (40.9) | 117 (32.1) |
| Lumbar puncture | 357 (44.7) | 157 (36.1) | 200 (54.9) |
| Neurosurgical intervention | 234 (29.3) | 100 (23.0) | 134 (36.8) |
| Any aspiration | 138 (17.3) | 67 (15.4) | 71 (19.5) |
| Resection | 107 (13.4) | 36 (8.3) | 71 (19.5) |
| Systemic complications | 142 (17.8) | 48 (11.0) | 94 (25.8) |
| Respiratory failure | 126 (15.8) | 41 (9.4) | 85 (23.4) |
| SIADH | 8 (1.0) | 3 (0.7) | 5 (1.4) |
| Intracranial complications | 236 (29.5) | 109 (25.1) | 127 (34.9) |
| Herniation | 94 (11.8) | 42 (9.7) | 52 (14.3) |
| Hydrocephalus | 84 (10.5) | 36 (8.3) | 48 (13.2) |
| Intracranial hemorrhage | 42 (5.3) | 24 (5.5) | 18 (4.9) |
| Ventriculitis | 35 (4.4) | 13 (3.0) | 22 (6.0) |
| Any complications | 312 (39.0) | 137 (31.5) | 175 (48.1) |
| Outcomes | |||
| Discharge by doctor’s orders | 653 (81.7) | 356 (81.8) | 297 (81.6) |
| Leave against medical advice | 135 (16.9) | 71 (16.3) | 64 (17.6) |
| In-hospital death | 11 (1.4) | 8 (1.8) | 3 (0.8) |
| LOSa | 26.6 (12.8, 41.0) | 21.9 (8.9, 38.0) | 29.9 (17.6, 44.7) |
| Hospitalization costsa | 7191.7 (3095.3, 12,287.2) | 6190.6 (1955.1, 10,724.5) | 8591.8 (4884.7, 14,440.5) |
aNon-normally distributed variables, presented as medians and interquartile ranges (IQRs)
The age distribution of the included children is shown in Online Resource 2 Supplementary Fig. S1 and Table 1. In the period-based comparisons, the proportion of children aged 1–3 years was lower in 2020–2023 than in 2016–2019 (OR, 0.65; 95% CI, 0.45–0.93; P = 0.017), whereas the proportions of those aged 10–12 years (OR, 1.56; 95% CI, 1.04–2.33; P = 0.033) and 13–17 years (OR, 2.01; 95% CI, 1.09–3.67; P = 0.024) were greater. In time-trend analyses across the years 2016–2023, the 1–3, 10–12, and 13–17-year age groups showed trends in the same direction (Online Resource 1 Supplementary Table S6).
Predisposing factors
Overall, 504 children (63.1%) had at least one predisposing factor (details in Table 1 and Fig. 3). In the period comparisons, meningitis was less common during the years 2020–2023 than during the years 2016–2019 (OR, 0.623; 95% CI, 0.42–0.92; P = 0.018), whereas contiguous-site infections were more common during the years 2020–2023 (OR, 1.997; 95% CI, 1.29–3.10; P < 0.001). CHD was stable between the two periods (OR, 1.21; 95% CI, 0.82–1.79; P = 0.364). Time-trend analyses across the years 2016–2023 revealed findings similar to those of the period comparisons (Online Resource 1 Supplementary Table S6).
Fig. 3.
Annual distribution of predisposing factors among children with brain abscess. Patients without a predisposing factor are not shown. Other underlying diseases included short bowel syndrome, pilonidal sinus, nephrotic syndrome, portal hypertension, cholestatic hepatitis, brain tumors, Marfan syndrome, intestinal obstruction, necrotizing enterocolitis, inborn errors of metabolism, and ventriculoperitoneal shunt infection. BA, brain abscess; CHD, congenital heart disease
The distribution of predisposing factors varied among different age groups (Fig. 4). Among children aged < 1 year, 41.8% of them had meningitis. In the 1–3, 4–6, and 7–9-year age groups, CHD was the most common predisposing factor, accounting for 9.4% (15/159), 18.9% (25/132), and 22.0% (28/127), respectively. As age increased, the proportion of children with contiguous-site infections also increased, from 1.3% (3/225) in the < 1 year group to 25.7% (28/109) in the 10–12-year group and 21.3% (10/47) in the 13–17-year group.
Fig. 4.
Distribution of predisposing factors by age group among children with brain abscess. Meningitis was the most common predisposing factor in infants. Congenital heart disease (CHD) was the most common predisposing factor in children aged 1–9 years. The proportion of children with contiguous-site infections increased with age and became the most common predisposing factor for children aged 10–17 years. Patients without a predisposing factor are not shown. Other underlying diseases included short bowel syndrome, pilonidal sinus, nephrotic syndrome, portal hypertension, cholestatic hepatitis, brain tumors, Marfan syndrome, intestinal obstruction, necrotizing enterocolitis, inborn errors of metabolism, and ventriculoperitoneal shunt infection. BA, brain abscess; CHD, congenital heart disease
Neurosurgical intervention
In total, 234 (29.3%) children underwent neurosurgical intervention, of whom 176 (75.2%) underwent a single procedure. Among the 132 children who underwent initial aspiration, 31 (23.5%) subsequently required repeat aspiration (n = 26) or abscess resection (n = 5). In the period-based comparisons, neurosurgical intervention was more frequent during the years 2020–2023 than during the years 2016–2019, both in the unadjusted analysis (OR, 1.95; 95% CI, 1.43–2.66; P < 0.001) and after adjustment for age, sex, and predisposing factors (adjusted OR, 2.10; 95% CI, 1.51–2.92; P < 0.001). Time-trend analyses across 2016–2023 revealed a consistent increasing trend in neurosurgical interventions (Online Resource 1 Supplementary Table S6).
Complications
In total, 236 (29.5%) of the included children had at least one intracranial complication, and 142 (17.8%) had at least one systemic complication (Table 1). In the period-based comparisons, complications were more frequent during the years 2020–2023 than during the years 2016–2019 in both the unadjusted analysis (OR, 2.01; 95% CI, 1.51–2.69; P < 0.001) and the analysis adjusted for age, sex, and predisposing factors (adjusted OR, 2.20; 95% CI, 1.63–2.98; P < 0.001). Time-trend analyses also revealed an increasing trend during the years 2016–2023 (Online Resource 1 Supplementary Table S6).
LOS and hospitalization costs
Among the included children, 653 (81.7%) were discharged by doctor’s orders, 135 (16.9%) left against medical advice, and 11 (1.4%) died in the hospital. The median LOS (IQR) was 26.6 (12.8, 41.0) days, and the median hospitalization cost was $7191.7 (3095.3, 12287.2). In the period–based comparisons, both the LOS and hospitalization costs were higher during the years 2020–2023 than during the years 2016–2019 in the unadjusted analyses (LOS: ratio, 1.26; 95% CI, 1.09–1.44; P = 0.001; cost: ratio, 1.37; 95% CI, 1.19–1.58; P < 0.001) and in the analysis adjusted for age, sex, and predisposing factors (LOS: adjusted ratio, 1.31; 95% CI, 1.14–1.52; P < 0.001; cost: adjusted ratio, 1.35; 95% CI, 1.17–1.56; P < 0.001). Time-trend analyses across 2016–2023 revealed consistent increasing trends in both the LOS and hospitalization costs (Online Resource 1 Supplementary Table S6).
Sensitivity analysis
In mixed-effects models including hospital as a random intercept, the findings were consistent with the primary analyses (Supplementary Table S7). In 2020–2023, meningitis-associated BA was less frequent, whereas contiguous-site infections, neurosurgical intervention, complications, LOS, and hospitalization costs were higher.
In sensitivity analyses excluding hospitals with high case contributions, contiguous-site infections, neurosurgical intervention, complications, LOS, and hospitalization costs remained higher in 2020–2023. After excluding BCH data, the decrease in meningitis-associated BA was attenuated and no longer statistically significant (Online Resource 1, Supplementary Tables S8–S16).
Discussion
In this large nationwide study, we characterized temporal changes in paediatric BA in China during the years 2016–2023. We observed a shift in predisposing conditions: the proportion of patients with contiguous-site infections increased. Although the overall proportion of BA patients among hospitalized children remained stable, the LOS and hospitalization costs increased significantly. This study, which included 799 children, is, to our knowledge, the largest epidemiological investigation of paediatric BA.
BA is a rare complication of paediatric bacterial meningitis, with an incidence rate of 1.4% [13]. Owing to the heavy burden of bacterial meningitis in China, it remained the most important predisposing factor for paediatric BA from 2006 to 2016 [3]. In this study, we observed a decrease in meningitis-related BA cases in the total dataset. This trend was broadly consistent with the decrease in bacterial meningitis recorded in the FUTURE database, from approximately 3,000 cases per year in 2016–2019 to 1,900 in 2020–2021 [13] and 1,600 in 2022–2023 (unpublished data). However, this trend became not statistically significant after the data from BCH were removed. One possible reason for this is that the BCH data accounted for more than 17% of the total cases. Removing the BCH data led to a significant reduction in sample size, which may have reduced statistical power. Additionally, differences in case composition between BCH and local hospitals could have contributed to this change. As a national paediatric medical centre, BCH receives complex and severe cases, often referred from multiple regions across China, while local hospitals primarily treat local children. These differences in case composition may have resulted in a different trend during 2020–2023, when the COVID-19 pandemic and referral patterns may have differed from those in earlier years.
In this study, we observed that the proportion of contiguous-site infections among children with BA tended to increase overall despite year–to–year variation (Fig. 3). One possible related factor is the changing epidemiology of respiratory viral infections, which may have been associated with changes in bacterial sinusitis and other local infections preceding BA. A study based on the same FUTURE database reported that the number of community viral pneumonia cases in China increased during 2016–2019 from 4,946 children in 2016 to 13,755 in 2019, followed by a slight decrease to 10,439 in 2020 and a significant increase to 18,435 in 2021 [18]. This pattern was broadly consistent with the temporal trend of BA associated with contiguous-site infections in this study. Moreover, recent studies on paediatric rhinosinusitis have shown an increased incidence of complications, particularly intracranial complications, during the COVID–19 pandemic period, and these changes were temporally associated with the incidence of viral infections [19, 20].
CHD was an important predisposing factor for paediatric BA. The percentage of children with CHD-related BA was 33.3% in Turkey [21] and 49.5% in Thailand [22]. However, it was only 7.6% in Italy, and early diagnosis and treatment of CHD were thought to explain the low percentage of children with CHD [23]. In this study, the proportion of children with CHD was stable. Given the decline in CHD incidence and CHD–related hospitalizations among children in China over the past decade, a longer observation period is needed to clarify the impact of these changes on the incidence of BA.
In this study, the overall proportion of children with BA among hospitalized children remained stable over time, whereas the LOS and hospitalization costs increased. Additional regression analyses revealed that neurosurgical intervention was associated with longer LOS and higher hospitalization costs and that complications were associated with higher hospitalization costs (Online resource 1 Supplementary Table S17–S20). Because both neurosurgical intervention and complications were more frequent during the years 2020–2023, the two factors may partly be associated with the increases in the LOS and hospitalization costs in 2020–2023. Additionally, LOS and hospitalization costs may have been influenced by clinical management, healthcare policy changes, resource utilization patterns, and economic factors. However, the present database did not include detailed information on these factors. Further studies with more detailed clinical and health-economic data are needed to clarify these findings.
There are several limitations inherent to a nationwide retrospective database study. First, the reliance on FSMRs data from multiple institutions may introduce inconsistencies, missing data, and coding errors. While the FUTURE database does not capture all clinical details for each BA patient, we utilized the detailed clinical information available from the MRS of BCH to validate and refine our case-identification strategy. By evaluating the PPVs and sensitivities of different search strategies, we selected the most appropriate strategy to identify children with BA in the FUTURE database, thereby strengthening the data quality. This method was also used to evaluate the Danish National Patient Registry [17], forming the basis for a series of population-based studies on BA in Denmark [4–7]. Second, in the validation analysis, we used only samples from BCH to assess the PPV and sensitivity in this study, which may not have captured the interhospital variability in coding practices. In this study, detailed clinical information and the numbers of confirmed BA cases at the other hospitals were unavailable, which were necessary for calculating PPV and sensitivity. To assess the impact of hospitals with high case contributions on the results, we performed additional sensitivity analyses. The results of sensitivity analyses showed consistent increases in the proportions of contiguous-site infections, neurosurgical interventions, complications as well as LOS, and hospitalization costs during the years 2020–2023 (Online Resource 1 Supplementary Tables S8–S16). Third, our cohort was restricted to children hospitalized in tertiary children's hospitals in China. This may result in missing severe ill children who died before hospitalization, as well as those with milder or misdiagnosed conditions treated in outpatient settings or in the paediatric departments of comprehensive tertiary hospitals. Fourth, as an administrative database study, FUTURE prioritizes demographic, diagnostic, and discharge information, which is valuable for epidemiological research but lacks detailed clinical details and microbiological data. Furthermore, this study could capture only in-hospital deaths recorded on the discharge face-sheet. Some children who left against medical advice may have died after discharge, and these deaths could not be captured in this study. Therefore, the reported in-hospital mortality may be lower than the overall mortality. Finally, heterogeneity in care across countries as large as China, as well as changing medical practices over the 8-year study period, may affect the interpretation of trends and limit the generalizability of the findings to other settings.
In conclusion, although the proportion of BA among hospitalized children in China has remained stable, the distribution of its predisposing factors has changed, and increases in the LOS and hospitalization costs have been observed. These changes highlight the need for greater clinical attention to contiguous-site infections in children with neurological symptoms.
Supplementary Information
Below is the link to the electronic supplementary material.
List of Abbreviations
- BA
Brain abscess
- CHD
Congenital heart disease
- CNS
Central nervous system
- COVID-19
Coronavirus disease 2019
- FSMRs
Face-sheets of discharge medical records
- FUTURE
FuTang Update Medical Records database
- ICD–10
International Classification of Diseases, 10th Revision
- IQR
Interquartile range
- LOS
Length of stay
- PPV
Positive predictive value
Authors' contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Zhenzhen Dou, Xinyu Wang, Xuekai Kang, Guoshuang Feng, Lingyun Ma, and Yutao Tang, Lingyun Guo, Bing Hu, Tianming Chen and Gang Liu. The first draft of the manuscript was written by Zhenzhen Dou and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Funding
This work is supported by Beijing Natural Science Foundation-Capital Medical University Joint Fund Project (L2510023), Capital's Funds for Health Improvement and Research [2024–1–2092], Training Plan for High level Public Health Technical Talents Construction Project [Discipline Leader–02–02], and Beijing Municipal Administration of Hospitals Incubating Program [PX2024042].
Data availability
No datasets were generated or analysed during the current study.
Declarations
Ethics approval
This study was approved by the Ethics Committee of Beijing Children’s Hospital (Approval Number: [2025]–E–124–R) and was conducted in accordance with the Declaration of Helsinki. The requirement for informed consent was waived because this was a retrospective database study using de-identified data. The ethics approval covered the use of de-identified multicenter discharge face-sheet data from the FUTURE database. The approval also covered the validation analysis using the electronic medical record system of Beijing Children’s Hospital. Direct personal identifiers were removed before statistical analysis.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Gilard V, Beccaria K, Hartley JC, Blanot S, Marqué S, Bourgeois M, Puget S, Thompson D, Zerah M, Tisdall M (2020) Brain abscess in children, a two-centre audit: outcomes and controversies. Arch Dis Child 105:288–291. 10.1136/archdischild-2018-316730 [DOI] [PubMed] [Google Scholar]
- 2.Bodilsen J, Dalager-Pedersen M, Nielsen H (2020) Long-term mortality and risk of epilepsy in children and young adults with brain abscess. Pediatr Infect Dis J 39:877–882. 10.1097/INF.0000000000002736 [DOI] [PubMed] [Google Scholar]
- 3.Dou ZZ, Guo LY, Liu LL, Li MH, Hu HL, Hu B, Guo X, Chen TM, Chen HY, Ge M et al (2021) Clinical characteristics and outcome analysis of 94 children with brain abscess in Beijing: a single-center retrospective study. Pediatr Infect Dis J 40:109–115. 10.1097/INF.0000000000002933 [DOI] [PubMed] [Google Scholar]
- 4.Bodilsen J, Dalager-Pedersen M, van de Beek D, Brouwer MC, Nielsen H (2020) Risk factors for brain abscess: a nationwide, population-based, nested case-control study. Clin Infect Dis 71:1040–1046. 10.1093/cid/ciz890 [DOI] [PubMed] [Google Scholar]
- 5.Bodilsen J, Dalager-Pedersen M, van de Beek D, Brouwer MC, Nielsen H (2020) Incidence and mortality of brain abscess in Denmark: a nationwide population-based study. Clin Microbiol Infect 26:95–100. 10.1016/j.cmi.2019.05.016 [DOI] [PubMed] [Google Scholar]
- 6.Bodilsen J, Dalager-Pedersen M, van de Beek D, Brouwer MC, Nielsen H (2020) Long-term mortality and epilepsy in patients after brain abscess: a nationwide population-based matched cohort study. Clin Infect Dis 71:2825–2832. 10.1093/cid/ciz1153 [DOI] [PubMed] [Google Scholar]
- 7.Bodilsen J, Duerlund LS, Mariager T, Brandt CT, Wiese L, Petersen PT, Larsen L, Hansen BR, Omland LH, Tetens MM et al (2023) Risk factors and prognosis of epilepsy following brain abscess: a nationwide population-based cohort study. Neurology 100:e1611–e1620. 10.1212/WNL.0000000000206866 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Yildiz N, Gayretli Aydin ZG (2025) Single-center experience and evaluation of rare intracranial abscesses in childhood. Ital J Pediatr 51:30. 10.1186/s13052-025-01895-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lu KH, Bruns N, Pentek C, Della Marina A, Gangfuß A, Kölbel H, Dogan B, Dziobaka J, Ahmadipour Y, Damman P et al (2024) Pediatric community-acquired brain abscesses: a single-center retrospective study. Pediatr Infect Dis J 43:e214–e217. 10.1097/INF.0000000000004308 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hoyer EA, Joseph M, Dunn J, Weiner HL, Dimachkieh A, Flores AR, Sanson MA, Ayele H, Hanson BM, Kaplan SL et al (2024) Increasing incidence of Streptococcus anginosus group intracranial infections associated with sinusitis, otitis media, and mastoiditis in children. Pediatr Infect Dis J 43:e261–e267. 10.1097/INF.0000000000004346 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Felsenstein S, Williams B, Shingadia D, Coxon L, Riordan A, Demetriades AK, Chandler CL, Bassi S, Koutoumanou E, Stapleton S et al (2013) Clinical and microbiologic features guiding treatment recommendations for brain abscesses in children. Pediatr Infect Dis J 32:129–135. 10.1097/INF.0b013e3182748d6e [DOI] [PubMed] [Google Scholar]
- 12.Chen Y, Zhao X, Ye C, Zhou J, Wang J, Ye X (2024) Epidemiology and viral loads of respiratory syncytial virus in hospitalized children prior to and during COVID-19 pandemic in Hangzhou, China. J Med Virol 96:e29855. 10.1002/jmv.29855 [DOI] [PubMed] [Google Scholar]
- 13.Liu L, Wang X, Guo L, Chen T, Hu B, Jian B, Feng G, Zeng Y, Liu G (2025) Epidemiological characteristics and disease burden of bacterial meningitis in hospitalized children in China: a 6-year nationwide retrospective study. BMC Infect Dis 25:592. 10.1186/s12879-025-10928-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Su Z, Zhang Y, Cai X, Li Q, Gu H, Luan Y, He Y, Li S, Chen J, Zhang H (2023) Improving long-term care and outcomes of congenital heart disease: fulfilling the promise of a healthy life. Lancet Child Adolesc Health 7:502–518. 10.1016/S2352-4642(23)00053-6 [DOI] [PubMed] [Google Scholar]
- 15.Zhang Y, Wang J, Zhao J, Huang G, Liu K, Pan W, Sun L, Li J, Xu W, He C et al (2023) Current status and challenges in prenatal and neonatal screening, diagnosis, and management of congenital heart disease in China. Lancet Child Adolesc Health 7:479–489. 10.1016/S2352-4642(23)00051-2 [DOI] [PubMed] [Google Scholar]
- 16.Wang X, Zeng Y, Tian J, Xu H, Song F, Guo Y, Xu X, Feng G, Ni X (2021) A brief introduction to the FUTang Updating Medical Records (FUTURE) database. Pediatr Investig 5:247–248. 10.1002/ped4.12297 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Bodilsen J, Dalager-Pedersen M, Kjærgaard N, van de Beek D, Brouwer MC, Nielsen H (2018) Positive predictive value of ICD-10 diagnosis codes for brain abscess in the Danish National Patient Registry. Clin Epidemiol 10:1503–1508. 10.2147/CLEP.S176072 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Feng Q, Wang J, Wang X, Tian J, Zhang L, Dilmurat D, Liu M, Ai J, Feng G, Zeng Y et al (2025) Clinical epidemiological characteristics of hospitalized pediatric viral community-acquired pneumonia in China. J Infect 90:106450. 10.1016/j.jinf.2025.106450 [DOI] [PubMed] [Google Scholar]
- 19.Bhat AN, Wang J, Yang A, Molter D, Dunsky KA, Menezes M, Lieu JEC (2025) Trends in complications of pediatric rhinosinusitis before and during the COVID-19 era. Otolaryngol Head Neck Surg 172:2090–2097. 10.1002/ohn.1196 [DOI] [PubMed] [Google Scholar]
- 20.Kalavacherla S, Hall M, Jiang W, Carvalho D (2024) Temporal trends in pediatric acute sinusitis surrounding the COVID-19 pandemic. Otolaryngol Head Neck Surg 170:913–918. 10.1002/ohn.578 [DOI] [PubMed] [Google Scholar]
- 21.Ozsürekci Y, Kara A, Cengiz AB, Celik M, Ozkaya-Parlakay A, Karadağ-Oncel E, Ceyhan M (2012) Brain abscess in childhood: a 28-year experience. Turk J Pediatr 54:144–149 [PubMed] [Google Scholar]
- 22.Auvichayapat N, Auvichayapat P, Aungwarawong S (2007) Brain abscess in infants and children: a retrospective study of 107 patients in northeast Thailand. J Med Assoc Thai 90:1601–1607 [PubMed] [Google Scholar]
- 23.Raffaldi I, Garazzino S, CastelliGattinara G, Lipreri R, Lancella L, Esposito S, Giannini AM, Montagnani C, Marseglia GL, Pignata C et al (2017) Brain abscesses in children: an Italian multicentre study. Epidemiol Infect 145:2848–2855. 10.1017/S0950268817001583 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
No datasets were generated or analysed during the current study.