Serial cerebrospinal fluid concentrations of high mobility group box 1 in bacterial meningitis: a retrospective cohort study | BMC Infectious Diseases

BMC Infectious Diseases volume 25, Article number: 107 (2025 ) Cite this article

Bacterial meningitis (BM) is a life-threatening central nervous system infection with potential for severe neurological sequelae. High mobility group box 1 (HMGB1) is known as a late inflammatory mediator associated with lethal pathology. This study aims to investigate the serial cerebrospinal fluid (CSF) concentrations of HMGB1 in children with BM and its relationship to neurological prognosis. Flexible Operation

Serial cerebrospinal fluid concentrations of high mobility group box 1 in bacterial meningitis: a retrospective cohort study | BMC Infectious Diseases | Full Text

This retrospective cohort study included children with BM, aseptic meningitis (AM), and controls. CSF samples were collected serially from patients with BM and once from those with AM and controls. HMGB1 and interleukin-6 (IL-6) concentrations were measured using ELISA and bead-based multiplex assays, respectively. Statistical analyses included Mann–Whitney U tests, Kruskal–Wallis tests, and three-way ANOVA to evaluate differences among groups and over time.

HMGB1 levels in the CSF of children with BM were significantly higher than in those with AM and controls (p < 0.001). Inflammatory cytokine IL-6 levels decreased after treatment; however, HMGB1 levels remained elevated in half of the BM patients. Notably, a patient with neurological sequelae exhibited a delayed elevation of HMGB1 until the latest time points. Three-way ANOVA revealed significant differences in the time course of IL-6 and HMGB1 among individuals (p = 0.018).

Elevated CSF HMGB1 levels persist in some children with BM even after treatment, particularly in those with poor neurological outcomes. These findings suggest that delayed elevation of HMGB1 may contribute to severe inflammation and poor prognosis in BM. Further research into HMGB1 as a potential therapeutic target in BM is warranted.

Bacterial meningitis (BM) is a serious and life-threatening infectious disease of central nervous system. Symptoms of BM include fever, convulsion, altered consciousness, headache, and vomiting. It is indispensable to diagnose and start the empiric antibiotic therapy as soon as possible. It has severe neurological sequelae in 12 to 17% of survivors, and milder impairment of neurological function occurs in 20 to 32% [1,2,3]. BM is primarily caused by bacterial invasion to central nervous system through the blood brain barrier. High levels of inflammatory response, so called cytokine storm, play an important role in the pathogenesis of BM, and adjunctive treatment of dexamethasone improves neurological outcomes [4]. However, its use is not universally accepted as a standard of care, as its effects on mortality remain uncertain [4].

High mobility group box 1 (HMGB1) is a main component of non-histone DNA-binding protein and is widely distributed in all mammalian tissues [5]. Previously, HMGB1 was known to be involved in maintaining nucleosome structure and regulation of gene transcription [6, 7]. Recently, HMGB1 has been identified as a late-phase inflammatory mediator in sepsis, persisting during prolonged inflammation and contributing to poor outcomes [8,9,10,11,12], while cytokines such as IL-6 act as early-phase mediators and play a key role in the cytokine storm associated with BM. Elevated serum HMGB1 concentrations have been reported in septic shock, acute lung injury, disseminated intravascular coagulation, and surgical operation [10, 13]. Notably, HMGB1 levels were found to increase after IL-6 levels in sepsis [11], and anti-HMGB1 antibody have been shown to reduce mortality, suggesting HMGB1 as a potential therapeutic target [14,15,16].

We determined cerebrospinal fluid (CSF) HMGB1 concentrations and kinetics to evaluate the pathogenesis of BM and compared CSF HMGB1 and IL-6 levels among BM, aseptic meningitis (AM), and controls. By analyzing both IL-6 and HMGB1 serially, we aimed to better understand their temporal relationship and potential contributions to disease progression and neurological outcomes in BM.

Informed consent was obtained from the parents of the patients and controls enrolled in this study. This study was approved by the Institutional Review Board at Yamaguchi University Hospital (H26-9).

This study included two analysis groups: the Baseline Cohort Group and the Longitudinal Cohort Group, with the Longitudinal Cohort Group representing a subset of the Baseline Cohort Group (Fig. 1). Each group was defined by specific inclusion and exclusion criteria to address different study objectives.

Flowchart of the selection of patients with bacterial meningitis (BM)

To compare HMGB1 and cytokine levels among BM, AM, and control groups, and within the BM group by bacterial species, neurological outcomes, and correlations between biomarkers.

Admission to Yamaguchi University Hospital between 1984 and 2009.

Insufficient clinical information or incomplete samples.

Initially, 36 children aged 0–18 years with BM were identified. After excluding 10 patients due to insufficient clinical information or incomplete samples, 26 patients (median age 0.8 years [interquartile range (IQR): 0.3–1.2], range 0.1–3.5 years) were included for analysis. CSF cultures from 26 patients with BM yielded Haemophilus influenzae (H. influenzae, n = 12), Streptococcus pneumoniae (S. pneumoniae, n = 6), Escherichia coli (E. coli, n = 4), Group B Streptococcus (GBS, n = 1), Listeria monocytogenes (n = 1), Proteus mirabilis (P. mirabilis, n = 1), and Methicillin-Resistant Staphylococcus epidermidis (n = 1). The patients were treated with empiric antibiotics in adequate doses according to ages and basic diseases. Adjunctive dexamethasone therapy was administered according to national clinical guidelines in Japan (0.6 mg/kg/day in four intravenous doses for two or four days, started before antibiotics) [17]. One patient died and 5 patients had neurological sequelae (2 epilepsy, 3 intellectual disability, 3 sensorineural hearing impairment, and 1 hydrocephalus, which were overlapped).

To evaluate longitudinal changes in HMGB1, cytokines, fever, and CRP levels, and to investigate their relationship with treatment and outcomes.

Availability of a sample collected on day 1–2.

At least two additional samples collected within three predefined periods (day 3–6, day 7–10, and day 11–16).

Lack of sufficient serial samples meeting the above criteria.

From the 26 patients in the Analysis Cohort Group, 16 were excluded due to insufficient serial samples. The remaining 10 patients (median age 0.9 years [IQR: 0.4–1.1], range 0.2–3.2 years) were included in the Longitudinal Cohort Group.

The medical records were retrospectively reviewed to investigate the clinical and laboratory findings, treatment, and prognosis (Table 2), and serial changes in fever and C-reactive protein (CRP) up to day 16 (Supplementary Fig. 1). The day of onset of fever was considered as the first day of illness, as it was the most consistently documented symptom in retrospective medical record reviews and closely aligned with the initiation of treatment. Samples were stored at −80˚C until assay.

Thirteen CSF samples were obtained on the day of admission from 13 children with AM (6 males and 7 females, 3.4 years as median age [IQR 0.5–4.4], range 0.1–7.0 years) (Table 1). The causative agents were enterovirus (n = 3), mumps virus (n = 1), Mycoplasma pneumoniae (n = 1), and unknown (n = 8). They all had fully recovered.

The control subjects were 12 afebrile and noninfectious children (6 males and 6 females, 1.4 years as median age [IQR 1.7–4.6], range, 0.1–5.1 years) with neuromuscular or psychological disorders, such as epilepsy, global developmental delay, and so on. CSF samples were obtained from them as routine analysis to diagnose their diseases and they all had normal CSF cell counts, protein, and glucose.

The CSF concentrations of HMGB1 were measured with HMGB1 ELISA Kit II (Shino-test corporation, Tokyo, Japan). An anti-HMGB1 monoclonal coating antibody was adsorbed onto polystyrene microwells. HMGB1 present in the samples or the standard bound to the adsorbed antibodies, and the HMGB1/antibody complex was detected with an alkaline phosphatase-conjugated secondary antibody. The amount of captured HMGB1 was measured by determining the color produced by reagents using iMark Microplate Absorbance Reader (BIO RAD, Hercules, CA, USA). The limit of detection (LOD) was 1.0 ng/mL.

The CSF concentrations of interleukin (IL)−2, IL-4, IL-6, IL-10, tumor necrosis factor (TNF), and interferon-gamma (IFN-γ) were measured with BD Cytometric Bead Array Human Th1/Th2 Cytokine Kit II (BD Biosciences, San Jose, CA, USA). The LODs of IL-2, IL-4, IL-6, IL-10, TNF, and IFN-γ were 2.6, 2.6, 3.0, 2.8, 2.8, and 7.1 pg/mL, respectively.

Differences between two groups for continuous variables were compared using Mann–Whitney U test. Three or more group comparisons were performed using Kruskal–Wallis test followed by Bonferroni-corrected Mann–Whitney U test. Categorical variables were compared using a Chi square test. Correlations among parameters were calculated applying Spearman correlation coefficient method. To identify differences associated with time course (day 1–2, 3–6, 7–10, and 11–16), individual (ten patients with BM), and parameters (HMGB1 and IL-6), we applied three-way ANOVA. In preparation for the analysis, values of HMGB1 and IL-6 were logarithmically transformed to correct their skewness, and then they were standardized by use of mean and SD calculated under the transformed scale to make them in uniform scale. If there was no sample in one of the four periods, it was supplemented with the mean value of the samples in the other three periods: i.e., the mean imputation method [18]. P-values less than 0.05 were taken to be significant. Statistical analyses were performed using StatFlex version 6.0 (Artech Co., Osaka, Japan) and Prism version 8.0 (GraphPad Software Inc., La Jolla, CA, USA).

The comparison of HMGB1 concentrations among three groups in Table 1 was shown in Fig. 2. There was no significant difference in sex among the three groups, but the ages in the AM group were significantly higher than those in the BM group. The data of BM and AM patients were taken by the first procedures of lumbar puncture, which were mainly on admission. The HMGB1 levels in the BM group ranged from < LOD to 97.7 ng/mL, with a median of 9.6 ng/mL (IQR: 2.4–32.2), and were significantly higher than those in the AM group (< LOD to 18.9 ng/mL, median 1.4 ng/mL, IQR: < LOD–4.4) and the control group (< LOD to 1.1 ng/mL, median < LOD, IQR: < LOD– < LOD) (p = 0.038 and p < 0.001, respectively). The HMGB1 levels of AM were significantly higher than those of controls (p = 0.0099). There were no significant differences in HMGB1 levels in the first sample between patients with and without neurological sequelae in the BM group (Supplementary Fig. 2). Similarly, no significant differences were observed when comparing HMGB1 levels among BM cases caused by H. influenzae, S. pneumoniae, E. coli, and other pathogens (Supplementary Fig. 2). The patients with BM caused by S. pneumoniae and GBS had neurological sequelae (4 and 1 cases, respectively), and the one by P. mirabilis died.

The cerebrospinal fluid (CSF) high mobility group box 1 (HMGB1) concentrations of patients with bacterial meningitis in the baseline cohort group, aseptic meningitis, and control subjects. All samples were obtained on admission. Horizontal lines indicate median values. Statistics of multiple comparisons were conducted using Bonferroni-corrected Mann–Whitney U test

The CSF concentrations of IL-6, IL-10, TNF, IL-2, IL-4, and IFN-γ in patients with BM were 14,255.9 pg/mL as median (IQR: 7,894.7–19,978.8; range: 205.0–577,037.0), 269.9 (IQR: 18.3–704.0; range: < LOD–2,600.7), 87.9 (IQR: 5.2–251.7; range: < LOD–18,964.0), < LOD (IQR: < LOD– < LOD; range: < LOD–201.5), < LOD (IQR: < LOD– < LOD; range: < LOD–145.4), and 17.9 (IQR: < LOD–73.6; range: < LOD–1,899.8), respectively (Supplementary Table 1). CSF IL-2, IL-4, and FN-γ levels in 70%, 78%, and 39% of patients with BM were under LOD, respectively. There were significant correlations between IL-6 and IL-10, between IL-6 and TNF, between IL-10 and TNF, and between TNF and IFN-γ (r = 0.53, 0.50, 0.58, 0.42; p = 0.0058, 0.0096, 0.0019, 0.046, respectively) (Fig. 3). There were no significant correlations between HMGB1 and cytokines (IL-6, IL-10, TNF, and IFN-γ) (Supplementary Fig. 3). There were no significant differences in cytokine levels on first sample between with and without neurological sequelae (Supplementary Fig. 2).

The correlations between inflammatory cytokines in 26 patients with BM in the baseline cohort group. a interleukin (IL)−6 and IL-10, b IL-6 and tumor necrosis factor (TNF), c IL-6 and interferon (IFN)-γ, d IL-10 and TNF, e IL-10 and IFN-γ, f TNF and IFN-γ. All samples were obtained on admission. The correlation line represents the linear regression fit. Statistics were conducted using Spearman correlation coefficient method

A total of 30 CSF samples were obtained from 10 patients in Table 2 as a routine examination on the 1st to 16th day of diseases. Numbers of sampling from 10 patients during 1–2, 3–6, 7–10, and 11–16 were 10, 8, 6, and 6, respectively (Fig. 1). The serial concentrations of HMGB1 and IL-6 were shown in Fig. 4. The peak concentrations of IL-6 were observed at day 1–2 in all patients, which once dropped, and then second peaks were observed in 5 patients. On the other hand, the peak concentrations of HMGB1 were after day 3 in 5 patients. HMGB1 levels in Patient 7 showed a continuous upward trend across the three sampling points, peaking in the final period. A three-way ANOVA revealed significant effects for time course (p < 0.001), between-individual differences (p = 0.0058), and the interaction between time course and parameters (p = 0.018), indicating that the patterns of time-dependent changes in HMGB1 and IL-6 were significantly different across patients. The patient 7 had neurological sequelae of intellectual disability, hearing impairment, and epilepsy, and the patient 6 had hearing impairment. The other patients had no neurological sequelae. The clinical data of sequential transitions of fever and CRP were obtained from these patients (Supplementary Fig. 1). Six patients (60%) had fever again once after they fell during day 3–5, and they all received dexamethasone treatment. The rise of CRP for the second time was observed in 6 patients (60%), and the days in which CRP began to go up again were from 5 to 10.

Previously, HMGB1 levels in human CSF have been reported in several diseases, including meningitis [19, 20], encephalopathy [21], traumatic brain injury [22], subarachnoid hemorrhage [23], and neuromyelitis optica [24, 25]. And previous reports of HMGB1 in human meningitis have been considered only for specimens with a relatively small sample size and at early time points of the disease. Here, we provide the first report of serial CSF HMGB1 levels in human BM.

Our data show that CSF HMGB1 levels were elevated in most patients with BM. However, HMGB1 levels at admission were not consistently higher in patients with neurological sequelae or fatal outcomes. Three-way ANOVA performed by using the uniform scale transformed data revealed that the pattern of time-dependent changes in IL-6 and HMGB1 levels was significantly different. Additionally, no significant correlations were observed between HMGB1 and cytokine levels, which may reflect HMGB1's role as a late-phase inflammatory mediator [11]. Unfortunately, due to the limited availability of serial CSF samples, including the fatal case, only two cases with neurological sequelae were available for the time-course analysis, preventing a full evaluation of HMGB1 dynamics in the most severely affected patients.

In the BM, antibiotics cause rapid lysis of bacteria and enhanced release of endotoxin and exotoxin into the CSF, which can stimulate astrocytes, microglia, and cerebral capillary endothelia, to produce cytokines such as TNF-α, IL-1, IL-6, and IL-8 [26]. The cytokines make attractant of leukocytes to the CSF through activation of adhesion molecules, and leukocytes release proteolytic products and toxic oxygen radicals [26]. A complex network of cytokines, chemokines, proteolytic enzymes, excitatory aminoacids and oxidants take part in the inflammatory cascade that leads to brain edema, ischemia, and neuronal injury [26]. Intracerebroventricular administration of HMGB1 increases brain TNF, IL-1, and IL-6 expression and induces fever, anorexia, and weight loss in mice [27, 28]. Thus, elevated HMGB1 concentrations in CSF of patients might be involved in inflammatory mechanisms of BM.

The fact that CSF HMGB1 levels in the majority of the patients with BM have increased for the first several days is not inconsistent with previous reports. HMGB1 is thought to act as a late mediator because the kinetics of its release is delayed compared with most other cytokines such as IL-6 in sepsis [11]. IL-6 is a cytokine to play an important role as a primary mediator in the inflammatory responses [29, 30]. In our data, inflammatory (IL-6, TNF) and anti-inflammatory (IL-10) cytokines had positive correlations among the first samples. And IL-6 decreased promptly after treatment. Although IL-6 levels are often highly elevated in patients with BM, some studies suggest that TNF-α, rather than IL-6, may be more closely associated with neurological sequelae [29]. Similarly, serum HMGB1 levels in sepsis patients have been reported to be higher in non-survivors compared to survivors [8], suggesting that HMGB1 could be associated with severe disease progression. In our study, CSF HMGB1 levels showed a delayed increase after day 7 only in Patient 7, who developed severe neurological sequelae. Although this finding is based on a single case, it may reflect HMGB1’s role as a late-phase inflammatory mediator.

The release of HMGB1 occurs in two different ways, active secretion from living inflammatory cells and passive release from necrotic cells [32, 33]. The elevated CSF HMGB1 concentrations in patients with BM may originate in both ways because the pathogenesis of BM includes strong inflammation and neuronal cell death [34]. As shown in Fig. 4a, CSF IL-6 concentrations peaked at day 1–2 in all patients, with a secondary peak in half of the patients. Similarly, fever and CRP levels also rose again in approximately half of patients (Supplementary Fig. 1), suggesting persistent inflammation. This secondary increase could result from renewed active secretion by living inflammatory cells after steroid discontinuation. However, based on the CSF HMGB1 dynamics observed during the first several days, we considered that steroid therapy alone might not be sufficient to suppress HMGB1 levels, as it primarily targets inflammatory responses and may not prevent HMGB1 release from necrotic cells. This persistent release could maintain or exacerbate inflammation even after steroid treatment. Passive immunization with anti-HMGB1 antibodies significantly protects against lethal endotoxemia in mice, even when treatment was delayed 2 h after LPS exposure [8, 9]. The microRNA, miR-141-3p, suppressed inflammatory response through the downregulation of HMGB1 in astrocytes of BM rat model [35]. These reports suggested that HMGB1 was expected to be potential therapeutic targets [15, 16]. There might be a room for improve neurological outcome to use an adjunctive anti-HMGB1 therapy.

The present study has several limitations. First, the sample size was too small to prove the correlation between delayed elevation of HMGB1 and poor prognosis statistically. Due to the success of vaccination programs, the number of BM cases caused by vaccine-preventable pathogens has decreased, limiting the size of our study population. Second, we could not obtain consecutive samples from all the BM patients because of the retrospective study, ethical issues, and a various clinical need for collection of CSF samples. A further large-scale prospective study is necessary to clarify them. Third, the control group included patients with epilepsy and other conditions due to the limited availability of CSF samples from healthy children, which may affect the consistency of the control data. However, most HMGB1 levels in the control group were below the detection limit, minimizing potential bias. Finally, this study does not directly prove a causal relationship of elevated HMGB1 for neurological sequelae in human, not just the result of tissue damage. This problem can be indirectly complemented by the in vivo experiments mentioned above.

In conclusion, CSF HMGB1 levels are elevated in most patients with BM. Although delayed HMGB1 elevation was observed in only one patient in our study, this finding suggests a potential association with poor prognosis in BM, warranting further investigation in larger cohorts.

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Baraff LJ, Lee SI, Schriger DL. Outcomes of bacterial meningitis in children: a meta-analysis. Pediatr Infect Dis J. 1993;12(5):389–94.

Oostenbrink R, Maas M, Moons KG, Moll HA. Sequelae after bacterial meningitis in childhood. Scand J Infect Dis. 2002;34(5):379–82.

Koomen I, Grobbee DE, Roord JJ, Jennekens-Schinkel A, van der Lei HD, Kraak MA, et al. Prediction of academic and behavioral limitations in school-age survivors of bacterial meningitis. Acta Paediatr. 2004;93(10):1378–85.

Brouwer MC, McIntyre P, Prasad K, van de Beek D. Corticosteroids for acute bacterial meningitis. Cochrane Database Syst Rev. 2015(9):CD004405.

Bustin M, Lehn DA, Landsman D. Structural features of the HMG chromosomal proteins and their genes. Biochim Biophys Acta. 1990;1049(3):231–43.

Goodwin GH, Sanders C, Johns EW. A new group of chromatin-associated proteins with a high content of acidic and basic amino acids. Eur J Biochem. 1973;38(1):14–9.

Landsman D, Bustin M. A signature for the HMG-1 box DNA-binding proteins. BioEssays. 1993;15(8):539–46.

Wang H, Bloom O, Zhang M, Vishnubhakat JM, Ombrellino M, Che J, et al. HMG-1 as a late mediator of endotoxin lethality in mice. Science. 1999;285(5425):248–51.

Yang H, Ochani M, Li J, Qiang X, Tanovic M, Harris HE, et al. Reversing established sepsis with antagonists of endogenous high-mobility group box 1. Proc Natl Acad Sci U S A. 2004;101(1):296–301.

Ueno H, Matsuda T, Hashimoto S, Amaya F, Kitamura Y, Tanaka M, et al. Contributions of high mobility group box protein in experimental and clinical acute lung injury. Am J Respir Crit Care Med. 2004;170(12):1310–6.

Wang H, Yang H, Tracey KJ. Extracellular role of HMGB1 in inflammation and sepsis. J Intern Med. 2004;255(3):320–31.

Yang H, Wang H, Czura CJ, Tracey KJ. The cytokine activity of HMGB1. J Leukoc Biol. 2005;78(1):1–8.

Abraham E, Arcaroli J, Carmody A, Wang H, Tracey KJ. HMG-1 as a mediator of acute lung inflammation. J Immunol. 2000;165(6):2950–4.

Tracey KJ, Fong Y, Hesse DG, Manogue KR, Lee AT, Kuo GC, et al. Anti-cachectin/TNF monoclonal antibodies prevent septic shock during lethal bacteraemia. Nature. 1987;330(6149):662–4.

Andersson U, Tracey KJ. HMGB1 is a therapeutic target for sterile inflammation and infection. Annu Rev Immunol. 2011;29:139–62.

Article PubMed PubMed Central Google Scholar

Mantell LL, Parrish WR, Ulloa L. Hmgb-1 as a therapeutic target for infectious and inflammatory disorders. Shock. 2006;25(1):4–11.

Japanese Society of Neurological Therapeutics JSoN, Japanese Society for Neuroinfectious Diseases (Supervision); Committee for the Development of Guidelines for the Management of Bacterial Meningitis (Ed.). Saikinsei zuimakuen no shinryo guideline (Guidelines for the Management of Bacterial Meningitis). Tokyo: Igaku-Shoin; 2007. 60–1 p.

Austin PC, White IR, Lee DS, van Buuren S. Missing Data in Clinical Research: A Tutorial on Multiple Imputation. Can J Cardiol. 2021;37(9):1322–31.

Tang D, Kang R, Cao L, Zhang G, Yu Y, Xiao W, et al. A pilot study to detect high mobility group box 1 and heat shock protein 72 in cerebrospinal fluid of pediatric patients with meningitis. Crit Care Med. 2008;36(1):291–5.

Asano T, Ichiki K, Koizumi S, Kaizu K, Hatori T, Mashiko K, et al. High mobility group box 1 in cerebrospinal fluid from several neurological diseases at early time points. Int J Neurosci. 2011;121(8):480–4.

Momonaka H, Hasegawa S, Matsushige T, Inoue H, Kajimoto M, Okada S, et al. High mobility group box 1 in patients with 2009 pandemic H1N1 influenza-associated encephalopathy. Brain Dev. 2014;36(6):484–8.

Au AK, Aneja RK, Bell MJ, Bayir H, Feldman K, Adelson PD, et al. Cerebrospinal fluid levels of high-mobility group box 1 and cytochrome C predict outcome after pediatric traumatic brain injury. J Neurotrauma. 2012;29(11):2013–21.

Article PubMed PubMed Central Google Scholar

Nakahara T, Tsuruta R, Kaneko T, Yamashita S, Fujita M, Kasaoka S, et al. High-mobility group box 1 protein in CSF of patients with subarachnoid hemorrhage. Neurocrit Care. 2009;11(3):362–8.

Uzawa A, Mori M, Masuda S, Muto M, Kuwabara S. CSF high-mobility group box 1 is associated with intrathecal inflammation and astrocytic damage in neuromyelitis optica. J Neurol Neurosurg Psychiatry. 2013;84(5):517–22.

Wang H, Wang K, Wang C, Xu F, Zhong X, Qiu W, et al. Cerebrospinal fluid high-mobility group box protein 1 in neuromyelitis optica and multiple sclerosis. NeuroImmunoModulation. 2013;20(2):113–8.

Saez-Llorens X, McCracken GH Jr. Bacterial meningitis in children. Lancet. 2003;361(9375):2139–48.

Agnello D, Wang H, Yang H, Tracey KJ, Ghezzi P. HMGB-1, a DNA-binding protein with cytokine activity, induces brain TNF and IL-6 production, and mediates anorexia and taste aversion. Cytokine. 2002;18(4):231–6.

O’Connor KA, Hansen MK, Rachal Pugh C, Deak MM, Biedenkapp JC, Milligan ED, et al. Further characterization of high mobility group box 1 (HMGB1) as a proinflammatory cytokine: central nervous system effects. Cytokine. 2003;24(6):254–65.

Heinrich PC, Castell JV, Andus T. Interleukin-6 and the acute phase response. Biochem J. 1990;265(3):621–36.

Article PubMed PubMed Central Google Scholar

Akira S, Taga T, Kishimoto T. Interleukin-6 in biology and medicine. Adv Immunol. 1993;54:1–78.

Ichiyama T, Hayashi T, Nishikawa M, Furukawa S. Levels of transforming growth factor beta 1, tumor necrosis factor alpha, and interleukin 6 in cerebrospinal fluid: association with clinical outcome for children with bacterial meningitis. Clin Infect Dis. 1997;25(2):328–9.

Gardella S, Andrei C, Ferrera D, Lotti LV, Torrisi MR, Bianchi ME, et al. The nuclear protein HMGB1 is secreted by monocytes via a non-classical, vesicle-mediated secretory pathway. EMBO Rep. 2002;3(10):995–1001.

Article PubMed PubMed Central Google Scholar

Scaffidi P, Misteli T, Bianchi ME. Release of chromatin protein HMGB1 by necrotic cells triggers inflammation. Nature. 2002;418(6894):191–5.

Gerber J, Nau R. Mechanisms of injury in bacterial meningitis. Curr Opin Neurol. 2010;23(3):312–8.

Fang X, Wang H, Zhuo Z, Tian P, Chen Z, Wang Y, et al. miR-141-3p inhibits the activation of astrocytes and the release of inflammatory cytokines in bacterial meningitis through down-regulating HMGB1. Brain Res. 2021;1770: 147611.

We thank the patients and their parents for their participation in this study. We also thank Midori Wakabayashi and Takako Waniishi for technical support with the measurements.

This work was supported by grants from the Morinaga Foundation for Health and Nutrition.

Department of Pediatrics, Yamaguchi University Graduate School of Medicine, 1-1-1 Minamikogushi, Ube, Yamaguchi, 755-8505, Japan

Takeshi Matsushige, Hirofumi Inoue, Madoka Hoshide, Fumitaka Kohno, Hikaru Kobayashi & Shunji Hasegawa

Department of Clinical Laboratory Sciences, Faculty of Health Sciences, Yamaguchi University Graduate School of Medicine, 1-1-1 Minamikogushi, Ube, Yamaguchi, 755-8505, Japan

Division of Pediatrics, Tsudumigaura Medical Center for Children With Disabilities, 752-4 Kume, Shunan, Yamaguchi, 745-0801, Japan

You can also search for this author in PubMed Google Scholar

TM conceived and designed the study. HI, MH, FK, and HK were involved in the clinical care of the patients and collected the data. KI performed the data analysis and contributed to the interpretation of the results. TM drafted the manuscript and performed revisions. TI provided critical advice and assisted in revising the manuscript. SH supervised the overall study, provided guidance throughout the process, and contributed to the final revision of the manuscript. All authors read and approved the final manuscript.

This study was approved by the Institutional Review Board at Yamaguchi University Hospital (H26-9). Informed consent was obtained from the parents of all patients and controls included in this study.

Not applicable. This manuscript does not contain any individual person’s data in any form.

The authors declare no competing interests.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Matsushige, T., Inoue, H., Hoshide, M. et al. Serial cerebrospinal fluid concentrations of high mobility group box 1 in bacterial meningitis: a retrospective cohort study. BMC Infect Dis 25, 107 (2025). https://doi.org/10.1186/s12879-025-10476-7

DOI: https://doi.org/10.1186/s12879-025-10476-7

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By using this website, you agree to our Terms and Conditions, Your US state privacy rights, Privacy statement and Cookies policy. Your privacy choices/Manage cookies we use in the preference centre.

Serial cerebrospinal fluid concentrations of high mobility group box 1 in bacterial meningitis: a retrospective cohort study | BMC Infectious Diseases | Full Text