Publication

• Title: Dexamethasone in adults with bacterial meningitis
• Year: 2002
• Journal published in: New England Journal of Medicine
• Citation: de Gans J, van de Beek D. Dexamethasone in adults with bacterial meningitis. N Engl J Med. 2002 Nov 14;347(20):1549-1556.

Context & Rationale

• Background

  • Adult community-acquired bacterial meningitis carries substantial risks of death and long-term neurological disability despite appropriate antimicrobial therapy.
  • Inflammation triggered by bacterial components and antibiotic-mediated lysis contributes to blood–brain barrier breakdown, cerebral oedema, vasculopathy, and neuronal injury.
  • Pre-2002 evidence for adjunctive corticosteroids was stronger in paediatric disease (particularly Haemophilus influenzae type b) than in adults, and adult data were limited and inconsistent.
  • There was a need for a rigorously conducted, adequately powered, double-blind adult trial using early corticosteroid administration around the first antibiotic dose.
• Research Question/Hypothesis

  • In adults with suspected acute bacterial meningitis, does adjunctive dexamethasone started before (or with) initial antibiotic therapy improve functional outcome and survival compared with placebo?
• Why This Matters

  • Dexamethasone is inexpensive, widely available, and can be administered immediately alongside empiric antibiotics, making even modest absolute benefits clinically and systemically important.
  • A convincing benefit in adults—particularly pneumococcal meningitis—would justify routine early use and influence guideline recommendations internationally.

Design & Methods

• Research Question: Among adults with suspected acute bacterial meningitis, does adjunctive dexamethasone (given before/with antibiotics) reduce unfavourable functional outcome and mortality compared with placebo?
• Study Type: Randomised, double-blind, placebo-controlled, multicentre, investigator-initiated trial conducted across 52 hospitals in five European countries (Netherlands, Belgium, Germany, Denmark, Austria), June 1993 to December 2001.
• Population:
  • Inclusion: Adults >17 years with suspected meningitis and at least one CSF criterion: cloudy CSF; bacteria on Gram stain; or CSF leucocyte count >1000/mm³.
  • Key exclusions: Hypersensitivity to β-lactams or corticosteroids; pregnancy; CSF shunt; antibiotic therapy within the preceding 48 hours; suspected tuberculosis or fungal infection; recent head trauma/neurosurgery; peptic ulcer disease; participation in another trial.
• Intervention:
  • Dexamethasone sodium phosphate 10 mg intravenously every 6 hours for 4 days.
  • Timing: initially 15–20 minutes before the first antibiotic dose; after interim analysis, permitted to be given with the first antibiotic dose.
• Comparison:
  • Matching placebo administered intravenously every 6 hours for 4 days, with the same timing rules.
  • Antimicrobial therapy was provided to all patients; initial protocol specified IV amoxicillin 2 g every 4 hours for 7–10 days with modifications based on Gram stain and local practice, and the protocol was later amended to allow use of each hospital’s usual empiric regimen.
• Blinding: Double-blind (patients, treating teams, and outcome assessors); emergency unblinding via sealed envelopes occurred twice; allocation code was broken after the final patient completed follow-up.
• Statistics: A total of 300 patients (150 per group) were required to detect an absolute reduction in unfavourable outcome from 40% to 25% with 80% power at a two-sided α=0.05; primary analysis was intention-to-treat with relative risks and 95% confidence intervals; missing 8-week outcomes were imputed using last observation carried forward.
• Follow-Up Period: 8 weeks (primary functional outcome), with in-hospital clinical course and adverse events additionally reported.

Key Results

This trial was not stopped early. Recruitment was temporarily paused after an interim analysis due to slow enrolment and later restarted with protocol amendments; the planned sample size was achieved (301 randomised).

Outcome	Dexamethasone	Placebo	Effect	p value / 95% CI	Notes
Unfavourable outcome at 8 weeks	23/157 (15%)	36/144 (25%)	RR 0.59	95% CI 0.37 to 0.94; P=0.03	Primary outcome: Glasgow Outcome Scale 1–4 (death to moderate disability)
Death at 8 weeks	11/157 (7%)	21/144 (15%)	RR 0.48	95% CI 0.24 to 0.96; P=0.04	Glasgow Outcome Scale score 1
Unfavourable outcome at 8 weeks (S. pneumoniae)	15/58 (26%)	26/50 (52%)	RR 0.50	95% CI 0.30 to 0.83; P=0.006	Pre-specified reporting by aetiology; largest observed effect
Death at 8 weeks (S. pneumoniae)	8/58 (14%)	17/50 (34%)	RR 0.41	95% CI 0.19 to 0.86; P=0.02	Pneumococcal subgroup mortality
Focal neurological abnormalities at 8 weeks (survivors assessed)	18/143 (13%)	24/119 (20%)	RR 0.62	95% CI 0.36 to 1.09; P=0.13	Assessed in survivors attending follow-up neurological examination
Hearing loss at 8 weeks (survivors assessed)	13/143 (9%)	14/119 (12%)	RR 0.77	95% CI 0.38 to 1.58; P=0.54	Assessed in survivors attending follow-up neurological examination
Seizures during admission	8/157 (5%)	17/144 (12%)	Not reported	P=0.04	In-hospital clinical course outcome
Cardiorespiratory failure during admission	16/157 (10%)	29/144 (20%)	Not reported	P=0.02	In-hospital clinical course outcome
Hyperglycaemia (>144 mg/dL)	50/157 (32%)	37/144 (26%)	Not reported	P=0.24	Adverse event monitoring
Gastrointestinal bleeding	2/157 (1%)	5/144 (3%)	Not reported	P=0.27	Adverse event monitoring

• Across all randomised patients, dexamethasone reduced unfavourable outcome at 8 weeks (15% vs 25%) and reduced 8-week mortality (7% vs 15%).
• Observed benefit was most pronounced in pneumococcal meningitis (unfavourable outcome 26% vs 52%; death 14% vs 34%).
• Major adverse events were not statistically increased; key monitored harms included hyperglycaemia (32% vs 26%) and gastrointestinal bleeding (1% vs 3%).

Internal Validity

• Randomisation and allocation:
  • Randomisation used a computer-generated list (blocks of six), stratified by centre; allocation was implemented by hospital pharmacists preparing indistinguishable study medication.
  • Emergency unblinding procedures were prespecified (sealed envelopes) and used twice; the code was otherwise kept until follow-up completion.
• Dropout, exclusions, and missing data:
  • Randomised: 157 dexamethasone vs 144 placebo; all received study medication initially.
  • Study medication discontinued early: 11/157 vs 11/144 (including inclusion violations 3 vs 1; adverse events 4 vs 1; accidental under-treatment 2 vs 2; withdrawal of consent 0 vs 1; open-label corticosteroids for clinical deterioration 2 vs 6).
  • Loss to follow-up for the 8-week outcome: 3 vs 4; imputed using last observation carried forward (sensitivity analyses described as consistent).
  • Neurological follow-up examination among survivors: 262/269 (97%).
• Performance and detection bias:
  • Double-blinding of participants and clinicians was designed to minimise differential co-interventions and outcome assessment bias.
  • Primary outcome (Glasgow Outcome Scale at 8 weeks) is a validated global functional measure but includes clinically adjudicated categories (not purely laboratory-based).
• Protocol adherence and separation of the variable of interest:
  • Planned regimen: 10 mg IV every 6 hours for 4 days; early discontinuation occurred equally (11 vs 11), reducing but not eliminating exposure separation.
  • Open-label corticosteroid “rescue” for deterioration was more frequent in placebo (6) than dexamethasone (2), plausibly biasing estimated benefit towards the null.
• Baseline characteristics and illness severity:
  • Groups were broadly balanced: mean age 45±17 vs 44±18 years; coma (Glasgow Coma Scale <8) 25/157 (16%) vs 23/144 (16%).
  • Notable imbalance: seizures before admission occurred in 15/157 (10%) vs 7/144 (5%).
  • Median Glasgow Coma Scale score on admission was 14 in both groups (range 3–15).
• Heterogeneity and subgroup interpretation:
  • Pathogen mix differed in prognosis: pneumococcal meningitis comprised 58/157 vs 50/144; meningococcal 50/157 vs 47/144; other bacteria 20/157 vs 16/144; negative cultures 29/157 vs 31/144.
  • Subgroup by admission Glasgow Coma Scale showed heterogeneity: among patients with coma (score <8), unfavourable outcome was 5/25 (20%) vs 14/26 (54%) (RR 0.37; 95% CI 0.15 to 0.93; P=0.04), whereas moderate impairment (score 8–13) did not show benefit (15/22 [68%] vs 15/26 [58%]; RR 1.18; 95% CI 0.75 to 1.86; P=0.57).
• Timing and dose:
  • Intended timing was before the first antibiotic dose; after protocol amendment, administration with antibiotics was permitted.
  • Exact time intervals from presentation to antibiotics and from antibiotics to study drug were not reported.
• Statistical rigour:
  • Planned sample size was achieved (301 randomised), supporting the prespecified power assumptions for the primary outcome.
  • Primary analysis was intention-to-treat with relative risks and 95% CIs; adjustment for prognostic factors was reported as supportive of the primary findings.

Conclusion on Internal Validity: Overall, internal validity appears moderate-to-strong, supported by concealed randomisation, double-blinding, high follow-up completeness, and intention-to-treat analysis; key threats include protocol amendments over a long recruitment period, small imbalances (e.g., baseline seizures), and use of last observation carried forward for a small amount of missing 8-week outcome data.

External Validity

• Population representativeness:
  • Participants were adults with suspected bacterial meningitis undergoing lumbar puncture in European hospitals; severity included a meaningful comatose subgroup (16% in each arm).
  • Exclusion of patients who had received antibiotics in the previous 48 hours limits applicability to settings with high rates of pre-hospital antibiotic exposure.
• Healthcare system and microbiology context:
  • Empiric antibiotic strategies and pathogen resistance patterns have evolved since 1993–2001; the pneumococcal isolates available for susceptibility testing in this trial were reported as penicillin susceptible.
  • Transferability may be reduced in regions with higher prevalence of penicillin/cephalosporin non-susceptible pneumococci or where empiric regimens rely heavily on vancomycin.
• Applicability to other subpopulations:
  • Patients with suspected tuberculosis or fungal meningitis were excluded; findings should not be extrapolated to these groups.
  • Findings may be less applicable in settings with delayed presentation, high HIV prevalence, and different pathogen spectra.

Conclusion on External Validity: Generalisability is moderate for adults with suspected acute bacterial meningitis presenting early to high-resource hospital settings, but is more limited in populations with frequent pre-treatment, high antimicrobial resistance, high HIV prevalence, or materially different empiric antibiotic pathways.

Strengths & Limitations

• Strengths:
  • Randomised, double-blind, placebo-controlled design with concealment procedures and limited unblinding.
  • Clinically meaningful primary endpoint (8-week functional outcome) with high follow-up completeness.
  • Early, protocolised corticosteroid dosing aligned to mechanistic rationale (around first antibiotics).
  • Multicentre, international conduct across 52 hospitals, enhancing within-Europe generalisability.
• Limitations:
  • Long recruitment period (1993–2001) with protocol amendments (including antibiotic strategy and timing of study drug relative to antibiotics), complicating interpretation in modern practice settings.
  • Exclusion of patients with antibiotics in the previous 48 hours may select for a subgroup with earlier presentation and different risk profile.
  • Subgroup analyses (pathogen- and severity-based) were underpowered for many strata, and some showed heterogeneity in direction/magnitude of effect.
  • Primary outcome and several secondary outcomes rely on clinical categorisation; a small amount of missing primary outcome data was imputed.

Interpretation & Why It Matters

• Clinical practice

  • Adjunctive dexamethasone (10 mg IV every 6 hours for 4 days) given before/with antibiotics improved overall 8-week functional outcomes and reduced mortality in this adult cohort.
  • Greatest benefit was observed in pneumococcal meningitis, supporting pathogen- and setting-aware application rather than indiscriminate use.
  • Therapy must not delay antibiotics; steroid benefit is mechanistically and empirically linked to early administration at the start of antimicrobial therapy.
• Pathophysiological insight

  • The findings support the concept that modulating early host inflammatory injury can change hard clinical outcomes (mortality and global disability), not merely surrogate markers.
• Systems impact

  • The trial influenced international guideline pathways and emergency department protocols by embedding “steroids with first antibiotics” into early meningitis bundles.

Controversies & Subsequent Evidence

• Antibiotic resistance and antimicrobial penetration: Concern was raised that dexamethasone may reduce CSF penetration of vancomycin (and potentially other agents), which could be clinically relevant in penicillin-resistant pneumococcal meningitis and might limit generalisability beyond low-resistance settings.¹²
• Context-dependence across settings: Large adult RCTs published after the Dutch trial produced divergent results—Vietnam showed no overall benefit but suggested benefit in microbiologically proven bacterial meningitis, whereas an adult trial in sub-Saharan Africa (high HIV prevalence) showed no mortality benefit and raised concern for harm—highlighting effect modification by pathogen mix, host factors, and health-system context.³⁴
• Timing, selection, and mechanism of benefit: Correspondence queried whether written consent and pre-lumbar puncture CT imaging could delay antibiotics and whether the observed mortality difference reflected systemic (shock/MODS) rather than neurological mechanisms; the authors noted that exact delays were not recorded and reported that systemic complications accounted for 4/11 vs 9/21 deaths (dexamethasone vs placebo) in interim data reviewed by the monitoring committee.²
• Long-term outcomes and safety signals: A long-term follow-up study of adult bacterial meningitis survivors reported persistent neurological and cognitive sequelae but did not identify an excess of long-term cognitive impairment attributable to dexamethasone, informing longer-horizon safety considerations beyond the 8-week endpoint.⁵
• Evidence synthesis: Systematic reviews conclude that adjunctive corticosteroids reduce hearing loss and neurological sequelae overall, with mortality benefits most apparent in high-income settings and pneumococcal meningitis; an individual patient data meta-analysis emphasised heterogeneity and did not demonstrate a uniform mortality/disability benefit across patient types.⁶⁷
• Guideline integration: Major international guidelines incorporated early adjunctive dexamethasone into adult bacterial meningitis care pathways, typically emphasising administration before/with first antibiotics and linking continuation to likely/confirmed pneumococcal meningitis and local epidemiology.⁸⁹¹⁰¹¹

Summary

• In 301 adults with suspected bacterial meningitis, adjunctive dexamethasone (10 mg IV every 6 hours for 4 days) reduced unfavourable 8-week functional outcome (15% vs 25%; RR 0.59; 95% CI 0.37 to 0.94) and reduced 8-week mortality (7% vs 15%; RR 0.48; 95% CI 0.24 to 0.96).
• The largest observed benefit was in pneumococcal meningitis (unfavourable outcome 26% vs 52%; death 14% vs 34%).
• Secondary neurological outcomes at 8 weeks (hearing loss; focal deficits) were not statistically reduced, though confidence intervals were compatible with clinically important effects.
• Major monitored adverse events were not statistically increased (hyperglycaemia 32% vs 26%; gastrointestinal bleeding 1% vs 3%).
• Internal validity is supported by blinding and follow-up completeness, but interpretation requires attention to protocol amendments, subgroup heterogeneity, and contextual generalisability.

Further Reading

Other Trials

• 2007Nguyen THM, Tran THC, Thwaites GE, et al. Dexamethasone in Vietnamese adolescents and adults with bacterial meningitis. N Engl J Med. 2007;357(24):2431-2440.
• 2007Scarborough M, Gordon SB, Whitty CJM, et al. Corticosteroids for bacterial meningitis in adults in sub-Saharan Africa. N Engl J Med. 2007;357(24):2441-2450.
• 1999Thomas R, Le Tulzo Y, Bouget J, et al. Trial of dexamethasone treatment for severe bacterial meningitis in adults. Intensive Care Med. 1999;25(5):475-480.
• 2002Molyneux EM, Walsh AL, Phiri AJ, et al. Dexamethasone treatment in childhood bacterial meningitis in Malawi: a randomised controlled trial. Lancet. 2002;360(9328):211-218.
• 1991Odio CM, Faingezicht I, Paris M, et al. The beneficial effects of early dexamethasone administration in infants and children with bacterial meningitis. N Engl J Med. 1991;324(22):1525-1531.

Systematic Review & Meta Analysis

• 2015Brouwer MC, McIntyre P, Prasad K, van de Beek D. Corticosteroids for acute bacterial meningitis. Cochrane Database Syst Rev. 2015;(9):CD004405.
• 2010van de Beek D, Farrar JJ, de Gans J, et al. Adjunctive dexamethasone in bacterial meningitis: a meta-analysis of individual patient data. Lancet Neurol. 2010;9(3):254-263.
• 2009Assiri AM, Alasmari FA, Zimmerman VA, Baddour LM, Erwin PJ, Tleyjeh IM. Corticosteroid administration and outcome of adolescents and adults with acute bacterial meningitis: a meta-analysis. Mayo Clin Proc. 2009;84(5):403-409.
• 1997McIntyre PB, Berkey CS, King SM, et al. Dexamethasone as adjunctive therapy in bacterial meningitis. A meta-analysis of randomized clinical trials. JAMA. 1997;278(11):925-931.

Observational Studies

• 2010Brouwer MC, Heckenberg SG, de Gans J, Spanjaard L, Reitsma JB, van de Beek D. Nationwide implementation of adjunctive dexamethasone therapy for pneumococcal meningitis. Neurology. 2010;75:1533-1539.
• 2012Heckenberg SG, Brouwer MC, van der Ende A, van de Beek D. Adjunctive dexamethasone in adults with meningococcal meningitis. Neurology. 2012;79(15):1563-1569.
• 2012Fritz D, Brouwer MC, van de Beek D. Dexamethasone and long-term survival in bacterial meningitis. Neurology. 2012 Nov 20;79(21):2177-2179.
• 2006Weisfelt M, Hoogman M, van de Beek D, de Gans J, Schmand BA. Dexamethasone and long-term outcome in adults with bacterial meningitis. Ann Neurol. 2006;60(4):456-468.
• 2022Ellis J, Watson C, and Meningitis in UK and Ireland Study Collaborators. Clinical care for patients with meningitis and meningococcal septicaemia in UK and Ireland hospitals: how close are we to current guidelines? BMJ Open. 2022;12:e062698.

Guidelines

• 2023Klein M, Abdel-Hadi C, Bühler R, et al. German guidelines on community-acquired acute bacterial meningitis in adults. Neurol Res Pract. 2023;5:44.
• 2016van de Beek D, Cabellos C, Dzupova O, et al. ESCMID guideline: diagnosis and treatment of acute bacterial meningitis. Clin Microbiol Infect. 2016;22(Suppl 3):S37-S62.
• 2016McGill F, Heyderman RS, Michael BD, et al. The UK joint specialist societies guideline on the diagnosis and management of acute meningitis and meningococcal sepsis in immunocompetent adults. J Infect. 2016;72(4):405-438.
• 2004Tunkel AR, Hartman BJ, Kaplan SL, et al. Practice guidelines for the management of bacterial meningitis. Clin Infect Dis. 2004;39(9):1267-1284.

Notes

• This Dutch trial’s effect estimates reflect a high-income European context with early lumbar puncture and low reported pneumococcal penicillin non-susceptibility; subsequent evidence supports context- and pathogen-aware implementation.

Overall Takeaway

This Dutch multicentre RCT established that early adjunctive dexamethasone can improve survival and global functional outcome in adults with suspected bacterial meningitis, with the clearest signal in pneumococcal disease. It became a cornerstone for guideline recommendations to give dexamethasone at the time of first antibiotics, while subsequent trials and meta-analyses clarified that benefit is context-dependent and should be applied with awareness of local epidemiology and antimicrobial strategies.

Bibliography

• 1Tunkel AR, Scheld WM. Corticosteroids for everyone with meningitis? N Engl J Med. 2002;347(20):1613-1615.
• 2Tabas JA, Chambers HF, Tancredi DN, et al. Dexamethasone in adults with bacterial meningitis. N Engl J Med. 2003;348(10):954-957.
• 3Nguyen THM, Tran THC, Thwaites GE, et al. Dexamethasone in Vietnamese adolescents and adults with bacterial meningitis. N Engl J Med. 2007;357(24):2431-2440.
• 4Scarborough M, Gordon SB, Whitty CJM, et al. Corticosteroids for bacterial meningitis in adults in sub-Saharan Africa. N Engl J Med. 2007;357(24):2441-2450.
• 5Weisfelt M, Hoogman M, van de Beek D, de Gans J, Schmand BA. Dexamethasone and long-term outcome in adults with bacterial meningitis. Ann Neurol. 2006;60(4):456-468.
• 6Brouwer MC, McIntyre P, Prasad K, van de Beek D. Corticosteroids for acute bacterial meningitis. Cochrane Database Syst Rev. 2015;(9):CD004405.
• 7van de Beek D, Farrar JJ, de Gans J, et al. Adjunctive dexamethasone in bacterial meningitis: a meta-analysis of individual patient data. Lancet Neurol. 2010;9(3):254-263.
• 8van de Beek D, Cabellos C, Dzupova O, et al. ESCMID guideline: diagnosis and treatment of acute bacterial meningitis. Clin Microbiol Infect. 2016;22(Suppl 3):S37-S62.
• 9McGill F, Heyderman RS, Michael BD, et al. The UK joint specialist societies guideline on the diagnosis and management of acute meningitis and meningococcal sepsis in immunocompetent adults. J Infect. 2016;72(4):405-438.
• 10Klein M, Abdel-Hadi C, Bühler R, et al. German guidelines on community-acquired acute bacterial meningitis in adults. Neurol Res Pract. 2023;5:44.
• 11Tunkel AR, Hartman BJ, Kaplan SL, et al. Practice guidelines for the management of bacterial meningitis. Clin Infect Dis. 2004;39(9):1267-1284.