Abstract
Background: In this nationwide prospective cohort study, we evaluated the implementation of adjunctive dexamethasone therapy in Dutch adults with pneumococcal meningitis.
Methods: From March 2006 through January 2009, all Dutch patients over 16 years old with community-acquired pneumococcal meningitis were prospectively evaluated. Outcome was classified as unfavorable (defined by a Glasgow Outcome Scale score of 1 to 4 points at discharge) or favorable (a score of 5). Clinical characteristics and outcome were compared with a similar nationwide cohort of 352 patients with pneumococcal meningitis from a previous period before guidelines recommended dexamethasone therapy (1998–2002). A multivariable prognostic model was used to adjust for differences in case mix between the 2 cohorts.
Results: We evaluated 357 episodes with pneumococcal meningitis in 2006–2009. Characteristics on admission were comparable with the earlier cohort (1998–2002). Dexamethasone was started with or before the first dose of antibiotics in 84% of episodes in 2006–2009 and 3% in 1998–2002. At discharge, unfavorable outcome was present in 39% in 2006–2009 and 50% in 1998–2002 (odds ratio [OR] 0.63; 95% confidence interval [CI] 0.46–0.86; p = 0.002). Rates of death (20% vs 30%; p = 0.001) and hearing loss (12% vs 22%; p = 0.001) were lower in 2006–2009. Differences in outcome remained after adjusting for differences in case mix between cohorts.
Conclusions: Dexamethasone therapy has been implemented on a large scale as adjunctive treatment of adults with pneumococcal meningitis in the Netherlands. The prognosis of pneumococcal meningitis on a national level has substantially improved after the introduction of adjunctive dexamethasone therapy.
Classification of evidence: This study provides Class III evidence that dexamethasone (10 mg IV, given every 6 hours for 4 days started before or with the first dose of parenteral antibiotics) reduced the proportion of patients with unfavorable outcomes (Glasgow Outcome Scale score of 1 to 4) in the 2006–2009 cohort, as compared to the 1998–2002 cohort (39% vs 50%; OR 0.63; 95% CI 0.46–0.86; p = 0.002). Mortality rate (20% vs 30%; absolute risk difference 10%; 95% CI 4%–17%; p = 0.001) was also lower in 2006–2009.
In 2004, we published a nationwide prospective cohort study in adults with bacterial meningitis from 1998 through 2002.1 In this study, 696 adults with community-acquired bacterial meningitis were included and most common pathogens were Streptococcus pneumoniae (51%) and Neisseria meningitidis (37%). The case fatality rate was 21% and half of surviving patients had neurologic sequelae.1,2 The mortality rate was highest among patients with pneumococcal meningitis (30%). This study was performed before routine dexamethasone therapy was introduced.1
Experimental models have shown that treatment with corticosteroids resulted in a reduction of the inflammatory response in the subarachnoid space and improved outcome.3,4 In 2002, a European multicenter randomized clinical trial showed a beneficial effect of adjunctive dexamethasone therapy in adults with bacterial meningitis.5 In this clinical trial, treatment with dexamethasone was associated with a reduction in the risk of an unfavorable outcome (relative risk 0.59; 95% confidence interval [CI] 0.37–0.94; p = 0.03). The effect was most apparent in the pneumococcal subgroup (relative risk 0.50; 95% CI 0.30–0.83; p = 0.006).
Four large randomized clinical trials on adjunctive dexamethasone in bacterial meningitis showed conflicting results.6,–,9 A recent meta-analysis of individual patient data of 5 recent randomized controlled trials showed no effect of adjunctive dexamethasone in pneumococcal meningitis.10 Therefore, the use of dexamethasone in pneumococcal meningitis remains controversial. Nevertheless, guidelines recommend routine use of adjunctive dexamethasone in adults with pneumococcal meningitis in high-income countries.11,–,13 We assessed the implementation of adjunctive dexamethasone therapy in adults with pneumococcal meningitis and its impact on outcome.
METHODS
We identified adults (defined as patients older than 16 years of age) who had pneumococcal meningitis defined by positive CSF culture and were listed in the database of the Netherlands Reference Laboratory for Bacterial Meningitis from March 2006 to January 2009. This laboratory receives CSF isolates from approximately 85% of all patients with bacterial meningitis in the Netherlands (population, 16.2 million).14 Daily updates were provided of hospitals where patients with bacterial meningitis had been admitted in the preceding 2 to 6 days. Physicians were informed about the study by telephone. Physicians could also contact investigators 24/7 to include patients, without preceding report of the reference laboratory. Patients or their legal representatives received written information concerning the study and were asked to give written informed consent for participation. Online case record forms were used to collect data. Patients with negative CSF cultures or hospital-associated meningitis were excluded. Patients with an altered immune status owing to the use of immunosuppressive drugs or splenectomy, diabetes mellitus, or alcoholism were considered immunocompromised, as were patients infected with HIV.
Outcome was graded according to the Glasgow Outcome Scale. A score of 1 on this scale indicates death; a score of 2 a vegetative state (the patient is unable to interact with the environment); a score of 3 severe disability (the patient is unable to live independently but can follow commands); a score of 4 moderate disability (the patient is capable of living independently but unable to return to work or school); and a score of 5 mild or no disability (the patient is able to return to work or school). A favorable outcome was defined as a score of 5, and an unfavorable outcome as a score of 1 to 4. The Glasgow Outcome Scale is a well-validated instrument with good interobserver agreement.15 At discharge, all surviving patients underwent a neurologic examination performed by a neurologist which included the assessment of the Glasgow Outcome Scale.
We compared our results with historical controls from a study with similar design that included 352 patients with pneumococcal meningitis from 1998 through 2002, before guidelines recommended routine dexamethasone therapy.1,2
The Mann-Whitney U test (continuous variables) and χ2 test (categorical variables) were used to identify differences in demographic and clinical characteristics between the 2 cohorts. In the earlier cohort we developed a prediction model with 18 potentially relevant prognostic factors for unfavorable outcome (table e-1 on the Neurology® Web site at www.neurology.org). We used logistic regression analysis to calculate odds ratios (ORs) and 95% CIs to assess the strength of the association between potential prognostic factors and the probability of an unfavorable outcome. Missing values were imputed by use of multivariate normal distributions and coefficients of 10 rounds of imputation were combined to obtain the final estimates from the multivariate logistic regression model. The coefficients of the multivariable prediction model were applied to obtain a risk score for each patient in the recent cohort. This risk score incorporates all available data on risk factors given an individual profile on clinical and demographic characteristics. We used these risk scores to calculate the expected number of events (if dexamethasone would not have been introduced) and compared these with the actually observed number of patients with unfavorable outcome in the recent cohort. Expected vs observed numbers were calculated for the recent cohort as a whole, across tertiles of predicted risk (i.e., severity of disease), and for various clinical subgroups. All statistical tests were 2-tailed, and a p value less than 0.05 was regarded as significant. Analyses were undertaken with SAS software version 9.1.
The primary research question was if the introduction of adjunctive dexamethasone in the Netherlands has improved outcome in pneumococcal meningitis. The study design provides a Class III level of evidence.
The funding source had no role in study design, collection, analysis, and interpretation of data, writing the report, or the decision to submit the paper for publication.
Standard protocol approvals, registrations, and patient consents.
The study was approved by all Dutch local ethics committees. All participating hospitals and local investigators are presented in appendix e-1.
RESULTS
A total of 787 episodes of bacterial meningitis were identified from March 2006 through January 2009 (figure 1). The cohort consisted of 518 episodes of community-acquired bacterial meningitis, including 357 episodes of pneumococcal meningitis in 354 patients. Classic symptoms and signs of meningitis were present in a large proportion of the patients (table 1). The classic triad of neck stiffness, fever, and altered mental status (defined as a score on the Glasgow Coma Scale <14) was present in 54% and coma (defined as a score on the Glasgow Coma Scale <8) in 18%. At least one individual CSF finding predictive of bacterial meningitis (a glucose level of less than 34 mg/dL [1.9 mmol/L], a ratio of CSF glucose to blood glucose of less than 0.23, a protein level of more than 220 mg/dL, or a white-cell count of more than 2,000 per mm3)16 was present in 328 of 348 episodes (94%). At admission, clinical characteristics and results of laboratory tests between cohorts were similar, although more episodes had positive blood cultures (85% vs 74%; p = 0.002) and less episodes of cranial nerve palsies (7% vs 12%; p = 0.02) in 2006–2009. The proportions of patients included in the 8 Dutch academic hospitals were similar between 2 time periods (11.7% vs 11.8%; p = 0.96).
Characteristics of Dutch adults with pneumococcal meningitis in 2 nationwide cohort studiesa
Cranial CT was performed on admission in 320 episodes (90%); results were normal in 52%. Abnormalities found were mastoid or sinus opacification in 37%, generalized brain edema in 18%, recent brain infarction in 7%, and other abnormalities in 16%. Imaging preceded lumbar puncture in 251 episodes (78%). The proportion of patients in the 2006–2009 cohort who experienced a delay in therapy due to cranial CT was unchanged compared to patients from 1998 to 2002 (155 of 357 [43%] vs 149 of 352 [42%]; p = 0.83).
Antimicrobial treatment consisted of penicillin or amoxicillin in 33% of episodes, third-generation cephalosporins in 28%, and a combination of penicillin or amoxicillin and third-generation cephalosporins in 34% of episodes; another regimen was used in 5%. Antibiotic treatment was in compliance with Dutch guidelines in 33% of episodes. The Dutch guideline recommends empirical therapy consisting of penicillin for adults between 16 and 60 years old, and empirical therapy consisting of amoxicillin plus a third-generation cephalosporin for patients over 60 years old or those with risk factors (defined as altered immune status, alcohol abuse, CSF leak, or recent head trauma).17 Guidelines for antibiotic use for meningitis in the Netherlands have not been changed from 1998 to 2009. Adherence to antibiotic guidelines between cohorts was similar (cohort 1998–2002, 32%).17
Antibiotic susceptibility testing was performed in 327 episodes; 2 strains showed intermediate susceptibility to penicillin (MIC 1.0 mg/L and 0.125 mg/L). The most common serotypes were type 3 and 14 (each 10%; table e-2); type 19F, 7F, and 9V (each 8%); and type 6B and 10A (each 6%). The proportions of disease due to strains that were covered by the 7-valent pneumococcal conjugate vaccine (PCV7) serotypes were unchanged in 2006–2009, compared to 1998–2002 (42% in 2006–2008 vs 38% in 1998–2002; p = 0.28).
Adjunctive dexamethasone was administered in 92% of episodes (table 2). Dexamethasone, 10 mg IV, given every 6 hours for 4 days was started before or with the first dose of parenteral antibiotics in 276 of 357 episodes (77%). Dexamethasone was given after the first dose of antibiotics in 28 episodes (8%); in 3 episodes this was prompted by clinical deterioration. Clinical deterioration in these 3 patients was caused by brain edema in 2 patients and respiratory failure due to bronchiolitis obliterans organizing pneumonia in 1 patient. There were no differences between patients treated with or without early dexamethasone with respect to antibiotic pretreatment (14% vs 11%; p = 0.79), immunocompromised state (22% vs 29%; p = 0.31), diastolic blood pressure (median, 80 vs 80 mm Hg; p = 0.61), or heart rate (median, 100 vs 100 beats per minute; p = 0.77). Adjunctive dexamethasone was administered in 59 episodes (17%) in 1998–2002. Eleven of these patients were included in the European dexamethasone in adulthood bacterial meningitis study and received dexamethasone 10 mg IV, given every 6 hours for 4 days, started before or with first dose of parenteral antibiotics; dexamethasone was initiated after clinical deterioration in all other episodes.5
Characteristics of intravenous dexamethasone treatmenta
During clinical course, neurologic complications (impairment of consciousness, seizures, or focal neurologic abnormalities) occurred in 60% of episodes and cardiorespiratory failure in 37% (table 3). Neurologic complications, including epileptic seizures, were less likely to occur in 2006–2008 as compared with 1998–2002 (60% vs 75%; p < 0.001). The rate of cardiorespiratory failure between cohorts was similar.
Clinical course, mortality, disability, and neurologic findings at dischargea
The mortality rate was 20% (table 3) and 39% of episodes had an unfavorable outcome. Neurologic examination was performed at discharge in 280 of 285 surviving patients (98%); most common abnormalities were hearing impairment (12%) and focal cerebral deficits (11%). The proportion of patients with unfavorable outcome (Glasgow Outcome Scale score of 1 to 4) was lower in the 2006–2009 cohort, as compared to the 1998–2002 cohort (39% vs 50%; OR 0.63; 95% CI 0.46–0.86; p = 0.002). Mortality rate (20% vs 30%; absolute risk difference 10%; 95% CI 4%–17%; p = 0.001) was also lower in 2006–2009.
On average, 3% of the values were missing and had to be imputed in the multivariate prediction model (table e-1). The observed unfavorable outcome in 2006–2009 of 39% was lower than the predicted 49% based on the multivariable prognostic model (p = 0.007). Figure 2A shows observed and predicted risks of unfavorable outcome within 3 groups of patients with increasing disease severity. Observed numbers of patients with unfavorable outcome in 2006–2009 were lower than the expected numbers in the middle and high-risk groups, whereas no differences were observed in the low-risk group. The improved outcome was primarily observed in those patients receiving the standard regimen of dexamethasone (figure 2B). There was no difference in observed and predicted outcome in patients who did not receive dexamethasone therapy. We explored the differences between observed and predicted outcome for various clinical subgroups of patients (figure e-1). The difference in observed vs predicted unfavorable outcome rate was largest in the subgroup of patients over aged 55 years, those without antibiotic pretreatment, and those with high CSF protein levels indicating severe CSF inflammation. This suggests that dexamethasone is most effective in patients with these characteristics.
Panel A shows predicted and observed rates of unfavorable outcome for groups with low, middle, and high risk for unfavorable outcome (groups based on tertiles). Panel B shows predicted and observed rates of unfavorable outcome for patients not treated with dexamethasone, those who received the recommended standard dexamethasone regimen (10 mg IV, given every 6 hours for 4 days, started before or with the first dose of parenteral antibiotics), and those who received an alternative regimen of dexamethasone.
DISCUSSION
Dexamethasone therapy has been implemented on a large scale as adjunctive treatment of adults with pneumococcal meningitis in the Netherlands. The drug was administered in 92% of episodes in 2006–2009. The large majority of physicians adhered to current guidelines recommending a standard regimen of dexamethasone, 10 mg IV, given every 6 hours for 4 days, started before or with the first dose of parenteral antibiotics.
The outcome of adults with community-acquired pneumococcal meningitis on a national level has significantly improved over the last few years. We found a decline in case fatality from 30% to 20%. This observation cannot be attributed to a change in disease severity as we corrected for a large set of prognostic factors. The main difference between cohorts was the successful introduction of adjunctive dexamethasone therapy in the Netherlands. The decline in case fatality that we observed matched the results of a randomized clinical trial that we performed in a comparable population.5
The use of observational data in the evaluation of treatment effects raises debates.18,19 The greatest concern with observational studies is the issue known as confounding by indication.20,21 Confounding by indication refers to the situation in daily clinical practice that prescribing will be guided by the prognosis of the patient: the worse the prognosis, the more or stronger therapy will be given. This means mixing treatment decisions with prognosis and that correction for important prognostic factors may only remove part of this bias. For several reasons, we believe that our observational data may provide valuable evidence in the controversy about the effectiveness of dexamethasone. First, our key analysis is based on comparing 2 national cohorts on an intention-to-treat basis (one from a period in which hardly any dexamethasone was used, compared to a cohort in which dexamethasone was generally prescribed). The bias due to prescribing dexamethasone to patients who are systematically in poorer or better condition does not apply when comparing 2 national cohorts as a whole. Second, we applied an extensive adjustment for differences in case mix between the 2 cohorts based on a large and independent body of data on prognostic factors in bacterial meningitis.1,2 Third, there are no indications of improvements in other (supportive) treatment options for bacterial meningitis that could explain such a large improvement. Fourth, the treatment benefit observed in our observational study was similar in magnitude as reported in the randomized clinical trial on dexamethasone. The vast majority of patients included in this trial were Dutch patients. Finally, the benefit of dexamethasone was observed across the whole study population, but was more prominent in patients actually receiving dexamethasone (per-treatment analysis).
Dexamethasone appears to be more effective in patients aged older than 55 years. This is consistent with findings of a recent individual patient data meta-analysis including 2,029 patients from 5 randomized controlled trials.10 Dexamethasone was not associated with a reduction in death in this meta-analysis (OR 0.97, 95% CI 0.79–1.19), but was effective in patients aged older than 55 years (OR for death 0.41 [95% CI 0.20–0.84], p = 0.01). In the meta-analysis the apparent benefit in adults aged over 55 years was interpreted as having occurred by chance, since there was no clear evidence of heterogeneity between the different age groups. Previous studies showed that induction of proinflammatory cytokines after septic stimuli is not adequately controlled by anti-inflammatory mechanisms in elderly persons.22 An age-related beneficial effect of dexamethasone could be an explanation for the apparent conflicting results of recent randomized controlled trials.5,6,8,9,23 Perhaps dexamethasone is most effective in older patients with severe CNS inflammation without antibiotic pretreatment.
From 1950 onwards, the introduction of modern hospital facilities, intensive care units, cranial CT, and evidence-based guidelines may all have contributed to the steady and gradual decrease from 40% to 30% in mortality of community-acquired pneumococcal meningitis.24 We now observed a further decrease in mortality of 10%, within a 4-year period, that could not be explained by differences in case mix. This study provides Class III evidence that dexamethasone given every 6 hours for 4 days reduced the proportion of patients with unfavorable outcome and reduced mortality in pneumococcal meningitis in adults. Our observation supports the use of adjunctive dexamethasone in adult pneumococcal meningitis in high-income countries.
DISCLOSURE
Dr. Brouwer, Dr. Heckenberg, Dr. de Gans, Dr. Spanjaard, and Dr. Reitsma report no disclosures. Dr. van de Beek has received research support from The Netherlands Organization for Health Research and Development and the Academic Medical Center.
ACKNOWLEDGMENT
The authors thank many physicians in the Netherlands for their cooperation.
Footnotes
Study funding: Supported by the Netherlands Organization for Health Research and Development (ZonMw; NWO-Veni grant 2006 [916.76.023] to D.v.d.B.) and the Academic Medical Center (AMC Fellowship 2008 to D.v.d.B.).
-
- CI
- confidence interval
- OR
- odds ratio
-
Supplemental data at www.neurology.org
- Received March 3, 2010.
- Accepted June 2, 2010.
- Copyright © 2010 by AAN Enterprises, Inc.
REFERENCES
Letters: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
You May Also be Interested in
More Online
Hemiplegic Migraine Associated With PRRT2 Variations A Clinical and Genetic Study
Dr. Robert Shapiro and Dr. Amynah Pradhan