111402 dexamethasone in adults with bacterial meningitis

C o py r ig ht 2 0 0 2 by t he Ma s s ac h u s e t t s Me d ic a l S o c ie t y V O L U M E 3 4 7
N U M B E R 2 0
DEXAMETHASONE IN ADULTS WITH BACTERIAL MENINGITIS
JAN DE GANS, PH.D., AND DIEDERIK VAN DE BEEK, M.D., FOR THE EUROPEAN DEXAMETHASONE IN ADULTHOOD BACTERIAL MENINGITIS STUDY INVESTIGATORS* ABSTRACT
HE mortality rate among adults with acute Background
bacterial meningitis and the frequency of among adults with acute bacterial meningitis, espe- cially those with pneumococcal meningitis. In studies vive are high, especially among patients with of bacterial meningitis in animals, adjuvant treatment pneumococcal meningitis.1,2 Unfavorable neurologic with corticosteroids has beneficial effects.
outcomes are not the result of treatment with inappro- Methods
priate antimicrobial agents, since cerebrospinal fluid double-blind, multicenter trial of adjuvant treatment cultures are sterile 24 to 48 hours after the start of an- with dexamethasone, as compared with placebo, in tibiotic therapy.3 Studies in animals have shown that adults with acute bacterial meningitis. Dexametha- bacterial lysis, induced by treatment with antibiotics, sone (10 mg) or placebo was administered 15 to 20 leads to inflammation in the subarachnoid space, which minutes before or with the first dose of antibiotic and was given every 6 hours for four days. The primary may contribute to an unfavorable outcome. These outcome measure was the score on the Glasgow Out- studies also show that adjuvant treatment with antiin- come Scale at eight weeks (a score of 5, indicating a flammatory agents, such as dexamethasone, reduces favorable outcome, vs. a score of 1 to 4, indicating an both cerebrospinal fluid inflammation and neurologic unfavorable outcome). A subgroup analysis according to the causative organism was performed. Analyses Many controlled trials have been performed to de- were performed on an intention-to-treat basis.
termine whether adjuvant corticosteroid therapy is Results
A total of 301 patients were randomly as- beneficial in children with acute bacterial meningitis.
signed to a treatment group: 157 to the dexametha- The results, however, do not point unequivocally to a sone group and 144 to the placebo group. The base- beneficial effect. A meta-analysis of randomized con- line characteristics of the two groups were similar.
Treatment with dexamethasone was associated with trolled trials performed since 1988 showed a beneficial a reduction in the risk of an unfavorable outcome (rel- effect of adjunctive dexamethasone therapy in terms of ative risk, 0.59; 95 percent confidence interval, 0.37 to severe hearing loss in children with Haemophilus influ- 0.94; P=0.03). Treatment with dexamethasone was enzae type b meningitis and suggested a protective ef- also associated with a reduction in mortality (relative fect in those with pneumococcal meningitis if the drug risk of death, 0.48; 95 percent confidence interval, 0.24 was given before or with parenteral antibiotics.6 There to 0.96; P=0.04). Among the patients with pneumo- are few data on the use of adjunctive dexamethasone coccal meningitis, there were unfavorable outcomes therapy in adults with bacterial meningitis. One large, in 26 percent of the dexamethasone group, as com- prospective, randomized trial (neither placebo-con- pared with 52 percent of the placebo group (relative trolled nor double-blind) showed a benefit of dex- risk, 0.50; 95 percent confidence interval, 0.30 to 0.83;P=0.006). Gastrointestinal bleeding occurred in two amethasone therapy in a subgroup of patients with patients in the dexamethasone group and in five pa- pneumococcal meningitis. The paucity of data pre- cludes a recommendation that dexamethasone be ad- Conclusions
improves the outcome in adults with acute bacterial From the Department of Neurology, Academic Medical Center, Amster- meningitis and does not increase the risk of gastroin- dam. Address reprint requests to Dr. de Gans at the Academic Medical Cen- testinal bleeding. (N Engl J Med 2002;347:1549-56.) ter, University of Amsterdam, Department of Neurology H2, P.O. Box 22660, Copyright 2002 Massachusetts Medical Society.
1100 DD Amsterdam, the Netherlands, or at [email protected].
*The investigators who participated in the European Dexamethasone in Adulthood Bacterial Meningitis Study are listed in the Appendix.
N Engl J Med, Vol. 347, No. 20 · November 14, 2002 · www.nejm.org · 1549
Downloaded from www.nejm.org by MATTHEW N. GRABER MD on March 24, 2007 . Copyright 2002 Massachusetts Medical Society. All rights reserved. The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne ministered routinely in adults with bacterial menin- or no disability (the patient is able to return to work or school).12 gitis.8,9 We conducted a study to determine whether A favorable outcome was defined as a score of 5, and an unfavorable adjunctive dexamethasone treatment improves the out- outcome as a score of 1 to 4. The Glasgow Outcome Scale has fre-quently been used in trials involving stroke and other brain injuries.
It is a well-validated scale with good interobserver agreement.13,14 Secondary outcome measures were death, focal neurologic ab- normalities (defined as aphasia, cranial-nerve palsy, monoparesis,hemiparesis, and severe ataxia), hearing loss, gastrointestinal bleed- Eligible Patients
ing (clinically relevant bleeding with a decreased serum hemoglobin Patients referred to one of the participating centers (listed in the level), fungal infection, herpes zoster, and hyperglycemia (a blood Appendix) were eligible for the study if they were 17 years of age or glucose level higher than 144 mg per deciliter [8.0 mmol per li- older, had suspected meningitis in combination with cloudy cer- ter]). Audiologic examination was performed in patients with clin- ebrospinal fluid, bacteria in cerebrospinal fluid on Gram’s staining, ical hearing loss. Subgroup analyses were performed for patients or a cerebrospinal fluid leukocyte count of more than 1000 per with prospectively defined causes of meningitis: Neisseria menin- cubic millimeter. Patients were excluded if they had a history of hy- gitidis, Streptococcus pneumoniae, other bacteria, and an unidenti- persensitivity to b-lactam antibiotics or corticosteroids; if they were fied cause (indicated by a negative cerebrospinal fluid culture).
pregnant; if they had a cerebrospinal shunt, had been treated withoral or parenteral antibiotics in the previous 48 hours, had a history Statistical Analysis
of active tuberculosis or fungal infection, or had a recent history ofhead trauma, neurosurgery, or peptic ulcer disease; or if they were Calculation of the required sample size was based on the assump- tion that dexamethasone would reduce the proportion of patients The study protocol was approved by the institutional review board with an unfavorable outcome from 40 to 25 percent. With a two- of each participating hospital. All patients or their legally author- sided test, an alpha level of 0.05, and a power of 80 percent, the ized representatives gave written informed consent before enroll- analysis required 150 patients per group. The analysis of outcomes ment. Patients were enrolled between June 1993 and December was performed on an intention-to-treat basis with the use of a last- 2001. The study was designed, conducted, and analyzed independ- observation-carried-forward procedure. An additional analysis in which data for patients lost to follow-up were defined as missingwas also performed. The results of these two analyses were similar.
Treatment
Proportions of patients in the two groups were compared with Fisher’s exact test. Two-tailed P values of less than 0.05 were con- Patients were randomly assigned to receive dexamethasone sodi- sidered to indicate statistical significance. Parametric and nonpara- um phosphate (Oradexon), at a dose of 10 mg given every six hours metric values were tested with Student’s t-test and the Mann–Whit- intravenously for four days, or placebo that was identical in appear- ney U test, respectively. The results are expressed as relative risks ance to the active drug. The study medication was given 15 to 20 for the dexamethasone group as compared with the placebo group, minutes before the parenteral administration of antibiotics. After with a relative risk of less than 1.0 indicating a beneficial effect. Lo- the interim analysis, the protocol was amended to allow adminis- gistic-regression analysis of base-line variables (sex; age; duration tration of the study medication with the antibiotics.
of symptoms; presence or absence of seizures, coma, and hypoten- Balanced treatment assignments within each hospital were sion on admission; results of blood culture; cerebrospinal fluid achieved with the use of a computer-generated list of random num- white-cell count; and causative organism) was performed to iden- bers in blocks of six. The code was not broken until the last patient tify risk factors for an adverse outcome other than the group assign- to be enrolled had completed eight weeks of follow-up. Treatment ment. Ninety-five percent confidence intervals, calculated with the assignments were concealed from all investigators, but in an emer- use of Confidence Interval Analysis, are reported.15 gency, investigators had access to the sealed, opaque envelopes con- A three-member independent data-monitoring committee per- taining the assignments; two emergencies occurred. Patients were formed an interim analysis after 150 patients had been enrolled.
initially treated with amoxicillin (2 g given intravenously every four The study would have been stopped if any significant differences hours) for 7 to 10 days, depending on the cause of the meningitis in efficacy or safety had been found. On January 10, 1997, the data- and the clinical response. This regimen was based on the available monitoring committee recommended early termination of the trial data on susceptibility to antibiotics of cerebrospinal fluid isolates because the enrollment rate was too slow for completion within in the Netherlands.10 The initial antibiotic treatment was maintained a reasonable time. The committee subsequently reconsidered its de- or changed according to the results of Gram’s staining of cerebro- cision and recommended that the trial be restarted if the enrollment rate could be improved. To increase the enrollment rate, two amend- Laboratory Studies
ments of the protocol were made. First, the protocol was amendedto allow administration of the study medication with the antibiot- Routine examination and cultures of blood and cerebrospinal flu- ics. This decision was based on the results of a meta-analysis of trials id were performed before the initiation of antibiotic treatment. On of dexamethasone in children with acute bacterial meningitis.6 day five, routine blood chemical tests were performed, including Second, the protocol was amended to allow investigators to follow measurement of glucose and hemoglobin levels. As part of routine local guidelines for administering empirical antibiotic therapy. This surveillance, the Netherlands Reference Laboratory for Bacterial change was made because of the participation of centers in coun- Meningitis performed in vitro testing of cerebrospinal fluid iso- tries where highly resistant pneumococcal strains are more com- lates for susceptibility to penicillin.11 Assessment of Outcome
The primary outcome measure was the score on the Glasgow Outcome Scale eight weeks after randomization, as assessed by the A total of 301 patients were randomly assigned to patient’s physician. A score of 1 indicates death; 2, a vegetative state a study group: 157 to the dexamethasone group and (the patient is unable to interact with the environment); 3, severe 144 to the placebo group. Two patients (one in each disability (the patient is unable to live independently but can followcommands); 4, moderate disability (the patient is capable of living group) did not meet the inclusion criteria because independently but unable to return to work or school); and 5, mild they were too young. Seven patients in the dexameth- 1550 · N Engl J Med, Vol. 347, No. 20 · November 14, 2002 · www.nejm.org
Downloaded from www.nejm.org by MATTHEW N. GRABER MD on March 24, 2007 . Copyright 2002 Massachusetts Medical Society. All rights reserved. D EX A M ET H AS O N E I N BAC T E R I A L M E N I N G I T I S
asone group and nine in the placebo group each met tients with herniation and hypodense lesions in the one exclusion criterion; one patient in the dexametha- patient with acute disseminated encephalomyelitis.
sone group met two exclusion criteria. Eleven patients Eight weeks after admission, neurologic examina- in each group were withdrawn from treatment early, tions were performed in 262 of 269 patients (97 per- but all 301 patients received the assigned treatment, cent). Seven patients were lost to follow-up, three in at least initially (Fig. 1). Four patients were withdrawn the dexamethasone group and four in the placebo because they did not meet the inclusion criteria (three group. At discharge, six of these seven patients had a in the dexamethasone group and one in the placebo score of 5 on the Glasgow Outcome Scale, and one group), and five because of adverse events (four in the had a score of 4. These last-observation scores were dexamethasone group and one in the placebo group).
carried forward to eight weeks, so that all 301 patients Thirteen patients were withdrawn for other reasons: were included in the analyses of the primary outcome four were accidentally not treated for four days (two in each group; all four received the assigned studymedication the first day or the first two days), one pa- Base-Line Characteristics of the Patients
tient in the placebo group withdrew consent, and Classic symptoms and signs of meningitis were eight had clinical deterioration and were treated with present in a large proportion of the patients (head- corticosteroids (two in the dexamethasone group and ache in 94 percent, fever in 81 percent, and neck stiff- six in the placebo group). The reasons for corticoster- ness in 94 percent). At base line, the clinical charac- oid treatment were brain herniation (in three patients), teristics and the results of laboratory tests were similar pulmonary problems (in three), disseminated intra- in the dexamethasone and placebo groups, although vascular coagulation (in one), and acute disseminated a higher percentage of patients in the dexamethasone encephalomyelitis (in one). Cranial computed tomog- group had seizures (Table 1). The mean cerebrospinal raphy (CT) showed diffuse brain swelling in the pa- fluid pressure was also similar in the two groups, as was 11 Patients withdrawn early from treatment 11 Patients withdrawn early from treatment Figure 1. Random Assignment to Treatment, Withdrawal from Treatment, and Follow-up among 301 Adults with Bacterial Meningitis.
N Engl J Med, Vol. 347, No. 20 · November 14, 2002 · www.nejm.org · 1551
Downloaded from www.nejm.org by MATTHEW N. GRABER MD on March 24, 2007 . Copyright 2002 Massachusetts Medical Society. All rights reserved. The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne TABLE 1. BASE-LINE CHARACTERISTICS OF THE STUDY POPULATION.*
DEXAMETHASONE GROUP
PLACEBO GROUP
CHARACTERISTIC
No bacteria in CSF on Gram’s staining but Duration of symptoms before admission — hr *Plus–minus values are means ±SD. CSF denotes cerebrospinal fluid.
†CSF pressure was measured in 157 patients (81 in the dexamethasone group and 76 in the pla- ‡Scores on the Glasgow Coma Scale range from 3 to 14, with 14 indicating a normal level of con- sciousness (abnormal flexion was omitted from the scale).
§A total of 86 patients in the dexamethasone group and 89 in the placebo group were examined ¶CSF cultures were performed in 155 patients in the dexamethasone group and 144 in the placebo group. The category of other bacteria included various streptococcal species other than S. pneumoniaein 12 patients, Listeria monocytogenes in 6, Haemophilus influenzae in 4, Staphylococcus aureus in 3,Escherichia coli in 1, Klebsiella pneumoniae in 1, corynebacterium species in 1, and Capnocytophagacanimorsus in 1.
¿To convert the values for glucose to millimoles per liter, multiply by 0.05551.
**Blood culture was performed in 264 patients (136 in the dexamethasone group and 128 in the the proportion of patients in the two groups who had per liter], a glucose ratio [the ratio of glucose in the very high pressure (40 cm of water or higher). Gram’s cerebrospinal fluid to that in blood] below 0.23, a pro- staining of cerebrospinal fluid specimens, performed tein level above 220 mg per deciliter, a white-cell count in 290 patients, showed bacteria in 215 patients (74 above 2000 per cubic millimeter, or a neutrophil count percent). Cerebrospinal fluid culture yielded bacteria in 234 of 299 patients (78 percent). Forty-three of the65 patients (66 percent) with negative cerebrospinal Efficacy
fluid cultures had at least one individual cerebrospinal Eight weeks after enrollment, the percentage of pa- fluid finding that was predictive of bacterial meningitis tients with an unfavorable outcome was significantly (a glucose level below 34 mg per deciliter [1.9 mmol smaller in the dexamethasone group than in the pla- 1552 · N Engl J Med, Vol. 347, No. 20 · November 14, 2002 · www.nejm.org
Downloaded from www.nejm.org by MATTHEW N. GRABER MD on March 24, 2007 . Copyright 2002 Massachusetts Medical Society. All rights reserved. D EX A M ET H AS O N E I N BAC T E R I A L M E N I N G I T I S
cebo group (15 percent vs. 25 percent; relative risk, cebo died. Adjuvant treatment with dexamethasone 0.59; 95 percent confidence interval, 0.37 to 0.94; P= did not have a significant beneficial effect on neuro- 0.03) (Table 2); the absolute reduction in the risk of logic sequelae, including hearing loss. During admis- an unfavorable outcome was 10 percent. Predictors sion, audiologic examination was performed in 28 pa- of an unfavorable outcome were coma on admission tients, 14 of whom had severe hearing loss (60 dB or (P=0.002), hypotension (P=0.03), and meningitis more in one or both ears). At eight weeks, 27 patients due to S. pneumoniae (P=0.02). The benefit of dex- had hearing loss. The distribution of scores on the amethasone remained substantial in an analysis adjust- Glasgow Outcome Scale is shown in Table 3. The low- ed for other risk factors (adjusted odds ratio, 0.45; P= er mortality in the dexamethasone group did not re- 0.02). Among the patients with pneumococcal men- sult in an increased rate of severe neurologic sequelae ingitis, 26 percent in the dexamethasone group had an unfavorable outcome, as compared with 52 percent Table 4 shows the relative risk of an unfavorable in the placebo group. Among the patients with men- outcome according to the severity of disease (as indi- ingitis due to N. meningitidis, however, adjuvant treat- cated by the score on the Glasgow Coma Scale on ad- ment with dexamethasone did not provide a signifi- mission). Dexamethasone appeared to be most ben- eficial in patients with moderate or severe disease.
The proportion of patients who died was signifi- cantly smaller in the dexamethasone group than in Adverse Events
the placebo group (7 percent vs. 15 percent; relative Adverse events resulted in the early withdrawal of risk, 0.48; 95 percent confidence interval, 0.24 to four patients in the dexamethasone group and one 0.96; P=0.04). Among the patients with pneumococ- in the placebo group (Fig. 1). In the dexamethasone cal meningitis, 14 percent of those who received dex- group, two patients were withdrawn because of se- amethasone and 34 percent of those who received pla- vere hyperglycemia, one because of suspected stom- TABLE 2. OUTCOMES EIGHT WEEKS AFTER ADMISSION,
DEXAMETHASONE
RELATIVE RISK
OUTCOME AND CULTURE RESULTS
(95% CI)†
*The analyses of unfavorable outcome and death included all patients and were performed with a last-observation-carried-forward procedure. The analyses of neurologic abnormalities and hearingloss included all surviving patients who underwent neurologic examination at eight weeks.
‡Included in this category are two patients in whom cerebrospinal fluid culture was not performed.
N Engl J Med, Vol. 347, No. 20 · November 14, 2002 · www.nejm.org · 1553
Downloaded from www.nejm.org by MATTHEW N. GRABER MD on March 24, 2007 . Copyright 2002 Massachusetts Medical Society. All rights reserved. The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne Antibiotic Treatment and Susceptibility Testing
TABLE 3. DISTRIBUTION OF SCORES ON THE GLASGOW OUTCOME
The most frequently prescribed initial antibiotics were amoxicillin and penicillin (in 77 percent of thepatients), third-generation cephalosporin (in 8 per- DEXAMETHASONE GROUP
PLACEBO GROUP
cent), and penicillin or amoxicillin combined with a cephalosporin (in 8 percent). The Reference Labora- tory for Bacterial Meningitis received cerebrospinal fluid isolates from 78 of 108 patients with S. pneumo- niae meningitis (72 percent); all the isolates were sus- ceptible to penicillin (minimal inhibitory concentra- tion, less than 0.1 µg per milliliter). Isolates from 80 of the 97 patients with meningococcal meningitis (82percent) were tested by the reference laboratory; only1 showed intermediate resistance to penicillin (mini-mal inhibitory concentration, between 0.1 and 1.0 µg ach perforation (which was not the case), and one per milliliter). The initial antibiotic regimen provided because of agitation and flushing. One patient in the adequate microbiologic coverage in 116 of the 120 pa- placebo group was withdrawn because of suspected tients (97 percent) with positive cerebrospinal fluid cerebral abscess. Overall, treatment with dexameth- cultures in the dexamethasone group and in 112 of asone did not result in an increased risk of adverse the 114 (98 percent) in the placebo group.
events (Table 5). In one patient in the dexamethasone DISCUSSION
group, gastrointestinal bleeding was complicated bystomach perforation, which required surgery.
The results of our controlled prospective trial show that early treatment with dexamethasone improves the Clinical Course
outcome in adults with acute bacterial meningitis. Ad- Impairment of consciousness was significantly less junctive treatment with dexamethasone reduced the likely to develop in the patients who received dexa- risks of both an unfavorable outcome and death. Dex- methasone than in those who received placebo (18 amethasone did not have a beneficial effect on neuro- of 157 patients [11 percent] vs. 36 of 144 [25 per- logic sequelae, including hearing loss. However, neu- cent], P=0.002). The patients in the dexamethasone rologic sequelae were seen predominantly in the most group were also significantly less likely to have seizures severely ill patients, and the proportion of severely ill (8 [5 percent] vs. 17 [12 percent], P=0.04) and car- patients who survived to be tested was larger in the diorespiratory failure (16 [10 percent] vs. 29 [20 per- dexamethasone group than in the placebo group.
The beneficial effect of dexamethasone was most TABLE 4. UNFAVORABLE OUTCOME AT EIGHT WEEKS ACCORDING TO THE SCORE
COMA SCORE AND CULTURE
DEXAMETHASONE
RELATIVE RISK (95% CI)
*Higher scores indicate a better level of consciousness. CI denotes confidence interval.
1554 · N Engl J Med, Vol. 347, No. 20 · November 14, 2002 · www.nejm.org
Downloaded from www.nejm.org by MATTHEW N. GRABER MD on March 24, 2007 . Copyright 2002 Massachusetts Medical Society. All rights reserved. D EX A M ET H AS O N E I N BAC T E R I A L M E N I N G I T I S
bar puncture in all patients with suspected meningitis TABLE 5. ADVERSE EVENTS.
Two important issues are the duration and timing DEXAMETHASONE
of dexamethasone therapy. Although data suggest PLACEBO GROUP
that two-day and four-day regimens are equally effec- tive,20,21 the four-day regimen has been used in most clinical trials involving children with bacterial men-ingitis.6 Dexamethasone has been shown to have a beneficial effect in children with pneumococcal men- ingitis only if it is given before or with the first dose of antibiotics.6 In our study, we used the four-day reg- imen and also started it early. Therefore, a four-day regimen is recommended, with dexamethasone thera-py started before or with the first dose of antibiotics.
Most patients initially received monotherapy with amoxicillin. In the first half of the study, amoxicillinwas standard treatment in all patients. Rates of antibi- apparent in the patients with pneumococcal meningi- otic resistance among meningococcal and pneumo- tis. However, a beneficial effect in the patients with coccal isolates were very low. Similar rates were found meningococcal meningitis cannot be ruled out, since in nationwide studies in the Netherlands.10,11 In a pro- the number of patients in this subgroup was small.
spective audit of empirical therapy in adults with bac- Therefore, we recommend dexamethasone treatment terial meningitis in the Netherlands, monotherapy for all patients with acute bacterial meningitis.
with amoxicillin or penicillin appeared to be prescribed The possibility of selection bias was a matter of most frequently.11 Although dexamethasone is not concern in the study. To control for selection bias, we associated with adverse events, concern has been ex- compared the base-line characteristics of patients en- pressed that because the drug reduces blood–brain rolled in the study with prospective data from our permeability, it may impede the penetration of vanco- nationwide cohort of 634 adults with acute bacterial mycin into the subarachnoid space.9,22 With the world- meningitis. Patients in that cohort, for whom data wide increase in the prevalence of penicillin-resistant were collected in the period from 1998 to 2002, were pneumococci, combination therapy that includes van- not included in the present study. There were no sig- comycin has become more important.8 In children nificant differences between the two groups with re- with bacterial meningitis, treatment with dexameth- spect to the score on the Glasgow Coma Scale on ad- asone did not reduce vancomycin levels in cerebro- mission. Furthermore, mortality rates among patients spinal fluid.23 However, treatment failures have been in the placebo group in this study and the nationwide reported in adults who received standard doses of van- cohort were similar. Therefore, we conclude that se- comycin and adjunctive dexamethasone.24 Therefore, lection bias did not confound the results.
patients with pneumococcal meningitis who are treat- A delay in initiating antibiotic therapy was also a ed with vancomycin and dexamethasone should be matter of concern. Informed-consent procedures can carefully observed throughout therapy.
delay the initiation of antimicrobial therapy, which Cognitive impairment occurs frequently in adults may lead to a poor outcome.17 In addition, cranial CT who survive bacterial meningitis.25 Because cortico- should be performed before lumbar puncture in or- steroids may potentiate ischemic injury to neurons, it der to rule out brain shift in patients with coma, pap- is important to know whether dexamethasone prevents illedema, or hemiparesis in whom meningitis is sus- death but worsens cerebral cortical functioning.26 Al- pected. Lumbar puncture increases the risk of brain though in our study the reduction in mortality among herniation if an intracranial mass is present.18,19 In this the patients treated with dexamethasone did not re- setting, empirical antibiotic therapy should be started sult in an increased rate of neurologic sequelae, a cog- before cranial CT is performed.18,19 In our study, treat- nitive evaluation of adults treated with dexamethasone ment may have been delayed in patients who under- and those treated without it is needed.
went cranial CT before lumbar puncture, because the The results of our study show that adjunctive dex- study medication was administered before or with the amethasone therapy improves the outcome in adults first dose of antibiotics and the inclusion of patients with acute bacterial meningitis. Dexamethasone (10 mg in the study depended on the presence of cerebrospi- every six hours for four days) should be given to all nal fluid abnormalities. Since early therapy reduces such adults, and the regimen should be initiated be- morbidity and mortality,17 treatment with dexameth- fore or with the first dose of antibiotics. This treatment asone and antibiotics should be initiated before lum- does not increase the risk of gastrointestinal bleeding.
N Engl J Med, Vol. 347, No. 20 · November 14, 2002 · www.nejm.org · 1555
Downloaded from www.nejm.org by MATTHEW N. GRABER MD on March 24, 2007 . Copyright 2002 Massachusetts Medical Society. All rights reserved. The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne Supported in part by a grant from NV Organon, which also supplied the ebrospinal fluid outflow resistance in rabbits with experimental meningitis: alterations with penicillin and methylprednisolone. J Clin Invest 1980;66:243-53.
We are indebted to C.J.J.M. Schouten for her help with the data 5. Tauber MG, Khayam-Bashi H, Sande MA. Effects of ampicillin and cor-
management and to P. McIntyre, Ph.D., and R. Booy, Ph.D., for ticosteroids on brain water content, cerebrospinal fluid pressure, and cere-brospinal fluid lactate levels in experimental pneumococcal meningitis. J Infect Dis 1985;151:528-34.
6. McIntyre PB, Berkey CS, King SM, et al. Dexamethasone as adjunctive
APPENDIX
therapy in bacterial meningitis: a meta-analysis of randomized clinical trials The following centers and investigators participated in the European Dexamethasone in Adulthood Bacterial Meningitis Study: the Netherlands
7. Girgis NI, Farid Z, Mikhail IA, Farrag I, Sultan Y, Kilpatrick ME. Dex-
— Academisch Medisch Centrum, Amsterdam: J. de Gans, D. van de Beek, amethasone treatment for bacterial meningitis in children and adults. Pe- R.H. Enting; Medisch Centrum Alkmaar, Alkmaar: R. ten Houten; Flevo- ziekenhuis, Almere: G.N. Mallo; Bovenij Ziekenhuis, Amsterdam: P.M.S.
8. Quagliarello VJ, Scheld WM. Treatment of bacterial meningitis. N Engl
Gerkens; Sint Lucas Andreas Ziekenhuis, Amsterdam: J. Vos, J.A.L. Vanneste; Onze Lieve Vrouwe Gasthuis, Amsterdam: H.K. van Walbeek; Slotervaartz- 9. Saez-Llorens X, McCracken GH Jr. Antimicrobial and anti-inflammato-
iekenhuis, Amsterdam: J.J. van der Sande; Gelre Ziekenhuizen, Apeldoorn: ry treatment of bacterial meningitis. Infect Dis Clin North Am 1999;13: R.B. van Leeuwen; Ziekenhuis Rijnstate, Arnhem: Q.H. Leyten; Stichting Ziekenhuisvoorzieningen Gelderse Vallei, Ede: M.G. Smits; Ziekenhuis 10. Enting RH, Spanjaard L, van de Beek D, Hensen EF, de Gans J,
Leijenburg, Den Haag: R.W.M. Keunen, J. Blankevoort; Medisch Centrum Dankert J. Antimicrobial susceptibility of Haemophilus influenzae, Neisseria Haaglanden, Den Haag: W.V.M. Perquin, P. Bienfait; Ziekenhuis Bronovo, meningitidis and Streptococcus pneumoniae isolates causing meningitis in the Den Haag: P.C.L.A. Lambregts; Albert Schweitzer Ziekenhuis, Dordrecht: Netherlands, 1993-1994. J Antimicrob Chemother 1996;38:777-86.
L.I. Hertzberger, Ziekenhuis Nij Smellinghe, Drachten: J.A. Hilbers, H.L.
11. van de Beek D, de Gans J, Spanjaard L, Vermeulen M, Dankert J. An-
van der Wiel; Catharina Ziekenhuis, Eindhoven: J.N. Berendes; Diaconessen- tibiotic guidelines and antibiotic use in adult bacterial meningitis in the huis, Eindhoven: A.J. Vermeij; Medisch Spectrum Twente, Enschede: G.
Netherlands. J Antimicrob Chemother 2002;49:661-6.
Hageman; Oosterscheldeziekenhuizen, Goes: A.M. Boon; Beatrixzieken- 12. Jennett B, Teasdale G. Management of head injuries. Vol. 20 of Con-
huis, Gorinchem: R.B. Alting van Geusau; Academisch Ziekenhuis Gro- temporary neurology. Philadelphia: F.A. Davis, 1981.
ningen, Groningen: A. Bollen, H.J.G. Dieks, A.E.J. de Jager; Atrium Medisch 13. Wilson JTL, Pettigrew LEI, Teasdale GM. Structured interviews for
Centrum, Heerlen: M.J. Wennekes; Westfries Gasthuis, Hoorn: F.E.A.M.
the Glasgow Outcome Scale and the extended Glasgow Outcome Scale: Bussemaker; Atrium Medisch Centrum, Kerkrade: A.J.H. van Diepen; Di- guidelines for their use. J Neurotrauma 1998;15:573-85.
aconessenhuis, Leiden: P.E. Briët; Rijnland Ziekenhuis, Leiderdorp: R.J.W.
14. Wade DT. Measurement in neurological rehabilitation. Oxford, Eng-
Witteveen; Medisch Centrum Haaglanden, Leidschendam: R.J. Groen; IJs- land: Oxford University Press, 1992.
selmeerziekenhuizen, Lelystad: J.P. Geervliet; Ziekenhuis Canisius–Wilhelm- 15. Gardner MJ, Altman DG. Confidence intervals rather than P values:
ina, Nijmegen: C.W.G.M. Frenken; Academisch Ziekenhuis Nijmegen Sint estimation rather than hypothesis testing. Br Med J (Clin Res Ed) 1986; Radboud, Nijmegen: P.E. Vos, A.J.M. Keyser; Amphia Ziekenhuis, Oost- erhout: A.H. Temmink; Medisch Centrum Rijnmond-Zuid, Rotterdam: 16. Spanos A, Harrell FE Jr, Durack DT. Differential diagnosis of acute
C.A. van Donselaar; Academisch Ziekenhuis Rotterdam Dijkzigt, Rotter- meningitis: an analysis of the predictive value of initial observations. JAMA dam: D. Hasan; Sint Elisabeth Ziekenhuis, Tilburg: C.C. Tijssen; Universitair Medisch Centrum Utrecht, Utrecht: A. Elderson, G. van Dijk; Mesos Me- 17. Aronin SI, Peduzzi P, Quagliarello VJ. Community-acquired bacterial
disch Centrum, Utrecht: R.P.M. Bruyn; Sint Joseph Ziekenhuis, Veldhoven: meningitis: risk stratification for adverse clinical outcome and effect of an- B.J. van Kasteren; Vlietland Ziekenhuis, Vlaardingen: J.J.M. Driesen; Reinier tibiotic timing. Ann Intern Med 1998;129:862-9.
de Graaf Groep, Voorburg: J.L. van Doorn; Sint Lucas Ziekenhuis, Winscho- 18. van Crevel H, Hijdra A, de Gans J. Lumbar puncture and the risk of
ten: M.C. Wittebol; Hofpoort Ziekenhuis, Woerden: R. Wielaard, E.J. Wier- herniation: when should we first perform CT? J Neurol 2002;249:129-37.
inga; Kennemer Gasthuis, IJmuiden: J.A. Don; Gelre Ziekenhuizen, Zut- 19. Hasbun R , Abrahams J, Jekel J, Quagliarello VJ. Computed tomogra-
phen: H.J.D. de Zwart; Isala Klinieken, Zwolle: P.L.J.M. Bos; Belgium —
phy of the head before lumbar puncture in adults with suspected meningi- Algemeen Ziekenhuis Sint-Jan, Brugge: M. D’Hooghe; Sint Blasius Ziek- enhuis, Dendermonde: E. Van Buggenhout; Universitair Ziekenhuis Leu- 20. Syrogiannopoulos GA, Lourida AN, Theodoridou MC, et al. Dexa-
ven, Leuven: A. Govaerts; Algemeen Ziekenhuis Middelheim, Antwerp: R.
methasone therapy for bacterial meningitis in children: 2- versus 4-day reg- Crols; Germany: Städtisches Klinikum St. Georg, Leipzig: B.R. Ruf, S.
Fischer, T. Grünewald; Universitäts-Krankenhaus Eppendorf, Hamburg: K.
21. Schaad UB, Lips U, Gnehm HE, Blumberg A, Heinzer I, Wedgwood
Kunze, H.C. Hansen; Denmark — Odense Universitetshospital, Odense:
J. Dexamethasone therapy for bacterial meningitis in children. Lancet S. Stenvang Pedersen; Austria — Universitätsklinik für Neurologie, Inns-
brück: E. Schmutzhard, H.K. Spiss. Steering Committee: J. Dankert, J. de
22. Coyle PK. Glucocorticoids in central nervous system bacterial infec-
Gans, L. Spanjaard, P. Speelman, M. Vermeulen. Data Monitoring Com-
mittee: H. van Crevel (chairman), P. Speelman, J.G.P. Tijssen. Clinical Ep-
23. Klugman KP, Friedland IR , Bradley JS. Bactericidal activity against
idemiology and Biostatistics: R. de Haan. Writing Committee: J. de Gans,
cephalosporin-resistant Streptococcus pneumoniae in cerebrospinal fluid of children with acute bacterial meningitis. Antimicrob Agents Chemother 1995;39:1988-92.
REFERENCES
24. Viladrich PF, Gudiol F, Linares J, et al. Evaluation of vancomycin for
therapy of adult pneumococcal meningitis. Antimicrob Agents Chemother
1. Bohr V, Hansen B, Jessen O, et al. Eight hundred and seventy-five cases
of bacterial meningitis. I. Clinical data, prognosis, and the role of special- 25. van de Beek D, Schmand B, de Gans J, et al. Cognitive impairment in
ized hospital departments. J Infect 1983;7:21-30.
adults with good recovery after bacterial meningitis. J Infect Dis 2002;186: 2. Durand ML, Calderwood SB, Weber DJ, et al. Acute bacterial menin-
gitis in adults: a review of 493 episodes. N Engl J Med 1993;328:21-8.
26. Sapolsky RM, Pulsinelli WA. Glucocorticoids potentiate ischemic in-
3. Quagliarello V, Scheld WM. Bacterial meningitis: pathogenesis, patho-
jury to neurons: therapeutic implications. Science 1985;229:1397-400.
physiology, and progress. N Engl J Med 1992;327:864-72.
4. Scheld WM, Dacey RG, Winn HR , Welsh JE, Jane JA, Sande MA. Cer-
Copyright 2002 Massachusetts Medical Society.
1556 · N Engl J Med, Vol. 347, No. 20 · November 14, 2002 · www.nejm.org
Downloaded from www.nejm.org by MATTHEW N. GRABER MD on March 24, 2007 . Copyright 2002 Massachusetts Medical Society. All rights reserved.

Source: http://www.ttuem.com/page2/files/Dex%20and%20meningitis.pdf