Dexamethasone treatment for bacterial meningitis

In the pre-antibiotic era more than 9096 of children with bacterial meningitis died from their disease. The introduction of antibiotics and intensive care reduced mortality to approximately 5%, but significant morbidity remained. A prospective study of 50 children who recovered from Haemophilus influenzae type b meningitis reported that 28% had significant physical or intellectual disabilities 3 years after their illness.¹ The failure of new and more powerful antibiotics to improve the outcome of meningitis meant that other treatment modalities needed to be evaluated.     

Dexamethasone treatment for bacterial meningitis in children and adults

Four hundred twenty-nine patients with bacterial meningitis were assigned on a nonselective alternating basis into one of two therapeutic regimens. Patients in Group I received dexamethasone in addition to standard antibacterial chemotherapy of ampicillin and chloramphenicol whereas those in Group II received antibacterial chemotherapy alone. Dexamethasone was given intramuscularly (8 mg to children younger than 12 years and 12 mg to adults every 12 hours for 3 days). Both treatment groups were comparable with regard to age, sex, duration of symptoms and state of consciousness at the time of hospitalization. A significant reduction in the case fatality rate (P less than 0.01) was observed in patients with pneumococcal meningitis receiving dexamethasone; only 7 of 52 patients died compared with 22 of 54 patients not receiving dexamethasone. A reduction in the overall neurologic sequelae (hearing impairment and paresis) was observed in patients receiving dexamethasone. This reduction was significant only in patients with Streptococcus pneumoniae meningitis; none of the 45 surviving patients receiving steroids had hearing loss whereas 4 of 32 patients not receiving dexamethasone had severe hearing loss (P less than 0.05). No significant difference was observed between the two groups with regard to time for patients to become afebrile or to regain consciousness or in the mean admission and 24- to 36-hour cerebrospinal fluid leukocyte count, glucose or protein content.     

Dexamethasone as an adjunctive treatment of bacterial meningitis

This study was conducted on 77 Libyan infants and children aged month to 10 years with acute bacterial meningitis. Upon admission, the patients were divided randomly in two groups. Group I (38 patients) received ceftriaxone plus dexamethasone I.V. Group II (39 patients) received ceftriaxone alone. Both groups were compared for mean changes in CSF sugar, CSF protein and CSF polymorph count at 4th day of treatment. There was a significant difference between the two groups in CSF sugar and protein changes (P < 0.05) but not in CSF polymorph (P > 0.05). Both groups showed prompt clinical response and similar occurrence of acute complications, fatality rate and permanent neurological sequelae. However, group I manifested shorter duration of fever (P < 0.05). Dexamethasone improved the inflammatory reaction in acute bacterial meningitis and shortened the duration of fever but it did not have any significant effect on the fatality and the occurrence of neurological sequelae of this disease.     

Dexamethasone adjunctive treatment for tuberculous meningitis

During a 5-year period, 280 of 2010 patients admitted to the meningitis ward of a referral hospital in Cairo, Egypt, were clinically diagnosed as having tuberculous meningitis and were treated with either antituberculous chemotherapy and dexamethasone or antituberculous chemotherapy alone. Fatality rates and neurologic sequelae were compared for the 2 treatment groups in the 160 patients who had cerebrospinal fluid cultures positive for Mycobacterium tuberculosis. The overall mortality rate of 51% reflects the delay in receiving appropriate therapy (79% with symptoms for more than 2 weeks) and the severity of illness on admission (56% in coma, 39% drowsy). The fatality rate was significantly lower in the group receiving dexamethasone (43% vs. 59%, P less than 0.05), particularly in the drowsy patients (15% vs. 40% P less than 0.04), and in patients surviving long enough to receive at least 10 days of treatment (14% vs. 33%, P less than 0.02). Development of neurologic complications after initiation of therapy (4 vs. 10) and permanent sequelae (6 vs. 13) were significantly lower in the dexamethasone-treated group (P less than 0.02).     

Neonatal bacterial meningitis

Despite major improvements in infant intensive care, neonatal meningitis remains a devastating disease. Survivors of bacterial meningitis are at high-risk for life-long neurological handicaps, and despite a reduction in mortality, the morbidity of neonatal meningitis has not changed substantially over the last thirty years. A substantial improvement in outcome is unlikely to result from further refinements in ICU technology or new antibiotics. However, recent advancements in our understanding of the pathogenesis of meningitis and the pathophysiology of brain injury in meningitis may provide the opportunity to interrupt the mechanisms that allow bacteria to enter the central nervous system and initiate the inflammatory response. Strategies aimed at modulating the inflammatory response must be chosen carefully, so as not to disrupt normal host responses needed for the infant to recover from the infectious episode.     

Recurrent bacterial meningitis

To characterize recurrent bacterial meningitis in children, we reviewed the charts of all patients treated for more than one episode of bacterial meningitis at the Würzburg University Children's Hospital from 1980 to June 1995. Twenty-five children suffered 2–13 episodes of bacterial meningitis. Most patients were referred from other hospitals to our paediatric neurosurgical service. No immunodeficiency was found. In all patients, the cause of recurrent meningitis was an anatomical lesion with 13 intracranial defects including encephaloceles, skull fractures, Mondini dysplasias, neurenteric cyst, fibrous dysplasia, persistent craniopharyngeal duct, and 12 lumbosacral defects with a dermoid cyst within the lumbosacral spine. A first episode of meningitis at school age did not exclude a congenital defect. In total, 84 episodes of meningitis were treated, a pathogen was isolated in 77%. The most common pathogen wasStreptococcus pneumoniae, followed byEscherichia coli, Staphylococci and others. The pathogen isolated often gave a clue to the location of the defect. Personal history was often unrewarding and in some cases the search for the anatomical lesion required repeated imaging and explorative surgery. In 24 of 25 cases, final treatment of recurrent meningitis was by surgical intervention.     

Changes in bacterial meningitis.

In 1964, one of us (WHG) undertook a retrospective study of bacterial meningitis in childhood in the north east of Scotland during the period 1946-61. We have recently carried out a similar review of cases occurring during 1971-86, to compare the incidence, mortality, and bacteriological patterns. During the earlier period 285 cases occurred, a total incidence of 16.9/100,000 children per year. In the later period 274 children were affected, an annual incidence of 17.8/100,000. The overall mortality rate fell dramatically from 11.9% to 1.8%, the latter figure comparing favourably with recent published studies from Scandinavia and the United States. There was a change in the bacteriological profile in the second period with a significant rise in cases due to Haemophilus influenzae at all ages. A fall in cases of meningococcal meningitis was significant in infants under 1 year of age only. Possible reasons for the change in the bacteriological pattern are discussed.     

Movement disorders in bacterial meningitis

Movement disorders developed in five children, ages 6 to 21 months, during the course of bacterial meningitis caused by Hemophilus influenzae (one), Streptococcus pneumoniae (one), Neisseria meningitidis (one), or Mycobacterium tuberculosis (two). Athetosis, choreoathetosis, and hemiballismus occurred, ranging in duration from hours to months. Cranial computed tomography, performed in four cases, showed no lesion of the basal ganglia. The movements were of such abrupt onset and severity that in four cases they were initially misinterpreted as seizures, and anticonvulsant therapy was contemplated. It is important to recognize the potential development of movement disorders during the acute phase of bacterial meningitis to preclude the inappropriate administration of anticonvulsant medication.     

Cefuroxime in bacterial meningitis.

In order to find an alternative antimicrobial treatment for childhood bacterial meningitis 30 infants and children with meningitis, due to Haemophilus influenzae (n = 13), Neisseria meningitis (n = 9), Streptococcus pneumoniae (n = 5), or meningitis of unknown aetiology (n = 3), were treated with cefuroxime, 200 mg/kg a day, as the only antibiotic. Prompt clinical and bacteriological responses were noted and every patient was cured. Cefuroxime concentrations in cerebrospinal fluid ranged from 1.1 to 18.8 (mean 7.0) mg/l at the beginning and from 0.5 to 4.1 (mean 1.6) mg/l at the end of treatment. Three infants developed symptomatic sterile subdural effusions which were managed by repeated subdural aspirations while still on antibiotics. Cefuroxime concentrations in the subdural fluid ranged from 17.4 to 32.4 mg/l. At follow-up 2 patients had moderate unilateral hearing loss and one had mild ataxia. We conclude that cefuroxime is effective and safe for the treatment of childhood bacterial meningitis caused by any of these common organisms.     

C-reactive protein and bacterial meningitis

We have studied prospectively the C-reactive protein values in the cerebrospinal fluid of 54 patients with bacterial meningitis, tuberculous meningitis, and severe malarial infection and convulsions without infections of the central nervous system. CSF CRP above 1 mg/l was observed in 23 out of 28 patients with bacterial meningitis (sensitivity of 82%). The specificity was 73% at the 1 mg/l level. Five out of 19 patients with severe malarial infection had CSF CRP levels above 1 mg/l. Two patients with TB meningitis were also studied. Both of them had CSF CRP above 1 mg/l. Five patients with febrile convulsions or sepsis without meningitis had CSF CRP below 1 mg/l. It is concluded that CSF CRP would not be used as a useful discriminatory test in areas where malaria and TB meningitis are common.