Q J Med 1999; 92: 719-725
© 1999 Association of Physicians
Acute bacterial meningitis in adults: a hospital-based epidemiological study
From the Department of Neurology, Chang Gung Memorial Hospital and Medical College, Chang Gung University, Taipei, Taiwan
Received 23 June 1999 and in revised form 29 September 1999
Dr L.-M. Tang, Department of Neurology, Chang Gung Memorial Hospital, 199 Tung Hwa North Road, Taipei, Taiwan
 |
Summary |
|---|
 Top Summary IntroductionMethodsData analysisResultsDiscussionReferences |
|---|
Bacterial meningitis, a world-wide disease, has to be reviewed periodically because the specific micro-organisms responsible for the infection vary with time, geography and patient age. To determine its incidence and clinical features in Taiwan, we reviewed the microbiological records for cerebrospinal fluid (CSF) and blood cultures, and the medical records of patients with bacterial meningitis admitted between 1981 and mid-1998. Bacterial micro-organisms were demonstrated in the CSF and/or blood in 395 patients with 418 episodes of bacterial meningitis. Streptococcus species were the most common causative micro-organism group, at 23.21% of all episodes. Its prevalence rate significantly decreased from the first 7 years of study (41.9%) to the last 10.5 years (19.2%). However, Klebsiella meningitis and Staphylococcal meningitis were more frequently noted after 1987. More than 70% of patients had at least one underlying disease or condition. Poor prognostic factors indicated by univariable analysis were: age >60 years; diabetes mellitus; severe neurological deficits on the first day of treatment; infection with Gram-negative bacilli; CSF WBC count >5000x106/l; malignancy; seizure; and bacteraemia. The overall mortality rate was 29.4%, 29.7% in the first 7 years of study and 29.4% in the last 10.5 years. The use of new antibiotics has not reduced the mortality rate in our patients with bacterial meningitis.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsData analysisResultsDiscussionReferences |
|---|
Most epidemiological data on bacterial meningitis come from the developed countries, in which the four most common infecting organisms are Streptococcus pneumoniae, Neisseria meningitidis, Listeria monocytogenes, and Haemophilus influenzae.1–6 The frequency of Gram-negative bacilli other than H. influenzae has increased over time. In the developing countries, these micro-organisms are also common, but the relative frequencies are unknown.7
Acute bacterial meningitis remains a major cause of death and long-term neurologic sequelae world-wide, and needs to be reviewed periodically, as the specific micro-organisms responsible for the infection vary with time, geography and patient age. In a recent Taiwan study, N. meningitidis was no longer a common organism responsible for bacterial meningitis in infants and children, accounting for only two of 41 episodes of meningitis.8 In adults, Gram-negative bacilli other than H. influenzae have become significant in both spontaneous meningitis and post-neurosurgical meningitis.9,10 In our study on Klebsiella meningitis, the species was important in both community-acquired meningitis and nosocomial meningitis in children and adults.11
There are a few recent studies on acute bacterial meningitis in adults, mainly by investigators from the US and Europe.6,12–14 In Taiwan, the information available on the epidemiology of the infection in adults is scanty. In this study, we reviewed our experience in adult patients with acute bacterial meningitis at Chang Gung Memorial Hospital over a 17.5-year period, from 1981 through mid-1998. We aimed to determine the relative frequency of each pathogen in adult patients with the infection, to evaluate whether there is a changing pattern over time, to identify the prognostic factors in adult meningitis, and to examine whether the mortality rate has changed in recent years.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsData analysisResultsDiscussionReferences |
|---|
We reviewed the microbiological records of cerebrospinal fluid (CSF) and blood cultures and the medical records of all patients 15 years of age or older in whom acute bacterial meningitis was diagnosed at Chang Gung Memorial Hospital between 1981 and mid-1998. This hospital is a 3600-bed teaching medical centre in the northern part of Taiwan, and a referral center for patients from other parts of the country. Criteria of bacterial meningitis for inclusion were (i) demonstration of a specific bacterial pathogen in CSF culture, with history of acute disease and clinical findings of meningeal inflammation, or (ii) demonstration of a specific bacterial pathogen in blood culture, history of acute disease, presence of clinical findings such as headache, fever and nuchal rigidity, and typical CSF features including decreased glucose level, increased protein concentration and pleocytosis with predominant polymorphonuclear cells.
Data on sex, age, duration of illness, clinical manifestations, associated diseases, head injury or neurosurgical procedures, laboratory findings, use of antibiotics, and outcome were recorded and comparison made between non-fatal and fatal cases. Meningitis was defined as nosocomial according to the 1988 guidelines of the Centers for Disease Control;15 the infection is acquired in the hospital and the infection itself is not a complication or extension of infection already present on admission, unless a change of pathogen suggests the acquisition of a new infection. The clinical severity of the illness on admission was classified into two stages: stage I, patients with mildly or non-altered consciousness; and stage II, patients in a state of delirium or coma. Initial antibiotic therapy was considered appropriate when (a) the antibiotic administered on the day of admission or diagnosis was demonstrated to be effective against the pathogen by sensitivity testing and (b) the antibiotic used was capable of reaching the central nervous system in an adequate amount. Reappearance of symptoms and signs of bacterial meningitis was considered as relapse if it resulted from persistence of the initial infection and occurred within 4 weeks of the completion of antibiotic therapy for the initial episode, or recurrence if it was caused by reinfection with the same or different bacteria but occurred more than 4 weeks after the completion of therapy for the initial episode.16
|
Data analysis |
|---|
|
TopSummaryIntroductionMethodsData analysisResultsDiscussionReferences |
|---|
All statistical analyses were performed on an IBM-PC compatible computer, using the Stata software package.17 Comparison between groups was made by means of the unpaired Student's t test for normally distributed continuous variables, the Mann-Whitney U test for non-normally distributed continuous variables, and the Pearson
2 or Fisher exact test for nominal variables. All p values were two-sided. The level of significance was 0.05.
For the predictors of mortality, univariable analysis was calculated by means of Pearson 2 test. Interaction was initially assessed in the stratification analysis with Mantel-Haenszel test for heterogeneity. To select a model predicting mortality, variables or interaction sets were sequentially deleted from a full model of multivariable logistic regression by using likelihood ratio test until all remaining variables had a significance level of <0.2.
|
Results |
|---|
|
TopSummaryIntroductionMethodsData analysisResultsDiscussionReferences |
|---|
Bacterial micro-organisms were demonstrated in the CSF and/or blood in 395 patients with 418 episodes of bacterial meningitis. There were 74 episodes diagnosed between 1981 and 1987, and 344 episodes between 1988 and mid-1998. In these two periods, the proportion of the infection was significantly different with respect to other neurological diseases (6.49 vs. 9.9 per 1000 patients admitted, p=0.0067) (Figure 1
) although there has been no change in the referral practice of patients sent to us. There were 274 (65.6%) men and 144 (34.5%) women, aged between 15 and 90 years (mean±SD 45.5±17.2 years). Of the 395 patients, 376 had a single episode of infection, whereas 16 had two episodes, two had three episodes, and one had four episodes of meningitis. This patient, who had Evan's syndrome and ventriculoperitoneal shunt for hydrocephalus, had relapsing episodes of meningitis due to Salmonella serogroup B, and has been previously reported;18 her fourth episode, 3 years from the last one, was caused by coagulase-positive Staphylococcus.
|
Nearly half of the episodes of bacterial meningitis were caused by Gram-negative bacilli, of which K. pneumoniae was the commonest (Table 1
). There were more cases of meningitis caused by Klebsiella and Staphylococcus species with respect to the total number of culture-proven bacterial meningitis cases after 1987. Of the 88 episodes of Klebsiella meningitis, only 13 occurred between 1981 and 1987, accounting for 17.6% of 74 cases. In the following years (1988 to mid-1998), there were 75 episodes, 21.8% of 344 bacteriologically proven cases (p=0.418, Pearson 2 test). Of the 75 episodes of Staphylococcal meningitis, 41 (54.7%) were caused by coagulase-positive Staphylococcus and 34 (45.3%) by coagulase-negative Staphylococcus. Only 8 (10.8%) of the 75 episodes were noted in the first 7 years of study, and there were 67 (19.5%) in the last 10.5 years (p=0.078). Overall, Streptococcus species composed the most common pathogen group, causing 97 episodes of meningitis: S. pneumoniae (71.1%), Viridans streptococci (18.5%), group B streptococcus (5.2%) and ß-haemolytic groups other than A, B, and D (5.2%) were isolated. Its prevalence rate significantly decreased from the first 7 years of study (41.9%) to the last 10.5 years (19.2%) (p<0.001). There were 155 episodes of nosocomial meningitis of which Gram-negative bacilli and Staphylococcus species were responsible for 91 (58.7%) episodes and 39 (25.2%) episodes respectively.
|
Of the 395 patients, 323 (71.8%) had at least one underlying disease or condition. The major underlying conditions were diabetes mellitus (DM), head injury and/or neurosurgical procedure, and malignancy. All 13 patients who had mixed bacterial meningitis were associated with conditions such as head injury, neurosurgical procedures, malignancy and DM.
Seasonal variation was not noted in the 418 episodes of bacterial meningitis; the frequencies for total cases were 27.8% from March to May, 25.4% from June to August, 21.8% from September to November, and 25.1% from December to February (p=0.673, goodness-of-fit 2 test).
On the day of admission or diagnosis, 89.5% of patients had a history of headaches, fever and/or neck stiffness. Disturbance of consciousness was noted initially in 191 (45.7%) of patients and seizure in 60 (14.4%). A CSF study was performed on every patient (Table 2). The CSF findings were incomplete in some cases. A difference was observed regarding WBC count, protein and lactate between the fatal and non-fatal groups; however, there was a large overlap.
|
Of the 395 patients, 123 died. The overall case-fatality rate was 29.4%; 29.7% during the first 7 years of study and 29.4% in the last 10.5 years. In the first 7 years of study, 22 patients died; initial antibiotic therapy was appropriate in 14 patients and inappropriate in 8, despite subsequent use of appropriate therapy. Appropriate therapy included chloramphenicol alone or in combination with gentamicin in 4/14 patients, third-generation cephalosporins such as cefotaxime, moxalactam or ceftriaxone in six and penicillin or oxacillin in four. Inappropriate therapy included use of chloramphenicol in 5/8 patients when a Gram-negative bacillus was subsequently shown to be resistant to the drug, use of penicillin and chloramphenicol in two patients with Staphylococcus species and an inadequate dose of penicillin in one patient with S. pneumoniae. Of the 101 patients who died in the last 10.5 years, initial antibiotic therapy was appropriate in 75 patients, of whom seven received chloramphenicol alone or in combination with gentamicin, 25 had penicillin, oxacillin, vancomycin alone or in combination with third-generation cephalosporins, and 43 had third-generation cephalosporins. Of the 26 patients who had inappropriate initial antibiotics, nine had chloramphenicol with or without gentamicin, nine had moxalactam, ceftizoxime or ceftriaxone, and seven had penicillin or oxacillin.
A higher mortality was observed in the elderly (Table 3); the mean age (53.9±15.7 years) of the fatal group was higher than that (42.0±16.6 years) of the non-fatal group (p<0.001). Predictors of outcome also included DM (p<0.001), seizure (p=0.012), infection with Gram-negative bacilli (p=0.002), bacteraemia (p=0.027), malignancy (p=0.024), CSF WBC count >5000x106/l (p=0.021) and clinical stage on the first day of treatment (p<0.001). The mortality rate was not significantly different between patients with community-acquired meningitis and patients with nosocomial meningitis (case fatality rates 70/263 vs. 53/155, p=0.101).
|
In the multivariate analysis (Table 4
), elderly patients (
60 years) with CSF sugar level <2.5 mmol/l had higher mortality; conversely, the association was reversed in younger patients, but did not reach the significance level of 0.05. In addition, advanced clinical stage predicted higher mortality only in men, not in women. Patients with bacteraemia had poorer prognosis in community-acquired infection than those without bacteraemia; the association was not observed in nosocomial infection. Similarly, bacteraemia was associated with a higher mortality in the patients with malignancy than in those without malignancy. In predicting model selection using multivariable logistic regression, the full model consisted of all variables and their interaction sets in Tables 3 and 4
. Neurosurgical procedure, alcoholism, jaundice, seizure, infection with Gram-negative bacilli and source of infection and its interaction term were sequentially removed from the model. The final model was as shown in Table 5.
|
|
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsData analysisResultsDiscussionReferences |
|---|
Bacterial meningitis is more common in the very young and the elderly. In the US, bacterial meningitis has become predominantly a disease of adults rather than of infants and young children, because of the vaccine-related decline in H. influenzae disease.5 In Taiwan, life expectancy is increasing and therefore so will the elderly population. Until bacterial meningitis can be effectively prevented, we can expect this life-threatening infection to continue to cause medical difficulties. In this report, the proportion of bacterial meningitis with respect to other neurological diseases noted in our hospital significantly increased from the first 7 years of study to the last 10.5 years.
In bacterial meningitis, there is a slight male predominance. Men outnumbered women by a ratio of 6 : 5 in a German study of bacterial meningitis in adults.12 However, one Taiwan study of Gram-negative bacillary meningitis in adults found a ratio was 6 : 1.10 The reason for this high rate of occurrence in male Taiwanese is unknown. It may be that the study was performed in a military hospital which serves more men than women. Our study, in a tertiary case centre for the entire country, found a male predominance of 2 : 1.
Streptococcus species have been the major cause of bacterial meningitis in many countries.4–6 As a whole, it was the most common causative micro-organism group demonstrated in our patients. However, its prevalence rate significantly decreased from the first 7 years of study (42%) to the last 10.5 years (21.6%). More cases of Klebsiella meningitis and staphylococcal meningitis were noted after 1987. This indicates that in Taiwan the pathogenic micro-organisms causing bacterial meningitis have been changing.
More than 70% of our patients had at least one underlying disease or condition. Predisposing factors for acute bacterial meningitis include otitis media, sinusitis, pneumonia, endocarditis, head injury, neurosurgical procedures, immunodeficiency, DM, alcoholism and CSF leak.13 In Taiwan, neurosurgical procedures and head injuries are important conditions associated in patients with bacterial meningitis. This may be due to the fact that neurosurgical procedures are more frequently performed, because of the increasing number of neurosurgeons and the improvement of surgical facilities, and the large number of patients with head injuries from motorcycle accidents. The increasing incidence of staphylococcal meningitis, which is also the second most frequent nosocomial bacterial meningitis, reflects the increased incidence of head injuries and increased practice of neurosurgery. Our recent study showed that Klebsiella oxytoca meningitis, a rare infection of the central nervous system, is now frequently associated with neurosurgical procedures.11 Furthermore, we have seen Citrobacter meningitis19 and Klebsiella ozaenae meningitis11 in patients with neurosurgical procedures; these infections were previously unknown in Taiwan. Immunodeficiency is well known to occur in patients with malignancy. We have recently reported a study of bacterial meningitis in 12 patients with nasopharyngeal carcinoma. These accounted for 0.65% of 1850 patients with cancer which ranks sixth in incidence and cause of death among Taiwanese.20
While Staphylococcus species are a major pathogen of nosocomial bacterial meningitis, Klebsiella species caused only 19 episodes of nosocomial infection. These Klebsiella nosocomial infections accounted for only 4.54% of 418 episodes of culture-proven bacterial meningitis or 21.6% of 88 episodes of Klebsiella meningitis. Thus, nosocomial infection is unlikely to be the main reason for the increasing cases of Klebsiella meningitis. However, studies from both northern and southern parts of Taiwan have shown an increased rate of the infection.11 In our hospital, a medical centre in the northern part of Taiwan, the increase in absolute numbers of cases may be partly due to the increased number of patient beds. There has been no change in the referral practice regarding patients sent to us.
The incidences of recrudescence, recurrence and relapse of bacterial meningitis are low in both paediatric and adult patients. Neurosurgical procedures are a predisposing factor for the relapsing infection in adults.18 Our previous study on relapsing bacterial meningitis demonstrated that the infection is a rare condition, accounting for 3.4% of 234 adult patients with bacterial meningitis.18 In this study, reinfection, including recrudescence, recurrence and relapse of bacterial meningitis, occurred with a frequency of 5.8%. However, a recent study of meningitis from the US showed a high frequency (9%) of recurrent meningitis in adults.13
Previous studies, as well as our own, have shown that certain prognostic factors correlate with a poor outcome in patients with bacterial meningitis.11,13 In this study, a higher mortality was observed in the elderly. Predictors of outcome also included diabetes mellitus, severe neurological deficits on the first day of treatment, infection with Gram-negative bacilli, CSF WBC count >5000x106/l, malignancy, seizure, and bacteraemia.
With the use of new antibiotics in recent years, such as third-generation cephalosporins to replace chloramphenicol in cases of Gram-negative bacillary meningitis, the mortality rate of adult meningitis was 58% in one study from Taiwan.10 The overall mortality rate in our patients with culture-proven bacterial meningitis was high (29.4%), probably due to the fact that nearly half of the episodes of meningitis were caused by Gram-negative bacilli. It remained high, as 29.7% patients died in the first 7 years of study and 29.4% in the last 10.5 years, despite the frequent use of third-generation cephalosporins instead of chloramphenicol in the latter period. Thus, the use of more highly bactericidal antibiotics has not reduced the mortality rate in patients with bacterial meningitis. Other than new antibiotics, progress in the management of the pathophysiological mechanisms of bacterial meningitis may perhaps improve outcome.
|
References |
|---|
|
TopSummaryIntroductionMethodsData analysisResultsDiscussionReferences |
|---|
1. Luby JP. Southwestern Internal Medicine Conference: Infections of the central nervous system. Am J Med Sci 1992; 304:379–91.[Web of Science][Medline]
2. Schlech WF III. The epidemiology of bacterial meningitis. In: Schonfeld H, Helwig H, eds. Bacterial meningitis. Basel, Karger, 1992:5–17.
3.
Urwin G, Yuan MF, Feldman RA. Prospective study of bacterial meningitis in North East Thames region, 1991–3, during introduction of Haemophilus influenzae vaccine. Br Med J 1994; 309:1412–14.
4. Wenger JD, Broome CV. Bacterial meningitis: epidemiology. In: Lambert HP, ed. Infections of the central nervous system. Philadelphia, BC Decker, 1991:16–31.
5.
Schuchat A, Robinson K, Wenger JD, Harrison LH, Farley M, Reingold AL, Lefkowitz L, Perkins BA. Bacterial meningitis in the United States in 1995. N Engl J Med 1997; 337:970–6.
6.
Sigurdardottir B, Bjornsson OM, Jonsdottir KE, Erlendsdottir H, Gudmundsson S. Acute bacterial meningitis in adults. Arch Intern Med 1997; 157:425–30.
7. Bharucha NE, Bhabha SK, Bharucha EP. Bacterial infections. In: Bradley WG, Daroff RB, Fenichel GM, Marsden CD, eds. Neurology in clinical practice. London, Butterworth-Heinemann, 1991:1049–84.
8. Liu CC, Chen JS, Lin CH, Chen YJ, Huang CC. Bacterial meningitis in infants and children in southern Taiwan: emphasis on Haemophilus influenzae type b infection. J Formos Med Assoc 1993; 92:884–8.[Medline]
9. Hsu GJ, Young TG, Chou JW, Peng MY, Chang FY, Chou MY. Gram-negative bacillary meningitis in adults. J Formos Med Assoc 1993; 92:317–23.[Medline]
10. Jang TN, Wang FD, Wang LS, Yu KW, Liu CY. Gram-negative bacillary meningitis in adults: a recent six-year experience. J Formos Med Assoc 1993; 92:540–6.[Medline]
11. Tang LM, Chen ST, Hsu WC, Chen CM. Klebsiella meningitis in Taiwan: an overview. Epidemiol Infect 1997; 119:135–42.[Medline]
12.
Pfister HW, Feiden W, Einhaupl KM. Spectrum of complications during bacterial meningitis in adults: results of a prospective clinical study. Arch Neurol 1993; 50:575–81.
13.
Durand ML, Calderwood SB, Weber DJ, Miller SI, Southwick FS, Caviness VS, Jr, Swartz MN. Acute bacterial meningitis in adults: a review of 493 episodes. N Engl J Med 1993; 328:21–8.
14. Kragsbjerg P, Kallman J, Olcen P. Pneumococcal meningitis in adults. Scand J Infect Dis 1994; 26:659–66.[Web of Science][Medline]
15. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for nosocomial infections, 1988. Am J Infect Control 1988; 16:128–40 (Erratum, Am J Infect Control 1988; 16:177).[Web of Science][Medline]
16. Kline MW. Review of recurrent bacterial meningitis. Pediatr Infect Dis J 1989; 8:630–4.[Web of Science][Medline]
17. Computing Resource Center. Stata Reference Manual: Release 3, 5th edn. Santa Monica, 1992.
18.
Tang LM, Chen ST. Relapsing bacterial meningitis in adults. Q J Med 1994; 87:511–18.
19. Tang LM, Chen ST, Lui TN. Citrobacter meningitis in adults. Clin Neurol Neurosurg 1994; 96:52–7.[Web of Science][Medline]
20.
Tang LM, Chen ST, Ng SH. Bacterial meningitis in patients with nasopharyngeal carcinoma. Q J Med 1996; 89:71–6.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  Mei Feng, Ruopeng Sun, Chengmei Zhang, Enhua Sun, Shaochun Wei, Jiaqing Wan, and Ruopeng Sun Increased Expression of Tyrosine Phosphatase SHP2 in Experimental Pneumococcal Meningitis: Correlation With Tumor Necrosis Factor-Alpha and Cerebrospinal Fluid Pleocytosis J Child Neurol, March 1, 2008; 23(3): 287 - 292. [Abstract] [PDF]
|
|
|
|
 |
|
 P.-Y. Lee, W.-N. Chang, C.-H. Lu, M.-W. Lin, B.-C. Cheng, C.-C. Chien, C.-J. Chang, and H.-W. Chang Clinical features and in vitro antimicrobial susceptibilities of community-acquired Klebsiella pneumoniae meningitis in Taiwan J. Antimicrob. Chemother., April 1, 2003; 51(4): 957 - 962. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||