Bacterial Meningitis — A View of the Past 90 Years
http://www.100md.com
《新英格兰医药杂志》
The history of community-acquired bacterial meningitis arguably represents the best example of the salutary effect of the introduction of antimicrobial agents. Before the use of specific antiserums, the outlook for patients with bacterial meningitis was dismal (see Figure). In the 1920s, 77 of 78 children at Boston Children's Hospital who had Haemophilus influenzae meningitis died. The prognosis for untreated pneumococcal meningitis was equally bleak: of 300 patients, all died. In the first decade of the 20th century, untreated meningococcal meningitis was associated with a mortality rate of 75 to 80 percent.
Figure. Mortality Rates Associated with Community-Acquired Bacterial Meningitis over the Past 90 Years.
In 1913, Simon Flexner was the first to report some success in treating bacterial meningitis with intrathecal equine meningococcal antiserum: among 1300 patients with epidemic meningitis, mortality was reduced to 31 percent.1 Among 169 children with meningococcal meningitis treated with intrathecal antiserum at Bellevue Hospital, New York, between 1928 and 1936, the outcome was even more favorable, with mortality of about 20 percent. Fothergill reported in 1937 that treatment of H. influenzae meningitis with combined intravenous and intrathecal antiserums reduced mortality among 201 children to 85 percent. The prognosis for patients with pneumococcal meningitis remained extremely grave even after the introduction of specific antiserums. There were only anecdotal reports of recovery after treatment with systemic and intrathecal antipneumococcal serum.
In the 1930s, with the introduction of sulfonamides, the mortality associated with meningococcal meningitis decreased to 5 to 15 percent. By 1944, Alexander had reported that treatment with both a sulfonamide and intravenous rabbit antiserum in 87 children with H. influenzae type b meningitis had reduced mortality to 22 percent. In the early 1950s, chloramphenicol treatment (with sulfadiazine) reduced the fatality rate of H. influenzae meningitis to 5 to 10 percent and made the use of antiserum unnecessary. The results of sulfonamide treatment of pneumococcal meningitis were less favorable, with mortality ranging from 45 to 95 percent.
The use of penicillin therapy for pneumococcal meningitis began in the mid-1940s, with combined systemic and intrathecal administration, and resulted in a fatality rate of 49 percent. Modern treatment with "meningeal" doses without concomitant intrathecal drug administration began with the 1949 report by Dowling et al. of a study in which 21 patients with pneumococcal meningitis were treated with 1 million units of penicillin intramuscularly every two hours, with a resulting improvement in mortality to 38 percent.2
For the past 15 years, therapy for community-acquired bacterial meningitis has consisted of intravenous penicillin (or ampicillin), a third-generation cephalosporin, or both; mortality rates for meningococcal meningitis have hovered around 10 percent, those for H. influenzae meningitis have been reduced to less than 5 percent, and those for pneumococcal meningitis have remained around 20 percent. In this issue of the Journal, van de Beek and colleagues (pages 1849–1859) confirm these rates in a large series of patients in the Netherlands, with mortality of 7 percent associated with meningococcal disease and 30 percent associated with pneumococcal disease.
The demonstration that bacterial components within the cerebrospinal fluid could cause a release of inflammatory cytokines led to trials of dexamethasone as adjuvant therapy. In four trials in children with H. influenzae meningitis conducted in the early 1990s, dexamethasone use did not change mortality but did reduce the frequency of neurologic sequelae — principally, sensorineural hearing loss. In 2002, Dutch investigators showed that, in adults, adjunctive dexamethasone therapy reduced the rate of unfavorable outcomes from 25 to 15 percent; beneficial effects were most obvious in pneumococcal meningitis.
In the past 20 years, cranial computed tomography (CT) has been another major advance, allowing the differentiation of other types of pyogenic intracranial infection and detection of brain swelling and mass lesions. The downside of its availability has been the frequent delay in antimicrobial therapy while the physician awaits the results of the CT scan and lumbar puncture. In the study by van de Beek et al., CT was performed before lumbar puncture in 337 of 696 patients, and in two thirds of these patients, initial therapy was not started before the CT results had been obtained. In a study from Yale University involving 301 adults with suspected meningitis at an urban hospital emergency room, certain clinical features (e.g., prior central nervous system disease, abnormal sensorium, limb drift, aphasia, and abnormal visual fields) were associated with abnormalities on CT. Tellingly, among the 96 patients who had none of these findings, fully 97 percent of the scans were normal (a mild mass effect was found in only one case), emphasizing that clinical features can be used to identify patients who do not need to undergo CT before lumbar puncture.
The relative frequency of the various causes of community-acquired bacterial meningitis has changed over the decades. When all cases occurring after the neonatal period are taken into account, the leading cause 20 to 30 years ago was H. influenzae (45 percent), followed by Streptococcus pneumoniae (18 percent) and Neisseria meningitidis (14 percent). By 10 years ago, considerable change had occurred as a result of the immunization of infants with H. influenzae type b protein–polysaccharide conjugate vaccines, reducing the total number of annual cases of H. influenzae type b disease in the United States by 55 percent and the number of cases of H. influenzae meningitis by 94 percent.3 As a consequence, S. pneumoniae became the leading species (47 percent), followed by N. meningitidis (25 percent) and Listeria monocytogenes (8 percent) — a distribution quite similar to that currently seen in the Netherlands. Another change has been an increase in the percentage of adults with bacterial meningitis at large urban hospitals in whom the origin of the disease is nosocomial — about 40 percent, for example, at Massachusetts General Hospital in Boston.
Finally, since the beginning of the antibiotic era, the emergence of antimicrobial resistance in species that cause community-acquired bacterial meningitis has been a continuing problem. Examples include sulfonamide resistance in N. meningitidis arising in the 1960s, resistance of H. influenzae to ampicillin arising in the 1970s, and resistance to penicillin in S. pneumoniae meningitis isolates in the 1990s in the United States (21 percent with intermediate-level resistance and 14 percent with high resistance), necessitating initial therapy with a combination of a third-generation cephalosporin and vancomycin. Most of the patients described by van de Beek et al. were treated initially with amoxicillin (or penicillin), a third-generation cephalosporin, or both. Only 2 of 351 pneumococcal meningeal isolates showed resistance (intermediate-level) to penicillin, so the use of vancomycin in initial therapy was not required, as it would have been in the United States for suspected pneumococcal meningitis.
How can the progress of the 20th century be extended into the 21st? In my view, future progress will come more from preventive than from therapeutic measures. For example, the introduction in early 2000 of the 7-valent pneumococcal protein–polysaccharide conjugate vaccine appears to be preventing invasive infections in young children for whom the vaccine is indicated and may be having a spillover effect in reducing the rate of invasive disease in older adults.4 The development of quadrivalent conjugate vaccines (A/C/Y/W-135) for immunoprophylaxis against meningococcal infection in at-risk groups may provide more effective and longer-lasting immunity than the current polysaccharide vaccine. In addition, improvements in food processing and food-safety measures may reduce the incidence of L. monocytogenes meningitis, an increasing problem that is most common in, though not exclusive to, the immunocompromised.
Source Information
From the Department of Medicine, Massachusetts General Hospital, Boston.
References
Flexner S. The results of the serum treatment in thirteen hundred cases of epidemic meningitis. J Exp Med 1913;17:553-576.
Dowling HF, Sweet LK, Robinson JA, et al. The treatment of pneumococcic meningitis with massive doses of systemic penicillin. Am J Med Sci 1949;217:149-156.
Schuchat A, Robinson K, Wenger JD, et al. Bacterial meningitis in the United States in 1995. N Engl J Med 1997;337:970-976.
Whitney CG, Farley MM, Hadler J, et al. Decline in invasive pneumococcal disease after the introduction of protein-polysaccharide conjugate vaccine. N Engl J Med 2003;348:1737-1746.
Related Letters:
Prognostic Factors in Adults with Bacterial Meningitis
Joffe A. R., ?stergaard C., Klussmann J. P., Guntinas-Lichius O., Altschuler E. L., van de Beek D., de Gans J., Swartz M. N.(Morton N. Swartz, M.D.)
Figure. Mortality Rates Associated with Community-Acquired Bacterial Meningitis over the Past 90 Years.
In 1913, Simon Flexner was the first to report some success in treating bacterial meningitis with intrathecal equine meningococcal antiserum: among 1300 patients with epidemic meningitis, mortality was reduced to 31 percent.1 Among 169 children with meningococcal meningitis treated with intrathecal antiserum at Bellevue Hospital, New York, between 1928 and 1936, the outcome was even more favorable, with mortality of about 20 percent. Fothergill reported in 1937 that treatment of H. influenzae meningitis with combined intravenous and intrathecal antiserums reduced mortality among 201 children to 85 percent. The prognosis for patients with pneumococcal meningitis remained extremely grave even after the introduction of specific antiserums. There were only anecdotal reports of recovery after treatment with systemic and intrathecal antipneumococcal serum.
In the 1930s, with the introduction of sulfonamides, the mortality associated with meningococcal meningitis decreased to 5 to 15 percent. By 1944, Alexander had reported that treatment with both a sulfonamide and intravenous rabbit antiserum in 87 children with H. influenzae type b meningitis had reduced mortality to 22 percent. In the early 1950s, chloramphenicol treatment (with sulfadiazine) reduced the fatality rate of H. influenzae meningitis to 5 to 10 percent and made the use of antiserum unnecessary. The results of sulfonamide treatment of pneumococcal meningitis were less favorable, with mortality ranging from 45 to 95 percent.
The use of penicillin therapy for pneumococcal meningitis began in the mid-1940s, with combined systemic and intrathecal administration, and resulted in a fatality rate of 49 percent. Modern treatment with "meningeal" doses without concomitant intrathecal drug administration began with the 1949 report by Dowling et al. of a study in which 21 patients with pneumococcal meningitis were treated with 1 million units of penicillin intramuscularly every two hours, with a resulting improvement in mortality to 38 percent.2
For the past 15 years, therapy for community-acquired bacterial meningitis has consisted of intravenous penicillin (or ampicillin), a third-generation cephalosporin, or both; mortality rates for meningococcal meningitis have hovered around 10 percent, those for H. influenzae meningitis have been reduced to less than 5 percent, and those for pneumococcal meningitis have remained around 20 percent. In this issue of the Journal, van de Beek and colleagues (pages 1849–1859) confirm these rates in a large series of patients in the Netherlands, with mortality of 7 percent associated with meningococcal disease and 30 percent associated with pneumococcal disease.
The demonstration that bacterial components within the cerebrospinal fluid could cause a release of inflammatory cytokines led to trials of dexamethasone as adjuvant therapy. In four trials in children with H. influenzae meningitis conducted in the early 1990s, dexamethasone use did not change mortality but did reduce the frequency of neurologic sequelae — principally, sensorineural hearing loss. In 2002, Dutch investigators showed that, in adults, adjunctive dexamethasone therapy reduced the rate of unfavorable outcomes from 25 to 15 percent; beneficial effects were most obvious in pneumococcal meningitis.
In the past 20 years, cranial computed tomography (CT) has been another major advance, allowing the differentiation of other types of pyogenic intracranial infection and detection of brain swelling and mass lesions. The downside of its availability has been the frequent delay in antimicrobial therapy while the physician awaits the results of the CT scan and lumbar puncture. In the study by van de Beek et al., CT was performed before lumbar puncture in 337 of 696 patients, and in two thirds of these patients, initial therapy was not started before the CT results had been obtained. In a study from Yale University involving 301 adults with suspected meningitis at an urban hospital emergency room, certain clinical features (e.g., prior central nervous system disease, abnormal sensorium, limb drift, aphasia, and abnormal visual fields) were associated with abnormalities on CT. Tellingly, among the 96 patients who had none of these findings, fully 97 percent of the scans were normal (a mild mass effect was found in only one case), emphasizing that clinical features can be used to identify patients who do not need to undergo CT before lumbar puncture.
The relative frequency of the various causes of community-acquired bacterial meningitis has changed over the decades. When all cases occurring after the neonatal period are taken into account, the leading cause 20 to 30 years ago was H. influenzae (45 percent), followed by Streptococcus pneumoniae (18 percent) and Neisseria meningitidis (14 percent). By 10 years ago, considerable change had occurred as a result of the immunization of infants with H. influenzae type b protein–polysaccharide conjugate vaccines, reducing the total number of annual cases of H. influenzae type b disease in the United States by 55 percent and the number of cases of H. influenzae meningitis by 94 percent.3 As a consequence, S. pneumoniae became the leading species (47 percent), followed by N. meningitidis (25 percent) and Listeria monocytogenes (8 percent) — a distribution quite similar to that currently seen in the Netherlands. Another change has been an increase in the percentage of adults with bacterial meningitis at large urban hospitals in whom the origin of the disease is nosocomial — about 40 percent, for example, at Massachusetts General Hospital in Boston.
Finally, since the beginning of the antibiotic era, the emergence of antimicrobial resistance in species that cause community-acquired bacterial meningitis has been a continuing problem. Examples include sulfonamide resistance in N. meningitidis arising in the 1960s, resistance of H. influenzae to ampicillin arising in the 1970s, and resistance to penicillin in S. pneumoniae meningitis isolates in the 1990s in the United States (21 percent with intermediate-level resistance and 14 percent with high resistance), necessitating initial therapy with a combination of a third-generation cephalosporin and vancomycin. Most of the patients described by van de Beek et al. were treated initially with amoxicillin (or penicillin), a third-generation cephalosporin, or both. Only 2 of 351 pneumococcal meningeal isolates showed resistance (intermediate-level) to penicillin, so the use of vancomycin in initial therapy was not required, as it would have been in the United States for suspected pneumococcal meningitis.
How can the progress of the 20th century be extended into the 21st? In my view, future progress will come more from preventive than from therapeutic measures. For example, the introduction in early 2000 of the 7-valent pneumococcal protein–polysaccharide conjugate vaccine appears to be preventing invasive infections in young children for whom the vaccine is indicated and may be having a spillover effect in reducing the rate of invasive disease in older adults.4 The development of quadrivalent conjugate vaccines (A/C/Y/W-135) for immunoprophylaxis against meningococcal infection in at-risk groups may provide more effective and longer-lasting immunity than the current polysaccharide vaccine. In addition, improvements in food processing and food-safety measures may reduce the incidence of L. monocytogenes meningitis, an increasing problem that is most common in, though not exclusive to, the immunocompromised.
Source Information
From the Department of Medicine, Massachusetts General Hospital, Boston.
References
Flexner S. The results of the serum treatment in thirteen hundred cases of epidemic meningitis. J Exp Med 1913;17:553-576.
Dowling HF, Sweet LK, Robinson JA, et al. The treatment of pneumococcic meningitis with massive doses of systemic penicillin. Am J Med Sci 1949;217:149-156.
Schuchat A, Robinson K, Wenger JD, et al. Bacterial meningitis in the United States in 1995. N Engl J Med 1997;337:970-976.
Whitney CG, Farley MM, Hadler J, et al. Decline in invasive pneumococcal disease after the introduction of protein-polysaccharide conjugate vaccine. N Engl J Med 2003;348:1737-1746.
Related Letters:
Prognostic Factors in Adults with Bacterial Meningitis
Joffe A. R., ?stergaard C., Klussmann J. P., Guntinas-Lichius O., Altschuler E. L., van de Beek D., de Gans J., Swartz M. N.(Morton N. Swartz, M.D.)