Bacterial Meningitis in the United States,

Transcription

1 T h e n e w e ngl a nd j o u r na l o f m e dic i n e original article Bacterial Meningitis in the United States, Michael C. Thigpen, M.D., Cynthia G. Whitney, M.D., M.P.H., Nancy E. Messonnier, M.D., Elizabeth R. Zell, M.Stat., Ruth Lynfield, M.D., James L. Hadler, M.D., M.P.H., Lee H. Harrison, M.D., Monica M. Farley, M.D., Arthur Reingold, M.D., Nancy M. Bennett, M.D., Allen S. Craig, M.D., William Schaffner, M.D., Ann Thomas, M.D., Melissa M. Lewis, M.P.H., Elaine Scallan, Ph.D., and Anne Schuchat, M.D., for the Emerging Infections Programs Network A BS TR AC T From the Centers for Disease Control and Prevention (M.C.T., C.G.W., N.E.M., E.R.Z., M.M.L., E.S., A.S.) and the Georgia Department of Human Resources (M.M.F.) both in Atlanta; Minnesota Department of Health, Minneapolis (R.L.); Connecticut Department of Public Health, Hartford (J.L.H.); Johns Hopkins University Bloomberg School of Public Health, Baltimore (L.H.H.); School of Public Health, University of California at Berkeley, Berkeley (A.R.); University of Rochester School of Medicine and Dentistry, Rochester, NY (N.M.B.); Vanderbilt University School of Medicine, Nashville (A.S.C., W.S.); and Oregon Public Health Division, Portland (A.T.). Address reprint requests to Dr. Thigpen at: 3150 Rampart Rd., Fort Collins, CO 80521, or at mthigpen@cdc.gov or thigpenm@kh.cdc.gov. N Engl J Med 2011;364: Copyright 2011 Massachusetts Medical Society. Background The rate of bacterial meningitis declined by 55% in the United States in the early 1990s, when the Haemophilus influenzae type b (Hib) conjugate vaccine for infants was introduced. More recent prevention measures such as the pneumococcal conjugate vaccine and universal screening of pregnant women for group B streptococcus (GBS) have further changed the epidemiology of bacterial meningitis. Methods We analyzed data on cases of bacterial meningitis reported among residents in eight surveillance areas of the Emerging Infections Programs Network, consisting of approximately 17.4 million persons, during We defined bacterial meningitis as the presence of H. influenzae, Streptococcus pneumoniae, GBS, Listeria monocytogenes, or Neisseria meningitidis in cerebrospinal fluid or other normally sterile site in association with a clinical diagnosis of meningitis. Results We identified 3188 patients with bacterial meningitis; of 3155 patients for whom outcome data were available, 466 (14.8%) died. The incidence of meningitis changed by 31% (95% confidence interval [CI], 33 to 29) during the surveillance period, from 2.00 cases per 100,000 population (95% CI, 1.85 to 2.15) in to 1.38 cases per 100,000 population (95% CI 1.27 to 1.50) in The median age of patients increased from 30.3 years in to 41.9 years in (P<0.001 by the Wilcoxon rank-sum test). The case fatality rate did not change significantly: it was 15.7% in and 14.3% in (P = 0.50). Of the 1670 cases reported during , S. pneumoniae was the predominant infective species (58.0%), followed by GBS (18.1%), N. meningitidis (13.9%), H. influenzae (6.7%), and L. monocytogenes (3.4%). An estimated 4100 cases and 500 deaths from bacterial meningitis occurred annually in the United States during Conclusions The rates of bacterial meningitis have decreased since 1998, but the disease still often results in death. With the success of pneumococcal and Hib conjugate vaccines in reducing the risk of meningitis among young children, the burden of bacterial meningitis is now borne more by older adults. (Funded by the Emerging Infections Programs, Centers for Disease Control and Prevention.) 2016 n engl j med 364;21 nejm.org may 26, 2011

2 Bacterial Meningitis in the United States Studies in the 1970s and 1980s showed that five pathogens (Haemophilus influenzae, Streptococcus pneumoniae, Neisseria meningitidis, group B streptococcus [GBS], and Listeria monocytogenes) caused more than 80% of cases of bacterial meningitis. 1-4 Between 1986 and 1995, the incidence of bacterial meningitis from these five pathogens declined by 55%, largely owing to the use of the H. influenzae type b (Hib) conjugate vaccine for infants, which was introduced in the United States in Since then, additional interventions to prevent invasive disease from these pathogens have been introduced 6-8 (see also Table 1 in the Supplementary Appendix, available with the full text of this article at NEJM.org). With the introduction of the heptavalent protein-polysaccharide pneumococcal conjugate vaccine (PCV7) in 2000, invasive pneumococcal disease declined by 75% among children under 5 years of age and by 31% among adults 65 years or older. 9,10 Since the age of the patient guides empirical antimicrobial therapy for purulent meningitis, 11 the effect of prevention strategies on the current epidemiology of bacterial meningitis remains important to define. We used data from two active laboratory- and population-based surveillance systems of the Emerging Infections Programs (EIP) Network the Active Bacterial Core surveillance (ABCs) and the Foodborne Diseases Active Surveillance Network (FoodNet) to describe trends in the incidence of bacterial meningitis from 1998 to 2007 and to describe the epidemiology of meningitis for in order to provide a baseline for the evaluation of future interventions. Me thods Data Collection To describe trends in the incidence of bacterial meningitis, we analyzed ABCs surveillance data on culture-confirmed invasive infection with H. influenzae, S. pneumoniae, N. meningitidis, or GBS and FoodNet surveillance data on culture-confirmed invasive infection with L. monocytogenes. The infections were reported between January 1, 1998, and December 31, 2007, at the following EIP sites: California (San Francisco county), Connecticut (entire state), Georgia (20-county Atlanta area), Maryland (6-county Baltimore area), Minnesota (7-county Minneapolis St. Paul area), New York (7-county Rochester area), Oregon (3-county Portland area), and Tennessee (5 urban counties). These eight sites encompassed an estimated 17,383,935 persons (6.4% of the total U.S. population), according to the 1998 census. Between 1998 and 2003, additional counties were added to EIP sites in Minnesota, New York, and Tennessee. To assess the epidemiology of bacterial meningitis from less-common pathogens across a 5-year period (between January 1, 2003, and December 31, 2007), we used this expanded surveillance area, with an estimated population of 22,870,454 persons (7.9% of the total U.S. population), according to the 2003 census. The methods of ABCs have been published previously. 12,13 ABCs defines a case of bacterial meningitis as the presence of H. influenzae, S. pneumoniae, N. meningitidis, or GBS in cerebrospinal fluid or another normally sterile site in association with a clinical diagnosis of meningitis made by the primary health care provider in a resident of a surveillance area. FoodNet receives data on all culture-confirmed cases of L. monocytogenes infection. 14,15 Since FoodNet does not collect information on clinical meningitis, we included in this study only cases from the FoodNet database in which L. monocytogenes was isolated from a cerebrospinal fluid specimen. All surveillance methods remained consistent between 1998 and ABCs sites sent available isolates to reference laboratories for organismspecific subtyping; pneumococcal and GBS isolates underwent antimicrobial-susceptibility testing. 12 PCV7 strains included pneumococcal serotypes 4, 6B, 9V, 14, 8C, 19F, and 23F; all other serotypes were considered non-pcv7 types. Strains considered to be 13-valent pneumococcal conjugate vaccine (PCV13) strains included the PCV7 serotypes as well as serotypes 1, 3, 5, 6A, 7F, and 19A; all other serotypes were considered to be non-pcv13 types. Serogroups found in quadrivalent meningococcal conjugate vaccine (MCV4) or quadrivalent meningococcal polysaccharide vaccine (MPSV4) included A, C, W135, and Y; all others were considered to be nonvaccine types. Statistical Analysis We used SAS software, version 9.1 (SAS Institute), for analyses. We calculated the rates of bacterial meningitis for each year from 1998 through 2007, expressed as the number of cases per 100,000 population, by using U.S. Census annual population estimates, adjusted for race and age, for the n engl j med 364;21 nejm.org may 26,

3 T h e n e w e ngl a nd j o u r na l o f m e dic i n e surveillance sites. We used chi-square analyses to test for significant linear trends over time and the Wilcoxon rank sum test to compare medians across years. To estimate the burden of bacterial meningitis in the United States in , we used observed rates of bacterial meningitis for each age group or race, calculated from the ABCs and FoodNet data, and applied the observed rates to the total U.S. population. Race was categorized as black, white, or other (including American Indian or Alaska Native and Asian or Pacific Islander). For each surveillance site and age group, cases for which race was unknown and cases caused by S. pneumoniae or N. meningitidis infection for which serotype or serogroup information was unknown were assigned according to the distributions of cases for which race or serotype or serogroup, respectively, were known. We calculated case fatality rates using only data from patients with a known outcome (98.9% of patients). A P value of less than 0.05 was considered to indicate statistical significance. Methods are described further in the Supplementary Appendix. R esult s Rates of Bacterial Meningitis, From 1998 through 2007, we identified 3188 cases of bacterial meningitis due to H. influenzae, S. pneumoniae, N. meningitidis, GBS, or L. monocytogenes. The incidence of bacterial meningitis caused by these pathogens changed during this period by 31% (95% confidence interval [CI], 33 to 29), from 2.00 cases per 100,000 population (95% CI, 1.85 to 2.15) in to 1.38 cases per 100,000 population (95% CI, 1.27 to 1.50) in (Table 1). Throughout the surveillance period, incidences remained highest for patients under 2 months of age and for black patients of any age. Incidences declined significantly over the surveillance period within all age groups except for patients under 2 months of age. The median age of patients increased from 30.3 years during to 41.9 years during (P<0.001). Between and , no significant change in the case fatality rate was observed (15.7% and 14.3%, respectively; P = 0.50). Infection with S. pneumoniae accounted for 1813 cases of bacterial meningitis and for 321 deaths among the 1796 cases for which outcome data were available (case fatality rate, 17.9%). Between and , the incidence of S. pneumoniae meningitis changed by 26% (95% CI, 29 to 23), from 1.09 cases per 100,000 population (95% CI, 0.98 to 1.20) in to 0.81 cases per 100,000 population (95% CI, 0.72 to 0.90) in (Table 2). This overall decline included a change of 62% (95% CI 66 to 58) among children 2 to 23 months of age, from 9.7 cases per 100,000 population (95% CI, 7.7 to 11.9) in to 3.7 cases per 100,000 population (95% CI, 2.5 to 5.1) in The overall case fatality rate did not change significantly; it was 17.9% in and 14.7% in (P = 0.26). The incidence of meningitis from S. pneumoniae PCV7 serotypes changed by 92% overall (95% CI, 93 to 91), from 0.61 cases per 100,000 population (95% CI, 0.48 to 0.64) in to 0.05 cases per 100,000 population (95% CI, 0.03 to 0.07) in (Table 2). However, the incidence of bacterial meningitis from non-pcv7 serotypes increased by 61% (95% CI, 54 to 69), from 0.48 cases per 100,000 population (95% CI, 0.45 to 0.60) in to 0.77 cases per 100,000 population (95% CI, 0.67 to 0.84) in The pathogen N. meningitidis was identified in 549 cases of bacterial meningitis; of the 547 cases for which outcome data were available, 55 were fatal (case fatality rate, 10.1%). The incidence of N. meningitidis meningitis changed by 58% overall (95% CI, 61 to 54), from 0.44 cases per 100,000 population (95% CI, 0.37 to 0.51) in to 0.19 cases per 100,000 population (95% CI, 0.14 to 0.24) in The rates of meningococcal meningitis caused by serogroups B, C, and Y changed by 55% (95% CI, 60 to 49), 65% (95% CI 69% to 60%), and 52% (95% CI 59% to 45%), respectively. The number of cases per 100,000 population that were due to serogroup B declined from 0.13 (95% CI, 0.09 to 0.16) in to 0.06 (95% CI, 0.04 to 0.09) in ; the number of cases due to serogroup C declined from 0.19 (95% CI, 0.14 to 0.23) to 0.07 (95% CI, 0.04 to 0.09); and the number of cases due to serogroup Y declined from 0.10 (95% CI, 0.05 to 0.11) to 0.05 (95% CI, 0.02 to 0.05). Between 1998 and 2007, a total of 534 cases of meningitis from GBS infection were reported to ABCs; of the 522 cases for which outcome data were available, 58 were fatal (case fatality rate, 11.1%). Overall, rates of GBS meningitis did not change significantly during the surveillance pe n engl j med 364;21 nejm.org may 26, 2011

4 Bacterial Meningitis in the United States Table 1. Incidence of Bacterial Meningitis in the United States, , Stratified According to Age Group, Race, and Pathogen.* Characteristic Percent Change, vs (95% CI) no. of cases per 100,000 population (95% CI) Age group <2 Mo (56.45 to 94.35) (69.69 to ) (42.13 to 74.45) (60.58 to 96.90) (63.53 to ) 10 (1 to 20) 2 23 Mo (11.85 to 16.91) (9.45 to 13.92) 6.56 (5.06 to 8.38) 6.95 (5.47 to 8.89) 6.91 (5.30 to 8.77) 51 ( 55 to 48) (1.20 to 1.96) 1.48 (1.16 to 1.88) 0.94 (0.68 to 1.27) 1.07 (0.79 to 1.43) 0.56 (0.36 to 0.82) 64 ( 68 to 59) (0.71 to 1.43) 0.87 (0.60 to 1.22) 0.62 (0.39 to 0.94) 0.56 (0.34 to 0.86) 0.43 (0.25 to 0.71) 58 ( 64 to 51) (0.79 to 1.22) 0.86 (0.68 to 1.07) 0.70 (0.54 to 0.89) 0.76 (0.59 to 0.97) 0.66 (0.50 to 0.86) 33 ( 38 to 27) (1.01 to 1.48) 1.30 (1.08 to 1.55) 1.08 (0.89 to 1.31) 0.91 (0.74 to 1.13) 0.95 (0.76 to 1.16) 23 ( 29 to 17) (1.75 to 2.57) 1.83 (1.49 to 2.21) 2.09 (1.75 to 2.48) 1.79 (1.49 to 2.14) 1.73 (1.44 to 2.06) 19 ( 25 to 14) (2.13 to 3.16) 2.20 (1.76 to 2.72) 2.21 (1.78 to 2.71) 1.51 (1.16 to 1.94) 1.92 (1.53 to 2.38) 27 ( 32 to 22) All ages 2.00 (1.85 to 2.15) 1.82 (1.69 to 1.97) 1.49 (1.38 to 1.62) 1.41 (1.30 to 1.54) 1.38 (1.27 to 1.50) 31 ( 33 to 29) Race White 1.71 (1.55 to 1.87) 1.58 (1.43 to 1.73) 1.28 (1.15 to 1.42) 1.27 (1.14 to 1.41) 1.28 (1.14 to 1.40) 25 ( 28 to 23) Black 4.07 (3.57 to 4.62) 3.85 (3.40 to 4.35) 3.12 (2.72 to 3.57) 2.62 (2.28 to 3.03) 2.41 (2.13 to 2.84) 41 ( 44 to 37) Other 1.55 (0.98 to 2.23) 0.68 (0.37 to 1.18) 0.76 (0.44 to 1.25) 0.67 (0.39 to 1.14) 0.46 (0.25 to 0.86) 70 ( 75 to 64) Pathogen Streptococcus pneumoniae 1.09 (0.98 to 1.20) 1.03 (0.93 to 1.13) 0.93 (0.83 to 1.03) 0.76 (0.68 to 0.85) 0.81 (0.72 to 0.90) 26 ( 29 to 23) Neisseria meningitidis 0.44 (0.37 to 0.51) 0.37 (0.31 to 0.44) 0.23 (0.19 to 0.29) 0.22 (0.17 to 0.27) 0.19 (0.14 to 0.24) 58 ( 61 to 54) Group B streptococcus 0.24 (0.20 to 0.30) 0.30 (0.25 to 0.36) 0.21 (0.17 to 0.26) 0.27 (0.22 to 0.32) 0.25 (0.21 to 0.31) 4 ( 3 to 12) Haemophilus influenzae 0.12 (0.09 to 0.17) 0.10 (0.07 to 0.14) 0.10 (0.07 to 0.13) 0.10 (0.07 to 0.14) 0.08 (0.05 to 0.11) 35 ( 42 to 27) Listeria monocytogenes 0.10 (0.08 to 0.16) 0.03 (0.01 to 0.05) 0.03 (0.01 to 0.05) 0.05 (0.04 to 0.10) 0.05 (0.03 to 0.08) 46 ( 53 to 39) * CI denotes confidence interval. Race was obtained from medical records. Other includes American Indian or Alaska Native, Asian or Pacific Islander, or other race. Within a site and age group, cases with missing data for race were assumed to have a distribution of race similar to that among cases with available data. n engl j med 364;21 nejm.org may 26,

5 T h e n e w e ngl a nd j o u r na l o f m e dic i n e Table 2. Incidence of Bacterial Meningitis from Streptococcus pneumoniae in the United States, , Stratified According to Age Group or Age Group and Serotype.* Characteristic Percent Change, vs (95% CI) no. of cases per 100,000 population (95% CI) Age group 2 23 Mo 9.69 (7.68 to 11.87) 7.24 (5.51 to 9.04) 3.32 (2.29 to 4.67) 3.59 (2.51 to 4.96) 3.67 (2.53 to 5.12) 62 ( 66 to 58) (0.34 to 0.80) 0.80 (0.57 to 1.10) 0.47 (0.29 to 0.71) 0.54 (0.35 to 0.81) 0.36 (0.21 to 0.57) 34 ( 43 to 22) (0.08 to 0.42) 0.26 (0.12 to 0.49) 0.22 (0.10 to 0.43) 0.25 (0.12 to 0.47) 0.21 (0.09 to 0.42) 7 ( 18 to 38) (0.28 to 0.55) 0.37 (0.26 to 0.52) 0.37 (0.25 to 0.51) 0.38 (0.26 to 0.54) 0.27 (0.17 to 0.40) 33 ( 41 to 24) (0.65 to 1.04) 0.90 (0.72 to 1.11) 0.80 (0.63 to 1.00) 0.65 (0.50 to 0.83) 0.76 (0.60 to 0.96) 8 ( 15 to 0) (1.16 to 1.84) 1.36 (1.08 to 1.70) 1.69 (1.39 to 2.05) 1.37 (1.10 to 1.67) 1.34 (1.09 to 1.63) 9 ( 15 to 1) ³ (1.48 to 2.36) 1.63 (1.26 to 2.09) 1.77 (1.39 to 2.22) 0.87 (0.61 to 1.20) 1.43 (1.10 to 1.83) 24 ( 30 to 18) All ages 1.09 (0.98 to 1.20) 1.03 (0.93 to 1.13) 0.93 (0.83 to 1.03) 0.76 (0.68 to 0.85) 0.81 (0.72 to 0.90) 26 ( 29 to 23) Age group and serotype <5 PCV7 serotype 3.37 (2.46 to 3.91) 2.56 (1.61 to 2.78) 0.51 (0.27 to 0.86) 0.31 (0.13 to 0.60) 0.07 (0.01 to 0.30) 98 ( 98 to 97) Non-PCV7 serotype 0.87 (0.76 to 1.65) 1.17 (1.11 to 2.15) 0.94 (0.61 to 1.41) 1.68 (1.23 to 2.26) 1.67 (1.17 to 2.19) 92 (68 to 119) 65 PCV7 serotype 0.77 (0.49 to 1.05) 0.78 (0.54 to 1.12) 0.49 (0.30 to 0.74) 0.23 (0.10 to 0.41) 0.11 (0.04 to 0.27) 85 ( 88 to 81) Non-PCV7 serotype 1.12 (0.84 to 1.53) 0.85 (0.58 to 1.18) 1.28 (0.95 to 1.67) 0.63 (0.43 to 0.94) 1.31 (1.00 to 1.70) 18 (7 to 29) Any age PCV7 serotype 0.61 (0.48 to 0.64) 0.58 (0.44 to 0.59) 0.30 (0.23 to 0.33) 0.16 (0.11 to 0.19) 0.05 (0.03 to 0.07) 92 ( 93 to 91) Non-PCV7 serotype 0.48 (0.45 to 0.60) 0.45 (0.43 to 0.57) 0.63 (0.56 to 0.72) 0.60 (0.53 to 0.69) 0.77 (0.67 to 0.84) 61 (54 to 69) * There were too few cases in the age group of under 2 months to determine trends: five cases in , five cases in , two cases in , three cases in , and two cases in CI denotes confidence interval, and PCV7 heptavalent protein-polysaccharide pneumococcal conjugate vaccine n engl j med 364;21 nejm.org may 26, 2011

6 Bacterial Meningitis in the United States riod; there were 0.24 cases per 100,000 population (95% CI, 0.20 to 0.30) in and 0.25 cases (95% CI, 0.21 to 0.31) in Rates of GBS meningitis among patients under 2 months of age did not change significantly after the introduction of universal GBS screening of pregnant women; the rate was 65.2 cases per 100,000 population (95% CI, 53.8 to 78.3) in and 62.5 (95% CI, 53.6 to 72.5) in (relative change, 4%; 95% CI 10 to 2), although most cases (86.5%) in were late-onset ( 7 days of age) and would not have been affected by the intrapartum antimicrobial prophylaxis. ABCs received 187 reports of H. influenzae meningitis cases, 13 of which were fatal (case fatality rate, 7.0%); 9.4% of cases were caused by serotype b. The overall incidence of H. influenzae meningitis changed by 35% (95% CI, 42 to 27) between and , from 0.12 cases per 100,000 population (95% CI, 0.09 to 0.17) to 0.08 (95% CI, 0.05 to 0.11). During the surveillance period, L. monocytogenes caused 105 cases of bacterial meningitis and 19 deaths (case fatality rate, 18.1%). The incidence changed by 46% (95% CI, 53 to 39), from 0.10 cases per 100,000 population (95% CI, 0.08 to 0.16) in to 0.05 (95% CI, 0.03 to 0.08) in The change among patients under 2 months of age was 36% (95% CI, 51 to 16), from 10.1 cases per 100,000 population (95% CI, 4.6 to 19.8) in to 6.5 (95% CI, 2.3 to 14.7) in Epidemiology of Bacterial Meningitis, EIP sites reported 1670 cases of bacterial meningitis from 2003 through Death occurred in 215 cases (13.0%); S. pneumoniae accounted for 70.7% of these 215 deaths. We estimate that an average of 4100 cases of bacterial meningitis, including 500 that were fatal, occurred annually in the United States during (Fig. 1 in the Supplementary Appendix). Pediatric Cases We identified 587 cases of bacterial meningitis among children. GBS accounted for 86.1% of cases among those under 2 months of age, and N. meningitidis caused 45.9% of cases among those 11 to 17 years of age. Among the other pediatric age groups, S. pneumoniae was the most common cause (Fig. 1A). The case fatality rate was 6.9% among pediatric patients on average; nearly 10% had underlying immunocompromising or chronic medical conditions (Table 3). Isolates were available for serotyping in 187 of the 203 pediatric cases (92.1%) caused by S. pneumoniae. PCV7 serotypes accounted for 15.5% of these pediatric cases overall and 13.0% of cases in the 2-to-23-month age group. The case fatality rate was similar among children infected with PCV7 isolates and those infected with non-pcv7 isolates (10.7% and 7.6%, respectively; P = 0.58). Among patients under 5 years of age, the most common serotypes were 19A (in 26.1% of cases), 7F (11.2%), 10A (6.7%), and 22F (6.0%). PCV13 serotypes accounted for 60.0% of cases in children between 2 and 23 months of age and 57.2% in children of any age. Isolates were available for serogroup testing in 105 of the 107 pediatric cases (98.1%) caused by N. meningitidis. Serogroups B, C, and Y were most commonly identified (in 59.1%, 21.0%, and 11.4% of cases, respectively). Serogroup B caused 70.5% of the cases among children under 11 years of age. Serogroups included in a quadrivalent meningococcal vaccine (MCV4 or MPSV4) accounted for 66.7% of infections among children 11 to 17 years of age. ABCs received 222 case reports of GBS meningitis, and isolates were available for serotyping in 139 of the cases (62.6%). Serotypes III, IA, and V accounted for 43.0%, 30.8%, and 15.9% of cases in the first 2 months of life, respectively. H. influenzae isolates were available for serotyping in 41 of 42 cases of meningitis (97.6%); 11 of these cases (26.8%) were nontypable. Serotypes f and b were identified in 14 and 5 cases, respectively. A total of 13 cases of L. monocytogenes meningitis were reported in children, including 10 (76.9%) in children under 2 months of age. Adult Cases We identified 1083 cases of bacterial meningitis in adults; S. pneumoniae was the most common pathogen (Fig. 1B). The overall adult case fatality rate was 16.4%, and the rate increased linearly with increasing age (8.9% among patients 18 to 34 years of age vs. 22.7% among those 65 years, P<0.001). Isolates were available for serotyping in 680 of the 765 adult cases (88.9%) caused by S. pneumoniae. PCV7 and PCV13 serotypes accounted for 16.0% n engl j med 364;21 nejm.org may 26,

7 T h e n e w e ngl a nd j o u r na l o f m e dic i n e A Children B Adults Percentage of Total Cases <2 Mo 2 23 Mo Age Group All pediatric cases No. of Cases Percentage of Total Cases Listeria monocytogenes Neisseria meningitidis GBS Age Group Haemophilus influenzae Streptococcus pneumoniae All adult cases No. of Cases Figure 1. Proportions of the 1670 Cases of Bacterial Meningitis Reported in Caused by Each Pathogen, According to Age Group. Panel A shows data for children, and Panel B shows data for adults. Overall, Streptococcus pneumoniae was the predominant cause of bacterial meningitis (accounting for 58.0% of cases), followed by group B streptococcus (GBS) (18.1%), Neisseria meningitidis (13.9%), Haemophilus influenzae (6.7%), and Listeria monocytogenes (3.4%). and 41.6% of the meningitis cases, respectively, and meningitis from PCV7 serotypes of S. pneumoniae had a higher case fatality rate than those caused by non-pcv7 serotypes (25.9% vs. 16.2%, P = 0.02). Among patients 65 years of age or older, the most common S. pneumoniae serotypes were 3 (in 9.2% of cases), 11A (7.0%), 19A (7.0%), and 23A (7.0%); PCV13 accounted for only 39.4% of the serotypes causing meningitis in this older age group. Isolates were available for confirmation in 117 of the 125 adult cases of bacterial meningitis (93.6%) caused by N. meningitidis. Serogroups B and C were the predominant causes of cases among adults 18 to 34 years of age (accounting for 34.4% and 45.9% of cases, respectively); in cases in adults 35 years of age or older, serogroups B and Y were the most common (each serogroup accounted for 30.4% of cases). Serogroups included in MCV4 or MPSV4 accounted for 62.8% of infections among patients 18 to 55 years of age. Among the 80 adult cases caused by GBS, isolates were available for laboratory analysis in 43 cases (53.8%). Serotypes IA and V accounted for 37.2% and 25.6% of these cases, respectively. Of the 69 adult cases from H. influenzae, 61 (88.4%) had isolates available for serotyping. A majority (73.8%) of the H. influenzae isolates were not able to be serotyped. Serotypes e and f were the most common serotypes identified (with each found in 11.5% of cases). The case fatality rate was significantly higher among adult meningitis cases from typable H. influenzae than among cases from nontypable H. influenzae (18.8% vs. 2.2%, P = 0.02). In 13 of the 44 reported cases of adult L. monocytogenes meningitis (29.5%), serotyping could be performed, and 8 isolates were of serotype 4B. Only one case of L. monocytogenes meningitis occurred in a pregnant woman, and none occurred in a patient with human immunodeficiency virus (HIV) infection. Discussion Our findings indicate that the incidence of bacterial meningitis caused by H. influenzae, S. pneumoniae, GBS, L. monocytogenes, or N. meningitidis decreased in the past decade, primarily due to declines in the rate of S. pneumoniae meningitis. Rates of bacterial meningitis decreased most sharply among children, causing the median age at diagnosis of bacterial meningitis to increase. However, the overall case fatality rate was not significantly reduced, since there has been little change in the case fatality rate of pneumococcal meningitis. The timing of the decline in the incidence of bacterial meningitis, as well as the change in causative serotypes, suggests that the use of PCV7 contributed to the changes observed, as has been suggested previously in reports about pathogen-specific causes of meningitis Despite significant declines in the incidence of pediatric bacterial meningitis, the incidence among infants under 2 months of age, which is the group 2022 n engl j med 364;21 nejm.org may 26, 2011

8 Bacterial Meningitis in the United States Table 3. Characteristics of Patients with Bacterial Meningitis Identified by the Emerging Infections Programs Network, Neisseria meningitidis Haemophilus influenzae Group B Streptococcus Listeria monocytogenes Streptococcus pneumoniae Characteristic All percent of patients Pediatric patients N = 107 N = 42 N = 222 N = 13 N = 203 N = 587 Male sex Race* White Black Other Underlying medical condition Immunocompromising condition Chronic condition Prematurity only None Case fatality rate All pediatric patients Pediatric patients <2 yr Adult patients N = 125 N = 69 N = 80 N = 44 N = 765 N = 1083 Male sex Race* White Black Other Underlying medical condition or risk group Immunocompromising condition Chronic condition Smoking Age 65 yr only None Case fatality rate All adult patients Adult patients 50 yr * Race was obtained from medical records. Other includes American Indian or Alaska Native, Asian or Pacific Islander, and other race. Data on race were not available for some patients; therefore the percentages do not sum to 100%. For underlying medical conditions, immunocompromising conditions include multiple myeloma, sickle cell disease, asplenia, organ transplantation, immunoglobulin deficiency, immunosuppressive therapy, human immunodeficiency virus or the acquired immunodeficiency syndrome (HIV AIDS), leukemia, Hodgkin s disease, lupus, the nephrotic syndrome, and chronic kidney disease. Chronic conditions include asthma or chronic obstructive pulmonary disease, diabetes, cirrhosis, alcohol abuse, atherosclerotic cardiovascular disease, congestive heart failure, burns, cerebrospinal fluid leak, injection-drug use, and cerebrovascular accident (as well as presence of hydrocephalus or ventriculoperitoneal shunt in children). Some conditions were added for study during the surveillance period; not all were identified a priori. Patients with more than one condition were counted for only one, according to the following hierarchy: immunocompromising condition, chronic condition, smoker only (if adult), and prematurity or age of 65 years or older only. Data from the New York site are not included, since cases of HIV AIDS are not reported at that site. Data for patients with bacterial meningitis from L. monocytogenes infection are also not reported, since FoodNet does not consistently record underlying medical conditions for these patients. at greatest risk for bacterial meningitis, did not decrease. The major causative organism in this vulnerable age group remains GBS, with infection manifested as late-onset disease 7 or more days after birth. Although intrapartum antibiotic prophylaxis has markedly reduced the risk of early-onset infection, such measures have had no effect on the risk of late-onset disease. 19 One hopeful observation in the pediatric population is the 36% decrease in the rate of L. mono- n engl j med 364;21 nejm.org may 26,

9 T h e n e w e ngl a nd j o u r na l o f m e dic i n e cytogenes meningitis during this period. In contrast to other causes of bacterial meningitis, almost all listeriosis cases are foodborne, most commonly associated with ready-to-eat meat products The incidence of listeriosis during pregnancy, which may result in adverse fetal outcomes (e.g., spontaneous abortion or preterm delivery), has shown a marked decrease in recent years, possibly because of reductions in L. monocytogenes contamination of ready-to-eat foods 21,24,25 or decreased consumption of high-risk foods by pregnant women due to educational efforts. 26 Among older children and young adults, N. meningitidis remained a major cause of bacterial meningitis, despite the fact that rates declined significantly between 1998 and Since the declines were similar among all major serogroups, and since MCV4 and MPSV4 do not include N. meningitidis serogroup B, these declines most likely represent secular trends rather than a vaccine effect. 17 As the proportion of children receiving MCV4 continues to increase, we expect additional reductions in the incidence of meningococcal disease (especially from N. meningitidis serogroups C and Y). The incidence of bacterial meningitis also declined among adults, including those 65 years of age or older. Rates of adult meningitis may decline further, as children are vaccinated with the newly licensed PCV13. 27,28 However, in contrast to the findings for cases of meningitis in children, chronic and immunocompromising conditions were common among adults with bacterial meningitis. Given that medical conditions such as HIV infection (a risk factor for some causes of meningitis) show no signs of decline in the United States, 29 reducing the meningitis burden among adults may prove difficult without consideration of the use of new pneumococcal vaccines for adults. We present the results for the most common causes of community-acquired bacterial meningitis over a 10-year period in a population of more than 17 million persons, yet these findings most likely underestimate the true burden of bacterial meningitis, for three main reasons. First, EIP sites only identify patients with culture-confirmed cases of meningitis. Therefore, identification of cases depends on the diagnostic and therapeutic practices of those caring for patients (e.g., the frequency of lumbar punctures or the timing of initiation of antibiotic therapy). Diagnostic methods for identifying culture-negative cases of meningitis, such as polymerase-chain-reaction assays, are not uniformly available at all EIP sites, and therefore such data are not included here. Second, ABCs and FoodNet do not cover all possible causal pathogens of bacterial meningitis; for instance, Escherichia coli and staphylococcus species are not included yet have been identified previously as clinically significant causes. 3 Although techniques for antimicrobial testing and for serotype and serogroup testing used by the Centers for Disease Control and Prevention (CDC) remained consistent during the surveillance period, individual hospitals may have adopted more sensitive antibacterial isolation techniques (i.e., liquid-based rather than agar-based culture). However, this var iation would result in underestimation of the decline in the incidence of meningitis during the surveillance period. Third, the surveillance systems used in the study do not routinely collect information on bacterial meningitis acquired in health care settings, which may account for up to 40% of cases. 30 Although the epidemiology of bacterial meningitis has evolved from 1998 to 2007, the rank order of causative pathogens has changed relatively little. For clinicians, these results suggest that current treatment guidelines for bacterial meningitis targeting the major pathogenic causes are still appropriate. 11 After the introduction of PCV7, rates of bacterial meningitis among children 1 to 23 months of age have declined significantly, ranging from 7.7 to 8.4 cases per 100,000 population between 2002 and 2007, surpassing the Healthy People 2010 goal of reducing the incidence to 8.6 cases per 100,000 population. 31 However, this achievement only reemphasizes the need for interventions targeting neonates and elderly persons, the two populations in which the meningitis burden remains greatest. Administration of GBS vaccines and new meningococcal vaccines could reduce the risk of bacterial meningitis among infants. However, these vaccines are still in the early stages of development. PCV13 has recently been licensed in the United States for pediatric use and may be licensed in the future for adults if shown to be efficacious, safe, and immunogenic. 32 However, because PCV13 and newer vaccines being developed against meningitis caused by GBS or N. meningitidis infection may not cover many causal isolates, other approaches will most likely be required to markedly reduce the meningitis burden in the very young and very old n engl j med 364;21 nejm.org may 26, 2011

10 Bacterial Meningitis in the United States Presented in part at the 43rd annual meeting of the Infectious Diseases Society of America, San Francisco, October 6 9, Supported by the Emerging Infections Programs, Centers for Disease Control and Prevention, Atlanta. Disclosure forms provided by the authors are available with the full text of this article at NEJM.org. We thank Jessica MacNeil, Tracy Pondo, Carolyn Wright, and Tami Skoff for their contributions to the ABCs system for purposes of this study; Delois Jackson, Varja Sakota, and other members of the CDC s Streptococcus Laboratory for streptococcal typing; and the participating clinical laboratories and ABCs and FoodNet surveillance staff within each EIP site who made this study possible. References 1. Fraser DW, Geil CC, Feldman RA. Bacterial meningitis in Bernalillo County, New Mexico: a comparison with three other American populations. Am J Epidemiol 1974;100: Fraser DW, Mitchell JE, Silverman LP, Feldman RA. Undiagnosed bacterial meningitis in Vermont children. Am J Epidemiol 1975;102: Schlech WF III, Ward JI, Band JD, Hightower A, Fraser DW, Broome CV. Bacterial meningitis in the United States, 1978 through 1981: the National Bacterial Meningitis Surveillance Study. JAMA 1985; 253: Wenger JD, Hightower AW, Facklam RR, Gaventa S, Broome CV. Bacterial meningitis in the United States, 1986: report of a multistate surveillance study. J Infect Dis 1990;162: Schuchat A, Robinson K, Wenger JD, et al. Bacterial meningitis in the United States in N Engl J Med 1997;337: Preventing pneumococcal disease among infants and young children: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep 2000;49(RR-9): Prevention of perinatal group B streptococcal disease: revised guidelines from the CDC. MMWR Recomm Rep 2002;51 (RR-11): Prevention and control of meningococcal disease: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep 2005;54(RR-7): 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: Direct and indirect effects of routine vaccination of children with 7-valent pneumococcal conjugate vaccine on incidence of invasive pneumococcal disease United States, MMWR Morb Mortal Wkly Rep 2005;54: Tunkel AR, Hartman BJ, Kaplan SL, et al. Practice guidelines for the management of bacterial meningitis. Clin Infect Dis 2004;39: Schuchat A, Hilger T, Zell E, et al. Active Bacterial Core surveillance of the Emerging Infections Program Network. Emerg Infect Dis 2001;7: Active Bacterial Core surveillance (ABCs): methodology. Atlanta: Centers for Disease Control and Prevention. ( index.html.) 14. Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food 10 states, United States, MMWR Morb Mortal Wkly Rep 2006;55: Hardnett FP, Hoekstra RM, Kennedy M, Charles L, Angulo FJ. Epidemiologic issues in study design and data analysis related to FoodNet activities. Clin Infect Dis 2004;38:Suppl 3:S121-S Hsu HE, Shutt KA, Moore MR, et al. Effect of pneumococcal conjugate vaccine on pneumococcal meningitis. N Engl J Med 2009;360: Cohn AC, MacNeil JR, Harrison LH, et al. Changes in Neisseria meningitidis disease epidemiology in the United States, : implications for prevention of meningococcal disease. Clin Infect Dis 2010;50: Tsai CJ, Griffin MR, Nuorti JP, Grijalva CG. Changing epidemiology of pneumococcal meningitis after the introduction of pneumococcal conjugate vaccine in the United States. Clin Infect Dis 2008; 46: Phares CR, Lynfield R, Farley MM, et al. Epidemiology of invasive group B streptococcal disease in the United States, JAMA 2008;299: Mead PS, Slutsker L, Dietz V, et al. Food-related illness and death in the United States. Emerg Infect Dis 1999;5: Voetsch AC, Angulo FJ, Jones TF, et al. Reduction in the incidence of the invasive listeriosis in Foodborne Diseases Active Surveillance Network sites, Clin Infect Dis 2007;44: Varma JK, Samuel MC, Marcus R, et al. Listeria monocytogenes infection from foods prepared in a commercial establishment: a case-control study of potential sources of sporadic illness in the United States. Clin Infect Dis 2007;44: Gottlieb SL, Newbern EC, Griffin PM, et al. Multistate outbreak of listeriosis linked to turkey deli meat and subsequent changes in US regulatory policy. Clin Infect Dis 2006;42: Food Safety and Inspection Service. Revised action plan for control of Listeria monocytogenes for the prevention of foodborne listeriosis. Washington, DC: Food Safety and Inspection Service, ( topics/lm_action.htm.) 25. Idem. FSIS regulatory testing for LM in RTE products by calendar year Washington, DC: Food Safety and Inspection Service, ( RTE_ pdf.) 26. Jackson KA, Iwamoto M, Swerdlow D. Pregnancy-associated listeriosis. Epidemiol Infect 2010;138: Flannery B, Hefferman RT, Harrison LH, et al. Changes in invasive pneumococcal disease among HIV-infected adults living in the era of childhood pneumococcal immunization. Ann Intern Med 2006; 144: Licensure of a 13-valent pneumococcal conjugate vaccine (PCV13) and recommendations for use among children Advisory Committee on Immunization Practices (ACIP), MMWR Morb Mortal Wkly Rep 2010;59: Hall HI, Song R, Rhodes P, et al. Estimation of HIV incidence in the United States. JAMA 2008;300: Durand ML, Calderwood SB, Weber DJ, et al. Acute bacterial meningitis in adults. N Engl J Med 1993;328: Active Bacterial Core surveillance (ABCs): Healthy People 2010 objectives related to ABCs pathogens. Atlanta: Centers for Disease Control and Prevention. ( healthy-people-2010.html.) 32. Advisory Committee on Immunization Practices. Summary report, February 25-26, Atlanta: ACIP, 2009: ( downloads/min-archive/min-feb09.pdf.) Copyright 2011 Massachusetts Medical Society. n engl j med 364;21 nejm.org may 26,