RSS Feeds
Sporadic Meningococcal Disease in Adults: Results of a 5-Year Population-Based Study
- David S. Stephens, MD;
- Rana A. Hajjeh, MD;
- Wendy S. Baughman, MPH;
- R. Christopher Harvey, MPH;
- Jay D. Wenger, MD; and
- Monica M. Farley, MD
- From Emory University School of Medicine, Veterans Affairs Medical Center (Atlanta), and Centers for Disease Control and Prevention, Atlanta, Georgia. Acknowledgments: The authors thank David Rimland, MD; Lane Pucko; Tina Stull, MD; Barbara Marston, MD, Mary Susan Bardsley, MPH; Brian Plikaytis; the Georgia Department of Human Resources; and the staff and hospitals of the Atlanta Metropolitan Surveillance Project for their invaluable assistance.
Abstract
Objective: To define the incidence, demographics, clinical features, and risk factors for sporadic meningococcal disease in adults (≥ 18 years) residing in metropolitan Atlanta.
Design: Prospective, population-based surveillance, with retrospective review of clinical and laboratory records.
Setting: Eight-county metropolitan Atlanta area.
Patients: All adult patients in whom Neisseria meningitidis was isolated from normally sterile sites (blood, cerebrospinal fluid) during the period 1 December 1988 to 30 November 1993.
Measurements: Incidence, relative risk, clinical and laboratory parameters, and serogroup of meningococcal isolates.
Results: For the 5-year period, 44 (33%) of 132 cases of meningococcal disease in Atlanta occurred in adults (annual incidence, 0.50/100 000 adults per year). Twenty-three (52%) of the 44 adults presented without rash or meningitis, the two most obvious signs of meningococcal disease. Pneumonia, sinusitis, or purulent tracheobronchitis, but without rash, were the likely sources of meningococcal bacteremia in 15 (34%) of the 44 adults. Twelve of the 15 patients with meningococcal respiratory infection were older than 50 years of age or were immunocompromised (or both), and three fourths of the 15 patients had disease caused by serogroups B, Y, and W-135. Overall, two thirds of adults older than 24 years of age with meningococcal disease had one or more immunocompromising conditions (for example, low complement 50 level [CH50], corticosteroid use, congestive heart failure, multiple myeloma, human immmunodeficiency virus infection). Meningococcemia or meningococcal meningitis, often caused by serogroup C, were the presentations in 14 of 15 adults 18 to 24 years old; only 2 had an identified underlying condition.
Conclusions: In this 5-year population-based study, one third of all cases of sporadic meningococcal disease occurred in adults. Over half of the adults presented without rash or meningitis. Pneumonia, sinusitis, and tracheobronchitis are important sources of bacteremic meningococcal disease, especially in immunocompromised patients and elderly persons.
Infections due to Neisseria meningitidis continue to cause significant mortality and morbidity in the United States [1]. About 3000 cases of meningococcal disease occur each year; recent increases in disease caused by serogroups C and B have been observed [2, 3], and meningococci that are relatively resistant to penicillin have been reported [4].
Most meningococcal disease in the United States is sporadic. The highest attack rates are seen in young children, but many cases occur in adults [1]. Despite a substantial number of cases and a 10% to 20% fatality rate [1], a comprehensive study of sporadic meningococcal disease in adults has not been done. Case reports suggest that presentations other than meningococcemia and meningococcal meningitis (for example, pneumonia) may occur in adults with sporadic meningococcal disease [5, 6]. Risk factors for adults, other than complement or immunoglobulin deficiencies [7], are not understood.
Since 1988, we have investigated the epidemiology and clinical spectrum of invasive bacterial pathogens in metropolitan Atlanta through active population-based surveillance [8-10]. An important benefit of this approach is the identification of all disease in a defined population, which avoids the selection bias associated with the more commonly used hospital-base case series approach. In our study, we analyzed cases of meningococcal disease in adults (≥ 18 years of age) who were identified during 5 years of prospective, population-based surveillance.
Methods
Cases were defined by isolation of N. meningitidis from blood or cerebrospinal fluid and were identified by a prospective laboratory-based surveillance in metropolitan Atlanta [8-10], an eight-county area with 32 hospitals and a total population of 2 344 514 persons, including 1 743 276 adults 18 years of age or older (1990 census). All patients were living in the surveillance area at the time of infection. Surveillance was done from 1 December 1988 to 30 November 1993. Among the adults, 70% were white, 27% were black, and 3% were of other racial backgrounds; 47% were men, and 53% were women. The numbers of adults estimated in the surveillance area to be alive and infected with the human immunodeficiency virus (HIV) were as follows: in 1989, 3250; in 1990, 4146; in 1991, 4776; in 1992, 6660; and in 1993, 7923 [9, 10]. Isolates were identified as N. meningitidis at hospital laboratories and serogrouped using standard techniques. Strain identification, serogrouping, and penicillin susceptibility of all available isolates were confirmed at the Centers for Disease Control and Prevention.
We obtained initial case reports from hospital laboratories and infection-control practitioners. We did laboratory audits at all hospitals every 6 months to evaluate reporting accuracy and to identify cases that were not reported originally (< 4% of cases). We reviewed the medical records of all adult patients to determine clinical features and underlying conditions. Incidence calculations were based on population data from the 1990 census. Relative risk was calculated as previously described [8]. Patients were classified as having meningococcemia, chronic meningococcemia, meningococcal meningitis, or meningococcal bacteremia without sepsis on the basis of of meningococcal disease (Table 1) described by Wolf and Birbara [11], Apicella [12], and Benoit [13].
Results
From December 1988 to December 1993, 132 cases of sporadic meningococcal disease occurred in metropolitan Atlanta. Forty-four of the 132 cases (33%) occurred in adults 18 years of age or older. The annual incidence of meningococcal disease per 100 000 persons (95% Poisson CIs) in the total population was 1.13 (CI, 0.94 to 1.34; n = 132). The incidence in children 0 to 4 years old was 6.80 (CI, 5.2 to 8.73; n = 61) and in children 5 to 17 years old, 1.28 (CI, 0.84 to 1.86; n = 27). The incidence in all adults was 0.50 (CI, 0.37 to 0.68; n = 44); in adults 18 to 24 years old, 1.17 (CI, 0.65 to 1.93; n = 15); in adults 25 to 49 years old, 0.25 (CI, 0.13 to 0.43; n = 13); and in adults 50 years of age or older, 0.72 (CI, 0.41 to 1.17, n = 16). The incidence of meningococcal disease in black adults was 0.68 (CI, 0.39 to 1.10; n = 16); in white adults, 0.46 (CI, 0.30 to 0.66; n = 28); in women, 0.45 (CI, 0.28 to 0.69; n = 21); and in men, 0.56 (CI, 0.36 to 0.84; n = 23).
In adults, most cases (33 of 44) occurred from November through April. No epidemic outbreaks or case clusters were reported in the surveillance area during the study, nor were any cases in adults secondary cases. However, two cases were associated with a defined exposure (one in a laboratory researcher working with large quantities of serogroup B N. meningitidis and one in a student exposed to an outbreak outside the surveillance area).
Clinical features of the 44 adults with sporadic meningococcal disease are shown in Table 1. Fourteen adults (32%) presented with acute meningococcemia; 7 of these 14 were hypotensive, and 5 presented with a petechial, purpuric, or maculopapular rash. Seven of these 14 adults presented with rhinorrhea, cough, or sore throat, but there was no evidence of a defined respiratory infection, except in one adult infected with HIV, who presented with meningococcal pneumonia, respiratory failure, and sepsis. Four percent (2 of 44) of adults presented with chronic meningococcemia.
Fourteen of the 44 adults (32%) presented with meningitis. One had maxillary sinusitis, and 5 had sore throat, cough, or rhinorrhea at presentation. A generalized rash was noted at presentation in 10 of 44 adults (71%) with meningitis.
Fourteen of the 44 adults (32%) presented with bacteremia without sepsis. None of these patients had or developed a rash or meningitis. All but 1 presented with a defined respiratory infection (Table 1). Overall, 34% of adults (15 of 44) had pneumonia, sinusitis, or tracheobronchitis as the likely source for the meningococcal bacteremia (Table 1), with pneumonia present in 10 of 44 cases (23%). Seven of the 10 adults with pneumonia were 50 years of age or older (4 were older than 75 years old). Pneumonia was not present in adults 18 to 24 years old.
A petechial or purpuric rash was noted at presentation in 13 of the 44 adults with sporadic meningococcal disease (32%) (Table 1); an additional 3 (7%) had a maculopapular rash, and 27 (61%) had no rash (13 of these 27 were black). Overall, 3 of 16 black persons (19%) compared with 14 of 28 white persons (50%) were noted to have a rash. No adults developed a rash after admission.
Underlying conditions (Table 2) were identified in 21 of the 44 adults with meningococcal disease. Of the 15 cases among persons 18 to 24 years old, 14 of whom presented with meningococcemia or meningococcal meningitis, 2 (13%) had an underlying risk factor. However, complement screening assays were not done in 7 of the 15. (Overall, complement screening was done in 36% of adults [16 of 44].)
Among persons 25 to 49 years old, 46% (6 of 13) had an underlying illness or risk factor (Table 2). Three were infected with HIV. Based on the estimates of HIV infection in the Atlanta population, the risk for meningococcal disease was increased in adults with HIV infection (relative risk, 23.8; CI, 7.4 to 76.7; P < 0.001), and the estimated annual incidence of meningococcal disease in adults with HIV infection was 11.2/100 000 adults. In adults 50 years of age or older, 81% (13 of 16) had at least one chronic underlying illness (Table 2). Overall, 19 of 29 adults older than 24 years of age had an immunocompromising condition compared with 2 of 15 adults younger than 24 years of age (relative risk, 4.91 [CI, 1.32 to 18.34]; P = 0.001, Fisher exact test).
The association of capsular serogroup and meningococcal disease in adults is shown in Table 1 and Table 2. Serogroup data were available for 38 of 44 isolates: Eighteen (47%) were serogroup C; 14 (37%) were serogroup B; 4 (11%) were serogroup Y, and 2 (5%) were serogroup W-135. Isolates from cases of meningitis were either serogroup C (53%) or B (47%). Serogroup C was the identified serogroup in 10 of 12 cases with major sequelae (Table 1) and was the responsible serogroup in 9 of 13 isolates from adults 18 to 24 years old (Table 2). In contrast, bacteremia without sepsis was caused by serogroups B, Y, and W-135 in 8 of 11 cases with isolates. Similarly, pneumonia was caused by serogroups B, Y, and W-135 in eight of nine cases with isolates, and these serogroups were found in 80% (16 of 20) of isolates from cases with underlying conditions (Table 2). All 38 isolates available for testing were sensitive to penicillin.
Discussion
During our 5-year study, one third of blood or cerebrospinal fluid culture-confirmed meningococcal cases in Atlanta occurred in adults, a result consistent with previous estimates [1]. Rash and meningitis, the two most obvious signs of meningococcal infection, were absent in more than half of the adults. This emphasizes the need for heightened awareness of the different clinical presentations of meningococcal disease in adults in order to implement infection control measures designed to decrease secondary transmission [1]. For example, nosocomial outbreaks have occurred because of transmission of N. meningitidis from patients with meningococcal pneumonia to hospital staff and to other patients [14].
Upper respiratory symptoms, which may be caused by co-infections [15] or mucosal damage produced directly by meningococci [12], were noted in most of the adults in our series. In addition, more than one third of adults with meningococcal bacteremia presented with clinical pneumonia, purulent tracheobronchitis, or sinusitis.
Meningococcal lobar and bronchopneumonia was first reported during the 1918-1919 influenza pandemic. Subsequently, several case reports, predominantly in adults and adolescents, indicated that the meningococcus can be a primary lower respiratory pathogen [5, 12, 14, 16]. Based on the frequency of cases of meningococcal pneumonia found in our series and data [5]—namely, that approximately 30% of cases of meningococcal pneumonia are bacteremic—the estimated annual incidence of meningococcal pneumonia, both bacteremic and nonbacteremic, is 0.4 cases/100 000 adults per year. Serogroups B, Y, and W-135 have most frequently been identified in case reports of sporadic meningococcal pneumonia [5, 12, 14, 16], and this was confirmed in our surveillance.
Skin lesions are reported in 40% to 90% of patients with invasive meningococcal disease [12, 16]. In our series, 61% of adults had no skin rash, and we found no rash in adults with meningococcal bacteremia without sepsis, many of whom had pneumonia. Although the absence of skin lesions in patients with meningococcal pneumonia has been attributed to the absence of bacteremia [16], all of our patients were bacteremic, suggesting that other factors (for example, lower levels of bacteremia or less virulent strains) may account for the absence of rash. Hyperpigmented skin also appears to contribute to the absence or detection of a rash.
Several groups have been identified as having increased risk for meningococcal disease, including those in close contact with patients with the disease, urban populations, military recruits [11, 12], smokers [17], and patients with immunoglobulin or complement deficiencies [7]. In our study, two thirds of adults older than 24 years of age had underlying conditions known to impair the integrity of complement and immunoglobulin (Table 2), local respiratory defenses, or neutrophil function. Complement or immunoglobulin deficiency may have been unrecognized in others, emphasizing that screening assays need to be done for these conditions in all adults with meningococcal disease. Our data also indicate that patients infected with HIV are at increased risk for sporadic meningococcal disease. Although meningococcal infection in persons with HIV infection has been previously reported [18], no increased risk with HIV infection was found in epidemic serogroup A meningococcal disease in sub-Saharan Africa [19].
Advanced age also may be a risk factor. In young children, the risk for meningococcal disease increases as the maternal antibody wanes. Decreases in humoral immunity that occur with aging [20] may similarly increase the risk for meningococcal disease in elderly persons, especially when such persons have conditions that increase the risk for respiratory infection (for example, congestive heart failure). Further studies are needed to characterize the magnitude of increased risk and the appropriateness of meningococcal vaccine use in various adult populations.
- Copyright ©2004 by the American College of Physicians
