Indications for the Immunological Evaluation of Patients with Meningitis

Transcription

1 INVITED ARTICLE CLINICAL PRACTICE Ellie J. C. Goldstein, Section Editor Indications for the Immunological Evaluation of Patients with Meningitis Gary D. Overturf Department of Pediatrics and Pathology, Division of Pediatric Infectious Diseases, University of New Mexico Health Sciences Center, Albuquerque Although people with bacterial meningitis lack adequate protective antibody against the invading pathogen, most do not have an underlying immunodeficiency. Certain comorbid conditions increase the risk for development of bacterial sepsis and meningitis. In addition, certain congenital complement deficiencies, defects of antibody production, or asplenia may be first recognized by the occurrence of bacterial meningitis, particularly when it occurs in infants or young children. Deficiencies of the terminal components of complement (C5 C9) or properdin have been associated with recurrent or invasive neisserial infections, and asplenia, agammaglobulinemia, and deficiencies of the early components of complement (e.g., C1 C3) are associated with risks of infections caused by Streptococcus pneumoniae, Haemophilus influenzae, and meningococci. The presence of congenital or acquired immunodeficiencies should be considered in persons who present with bacterial meningitis on the basis of the etiology, clinical epidemiology, and presence of other risk factors. Bacterial meningitis is a rare presentation for acquired or congenital immunodeficiency; thus, most patients who present with bacterial meningitis do not have an identifiable deficiency of immune function. Rarely, however, certain congenital or acquired immunodeficiencies may be first recognized by the occurrence of bacterial meningitis. Among the critical host factors that contribute to an enhanced susceptibility to sepsis and consequent meningeal infection, the most frequently recognized is the lack of adequate acquired protective antibody against the infecting pathogen, thus explaining the frequent occurrence of meningitis in the very young child or in the elderly individual. Other factors that are specific to the host, such as disease-associated immune compromise (e.g., diabetes or chronic alcoholism) or specific acquired immune dysfunction (e.g., HIV infection), may also play a role in susceptibility to bacterial meningitis. In addition, a predisposing infection, such as endocarditis, sinusitis, or other chronic suppurative Received 27 November 2001; accepted 2 October 2002; electronically published 31 December Reprints or correspondence: Dr. Gary D. Overturf, Dept. of Pediatrics and Pathology, Div. of Pediatric Infectious Diseases, 2211 Lomas Ave. NE, ACC3, University of New Mexico Health Sciences Center, Albuquerque, NM (goverturf@salud.unm.edu). Clinical Infectious Diseases 2003; 36: by the Infectious Diseases Society of America. All rights reserved /2003/ $15.00 pericranial infection, may lead to direct or hematogenous infection of the meninges. The incidence of bacterial meningitis due to its 3 major pathogenic causes, Streptococcus pneumoniae, Neisseria meningitidis, and Haemophilus influenza type b (table 1), remained relatively constant during the past 3 decades, until the licensure of the Haemophilus conjugate vaccine [1]. During the past decade, since the introduction of routine immunization of infants during the first year of life, the incidence of invasive infection due to Haemophilus influenza type b has decreased by 198% [2]. Characteristically, the incidence of the communityacquired infections peak after 5 6 months of age and before 2 years of age, which is the period of time before infants acquire active immunity and after the transplacental maternal antibody level has decreased to nonprotective levels. The incidence of invasive pneumococcal infection peaks in young children (age,!2 years), and there is a second equal peak in adults aged 165 years. For persons of all age groups, the lack of adequate specific protective antibody associated with opsonophagocytic deficiency is the major risk factor. Similarly, although meningococcal infection may occur at all ages, it also has a peak incidence in children aged!2 years, and the lack of protective antibody is an important risk factor. It is likely that protein conjugate meningococcal and pneumococcal polysaccharide vaccines will have a marked impact on the occurrence of these CLINICAL PRACTICE CID 2003:36 (15 January) 189

2 2 infections in the future, similar to that experienced after the introduction of Haemophilus conjugate vaccines. ACQUIRED OR CONGENITAL IMMUNODEFICIENCY AS A FACTOR IN SUSCEPTIBILITY TO BACTERIAL MENINGITIS Certain acquired or congenital conditions affecting phagocytic function (e.g., diabetes or chronic alcoholism), antibody deficiency or dysfunction (e.g., young or old age or congenital or acquired immunodeficiency), congenital or acquired splenic dysfunction, or complement deficiency or dysfunction increase the risk of meningeal infection, particularly infections caused by polysaccharide-encapsulated pathogens. Congenital immunodeficiencies, such as agammaglobulinemia, may present with bacterial meningitis caused by one of the community-acquired pathogens, but, in general, bacterial meningitis is a relatively rare presenting illness for most congenital immunodeficiencies. In addition, opportunistic infections, such as those caused by pneumococci, may herald the presence of congenital or acquired immunodeficiency, such as HIV infection. For most patients with bacterial meningitis, the usual age of presentation and the lack of a history of significant previous invasive bacterial infections suggestive of immunodeficiency, as well as the absence of risk factors for acquired immunodeficiency, will exclude a consideration for a specific immunodeficiency. However, for some infections, particularly those caused by N. meningitidis, the presentation of bacterial sepsis and meningitis may be the first and only clinical signal of an underlying, previously undetermined immunodeficiency. In addition, certain populations experience higher rates of invasive bacterial infections due to Haemophilus species and pneumococci, particularly Native American, Native Alaskan, and African American children. Some members of these populations also have high rates of congenital immunodeficiency (e.g., common variable or severe combined immunodeficiency), which may predispose them to certain invasive bacterial infections. Table 1. Causes of 248 cases of bacterial meningitis and overall case-fatality rates, according to organism, Organism No. (%) of cases Incidence per 100,000 persons Case-fatality rate, % Haemophilus influenzae 18 (7) Streptococcus pneumoniae 117 (47) Neisseria meningitidis 62 (25) Group B Streptococcus species 31 (13) Listeria species 20 (8) NOTE. From [1]. RISK OF MENINGOCOCCAL INVASIVE DISEASE WITH COMPLEMENT AND PROPERDIN DEFICIENCIES Development of systemic infections with N. meningitidis is usually due to an absence of specific protective antibodies, or, rarely, susceptibility may be caused by the presence of IgA antibodies that block serum antineisserial bactericidal activity. Meningococcal infection has been associated with several immunodeficiencies, particularly those caused by low levels or dysfunction of the terminal components of complement or properdin proteins [3 5]. Also, although most cases of meningococcemia occur among individuals without previously known complement deficiency, this infection may complicate the course of disease-associated complement deficiency such as systemic lupus erythematosus [6]. Meningococci colonize the nasopharynx of 5% 15% of individuals in areas of nonendemicity, and a larger proportion of individuals may be colonized during epidemics of invasive disease. Meningitis occurs in 39% of individuals with late component complement deficiencies [3 5] and 6% of those with properdin deficiencies [7]. Previously, investigators have documented complement deficiencies in persons who presented with meningococcal sepsis or meningitis, which occurred either singly or in various combinations of deficiencies of C5, C6, C7, and C8, with concomitant loss of serum bactericidal activity against N. meningitidis. Recently, investigators in Japan studied a large group of C9- deficient individuals and found that this defect is also associated with a higher prevalence of meningococcal infection [8]. In a study of patients who presented with meningococcal invasive disease, 6 of 20 patients had complement deficiencies, which were determined on the basis of a CH50 level 12 SD below the normal mean level [3]. Three of these 6 subjects had a deficiency of C6 or C8, and the other 3 had multiple complement deficiencies associated with underlying systemic lupus erythematosus. In another study of patients with meningococcal diseases due to uncommon serotypes (X, Y, Z, W-135, or 29E), the frequency of complications was higher in those patients with complement deficiencies, occurring in 125% of patients with complement deficiencies versus 3.3% of those without complement deficiencies [4]. Complement deficiencies were detected in 8 (26.7%) of 30 patients who had localization of deficits to either C8 or C7. Deficiency or dysfunction in the properdin (i.e., alternate) pathway may also be associated with invasive or severe infections caused by Neisseria species. Properdin promotes activation of the alternative pathway of complement by stabilizing C3 convertase. Previous studies have suggested that properdin dysfunction is X-linked, and serogroups B, C, Y, and W-135 have all been isolated from patients with properdin dysfunction during episodes of meningitis or sepsis due to N. meningitidis [7, 8]. 190 CID 2003:36 (15 January) CLINICAL PRACTICE

3 RISK OF PNEUMOCOCCAL INFECTION WITH IMMUNODEFICIENCY OR ASPLENIA The primary risk factor for development of invasive disease and meningitis caused by S. pneumoniae is related to age and deficiency of protective antibody against the infecting strain [1, 9]. However, conditions associated with congenital agammaglobulinemia, or other acquired deficiencies of antibody production (e.g., severe combined or common variable immunodeficiency), may be associated with invasive pneumococcal infection. In addition, invasive pneumococcal infection may be the first presenting infection for HIV-infected individuals. However, acquired or congenital asplenia is the most common host defect to lead to severe invasive pneumococcal infection. Rates of invasive pneumococcal infection among children aged!5 years with acquired or congenital asplenia, including sickle-cell disease, exceed those of healthy children by fold. The rate of infection for children with sickle-cell disease (determined before 1985) was cases per 100,000 children aged!5 years, and the rate decreased to cases per 100,000 children for those aged 5 years [10, 11]. Rates of invasive pneumococcal infection in healthy children aged!5 years peak during the second year of life, increasing to a peak incidence of 1200 cases per 100,000 children. Acquired splenic dysfunction and loss of the spleen at any age presumably carry an increased risk of severe pneumococcal infection and fulminant sepsis with death, although the magnitude of these risks has not been determined prospectively [12]. Although the incidence of invasive pneumococcal infection markedly decreases for asplenic persons aged 5 years, rates continue to exceed other age-matched comparative populations by 110-fold, and fulminant pneumococcal sepsis continues to occur. Congenital asplenia also is associated with a high risk of invasive pneumococcal infection. Children with congenital heart disease and associated congenital asplenia have rates of invasive pneumococcal disease that parallel those of children with sickle-cell disease [13]. ANTIBODY DEFICIENCY, HIV INFECTION, AND COMPLEMENT DYSFUNCTION AND BACTERIAL MENINGITIS Lack of specific antibodies also markedly increases the risk of acquiring invasive infections caused by S. pneumoniae, as well as increasing the risk of acquiring infections due to Neisseria and Haemophilus species. Thus, congenital deficiencies of immunoglobulin or complement production may present with bacterial meningitis due to any of the major meningeal pathogens. Immunodeficiencies characterized by absent or inadequate antibody production include Bruton agammaglobulinemia, X-linked hypogammaglobulinemia with normal-to-increased IgM levels, common variable immunodeficiencies, combined antibody and cellular immunodeficiencies, and Wiskott-Aldrich syndrome. In addition, a number of complement deficiencies of early-phase components in the classical pathway, such as C1q, C1rs, C2, C4, and C3, or the alternative pathway of complement fixation, are associated with opsonophagocytic dysfunction, which increases the risk of acquiring invasive infections caused by polysaccharide-encapsulated organisms, particularly pneumococcal infections (table 2). Approximately 10% 15% of children with Bruton agammaglobulinemia experience episodes of sepsis or meningitis before they received treatment with immunoglobulin. In a review of 154 patients with Wiskott-Aldrich syndrome [14], 7 (5%) were reported to have experienced bacterial meningitis, whereas 5 (8.9%) of 56 patients with X-linked hyper-igm syndrome experienced meningitis or encephalitis [15]. Thus, although CNS infection is not a common occurrence in most antibody deficiency syndromes, it may be the presenting manifestation of these immunodeficiencies. Some authorities have recommended that an evaluation for immunodeficiency be considered with the second episode of bacterial meningitis [16]. A complete discussion of congenital antibody and complement deficiencies is beyond the scope of this article [16 18]. Acquired immunodeficiencies (particularly HIV infection) are also associated with invasive infections due to pneumococcal organisms. Less frequently, HIV infection and other acquired immunodeficiencies have been associated with invasive meningococcal or Haemophilus infections. HIV infection is associated with a striking increase in the incidence of invasive pneumococcal infections in adults and children [19 21]. HIV infection in children may be associated with bacterial meningitis due to pneumococci and other bacteria in up to 20% of children. HIV-infected children have high rates of invasive pneumococcal disease, and S. pneumoniae is the most common cause of invasive bacterial infection, accounting for 35% 50% of such episodes. The relative risk of pneumococcal disease in HIV-infected children ( ,000 cases per 100,000 children) is 3 22-fold higher than is the relative risk for children without HIV infection [20, 22]. Similarly, the rate of pneumococcal invasive infection in adults with AIDS in San Francisco County, California, was 802 cases per 100,000 persons, which is 146- fold the rate for age-matched adults without AIDS [21]. OTHER CAUSES OF BACTERIAL MENINGITIS AND THE RISK OF IMMUNODEFICIENCY Haemophilus infection. Disseminated Haemophilus infections (e.g., meningitis, pneumonia, and septic arthritis) may be either a presenting infection in a child with congenital deficiency that involves antibody production or an early complement component deficiency. Such infections may be due to any one of the capsular types of Haemophilus species (most CLINICAL PRACTICE CID 2003:36 (15 January) 191

4 Table 2. Congenital immunodeficiencies that may predispose persons to acquire bacterial meningitis and other invasive bacterial infections. Condition Mode of inheritance Antibody (B cell) immunodeficiency X-linked; autosomal recessive and/or dominant X-linked (Bruton) agammaglobulinemia Common variable immunodeficiency X-linked hypogammaglobulinemia with increased IgM level Combined antibody and cellular immunodeficiency Variable Severe combined immunodeficiency Autosomal recessive or dominant Immunodeficiency with adenosine deaminase Autosomal recessive or nucleoside phosphorylase deficiency Immunodeficiency with ataxia telangiectasia Autosomal recessive Wiskott-Aldrich syndrome X-linked Complement deficiencies Variable C1, C2, and C3 (early complement) deficiencies Autosomal recessive or unknown C4, C5, C6, C7, and C8 deficiencies Autosomal recessive or unknown Defects in alternate pathway, properdin X-linked; sickle-cell disease Splenic dysfunction Variable Sickle-cell disease Congenital acquired (elective or traumatic) dysfunction Splenectomy commonly type b, but occasionally a, e, or f) [23], or occasionally, these infections may be caused by nontypeable Haemophilus organisms. The rate of Haemophilus infection may also be increased in patients with splenectomy or sickle-cell disease, but these infections occur less frequently in such patients than does invasive pneumococcal infection. The rate of Haemophilus infection among healthy children has markedly decreased since the institution of routine infant immunization with Haemophilus conjugate vaccines after The number of cases of Haemophilus meningitis in the United States deceased from 112,000 cases per year in 1985 ( cases per 100,000 persons per year) to only 669 cases in 1994 and 1995 ( cases per 100,000 persons per year) [2]. However, these infections continue to occur among children from families who decline immunization and among children with acquired or congenital immunodeficiencies. In addition, certain racial or ethnic populations experience high case rates of Haemophilus infection, such as children of Native American or Native Alaskan descent, and, in immunized children, Haemophilus infection may be caused by previously unusual types, such as types a and f, rather than type b. Neonatal meningitis. Both Escherichia coli and group B streptococcal infections of neonates have been associated with a lack of specific maternal antibodies to either the K-1 capsule or capsular polysaccharide of streptococci, respectively. In addition, E. coli infection has been associated with certain congenital metabolic disorders, most frequently galactosemia [24, 25]. Before the routine use of chemoprophylaxis for colonized women, invasive group B streptococcal infections (Streptococcus agalactiae), including meningitis, were the most common cause of neonatal meningitis, affecting 1.8 million infants aged!90 days per 1000 live births (or 9600 infants annually) in the United States [26]. The susceptibility of neonates to infection with group B streptococci is correlated with deficiency of maternal (transplacental) specific antibody [27]. Listeria infection. Infection with Listeria monocytogenes has long been associated with conditions of immune compromise. In healthy populations, the highest rates of infections are observed in infants aged!1 month and adults aged 160 years. Pregnant women account for 27% of all cases and more than one-half of all infections in persons years of age. Almost 70% of nonperinatal infections occur in patients with hematologic malignancies, patients with HIV infection, organ transplant recipients, or patients who are receiving corticosteroid therapy [28]. The intracellular life cycle of Listeria species may explain the predilection for neonates, pregnant women, and immunocompromised hosts; thus, Listeria species avoid the extracellular immune environment (consisting of immunoglobulins and complement) characteristic of other meningitis pathogens. Patients with AIDS are most likely to contract listeriosis when CD4 T lymphocyte counts decrease to!40 cells/ml. 192 CID 2003:36 (15 January) CLINICAL PRACTICE

5 Finally, other immunodeficiencies, such as those associated with phagocyte dysfunction, were recently reviewed elsewhere [29]. These disorders share a common clinical pattern of recurrent mucocutaneous infections, pneumonia, and otitis media, with occasional suppurative infections of the viscera. However, meningitis is rarely associated with disorders of phagocytic dysfunction. Similarly, IgA deficiency, although it is a common disorder, is rarely associated with bacterial meningitis. IMMUNE EVALUATION OF PATIENTS WITH BACTERIAL MENINGITIS Evaluation for immunodeficiency is not indicated for most patients with bacterial meningitis, particularly when the disease occurs in a previously healthy individual, when there is not a history of recurrent infections before the episode of meningitis, or risk factors for HIV infection are not present. The single exception is meningococcal disease, which may be the presenting infection in a patient with an underlying deficiency or dysfunction of complement or properdin proteins. Some patients may have a history of previous neisserial infections, including localized or disseminated gonococcal infections. Because previous studies have demonstrated an incidence as high as 39% in populations of patients with meningococcal infections, at a minimum, a screening test for complement function (CH50) should be performed for all patients who have invasive meningococcal infections, and direct assessment of complement (C5, C6, C7, C8, and C9) and properdin proteins should be considered. Levels of IgG, IgM, and IgA should be measured in children who have a history of recurrent infections, including meningitis, sepsis, or recurrent upper and lower respiratory tract infections, before an episode of bacterial meningitis. All children who have a second episode of bacterial meningitis should be screened for congenital immunoglobulin or complement deficiencies. Elevated IgE levels may be associated with some immunodeficiencies (e.g., Wiskott-Aldrich syndrome), and isolated IgA deficiency may be associated with IgG-subclass deficiencies, such as IgG2 or IgG4 deficiencies. Such deficiencies have been associated with increased rates of pneumococcal and other encapsulated bacterial infections. In addition, serological testing for HIV infection should be considered for children who have bacterial meningitis if additional risk factors are present in the parents. Adults aged!50 years who present with pneumococcal disease or Listeria infection should also be screened for HIV infection. In general, some experts recommend an immune evaluation if a single episode of meningitis occurs at an age that is either earlier (e.g.,!6 months of age) or later (e.g., 3 4 years of age) than is characteristic for the causative meningeal pathogen. Quantitative assessment of B cells should be performed if total levels of IgM and/or IgG are low. In addition, early component complement deficiencies, particularly C2 or C3 deficiencies, may predispose children to pneumococcal or Haemophilus infections, and screening can be accomplished by an assessment of total complement function (CH50). However, if suspicion is high, measurement of individual complement components and properdin proteins should be considered. For children who are at risk for development of splenic dysfunction (e.g., sickle-cell disease) or congenital asplenia (e.g., congenital heart disease), splenic function should be evaluated by assessment for RBC pitting (e.g., the presence of Howell- Jolly bodies) and by direct functional-anatomic assessment with radionuclide splenic scanning with use of 99m Tc sulfur colloid or tagged heat-damaged RBCs. Careful assessment of the peripheral RBC smear may demonstrate pitting of the RBCs. In addition, the size and number of platelets should be noted, because patients with certain immunodeficiencies, such as Wiskott-Aldrich syndrome, may present with thrombocytopenia and abnormally small platelets. References 1. Schuchat A, Robinson K, Wenger JD, et al. Bacterial meningitis in the United States in N Engl J Med 1997; 337: Bisgard KM, Kao A, Leake J, et al. Haemophilus influenzae invasive diseases in the United States, : near disappearance of a vaccine-preventable childhood disease. Emerg Infect Dis 1998; 4: Ellison RT, Kohler PF, Curd JG, et al. Prevalence of congenital or acquired complement deficiency in patients with sporadic meningococcal disease. N Engl J Med 1983; 308: Mayatepek E, Grauer M, Hansch G, Sonntag HG. Deafness complement deficiencies and immunoglobulin status in patients with meningococcal diseases due to uncommon serogroups. Pediatr Infect Dis J 1993; 12: Ross SC, Densen P. Complement deficiency states and infection: epidemiology, pathogenesis and consequences of neisserial and other infections in an immune deficiency. Medicine (Baltimore) 1984; 63: Lehman TJA, Bernstein B, Hanson V, et al. Meningococcal infection complicating systemic lupus erythematosus. J Pediatr 1981; 99: Sjoholm AG, Kuijper EJ, Tijssen CC, et al. Dysfunctional properdin in a Dutch family with meningococcal disease. N Engl J Med 1988; 319: Densen P. Complement deficiencies and infection. In: Frank MM, Volanakis JE, eds. The human complement system in health and disease. New York: Marcel Dekker, 1998: Overturf GD. Technical report: prevention of pneumococcal infections, including the use of pneumococcal conjugate and polysaccharide vaccines and antibiotic prophylaxis. Committee on Infectious Diseases. Pediatrics 2000; 106: Overturf GD. Infections and immunizations of children with sickle cell disease. Adv Pediatr Infect Dis 1999; 14: Zarkowsky HS, Gallagher D, Gill FM, et al. Bacteremia in sickle hemoglobinopathies. J Pediatr 1986; 109: Styrt B. Infection associated with asplenia: risks, mechanisms, and prevention. Am J Med 1990; 88:33N 43N. 13. Waldman JD, Rosenthal A, Smith AL, et al. Sepsis and congenital asplenia. J Pediatr 1977; 90: Sullivan KE, Mullen CA, Blaese RM, Winklestein JA. A multiinstitu- CLINICAL PRACTICE CID 2003:36 (15 January) 193

6 tional survey of the Wiskott-Aldrich syndrome. J Pediatr 1994; 125: Levy J, Espanol-Boren T, Thomas C, et al. Clinical spectrum of X- linked hyper-igm syndrome. J Pediatr 1997; 131: Iseki M, Heiner DC. Immunodeficiency disorders. Pediatr Rev 1993; 14: Berthet F, Le Deist F, Duliege AM, et al. Clinical consequences and treatment of primary immunodeficiency syndromes characterized by functional T and B lymphocytes anomalies (combined immune deficiency). Pediatrics 1994; 93: Sorenson RU, Moore C. Antibody deficiency syndromes. Pediatr Clin North Am 2000; 47: Berger BJ, Hussain F, Roistacher K. Bacterial infections in HIV-infected patients. Infect Dis Clin North Am 1994; 8: Bernstein LJ, Krieger BZ, Novick B, et al. Bacterial infection in the acquired immunodeficiency syndrome of children. Pediatr Infect Dis 1985; 4: Nuorti JP, Butler JC, Gelling L, et al. Epidemiologic relation between HIV and invasive pneumococcal disease in San Francisco County, California. Ann Intern Med 2000; 132: Andimer WA, Mezger J, Shapiro E. Invasive bacterial infections in children born to women infected with human immunodeficiency virus. J Pediatr 1994; 124: Waggoner-Fountain LA, Hendley JO, Cody EJ, et al. The emergence of Haemophilus influenzae types e and f as significant pathogens. Clin Infect Dis 1995; 21: Levy HL, Sepe SJ, Shih VE, et al. Sepsis due to E. coli in neonates with galactosemia. N Engl J Med 1977; 297: Bingen E, Picard B, Brahimi N, et al. Phylogenetic analysis of Escherichia coli strains causing neonatal meningitis suggests horizontal gene transfer from a predominant pool of highly virulent B2 strains. J Infect Dis 1998; 177: Farley MM, Harvey C, Stull T, et al. A population-based assessment of invasive disease due to group B Streptococcus in nonpregnant adults. N Engl J Med 1993; 328: Baker CJ, Kasper DJ. Correlation of maternal antibody deficiency with susceptibility to neonatal group B streptococcal infection. N Engl J Med 1976; 294: Lorber B. Listeriosis. Clin Infect Dis 1997; 24: Lekstrom-Himes JA, Gallin JI. Immunodeficiency diseases caused by defects in phagocytes. N Engl J Med 2000; 343: CID 2003:36 (15 January) CLINICAL PRACTICE