PEDIATRICS Vol. 120 No. 4 October 2007, pp. e912-e921 (doi:10.1542/peds.2006-3150)
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ARTICLE |
Immune Function in Young Children With Previous Pulmonary or Miliary/Meningeal Tuberculosis and Impact of BCG Vaccination
a Division of Infectious Diseases, Department of Medicine
i Department of Biostatistics, Vanderbilt University School of Medicine, Nashville, Tennessee
b Academic Tuberculosis Program, Clementino Fraga Filho University Hospital
e Martagão Gesteira Pediatric Institute, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil
c Division of Pediatric Pneumology, Universidade Federal do Estado do Rio de Janeiro, Rio de Janeiro, Brazil
d Laboratory of Clinical Infectious Diseases, National Institutes of Health, Bethesda, Maryland
f Department of Pediatrics, Hospital San Sebastian, Rio de Janeiro, Brazil
g Department of Pediatric Neurosurgery, Hospital Sao Goncalo, Sao Goncalo, Brazil
h Department of Pediatric Neurosurgery, Hospital Municipal Jesus, Rio de Janeiro, Brazil
 |
ABSTRACT |
|---|
 TOP ABSTRACT PATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
OBJECTIVE. Children <5 years old are at increased risk of miliary/meningeal tuberculosis, but the immunologic factors that place them at risk are unknown. BCG vaccine protects against miliary/meningeal tuberculosis, but the mechanism of protection is unknown. We assessed for abnormalities in immune response associated with miliary/meningeal or pulmonary tuberculosis in young children.
PATIENTS AND METHODS. We conducted a case-control study among HIV-seronegative Brazilian children who were <5 years old. Case subjects had previous culture-confirmed or clinical miliary/meningeal tuberculosis. There were 2 sets of control subjects: those with culture-confirmed pulmonary tuberculosis and purified protein derivative–positive household contacts. All of the children had completed treatment. Peripheral blood mononuclear cells were stimulated (phytohemagglutinin, phytohemagglutinin + interleukin 12, lipopolysaccharide, lipopolysaccharide + interferon-
, and purified protein derivative), and cytokine responses (interleukin 1β, interleukin-4, interleukin-6, interleukin-8, interleukin 10, interleukin 12, interferon-, tumor necrosis factor-
, and monocyte chemoattractant protein 1) were quantified by bead-based assay. Median cytokine responses were compared by the Kruskal-Wallis test. Multivariate analysis of variance accounted for multiple comparisons.
RESULTS. There were 18 case subjects with miliary/meningeal tuberculosis, 28 pulmonary control subjects, and 29 purified protein derivative–positive control subjects. The median age was 4.2 years. There was no difference in case and control subjects by age, gender, race, BMI, or median CD4 count. Twelve (67%) of 18 case subjects, 26 (93%) of 28 pulmonary control subjects, and 28 (97%) of 29 purified protein derivative–positive subjects had received BCG vaccine. No cytokine defects were identified in case subjects with miliary/meningeal tuberculosis compared with either set of control subjects. Pulmonary control subjects had uniformly higher monocyte chemoattractant protein 1 levels than case subjects with miliary/meningeal tuberculosis and purified protein derivative–positive control subjects, both at rest and with lipopolysaccharide, lipopolysaccharide + interferon-, and purified protein derivative stimulation. Pulmonary control subjects did not have a higher frequency of allele G in the –2518 monocyte chemoattractant protein 1 promoter polymorphism. Case subjects with miliary/meningeal tuberculosis who had received BCG vaccine (n = 12) had lower stimulated interleukin 8 production than children who did not receive BCG vaccine (n = 6).
CONCLUSIONS. Children with previous miliary/meningeal tuberculosis did not have a major defect in the cytokine pathways studied. Increased monocyte chemoattractant protein 1 levels were associated with pulmonary disease, occurred despite BCG vaccination, and were not associated with a polymorphism in the monocyte chemoattractant protein 1 promoter.
Key Words: M tuberculosis • tuberculosis • tuberculosis meningitis • miliary tuberculosis • BCG vaccine • MCP-1
Abbreviations: IL—interleukin • MCP—monocyte chemoattractant protein • PPD—purified protein derivative • PBMC—peripheral blood mononuclear cell • IFN—interferon • TNF—tumor necrosis factor • MANOVA—multivariate analysis of variance • IQR—interquartile range
It is estimated that the lifetime risk of developing tuberculosis after infection with Mycobacterium tuberculosis in childhood is 
10%.1 Therefore, the human immune response to M tuberculosis infection prevents the development of disease in most persons. The factors that allow for progression to active disease among infected persons are not fully understood but are likely immunologic based on the increased rates of disease in persons with varying forms of immune compromise.2–5 The factors that predispose to the development of tuberculosis must be better understood so that persons at increased risk can be identified and receive interventions to prevent tuberculosis.
Extrapulmonary tuberculosis seems to be a marker of an underlying immune defect. The risk of extrapulmonary disease is increased in persons infected with HIV,2–4 particularly those with advanced immune suppression.5 Children are also at increased risk for extrapulmonary tuberculosis, presumably because of immature innate and adaptive immune responses, although the defects that increase tuberculosis risk in children are not well understood.2,3,6–8 Our current understanding of the immune response to M tuberculosis has been summarized in several recent reviews.9,10 Because the most severe manifestations of extrapulmonary tuberculosis include miliary and meningeal tuberculosis and because young (<5 years old) children are particularly prone to this form of disease,1,11–13 we hypothesized that if there is indeed an immunologic defect that predisposes to tuberculosis, it would most likely be identified among children with miliary or meningeal disease.
We previously noted defects in unstimulated interleukin 8 (IL-8) levels in adults who were seronegative for HIV with previous extrapulmonary tuberculosis.14 In a subsequent study we noted decreased CD4+ lymphocytes and a global defect in unstimulated cytokine production in adults who were seronegative for HIV with previous extrapulmonary tuberculosis compared with persons with previous pulmonary tuberculosis or latent M tuberculosis infection.15 This has not been assessed in young children, however. The pathogenesis of tuberculosis differs in children compared with adults, with children being more likely to have progressive primary disease than reactivation of latent M tuberculosis infection.16–18
BCG vaccine protects against miliary and meningeal disease in young children, but it may be less effective in protecting against pulmonary disease in either children or adults.19–24 The mechanism of protection against miliary/meningeal tuberculosis in young children is unknown, although BCG vaccine likely prevents dissemination of M tuberculosis after infection.25 A recent population-based study suggested that BCG vaccine may also prevent M tuberculosis infection in children.26 It is unknown why BCG may confer less protection against pulmonary disease than disseminated tuberculosis.
One of the immunologic factors associated with an increased risk of pulmonary tuberculosis is monocyte chemoattractant protein 1 (MCP-1; also known as CCL-2), a chemokine that recruits monocytes and T lymphocytes and is important in granuloma formation. In a recent study among persons with active tuberculosis, those with pulmonary disease had higher serum MCP-1 levels than those with extrapulmonary disease.27 In a study of adults with active pulmonary tuberculosis in Mexico and Korea, high levels of MCP-1, which seemed to suppress IL-12p40 production, were found both in plasma and after stimulation of monocytes with M tuberculosis antigens because of MCP-1 promoter genotypes AG and GG.28
We sought to determine whether abnormalities in immune response might be associated with miliary/meningeal and/or pulmonary tuberculosis in young children. Because of the profound defects in cellular immunity that can occur because of HIV infection, we restricted our study to persons seronegative for HIV. Because of the impact of active disease on cytokine responses, we studied only children who had received curative therapy and were not acutely ill with tuberculosis. To assess both innate and adaptive immune responses, both specific and nonspecific stimuli were used.
|
PATIENTS AND METHODS |
|---|
|
TOPABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study Population
Case and control subjects were identified and enrolled at Clementino Fraga Filho University Hospital (n = 37), San Sebastian Hospital (n = 15), Gafree Guinle Hospital (n = 13), Sao Goncalo Hospital (n = 5), and Martagão Gesteira Pediatric Institute (n = 5). All of the sites contributed both case and control subjects.
Eligibility criteria for patients with miliary/meningeal tuberculosis included children seronegative for HIV with a history of treated miliary or meningeal tuberculosis who were <5 years old at the time of diagnosis. Tuberculosis was either culture confirmed or based on the following clinical criteria: signs and symptoms of meningitis (headache, fever, and nuchal rigidity), close contact with a smear- or culture-positive tuberculosis case, abnormal computed tomography or MRI of the brain, funduscopic examination demonstrating granulomas, abnormal cerebrospinal fluid (elevated protein, decreased glucose, and presence of leukocytes), and clinical response to antituberculosis therapy alone. Exclusion criteria included serum creatinine >2 mg/dL, use of corticosteroids or other immunosuppressive agents before the time of diagnosis or at the time of study entry, malignancy, or diabetes mellitus. The criteria for pulmonary tuberculosis control patients included children seronegative for HIV who had completed treatment for culture-confirmed pulmonary tuberculosis and were <5 years old at diagnosis. Positive cultures of sputum, bronchoalveolar or gastric lavage, or pulmonary parenchyma were required. Control subjects with latent M tuberculosis infection were household contacts of smear- or culture-positive tuberculosis case subjects, <5 years old, HIV seronegative, and had a positive tuberculin skin test (defined as 
10 mm in duration after intradermal placement of 5 tuberculin units of purified protein derivative [PPD]) without evidence of active tuberculosis. Exclusion criteria for both control groups were the same as for the case group.
This study was approved by the institutional review boards of the Federal University of Rio de Janeiro (for Clementino Fraga Filho University Hospital, San Sebastian Hospital, Sao Goncalo Hospital, and Martagão Gesteira Pediatric Institute), the University of Rio de Janeiro (for Gafree Guinle Hospital), the Johns Hopkins Medical Institutions, Vanderbilt University Medical Center, and the National Institutes of Allergy and Infectious Diseases, National Institutes of Health. Parents or guardians of all of the study participants provided written informed consent.
Laboratory Methods
Peripheral blood mononuclear cells (PBMCs) were purified on site in Brazil within 24 hours of obtaining the specimens from study participants using density gradient separation from heparinized whole blood; 106 cells per milliliter were plated in 1 mL of complete RPMI 1640 medium (Mediatech Inc, Herndon, VA).14 Selected wells of the PBMC were stimulated for 48 hours at 37°C in 5% CO2 with phytohemagglutinin 1% (Life Technologies, Gaithersburg, MD); 1 ng/mL of phytohemagglutinin plus IL-12p70 heterodimer (R&D Systems, Minneapolis, MN); 200 ng/mL of Escherichia coli–derived lipopolysaccharide (Sigma, St Louis, MO); and 1000 U/mL of lipopolysaccharide plus interferon (IFN)- (Genentech, South San Francisco, CA). There was also a 48-hour unstimulated condition. PBMCs were stimulated with 10 µg/mL of PPD (Statens Serum Institut, Copenhagen, Denmark) for 96 hours,29 and a 96-hour unstimulated condition was also performed on the same number of cells. Culture supernatants from all of the above conditions were obtained and frozen at –70°C for subsequent cytokine determinations.
Cytokine Quantification
Cell culture supernatants were thawed once and examined for IL-1β, IL-4, IL-6, IL-10, IL-12p70, IFN-, tumor necrosis factor (TNF) , and MCP-1 concentrations in duplicate by multiplex cytokine array analysis performed using the Bio-Plex protein multiarray system, which uses Luminex-based technology as specified by the manufacturer (Bio-Rad Laboratories, Hercules, CA). IL-8 levels were determined by a commercial enzyme-linked immunosorbent assay (Meso Scale Discovery, Gaithersburg, MD) run in parallel. All of the cytokine determinations were done with the same lots of reagents. Laboratory personnel were blinded to the case-control status of the specimens.
Assessment for Polymorphism in MCP-1
The promoter polymorphism for MCP-1 (-2518 G/A; reference single-nucleotide polymorphism No. 1024611) was determined by using a predeveloped TaqMan allelic discrimination assay (No. C_2590362_10) from Applied Biosystems (Foster City, CA) as per the manufacturer's protocol. Assays were performed on the ABI 7500 real-time polymerase chain reaction system. In addition, the genotypes of 15 random samples were confirmed by sequencing. Polymerase chain reaction products were amplified by using the primers 5'-CCAGTATCTGGAATGCAGGC-3' (forward) and 5'-ACAGGGAAGGTGAAGGGTAT-3' (reverse). The sequences were analyzed with Sequencher 4.2 (Gene Codes, Ann Arbor, MI).
Statistical Analysis
The sample size was determined to detect a twofold difference in median cytokine production between case and control subjects with 80% power and a 2-tailed of .05. Clinical and demographic characteristics were compared among the 3 groups (miliary/meningeal tuberculosis, pulmonary tuberculosis, and PPD-positive) using the Kruskal-Wallis test for continuous variables and the 
2 and Fisher's exact tests for categorical variables. For the analysis of cytokine responses, cytokine concentrations read as "low" were counted as the lowest reading, –1. Concentrations read as "high" were counted as the highest reading, +1.
Given the possibility of type 1 error because of multiple comparisons of cytokine responses, we conducted a global test for the combined effect of each specific cytokine response across the five 48-hour stimulus conditions and separately across the two 96-hour conditions, comparing the 3 patient groups using multivariate analysis of variance (MANOVA).30 MANOVA does not require that the cytokine responses all be in the same direction for each stimulus condition. Because many variables were skewed, Box-Cox transformations were performed for all of the cytokine responses in the MANOVA to satisfy the multivariate normality assumption of regression residuals. The results of Kruskal-Wallis tests are presented for stimulus condition-cytokine response pairs only when the global test for the specific cytokine detected a significant effect among the 3 groups. Posthoc pairwise comparisons of the Kruskal-Wallis tests were not performed to minimize the risk of type 1 error.
2 analysis tested for deviation of genotype distribution from the Hardy-Weinberg equilibrium and compared allele or genotype frequencies between case and control subjects. MCP-1 genotype proportions were compared among pulmonary control subjects, case subjects with miliary/meningeal tuberculosis, and PPD-positive control subjects using Fisher's exact test. Allele frequencies for these comparisons were assessed by the 2 test. The Kruskal-Wallis test compared the distribution of MCP-1 responses according to MCP-1 genotype.
A 2-sided significance level of .05 was used for statistical inference. Statistical analyses were performed using Stata 8.2 (Stata Corp, College Station, TX), SAS 9.1 (SAS Institution, Cary, NC), and R 2.1.0 (R Project for Statistical Computing, Vienna, Austria).
|
RESULTS |
|---|
|
TOPABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
There were 18 case subjects with miliary/meningeal tuberculosis, 28 pulmonary tuberculosis control subjects, and 29 control subjects with latent M tuberculosis infection. The clinical and demographic characteristics of the study participants are in Table 1. The median age of all 75 of the participants at the time of study entry was 4.2 years (interquartile range [IQR]: 3.0–5.3 years). The median age at diagnosis was 5.4 months (IQR: 3.6–10.2 months) among case subjects with miliary/meningeal tuberculosis and 17.5 months (IQR: 13.7–24.5 months) among pulmonary control subjects. The median time between diagnosis and study entry was 47.2 months (IQR: 30.7–57.1 months) among case subjects with miliary/meningeal tuberculosis and 25.0 months (IQR: 17.6–37.3 months) in pulmonary control subjects. There were no differences in case and control subjects by age, gender, race, BMI, or median CD4+ or CD8+ lymphocyte count. Of the 18 case subjects with miliary/meningeal tuberculosis, 12 (67%) had been vaccinated with BCG; 26 (93%) of 28 pulmonary tuberculosis control subjects and 28 (97%) of 29 PPD-positive control subjects had received BCG vaccine (P = .006). Vaccination status according to site of disease is in Table 2. Half of the case subjects with miliary/meningeal tuberculosis had culture-confirmed disease.
|
|
Of the 75 study participants, all had 48-hour cytokine testing performed; all but 4 PPD-positive control subjects also had 96-hour cytokine testing performed. There were very few conditions with a substantial number of low or high cytokine responses. The conditions in which there were the most low cytokine responses were 48- and 96-hour IFN-
levels without stimulation. IFN- levels after stimulation with lipopolysaccharide plus IFN- were high because of the addition of IFN- and were omitted. IL-12 levels after stimulation with phytohemagglutinin plus IL-12 were also omitted.
Cytokine responses after 48 hours are provided in Table 3. There were no cytokine defects in case subjects with miliary/meningeal tuberculosis compared with either set of control subjects. MCP-1 levels were elevated in pulmonary tuberculosis control subjects under all of the test conditions, but the difference was statistically significant in the unstimulated condition and after stimulation with lipopolysaccharide and lipopolysaccharide plus IFN-.
|
At 96 hours, MCP-1 levels remained elevated in pulmonary tuberculosis control subjects, in both the unstimulated condition and after stimulation with PPD (Table 3). There were no other statistically significant differences in cytokine responses among the 3 patient groups. Exclusion of low and high cytokine readings resulted in median cytokine responses similar to those reported in Table 3, and statistical significance was not affected (data not shown).
Cytokine responses of miliary versus meningeal tuberculosis versus both sets of controls were then assessed. For this analysis, all of the persons with miliary disease were combined (n = 10) and compared against 8 persons with meningeal disease. In general there were no pronounced discrepancies in cytokine production between persons with these 2 clinical manifestations of tuberculosis. However, median MCP-1 levels were increased in children with previous meningeal disease after stimulation with lipopolysaccharide (P = .01) and lipopolysaccharide plus IFN- (P = .03), with levels that were comparable to those seen in pulmonary tuberculosis control subjects (data not shown).
We then assessed cytokine responses of case subjects with miliary/meningeal tuberculosis according to BCG vaccination status (Table 4). The only cytokine or chemokine with statistically significant differences in production was IL-8, which was uniformly lower in persons with previous BCG vaccination. These differences were statistically significant after stimulation with lipopolysaccharide (P = .01), lipopolysaccharide plus IFN- (P = .01), and PPD (P = .03).
|
In an analysis of the -2518 MCP-1 promoter polymorphism among pulmonary tuberculosis control subjects, there was no statistically significant difference in AG or GG genotype or the G allele compared with either patients with miliary/meningeal tuberculosis or PPD-positive control subjects (Tables 5 and 6). The genotype at the -2518 promoter polymorphism was in Hardy-Weinberg equilibrium (P = .20). In an analysis of MCP-1 levels according to MCP-1 genotype among all of the study patients regardless of case-control status, there were no statistically significant differences in the distribution of MCP-1 level according to AA, AG, or GG genotype for each of the conditions tested (Table 7).
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In this study of young children with previous miliary/meningeal tuberculosis, we found no defect in CD4+ lymphocyte number, unstimulated cytokine production, or stimulated cytokine production compared with either pulmonary tuberculosis or PPD-positive control subjects. In contrast, in a study in adults who were seronegative for HIV with previous extrapulmonary tuberculosis, we found mildly decreased CD4+ lymphocyte counts and uniformly lower unstimulated cytokine production.15 That the young children in this study differed from adults with extrapulmonary tuberculosis is not surprising, given that tuberculosis pathogenesis in young children likely differs significantly from that in adults. Young children usually have progressive primary disease after exposure to M tuberculosis, whereas adults are more likely to have reactivation of latent M tuberculosis infection.16–18,31–34 Therefore, those in whom tuberculosis reactivation occurs later in life likely successfully contained M tuberculosis infection for many years before disease occurrence, unlike the children in this study who probably developed disease as their first manifestation of infection.
Genetic disorders affecting the IFN-–IL-12–IL-23 synthesis and response pathways predispose to severe and disseminated mycobacterial disease due to nontuberculous mycobacteria.35 However, their role in severe tuberculosis has only been anecdotally reported and is less clear. We found that neither miliary/meningeal nor pulmonary tuberculosis was associated with functional defects in the IFN-–IL-12–IL-23 synthesis and response pathway. Furthermore, other pathways involved in the generation and regulation of the inflammatory response seemed to be intact. This is consistent with a recent study that found that intracerebral cytokine responses in persons seronegative for HIV with tuberculous meningitis were not closely associated with disease outcome.36 The clinical benefit in patients with tuberculous meningitis treated with corticosteroids was not associated with attenuation of peripheral or local immune responses to M tuberculosis antigens.37 Therefore, the predisposition to tuberculous meningitis, as well as interventions that ameliorate its course, may not be directly related to tested cytokine responses. Miliary tuberculosis is caused by lymphohematogenous dissemination of M tuberculosis. The pathophysiology is not completely understood but is felt to be associated with abnormalities in CD4+ lymphocytes, TNF-, and IFN-.38 Thus, the lack of abnormalities in peripheral CD4+ lymphocyte counts or cytokine production among the children in this study is intriguing.
We found consistently elevated MCP-1 levels in children with cured pulmonary tuberculosis under all of the conditions (Table 3). Hasan et al27 reported elevated lipopolysaccharide- and BCG vaccine–induced MCP-1 secretion in vitro in Pakistani adults with active pulmonary tuberculosis compared with adults with extrapulmonary tuberculosis and PPD+ controls. Flores-Villanueva et al28 studied a functional polymorphism in the MCP-1 promoter (-2518A/G) in Mexican and Korean adults with pulmonary tuberculosis compared with PPD-positive and PPD-negative control subjects. They found that the -2518G allele was associated with higher in vitro MCP-1 production and consequent IL-12p40 suppression. Both of these studies examined cells and plasma from adults with active tuberculosis compared with persons without acute illness, raising the possibility that acute illness, not tuberculosis, was responsible for the cytokine abnormalities. In contrast to the above studies, we focused on children, and studied only those who were cured of tuberculosis. Therefore, our demonstration of elevated MCP-1 levels in children with previous pulmonary but not miliary/meningeal tuberculosis extends the previous findings in acutely infected adults and suggests that MCP-1 levels may be inherently elevated in children who develop pulmonary tuberculosis (ie, a possible abnormality in innate immunity) compared with miliary/meningeal disease. It also suggests that the pathophysiology of pulmonary and extrapulmonary tuberculosis may differ.
However, in contrast to the study by Flores-Villanueva et al,28 the MCP-1 -2518G allele was not associated with pulmonary tuberculosis or increased MCP-1 levels in our study. Although our sample size was small, the frequency of the GG genotype among pulmonary tuberculosis patients was substantially lower among children in our study. This may reflect differences in distribution of the genotype in the study populations: Mexico and Korea versus Brazil. Interestingly, MCP-1 production from cells carrying the GG -2518 allele seemed to be more downregulated by IL-12 than either the AA or AG alleles (MCP-1 levels after stimulation with phytohemagglutinin versus phytohemagglutinin plus IL-12; Table 7).
Because 93% of the children with pulmonary tuberculosis had received BCG vaccines, BCG vaccines did not protect them against this form of the disease. If the elevated MCP-1 levels seen after developing tuberculosis were also present before vaccination, it might explain in part the poor BCG immunogenicity. Assuming that cytokine levels after recovery from tuberculosis are comparable to levels before developing disease, the elevated MCP-1 levels in our patients likely represent an abnormality in innate immune function that may have predisposed to the development of pulmonary tuberculosis. If true, it would be important to examine ways to create a tuberculosis vaccine that circumvented or reduced elevated MCP-1 levels. It is unlikely that BCG vaccination itself increased MCP-1 levels, because MCP-1 was not increased in the PPD-positive control subjects, 97% of whom received BCG vaccines.
Significantly fewer case subjects with miliary/meningeal tuberculosis received BCG vaccine than the other 2 patient groups, supporting the hypothesis that BCG vaccine protects against severe forms of tuberculosis in young children. Of the 18 case subjects with miliary/meningeal tuberculosis, only 12 had received BCG vaccine. Although not a primary end point of the study and although statistical power was limited, we assessed cytokine responses according to BCG vaccination status to generate hypotheses to be tested in subsequent studies (Table 4). IL-8 production was uniformly lower in children who had been BCG vaccinated than in those who had not, and the difference was statistically significant after stimulation with lipopolysaccharide, lipopolysaccharide plus IFN-, and PPD, all agents with strong myeloid agonism. These findings suggest that BCG vaccination may have exerted a persistent shift in IL-8 production or that the children who developed miliary/meningeal tuberculosis despite BCG vaccination had an underlying defect in IL-8 production that the BCG vaccine was unable to overcome. The Moreau BCG strain is used in Brazil; findings could differ with the Pasteur or Tokyo strains.
There were several limitations of this study. First, approximately half of the tuberculosis meningitis case subjects were culture negative. Ideally, only culture-confirmed case subjects would have been included to ensure that persons indeed had tuberculosis. However, current standards for the diagnosis of meningeal tuberculosis are often based on clinical criteria, such as those used for this study. Second, there is the issue of multiple comparisons because of the analysis of several cytokine responses. However, MANOVA was used to identify those cytokine measurements that were statistically significantly different across all of the stimuli tested and avoid inflation of type 1 error. It is possible that small differences in cytokine responses between groups could have become statistically significant with a larger sample size. Third, children had to survive their episode of tuberculosis to be included in this study. Therefore, we may have missed children with the most severe disease and immunologic defects. However, given the confounding impact of active tuberculosis on cytokine production and responses, we chose to eliminate the known bias of active disease to be able to identify fundamental underlying abnormalities. Fourth, we did not assess for helminth or other concomitant infections in study patients. Although such concomitant infections could have affected cytokine responses, study patients were not acutely ill at the time of the study. Fifth, systemic immune responses of PBMCs were assessed in this study rather than local immune responses in the lung or meninges. Cytokine responses at these sites of disease could differ from peripheral responses. Finally, data regarding the M tuberculosis strain that caused disease in study patients were not available; host immune response may vary according to M tuberculosis strain.
This study had several interesting findings. First, we found no defect in cytokine production or response, including the IFN-–IL-12–IL-23 pathway, in young children with the most severe form of extrapulmonary tuberculosis: miliary/meningeal disease. Therefore, factors other than those defects in the host immune response must be responsible for this disease manifestation. Second, MCP-1 levels were increased in children with previous pulmonary tuberculosis, consistent with reports in adults. However, this was not associated with the -2518G allele of MCP-1, which has been described in pulmonary tuberculosis with increased MCP-1 levels in adults. Taken together, these findings suggest that the pathogenesis of extrapulmonary versus pulmonary tuberculosis in young children may differ, but the pathogenesis of pulmonary tuberculosis in children may be similar to that in adults, with elevated MCP-1 levels predisposing to pulmonary disease. Large studies among in children and adults in the same population are needed. Third, almost all of the children who developed pulmonary tuberculosis had received BCG vaccination. Increased MCP-1 levels could potentially explain why BCG vaccine did not confer protection against pulmonary disease. This warrants further study, because it has potential implications for developing an improved tuberculosis vaccine.
|
ACKNOWLEDGMENTS |
|---|
This work was supported by the extramural (grant K23-AI01654 [to Drs Sterling, Martire, and Silva de Almeida and Ms Torres]) and intramural (to Drs Ding, Greenberg, Elloumi, and Holland) programs of the National Institutes of Allergy and Infectious Diseases, National Institutes of Health.
We thank Richard E. Chaisson, MD (Johns Hopkins University). We also thank Elizabeth Souza, MD (Rio de Janeiro), Sheila Lucena, MD (Clementino Fraga Filho University Hospital), Anna Marques, MD (Hospital Municipal Jesus), Laurinda Higa, MD (Clementino Fraga Filho University Hospital), Monica Tura (Martagão Gesteira Pediatric Institute), and Ms Conceição (Gafree Guinle Hospital), all of whom assisted in patient recruitment.
|
FOOTNOTES |
|---|
Accepted Mar 3, 2007.
Address correspondence to Timothy R. Sterling, MD, Division of Infectious Diseases, Vanderbilt University Medical Center, A2209 Medical Center North, 1161 21st Ave South, Nashville, TN 37232-2605. E-mail: timothy.sterling{at}vanderbilt.edu
The authors have indicated they have no financial relationships relevant to this article to disclose.
This work was presented in part at the American Thoracic Society International Conference; May 19–24, 2006; San Diego, CA. Abstract A391.
Drs Sterling and Martire contributed equally to this work.
|
REFERENCES |
|---|
|
TOPABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1. Comstock GW, Livesay VT, Woolpert SF. The prognosis of a positive tuberculin reaction in childhood and adolescence. Am J Epidemiol. 1974;99 :131 –138
2. Gonzalez OY, Adams G, Teeter LD, Bui TT, Musser JM, Graviss EA. Extra-pulmonary manifestations in a large metropolitan area with a low incidence of tuberculosis. Int J Tuberc Lung Dis. 2003;7 :1178 –1185[Web of Science][Medline]
3. Ong A, Creasman J, Hopewell PC, et al. A molecular epidemiological assessment of extrapulmonary tuberculosis in San Francisco. Clin Infect Dis. 2004;38 :25 –31[CrossRef][Web of Science][Medline]
4. Yang Z, Kong Y, Wilson F, et al. Identification of risk factors for extrapulmonary tuberculosis. Clin Infect Dis. 2004;38 :199 –205[CrossRef][Web of Science][Medline]
5. Jones BE, Young SMM, Antoniskis D, Davidson PT, Kramer F, Barnes PF. Relationship of the manifestations of tuberculosis to CD4 counts in patients with human immunodeficiency infection. Am Rev Respir Dis. 1993;148 :1292 –1297[Web of Science][Medline]
6. Wallgren A. The time-table of tuberculosis. Tubercle. 1948;29 :245 –251[Medline]
7. Rieder HL, Snider DE Jr, Cauthen GM. Extrapulmonary tuberculosis in the United States. Am Rev Respir Dis. 1990;141 :347 –351[Web of Science][Medline]
8. Lewinsohn DA, Gennaro ML, Scholvinck L, Lewinsohn DM. Tuberculosis immunology in children: diagnostic and therapeutic challenges and opportunities. Int J Tuberc Lung Dis. 2004;8 :658 –674[Web of Science][Medline]
9. van Crevel R, Ottenhoff TH, van der Meer JW. Innate immunity to Mycobacterium tuberculosis.
Clin Microbiol Rev. 2002;15
:294
–309
10. Orme I. Adaptive immunity to mycobacteria. Curr Opin Microbiol. 2004;7 :58 –61[CrossRef][Web of Science][Medline]
11. Lincoln EM, Sordillo SVR, Davies PA. Tuberculous meningitis in children. J Pediatr. 1960;57 :807 –823[CrossRef][Web of Science][Medline]
12. Kennedy DH, Fallon RJ. Tuberculous meningitis.
JAMA. 1979;241
:264
–268
13. Hussey G, Chisolm T, Kibel M. Miliary tuberculosis in children: a review of 94 cases. Pediatr Infect Dis J. 1991;10 :832 –836[Web of Science][Medline]
14. Sterling TR, Dorman SE, Chaisson RE, et al. HIV-seronegative adults with extrapulmonary tuberculosis have abnormal innate immune responses. Clin Infect Dis. 2001;33 :976 –982[CrossRef][Web of Science][Medline]
15. Antas PRZ, Ding L, Hackman J, et al. Decreased CD4+ lymphocytes and innate immune responses in adults with previous extrapulmonary tuberculosis. J Allerg Clin Immunol. 2006;117 :916 –923[CrossRef][Web of Science][Medline]
16. Feja K, Saiman L. Tuberculosis in children. Clin Chest Med. 2005;26 :295 –312, vii[CrossRef][Web of Science][Medline]
17. Alcais A, Fieschi C, Abel L, Casanova JL. Tuberculosis in children and adults: two distinct genetic diseases.
J Exp Med. 2005;202
:1617
–1621
18. Starke JR. Tuberculosis in children. Semin Respir Crit Care Med. 2004;25 :353 –364[CrossRef][Web of Science][Medline]
19. Rodrigues LC, Diwan VK, Wheeler JG. Protective effect of BCG against tuberculous meningitis and miliary tuberculosis: a meta-analysis.
Int J Epidemiol. 1993;22
:1154
–1158
20. Colditz GA, Berkey CS, Mosteller F, et al. The efficacy of bacillus Calmette-Guerin vaccination of newborns and infants in the prevention of tuberculosis: meta-analysis of the published literature.
Pediatrics. 1995;96
:29
–35
21. Colditz GA, Brewer TF, Berkey CS, et al. Efficacy of BCG vaccine in the prevention of tuberculosis.
JAMA. 1994;271
:698
–702
22. Comstock GW. Field trials of tuberculosis vaccines: how could we have done them better. Control Clin Trials. 1994;15 :247 –276[CrossRef][Web of Science][Medline]
23. ten Dam HG, Pio A. Pathogenesis of tuberculosis and effectiveness of BCG vaccination. Tubercle. 1982;63 :225 –233[Web of Science][Medline]
24. Aronson NE, Santosham M, Comstock GW, et al. Long-term efficacy of BCG vaccine in American Indians and Alaska Natives: a 60-year follow-up study.
JAMA. 2004;291
:2086
–2091
25. Sutherland I, Lindgren I. The protective effect of BCG vaccination as indicated by autopsy studies. Tubercle. 1979;60 :225 –231[CrossRef][Web of Science][Medline]
26. Soysal A, Millington KA, Bakir M, et al. Effect of BCG vaccination on risk of Mycobacterium tuberculosis infection in children with household tuberculosis contact: a prospective community-based study. Lancet. 2005;366 :1443 –1451[CrossRef][Web of Science][Medline]
27. Hasan Z, Zaidi I, Jamil B, Khan MA, Kanji A, Hussain R. Elevated ex vivo monocyte chemotactic protein-1 (CCL2) in pulmonary as compared with extra-pulmonary tuberculosis. BMC Immunol. 2005;6 :14[Medline]
28. Flores-Villanueva PO, Ruiz-Morales JA, Song CH, et al. A functional promoter polymorphism in monocyte chemoattractant protein-1 is associated with increased susceptibility to pulmonary tuberculosis.
J Exp Med. 2005;202
:1649
–1658
29. Lopez-Maderuelo D, Arnalich F, Serantes R, et al. Interferon-gamma and interleukin-10 gene polymorphisms in pulmonary tuberculosis.
Am J Respir Crit Care Med. 2003;167
:970
–975
30. Cooley WW, Lohnes PR. Multivariate Data Analysis. New York, NY: John Wiley & Sons; 1971
31. Small PM, Hopewell PC, Singh SP, et al. The epidemiology of tuberculosis in San Francisco: a population-based study using conventional and molecular methods.
N Engl J Med. 1994;330
:1703
–1709
32. Alland D, Kalkut GE, Moss AR, et al. Transmission of tuberculosis in New York City: an analysis by DNA fingerprinting and conventional epidemiologic methods.
N Engl J Med. 1994;330
:1710
–1716
33. Bishai WR, Graham NMH, Harrington S, et al. Molecular and geographic patterns of tuberculosis transmission after 15 years of directly observed therapy.
JAMA. 1998;280
:1679
–1684
34. Jasmer RM, Hahn JA, Small PM, et al. A molecular epidemiologic analysis of tuberculosis trends in San Francisco, 1991–1997.
Ann Intern Med. 1999;130
:971
–978
35. Rosenzweig SD, Holland SM. Defects in the interferon-gamma and interleukin-12 pathways. Immunol Rev. 2005;203 :38 –47[CrossRef][Web of Science][Medline]
36. Simmons CP, Thwaites GE, Quyen NT, et al. Pretreatment intracerebral and peripheral blood immune responses in Vietnamese adults with tuberculous meningitis: diagnostic value and relationship to disease severity and outcome.
J Immunol. 2006;176
:2007
–2014
37. Simmons CP, Thwaites GE, Quyen NT, et al. The clinical benefit of adjunctive dexamethasone in tuberculous meningitis is not associated with measurable attenuation of peripheral or local immune responses.
J Immunol. 2005;175
:579
–590
38. Sharma SK, Mohan A, Sharma A, Mitra DK. Miliary tuberculosis: new insights into an old disease. Lancet Infect Dis. 2005;5 :415 –430[CrossRef][Web of Science][Medline]
PEDIATRICS (ISSN 1098-4275). ©2007 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
