The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Simmons, C. P.
Right arrow Articles by Farrar, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Simmons, C. P.
Right arrow Articles by Farrar, J.
The Journal of Immunology, 2006, 176: 2007-2014.
Copyright © 2006 by The American Association of Immunologists

Pretreatment Intracerebral and Peripheral Blood Immune Responses in Vietnamese Adults with Tuberculous Meningitis: Diagnostic Value and Relationship to Disease Severity and Outcome

Cameron P. Simmons1,*, Guy E. Thwaites*, Nguyen Than Ha Quyen*, Estee Torok*, Dang Minh Hoang*, Tran Thi Hong Chau{ddagger}, Pham Phuong Mai{ddagger}, Nguyen Thi Ngoc Lan{dagger}, Nguyen Huy Dung{dagger}, Hoang Thi Quy{dagger}, Nguyen Duc Bang{dagger}, Tran Tinh Hien{ddagger} and Jeremy Farrar*

* Oxford University Clinical Research Unit, Hospital for Tropical Diseases, {dagger} Pham Ngoc Thach Hospital for Tuberculosis and Lung Disease, and {ddagger} Hospital for Tropical Diseases, Ho Chi Minh City, Vietnam


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Tuberculous meningitis (TBM) is the most devastating form of tuberculosis. Both intracerebral and peripheral blood immune responses may be relevant to pathogenesis, diagnosis, and outcome. In this study, the relationship between pretreatment host response, disease phenotype, and outcome in Vietnamese adults with TBM was examined. Before treatment, peripheral blood IFN-{gamma} ELISPOT responses to the Mycobacterium tuberculosis Ags ESAT-6, CFP-10, and purified protein derivative (PPD) were a poor diagnostic predictor of TBM. Cerebrospinal fluid IL-6 concentrations at presentation were independently associated with severe disease presentation, suggesting an immunological correlate of neurological damage before treatment. Surprisingly however, elevated cerebrospinal fluid inflammatory cytokines were not associated with death or disability in HIV-negative TBM patients at presentation. HIV coinfection attenuated multiple cerebrospinal fluid inflammatory indices. Low cerebrospinal fluid IFN-{gamma} concentrations were independently associated with death in HIV-positive TBM patients, implying that IFN-{gamma} contributes to immunity and survival. Collectively, these results reveal the effect of HIV coinfection on the pathogenesis of TBM and suggest that intracerebral immune responses, at least in HIV-negative cases, may not be as intimately associated with disease outcome as previously thought.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Tuberculous meningitis (TBM)2 is the most devastating form of Mycobacterium tuberculosis infection and is associated with ~30% mortality and disabling neurological sequelae in ~25% of survivors (1), despite appropriate antibiotic and adjunctive therapy. Although the pathophysiological processes leading to TBM are incompletely understood, the disease is presumed to result from the hematogenous dissemination of M. tuberculosis following primary infection of lung tissue to the CNS. However, TBM does not arise by direct hematogenous invasion of the meninges by the bacilli, but by inoculation from contiguous lesions resulting from earlier bacillemia (2). These lesions, called Rich foci, are usually subpial or subependymal and most commonly situated in the sylvian fissure; their rupture introduces bacteria into the subarachnoid space and heralds the onset of TBM. The consequent cerebral pathology and high case fatality has long been considered dependant upon the severity of the inflammatory response. The ‘immunopathogenesis’ hypothesis has formed the rationale for adjunctive corticosteroid therapy for TBM for >50 years (3); the diverse immunomodulatory effects of corticosteroids might attenuate inflammation in the subarachnoid space and improve clinical outcomes. Accordingly, we recently demonstrated that adjunctive dexamethasone improved mortality, but not disability, in Vietnamese adults with TBM (1). Surprisingly, however, adjunctive corticosteroids were not associated with significant attenuation of inflammatory indices in the cerebrospinal fluid (CSF) or peripheral blood (4). These findings suggested other physiological events in the intracranial space could be better correlates of disease presentation or outcome.

Nonetheless, it remains clear that a long-lived inflammatory and anti-inflammatory response exists in the subarachnoid space of TBM patients (5, 6, 7, 8). Acutely, there is a cellular pleocytosis consisting predominantly of activated CD3+CD4+ T cells in the CSF (4). Concentrations of inflammatory indices such as TNF, IFN-{gamma}, IL-8, IL-10, matrix metalloproteases, and tissue inhibitors of matrix metalloproteases are elevated acutely in the CSF of TBM patients and remain persistently high during the first weeks of therapy (6, 7, 8, 9). Of these inflammatory markers, TNF has been implicated as determining, at least in part, the extent of disease pathogenesis (10). The blood-brain barrier is also compromised, and raised CSF lactate and low CSF glucose concentrations suggest anaerobic metabolic activity in cerebral tissues (8). Before treatment, the relationship between this spectrum of inflammatory indices and clinical presentation or outcome has not been well established.

Immune responses to M. tuberculosis are not restricted to the subarachnoid space during TBM. T cell responses to mycobacterial Ags, which are crucial in mediating resistance to M. tuberculosis infection, are detectable in the peripheral blood of TBM patients (4). Of particular importance are T cells specific to the M. tuberculosis-secreted Ags 6-kDa early secretory antigenic target 6 (ESAT-6) and culture filtrate protein 10 (CFP-10). These Ags are major and possibly protective targets of T cell responses in M. tuberculosis-infected individuals (11, 12). Epidemiological studies have indicated that detectable IFN-{gamma} ELISPOT responses to these Ags are strongly and specifically associated with both latent M. tuberculosis infection and clinically apparent pulmonary and pleural tuberculosis (TB) (13, 14, 15, 16). In the setting of TBM, a better understanding of the frequency and magnitude of ESAT-6- and CFP-10-specific T cells in peripheral blood could potentially be useful in the diagnosis, monitoring of treatment, and prediction of outcome.

The aim of the current study was 2-fold. First, to assess the utility of peripheral blood ESAT-6- and CFP-10-specific IFN-{gamma} ELISPOT assays in the diagnosis of TBM. Second, to identify CSF immunological and biochemical correlates of disease severity and outcome in HIV-seropositive and -seronegative TBM patients. The results suggest that ESAT-6 and CFP-10 IFN-{gamma} ELISPOT assays have limited value in the diagnosis of TBM. In addition, CSF inflammatory cytokine concentrations in HIV-negative TBM patients were not strongly associated with disease severity at presentation or with death and disability. However, an important finding was the relationship between CSF concentrations of IFN-{gamma} and outcome in HIV-associated TBM cases.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Patient population and setting

Study subjects were recruited from Pham Ngoc Thach Hospital and the Hospital for Tropical Diseases (HTD) in Ho Chi Minh City, Vietnam. Both hospitals serve the local community and provide tertiary care for cases of severe TB (Pham Ngoc Thach Hospital) and infectious disease (HTD) in southern Vietnam. Studies on pretreatment CSF were performed in patients (>14 years) who were recruited into a randomized controlled trial of adjunctive dexamethasone in adults with TBM between April 2001 and April 2003 (1). Only patients over 14 years of age with clinical evidence of meningitis (defined as the combination of nuchal rigidity and CSF abnormalities) were eligible to enter the study. TMB was defined as "definite" if acid-fast bacilli were seen in the CSF. It was defined as "probable" in patients with one or more of the following: suspected active pulmonary TB on chest radiography, acid-fast bacilli found in any specimen other than the CSF, and clinical evidence of other extrapulmonary TB. TMB was defined as "possible" in patients with at least four of the following: a history of TB, predominance of lymphocytes in the CSF, a duration of illness of more than 5 days, a ratio of CSF glucose to plasma glucose of <0.5, altered consciousness, yellow CSF, or focal neurological signs. Patients were reclassified on discharge as having definite TMB if acid-fast bacilli were seen or M. tuberculosis was cultured from the CSF or as not having TMB if another diagnosis was confirmed by microbiological or histopathological evaluation.

Patients were not eligible to enter the trial if the enrolling physician believed that corticosteroids were contraindicated, if the patient had received more than one dose of any corticosteroid or >30 days of anti-TB chemotherapy immediately before study entry, or if the consent of either the patient or the patient’s relatives was not obtained.

In total, 545 patients entered the clinical trial (73 eligible cases excluded for reasons defined previously (1)). Pretreatment investigations of CSF were performed in 497 of the 545 enrolled cases. However, not all of the biochemical, hematological, or immunological parameters measured as part of this study could be investigated in every sample, primarily because there was insufficient CSF volume. Similarly, some clinical parameters (e.g., opening pressure) were not recorded for all cases.

Studies of peripheral blood T cell responses ELISPOTs were performed in a separate group of 47 HIV-seropositive and -seronegative TBM patients admitted consecutively to the HTD between June 2003 and May 2005. These cases were recruited into descriptive and pharmacokinetic studies of TBM. The enrollment criteria and definition of possible, probable, and definite TBM were as described above for the steroid trial with the addition that probable TBM also included computed tomography or magnetic resonance imaging brain scan evidence of TBM. All adults were tested for Abs to HIV 1/2.

The ethical and scientific committees of both hospitals, the Health Services of Ho Chi Minh City, and the Oxford Tropical Research Ethics Committee approved the study protocols. Written informed consent to participate in the study was obtained from all patients or their relatives. In unconscious HIV-positive patients, the consent of two independent physicians was deemed acceptable.

Treatment

All patients in whom CSF parameters were investigated received anti-TB drugs and were randomized to adjunctive dexamethasone according to previously described treatment regimens and protocols (1). None of the patients received antiretroviral drugs. All adults recruited to this study after April 2003 received adjunctive dexamethasone.

Clinical investigations and assessment of outcome

Disease severity at presentation was assessed by the British Medical Research Council (BMRC) grading system (17): Grade I had a Glasgow coma score (GCS) of 15/15 with no focal neurological signs, grade II either had a GCS 11–14 or GCS 15 with focal neurological signs, grade III had a GCS of ≤10. Serial, paired, CSF and peripheral blood samples were collected before treatment. Concentrations of total and differential CSF leukocytes, lactate, glucose, and protein were measured by standard methods. Routine clinical investigations were performed on the same day the specimen was taken; all other measurements were performed on aliquots of CSF supernatant frozen at –70°C. CSF supernatants samples were typically frozen within 3 h of collection. Absolute CD4 counts were performed on a flow cytometer (BD Biosciences).

The primary outcome measure was death or severe neurological disability 9 mo after study entry. Disability was assessed after 1, 2, 6, and 9 mo of treatment by the Rankin and Simple questions scores (1).

Cytokine and chemokine measurements and assessment of the blood-brain barrier

Cytokines (IL-6, IL-10, IL-1beta, TNF, IL-8, IL-12p70) and chemokines (IFN-{gamma}-inducible protein 10 (IP-10), MCP-1, RANTES, and monokine induced by IFN-{gamma} (MIG) were measured using a cytometric bead array assay (BD Biosciences) according to the manufacturer’s instructions, with one modification: all samples were fixed in 4% paraformaldehyde before analysis. All measurements were performed in duplicate and the results expressed as the mean value. The mean limits of detection for the cytokines and chemokines were: IL-6, 4.3 pg/ml; IL-10, 4.8 pg/ml; IL-1beta, 2.8 pg/ml; TNF, 4.2 pg/ml; IL-8, 5 pg/ml; IL-12p70, 5.1 pg/ml; IP-10, 6 pg/ml; MCP-1, 5.4 pg/ml; RANTES, 4.6 pg/ml; and MIG, 4.8 pg/ml. For the purposes of analysis, samples with cytokine concentrations below the detectable limit were given the value 0.1 pg/ml. To measure interassay variation, three different control samples (derived from pooled CSF) were tested in every assay to gauge the extent of variation. The maximum coefficient of variation recorded for any one cytokine was 12.3%. Blood-brain barrier integrity was assessed by measurement of paired CSF and plasma albumin concentrations by standard methods with calculation of the albumin index using the formula [albumin csf]/[albumin plasma].

ELISPOT assays

IFN-{gamma} ELISPOT assays were used to measure the frequency of peripheral blood T cells producing IFN-{gamma} in response to PPD (Serum Staten) and overlapping 15-mer peptides (overlapping by 10 aa) spanning the M. tuberculosis ESAT-6 and CFP-10 Ags. The assay was performed according to the manufacturer’s instructions (Mabtech). Briefly, a total of 1–3 x 105 PBMC was added in 100 µl of culture medium per well along with either PPD (5 µg/ml), pooled ESAT-6, or CFP-10 peptides (final concentration per peptide: 1 µg/ml) or the positive control PHA (5 µg/ml). After overnight incubation and assay development, the number of spot forming units (SFU) in each well was counted using a dissecting microscope. The cutoff for a positive response in the ELISPOT assay was calculated using the formula: mean SFU from wells without Ag + (3.2 x SE) to achieve a confidence interval of 99% (18). ELISPOT responses in positive cases were expressed as SFU per million PBMC after subtraction of background responses (no Ag stimulation).

Statistics

Immunological variables associated with particular clinical presentations or outcomes were assessed by univariate and multivariate analysis. The Mann-Whitney U test was used to compare continuous parameters between two or more groups of patients; the {chi}2 test (or Fisher’s exact test) was used for comparison of categorical variables. Values of p were corrected for multiple testing by multiplying the p value by the number of tests performed (Bonferroni’s correction). Spearman’s rank correlation was used to measure correlations between ordinal variables. Variables identified in univariate analysis as associated (p < 0.1) with the outcome variable were then incorporated into multivariate logistic regression. Forward stepwise variable selection procedure was used to identify independent predictors of outcome (with p-to-enter 0.05, p-to-remove 0.01). All analyses were performed using SPSS version 10.0 software (Microsoft; SPSS).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
ESAT-6- and CFP-10-specific IFN-{gamma} ELISPOT peripheral blood responses in the immunodiagnosis of TBM

Measurable IFN-{gamma} ELISPOT responses to ESAT-6 are considered a sensitive and specific diagnostic measure of latent and some forms of clinically apparent M. tuberculosis infection (13, 14, 15, 16). The relevance of ESAT-6-specific IFN-{gamma} ELISPOT responses in the peripheral blood for the diagnosis of TBM is unknown. To address this, PPD and overlapping 15-mer peptides spanning the M. tuberculosis ESAT-6 and CFP-10 Ags were used in IFN-{gamma} ELISPOT assays against pretreatment PBMC samples from TBM cases. Subjects in this study were recruited as part of a pharmacokinetic study and a clinical descriptive study at the Hospital for Tropical Diseases. TBM patients comprised 19 HIV-negative cases, 14 (74%) of whom had M. tuberculosis cultured from their CSF, and 28 HIV-positive cases, of whom 25 (89%) were culture confirmed. The baseline clinical details of these cases are shown in Table I. As controls, we used healthy Vietnamese blood donors (university students; n = 53).


View this table:
[in this window]
[in a new window]
  		Table I. Baseline clinical variables of 47 adults with TBM in whom ELISPOT responses were measured

 
Before treatment, <60% of HIV-negative TBM cases made IFN-{gamma} ELISPOT responses to ESAT-6, CFP-10, or PPD alone (Fig. 1A). When the number of responders to either ESAT-6 and/or CFP-10 was combined, the sensitivity was 58% (Fig. 1A). Among culture-confirmed HIV-negative TBM cases, just 50% of cases made responses to ESAT-6 and/or CFP-10 (Fig. 1A). In HIV-positive TBM cases, <60% made IFN-{gamma} ELISPOT responses to ESAT-6, CFP-10, or PPD alone (Fig. 1B). Among culture-confirmed HIV-positive TBM cases, 64% of cases made responses to ESAT-6 and/or CFP-10 (Fig. 1B). Among healthy blood donors, ESAT-6-, CFP-10-, and PPD-specific responses were detected in 54, 49, and 62% of donors, respectively. When the number of healthy responders to either ESAT-6 or CFP-10 was combined, the response rate was 62%.


Figure 1
View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 1. IFN-{gamma} ELISPOT responses to ESAT-6, CFP-10, and PPD. The graphs show the percentage of (A) HIV-negative (n = 19) and (B) HIV-positive (n = 28) TBM cases with responses to ESAT-6 and/or CFP-10, ESAT-6 alone, CFP-10 alone, or PPD. In each graph, the percentage of responders is shown as a proportion of clinically defined cases ({square}) and microbiologically confirmed cases ({blacksquare}). There were 25 culture-confirmed HIV-positive cases and 14 culture-confirmed HIV-negative cases.

 
Comparison of the frequency of ESAT-6-, CFP-10-, or PPD-specific T cells in individuals who responded to these Ags did not reveal significant differences in the magnitude of responses detected in TBM patients or healthy donors (Fig. 2).


Figure 2
View larger version (16K):
[in this window]
[in a new window]
  		FIGURE 2. Frequency of IFN-{gamma} ELISPOT responses to ESAT-6, CFP-10, and PPD. The graph depicts the frequency of ESAT-6-, CFP-10-, and PPD-specific IFN-{gamma} SFU per million PBMC detected in individual healthy donors, HIV-negative, and HIV-positive TBM cases.

 
The collection of blood lymphocyte data in HIV-positive cases (but not HIV-negative cases since it was not clinically indicated) allowed us to relate ELISPOT responses to T cell counts. Blood CD3+CD4+ and CD3+CD8+ T cell counts in HIV-positive patients that responded to ESAT-6, CFP-10, or PPD were generally higher than in patients who did not respond to these Ags (Table II). In particular, there were significantly more CD3+CD4+ and CD3+CD8+ T cells in cases that responded to PPD and significantly more CD3+CD8+ T cells in cases that responded to CFP-10 (Table II). These data suggest profound T lymphopenia may diminish the sensitivity of the ELISPOT assay.


View this table:
[in this window]
[in a new window]
  		Table II. Baseline CD4+ T cell counts (mean (±SE)) in HIV-positive TBM patients stratified according to whether they responded to Ag stimulation in the ELISPOT assaya

 
Relationship between clinical presentation and the acute intracerebral inflammatory response in HIV-negative TBM cases

Subjects in this study were recruited as part of a randomized controlled trial of adjunctive dexamethasone in adults with TBM (1). In this study, the severity of TBM at presentation was assessed by BMRC grade (17), which has been shown in numerous studies in different populations to be strongly predictive of outcome (1, 8, 19). We hypothesized that an excessive intracerebral inflammatory immune response would be associated with BMRC grade in both HIV-negative and HIV-positive TBM patients. To address this hypothesis, inflammatory indices in pretreatment CSF samples from 497 adults with a clinical diagnosis of TBM were measured. The baseline characteristics of the study population are shown in Table III. Immune responses in HIV-negative and HIV-positive patients were analyzed separately.


View this table:
[in this window]
[in a new window]
  		Table III. Baseline clinical variables of 497 adults with TBM

 
At presentation, CSF concentrations of lactate and protein were positively associated with increasing BMRC disease grade in HIV-negative patients (Fig. 3, A and B). There was also a trend for CSF IL-6 and disease grade (Fig. 3C). Other intracerebral parameters, such as opening pressure (n = 272), CSF:blood glucose ratio (n = 405), CSF white cell count (n = 408), CSF IL-10 (n = 141), CSF IL-12p70 (n = 141), IL-1beta (n = 141), TNF (n = 141), IL-8 (n = 142), IFN-{gamma} (n = 142), IP-10 (n = 24), MCP-1 (n = 23), RANTES (n = 23), and MIG (n = 24) concentrations were unrelated to disease grade. Among clinical parameters, the length of illness before presentation (Fig. 3D) and evidence of active, pulmonary TB was also significantly associated with BMRC grade III TBM (p = 0.02, {chi}2). Other parameters, such as radiographic evidence of miliary TB, clinical extrapulmonary TB, or a history of TB therapy, were not significantly associated (by univariate analysis) with presentation in BMRC grade III.


Figure 3
View larger version (29K):
[in this window]
[in a new window]
  		FIGURE 3. Correlation between disease grade at presentation and select CSF parameters and length of illness prior to presentation. Shown are the magnitude of CSF protein (A), CSF lactate (B), CSF IL-6 (C), and the duration of illness in HIV-negative patients (D) presenting in BMRC grade I, II, or III. The linear regression line is shown in each graph. Correlation coefficients (r) and p values were defined using Spearman’s rank correlation analysis.

 
Logistic regression was used to determine whether length of history before presentation, evidence of pulmonary TB, and CSF concentrations of IL-6 and protein were independently associated with grade III TBM. Lactate was omitted from the model because there were insufficient measurements. Logistic regression using data from 140 cases indicated only IL-6 was independently associated with grade III (p = 0.017, odds ratio (95% confidence interval (CI)), 2.1 (1.1–3.8), n = 140).

Relationship between the acute intracerebral inflammatory response in HIV-negative TBM patients and death

Excessive immune activation in the subarachnoid space is thought to be associated with poor outcome in TBM; hence, the use of steroids as adjunctive therapy. Previous work from our group has documented the significant relationship between death from TBM and a relatively low CSF white cell count and a low CSF:blood glucose ratio at presentation (1). In this study, the relationship between death and other CSF inflammatory indices was investigated. The frequency with which IL-12p70 was detected in CSF at presentation was significantly greater in patients who died (4 of 33, 12%) than in those who survived ((3 of 112, 2.5%; p < 0.01, {chi}2 test,), although this was not significant after correction for multiple testing. However, other parameters, such as opening pressure (n = 272), CSF protein (n = 410), CSF lactate (n = 79), IFN-{gamma} (n = 145), CSF IL-10 (n = 145), CSF IL-12p70 (n = 145), IL-1beta (n = 145), TNF (n = 145), IL-8 (n = 145), IP-10 (n = 24), MCP-1 (n = 23), RANTES (n = 24), and MIG (n = 24) concentrations were not associated with death by univariate analysis.

Differences in the intracerebral immune response between HIV-positive and HIV-negative TBM patients

There are scant data on the impact of HIV coinfection on the intracerebral immune response in TBM patients. In this study, the mean peripheral blood CD4+ T cell count in HIV-positive cases (n = 77) was 117 per µl (range, 9–694) at presentation, and no patient received antiretrovirals during their acute illness. In accordance with their immunocompromised state, there was a trend for many, but not all markers of CSF inflammation to be attenuated in HIV-positive patients relative to HIV-negative patients. Thus, CSF opening pressures, CSF pleocytosis values, and CSF IFN-{gamma} concentrations were all lower in HIV-positive patients relative to HIV-negative patients (Table IV). Conversely, CSF lactate concentrations were less in HIV-positive patients than in HIV-negative patients (Table IV). Interestingly, the most affected CSF parameter was the anti-inflammatory molecule IL-10, concentrations of which were ~6-fold less in HIV-positive patients than in HIV-negative patients (Table IV). Consequently, the CSF IFN-{gamma}:IL-10 ratio, a useful index measuring the balance of candidate proinflammatory vs anti-inflammatory cytokines, was tilted toward excess IFN-{gamma} in HIV-positive patients relative to HIV-negative patients (Table IV). Logistic regression on 144 (112 HIV-negative and 32 HIV-positive) cases in which appropriate data were available was used to identify immunological and clinical features at presentation that independently associated with HIV infection. Included in the model were factors that by univariate analysis were associated with HIV in this patient subset. These parameters were sex, age, evidence of extrapulmonary/neural TB, CSF white blood cell count, CSF:blood glucose ratio, CSF IL-10 concentration, and CSF IFN-{gamma}:IL-10 ratio. Logistic regression indicated male sex (p = 0.014, odds ratio (95% CI), 4.5 (1.4–15)), age (p = 0.003, odds ratio (95% CI), 0.93 (0.89–0.98)), CSF white blood cell count ((p = 0.001), odds ratio (95% CI), 0.24 (0.11–0.57)), CSF IFN-{gamma}:IL-10 ratio (p = 0.009, odds ratio (95% CI), 2.47 (1.25–4.88)), and evidence of extrapulmonary/neural TB (p = 0.009, odds ratio (95% CI), 4.4 (1.1–17)) were independently associated with HIV.


View this table:
[in this window]
[in a new window]
  		Table IV. Attenuated intracerebral immune responses in HIV-positive TBM patients at presentation

 
Relationship between the acute intracerebral inflammatory response in HIV-positive TBM cases and death

The relationship between intracerebral immune response and disease grade at presentation in HIV-positive cases was also examined. However, no CSF indices in HIV-positive cases were significantly correlated with disease grade at presentation (data not shown). The relationship between CSF inflammatory indices at presentation was also related to final outcome. Death as a final outcome in HIV-positive patients was associated by univariate analysis with significantly diminished CSF concentrations of IFN-{gamma} and, to a lesser extent, IL-8 and IL-10, at presentation (Table V). Peripheral blood CD4 counts were also significantly less in those who died (Table V). Low IFN-{gamma} and CD4 count remained associated with death after correction for multiple testing. Other parameters, such as opening pressure (n = 63), CSF protein (n = 89), CSF:blood glucose ratio (n = 89), CSF IL-12p70 (n = 89), CSF IL-1beta (n = 89), and CSF TNF (n = 89) concentrations were not associated with death as an outcome.


View this table:
[in this window]
[in a new window]
  		Table V. Presentation CSF indices associated with outcome in HIV-positive TBM patients

 
The significance of clinical parameters such as type of adjunctive treatment (placebo or dexamethasone), GCS, previous TB treatment, evidence of extrapulmonary TB, interruption in TB therapy, and CSF:blood glucose ratio to death in HIV-positive patients was also examined. Univariate analysis revealed only the presence of extrapulmonary TB and a low GCS at admission was significantly associated with death.

Logistic regression was used to determine whether IFN-{gamma}, IL-8, IL-10, extrapulmonary TB, and GCS at presentation were independently associated with death in HIV-positive TBM cases. Logistic regression indicated only IFN-{gamma} was independently associated with death as an outcome (p = 0.018, odds ratio (95% CI), 0.28 (0.09–0.8), n = 41). When CD4 count was included in the logistic regression, the number of eligible cases for analysis became smaller (n = 32), and IFN-{gamma} was only weakly associated with death as an outcome (p = 0.06, odds ratio (95% CI), 0.31 (0.09–1.1), n = 32).


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
Understanding which aspects of the intracerebral and peripheral blood host response in TBM patients are most relevant to diagnosis, clinical presentation, and outcome could facilitate improved clinical management of this disease. In this study, we found that ESAT-6- and CFP-10-based ELISPOT assays using peripheral blood had poor sensitivity and specificity for the diagnosis of TBM in Vietnamese adults. CSF IL-6 was identified as the only inflammatory parameter that was independently associated with severe TBM disease presentation (BMRC grade III). We also found that TBM patients coinfected with HIV had an attenuated intracerebral host response and death was independently associated with lower CSF IFN-{gamma} levels at presentation.

In some epidemiological settings, peripheral blood T cell responses to the RD1-asociated Ags ESAT-6 and CFP-10 are considered useful markers of infection with M. tuberculosis and are probably more specific than traditional skin tests (13, 15, 20). However, their role in the diagnosis of TB disease is less well established. Responses to these Ags have not previously been described in a large number of patients with culture-confirmed TBM. We found that measurable ESAT-6- and/or CFP-10-specific ELISPOT responses were a poor correlate of TBM in HIV-positive and HIV-negative Vietnamese adults and imply that conventional methods of microbiological diagnosis remain essential (21). In part, T cell lymphopenia and anergy resulting from disseminated TB, or advanced HIV infection, may explain the lack of peripheral blood ELISPOT responses in some TBM patients. In HIV-positive cases, there was a strong suggestion that responders to these Ags had relatively higher CD4+ and CD8+ T cell counts (Table II).

The presence of circulating ESAT-6- and/or CFP-10-specific T cells in 62% of healthy young Vietnamese adults is consistent with a high level of exposure in this population, all of whom live in an urban setting where TB is highly prevalent. The frequency of positive ESAT-6 and/or CFP-10 responses among healthy Vietnamese adults, similar to that reported previously from healthy adults in urban India (22), underlines the difficulties in using immunodiagnostic methods to diagnose active TB in endemic settings. An alternative approach to the immunodiagnosis of TBM would entail detection of ESAT-6- and/or CFP-10-specific T cells in CSF. However, we have previously shown that CSF T cells from TBM cases fail to respond to antigenic or mitogenic stimulation ex vivo, rendering this approach inappropriate (4).

TBM patients presenting with significant neurological deficit (BMRC grade III) have a poor prognosis (1). Immune-driven inflammation in the confines of the subarachnoid space has long been considered as central to the cerebral pathology and outcome from TBM. In this study, CSF concentrations of IL-6 were independently associated with BMRC grade III. IL-6 is an acute phase cytokine with diverse actions and has been implicated in the pathophysiology of many neurological and inflammatory disorders (23). In the context of TB, murine studies suggest IL-6 is involved in stimulating early IFN-{gamma} production, but is not essential for the development of protective immunity against M. tuberculosis (24). However, IL-6 may also play an anti-inflammatory role by suppressing gene expression of proinflammatory cytokines (25). Whether IL-6 is inflammatory, anti-inflammatory, or both, in the setting of TBM in humans is uncertain. Nevertheless, the positive association between CSF IL-6 and clinical severity is consistent with a pathogenic process in which the acute phase response to mycobacterial Ags is associated with cerebral pathology before therapy.

In agreement with previous studies (5, 6, 7, 8), we found high baseline CSF concentrations of IL-6, IL-8, IFN-{gamma}, and IL-10 in HIV-negative TBM patients. TNF and IL-12p70 were also detectable at baseline, but not in all patients. Significant correlations were observed between different cytokine pairs (data not shown), as might be expected in a regulated cytokine cascade, but not with cytokines and CSF cell counts, suggesting resident cells were the most likely source of these molecules. Importantly, none of the CSF cytokines or chemokines that we measured were associated with outcome.

HIV-positive individuals with TBM generally have a poor prognosis (1, 26), although the neurological features of the disease at presentation are probably not influenced by HIV infection (27). However, this study suggests HIV infection alters the intracerebral inflammatory response. Compared with HIV-negative TBM patients, there was a strong trend for HIV-positive individuals to have a reduced CSF pleocytosis and opening pressure at presentation. Most dramatically, HIV-positive individuals had diminished CSF IL-10 concentrations. Since IL-10 is an inhibitor of cytokine secretion, e.g., IFN-{gamma}, by Th1 cells and NK cells, the overall effect of reduced CSF IL-10 concentrations would be a cytokine milieu skewed toward cellular activation. This was reflected in the IFN-{gamma}:IL-10 ratio, which was heavily polarized in favor of IFN-{gamma} in HIV-positive individuals. The basis for an attenuated CSF IL-10 response in HIV-positive individuals is uncertain, although clearly HIV infection in both the periphery and CNS could have significant modulatory effects on quantitative and qualitative aspects of the intracerebral immune response.

When HIV-positive TBM cases were stratified by outcome, death was independently associated with relatively low CSF concentrations of IFN-{gamma} at presentation. This suggests IFN-{gamma}, despite its inflammatory activity, plays a beneficial role in the intracerebral immune response, an observation consistent with a body of literature describing a crucial role for IFN-{gamma} and related molecules in host resistance to mycobacterial infection (28, 29). The basis for the diminished CSF IFN-{gamma} concentrations in HIV-positive TBM patients that subsequently died is unclear. Potentially, it reflects the underlying level of immunocompromise in these individuals; CD4 counts were significantly less in patients who died than in those who survived, although CD4 counts were not correlated with CSF IFN-{gamma} concentrations (data not shown). Alternatively, host genetic, nutritional factors, differences in pathogen burden,or other concomitant illnesses may attenuate the IFN-{gamma} response in these subjects. Potentially, HIV-positive TBM patients might benefit from IFN-{gamma} as an immunoadjuvant to chemotherapy and antiretrovirals. Aerosolized rIFN-{gamma} has been safely administered to patients with refractory pulmonary TB (30), though there are scant data on its clinical effectiveness (31).

Collectively, these data on TBM cases highlight a conceptual paradox in our understanding of TBM. On the one hand, prevailing dogma suggests that a strong intracerebral immune response is associated with disease severity and is also deleterious to outcome. Consistent with this argument is the observation that disease severity at presentation was positively correlated with CSF IL-6 concentrations (this study) and that adjunctive corticosteroids provide clinical benefit to HIV-negative TBM cases (1). In contrast, poor outcome from TBM in HIV-negative patients is not significantly associated with elevated CSF inflammatory cytokine concentrations at presentation (this study). In addition, our earlier work revealed that aggressive adjunctive corticosteroid therapy does not profoundly attenuate markers of CSF inflammation, despite improving outcome (4). Furthermore, a poor outcome from TBM in HIV-negative cases is associated with an attenuated CSF pleocytosis (1), and, in HIV-positive cases, a weak CSF IFN-{gamma} response (this study). Collectively, these data serve to highlight the complexity of TBM and suggest the immunopathogenesis model, whereby intracerebral immune responses are pivotal to presentation and outcome, may be overly simplistic. Clearly, other factors contribute to TBM disease pathogenesis and outcome. The anatomical location in the cerebral tissue of tuberculous lesions is likely to be important, as probably are host and pathogen genetic determinants. For example, TBM associated with rifampicin and isoniazid-resistant M. tuberculosis strains is clinically devastating (32). Concomitant infections, such as hepatitis B, may also complicate treatment and affect outcome (1).

Collectively, the results described here represent the most comprehensive description of the peripheral and intracerebral immune response in either HIV-negative or HIV-positive TBM patients. Future studies that define more thoroughly the nature of the intracerebral host response and its relationship to outcome are warranted. Ultimately, a better understanding of TBM pathogenesis should help improve clinical management in one of the most devastating forms of TB.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflict of interest.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 Address correspondence and reprint requests to Dr. Cameron Simmons, Oxford University Clinical Research Unit, Hospital for Tropical Diseases, 190 Ben Ham Tu, District 5, Ho Chi Minh City, Vietnam. E-mail address: csimmons{at}hcm.vnn.vn Back

2 Abbreviations used in this paper: TBM, tuberculous meningitis; CSF, cerebrospinal fluid; GCS, Glasgow Cancer Score; MIG, monokine induced by IFN-{gamma}; IP-10, IFN-{gamma}-inducible protein 10; PPD, purified protein derivative; SFU, spot-forming unit; CI, confidence interval. ESAT-6, 6-kDa early secretory antigenic target; CFP-10, culture filtrate protein 10. Back

Received for publication August 25, 2005. Accepted for publication November 14, 2005.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Thwaites, G. E., D. B. Nguyen, H. D. Nguyen, T. Q. Hoang, T. T. Do, T. C. Nguyen, Q. H. Nguyen, T. T. Nguyen, N. H. Nguyen, T. N. Nguyen, et al 2004. Dexamethasone for the treatment of tuberculous meningitis in adolescents and adults. N. Engl. J. Med. 351: 1741-1751. [Abstract/Free Full Text]
  2. Rich, A. R., H. A. McCordock. 1933. The pathogenesis of tuberculous meningitis. Bull. John Hopkins Hosp. 52: 5-37.
  3. Shane, S. J., C. Riley. 1953. Tuberculous meningitis: combined therapy with cortisone and antimicrobial agents. N. Engl. J. Med. 249: 829-834. [Medline]
  4. Simmons, C. P., G. E. Thwaites, N. T. Quyen, T. T. Chau, P. P. Mai, N. T. Dung, K. Stepniewska, N. J. White, T. T. Hien, J. Farrar. 2005. The clinical benefit of adjunctive dexamethasone in tuberculous meningitis is not associated with measurable attenuation of peripheral or local immune responses. J. Immunol. 175: 579-590. [Abstract/Free Full Text]
  5. Akalin, H., A. C. Akdis, R. Mistik, S. Helvaci, K. Kilicturgay. 1994. Cerebrospinal fluid interleukin-1beta/interleukin-1 receptor antagonist balance and tumor necrosis factor-{alpha} concentrations in tuberculous, viral and acute bacterial meningitis. Scand. J. Infect. Dis. 26: 667-674. [Medline]
  6. Donald, P. R., J. F. Schoeman, N. Beyers, E. D. Nel, S. M. Carlini, K. D. Olsen, G. H. McCracken. 1995. Concentrations of interferon {gamma}, tumor necrosis factor {alpha}, and interleukin-1beta in the cerebrospinal fluid of children treated for tuberculous meningitis. Clin. Infect. Dis. 21: 924-929. [Medline]
  7. Mastroianni, C. M., F. Paoletti, M. Lichtner, C. D’Agostino, V. Vullo, S. Delia. 1997. Cerebrospinal fluid cytokines in patients with tuberculous meningitis. Clin. Immunol. Immunopathol. 84: 171-176. [Medline]
  8. Thwaites, G. E., C. P. Simmons, N. Than Ha Quyen, T. Thi Hong Chau, P. Phuong Mai, N. Thi Dung, N. Hoan Phu, N. P. White, T. Tinh Hien, J. J. Farrar. 2003. Pathophysiology and prognosis in Vietnamese adults with tuberculous meningitis. J. Infect. Dis. 188: 1105-1115. [Medline]
  9. Price, N. M., J. Farrar, T. T. Tran, T. H. Nguyen, T. H. Tran, J. S. Friedland. 2001. Identification of a matrix-degrading phenotype in human tuberculosis in vitro and in vivo. J. Immunol. 166: 4223-4230. [Abstract/Free Full Text]
  10. Tsenova, L., A. Bergtold, V. H. Freedman, R. A. Young, G. Kaplan. 1999. Tumor necrosis factor {alpha} is a determinant of pathogenesis and disease progression in mycobacterial infection in the central nervous system. Proc. Natl. Acad. Sci. USA 96: 5657-5662. [Abstract/Free Full Text]
  11. Brodin, P., L. Majlessi, R. Brosch, D. Smith, G. Bancroft, S. Clark, A. Williams, C. Leclerc, S. T. Cole. 2004. Enhanced protection against tuberculosis by vaccination with recombinant Mycobacterium microti vaccine that induces T cell immunity against region of difference 1 antigens. J. Infect. Dis. 190: 115-122. [Medline]
  12. Brodin, P., I. Rosenkrands, P. Andersen, S. T. Cole, R. Brosch. 2004. ESAT-6 proteins: protective antigens and virulence factors?. Trends Microbiol. 12: 500-508. [Medline]
  13. Wilkinson, K. A., R. J. Wilkinson, A. Pathan, K. Ewer, M. Prakash, P. Klenerman, N. Maskell, R. Davies, G. Pasvol, A. Lalvani. 2005. Ex vivo characterization of early secretory antigenic target 6-specific T cells at sites of active disease in pleural tuberculosis. Clin. Infect. Dis. 40: 184-187. [Medline]
  14. Ewer, K., J. Deeks, L. Alvarez, G. Bryant, S. Waller, P. Andersen, P. Monk, A. Lalvani. 2003. Comparison of T-cell-based assay with tuberculin skin test for diagnosis of Mycobacterium tuberculosis infection in a school tuberculosis outbreak. Lancet 361: 1168-1173. [Medline]
  15. Liebeschuetz, S., S. Bamber, K. Ewer, J. Deeks, A. A. Pathan, A. Lalvani. 2004. Diagnosis of tuberculosis in South African children with a T-cell-based assay: a prospective cohort study. Lancet 364: 2196-2203. [Medline]
  16. Richeldi, L., K. Ewer, M. Losi, B. M. Bergamini, P. Roversi, J. Deeks, L. M. Fabbri, A. Lalvani. 2004. T cell-based tracking of multidrug resistant tuberculosis infection after brief exposure. Am. J. Respir. Crit. Care Med. 170: 288-295. [Abstract/Free Full Text]
  17. MRC. 1948. Streptomycin treatment of tuberculous meningitis. Br. Med. J. 1: 582-597. [Free Full Text]
  18. Vordermeier, H. M., A. Whelan, P. J. Cockle, L. Farrant, N. Palmer, R. G. Hewinson. 2001. Use of synthetic peptides derived from the antigens ESAT-6 and CFP-10 for differential diagnosis of bovine tuberculosis in cattle. Clin. Diagn. Lab. Immunol. 8: 571-578. [Abstract/Free Full Text]
  19. Garg, R. K.. 1999. Tuberculosis of the central nervous system. Postgrad. Med. J. 75: 133-140. [Abstract/Free Full Text]
  20. Lalvani, A., A. A. Pathan, H. Durkan, K. A. Wilkinson, A. Whelan, J. J. Deeks, W. H. Reece, M. Latif, G. Pasvol, A. V. Hill. 2001. Enhanced contact tracing and spatial tracking of Mycobacterium tuberculosis infection by enumeration of antigen-specific T cells. Lancet 357: 2017-2021. [Medline]
  21. Thwaites, G. E., T. T. Chau, J. J. Farrar. 2004. Improving the bacteriological diagnosis of tuberculous meningitis. J. Clin. Microbiol. 42: 378-379. [Abstract/Free Full Text]
  22. Lalvani, A., P. Nagvenkar, Z. Udwadia, A. A. Pathan, K. A. Wilkinson, J. S. Shastri, K. Ewer, A. V. Hill, A. Mehta, C. Rodrigues. 2001. Enumeration of T cells specific for RD1-encoded antigens suggests a high prevalence of latent Mycobacterium tuberculosis infection in healthy urban Indians. J. Infect. Dis. 183: 469-477. [Medline]
  23. Van Snick, J.. 1990. Interleukin-6: an overview. Annu. Rev. Immunol. 8: 253-278. [Medline]
  24. Saunders, B. M., A. A. Frank, I. M. Orme, A. M. Cooper. 2000. Interleukin-6 induces early {gamma} interferon production in the infected lung but is not required for generation of specific immunity to Mycobacterium tuberculosis infection. Infect. Immun. 68: 3322-3326. [Abstract/Free Full Text]
  25. Xing, Z., J. Gauldie, G. Cox, H. Baumann, M. Jordana, X. F. Lei, M. K. Achong. 1998. IL-6 is an antiinflammatory cytokine required for controlling local or systemic acute inflammatory responses. J. Clin. Invest. 101: 311-320. [Medline]
  26. Katrak, S. M., P. K. Shembalkar, S. R. Bijwe, L. D. Bhandarkar. 2000. The clinical, radiological and pathological profile of tuberculous meningitis in patients with and without human immunodeficiency virus infection. J. Neurol. Sci. 181: 118-126. [Medline]
  27. Guy E Thwaites, N. D. B., Nguyen Huy Dung, Hoang Thi Quy, DoThi Tuong Oanh, Nguyen Thi Cam Thoa, Nguyen Quang Hien, Nguyen Tri Thuc, N. T. N. L. Nguyen Ngoc Hai, Nguyen Ngoc Lan, et al 2005. The influence of HIV infection on clinical presentation, response to treatment, and outcome in adults with tuberculous meningitis. J. Infect. Dis. 192: 2134-2141. [Medline]
  28. Flynn, J. L., J. Chan. 2001. Immunology of tuberculosis. Annu. Rev. Immunol. 19: 93-129. [Medline]
  29. Flynn, J. L., J. Chan, K. J. Triebold, D. K. Dalton, T. A. Stewart, B. R. Bloom. 1993. An essential role for interferon {gamma} in resistance to Mycobacterium tuberculosis infection. J. Exp. Med. 178: 2249-2254. [Abstract/Free Full Text]
  30. Condos, R., W. N. Rom, N. W. Schluger. 1997. Treatment of multidrug-resistant pulmonary tuberculosis with interferon-{gamma} via aerosol. Lancet 349: 1513-1515. [Medline]
  31. Suarez-Mendez, R., I. Garcia-Garcia, N. Fernandez-Olivera, M. Valdes-Quintana, M. T. Milanes-Virelles, D. Carbonell, D. Machado-Molina, C. M. Valenzuela-Silva, P. A. Lopez-Saura. 2004. Adjuvant interferon {gamma} in patients with drug - resistant pulmonary tuberculosis: a pilot study. BMC Infect. Dis. 4: 44[Medline]
  32. Thwaites, G. E., N. T. Lan, N. H. Dung, H. T. Quy, D. T. Oanh, N. T. Thoa, N. Q. Hien, N. T. Thuc, N. N. Hai, N. D. Bang, et al 2005. Effect of antituberculosis drug resistance on response to treatment and outcome in adults with tuberculous meningitis. J. Infect. Dis. 192: 79-88. [Medline]



This article has been cited by other articles:


Home page
 CVIHome page
X. Chen, Q. Yang, M. Zhang, M. Graner, X. Zhu, N. Larmonier, M. Liao, W. Yu, Q. Deng, and B. Zhou
Diagnosis of Active Tuberculosis in China Using an In-House Gamma Interferon Enzyme-Linked Immunospot Assay
Clin. Vaccine Immunol., June 1, 2009; 16(6): 879 - 884.
[Abstract] [Full Text] [PDF]

Home page
 CVIHome page
S. Murakami, M. Takeno, H. Oka, A. Ueda, T. Kurokawa, Y. Kuroiwa, and Y. Ishigatsubo
Diagnosis of Tuberculous Meningitis Due to Detection of ESAT-6-Specific Gamma Interferon Production in Cerebrospinal Fluid Enzyme-Linked Immunospot Assay
Clin. Vaccine Immunol., May 1, 2008; 15(5): 897 - 899.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Microbiol. Rev.Home page
R. B. Rock, M. Olin, C. A. Baker, T. W. Molitor, and P. K. Peterson
Central Nervous System Tuberculosis: Pathogenesis and Clinical Aspects
Clin. Microbiol. Rev., April 1, 2008; 21(2): 243 - 261.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Microbiol.Home page
G. Thwaites, M. Caws, T. T. H. Chau, A. D'Sa, N. T. N. Lan, M. N. T. Huyen, S. Gagneux, P. T. H. Anh, D. Q. Tho, E. Torok, et al.
Relationship between Mycobacterium tuberculosis Genotype and the Clinical Phenotype of Pulmonary and Meningeal Tuberculosis
J. Clin. Microbiol., April 1, 2008; 46(4): 1363 - 1368.
[Abstract] [Full Text] [PDF]

Home page
 PediatricsHome page
T. R. Sterling, T. Martire, A. S. de Almeida, L. Ding, D. E. Greenberg, L. A. Moreira, H. Elloumi, A. P.V. Torres, C. C. Sant'Anna, E. Calazans, et al.
Immune Function in Young Children With Previous Pulmonary or Miliary/Meningeal Tuberculosis and Impact of BCG Vaccination
Pediatrics, October 1, 2007; 120(4): e912 - e921.
[Abstract] [Full Text] [PDF]

Home page
 J. Leukoc. Biol.Home page
J. E. Harris, J. A. Green, P. T. Elkington, and J. S. Friedland
Monocytes infected with Mycobacterium tuberculosis regulate MAP kinase-dependent astrocyte MMP-9 secretion
J. Leukoc. Biol., February 1, 2007; 81(2): 548 - 556.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
M. Eisenhut
Comment on "Pretreatment Intracerebral and Peripheral Blood Immune Responses in Vietnamese Adults with Tuberculous Meningitis: Diagnostic Value and Relationship to Disease Severity and Outcome"
J. Immunol., May 1, 2006; 176(9): 5137 - 5137.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Simmons, C. P.
Right arrow Articles by Farrar, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Simmons, C. P.
Right arrow Articles by Farrar, J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS