© 2007 American Thoracic Society doi: 10.1164/rccm.200611-1653UP
Update in Tuberculosis 20061 Tuberculosis and Chest Unit, Grantham Hospital, Hong Kong, China; and 2 Tuberculosis and Chest Service, Centre for Health Protection, Department of Health, Hong Kong, China Correspondence and requests for reprints should be addressed to Win Wai Yew, M.B., Tuberculosis and Chest Unit, Grantham Hospital, Hong Kong, China. E-mail: yewww{at}ha.org.hk EPIDEMIOLOGY OF TUBERCULOSIS Because little was known concerning the scope of treatment of latent tuberculosis infection (LTBI) in the United States and Canada, identification of the types of clinics that administered such treatment, and patients who received it, would guide resource utilization and improve treatment initiation and completion. Sterling and coworkers, on behalf of the Tuberculosis Epidemiologic Studies Consortium, Centers for Disease Control and Prevention, surveyed 244 clinics, each having initiated LTBI treatment for 10 or more patients in 2002, at 19 U.S. and 2 Canadian sites (1). An estimated 37,857 patients started LTBI treatment in 2002, including 37,145 from the U.S. sites, with 79% at general public health clinics, 6.4% at immigrant/refugee clinics, and 6.1% at correction/detention facilities. Study catchment areas for the 19 U.S. sites represented 8.6% of the U.S. population and 12.7% of all tuberculosis (TB) cases in 2000. On extrapolation to the entire U.S. population, the estimated total number of LTBI treatment starts was approximately 291,000 to 433,000. Assuming a 5% lifetime risk of TB without treatment, and 20 to 60% treatment effectiveness, approximately 4,000 to 11,000 cases of TB were prevented in the United States. Thus, Sterling and coworkers concluded that treatment for LTBI was initiated among a substantial number of persons in the United States and Canada, primarily in the public sector, and such treatment could significantly decrease the disease burden in these countries. Targeted screening and treatment of latently infected subjects are central to strategies aimed at eliminating TB. Unfortunately, there appear to be few specific criteria, other than medical factors, in designating groups as high risk for developing TB. Moonan and coworkers conducted location-based screenings in partnership with multiple community-based organizations in communities previously demonstrated by geographic information system to have genotypically clustered Mycobacterium tuberculosis isolates (2). One person with TB was found for every 83 screened, and one person with LTBI for every five screened, far exceeding the expected yield of untargeted screening for a county with a TB incidence of only 5.7 per 100,000. Male subjects were more commonly identified (odds ratio [OR], 4.8). Thus, it appeared that combining genotyping and geographic information systems could potentially help in identifying high-risk status and in determining areas for location-based TB screening. In an editorial accompanying Moonan and colleagues' article (2), it was pointed out that these data could be used as a tool for garnering the critical support of community-based organizations (3). Both short- and long-term benefits of such partnerships as well as the resulting interventions are important to measure. In addition to following up on the outcome of the intervention, an analysis of the services received by the screened individuals would clarify the relative roles of housing, correctional care facilities, as well as other settings, in contributing to the high rates of ongoing transmission of TB in the community. Treatment completion is of paramount importance to the success of LTBI therapy. However, the reasons for not completing treatment are not fully known. Shieh and coworkers investigated this issue in the United States by conducting a survey in English, Chinese, or Spanish for subjects with LTBI at their first TB clinic visits (4). A total of 217 patients, 90% foreign-born, completed the survey and 28.6% also finished at least 6 months of isoniazid treatment under usual clinic settings. Multivariate analysis identified low-risk perception of progressing to active TB without treatment (OR, 0.31; p = 0.007) and not wanting venipuncture (OR, 0.43; p = 0.015) as the two independent predictors for noncompletion. Targeting individuals at a high risk for TB, minimizing inconveniences, utilizing further education, and using diagnostic tests with improved specificity for LTBI may help to address these critical concerns. The HIV epidemic is often blamed for the failure of TB control in sub-Saharan Africa, but it is not totally clear why the directly observed therapy, short-course (DOTS) strategy is performing suboptimally in this setting. Wood and coworkers conducted a cross-sectional survey in a community of 13,000 persons with high HIV prevalence, high TB notification rate, and a well-functioning DOTS TB control program (5). Active case finding for pulmonary TB was performed in 762 adults using sputum microscopy and culture. HIV status was assessed, and symptoms and risk factors for TB were surveyed by questionnaire. Of those studied, 174 (23%) tested HIV positive, 11 (7 HIV positive) were receiving TB therapy, 6 (5 HIV positive) had previously undiagnosed smear-positive TB, and 6 (4 HIV positive) had smear-negative/culture-positive TB. Symptoms were not useful for screening TB. Among HIV-positive subjects, the prevalence rates of notified and undiagnosed smear-positive TB were 1,563 of 100,000 and 2,837 of 100,000, respectively, with a case-finding proportion of 37%. The corresponding figures for HIV-negative subjects were 352 of 100,000, 175 of 100,000, and 67%, respectively. Contrary to the findings of an earlier study (6), the estimated duration of infectiousness was similar for HIV-positive and HIV-negative subjects. However, 87% of total person-years of undiagnosed smear-positive TB in the community were among HIV-infected individuals. Thus, DOTS strategy based on passive case finding (centering on symptoms) should be supplemented by active case finding, targeting HIV-infected subjects. GENETICS AND IMMUNOLOGY OF TB
Interferon (IFN)-
Two recent publications from Croatia have addressed important genetic polymorphisms of IFN- Two recent publications on genetics and TB focused on iron acquisition. Krithika and coworkers reported on a biosynthetic locus that could incorporate a variety of aliphatic chains on the mycobactin skeleton involved in iron acquisition (10). McDermid and coworkers reviewed the putative role of iron-related host genes by focusing on two candidate iron-regulatory genes, haptoglobin and NRAMP1, with common polymorphic variants leading to functionally distinct biochemical phenotypes that would be predicted to influence the course of TB infection in humans (11). Because iron plays a key role in the development of many infectious diseases, including TB, these findings provide a potentially attractive target for development of new anti-TB drugs.
The cell-mediated immune response in TB originates predominantly from IFN-
Banaiee and coworkers demonstrated that macrophages lacking Toll-like receptor 2 (TLR2) were more resistant to inhibition by M. tuberculosis, possibly due to inhibition of selected IFN-
CD8 T cells are among the key cells in the adaptive immune response to intracellular pathogens. Once stimulated, these T cells can provide a variety of effector functions, aimed at clearance or containment of the pathogens. Thus, delineation of their roles is of great importance in understanding cell-mediated immunity in TB. Sud and coworkers, using published and unpublished data, built and tested a mathematical model of the immune response to M. tuberculosis in the lung (16). The model was then used to perform simulations mimicking various experimental scenarios. Selective deletion of CD8 T-cell subsets suggested a differential contribution from CD8 T-cell effectors with cytotoxicity and those with production of IFN- DIAGNOSIS OF LTBI AND TB
Detection of LTBI is traditionally based on a positive tuberculin skin test (TST) reaction, a delayed type of hypersensitivity relying on T-cell recognition of the M. tuberculosis antigens from purified protein derivative (PPD). Advancement in the diagnosis of such infection is now possible with in vitro IFN-
Judging from currently accumulated research experience, IFN-
It would be scientifically important to assess the dynamics of M. tuberculosis–specific Th1 T cells as these might be linked to clinical outcome. Ewer and coworkers used the ex vivo IFN-
Pai and coworkers have shown that, in a nosocomial environment with likely ongoing intensive exposure to M. tuberculosis, Indian health care workers had strong IFN-
Two important studies have provided further insights into the differing performance of two commercially available IFNGRAs. Ferrara and coworkers did a prospective study on 393 consecutively enrolled patients by testing with T-SPOT.TB and QFT for suspected LTBI or active TB (22). Overall agreement with TST was similar ( Lee and coworkers studied TST, QFT, and T-SPOT.TB responses prospectively in patients with TB and also in low-risk subjects in a tertiary referral hospital in South Korea (23). They found that T-SPOT.TB was more sensitive than TST and QFT (p < 0.001). Although QFT was more specific than T-SPOT.TB, the difference was not significant (p = 0.13). Taken together, these studies would suggest that IFNGRA is more sensitive and specific in detection of LTBI compared with TST. The differing performances of the two commercial assays, in head-to-head comparisons, raise the possibility of dissimilar results in some clinical settings. Although the superiority of IFNGRA over TST seems likely, there might still be some evidence to the contrary. Hill and coworkers have found that, in Gambian children, the ELISPOT was slightly less sensitive than TST in the diagnosis of LTBI after recent exposure, and neither test was confounded by previous bacillus Calmette-Guérin vaccination (24). Jeffries and coworkers further explored the identification of suitable ELISPOT and TST cutoffs for the diagnosis of M. tuberculosis infection in the Gambia for comparison purposes (25). M. tuberculosis–specific DNA amplification techniques on sputum or respiratory secretions still lack good sensitivity in the diagnosis of smear-negative pulmonary TB. Jafari and coworkers investigated the utility of enumeration of M. tuberculosis–specific mononuclear cells in bronchoalveolar lavage (BAL) fluid for diagnosing this form of TB (26). An ELISPOT assay (T-SPOT.TB) using ESAT-6 and CFP-10 peptides was performed on mononuclear cells in peripheral blood (PBMCs) and BAL fluid. Of 37 patients with suspected smear-negative TB, 12 had TB, whereas 25 had alternative diseases. Patients with TB had a median number of 17 ESAT-6– and 24.5 CFP-10–specific cells per 200,000 PBMCs, and 37.5 ESAT-6– and 49.5 CFP-10–specific cells per 200,000 mononuclear cells in the BAL fluid, whereas control patients had a median of 1 ESAT-6– and 1 CFP-10–specific cells per 200,000 PBMCs and 0 ESAT-6– and 0 CFP-10–specific cells per 200,000 cells in the BAL fluid (all p < 0.0001). All patients with TB, but none of the control subjects, had more than 5 spot-forming cells per 200,000 mononuclear cells with either peptide in the BAL ELISPOT assay. It therefore appears that smear-negative pulmonary TB can be rapidly diagnosed by identification of M. tuberculosis–specific cells in the BAL fluid.
Goletti and coworkers evaluated the diagnostic accuracy of an IFN- Aiken and coworkers (28) found that 82% of HIV-negative patients with smear-positive TB were ESAT-6 or CFP-10 ELISPOT positive, and 90% were PPD ELISPOT positive. Among those who successfully completed treatment for TB, 55% were ELISPOT negative with the former two antigens, and 21% were PPD ELISPOT negative. Seventy-three percent of cured cases had a decreased CFP-10 ELISPOT count, 78% a decreased ESAT-6 ELISPOT count, and 70% a decreased PPD ELISPOT count. Thus, successful treatment for TB can be accompanied by a significant reduction in M. tuberculosis–specific antigen ELISPOT count (28). It appears that ELISPOT has the potential to act as a proxy measure of treatment outcome. TREATMENT OF TB AND OTHER MYCOBACTERIAL DISEASE HIV-negative patients who were underweight at baseline had increased risk for TB relapse in a Tuberculosis Trials Consortium study (29). A follow-up study by Khan and coworkers demonstrated the positive prognostic impact of early weight gain among underweight patients with TB in the same trial (30). In their stratified analysis, the 2-year relapse rates were 4.2%, 11.9%, and 20.3% among those not underweight, those underweight but gaining more than 5% weight after 2 months of therapy, and those underweight and not gaining weight, respectively. In a multivariate analysis, 5% or less weight gain between diagnosis and completion of 2-month intensive phase therapy among underweight patients at baseline was significantly associated with increased relapse risk (OR, 2.4; p = 0.03), after controlling for other risk factors including cavity and sputum culture positivity at 2 months. Body weight is a simple and inexpensive clinical marker, much easier to obtain than most laboratory outcome surrogates. If such a result could be reproduced in large TB programs in diverse settings, there would be significant clinical impact, especially in resource-limited conditions. It would appear relevant to explore other means to reduce the relapse risk in underweight patients, aside from prolongation of the duration of chemotherapy (31). Possible areas might include nutritional enhancement and adjunctive steroid therapy. It would also be interesting to assess whether weight gain achieved through nutritional or pharmacologic intervention could result in a reduction of relapse risk similar to weight gain during chemotherapy. Because randomized controlled chemotherapy trials to address reduction of relapse of TB are logistically demanding, Chang and coworkers performed a systematic review of published clinical trials involving adult cohorts with pulmonary TB treated with 6-month rifamycin-containing regimens (32). A static deterministic model was applied in the analysis. Two hundred cases of bacteriologic relapse were identified (out of 5,208 patients). A logistic risk model showed a significant dose–response relationship between dosing schedules and relapse, with the following odds of relapse relative to daily regimens: 1.6 for daily initial phase plus thrice-weekly continuation phase; 2.8 for daily initial phase plus twice-weekly continuation phase; 2.8 for thrice weekly throughout the therapeutic period; 5.0 for daily initial phase plus once-weekly rifapentine; and 7.1 for thrice-weekly initial phase plus once-weekly rifapentine. In the presence of cavitation, only 6-month daily or daily plus thrice weekly attained best-estimated relapse risks of less than 5%; they reached 6% when the 2-month culture was also positive. Thus, cavitary TB would be best treated with 6-month regimens comprising a daily dosing initial phase, and a thrice-weekly continuation phase, which should then be extended if the 2-month culture turns out to be positive (32). These findings have potential therapeutic implications, and generally concur with the recommendations of the American Thoracic Society/Centers for Disease Control and Prevention/Infectious Disease Society of America (33). Rifamycins, as distinguished by their sterilizing activity, have greater potency than isoniazid against dormant and semidormant tubercle bacilli responsible for latent infection. Trials of rifapentine, given together with isoniazid on a once-weekly basis during the continuation phase of treatment, showed satisfactory efficacy in HIV-negative patients with TB, especially in those with low bacterial burdens (34). Studies have also shown the effectiveness of a combination of rifapentine and isoniazid in murine models of latent TB (35). Schechter and coworkers conducted a trial comparing the efficacy of weekly rifapentine and isoniazid versus daily rifampin and pyrazinamide in the treatment of LTBI in household contacts (36). Such contacts were randomized to receive rifapentine 900 mg/isoniazid 900 mg once weekly for 12 weeks or rifampin 450–600 mg/pyrazinamide 750–1,500 mg once daily for 8 weeks, and were monitored for at least 2 years. Among 206 subjects in the former arm and 193 in the latter arm, only 1 patient was HIV infected. The rifapentine/isoniazid combination was well tolerated, but the trial was halted by the investigators before completion because of unanticipated hepatotoxicity in the rifampin/pyrazinamide arm: 20 of 193 (10%) participants in this arm experienced grade 3 or 4 hepatotoxicity compared with 2 of 206 (1%) subjects in the rifapentine/isoniazid arm (p < 0.001). These data demonstrated better tolerance of rifapentine/isoniazid than rifampicin/pyrazinamide. During follow-up, four cases of active TB developed, three in the rifapentine/isoniazid group and one in the rifampin/pyrazinamide arm (1.46 vs. 0.52%: difference, 0.94%; 95% confidence interval [CI], –1.6–3.7%; p = 0.66). The incidence of TB during follow-up was 0.5 per 100 person-years for the rifapentine/isoniazid regimen and 0.2 per 100 person-years for the rifampin/pyrazinamide regimen (risk ratio, 2.8; 95% CI, 0.3–26.8; p = 0.66). Thus, rifapentine/isoniazid weekly for 12 weeks is a promising therapy for LTBI. Recent studies have shown that intermittent administration of rifamycin-based regimens in TB treatment might result in higher rates of failure and relapse when compared with daily dosing (37, 38). Rosenthal and coworkers (39) hypothesized that administration of twice-weekly rifapentine, a long-acting rifamycin, would improve chemotherapeutic efficacy. They compared the activity of conventional daily and twice-weekly rifampin plus isoniazid–based regimens with those of twice-weekly rifapentine plus isoniazid– or moxifloxacin–containing regimens in a murine model of TB. Relapse rates were also measured after 4, 5, and 6 months of therapy to assess stable cure. After 2 months of treatment, twice-weekly therapy with rifapentine (15–20 mg/kg), moxifloxacin, and pyrazinamide proved to be significantly more active than daily or twice-weekly therapy with rifampin, isoniazid, and pyrazinamide. Stable cure was achieved after 4 months of rifapentine plus isoniazid- or moxifloxacin-containing treatment, but only after 6 months of standard daily therapy. Furthermore, twice-weekly rifapentine (15 mg/kg) was found to display more favorable pharmacodynamics than did rifampin (10 mg/kg). Thus, twice-weekly rifapentine regimens, by virtue of their potential for shortening anti-TB therapy, merit clinical evaluation. Moxifloxacin has been shown to possess good in vitro activity against M. tuberculosis, as well as efficacy in murine models of TB, presumably related to its potent sterilizing capacity (40). However, these qualities have not been formally evaluated in clinical trials. Burman and coworkers attempted to compare the impact of moxifloxacin versus ethambutol, both agents in combination with isoniazid, rifampin, and pyrazinamide, on sputum culture conversion to negativity at 2 months to assess the potential sterilizing activity of moxifloxacin (41). The rates were found to be 71% for both of the regimens. However, patients given moxifloxacin more often had negative cultures after 4 weeks of treatment. More patients on moxifloxacin had nausea (22 vs. 9%), but both groups had similar proportions of patients who completed treatment (88 vs. 89%). Thus, the addition of moxifloxiacin to isoniazid, rifampin, and pyrazinamide did not affect 2-month sputum culture status but did show increased activity at earlier time points. There are limitations to this study, including the following: (1) there was a lack of relapse data (which would give more concrete information on sterilizing activity of the drugs); (2) potentially advantageous combinations of moxifloxacin, for example without isoniazid, were not explored; and (3) there was a lack of standardized methodology for mycobacterial culture at different study sites. Although the study did not allow unequivocal conclusions to be made on the sterilizing activity of moxifloxacin and its ability to shorten chemotherapy duration, the increased early activity and tolerability of moxifloxacin would warrant its further evaluation in treatment of TB. A smaller scale study by Codecasa and coworkers (42) has addressed the long-term tolerance and safety of moxifloxacin in patients with complicated TB who received the 8-methoxy fluoroquinolone for prior adverse reactions or bacillary resistance to first-line drugs. Among 38 patients, 81.6% reported treatment success, including 51.7% of patients with multidrug-resistant TB. At least one adverse effect due to moxifloxacin was reported in 31.6% of patients, with most adverse effects being gastrointestinal in origin. In 10.5%, the drug was withdrawn for major intolerance. Recently published articles have addressed the important issues of pharmacologic interaction of rifapentine with other anti-TB drugs, and the pharmacokinetics of rifamycin in children. Prasad and coworkers (43) studied the interaction between rifapentine and isoniazid under acid conditions; their findings suggest that coadministration of rifapentine and isoniazid should be avoided under such situations. The findings by Blake and coworkers (44) of lower rifapentine exposure estimates in children, given a comparable weight-normalized dose as for adults, would suggest the need for a larger milligram/kilogram dose of rifapentine in pediatric patients. With increasing likelihood of moxifloxacin being useful in the treatment of TB, there has been an ongoing search for a better formulation to enhance efficacy and feasibility of intermittent dosing. Thus, Schwartz and coworkers explored the use of moxifloxacin-conjugated dansylated carboxymethylglucan (45). It was found that the conjugated glucan was concentrated in the macrophages of lungs and spleen. Analysis of its pharmacokinetics also demonstrated more rapid and persistent accumulation in tissues than free moxifloxacin. Furthermore, the conjugated glucan was significantly more potent than free moxifloxacin in therapeutic studies on mycobacterial growth in C57BL/6 mice.
An increase in pulmonary disease caused by Mycobacterium avium complex (MAC) has been noted since the late 1990s. Standard present therapy should include a macrolide or an azalide. Two previously published studies suggested that a three-times-weekly regimen incorporating clarithromycin or azithromycin might be as effective as a daily regimen (46, 47). A three-times-weekly regimen holds promise because it potentially reduces costs and adverse effects, and enables better adherence to therapy. Lam and coworkers reported on a 1-year prospective noncomparative trial of three-times-weekly treatment with regimens including a macrolide or an azalide, together with other accompanying agents, principally rifamycins and ethambutol (48), among 91 HIV-negative adults from 17 U.S. cities who initially participated in a trial of inhaled IFN-
Although the advent of macrolides has certainly improved results in the therapy of MAC lung disease, macrolide-resistant MAC disease remains poorly characterized, and its outcome has not been fully described. Griffith and coworkers identified 51 such patients (minimum inhibitory concentration [MIC] for clarithromycin Watanabe and coworkers performed lung resections on 22 patients with localized pulmonary MAC disease who failed state-of-the-art antimicrobial chemotherapy (50). The organisms disappeared from sputum in all patients postoperatively, although one patient experienced bacteriologic relapse 4 months after resection (with subsequent conversion to negativity after further chemotherapy). As a result of such excellent long-term outcome (median follow-up duration, 46 mo), Watanabe and coworkers made strong recommendations regarding early surgery for selected patients before disease becomes too extensive for resection (50). FOOTNOTES Conflict of Interest Statement: Neither author has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form November 17, 2006; accepted in final form December 8, 2006 REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
mediated pathways are of central importance in cell-mediated protective immunity against M. tuberculosis. In a large-scale study involving 1,301 West African subjects, Cooke and coworkers investigated genetic polymorphisms within the IFN-
1 were also detected, but these were not correlated with the regulatory T-cell markers. Ex vivo depletion of CD4+CD25high cells from PBMCs resulted in increased M. tuberculosis–specific IFN-
and interleukin-12p40 (cytokines associated with the development of delayed-type hypersensitivity) from monocytes of blood buffy coats, as compared with two reference strains (H37Rv and CDC 1551). These data therefore suggest that M. tuberculosis infection undetected by TST could occur, and this might be partially related to the difference in immunogenicity of bacillary strains. 
0.35 IU/ml, and 11 (7.5%) using IFN-
0.50), but fewer bacillus Calmette-Guérin–vaccinated individuals were identified as positive by the two assays than by TST (p < 0.005). Indeterminate results were more frequent with QFT than with T-SPOT.TB (11 vs. 3%, p < 0.0001) and were associated with immunosuppressive therapy. Young age (< 5 yr) was also associated with indeterminate results in QFT, but not T-SPOT.TB. Overall, T-SPOT.TB produced significantly more positive results than QFT (38 vs. 26%, p < 0.0001), and close contacts of patients with active TB were more likely to be positive with the former than with the latter (p = 0.0010). 
