Issues in the treatment of Mycobacterium tuberculosis in patients with human immunodeficiency virus infection.

Abstract

Introduction The HIV/AIDS pandemic has led to a rise in the incidence of tuberculosis and an epidemic of co-infection in many developing countries. Treatment of Mycobacterium tuberculosis in persons with HIV infection presents several challenges to the clinician, particularly in resource-poor countries. As will be discussed in this paper, diagnosis of latent tuberculosis relies on tuberculin skin testing, which has poor sensitivity and reproducibility in immunocompromised patients. The World Health Organization (WHO) recommends treatment of active tuberculosis as the primary means of global tuberculosis control. In practice, treatment of active tuberculosis typically requires that a symptomatic patient self-report to a health service for evaluation and management. Even if this approach to tuberculosis control were sufficient, many logistic and clinical problems remain involving tuberculosis diagnosis and therapy in the patient with HIV/AIDS. Recognizing the significant clinical and public health challenges surrounding the treatment of tuberculosis in patients with HIV infection, this paper will address a number of issues relevant to the care of co-infected patients. These include current guidelines for the treatment of active tuberculosis, as well as the diagnosis and treatment of latent tuberculosis in HIV-positive patients. The paper concludes with a discussion of promising new drugs for tuberculosis treatment. Epidemiology of tuberculosis and HIV co-infection It is estimated that one-third of the world population is infected with M. tuberculosis, the large majority of whom live in the developing world. The HIV pandemic of the past two decades has led to a rise in the incidence of tuberculosis, particularly in sub-Saharan Africa. There is now an emerging pandemic of patients with HIV infection who are co-infected with tuberculosis. As of December 2000, the WHO estimated that approximately 36.1 million persons worldwide are living with HIV and nearly one-third of these persons are co-infected with M. tuberculosis[1]. Approximately 68% of persons co-infected with HIV and tuberculosis live in sub-Saharan Africa, while 22% live in Southeast Asia. In the United States, the Centers for Disease Control and Prevention (CDC) estimates that approximately 40% of new tuberculosis cases among persons aged 15-44 years occur among individuals with HIV infection or AIDS. Tuberculosis rates among HIV-infected individuals in the United States, however, vary significantly among different groups, with highest rates among intravenous drug users and those who are foreign-born. Today, the burden of tuberculosis and HIV infections largely impacts the developing world, as well as the minority and low socio-economic individuals within industrialized countries. This paper will discuss issues in the treatment of M. tuberculosis in patients with HIV/AIDS; however, it is important to recognize that many patients co-infected with HIV and M. tuberculosis have limited or no access to essential diagnostic and therapeutic strategies. Escalating tuberculosis case rates over the past decade are largely attributable to HIV. Immunity to M. tuberculosis is partly under the control of the MHC class II restricted CD4 cells. With the progressive loss of CD4 cells, patients with HIV infection are at increased risk of reactivation of latent tuberculosis, as well as primary tuberculosis infection [2]. In turn, active tuberculosis infection appears to upregulate HIV replication, resulting in further immune compromise and accelerated HIV disease progression [3,4]. As a result, patients with HIV infection and active tuberculosis are at increased risk of opportunistic infections and associated mortality. The case fatality rate by the end of tuberculosis treatment is approximately 20% for new sputum smear-positive cases and up to 50% for new smear-negative cases [5]. Tuberculosis is the leading cause of death among persons with HIV/AIDS worldwide [6]. Guidelines for the treatment of active tuberculosis This section will review recent guidelines for the treatment of tuberculosis published by the Tuberculosis Committee of the Infectious Disease Society of America (IDSA) in conjunction with the American Thoracic Society (ATS) and the CDC [7,8]. In addition, we will review the Directly Observed Treatment Short-course (DOTS) strategy of the WHO for tuberculosis control worldwide. These published guidelines pertain to the treatment of tuberculosis without respect to the patient's HIV status. Modified recommendations for the treatment of tuberculosis in HIV-seropositive patients will be discussed in the following section. In April 2000, the IDSA published practice guidelines for the treatment of tuberculosis [7]. Table 1, adapted from the IDSA publication, lists the 10 essential recommendations for the treatment of patients with tuberculosis. Readers are referred to the original publication for detailed comments pertaining to these recommendations, as well as performance indicators. It should be noted that the IDSA recommendations were developed for use in industrialized nations such as the United States and are currently not feasible in many countries of the world. In Table 1, therefore, we have juxtaposed the WHO DOTS strategy guidelines for management of patients with tuberculosis.Table 1: Infectious Disease Society of America (IDSA) recommendations and World Health Organization Directly Observed Treatment Shortcourse (WHO DOTS) strategy for the management of patients with tuberculosis (TB).In geographic areas where ≥ 4% of the M. tuberculosis isolates are resistant to isoniazid, the IDSA, ATS, and CDC recommend that the usual three-drug regimen of isoniazid, rifampin and pyrazinamide be augmented with a fourth drug, either ethambutol or streptomycin. Clinicians will therefore need to be aware of the susceptibility patterns in their geographic area. In 1997, approximately 84% of the US population lived in states that had ≥ 4% of tuberculosis isolates resistant to isoniazid. Therefore, most patients in the United States should be started on an initial four-drug regimen. This is followed by isoniazid and rifampin for 18 weeks. Although a 6-month course of treatment is recommended, this should be extended to 9 months if there is a delay in AFB, culture conversion or clinical improvement for 8 weeks. The WHO Global Tuberculosis Programme (WHO/GTP) assists over 60 countries with national tuberculosis control and prevention. The priority of the WHO program has been active case finding and cure of infectious tuberculosis cases. In 1993, the WHO/GTP declared tuberculosis a global emergency and began promoting the DOTS strategy. The DOTS strategy consists of five key components: "1) Government commitment to sustained TB control activities; 2) Case detection by sputum smear microscopy among symptomatic patients self-reporting to health services; 3) Standardized treatment regimen of six to eight months for at least all confirmed sputum smear positive cases, with directly observed treatment (DOT) for at least the initial two months; 4) A regular, uninterrupted supply of all essential anti-tuberculosis drugs; and 5) A standardized recording and reporting system that allows assessment of treatment results for each patient and of the TB control programme performance overall" [9] (see Table 1). According to the WHO/GTP, as of 1999, 127 countries had accepted the DOTS strategy and were implementing it to varying degrees [10]. Nevertheless, the WHO/GTP estimate that, in 1999, only 45% of the world population had access to DOTS and 23% of new smear-positive cases were referred to DOTS programs. In addition, there are often a complex array of political, financial, and infrastructure problems that impede local DOTS programs. The current WHO DOTS strategy does not incorporate the diagnosis and treatment of latent tuberculosis. Some experts believe that the DOTS strategy, which focuses exclusively on the treatment of active cases, is insufficient for the control and elimination of tuberculosis, particularly in the HIV/AIDS era [11]. It has been recommended that targeted diagnosis and treatment of latent tuberculosis infection among specific populations be added to national tuberculosis control programs. At present, WHO recommends treatment of latent tuberculosis in HIV-positive patients; but only in settings where it is possible to provide HIV testing and counseling, and where it is possible to exclude cases of active tuberculosis and ensure proper follow-up. The DOTS strategy also employs empiric anti-tuberculous therapy without mycobacterial cultures and drug susceptibility testing. Without drug susceptibility information, it is impossible to identify cases of drug-resistant tuberculosis and to avoid treatment failure and further transmission of drug-resistant strains. The WHO recognizes the threat of multidrug-resistant tuberculosis (MDR-TB) and, in July 1999, convened a working group on DOTS-Plus for the treatment of MDR-TB [12]. DOTS-Plus is a pilot program to provide second-line drugs (i.e., fluoroquinolones, amikacin, kanamycin, capreomycin, cycloserine, para-aminosalicylic acid, and ethionamide) to manage MDR-TB in resource-limited countries. This strategy does not, however, address the role of mycobacterial cultures and drug susceptibility testing for individualized drug therapy and the prevention of drug-resistant tuberculosis. Finally, because DOTS relies on patients self-reporting to health services, additional tuberculosis transmission can take place before the patient is evaluated and receives appropriate therapy. Modeling of the tuberculosis epidemic in Tanzania suggests that DOTS may slow the incidence rate, but in the face of the HIV epidemic is unlikely to reverse the upward trend [13]. Again, some experts advocate targeted case finding, particularly among high-risk groups [11,14]. Diagnosis and treatment of active tuberculosis infection in patients with HIV infection This section deals specifically with the treatment of active tuberculosis in HIV-infected individuals as recommended by the CDC [15,16]. Again, many people co-infected with HIV and tuberculosis in the developing world do not have access to the relevant diagnostic tests and anti-tuberculosis and antiretroviral therapies. To impact tuberculosis-associated morbidity and mortality worldwide, developing countries will require both the skills and commodities to diagnose and treat individuals effectively. At present, tuberculosis treatment for the HIV-positive patient as recommended by the CDC is not applicable in most resource-poor countries. Treatment of active and latent tuberculosis infection in patients with HIV depends on the application of both clinical judgment and appropriate diagnostic tests. Active tuberculosis can occur at any CD4 cell count but atypical presentations are more likely with advanced HIV disease or AIDS. Clinicians should be alert to the sometimes atypical presentations of pulmonary and extrapulmonary tuberculosis in HIV-infected patients. Sputum acid fast staining, mycobacterial cultures, and drug susceptibility testing are recommended in all patients suspected of having tuberculosis. However, patients with HIV are slightly less likely to have positive sputum smears than non-HIV-infected individuals [17]. Likewise, chest radiographic findings can vary depending on the degree of immunosuppression. Patients with CD4 cell counts greater than 200 are more likely to have classic findings of upper lobe infiltrates with cavitary lesions, while those patients with AIDS may more likely have hilar adenopathy and pleural effusions without cavitations [18]. Mycobacteremia and extrapulmonary tuberculosis, especially meningitis and adenopathy, also correlate with diminishing numbers of CD4 cells and degree of immunosuppression. Thus, for patients with HIV infection, the diagnosis of active tuberculosis is more challenging. Clinical suspicion of tuberculosis in a patient known or suspected of being HIV-infected should result in prompt initiation of anti-tuberculosis therapy regardless of sputum staining or radiograph findings. The 1998 CDC recommendations for the treatment of tuberculosis among patients infected with HIV are summarized in Table 2[15]. The treatment of tuberculosis in the HIV-seropositive patient may differ from the standard treatment in the following ways: (i) choice of anti-tubercular regimen and dose adjustments; (ii) duration of treatment (ideally with directly observed therapy); (iii) promotion of antiretroviral therapy; and (iv) monitoring requirements. Due to rapid advances in the management of HIV disease, it is recommended that all patients co-infected with HIV and tuberculosis should be evaluated by a specialist to ensure optimal management.Table 2: Tuberculosis (TB) treatment recommendations for the HIV-seropositive patient.The treatment algorithm begins with establishing the patient's HIV status and whether the patient is on optimal antiretroviral therapy. All patients diagnosed with active tuberculosis should be HIV tested and, if seropositive, evaluated for antiretroviral therapy. It was previously felt that the diagnosis of active tuberculosis should result in the deferral of antiretroviral therapy. Early initiation of antiretroviral therapy is now recommended. While being treated for tuberculosis, the HIV-positive patient not receiving concurrent HIV therapy should be reassessed every 3 months for initiation of antiretroviral therapy. Known or suspected HIV-positive patients should receive prompt initiation of effective anti-tuberculosis therapy. If antiretroviral therapy is not started, the patient can typically receive standard anti-tuberculosis therapy: isoniazid, rifampin, pyrazinamide, and ethambutol. If the patient is to receive simultaneous anti-tuberculosis and antiretroviral therapy, the selected regimens and doses must account for significant drug-drug interactions between the rifamycins (rifampin, rifabutin, rifapentine) and the protease inhibitors and non-nucleoside reverse transcriptase inhibitors (NNRTI). Rifamycins, particularly rifampin, induce the hepatic cytochrome P-450 (CYP450) and reduce the serum levels of protease inhibitors, NNRTI, and other drugs metabolized by the CYP450 system. The CDC generally recommends the substitution of rifabutin, a less potent CYP450 inducer, for rifampin to allow simultaneous use of protease inhibitors and NNRTI. The use of rifampin with protease inhibitors or NNRTI is contraindicated, except in three antiretroviral combinations: (i) the NNRTI efavirenz and two nucleoside reverse transcriptase inhibitors (NRTI); (ii) the protease inhibitor ritonavir and one or more NRTI; and (iii) the combination of ritonavir and saquinavir, either hard-gel or soft-gel capsules [16]. Table 3 lists the recommended anti-tubercular drug doses, depending on frequency of administration and concurrent antiretroviral use.Table 3: Anti-tubercular drug doses.If the HIV-positive patient is already on an effective anti-retroviral regimen at the time of tuberculosis diagnosis, it is desirable to continue the patient on the same antiretroviral regimen with appropriate dose adjustments. If the patient is starting a new antiretroviral regimen, options include: (i) a rifabutin-based regimen with the necessary protease inhibitor or NNRTI dose adjustments; (ii) a non-rifamycin-containing regimen such as isoniazid, streptomycin, pyrazinamide, ethambutol for 2 months, then isoniazid, streptomycin, ethambutol for 7 months; or (iii) a regimen that does not contain a protease inhibitor or NNRTI. Table 4 lists the recommended dose adjustments for rifabutin-based regimens. Given the complexity of these drug interactions, it is recommended that the selection of dual anti-tubercular and antiretroviral therapies be made following consultation with a specialist.Table 4: Protease inhibitor (PI) or non-nucleoside reverse transcriptase inhibitor (NNRTI) dose adjustments with rifabutin.Among the protease inhibitors, ritonavir has the highest potency in inhibiting the CYP450 pathway. With any dose of ritonavir, including low-dose ritonavir 100 mg twice a day, a reduced dose of rifabutin (150 mg two or three times per week) is recommended. According to the 2000 CDC guidelines for the use of rifabutin or rifampin among patients taking protease inhibitors or NNRTI, co-administration of ritonavir with the usual dose of rifampin (600 mg daily or two or three times per week) may be an option but pharmokinetic and clinical data are limited [16]. As previously stated, a patient taking the combination of saquinavir (either the soft-gel or hard-gel capsule) and ritonavir should take the reduced dose of rifabutin (150 mg two or three times per week). The saquinavir-ritonavir combination may possibly be given with the usual dose of rifampin but, again, limited pharmacokinetic and clinical data are available. Saquinavir, as a sole protease inhibitor, is generally not recommended in combination with rifabutin because the serum levels of saquinavir may be decreased as much as 45%. Indinavir, nelfinavir, and amprenavir should not be used in combination with rifampin, but all three protease inhibitors can be administered with a reduced daily dose of rifabutin (150 mg daily) or the usual dose of rifabutin (300 mg two or three times per week). Efavirenz induces CYP450 and accelerates rifamycin metabolism; therefore, efavirenz should be co-administered with an increased dose of rifabutin (450 or 600 mg daily, or 600 mg two or three times per week). According to the recent CDC guidelines, efavirenz may be combined with the usual dose of rifampin (600 mg daily or two or three times per week). The NNRTI nevirapine should typically be given with the usual dose of rifabutin (300 mg daily or two or three times per week). If co-administration of nevirapine with rifampin is clearly indicated, careful monitoring is recommended. The use of delavirdine is contraindicated during the treatment of tuberculosis because the drug levels are markedly decreased with both rifampin and rifabutin. HIV-infected patients have a higher incidence of drug-resistant tuberculosis isolates than non-HIV-infected patients. There have been several reports of increased risk of rifampin resistance among HIV-positive patients [19-21]. Higher rates of drug-resistant tuberculosis in HIV-infected patients may be associated with biological, behavioral, and societal factors including drug malabsorption, non-adherence, nosocomial outbreaks, and inadequate drug therapy in countries with high rates of co-infection. Directly observed therapy, regarded as the best strategy for ensuring adherence and limiting drug-resistant tuberculosis, is recommended for all patients with HIV infection [15]. In a randomized, controlled trial of anti-tuberculosis therapy among HIV-positive patients in Baltimore, Maryland, patients who received supervised therapy for tuberculosis had better survival than those who self-administered therapy [22]. Treatment of drug-resistant tuberculosis involves initiation of a multi-drug regimen tailored to the susceptibility profile of the organism. Due to the diversity of resistance patterns, it is not possible to recommend standardized protocols for therapy. Any regimen should include two or more drugs to which the isolate is susceptible. For HIV-positive patients at risk of MDR-TB, defined as resistance to both isoniazid and rifampin, initial empiric treatment should include second-line tuberculosis drugs to which resistance is uncommon. The 1999 WHO Essential Drug List was modified to include the following second-line tuberculosis drugs: fluoroquinolones (ciprofloxacin, ofloxacin, levofloxacin), amikacin, kanamycin, capreomycin, cycloserine, paraaminosalicylic acid, and ethionamide. Cultures and drug susceptibility testing should be routinely performed on all tuberculosis isolates and the anti-tuberculosis regimens adjusted accordingly. Although potentially curable, MDR-TB is associated with a high mortality rate (35-50%), especially in resource-poor countries where the detection of drug resistance takes weeks to months, if at all, and the second-line drugs are often unavailable. Six months is the minimum duration of treatment for pan-susceptible, pulmonary tuberculosis among HIV-positive patients in the United States when a rifamycin-based regimen is used [15]. Prolonged treatment is indicated for patients with a delayed clinical response or delayed conversion of sputum cultures from positive to negative. Some experts recommend the use of 9-month treatment regimens in all patients with HIV infection, especially among patients with advanced HIV-related immunosuppression [17]. Extended duration of treatment is clearly recommended in patients with slow clinical response or with meningeal or bone/joint involvement. HIV-positive individuals with active tuberculosis appear to be at increased risk of tuberculosis recurrence after successful completion of a 6-month rifampin-containing regimen. A randomized trial of post-treatment isoniazid prophylaxis was conducted among HIV-positive patients in Port au Prince who completed a 6-month rifampin-containing regimen [23]. One year of post-treatment isoniazid decreased the risk of a second episode of tuberculosis among HIV-positive patients. This study did not analyze whether these episodes were due to recurrent M. tuberculosis or exogenous re-infection, both of which have previously been described [24]. The investigators recommend consideration of isoniazid prophylaxis for HIV-positive patients after completion of tuberculosis therapy. At present, the WHO and the CDC recommend a minimum 6 months of directly observed tuberculosis therapy without post-treatment prophylaxis. Treatment of latent tuberculosis infection in patients with HIV infection It is estimated that 2 billion people worldwide are infected with latent tuberculosis. Patients with HIV infection are at increased risk of progressing to active disease [2]. In 2000, the ATS and the CDC issued new guidelines for the diagnosis and treatment of latent tuberculosis infection [25]. This preventive public health strategy encompasses both secondary prevention by treating the latent infection before it progresses to active infection and primary prevention by halting further tuberculosis transmission. In addition, the use of highly active antiretroviral therapy has been shown to reduce the incidence of tuberculosis among persons with HIV infection [26]. The new ATS/CDC guidelines recommend targeted screening of populations and patients at increased risk of tuberculosis infection who would benefit from treatment to prevent active disease. Populations listed for targeted screening include organ transplant patients, injection drug users, health care workers, residents of homeless shelters, and all patients with HIV infection. Diagnosis of latent tuberculosis infection is based on the tuberculin skin test, although new diagnostics, including cytokine detection assays, are being developed [27,28]. A commercially available interferon-gamma release assay (QuantiFERON™ Cellestis Limited, St Kilda, Victoria, Australia) has modest correlation with tuberculin skin testing in humans [27,28]. Large multicenter studies are ongoing to test the diagnostic efficacy of this assay in the United States and in populations with varying degrees of risk for latent tuberculosis. There would be two significant advantages of developing blood tests for the detection of latent M. tuberculosis infection. Blood tests would alleviate the need for a return health service visit at 48-72 h and their interpretation would be less subjective than tuberculin skin testing. Detection of latent tuberculosis in an HIV-infected patient can be a particular diagnostic challenge. It has long been appreciated that the sensitivity of tuberculin skin testing may be depressed in patients with critical illness. Reactivity to tuberculin, as well as to mumps and Candida antigens, can fluctuate in persons with HIV infection [29], and instability of the anergic state is associated with higher CD4 cell counts. For these reasons, anergy testing is no longer recommended for the diagnosis of latent tuberculosis. Current ATS/CDC guidelines are to treat latent tuberculosis infection in HIV-infected persons with ≥5 mm induration. Persons at high-risk of latent tuberculosis or with recent close contact to a case of active tuberculosis should receive preventive therapy regardless of tuberculin reactivity. Isoniazid has long been the mainstay of treatment of latent tuberculosis [30]. Several clinical trials have demonstrated a 50-90% reduction in the risk of progressing to active tuberculosis following 6-12 months of isoniazid therapy. The optimal duration of isoniazid therapy has also been investigated, including the landmark trial of the International Union Against Tuberculosis and Lung Disease. Conducted in Eastern Europe in the 1970s and 1980s, this study compared 3, 6, or 12 months of isoniazid therapy versus placebo. Twelve months of isoniazid therapy reduced the tuberculosis incidence by 75% compared with 66% for 6 months and 20% for 3 months; however, compliance with 12 months therapy was considerably poorer than for shorter courses of therapy [31]. A recent analysis of isoniazid therapy in Bethel, Alaska revealed that maximal efficacy from isoniazid therapy occurred after 9 months, with no additional benefit associated with longer therapy [32]. The new ATS and CDC guidelines recommend isoniazid therapy for 9 months (300 mg isoniazid daily plus 50 mg pyridoxine daily), with 6 months of therapy as a less preferred alternative. The disadvantages to 9-month isoniazid therapy include potential hepatotoxicity, cost, and poor adherence. Directly observed preventive therapy has been shown to improve adherence but is not widely available [33]. An important advance in the treatment of latent tuberculosis has been the development of short-course regimens that may improve adherence and reduce cost. Several studies have assessed the efficacy of short-course regimens for the treatment of latent tuberculosis in HIV-infected, tuberculin-positive individuals [34-36]. A multinational trial found that daily rifampin and pyrazinamide for 2 months was equivalent to isoniazid for 12 months for the treatment of latent tuberculosis in HIV-infected persons [37]. On the basis of these studies, the new ATS/CDC guidelines endorse short-course regimens for the treatment of latent tuberculosis as summarized in Table 5 from their publication [25]. Daily rifampin and pyrazinamide for 2 months (600 mg rifampin daily plus 20 mg/kg pyrazinamide daily) is the preferred short-course regimen for HIV-infected patients and would be indicated for suspected cases of isoniazid-resistant tuberculosis. Rifabutin should be substituted for rifampin in patients receiving protease inhibitors or NNRTI, although further research is needed to establish equivalent efficacy in the treatment of latent tuberculosis. The same drug-drug interactions and dose adjustments for antiretroviral drugs and rifamycins apply. It is noteworthy that the ATS/CDC expert panel also recommends the 2-month rifampin/pyrazinamide regimen for HIV-negative individuals despite the fact the clinical trials for this combination have been carried out only in HIV-positive patients. If MDR-TB is suspected, the recommended preventive therapy is pyrazinamide and ethambutol or pyrazinamide and a fluoroquinolone (i.e., levofloxacin or ofloxacin) for 6-12 months. Treatment for suspected exposure to MDR-TB should be routinely extended to 12 months for HIV-positive individuals.Table 5: American Thoracic Society and Centers for Disease Control and Prevention guidelines for the treatment of latent tuberculosis infection.Recently, the use of rifampin plus pyrazinamide preventive therapy in non-HIV infected patients has been associated with a concerning rate of hepatotoxicity. Twenty-one patients with liver injury resulting from the use of rifampin/pyrazinamide preventive therapy have been reported to the CDC; five of these patients died. Based on these findings, CDC and ATS have revised their earlier 2000 recommendations [38]. The revisions recommend: (i) that rifampin/pyrazinamide be used with caution particularly in patients with pre-existing liver disease, dependency on alcohol, or those taking hepatotoxic drugs; and (ii) that patients who are treated with rifampin/pyrazinamide be seen by a health care provider at 2, 4, and 6 weeks for a determination of liver function tests and medical follow-up evaluation with a final visit at 8 weeks to document treatment completion. Preventive therapy with the regimen should be terminated for asymptomatic transaminase elevation. Surprisingly, significant rates of liver injury have not been observed in HIV-infected patients treated with this regimen, either in clinical trials or subsequent to the expanded use of the regimen in 2000. For HIV-infected patients exposed to persons with active tuberculosis, presumptive treatment for latent tuberculosis should be instituted despite a negative tuberculin test. In addition, some experts advocate presumptively treating HIV-infected individuals who reside in high-risk environments. Patient evaluation for latent tuberculosis treatment should include a careful history and physical examination and, whenever possible, chest radiograph to exclude active tuberculosis. New anti-tubercular drugs in development Future treatment of tuberculosis, particularly for MDR-TB, will depend on the development of new anti-tubercular drugs. Currently, there are several drugs under investigation that have in vitro activity against M. tuberculosis. In some cases the drug has also demonstrated activity in a murine model of tuberculosis, and some have received Food and Drug Administration approval for other indications. Drugs under study include rifapentine, moxifloxacin, linezolid and other oxazolidinones, and nitroimidazoles. Rifapentine is a long-lasting rifamycin derivative, the serum half-life of which is three times longer than that of the parent compound rifampin [39]. The minimal inhibitory concentration of rifapentine against M. tuberculosis is similar to or one dilution inferior to that of rifampin. Given the cross-resistance between rifapentine and other rifamycin derivatives, the advantage of rifapentine over rifampin lies its pharmokinetic properties. In the immunocompetent mouse infected with tuberculosis [40] and in immunocompetent tuberculosis patients, a rifapentine-containing isoniazid-combined regimen once weekly is marginally less active than a daily rifampin-containing isoniazid-combined regimen, both rifamycins being given at 10 mg/kg. In immunodeficient tuberculosis patients, however, the rifapentine-containing regimen once weekly is significantly less active than the daily rifampin-containing regimen [41]. In the future, weekly rifapentine-containing regimens during the continuation phase of tuberculosis therapy may provide increased convenience to patients, as well as cost savings to health care delivery programs. Moxifloxacin is an 8-methoxyquinolone approved for pneumonia and other respiratory tract infections that has in vitro activity against M. tuberculosis, equivalent to that of sparfloxacin and clinafloxacin [42]. Two in vivo studies in a murine model of tuberculosis found that moxifloxacin at 100 mg/kg/day is as bactericidal as isoniazid at 25 mg/kg/day, and more strongly bactericidal than sparfloxacin [42,43]. Several clinical trials are being organized to test the role of moxifloxacin in combination therapy against active tuberculosis. Two oxazolidinones, linezolid and PNU-100480, also show promising activity in a murine model of tuberculosis [44]. Oxazolidinones are a class of antibacterial protein synthesis inhibitors with activity against methicillin-resistant Staphylococcus aureus, penicillin-resistant Streptococcus pneumoniae, and vancomycin-resistant enterococci. The activity of PNU-100480 at 100 mg/kg/day is comparable with isoniazid at 25 mg/kg per day. Linezolid, an oral agent approved by the Food and Drug Administration for the treatment of vancomycin-resistant enterococci, appears to be less active against M. tuberculosis than PNU-100480. It has been recommended, however, for further study at higher doses. PA-824 is a new nitroimidazopyran related to metronidazole that has been found to possess potent activity in vitro against M. tuberculosis[45]. Multidrug-resistant strains in vitro were also found to be sensitive to this agent with minimal inhibitory concentrations of 0.03-0.25 μg/ml. In a murine model, administration of this drug at a dose of 25-100 mg/kg per day led to reduction of disease burden in spleen and lung tissue comparable with that achieved with 25 mg/kg isoniazid per day. Social and behavioral issues in the treatment of tuberculosis in patients with HIV We have previously alluded to some of the social and economic issues that severely limit the treatment of tuberculosis in resource-poor countries. At the individual and community level, there may also exist some important knowledge and behavioral constraints to effective tuberculosis treatment. A study among tuberculosis patients in Mwanza, Tanzania found that only 30% of the patients had satisfactory knowledge of the disease and its treatment [46]. In a recent study from Chiang Rai, Thailand, people were generally well informed about HIV but knew little about tuberculosis [47]. Symptoms of weight loss, fever, and cough were attributed to AIDS rather than tuberculosis. HIV-negative patients with tuberculosis were often perceived as having AIDS. Due to the stigma associated with HIV, the researchers conclude that patients with tuberculosis may not seek or adhere to appropriate care. These studies suggest the important impact that the HIV epidemic may have on the public perception and response to tuberculosis. Conclusions Recent trends suggest a rising epidemic of tuberculosis and HIV co-infection in many parts of the developing world. The increased prevalence of tuberculosis and HIV/AIDS and the rise of MDR-TB pose a serious health threat to all societies. While significant improvements have been made in developing standard, efficient, short-course regimens for the treatment of active and latent tuberculosis, therapy is still prolonged and few second-line drugs are available when drug resistance is detected. Two reasons for optimism are the cross-over anti-tuberculosis activity of drugs such as linezolid and moxifloxacin and the development of rifapentine, which may permit once-weekly treatment. The optimal use of these drugs in combination therapy is a promising area of active research. In addition, we advocate the development of new and affordable tuberculosis drugs that will be made accessible to patients in developing countries.