© 2007 American Thoracic Society doi: 10.1164/rccm.200702-248ED
Treatment of Tuberculosis in HIV-infected Patients: We Need to Know MoreBeth Israel Medical Center, New York, New York
Tuberculosis and Chest Service, Centre for Health Protection, Department of Health, Hong Kong, China
Grantham Hospital, Hong Kong, China Varying failure-relapse rates (0–10%) have been reported among HIV-infected tuberculosis (TB) patients treated with 6-mo standard short-course chemotherapy regimens (1). Selective attrition due to high mortality and heterogeneity in study design precluded more precise characterization of outcomes (1). In this issue of the Journal, Nahid and coworkers (2) found that HIV-infected TB patients had a significantly higher relapse rate than patients not infected with HIV (9.3 versus 1.0 per 100 person-years). In contrast to some earlier studies (1), sputum cultures were regularly performed. Molecular analysis confirmed that recurrences were true relapses, rather than reinfection. Despite the current recommendation of the same 6-mo anti-TB regimen for both HIV-infected and uninfected subjects (3), longer durations of treatment were actually observed in program settings (2). Similar to a previous study (4), HIV-infected patients given 6 mo of anti-TB treatment had a higher relapse risk than those treated longer (adjusted hazard ratio [AHR] 4.33, p = 0.02). Intermittent therapy was also associated with relapse (AHR 4.12, p = 0.04), corroborating previously published data regarding intermittent use of rifabutin and rifampin (5, 6), largely during the intensive phase. Furthermore, acquired drug resistance was much more common (4.2% in HIV-infected versus 0.5% in HIV-uninfected, p < 0.001), and 9 of 11 cases of acquired resistance involved rifamycins. Indeed, in this (2) and other recent series (5), there were treatment failures with acquired multidrug-resistant TB (MDR-TB), highlighting the need to evaluate regimens not solely based on failure-relapse rates, but perhaps even more importantly on rates of acquired resistance. From a program perspective, the ability to use the same 6-mo short-course anti-TB regimen in both HIV-infected and uninfected patients is attractive, especially in resource-limited areas. However, given the global estimate of 741,000 HIV-TB cases annually, even a low rate of acquired rifamycin resistance can have significant consequences. With the findings of the current study (2) and the recent eye-catching report of extensively drug-resistant TB (XDR-TB) among HIV-coinfected subjects (7), it seems timely for a critical reappraisal. Given the profound degree of immune compromise and even malabsorption of rifamycins and isoniazid/ethambutol (8, 9) in the advanced stages of HIV infection, it is perhaps counter-intuitive to conceive that TB can be managed in exactly the same way as in immunocompetent individuals. Further, both TB and HIV are associated with multiple adverse social factors, and social inequities are commonly associated with suboptimal health care infrastructures in many resource-limited areas. A standard short-course regimen administered under conditions that promote adherence and deter acquired resistance, such as directly observed therapy (DOTS), will likely be crucial in program implementation in highly affected resource-limited areas. In the study by Nahid and coworkers (2), longer duration of treatment did not, by itself, prevent emergence of drug resistance. Program operational efficiency, medication adherence, and the intrinsic ability of a regimen to deter the selection of drug-resistant mutants are all critical determinants for the prevention of drug resistance. Therefore, the search for an optimal, and preferably intermittent, regimen suitable for field application must be continued. Rifamycin monoresistance has been associated with once-weekly rifapentine (600 mg) and isoniazid (900 mg) during the continuation phase of TB treatment in HIV-infected subjects (10). A 900 mg dose of rifapentine has a more favorable pharmacokinetic profile than a 600 mg dose (11), and a 1,200 mg dose of rifapentine has also been shown to produce an optimal pulse size and postantibiotic lag on the growth of Mycobacterium tuberculosis (12). In a murine model, twice-weekly regimens containing rifapentine (15–20 mg/kg) have shown marked anti-TB potency by enhancing rifamycin exposure (13). Currently, there are only preliminary data on the safety and tolerance of rifapentine dosed at 900 mg once weekly (14). Further evaluation of more frequent or higher rifapentine dosing appears warranted. However, whether the answers are to be found in the use of higher dose rifamycins, or in ensuring adequate isoniazid exposure (9) remains to be proven, and HIV-related TB is proving to be very unforgiving of reductions in dose, frequency, or duration. Further data on more intensive intermittent regimens, longer multidrug induction periods, perhaps with longer total therapy, are also needed. With uncertainty over the optimal treatment duration, there is also a need for a simple prognostic marker that can be utilized in resource-limited areas. The rising proportion of pulmonary TB cases that are smear-negative (linked to HIV-infection) increasingly complicates diagnosis and undermines the utility of the sputum smear as an epidemiological or clinical monitoring tool. Sputum culture is not suitable as a bedside indicator, and has reduced sensitivity in extrapulmonary disease. There is some evidence of the possible prognostic value of early body weight gain among underweight HIV-negative subjects during TB treatment (15). It will be useful to explore whether the same applies to HIV-infected patients. In the article by Nahid and coworkers in this issue (pp. 1199–1206), use of highly active antiretroviral therapy (HAART) during TB treatment significantly protected HIV-infected individuals against mortality (2). Furthermore, HIV-infected patients who received HAART during TB treatment had more rapid bacteriologic conversion than those not so treated. This association suggests that HAART may facilitate an early reduction in the mycobacterial burden, and therefore suggests a potential for HAART to reduce the risk of acquired resistance that should be further examined. None of those who received concomitant HAART relapsed, although the small sample size limits the ability to draw conclusions. Despite another similar report of reduced mortality by initiation of antiretroviral therapy within 6 mo of TB diagnosis (16), the optimal timing of HAART initiation remains controversial. Concerns about immune reconstitution and other adverse events must be carefully balanced against the risk of early mortality, disease progression, and acquired resistance. Prospective clinical studies on the timing of HAART initiation are underway and should be informative. Access to secondline drugs in well-functioning DOTS-plus programs is central to the control of MDR-TB. Fluoroquinolones play a pivotal role in existing secondline regimens (17), but resistance to fluoroquinolones can occur easily in the absence of adequate companion drugs. Erratic access to secondline anti-TB drugs in areas with high HIV prevalence has proven a fertile ground for the emergence of XDR-TB (7). However, in the treatment of HIV-related TB, access to essential drugs alone is far from sufficient. To avert a looming disaster, there is an urgent need to strengthen both the HIV and TB control programs in areas with high rates of HIV-related TB. FOOTNOTES Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
Related articles in AJRCCM:
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
