© 2004 American Thoracic Society doi: 10.1164/rccm.2409005
In the Treatment of Tuberculosis, You Get What You Pay for...U.S. Centers for Disease Control and Prevention Atlanta, Georgia In this issue of the Journal, Chang and colleagues (pp. 1124–1130) present a well executed case–control study of risk factors for early relapse after treatment of tuberculosis (1). This is the latest of several recent studies in HIV-negative persons that have assessed associations with relapse after the completion of apparently adequate short-course treatments. A consistent theme has begun to appear: more extensive disease requires more treatment, and the fewer total doses, the higher the risk that treatment will prove inadequate (1–7). In other words, in the treatment of tuberculosis, you get what you pay for... In their analysis, Chang and coworkers have reviewed 12,946 tuberculosis cases seen from 1998 to 2000 in the public clinics in Hong Kong, and who had completed therapy within 12 months. Among these, they identified 113 (0.93%) who relapsed within 30 months, and matched each to two control subjects by sex, age, year of treatment initiation, clinic at which they were treated, and baseline sputum culture result. In conditional logistic regression analysis, relapse was statistically significantly associated with thrice-weekly (versus daily) therapy, and with the presence of cavitation at diagnosis. Prolongation of therapy significantly lowered the estimated odds of relapse. Other factors associated with relapse included the presence of extrapulmonary site of disease at diagnosis, and the presence of conditions predisposing to tuberculosis (e.g., diabetes, silicosis, steroid therapy). Three quarters of the relapse patients were tested for HIV, and all had negative results. A reasonable definition of tuberculosis was used, and an analysis limited to the 75% with positive culture results produced similar findings. The overall reported rate of relapse of approximately 1% is low compared with what has been seen in recent clinical trials, but it seems unlikely that missed cases of relapse would lead to serious bias in the findings. The mean time to relapse is high (14.9 months) compared with recent studies, but this may be due simply to the diagnostic delay experienced by programs around the world. Importantly, this study is adequately powered to compare the two approaches of daily versus thrice-weekly therapy. One of the earliest regression analyses of risk factors for tuberculosis relapse was published 26 years ago by Aber and Nunn (2). Over the next 20 years, only modest attention was paid to the role of such factors, while attention was focused on the important role of variations in the treatment regimen such as choice of drugs, duration of intensive phase, and the role of pyrazinamide in short-course chemotherapy (8). More recent studies have searched for features that portend an increased risk of relapse after short-course treatment. While some studies have identified a role for individual characteristics such as sex (4) or comorbidity (9), all have shown that the extent of disease at diagnosis (e.g., presence of cavitation, extent of disease on chest X-ray), the early response to therapy (e.g., sputum culture positivity at 2 months), and the amount of therapy administered (i.e., number of doses, duration of therapy, intermittency of therapy) are highly significant factors predicting relapse (Table 1).
So why is this news? Isn't it rather obvious that more is better than less? In fact, the dogma for the past 15 years has been that intermittent regimens are highly efficacious and more cost-effective for tuberculosis control programs, especially for those committed to use of directly observed therapy. In vitro evidence supported this approach (10), as well as clinical trial and program data (11, 12). It was thus surprising in Tuberculosis Trials Consortium Study 22 to find a suggestion that the standard twice-weekly regimen performed more poorly than the daily or thrice-weekly regimens, and it was even more surprising that two potent risk factors (i.e., positive 2-month sputum culture results and cavitation) could identify a group whose risk of relapse exceeded 20 percent. (3) Furthermore, tuberculosis control staff in the field confirmed the Study 22 findings when they analyzed recent program data (13). Although Study 22 was not powered to allow a comparison of daily versus thrice-weekly treatment, the study by Chang and colleagues is, and leads us to important reflection. What should we conclude? First, more is more and less is less. That is, for regimens with the same drugs, more treatment means more cures, and vice versa. Second, programs need to consider some individualization of therapy—not for the routine patient, but rather for the rarer patient with very extensive disease or cavitation who is slow to respond to therapy (14). Third, these considerations need not discourage us from use of intermittent therapy, but should remind us that sophisticated management based on case-specific circumstances is still needed. We should not be surprised, in an era in which we are learning weekly about the advances of host and bacterial genomics, that individuals may differ in their responses to therapy, and that these differences may sometimes have implications for treatment. Programs must evaluate the value of reductions in relapse and transmission relative to the costs of directly observed therapy (DOT) and drugs. We should be cautious about accepting dogma, and always eager to evaluate our own programmatic and clinical experience: Nothing has such power to broaden the mind as the ability to investigate systematically and truly all that comes under thy observation in life. (15) FOOTNOTES Conflict of Interest Statement: A.A.V. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; M.F.I. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
