New CDC guidelines on co-infected patients

Stopping protease inhibitors not recommended

Nearly two years after updating recommendations on treating patients co-infected with tuberculosis and HIV, the Centers for Disease Control and Prevention (CDC) soon will publish a comprehensive set of recommendations that take into account the latest studies on treating this population. Most of the 1996 updated guidelines still stand, except clinicians are no longer advised to stop protease inhibitor therapy to allow the use of rifampin for TB therapy.

"We have realized that a lot of users are often not going to have ready access to the multiple references on this issue, and therefore, this is an attempt to put everything in one place," says Kenneth Castro, MD, director of the CDC's division of TB elimination.

A draft of the guidelines was presented recently to the CDC's Advisory Committee for the Elimin ation of TB for review and comments. The members suggested only minor changes, and the guidelines are expected to be published this summer in the Morbidity and Mortality Weekly Report.

"Even though we are going through some updates, I don't think it is really updating any of the major standing recommendations," says Elsa Villarino, MD, MPH, chief of the therapeutic and diagnostic studies section of the CDC's division of tuberculosis elimination. "The main purpose of this document is to review the newest information that is still in question."

The guidelines summarize new information on using directly observed therapy for co-infected patients, prescribing nonrifampin-containing regimens for treating TB patients taking antiretroviral drugs that have significant interaction with rifampin, monitoring responses to treatment for deciding whether to extend TB treatment beyond six months, and prescribing short-course, multidrug TB preventive therapy.

The document provides a list of treatment options in table form and will be graded according to the strength of study data backing them. A treatment algorithm also has been created, providing a simple flowchart for managing co-infection.

The guidelines follow a seven-year decline in national TB rates. Nonetheless, the rate of co-infection remains significant; matches of state AIDS and TB registries in 1994 found 14% of TB cases with an AIDS match. The annual risk of TB disease in tuberculin skin-test-positive, HIV-infected Americans ranges from 2% to 5%, which is 10 times higher than in HIV-negative groups, the document notes.

Also, the risk of drug-resistant TB in HIV-positive patients remains high. A population-based study by the CDC from 1993 to 1996 found that among U.S.-born patients ages 25 to 44, resistance rates were significantly higher among those with HIV infection: 11% were resistant to isoniazid, 9% to rifampin, and 6% to both. While the reasons for the disparity are unclear, one explanation is that HIV-positive patients are at increased risk for exposure to patients with drug-resistant TB.

New findings in co-pathogenesis

Recent research on the effects of co-infection with TB and HIV underscores the importance of early diagnosis and treatment. In summarizing new findings, the draft notes recent in vivo data showing a fivefold increase of plasma HIV RNA in HIV-positive patients infected with TB vs. those without TB infection. There also is evidence anti-TB treatment leads to a reduction in viral load in dually infected patients. "The impact of HIV on the natural history of TB and the effect of active TB on the clinical course of HIV infection is bi-directional in nature and support the recommendation for early and aggressive treatment of both diseases," the document states.

One area of research that remains equivocal is whether co-infected patients have a greater risk of TB relapse than patients not infected with HIV. The CDC recommends six months of therapy for drug-susceptible TB disease in patients co-infected with HIV, but the agency also suggests treating for six months after documented culture conversion if the therapeutic response has been slow.

Villarino says of the five published trials study ing therapy duration and relapse in co-infected patients, three suggest six months of effective TB therapy cures most HIV-positive patients, and relapse remains relatively low.

The awareness that rifampin would be contraindicated with most protease inhibitors led the CDC to issue its 1996 treatment update in which the dilemma of whether to treat HIV or TB first or exclusively was worked out with several options. Since then, research results have strengthened the appropriateness of using nonrifampin-containing regimens in patients who already are being treated for HIV. The reasons for changing this recommendation include that Rifabutin may be absorbed more reliably than rifampin in patients with advanced disease, and it may have fewer drug-drug interactions with agents used to treat opportunistic infections (i.e., azole antifungal drugs, anticonvulsants, and steroids).

Three trials have shown the drug's efficacy in TB. The largest found that six-month regimens containing rifabutin at 150 mg or 300 mg daily were as effective and safe as a control regimen of 600 mg of rifampin. Villarino also related the experience of a Florida researcher who treated more than 50 co-infected patients for six months with twice-weekly INH and rifabutin in the maintenance phase and reported no relapses after a year of follow-up.

The draft states a rifampin-containing regimen in HIV-positive patients is reasonable for patients who haven't started or aren't candidates for immediate initiation of HIV therapy with protease inhibitors or non-nucleoside reverse transcriptase inhibitors (NNRTIs), and for those who have failed combination antiretroviral regimens. In the latter case, the two types of drugs will not be used in salvage therapy.

If a patient has begun therapy with protease inhibitors or NNRTIs, the CDC says, interruption of therapy is no longer an option because the risks of HIV resistance are too high. Consequently, those patients needing concurrent HIV and TB therapy should be on TB regimens including rifabutin instead of rifampin, or they should be on a nine- to 12-month TB regimen without rifampin but including streptomycin for at least four months.

"HIV therapy with drug combinations containing any of the available protease inhibitors or NNRTIs would not be affected by any TB treatment regimens that do not contain a rifamycin or would be minimally affected by most anti-TB regimens containing rifabutin," the draft guidelines state. "However, the administration of Ritonavir or Delavirdine with rifabutin, and of rifampin with any of the protease inhibitors or NNRTIs, is contraindicated."

For preventive therapy in HIV-positive patients, the CDC will issue updated recommendations possibly before the end of the year. Eight trials have evaluated such patients, and the majority suggest daily INH for six to 12 months may reduce TB incidence by up to 80% for PPD-positive, HIV-positive patients, Villarino says. More recent studies have shown preventive therapy with two drugs - rifampin plus pyrazinamide, and rifampin plus INH - can cut treatment time to three months.