Antituberculosis Agents General Statement

Aminosalicylic Acid, Aminosalicylate Sodium, Capreomycin Sulfate, Clofazimine, Cycloserine, Ethambutol Hydrochloride, Ethionamide, Isoniazid, Kanamycin Sulfate, Pyrazinamide, Rifabutin, Rifampin, Rifapentine, Streptomycin Sulfate

General Principles in Antituberculosis Therapy

Antituberculosis agents are antibiotics and synthetic anti-infectives used in the treatment of tuberculosis and other diseases caused by organisms of the genus Mycobacterium. Isoniazid, rifampin, ethambutol, and pyrazinamide are the drugs used most frequently in the treatment of tuberculosis and are considered first-line agents for use in antituberculosis regimens. Rifapentine and rifabutin, like rifampin, are rifamycin derivatives; these drugs also are considered first-line agents and are used as alternatives to rifampin in antituberculosis regimens. Other antituberculosis agents currently available in the US are considered second-line agents and include aminosalicylic acid, capreomycin, cycloserine, ethionamide, and certain aminoglycosides (streptomycin, amikacin, kanamycin).

Certain fluoroquinolones (e.g., gatifloxacin, levofloxacin, moxifloxacin) also are considered second-line agents for the treatment of tuberculosis when first-line agents cannot be used because of resistance or intolerance. In general, second-line antituberculosis agents may be more toxic and less effective than the first-line antituberculosis agents and are used when the first-line agents are contraindicated or are ineffective because of bacterial resistance.

Latent Tuberculosis Infection versus Active Tuberculosis

Tuberculosis has 2 stages: asymptomatic (latent) infection with Mycobacterium tuberculosis and clinical disease. Previously, the use of a simple drug regimen (e.g., isoniazid monotherapy) to prevent the development of active tuberculosis disease in individuals known or likely to be infected with M. tuberculosis was termed “preventive therapy” or “chemoprophylaxis”. If infection with M. tuberculosis has occurred but there is no clinical evidence of tuberculosis, daily administration of isoniazid alone for 6-12 months prevents development of active disease in a high percentage of patients. However, since use of such a regimen rarely results in true primary prevention (i.e., prevention of infection in individuals exposed to infectious tuberculosis), the American Thoracic Society (ATS) and US Centers for Disease Control and Prevention (CDC) state that “treatment of latent tuberculosis infection” rather than “preventive therapy” more accurately describes the intended intervention and potentially will result in greater understanding and more widespread implementation of this tuberculosis control strategy.

If untreated, latent M. tuberculosis infection may progress to pulmonary and/or extrapulmonary tuberculosis that requires more aggressive treatment than that used for treatment of latent tuberculosis infection. Multiple-drug regimens are necessary to treat clinical tuberculosis and prevent relapse; no antituberculosis agent should be used alone in the treatment of active tuberculosis. The use of multiple agents rapidly decreases infectiousness and may delay or prevent emergence of resistant organisms. The number of antituberculosis agents and the specific agents used depend on the severity of the disease, history of prior antituberculosis agent therapy, and in vitro susceptibility of the infecting organism to the agents available.

Patient Compliance

Patient compliance is crucial to the success of antimycobacterial therapy, and patients should be taught the necessity of adhering to the drug regimen for the full duration of treatment. The principal reason why cures are not achieved with available drug regimens for tuberculosis is patient noncompliance or failure to complete the prescribed regimen. As a potential means of combating this problem, regimens currently recommended by the ATS, CDC, and Infectious Diseases Society of America (IDSA) for treatment of uncomplicated pulmonary and most cases of extrapulmonary tuberculosis have a duration of 6-9 months (minimum duration 6 months) and are shorter than the more prolonged regimens previously recommended (18-24 months). In addition, the ATS, CDC, and IDSA recommend that directly observed (supervised) therapy (DOT) be used whenever possible to ensure compliance. Current evidence suggests that treatment completion rates exceeding 90% are possible when DOT is used with multiple incentives and enablers (e.g., intermittent regimens designed around a patient’s lifestyle; social and economic incentives such as food, clothing, and transportation; and culturally appropriate outreach). Fixed-combination preparations containing isoniazid, rifampin, and pyrazinamide (Rifater®) or isoniazid and rifampin (Rifamate®) are commercially available for the treatment of active tuberculosis, and use of these fixed-combination preparations can enhance patient adherence and are especially useful when DOT is not possible.

Duration of Therapy

Individualization of therapy, possibly including extension of the duration of treatment, is particularly important if patient compliance with the multiple-drug antituberculosis regimen is seriously questioned, if there have been complicating medical conditions (e.g., HIV infection, silicosis, diabetes, hematologic or reticuloendothelial malignancies, immunosuppressive therapy, chronic renal failure, malnutrition), or if there is evidence of serious extrapulmonary (e.g., meningeal) or complicated pulmonary (e.g., empyema) tuberculosis.

Interruptions in treatment may have a substantial effect on the duration of therapy needed to complete treatment. Factors to consider when establishing the date of completion include the total number of doses given, length of any interruptions in therapy, time during therapy when interruptions occurred (early or late in therapy), and the patient’s clinical, radiographic, and bacteriologic status before, during, and after interruption of therapy. In general, the earlier in treatment and the longer the duration of the interruption, the more serious the effect and the greater the need to restart therapy from the beginning.

The usual duration of treatment for most cases of pulmonary and extrapulmonary tuberculosis (except disseminated infections and tuberculous meningitis) is 6-9 months. However, the ATS, CDC, and IDSA state that completion of treatment is determined more accurately by the total number of doses and should not be based solely on the duration of therapy.

Although the optimal drug regimen and duration of treatment for tuberculosis meningitis has not been established, the 4-drug regimens currently recommended for most patients with tuberculosis (i.e., isoniazid-rifampin-pyrazinamide-ethambutol) generally are considered adequate, but the duration of therapy is extended to 9-12 months.

Patient Monitoring

Microscopic examination of sputum for acid-fast bacilli and mycobacterial cultures of appropriate specimens should be performed prior to initiation of antituberculosis therapy. During treatment of pulmonary tuberculosis, sputum specimens for microscopic examination and culture should be obtained at least once monthly until 2 consecutive culture specimens are negative. More frequent acid-fast sputum smears (e.g., every 2 weeks) may be useful to assess the early response to treatment and provide an indication of infectiousness in those who had positive smears at the time of diagnosis.

Sputum exam and culture are particularly important after 2 months of treatment has been completed (end of initial phase of treatment) since results of these tests at this time will determine the length of the continuation phase of treatment. Patients with cavitation on initial chest radiograph and positive cultures at completion of 2 months of therapy should receive a 7-month continuation phase of treatment for a total duration of 9 months of therapy. More than 85% of patients with positive sputum cultures who are treated initially with regimens containing isoniazid and rifampin should have negative sputum cultures after 2 months of treatment.

Sputum conversion is accomplished more rapidly with a 4-drug than with a 3-drug antituberculosis regimen, even with susceptible organisms. In vitro drug susceptibility testing should be performed on isolates from patients who have positive cultures after 3 months of treatment; those with positive cultures after 4 months of treatment should be considered treatment failures and be treated accordingly.Generally, routine follow-up evaluations are unnecessary in patients who have completed a standard period of treatment, but patients should be instructed to contact their clinician if signs or symptoms recur.

The ATS, CDC, and IDSA state that a repeat chest radiograph at completion of the initial 2 months of antituberculosis therapy may be useful in patients who had positive cultures at diagnosis, but are not essential. A chest radiograph at completion of therapy also is not considered essential, but can be used to provide a baseline against which subsequent examinations can be compared. The AAP recommends that chest radiographs be performed after 2-3 months of therapy in pediatric patients with pulmonary tuberculosis to evaluate the response to antituberculosis therapy.

Prior to initiating a multiple-drug antituberculosis regimen, the ATS, CDC, and IDSA recommend baseline measurements of serum creatinine, ALT (SGPT), AST (SGOT), bilirubin, alkaline phosphatase, and platelet count. Baseline examination of visual acuity and color vision should be obtained for those who will receive ethambutol. Routine monitoring of hepatic and renal function and platelet counts generally is not necessary during treatment unless the patient has abnormalities at baseline or is at increased risk for hepatotoxicity (e.g., history of hepatitis B or C virus infection or alcohol abuse).

However, patients receiving ethambutol should be questioned during monthly visits regarding possible visual disturbances including blurred vision or scotomata; monthly testing of visual acuity and color discrimination is recommended for those receiving higher than usually recommended dosages of ethambutol or those receiving the drug for longer than 2 months. Patient monitoring should take into account the observation that HIV-infected patients appear to experience a greater frequency of adverse effects to antituberculosis drugs than those not infected with HIV. The AAP states that although routine determination of serum transaminase concentrations in pediatric patients receiving antituberculosis therapy usually is not necessary, serum transaminase concentrations should be monitored approximately monthly during the first several months of therapy in those with severe tuberculosis (especially those with meningitis or disseminated disease).

Monitoring also may be indicated in patients with concurrent or recent liver or biliary disease, pregnant women, and whenever there is clinical evidence of hepatotoxicity or when other hepatotoxic drugs are used concomitantly.

Active Tuberculosis

A decision to initiate antituberculosis therapy in a patient with suspected active tuberculosis should be based on clinical, pathologic, and radiographic findings in the patient, including results of the initial series of microscopic examination of acid-fast bacilli and cultures for mycobacteria, as well as epidemiologic information. All patients suspected of having tuberculosis should have appropriate specimens collected for microscopic examination and mycobacterial cultures; in vitro susceptibility testing using isoniazid, rifampin, and ethambutol should be performed on all positive initial cultures, regardless of the specimen source. In patients with suspected pulmonary tuberculosis, a series of 3 sputum specimens should be obtained 8-24 hours apart. Although a tuberculin skin test can be performed at the time of initial evaluation, a negative tuberculin skin test reaction does not exclude a diagnosis of active tuberculosis. A recommended multiple-drug regimen should be initiated promptly (possibly even before acid-fact sputum smear results are available) if there is a high suspicion of tuberculosis or the patient is seriously ill with a pulmonary or extrapulmonary disease thought to be tuberculosis. A positive acid-fast sputum smear provides strong inferential evidence for a diagnosis of active tuberculosis and isolation of M. tuberculosis in culture or a positive nucleic acid amplification test confirms the diagnosis.

Active Tuberculosis in Adults

For the initial treatment of uncomplicated culture-positive pulmonary tuberculosis caused by drug-susceptible M. tuberculosis, the ATS, CDC, and IDSA currently recommend 4 possible multiple-drug regimens that include the first-line agents (isoniazid, rifampin, pyrazinamide, ethambutol, rifapentine). These regimens have a minimum duration of 6 months (26 weeks), and consist of an initial intensive phase (2 months) and a continuation phase (usually either 4 or 7 months). 

Because a large proportion of adults with tuberculosis have infections caused by M. tuberculosis strains resistant to isoniazid, 3 of the currently recommended regimens include 4 drugs in the initial phase of therapy to ensure that a 6-month regimen is maximally effective (i.e., regimens 1, 2, and 3). Therefore, most adults with previously untreated tuberculosis should receive a 2-month initial regimen that consists of isoniazid-rifampin-pyrazinamide-ethambutol; this initial intensive phase is then followed by one of several possible regimens for the continuation phase.

If in vitro susceptibility test results are available and M. tuberculosis is shown to be fully susceptible to isoniazid and rifampin, ethambutol does not need to be included during the initial phase and a 3-drug regimen that includes isoniazid-rifampin-pyrazinamide can be used. The ATS, CDC, and IDSA state that the only other circumstance when a 3-drug regimen would be acceptable for the initial phase of treatment is when pyrazinamide cannot be used because of contraindications or intolerance or when the strain is resistant to pyrazinamide; in these cases, the initial 3-drug regimen should consist of isoniazid-rifampin-ethambutol (i.e., regmen 4).

Initial Treatment Phase

In the initial intensive phase of antituberculosis therapy, the appropriate 4- or 3-drug regimen can be administered daily (7 or, alternatively, 5 days per week) for the entire 2 months (8 weeks) (i.e., regimens 1 and 4); the 4-drug regimen can be administered 3 times weekly for the entire 2 months (8 weeks) (i.e., regimen 3); or the 4-drug regimen can be administered daily (7 days per week) during the first 2 weeks then twice weekly for the next 6 weeks or, alternatively, 5 times per week for 2 weeks then twice for 6 weeks (i.e., regimen 2). The ATS, CDC, and IDSA state that clinical experience suggests that patients receiving treatment regimens 5 days per week using DOT have a success rate equivalent to those receiving the regimen 7 days per week. Therefore, “daily” may also be interpreted to mean DOT given 5 days per week and the required number of doses is adjusted accordingly.

All patients suspected of having tuberculosis should have appropriate specimens collected for microscopic examination and mycobacterial culture. In vitro susceptibility testing using isoniazid, rifampin, and ethambutol should be performed on all initial M. tuberculosis isolates. If the strain of M. tuberculosis isolated is found to be susceptible to both isoniazid and rifampin and ethambutol was included the initial 4-drug regimen, this drug can be discontinued in those receiving a daily regimen (i.e., regimen 1). If the patient is receiving an intermittent regimen (i.e., regimen 2 or 3), some experts suggest that ethambutol also can be safely discontinued from these regimens as soon as in vitro susceptibility test results are available, but there is no evidence to support this approach. If resistant organisms are found, the drug regimen should be changed accordingly; patients in whom drug-resistant M. tuberculosis are isolated should be managed in consultation with an expert in the treatment of tuberculosis.

Although streptomycin can be as effective as ethambutol when used in the initial phase of antituberculosis treatment and was previously included in recommendations for this phase of treatment, resistance to streptomycin has been reported with increasing frequency worldwide, which makes the drug less useful. Therefore, the ATS, CDC, and IDSA state that streptomycin is no longer recommended as being interchangeable with ethambutol unless the strain is known to be susceptible to the drug or the patient is from a population in which streptomycin resistance is unlikely.

Continuation Treatment Phase

The ATS, CDC, and IDSA currently recommend that after the initial 2-month intensive phase of treatment of tuberculosis, a continuation phase of treatment should be given for either 4 or 7 months. A 2-drug regimen of isoniazid and a rifamycin (rifampin or rifapentine) is recommended for this phase of treatment when tuberculosis is caused by drug-susceptible organisms. Most patients can receive a 4-month (18 weeks) continuation regimen of isoniazid-rifampin or isoniazid-rifapentine. However, a 7-month continuation regimen should be used in patients with cavitary pulmonary tuberculosis caused by drug-susceptible organisms who still have positive sputum culture at completion of the 2-month initial treatment phase; in patients whose initial phase of treatment did not include pyrazinamide (i.e., regimen 4); and in patients who had positive sputum cultures at the end of the initial treatment phase and are receiving a once-weekly regimen of isoniazid-rifapentine for the continuation phase.

In the continuation phase of treatment, isoniazid-rifampin may be given daily (7 or, alternatively, 5 days per week); twice weekly using DOT; or 3 times weekly using DOT. Alternatively, a regimen of isoniazid-rifapentine can be given once weekly for 4 months using DOT; however, this once-weekly regimen should only be used in HIV-negative patients with noncavitary pulmonary tuberculosis (as determined by chest radiography) who had negative sputum smears at completion of the initial 2-month regimen.

Completion of Treatment and Total Number of Doses

The ATS, CDC, and IDSA state that completion of treatment of active tuberculosis is determined more accurately by the total number of doses and should not be based solely on the duration of therapy. Therefore, 6 months is the minimum duration of treatment in patients with active tuberculosis and would accurately indicate the length of treatment only if there are no interruptions in drug administration. The goal is to deliver the specified number of doses within a recommended maximum time. For the 6-month daily multiple-drug regimen recommended for the treatment of pulmonary tuberculosis (regimen 1), a total of 182 doses should be administered within 9 months after beginning treatment if the drugs are given 7 days per week (56 doses during the initial phase and 126 doses during the continuation phase) or, alternatively, a total of 130 doses should be administered within 9 months if the drugs are administered 5 days per week (40 doses during the initial phase and 90 doses during the continuation phase).

When treatment interruptions occur (e.g., because of nonadherence or drug toxicity), the specified number of doses cannot be administered within the target period. In such cases, the patient should be assessed to determine the most appropriate action, including continuing treatment for a longer duration or restarting treatment from the beginning. Continuous (uninterrupted) treatment is more important during the initial phase when the mycobacterial population and risk of developing resistance are greatest than during the continuation phase of treatment when the mycobacterial population is lower and the goal of treatment is to kill persisting organisms. Some experts suggest that treatment should be restarted from the beginning if an interruption occurs during the initial phase of treatment and the lapse in dosing is 14 days or longer, but usually can be continued if the lapse in dosing during the initial phase was shorter than 14 days. In either case, the total number of doses targeted for the initial phase should be given. In addition, these experts recommend that treatment be restarted from the beginning if an interruption occurs during the continuation phase of treatment and the patient has received less than 80% of the planned total number of doses and the lapse is 3 months or longer in duration. The ATS, CDC, and IDSA recommend consultation with an expert to assist in managing treatment interruptions.

Active Tuberculosis in Pediatric Patients

Infants and Children

Because there is a high risk of disseminated tuberculosis in infants and children younger than 4 years of age, treatment should be initiated as soon as a diagnosis of tuberculosis is suspected. In general, treatment regimens recommended for the treatment of active tuberculosis in adults also are recommended for treatment of the disease in infants, children, and adolescents (with appropriately adjusted drug dosages); however, ethambutol is not used routinely in children.

Therefore, the ATS, CDC, IDSA, and American Academy of Pediatrics (AAP) recommend that uncomplicated pulmonary and most cases of extrapulmonary tuberculosis in children be treated with an initial 3-drug regimen of isoniazid-rifampin-pyrazinamide for the first 2 months followed by a continuation phase regimen of isoniazid-rifampin, provided adult-type tuberculosis (upper lobe infiltration, cavitation, sputum production) is not present and drug-resistant organisms are not involved.

Most studies in children have used this 3-drug initial treatment regimen with a continuation phase of 4 months for a total duration of 6 months of treatment; this regimen has been effective in more than 95% of children and has been associated with a low incidence of adverse reactions. However, an initial 4-drug regimen is recommended when drug-resistant organisms are known or suspected to be involved or when the infection is life-threatening.

Although a 6-month regimen of only isoniazid and rifampin may be effective in children with hilar adenopathy and pulmonary disease in whom drug resistance is not a consideration, the ATS, CDC, and IDSA state that a 6-month regimen that includes a 3-drug regimen of isoniazid-rifampin-pyrazinamide during the initial phase has been used more frequently and is recommended in most patients. Even with an effective 6-month treatment regimen, hilar adenopathy may persist for 2 or 3 years and the AAP states that normal radiographic findings are not necessary in these patients prior to discontinuance of treatment.

The ATS, CDC, and IDSA state that antituberculosis therapy in pediatric patients should always be administered using DOT and children should be closely monitored for adverse effects. These experts also state that parents should not be relied on to supervise DOT. The fact that lack of pediatric dosage forms of many antituberculosis agents necessitates using crushed tablets and extemporaneous suspensions should be considered. Intermittent dosing regimens can be used in children, usually after an initial daily regimen given for 2 weeks to 2 months. However, although the AAP states that an intermittent regimen given 2- or 3-times weekly can be administered using DOT after the initial daily phase, the ATS, CDC, and IDSA state that a 3-times weekly regimen is not recommended for pediatric patients.

The AAP states that a 4-drug regimen should be used initially in the treatment of life-threatening tuberculosis because of the possibility of drug resistance and the severe consequences of treatment failure. Drug-susceptible tuberculous meningitis in pediatric patients should be treated using an initial 4-drug daily regimen of isoniazid-rifampin-pyrazinamide and either ethambutol, streptomycin (or another aminoglycoside), or ethionamide given for the first 1-2 months followed by 7-10 months of isoniazid-rifampin given once daily or twice weekly using DOT for a total duration of 9-12 months.

Because there is a lower bacillary burden in childhood-type tuberculosis, there is a lower risk of acquired drug resistance, treatment failure, and relapse in pediatric patients compared with adult patients.

However, adult-type tuberculosis (upper lobe infiltration, cavitation, sputum production) can occur in children and, more frequently, in adolescents; and the ATS, CDC, and IDSA recommend use of an initial 4-drug regimen in these cases until results of in vitro susceptibility data are available. Although ethambutol should be used with caution in children in whom it may be difficult to monitor visual acuity (e.g., those younger than 5 years of age), ethambutol-associated optic neuritis is exceedingly rare in children with normal renal function and the ATS, CDC, and IDSA state that use of ethambutol as the fourth drug in the treatment regimen can be considered even in young children when the child has adult-type (upper lobe infiltration, cavity formation) tuberculosis or when M. tuberculosis is known or suspected to be resistant to isoniazid.

Alternatively, when necessary, an aminoglycoside (streptomycin, kanamycin, amikacin) can be used for the fourth agent in the multiple-drug regimen. For patients who acquired tuberculosis in areas where resistance to streptomycin is common, the AAP recommends use of capreomycin, kanamycin, or amikacin rather than streptomycin as the fourth drug.

When selecting a treatment regimen for a child with pulmonary tuberculosis, it may be necessary to rely on results of in vitro susceptibility tests of the organisms isolated from the presumed source case since it often is difficult to isolate M. tuberculosis from children. In cases of suspected drug-resistant tuberculosis in a child or when an isolate is not available from the source case, consideration should be given to obtaining specimens for microbiologic evaluation via 3 early morning gastric aspiration (optimally performed during hospitalization), bronchoalveolar lavage, or biopsy.

While hepatotoxic reactions to isoniazid alone are extremely rare in children, the frequency of hepatotoxic reactions to rifampin alone or in combination with isoniazid appears to be higher in children than in adults and may be influenced by several factors, including severity of tuberculosis, nutritional status, and drug dosage; some evidence suggests that underlying viral hepatitis has been responsible in part for the reported higher incidence of hepatotoxicity in children in developing countries receiving these drugs.

The AAP states that use of an isoniazid dosage exceeding 10 mg/kg daily in conjunction with rifampin may increase the incidence of hepatotoxicity and some clinicians suggest that the dosage of isoniazid and rifampin in children should be limited to 10 and 15 mg/kg, respectively. The AAP also recommend concomitant use of pyridoxine therapy in children and adolescents receiving isoniazid who have an abnormally low milk and meat intake, those with nutritional deficiencies (including all symptomatic HIV-infected children), pregnant adolescents and women, and breast-feeding infants.

Neonates

Congenital tuberculosis is rare since pregnant women with pulmonary tuberculosis are unlikely to transmit the infection to the fetus until after delivery, although in utero infections can occur if the pregnant woman has M. tuberculosis bacillemia. If a neonate is suspected of having congenital tuberculosis, the AAP states that a Mantoux tuberculin skin test, chest radiograph, lumbar puncture, and appropriate cultures should be performed promptly and, regardless of the skin test results, treatment of the infant should be promptly initiated using a 4-drug initial regimen of isoniazid-rifampin-pyrazinamide-streptomycin (or kanamycin). The placenta should be examined histologically and cultured for M. tuberculosis. The mother should be evaluated for the presence of pulmonary or extrapulmonary (including uterine) tuberculosis. If her physical examination or chest radiograph supports the diagnosis of active tuberculosis, the AAP recommends that the neonate be treated with the regimen recommended for active tuberculosis. The organism recovered from the mother and/or infant should be tested for in vitro drug susceptibility.

Active Tuberculosis During Pregnancy

Untreated tuberculosis represents a far greater hazard to a pregnant woman and her fetus than does treatment of the disease, and it is essential that effective therapy be administered promptly to a pregnant woman whenever active tuberculosis is suspected. Although safe use of antituberculosis agents during pregnancy has not been definitely established, the ATS, CDC, IDSA, and AAP recommend that active tuberculosis during pregnancy be treated with an initial 3-drug regimen of isoniazid-rifampin-ethambutol. These experts consider isoniazid, rifampin, and ethambutol safe for use during pregnancy. However, they recommend concomitant use of pyridoxine (25 mg daily) in all pregnant and breast-feeding women receiving isoniazid; the amount of pyridoxine in commercially available multivitamins is variable but generally lower than what is needed for patients receiving isoniazid.

Although pyrazinamide has been used in pregnant women with tuberculosis, the risk of teratogenicity has not been fully determined to date. The ATS, CDC, IDSA, and AAP state that the benefits of pyrazinamide may outweigh the possible (but unquantified) risk in some pregnant women with tuberculosis, especially when resistance to other drugs but susceptibility to pyrazinamide is likely. If pyrazinamide is included in the initial treatment regimen in pregnant women, a total duration of 6 months is recommended; if pyrazinamide is not used, a minimum duration of 9 months is recommended.

Because use of aminoglycosides in pregnant women can cause congenital ototoxicity and/or fetal nephrotoxicity, streptomycin, amikacin, and kanamycin are contraindicated during pregnancy. Capreomycin also should be avoided during pregnancy because of the risk of fetal nephrotoxicity and congenital ototoxicity. Because cycloserine crosses the placenta and data are limited regarding safety of the drug in pregnant women, the ATS, CDC, and IDSA state that the drug should be used for the treatment of tuberculosis during pregnancy only when there are no suitable alternatives. Ethionamide also crosses the placenta and has been shown to be teratogenic in animals; therefore, the drug should not be used during pregnancy. Although aminosalicylic acid has been used safely in some pregnant women, the ATS, CDC, and IDSA state that the drug should be used in pregnant women only when there are no other alternatives for the treatment of multidrug-resistant tuberculosis. The ATS, CDC, and IDSA state that there are insufficient data to date to recommend use of rifabutin or rifapentine in pregnant women; the drugs should be used during pregnancy only when the potential benefits justify the possible risks to the fetus.

In a prospective study in pregnant women with extrapulmonary tuberculosis and tuberculous lymphadenitis (the most common form of extrapulmonary tuberculosis) there was no evidence of adverse effects on maternal and fetal outcome; however, extrapulmonary tuberculosis at other sites was associated with an increased frequency of maternal disability, hospitalization during pregnancy, fetal-growth retardation, and infants with low Apgar scores soon after birth.

Congenital tuberculosis is rare since pregnant women with pulmonary tuberculosis are unlikely to transmit the infection to the fetus until after delivery, although in utero infections can occur if the pregnant woman has M. tuberculosis bacillemia.

Drug-Resistant Tuberculosis

The ATS, CDC, and IDSA state that treatment of tuberculosis caused by drug-resistant M. tuberculosis requires the expertise of or consultation with an expert in the management of these infections since inappropriate management can have life-threatening consequences.

If drug resistance is acquired during treatment, it usually occurs because there is a large bacillary population (e.g., in pulmonary cavities), an inadequate antituberculosis regimen is prescribed (e.g., inappropriate drugs, inappropriate dosages), or there is a combined failure of both the patient and provider to ensure that an adequate regimen is taken. Drug resistance in a patient with newly diagnosed tuberculosis may be suspected on the basis of historical (previous treatment) or epidemiologic information (contact with a known drug-resistant case or a region where drug resistance is common).

Patients with tuberculosis caused by M. tuberculosis resistant to isoniazid and rifampin (multiple-drug resistant strains) are at high risk for treatment failure and acquiring further drug resistance. The ATS, CDC, and IDSA recommend that such patients be referred to or consultation be obtained from a specialized treatment center as identified by local or state health departments or the CDC. Although individuals with M. tuberculosis resistant to rifampin alone have a better prognosis than those with multiple-drug resistant strains, these individuals also are at increased risk for treatment failure and additional resistance and should be managed in consultation with an expert.

The role of resectional surgery in the management of patients with extensive pulmonary tuberculosis caused by multiple-drug resistant strains has not been established in randomized studies and results have been conflicting. Surgery should be performed by surgeons with experience in these situations and only after the patient has received several months of intensive treatment. The ATS, CDC, and IDSA state that expert opinion suggest that, even with successful resection, antituberculosis therapy should be continued for 1-2 years postoperatively to prevent relapse.

Results of randomized or controlled studies are not available to guide selection of the most appropriate regimens for the treatment of tuberculosis caused by the various patterns of drug-resistant M. tuberculosis. Therefore, treatment regimens for these patients are based on general principles of antituberculosis therapy and expert opinion. Table 2 contains information on ATS, CDC, and IDSA recommended regimens for use in patients with pulmonary tuberculosis who have various patterns of drug-resistant M. tuberculosis.

The AAP states that when drug-resistant tuberculosis is suspected, a 4-drug regimen should be used in pediatric patients during the initial phase of treatment until results of in vitro susceptibility testing are available. The 4-drug regimen should include at least 2 bactericidal drugs such as isoniazid and rifampin, pyrazinamide, and an aminoglycoside (also bactericidal) or ethambutol. The ATS, CDC, and IDSA suggest that use of a fourth drug in the initial regimen (usually ethambutol) should be considered in children with active tuberculosis if the child traveled in an area with a high prevalence of drug-resistant tuberculosis or was exposed to an individual with known drug-resistant tuberculosis; was exposed to an individual with active tuberculosis who has had prior treatment (treatment failure or relapse) and whose in vitro susceptibility test results are unknown; was exposed to an individual with active tuberculosis from areas with a high prevalence of drug-resistant tuberculosis; was exposed to an individual who continues to have positive sputum smears after 2-3 months of multiple-drug therapy. The AAP states that 6-month drug regimens are not recommended in pediatric patients with isoniazid- or rifampin-resistant tuberculosis and that a 12- to 18-month regimen usually is necessary to effect a cure. In addition, the AAP states that twice-weekly intermittent regimens are not recommended for children with drug-resistant tuberculosis and that use of DOT is critical to prevent emergence of further resistance. For patients who acquired tuberculosis in geographic areas where resistance to streptomycin is common, the AAP recommends use of capreomycin, kanamycin, or amikacin rather than streptomycin.

Relapse or Treatment Failure

Relapse may occur after an apparently successful antituberculosis regimen (patient becomes and remains culture-negative while receiving antituberculosis treatment, but at some point after completion of therapy becomes culture-positive again or experiences clinical or radiographic deterioration consistent with active tuberculosis). In addition, treatment failure (continued or recurrently positive sputum cultures in a patient receiving an appropriate antituberculosis regimen) may occur. Relapse and treatment failure may be the result of poor patient compliance with the drug regimen and/or the emergence of resistance to one or more of the antituberculosis agents administered; both require retreatment with antituberculosis agents.

Some data based on the use of DNA fingerprinting to characterize the genotype of M. tuberculosis suggest that exogenous reinfection, rather than reactivation, is a major cause of postprimary tuberculosis in an area with a high incidence of M. tuberculosis infection. In a study conducted in an area of South Africa with endemic tuberculosis, 12 of 16 patients with a relapse of pulmonary tuberculosis after curative treatment of the disease had isolates of M. tuberculosis identified by restriction-fragment-length polymorphism (RFLP) analysis that were different from those of the initial disease. While the study included only 16 of 698 patients with a culture available for analysis during the study period and thus must be interpreted cautiously, the low rate of contaminated cultures (3.%) suggests that laboratory error was not responsible for the results. In addition, 11 of the 12 isolates were identified as being among those strains of M. tuberculosis already present in the community. If exogenous reinfection is common, the use of tuberculosis prophylaxis in individuals who recently have been exposed to infectious tuberculosis would become even more important regardless of whether such individuals had prior M. tuberculosis infection.

Relapse

The ATS, CDC, and IDSA state that when relapse of tuberculosis is suspected because there is clinical or radiologic deterioration, rigorous efforts should be made to establish a diagnosis and to obtain microbiologic confirmation of the relapse and evaluate the possibility of drug resistance. Relapses usually are caused by failure of the antituberculosis regimen to sterilize tissues, thereby enabling endogenous recrudescence of the original infection; however, in hyperendemic settings, exogenous reinfection with a new strain of M. tuberculosis may be responsible for the relapse. Relapses generally occur within the first 6-12 months after completion of antituberculosis therapy. In most cases when relapse occurs in patients who originally had M. tuberculosis susceptible to first-line agents and who were treated using DOT and an antituberculosis regimen containing a rifamycin (rifampin, rifabutin, rifapentine), relapse will be caused by strains susceptible to the first-line agents. However, the risk of drug-resistant strains causing relapse is substantially greater in patients who received a self-administered regimen or a multiple-drug regimen that did not contain a rifamycin. Drug-resistant strains also are likely if initial drug susceptibility testing was not performed and the patient fails or relapses after a rifamycin-containing regimen administered using DOT; in these cases, the organisms probably were resistant from the outset.

An empiric regimen for retreatment of patients with relapse of tuberculosis should be based on the prior treatment regimen and disease severity. If tuberculosis was originally caused by drug-susceptible organisms and the individual originally received an appropriate regimen using DOT, the standard 4-drug treatment regimen can be initiated pending results of in vitro susceptibility testing; however, those who have life-threatening forms of the disease should also receive at least 3 additional agents to which the organisms are likely to be susceptible.

Individuals with relapse who did not originally receive an antituberculosis regimen administered using DOT, did not originally receive a regimen containing a rifamycin, or who are known or presumed to have had an irregular treatment should be presumed to have drug-resistant strains of M. tuberculosis. In these cases, a regimen of isoniazid-rifampin-pyrazinamide and 2 or 3 additional agents (based on results of in vitro susceptibility testing) should be used. These additional agents could be ethambutol, a fluoroquinolone (levofloxacin, moxifloxacin, gatifloxacin), a parenteral drug (e.g., streptomycin [if not used in original regimen and susceptibility established with in vitro testing], amikacin, kanamycin, capreomycin), with or without an additional oral drug. This expanded regimen is particularly important in patients with immunodeficiency, limited respiratory reserve, CNS involvement, or other life-threatening circumstances in whom an inadequate regimen could have severe consequences.

If exogenous reinfection is the principal cause of postprimary tuberculosis in areas with a high incidence of the disease, relapse cannot be assumed to result from failure of a drug regimen unless RFLP analysis of bacterial isolates is performed. In populations with a low risk of infection, the likelihood of reexposure and reinfection is small and repeat cases of tuberculosis therefore probably result from reactivation. In those rare cases when exogenous reinfection is strongly suspected as the cause of relapse, the retreatment regimen should be chosen based on drug susceptibility of the presumed source case. If the likely source is known to have drug-resistant M. tuberculosis, the empiric regimen should be expanded based on the resistance profile of that case.

Treatment Failure

About 90-95% of patients with pulmonary tuberculosis caused by drug-susceptible M. tuberculosis (even those with extensive lung cavitation) will have negative sputum cultures and clinical improvement after 3 months of an appropriate multiple-drug antituberculosis regimen containing isoniazid and rifampin. Therefore, patients with persistently positive cultures (with or without ongoing symptoms) after 3 months of treatment should be evaluated carefully to identify the cause of delayed conversion and those with positive cultures after 4 months of treatment should be considered treatment failures. The most frequent cause of treatment failure in individuals receiving appropriate antituberculosis regimens without DOT is nonadherence to the drug regimen. Treatment failure in those receiving an appropriate regimen with or without DOT may be caused by unrecognized drug resistance, malabsorption of the drugs (e.g., because of prior resectional surgery of the stomach or small intestine, taking antituberculosis agents with antacids or other drugs that interfere with absorption), laboratory error, and extreme biologic variation in response.

The ATS, CDC, and IDSA state that early consultation with a specialty center is strongly advised in cases of treatment failure. M. tuberculosis isolates from patients with treatment failure should be promptly sent to a reference laboratory for in vitro drug susceptibility testing using both first- and second-line agents. If treatment failure is likely to be caused by drug resistance and the patient is not seriously ill, an empiric retreatment regimen can be started pending results of in vitro susceptibility testing or, alternatively, initiation of a retreatment regimen can be deferred until results are available. If the patient is seriously ill or sputum smears are positive, an empiric regimen should be started immediately and then adjusted accordingly when in vitro susceptibility test results are available.

In patients with treatment failure, a single new drug should never be added to a failing regimen since this may lead to acquired resistance to the new drug. Instead, at least 2, and preferably, 3 new drugs to which susceptibility is expected should be added to the prior regimen. The ATS, CDC, and IDSA state that empiric retreatment regimens might include a fluoroquinolone (levofloxacin, moxifloxacin, gatifloxacin), a parenteral drug (e.g., streptomycin [if not used in original regimen and the patient is not from an area having high rates of streptomycin resistance], amikacin, kanamycin, capreomycin), and an additional oral agent (e.g., aminosalicylic acid, cycloserine, ethionamide). The regimen should be adjusted accordingly after results of in vitro susceptibility testing are available.

Extrapulmonary Tuberculosis

In general, the general principles concerning use of antituberculosis agents in the treatment of pulmonary tuberculosis also apply to extrapulmonary tuberculosis.

Adults

Short-course antituberculosis regimens (at least 6 months) have been effective in adults for the treatment of extrapulmonary tuberculosis (e.g., pleural, genitourinary, lymphatic, pericardial, bone and joint, vertebral, meningeal) or pulmonary tuberculosis complicated by pneumoconiosis. The ATS, CDC, IDSA, and other clinicians state that 6- to 9-month multiple-drug regimens that include isoniazid and rifampin generally are effective for extrapulmonary tuberculosis. Therefore, a standard 6-month multiple-drug regimen that involves use of isoniazid-rifampin-pyrazinamide-ethambutol during the initial 2 months and isoniazid-rifampin for an additional 4 months (see Table 1) is recommended for most adults with extrapulmonary tuberculosis (except those with meningitis) unless the organisms are known or strongly suspected of being resistant to first-line agents. This includes adults with lymph node tuberculosis, bone and joint tuberculosis (some experts prefer a 9-month regimen), pericardial tuberculosis, pleural tuberculosis, disseminated or miliary tuberculosis, genitourinary tuberculosis, and abdominal tuberculosis. However, treatment should be extended whenever there is a slow response to the antituberculosis regimen. If pyrazinamide cannot be used in the initial treatment phase, then the continuation phase should be given for 7 months for a total duration of treatment of 9 months. It also has been recommended that the once-weekly isoniazid-rifapentine regimen not be used in the continuation phase of treatment in patients with extrapulmonary tuberculosis.

Pediatric Patients

Children with extrapulmonary tuberculosis (except those with disseminated disease or meningitis) generally can be treated with the same regimens currently recommended for pulmonary tuberculosis, unless drug-resistant M. tuberculosis is involved. The ATS, CDC, IDSA, and AAP recommend a 6-month treatment regimen for most children with extrapulmonary tuberculosis; however, a duration of 9- to 12-months is recommended for those with disseminated disease or meningitis.

Tuberculous Meningitis

Although the optimal duration of therapy for the treatment of meningitis caused by M. tuberculosis has not been established, the ATS, CDC, IDSA, and AAP generally recommend that a 9- to 12-month regimen be used. Tuberculous meningitis is associated with high morbidity and mortality, despite prompt initiation of an appropriate antituberculosis regimen.

The ATS, CDC, and IDSA recommend that treatment of tuberculous meningitis be initiated with a 4-drug regimen of isoniazid-rifampin-pyrazinamide-ethambutol given for 2 months followed by a regimen of isoniazid-rifampin given for an additional 7-10 months; use of parenteral antituberculosis agents (isoniazid, rifampin, aminoglycosides, capreomycin, fluoroquinolones) may be necessary in those with altered mental status who may not be able to take oral drugs. Repeat lumbar punctures should be considered to monitor changes in CSF cell count, glucose, and protein, especially early in the course of treatment.

The AAP recommends that drug-susceptible tuberculous meningitis in pediatric patients be treated using an initial 4-drug daily regimen of isoniazid-rifampin-pyrazinamide and either ethambutol, streptomycin (or another aminoglycoside), or ethionamide given for the first 1-2 months followed by isoniazid-rifampin once daily or twice weekly using DOT for a total duration of 9-10 months.

Currently available antituberculosis agents that distribute into the CSF in clinically important concentrations include isoniazid, rifampin, rifabutin, pyrazinamide, cycloserine, and ethionamide. While ethambutol penetrates the meninges in the presence of inflammation, efficacy in the treatment of tuberculous meningitis has not been demonstrated. Only low concentrations of the aminoglycosides (streptomycin, amikacin, kanamycin) or aminosalicylic acid diffuse into CSF; capreomycin is not distributed into CSF.

Adjunctive Use of Corticosteroids

While data from randomized, controlled trials are limited and principally consist of studies conducted before the use of 4-drug antituberculosis regimens, available evidence suggests that adjunctive corticosteroid therapy may enhance short-term resolution of disease manifestations (e.g., clinical and radiographic abnormalities) in patients with severe pulmonary or extrapulmonary tuberculosis and, in some cases, reduces mortality associated with certain forms of extrapulmonary (e.g., meningitis, pericarditis) disease. (See Respiratory Diseases: Advanced Pulmonary and Extrapulmonary Tuberculosis, in Uses in the Corticosteroids General Statement 68:04.) The ATS, CDC, and IDSA suggest that adjunctive use of corticosteroid therapy be considered in patients with tuberculous meningitis and certain other forms of extrapulmonary disease (e.g., tuberculous pericarditis).

Although evidence supporting the adjunctive use of corticosteroids in pediatric patients with tuberculosis is incomplete, the AAP recommends use of corticosteroid therapy as an adjunct to antituberculosis therapy in children with tuberculous meningitis (to decrease mortality rate and long-term neurologic impairment). The AAP also states that adjunctive use of corticosteroids can be considered in children with pleural and pericardial effusions (to hasten fluid reabsorption), severe miliary disease (to mitigate alveolocapillary block), and endobronchial disease (to relieve obstruction and atelectasis).

Active Tuberculosis in HIV-Infected Individuals

Patients with human immunodeficiency virus (HIV) infection who previously were infected or recently exposed to tuberculosis are much more likely to develop the disease than are non-HIV-infected individuals. In addition, the observed mortality rate for HIV-infected patients with tuberculosis is approximately 4 times as great as that for patients with tuberculosis who do not have HIV infection. Increasing immunologic and virologic evidence indicates that the host immune response to M. tuberculosis enhances HIV replication and might accelerate the natural progression of HIV infection. Treatment of tuberculosis leads to reductions in the viral load in patients who have coinfection with HIV and tuberculosis. Prompt initiation of effective antituberculosis therapy increases the probability that a patient with HIV infection who develops tuberculosis will be cured of the disease. Effective antituberculosis therapy quickly renders the patient noninfectious, resulting in a reduction in transmission of M. tuberculosis to others, and also minimizes the patient’s risk of death from tuberculosis. Therefore, clinicians must immediately and thoroughly investigate the possibility of tuberculosis when a patient with HIV has symptoms consistent with the disease. Early clinical response to therapy and time required for conversion of sputum cultures from positive to negative appear to be similar in patients with or without HIV infection. However, whether HIV infection coinfection influences the rate of relapse in patients with tuberculosis has not been determined.

The CDC states that antituberculosis therapy should be initiated immediately in HIV-infected patients suspected of having tuberculosis disease and that such patients should be placed in isolation if necessary. The fact that concomitant use of some antituberculosis agents (e.g., rifabutin, rifampin) and certain antiretroviral agents (e.g., HIV protease inhibitors, nonnucleoside reverse transcriptase inhibitors [NNRTIs]) can affect plasma concentrations of the antituberculosis agent and/or the antiretroviral agents must be considered when antituberculosis therapy is indicated for the treatment of active tuberculosis or latent tuberculosis infection in HIV-infected patients. (See Patients Receiving Concurrent Antiretroviral Therapy under: Active Tuberculosis: Active Tuberculosis in HIV-infected Patients.) Some of these pharmacokinetic interactions contraindicate concomitant use of the drugs; however, with some of the drugs, concomitant administration is a possibility with appropriate dosage adjustments. Because the management of these patients is complex and must be individualized, experts in the management of mycobacterial infections in HIV-infected patients should be consulted. In addition, the CDC states that all patients with HIV-related tuberculosis should be treated using directly observed therapy (DOT).

Patients Receiving Concurrent Antiretroviral Therapy

Pharmacokinetic interactions between certain antimycobacterial agents (e.g., rifabutin, rifampin) and HIV protease inhibitors (e.g., amprenavir, indinavir, lopinavir, nelfinavir, ritonavir, saquinavir) and NNRTIs (e.g., delavirdine, efavirenz, nevirapine) have been reported and may complicate drug therapy for mycobacterial infections in HIV-infected patients receiving concurrent antiretroviral therapy. Limited data suggest that rifamycin derivatives (e.g., rifampin, rifabutin) accelerate the metabolism of HIV protease inhibitors and some NNRTIs (e.g., delavirdine) (by induction of hepatic P-450 cytochrome oxidases), which may result in subtherapeutic plasma concentrations of some of these antiretroviral agents. In addition, HIV protease inhibitors and some NNRTIs (e.g., delavirdine) reduce the metabolism of rifamycins, leading to increased plasma concentrations of rifamycins and an increased risk of toxicity and some other NNRTIs (e.g., efavirenz) can decrease plasma concentrations of rifabutin.If an antiretroviral regimen contains 2 HIV protease inhibitors, the complexity of drug interactions is amplified and recommendations regarding dosage modifications are difficult when rifamycins also are administered. If an antiretroviral regimen contains both an inhibitor and an inducer of cytochrome P-450 (CYP) enzymes (e.g., an HIV protease inhibitor and a NNRTI), a different complex interaction occurs and the appropriate dosage modifications necessary to ensure optimum levels of the antiretroviral agents and the rifamycins are unknown. Because nucleoside reverse transcriptase inhibitor antiretroviral agents (e.g., abacavir, didanosine, lamivudine, stavudine, zalcitabine, zidovudine) are not metabolized by CYP isoenzymes, concomitant use of rifamycins and these antiretrovirals is not expected to result in pharmacokinetic interactions and dosage modifications are not required. In addition, concomitant use of isoniazid, ethambutol, pyrazinamide, or streptomycin with HIV protease inhibitors, NNRTIs, or nucleoside reverse transcriptase inhibitors does not require dosage modifications.

Because of the pharmacokinetic interactions between rifamycins and HIV protease inhibitors or NNRTIs and because rifabutin is a less potent inducer of CYP isoenzymes, the CDC and other experts previously stated that use of rifampin was contraindicated in patients receiving HIV protease inhibitors or NNRTIs and use of rifabutin-containing regimens was the preferred alternative for the treatment of active tuberculosis in HIV-infected patients receiving these antiretroviral agents. However, the CDC and some experts now suggest that there are specific circumstances when HIV-infected patients with active tuberculosis can receive rifampin concomitantly with certain HIV protease inhibitors or certain NNRTIs. Rifampin can be used in patients receiving the following antiretroviral regimens (with appropriate dosage adjustments): efavirenz and 2 nucleoside reverse transcriptase inhibitors; ritonavir and 1 or 2 nucleoside reverse transcriptase inhibitors; or an antiretroviral regimen that includes both ritonavir and saquinavir. Concomitant use of rifampin with other HIV protease inhibitors (amprenavir, indinavir, lopinavir, nelfinavir) is contraindicated. In addition, concomitant use of rifampin with delavirdine is contraindicated. Although there is no published clinical experience, rifampin can be administered concomitantly with efavirenz; some experts recommend using an increased dosage of efavirenz in patients receiving concomitant rifampin. Data are insufficient to assess whether dosage adjustments are necessary when rifampin is administered concomitantly with nevirapine, and this combination should be used only if clearly indicated and with careful monitoring. For specific information on the pharmacokinetic interactions between antiretroviral agents and rifampin and recommendations regarding these interactions.

If rifabutin-containing regimens are used in HIV-infected patients receiving amprenavir, indinavir, lopinavir, nelfinavir, ritonavir, or saquinavir, a reduced rifabutin dosage may be necessary. In addition, an increase in indinavir or nelfinavir dosage also may be necessary. Although there is no published clinical experience, some experts recommend that if rifabutin-containing regimens are used in HIV-infected patients receiving efavirenz, an increase in rifabutin dosage may be necessary. Nevirapine can be used concomitantly with usual dosage of rifabutin, but concomitant use of delavirdine and rifabutin is contraindicated.

Use of antituberculosis regimens that do not contain rifamycins can be considered as an alternative for patients receiving complex antiretroviral combination regimens that contain HIV protease inhibitors or NNRTIs. However, for HIV-infected patients with active tuberculosis, use of a treatment regimen that does not contain a rifamycin, although possible, may be suboptimal and usually is not recommended. The safety and efficacy of rifapentine, a new long-acting rifamycin, have not been established in patients with HIV infection, and the CDC states that use of rifapentine in antituberculosis regimens in such patients currently is not recommended. Because current CDC recommendations strongly advise against interruption of antiretroviral therapy, the previously recommended practice of stopping protease inhibitor therapy to allow the use of rifampin in an antituberculosis regimen is no longer recommended for patients with HIV-related tuberculosis.

When determining the time to initiate antiretroviral therapy in HIV-infected patients who are acutely ill with tuberculosis, clinicians and patients should consider existing clinical issues such as drug interactions and toxicities, the patient’s ability to adhere to 2 complex treatment regimens, and laboratory abnormalities. A staggered initiation of antituberculosis and antiretroviral therapies, which might consist of starting antiretroviral therapy either at the end of the 2-month induction phase of antituberculosis therapy or after completion of such therapy, might promote greater adherence to both treatment regimens and reduce associated drug toxicities. When initiation of antiretroviral therapy is delayed, the patient’s condition should be monitored clinically and with the use of plasma HIV RNA levels (viral load) and CD4+ T-cell counts at least every 3 months to aid in determining the appropriate time to initiate antiretroviral therapy. Because of the potent effect of rifampin as a hepatic cytochrome P-450 enzyme inducer, which reduces the serum concentrations of protease inhibitors and NNRTIs, clinicians should plan to allow a 2-week period between the last dose of rifampin and the first dose of protease inhibitors or NNRTIs.

For patients who are receiving therapy with protease inhibitors or NNRTIs, the initial phase of a 6-month antituberculosis regimen consists of isoniazid, rifabutin, pyrazinamide, and ethambutol administered daily for 8 weeks, or daily for at least 2 weeks followed by twice-weekly dosing for 6 weeks, to complete the 2-month induction phase. The second phase of treatment consists of isoniazid and rifabutin administered daily or twice weekly for a minimum of 4 months. However, there is evidence that use of antituberculosis regimens that include once- or twice-weekly administration of rifamycins (e.g., rifabutin, rifampin, rifapentine) in HIV-infected patients with CD4+ T-cell counts less than 100/mm3 is associated with an increased risk of acquired rifamycin resistance. It is not known whether the risk for acquired rifamycin resistance is greater with rifabutin than with rifampin. Therefore, until additional data are available regarding this issue, the CDC recommends that HIV-infected individuals with CD4+ T-cell counts less than 100/mm3 not receive rifamycin regimens for the treatment of active tuberculosis that involve once- or twice-weekly administration. These individuals should receive daily therapy during the induction phase, and daily or 3-times weekly rifamycin regimens during the second phase; directly observed therapy also is recommended for both the daily and 3-times weekly regimens. Although no further action is recommended at this time for patients with advanced HIV disease who have completed treatment of active tuberculosis with intermittent regimens and are clinically stable, suspected relapse in these individuals should be treated with regimens active against rifamycin-resistant M. tuberculosis until results of susceptibility testing are available.

For patients in whom the use of rifamycins is limited or contraindicated for any reason (e.g., intolerance, patient/clinician decision not to use antiretroviral therapy concomitantly with rifabutin), the initial phase of a 9-month antituberculosis regimen consists of isoniazid, streptomycin, pyrazinamide, and ethambutol administered daily for 8 weeks, or daily for at least 2 weeks followed by twice-weekly dosing for 6 weeks, to complete a 2-month induction phase. The second phase of treatment consists of isoniazid, streptomycin, and pyrazinamide administered 2-3 times a week for 7 months. Every effort should be made to continue streptomycin therapy for the total duration of treatment or at least for 4 months after culture conversion (approximately 6-7 months from the start of treatment). Some experts suggest that when streptomycin is not included in the regimen for the entire 9 months of therapy, ethambutol should be added to the regimen to replace streptomycin, and the duration of treatment should be prolonged from 9 to 12 months. In addition, for patients with a delayed response to therapy, the duration of streptomycin-containing regimens should be prolonged from 9 to 12 months (or to 6 months after documented culture conversion). Alternatives to streptomycin in such regimens are amikacin, kanamycin, or capreomycin.

Clinicians should consider the patient’s response to treatment when making final decisions regarding duration of antituberculosis therapy. The CDC states that completion of such therapy is determined by the total number of administered doses of medication, not the duration of therapy alone. Interruptions in antituberculosis therapy because of drug toxicity or other reasons should be considered when calculating the point at which such therapy is to be discontinued.

The minimum duration of short-course therapy with rifabutin-containing antituberculosis regimens is 6 months, which consists of at least 180 doses when therapy is given daily (one dose daily for 6 months); alternatively, 62 doses can be administered in 6 months as 14 induction doses (one dose daily for 2 weeks) followed by 12 induction doses (2 doses weekly for 6 weeks) plus 36 continuation doses (2 doses weekly for 18 weeks).

The minimum duration of short-course therapy with rifampin-containing antituberculosis regimens is 6 months, which consists of at least 180 doses when therapy is given daily (one dose daily for 6 months); alternatively, 62 or 86 doses can be administered in 6 months as 14 induction doses (one dose daily for 2 weeks) followed by 12 or 18 induction doses (2 or 3 doses weekly for 6 weeks) plus 36 or 54 continuation doses (2 or 3 doses weekly for 18 weeks).

The minimum duration of therapy with nonrifamycin-containing antituberculosis regimens is 9 months, consisting of at least 60 induction doses (one dose daily for 2 months) or, alternatively, 14 induction doses (one dose daily for 2 weeks) followed by 12 or 18 induction doses (2 or 3 doses weekly for 6 weeks); 60 or 90 continuation doses (2 or 3 doses weekly for 30 weeks) should then be administered to complete the 9-month course of therapy.

Reinstitution of therapy in patients with interrupted antituberculosis therapy might require a continuation of the regimen originally prescribed (as long as needed to complete the recommended duration of the particular regimen) or a complete renewal of the regimen. In either situation, when therapy is resumed after an interruption of 2 months or longer, sputum samples (or other clinical samples as appropriate) should be taken for smear, culture, and drug-susceptibility testing.

Antituberculosis regimens that do not contain a rifamycin, an aminoglycoside, or capreomycin (e.g., a regimen consisting of isoniazid, ethambutol, and pyrazinamide) generally should not be used in patients with HIV-related tuberculosis; if such regimens are used, the CDC states that the minimum duration of antituberculosis therapy should be 18 months (or 12 months after documented culture conversion).

The frequency and type of adverse effects with antituberculosis therapy appear to be similar in patients with or without HIV infection. However, several considerations apply to the care of patients with HIV-related tuberculosis: (a) patients with HIV infection may be more likely to experience isoniazid-related peripheral neuropathy; evaluation of dermatologic reactions related to antituberculosis therapy may be complicated because of dermatologic diseases related to HIV disease or to other drug therapy used in these patients; and patients undergoing concurrent therapy with rifabutin and protease inhibitors or NNRTIs are at risk for rifabutin toxicity (e.g., arthralgias, uveitis, leukopenia) associated with elevated serum concentrations of this drug. (See Cautions: Precautions and Contraindications, in Rifabutin 8:16.04.)

The CDC recommends that pyridoxine (25-50 mg daily or 50-100 mg twice weekly) be administered to all HIV-infected patients receiving antituberculosis therapy with isoniazid to reduce the occurrence of isoniazid-induced adverse effects in the CNS or peripheral nervous system.

Paradoxical reactions (temporary exacerbation of tuberculosis symptoms and lesions after initiation of antituberculosis therapy) have occurred rarely in patients without HIV infection receiving antituberculosis therapy and have been attributed to recovery of the patient’s delayed hypersensitivity response and an increase in exposure and reaction to mycobacterial antigens after such therapy.

Similar reactions have occurred in patients with HIV-related tuberculosis, although these reactions appear to be related more often to initiation of potent combination antiretroviral therapy in patients receiving antituberculosis therapy and occur with greater frequency than those associated with initiation of antituberculosis therapy alone. Because an association between paradoxical reactions and concomitant antiretroviral and antituberculosis therapy has been observed, clinicians should be aware of this possibility and discuss the risks with patients undergoing such combined therapy.

Some experts suggest that to avoid paradoxical reactions, clinicians should delay initiation of or changes in antiretroviral therapy until the manifestations of tuberculosis are well controlled (possibly 4-8 weeks following initiation of antituberculosis therapy). For patients with a paradoxical reaction in whom the symptoms are not severe or life-threatening, the CDC states that management of these reactions might consist of symptomatic therapy and no change in antituberculosis or antiretroviral therapy. For patients who have severe or life-threatening manifestations (e.g., uncontrollable fever, airway compromise from enlarging lymph nodes, enlarging serosal fluid collections [pleuritis, pericarditis, peritonitis], sepsis-like syndrome) associated with such reactions, management might include hospitalization and possibly a short course of corticosteroids (e.g., prednisone 60-80 mg daily with reduction in the dosage after 1-2 weeks according to resolution of symptoms); the CDC states that in most cases, corticosteroid therapy should not last longer than 4-6 weeks.

Malabsorption of antituberculosis drugs has been demonstrated in some patients with HIV infection and has been associated in some cases with treatment failures and the selection of drug-resistant M. tuberculosis strains. While therapeutic drug monitoring has been advocated by some clinicians as an adjunct in the management of HIV-related tuberculosis and may be useful in patients with failure or relapse of antituberculosis therapy or those with multidrug-resistant tuberculosis, the CDC states that the role of therapeutic drug monitoring in the routine management of tuberculosis in HIV-infected patients has not been established and currently is not recommended.

Patients Not Receiving Concurrent Antiretroviral Therapy

For the initial treatment of tuberculosis in patients with HIV infection or acquired immunodeficiency syndrome (AIDS) who are not candidates for or who have not started antiretroviral therapy or in patients whose current antiretroviral regimen does not include a protease inhibitor or an nonnucleoside reverse transcriptase inhibitor (NNRTI), the ATS and CDC currently suggest that therapy be initiated with the same 4-drug, 6-month regimens used for tuberculosis in nonimmunocompromised individuals (i.e., isoniazid-rifampin-pyrazinamide plus ethambutol or streptomycin). However, the ATS and CDC state that it is critically important to assess clinical and bacteriologic response in HIV-infected patients on a case-by-case basis, and treatment should be prolonged in patients who respond slowly or otherwise suboptimally. The CDC states that a delayed response to treatment should be suspected (and in most cases antituberculosis therapy should be prolonged) in patients who, after completing the 2-month induction phase of antituberculosis therapy, continue to be culture-positive for M. tuberculosis, do not experience resolution of the manifestations of tuberculosis, or experience progression of tuberculosis manifestations (e.g., persistent fever, progressive weight loss, increase in the size of lymph nodes, abscesses, or other tuberculous lesions) that cannot be accounted for by diseases other than tuberculosis.

Drug-Resistant Tuberculosis in HIV-Infected Individuals

Consultation with experts is recommended when treating patients with drug-resistant tuberculosis. In patients with resistance to isoniazid alone, the CDC recommends administration of a rifamycin (rifampin or rifabutin), pyrazinamide, and ethambutol for the duration of treatment. Therapy with the rifamycin-pyrazinamide-ethambutol regimen may be administered twice weekly following at least 2 weeks (14 doses) of daily induction therapy. The recommended duration of treatment is 6-9 months or 4 months after culture conversion. The CDC states that isoniazid generally is discontinued when high-level resistance (greater than 1% of bacilli resistant to 1 mcg/mL of drug) is present but that some experts recommend continuing isoniazid in cases of low-level resistance (greater than 1% of bacilli resistant to 0.2 mcg/mL of drug but no resistance to 1 mcg/mL). Because the development of acquired rifamycin resistance would result in multidrug-resistant tuberculosis, clinicians should carefully supervise and manage tuberculosis treatment in these patients.

In patients whose tuberculosis is resistant only to rifampin, a 9-month treatment regimen consisting of an initial 2-month phase of isoniazid, streptomycin, pyrazinamide, and ethambutol is recommended. The second phase of treatment should consist of isoniazid, streptomycin, and pyrazinamide administered for 7 months. Because the development of acquired isoniazid resistance would result in multidrug-resistant tuberculosis, clinicians should carefully supervise and manage tuberculosis treatment in these patients.

Patients with tuberculosis that is resistant to both isoniazid and rifampin (multidrug-resistant tuberculosis) should be managed by or in consultation with physicians who are experienced in the management of multidrug-resistant tuberculosis. Early aggressive treatment of multidrug-resistant tuberculosis with appropriate regimens (based on known or suspected drug-resistance pattern of the M. tuberculosis isolate) appears to markedly reduce the incidence of death associated with this form of tuberculosis. The CDC states that most drug regimens currently used to treat multidrug-resistant tuberculosis include an aminoglycoside (e.g., streptomycin, kanamycin, amikacin) or capreomycin and a fluoroquinolone.

The recommended duration of treatment for multidrug-resistant tuberculosis in HIV-seropositive patients is 24 months after culture conversion, and posttreatment follow-up should be conducted every 4 months for 24 months to monitor for relapse. Because of the serious personal and public health concerns associated with multidrug-resistant tuberculosis, health departments should always use directly observed therapy (DOT) for these patients and take whatever steps are necessary to ensure adherence to therapy.

Tuberculosis in HIV-Infected Pediatric Patients

Children with HIV infection who are suspected of having tuberculosis disease should be treated without delay. The CDC states that treatment regimens for such children, even those too young to be evaluated for visual acuity and red-green perception, should include ethambutol (15 mg/kg of body weight) unless the infecting strain of M. tuberculosis is known or suspected of being susceptible to isoniazid and rifampin. If the results of drug-susceptibility tests are not available, a 4-drug regimen (e.g., isoniazid, a rifamycin, pyrazinamide, and ethambutol) given for 2 months, followed by intermittent administration of isoniazid and a rifamycin for 4 months, is recommended.

The AAP states that optimal therapy for treatment of active tuberculosis in children with HIV infection has not been established. Therapy in such children should always be initiated using at least 3 drugs and should be continued for a total duration of at least 9 months. The AAP recommends therapy with isoniazid, rifampin, and pyrazinamide with or without ethambutol or an aminoglycoside for at least the first 2 months; the fourth drug can be discontinued if the infection is found to be caused by susceptible organisms. The AAP states that consultation with a specialist who has experience in the management of tuberculosis in HIV-infected patients is recommended.

Tuberculosis during Pregnancy in HIV-Infected Patients

The CDC states that pregnant women with HIV infection who have a positive culture for M. tuberculosis or who are suspected of having tuberculosis disease should be treated without delay. Recommended treatment regimens for HIV-infected pregnant women are those that include a rifamycin (e.g., rifampin or rifabutin plus isoniazid, ethambutol, and pyrazinamide). Routine use of pyrazinamide during pregnancy is recommended by international organizations but has not been recommended in the US because of inadequate teratogenicity data; however, the CDC states that the benefits of a pyrazinamide-containing antituberculosis regimen outweigh the potential risks of pyrazinamide to the fetus. Aminoglycosides are contraindicated in all pregnant women because of potential adverse effects on the fetus.

• Extrapulmonary Tuberculosis in HIV-infected Patients

Most extrapulmonary forms of tuberculosis (including tuberculous meningitis or lymphadenitis and pericardial, pleural, disseminated, or miliary tuberculosis) are more common among individuals with advanced HIV disease than among those with asymptomatic HIV infection. The basic principles supporting the treatment of pulmonary tuberculosis in patients with HIV infection also apply to extrapulmonary tuberculosis. (See Active Tuberculosis: Active Tuberculosis in HIV-Infected Individuals.) However, for certain forms of extrapulmonary disease (e.g., meningioma, bone or joint tuberculosis), the CDC recommends therapy with a rifamycin-containing antituberculosis regimen for at least 9 months.

Corticosteroids have been used as adjunctive therapy in non-HIV-infected patients to accelerate resolution of symptoms and improve survival in patients with some forms of extrapulmonary tuberculosis (e.g., meningitis, pericarditis); however, current data are insufficient to determine the potential benefits and risks of adjunctive corticosteroid therapy in HIV-infected patients with tuberculosis.

Latent Tuberculosis Infection

Management of Other Mycobacterial Diseases

Some antituberculosis agents also are used in the prevention and/or treatment of diseases caused by mycobacteria other than M. tuberculosis. A variety of terms has been used to designate these mycobacteria, including atypical mycobacteria, mycobacteria other than tubercule bacilli (MOTT), and nontuberculous mycobacteria (NTM). In general, antituberculosis agents are not as effective against NTM as they are against M. tuberculosis and, in some cases, these organisms are resistant to antituberculosis agents.

Many species of NTM (e.g., M. abscessus, M. avium complex [MAC], M. chelonae, M. fortuitum, M. gastri, M. gordonae, M. kansasii, M. malmoense, M. marinum, M. phlei, M. scrofulaceum, M. simiae, M. smegmatis, M. terrae, M. triviale, M. vaccae, M. xenopi) are widely distributed in water and/or soil and the source for most human infections caused by many of these NTM appears to be the environment, especially water (e.g., tap water, salt water, fresh water, fish tanks, swimming pools). Other species of NTM (e.g., M. celatum, M. conspicuum, M. genavense, M. haemophilum, M. szulgai, M. ulcerans) have not yet been isolated from the environment, although an environmental source is highly likely. Person-to-person transmission of NTM infections appears to be rare, and most individuals are infected by environmental exposures.

NTM infections usually involve chronic pulmonary disease (e.g., M. abscessus, MAC, M. kansasii, M. mammoense, M. xenopi), lymphadentis (e.g., MAC, M. malmoense, M.scrofulaceum), localized skin, soft tissue, or skeletal disease (e.g., M. abscessus, MAC, M. chelonae, M. fortuitum, M. marinum, M. ulcerans), or disseminated infection (e.g., M. abscessus, MAC, M. chelonae, M. haemophilum, M. kansasii, M. scrofulaceum). In the absence of specific diagnostic features obtained from patient history, physical exam, chest radiograph, and differential skin testing, isolation of NTM in culture is essential for diagnosis. However, since many NTM are found in the environment, contamination of culture material or transient infection can occur and single or intermittent recovery of small numbers of one of these organisms in a patient without evidence of disease may not be clinically important.

HIV-infected individuals are at especially high risk of disease caused by NTM. Disseminated MAC infection is the most common bacterial infection in patients with acquired immunodeficiency syndrome (AIDS), and MAC disease in these patients is highly correlated with severe immunosuppression. Other NTM that have been reported as causes of pulmonary and/or disseminated disease in patients with AIDS include M. celatum, M. conspicuum, M. fortuitum, M. genavense, M. gordonae, M. haemophilum, M. kansasii, M. malmoense, M. marinum, M. scrofulaceum, M. simiae, and M. xenopi. In addition, disseminated infections caused by MAC, M. chelonae, M. haemophilum,M. kansasii, and M. scrofulaceum have been reported to cause disease in adults with other forms of immunosuppression (e.g., renal or cardiac transplant, leukemia, chronic immunosuppressive therapy).

Susceptibility Testing

Prior to initiation of therapy, appropriate specimens should be collected for identification of the causative organism and, in some cases, for in vitro susceptibility testing. Optimal methods for in vitro susceptibility testing of mycobacteria other than M. tuberculosis are still being investigated Many nontuberculous mycobacteria (NTM) are resistant in vitro to most currently available antituberculosis agents and in vitro susceptibility does not always correlate to in vitro activity. Therefore, routine in vitro susceptibility testing of all mycobacterial isolates is not indicated but may be beneficial depending on the specific species involved. In addition, there may be some circumstances where in vitro susceptibility testing may be warranted to provide baseline data in case the patient fails to respond to therapy or relapses.

In vitro susceptibility testing of MAC isolates using rifabutin and other antituberculosis agents generally is not recommended. Routine in vitro susceptibility testing of pretreatment isolates of MAC using clarithromycin also is not generally recommended, but should be performed on isolates from patients who have received prior macrolide therapy and have failed treatment or prophylaxis. There is some evidence that radiometric broth methods are more reliable than agar methods for testing in vitro susceptibility of MAC.

Routine in vitro susceptibility testing of M. kansasii using rifampin should be performed prior to treatment and may provide useful information since acquired resistance to rifampin has been reported in some strains of this organism. In addition, in vitro susceptibility testing using rifampin is indicated in patients who fail to respond to treatment or relapse. Routine in vitro susceptibility testing of rifampin-susceptible M. kansasii using other drugs (e.g., ethambutol, isoniazid) is not recommended. However, in vitro susceptibility testing using ciprofloxacin (or ofloxacin), clarithromycin, ethambutol, streptomycin, and a sulfonamide (e.g., sulfamethoxazole) may be useful if rifampin-resistant isolates are identified.

In vitro susceptibility tests may be useful in cases of M. marinum infections that have relapsed or failed to respond to regimens containing rifampin. In vitro susceptibility testing of M. marinum using ethambutol, rifampin, doxycycline (or minocycline), clarithromycin, and a sulfonamide may be useful.

In vitro susceptibility testing of M. haemophilum, M. malmoense, M. simiae, M. szulgai, and M. xenopi using ethambutol, isoniazid, rifampin, clarithromycin, and ciprofloxacin may be useful since information on susceptibility patterns of these species is limited.

In vitro susceptibility testing of rapidly growing mycobacteria (M. abscessus, M. chelonae, M. fortuitum) should not be performed using antituberculosis agents. These organisms should be tested for in vitro susceptibility to amikacin, doxycycline, imipenem, ciprofloxacin, sulfonamides, cefoxitin, and clarithromycin.

Specialized references should be consulted for further information on in vitro susceptibility testing for mycobacteria other than M. tuberculosis and for information on alternative methods of testing susceptibility of MAC.

Mycobacterium avium Complex (MAC) Infections

Treatment of Pulmonary and Localized Extrapulmonary MAC Infections

Mycobacterium avium complex (MAC) represents 2 closely related organisms, M. avium and M. intracellulare. MAC strains frequently are resistant in vitro to most of the available antituberculosis agents, and treatment of MAC infections has varied depending on the type and severity of infection and the immune status of the patient. Some immunocompetent patients with stable MAC infections may do well with no specific drug therapy and only close medical observation. Surgical excision without chemotherapy may be adequate for immunocompetent patients with either localized lymphadenitis or a solitary, resectable pulmonary nodule. However, the ATS states that with the availability of more effective and tolerable treatment regimens for MAC infection, there should be less reluctance to treat MAC lung disease at an earlier stage. Because the treatment of MAC disease involves multiple drugs with their associated toxicities and an optimal regimen has not yet been established, such treatment may best be accomplished by clinicians experienced in pulmonary or mycobacterial diseases.

In patients receiving treatment for disseminated MAC disease, the risk of rifabutin-associated uveitis reportedly may be increased because clarithromycin inhibits the metabolism of rifabutin and can increase plasma (and presumably tissue) concentrations of rifabutin. (See Cautions: Precautions and Contraindications, in Rifabutin 8:16.04.) While the clinical importance has not been established, concomitant administration of clarithromycin and rifampin or rifabutin may also affect metabolism of clarithromycin. The potential for alterations in the plasma concentrations of the antimycobacterial agent(s) and/or antiretroviral agents must be considered when antimycobacterial agents are indicated for the prophylaxis or treatment of MAC infections in HIV-infected patients who are receiving or are being considered for antiretroviral therapy.

The ATS and some clinicians currently recommend that symptomatic or progressive disease in immunocompetent adults with mild to moderately advanced pulmonary MAC infections be treated with a regimen that includes clarithromycin (500 mg twice daily) or azithromycin (250 mg daily or 500 mg 3 times weekly), rifabutin (300 mg daily) or rifampin (600 mg daily), and ethambutol (25 mg/kg daily for 2 months, then 15 mg/kg daily). The ATS states that the addition of streptomycin given intermittently (2 or 3 times weekly) for at least 2 months may be considered for patients with extensive disease; longer therapy may be desirable in patients with very extensive disease or in those who do not tolerate other agents. Studies evaluating the efficacy and tolerability of azithromycin- and clarithromycin-containing regimens given intermittently (i.e., 3 times weekly) for the treatment of MAC pulmonary infections currently are ongoing. The ATS states that the optimal duration of therapy for MAC pulmonary disease has not been established; however, limited data indicate that combined therapy with clarithromycin, ethambutol, and rifampin or rifabutin, with an initial period of streptomycin, can achieve high response rates (conversion of sputum cultures to negative in up to 92% of patients) for prolonged periods (e.g., mean duration of culture negativity while on therapy: 12 months). The ATS states that if clinical or microbiologic improvement is not seen in 3-6 months or sputum conversion does not occur within 12 months in patients with MAC infections receiving macrolide-containing regimens, treatment should be reassessed.

For patients whose MAC isolate becomes resistant to macrolides or in whom initial therapy fails or is not tolerated, subsequent therapy is complicated, may be associated with a high incidence of adverse effects, and should be undertaken only under the guidance of health care workers experienced in handling these patients. The ATS states that a 4-drug regimen consisting of isoniazid, rifabutin, ethambutol, and initial (first 3-6 months) streptomycin therapy may be reasonable in patients who have failed therapy with macrolide-containing regimens. The ATS states that other drugs used in multiple-drug regimens are limited by toxicity (e.g., cycloserine, ethionamide) or little or no evidence of efficacy (e.g., clofazimine, newer quinolones, capreomycin). Although a regimen that includes ciprofloxacin (750 mg twice daily) or ofloxacin (400 mg twice daily), clofazimine (100 mg daily), ethionamide (250 mg twice daily, increased to 3 times daily as tolerated), and prolonged use of streptomycin or amikacin (3-5 times weekly) may be associated with at least short-term conversion of sputum to negative, the ATS states that the long-term success of such “salvage” regimens is unknown but likely very low.

For the treatment of extrapulmonary MAC soft-tissue infection, arthritis, or osteomyelitis confined to a single site in otherwise healthy patients, drainage or debridement combined with the above-mentioned initial drug regimen (i.e., clarithromycin, ethambutol, rifampin or rifabutin, and streptomycin) has been recommended.

Primary Prevention of Disseminated MAC Infection

Prevention of disseminated MAC disease is an important goal in the management of patients with HIV infection and low helper/inducer (CD4+, T4+) T-cell counts because of the frequency with which the disease occurs in such patients and its associated morbidity. Current evidence from controlled studies indicates that primary prophylaxis with azithromycin, clarithromycin, or rifabutin or the combination of azithromycin and rifabutin can substantially reduce the frequency of MAC bacteremia and ameliorate clinical manifestations of the disease in patients with AIDS. Primary prophylaxis with clarithromycin or rifabutin also has been shown to improve survival in patients with advanced HIV infection; although similar results with azithromycin prophylaxis might be presumed, whether azithromycin-containing regimens improve survival in such patients has not been established to date. BR>

Results of a limited number of controlled studies in patients with advanced HIV infection indicate that clarithromycin is more effective than placebo in preventing disseminated MAC disease. In a randomized, double-blind study in patients with acquired immunodeficiency syndrome (AIDS) and baseline median CD4+ T-cell counts of 25-30 cells/mm3, the risk of MAC infection was reduced by 69% in patients receiving clarithromycin 500 mg twice daily compared with that in patients receiving placebo. Although clarithromycin-resistant MAC isolates were found in many patients who developed MAC infection despite prophylaxis, clarithromycin prophylaxis was still associated with a 26% reduction in mortality compared with placebo.

Primary prophylaxis with azithromycin alone or in combination with rifabutin also has been shown to be effective against disseminated MAC infection in patients with advanced HIV infection. In a randomized, comparative study in patients with advanced HIV (CD4+ counts less than 100/mm3), the risk of disseminated MAC infection (after adjustment for baseline CD4+ counts) in patients receiving azithromycin prophylaxis (1. g once weekly) was 47% lower than that with rifabutin prophylaxis (300 mg daily), while prophylaxis with both drugs reduced the risk by 72% compared with rifabutin alone; survival among the 3 groups was similar. In addition to reducing the risk of disseminated MAC infection, prophylaxis with azithromycin alone or in combination with rifabutin provided additional protection (45% risk reduction) against Pneumocystis carinii pneumonia compared with that provided by rifabutin alone in patients without previous P. carinii episodes (primary prophylaxis).Among patients in whom prophylaxis with azithromycin was unsuccessful, resistance to azithromycin (and clarithromycin) was found in 11%. The overall incidence of adverse effects was similar among the 3 groups, although dose-limiting adverse effects (principally GI effects) occurred more frequently with combined azithromycin-rifabutin prophylaxis than with either drug alone.

In 2 controlled studies, patients with AIDS and CD4+ T-cell counts of 200/mm3 or less who received rifabutin 300 mg daily were one-half to one-third as likely to develop MAC bacteremia and its clinical manifestations as those receiving placebo.Rifabutin also was well tolerated in these studies; the incidence of adverse effects was similar in both the treatment and placebo groups. Although mortality rates in the individual studies were not significantly reduced with rifabutin prophylaxis, an analysis of the combined double-blind and open follow-up periods of these studies indicated that prophylaxis with the drug was associated with improved survival over a period of approximately 700 days of follow-up.

Current evidence suggests that primary prophylaxis against disseminated MAC infection is superior to efforts aimed at early detection and treatment of the disease in terms of survival benefit. The Prevention of Opportunistic Infections Working Group of the US Public Health Service and the Infectious Diseases Society of America (USPHS/IDSA) currently recommends that primary prophylaxis against MAC disease be given to HIV-infected adults and adolescents (13 years and older) who have absolute helper/inducer (CD4+, T4+) T-cell counts less than 50/mm3. Severely immunocompromised HIV-infected children younger than 13 years of age also should receive primary prophylaxis against MAC disease according to the following age-specific CD4+ T-cell counts: children 6-13 years of age, less than 50 cells/mm3; children 2-6 years of age, less than 75 cells/mm3; children 1-2 years of age, less than 500 cells/mm3; and children younger than 1 year of age, less than 750 cells/mm3. The USPHS/IDSA currently states that clarithromycin or azithromycin is the preferred agent for prophylaxis; alternatively, if these drugs cannot be tolerated, rifabutin may be used. Patients in whom primary MAC prophylaxis is considered must be evaluated (e.g., using chest radiography, tuberculin skin tests, blood cultures for MAC) prior to initiation of such prophylaxis to rule out the presence of active infection with M. tuberculosis or disseminated MAC infection, since monotherapy is not adequate for the treatment of either disease and results in drug resistance and clinical failure.

In selecting a prophylactic regimen, consideration should be given to the potential for clinically important interactions between rifabutin, macrolide antibiotics (e.g., azithromycin, clarithromycin), and other drugs commonly used in HIV-infected patients (e.g., HIV protease inhibitors, NNRTIs). (See Patients Receiving Concurrent Antiretroviral Therapy, under Active Tuberculosis: Active Tuberculosis in HIV-Infected Individuals.) While azithromycin has not been directly compared with clarithromycin for prophylaxis of MAC infection, advantages of azithromycin compared with clarithromycin or rifabutin include once-weekly rather than daily administration (resulting in comparatively less costly prophylaxis) and, based on limited data/experience, a relatively low potential for interaction with other drugs commonly used in patients with AIDS (e.g., HIV protease inhibitors such as indinavir, ritonavir, saquinavir). In addition, limited data suggest that the potential for development of resistance in patients in whom azithromycin prophylaxis fails is relatively low. Although the combination of azithromycin and rifabutin is more effective than azithromycin alone, the USPHS/IDSA currently does not recommend routine primary prophylaxis with the combination because of additional cost, increased incidence of adverse effects, and absence of a difference in survival in patients receiving the combination compared with azithromycin alone. In addition, the USPHS/IDSA states that the combination of clarithromycin and rifabutin should not be used since such combined therapy is no more effective than clarithromycin alone and is associated with a higher incidence of adverse effects. However, some clinicians state that combined azithromycin-rifabutin prophylaxis may have a role in patients with very low CD4+ counts because of their high risk of developing disseminated MAC disease.

HIV-infected pregnant women are at risk for MAC disease, and primary prophylaxis should be given to such women who have T-cell counts less than 50/mm3. However, some clinicians may choose to withhold prophylaxis during the first trimester of pregnancy because of general concerns regarding drug administration during this period. Of the available agents, the USPHS/IDSA considers azithromycin the drug of choice for MAC disease primary prophylaxis in HIV-infected pregnant women because of the drug’s safety profile in animal studies and anecdotal information on safety in humans. Experience with rifabutin is limited. Clarithromycin has demonstrated adverse effects on pregnancy outcome and/or embryo-fetal development in animals and should be used during pregnancy only in clinical circumstances where no alternative therapy is appropriate.

Current evidence indicates that primary MAC prophylaxis can be discontinued with minimal risk of developing disseminated MAC disease in HIV-infected adults and adolescents who have responded to highly active antiretroviral therapy (HAART) with an increase in CD4+ T-cell counts to greater than 100/mm3 that has been sustained for at least 3 months. The USPHS/IDSA states that discontinuance of primary prophylaxis is recommended in adults and adolescents meeting these criteria because prophylaxis appears to add little benefit in terms of disease prevention for MAC or bacterial infections, and discontinuance reduces the medication burden, the potential for toxicity, drug interactions, selection of drug-resistant pathogens, and cost. However, the USPHS/IDSA states that primary MAC prophylaxis should be restarted in adults and adolescents if CD4+ T-cell counts decrease to less than 50-100/mm3. The safety of discontinuing MAC prophylaxis in children whose CD4+ T-cell counts have increased as a result of highly active antiretroviral therapy has not been studied to date.

Treatment and Prevention of Recurrence of Disseminated MAC Infection

Disseminated MAC disease characteristically occurs in patients with advanced HIV infection who have absolute helper/inducer (CD4+, T4+) T-cell counts less than 100/mm3. Limited data from retrospective and prospective analyses indicate that disseminated MAC infection is an independent predictor of mortality in patients with AIDS and that treatment of the infection may increase survival. However, death occurs rapidly following diagnosis of disseminated MAC infection, underscoring the importance of early diagnosis and of antimycobacterial prophylaxis for this infection.

Currently available data suggest that multiple-drug regimens containing a macrolide (e.g., clarithromycin, azithromycin) are superior to non-macrolide-containing regimens for the treatment or prevention of recurrence of MAC disease. Monotherapy with a macrolide antibiotic (e.g., azithromycin, clarithromycin) in a limited number of studies has been effective in reducing MAC bacteremia and improving constitutional symptoms of MAC disease (e.g., fever, night sweats). However, in a dose-ranging study in which clarithromycin dosages of 500 mg, 1 g, or 2 g twice daily were used, clinical relapse and clarithromycin-resistant MAC isolates developed after several months in a large proportion of patients. Although not fully understood, survival in patients receiving high clarithromycin dosages (e.g., 1 or 2 g twice daily) for the treatment of disseminated MAC infection in some studies has been reported to be shorter compared with that in patients receiving 500 mg twice daily; therefore, clarithromycin dosages higher than 500 mg twice daily are not recommended for treatment of disseminated MAC infection.

Results of a randomized, comparative study in patients with AIDS and MAC bacteremia demonstrated improved functional status, decreased weight loss, and increased survival in patients receiving a 3-drug, clarithromycin-containing regimen compared with a 4-drug regimen that did not include a macrolide antibiotic. In this study, MAC bacteremia was cleared in 69% of evaluable patients receiving clarithromycin (1 g twice daily), ethambutol (approximately 15 mg/kg daily), and rifabutin (300 or 600 mg daily) compared with 29% of those receiving rifampin (600 mg daily), ethambutol (approximately 15 mg/kg daily), clofazimine (100 mg daily), and ciprofloxacin (750 mg twice daily). Median survival was 8.6 months with the clarithromycin-containing regimen versus 5.2 months for the 4-drug regimen. Among patients treated for at least 4 weeks, MAC bacteremia resolved more frequently with the 3-drug regimen. The dosage of rifabutin used in the 3-drug regimen was reduced from 600 to 300 mg daily in the latter part of the study following an unacceptably high incidence of uveitis in patients receiving this regimen. However, although the 600-mg dosage of rifabutin was more effective in clearing MAC bacteremia, the 3-drug regimen that included rifabutin 300 mg daily was still more effective than the 4-drug, non-macrolide-containing regimen. Poor response of MAC infections in patients with underlying HIV infection may be secondary to ineffective antimycobacterial therapy, severity of mycobacterial infection, severity of immunodeficiency, presence of other opportunistic infection, or a combination of these factors.

Most authorities currently recommend the use of multiple-drug regimens that include clarithromycin or azithromycin for the treatment of disseminated MAC infections. The ATS and some clinicians currently recommend therapy that includes either clarithromycin or azithromycin combined with ethambutol and rifabutin for the treatment of disseminated MAC infections in HIV-infected patients. The choice of a drug regimen should be made in consultation with an expert and treatment may need to be continued for the duration of the patient’s life if such therapy is associated with clinical and microbiologic improvement. Limited data from comparative trials suggest that inclusion of clofazimine in multiple-drug regimens containing clarithromycin (e.g., with or without ethambutol) does not add to the efficacy (e.g., in terms of prevention of clarithromycin resistance) of such regimens and may even be associated with reduced survival; therefore, the USPHS/IDSA states that clofazimine should not be used for the treatment or prevention of recurrence of disseminated MAC disease.

Clinical manifestations of disseminated MAC disease, such as fever, weight loss, and night sweats, should be monitored during the initial weeks of therapy; periodic (e.g., every 4 weeks) blood cultures also may be useful for determining response to therapy. Most patients with disseminated MAC disease who respond have substantial clinical improvement within the first 4-6 weeks of therapy. In patients who have a recurrence of MAC bacteremia following an initial clinical and microbiologic response, determination of MICs of azithromycin and clarithromycin may be useful for guiding decisions regarding the use of these agents in subsequent therapy for disseminated MAC disease.

To prevent recurrence of MAC disease in HIV-infected adults and adolescents who have previously been treated for an acute episode of MAC infection, the USPHS/IDSA currently recommends a regimen consisting of clarithromycin (500 mg twice daily) plus ethambutol (15 mg/kg once daily) with or without rifabutin (300 mg once daily). Alternatively, the USPHS/IDSA states that azithromycin (500 mg once daily) may be used instead of clarithromycin in such a regimen. In HIV-infected infants and children, the USPHS/IDSA currently recommends a regimen consisting of clarithromycin (7. mg/kg twice daily, not to exceed 500 mg twice daily) plus ethambutol (15 mg/kg once daily, not to exceed 900 mg daily), with or without rifabutin (5 mg/kg once daily, not to exceed 300 mg daily). Alternatively, azithromycin (5 mg/kg once daily, not to exceed 250 mg daily) can be used instead of clarithromycin in such a regimen. The USPHS/IDSA considers azithromycin plus ethambutol the preferred drug regimen for secondary prevention of disseminated MAC infections in pregnant women.

Secondary MAC prophylaxis generally is administered for life in adults and adolescents unless immune recovery has occurred as a result of potent antiretroviral therapy. Limited data indicate that the risk of recurrence of MAC is low if secondary MAC prophylaxis is discontinued in HIV-infected adults and adolescents who have successfully completed at least 12 months of MAC therapy, have remained asymptomatic with respect to MAC, and have CD4+ T-cell counts greater than 100/mm3 as the result of potent antiretroviral therapy and this increase has been sustained (e.g., for 6 months or longer). Based on these data and more extensive cumulative data regarding safety of discontinuing secondary prophylaxis for other opportunistic infections, the USPHS/IDSA states that it is reasonable to consider discontinuance of secondary MAC prophylaxis in adults and adolescents meeting these criteria. Some experts would obtain a blood culture for MAC (even in asymptomatic patients) prior to discontinuing secondary MAC prophylaxis to substantiate that the disease is no longer active. The USPHS/IDSA recommends that secondary MAC prophylaxis should be restarted in adults of adolescents if CD4+ T-cell counts decrease to less than 100/mm3.

Children with a history of disseminated MAC should receive lifelong secondary prophylaxis. The safety of discontinuing secondary MAC prophylaxis in HIV-infected children receiving potent antiretroviral therapy has not been studied.

Mycobacterium kansasii Infections

M. kansasii generally is susceptible in vitro and in vivo to rifampin, isoniazid, ethambutol, ethionamide, streptomycin, and clarithromycin. Although the clinical importance is unclear, the organism also may be susceptible in vitro to sulfamethoxazole, amikacin, some quinolones, and rifabutin. M. kansasii is resistant to aminosalicylic acid, capreomycin, and pyrazinamide.

For the treatment of pulmonary or extrapulmonary M. kansasii infections, 3 primary antituberculosis agents (usually isoniazid-ethambutol-rifampin) are administered daily for 18 months with at least 12 months of negative cultures. There are no published controlled studies to date comparing drug regimens used in the treatment of M. kansasii infections. The ATS currently recommends that pulmonary or extrapulmonary M. kansasii infections be treated with isoniazid (300 mg daily), rifampin (600 mg daily), and ethambutol (25 mg/kg daily for the first 2 months, then 15 mg/kg daily) for 18 months. In patients unable to tolerate one of these 3 drugs, the ATS suggests that clarithromycin may be substituted in the regimen; however, efficacy of clarithromycin in the treatment of M. kansasii infections has not been established in clinical trials. Use of intermittent or short-course regimens for the treatment of M. kansasii infections currently are being investigated, but have not been adequately studied to date. Pyrazinamide is inactive against M. kansasii and should not be used for the treatment of infections caused by the organism.

In patients with infections caused by rifampin-resistant M. kansasii, a regimen of isoniazid (900 mg daily), pyridoxine (50 mg daily), ethambutol (25 mg/kg daily), and sulfamethoxazole (1 g 3 times daily) given until the patient is culture negative for 12-15 months is being investigated and appears to be effective in some patients. In addition, some patients have received streptomycin or amikacin therapy (daily or 5 times weekly) for the initial 2-3 months followed by intermittent streptomycin or amikacin therapy for a total of at least 6 months in conjunction with the above oral regimen. The ATS states that the excellent in vitro activity of clarithromycin against M. kansasii suggests that this drug also may be useful in retreatment regimens, perhaps allowing for omission of the aminoglycoside.

The optimum regimen for the treatment of M. kansasii infections in patients with HIV infection has not been determined. Pharmacokinetic interactions between certain antimycobacterial agents (e.g., rifampin, rifabutin) and HIV protease inhibitors or NNRTIs have been reported and may complicate drug therapy for mycobacterial infections in HIV-infected patients. Because rifampin is a potent inducer of the P-450 CYP3A isoenzyme and has been shown to markedly reduce plasma concentrations of HIV protease inhibitors, rifampin and HIV protease inhibitors generally should not be administered concomitantly. The ATS states that options for treating M. kansasii disease in patients with HIV infection who are receiving a protease inhibitor include substitution of clarithromycin for rifampin in the standard regimen (i.e., isoniazid-rifampin-ethambutol) or substitution of rifabutin 150 mg daily for rifampin. The ATS states that while none of these regimens has been studied clinically, they appear likely to be successful. For further information on drug interactions involving the antiretroviral agents and recommendations regarding these interactions.

Relapse after successful treatment of M. kansasii infections may rarely occur, but use of rifampin in the initial treatment regimen appears to reduce the relapse rate.

Mycobacterium marinum Infections

M. marinum generally are susceptible in vitro to rifampin and ethambutol, but are resistant to isoniazid and pyrazinamide; the organism may be susceptible to high concentrations of streptomycin. M. marinum also may be susceptible in vitro to clarithromycin, sulfonamides, co-trimoxazole, and high concentrations of doxycycline or minocycline.

Some M. marinum infections may be self-limited and resolve spontaneously without treatment. Granulomas of the skin caused by M. marinum that require treatment may respond to daily administration of rifampin or ethambutol-rifampin; surgical excision may also be indicated. Although rifampin may be effective when used alone for the treatment of M. marinum infections, experience with use of the drug alone is limited 268 and the ATS and some clinicians suggest that rifampin (600 mg) and ethambutol (15 mg/kg) should be used concomitantly for the treatment of these infections to delay or prevent the emergence of rifampin resistance. Tetracyclines (i.e., doxycycline, minocycline) and other antibiotics (e.g., clarithromycin, co-trimoxazole) also are recommended for the treatment of granulomas caused by M. marinum. Optimum duration of treatment for M. marinum infections is not well established. Some clinicians suggest that 6-12 weeks of therapy may be adequate, but other clinicians recommend that therapy be continued for up to 6-12 months. The ATS recommends a minimum of 3 months of therapy for the treatment of M. marinum infections.

Mycobacterium abscessus, M. chelonae, M. fortuitum, and M. smegmatis Infections

M. abscessus, M. chelonae, M. fortuitum, and M. smegmatis generally are resistant in vitro to currently available antituberculosis agents; however, M. smegmatis may be susceptible to ethambutol. These organisms may be susceptible in vitro to other anti-infective agents including amikacin, cefoxitin, quinolones, sulfonamides, clarithromycin, doxycycline, and imipenem.

Cutaneous M. fortuitum or M. chelonae infections may slowly resolve spontaneously or be severe and persistent; disseminated M. chelonae infections are most common in immunocompromised patients and neutropenia appears to be a predisposing factor in such dissemination. Chronic pulmonary M. fortuitum or M. chelonae infections generally are indistinguishable from pulmonary infections caused by MAC or M. kansasii. The optimum regimen for the treatment of M. fortuitum or M. chelonae infections has not been determined to date; the ATS states that primary antituberculosis agents have no role in the treatment of these infections. Some clinicians recommend that serious M. fortuitum or M. chelonae infections be treated with amikacin and cefoxitin for at least 2-6 weeks, followed by therapy with an oral sulfonamide, tetracycline, or erythromycin; therapy should be continued for 4-6 weeks after wound healing is complete.

The ATS recommends that a regimen of IV amikacin (10-15 mg/kg in 2 divided doses in adults with normal renal function [average 400 mg twice daily] to provide serum concentrations in the low 20 mcg/mL range) and IV cefoxitin (12 g daily) be given for at least 2 weeks until clinical improvement occurs; the lower daily dosage of amikacin (10 mg/kg) should be used in patients older than 50 years of age. The ATS states that a minimum of 4 months of therapy is necessary for the treatment of serious infections; 6 months of therapy is recommended for bone infections.

Although experience in the treatment of M. smegmatis infections is limited, osteomyelitis, cellulitis, bacteremia, and wound infections caused by the organism have been effectively treated with regimens based on results of in vitro susceptibility tests using amikacin, cefoxitin, doxycycline, co-trimoxazole, and/or imipenem; at least one case of pleuropulmonary disease caused by M. smegmatis responded well to a regimen of doxycycline, co-trimoxazole, and ethambutol.

Mycobacterium malmoense, M. simiae, M. szulgai, and M. xenopi

Optimum regimens for the treatment of pulmonary or disseminated infections caused by M. malmoense, M. simiae, M. szulgai, or M. xenopi have not been identified. The ATS and some clinicians suggest that multiple-drug regimens used for the treatment of MAC infections (e.g., clarithromycin, ethambutol, and rifabutin with or without streptomycin) can be considered for initial treatment of infections caused by these NTM with subsequent modification based on results of in vitro susceptibility testing. The ATS recommends that patients with M. xenopi infections receive initial therapy with a macrolide, rifampin or rifabutin, and ethambutol without or without initial streptomycin and that surgery should be considered for patients who do not respond to therapy or who relapse after completion of therapy.

Leprosy

Rifampin is used in conjunction with other anti-infectives, including dapsone, clofazimine, minocycline, and ofloxacin, for the treatment of leprosy. For information on the treatment of leprosy.

For information on chemistry and stability, mechanism of action, spectrum, resistance, pharmacokinetics, cautions, acute toxicity, drug interactions, and dosage and administration of the antituberculosis agents.