CHEMOTHERAPY OF TUBERCULOSIS
Common Side Effects of Drugs Used in Short-course Chemotherapy
The potential to control tuberculosis has existed only since the late 1940s, when antituberculous medications became available. The impact of such medications became obvious quickly. Between 1932 and 1983, the number of new active cases declined from 96,500 to 23,500 per year, and mortality dropped from 71,000 to 1,980 per year. In 1986, however, the number of new cases increased by 2.6%. In 1991, 26,283 new cases were reported and were associated with direct health care expenditures estimated at $703 million. In 1997, 19,855 cases of tuberculosis were reported, representing a 7% decrease from what was seen the year before and a 26% decrease from what was noted in 1992—the peak year of the tuberculosis resurgence. Much of the increase seen since 1986 represents disease in persons simultaneously infected with HIV and tuberculosis. Outbreaks of multidrug-resistant tuberculosis have also occurred in this population and have spread to health care workers, resulting in significant morbidity and mortality. The problem has been compounded by the limited availability of streptomycin.
Through the 1970s, two-drug therapy administered for up to 2 years provided excellent results, with relapse rates predictably less than 5%. However, treatment of this length was associated with higher costs, increased side effects, and problems with compliance. During the past one to two decades, studies documented the efficacy and safety of shorter courses of chemotherapy using at least two effective agents. Treatment combining isoniazid (INH) and rifampin became commonplace. As an example, for sensitive organisms, daily therapy for 9 months with 300 mg of INH orally plus 600 mg of rifampin orally was associated with relapse rates of less than 3% and became a standard of care for uncomplicated pulmonary and extrapulmonary tuberculosis. However, the recent emergence of multidrug-resistant tuberculosis has required a reassessment of this issue.
In 1998, the FDA approved the first new antituberculous agent in almost 25 years. Rifapentine (Priftin; Hoechst Marion Roussel) functions as a long-active form of rifampin and has been approved for use in patients with tuberculosis but without HIV/AIDS. Doses may range from 600 mg/d to 300 mg twice weekly. Dosing with this drug can be as infrequent as once weekly, which, it is hoped, will decrease noncompliance. Preliminary data, however, suggest slightly higher relapse rates than are seen with rifampin. This new product is also likely to interact with protease inhibitors.
Animal studies have demonstrated that INH plus a second bactericidal agent can rapidly cure murine TB. Other in vitro investigations have shown that tubercle bacilli can be killed only when dividing, and that organisms at different sites vary in their frequency of replication. Those that are dormant are most resistant to killing. Mycobacterium tuberculosis infection can be classified by the location of lesions as follows:
Cavitary lesions: M. tuberculosis replicates best at the high oxygen tensions found in body cavities. Numbers of bacteria approach 108/g of tissue. Organisms in cavitary lesions are usually extracellular and in a neutral or alkaline pH. Streptomycin, INH, and rifampin are active against organisms at these sites and rapidly kill them.
Closed caseous lesions (organisms usually extracellular): Antituberculous drugs are less effective at these sites than within cavities. Organisms are metabolically less active and tend to number only 104 to 105/g of tissue. The pH remains neutral or slightly alkaline. Rifampin and INH are most effective at these locations.
Within macrophages (intracellular organisms): Organisms within macrophages have a low rate metabolic activity, may be intermittently dormant, and exist at an acid pH in numbers of 104 to 105/g of tissue. Those that are dormant are unresponsive to all medications. Although both INH and rifampin have modest activity within this acidic environment, pyrazinamide is most effective.
Patients infected with HIV are at increased risk for infections with M. tuberculosis. This pathogen is noted in 2% to 10% of patients with AIDS and often occurs in extrapulmonary locations. The likelihood of extrapulmonary disease increases with more advanced HIV infection. In recent studies of patients in whom tuberculosis has been newly diagnosed, up to 30% are simultaneously infected with HIV. Additionally, in patients with HIV infection, reactivation of latent infection is more likely, extrapulmonary disease is seen more often, and disease progresses more rapidly to active tuberculosis. The rate of development of active disease in patients with HIV is approximately 8% annually, compared with 10% in a lifetime for persons not infected with HIV. Fortunately, in the absence of multidrug-resistant tuberculosis, response to antituberculous therapy is similar to that seen in persons not infected with HIV.
Multidrug-resistant tuberculosis is being increasingly observed and is strongly associated with both HIV infection and increased mortality; fatality rates have approached 90% in some studies. The possibility of improved outcomes has been acknowledged in recent studies. These demonstrate that early identification of patients at risk and use of effective combination therapies can result in earlier bacteriologic conversion and prolonged survival. With appropriate management, survival of more than 1 year was seen in 59% of patients with multidrug-resistant tuberculosis. Factors that have been identified as increasing survival in patients with multidrug-resistant tuberculosis include initiation of treatment within 4 weeks of diagnosis, treatment with at least two effective agents for at least 2 consecutive weeks, and disease limited to the lungs. Appropriate treatment for at least 2 weeks consecutively is probably the most important of these. Multidrug resistance often extends beyond resistance to INH and rifampin. Consideration for multidrug-resistant tuberculosis is made on epidemiologic grounds, as most cases have occurred in closed populations within hospitals or prisons and in localized geographic areas, including Florida and New York. Health care workers exposed to actively infected persons have also become ill. Recent national data suggest that 7% to 8% of M. tuberculosis isolates are now INH-resistant, and that 1% to 2% of strains are multidrug-resistant. Forty-three states plus the District of Columbia reported strains of multidrug-resistant tuberculosis between 1993 and 1997. Other data from New York, an area with increased numbers of patients with AIDS, demonstrate that approximately 33% of isolates of M. tuberculosis are resistant to one drug and 19% are multiply resistant. Knowledge of applicable resistance patterns is mandatory to effect treatment.
Recommendations for empiric therapy of M. tuberculosis before the availability of sensitivities must take into account the HIV status of the patient and the likelihood of resistant organisms being present. Table 67-1 provides current recommendations for initial empiric therapy in HIV-negative patients. A four-drug regimen that usually consists of INH (300 mg daily), rifampin (600 mg daily), pyrazinamide (15 to 30 mg/kg daily, 2 g maximum), and ethambutol (15 mg/kg daily) or streptomycin (20 to 30 mg/kg daily, 1 g maximum) is generally recommended. A three-drug regimen that does not include ethambutol or streptomycin can be used if the likelihood of INH resistance is less than 4% in a community, and if there are no risk factors for multidrug-resistant tuberculosis. The same regimens can be used in patients who are HIV-positive; however, the length of therapy must be extended to at least 9 months, or 6 months beyond conversion of sputum cultures.
Table 67-1. Initial empiric treatment of tuberculosis in HIV-negative patients
Fluoroquinolones have been studied in the treatment of tuberculosis and represent alternatives for some patients with multidrug-resistant tuberculosis. Both ciprofloxacin and ofloxacin possess significant antituberculous activity. However, INH plus rifampin plus ciprofloxacin (750 mg/d) was less effective than a traditional four-drug regimen in regard to time to sputum negativity and was associated with a higher relapse rate. Problems with efficacy of fluoroquinolones may be more noted in the HIV-infected population.
Directly observed therapy (DOT), generally twice or thrice weekly, is strongly advised for potentially noncompliant persons. Under DOT, the usual adult doses of drugs are as follows:
INH: 900 mg two to three times per week
Rifampin: 600 mg two to three times per week
Ethambutol: 2.5 g two to three times per week
Pyrazinamide: 3 to 4 g two to three times per week
Streptomycin: 1.0 to 1.5 g two to three times per week
DOT has been compared with the traditional treatment of tuberculosis. This investigation documented a decrease in relapse from 21% to 5.5%, and a decrease in primary and acquired drug resistance from 13% and 14% to 7% and 2%, respectively. The use of DOT should be expanded, and it may be cost effective for all populations.
The likelihood of adverse reactions to antituberculous medications is probably higher in HIV-infected patients, so these patients should be closely monitored. Additionally, other medications taken by patients with HIV infection (e.g., ketoconazole, fluconazole) may interact with antituberculous agents. For patients infected with susceptible organisms but incapable of tolerating both INH and rifampin, current recommendations are for at least 18 months of either INH or rifampin plus both ethambutol and pyrazinamide. Similar recommendations apply to patients infected with organisms resistant to either INH or rifampin.
When multidrug-resistant tuberculosis is considered, therapy should be initiated with four to six agents, one of which should be a quinolone. Isolates of M. tuberculosis should be tested against an expanded spectrum of antituberculous agents, including ciprofloxacin, ofloxacin, amikacin, and amoxicillin-clavulanate. The length of therapy is not defined but should probably be 18 to 24 months if multidrug-resistant tuberculosis is identified.
Common Side Effects of Drugs Used in Short-course Chemotherapy
Peripheral neuropathy and central nervous system events are rarely seen. Pyridoxine (vitamin B6) deficiency, occurring especially in alcoholics, may provoke these problems. Prevention consists of daily administration of 50 to 100 mg of pyridoxine. Hepatitis is an uncommon but serious complication of INH administration. Incidence peaks at 2% to 3% for patients 50 years of age and may be enhanced by alcohol and rapid acetylation. Continued administration can lead to death. The problem is complex because transient and reversible elevations of hepatic enzymes develop in up to 20% of patients receiving INH. All persons taking INH should have a baseline measurement of hepatic function and should be informed of symptoms of hepatitis; use of INH should be discontinued if such symptoms arise. Persons over 35 years of age should be monitored regularly during therapy. Treatment should be discontinued if elevations three to five times normal are noted. Important drug interactions exist between INH and disulfiram and diphenylhydantoin; such interactions can result in psychosis or phenytoin toxicity.
Rifampin imparts an orange color to body secretions and may cause liver disease, especially in persons receiving INH. When used intermittently in doses above 600 mg/d, an influenza-like syndrome and thrombocytopenia may occur. Rifampin can induce hepatic enzymes that increase the metabolism of other drugs. Blood levels of birth control pills, warfarin, digoxin, and oral antidiabetic drugs are all decreased when rifampin is simultaneously administered. Ketoconazole, an agent often employed in patients with AIDS, may decrease rifampin absorption.
Major clinical complications of this drug are hepatitis and arthralgia. The latter may result in severe but usually self-limited involvement of small and large joints. The cause of these complications is unclear, but it may be related to the hyperuricemia commonly noted in patients taking pyrazinamide. Therapy with salicylates is usually effective. Hepatitis can be observed in 1% to 4% of patients, but the use of pyrazinamide with INH does not appear to increase the risk. Hepatitis is dose-related and occurs less commonly with intermittent than with daily therapy. As with INH, hepatic enzymes may be transiently elevated, but this does not necessarily require discontinuation of therapy. (R.B.B.)
Ad Hoc Committee of the Scientific Assembly on Microbiology, Tuberculosis, and Pulmonary Infections. Treatment of tuberculosis and tuberculous infections in adults and children. Clin Infect Dis 1995;21:9–27.
This publication (reprinted from Am J Respir Crit Care Med 1994;149:1359–1374) reviews relevant data concerning drugs used in therapy, regimens, and monitoring of patients with tuberculosis. Specific measures for multidrug-resistant tuberculosis are given, and issues related to prophylaxis are addressed. An excellent source of information for most clinicians.
Alangaden GJ, Lerner SA. The clinical use of fluoroquinolones for the treatment of mycobacterial diseases. Clin Infect Dis 1997;25:1213–1221.
The authors review the rationale and clinical investigations regarding quinolones in the treatment of mycobacterial diseases. With regard to tuberculosis, ciprofloxacin and ofloxacin have been best studied, and both possess rapid bactericidal activity. In most instances, doses of ciprofloxacin or ofloxacin were 500 to 750 mg/d or 300 to 800 mg/d, respectively. Based on available data, the quinolones should not be considered front-line agents, but they can prove valuable in selected cases (especially multidrug-resistant tuberculosis) based on susceptibility studies. They do appear to be well tolerated when given for months at a time.
Centers for Disease Control. Initial therapy for tuberculosis in the era of multidrug resistance. MMWR Morb Mortal Wkly Rep 1993;42:1–8.
These recommendations of the Advisory Council for the Elimination of Tuberculosis take into account what is known about multidrug-resistant tuberculosis and the risks associated with HIV infection. A four-drug regimen is recommended as initial empiric therapy unless it is clear that the risk of INH resistance in a community is under 4%. Initial regimens should always employ at least three drugs.
Centers for Disease Control. Tuberculosis morbidity—United States, 1997. MMWR Morb Mortal Wkly Rep 1998;47:253–257.
Updated information is provided concerning the scope of the tuberculosis problem in the United States. Since 1992, the annual number of new cases of tuberculosis has declined. However, major geographic variations exist. Currently, fewer than 2% of M. tuberculosis isolates were multidrug-resistant, and 7% to 8% were INH-resistant. The editors attribute these encouraging trends to more thorough and rapid identification of patients with tuberculosis, and better and more complete treatment.
Combs DL, O’Brien RJ, Geiter LJ. USPHS tuberculosis short-course chemotherapy trial 21: effectiveness, toxicity, and acceptability. The report of final results. Ann Intern Med 1990;112:397–406.
This investigation represents a multicenter trial comparing 6 months of INH plus rifampin (pyrazinamide for first 8 weeks) with 9 months of INH plus rifampin. Patients receiving 6 months of treatment demonstrated quicker sputum conversions, better acceptance and completion of therapy, and similar rates of relapse. This study provides compelling evidence to use INH, rifampin, and pyrazinamide as standard therapy for pansensitive strains of M. tuberculosis for 6 months.
Fischl MA, et al. An outbreak of tuberculosis caused by multiple-drug-resistant tubercle bacilli among patients with HIV infection. Ann Intern Med 1992;117:177–183.
This investigation is a case-control study of patients with tuberculosis caused by either multidrug-resistant or sensitive strains. Patients with multidrug-resistant tuberculosis were more likely to be infected with HIV, to have had contact with HIV clinics, and to have received inhaled pentamidine. Many patients had had previous contact with patients harboring multidrug-resistant strains. The study documents potential nosocomial transmission.
Goble M, et al. Treatment of 171 patients with pulmonary tuberculosis resistant to isoniazid and rifampin. N Engl J Med 1993;328:527–532.
The authors describe their extensive experience with multidrug-resistant tuberculosis between 1973 and 1983, at which time most patients were probably not infected with HIV. Thirty-five percent failed to respond to multiple-drug regimens, and more than 50% ultimately relapsed. Deaths were noted in 37% of patients.
Kennedy N, et al. Randomized controlled trial of a drug regimen that includes ciprofloxacin for the treatment of pulmonary tuberculosis. Clin Infect Dis 1996;22: 827–833.
This study, conducted in Tanzania, compared INH-rifampin-ciprofloxacin with standard four-drug treatment of pulmonary tuberculosis. The smears of all 168 patients who could be assessed were negative at 6-month follow-up, but the smears were slower to convert and the relapse rate was higher in those patients, both HIV-positive and HIV-negative, who received the quinolone-containing regimen. Also, INH-resistant strains developed in two patients in the quinolone arm of the study. It is known that several quinolones are bactericidal against M. tuberculosis, but the best dose and combination of other agents need to be defined better.
Liu S, Shilkret KL, Finelli L. Initial drug regimens for the treatment of tuberculosis. Chest 1998;113:1446–1451.
The authors reviewed the treatment regimens of 1,230 culture-positive patients residing in New Jersey to determine the adequacy of treatment based on Centers for Disease Control guidelines. Disturbingly, almost 40% of patients were initially placed on regimens with fewer than four effective drugs, despite the fact that almost all were in an area with substantial resistance. At least 6% to 7% of patients received fewer than two effective agents. However, outcome data were not presented. Treatment was much better at the hospital where tuberculosis experts were overtly available. The authors conclude that information about tuberculosis treatment needs to be disseminated better.
Saloman N, et al. Predictors and outcome of multidrug-resistant tuberculosis. Clin Infect Dis 1995;21:1245–1252.
The investigators identified 88 HIV-positive patients, 18 of whom had multidrug-resistant tuberculosis. Use of at least two agents effective in vitro resulted in enhanced 1-year survivorship and more rapid sputum conversion. The presence of multidrug-resistant tuberculosis did not predict poor outcome. This study reemphasizes the need for earlier and better identification of patients, use of multiple-drug regimens, and expanded susceptibility testing of patients at risk for multidrug resistance.
Simone PM, Iseman MD. Drug-resistant tuberculosis: a deadly and growing danger. J Respir Dis 1992;13:960–971.
This excellent review article summarizes the epidemiology and treatment of drug-resistant tuberculosis. The authors recommend continuation of treatment for at least 18 to 24 months after sputum conversion, although little data substantiate this. The article also provides short vignettes of first- and second-line antituberculous agents.
Small PM, et al. Treatment of tuberculosis in patients with advanced human immuno-deficiency virus infection. N Engl J Med 1991;324:289–294.
Retrospective study of 132 patients in San Francisco with both tuberculosis and HIV infection. Tuberculosis was found to develop both before and after HIV diagnosis and was often extrapulmonary. Standard treatment regimens included INH plus rifampin (9 months) and INH plus rifampin with the addition of either ethambutol or pyrazinamide (6 months). Patients who adhered to the regimen did well. A single treatment failure occurred in a noncompliant person with multidrug-resistant tuberculosis. This investigation documents that standard therapy is effective for most patients with HIV infection and tuberculosis.
Weis SE, et al. The effect of directly observed therapy on the rates of drug resistance and relapse in tuberculosis. N Engl J Med 1994;330:1179–1184.
The authors conducted a retrospective study of patients treated for tuberculosis during two time frames, with DOT used in the latter in 90% of cases. Almost 1,000 patients comprised the study group. Despite the fact that many patients were substance abusers and homeless, use of DOT resulted in significantly fewer relapses and fewer numbers of resistant strains of M. tuberculosis. The authors conclude that DOT is cost effective, and that its use should be expanded in most populations.
Wolinsky E. Statement of the Tuberculosis Committee of the Infectious Disease Society of America. Clin Infect Dis 1993;16: 627–628.
This commentary reviews the rationale behind using at least three agents as initial therapy of tuberculosis and provides practical guidelines for the selection of agents when multidrug-resistant tuberculosis is suspected. The committee suggests that an injectable agent (amikacin, capreomycin, or kanamycin), a quinolone (ofloxacin or ciprofloxacin), pyrazinamide, and ethambutol be used. A fifth agent such as rifabutin can be added if tolerated.