National survey to measure rates of liver injury, hospitalization, and death associated with rifampin and pyrazinamide for latent tuberculosis infection

Abstract

Background. Cases of severe and fatal liver injury were reported after a 2-month course of rifampin-pyrazinamide therapy was recommended in 2000 as an alternative to isoniazid for treatment of latent tuberculosis infection. We estimated rates of rifampin-pyrazinamide-associated liver injury and compared these with historical rates for isoniazid. Methods. We conducted a survey of state and city tuberculosis programs and other health care settings in the United States where rifampin-pyrazinamide was prescribed. The number of rifampin-pyrazinamide therapy initiations was collected, as well as the number of occurrences of (1) asymptomatic aspartate aminotransferase serum concentration >5 times the upper limit of normal, (2) symptomatic hepatitis (in which the patient was not hospitalized), (3) hospitalization for liver injury, (4) death with liver injury, and (5) treatment completion. We also searched a national pharmacy claims database (Verispan). Rates of these events were calculated. Results. Among 139 programs, 110 (79%) responded; 87 (79%) had initiated rifampin-pyrazinamide therapy for a total of 8087 patients between January 2000 and June 2002. Rates per 1000 rifampin-pyrazinamide therapy initiations during this period were 25.6 (95% confidence interval [CI], 22.3-29.3) for asymptomatic aspartate aminotransferase level >5 times the upper limit of normal and 18.7 (95% CI, 15.9-21.9) for hepatitis. Seven fatalities and 23 hospitalizations occurred, with rates of 0.9 (95% CI, 0.4-1.9) and 2.8 (95% CI, 1.8-4.3) per 1000 rifampin-pyrazinamide therapy initiations, respectively. Of 8087 patients, 64% completed rifampin-pyrazinamide therapy. The Verispan search revealed 1 rifampin-pyrazinamide-associated hospitalization (2.9 hospitalizations per 1000 rifampin-pyrazinamide therapy initiations; 95% CI, 0.1-18.4) and no deaths. Articles on the use of isoniazid therapy for latent tuberculosis infection that were published after 1990 reported fatality rates of 0.0-0.3 deaths per 1000 persons. Conclusions. Rates of liver injury, hospitalization, and death associated with rifampin-pyrazinamide therapy exceed rates reported for isoniazid therapy. Because earlier randomized trials of rifampin-pyrazinamide lacked adequate statistical power to detect fatal events, the Centers for Disease Control and Prevention recommends that rifampin-pyrazinamide generally should not be used for treatment of latent tuberculosis infection. © 2005 by the Infectious Diseases Society of America. All rights reserved.