Examining Initial Treatment Regimens for Tuberculosis Cases in New York City

BACKGROUND: Initiating appropriate therapy for tuberculosis (TB) is important to rapidly kill Mycobacterium tuberculosis, stop transmission, and prevent acquired antibiotic resistance. The Centers for Disease Control and Prevention (CDC) recommends initial therapy using rifampin, isoniazid, pyrazinamide, and ethambutol, referred to as RIPE. CDC’s national target for initiating patients on RIPE is 93.4% by 2015. In this study, we assessed New York City (NYC) Department of Health and Mental Hygiene’s ability to meet the national target.

METHODS:   Data was obtained from the NYC TB case registry for all cases reported in 2011. Patients alive at diagnosis and who initially received at least one drug for TB were included in the analysis. Patients who received all four recommended drugs for at least two weeks were considered to have started RIPE, while those who received an alternative drug regimen were considered to have not started RIPE. Demographic, social, and clinical characteristics of each group were compared using Pearson’s chi-square or Fisher’s exact test. Reasons for starting alternative drug regimens were also examined.

RESULTS:   There were 688 cases of tuberculosis in NYC in 2011. Of these, 655 patients started therapy, including 564 (86%) with a RIPE-containing regimen. Compared to patients who started RIPE, the 91 non-RIPE patients were more likely to be infected with human immunodeficiency virus (HIV) (28% vs. 7%, p<0.001), infected with a TB strain resistant to at least one RIPE-drug (30% vs. 17%, p=0.029), and to have a previous history of TB disease (13% vs. 4%, p=0.001). No differences were found when examining acid-fast bacilli smear positivity, site of disease, provider type, age, sex, and US-birth. Reasons for not starting RIPE included: 1) history of, or concerns about future side effects from at least one of the RIPE-drugs (n=44, 48%); 2) drug interactions with rifampin in HIV-infected patients on anti-retroviral therapy (n=19, 21%); 3) pregnancy, which precludes the use of pyrazinamide (n=11, 12%); 4) availability of drug susceptibility results (n=10, 11%); 5) physician’s discretion (n=5, 6%); and 6) patients lost to follow-up (n=2, 2%).

CONCLUSIONS:   There are clinical reasons for which RIPE may not be the most appropriate initial therapy for TB patients. Further investigation is needed to examine whether patients with complex clinical conditions will preclude NYC from reaching the national target for initiating RIPE.