| |||||||||||||||||||||||||||||||
Rifabutin is a derivative of rifamycin S that is recommended and widely used as prophylaxis against Mycobacterium avium complex infection in patients infected with the human immunodeficiency virus (HIV) who have low CD4 lymphocyte counts.9,10 Since rifabutin was introduced in 1993, concern has been expressed about its cross-activity against M. tuberculosis and the possibility that prophylactic use of the agent might induce drug-resistant tuberculosis.11 A single case of rifampin-resistant tuberculosis has been reported in a patient taking rifabutin prophylactically.12
In this report we describe a patient with the acquired immunodeficiency syndrome (AIDS) and fully susceptible M. tuberculosis disease who was treated with directly observed therapy followed by rifabutin prophylaxis, who then had a relapse of tuberculosis caused by a rifampin-resistant strain that was otherwise indistinguishable from the initial infecting organism.
Case Report
A 35-year-old man with HIV infection was hospitalized with a two-month history of fever, night sweats, cough, dyspnea, and an 18-kg weight loss. He had a history of thrush and peripheral neuropathy, and his CD4 lymphocyte count was 7 per cubic millimeter. He was taking fluconazole and nortriptyline. A chest radiograph revealed bilateral hilar adenopathy without infiltrates. A sputum sample showed many acid-fast bacilli, and culture of the sputum sample yielded M. tuberculosis susceptible to isoniazid, rifampin, pyrazinamide, ethambutol, and streptomycin. The patient was given isoniazid, rifampin, pyrazinamide, and ethambutol daily for 16 days and then twice weekly for 6 weeks, followed by isoniazid and rifampin twice weekly for 33 doses over a 5-month period. All therapy was directly observed by a nurse.13 After one month smears and cultures of the patient's sputum remained positive for M. tuberculosis, but they became negative after two months of therapy.
After five months pneumonia with diffuse infiltrates developed that resolved with empirical treatment with trimethoprim and sulfamethoxazole; smears and cultures of the sputum were negative for mycobacteria. A chest radiograph obtained nine months after the diagnosis of tuberculosis was normal. Ten months after the diagnosis smears and cultures of a sputum sample were negative for mycobacteria. At this time the patient saw a private physician and began taking 300 mg of rifabutin daily as prophylaxis against M. avium complex infection.
Two months later — one year after the initial diagnosis of tuberculosis — fever, a productive cough, and severe dyspnea developed. A chest radiograph showed diffuse interstitial infiltrates. The patient was taking rifabutin, fluconazole, and nortriptyline. Rifabutin was stopped two days after admission. A smear of bronchoalveolar-lavage fluid was negative for acid-fast bacilli and Pneumocystis carinii. The patient was given a combination of trimethoprim and sulfamethoxazole and broad-spectrum antibiotics but died of respiratory failure after 10 days. A strain of M. tuberculosis resistant to rifampin and rifabutin but susceptible to isoniazid, pyrazinamide, ethambutol, and streptomycin subsequently grew in cultures of bronchoalveolar-lavage fluid and blood.
Methods
The susceptibility of M. tuberculosis isolates to rifampin, rifabutin, and other drugs was determined by the BACTEC radiometric culture method14 and the modified proportion method.15 Analysis involving restriction-fragment–length polymorphisms (RFLPs) was performed. Chromosomal DNA from M. tuberculosis was extracted, digested with PvuII, and hybridized by Southern blot techniques with a 245-bp portion of the IS6110 repeat element labeled with phosphorus-32.16,17 Alternatively, chromosomal DNA from M. tuberculosis was extracted, digested with AluI, and hybridized by Southern blot techniques with 32P-labeled pTBN12 containing the M. tuberculosis polymorphic guanosine–cytosine (GC)–rich repetitive sequence.18,19 A segment of the RNA polymerase 
subunit (rpoB) gene from M. tuberculosis chromosomal DNA was amplified with the polymerase chain reaction (PCR) with the use of primers that amplify a 411-bp product 20; TaqI polymerase was used in standard reaction mixes for 30 cycles. A biologic safety cabinet and aerosol tips were used to set up solitary PCRs for each isolate on separate days, and each amplification was accompanied by a control reaction in which the DNA template was omitted. After gel purification, both strands of the PCR products were sequenced according to the dye-terminator method. Amplification and sequencing of the rpoB segment were performed twice for each isolate.
Results
Analysis of the M. tuberculosis isolates from the patient's first and second episodes of tuberculosis showed that the strains were identical except in their susceptibility to rifampin and rifabutin and in their rpoB sequences. DNA fingerprinting by RFLP analysis with PvuII and IS6110 revealed an identical two-banded pattern in the two isolates (Figure 1A and Figure 1B), suggesting a common clonal origin. Because of the limited ability of RFLP analysis to differentiate between strains when the number of IS6110 bands is small, we repeated the test using AluI and the polymorphic GC-rich repetitive sequence as a probe.19 This analysis revealed identical 10-banded patterns in each isolate (Figure 1A and Figure 1B), confirming the probable clonal origin of the two isolates. Drug-susceptibility testing of the two M. tuberculosis isolates, however, revealed that the original was susceptible to all drugs tested, whereas the strain isolated during the relapse was resistant to rifampin and rifabutin.
|
To investigate whether our patient might have been reinfected with an M. tuberculosis strain prevalent in Baltimore, we reviewed our data base of RFLP patterns in 101 isolates for the past two years.24 Strains with matching RFLP patterns on analysis with PvuII and IS6110 or AluI and the polymorphic GC-rich repetitive sequence have been identified in only two other cases. Both were diagnosed within four months of our patient's first illness, and both isolates were susceptible to rifampin. No rifampin-resistant isolates with the same RFLP patterns were identified, and no other isolates of this strain were found in the period after our patient completed treatment. Despite an extensive investigation of the patient's contacts, we were unable to establish an epidemiologic link between our patient and the other patients with rifampin-susceptible M. tuberculosis with the same DNA fingerprint.
Discussion
We report a relapse of tuberculosis characterized by resistance to rifampin and rifabutin after exposure to both drugs in a patient whose infection was initially drug-susceptible. The matching RFLP patterns of the two M. tuberculosis isolates indicate that the strains were virtually identical. The small number of other cases of tuberculosis caused by this strain in the community at the time of the relapse makes it highly unlikely that the patient was reinfected with either a drug-susceptible or a drug-resistant M. tuberculosis variant of the same clone.25 Relapse of the initial tuberculosis infection with acquired resistance to rifamycin is the best explanation of the patient's second illness.
Our patient received six months of directly observed antituberculosis therapy, the regimen recommended by the American Thoracic Society and the Centers for Disease Control and Prevention (CDC).26 Although some authorities, including the CDC Advisory Committee for the Elimination of Tuberculosis, have recommended nine months of treatment for HIV-associated tuberculosis,27 six-month regimens have resulted in very low rates of relapse.28,29 Moreover, our patient took the prescribed antituberculosis drugs under direct observation, an approach that reduces the risk of relapse and the emergence of drug-resistant disease, even in patients with HIV infection.30,31 Nonetheless, a small percentage of patients with adequately treated tuberculosis will relapse, and the proportion of HIV-infected patients who relapse may be decreased by prolonging therapy.32 Our patient was severely immunocompromised at the time of his initial therapy. In these circumstances, the probability of relapse is likely to be increased. This is the only relapse among the more than 80 HIV-infected patients with tuberculosis we have treated with a six-month supervised regimen (unpublished data).
It is not clear whether the strain of tuberculosis became resistant while the patient was taking rifabutin or whether it was already resistant before prophylaxis with rifabutin was begun. Sequencing of the rpoB gene of the two M. tuberculosis isolates revealed a point mutation resulting in the substitution of a single amino acid at codon 531. The substitution of leucine for serine at this position has been noted in a large proportion of rifampin-resistant strains of M. tuberculosis.20 This mutation is also associated with high-level resistance to rifabutin in vitro.33 Relapse of rifampin-resistant tuberculosis with substitution of a single amino acid in the rpoB gene product has been reported by Nolan and coworkers in three HIV-infected patients who were not treated with rifabutin but who had received rifampin.34 They postulated that in addition to poor compliance with therapy, malabsorption of isoniazid or altered drug metabolism could have contributed to the appearance of rifampin resistance. Our patient was compliant with therapy and had no apparent gastrointestinal or hepatic disease that would alter drug absorption or metabolism.
Weltman and colleagues have reported a case of primary rifampin-resistant tuberculosis in a prison inmate taking rifabutin as prophylaxis against M. avium complex infection.12 In that case, exposure to an infectious inoculum of M. tuberculosis that was susceptible to rifampin while the patient was taking rifabutin apparently resulted in the emergence of resistance to rifampin and rifabutin. Although our patient's compliance with rifampin therapy was closely monitored, it is possible that he took rifabutin irregularly. If he began taking rifabutin only after the occurrence of symptoms, it is very likely that this would have selected for rifamycin-resistant mutants. Alternatively, rifampin resistance may have been present from the outset of the relapse, and rifabutin may therefore have been ineffective in suppressing disease.
In patients with active tuberculosis, multidrug therapy is necessary to prevent the emergence of drug resistance because of the presence of innately resistant mutants in populations of replicating M. tuberculosis. The prevalence of rifampin-resistant mutants is approximately 1 in 108 to 1 in 1010 bacteria.35 Exposure to rifampin or rifabutin alone when the number of tuberculosis organisms exceeds this level would select for resistant mutants. Conversely, the use of these agents when the mycobacterial burden is less than 108 should effectively suppress bacterial replication. In fact, monotherapy with either rifampin or rifabutin is effective in animal models of chronic tuberculosis,36,37 and rifampin prophylaxis has been efficacious in men with silicosis.38 Thus, the emergence of rifampin-resistant tuberculosis during preventive therapy with rifabutin suggests either the presence of a heavy preexisting M. tuberculosis infection or failure to achieve consistently adequate levels of rifabutin in the blood.
Despite physicians' fears, few cases of rifampin-resistant tuberculosis have been reported in patients taking rifabutin,12 and it is possible that rifabutin protects against tuberculosis as well as M. avium complex infection. Clarithromycin and azithromycin are possible alternatives to rifabutin as prophylaxis against M. avium complex infection that would not increase the risk of drug-resistant tuberculosis.39,40 Guidelines issued by the Public Health Service and the Infectious Disease Society of America41 recommend screening all patients for active tuberculosis and obtaining mycobacterial blood cultures before rifabutin prophylaxis is begun. Our patient did not undergo any such screening before his physician prescribed rifabutin. Adherence to these guidelines might have detected the relapse of tuberculosis and averted the initiation of rifabutin therapy.42 Careful follow-up of HIV-infected patients with treated tuberculosis is essential, and physicians should follow the published guidelines41 for the use of rifabutin as prophylaxis against M. avium complex disease.
Supported in part by a grant (AI36973) from the National Institute of Allergy and Infectious Diseases, a grant from the National Institute on Drug Abuse (DA 07061), and a Cooperative Agreement with the Centers for Disease Control and Prevention (U300466-10).
We are indebted to the staff of the Baltimore City Health Department Eastern Chest Clinic and the Maryland State Tuberculosis Laboratory for their assistance in the investigation of this case, and to Colleen Moore for secretarial assistance.
Source Information
From the Departments of Molecular Microbiology and Immunology (W.R.B., C.P.) and Epidemiology (N.M.H.G., R.E.C.), Johns Hopkins University School of Hygiene and Public Health; the Departments of Medicine (W.R.B., R.E.C.) and Pathology (S.H.), Johns Hopkins University School of Medicine; the Baltimore City Health Department (K.M.-R.); and the Maryland Department of Health and Mental Hygiene (N.H.) — all in Baltimore.
Address reprint requests to Dr. Chaisson at Johns Hopkins University School of Medicine, Carnegie 292, 600 N. Wolfe St., Baltimore, MD 21287-6220.
References
gene that lead to rifampicin resistance. J Mol Biol 1988;202:45-58. [CrossRef][Medline]
gene of Mycobacterium tuberculosis. Antimicrob Agents Chemother 1994;38:805-811. -subunit of RNA polymerase influence in-vitro activity of rifamycins in rifampicin-resistant Mycobacterium tuberculosis. J Antimicrob Chemother 1995;35:345-348.
| |||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
