Skip Navigation LinksSkip Navigation Links
Centers for Disease Control and Prevention
Safer Healthier People
Blue White
Blue White
bottom curve
CDC Home Search Health Topics A-Z spacer spacer
Blue curve MMWR spacer

Meeting the Challenge of Multidrug-Resistant Tuberculosis: Summary of a Conference

This report was prepared by the following CDC staff:

Alan R. Hinman, M.D., M.P.H. James M. Hughes, M.D. Dixie E. Snider, Jr., M.D., M.P.H. Mitchell L. Cohen, M.D.


On January 22-23, 1992, a conference on "Meeting the Challenge of Multidrug-Resistant Tuberculosis" (MDR-TB) was held at the Centers for Disease Control with >400 attendees. Plenary presentations summarized the nature and magnitude of the problem, and work groups were formed to address six issues: public health, epidemiology and surveillance, infection control and environment, laboratory diagnosis and research, therapy, and training and education. Participants concluded that existing tools for diagnosis, patient management, and infection control must be more thoroughly implemented and that research to develop new approaches to diagnosis and therapy is critically needed. Discussions at the conference provided information that has been used to develop the National Action Plan to Combat Multidrug-Resistant Tuberculosis.


After decades of decline, the incidence of tuberculosis (TB) in the United States has increased in recent years. Furthermore, a number of TB outbreaks due to multidrug-resistant strains of Mycobacterium tuberculosis have been reported. Therefore, on January 22-23, 1992, a conference on "Meeting the Challenge of Multidrug-Resistant Tubercu- losis" (MDR-TB) was held at CDC. The conference was attended by >400 persons from 47 state health departments, CDC, other federal agencies, and many professional and voluntary organizations. After plenary sessions presented the nature and magnitude of the problem, attendees divided into six work groups to address the following issues: public health, epidemiology and surveillance, infection control and environ- ment, laboratory diagnosis and research, therapy, and training and education.


In recent years, TB had become confined to definable population groups, such as disadvantaged populations, immigrants from countries with a high prevalence of TB, the elderly, substance abusers, persons in correctional facilities and nursing homes, and the homeless. After years of steady decline, the reported incidence of TB began to plateau in 1985 and then began to increase in 1989. Compared with the previous downward trend, an estimated 28,000 "excess" cases of TB occurred during the period 1985-1990. Most of the increase is believed to be a result of TB among persons infected with human immunodeficiency virus (HIV), although other factors, such as immigration and transmission of TB in group or institutional settings, have also contributed to the increase.

The Strategic Plan for the Elimination of Tuberculosis in the United States (1), developed by a CDC advisory committee, calls for full implementation of existing technologies and the development and subsequent implementation of new technologies. The elimination plan has not yet been fully implemented. Because failure to complete a curative course of therapy is a major cause of the emergence of drug- resistant organisms, the plan recommends directly observed therapy (DOT) to ensure patient compliance. At present, approximately 15% of TB patients in the United States receive DOT, which greatly increases the likelihood of cure. Nonetheless, approximately 20% of TB patients who begin a course of therapy do not complete it within 12 months.

TB Infection and HIV

Persons who have TB infection and subsequently acquire HIV infection are more likely to progress to TB disease than those who are HIV negative. Furthermore, persons who are HIV infected and subsequently are exposed to TB are more likely to progress from infection to disease than those who are HIV negative -- and they are likely to have a rapid, severe progression of illness. Approximately 10% of the 1 million persons in the United States infected with HIV are also infected with TB. The prevalence of HIV infection among patients with TB disease varies around the country but may exceed 40% in some areas.

Recently, outbreaks of MDR-TB have occurred in health-care settings in Miami and New York City and in correctional institutions and a hospital in upstate New York. Transmission of MDR-TB has occurred among patients with acquired immunodeficiency syndrome (AIDS) in at least six hospitals (in each of which lapses in implementation of infection control procedures were noted), from AIDS patients to health-care workers (HCWs), and from AIDS patients to other patients and family members. Rates of skin-test conversions among HCWs in two hospitals were 33% and 50%. At least eight HCWs have developed TB disease; four (all HIV positive) have died. Case-fatality rates among AIDS patients in these outbreaks have ranged from 72% to 89%, and the median intervals between diagnosis and death have been 4-16 weeks. Some patients with rapidly progressive disease have died before laboratory results were available, indicating that they had MDR-TB. Some of the organisms were resistant to as many as seven antituber- culosis drugs (isoniazid {INH}, rifampin {RIF}, kanamycin, ethambutol, ethionamide, streptomycin {SM}, and rifabutin).

Related to the outbreaks in New York City hospitals, an outbreak of MDR-TB in the New York State correctional system involved prisoners in several different facilities, as well as transmission to at least 50 HCWs in an upstate New York hospital. A prison employee who guarded prisoners in the hospital died from MDR-TB. He was not HIV infected but was receiving radiation therapy for laryngeal cancer.

Role of Laboratory Diagnosis of TB

The critical role of the laboratory in the diagnosis of active TB was described. Isolation and identification of M. tuberculosis by commonly used methods require approximately 30 days, and drug susceptibility testing of the isolate requires an additional 2-3 weeks. This extremely slow growth rate, other unique properties of the organism, and the need for special laboratory containment facilities have complicated laboratory diagnosis. More rapid and reliable methods are available but not widely used.

Although there is currently no ongoing nationwide surveillance of the antibiotic susceptibility of M. tuberculosis isolates, a CDC survey of isolates from the first quarter of 1991 indicated that approximately 60% of the isolates had been tested. Of new cases, >8% were resistant to at least one drug; 3.1% were resistant to both INH and RIF. Considerable variation existed in the frequency of resis- tance: only 11 states reported INH/RIF-resistant isolates, with approximately 40% of isolates from New York City having this pattern, followed by New Jersey with 6%.

The problems associated with treatment of cases caused by MDR-TB are evident. Although resistance data on isolates can allow therapy to be individually tailored to patients, this information may not be reported to physicians until weeks or months after the specimen initially was obtained. Chemoprophylaxis is another problem. Most persons infected with TB have only a 10% lifetime risk of developing disease, but for HIV-infected persons the risk may be as high as 10% per year. No data exist on the effectiveness of regimens other than INH in preventing the progression from infection to disease.

Ensuring a continuing supply of the full range of antituberculosis drugs has been a problem, with interruptions in the supplies of INH, SM, para-aminosalicylic acid (PAS), and other drugs during the past year. Although there is currently no U.S. supplier of SM and PAS, CDC and the Food and Drug Administration are making these drugs available to individual patients under an Investigational New Drug Application, and the drugs should again be commercially available during 1992.

Relatively little research has been carried out on TB during the past 20-30 years in the United States, in part because of the success of TB control efforts. Consequently, many of the newer molecular approaches and microbiologic techniques have not been applied to TB. In the past few years, research interest has been rekindled, with both the National Institutes of Health and CDC devoting increased resources to TB research. Nonetheless, a major need exists for increased research in the biology and pathophysiology of M. tuberculosis, drug resistance, diagnosis, subtyping methods, therapy (both preventive and curative) for host immune factors, behavioral factors, infection control, and vaccines.


Public Health Operational Issues

Many participants thought that the increase in TB is, in part, a manifestation of a broad decline in public health activities nation- wide. Most believed that it was essential to integrate TB program activities into institutional and other health-care settings, with the first priority being to provide preventive and curative services for TB in correctional facilities. Some participants called for development of guidelines for an "ideal" TB program that would include a description of minimum acceptable program standards as a means of helping to secure funding for a minimum level of program activities and of achieving a degree of uniformity across the country. Cost- effectiveness studies of a range of program activities were suggested as a way to prioritize these efforts. Recommendations from advisory bodies should address both long-term "ideal" approaches and short-term practical approaches to be undertaken at the state and local levels.

Case management, a minimum standard for all TB control programs, should incorporate both public health management and the provision of clinical care from whatever source. DOT, which should be more widely applied, was considered an important part of case management. DOT potentially could be provided in sites such as drug treatment centers and shelters that do not currently provide TB services. Although many opportunities exist for integration of programs (e.g., with HIV counseling and testing, community and migrant health centers, community-based organizations, private hospitals and clinics, shelters for the homeless, drug treatment centers, parole programs), most participants agreed that effective identification of TB infection and disease, preventive therapy, and curative therapy for prisoners and detainees should be the first priority; these measures would be very cost-effective. Concerns were also expressed about rates of TB among immigrants (both legal and illegal) and migrant workers.

Updating state laws and regulations to address the personal liberty and public safety concerns of the 1990s was seen as an immediate need. Issues of confidentiality, quarantine standards, laboratory standards, TB control in institutions, protection of HCWs, and reporting by health-care providers and laboratories should be addressed. In managing "noncompliant" patients, a continuum of approaches should be used, with the least restrictive approaches tried before involuntary detention is used. Optimal case management requires clarification of confidentiality issues and removal of barriers to information-sharing between programs such as drug treatment centers, health departments, Veterans Administration hospitals, and correc- tional institutions. Related to information-sharing is the issue of who should be included in a TB registry (e.g., cases, suspected cases, persons on preventive therapy).

Several problems with the ability of the health-care system to deal with MDR-TB were identified, including deficiencies in the physical plants of treatment sites with respect to ventilation and isolation capabilities. Several persons suggested that regional centers of excellence be established to provide appropriate isolation as well as the most skilled clinical management. Concerns were expressed about appropriate methods to protect HCWs from MDR-TB, as well as appropriate chemoprophylaxis for persons infected with MDR-TB.

Epidemiology and Surveillance Issues

Systematic local, state, and national surveillance of M. tuberculosis isolates for drug susceptibility could

  • identify where drug resistance is already a problem;

  • monitor the epidemiology of drug resistance as to place, person, and time;

  • detect outbreaks of MDR-TB; and

  • establish a bank of drug-resistant isolates for further study. Systematic local, state, and national surveillance of the HIV

infection status of each patient with TB would

  • permit adequate case management of HIV-infected TB cases;

  • enable priorities to be set for contact investigation (if necessary); and

  • monitor the extent to which HIV infection is contributing to TB and MDR-TB.

Additional information on individual TB patients that should be included in the surveillance system includes institutional experience (e.g., hospital, prison, shelter), occupation, initial therapy regimen, insurance/Medicaid status, and behavioral factors that might predict poor compliance with therapy. Information from the surveil- lance system should enable identification of geographic areas where drug-resistant strains are more likely to emerge as a result of problems with the appropriateness and quality of TB therapy and compliance. Local registries should enable estimation of the number of infections with MDR-TB (not just cases) and should also provide information about the extent of coinfection with HIV and TB.

An important new focus for surveillance is new TB infections among hospital workers, prisoners, and correctional employees, so that indices of the risk of TB transmission within hospitals and correc- tional institutions can be developed and current and future methods of infection control can be evaluated.

Development of the surveillance systems described above will require assessment of current surveillance systems, development of practical software for local analysis of epidemiologic and program- matic data and transmission to CDC, and improved cooperation and communication between TB and HIV programs at local and state levels. A national TB registry would enable overall assessment of MDR-TB and facilitate interstate communication for case-management purposes.

Infection Control and Environmental Issues

Given the circumstances of recent MDR-TB outbreaks -- in hospital and correctional institutions -- development and dissemination of detailed guidelines for reducing the risk of transmission in these settings is of highest priority. Simple, logical procedures should be instituted immediately, especially in facilities now having problems with TB transmission. A survey of current isolation capabilities in hospitals was thought essential.

Data are urgently needed to assess the efficacy of the various isolation procedures currently recommended in facilities. The effec- tiveness and relative importance of ventilation, ultraviolet lights, particulate respirators, and isolation booths must be determined. In the absence of definitive data, "best judgment" recommendations should be developed, perhaps with assessment of the category of proof (strength of evidence) of efficacy, as in the current CDC guidelines for infection control and isolation (2).

Screening and monitoring for TB infection among HCWs are essential. Such programs should include trainees as well as employees and other staff who have contact with patients.

To reduce the risk of nosocomial/institutional transmission, physicians must be educated to have a high index of suspicion for TB and to be aware of atypical patterns of the disease among HIV-infected persons. Rapid laboratory support is critical -- acid-fast bacillus (AFB) smear results must be available within 24 hours. Physicians must also be educated about the need for early and appropriate therapy.

Laboratory Issues

Definitive diagnosis of TB requires that M. tuberculosis be isolated, and current technology for determining drug susceptibility also requires an isolate. To contain current MDR-TB outbreaks and prevent others, laboratory procedures must be improved. Clinical specimens should reach the laboratory within 24 hours of collection; smear reports should be in the hands of physicians within 24 hours of specimen receipt by the laboratory; positive cultures should be identified within 14 days of receipt of specimen; isolates should be definitively identified as to species within 17-21 days of receipt of specimen; and antibiotic susceptibility results should be in the hands of physicians within 28 days of receipt of specimen.

To achieve these goals, many changes and improvements are required in clinical/public health laboratories, including

  • widespread use of the most specific, sensitive, and rapid methods available (e.g., fluorochrome microscopy; rapid culture techniques, such as radiometric methods, biphasic methods; microcolony morphology);

  • rapid methods for identification, such as DNA probes and high- performance liquid chromatography; and

  • use of a primary susceptibility test panel of five drugs -- INH, RIF, pyrazinamide, ethambutol, and SM.

New equipment, training courses, and information systems in labora- tories are needed to effect these improvements.

Research is a critical element of this issue. Increased knowledge of the basic genetics and biology of M. tuberculosis is needed so that the pathogenic process of TB can be understood and more targeted diagnostic assays, drugs, and vaccines can be developed. There is also a tremendous need for more information on the host immune response to M. tuberculosis; new data could profoundly affect the diagnosis and treatment of persons with TB. The immediate need is to ensure the ability of laboratories to use the best existing laboratory methods and procedures to achieve rapid turnaround time, with priority placed on areas where MDR-TB outbreaks are occurring or are likely to occur.

Drug and Therapy Issues

To ensure the availability of adequate supplies of currently manufactured antituberculosis drugs, a system must be developed to monitor drug production, supplies, sources, and prices and to dissem- inate the information to state, local, and private agencies. More information is also needed about the behavioral factors that affect patient compliance with therapy.

Beyond these immediate needs, new antituberculosis drugs are required and their development should begin immediately. The following measures will be entailed:

  • screening existing drugs for antimycobacterial activity in vitro and in animal models;

  • developing new drug delivery systems, such those involving the transdermal or subcutaneous routes;

  • improving basic understanding of mycobacterial biology; and

  • encouraging drug companies to pursue the development and marketing of antituberculosis drugs (through the Orphan Drug Act or other mechanisms).

Training and Education Issues

High-risk populations who should be targeted to receive education include persons working in hospitals, drug treatment centers, homeless shelters, HIV clinics, and correctional and other institutions with close living quarters; their clients; and refugees and immigrants. Emphasis should be placed on visual presentation as well as on material that can be taken home. Educational materials should be written in simple language that is culturally sensitive and linguistically appropriate.

Professional education of those involved in the prevention, control, diagnosis, and treatment of TB is also important. Both governmental policymakers and the general public need to understand TB as a disease, its threat to the general population, and the prospects for its control and elimination.


The discussions at the conference provided information that has been used by a federal Tuberculosis Task Force to develop the National Action Plan to Combat Multidrug-Resistant Tuberculosis. However, as pointed out in a recent editorial, "This is not just a governmental issue. Preventing the reemergence of tuberculosis requires the efforts of the entire medical community, acting in concert with the public" (3).


This conference was convened on short notice and planned by a committee chaired by Walter W. Ihle, Jr., and Barbara A. Meek, to whom the authors are deeply grateful.


  1. Advisory Committee for the Elimination of Tuberculosis. A strategic plan for the elimination of tuberculosis in the United States. MMWR 1989;38(Suppl. No. S-3):{inclusive page numbers}.

  2. CDC. Guidelines for infection control in hospital personnel. Infect Control 1983;4(Suppl):328-49.

  3. Snider DE Jr., Roper WL. The new tuberculosis. N Engl J Med 1992;326:703-5.

Disclaimer   All MMWR HTML documents published before January 1993 are electronic conversions from ASCII text into HTML. This conversion may have resulted in character translation or format errors in the HTML version. Users should not rely on this HTML document, but are referred to the original MMWR paper copy for the official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.

**Questions or messages regarding errors in formatting should be addressed to

Page converted: 08/05/98


Safer, Healthier People

Morbidity and Mortality Weekly Report
Centers for Disease Control and Prevention
1600 Clifton Rd, MailStop E-90, Atlanta, GA 30333, U.S.A


Department of Health
and Human Services

This page last reviewed 5/2/01