Image of CDC Logo Image containing title: Tuberculosis Information CD-ROM   Image of people
jump over main navigation bar to content area
TB Guidelines
Surveillance Reports
Slide Sets
TB-Related MMWRs and Reports
Education/Training Materials
Ordering Information

U.S. Department of Health and Human Services logo

Publications > Multidrug-Resistant Tuberculosis

Multidrug-Resistant Tuberculosis, 1994

Patricia M. Simone, MD, and Samuel W. Dooley, MD

arrow right Mulitdrug-Resistant Tuberculosis PDF

Drug-resistant tuberculosis is not a new phenomenon: resistance to antituberculosis drugs has been noted since the drugs were first introduced, and occasionally outbreaks of drug-resistant tuberculosis have been reported. But recent outbreaks of multidrug-resistant tuberculosis have differed considerably from the previous outbreaks. This article will review the causes, the treatment, and the control of drug-resistant tuberculosis and describe the recent outbreaks of multidrug-resistant tuberculosis.


Tuberculosis has afflicted humans since ancient times. Tuberculous lesions have been found in Egyptian mummies, and the disease was well described by Hippocrates. During the Industrial Revolution, substantial increases in tuberculosis accompanied urban crowding and malnutrition. In the 19th century, an estimated one quarter of the adult population of Europe died of tuberculosis.

In the United States, approximately 20% of deaths in the early 1800s were attributed to tuberculosis; this figure had decreased to 11% by midcentury. Mortality declined throughout the century probably because of improvements in living conditions and nutrition. The realization in the late 1800s that tuberculosis was infectious eventually led to the isolation of tuberculosis patients in treatment centers, or sanatoria. Treatment in sanatoria consisted of bed rest, enhanced nutrition, and various surgical procedures aimed at closing lung cavities. The discovery of antituberculosis medications in the mid–20th century further reduced tuberculosis incidence and mortality and rendered surgical resection unnecessary.

Soon after antituberculosis medications were introduced, researchers recognized that tuberculosis was unique with respect to the frequency and importance of the emergence of drug resistance during therapy. They discovered that drug-resistant mutants existed in wild strains of Mycobacterium tuberculosis, that selection of these mutants could occur under certain treatment conditions, and that drug-resistant tuberculosis was more difficult to cure. Once additional drugs were available, controlled trials were conducted and demonstrated that combined regimens were more efficient than single-drug regimens in treating tuberculosis and preventing the emergence of drug resistance.

Transmission and Pathogenesis of Drug-Resistant Tuberculosis

Drug-resistant and drug-susceptible tuberculosis are transmitted in the same way. For many years, drug-resistant tuberculosis was believed to be less infectious than drug-susceptible tuberculosis. This belief was largely based on animal studies that showed that isoniazid-resistant bacilli were less virulent than isoniazid-susceptible bacilli. Reports of outbreaks of drug-resistant tuberculosis in the 1970s and 1980s did not completely dispel this notion. In 1985 Snider and colleagues compared the risk of infection among persons exposed to drug-resistant bacilli with the risk among persons exposed to drug-susceptible bacilli (1). They found no evidence that drug-resistant bacilli were less infectious than drug-susceptible bacilli. In fact, contacts of previously untreated patients had a similar risk of infection, regardless of whether the bacilli were drug-resistant or drug-susceptible. They did find, however, an increased risk of infection in contacts of patients with drug-resistant tuberculosis who had been previously treated. They suggested that the increased risk resulted from prolonged exposure rather than increased infectiousness of the drug-resistant bacilli. Patients with drug-resistant tuberculosis who have a history of prior treatment are more likely to have been nonadherent to therapy and infectious for longer periods of time than patients with drug-susceptible tuberculosis. The recent outbreaks of multidrug-resistant tuberculosis support these findings that drug-resistant tuberculosis is no less infectious than drug-susceptible tuberculosis and that, in fact, prolonged periods of infectiousness may facilitate transmission.

Drug-resistant tuberculosis occurs when drug-resistant bacilli outgrow drug-susceptible bacilli. The drug-resistant organisms are produced by random mutations in the bacterial chromosome which occur spontaneously in wild-type strains even before the strains come in contact with an antituberculosis drug (2). Mutations can produce bacilli resistant to any of the antituberculosis drugs, although they occur more frequently for some drugs than others. The average mutation rate in M. tuberculosis for resistance to isoniazid is 2.56 x 10-8 mutations per bacterium per generation; for rifampin, 2.25 x 10-10; for ethambutol, 1.0 x 10-7; and for streptomycin, 2.95 x 10-8. The mutation rate for resistance to more than one drug is calculated by multiplying the rates for the individual drugs. For example, the mutation rate for resistance to both isoniazid and rifampin is approximately 2.56 x 10-8 times 2.25 x 10-10, or 5.76 x 10-18. The expected ratio of resistant bacilli to susceptible bacilli in an unselected population of M. tuberculosis is about 1:106 each for isoniazid and streptomycin and 1:108 for rifampin. Mutants resistant to both isoniazid and rifampin should occur less than once in a population of 1014 bacilli. Pulmonary cavities contain about 107 to 109 bacilli; thus, they are likely to contain a small number of bacilli resistant to each of the antituberculosis drugs but unlikely to contain bacilli resistant to two drugs simultaneously (3).

Drug-resistant mutants are selected when therapy is inadequate, for example, when a single drug is used to treat a large population of bacilli. The treatment of a wild-type strain of M. tuberculosis with a single drug kills the majority of the bacilli in the population, but the small number of mutants resistant to the drug continue to multiply (Figure 1). After 2 weeks to several months of treatment with the single drug, the resistant bacilli will outgrow the susceptible bacilli, causing clinical drug resistance. Furthermore, in a large population of resistant mutants, additional mutations can occur resulting in doubly-resistant mutants (Figure 2).

Graphic showing the Pathogensis of Drug Resistance I

Figures 1 and 2. Empty circles represent tubercle bacilli susceptible to all antituberculosis drugs.  A circle with a letter in it represents a tubercle bacillus with resistance to the drug indicated by the letter: I = isoniazid resistance; R = rifampin resistance; P = pyrazinamide resistance.  Adapted with permission from Reichman LB: A looming public health nightmare. Lungs at Work 1992. Copyright 1992 American Lung Association.

Graphic showing the Pathogensis of Drug Resistance II

In contrast, the emergence of drug resistance is far less likely in most types of extrapulmonary tuberculosis, in which the bacillary population is much smaller. The bacillary population is even smaller in latent tuberculosis infection, so the chances that drug resistance will emerge during preventive therapy are negligible.

Drug resistance is divided into two types: primary resistance and secondary (or acquired) resistance. Primary resistance occurs in persons who do not have a history of previous treatment; these persons are initially infected with resistant organisms. Secondary resistance occurs during therapy for tuberculosis, either because the patient was treated with an inadequate regimen or because the patient did not take the prescribed regimen appropriately. Treatment with two or more drugs in combination can prevent the emergence of drug resistance, if the regimen includes at least two drugs to which the organisms demonstrate in vitro susceptibility.

Multidrug resistance is defined as in vitro resistance of a strain of M. tuberculosis to two or more of the antituberculosis drugs. Clinically, the most important pattern of multidrug resistance is resistance to both isoniazid and rifampin. This combination has been associated with an overall cure rate of less than 60% in some reports, similar to the rate of cure before the discovery of antituberculosis drugs (4,5).

The two main causes of drug resistance are nonadherence to therapy and the use of inadequate treatment regimens. When medications are not taken as prescribed, the infecting bacilli may be exposed to a single drug for long periods of time, which allows drug-resistant organisms to emerge (Figure 3).

 Figure outlining the Emergence of Resistance for Nonadherence and Inappropriate Therapy

In addition, some regimens may contain multiple drugs but only one drug to which the infecting bacilli are susceptible. This can happen when primary drug resistance is not suspected or when a single drug is added to a failing regimen (Figure 4). These regimens are equivalent to single-drug therapy, and they can select multidrug-resistant organisms. Acquired multidrug resistance usually results from a combination of nonadherence and inappropriate therapy.

 Figure outlining the Emergence of Resistence for Inappropriate Therapy        


Tuberculosis in the United States
Since nationwide reporting first began, the number of reported tuberculosis cases declined by an average of 5.6% per year, from more than 84,000 cases in 1953 to 22,255 cases in 1984. In 1985 this decline ended abruptly; from 1985 through 1993, the number of cases increased by 14%. It is estimated that 64,000 more cases have occurred since 1985 than would have been expected had the trend of 1980 to 1984 continued (Figure 5) (6). The excess cases have been attributed to a number of factors including the human immunodeficiency virus (HIV) epidemic, tuberculosis occurring in foreign-born persons from countries with a high prevalence of tuberculosis, the transmission of tuberculosis in congregate settings (e.g., health care facilities, correctional facilities, drug treatment facilities, and shelters for the homeless), and a deterioration of the public health care infrastructure.

Graph showing the Expected and Observed TB Cases, United States, 1980-1993

Drug-resistant tuberculosis in the United States
Until recently, the national surveillance system for tuberculosis has not included the reporting of drug susceptibility. Therefore, information about trends in drug-resistant tuberculosis is limited. Several regional surveys and some larger national surveys provide some information, but methodological differences make comparing the surveys difficult. The Centers for Disease Control and Prevention (CDC) has conducted several surveys of primary drug resistance. The first survey examined drug susceptibility results among tuberculosis patients hospitalized between 1961 and 1968. Resistance to at least one drug was found in 3.5% of the 9350 strains tested, and resistance to two or more drugs was found in 1% of the strains tested (7,8).

In a second survey of city and state laboratories that was conducted between 1975 and 1982, 6.9% of the cases involved resistance to one or more drugs, and 2.3% of cases involved resistance to two or more drugs (9-11). The third survey, conducted between 1982 and 1986, found that 9% of strains isolated in 31 health department laboratories were resistant to one or more drugs (12).

The difference in primary resistance rates in the three surveys may be due to methodological differences; however, each survey showed a rate of primary drug resistance that was relatively low and that was stable or decreasing during the study period. Drug resistance did not appear to be an increasing problem, and the surveys were discontinued.

Prompted by outbreaks of multidrug-resistant tuberculosis, CDC conducted an additional survey of drug resistance among tuberculosis cases reported from January through March 1991 (13). CDC then conducted a second survey of cases reported in the first quarter of 1992, and compared the findings from that survey with the findings from the 1991 survey (CDC, unpublished data, 1995). The proportion of culture-positive patients with drug susceptibility test results increased from 81.8% in 1991 to 84.1% in 1992. In 1991 resistance to at least one drug was found in 13.4% of new cases, 26.6% of recurrent cases, and 14.2% of cases overall; in comparison, in 1992 resistance to at least one drug was found in 12.8% of new cases, 19.4% of recurrent cases, and 13.1% of all cases. Resistance to both isoniazid and rifampin was reported in 3.2% of new cases, 6.9% of recurrent cases, and 3.5% of all cases in 1991, and was reported in 3.4% of new cases, 3.7% of recurrent cases, and 3.3% of all cases in 1992.

The findings of these two surveys may not be comparable to the findings of the previous surveys because of significant methodological differences. In the previous surveys, susceptibility testing was performed at a single laboratory on isolates from only a sample of areas. In contrast, in the 1991 and 1992 surveys, susceptibility tests were performed at local laboratories, and researchers attempted to collect results from all areas for all of the cases reported during the study period. Furthermore, in the previous surveys, attempts were made to determine whether "new" cases had a history of previous treatment and if so, to exclude them from the study, whereas in the 1991 and 1992 surveys, classification of cases as "new" or "recurrent" was not verified, and a history of prior treatment was not sought. Although the full effect of these methodological differences cannot be determined, the most recent CDC survey suggests that drug resistance has increased in the United States after years of relative stability.

Some regional data suggest that the increases in drug-resistant tuberculosis may be concentrated in large urban areas. In the 1991 CDC survey, 67% of the cases of drug resistance were reported from only five areas: New York City, California, Texas, New Jersey, and Florida. In both the 1991 and the 1992 survey, 61% of the cases of multidrug-resistant tuberculosis were reported from New York City alone. Frieden and colleagues conducted a survey in New York City of drug susceptibility results on specimens from all patients with a positive culture for M. tuberculosis during April 1991 (14). They found that of patients with no previous treatment, 23% had organisms resistant to at least one drug and 7% had organisms resistant to both isoniazid and rifampin. Beginning in 1993, results of drug susceptibility tests were reported to CDC from all 50 state health departments and from New York City, the District of Columbia, and Puerto Rico through SURVS-TB, CDC’s national TB surveillance system. In an analysis of areas reporting susceptibility results for at least 75% of culture-positive cases, New York City accounted for 80% of the cases of multidrug-resistant tuberculosis (CDC, unpublished data, 1995).

Factors associated with drug resistance
One important risk factor for drug resistance is previous treatment with antituberculosis medications (Table 1). Studies have found that rates of drug-resistance increase as the duration of previous treatment increases (14,15). In most instances, drug resistance develops because of inadequate or erratic therapy, although it has been shown that persons previously treated for drug-susceptible tuberculosis can be reinfected with drug-resistant strains (16-18).

Table showing the Groups at Increased Risk for Drug-Resistant Tuberculosis

Another risk factor for drug-resistant tuberculosis is contact with a person who has infectious, drug-resistant tuberculosis. Recent nosocomial outbreaks demonstrate a strong correlation between previous exposure to a patient who has infectious, multidrug-resistant tuberculosis and the subsequent development in the contact of multidrug-resistant tuberculosis (19-23). Drug resistance occurs more frequently in persons from areas of the world with a high prevalence of drug-resistant tuberculosis, such as Southeast Asia, Latin America, Haiti and the Phillippines. Several studies have found high rates of primary drug resistance in these persons, indicating transmission of drug resistant tuberculosis in the country of origin (24-27). However, because an accurate history of treatment may be difficult to obtain, some previously treated patients may be misclassified as having primary resistance. In the CDC survey conducted from 1982 to 1986, the rates of both primary and acquired drug resistance were two times higher among foreign-born persons than among persons born in the United States (12).

Outbreaks of Drug-Resistant Tuberculosis
Drug-resistant tuberculosis was described soon after antituberculosis drugs were introduced, but the first documented outbreak of drug-resistant tuberculosis was not reported until 1970. Between 1970 and 1990, only a few outbreaks of isoniazid-resistant tuberculosis and five outbreaks of multidrug-resistant tuberculosis were reported. In general, these outbreaks involved small numbers of cases among close contacts who had prolonged or repeated exposure to source patients (16,17,28-33).

From 1990 through August 1992, in collaboration with state and local health departments, CDC investigated outbreaks of multidrug-resistant tuberculosis in seven hospitals in Florida, New York, and New Jersey and in the New York State correctional system (Tables 2 and 3) (19-23, 34-40). The recent outbreaks differ considerably from previous outbreaks of drug-resistant tuberculosis in several ways. First, the recent outbreaks involved large numbers of patients; nearly 300 cases have been identified. In addition, in the recent outbreaks tuberculosis was transmitted not only from patient to patient but also from patient to health care worker. The epidemiologic evidence of nosocomial transmission was confirmed by DNA fingerprinting data: strains from epidemiologically linked cases were found to have identical patterns by restriction fragment length polymorphism (RFLP) analysis. RFLP analysis also suggested that several of the outbreaks in New York State were connected.

Table: Nosocomial HIV-Related Multidrug-Resistant Tuberculosis Outbreaks as of October 1992


Table: HIV Prevalence and Mortality of Patients with Multidrug-Resistant Tuberculosis as of October 1992

Finally, the recent outbreaks involved a large percentage of highly susceptible patients who became infected with highly drug-resistant organisms that are more difficult to treat. More than 80% of the cases occurred in persons infected with HIV, and all but six of the patients had organisms resistant to both isoniazid and rifampin. Mortality in these outbreaks was very high. In seven of the eight outbreaks, more than 70% of the patients died. The median interval between diagnosis and death ranged from 4 to 16 weeks.

Factors associated with the outbreaks of multidrug-resistant tuberculosis
Most of the outbreaks were centered in wards or outpatient clinics where HIV-infected persons received care. There is no evidence that persons with HIV infection are more likely to be infected with tuberculosis if exposed, but it is clear that, once infected with tuberculosis, persons who are infected with HIV have a much higher risk for the development of active disease than persons who are not infected with HIV. In addition, infection may progress to active disease rapidly. The transmission of tuberculosis to HIV-infected persons in these outbreaks was followed by the rapid development of new cases of active disease, which were sources of further transmission.

Prolonged infectiousness also promoted transmission. The diagnosis of tuberculosis in HIV-infected persons was sometimes delayed because of the unusual radiographic presentations of tuberculosis, coinfection with other pulmonary pathogens to which patients' symptoms were attributed, and the overgrowth of M. tuberculosis in the laboratory by other mycobacteria. Delays in diagnosis lead to delays in the initiation of isolation and treatment. In addition, drug resistance was not recognized promptly because of lengthy laboratory delays, which postponed the initiation of effective treatment and resulted in prolonged infectiousness.

Inadequate infection control practices also facilitated transmission. Isolation rooms were found to be at positive pressure relative to other parts of the facility. Patients who had been assigned to isolation rooms were found in hallways, patient lounges, or other common areas. Furthermore, isolation precautions were discontinued prematurely, after an arbitrary number of days, rather than when there was clinical or laboratory evidence of decreased infectiousness.

Finally, some patients who were given appropriate therapy in the hospital were lost to follow-up after discharge. A lack of consistent follow-up after discharge promotes lapses in therapy and heightens the potential for transmission both in the hospital and in the community. This problem was not specifically addressed in the outbreak investigations, but it was described in an unrelated report by Brudney and Dobkin. They studied 224 tuberculosis patients admitted to Harlem Hospital in 1988. Of 178 patients discharged on tuberculosis treatment, 89% were lost to follow-up and did not complete therapy. Forty-eight patients were readmitted to the hospital with infectious tuberculosis within 1 year (41).

The Advisory Council for the Elimination of Tuberculosis has issued recommendations aimed at preventing drug-resistant tuberculosis (42). Drug susceptibility tests should be performed on all initial isolates of M. tuberculosis. Additional isolates should be tested for patients whose cultures remain positive after 3 months of therapy or who have signs of failure or relapse. Susceptibility testing not only guides clinical therapeutic decisions, but also allows tuberculosis control programs to monitor drug resistance trends and evaluate tuberculosis control efforts.

Second, initial regimens for treating tuberculosis should include four drugs: isoniazid, rifampin, pyrazinamide, and either ethambutol or streptomycin. When drug susceptibility results become available, the regimen can be adjusted accordingly. More drugs may be required for initial regimens in institutions where outbreaks of multidrug-resistant tuberculosis are occurring. In areas where resistance to isoniazid or rifampin is less than 4%, an initial regimen of three drugs may be adequate. The continued surveillance of drug resistance is necessary to detect changes in local drug resistance patterns and to evaluate the appropriateness of initial regimens in given areas.

Third, therapy for tuberculosis should be directly observed by a health care provider. Nonadherence to therapy is a major cause of treatment failure, and it may contribute to the emergence of drug resistance. Directly observed therapy (DOT) is an effective way to ensure adherence, but the method must be tailored to each patient's needs and preferences.

Tuberculosis that is resistant to isoniazid alone can be treated successfully by using rifampin and ethambutol for a minimum of 12 months, preferably supplemented by pyrazinamide for the first 2 months. In addition, a four-drug regimen has been shown to be highly effective for isoniazid-resistant organisms. Rifampin resistance, however, significantly reduces cure rates and increases the complexity and the required length of treatment (43).

The treatment of multidrug-resistant tuberculosis should be based on the in vitro susceptibility test results and the patient's treatment history (5,44,45). The regimen should include at least three drugs — preferably drugs that the patient has not received before — to which the patient's organism is susceptible. The regimen should include an injectable medication whenever possible. If the regimen must be changed for any reason, two medications should be added simultaneously to avoid selecting for drug-resistant mutants.

Medications used to treat multidrug-resistant tuberculosis are less effective, more costly, and more likely to cause adverse reactions than the five first-line drugs (44-48). Treatment is further complicated because of the length of therapy needed to prevent relapse: generally, therapy for multidrug-resistant tuberculosis should continue an additional 18 to 24 months after culture results convert to negative.

Because of the high failure and relapse rates associated with multidrug-resistant tuberculosis, surgical resection has been used by some clinicians to supplement medical therapy. The surgical resection of a major pulmonary focus is best performed once aggressive medical therapy has achieved a clinical response. The administration of antituberculosis medications should continue for 18 to 24 months after culture results convert to negative. Iseman and associates reported that in a series of 99 patients being treated for multidrug-resistant tuberculosis, the combination of surgical resection and medical therapy produced lower rates of failure and relapse compared with that of historical controls (49).

Preventive Therapy for Persons Exposed to Multidrug-Resistant Tuberculosis
When taken appropriately, isoniazid preventive therapy is very effective in preventing the development of active tuberculosis in persons infected with susceptible strains of M. tuberculosis, even in persons coinfected with HIV. Rifampin is recommended for persons infected with isoniazid-resistant strains (43,50).

In the outbreaks of multidrug-resistant tuberculosis, numerous tuberculin skin test conversions were documented among persons exposed to multidrug-resistant tuberculosis. Preventive therapy for persons exposed to multidrug-resistant tuberculosis has not been studied, but CDC has published guidelines for the management of such contacts (51). Decisions regarding preventive therapy should be based on the likelihood that the contact has been newly infected, that the infecting organisms are multidrug-resistant, and that active tuberculosis will develop. If the likelihood of these factors is low, standard preventive therapy is recommended. Alternative regimens should be considered for persons who are likely to be newly infected with a multidrug-resistant strain and who are at high risk for the development of active disease if infected. CDC recommends that alternative regimens include at least two drugs to which the infecting organism is known to be susceptible. Potential alternative regimens include pyrazinamide and ethambutol or pyrazinamide and a quinolone (e.g., ciprofloxacin) for 6 to 12 months.

Control Measures
Controlling the tuberculosis epidemic and preventing drug-resistant tuberculosis necessitate that health care providers diagnose tuberculosis early and initiate effective therapy promptly and that tuberculosis patients successfully complete therapy. Specifically, directly observed therapy should be used more widely, initial treatment regimens should be used that prevent the development of drug-resistant tuberculosis, and treatment activities should be coordinated between public health departments and other facilities that provide care for patients with tuberculosis. Surveillance of drug resistance must be done to determine the appropriate initial regimen for tuberculosis in given areas, define risk factors for drug resistance, and detect transmission of drug-resistant tuberculosis.

Furthermore, appropriate infection control practices must be used to prevent the transmission of tuberculosis in health care settings (52). Different infection control measures interrupt the transmission of tuberculosis at different points in the transmission process. The most effective infection control measures curb transmission at the source by preventing the generation of infectious droplet nuclei. Examples of administrative control methods include the early diagnosis of disease and prompt initiation of effective therapy for persons with active tuberculosis, preventive therapy for persons with tuberculosis infection, and instruction of patients to cover their noses and mouths with a tissue when coughing. Booths and hoods used for aerosolized pentamidine or sputum induction are other examples of source control measures.

Potentially infectious patients should be placed in tuberculosis isolation in a private room that has negative pressure relative to the rest of the facility. The air from the room should be exhausted directly to the outside or through a high-efficiency particulate air (HEPA) filter, if recirculation of air into the general ventilation system from the room is unavoidable. Isolation should be continued until the patient is no longer infectious. Patients known or suspected to have active tuberculosis should be considered infectious if they are coughing, if they are undergoing cough-inducing procedures, if their sputum smears contain acid-fast bacilli, if they are not receiving antituberculosis chemotherapy, if they recently started receiving chemotherapy, or if they are not responding to therapy. Most patients with drug-susceptible disease become noninfectious after 2 to 3 weeks of therapy, but the period of infectiousness varies from patient to patient. Patients with drug-susceptible disease who are receiving effective therapy and who show signs of response to therapy (i.e., a reduction in cough, the resolution of fever and a decreasing number of bacilli on sputum smears) are probably no longer infectious. However, patients with drug-resistant disease may take longer to respond to therapy, especially if drug resistance is not suspected and they are not receiving effective therapy. Therefore, patients should be considered infectious until their sputum smears are free of bacilli on 3 consecutive days.

Ventilation and other engineering controls can be used to eliminate infectious droplet nuclei once they are released into the air, but these infection control measures are less efficient and less effective than administrative control methods. Ventilation reduces airborne contaminants by introducing uncontaminated air into the room, thereby diluting the concentration of contaminants. The contaminants are removed by exhausting the air from the room directly to the outside. Potential supplemental measures include HEPA filters and germicidal ultraviolet irradiation. These supplemental measures require proper installation and regular maintanence by qualified personnel, and their use in preventing tuberculosis transmission has not been well studied. Furthermore, exposure to ultraviolet irradiation may pose substantial health risks. Short-term exposure to germicidal ultraviolet irradiation is known to cause keratoconjunctivitis and erythema, and long-term exposure has been associated with an increased risk of basal cell carcinoma.

Of all infection control methods, the use of personal protective equipment is the least efficient and least effective. These devices protect only the wearer, and they function only when they fit properly and are worn correctly. Administrative controls and engineering controls are necessary to protect all persons in the facility from airborne contaminants. Personal protective devices should be worn by health care workers in areas where the air is likely to be contaminated with a higher concentration of infectious droplet nuclei such as isolation rooms or rooms used for cough-inducing procedures.

A tuberculosis screening and prevention program for health care workers is an essential component of infection control programs in health care facilities. All persons working in health care facilities should be skin tested upon employment. Persons who have negative skin test results should be retested periodically. Screening will detect tuberculosis infection in persons for whom preventive therapy may be appropriate. In addition, skin test results must be analyzed to evaluate the effectiveness of an infection control program by determining whether tuberculosis is being transmitted in the facility.

Most cases of drug-resistant tuberculosis in the United States can be prevented by using drug-susceptibility surveillance to monitor trends, directly observed therapy to ensure adherence to and completion of therapy, and initial regimens that include four drugs. Most of the transmission of drug-resistant tuberculosis in health care settings could be prevented by fully implementing the current CDC guidelines. Additional research is needed to identify more rapid diagnostic techniques, develop new drugs that are less toxic and require shorter courses of treatment, and evaluate the role of supplemental infection control measures.


  1. Snider DE Jr., Kelly GD, Cauthen GM, Thompson NJ, Kilburn JO. Infection and disease among contacts of tuberculosis cases with drug resistant and drug susceptible bacilli. Am Rev Respir Dis 1985;132(1):125-132.

  2. David HL. Probability distribution of drug-resistant mutants in unselected populations of Mycobacteria tuberculosis. Applied Microbiology 1970;20:810-814.

  3. Canetti G. The J. Burns Amberson Lecture. Present aspects of bacterial resistance in tuberculosis. Am Rev Respir Dis 1965;92:687-703.

  4. Mitchison DA, Nunn AJ. Influence of initial drug resistance on the response to short course chemotherapy of pulmonary tuberculosis. Am Rev Respir Dis 1986;133:423-430.

  5. Goble M, Horsburgh CR Jr., Waite D, et al. Treatment of isoniazid and rifampin-resistant tuberculosis. Am Rev Respir Dis 1988;137(Suppl):24.

  6. CDC. Expanded tuberculosis surveillance and tuberculosis morbidity — United States, 1993. MMWR 1994;43:361-366.

  7. U.S. Public Health Service Cooperative Investigation. Prevalence of drug resistance in previously untreated patients. Am Rev Respir Dis 1964;89:327-336.

  8. Doster B, Caras GJ, Snider DE Jr. A continuing survey of primary drug resistance in tuberculosis, 1961 to 1968. A U.S. Public Health Service cooperative study. Am Rev Respir Dis 1976;113:419-425.

  9. Kopanoff DE, Kilburn JO, Glassroth JL, et al. A continuing survey of tuberculosis primary drug resistance in the United States: March 1975-November 1977. A U.S. Public Health Service cooperative study. Am Rev Respir Dis 1978;118:835-842.

  10. CDC. Primary resistance to antituberculosis drugs — United States. MMWR 1980;29:345-346.

  11. CDC. Primary resistance to antituberculosis drugs — United States. MMWR 1983;32:521-523.

  12. Snider DE Jr., Cauthen GM, Farer LS, et al. Drug resistant tuberculosis (letter). Am Rev Respir Dis 1991;144:732.

  13. Bloch AB, Cauthen GM, Onorato IM, et al. Nationwide survey of drug-resistant tuberculosis in the United States. JAMA 1994;271:665-671.

  14. Frieden TR, Sterling T, Pablos-Mendez A, Kilburn JO, Cauthen GM, Dooley SW. The emergence of drug-resistant tuberculosis in New York City. N Engl J Med 1993;328(8):521-526.

  15. Costello HD, Caras GJ, Snider DE Jr. Drug resistance among previously treated tuberculosis patients, a brief report. Am Rev Respir Dis 1980;121:313-316.

  16. CDC. Drug resistant tuberculosis among the homeless — Boston. MMWR 1985;34:429-431.

  17. Nardell E, McInnes B, Thomas B, Weidhaas S. Exogenous reinfection with tuberculosis in a shelter for the homeless. N Engl J Med 1986;315:1570-1575.

  18. Small PM, Shafer RW Hopewell PC, et al. Exogenous reinfection with multidrug-resistant Mycobacterium tuberculosis in patients with advanced HIV infection. N Engl J Med 1993;328(16):1137-1144.

  19. CDC. Nosocomial transmission of multidrug resistant TB to health-care workers and HIV-infected patients in an urban hospital — Florida. MMWR 1990;39:718-722.

  20. CDC. Nosocomial transmission of multidrug resistant tuberculosis among HIV-infected persons — Florida and New York, 1988-1991. MMWR 1991;40:585-591.

  21. Edlin BR, Tokars JI, Grieco MH, et al. An outbreak of multidrug resistant tuberculosis among hospitalized patients with the acquired immunodeficiency syndrome. N Engl J Med 1992;326:1514-1521.

  22. Pearson ML, Jereb JA, Friedman TR, et al. Nosocomial transmission of multidrug resistant Mycobacterium tuberculosis. A risk to patients and health care workers. Ann Intern Med 1992;117:191-196.

  23. Beck-Sague C, Dooley SW, Hutton MD, et al. Outbreak of multidrug resistant Mycobacterium tuberculosis infections in a hospital: transmission to patients with HIV infection and staff. JAMA 1992;268:1280-1286.

  24. Barnes PF. The influence of epidemiologic factors on drug resistance rates in tuberculosis. Am Rev Respir Dis 1987;136:325-328.

  25. Riley LW, Arathoon E, Loverde VD. The epidemiologic patterns of drug resistant Mycobacterium tuberculosis: a community-based study. Am Rev Respir Dis 1989;19:1282-1285.

  26. CDC. Drug resistance among Indochinese refugees with tuberculosis. MMWR 1981;30:273-275.

  27. Pitchenik AE, Russell BW, Cleary T, et al. The prevalence of tuberculosis and drug resistance among Haitians. N Engl J Med 1982;307:162-165.

  28. CDC. INH-resistant tuberculosis in an urban high school — Oregon. MMWR 1980;29:194-196.

  29. CDC. Multidrug-resistant tuberculosis North Carolina. MMWR 1987;35(51,52):785.

  30. CDC. Outbreak of multidrug resistant tuberculosis — Texas, California, and Pennsylvania. MMWR 1990;39(22):369.
  31. Reeves R, Blakey D, Snider DE Jr., Farer LS. Transmission of multiple drug-resistant tuberculosis: report of a school and community outbreak. Am J Epidemiol 1981;113:423.

  32. Steiner M, Chaves AD, Lyons JA, et al. Primary drug-resistant tuberculosis: report of an outbreak. N Engl J Med 1970;283:1353.

  33. CDC. Interstate outbreak of drug resistant tuberculosis involving children — California, Montana, Nevada, Utah. MMWR 1983;32:516-518.

  34. Fischl M, Uttamchandani RB, Daikos GL, 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.

  35. Fischl M, Daikos GL, Uttamchandani RB, et al. Clinical presentation and outcome of patients with HIV infection and tuberculosis caused by multiple-drug resistant bacilli. Ann Intern Med 1992;117:184-190.

  36. CDC. Transmission of multidrug resistant tuberculosis among immunocompromised persons in a correctional system — New York, 1991. MMWR 1992;41:507-509.

  37. Dooley SW, Jarvis WR, Martone WJ. Multidrug resistant tuberculosis (editorial). Ann Intern Med 1992;117:257-258.

  38. Villarino ME, Geiter LJ, Simone PM. The multidrug resistant tuberculosis challenge to public health efforts to control tuberculosis. Public Health Rep 1992;107:616-625.

  39. Valway SE, Greifinger RB, Papania M, et al. Multidrug-resistant tuberculosis in the New York State prison system, 1990-1991. J Infect Dis 1994;170:151-156.

  40. Valway SE, Richards SB, Kovacovich J, Greifinger RB, Crawford JT, Dooley SW. Outbreak of multidrug-resistant tuberculosis in a New York State prison, 1991. Am J Epidemiol 1994;140(2):113-122.

  41. Brudney K, Dobkin J. Resurgent tuberculosis in New York City. Human immunodeficiency virus, homelessness and the decline of tuberculosis control programs. Am Rev Respir Dis 1991;144:745-749.

  42. CDC. Initial therapy for tuberculosis in the era of multidrug-resistance: recommendations of the Advisory Council for the Elimination of Tuberculosis. MMWR 42(RR-7):1-8.

  43. American Thoracic Society/CDC. Treatment of tuberculosis and tuberculosis infection in adults and children. Am J Respir Crit Care Med 1994;149:1359-1374.

  44. Iseman MD, Madsden LA. Drug-resistant tuberculosis. Clin Chest Med 1989;10:341-353.

  45. Goble M. Drug resistant tuberculosis. Semin Respir Infect 1986;1:220-229.

  46. Girling DJ. Adverse effects of antituberculosis drugs. Drugs 1982;23:56-74.

  47. Moulding T, Davidson PT. Tuberculosis II: toxicity and intolerance to antituberculosis drugs. Drug Ther 1974:39-43.

  48. Lefkowitz MS. The antimycobacterial drugs. Semin Respir Med 1981;2:196-201.

  49. Iseman MD, Madsden LA, Goble M, Pomerantz M. Surgical intervention in the treatment of pulmonary disease caused by drug resistant Mycobacterium tuberculosis. Am Rev Respir Dis 1990;141:623-625.

  50. CDC. The use of preventive therapy for tuberculous infection in the United States: recommendations of the Advisory Committee for the Elimination of Tuberculosis. MMWR 1990;39(RR-8):9-12.

  51. CDC. Management of persons exposed to multidrug-resistant tuberculosis. MMWR 1992;41(RR-11):61-71.

  52. CDC. Guidelines for preventing the transmission of Mycobacterium tuberculosis in health-care facilities, 1994. MMWR 1994;43(No. RR-13):1-132.

Please refer to the electronic PDF version and/or the original paper copy for the graphics, figures, and tables.


Released October 2008
Centers for Disease Control and Prevention
National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention
Division of Tuberculosis Elimination -

Please send comments/suggestions/requests to:, or to
CDC/Division of Tuberculosis Elimination
Communications, Education, and Behavioral Studies Branch
1600 Clifton Rd., NE - Mailstop E-10, Atlanta, GA 30333