The Global Problem of Multidrug-Resistant Tuberculosis: Title and subTitle BreakThe Genie Is out of the Bottle

Drug-resistant tuberculosis (TB) was first observed in 1948, subsequent to the first trials of streptomycin for TB treatment.¹ Ever since then, drug-resistant TB has been recognized to occur as the result of suboptimal TB treatment. In a recent survey of 35 countries, 12.6% of Mycobacterium tuberculosis isolates were resistant to at least 1 drug, and 2.2% were resistant to both of the primary drugs used to treat TB, isoniazid and rifampin.² Thus, today most cases of TB are drug susceptible at the time of diagnosis, only becoming drug resistant through suboptimal therapy. Therefore, the World Health Organization has focused its efforts to combat multidrug-resistant TB on a strategy of preventing the generation of new multidrug-resistant TB cases. This program, directly observed therapy short-course (DOTS), relies on local government commitment to a sustained effort using case detection by sputum microscopy, directly observed treatment with a standard therapeutic regimen, maintenance of an uninterrupted drug supply, and monitoring outcome with a standard reporting system.

A strategy of preventing TB through a carefully designed and supervised TB program that incorporates the elements of DOTS has been demonstrated to be effective in reducing rates of multidrug-resistant TB. For example, in Tarrant County, Texas, standard TB therapy without direct observation was used from 1980 to 1986. In 1986, directly observed therapy was introduced and used exclusively until 1992. After the institution of directly observed therapy (with effective treatment for multidrug-resistant TB during both periods), primary drug resistance declined from 14% to 2.1% of cases; moreover, among relapses, multidrug-resistant TB was reduced by 80%.³ In sophisticated modeling studies, Blower and colleagues⁴ have shown that while a well-functioning TB control program can reduce the incidence of multidrug-resistant TB, a poorly functioning program is worse than none at all; such a program actually produces multidrug-resistant TB.

The DOTS program has been increasingly adopted throughout the world; 119 countries now have DOTS programs.⁵ The DOTS program has been successful in preventing increases in cases of multidrug-resistant TB in many countries in which preexisting levels of multidrug-resistant TB were low. Examples include Chile, in which only 0.4% of TB cases have multidrug-resistant TB, and Benin, in which the rate is 0.5%.⁶ In countries in which there are already substantial levels of multidrug-resistant TB, it was hoped that the DOTS strategy would lead to a decline in such cases; in theory, reduced generation of multidrug-resistant TB by an effective DOTS program and reduced spread of multidrug-resistant TB by persons with persistent disease could eventually lead to its disappearance.

Until now, it has been unclear whether a good DOTS program would be enough to control multidrug-resistant TB in countries in which multidrug-resistant TB was already established. The report in this issue of THE JOURNAL by Espinal and colleagues⁷ shows that achieving control of multidrug-resistant TB through establishment of DOTS alone in such countries was a false hope. In 2 of the countries described, Peru and South Korea, excellent DOTS programs have been operating for several years. Despite this, the prevalence of multidrug-resistant TB in these countries is still substantial among persons with newly diagnosed TB. Since these patients have not been previously treated, their disease cannot have resulted from failed therapy, and must be the result of primary transmission of multidrug-resistant TB from others in the community. Multidrug-resistant TB has not been markedly reduced or prevented by the DOTS program and it is unlikely that these rates will decline to low levels without further intervention.

As noted by the authors, the study was performed under field conditions. While this is the appropriate way to evaluate a TB control strategy, the approach introduces several limitations. First, some of the primary cases may have been previous treatment failures that were unrecognized. This would have the effect of overestimating the amount of multidrug-resistant TB disease that was being transmitted in the community. Second, treatment outcome results were only available for 54% of the subjects enrolled. While this could be a serious limitation, it resulted from low rates in only 2 of the 6 areas studied; the other 4 had an impressive 87% to 91% follow-up rate. Since the results were similar at all 6 sites, it is unlikely that substantial bias was introduced by the missing results. Third, smear positivity was used to identify treatment failures. This method underestimates the rate of failure by not detecting smear-negative, culture-positive cases, so that actual failure rates were likely even higher than those reported. Taken together, these limitations do not invalidate the conclusion that DOTS alone will not control multidrug-resistant TB in countries in which substantial numbers of these organisms are now circulating.

What is to be done? As Espinal and colleagues note, it is imperative that new treatment strategies be devised for patients with multidrug-resistant TB in these countries. However, it is also essential that an effective DOTS program be ensured before beginning a multidrug-resistant TB treatment program. If this were not done, a poor TB control program would generate multidrug-resistant TB cases more rapidly than a treatment program could treat them. The necessity of targeting multidrug-resistant TB treatment programs to areas with an effective DOTS program has led to the proposal that these new programs be known as DOTS Plus.^{8 - 9} Such DOTS Plus programs will require drug susceptibility testing, surveillance of trends in drug susceptibility, and treatment regimens tailored to the results of such testing.^{10 - 11} Both drug susceptibility testing and tailored treatment regimens will entail considerable expense.

Is this a problem for citizens of the developed countries and other countries in which effective TB control programs and individualized therapy for multidrug-resistant TB are already in place? The answer is an emphatic yes. Many of the isolates of M tuberculosis that cause disease in the United States already come from abroad, either in persons born abroad or in US residents who become infected with TB while traveling overseas.^{12 - 13} In 1 case, TB acquired abroad was highly drug resistant and nearly fatal; many persons in the United States were exposed to this organism.¹⁴ Such strains have been shown to spread rapidly in the United States once introduced.¹⁵ If drug-resistant TB increases in developing countries, rates in developed countries will increase as well. Such rates recently increased by 50% in Denmark and Germany.¹⁶ This is not someone else's problem.

The genie of multidrug-resistant TB is irreversibly out of the bottle. Containing multidrug-resistant TB is a public health emergency. Currently available second-line antibiotics to treat multidrug-resistant TB are 4 to 10 times more likely to fail than standard therapy for drug-susceptible TB, and about 100 times more costly.^{6 ,17 - 19} Moreover, no new drugs that might be effective in treatment of multidrug-resistant TB are currently undergoing clinical trials. A recent report predicted that effective new drugs for TB are at least a decade away.²⁰ Multidrug-resistant TB thus raises the specter of a return to the 18th and 19th centuries, when as many as 20% of all adult deaths in Europe and the United States resulted from TB.

The response to this emergency will require 3 commitments. First, there must be a commitment to establish effective DOTS programs in every county in which TB occurs. The World Health Organization and its partners in the Stop TB Initiative are working diligently toward this goal.²¹ Second, there must be a commitment to establish DOTS Plus in every country in which there is an effective DOTS program and multidrug-resistant TB persists. This will be an expensive undertaking and will require resources from governments of developed countries, voluntary agencies, and concerned individuals. Third, there must be a commitment to intensify efforts to develop new drugs for TB treatment. A recent meeting convened by the Rockefeller Foundation has begun the process of developing public-private collaborations to advance the development of such drugs.²² Without each of these 3 commitments, the future will be grim.