Tuberculosis chemoprophylaxis for diabetics: Are the benefits of isoniazid worth the risk?

The American Thoracic Society has recommended isoniazid chemoprophylaxis for tuberculin-positive but asymptomatic, chest x-ray negative, adult diabetics. This is despite little evidence that diabetics are at increased risk for active tuberculosis. Furthermore both the diabetes and tuberculosis mortality rates are declining and isoniazid therapy is not totally effective and carries risks of morbidity and death. We performed a decision analysis comparing chemoprophylaxis and no chemoprophylaxis, using isoniazid-hepatitis risks, Markov-process determined life expectancies, and tuberculosis risks. We reviewed the literature to obtain isoniazid-hepatitis morbidity and mortality rates, isoniazid effectiveness, tuberculosis activation and case-fatility rates, and diabetes mortality rates. Although the results show isoniazid chemoprophylaxis to be beneficial in almost all age, sex, and race cohorts, the benefits are small and measured in days. Combining all assumptions favoring chemoprophylaxis, life expectancy for 50-year-old white men, for example, is extended by 39 days (an increase of 0.7%). Combining all assumptions favoring the strategy of no chemoprophylaxis for the same cohort, life expectancy is the same for both strategies. Thus the isoniazid strategy for asymptomatic adult diabetics has a small effect. We suggest that the recommendations be qualified to encourage a joint decision by patient and physician.