Resurgent nosocomial tuberculosis: consequences and actions for hospital epidemiologists.

Nosocomial tuberculosis did not receive the respect it deserved during the 198Os, for several reasons. First, tuberculosis was thought to be a disappearing disease, to the extent that a Program for Elimination of Tuberculosis was created by the Centers for Disease Control (CDC).’ Second, rather than its previous pattern as a disease of the entire population, by the 198Os, tuberculosis had become primarily a disease of the disadvantaged-cases of tuberculosis became a rarity except in U.S. hospitals serving the homeless and the helpless. Because these hospitals and their patients tend to be ignored, tuberculosis was lost to general public and medical attention, especially in comparison to new problem infections like acquired immunodeficiency syndrome (AIDS). Third, virtually all tuberculosis isolates were susceptible to the two major drugs used for treatment: isoniazid and rifampin. The tuberculocidal action of these drugs in combination made treatment shorter and easier than ever before. When resistance to these drugs was found in occasional strains, these usually emerged from initially susceptible strains in patients who were noncompliant with therapy or in patients from other countries. Resistant strains seldom were found in clustered fashion and usually could be treated outside the hospital setting. Thus, the threat of nosocomial tuberculosis seemed diminished. During this period, tuberculosis control programs at the federal, state, and local level were cut back or eliminated.2 Similarly, hospital guidelines and policies developed to prevent nosocomial spread of tuberculosis fell into disuse and neglect. These polities had been developed when care of tuberculosis patients was moving from specialized tuberculosis sanitariums to general acute care hospitals. Included were extensive guidelines for respiratory precautions, laboratory studies, and proper chemotherapy3s4 Parts of these guidelines were based on minimal objective evidence. How long should a patient newly started on antituberculous chemotherapy remain on isolation precautions before the risk of contagion ends?” How many sputum cultures are needed to define presence or absence of tuberculosis? Should these sputum cultures be obtained as soon as possible, or should they be spread out over a period of days? How important is proper masking compared with airflow as a preventive measure? How can a diagnosis of tuberculosis be made more speedily, and how can drugresistant strains be identified earlier? As tuberculosis declined in attention, the research needed to deal with these and other questions never was funded and never was done.6 Guidelines for dealing with tuberculosis in hospitals also specified continuous surveillance of tuberculosis infection in hospital workers.4,7 In the 198Os, however, tuberculosis seemed to be disappearing, and AIDS and its complications became prominent. Hospitals in many communities where tuberculosis had become rare found that continued tuberculin skin testing of their employees appeared to be costly and of minimal benefit.7vs Many hospitals still encountering tuberculosis patients now were bearing the brunt of the AIDS epidemic. AIDS was almost always fatal, and tuberculosis was almost always treatable, so marshall-