before patient A was diagnosed. Patient A had complained of low-grade fevers and weight loss, but a lack of cough and pulmonary signs was specifically documented. Four months later patient C developed pulmonary tuberculosis caused by an isolate with the same fingerprint as that of the isolate from patient A. Despite a documented lack of pulmonary symptoms, nosocomial transmission was thought to be possible.
During the study period eight patients with pulmonary tuberculosis were not isolated before the diagnosis was made. Two had isolates in clusters: six did not. A total of 186 employees had follow-up testing with no PPD conversions. In fact, 28 of 70 (40 percent) health care workers with PPD conversions over the entire study period had no adult patient care responsibilities.
The authors state that their hospital follows guidelines consistent with the CDC guidelines, although details are not provided. This paper suggests that even in a hospital with a large number of tuberculosis admissions, the CDC guidelines are effective at preventing nosocomial transmission. Of the two possible cases of nosocomial transmission, no breakdown in following the guidelines occurred. This points out that unless every patient is isolated for every visit, some nosocomial transmission of tuberculosis may be unavoidable.
Jernigan and colleagues (28) reported on a retrospective questionnaire that was sent to 52 former residents who had done a total of 70 6-week (420 physician-weeks) rotations at a tuberculosis sanatorium affiliated with the University of Virginia. There were 10 unprotected exposures to tuberculosis patients during training reported by the former house staff, 2 of which occurred at the sanatorium. No PPD conversions were reported during residency. The sanatorium had tuberculosis isolation rooms with negative pressure as well as UVGI (details were not given), and only simple surgical masks were used at the facility. Since “administrative controls” are somewhat built in at a sanatorium (in that tuberculosis is presumably known in all patients prior to their arrival), this suggests that even in a potentially high-risk environment, routine engineering controls and simple personal respiratory protection are adequate.
The major role of administrative controls is to ensure that patients with pulmonary tuberculosis are promptly isolated. In most settings this requires isolating many patients who prove not to have tuberculosis for every patient who actually does have tuberculosis. As pointed out in the CDC guidelines, criteria for isolation must be derived locally, taking into