The transmission of Mycobacterium tuberculosis that we describe aboard a commercial aircraft involved a highly infectious passenger, a long flight, and close proximity of contacts to the index patient.

In January 1996, smear- and culture-positive tuberculosis (TB) was diagnosed for a 22-year-old black man after he had traveled on two U.S. passenger trains (29.1 hours) and a bus (5.5 hours) over 2 days. To determine if transmission had occurred, passengers and crew were notified of the potential exposure and instructed to undergo a tuberculin skin test (TST). Of the 240 persons who completed screening, 4 (2%) had a documented TST conversion (increase in induration of > or = 10 mm between successive TSTs), 11 (5%) had a single positive TST (> or = 10 mm), and 225 (94%) had a negative TST (< 10 mm). For two persons who underwent conversion, no other risk factors for a conversion were identified other than exposure to the ill passenger during train and/or bus travel. These findings support limited transmission of Mycobacterium tuberculosis from a potentially highly infectious passenger to other persons during extended train and bus travel.

Contacts exposed to tuberculosis patients with acquired immunodeficiency syndrome (AIDS) or human immunodeficiency virus (HIV) infection were compared with contacts of HIV-negative patients for evidence of Mycobacterium tuberculosis transmission, based on a review of records of tuberculin skin tests administered during routine health department follow-up investigations in Miami/Dade County, Florida, from 1985 through 1989. After an adjusted analysis designed to balance background prevalence, tuberculin positivity was 42.0% in 2,158 contacts of HIV-negative patients compared with 28.6% and 31.3% in 363 contacts of HIV-infected patients and 732 contacts of AIDS patients, respectively. Similar results were observed in a subset of 5- to 14-year-old contacts of United States-born black or white tuberculosis patients chosen to minimize the possibility of false-negative tuberculin tests in contacts due to undiagnosed HIV infection. Analysis of contacts as sets showed a more than expected number of sets with none or all contacts infected, but this did not differ by HIV/AIDS group. In this study, tuberculosis patients with AIDS or HIV infection were less infectious to their contacts and, in this community, exposed fewer contacts than HIV-negative tuberculosis patients.