As. Breathnach et al., AN OUTBREAK OF MULTI-DRUG-RESISTANT TUBERCULOSIS IN A LONDON TEACHINGHOSPITAL, The Journal of hospital infection, 39(2), 1998, pp. 111-117
We describe the epidemiology and control of a hospital outbreak of mul
ti-drug-resistant tuberculosis (MDR-TB). A human immunodeficiency viru
s (HIV)-negative patient with drug-sensitive tuberculosis developed MD
R-TB during a period of unsupervised therapy. She was admitted to an i
solation room in a ward with HIV-positive patients, but the room, unbe
known to hospital staff, was at positive-pressure relative to the main
ward. Seven HIV-positive contacts developed MDR-TB. The diagnosis in
the second patient was delayed, partly because acid-fast bacilli in hi
s sputum were assumed to be Mycobacterium avium-intracellulare. All th
e available Mycobacterium tuberculosis isolates were indistinguishable
by molecular typing. Nearly 1400 staff and patient contacts were offe
red screening, but the screening programme detected only one of the ca
ses. Despite therapy, the index patient and two of the contacts died.
HIV-positive patients are more likely than others to develop tuberculo
sis after exposure, and the disease may progress more rapidly. In thes
e patients the possibility that acid-fast bacilli may represent M. tub
erculosis must always be considered. Patients with tuberculosis (suspe
cted or proven) should not be nursed in the same wards as immunosuppre
ssed patients, and should be isolated. MDR-TB cases must be isolated i
n negative-pressure rooms. Hospital siderooms may be positive-pressure
as a fire safety measure; infection control teams must be aware of th
e airflows in all isolation rooms, and must be consulted during the de
sign of hospital buildings. Good communication between infection contr
ol teams and clinicians is important, and all medical and nursing staf
f must be aware of the principles of management of patients with prove
n or suspected tuberculosis and MDR-TB.