Pa. Jenkins et al., MYCOBACTERIUM-KANSASII PULMONARY INFECTION - A PROSPECTIVE-STUDY OF THE RESULTS OF 9 MONTHS OF TREATMENT WITH RIFAMPICIN AND ETHAMBUTOL, Thorax, 49(5), 1994, pp. 442-445
Background - Pulmonary disease caused by Myocobacterium kansasii is re
ported in approximately 50 new patients in Britain annually. Rifampici
n and ethambutol are effective in vitro but the optimal duration of tr
eatment, and whether isoniazid should also be given, are uncertain. Th
e British Thoracic Society has conducted a prospective, multicentre st
udy of the treatment of this condition with rifampicin and ethambutol
given for nine months. Methods - One hundred and seventy three patient
s with two or more positive cultures and radiological evidence of dise
ase were recruited via the Mycobacterium Reference Unit (PHLS) in Card
iff from 113 physicians in England, Scotland, and Wales. Rifampicin an
d ethambutol were given for nine months, other antituberculosis drugs
being discontinued once the culture was identified as M kansasii. Pati
ents were reviewed, sputum cultured, and chest radiographs performed b
efore, during, and at regular intervals for 51 months after chemothera
py. Results - The mean (SD) age was 55.5 (11.7) years, 73% were men, a
nd 50% had other lung problems. Cavitation was seen in 88%, bilateral
shadowing in 48%, and three or more lung zones were affected in 46%. A
ll cultures were sensitive to rifampicin and ethambutol but resistant
to isoniazid and pyrazinamide. One patient who took chemotherapy irreg
ularly still had positive cultures at seven and eight months. Fifteen
patients developed positive cultures after the end of chemotherapy; fa
ctors which might account for the relapse were identified in eight. Re
infection rather than relapse was suspected in three of the 15. Radiog
raphic improvement stabilised within three years in 80%. Conclusions -
M kansasii pulmonary infection responds well to nine months of treatm
ent with rifampicin and ethambutol but patients who contract this dise
ase have a high mortality rate from other causes. Isoniazid does not a
ppear to be a necessary part of the regimen.