P. Godfreyfaussett et al., RECURRENCE OF HIV-RELATED TUBERCULOSIS IN AN ENDEMIC AREA MAY BE DUE TO RELAPSE OR REINFECTION, Tubercle and lung disease, 75(3), 1994, pp. 199-202
Setting: Two Research Clinics within Nairobi, Kenya, one in the Infect
ious Diseases Hospital, the national referral centre for tuberculosis,
and one in a community based project in Pumwani district, and the Bac
terial Molecular Genetics Unit at the London School of Hygiene and Tro
pical Medicine. Objective: To determine whether recurrence of tubercul
osis after 'adequate' treatment was due to reinfection with a differen
t isolate of Mycobacterium tuberculosis or to relapse of the original
infection. Design: A retrospective comparison by DNA fingerprinting of
sets of isolates of M. tuberculosis from patients with recurrence of
tuberculosis and in whom isolates from the original episode had been s
tored was made. Five patients with recurrence of tuberculosis two to n
ineteen months after adequate treatment and documented clearance of di
sease were studied. Results: In one patient, fingerprints of the isola
tes of M. tuberculosis from the recurrence were quite different to tho
se from the original episode; in the other four, the fingerprints were
identical. Conclusion: Reinfection rather than relapse was the cause
of recurrence in at least one patient. The high 'relapse' rates seen i
n HIV-related tuberculosis in Africa may in part be due to increased s
usceptibility to reinfection and not to treatment failure.