Hm. Abdeldayem et al., SITES OF TUBERCULOUS INVOLVEMENT IN PATIENTS WITH AIDS - AUTOPSY FINDINGS AND EVALUATION OF GALLIUM IMAGING, Clinical nuclear medicine, 22(5), 1997, pp. 310-314
The purpose of this study was to review autopsy and gallium scan findi
ngs in two different acquired immune deficiency syndrome (AIDS) patien
t populations who had a confirmed diagnosis of tuberculosis (TB) to id
entify organs involved and accuracy of clinical diagnosis. The first g
roup was comprised of 29 autopsies between January 1982 and December 1
994, including only 18 patients who were diagnosed before death, Organ
s most commonly involved were the lymph nodes (59%), lungs (56%), sple
en (53%), liver (45%), and kidneys (37%). Other opportunistic infectio
ns were present in 18 (59%) of autopsies, with more than one opportuni
stic infection present in 11 (37%) of the autopsies. Lungs were involv
ed in 79% of all autopsies. The second population group included 94 pa
tients with AIDS with a proven diagnosis of TB, only 24 of whom had ga
llium scans in the period between January 1992 and December 1994. Ches
t x-ray results were negative in 4 patients (17%); gallium scan result
s were positive in 16 patients (66%), The reasons for false-negative g
allium scan results were due to anti-tuberculous treatment for periods
varying from 2-21 months in 7 patients or the presence of extra pulmo
nary tuberculosis, The sites of TB involvement in the chest were: lung
parenchyma in 5 patients (19%, 4 in mid and lower lung, and 1 in uppe
r lung fields), There was lymph node involvement in all 16 patients (2
4 locations) with mediastinal involvement in 23%, supraclavicular 23%,
axillary 11%, retroperitoneal 11%, and inguinal region in 4%. We conc
lude that (1) tuberculosis in patients with AIDS behaves similar to pr
imary tuberculosis; (2) the combination of chest x-ray and gallium ima
ging is sensitive for the diagnosis of pulmonary tuberculosis in patie
nts with AIDS; (3) the involvement of mediastinal lymph nodes in galli
um scans in the presence or absence of chest x-ray abnormalities shoul
d raise the possibility of TB involvement in patients with human immun
odeficiency virus; (4) anti-TB treatment decreases the sensitivity of
gallium scan.