We performed limited autopsy with histological examination of tissue cores
obtained percutaneously using the Tru-Cut needle and the Jamshidi trocar in
150 adult HIV-positive patients. Data were compared retrospectively with t
he antemortem clinical diagnosis. Eighty-one percent of the patients were m
ale, and 78% were intravenous drug users. Specimens were obtained from the
brain, liver, lung, bone marrow, and kidney of most patients. The main find
ings included liver cirrhosis in 22 cases (associated with HCV infection in
81%), Pneumocystis carinii pneumonia in 21, Cytomegalovirus (CMV) infectio
n in 19, Mycobacterium avium-intracellulaire (MAI) infection in 17, bacteri
al pneumonia in 14, tuberculosis in 12, and lymphoma in 13 cases. Forty-six
(30.6%) patients had at least one clinical diagnosis that was confirmed by
autopsy, i.e., there was 40.6% agreement between pre- and postmortem findi
ngs. Forty-six (30.6%) patients had at least one clinical diagnosis that wa
s not confirmed at autopsy, whereas 41 (27.3%) had at least one AIDS-relate
d or unrelated disease that was not suspected clinically. The results obtai
ned by limited autopsy are principally comparable to those achieved by full
necropsy, with the advantages of decreasing the contagious risk, saving co
st and time, including a rapid final diagnosis, and easily obtaining the co
nsent for postmortem examination so that necropsy studies may be performed
on a larger number of patients, thus contributing to a better understanding
of the spectrum of HIV infection in our environment.