The recent literature contains reports of thrombotic episodes occurring in
patients with human immunodeficiency virus (HIV) infection and various abno
rmalities predisposing to a hypercoagulable state have also been reported i
n such patients. To study the incidence of thrombosis in patients infected
with HIV, and to assess the correlation of thrombosis with the degree of im
munosuppression as well as the association with active illnesses and neopla
sms, we reviewed the charts of 131 patients, which include all the patients
with the diagnosis of HIV admitted or seen in the clinic between January 1
, 1993, and January 1, 1998. The diagnosis of thrombosis was based on docum
ented reports of venous plethysmography or venography for deep venous throm
bosis and ventilation-perfusion scan or pulmonary angiography for pulmonary
embolus. Risk factors for thrombotic disease were evaluated including gene
ral risk factors such as family history, ambulatory status, medications, an
d data were also collected regarding CD4 cell counts and the presence of co
ncurrent or remote opportunistic infections, acquired immune deficiency syn
drome (AIDS)-related malignancy or other AIDS-related diseases at the time
of diagnosis of the thrombotic event. We also reviewed the medical literatu
re via MEDLINE and found 45 cases of patients with HIV who developed thromb
oembolic complications. We found thrombotic complications in 9 of 37 patien
ts with a CD4 count less than 200 cells/mm(3) and in 1 of the remaining 94
patients with a CD4 count more than 200 cells/mm3. The difference was signi
ficant, with p = 0.00004, and the estimated odds of an event given CD4 cell
counts less than 200/mm(3) is 29.89 (95% confidence interval). Three patie
nts had abnormalities of anticoagulation proteins. There was a history of o
pportunistic infections in 5 patients and malignancy in 3 patients. Two pat
ients with autoimmune hemolytic anemia (AIHA) secondary to HIV-infection de
veloped PE upon transfusion of packed red blood cells. The results of this
study suggests that AIDS appears to predispose to thrombosis. It also revea
led a significant correlation between thrombotic disease and CD4 counts (<2
00/mm(3)) as well as the presence of opportunistic infections, AIDS-related
neoplasms, or autoimmune disorders associated with HIV such as AIHA, There
fore, clinicians caring for these patients should be aware of thromboemboli
c disease as a possible complication of AIDS. Further studies to elucidate
the mechanisms underlying this abnormal hemostatic profile, the epidemiolog
y, and to answer several questions such as should patients with risk factor
s for HIV infection who develop thromboembolic complications be further eva
luated including tests for HIV are warranted.