Thromboembolic disease and human immunodeficiency virus infection.

In a recent issue of Blood, Stahl et all suggested that human immunodeficiency virus (H1V)-associated free protein S deficiency might predispose infected patients to thromboembolic complications. Very few published studies deal with thrombosis during HIV in fe~ t ion .~ ,~ In a previous study: we found free protein S deficiency, with normal C4bBP levels, in 14 of 16 (87%) HIV-infected patients, an incidence close to that reported by Stahl et al. Two of our patients had a history of venous thrombosis. Because of these findings we recently conducted a study on thrombosis in HIV infection. We have retrospectively reviewed the charts of HIV-infected patients treated in three acquired immunodeficiency syndrome (AIDS) units in Pans area from 1983 to 1992. Twenty patients (15 men, 5 women) with 35 thromboembolic episodes were identified. They ranged in age from 28 to 68 years (mean age: 43.2 years). According to the Centers for Disease Control criteria, 1 patient was asymptomatic (group 11), 2 showed persistent lymphadenopathy (group III), 3 had AIDS-related complex (group IVA), and 14 had AIDS (groups IV C and IV D). Their mean CD4+ cell level was 152/mm3 (range IO to 824). Six of the patients had more than one episode. There were 33 episodes of venous thrombosis, including 19 episodes of deep vein thrombosis documented by venogram (1 1) and/or Doppler studies (1 2), and 14 episodes of pulmonary embolism, documented by angiography (8) and/or lung scans (9). In addition, two episodes of cerebral infarcts documented by angiogram were observed. In 25 of 35 episodes, the thromboembolic event occurred in outpatients. All episodes occurred after infection with HIV. Traditional therapy was effective in all patients, except one who required chronic anticoagulation to control rethrombosis. Because of the retrospective design of our study, only partial data on protein S were available. Decreased (range 31% to 57%) free protein S, with normal total protein S and C4bBP levels, was recorded in 6 of 8 patients. Our data confirm that HIV-infected patients may be predisposed to thrombotic complications. Thrombosis may occur without provocation in ambulatory patients. Furthermore, true incidence may be underestimated because signs are subtle in patients with phlebitis and because pulmonary embolism can mimic opportunistic infections (mainly Pneumocystis carinii pneumonia). Clinicians should be alert to unprovoked thrombosis as a possible complication of HIV infection. In cases of unexplained thrombotic complication, HIV testing seems mandatory.