Absolute risk of venous and arterial thrombosis in HIV-infected patients and effects of combination antiretroviral therapy.

Abstract

Several reports have noted an increased risk of venous thrombosis (VT) and arterial thrombosis (AT) in HIV‐infected patients who received combination antiretroviral therapy [1Sullivan P.S. Dworkin M.S. Jones J.L. Hooper W.C. Epidemiology of thrombosis in HIV‐infected individuals. The adult/adolescent spectrum of HIV disease project.AIDS. 2000; 14: 32-4Crossref Scopus (123) Google Scholar, 2Mary‐Krause M. Cotte L. Simon A. Partisani M. Costagliola D. Clinical Epidemiology Group from the French Hospital DatabaseIncreased risk of myocardial infarction with duration of protease inhibitor therapy in HIV‐infected men.AIDS. 2003; 17: 2479-86Crossref PubMed Scopus (494) Google Scholar, 3The Data Collection on Adverse Events of Anti‐HIV Drugs (DAD) Study GroupCombination antiretroviral therapy and the risk of myocardial infarction.N Engl J Med. 2003; 349: 1993-2003Crossref PubMed Scopus (1494) Google Scholar]. Whether HIV‐infected patients who are not on combination antiretroviral therapy are also at risk of thrombosis is largely unknown. To assess the absolute risk of VT and AT in HIV‐infected patients and the effect of antiretroviral therapy, we reviewed medical records of 519 HIV‐infected patients, who were registered in our hospital from January 1989 through December 2004. HIV infection was documented by HIV‐1 antibody ELISA and confirmatory Western blot analysis or by polymerase chain reaction‐detectable HIV load. VT included deep vein thrombosis (DVT), pulmonary embolism (PE) and vein thrombosis at other sites, established by objective techniques [4Brouwer J.L. Bijl M. Veeger N.J. Kluin‐Nelemans H.C. Van Der Meer J. The contribution of inherited and acquired thrombophilic defects, alone or combined with antiphospholipid antibodies, to venous and arterial thromboembolism in patients with systemic lupus erythematosus.Blood. 2004; 104: 143-8Crossref PubMed Scopus (109) Google Scholar]. AT was diagnosed when a patient had either myocardial infarction (MI), ischemic stroke, transient ischemic attack (TIA), or symptomatic peripheral artery occlusive disease, in accordance with earlier described diagnostic criteria [4Brouwer J.L. Bijl M. Veeger N.J. Kluin‐Nelemans H.C. Van Der Meer J. The contribution of inherited and acquired thrombophilic defects, alone or combined with antiphospholipid antibodies, to venous and arterial thromboembolism in patients with systemic lupus erythematosus.Blood. 2004; 104: 143-8Crossref PubMed Scopus (109) Google Scholar]. For each patient with a thrombotic event, we collected detailed information regarding history of malignancies, opportunistic infections, smoking, family history of thrombosis, medication history, cholesterol and CD4 counts. Patients with VT were tested for thrombophilia, including deficiencies of protein S, protein C, antithrombin, factor (F) V Leiden (FVL), prothrombin G20210A mutation and elevated levels (> 150%) of FVIII: C, FIX: C and FXI: C as previously described [4Brouwer J.L. Bijl M. Veeger N.J. Kluin‐Nelemans H.C. Van Der Meer J. The contribution of inherited and acquired thrombophilic defects, alone or combined with antiphospholipid antibodies, to venous and arterial thromboembolism in patients with systemic lupus erythematosus.Blood. 2004; 104: 143-8Crossref PubMed Scopus (109) Google Scholar]. A total of 519 consecutive patients were enrolled; 77% were male. Median age at HIV diagnosis was 35 years (range, 12–77). VT had occurred in 19 patients (4%); 12 had DVT, five PE and two cerebral vein thrombosis. Median age at onset of VT was 46 years (range, 17–64). AT had occurred in 13 patients (3%), median age 45 years (range, 37–56). Five patients had symptomatic peripheral artery occlusive disease, four ischemic stroke, three MI and one patient had a TIA. The median time from HIV diagnosis to VT was 1 year (range, 0–12) and to AT 5 years (0–9). Median CD4 counts at time of VT was 130 cells μL−1 (2–630) and at time of AT 252 cells μL−1 (4–465). Several risk factors for thrombosis were identified. Of 19 patients with VT, two had their event postpartum, one had a family history with VT in a first‐degree relative, two had malignancies. Of these patients, 11 were tested for thrombophilia. Abnormal findings were observed in nine patients, including free protein S deficiency (n = 7), elevated FVIII:C levels (n = 9) and FVL (n = 2). Of 13 patients with AT, 10 were active smokers, three had a family history, six hypercholesterolemia, and three hypertension. Annual incidences of VT and AT were 0.65% (95%CI, 0.39–0.92) and 0.45% (0.24–0.77), respectively (Table 1). Patients who were 45 years or older had a higher risk of VT or AT than younger patients. Although females showed lower annual incidences of VT and AT compared to males, the differences were not statistically significant. In patients who received combination antiretroviral therapy, the annual incidence of VT was 0.72% (0.39–1.29), vs. 0.58% (0.25–1.14) in patients who did not receive these drugs. For AT, annual incidences were 0.46% (0.18–0.95) and 0.43% (0.16–0.95); relative risk (RR) 1.1 (0.4–3.1). In patients who used a protease inhibitor, annual incidence of VT and AT increased to 0.80% (0.40–1.43) and 0.51% (0.20–1.05), respectively; RR were 1.4 (0.6–3.4) for VT and 1.2 (0.4–3.5) for AT.Table 1Risk of venous and arterial thrombosis associated with age, sex and drug therapyFactorVenous thrombosisArterial thrombosisObservation time* (years)Patients with event (n)Annual incidence (per 100 patient years) (95% CI)Relative risk (95% CI)Observation time (years)Patients with event (n)Annual incidence (per 100 patient years) (95% CI)Relative risk (95% CI)Age (years)All2911190.65 (0.39–0.92)–2905130.45 (0.24–0.77)–12–45213690.42 (0.19–0.80)Reference214450.23 (0.07–0.54)Reference≥ 45775101.29 (0.62–2.37)3.1 (1.2–7.5)76170.92 (0.37–1.89)4.0 (1.3–12.6)SexMale2355170.72 (0.42–1.16)Reference2350110.47 (0.23–0.84)ReferenceFemale55620.36 (0.04–1.30)0.5 (0.1–2.2)55520.36 (0.04–1.30)0.8 (0.2–3.5)Drug therapyNo combination ART138380.58 (0.25–1.14)Reference138060.43 (0.16–0.95)ReferenceCombination ART1528110.72 (0.36–1.29)1.3 (0.5–3.1)152570.46 (0.18–0.95)1.1 (0.4–3.1)Any PI combination1381110.80 (0.40–1.43)1.4 (0.6–3.4)137770.51 (0.20–1.05)1.2 (0.4–3.5)*Period from HIV‐diagnosis or start of combination ART until the first episode of thrombosis or the end of follow‐up.ART, antiretroviral therapy; PI, protease inhibitor. Open table in a new tab *Period from HIV‐diagnosis or start of combination ART until the first episode of thrombosis or the end of follow‐up. ART, antiretroviral therapy; PI, protease inhibitor. Overall, the absolute risk of VT in our patient group was 6.5 times higher than reported in the general population (i.e. 0.1%) [5Nordstrom M. Lindblad B. Bergqvist D. Kjellstrom T. A prospective study of the incidence of deep‐vein thrombosis within a defined urban population.J Intern Med. 1992; 232: 155-60Crossref PubMed Scopus (713) Google Scholar], and comparable with heterozygous carriers of FVL [6Middeldorp S. Meinardi J.R. Koopman M.M. Van Pampus E.C. Hamulyak K. Van Der Meer J. Prins M.H. Buller H.R. A prospective study of asymptomatic carriers of the factor V Leiden mutation to determine the incidence of venous thromboembolism.Ann Intern Med. 2001; 135: 322-7Crossref PubMed Scopus (163) Google Scholar]. Despite our finding that aging is an important confounder, median age at onset of VT was 16 years less than the median age at onset of VT in the community [7Heit J.A. Silverstein M.D. Mohr D.N. Petterson T.M. Lohse C.M. O'Fallon W.M. Melton III, L.J. The epidemiology of venous thromboembolism in the community.Thromb Haemost. 2001; 86: 452-63Crossref PubMed Scopus (535) Google Scholar]. Hence, our patients were at higher risk of VT, irrespective of their age. Decreased free protein S levels were demonstrated in seven of 11 tested patients with VT. This finding is in agreement with previous studies [8Stahl C.P. Wideman C.S. Spira T.J. Haff E.C. Hixon G.J. Evatt B.L. Protein S deficiency in men with long‐term human immunodeficiency virus infection.Blood. 1993; 81: 1801-7Crossref PubMed Google Scholar, 9Bissuel F. Berruyer M. Causse X. Dechavanne M. Trepo C. Acquired protein S deficiency: correlation with advanced disease in HIV‐1‐infected patients.J Acquir Immune Defic Syndr. 1992; 5: 484-9Crossref PubMed Google Scholar]. Moreover, nine of these patients had elevated levels FVIII:C. Because we do not routinely screen for thrombophilia in HIV‐infected patients it remains unclear whether abnormal levels of these proteins contributed to the risk of VT. In the Framingham study, age‐related annual incidences of AT in men were 0.07% (aged < 45 years), 0.15% (45–54 years), 0.26% (55–64 years), and 0.39% (65–74 years) [10Thom T.J. Kannel W.B. Silbershatz H. D'Agostino R.B. Incidence, prevalence, and mortality of cardiovascular diseases in the United States.in: Alexander RW Schlant RC Fuster V Hurst's the Heart. McGraw‐Hill, 1998: 3-17Google Scholar]. The risk of AT in our patients younger than 45 years was threefold higher, whereas it was 2‐ to 6‐fold higher in patients aged 45 years or older. Males almost had a 2‐fold increased risk of VT, compared to females. We have no explanation for this difference, which was also reported from a much larger study [1Sullivan P.S. Dworkin M.S. Jones J.L. Hooper W.C. Epidemiology of thrombosis in HIV‐infected individuals. The adult/adolescent spectrum of HIV disease project.AIDS. 2000; 14: 32-4Crossref Scopus (123) Google Scholar]. We observed a higher risk of VT and AT in patients who were on combination antiretroviral therapy. This was more pronounced when combination antiretroviral therapy contained a protease inhibitor. Thus, our data suggests an additional risk of VT and AT in patients who used combination antiretroviral therapy, particularly when it contained a protease inhibitor. Although not significant in our population, previous sufficiently powered studies showed a similar mild effect of combination antiretroviral therapy on the risk of VT and AT [1Sullivan P.S. Dworkin M.S. Jones J.L. Hooper W.C. Epidemiology of thrombosis in HIV‐infected individuals. The adult/adolescent spectrum of HIV disease project.AIDS. 2000; 14: 32-4Crossref Scopus (123) Google Scholar, 3The Data Collection on Adverse Events of Anti‐HIV Drugs (DAD) Study GroupCombination antiretroviral therapy and the risk of myocardial infarction.N Engl J Med. 2003; 349: 1993-2003Crossref PubMed Scopus (1494) Google Scholar]. The absolute risk of VT in patients not using combination antiretroviral therapy was still approximately 6‐fold higher compared to the general population [5Nordstrom M. Lindblad B. Bergqvist D. Kjellstrom T. A prospective study of the incidence of deep‐vein thrombosis within a defined urban population.J Intern Med. 1992; 232: 155-60Crossref PubMed Scopus (713) Google Scholar]. This result suggests a pathophysiological role for HIV infection in the risk development of VT, which needs to be further unraveled. As most of our patients with AT were active smokers and almost half of them had hypercholesterolemia, the estimated risk of AT should be adjusted for these classical risk factors. However, regression analyses could not be done, as information about these risk factors was often not complete in medical charts. Other studies are needed to show whether this association is fully explained by HIV‐infection itself. The authors state that they have no conflict of interest. J. van der Meer conceived the study idea and all authors contributed to the study design, and the data abstraction and interpretation. W.M. Lijfering undertook the statistical analysis and wrote the manuscript. All authors took part in the revision and the final version of the report.