HIV and risk of venous thromboembolism: a Danish nationwide population-based cohort study

Objective The association between HIV infection and the risk of venous thromboembolism (VTE) is controversial. We examined the risk of VTE in HIV-infected individuals compared with the general population and estimated the impact of low CD4 cell count, highly active antiretroviral therapy (HAART) and injecting drug use (IDU). Methods We identified 4333 Danish HIV-infected patients from the Danish HIV Cohort Study and a population-based age- and gender-matched comparison cohort of 43 330 individuals. VTE diagnoses were extracted from the Danish National Hospital Registry. Cumulative incidence curves were constructed for time to first VTE. Incidence rate ratios (IRRs) and impact of low CD4 cell count and HAART were estimated by Cox regression analyses. Analyses were stratified by IDU, adjusted for comorbidity and disaggregated by overall, provoked and unprovoked VTE. Results The 5-year risk of VTE was 8.0% [95% confidence interval (CI) 5.78-10.74%] in IDU HIV-infected patients, 1.5% (95% CI 1.14-1.95%) in non-IDU HIV-infected patients and 0.3% (95% CI 0.29-0.41%) in the population comparison cohort. In non-IDU HIV-infected patients, adjusted IRRs for unprovoked and provoked VTE were 3.42 (95% CI 2.58-4.54) and 5.51 (95% CI 3.29-9.23), respectively, compared with the population comparison cohort. In IDU HIV-infected patients, the adjusted IRRs were 12.66 (95% CI 6.03-26.59) for unprovoked VTE and 9.38 (95% CI 1.61-54.50) for provoked VTE. Low CD4 cell count had a minor impact on these risk estimates, while HAART increased the overall risk (IRR 1.93; 95% CI 1.00-3.72). Conclusion HIV-infected patients are at increased risk of VTE, especially in the IDU population. HAART and possibly low CD4 cell count further increase the risk.