You are here
HIV INFECTION AND RISK OF RECURRENT VENOUS THROMBOEMBOLISM: A NATIONAL COHORT STUDY
Casper Rokx1, Jaime Borjas Howard2, Colette Smit3, Ferdinand Wit4, Elise D. Pieterman1, Suzanne Cannegieter5, Willem Lijfering5, Karina Meijer2, Peter Reiss3, Wouter Bierman2, Vladimir Tichelaar2, Bart Rijnders1
1Erasmus University Medical Center, Rotterdam, Netherlands,2University Medical Center Groningen, Groningen, Netherlands,3Stichting HIV Monitoring, Amsterdam, Netherlands,4Amsterdam Institute for Global Health and Development, Amsterdam, Netherlands,5Leiden University Medical Center, Leiden, Netherlands
People with HIV (PWH) are at increased risk of a first venous thrombotic event (VTE). Whether this also translates into more recurrent VTE is unknown. We assessed VTE recurrence rates in PWH and compared these to VTE recurrence rates in HIV uninfected patients.
PWH with a first VTE between 2003-2018 were identified in the ATHENA cohort and compared to HIV uninfected patients with a first VTE in the MEGA cohort in the Netherlands. Provoked VTE were associated with cancer, major surgery, estrogen exposure, immobilization, or plaster cast use for fractures. The primary endpoint was recurrence of VTE following discontinuation of anticoagulant therapy for a first VTE. Multivariable Cox regression was used to estimate the VTE recurrence risk. Kaplan-Meier estimates (KME) of VTE recurrence accounted for death as competing risk and were stratified for provoked or unprovoked first VTE.
Of 201 PWH with a first VTE in ATHENA, 153 had observations after anticoagulant therapy withdrawal. Of these, 126 (95 unprovoked) were in men and 27 events (13 unprovoked) in women. In MEGA, 4005 patients had a first VTE, including 1813 (998 unprovoked) in men and 2192 (363 unprovoked) in women. In PWH, 40 recurrent VTE occurred during 772 person years of follow up (PYFU; median 4.7 years, 5.2/100 PYFU, 95%CI 3.8-7.0). In MEGA, 635 recurrent VTE occurred during 20,215 PYFU (median 6.1 years, 3.1/100 PYFU, 95%CI 2.9-3.4). KME were higher for PWH at 1 year following anticoagulant withdrawal (13% vs 6%), attenuating at 3 (20% vs 11%) and 5 (23% vs 15%) years of follow up. PWH were at higher risk of recurrent VTE during the first year following anticoagulant withdrawal (HR 1.86, 95%CI 1.16-3.01), but not thereafter (HR 1.06, 95%CI 0.65-1.73). KME at 1, 3 and 5 years in PWH and HIV uninfected patients with unprovoked first VTE were 16% vs 9%, 24% vs 17% and 27% vs 24%. Multivariable Cox regression showed that the CD4+ T-cell increase between the first VTE and anticoagulant therapy discontinuation was an independent predictor of a lower recurrent VTE risk (HR 0.73 per 100 CD4+ T-cells increase, 95%CI 0.60-0.89).
PWH were at increased risk of recurrent VTE, which might be driven by HIV-related immune deficiency, inflammation, and associated hypercoagulability. The increased risk attenuated over time, possibly reflecting the gradual recovery of these factors following initiation of effective antiretroviral therapy.