Thrombin Generation: Plasma Composition Imbalance and Mortality in HIV

Introduction: Effective treatment of HIV infection with antiretroviral therapy (ART) has prolonged survival and shifted causes of death to non-AIDS defining illnesses such as cardiovascular disease (CVD). Increased thrombin generation is evident from increased D-dimer levels in HIV; previously we have shown that inflammation and viral replication associate with imbalances in pro and anticoagulant factors resulting in increased thrombin generation profiles when mathematically modeled. We explore for the first time the hypothesis that factor compositional imbalance increases simulated thrombin generation and predicts mortality in HIV.

Methods: In a nested case-control study of the SMART and ESPRIT participants randomized to continuous ART, we evaluated cases of all-cause mortality and controls (1:3) matched on enrollment date, age (± 5yrs), study, and country. Thrombin generation in response to a 5 pM tissue factor initiator for each individual was calculated by a mathematical model incorporating levels of factors II, V, VII, VIII, IX, X, antithrombin, tissue factor pathway inhibitor and protein C. Thrombin generation metrics included time to clot, maximum rate (MaxR), maximum level (MaxL) and area under the curve (AUC). Conditional logistic regression was used to compare parameters between cases and controls, adjusted for age, gender, and race/ethnicity.

Results: Among cases (n=129) and controls (n=369), respectively, median CD4 count was 451 and 496 cells/mm³, and 65 and 77% had HIV viral suppression (<400 copies/mL). Levels of antithrombin (median: 125% vs. 128%, p=0.084) and protein C (median: 109% vs. 117%, p=0.005) were decreased and factor V (median: 112% vs 98%, p=0.014), FVIII (median: 145% vs 129%, p=0.006) and TFPI (median: 27 ng/mL vs. 26 ng/mL, p=0.073) were higher in cases versus controls. The overall effect is a more procoagulant simulated phenotype. In mortality cases versus controls, MaxR (median: 0.44 nM/s vs. 0.35 nM/s, p=0.064), MaxL (68 nM vs. 56 nM, p=0.037) and AUC (18 µM*s vs. 14 µM*s, p=0.016) were all more procoagulant, although clot time was unaffected. Among mortality cases thrombin generation (MaxL) was greater among those in the highest versus lowest quartile of IL-6 (p=0.006). In controls thrombin generation (MaxL) did not vary by IL-6 level (p=0.93).

Conclusions: Antithrombin, FV and FVIII, and PC were seen as the major contributors to the increased simulated thrombin generation among HIV+ mortality cases. Our results are consistent with the hypothesis that HIV is associated with an increase in thrombin generation via altered balance of pro and anticoagulant factors, and the resulting coagulopathy is associated with increased mortality.