Abstract 2635: Prothrombotic Disorders, Patent Foramen Ovale and Unexplained Cerebral Infarction: The THrombophilia in Cryptogenic StroKe (THICK) Study
The etiology of ischemic stroke is frequently indeterminate, particularly in younger adults. Although PFO is identified in over half, its pathophysiological significance is uncertain, and the role of thrombophilias remains insufficiently studied. We hypothesized that venous thrombophilias (VTB) are key determinants of PFO pathogenicity in premature brain infarction and that arterial thrombophilias (ATB) are major risk factors for this condition independent of PFO. We undertook a prospective case-control study of hypercoagulability and PFO in younger patients with cryptogenic stroke. Cases comprised patients ages 18 – 64 referred to our center (2002– 08) for evaluation of first cerebral infarction in whom a probable etiology could not be assigned. Controls included stroke-free volunteers from the area. Measurement of ATB (antiphospholipid Ab, homocysteine, Lp(a), FVIII) and VTB (proteins C & S, antithrombin III, FIX, FXI, FV Leiden, FII G20210A) was performed using standard methods. All cases and controls underwent contrast TTE. Cases (n=212) and controls (n=292) were similar in age (44±11 vs 43±11), sex (46 vs 53% women) and ethnicity (39 vs 39% non-white), but cases more frequently had HTN (24 vs 13%), migraine with aura (17 vs 9%) or estrogen use (30 vs 15%). Thrombophilias were more common in cases than controls (49 vs 31%), reflecting higher prevalence of ATB (38 vs 17%), as were multiple thrombophilias (14 vs 4%) and PFO (56 vs 22%) (all P<.001). Among cases with and without PFO, there was no difference in prevalence for any or multiple thrombophilias or ATB, but cases with PFO had two-fold more VTB (18% vs 8%, P=.038). The prevalence of PFO in cases vs. controls with VTB (OR=13.9, P<.001) was significantly greater than without VTB (OR=3.6, P<.001) (P=.048), as supported by a significant interaction (OR=3.5, P=.050) in multivariable models. In this study of younger adults, ATB were significantly associated with cryptogenic stroke independent of PFO, while VTB more than tripled the risk of PFO. Although VTB explain only a fraction of cryptogenic strokes associated with PFO, their detection strengthens the case for paradoxical embolism and may be useful for guiding management, but this requires further study.