Abstract 2376: Elevated Protein Z as a Risk Factor for Unexplained Cerebral Infarction: The THrombophilia In Cryptogenic stroKe (THICK) Study
Background: Protein Z (PZ) is a vitamin K-dependent plasma protein that exhibits both pro-and anticoagulant properties. There are conflicting data as to the role of PZ in cerebrovascular disease, with some studies showing a relation between high PZ levels and ischemic stroke, others highlighting an association for reduced PZ, and still others finding no relation.
Methods: We evaluated PZ in an ongoing prospective case-control study of prothrombotic risk factors for cryptogenic stroke in subjects ages 18 to 64. Cases consisted of patients referred to Weill-Cornell Medical Center (09/2002 – 05/2007) for evaluation of cerebral infarction in whom a probable stroke etiology could not be assigned despite neuroimaging and transesophageal echocardiography. The classification of cryptogenic stroke was based on modified TOAST criteria. Controls included stroke-free volunteers recruited from the area. PZ was measured, >2 months post-stroke in cases, using a commercial immunoassay.
Results: PZ level was available in 156 cases and 165 controls. Participants were similar in age [44 (24 – 64) vs. 42 (19 – 64)], gender (49 vs. 56% women) and race-ethnic distribution (15 vs. 13% black, 17 vs. 21% Hispanic), but cases were more frequently hypertensive (24 vs. 13%) than controls, and more often had a family history (27 vs. 18%), and personal history of stroke (5 vs. 0%) or coronary disease (2 vs. 0%). There was no difference in PZ level between cases and controls (1.85±0.8 vs. 1.80±0.6 μg/ml, P=.52). There were also no significant differences in prevalence of low or high PZ levels as defined by cutpoints at the 5th, 10th or 90th percentiles. At the 95th percentile (>2.70 μg/ml), however, significantly more cases than controls had a high PZ (12.8 vs. 4.2%, P=.006). After adjustment for clinical covariates, this association remained significant (OR=3.61, 95% CI=1.44 –9.04), and did not change when cases with prior stroke were excluded.
Conclusions: In this study of younger adults, PZ elevation, but not deficiency, was strongly and independently associated with cryptogenic stroke. Further research is necessary to determine whether this relationship has a causal basis, and to explore the role of PZ as a potential screening and therapeutic target in cerebral ischemia.