Circulating sGPVI and D-dimer are elevated in patients with AAA and are predictive for case status and growth rate in the American case-control study. (A-B) Plasma was assessed for sGPVI (A) and D-dimer (B) from healthy older control patients (n = 100), slow-growing AAAs (<4 mm/y; n = 89), or fast-growing AAAs (>4 mm/y; n = 29). The median (25th, 75th percentile) of sGPVI for controls was 15.8 ng/mL (10.8, 23.2), for slow-growing AAAs was 20.9 ng/mL (14.3, 35.2), and for fast-growing AAAs was 138.2 ng/mL (39.2, 176.4), represented by black bars (∗∗∗P < .001, by Kruskal-Wallis test). The median (25th, 75th percentile) of D-dimer for controls was 256 ng/mL (184, 340), for slow-growing AAAs was 795 ng/mL (551, 1139), and for fast-growing AAAs was 1123 ng/mL (854, 1633), represented by black bars (∗∗P = .002; ∗∗∗P < .001, by Kruskal-Wallis test). (C) The OR and 95% CI for the association between sGPVI and case status (top) and D-dimer and case status (bottom). Both sGPVI and D-dimer were modeled dichotomously (above/below median) and continuously (log base-2 transformed). (D) Unadjusted (blue) and adjusted (red) linear regression for the relationship between sGPVI (n = 118), D-dimer (n = 118), and the growth rate (mm/year) of AAA. Continuous sGPVI and D-dimer are log transformed; analysis restricted to cases only. (E) Comparative ROC analysis using sGPVI and D-dimer to predict the occurrence of fast-growing AAAs using patients with slow-growing AAAs used as reference (∗P = .04, as determined by ROC curve area comparison). (F) The effect of sGPVI adjusted for D-dimer (top) and D-dimer adjusted for sGPVI (middle) on case status was examined using ordinal logistic regression. The combined effect calculated using linear combination of sGPVI and D-dimer (bottom). (G) Unadjusted (blue) and adjusted (red) linear regression for the relationship between sGPVI, D-dimer, and the growth rate (mm/year) of AAA. Continuous sGPVI and D-dimer are log transformed; analysis restricted to cases only (n = 118).