Abstract

Objectives This project aimed to determine whether cranial ischaemic complications at the presentation of giant cell arteritis (GCA) were associated with pre-existing cardiovascular (CV) risk factors, CV disease or genetic risk of CV-related traits. Methods 1946 GCA patients with clinicodemographic data at GCA presentation were included. Associations between pre-existing CV-related traits (including Polygenic Risk Scores (PRS) for CV traits) and cranial ischaemic complications were tested. A model for cranial ischaemic complications was optimised using an elastic net approach. Positional gene mapping of associated PRS was performed to improve biological understanding. Results In a sample of 1946 GCA patients (median age=71, 68.7% female), 17% had cranial ischaemic complications at presentation. In univariable analyses, 10 variables were associated with complications (likelihood-ratio test p≤0.05). In multivariable analysis, the two variables with the strongest effects, with or without PRS in the model, were anticoagulant therapy (adjusted OR (95% CI)=0.21 (0.05 to 0.62), p=4.95×10−3) and age (adjusted OR (95% CI)=1.60 (0.73 to 3.66), p=2.52×10−3, for ≥80 years versus <60 years). In sensitivity analyses omitting anticoagulant therapy from multivariable analysis, age and hypertension were associated with cranial ischaemic complications at presentation (hypertension: adjusted OR (95% CI)=1.35 (1.03 to 1.75), p=0.03). Positional gene mapping of an associated transient ischaemic attack PRS identified TEK, CD96 and MROH9 loci. Conclusion Age and hypertension were risk factors for cranial ischaemic complications at GCA presentation, but in this dataset, anticoagulation appeared protective. Positional gene mapping suggested a role for immune and coagulation-related pathways in the pathogenesis of complications. Further studies are needed before implementation in clinical practice.