Context: Circulating insulin-like growth factor (IGF)-I is inversely associated with ischaemic heart disease incidence. Whether this association relates to alterations in plaque growth or stability, and the role of IGF-II and the major binding proteins (IGFBP-2 and -3), is unclear.

Objectives: To test the hypothesis that circulating IGF-I is inversely, and IGF-II is positively, associated with sub-clinical atherosclerosis and plaque stability.

Design, setting and participants: Cross-sectional analysis based on 310 participants in the UK-based Boyd Orr cohort aged 63–84 years. Cohort members from Aberdeen, Bristol, Dundee, Wisbech and London were invited to clinics for fasted venepuncture and arterial ultrasound examination.

Main outcomes: Arterial intima-media thickness (IMT); arterial plaque prevalence; computerised assessment of plaque echogenicity (a measure of stability), undertaken using the grey scale median (GSM).

Results: In total, 269 of 310 (86.8%) participants had at least one carotid or femoral plaque. In models controlling for IGFBP-3, there was a 44% (95% CI: 12%–64%) reduction in odds of any plaque and a 28% lower (0%–48%) odds of echolucent (unstable) plaques per standard deviation (SD) increase in IGF-I. IGFBP-3 was positively associated with plaque instability (odds ratio: 1.38; 0.99–1.93). IGF-II was positively associated (0.05 mm increase per SD; 95% CI: 0.01–0.09) and IGFBP-2 was inversely associated, with carotid bifurcation IMT. Neither IGF-II nor IGFBP-2 were associated with plaque prevalence or echogenicity.

Conclusion: High circulating IGF-I levels may promote arterial plaque stability. IGF-II and IGFBP-2 do not appear to play a role in plaque development or stability.