Clinical Endocrinology (2009) 71, 189–194 doi: 10.1111/j.1365-2265.2008.03445.x © 2009 The Authors Journal compilation © 2009 Blackwell Publishing Ltd 189 ORIGINAL ARTICLE Blackwell Publishing Ltd Risk factors for cardiovascular disease do not fully explain differences in carotid intima–media thickness between Indigenous and European Australians without diabetes L. Maple-Brown*§, A. Hodge†§, J. Cunningham*, D. S. Celermajer‡ and K. O’Dea† *Menzies School of Health Research and Institute of Advanced Studies, Charles Darwin University, Darwin, †University of Melbourne, Department of Medicine, St Vincent’s Hospital, Melbourne and ‡Department of Cardiology, Royal Prince Alfred Hospital, Sydney, Australia Summary Objective To investigate whether cardiovascular risk factors can explain the higher carotid intima–media thickness (CIMT) in Indigenous compared with European Australians. Design Cross-sectional study in three subgroups. Patients Non-diabetic urban European (n = 86), urban Indigenous (n = 69), and remote Indigenous (n = 60) Australians aged 25–64 years. Measurements CIMT, age, sex, anthropometry, blood pressure, smoking status, fasting glucose and insulin, haemoglobin (Hb)A1c, homocysteine, C-reactive protein (CRP), lipids, urinary albumin and creatinine. Results CIMT and levels of risk factors, except fasting glucose and total cholesterol, worsened across the three groups. Log n fasting insulin [β = 0·022, 95% confidence interval (CI) 0–0·0439], age (β = 0·006, 95% CI 0·004–0·007), gender (female β = –0·005 vs. male, 95% CI –0·084 to –0·026), mean arterial pressure (MAP) (β = 0·001, 95% CI 0·001–0·002) and ethnicity/location [urban Indigenous (β = 0·027, 95% CI –0·010 to 0·064 vs. European); remote Indigenous (β = 0·083, 95% CI 0·042–0·123 vs. European)] explained 41% of variance in CIMT. Significant interactions were seen for ethnicity/location with age (P = 0·014) and MAP (P = 0·018). Age was consistently associated with CIMT across the three populations, and was associated with larger increments in CIMT for the Indigenous subgroups (β = 0·007, 95% CI 0·005–0·009 urban; β = 0·007, 95% CI 0·004–0·010 remote) compared with Europeans (β = 0·003, 95% CI 0·002–0·006) in models including age, sex and MAP. MAP was only associated with CIMT in the remote Indigenous subgroup. Conclusion After adjusting for selected risk factors, CIMT in remote Indigenous participants was still higher than in Europeans. The slope of the association between age and CIMT steepened from urban Europeans to remote Indigenous. (Received 30 June 2008; returned for revision 19 July 2008; finally revised 17 August 2008; accepted 2 October 2008) Introduction Indigenous Australians have rates of ischaemic heart disease mortality 7–14 times higher at ages 35–54 years and overall mortality from diseases of the circulatory system around three times higher than nonindigenous. 1 Consistent with this, we have previously reported that, among both diabetic and nondiabetic participants, carotid intima–media thickness (CIMT), a surrogate for coronary atherosclerosis and marker of cardiovascular disease (CVD) risk 2 of remote Indigenous participants, was higher than that of urban Indigenous, which was in turn higher than urban European. This pattern remained after adjustment for age, gender, HbA1c, total cholesterol and smoking status. 3 In five studies among Indigenous Australians, maleness, diabetes, glucose or haemoglobin (Hb)A1c, blood pressure, waist circumference, high density lipoprotein (HDL)-cholesterol, low density lipoprotein (LDL)-cholesterol, smoking and C-reactive protein (CRP) have been found to be related to CIMT. 3–7 We aimed to see whether these and other risk factors can explain the differences in CIMT between urban and remote Indigenous groups and urban Europeans without diabetes. Research design and methods Participants The study was conducted in three groups without diabetes in the Northern Territory of Australia: urban European, urban Indigenous and remote Indigenous. Recruitment and assessment of the three groups has been described previously. 3 Remote Indigenous participants were part of a larger study on community-based interventions to reduce the risk of diabetes and CVD. Participation was voluntary and all community members aged ≥ 15 years were invited. Anthro- pometric and biochemical methods have been described previously. 8 Fasting biochemistry was performed on all participants. In those §Joint first authors. Correspondence: Allison Hodge, University of Melbourne, Department of Medicine, St Vincent’s Hospital, PO Box 2900, Fitzroy 3065, Australia. Tel.: +613 9288 2676; Fax: +613 9288 3652; E-mail: ahodge@medstv.unimelb.edu.au