METHODS: We conducted a case-control study nested within the population-based HUNT2 cohort in Norway. A total of 58,761 men and women free of known cardiovascular disease were followed for a first myocardial infarction (MI), and during 11.3 years of follow-up, 1587 incident MIs were registered. These cases were compared to 3959 age- and sex-matched controls.
RESULTS: Among MI cases, the median CXCL16 concentration was 9.9 ng/ml (interquartile range 7.2-12.6) compared to 9.6 ng/ml (interquartile range 6.9-12.3) among controls (p < 0.001). Although the difference in median value between cases and controls was small, MI risk was twice as high (OR, 2.08; 95% CI: 1.74-2.50) among participants in the highest quartile compared to participants in the lowest quartile of CXCL16 after adjustment for age and sex. Additional adjustment for serum lipids, body mass index, smoking habits, diabetes mellitus, serum creatinine, and high-sensitivity C-reactive protein attenuated the excess risk by about half, yielding an odds ratio of 1.46 (95% CI: 1.19-1.79).
CONCLUSIONS: Soluble CXCL16 may provide novel information in clinical cardiovascular risk assessment, but its importance needs to be verified in other prospective population studies.