Objective Epidemiological evidence suggests that different lengths of carbon chains might predict cardiovascular disease (CVD) events differently. However, little data exist concerning the effects of specific types of monounsaturated fatty acids (MUFAs) stratified by chain length. Therefore, the study aimed to explore whether the associations of long-chain MUFAs (LC-MUFAs: 16:1n-7 and 18:1n-9) and very long-chain MUFAs (VLC-MUFAs: 20:1n-9, 22:1n-9 and 24:1n-9) with mortality were different among patients with coronary artery disease (CAD).

Methods Erythrocyte membrane fatty acids were measured at baseline in 1320 Chinese patients with CAD (56.2% were newly diagnosed) in the Guangdong Coronary Artery Cohort from 2008 to 2011. Cox proportional hazards models were used to estimate the association of each MUFA with risk of all-cause mortality and CVD mortality.

Results During 4229 person-years of follow-up, 104 deaths occurred, 80 of which were due to CVD. There were no statistically significant associations between overall MUFAs and all-cause mortality and CVD mortality. When we stratified MUFAs, comparing with the lowest quartile, multivariable-adjusted HRs in the top quartile of LC-MUFAs were 0.40 (95% CI 0.21 to 0.75) for all-cause mortality and 0.41 (95% CI 0.20 to 0.85) for CVD mortality, whereas multivariable-adjusted HRs in the highest quartile of VLC-MUFAs were 2.72 (95% CI 1.47 to 5.01) for all-cause mortality and 2.58 (95% CI 1.30 to 5.10) for CVD mortality.

Conclusions This study showed an inverse association between LC-MUFAs and mortality and a positive association between VLC-MUFAs and mortality among patients with CAD. These findings may help explain some of the reported controversial effects of MUFAs.