Lipoprotein (a) is not significantly associated with type 2 diabetes mellitus: cross-sectional study of 1604 cases and 7983 controls.

AIMS: Lipoprotein (a) (Lp(a)), a well-established risk factor for coronary artery diseases (CAD), would also be anticipated to be associated in a similar manner with risk of type 2 diabetes mellitus (T2DM) based on the common soil hypothesis of etiology of T2DM and CAD. Unfortunately, there remains considerable uncertainty regarding the association of Lp(a) with the risk of T2DM. We aimed to examine the association of Lp(a) with T2DM.

METHODS: Cross-sectional study of 1604 cases and 7983 controls was performed for identifying the association of Lp(a) with T2DM, its possible interactions with risk factors and threshold effects on T2DM. The association of Lp(a) with CAD was also examined and compared within the same study.

RESULTS: On a continuous scale, 10 mg/L higher Lp(a) levels were insignificantly associated with a fully adjusted OR of 1.011, 95% CI 0.961-1.063 for T2DM. On a categorical scale, the fully adjusted ORs for T2DM were 0.733 (0.526-1.022), 0.554 (0.387-0.793), 0.848 (0.612-1.176), 0.727 (0.515-1.026), 0.692 (0.488-0.981), 0.696 (0.492-0.985), 0.719 (0.509-1.016), 0.74 (0.523-1.045), 0.809 (0.571-1.146), and 0.99 (0.962-1.019) for decile 2-10 in reference to decile 1. The magnitude of association did not increase with increasing decile (P for trend test = 0.990). In contrast, higher Lp(a) levels were significantly associated with prevalent CAD on a continuous or categorical scale in a fully adjusted model. No threshold effects were observed in terms of association of Lp(a) with T2DM or with CAD in Lp(a) <50 mg/dL.

CONCLUSIONS: The current study suggested that there was a lack of association of Lp(a) levels with prevalent T2DM. In contrast, Lp(a) levels were significantly associated with CAD in a dose-responding manner. Our findings provided evidence for differential approaches to higher Lp(a) levels in patients with T2DM or with CAD.