Abstract
We examined the effect of serum markers of cholesterol synthesis and absorption on the incidence of type 2 diabetes (T2D) in the randomized Finnish Diabetes Prevention Study (DPS). We also explored a possible interaction of ABCG8 rs4299376 on sterol levels and lifestyle intervention.
We conducted a prospective cohort study including overweight, middle-aged people with impaired glucose tolerance at baseline who participated in the randomized DPS. The primary outcome of the DPS was the diagnosis of T2D based on repeated oral glucose tolerance tests (OGTTs). After active intervention (median of four years, 1994-2001), non-T2D participants were further followed until T2D diagnosis, dropout or the end of 2009. Of these, 340 participants who had β-sitosterol, campesterol, lathosterol and desmosterol measured by gas chromatography-mass spectrometry during the active four-year follow-up and who were not using cholesterol lowering medications were analysed. Surrogate indexes of insulin sensitivity (IS) and secretion were calculated from an OGTT. In adjusted models, plant sterols during the four-year follow-up were associated with lower T2D incidence during the extended eight-year follow-up (HR for 1-SD change in β-sitosterol and campesterol: 0.76 [0.63-0.92], and 0.81 [0.67-0.99], respectively). Lathosterol levels were associated with higher T2D incidence (HR: 1.35 [1.13-1.62]). These associations, though, were not independent of IS. There was an interaction between rs4299376 and study group on β-sitosterol (p = 0.001) and campesterol (p = 0.004) levels during the follow-up.
Markers of low absorption and high synthesis of cholesterol were associated with the risk of developing T2D, mostly ascribed to IS.