The Enigma of Glucose and Lipid Metabolism

How dysglycemia and dyslipoproteinemia intersect, intertwine, and interact in patients with type 2 diabetes mellitus (T2DM) is, as Sir Winston Churchill put it, “like a riddle, wrapped in a mystery, inside an enigma.” Although T2DM is defined primarily as the dysregulation of serum glucose metabolism, we now understand that the dysregulation of lipoprotein metabolism may be, at least with regard to the rate at which vascular disease develops, just as prominent a feature of this disease process.¹ The apparent role that statins play in increasing the risk for T2DM only makes the story more convoluted and complex.

In this issue of JAMA Cardiology, Dugani et al² provide further evidence as to how interconnected glucose and lipid metabolism are. Ridker et al³ first called attention to the association between statin therapy and the risk for new-onset T2DM in the Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER study). Subsequently, Ridker et al⁴ reported that most of those patients affected were already on a metabolic trajectory toward T2DM and that statins likely only accelerated this transition. In the present study, Dugani et al examine the relationship of the lipoprotein insulin resistance (LPIR) score,⁵ a weighted composite of 6 measures of lipoprotein particles generated by nuclear magnetic resonance, with the incidence of T2DM in patients randomized to rosuvastatin therapy or placebo in the JUPITER study. The LPIR score is an advanced version of the ratio of triglycerides to high-density lipoprotein (HDL) cholesterol, which with triglycerides and apolipoprotein B (ApoB) have previously been shown to be significant predictors of the risk for incident T2DM.⁶

The authors find that the baseline hazard ratio for T2DM in the highest tertile of the LPIR score was substantially higher than that of the lowest tertile of the LPIR score in the placebo (4.08; 95% CI, 2.53-6.57) and statin (4.97; 95% CI, 3.16-7.82) groups. This relationship remained after adjustment for systolic blood pressure, body mass index, and levels of high-sensitivity C-reactive protein, hemoglobin A_1c, HDL cholesterol, low-density lipoprotein (LDL) cholesterol, triglycerides, and ApoB. Although statin therapy reduced LDL particles by 49.0%, reduced very low-density lipoprotein (VLDL) particles by 19.6%, and reduced LDL size, the LPIR score was not significantly affected by statin therapy. Thus, the association between LPIR and new-onset T2DM was similar in the treated and untreated groups. Therefore, the aggregate LPIR score masked the effect of statins on its individual components or was associated with the cases of T2DM that were fated to occur with or without statin therapy rather than with the smaller number of cases that occur consequent to statin therapy. This finding would indicate that LPIR cannot fully explain statin-induced T2DM and that some other factor must be responsible.

The LPIR score was designed to correlate as closely as possible with glucose clearance and insulin resistance,⁵ both of which are known to be in the pathophysiologic pathway of T2DM. Therefore, LPIR not surprisingly predicts incident T2DM in a model that does not adjust for these factors. Indeed, as the authors acknowledge, this analysis confirmed that LPIR is a marker of the development of T2DM, but does not elucidate whether LPIR is involved in the causal pathway of T2DM.

From a pathophysiologic perspective, the rate at which the liver secretes triglycerides can account for most of the features of the LPIR score. This process might work as follows.⁷ The number of large VLDL particles and the mean VLDL size are the 2 most heavily weighted inputs into the LPIR score. Each VLDL particle is composed mostly of triglycerides and as the mass of triglycerides secreted by the liver increases, so does the number of large VLDL particles and the mean VLDL particle size. Furthermore, an increase in triglyceride secretion from the liver also affects the number of large HDL particles and HDL size and the number of small LDL particles and LDL size, which are the other components of the LPIR score. This effect occurs because the core lipids of the VLDL particles, which are mainly triglyceride, exchange with the core lipids of HDL and LDL particles, which are mainly cholesterol ester. Thus, as triglyceride secretion from the liver and the amount of triglyceride in VLDL particles increase, more triglycerides are transferred to HDL and LDL particles in exchange for cholesterol ester. Once in the HDL and LDL particles, the triglycerides are hydrolyzed. As a result, the HDL and LDL particles now have less cholesterol ester than before and are smaller, which increases the LPIR score. Thus, most of the features that compose the LPIR score can be traced back to increased hepatic secretion of triglycerides.

Hepatic triglyceride secretion, in turn, is related to hepatic triglyceride synthesis, which is driven by the rates of uptake and synthesis of glucose and fatty acids. Fatty acid flux to the liver is related to the release of fatty acids from adipocytes, a process that is negatively regulated by insulin. Excess glucose in the liver drives fatty acid and triglyceride synthesis, and the rate at which large triglyceride particles are secreted by the liver is insulin sensitive. The net result is that as sensitivity to insulin is reduced, which is to say, as the threshold into T2DM is being crossed, the normal inhibitory effects of insulin on fatty acid release from adipocytes and lipid synthesis in the liver are reduced, and hepatic triglyceride synthesis and secretion increase.⁸

Accordingly, glucose, fatty acids, triglyceride lipoprotein particles, and insulin intersect, intertwine, and interact at multiple metabolic crossroads in multiple tissues, which is one reason why the pathogenesis of T2DM remains obscure. The LPIR score is one outcome in this convoluted array of metabolic processes, and we look forward to future research to determine whether LPIR is a cause or a consequence of the pathways that lead to T2DM.

Corresponding Author: Allan D. Sniderman, MD, McGill University Health Centre, 1001 Blvd Décarie, Montréal, QC H4A 3J1, Canada (allansniderman@hotmail.com).

Published Online: April 13, 2016. doi:10.1001/jamacardio.2016.0183.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Pencina reports receiving personal fees from McGill University Health Center Doggone Foundation during the conduct of the study and grants from Regeneron/Sanofi outside the submitted work. No other disclosures were reported.