Is insulin resistance: the link between TG-rich lipoproteins and excess death?


Gerald Reaven MD, Falk CVRC, Stanford Medical Center, Division of Cardiovascular Medicine, Stanford University School of Medicine, 300 Pasteur Drive, Stanford, CA 94305, USA.
(fax: 650-725-1599; e-mail:

Dear Sir,Nordestgaard et al. [1] have been one of the research groups emphasizing the importance of elevated nonfasting triglyceride (TG)-rich lipoproteins as cardiovascular disease (CVD) risk factors. Their recent publication in the Journal of Internal Medicine [2] extended these observations by demonstrating that ‘stepwise increasing levels of nonfasting cholesterol and nonfasting TGs were similarly associated with increasing risk of myocardial infarction’. These observations provide further support for the importance of TG-rich, remnant lipoproteins as risk factors for CVD. Not only were nonfasting TG concentrations comparable with nonfasting cholesterol concentrations as CVD risk factors, ‘only increasing levels of nonfasting TGs, and not levels of nonfasting cholesterol, were associated with total mortality’. Upon further analysis, it appeared that both measures of lipoprotein metabolism were associated with increased mortality from CVD, but only nonfasting TG concentrations were associated with increased mortality from cancer, as well as other causes.

The authors do not speculate upon this difference between the two lipoprotein species and total mortality but suggest that it ‘needs to be explored further in future studies’. In that context I would like to propose a testable explanation, focused on the role of insulin resistance. More specifically, there is a significant relationship between insulin resistance and TG metabolism. For example, the more insulin resistant an individual, the higher is their fasting TG concentration [3–5]. Furthermore, there is a significant relationship between insulin resistance/hyperinsulinemia and TG-rich remnant lipoprotein concentrations [6–10]. In contrast, there is no relationship between insulin resistance and LDL-cholesterol concentration [11].

There is considerable evidence that the metabolic abnormalities and clinical syndromes associated with insulin resistance extend beyond abnormal metabolism of TG-rich lipoproteins, and the phrase ‘insulin resistance syndrome’ was introduced to capture these relationships [12, 13]. The most obvious is the role played by insulin resistance in the aetiology of type 2 diabetes [14, 15]. Although the CVD risk of nonfasting TG-rich lipoproteins was adjusted for the presence of diabetes in their original study [1], I could not discern from the current manuscript if patients with type 2 diabetes were excluded from the analysis, nor could I find any information as to possible differences between the two lipoprotein classes in deaths associated with diabetes. If this information was present, and I missed it, I apologize for raising the possibility that the excess of total deaths in those with increasing nonfasting TG concentrations could be related to an increased prevalence of diabetes.

Death from cancer was increased in those with stepwise increases in nonfasting TG-rich lipoprotein concentrations, and there is accumulating evidence that the risk of developing a number of cancers may be increased in insulin resistant individuals [12, 13]. The evidence supporting this relationship is probably strongest in the case of breast cancer [16–20], but insulin resistance, or the cluster of abnormalities associated with insulin resistance/hyperinsulinemia, has also been associated with a number of other cancers, with colorectal cancer probably with the next strongest evidence of an association [21–24] In this context, the current controversy over the potential oncogenic role of exogenous insulin in the treatment of patients with type 2 diabetes may represent another example of the adverse effect of hyperinsulinemia [25, 26].

In summary, an increase in nonfasting TG-rich lipoproteins is just one of the components of the highly atherogenic lipoprotein profile that includes higher fasting TG and lower high-density lipoprotein cholesterol concentrations, along with smaller and denser LDL particles that characterizes insulin resistant individuals [27]. In addition, other CVD risk factors that are increased in insulin resistant persons include evidence of endothelial dysfunction and a procoagulant state [12, 13, 28]. Thus, I did not find it surprising that nonfasting TG-rich lipoproteins and nonfasting cholesterol concentrations were comparable in their association with CVD. Furthermore, given the protean role of insulin resistance as a risk factor for many clinical syndromes [12, 13], beyond CVD, with diabetes and cancer being prime examples, the presence of insulin resistance may explain why nonfasting TG concentrations were betters predictors of total death than nonfasting cholesterol concentrations. An approach to a preliminary evaluation of this formulation would be for the authors to compare fasting insulin concentrations in the two lipoprotein groups, as well as seeing whether step-wise increases in insulin concentration predicted total deaths. This analysis may lead to nothing useful, but it seems to me a reasonable beginning to examine the very interesting relationship between nonfasting TG-rich lipoproteins and total death described in this excellent study.

Conflict of interest statement

No conflicts of interest to declare.