Insulin Resistance in Relation to Lipids and Inflammation in Type-2 Diabetic Patients and Non-Diabetic People

Abstract

Background: We demonstrated in experimental studies that hypercholesterolaemia enhances the proliferation of haematopoietic stem cells and the subsequent differentiation to neutrophils, whereas HDL-cholesterol inhibits these processes. To translate our experimental findings to clinical practice, we investigated in Chinese type-2 diabetic patients and in Flemish non-diabetic people the independent and joint associations of insulin resistance with markers of dyslipidaemia and inflammation, while looking for consistency between ethnicities and across the spectrum of insulin resistance.

Methods: We studied 798 Chinese patients with type-2 diabetes (53.6% women; mean age, 60.6 years) admitted to a tertiary referral centre and 1060 white Flemish (50.5%; 51.1 years) randomly recruited in Northern Belgium. Fasting insulin resistance (HOMA-IR) was derived from C-peptide in Chinese and from insulin in Flemish using the Homeostasis Model of Assessment algorithm. In multivariable-adjusted analyses, HOMA-IR was regressed on triglycerides, HDL-cholesterol and neutrophil count.

Results: In Chinese patients, the percentage changes in HOMA-IR associated with triglycerides, HDL-cholesterol and neutrophils (per 1-SD increment) amounted to 8.1 (95% confidence interval, 3.0 to 13.4; p = 0.0015), -8.7 (-13.0 to -4.2; p = 0.0002) and 5.6 (1.0 to 10.4; p = 0.017). In non-diabetic Flemish, the corresponding estimates were 11.7 (8.3 to 15.1; p<0.0001), -1.7 (-4.6 to 1.4; p = 0.28) and 3.3% (0.5 to 6.3; p = 0.022), respectively. None of the interaction terms between the three explanatory variables reached significance in Chinese or Flemish (p≥0.10).

Conclusions: Insulin resistance increases with the serum level of triglycerides and the blood neutrophil count, but decreases with serum HDL-cholesterol concentration. These associations were consistent in Chinese type-2 diabetic patients and non-diabetic Flemish people and were independent from one another. The clinical implications are that future studies should focus on intervening with serum triglyceride and HDL-cholesterol levels or controlling inflammation as a way to prevent or treat insulin resistance.

Fig 1

Serum triglycerides (A, D), serum HDL-cholesterol (B, E) and neutrophil count (C, F) by thirds of the distribution of insulin resistance. Insulin resistance was computed by Homeostasis Model Assessment algorithm (HOMA-IR), using C peptide in diabetic Lu Hepatients (A, B, C) patients and insulin in non-diabetic FLEMENGHO participants (D, E, F). Significance of the difference with the adjacent lower third: ɫ p≤0.01; § p≤0.0001.

Fig 2

Insulin resistance (A) and C-peptide or fasting insulin (B) in relation to serum lipids and neutrophil count in Lu He patients and FLEMENGHO participants. Insulin resistance was computed by Homeostasis Model Assessment algorithm (HOMA-IR), using C peptide in diabetic Lu He patients (■) and fasting insulin (●) in non-diabetic FLEMENGHO participants. Estimates express the percentage change in the dependent variable associated with a 1-SD increment in the explanatory variables. All estimates were standardised to the average in each population (mean or ratio) of age, body mass index, mean arterial pressure, and use of antidiabetic medications (by class), lipid-lowering drugs (by class), antihypertensive drugs (by class), and nonsteroidal anti-inflammatory drugs (see S3 Table). The three explanatory variables were jointly entered in the models.

Fig 3. Insulin resistance as function of the serum levels of triglycerides and HDL-cholesterol and the neutrophil blood cell count.

Insulin resistance was computed by Homeostasis Model Assessment algorithm (HOMA-IR) and plotted along the vertical axis. Serum triglycerides and HDL-cholesterol are plotted along the horizontal axes. The colour gradient in the modelled two-dimensional plane represent the gradient of the neutrophil blood cell count. All estimates were standardised to the average in each population (mean or ratio) of age, body mass index, mean arterial pressure, and use of antidiabetic medications (by class), lipid-lowering drugs (by class), antihypertensive drugs (by class), and nonsteroidal anti-inflammatory drugs (see S3 Table).