INTRODUCTION: Systemic low-grade inflammation, as measured by high-sensitive C-reactive protein (hsCRP), may contribute to the risk of type 2 diabetes in patients with manifest arterial disease. METHODS: Cohort study in 4072 patients with manifest arterial disease without diabetes. The relation between quartiles of hsCRP and type 2 diabetes was assessed with Cox regression analyses, taking age, smoking and blood pressure-lowering medication and lipid-lowering medication into account. Insulin resistance was estimated with homeostasis model of insulin resistance (HOMA-IR). In exploratory models, adjustments were performed for body mass index (BMI) and visceral and subcutaneous adipose tissue thickness. RESULTS: During a median follow-up of 5·0 (IQR 2·5-8·2) years, 288 subjects developed diabetes. High hsCRP was independently associated with incident diabetes (Q4 vs. Q1 males: HR 1·62; 95% CI 1·06-2·48; females: HR 3·12; 95% CI 1·57-6·21). HOMA-IR at baseline is related to hsCRP plasma levels (Q4 vs. Q1: males: β 0·27; 95% CI 0·19-0·36; females: β 0·35; 95% CI 0·22-0·48). The risk of diabetes associated with hsCRP was abolished in males (Q4 vs. 1 HR 1·23; 95% CI 0·80-1·88) and attenuated in females (Q4 vs. 1 HR 2·32; 95% CI 1·14-4·75) after adding BMI to the model, but not modified by statin use (P for interaction: 0·61). CONCLUSIONS: Patients with manifest arterial disease with high hsCRP plasma levels are at increased risk to develop type 2 diabetes and are more insulin resistant as compared to those with low hsCRP levels. This increase in risk is more pronounced in females than in males and is not modified by statin use.
INTRODUCTION: Systemic low-grade inflammation, as measured by high-sensitive C-reactive protein (hsCRP), may contribute to the risk of type 2 diabetes in patients with manifest arterial disease. METHODS: Cohort study in 4072 patients with manifest arterial disease without diabetes. The relation between quartiles of hsCRP and type 2 diabetes was assessed with Cox regression analyses, taking age, smoking and blood pressure-lowering medication and lipid-lowering medication into account. Insulin resistance was estimated with homeostasis model of insulin resistance (HOMA-IR). In exploratory models, adjustments were performed for body mass index (BMI) and visceral and subcutaneous adipose tissue thickness. RESULTS: During a median follow-up of 5·0 (IQR 2·5-8·2) years, 288 subjects developed diabetes. High hsCRP was independently associated with incident diabetes (Q4 vs. Q1 males: HR 1·62; 95% CI 1·06-2·48; females: HR 3·12; 95% CI 1·57-6·21). HOMA-IR at baseline is related to hsCRP plasma levels (Q4 vs. Q1: males: β 0·27; 95% CI 0·19-0·36; females: β 0·35; 95% CI 0·22-0·48). The risk of diabetes associated with hsCRP was abolished in males (Q4 vs. 1 HR 1·23; 95% CI 0·80-1·88) and attenuated in females (Q4 vs. 1 HR 2·32; 95% CI 1·14-4·75) after adding BMI to the model, but not modified by statin use (P for interaction: 0·61). CONCLUSIONS:Patients with manifest arterial disease with high hsCRP plasma levels are at increased risk to develop type 2 diabetes and are more insulin resistant as compared to those with low hsCRP levels. This increase in risk is more pronounced in females than in males and is not modified by statin use.