Baseline glycaemic control, family diabetes history predict T2DM in chronic kidney disease
Patients with chronic kidney disease appear to be at greater risk of developing type 2 diabetes mellitus (T2DM) compared with the general population, with predictors including poor baseline glycaemic control and family history of diabetes mellitus, a study has found.
The analysis included 1,713 patients (mean age 56.2 years; 52.9 percent male) with reduced glomerular filtration rates enrolled in the Chronic Renal Insufficiency Cohort study. Researchers obtained data on kidney function and damage, fasting blood glucose, haemoglobin A1c (HbA1c), homeostatic model assessment of insulin resistance (HOMA-IR), demographics, family history of diabetes and smoking status, among others.
During a mean follow-up of 6.65 years, 203 patients (11.85 percent) developed T2DM (defined as fasting blood glucose ≥126 mg/dL or prescription of insulin or oral hypoglycaemic agents). The overall T2DM incidence rate was 17.81 cases/1,000 person-years.
Concordance was low between fasting blood glucose and HbA1c levels (κ=18 percent). In multivariable cause-specific hazards modelling, T2DM showed a significant association with fasting blood glucose level (adjusted hazard ratio [aHR], 1.407; 95 percent CI, 1.135–1.744; p=0.002) and family history of T2DM (aHR, 1.498; 1.033–2.172; p=0.03).
The adjusted association of T2DM with HOMA-IR (aHR, 1.442; 1.208–1.721; p<0.001) was comparable to that observed for fasting blood glucose level, while the association observed for HbA1c level was nonsignificant (aHR, 1.183; 0.966–1.449; p=0.1).
Measures of kidney function and damage were not associated with T2DM.
The results provide clear evidence that in a CKD population, measures of glycaemic control (fasting blood glucose and/or HOMA-IR) have a moderate predictive value for the risk of incident T2DM, even among individuals classified as normoglycaemic by fasting blood glucose level, researchers pointed out.
Limitations of the study include the small number of outcome events and predictors being restricted to measures taken at baseline, they added.