Estimation Of The Variation In Glomerular Filtration Rate Based On Glycosylated Hemoglobin, Serum Creatinine, And Age, In Type 2 Diabetic Patients With Or Without Chronic Kidney Disease

ObjectiveThe methods currently used to calculate GFR underestimate this measurement in the population of diabetic patients, so there is a need to look for more accurate methods of estimating GFR in this specific population. This study aims to evaluate a predictive model based on the use of HbA1c to estimate the variability of GFR in diabetic patients with or without CKD. MethodsWe analyzed data from diabetic patients belonging to a cohort of prospective follow-up of a renal health surveillance program attached to a Peruvian hospital. The following factors were included in the multiple linear regression model: age, sex, diastolic blood pressure (DBP), systolic blood pressure (SBP), body mass index (BMI), cholesterol, triglycerides, HDL, LDL, serum creatinine, Urinary creatinine, microalbuminuria, hemoglobin, basal glycemia and HbA1c. Results122 patients were included in the analysis. The final multivariate model, which included variation of HbA1c, age and creatinine variation, was very significant (p <0.0001) with an adjusted R2 of 80%. The other variables analyzed were not significant to predict the variation of the GFR despite showing some correlation. ConclusionsThe study shows that HbA1c, age and creatinine variation significantly predict the variation of GFR in diabetic patients with or without CKD and opens the possibility of use as a prognostic tool for this specific population.