Biomedical Research

- Biomedical Research (2012) Volume 23, Issue 4

Identifying patients with diabetic nephropathy based on serum creatinine in the presence of covariates in type-2 diabetes: A retrospective study

Diabetes affects more than 170 million people worldwide and the number will rise to 370 million people by 2030. About one third of those affected, will eventually have progressive deterioration of renal function. To estimate progression of renal disease among type -2 diabetic population, with Serum Creatinine (SrCr), in the presence of covariates: fasting blood glucose (FBG), systolic blood pressure (SBP), diastolic blood pressure (DBP) and low density lipoprotein (LDL), duration of disease and age at which diabetes was diagnosed. Retrospective data collected from 132 patients, who were diagnosed as diabetic as per ADA standards with or without diabetic complications. Multiple linear regression (MLR) and logistic regression models are adopted here to estimate and predict SrCr, a well-accepted marker for the progression of diabetic nephropathy (DN). The fitted multiple linear regression models are found to be statistically significant, with p <. 001, Fitted logistic models have 88.5% and 84.7% predictive power to assess the renal disease based on mean values of predictors and last record of predictors, respectively. We conclude from the models, which are based on mean values of records, that high blood glucose and high blood pressure along with duration of diabetes are the main contributors for estimating SrCr and predicting diabetic nephropathy. Similar results are concluded from the models which are based on last records of predictors except that LDL is also a significant factor for estimating renal health and DN.

Author(s): Gurprit Grover, A.K. Gadpayle, Alka Sabharwal

Abstract Full Text PDF