Impact of age, body mass index, insulin resistance and proteinuria on the kidney function in obese patients with Type 2 diabetes and renal insufficiency.

OBJECTIVE

To date, several different equations to predict the glomerular filtration rate (GFR) in patients with renal insufficiency have been developed for different patients groups. Our aim was to determine the prognostic factors of GFR in our homogenous patient group of obese, water-loaded patients with Type 2 diabetes and renal insufficiency, since we assumed that the endogenous creatinine clearance (ECC) alone may not be an accurate method to predict GFR.

METHODS

We recruited 46 obese patients (37 men) with Type 2 diabetes and renal insufficiency in our nephrology center in Mettmann (Germany). However, two male patients were excluded from the analysis due to an outlying insulin level or low inulin clearance. The inulin clearance as a measure of renal function performed by the single shot method was compared with the GFR estimated by ECC, Cystatin C, and MDRD formula. Several multiple regression models were built to test the impact of the prognostic factors age, sex, body mass index (BMI), insulin resistance according to the homeostasis model assessment (HOMA), body water (TBW), brain natriuretic peptide (BNP), and proteinuria on the inulin clearance. In the main regression model to predict the inulin clearance by ECC, only the statistically significant prognostic factors of these models were selected, as well as the interaction between GFR predicted by ECC (GFR_ECC) and BMI.

RESULTS

The prognostic factors GFR_ECC, age, BMI, HOMA and proteinuria had a statistically significant impact on the inulin clearance (the gold standard of the GFR) in our patient population (p < 0.05). However, the interaction of GFR_ECC and BMI was not significant (p = 0.06) in our model. The model was validated and considered well-fitted with a coefficient of determination (R2) of 0.69.

CONCLUSIONS

The independent prognostic factors to determine GFR in obese, water-loaded diabetic patients are GFR_ECC, age, BMI, HOMA and proteinuria. However, our model should be revalidated and tested in a larger sample size to probably detect an interaction between GFR_ECC and BMI as an additional prognostic factor.