SP220
MEASURED AND ESTIMATED GFR IN SEVERELY AND MORBIDELY OBESE PATIENTS: EFFECTS OF AGE, BMI AND DIABETIC STATUS

Introduction and Aims: The worldwide epidemic of obesity is now recognised as a risk factor for chronic renal disease. Adequate estimation of renal function in these obese patients is thus essential. Currently, previous reports have shown that formulas are not validated in this population when BMI ≥ 35. Therefore, the aim of this study was to analyse the performance of formulas and assess the determinants of glomerular filtration rate in a large cohort of obese population.

Methods: This study included 707 estimated GFR in 605 severely or morbidely obese patients referred to the Department of Renal Function Study at the University Hospital in Lyon between 2003 and 2015 because of suspected renal dysfunction or before organ donation. GFR was estimated with the Chronic Kidney Disease and Epidemiology (CKD-EPI), Modification of Diet in Renal Disease (MDRD), Berlin Initiative Study 1 (BIS1) equations (for patients over 70) and measured with a gold standard method (inulin or iohexol) indexed to body surface area determined by the Dubois and Dubois formula with either actual (mGFRr) or adjusted (mGFRa) body weight. Mean bias (eGFR-mGFRr) and accuracy of eGFR were compared in the whole population and between subgroups of age, BMI, diabetic status as potential factors influencing mGFR or eGFR.

Results: In the whole cohort, bias between eGFR_CKD-EPI and mGFRr and mGFRa were important (10.8-12.0 ± 11.0-12.0 ml/min/1.73m² ). eGFR_CKD-EPI had a better accuracy with mGFRa compared to mGFRr (82% versus 78% respectively, p<0.01). There was no difference of bias or accuracy between eGFR_CKD-EPI and mGFRr or mGFRa when comparing diabetic to non diabetic patients. For patients over 70, bias with mGFRr were lower for eGFR_BIS1 than eGFR_CKD-EPI and eGFR_MDRD (8.1 ± 6.0 versus 9,0±7.9 and 9.98 ± 9.0 ml/min/1.73m² respectively, p<0.01). And when compared to patients below 70, mGFRr and eGFR were significantly lower and bias were less important. For morbidely obese patients defined by a BMI > 40, bias were important (11.4-15.6 ± 12.1-15.5 ml/min.1.73 m²) but not different compared to the subgroup of patients with a BMI between 35 and 40 kg/m².

Conclusions: This study shows that indexation of mGFR with body surface area using adjusted body weight gives less bias with _eGFRCKD-EPI than mGFR scaled with body surface area using real body weight. For obese patients over 70, BIS1 had a better performance than MDRD or CKD-EPI equations when compared to mGFRr. Overall, estimation of GFR in obese population is not accurate whatever the severity of obesity, age or diabetic status.