#5 The adjustment of GFR by body surface area: a misleading procedure in trials focused on weight reduction and GFR changes

Abstract Background and Aims controversies exist on the benefits of estimated GFR (eGFR) by formulas (more uncertain) of gold standards (more accurate). Another problem is the adjustment of GFR for body surface area (BSA). The scientific background of the adjustment is weak and mostly relies on the fact that GFR always changes in parallel with weight. This is not true for all patients, and the changes of GFR according to weight are highly variable and unpredictable. Also, the need to adjust a value of GFR remains obscure. Finally, the adjustment to a “standard” BSA of 1.73 m2, which is the average BSA of US subjects of the east cost of 1920, seems difficult to justify today. In brief, the adjustment for BSA “artificially” reduces GFR in obese subjects and increases GFR in small and lean subjects. Also, the adjustment for BSA limits the observation of changes in GFR associated with weight loss due to pharmacological treatments, diets, or exercise. We evaluated the impact of the adjustment of BSA in the analysis of GFR changes in 30 patients with chronic kidney disease, obesity, and metabolic syndrome who underwent a 6-month therapy with exercise (EXRED study, abstract n°119). We investigated whether mGFR adjusted for BSA (mGFR-BSA) or eGFR would adequately reflect GFR changes after weight loss, compared to mGFR. Method we included 30 patients from the EXRED study. mGFR with plasma clearance of iohexol was performed at months 0, 3, and 6. On the same day, samples were collected for analytes, including serum creatinine for eGFR (CKD-EPI 2021). BSA was calculated through DuBois-DuBois formula. Changes in mGFR, mGFR-BSA, and eGFR were assessed using the non-parametric statistics. We performed the analysis as follows: Overall—we considered the 30 patients as a whole (Fig. 1A). Subgroups—in the EXRED study, mGFR changes were heterogeneous: over six months, half of the patients (n = 15) showed a significant reduction (>8%) in mGFR (“mGFR-decliners”), whereas the other half (n = 15), with similar weight loss and baseline characteristics, did not (“mGFR-stable”) (Fig. 1B). We compared the original mGFR with the GFR-BSA or eGFR values. Results after 6 months OVERALL (Fig. 1A)—mGFR significantly reduced from 59.3 ± 20.4 to 54.1 ± 19.4 ml/min (p 0.006). mGFR-BSA showed only a modest reduction from 49.7 ± 17.3 to 47.7 ± 17 ml/min/1.73 m2 (p 0.001), lower than with mGFR, P < 0.001. eGFR had a slight significative increase. SUBGROUPS—in “mGFR decliners” (n = 15), mGFR reduced from 62.9 ± 22.2 to 50.5 ± 18.3 ml/min (P < 0.001) (Fig. 1B). According to mGFR-BSA, the same 15 patients still had a significative, although less marked, GFR decline (53.3 ± 19 to 45.5 ± 16.8 ml/min/1.73 m2, P < 0.001) (Fig. 1C). eGFR did not change at all (Fig. 1D). mGFR values were significantly higher than mGFR-BSA and eGFR throughout the study. Conclusion GFR changes following weight loss were not adequately detected by mGFR-BSA or eGFR. The correction for BSA artificially reduces GFR in obese subjects and artificially increases GFR after they lose weight. This is particularly true for the patients who experience a significant GFR decline. In studies where GFR decline is the main outcome and weight loss is the therapeutic approach, the use of gold standard methods such as mGFR unadjusted for BSA is crucial for a correct interpretation of results. Furthermore, the consequences of the heterogeneous GFR response to weight loss need to be better investigated with long-term studies and larger numbers of patients.