Abstract: FR-PO554
Effect Modification of GFR and BMI on Insulin Sensitivity Among Nondiabetic Patients
Session Information
- Physical Activity, Body Composition, Metabolism: Clinical
October 26, 2018 | Location: Exhibit Hall, San Diego Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: CKD (Non-Dialysis)
- 1902 CKD (Non-Dialysis): Clinical, Outcomes, and Trials
Authors
- Akwo, Elvis A., Vanderbilt University Medical Center, Nashville, Tennessee, United States
- Alsouqi, Aseel, Vanderbilt University Medical Center, Nashville, Tennessee, United States
- Whitfield, Victoria, Vanderbilt University Medical Center, Nashville, Tennessee, United States
- Ikizler, Talat Alp, Vanderbilt University Medical Center, Nashville, Tennessee, United States
- Hung, Adriana, Veterans Affairs CSR&D, Nashville, Tennessee, United States
Background
Insulin resistance (IR) is highly prevalent in patients with CKD. The reason for this phenomenon is unknown. We studied the interaction between estimated glomerular filtration rate (eGFR) and body mass index (BMI) as determinants of peripheral and central insulin sensitivity (IS).
Methods
In a cross-sectional study of 150 patients, 56 CKD and 94 with normal GFR, we used hyperinsulinemic euglycemic glucose clamp to measure IS (peripheral or skeletal muscle IS) as insulin sensitivity index (ISI) and the homeostasis assessment of insulin resistance (HOMA-IR) (central or liver IS). eGFR estimated by CKD-EPI and body mass index (BMI) were estimated at baseline. Linear regression models with robust standard errors (to relax homoscedasticity assumptions) and interaction terms were used to investigate GFR and BMI as predictors of IS.
Results
The mean age was 53.9 (14.5) yrs; 50.7% were female and 36.7% African-American. Log ISI was positively correlated (r = 0.39, p < 0.0001) with eGFR and inversely correlated (- 0.30, p < 0.0001) with BMI. In multivariable models adjusted for age, sex and race, a 10 ml/min/1.73m2 increase in eGFR was associated with a greater increase in ISI among non-obese (0.48; 95% CI: 0.25, 0.70) compared to obese participants (0.18; 95% CI: 0.02, 0.35) (p-interaction = 0.04). Patients with low GFR had lower ISI (insulin resistant) even with normal BMI (Fig. 1a). Log HOMA-IR was inversely correlated with eGFR (r = - 0.49, p < 0.0001) and positively correlated with BMI (r = 0.52, p < 0.0001). HOMA-IR was higher for persons with lower GFR compared to higher GFR, at any BMI value. For example, at a BMI of 30 and a GFR of 30 ml/min/1.73m2, HOMA-IR was 4.8 compared to 1.2 at a GFR of 120 ml/min/1.73m2 (Fig. 1b).
Conclusion
GFR and BMI are both predictors of IS but the magnitude of the effect of BMI on IS varies across GFR levels and type of IS (peripheral versus central). The effect of BMI on central or liver IS (HOMA-IR) is more pronounced at lower GFR with small changes in BMI translating into greater variations in IS. Conversely, at low GFR, skeletal muscle IS is less affected by BMI.
Funding
- Veterans Affairs Support