MON-661 Association of 25-Hydroxyvitamin D Serum Levels with Markers of Glycemic Control in Diabetic and Nondiabetic Patients on Maintenance Hemodialysis

Introduction: In type 2 diabetic patients (T2DM) with end-stage renal disease (ESRD) on hemodialysis (HD), an inverse relationship between 25(OH)D and glycated hemoglobin (HbA1c) have been reported. However, because of anemia and altered red cells turnover, glycated albumin (GA) has been proposed as a more reliable glycemic marker in these subjects. Therefore, we have examined whether an association exists between 25(OH)D and GA in HD diabetic and nondiabetic (ND) subjects. Material and methods: A total of 121 HD (M/F: 69/31%; age 69±15), of whom 43 (35.5%) had T2DM, were analyzed. Median Hb level was 10.2 g/dL (IQR, 9.35-10.9) and median albumin was 3.44 g/dL (IQR, 3.09-3.77). The median dialysis vintage was 29.3 months (IQR, 11.6-67.44). In all subjects, 25(OH)D, HbA1c, and GA were determined on the morning of the dialysis. Results: Median 25(OH)D concentration in the whole study population was 25.5 mcg/mL (IQR, 17.7-33.75), while median HbA1c level was 37 mmol/mol (IQR, 32-43) and GA was 16.5% (IQR, 14.3-20.2). The median serum concentration of 25(OH)D was lower in T2DM group compared to ND (19.0 vs. 27.3 mcg/mL; P=0.002). In the entire cohort, there was an inverse relationship between 25(OH)D and GA (r: -0.255; P=0.005), while there was no significant correlation between HbA1c and 25(OH)D (r:-0.158; p=0.086). The relationship between GA and 25(OH)D levels was confirmed among T2DM patients (r: -0.375; P=0.013) but not in ND (r: -0.036; P=0.751), while HbA1c didn’t significantly correlate with 25(OH)D neither in T2DM (r: -0.255; P=0.099) nor in ND (r: -0.002; P=0.986). Both in the whole cohort as well in T2DM, the independent association of GA and 25(OH)D persisted upon adjustment for age, sex, BMI, dialysis duration, and vitamin D supplementation. By these regression analyses, it was calculated that 10 mcg/mL decline in 25(OH)D was associated with a 2.4% increase in GA. When T2DM individuals were considered, this increase was 75% greater (4.2%). Conclusions: In HD patients, serum 25(OH)D is more strongly associated with GA than HbA1c in both diabetic and nondiabetic subjects. This might reflect the higher reliability of GA in assessing glucose control in this category of subjects. Furthermore, lower serum 25(OH)D concentration was associated with poorer glycemic control among T2DM subjects, but not in ND. Since 25(OH)D deficiency was more prevalent in T2DM than ND, these findings could emphasize the importance of adequate vitamin D repletion in this category of patients. Whether correction of vitamin D insufficiency might affect GA levels remains to be explored.