Assessment of Hyperglycemia, Hypoglycemia and Inter-Day Glucose Variability Using Continuous Glucose Monitoring among Diabetic Patients on Chronic Hemodialysis.

Continuous glucose monitoring (CGM) facilitates the assessment of short-term glucose variability and identification of acute excursions of hyper- and hypo-glycemia. Among 37 diabetic hemodialysis patients who underwent 7-day CGM with the iPRO2 device (Medtronic Diabetes, Northridge, CA, USA), we explored the accuracy of glycated albumin (GA) and hemoglobin A1c (HbA1c) in assessing glycemic control, using CGM-derived metrics as the reference standard. In receiver operating characteristic (ROC) analysis, the area under the curve (AUC) in diagnosing a time in the target glucose range of 70-180 mg/dL (TIR) in <50% of readings was higher for GA (AUC: 0.878; 95% confidence interval (CI): 0.728-0.962) as compared to HbA1c (AUC: 0.682; 95% CI: 0.508-0.825) ( < 0.01). The accuracy of GA (AUC: 0.939; 95% CI: 0.808-0.991) in detecting a time above the target glucose range > 250 mg/dL (TAR) in >10% of readings did not differ from that of HbA1c (AUC: 0.854; 95% CI: 0.699-0.948) ( = 0.16). GA (AUC: 0.712; 95% CI: 0.539-0.848) and HbA1c (AUC: 0.740; 95% CI: 0.570-0.870) had a similarly lower efficiency in detecting a time below target glucose range < 70 mg/dL (TBR) in >1% of readings ( = 0.71). Although the mean glucose levels were similar, the coefficient of variation of glucose recordings (39.2 ± 17.3% vs. 32.0 ± 7.8%, < 0.001) and TBR (median (range): 5.6% (0, 25.8) vs. 2.8% (0, 17.9)) were higher during the dialysis-on than during the dialysis-off day. In conclusion, the present study shows that among diabetic hemodialysis patients, GA had higher accuracy than HbA1c in detecting a 7-day CGM-derived TIR < 50%. However, both biomarkers provided an imprecise reflection of acute excursions of hypoglycemia and inter-day glucose variability.