Continuous glucose monitoring in diabetes patients with chronic kidney disease on dialysis: a meta-analysis

Background This study evaluates the effectiveness of continuous glucose monitoring (CGM) in diabetes patients who were on routine dialysis. Methods Literature search was conducted in electronic databases and studies were selected by following precise eligibility criteria. Random-effects meta-analyses were performed to estimate a) correlations of CGM with other indicators including glycated hemoglobin A1c (HbA1c), glycated albumin and mean amplitude of glucose excursions (MAGE), and b) mean differences (MDs) in CGM and MAGE values between dialysis and off-dialysis days or during vs before dialysis. Results There was a good correlation between CGM and self-monitoring of blood glucose values (r=0.837 [95% confidence interval (CI): 0.67, 0.92]. Correlation coefficient between CGM and HbA1c was 0.523 [95% CI: 0.422, 604] in diabetes patients on dialysis and 0.592 [95% CI: 0.354, 0.757] in diabetes patients without renal failure. Correlation coefficient between CGM and glycated albumin in diabetes patients on dialysis was 0.544 [95% CI: 0.254, 0.744]. Average CGM and MAGE values on dialysis day were not significantly different from those of offdialysis day in diabetes patients (MDs: -0.40 millimole/liter (mmol/L) [95% CI: -1.06, 0.26]; p=0.23) and MAGE 0.50 mmol/L [95% CI: -0.01, 1.00]; p=0.05, respectively). Compared to predialysis values, average CGM and MAGE values were significantly lower during dialysis in diabetes patients (MDs: -2.11 mmol/L [95% CI: -3.25, -0.97]; p=0.0003 and MAGE -2.24 mmol/L [95% CI: -3.99, -0.50]; p=0.01, respectively). Conclusions CGM is an efficient method of glycemic monitoring in diabetes patients on dialysis. CGM had similar correlations with HbA1c and glycated albumin.