Predictive value of blood glucose variability and muscle mass index change in end-stage renal disease in diabetic nephropathy

Objective To investigate the prognostic value of blood glucose variability and muscle mass index change in diabetic nephropathy patients undergoing hemodialysis.

Methods A total of 46 patients undergoing hemodialysis, with or without type 2 diabetes mellitus (T2DM), were included in the study. Among them, 36 patients had end-stage diabetic nephropathy (ESDN group), while the remaining 10 patients had end-stage renal disease without diabetes mellitus (ESRD group). Demographic characteristics were collected at the time of enrollment. Blood samples were collected, and blood glucose fluctuations were monitored for 72 hours using a continuous glucose monitoring system (CGMS). Fasting blood glucose, glycated hemoglobin (HbA1c), and body mass index (BMI) were analyzed in the ESDN group and ESRD group. Spearman analysis was used to examine the correlation between fasting blood glucose, HbA1c, and muscle mass index change. The predictive value of fasting blood glucose, HbA1c, and BMI for the ESDN group was analyzed using ROC curve.

Results Compared with the ESRD group, the ESDN group had higher fasting blood glucose, HbA1c, and BMI (P ＜ 0.05 to P ＜ 0.01). During the 72-hour observation period, the ESDN group had significantly higher mean blood glucose, standard deviation (SD), maximum and mean amplitude glycemic excursions (MAGE), and the proportion of blood glucose readings ＞13.9 mmol/L. The average blood glucose value calculated from HbA1c did not reflect the actual average blood glucose measured by CGM in either group. Spearman analysis showed a positive correlation between fasting blood glucose, HbA1c, and BMI scores. ROC curve analysis of fasting blood glucose, HbA1c, and BMI in the ESDN group yielded AUC values of 0.874, 0.677, and 0.749, respectively, all of which were statistically significant (P ＜ 0.05 to P ＜ 0.01).

Conclusions Blood glucose fluctuations are greater in ESDN patients compared to ESRD patients. The HbA1c value in ESDN patients is inaccurate and does not reflect long-term blood glucose status. Additionally, fasting blood glucose is correlated with BMI, making it a potential marker for ESDN.