Acarbose-metformin is more effective in glycemic variability control than repaglinide-metformin in T2DM patients inadequately controlled with metformin: a retrospective cohort study

BACKGROUND: Acarbose and repaglinide are widely used either by themselves or in combination with other medications. However, their efficacy in diabetes control has not been compared when used in combination with metformin. METHODS: The present study aimed to compare their effects on glycemic variability (GV) control when taken with metformin for type 2 diabetes mellitus (T2DM) inadequately controlled with metformin alone. In this retrospective cohort study, T2DM patients who were treated with either acarbose-metformin or repaglinide-metformin combination were recruited. Either acarbose 100 mg or repaglinide 2 mg triple daily was taken for the subsequent 12 weeks in combination with metformin. Demographic data, biochemical data and 7-point glycemic self-monitoring conducted with capillary blood (SMBG) data were reviewed after one week and 12 weeks. The primary outcome including glucose control and changes in GV as well as other factors affecting GV and the incidence of hypoglycemia were also analyzed. RESULTS: Of the 305 T2DM patients enrolled, data from 273 subjects, 136 in the acarbose-metformin group (M+A) and 137 in the repaglinide-metformin group (M+R) were analyzed. Both regimens improved glycemic control at 12 weeks post commencement of new medications. GV, expressed as the mean amplitude of plasma glycemic excursions (MAGE, 5.0 ± 2.6 vs. 2.8 ± 1.6 mmol/L, p < 0.001 in M+A; 5.1 ± 2.5 vs. 2.9 ± 1.3 mmol/L, p < 0.001 in M+R), standard deviation of blood glucose (SDBG, 3.6 ± 1.3 vs. 2.0 ± 0.9 mmol/L, p < 0.001 in M+A; 3.7 ± 1.3 vs. 2.4 ± 1.3 p < 0.001 in M+R), coefficient of variation of blood glucose (CVBG, (0.30 ± 0.09 vs. 0.21 ± 0.1, p < 0.001 in M+A; 0.31 ± 0.09 vs. 0.24 ± 0.12, p < 0.001 in M+R), postprandial amplitude of glycemic excursions (PPGE, 5.2 ± 2.6 vs. 2.8 ± 1.6 mmol/L, p < 0.001 in M+A; 5.3 ± 2.5 vs. 2.9 ± 1.3 mmol/L, p < 0.001 in M+R) or largest amplitude of glycemic excursions (LAGE, 9.8 ± 3.6 vs. 5.4 ± 2.4 mmol/L, p < 0.001 in M+A; 10.1 ± 3.4 vs. 6.3 ± 3.2 mmol/L, p < 0.001 in M+R) decreased significantly after the addition of acarbose or repaglinide (p < 0.05 respectively). Compared with repaglinide-metformin, acarbose-metformin was more effective in GV control at 12 weeks post commencement of new medications (p < 0.05). This study indicates that both acarbose-metformin and repaglinide-metformin combinations could effectively reduce GV and the acarbose-metformin combination seems to be more effective than the repaglinide-metformin combination. However, this conclusion should be confirmed by future large-scaled and more comprehensive studies due to the limitations of the present study.