Determining insulin requirements for hemodialysis patients with end-stage renal disease (ESRD) is difficult. We performed a randomized crossover study among type 2 diabetes (T2DM) patients with ESRD on continuous hemodialysis and receiving standard insulin for glycemic control. The patients were randomized in 2 groups: daily insulin needed on the day after hemodialysis and a 25% decrease in daily insulin needed on the day after hemodialysis. A total of 51 T2DM patients with ESRD were enrolled. The adjusted-insulin group had higher plasma glucose levels at the 2nd hour of dialysis than those of the nonadjusted-insulin group. Incidence of hypoglycemia per dialysis session (3.3% vs. 0.7%, P=0.02) and symptoms related to hypoglycemia (6.9% vs. 0.7%, P=0.001) were more frequent in the nonadjusted-insulin group. A reduced insulin administration of 25% among T2DM patients undergoing hemodialysis on the day of dialysis was associated with sustained glycemic efficacy and the production of fewer hypoglycemic symptoms. This trial is registered with TCTR20180724002.