Abstract: Objective: To compare the effects of different methods on the glucose treat-to-target time,glucose fluctuation,hypoglycemia and insulin doses in the type 2 diabetic patients treated by insulin bump, and to find the best method to make the glucose to reach the standand level safely,fast and effectively in the type 2 diabetic patients treated by insulin bump.Methods: Sixty hospitalized type 2 diabetic patients inadequatly controlled by premix insulin treatment were randomly divided into convention group (n=20),Bolus Wizard group (n=20),and Bolus Wizard combined with monitoring(combination) group (n=20) according to the random number grouping method.The insulin doses of the patients in convention group were adjusted according to the glucose monitoring of fingertip and the doctor's experiences; the insulin doses of the patients in Bolus Wizard group were adjusted according to the Bolus Wizard software in insulin bump, and the glucose monitoring of fingertip; the insulin doses of the patients in combination group were adjusted according to the Bolus Wizard software combined with real time continuous glucose monitoring system(RTCGM). The level of fingertip glucose was tested.The standard deviation of blood glucose(SDBG)and largest amplitude of glycemic excursion(LAGE)were used to evaluate the glucose fluctuation of the patients in various groups. The treat-to-target time,glucose fluctuation,hypoglycemia and daily insulin doses of the patients in various groups were recorded.Results: Compared with convention group,the treat-to-target time of the patients in Bolus Wizard group was decreased(t=2.30,P<0.05);compared with Bolus Wizard group,the treat-to-target time of the patients in combination group was decreased(t=3.50,P<0.05).On the 3rd day of treatment, compared with convention group,the SDBG and LAGE of the patients in Bolus Wizard group were decreased(t_SDBG=3.11,t_LAGE=2.54,P<0.05); compared with Bolus Wizard group, the LAGE of the patients in combination group was decreased(t_LAGE=2.47,P<0.05).There were no significant differences in the incidence of total hypoglycemia events (χ²=2.192, P=0.532), significant hypoglycemia events (χ²=2.765, P=0.322) and nocturnal hypoglycemia events (χ²=2.192, P=0.532) among the patients in various groups; there were no significant differences in the average insulin dosage (F=2.102, P=0.131), the non-basic insulin dosage (χ²=2.328, P=0.107) and the percentage of non-basic insulin (χ²=2.104, P=0.131) among the patients in various groups.Conclusion: Bolus Wizard software combined with real-time dynamic RTCGM has better effect in the treatment of type 2 diabetes without increasing the risk of hypoglycemia and insulin dosage.

Key words: insulin bump, real time continuous glucose monitoring, Bolus Wizard, glucose fluctuation, insulin