Outpatient Discharge Therapy With Saxagliptin+MetforminXR vs GlipizideXL for Type 2 Diabetes With Severe Hyperglycemia

• The proportion of responders in each arm. Responder: FBG 70-300 and/or PPBG <400 mg/dl (week1-6), FBG 70-250 and/or PPBG <300 mg/dl (week 7-12) and without metabolic exclusion criteria, repeat ED visits, hospitalization or significant hypoglycemia. [ Time Frame: 12 weeks ] [ Designated as safety issue: Yes ]

  Non-responder:1 FBG >300 and/or PPBG >400 mg/dl (week 1-6) and FBG >250 and/or PPBG >300 mg/dl in 4 consecutive readings or more (week 7-12).

  2. A single confirmed BG of >450 mg/dl. 3. Significant hypoglycemia: Single episode of hypoglycemia with BG < 50 mg/dl or 2 episodes of BG between 50 and 70 mg/dl within 7 days or any episode of symptomatic hypoglycemia.

  4. Persistently positive large ketones in urine and/or electrolyte imbalances. 5. Revisit to ED or admission to hospital because of hypoglycemia or uncontrolled hyperglycemia.

  
  

• Proportion of patients achieving FBG goal of 70-130 mg/dl at 12 weeks in the 2 treatment arms [ Time Frame: 12 weeks ] [ Designated as safety issue: No ]

  The rate of decline in BG values (mg/dl) in the two groups over the period of twelve weeks will be analyzed using a mixed model (with random intercept) as a sensitivity analysis.

  The Kaplan-Meier (KM) curves, area under the curve,t-test and chi-square analysis will be used for analysis.

  
  
• Percentage of patients with symptomatic hypoglycemia [ Time Frame: 12 weeks ] [ Designated as safety issue: Yes ]
  Hypoglycemia and hospitalization rates will be compared between the 2 groups using either chi-square or Fisher exact test will be used. Binary logistic regression will be used to further analysis to identify predictors of hypoglycemia.
  
  
• To measure percentage compliance with medication in the two treatment arms. [ Time Frame: 12 weeks ] [ Designated as safety issue: No ]
  Medication compliance will be assessed by pill counting. Each patient will assigned a percentage compliance and the study groups will be compared using independent two sample t-test.
  
  
• The number of fold increase in beta cell function in the 2 arms. [ Time Frame: 12 weeks ] [ Designated as safety issue: No ]
  The early insulin response (EIR) will be calculated as the ratio of insulin to glucose response at 0 and 30 minutes (ΔI30pmol/l/ΔG30mmol/l,). The homeostasis model assessment to assess basal insulin secretion (HOMA-β cell) and insulin resistance (HOMA-IR) will be calculated. The beta cell response to OGTT will be calculated as area under the curve for glucose and insulin at 0, 30 and 60 minutes using the trapezoid rule.
  
  

There is very little information regarding diabetes discharge regimens for patients with recently diagnosed diabetes (<1 year duration) who present with severe hyperglycemia (blood glucose 300-450 mg/dl) to the ED or other clinical settings and who do not need to be admitted.

A combination of Saxagliptin+Metformin XR, could be a potential drug combination to be tested as an initial treatment in these circumstances compared to Glipizide XL which was shown to be effective in our previous study. We expect Saxagliptin to improve beta cell function and decrease glucagon levels as was shown for the DPP-IV class medications and in turn improve blood glucose levels, while Metformin XR may reduce insulin resistance and hepatic glucose output. Such discharge therapy may help to prevent deterioration into acute metabolic complications (DKA or hyperosmolar states) and avoid hospitalization. A high proportion of patients may achieve glycemic targets without significant hypoglycemia as measured by self glucose monitoring and objectively by continuous glucose monitoring system (CGMS). Such an easy regimen may safely bridge the time gap until patients will be seen by their providers.