occurred. Study population serum fructosamine was measured as a secondary variable. Pre-treatment means were compared to values at Days 30, 60, and 180 to determine if a significant reduction in fructosamine occurred.
11. Statistical Methodology: The measure of effectiveness was the change of the primary variables (VAS, blood glucose and blood glucose nadir) during the study compared with Day 0. For animals that withdrew from the study prior to completion, missing data for the primary effectiveness variables were populated using a last observation carried forward (LOCF) method. The endpoints were analyzed as a repeated measures general linear mixed model using the MIXED procedure of SAS Version 9.1.3.
The two-sided 90% confidence intervals on the proportion of cats whose average blood glucose was below 300 mg/dL and whose blood glucose nadir was below 200 mg/dL were constructed based on the binary response of whether or not a cat’s glucose was below the threshold and were analyzed as repeated measures generalized linear mixed models using the GLIMMIX procedure of SAS Version 9.1.3.
Four cats (5%) went into diabetic remission during the study. One cat went into remission prior to Day 60 and three between Day 60 and Day 180 of the study. A significant reduction of visual analogue scale score was observed when comparing Day 0 to Days 7, 14, 30, 60, and 180. The visual analogue scale score improved by Day 60 as summarized in Table 7.
Change: Day 0 – Day
*LS = Least Square
Blood glucose curve mean and mean nadir values between Day 0 and Days 7, 14, 30, 60 and approximately 180, were compared and tested for significant difference. The number and percentage of cats with blood glucose <300 mg/dL and the number and percentage of cats with blood glucose nadir <200 mg/dL at each time point were analyzed.
Analysis of visual analog scale