Statistical analysis

Hematology data for days where control dogs were sampled were evaluated by a nested repeated measures ANOVA of each hematologic parameter accounting for treatment group, day and treatment group by day interaction. For models of hematology outcome with a significant effect of treatment group (p<0.05), individual ANOVA models were evaluated for each day and if the treatment group effect was significant, day specific contrasts of the treatment groups to the control group were performed using a Bonferroni adjustment to account for multiple comparisons. Treatment group to control comparisons were considered different if the Bonferroni p value was less than 0.05. Commercial software (Stata Corp LP, College Station, TX) was used to perform these analyses. A two-tailed unpaired Student’s t test was performed for ELISA results using statistical software (GraphPad software, http://www.graphpad.com/, La Jolla, CA).