Fig. 2

Box plot of the change of Lameness Locator® scores compared to the baseline examination at day 0 and 29. Repeated measures analysis of variance (ANOVA) was used to compare temporal changes in the change of lameness from baseline for each pair of treatment groups (AUTO to FBS and AUTO to ALLO), with horse considered a random effect. These analyses were performed using PROC MIXED, and an autoregressive correlation structure was specified. Treatment group (AUTO, ALLO or FBS), time point, and their interaction were included as factors in the ANOVA. After the first injection, there was a marginal difference in the Lameness Locator® evaluation over time between the AUTO and FBS groups (p = 0.06; FBS had increased lameness) but not between the AUTO and ALLO groups (p = 0.2). After the second injection, there were significant differences between the AUTO and FBS groups (p = 0.046; FBS had increased lameness) and a marginal difference between the AUTO and ALLO groups (p = 0.09; ALLO had increased lameness). There was a significant treatment effect at day 30 between the AUTO and FBS groups (p = 0.0006; FBS had increased lameness). FBS autologous cells not depleted of fetal bovine serum (FBS), AUTO autologous cells depleted of FBS, ALLO allogeneic cells depleted of FBS