In this trial, the effect of MSCs in lame OA dogs was investigated and quantified by means of an objective evaluation with a force platform. The ground reaction forces-related aspects of the gait, such as the PVF and VI, which represent maximal weight bearing and distribution of forces through time, respectively, measured the clinical impact of MSC treatment on the function of the limb during the stance phase of walking.
The absence of a direct relationship between radiographic evidence of OA and force platform findings is well known; in any case, in our study, diseased dogs were selected both on the basis of the presence or absence of radiographic evidence of severe OA (D-E degrees of hip dysplasia) and for evident lameness objectively determined by platform gait analysis .
Voss et al.  and Evans et al.  reported that force platform gait analysis at trot was much more sensitive than at walk for low-grade hindlimb lameness, but not for severe lameness. In our case, lameness of the OA dogs group was evident by direct observation, even at walk.
Although each dog had bilateral lameness, we believe that confident data could only be obtained from the more-lame limbs (lesser PVF), in order to limit a possible bias caused by inconsistent weight redistribution to the less affected contra-lateral hindlimb. In fact, mean values showed how initially less-lame limbs seemed to be “better” than limbs from the control group.
We observed a substantial improvement both in PVF and VI values through the period of evaluation, with an indicative gaining of limb function. Evolution of VI seems to be something controversial since some authors suggest that recordings of improvement in VI may suffer from a delay; their dogs tended to decrease their stance time in comparison with baseline values after non-steroidal anti-inflammatory drug treatment because they were able to improve their performance. However, other authors affirm that stance time did not change or increase when limb function improved [18–20]. In our study we observed independent evolution of both PVF and VI values, we hypothesize that this fact occurs because PVF only depends on the force exerted by the limbs, whilst VI reflects the evolution of the force during the whole support phase. For this reason VI could vary when one or both variables change (force and support time).
In 1999 Budsberg et al. compared the efficacy of a treatment(Etodolac) with a placebo control group and demonstrated a decrease in lameness in the treatment group using force platform analysis. In contrast, the degree of lameness in the dogs receiving the placebo deteriorated during the study period, making the control group unable to provide fixed reference data . Based on their results, we designed the experimental study using a control group of sound dogs that were able to provide fixed reference data.
Different adipose tissue donor sites have been found in revised literature: retroperitoneal adipose tissue , lateral thoracic area , gluteous fat  or inguinal region [5, 25]. We preferred this location to others because it is easier to access, abundant quantities of fat, absence of surgical complications and production of a non visible scar.
Regarding cartilage healing, Murphy et al. published in 2003 that the use of MSCs deposited in a fibrin matrix would be useful . However, although a recent equine study demonstrated early benefit, no significant differences were noted when MSCs plus fibrin were compared to fibrin alone at 8 months . Based on that work, it is likely that modulation of the matrix or cells will need to be accomplished to observe long-term benefit of MSCs for cartilage repair. Treatment timing in relation to the degree of pathology could also be a factor contributing to the insignificant results of the equine study. Specifically, because MSCs appear to have a tropism for damaged cells, including fibrillated articular cartilage, it may be that at day 14 (day of treatment) the degree of fibrillation was not enough to allow an MSC treatment effect. However, more observation and studies on cases with more advanced fibrillation need to be conducted to answer this question. Following those results, our study was designed using animals with severe and chronic OA, where chondral degeneration and fibrillation were clearly present. These criteria could explain why our diseased dogs showed highly significant improvements in limb function.
Moreover, our study supports previously published data demonstrating that a single intraarticular administration of ADMSCs associated to PRGF-Endoret decreases pain and lameness in dogs with OA over at least a 6-month period , although our study was conducted in a different joint and our results were supported by objective kinetic data.
A follow up of six months could be considered as a standard for testing the evolution of a medical or surgical treatment, although our dogs always seemed to improve during the first month after treatment. Beyond this time lapse not significant changes were statistically detected. Recent clinical evaluation of this same treatment in a much larger group of animals with osteoarthritis in hip or/and elbow joints are showing that apparently improvement could be prolonged for about 10 month (JM Carrillo, unpublished observations) when dogs seemed to start to worsen. This is a fact that could be contrasted with a biomechanic evaluation of a homogeneous larger group of hip OA dogs of the same breed and should encourage researchers to objectively determine when a new cycle of treatment should be useful to stop a relapse.
Regarding statistical analysis, more complex models could have been considered but we chose this one because it offers an adequate compromise between complexity and ability to represent the relationships between the considered variables [28, 29].