The pattern of vertical ground reaction force redistribution among limbs during episodes of acute synovitis of the stifle in 12 mixed-breed dogs was investigated as an adjunct to a blinded nonsteroidal anti-inflammatory drug efficacy study. Without regard to drug efficacy groupings, the redistribution of vertical forces before and during the acute synovitis episode was evaluated by analysis of gait, using a force platform. Acute synovitis was induced by intrasynovial injection of sodium urate crystals. Simultaneously, each dog was given 1 of 4 treatment regimens, including IV injection of sterile saline solution (as a negative control), phenylbutazone (as a positive control), or 1 of 2 proprietary nonsteroidal anti-inflammatory drugs. Postinjection analyses took place at 2, 4, 8, 12, 24, and 36 hours. The peak vertical force redistribution in the 3 untreated limbs of the dogs was described. The greatest redistribution was observed 4 hours after substance injection when the synovitis was clinically at maximum. Thereafter, there was steady improvement and the dogs had a clinically normal gait 24 hours after substance injection. During synovitis, peak vertical force increased in the contralateral hind limb. During the more severe synovitis episodes, force was decreased in both forelimbs. There was good correlation between severity of lameness and peak vertical force response in the contralateral hind limb. Results of the study indicate that the untreated limbs of the same animal should not be used as a control during acute lameness studies.