METHODS

A retrospective analysis of clinical and radiographic records of cats treated for long bone fractures from 1990 to 2005 was carried out to ascertain the incidence of nonunion. All cats presented during this period were included in the survey, as long as sufficient radiographic data existed to confirm that healing had taken place. Healing was judged to have occurred in accordance with clinical and critically radiographic data. Clinical evidence of bone healing was based on stability, soundness, and lack of pain. For the purposes of this study, any long bone that was not healed on radiographs by 9 months after the original fracture was deemed to be a non-union. Cats operated by first opinion surgeons and second opinion surgeons were included in the analysis. In an effort to provide some enlightenment as to the highest associations with non-union, the data for injury complexity, multiple injuries, and treatment method were concentrated on specifically. Injury complexity specifically referred to open, comminuted, or infected fractures. If there was a significant other injury, in particular another bone fracture, it was also recorded. Treatment method was categorized to high-