The outcome of a spinal fusion depends upon a multifactorial process.
Therefore, it is difficult and impractical to study individual variabl
es clinically. Animal models have been useful in studying the paramete
rs that influence the spinal fusion process. Despite the knowledge gai
ned from previous animal studies, there exists a large void in our bas
ic understanding of this complex process. The purpose of this update i
s to review the concept of animal models of spine fusion by summarizin
g previous models, identifying strengths and deficiencies, reviewing t
he strategy for selecting a model, and highlighting knowledge gaps sui
table for future research. With recent advances in the knowledge regar
ding gene-specific therapies and the availability of osteoinductive pr
oteins, we have entered a new era of biologic manipulation of bone for
mation. Animal data, however, are only as applicable as the model from
which it is derived, acid future animal models must overcome the defi
ciencies highlighted in this review. A valid animal model will prove e
ssential in studing the use of growth factors for spinal fusion. The u
ltimate goal is that fusion nonunions will no longer be of clinical co
ncern.