Introduction: Cyclooxygenase-2 (cox-2) inhibitors have been shown to delay healing and decrease fracture strength in animal models that heal by secondary bone healing, but we are not aware of any studies examining their effect on primary bone healing. The purpose of this study was to investigate whether cox-2 inhibitors also impair primary bone healing in fractures that are anatomically reduced and rigidly fixed.

Methods: A transverse mid-diaphyseal tibial osteotomy was created in adult male rabbits. The tibia was reduced and internally fixed in compression using a 6-hole 2.7 mm dynamic compression plate. Animals were randomized to receive Rofecoxib 12.5 mg/day orally, or placebo for four weeks postoperatively. Animals were sacrificed at 4 weeks. Fracture healing was assessed with radiographs, histology, and mechanical testing.

Results: No differences were identified between the placebo and the Rofecoxib treated animals on histologic evaluation. There were no differences in the maximum torque or stiffness between the placebo and the Rofecoxib treated animals at four weeks. Eighteen animals were excluded from the study because of hardware failure. These failures were primarily acute fractures that occurred during animal handling for medication administration.

Conclusion and Discussion: The administration of a cox-2 specific inhibitor immediately after rabbit tibial osteotomy did not impair primary bone healing at the dose administered in this study. Subsequent reporting of major cardiovascular risks has lead to a withdrawal of rofecoxib from the market, and a marked decrease in the use of other cox-2 inhibitors. While there may be other reasons to avoid the use of cox-2 inhibitors in certain patients, the necessity for primary (direct) bone healing need not be a contraindication for their use. The high complication rate with this animal model leads us to discourage its use for other studies examining primary (direct) bone healing.