Long-term fate and effects of exercise on sternal cartilage autografts used for repair of large osteochondral defects in horses.

Bilateral osteochondral defects (10 mm2 x 3 mm deep) were created on the distal articular surface of the radial carpal bone of ten, 2- to 3-year-old horses. One defect of each horse was repaired, using a sternal cartilage autograft (treated), and the other was left untreated (control). The horses were exercised on a high-speed treadmill at incrementally increased speed and duration over the course of 12 months. Horses were evaluated arthroscopically at 6 to 7 weeks, and clinical examinations were conducted weekly at exercise. Twelve months after surgery, carpuses of each horse were radiographed and clinically examined prior to euthanasia. A gross pathologic evaluation of each joint was conducted, and samples were collected for histologic, histochemical, histomorphometric, and biochemical evaluation. Radiographically, the grafted joints had more extensive evidence of arthropathy, and clinically, 8 of the 10 horses were more lame in the grafted limb. On the basis of histomorphometry, the repair tissue of the grafted defects contained a greater median percentage of hyaline cartilage (45%) than that of control defects (4.5%), and the control defects contained a greater percentage of fibrocartilage (82%) than did grafted defects (28.5%). A greater median percentage of repair tissue stained with safranin-O in the grafted defects (24.5%) than in the control defects (3.5%). On gross pathologic and histologic evaluation, repair tissue of the control defects had better continuity and was more firmly attached to the subchondral bone than was repair tissue of the grafted defects. Repair tissue of the grafted defects had extensive fissure and flap formation. Histologically, subchondral bone reactivity and fibroplasia was extensive in grafted joints. Repair tissue of grafted defects had a greater percentage of type II collagen (mean +/- SEM, 83.5 +/- 2.95%) than did controls (mean, 79.4 +/- 3.87%) that was not statistically significant. Hexosamine content was significantly higher (P < 0.05) in repair tissue of the grafted defect (mean, 28.9 +/- 3.00 mg/g of dry weight) vs control (mean, 20.6 +/- 1.85 mg/g of dry weight). On the basis of this experimental model, sternal cartilage autografts cannot be recommended at this time for repair of osteochondral defects in athletic horses.