Behavior of Locked Intramedullary Implants Spanning the Distal Femoral Growth Plate: Results From an Ovine Model

Introduction: Safe retrograde femoral nail placement across the distal femoral growth plate could broaden fracture and reconstructive management options for skeletally immature patients, especially in resource-poor settings. Previous data suggested up to 7% of the distal femoral physis can be violated using a retrograde implant without subsequent growth arrest or inhibition. When the intramedullary implant is locked in the metaphysis, the bone will grow with the nail passing from the epiphysis to the metaphysis. However, the consequences of this growth across the physis are not well understood. Methods: Retrograde femoral nails measuring 8 mm in diameter with a single metaphyseal interlocking screw were inserted across the distal femoral physis of a single hind limb in ten 3-month-old sheep. Physeal violation from implant placement was calculated based on the ratio of the nail width to the physeal width on anteroposterior and lateral radiographs obtained preoperatively. Four sheep were sacrificed at 5 months, while 6 sheep were sacrificed at 9 months. Both the surgical and contralateral control hind limbs were harvested and femoral lengths measured, followed by histologic analysis at the physis. Results: When compared with the overall area of the distal femoral physis, mean growth plate violation was 5.4% (range, 3.4% to 7.4%). The nail migrated proximally across the physis in all specimens. In the specimens sacrificed at 5 months, the operative limb was a mean of 6.1 mm (range, 3 to 9 mm) shorter when compared with control limbs. In the specimens sacrificed at 9 months, there was a mean of 3.3 mm (range, 2 to 5 mm) of limb shortening. Histologic analysis demonstrated normal-appearing cellular activity at the growth plate adjacent to the path created by the nail in all specimens, with obliquity of the physis and small bony bridging appreciated across the physis in some 5-month specimens but no 9-month specimens. Conclusion: Placement of a retrograde femoral nail with a metaphyseal interlocking screw resulted in a mean violation of 5.4% of the growth plate, with proximal implant migration across the physis in all specimens. Shortening of the operative limb was greater and histologic findings more abnormal in specimens sacrificed at 5 versus 9 months, suggesting resumption of more normal growth plate activity with time. Level of Evidence: Level IV.