Incidence and risk factors of allograft bone failure after calcaneal lengthening.

BACKGROUND: Calcaneal lengthening with allograft is frequently used for the treatment of patients with symptomatic planovalgus deformity; however, the behavior of allograft bone after calcaneal lengthening and the risk factors for graft failure are not well documented. QUESTIONS/PURPOSES: (1) What proportion of the patients treated with allograft bone had radiographic evidence of graft failure and what further procedures were performed? (2) What are the risk factors for radiographic graft failure after calcaneal lengthening? (3) What patient factors are associated with the magnitude of correction achieved after calcaneal lengthening?
METHODS: Between May 2003 and January 2014, we performed 341 calcaneal lengthenings on 202 patients for planovalgus deformity, the etiology of which included idiopathic, cerebral palsy, and other neuromuscular disease. Of these, 176 patients (87%) had adequate followup for graft evaluation, defined as lateral radiographs taken before and at least 6 months after the index procedure (mean, 18 months; range, 6-100 months) and 117 patients (58%) had adequate followup for the assessment of the extent of correction, defined as weightbearing anteroposterior and lateral radiographs taken before and at least 1 year after the index procedure (mean, 24 months; range, 12-96 months). These patients' results were evaluated retrospectively. The Goldberg scoring system was chosen for demonstration of allograft behavior. A score lower than 6 at 6 months after surgery was defined as radiographic graft failure; the highest possible score was 7 points, and this represented graft incorporation with excellent reorganization of the graft and no loss of height. The patient age, sex, diagnosis, graft material, ambulatory status, and use of antiseizure medication were evaluated as possible risk factors, and we controlled for the interaction of potentially confounding variables using multivariate analysis. Additionally, six radiographic indices were analyzed for their effects on the extent of correction.
RESULTS: The mean estimated Goldberg score was 6 (SD, 1.14) at 6 months after calcaneal lengthening with 11 feet (4%) classified as radiographic graft failure (Goldberg score < 6). Of these, four feet (1%) underwent reoperation using an iliac autograft bone resulting from pain and loss of correction. Multivariate analysis showed that the tricortical iliac crest allograft was superior to the patellar allograft (odds ratio [OR], 3.2; 95% confidence interval [CI], 1.1-9.8; p = 0.038) and the possibility of radiographic graft failure was found to increase along with age (OR, 1.2; 95% CI, 1.0-1.3; p = 0.006). Radiographically, the extent of correction was found to decrease with patient age, as observed at the anteroposterior talus-first metatarsal angle (p < 0.001), lateral talocalcaneal angle (p < 0.001), lateral talus-first metatarsal angle (p < 0.001), and relative calcaneal length (p = 0.041).
CONCLUSIONS: Graft failure can occur after calcaneal lengthening using allograft. Our study showed that the tricortical iliac allograft was superior to the patellar allograft, and further studies are warranted to further elucidate the effects of age on radiographic graft failure. LEVEL OF EVIDENCE: Level III, therapeutic study.