Healing the Index Tibial Nonunion: Risk Factors for Development of a Recalcitrant Nonunion in 222 Patients.

OBJECTIVES:To compare and stratify the healing rates after our index nonunion surgery using contemporary methods of fixation, report the prevalence of recalcitrant non-union, and identify specific demographic, injury, and treatment-related risk factors for the development of a recalcitrant nonunion. DESIGN:Retrospective analysis of a prospectively collected database. SETTING:Academic Level 1 Trauma Center. PATIENTS/PARTICIPANTS:Two hundred twenty-two tibial nonunions treated with internal fixation by a single surgeon. INTERVENTION:Bivariate and multivariate regression analysis were performed to compare healing rates by the type of fixation and graft augmentation and to identify specific demographic, injury, and treatment-related risk factors for the development of a recalcitrant nonunion. RESULTS:Of the 222 patients, 162 (73%) healed as intended and 51 (23%) required 1 or more subsequent interventions to achieve union (96%). Nine fractures (4%) failed to unite. The 60 fractures (27%) that required a subsequent intervention(s) or failed to consolidate were defined as recalcitrant nonunions. There were no statistically significant differences in the recalcitrant rate when we compared plates versus nails or types of bone graft. Risk factors for developing a recalcitrant nonunion were multifactorial and included grade III open fractures, compartment syndrome, deep infection, and 2 or more prior surgical procedures. CONCLUSIONS:Internal fixation remains a successful method of treatment for most tibial nonunions. However, 27% of patients required a subsequent intervention because of failure to heal our index nonunion procedure. Factors that are associated with recalcitrant nonunions were a grade III open fracture, compartment syndrome, deep infection, and 2 or more or more prior surgical procedures. LEVEL OF EVIDENCE:Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.