• PURPOSE
    • To determine by systematic review the optimal treatment of scaphoid nonunion.
  • METHODS
    • We conducted a systematic review of the literature with a meta-analysis of proportions to investigate the comparative effectiveness of different surgical techniques.
  • RESULTS
    • A total of 48 publications (1,602 patients) met the eligibility criteria. Vascularized and nonvascularized bone grafts had an estimated union incidence of 92% and 88%, respectively. Distal radius and iliac crest bone grafts had similar union rates (89% and 87%, respectively) but harvesting of iliac crest bone grafts had more complications. Both screw and K-wire fixation had a higher incidence of union (88% and 91%, respectively) than no fixation (79%). No approach was statistically different. Patients fixed with screws were mobilized earlier than those with K-wire fixation.
  • CONCLUSIONS
    • Current evidence does not demonstrate a significantly superior method for the treatment of scaphoid nonunion. A multicenter randomized trial would be ideal but the large numbers that would be required may make this unrealistic. We recommend the continued reporting of series with specific assessments and outcome measures to optimize future comparisons in an attempt to determine the best management of scaphoid nonunion.
  • CLINICAL RELEVANCE
    • The use of bone grafts and the methods of their fixation for scaphoid nonunion are debated issues in hand and wrist surgery, with multiple methods employed. There is no current consensus on optimal treatment. A meta-analysis of proportions of available data from recent studies was deemed the most appropriate way to assimilate the available evidence with the view to inform surgeons of the optimal treatment according to the evidence base.