Graft choice in the management of unstable scaphoid nonunion: a systematic review.

PURPOSE

To systematically compare outcomes between corticocancellous (CC) and cancellous-only (C-only) bone grafts for unstable scaphoid nonunions.

METHODS

The English-language literature was searched using PubMed, the Cumulative Index to Nursing and Allied Health Literature, and the Cochrane Library with the following keywords: scaphoid, nonunion, malunion, unstable, collapse, humpback, dorsal intercalated segment instability, and deformity. Studies whose patient populations were characterized as possessing unstable scaphoid nonunion and/or collapse deformity or who met radiological definitions of such deformity were included. The union rate, interval to union, clinical assessments, strength, range of motion, correction of deformity, pain, and return to activity were analyzed.

RESULTS

Twenty-three studies published between the beginning of 1987 and the end of 2013 met the eligibility criteria and contained 604 patients. The union rate was statistically equivalent for C-only (95%) and CC grafts (92%). However, the frequency-weighted mean interval to union was significantly shorter for C-only (11 wk) than for CC grafts (16 wk). CC grafts were associated with a significantly higher frequency-weighted mean Mayo wrist score (86) than C-only grafts (80), whereas C-only grafting provided significantly greater improvement of wrist flexion. Of the 3 measures of carpal geometry analyzed, CC grafts significantly improved the scapholunate and radiolunate angles more than C-only grafts.

CONCLUSIONS

Based on retrospective data from uncontrolled studies, C-only grafts provide the shortest interval to union for unstable scaphoid nonunions. CC grafts are associated with consistent deformity correction and superior Mayo wrist scores.

TYPE OF STUDY/LEVEL OF EVIDENCE: Therapeutic IV.