Management of displaced fractures of the waist of the scaphoid: meta-analyses of comparative studies.

INTRODUCTION: Scaphoid fractures with displacement have a higher incidence of nonunion that can cause pain and reduced movement, strength and function. The aim of this study was to review the evidence available and establish the risk of nonunion associated with management of displaced fractures of the waist of the scaphoid.
METHODS: Electronic databases were searched using the Medical Subject Headings (MeSH) controlled vocabulary (scaphoid fractures, AND'd with displaced, or nonunion, or non-healing or cast immobilisation, or plaster or surgery). At present, there are no randomised, controlled trials or studies comparing fixation to plaster cast treatment of displaced fractures of the scaphoid. The search was therefore limited to observational studies of displaced fractures of the scaphoid treated in a plaster cast (non-operative group) or fixed surgically (operative group). The criterion for displacement was limited to gap or step of more than 1 mm. In the non-operative group, we compared the outcome of displaced and undisplaced fractures of the waist of the scaphoid treated in a plaster cast. In the operative group, contingency table analysis was used to calculate the odds ratio of nonunion with plaster treatment compared to surgery.
RESULTS: In the non-operative group, seven studies were included in a meta-analysis with a total of 1401 scaphoids. Ninety-three percent (1311 scaphoids) of these scaphoid fractures healed in a plaster cast. A total of 207 (15%) of all scaphoid fractures showed displacement of at least 1 mm (gap/step) between fracture fragments. Nonunion was identified in 18% (37/207) of displaced scaphoid fractures treated in a plaster cast. The pooled relative risk of fracture nonunion was 4.4 (95% confidence interval (CI): 2.3-8.7; p=0.00; I(2)=54.3%). In the surgical group, we identified six observational studies in which 157 'displaced' fractures of the scaphoid were surgically fixed. Only two of these fractures did not heal. The odds of nonunion were 17 times higher with plaster cast treatment than surgery.
CONCLUSIONS: Displaced fractures of scaphoid have a four times higher risk of nonunion than undisplaced fractures when treated in a plaster cast, and the patients should be advised of this risk. Nonunion is more likely if a displaced fracture of the scaphoid is treated in a plaster cast.