Efficacy of Bone Graft on Fusion Rates in Primary Open Ankle Arthrodesis: A Systematic Review

CATEGORY: Ankle Arthritis; Basic Sciences/Biologics; Trauma INTRODUCTION/PURPOSE: Success in ankle arthrodesis has been reported between 74% and 100%. Despite many recent reports with success rates >95%, the modern literature details a precipitous increase in the use of biological adjuvants for primary ankle arthrodesis. These additions raise the cost of the procedure and potentially incur morbidity related to the donor site or adverse graft response. For these reasons the routine use of bone graft has come under scrutiny with no consensus regarding its use. The cost - both direct and indirect - of biological adjuvant should be weighed against its added value to clinical outcomes. This systematic review aimed to assess the efficacy of bone autograft versus no graft on achieving a clinically successful primary open ankle arthrodesis fixated using cannulated screws. METHODS: Included studies were divided into 3 groups based on the use of bone graft: no bone graft (NBG), fibular onlay with local autograft (FOBG), and the use of autograft without fibular onlay (BG). Pooled data analysis was performed, and fusion rates were frequency weighted in order to appropriately judge the impact of each study. Cohen's d was used to evaluate effect size for the use of bone graft. Modified Coleman Methodology Score (MCMS) was calculated to analyze reporting quality. RESULTS: These selected studies described outcomes for a total of 884 patients with mean follow up of 46 months. The reviewed literature was of moderate quality, with an MCMS of 60.6. The weighted mean rate of fusion for the 3 groups was 96% in the NBG group (n=182), 96% in the FOBG group (n = 257), and 95% in the BG group (n = 445), with no significant difference between the groups (p=0.98). Use of bone graft resulted in a Cohen's d effect size of 0.10 (95% confidence interval 0.07- 0.26). Time to fusion for the 3 groups was 13.60 weeks for the NBG group, 14.51 weeks for the FOBG group, and 15.10 weeks for the BG group, with no significant difference between the groups (p=0.87). Number needed to treat was 7 with the event of interest defined as a fusion rate less than 95%. CONCLUSION: Bone graft use in primary open ankle arthrodesis fixated with cannulated screws did not have a significant effect (p=0.98) on union rates and yielded a small effect size. Further data demonstrated that 7 ankle arthrodesis patients need to be treated with bone graft in order to prevent 1 clinically significant nonunion. The current findings are both clinically and economically relevant, as commercial bone graft carries additional cost and autologous graft harvest is not consistently a benign endeavor. With increased cost and potential for complication, bone graft use should demonstrably improve clinical outcomes in order to validate these additional burdens.