A Comparison of Fusion Rates and Risk Factors in a Cohort of 2301 Foot and Ankle Fusions Comparing the Use of No Graft and Graft

Abstract

Category: Ankle, Ankle Arthritis, Hindfoot, Midfoot/Forefoot Introduction/Purpose: Bone graft and bone graft substitutes may affect fusion rates and have been advocated to reduce the risk of nonunion particularly in patients with risk factors for failure of fusion. However historically they have not been used in all patients but tend to be used in patients at risk for nonunion. This paper analyses the union rates of patients with and without graft for a large retrospective cohort in which no graft, autograft and bone graft substitute, and summarizes the risk factors for nonunion. Methods: 2301 fusions performed in 1320 patients were reviewed with a minimum of 2 years follow up to determine the risk of repeat surgery for nonunion. The records were reviewed to determine which patients underwent revision surgery of the primary site for nonunion. Demographic factors and risk factors for nonunion were recorded. The nonunion rate for each joint (ankle, subtalar, talonavicular, calcaneocuboid, navicular cuneiform and tarsometatarsal joints) was determined. The mean age of the patients was 56, BMI 27.83, 32% were male, diabetes was present in 8.9%, smoking present in 10.9%, rheumatoid arthritis present in 10%. No graft was used in 1249 fusions, Autograft in 644, PDGF with TCP (regular) in 113, and PDGF injectable (TCP and collagen) formula in 110. Combinations were used in 159 patients. Odds ratios were used with 95% confidence intervals and p values. Results: No graft was used in 1249 fusions with 34 revisions for nonunion (2.7%) and graft was used in 1052 patients with 59 non unions (5.6%). The odds ratio for nonunion with graft was 2.2x, and was not significant. The odds ratio for PDGF with TCP for nonunion was higher than the PDGF injection, at 5 x versus 1 x. However the difference was not significant. The demographics of patients receiving graft were different than those not receiving graft, and were more likely to be older, male, have a higher BMI, be smokers and have diabetes. The appended figure shows the odds ratios for risk factors for receiving graft. Conclusion: Bone graft substitute and bone grafts result in a similar union rate for patients undergoing foot and ankle fusions. However patients receiving graft have higher risk factors for nonunion compared to those not receiving graft. This paper demonstrates that the concept of using graft for at risk patients is merited, with similar fusion rates being achieved. Larger patient numbers would be required to reach statistical conclusions.