Purpose The purpose of this retrospective study was to evaluate the long-term results with a minimum of ten years follow-up of primary Bologna-Oxford (BOX) TAA. Methods Between December 2004 and December 2009, 80 patients (82 ankles) underwent a primary BOX TAA performed by a single senior surgeon, expert in foot and ankle surgery. Pain and functional outcomes were analysed using Visual Analogue Scale (VAS) for pain, American Orthopaedic Foot & Ankle Society (AOFAS) scoring system, Foot Functional Index Disability and Pain (FFI-Disability, FFI-Pain) score for comparative analysis. Results A total of 52 patients (54 implants) in a cohort of 80 (82 implants) were examined at a minimum ten years follow-up. Twenty implants out of 54 underwent implant failure (37 %) and 34 patients were enrolled in the present study. The mean VAS for pain decreased significantly from 8.5 +/- 1.2 to 2.9 +/- 2.2 (p<0.01) and the mean AFOAS score changed from 28.6 +/- 11.8 pre-operatively to 72.7 +/- 16.9 (p<0.01) at last follow-up. Ninety-seven percent was satisfied with a mean FFI-Disability score that improved from 77.6 +/- 19.3 to 26.7 +/- 25.4 (p<0.01) and FFI-Pain score that decreased from 76.2 +/- 14.2 to 31.4 +/- 25.6 (p<0.01). We calculated post-operative alignment using alpha, beta and gamma angles with no difference at long-term follow-up. The survival rate of the implant was 66% at ten years of follow-up. Conclusions Our data suggest that BOX TAA is an implant with a good patient satisfaction rate at long-term follow-up; therefore, it is a valid option to increase the quality of life in subjects with end-stage osteoarthritis; however, long-term survivorship is unsatisfactory when compared to modern knee and hip implant.