Joint fusion is an appropriate surgical treatment option for end-stage arthritis of the ankle joint. The purpose of ankle fusion procedures is to achieve a bony union between the tibia and talus with appropriate alignment to alleviate debilitating pain and improve mobility [1]. To assist in achieving fusion of the tibiotalar joint, bone graft augment is commonly utilised to facilitate bony union. Despite the best treatment, incidences of non-union can occur, particularly in the presence of known risk factors, including avascular necrosis, prior infection, smoking, varus malalignment and uncontrolled diabetes mellitus [2, 3]. There are several sources of autologous bone grafts, including iliac crest, tibia or calcaneus; however, they can result in a physical and monetary cost to the health system and the patient [4]. They are also associated with morbidity, and complications can include infection, neurological injury, hematomas, along with ongoing pain [5-7]. A composite peptide-enhanced bone graft (CPEBG) has been used successfully in spine surgery [8, 9]. There is potential for CPEBG to be used in foot and ankle surgery [10].

Recent studies have demonstrated similar union rates in ankle and hindfoot arthrodesis using rhPDGF/β-TCP (recombinant human platelet-derived growth factor with beta-tricalcium phosphate) compared to autologous bone graft [11]. To the best of our knowledge, there are no studies examining the union rates in ankle fusion procedures using CPEBG. We hypothesised that in patients undergoing both primary and secondary ankle fusion procedures with CPEBG bone augment, overall fusion rates would yield similar results to that described in the literature.

This was a retrospective study of patients over the age of 18 who underwent ankle fusion procedures using CPEBG. The CPEBG used in this study was an i-FACTOR putty (Cerapedics Inc, Westminister, Colorado) CPEBG. This specifically engineered form of bone graft comprises of an anorganic bone mineral (ABM) derived from hydroxyapatite enriched with a P-15 peptide. The ABM component acts like a scaffold for bony ingrowth. P-15 is a 15 amino acid peptide that is found in type 1 human collagen. This peptide acts as an attachment factor for osteogenic precursor cells to promote osteogenesis [8]. All operations were performed by a fellowship-trained orthopaedic surgeon between June 2016 and August 2020.

Upon ethics approval, data collected included baseline patient demographic data (age, sex, co-morbidities), operative data (side operated, operation performed), follow-up data (subjective pain, post-operative limp and use of mobility aids), adverse events and radiographic data at 6 and 12 months.

A total of 48 patients were included in the primary group and 3 patients in the secondary group.

Patient demographics are presented in Table 1. Within the primary group, 44% of patients had at least one co-morbidity, with hypertension most common at 31%. Other comorbidities prevalent within the primary group included inflammatory conditions (rheumatoid arthritis, Crohn's disease, ulcerative colitis, and systemic lupus erythema) and mental health conditions (e.g. depression, anxiety, bipolar disorder, and schizophrenia), which were present in 11% of patients respectively. Of the 3 patients in the secondary group, 1 patient had a prior DVT, and 1 patient had diabetes mellitus. There was 1 smoker in the primary group. No patients in the primary or secondary group had a diagnosis of Charcot neuroarthropathy.

Radiographic union rates for the primary group at at 6 & 12-month follow-up was 40/48 (90%) respectively. In our secondary group, 2/3 (67%) showed evidence of radiographic union at 6 months follow up. All secondary ankle fusion procedures united at 12 months follow up.

Complications for the primary and secondary groups are presented in Tables 2 and 3, respectively. In particular, the overall non-union rate in our primary group using clinical data was 8% after 12 months. The causes of repeat surgery included infections and non-union. In the primary group at 6 and 12 months, the pain was reported in 34/39 (87%) and 22/32 (69%), while 25/35 (71%) and 24/32 (75%) patients mobilised with a limp, respectively. The use of mobility aids decreased from 6/37 (16%) to 2/32 (6%).

There were no non-unions noted in the secondary group. All 3 patients reported pain at 6 months (3/3, 100%), with 1 patient with ongoing pain at 12 months (1/3,33%). No patients presented with a limp at follow-up (0/2, 0% & 0/3, 0%), and 1 patient required the use of a mobility aid at 6 months only (1/3, 33% & 0/3, 0%). There was no correlation between complications and patient demographics in both groups.

The use of bone autologous bone graft (ABG) and bone graft substitutes have long been used to reduce the risk of non-union when performing arthrodesis in the foot and ankle joint. Our rates of radiographic ankle fusion were 83% and 90% at 6 & 12 months, respectively. Our overall clinical non-union rate was 8%.

This is comparable to non-union rates described by Haddad et al. [13]. They conducted a meta-analysis of 1262 ankle arthrodesis procedures (use of bone graft not specified) with an overall non-union rate of 10%.

DiGiovanni et al. conducted a study of 414 patients that examined the utility of recombinant human platelet-derived growth factor with beta-tricalcium phosphate matrix in ankle and hindfoot fusions with radiographic union rates of 61% and 67% and 6 and 12 months, respectively. This was compared to the use of ABG, which demonstrated similar rates of 66% and 65% at respective time periods [14]. To our knowledge, DiGiovanni’s study is the largest multicentre randomised control trial to date examining rates of ankle and hindfoot fusion; however, the exact number of isolated ankle fusion procedures was not reported.

The utility of CPEBG is further substantiated by Arnold et al. noninferiority trial comparing the efficacy of i-FACTOR bone graft to autologous bone graft during anterior cervical discectomy and fusion procedures with both bone graft modalities showing similar rates of fusion at 12 and 24 months postoperatively [8]. The utility of a similar P-15 bone graft substitute has further been described for the management of long bone non-union by Gomar et al., with 90% (20 out of 22) patients demonstrating radiographic union [15].

Majority of patients had mild or minimal pain at both 6 and 12 months in both groups (as described in Tables 1-3 for the 6 and 12-month outcomes, respectively). These outcomes are similar to those described by Chou et al., assessing tibio-talo-calcaneal fusion. 84% of patients had mild/no pain after an average of 2 years follow. Chou et al. also noted that 64% of patients walked with a limp [16]. Davies et al. noted that 70% of patients had none or mild pain symptoms after a subtalar fusion [17]. Another study noted that the most patients were predominantly pain-free; however, this was after 5 years of follow-up following a triple arthrodesis [18].

There are various reasons for the increase in the number of patients presenting with a limp. These included patients undergoing repeat surgeries, delayed wound healing and delayed osseous union, and they were asked not to bear full weight. The reason for the increased percentage of patients limping at 12 months vs 6 months is the increased loss to follow-up at 12 months. We used the method of the last observation carried forward.

This is the first study to examine the efficacy of i-FACTOR CPEBG in ankle fusion procedures. To our knowledge, this is the largest single-centre retrospective analysis investigating isolated ankle fusion rates using CPEBG, demonstrating comparable results to traditional bone grafting options as described in the literature. The advantage of this study is that all procedures were performed by a single surgeon. All patients were followed up by a single surgeon post-operatively, allowing for consistency in reported patient clinical progress and operative outcomes.

This study experienced limitations with patient loss to follow-up and those associated with a retrospective study. Patients who were deemed radiographically united with minimal pain were not routinely followed up at 12 months. Of those patients that were followed up, 6- and 12-month radiographs were not always available. Due to the retrospective nature of this study, there is a lack of validated patient-reported outcomes, including overall satisfaction and functional outcomes. The inherent limitations with plain radiographs include operator-dependent views, beam obliquity, determining bony overlap and interruption of views with existing hardware. Previous studies have demonstrated poor correlation of CT scans and plain radiographs, with the latter imaging technique leading to an overestimation of the degree of bony union when compared to CT [19]. Our study did not conduct CT scans as routine practice, and thus, this was not included in the current study.

Our results show that using CPEBG as an adjunct in foot and ankle fusion procedures has similar union rates to other graft options. Further well-designed randomised control trials are warranted to confirm these findings.