Minimally invasive treatment approaches in the operative management of unstable ankle fractures

Abstract

Ankle fractures account for approximately 10% of the acute orthopaedic workload and patients with unstable fracture patterns are typically offered surgery. Soft-tissue complications are unfortunately common and can result in significant morbidity.

Consequently, there has been recent interest in minimally invasive treatment approaches, including intramedullary fixation of the fibula and treating well-reduced medial malleolar fractures non-operatively, following fibular stabilisation. This thesis aims to examine both concepts further through a series of retrospective, biomechanical and randomised controlled trial (RCT) investigations.

A prospective database of patients undergoing fibular intramedullary nail fixation of an unstable ankle fracture over an eight-year period was developed. This database was used to report on the outcome of fibular intramedullary nail fixation for these injuries and was also used to compare the outcome of those patients managed with or without medial malleolar fracture fixation after intramedullary fibular stabilisation. The primary short-term outcome was complications, including construct failure requiring revision surgery and medial sided complications. The primary mid-term outcome was the Olerud-Molander Ankle Score (OMAS). To further evaluate the performance of the fibular intramedullary nail against locking plate and interfragmentary screw fixation, a cadaveric biomechanical study was performed using six matched pairs of cadaveric lower limbs (mean age 86.5 years) with simulated AO/OTA 44-B type fractures. The specimens were randomised within each pair and tested to failure using a modification of a previously published protocol. Finally, a prospective RCT was carried out to compare fixation and non-fixation of an associated well reduced medial malleolar fracture (£2 millimetres) after fibular fracture stabilisation. The primary outcome measure was the OMAS one-year post randomisation. Secondary outcomes included further validated PROMs, complications, return to function and treatment satisfaction.

For the retrospective study of 342 patients (mean age of 64.6 years; range, 21- 96), there were 20 fibular intramedullary nail fixation cases (6%) that required revision surgery for construct failure, and these were reviewed to classify failure according to ‘surgeon error’ (n=13) or ‘device failure’ (n=7). Risk factors for failure were identified, including suprasyndesmotic injuries and those constructs with the proximal locking screw (PLS) inserted >20mm above the plafond, which have informed technical refinements to improve patient care. After a mean follow-up of five years, outcome scores were collected from 229 patients, who in general reported a ‘good’ outcome according to a median OMAS of 80/100. From this database a refined cohort of 247 patients was analysed; 193 (78.1%) had received medial malleolar fracture fixation and 54 (21.9%) had not. Following retrospective review, there was no difference between the groups with respect to fixation failure (p=0.634) or loss of talar reduction (p=0.157). Medial sided soft tissue complications were more frequent following fixation and 10% of patients required surgery to address these. Medial sided radiographic union at the point of discharge was lacking in 16 (29.6%) patients in the non-fixation compared with 22 (11.4%) in the fixation group (p=0.002). The median mid-term OMAS was comparable between the two groups (p=0.885). In a cadaveric setting, the fibular intramedullary nail and locking plate with interfragmentary screw fixation demonstrated comparable biomechanical properties including torque to failure, stiffness and energy absorbed (all p>0.05). The intramedullary nail failed primarily at the lateral ligament complex and at a greater angle of rotation (p=0.046), but the clinical significance of this latter finding is unclear.

Preliminary results from the prospective RCT are reported from 106 participants randomised to fixation (n=53) or non-fixation (n=53) of an associated well-reduced medial malleolar fracture after fibular stabilisation. The baseline demographics and injury characteristics were comparable (all p>0.05). One hundred patients (94.3%) were reviewed one year following surgery. There was no significant difference in the OMAS at any assessment point during the trial (all p<0.05) or any of the secondary outcomes at one year, including patients experiencing complications (64.0% vs. 62.0%; p=0.836). Whilst not statistically significant, more patients in the non-fixation group experienced a major complication (18.0% vs. 8.0%; p=0.137). Medial malleolar non-union was evident in three patients (6.0%) in the non-fixation group compared with no patients in the fixation group (p=0.242), and seven patients developed an asymptomatic pseudoarthrosis of the medial malleolus. Multivariate linear regression analysis identified smoking (p=0.006) and medial malleolar non-union (p=0.002) as predictive of a poor outcome according to the OMAS, but treatment group allocation was not (p=0.357).

The findings of this thesis support the use of the fibular intramedullary nail in the management of unstable ankle fractures. The mechanical failure rate, mid-term patient reported outcome and biomechanical properties are encouraging. Non-operative management of medial malleolar fractures as part of unstable fracture patterns is corroborated by both the retrospective and prospective data reported in this thesis. However, the prospective RCT remains under-powered and subsequent reporting of results upon trial completion is required to validate these findings.