Abstract
Category:
Ankle
Introduction/Purpose:
Ankle fractures are a common cause of functional mobility that have seen a dramatic rise incidence over the past decade1. The purpose of this study was to compare the outcomes and prognosis of various fracture sub-types by utilizing two validated patient reported outcome measures: the PROMIS Physical Function and Pain Interference Computer Adaptive Tests (CATs) for patients at a major tertiary facility and level I trauma center.
Methods:
12-month post-operative PROMIS Physical Function and Pain Interference CATs were collected for 125 ankle fracture patients presenting between 2014 and 2017. Patients were initially stratified by ankle fracture sub-type after examining pre- operative X-rays and further refined by the presence or absence of concomitant deltoid injury or posterior malleolar fracture.
Patients defined as members of vulnerable populations and patients presenting more than 2 weeks from time of injury/with prior acute ipsilateral fracture were excluded. The distribution of PROMIS physical function and pain interference T-scores were assessed via a Shapiro-Wilk test and a one-way ANOVA was used for further evaluation. If significant differences were found between groups, pair-wise comparisons were tested via Dwass, Steel, Critchlow-Fligner multiple comparison analysis.
Results:
Fracture types and corresponding n-value in this study: (44) isolated lateral malleolar, (12) isolated medial malleolar, (12) bimalleolar, (44) trimalleolar, (6) isolated syndesmotic injury, and (8) isolated posterior malleolar fractures. Mean values for the PROMIS physical function and pain interference domains for each fracture sub-type were calculated and compared to reference population mean T-scores of 50 (SD = 10): isolated lateral malleolar (PF:50/PI:51), isolated medial malleolar (PF:52/PI:49), bimalleolar (PF:48/PI:50), trimalleoar (PF:47/PI:51), isolated posterior malleolar (PF:53/PI:44), and isolated syndesmotic injury (PF:60/PI:46). The Shapiro-Wilk test indicated a non-normal distribution for the post-operative PROMIS physical function T-scores across all fracture patients (p-value = 0.0421). Pair-wise comparisons were done to compare relative prognosis between fracture sub-types but failed to yield statistically significant differences in physical function or pain domains.
Conclusion:
Surgical intervention is effective in returning patients to normal levels of physical function and pain interference across all sub-types according to post-op T-scores. Although our analysis indicated a significant difference when comparing across all fracture sub-types, none of the pairwise sub-type comparisons showed statistical significance. Analyses of fracture sub-types determined trimalleolar injuries report the poorest outcomes for both PROMIS domains while the isolated syndesmotic and posterior malleolar groups reported the best outcomes. Further investigation with a larger sample size for each group is needed to delineate whether a significant difference exists for PROMIS physical function and pain interference scores.