Abstract
Category:
Trauma
Introduction/Purpose:
Management of basilar 5th metatarsal fractures remains a controversial topic in orthopaedic surgery. Both operative and non-operative approaches have been described as potentially effective treatment strategies in the clinical setting. Non-operative approach is associated with higher non-union rates and longer times to union; however, operative intervention exposes patients to surgical risks. This confusion has led to non-standardized treatment recommendations for basilar 5th metatarsal fractures. This study was therefore designed to analyze whether or not orthopaedic trainees recommend treatment that differs from the treatments proposed by orthopaedic foot and ankle experts. Our hypothesis is that PGY-level does not improve concordance of treatment recommendation compared with orthopaedic foot and ankle specialists as a reflection of the varied treatment algorithms trainees are exposed to during residency training.
Methods:
An online survey containing 20 operative cases of proximal 5th metatarsal fractures were distributed to 92 orthopaedic residents in two ACGME-accredited programs. For each case, residents were provided relevant weight-bearing radiographs, patient’s age and gender, and then two questions regarding treatment recommendations. Residents were first asked to decide between operative and non-operative treatment. When the choice was operative, the second question focused on identifying operative reasoning (age, fracture pattern/location, or both). If the respondent chose non-operative treatment, however, the second question instead asked about a recommended weight-bearing status (weight-bearing as tolerated vs. non-weight-bearing). Resident’s recommended treatment was then matched against ultimate treatment by orthopaedic foot and ankle experts, which are all operative. This association was termed the rate of concordant treatment. Pearson’s rho is calculated for associations between the rate of concordant treatment with PGY and trainee foot and ankle experience. Fleiss’ kappa was used to assess the inter-observer agreement.
Results:
Seventy-two residents returned the survey. The overall concordance rate was 28.33% (32.06% for PGY-1, 32.86% for PGY-2, 31.79% for PGY-3, 19.67% for PGY-4, and 24.58% for PGY-5) with no correlation between agreement rate and PGY-years (Spearman’s r = -0.214, p=0.071). The overall concordance rate for treatment of Jones’ fractures was 34.44% (29.41% for PGY-1, 34.29% for PGY-2, 42.86% for PGY-3, 28% for PGY-4, and 40% for PGY-5) with no correlation between agreement rate and PGY-years (Spearman’s r = 0.073 p=0.540). No difference in agreement rate was observed between residents who had completed their foot and ankle rotation versus those who had not (24.71% vs. 31.58%, p=0.126). There was a slight inter-observer agreement in recommending operative treatment among all residents (? = 0.091, 95% CI 0.055-0.143).
Conclusion:
Our data demonstrated that there was no significant concordance between resident level in training regarding 5th MT fracture treatment decisions, nor between residents and subspecialty trained foot and ankle surgeons. The inter-observer agreement was in fact minimal amongst residents, suggesting that resident treatment recommendations reflect marked variability in clinical exposure and treatment decision making as trainees. Increased rotations with foot and ankle fellowship-trained surgeons throughout residency may be desirable to improve the quality of residency training.