Abstract
Background:
A number of controversies surround the evaluation of medial epicondyle fractures, including the use of advanced imaging at the time of diagnosis. In an era of increased focus on resource utilization, an analysis of diagnostic testing for this injury is timely. The purpose of this study is to evaluate national variation in diagnostic imaging for pediatric medial epicondyle fractures.
Methods:
The Pediatric Health Information System, a national database consisting of 49 children’s hospitals, was queried for all patients with a medial epicondyle fracture between 2004 and 2015. Demographic information was collected as well as data regarding diagnostic imaging at initial presentation. United States Census guidelines were used to categorize hospitals geographically into regions and divisions. Univariate analysis was followed by purposeful entry multivariate regression to control for confounders.
Results:
A total of 7799 patients underwent 9961 imaging procedures at initial presentation, of which 9627 were plain radiographs (1.2 x-rays per patient). Additionally, 286 patients (3.7%) obtained a computed tomography (CT) scan and 46 (0.6%) received magnetic resonance imaging (MRI). After controlling for confounders in a multivariate model, patients in cities with a pediatric population below 300,000 were 2.5 times more likely to undergo CT scan than in a city with over 1 million children (95% CI 1.7, 3.9; p<0.001). Geography was also predictive, with children in the Northeast at 3.1 times higher odds of receiving a CT scan (95% CI 2.2, 4.3; p<0.001). When region was further stratified into Census division, those in New England had 4.6 times higher odds of undergoing CT (95% CI 2.5, 8.7; p<0.001). Patients treated at a hospital with 300 or fewer beds were 2.6 times more likely to receive MRI than in larger hospitals (95% CI 1.2, 5.4; p=0.01). Older age was associated with higher odds of receiving a CT (p<0.001) while younger age was associated with higher likelihood of MRI (p=0.006).
Conclusions:
There is substantial variation in imaging practices across the United States when diagnosing a medial epicondyle fracture. CT scans are more likely to be ordered in smaller cities and in older children, while MRI is more likely to be obtained in smaller hospitals and in younger children. CT is ordered most frequently in the Northeast, especially in New England. Improved evidence relating to the initial evaluation of medial epicondyle fractures may help reduce the current variation in resource utilization.