How displaced are "nondisplaced" fractures of the medial humeral epicondyle in children? Results of a three-dimensional computed tomography analysis.

BACKGROUND: The management of fractures of the medial humeral epicondyle is controversial, but the primary issue is the relationship of outcomes to the extent of fracture displacement. This study compares the use of radiographs and three-dimensional computed tomography for determining the amount of displacement in medial humeral epicondylar injuries deemed to be minimally displaced (<5 mm) or nondisplaced.
METHODS: A retrospective review was performed on the cases of all patients with a fracture of the medial humeral epicondyle that had been diagnosed as minimally displaced or nondisplaced who were seen over a one-year period at our institution. Measurements of medial and anterior displacement on both the radiographs and three-dimensional computed tomography scan were recorded. Measurements of displacement were also recorded on internal oblique radiographs of the elbow, if available. Demographics, treatment, and any additional findings by computed tomography scans were noted. Means and Student t tests were utilized for statistical analysis.
RESULTS: The eleven patients who met the inclusion criteria had a mean age of 12.2 years (range, 7.3 to 15.4 years). One fracture that involved the medial condyle on the computed tomography scan was excluded from the analysis. Anterior displacement was immeasurable on all but one lateral radiograph and recorded as 0 mm; the mean was 0.9 mm, which was significantly less than the anterior displacement on the three-dimensional computed tomography scan (mean, 8.8 mm; range, 0 to 15 mm) (p ≤ 0.001). Conversely, mean medial displacement on anteroposterior radiographs was 3.5 mm (range, 0 to 8 mm), which was significantly more than that measured on three-dimensional computed tomography scans (mean, 0.3 mm; range, 0 to 1.9 mm) (p ≤ 0.001). Mean displacement on internal oblique radiographs of the elbow was 6.6 mm (range, 0 to 10.5 mm) and matched the anterior displacement measurement on the three-dimensional computed tomography scan in three of the six patients (p = 0.037). Five of the six fractures with >1 cm of displacement by three-dimensional computed tomography scan underwent surgical treatment.
CONCLUSIONS: Standard radiographs (anteroposterior and lateral views) are not sufficient to measure anterior displacement nor accurate enough to measure medial displacement of medial humeral epicondylar fractures. Internal oblique radiographs of the elbow appear to approximate the true anterior displacement, but three-dimensional computed tomography is the most accurate method to assess true displacement. The results of this study demonstrate that fractures that are found to be minimally displaced or nondisplaced by radiographs may have >1 cm of anterior displacement, for which surgery is usually recommended.