Can clinical parameters predict fractures in acute pediatric wrist injuries?

OBJECTIVE: Fractures around the wrist are common in pediatric patients presenting to the emergency department (ED). This pilot study was aimed at identifying clinical variables that are most likely to be associated with a fracture.
METHODS: This was a prospective blinded case series of patients 3-18 years of age presenting with an acute (<3 days) wrist injury, without obvious deformity. A team of five investigators blinded to the eventual radiographic findings evaluated patients. Physical examination variables included range of motion (ROM), site of maximal tenderness, and functional deficit. The latter was determined objectively, by recording any difference in grip strength between the injured and noninjured hands. Diagnostic radiographs were obtained for all patients. Univariate analysis using Wilks' log likelihood ratio test was performed to identify clinical variables associated with confirmed wrist fractures. Sample size was determined based on the ability to detect a difference of 15 degrees in the ROM variables, 20% point differences in grip strength, and 30% proportion differences in categorical variables using a power of 0.8 and a two-tailed of 0.05.
RESULTS: The ROMs were not significantly different between the fracture (Fx) and nonfracture (NFx) group. There was significant change in the grip strength between the Fx and NFx groups (t = 3.3, p = 0.0019). Tenderness over the distal radius was also associated with a greater likelihood of a fracture (G(2) = 5.0, p = 0.02). Sensitivity of clinical prediction was found to be 79%, and specificity was 63%. The false-negative rate was 0.21 and the false-positive rate was 0.37, while the positive predictive value was found to be 0.68 and negative predictive value 0.75.
CONCLUSIONS: Distal radius point tenderness and a 20% or more decrease in grip strength were predictive of fractures.