A new classification system for distal radius fractures: ideal - description and reliability compared with traditional classifications

Abstract

Distal radius fractures occur in approximately one of every 10,000 persons, accounting for 16% of skeletal fractures and 74% of forearm fractures. The fracture characteristics are directly related to the trauma energy, position of the hand at the moment of trauma, and bone quality. Appropriate fracture treatment requires a good understanding of these fracture characteristics. Currently, we lack a classification system with adequate reproducibility to provide evidence for treatment and prognosis. In this study we aim to describe the new method of classification of distal radius fractures-the IDEAL classification, assess its reliability (intraobserver and interobserver agreement), and compare IDEAL to established classification systems (AO, Frykman, and Fernandez classification systems). Methods: To evaluate the reproducibility of the IDEAL classification system, 60 radiographs (anteroposterior and lateral) from patients with distal radius fractures were assessed by six examiners-a hand surgery expert, a hand surgery resident, an orthopedic generalist, orthopedic resident, and two medical students-who each independently assessed the radiographs three different times. We compared the intraobserver and interobserver reliability of the IDEAL system and with those of the AO, Frykman, Fernandez and classifications using Cohen's kappa (x) for interobserver agreement (two observers) and Fleiss' x for intraobserver agreement and interobserver agreement (more than two observers), followed by measure of dispersion of spread (standard error) and inferential analysis. (AU)