CLASSIFICATION OF DISTAL RADIUS FRACTURES - AN ANALYSIS OF INTEROBSERVER RELIABILITY AND INTRAOBSERVER REPRODUCIBILITY

The Frykman, Melone, Mayo, and AO classification systems for distal ra dius fractures were evaluated for interobserver reliability and intrao bserver reproducibility in a clinical setting using initial plain radi ographs. Two attending orthopedic hand surgeons and two attending radi ologists classified 55 sets of distal radius fractures. kappa-statisti cs were used to establish a relative level of agreement between observ ers for the two readings and between separate readings by the same obs erver. Interobserver agreement was rated as moderate for the Mayo clas sification and fair for the Frykman, Melone, and AO classifications. I ntraobserver agreement was substantial for only one of four observers for each of the Frykman, Melone, and Mayo, while the remaining three o bservers achieved only fair to moderate reproducibility. Intraobserver agreement for the AO classification was fair for all four of the obse rvers. Neither interobserver or intraobserver agreement was affected b y combining similar subclasses in the Melone classification or by redu cing the number of categories in the AO system from 27 to 9. However, further reducing the AO system to its three main types brought agreeme nt to the ''substantial'' level. No difference was found in interobser ver agreement between the first and second readings or in interobserve r or intraobserver agreement between orthopedic hand surgeons and radi ologists. Understanding the limitations of fracture classifications ba sed solely on plain radiographs can help avoid undue reliance on them. Given the low degree of interobserver and intraobserver agreement for each of the distal radius fracture classifications in this study, the ir use as the sole means for determining the direction of treatment or for the direct comparison of results among different studies is not w arranted.