Reproducibility of classifications for olecranon fractures

Abstract

Introduction: Fractures of the olecranon are relatively common injuries in adults and are of great clinical importance. Classification systems have been developed as tools to assist surgeons in grouping different types of fractures, to facilitate communication and to standardise treatment, but none of the systems used today is universally accepted for olecranon fractures.

Methods: Fifty-nine olecranon fractures were classified according to the Schatzker, Colton, Mayo and AO/ASIF systems by four observers with different levels of expertise. Intra- and inter-observer agreement was assessed. Each observer analysed the images at three different times; the images were randomised and presented in a different sequence at each assessment.

Results: There was higher mean intra-observer agreement in the AO/ASIF (0.60) and Mayo (0.64) classifications compared with the Schatzker (0.49) and Colton (0.38) classifications. Inter-observer agreement was better with AO/ASIF and Mayo (0.35 and 0.32, respectively) than with Schatzker and Colton (0.29 and 0.12, respectively).

Conclusion: The results of this study indicate that the most commonly used classifications for olecranon fractures are associated with low reproducibility.

Keywords: Olecranon process; Reproducibility of results; Ulna fractures; Validation studies.