INTER-OBSERVER AND INTRA-OBSERVER RELIABILITY OF THREE DIFFERENT CLASSIFICATION SYSTEMS FOR TIBIAL PLATEAU FRACTURES

Abstract

Introduction: Fracture classification systems help in communication, treatment planning, assessing prognosis and form standards to report treatment results. The ideal classification system should be reliable, reproducible, all-inclusive, mutually exclusive, logical and clinically useful. The aim of our study was to assess the inter observer reliability and intra observer variability for the AO, Schatzker and Hohl and Moore classification systems.

Materials and Methods: We randomly selected fifty sets of radiographs of tibial plateau fractures occurred between 2000 and 2005. Exclusion criteria: Only one available view, inadequate films. Four orthopaedic surgeons at various level of experience i.e. one senior senior house officer, two registrars and a trauma consultant classified the fractures. Radiographs were blinded and each time the radiographs were presented to the observers in a different order. Radiographs were viewed at two separate sittings 8 weeks apart. The data was analysed using kappa statistics through SPSS version 14. The Kappa co-efficients were interpreted according to Landis and Koch grading. (< 0.00=poor; 0.0–0.2=slight, 0.21–0.4=fair, 0.41–0.60=moderate, 0.61–0.8=substanti al,> 0.8=excellent)

Results: For the AO classification the mean kappa co-efficients for inter-observer and intra-observer reliability were 0.36 (0.33–0.39) and 0.83(0.61–1.00) respectively. For the Schatzker classification the mean kappa co-efficients for inter-observer and intra-observer reliability were 0.47(0.45–0.49) and 0.90(0.75–1.00) respectively. For the Hohl& Moore classification mean kappa values for inter-observer and intra-observer variability were 0.14 and 0.81(0.59–1.00) respectively. According to Landis and Koch grading, AO classification is fair in terms of interobserver reliability, the Schatzker classification is moderate and the Hohl& Moore is slight.

Conclusions: None of the three systems fulfils the criteria for an ideal classification system. However, the Schatzker classification system was found to be superior. The Hohl& Moore system was least reliable of all. Hence, we recommend usage of Schatzker classification system in tibial plateau fractures both in clinical practice and clinical studies until a superior classification system evolves.