• Drago Gašpar, Tomislav Crnković, Dražen Durović, Dinko Podsednik, and Ferdinand Slišurić.
• Department of Traumatology, General County Hospital of Požega, Požega, Croatia. gaspar.drago@gmail.com
• Med Glas (Zenica). 2012 Aug 1; 9 (2): 243-7.

AimTo determine which of the classification systems for the femoral neck fracture between AO group, AO subgroup, Garden and Pauwels is much more reliable and reproducible to predict a method of treatment, radiological prediction of nonunion and prediction of outcomes.MethodsFive observers classified 77 randomly selected anterior- posterior (AP) and lateral view preoperative radiographs of the femoral neck fractures according to AO group, AO subgroup, Garden and Pauwels classification systems. The procedure was repeated on the same radiographs after three months. First classification is used to calculate interobserver agreement by kappa value between observers, while the first and second classification has served to calculate intraobserver kappa value for each examiner.ResultsOverall mean for classification system for interobserver agreement is: AO 0.44, AO subgroup 0.17, Garden 0.41 and Pauwels 0.19. Mean intraobserver agreement for AO group was 0.56, AO subgroup 0.38, Garden 0.49 and Pauwels 0.38 coefficient kappa value.ConclusionGarden and AO group are useful for the division of femoral neck fractures without displacement and with displacement, but they are not for clinical use. AO subgroup and Pauwels classification are not recommended for further use.

7 keywords...

hide…