27.
Intra- and interobserver reliability on Sanders classification system
Lonnie Froberg, Nikolaj Sode, Henrik Eckardt
Department of Orthopaedic Surgery Rigshospitalet, Copenhagen University
Hospital, Blegdamsvej 9, 2100 Copenhagen, Denmark; Department of
Orthopaedic Surgery, Bispebjerg Hospital, Copenhagen University Hospital,
Bispebjerg Bakke 23, 2400 Copenhagen, Denmark; Department of
Orthopaedic Surgery, Rigshospitalet, Copenhagen University Hospital,
Blegdamsvej 9, 2100 Copenhagen, Denmark
Background:
Sanders Classification System is widely accepted for evaluation
of calcaneus fractures. Previous studies have shown a poor inter- and
intraobserver reliability. This may cause inconsistency in choice of treatment.
Purpose / Aim of Study:
The purpose of this study was to assess inter- and
intraobserver reliability of Sanders classification and explore whether
optimization of the preoperative computer tomographies improved these
values.
Materials and Methods:
The preoperative CT of 30 arbitrarily chosen
calcaneus fractures, treated at Rigshospitalet in 2011, were included in the
study. In the first and second trial, we used CT-scans with the coronal plane
parallel to the tibial axis. The angle of the foot relative to the tibia is thus not
taken into consideration. In the third and fourth trial, we oriented the coronal
plane perpendicular to the posterior facet of the subtalar joint before
classifying fractures. Inter- and intraobserver reliabilities were assessed using
kappa coefficient.
Findings / Results:
With the coronal plane parallel to the tibia axis, kappa for
interobserver agreement was 0.41 and intraobserver agreement 0.61. With the
true coronal plane perpendicular to the subtalar joint, kappa for interobserver
agreement was 0.43 and intraobserver agreement 0.67.
Conclusions:
The inter- and intraobserver agreement of Sanders Classification
System is low, and correspond to previous studies. We demonstrated an
improvement when high quality CT-scans were used, and we suggest that
classification of calcaneal fractures should be based on CT-scans with a well
defined coronal plane.
