INTER-OBSERVER AND INTRA-OBSERVER AGREEMENT IN THE RADIOGRAPHIC DIAGNOSIS OF AVASCULAR NECROSIS OF THE FEMORAL HEAD FOLLOWING SLIPPED CAPITAL FEMORAL EPIPHYSIS

Abstract

Avascular necrosis (AVN) of the femoral head is a devastating complication of slipped capital femoral epiphysis (SCFE). The reported prevalence of AVN following unstable SCFE has ranged between fifteen and forty-seven per cent in the literature. The explanation for this discrepancy is not clear. The inter-observer and intra-observer agreement between Orthopaedic surgeons for the radiographic diagnosis of AVN following SCFE has not been reported. It is the objective of this study to estimate these parameters between two experienced pediatric Orthopaedic surgeons for the radiographic diagnosis of AVN following SCFE.

A retrospective review of all one hundred and three cases of SCFE treated at a Canadian pediatric referral center between 1995 and 2005 was performed. Of these, eight were diagnosed, by the treating surgeon, with AVN. Each of these eight children and a random sample of fifteen of the remaining children, who were not diagnosed with AVN, were included in this study. The most recent anteroposterior and lateral radiographs were digitised and presented to two experienced pediatric orthopaedic surgeons in a blinded, random order. Each surgeon reviewed the radiographs independently and recorded which radiographs they believed to be consistent with AVN. The surgeons were told that each patient had SCFE and that some developed AVN, however neither the classification of the slip, nor the proportion who developed AVN were divulged. Each observer repeated this process two weeks after the initial review in order to determine intra-observer agreement. The kappa value was determined to assess inter-observer and intra-observer agreement.

The first observer recorded eight cases of AVN at the initial and seven cases at the second observation time. The intra-observer agreement was 0.9. The second observer recorded six cases of AVN at the initial and five cases at the second observation time. The intra-observer agreement was 0.88. The inter-observer agreement was determined at the first observation time and was 0.79.

On the basis of the results of this study, both the inter-observer and intra-observer agreement for the radiographic diagnosis of AVN following SCFE, amongst experienced pediatric Orthopaedic surgeons, was very high. It is unlikely that the reported discrepancy in prevalence of AVN following SCFE is due to a lack of inter-observer agreement, on the basis of the findings of this study. The inter-observer agreement between less experienced observers requires further study to determine if this may be the source of the variability in the reported prevalence of AVN following SCFE.