Knee osteoarthritis: interpretation variability of radiological signs

The aim of this study was (1) to determine the variability in detecting radiological signs of knee osteoarthritis (OA) between an orthopaedics specialist, a fourth-year resident in the speciality and a recently qualified doctor and (2) to determine which of the existing criteria show the greatest va...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Clinical rheumatology 2004-12, Vol.23 (6), p.501-504
Hauptverfasser: Vilalta, C, Núñez, M, Segur, J M, Domingo, A, Carbonell, J A, Maculé, F
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The aim of this study was (1) to determine the variability in detecting radiological signs of knee osteoarthritis (OA) between an orthopaedics specialist, a fourth-year resident in the speciality and a recently qualified doctor and (2) to determine which of the existing criteria show the greatest variability when used by the three participants to detect the degree of evolution of the pathology. This observational study included radiographs of 95 patients with knee pain. Osteophytes, narrowing of joint space (excluding inter-osteophyte bridges) subchondral sclerosis, subchondral cysts, collapse of the central joint cortical bone and lateral deformity, according to the criteria of Kellgren and Lawrence, modified by Kallman et al. were evaluated. Anteroposterior radiographs were used. Knees that had undergone previous surgery were excluded. Cohen's kappa index was used to calculate the degree of agreement between observers. The concordance analysis showed a low level of agreement among the three observers of the radiological variables with a maximum of 50% in some parameters. The authors discuss the possible causes of this low level of agreement. The low degree of agreement of 50% among the three observers is in line with previous reports and suggests that better training of observers is necessary and that the use of any classification is problematic.
ISSN:0770-3198
1434-9949
DOI:10.1007/s10067-004-0934-3