Previous Page  124 / 245 Next Page
Information
Show Menu
Previous Page 124 / 245 Next Page
Page Background

124

· DOS Abstracts

Reliability of measurements on x-rays for knee

dysplasia and patella height

Anders Bøvling, Rune D Bech, Bertel Understrup, Bjarke Viberg

Department of Orthopaedic Surgery and Traumatology, Odense University

Hospital

Background:

Radiographic measurements are frequently used for the assess-

ment of knee dysplasia and patella height especially in patients with patella in-

stability but the literature is scarce on the reliability.

Purpose / Aim of Study:

To estimate the reliability of the Dejour Classification

(DC), Anterior Translation (AT), Trochlea Depth (TD), Caton-Dechamps Index

(CDI), Insall-Salvati Index (ISI), and the Blackburne-Peel Index (BPI).

Materials and Methods:

A search for DS830 was conducted in the county

database. X-rays were reviewed for applicability and 115 x-rays was acceptable

for measurements. 4 raters independently measured the x-rays 2 times with

a minimum of 14 days apart according to GRRAS guidelines. The raters were

one medical student, two residents, and one consultant in knee arthroscopi.

Unweighted kappa statistics was applied and prior to measurements a sample

size of minimum 105 x- rays was estimated. The Landis and Koch interpretation

of kappa values is <0 poor, 0.01- 0.20 slight, 0.21-40 fair, 0.41-0.60 moder-

ate, 0.61-0.80 substantial, and 0.81-1.00 almost perfect agreement

Findings / Results:

Intrarater kappa results ranged from 0.19-0.55 for DC,

-0.01-0.70 for AT, 0.31-0.85 for TD, 0.12-0.70 for CDI, 0.63-0.85 for ISI,

and 0.24-0.57 for BP. There was a tendency for the medical student in gen-

eral to have lower kappa values compared to the other raters. Interrater kappa

results ranged from 0.07-0.36 for DC, -0.08-0.60 for AT, 0.14-0.70 for TD,

0.10-0.58 for CDI, 0.59-0.71 for ISI, and 0.07-0.30 for BP. There was no ten-

dency for lower kappa values with higher experience.

Conclusions:

The only measurement with substantial reliability was ISI. All oth-

er measurements seem to be unreliable for use in clinical research.

No conflicts of interest reported

75.