Abstract
To report the interobserver and intraobserver reliability of 3 chondral damage classifications used to assess articular cartilage damage during hip arthroscopy.
A prospective multicenter study was performed during April and May 2013. Inclusion criteria were all patients who underwent hip arthroscopy for femoroacetabular impingement (FAI) and had evidence of chondral damage at the time of surgery. Intra-articular recordings were obtained during the operation in a standardized way. These recordings were obtained from 2 different hospitals in 2 countries by 3 different surgeons. Four fellowship-trained orthopaedic surgeons, with at least 2 years postfellowship experience in hip arthroscopy, independently analyzed the recordings 2 times in randomized order and 4 months apart. They classified the lesions according to the Outerbridge, Beck, and Haddad classifications of chondral damage. The values obtained were used for interobserver and intraobserver analysis. Percentage of agreement and weighted Cohen κ values were calculated.
Absolute agreement between observers was present in 12.5% of the cases for the Outerbridge classification, in 20% of the cases for the Beck classification, and in 40% of the cases for the Haddad classification. For interobserver reliability, the average weighted Cohen κ values were 0.28 (95% confidence interval [CI], 0.16 to 0.39), 0.33 (95% CI, 0.24 to 0.41), and 0.47 (95% CI, 0.42 to 0.51) for the Outerbridge, Beck, and Haddad classification systems, respectively. For intraobserver reliability, the mean Cohen κ values were 0.62, 0.63, and 0.68 for the Outerbridge, Beck, and Haddad classification systems, respectively.
In our series, the Haddad classification had the best interobserver reliability. There was no difference in the intraobserver reliability among the 3 classifications studied.
Level III, diagnostic study of nonconsecutive patients (without consistently applied reference gold standard).