The reliability of musculoskeletal ultrasound in the detection of cartilage abnormalities at the metacarpo-phalangeal joints

Objective: To assess the reliability of ultrasound (US) in detecting cartilage abnormalities at the metacarpo-phalangeal (MCP) joints in people with cartilage pathology.

Methods: Nine expert ultrasonographers initially achieved consensus on definitions and scanning protocols. They then examined the second to fifth MCP joints of the dominant hand of eight people with hand osteoarthritis (OA). US examinations were conducted in two rounds, with independent blinded evaluations of cartilage lesions. Global cartilage abnormalities were assessed by applying a dichotomous (presence/absence) score; in addition, the following lesions were evaluated using the same scoring system: loss of anechoic structure and/or thinning of the cartilage layer, and irregularities and/or loss of sharpness of at least one cartilage margin. Reliability was assessed using kappa (k) coefficients.

Results: Thirty-two joints were examined. Intra-observer k values ranged from 0.52 to 1 for global cartilage abnormalities; k values ranged from 0.54 to 0.94 for loss of anechoic structure and/or thinning of cartilage layer and from 0.59 to 1 for irregularities and/or loss of sharpness of at least one cartilage margin. Values of k for inter-observer reliability were 0.80 for global cartilage abnormalities, 0.62 for loss of anechoic structure and/or thinning of cartilage layer, and 0.39 for irregularities and/or loss of sharpness of at least one cartilage margin.

Conclusion: US is a reliable imaging modality for the detection of cartilage abnormalities in patients with cartilage pathology in the MCP joints. The analysis of specific cartilage measures showed more variable results that may be improved by modifying definitions and further standardization of US techniques.