Objectives To correlate cartilage thickness as measured by ultrasound (US) with joint space narrowing (JSN) and joint space width (JSW) as measured by conventional radiography in the metacarpophalangeal (MCP) joints of patients with rheumatoid arthritis. To validate US for measuring cartilage thickness, cadaver specimens of MCP joints were evaluated by US and by anatomical measurement.
Methods Bilateral MCP 2-5 joints of 35 consecutive patients (280 joints) recruited in our outpatient clinic who had conventional radiographs (<1 years) of both hands available as well as cadaver specimens of MCP 2-5 joints (n:12) were included in the analysis. The cartilage layer of the metacarpal heads and proximal phalangeal bases 2-5 of included patients as well as of cadavers were assessed bilaterally using a 15 Mhz linear transducer (GE Logic E9) from dorsal longitudinal and transverse views in midline by 2 sonographers blinded to clinical/radiographic data. Cartilage thickness was measured in cm with an integrated caliper on static US images. Both JSW and JSN were evaluated on conventional posterior-anterior radiographs. JSW was quantified as the shortest distance betweenthe subchondral bone plates using a standard measurementtool, while JSN was evaluated using the van der Heijde modified Sharp scoring method (vdHS). Ankylosed and luxated joints were excluded from the analysis. Anatomical metacarpal cartilage thickness (MCT) was also evaluated on static digital images of fresh frozen cadaver specimens, split in the mid-sagittal plane. Cartilage thickness was correlated with x-ray findings using Spearman’s or Pearson’s correlation. Intra- and interobserver reliability of US, and agreement between US and anatomic measurements were assessed by estimating the intraclass correlation coefficient (ICC).
Results In the patient population mean age was 63.1±11.3 years, mean disease duration was 10.6±7.8 years, mean CDAI was 8.1±7.3; 82.9% were female and 51% were rheumatoid factor positive. US measurement of MCT of MCP 2-5 was 0,42+/-0,19 mm. MCT of individual joints of the left and right hand correlated with individual MCP JSN (r=-0.799 and -0.702 respectively, p<0.01) and individual MCP JSW (r=0.588 and 0.401 respectively, p<0.01). The sum score of MCT for MCP joint 2-5 correlated with total MCP JSW (r=0.771, p<0.01), total JSN (r=-0.412, p<0.05), sum erosion score of the vdHS (r=-0.500, p<0.01) and the total vdHS (r=-0.576, p<0.01). No correlation was found between phalangeal cartilage thickness and JSN or JSW. Intra- and interobserver reliability of the US measurement of MCT in the clinical cohort and in the cadavers was 0.77 and 0.62, and 0.91 and 0.75 respectively. Agreement between US and anatomic measurement of MCT on cadaver specimens of MCP joints was 0.62.
Conclusions Both JSW and JSN by radiography indeed represent cartilage thickness at least in MCP joints. US is a valid tool for measuring MCT. Phalangeal cartilage thickness has no added value beyond the measurement of MCT. When radiographic scoring is not available, US measurement of MCT might be a feasible alternative to depict cartilage damage in patients with RA.
Disclosure of Interest None Declared