Elbow clinical, ultrasonographic and radiographic study in patients with inflammatory joint diseases.

The main objective of this cross-sectional observational study was to investigate the relationship between clinical, ultrasonographic (US) and radiographic elbow features in patients with inflammatory joint diseases (IJD). The secondary objective was to evaluate the association between regional clinical elbow diagnoses and imaging findings. Consecutive patients with IJD attending follow-up visits were assessed for elbow pain and standardized elbow examination. Seven regional clinical diagnoses were defined. Digital elbow radiographs were read for 9 abnormalities. A standardized elbow grayscale (GS) and power Doppler (PD) scan recorded 13 defined abnormalities. Analysis encompassed 361 clinical, 361 US and 340 radiographic elbow assessments from 181 patients. US and clinical assessments showed an overall higher agreement than radiographic and clinical assessments (68.8 vs 59.1%, p = 0.001). When structural US abnormalities were compared with radiographic findings, agreement was slightly higher than when comparing all US abnormalities with radiographic findings (77.3%, k 0.533 and 73.5%, k 0.492). Enthesophytes, the most common abnormalities, were not associated with clinical findings. Subclinical US-synovitis and US-enthesopathy were found, respectively, in 17.3 and 14.1% of the clinically normal elbows. Clinical elbow arthritis prevalence and bias-adjusted kappa (PABAK) agreement was good for radiographic fat pad sign, PD-synovitis and GS-synovitis. Clinical elbow enthesopathy PABAK agreement was moderate for GS-enthesopathy and radiographic calcifications. US showed acceptable agreement with clinical and radiographic assessments for detecting elbow inflammatory and structural abnormalities in patients with IJD. Because US detected more abnormalities than radiography and has the capability to detect more subclinical abnormalities, US may be potentially used as a first-line elbow diagnostic tool in this clinical setting.