Objectives: This study aims to evaluate the inter-observer reliability of the ultrasonographic examination of the wrist in RA patients between 3 examiners and 3 probe positions.
Methods: Fifty-three RA patients were recruited at the University Clinical Hospital of Santiago de Compostela in Spain for ultrasonographic examination of the wrist. Ultrasonography (US) was performed on both wrists using a GE LOGIQ 9 machine, using three probe positions: Lister's Tubercle to digit II (position 1), Lister's Tubercle to digit III (position 2) and ulnocarpal (position 3), from the anatomic medial orientation. Three examiners (2 experienced ultrasonographers and 1 junior ultrasonographer) scored synovitis according to a 0-3 semiquantitative scoring system. Inter-observer reliability was expressed using the ICC (A,1).
Results: For grey-scale ultrasound (GSUS) the inter-observer reliability (ICC(A,1)) (single measure, agreement definition) ranged from 0.35 for the ulnocarpal joint, position 3, to 0.60 in both position 1 for the radiocarpal joint and position 2 for the inter-carpal joint. Using power Doppler ultrasound (PDUS) the inter-observer reliability (ICC(A,1)) ranged from 0.36 in position 3, to 0.52 both in position 1 and 2 for the radiocarpal joint.
Conclusions: The reliability of the GSUS-examination of the wrist joints of RA patients with GSUS shows highest, moderate reliability using the anatomical landmarks Tubercle of Lister and digit III (position 2). The reliability of the PDUS examination was similar and moderate in both position 1 (Lister's Tubercle to digit II) and position 2 (Lister's Tubercle to digit III). The reliability was poorest for position 3 (the anatomic medial view of the ulnocarpal joint) in both the GSUS and PDUS examination. This study suggests that position 2 should be used in clinical trials and daily practice.