Keywords:
Musculoskeletal joint, Bones, MR, Diagnostic procedure, Arthritides, Oedema
Authors:
J. Neubauer, T. Yilmaz, D. Maier; Freiburg/DE
DOI:
10.26044/ecr2019/C-2446
Methods and materials
Over a period of 1 year this IRB approved prospective study included patients who presented to our outpatient clinic with shoulder pain of unknown origin and who had previously undergone MRI of the shoulder for this reason.
MRI had been performed in multiple different institutions.
An orthopedic surgeon performed O´Brien´s Test1,
Paxinos Test2,
Cross Body Adduction Test1 and AC joint tenderness test1 in all patients resulting in a score from 0 (no test positive) to 4 (all tests positive).
3 raters experienced in the evaluation of musculoskeletal MRI independently evaluated the anonymized MR images and assessed the amount of edema in the lateral clavicle,
osteophytes,
subchondral cysts,
effusion,
joint space narrowing with localization of bony contact and measured the extent of edema in the lateral clavicle and the thickness of the joint capsule.
Inter-rater reliability was analyzed with Fleiss´ kappa and intraclass correlation coefficient.
Features with at least moderate inter-rater reliability (>0.4) were considered reliable.
Correlation was analyzed with ordinal logistic regression and proportional odds ratios were calculated.
P-values <0.05 were considered to denote statistical significance.