ABSTRACT
AIM: To determine the causes and diagnostic utility of musculoskeletal (MSK) magnetic resonance imaging (MRI) recall examinations. MATERIALS AND METHODS: An institutional review board-approved retrospective review was conducted of all MSK MRI examinations performed at a single academic institution over 10 years where radiologists requested the patient return for additional imaging. The reason for the recall was documented. Recalls were reviewed in consensus by two MSK radiologists to determine whether additional sequences resulted in a change in the final report. Recall causes were divided into four categories: (1) radiologist-related: incorrect field of view (FOV) or incorrect protocol; (2) technologist-related: incorrect FOV or incorrect/incomplete protocol performed, or technically poor-quality images; (3) patient-related motion artefact; (4) unexpected lesion discovered. Fisher's exact test was used to assess for statistical significance. RESULTS: The recall rate was 0.25% (156/62,930). Of the total 129 recalls returning for imaging, 42 (33%) were radiologist-related, 45 (35%) were technologist-related, six (5%) were patient-related, and 36 (28%) had an unexpected lesion requiring additional sequences. For clinical utility, 42% resulted in a change from the initial report. Recalls due to radiologist error, incorrect FOV, or unexpected lesion caused a significant change in the final report; however, recalls due to technologist error, patient motion artefact, or incorrect protocol did not. CONCLUSION: MRI MSK recalls are uncommon, and the most common reasons are incorrect FOV, incorrect protocol, and unexpected lesion. Radiologist-related errors in protocols and FOV led to a significant change in the final report and should be targeted as areas for improvement to reduce recall examinations.
Subject(s)
Magnetic Resonance Imaging , Radiologists , Humans , Magnetic Resonance Imaging/methods , Retrospective StudiesABSTRACT
AIM: To determine the relationship between knee pain following anterior cruciate ligament (ACL) graft placement with morphological graft findings and dynamic contrast enhancement as assessed at MRI. MATERIAL AND METHODS: Following institutional review board approval, 37 consecutive patients with double-bundle ACL reconstruction were enrolled. Thirteen patients had pain and 24 were asymptomatic. Imaging was performed using a 1.5 T MRI machine an average of 7.6 months after surgery. Graft-related (increase signal intensity, abnormal orientation, discontinuity, cystic degeneration, anterior translation of lateral tibia, arthrofibrosis), and non-graft related causes of knee pain (meniscal tear, cartilage injury, loose bodies, and synovitis) were evaluated. During dynamic contrast enhancement analysis, peak enhancement (ePeak) was calculated by placing a region of interest at the osteoligamentous interface of each bundle. Student's t-test was used for continuous variables analysis and chi-square or Fisher's exact test was used for categorical variables analysis. RESULTS: There was no difference between symptomatic and asymptomatic patients regarding morphological graft-related or non-graft-related causes of knee pain. For dynamic contrast enhancement analysis, symptomatic patients had significantly lower ePeak values than asymptomatic patients in the anteromedial (p = 0.008) and posterolateral (p = 0.001) bundles or when using the higher ePeak value in either bundle (p = 0.003). CONCLUSION: Morphological ACL graft findings as assessed at MRI could not be used to distinguish between symptomatic and asymptomatic patients. However, lower ePeak values had a significant association with knee pain. This may indicate poor neovascularization of the graft, potentially leading to graft failure.
