Abstract Archives of the RSNA, 2011
LL-MKS-TU4B
Diagnostic Accuracy of Lateral and Axial Radiographs Compared with Arthroscopic Findings of Patellofemoral Cartilage Damage
Scientific Informal (Poster) Presentations
Presented on November 29, 2011
Presented as part of LL-MKS-TU: Musculoskeletal Imaging
Kirkland W. Davis MD, Presenter: Nothing to Disclose
Richard Kijowski MD, Abstract Co-Author: Consultant, Flex Biomedical, Inc
Donna Genette Blankenbaker MD, Abstract Co-Author: Nothing to Disclose
Alejandro Munoz Del Rio PhD, Abstract Co-Author: Nothing to Disclose
Ben K. Graf MD, Abstract Co-Author: Royalties, Smith & Nephew plc
Patellar and femoral cartilage defects (PFCD) may be a source of substantial pain. The purpose of this study was to determine the accuracy of radiographic findings of osteoarthritis on lateral and axial views for diagnosing cartilage defects in the patellofemoral compartment.
This IRB-approved, HIPAA compliant study included 200 subjects aged 20-69 who underwent arthroscopy of the knee for indications other than PFCD. All subjects had lateral and Laurin (axial) radiographs performed within 6 months prior to surgery. Three MSK radiologists independently assessed each subject’s radiographs for sclerosis, cysts and other lucencies, osteophytes, joint space narrowing (JSN), and reader confidence for chondrosis. Sensitivity and specificity were calculated. For this report, abnormal cartilage was considered to have at least deep (> 50% depth) focal partial-thickness defects in either the trochlea or patella; fraying and shallow defects were considered normal.
Seventy-nine subjects (39.5%) had substantial cartilage defects. The severity of trochlear defects was strongly correlated with the severity of patellar defects (p<0.0001). On the lateral view, sensitivity ranges for the 3 readers were: cysts and/or other lucencies: 0.25- 0.68; sclerosis: 0.20-0.70; JSN: 0.09-0.29; and patellar osteophytes: 0.65-0.76. On the Laurin view, sensitivity ranges were: cysts/lucencies: 0.21-0.61; sclerosis 0.06-0.38; JSN: 0.19-0.39; and patellar osteophytes: 0.66-0.91. Lateral view specificity ranges were: cysts/lucencies: 0.75-0.95; sclerosis 0.51-0.86; JSN: 0.97-1.00; and patellar osteophytes: 0.59-0.74. Laurin view specificity ranges were: cysts/lucencies: 0.73-0.95; sclerosis: 0.88-0.99; JSN:0.89-0.95; and patellar osteophytes: 0.47-0.61. High reader confidence in PFCD was highly specific.
None of the findings in either view was particularly sensitive for PFCD though patellar osteophytes were the most sensitive. However, specificity was high on both views for cysts/lucencies and JSN, and on the Laurin view for sclerosis. These high specificities suggest that patients without PFCD will have normal radiographs. Neither the lateral nor Laurin view was clearly superior for accuracy of findings.
Radiographs are a crude tool for detecting PFCD but positive findings, especially joint space narrowing, are likely indicators of PFCD, which may be the source of anterior knee pain.
Davis, K,
Kijowski, R,
Blankenbaker, D,
Munoz Del Rio, A,
Graf, B,
Diagnostic Accuracy of Lateral and Axial Radiographs Compared with Arthroscopic Findings of Patellofemoral Cartilage Damage. Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL.
http://archive.rsna.org/2011/11034225.html