Accuracy of Cross-Table Lateral Knee Radiography for Evaluation of Joint Effusions

OBJECTIVE. The purpose of our study was to investigate the efficacy of cross-table lateral knee radiography in the diagnosis of knee effusions compared with an MRI reference standard, to evaluate reader experience in effusion assessment, and to establish a new threshold for suprapatellar pouch measurement for the diagnosis of effusion. MATERIALS AND METHODS. First-and third-year radiology residents and an attending musculoskeletal radiologist retrospectively assessed 108 cross-table lateral knee radiographs for qualitative grading of joint fluid and quantitative measurement of the suprapatellar pouch. Qualitative and quantitative evaluation of ipsilateral knee MRI examinations performed within 1 week of radiography was performed by two attending musculoskeletal radiologists as a reference standard. RESULTS. Qualitative visual grading of cross-table lateral radiographs had a sensitivity of 90-92%, specificity of 39-54%, and accuracy of 69-76% for joint effusion. Extrapolating from previous work showing 4 mL of fluid distends the suprapatellar pouch to 4 mm on midline sagittal MRI, the corresponding measurement on cross-table lateral radiographs was predicted to be 7 mm. Using this new criterion of effusion, sensitivity, specificity, and accuracy compared with an MR midline sagittal reference standard were 76%, 83%, and 81%, respectively. Historical data for overhead lateral radiographs had a sensitivity of 78%, specificity of 80%, and accuracy of 79%. CONCLUSION. Qualitative visual assessment of cross-table lateral knee radiographs is highly sensitive for the detection of joint effusion. By performing quantitative evaluation with a new 7-mm criterion for suprapatellar pouch measurement, sensitivity, specificity, and accuracy are equivalent to that of overhead lateral radiography.