Objective: The objective was to determine the value of radiographs in young adults with non-acute knee symptoms who are scheduled for magnetic resonance imaging (MRI).
Materials and methods: Nine hundred and sixty-one consecutive patients aged between 16 and 45 years with knee symptoms of at least 4 weeks' duration were prospectively included in three participating hospitals. After applying exclusion criteria, 798 patients remained. Exclusion criteria were previous knee surgery (including arthroscopy) or MRI, history of rheumatoid arthritis, clinical diagnosis of retropatellar chondromalacia, contra-indication for MRI and recent trauma. We identified two groups: group A with no history of trauma (n = 332), and group B with an old (>4 weeks) history of trauma (n = 466). Patients had a standardized history taken, and underwent a physical exam, antero-posterior (AP) and lateral radiographs and MRI. We evaluated the radiographs and MRI for osseous lesions, articular surface lesions, fractures, osteoarthritis, loose bodies, bone marrow edema and incidental findings. Subsequently, patients with osseous abnormalities (Kellgren grade 1 and 2 excluded) on radiographs and a matched control group was evaluated again using MRI without radiographs.
Results: Median duration of symptoms was 20 weeks. In group A, radiographs showed 36 osseous abnormalities in 332 patients (10.8%). Only 13 of these, all Kellgren grade 1 osteoarthritis, were not confirmed on MRI. MRI showed 72 (21.7%) additional abnormalities not confirmed on radiographs. In group B, radiographs showed 40 osseous abnormalities (8.6%) in 466 patients. Only 15 of these, all Kellgren grade 1 osteoarthritis, were not confirmed on MRI. MRI showed 194 (41.6%) additional abnormalities not confirmed on radiographs. The second evaluation of MRI without radiographs in 34 patients was identical to the first MRI evaluation. Common lesions were significantly more often diagnosed with MRI than with radiographs.
Conclusion: Radiographs should not be obtained routinely when MRI is being performed in young adults with non-acute knee complaints because the yield and added value to MRI are low.