Traumatic hemarthrosis of the knee joint--clinical significance of nuclear magnetic resonance tomography

Abstract

A prospective study was performed on 94 patients with hemarthrosis of the knee to assess the value of MR imaging (MRI) in detection of bleeding sources and selection of therapy modalities. Patients were examined clinically within a week after knee trauma; the investigations performed included puncture of the joint to confirm hemarthrosis, a conventional X-ray to exclude fractures, MRI and arthroscopy. MRI was performed for diagnosis of acute lesions, definition of bleeding sources by morphological criteria, and allocation of patients to conservative or surgical therapy. Arthroscopy was performed to control MRI, to visualize bleeding sources, and to maintain therapy if necessary. The 94 patients were found on arthroscopy to have a total of 123 bleeding sources, which were correctly defined by MRI in each of 107 cases as an acute lesion communicating with the joint space; 16 bleeding sources were not found and there were 10 false-positive reports. In keeping with our treatment strategies, arthroscopy disclosed lesions requiring surgical therapy in 77 of 94 patients (82%) and lesions that would be adequately treated by conservative therapy in 17 of 94 patients (18%). MRI predicted surgical or conservative therapy correctly before arthroscopy in 83 of 94 patients (88%). In conclusion, MRI has a high predictive value in definition of acute lesions that will lead to hemarthrosis of the knee. This noninvasive method allows screening out of the relatively small portion of patients without severe lesions, who then do not have to be subjected to further invasive therapy.