Correlation between preprocedural MRI findings and clinical outcomes in the treatment of chronic symptomatic vertebral compression fractures with percutaneous vertebroplasty

OBJECTIVE. The purpose of our study was to correlate findings of prevertebroplasty MRI with outcomes in the treatment of chronic osteoporotic compression fractures. MATERIALS AND METHODS. Forty-five patients with osteoporotic spinal compression fractures of more than I year's duration were treated with vertebroplasty. Changes in pain and mobility were assessed by follow-up of 1-28 months. Preprocedural MR images were reviewed using the Modic criteria and were correlated with outcomes. RESULTS. Fifteen patients (33%) had marrow edema on MRI and 30 (67%) of the 45 patients did not. All 15 of the patients with edema had clinical benefit: six patients (40%) achieved complete relief and nine (60%) experienced symptom improvement. Ten patients (67%) had improvement in mobility, and the remaining five patients (33%) had no change. Of patients with no marrow edema (n = 30), five (17%) had complete resolution of pain, 19 (63%) were improved, and six (20%) were unchanged. None had worsening of their symptoms. Mobility was improved in 17 (57%) and unchanged in 10 (33%). Mobility was diminished in three patients (10%). In two cases, impaired mobility was due to causes other than spine disorders. CONCLUSION. Most (87%) of the 45 patients with compression fractures older than I year derived clinical benefit from vertebroplasty irrespective of MRI findings. Although 100% of patients with bone marrow edema had clinical benefit, no direct correlation was seen between symptom resolution and the presence of edema on preprocedural MRI. In our experience, absence of abnormal marrow signal does not definitively predict the outcome of vertebroplasty in chronic fractures.