Contrast-enhanced MRI for detecting intravertebral cleft formation: relation to the time since onset of vertebral fracture.

Abstract

The purpose of this study was to investigate the advantages of contrast-enhanced MRI (CE-MRI) for detecting intravertebral clefts over unenhanced MRI in relation to the time since onset of vertebral fracture. In this retrospective study, a total of 115 patients (88 women, 27 men; mean age, 77.1 years) who underwent percutaneous vertebroplasty for a single-level compression fracture within 7 days of preprocedural MRI were enrolled. Two radiologists evaluated preprocedural unenhanced MRI (T1-weighted and STIR) and CE-MRI examinations for intravertebral clefts on separate days by consensus. The time from the onset of fracture to MRI was classified into three groups: early phase fractures (< 1 month), late phase fractures (1-3 months), and chronic phase fractures (≥ 3 months). The cement distribution during percutaneous vertebroplasty was used as a reference standard, and detectability of clefts was compared between unenhanced MRI and CE-MRI in relation to the time frame. An analog of the McNemar test was used for analyses. There were 104 patients (90.4%) with and 11 patients (9.6%) without a cleft. The sensitivities of unenhanced MRI and CE-MRI, respectively, for detecting clefts were 60.9% and 91.3% for early phase fractures (p = 0.02); 78.6% and 100% for late phase fractures (p = 0.014); and 92.5% and 94.3% for chronic phase fractures (p = 0.3). The specificities were the same for both groups (100% for each time frame). CE-MRI is more sensitive than unenhanced MRI for detecting intravertebral clefts in patents with benign compression fractures less than 3 months old.