Magnetic resonance evaluation of multiple myeloma at 3.0 Tesla: how do bone marrow plasma cell percentage and selection of protocols affect lesion conspicuity?

Abstract

To compare various pulse sequences in terms of percent contrast and contrast-to-noise ratio (CNR) for detection of focal multiple myeloma lesions and to assess the dependence of lesion conspicuity on the bone marrow plasma cell percent (BMPC%). Sagittal T1-weighted FSE, fat-suppressed T2-weighted FSE (FS- T2 FSE), fast STIR and iterative decomposition of water and fat with echo asymmetry and least-squares estimation (IDEAL) imaging of the lumbar spine were performed (n = 45). Bone marrow (BM)-focal myeloma lesion percent contrast and CNR were calculated. Spearman rank correlation coefficients were obtained between percent contrast, CNR and BMPC%. Percent contrasts and CNRs were compared among the three imaging sequences. BM-focal lesion percent contrasts, CNRs and BMPC% showed significant negative correlations in the three fat-suppression techniques. Percent contrast and CNRs were significantly higher for FS- T2 FSE than for STIR (P<0.01, P<0.05, respectively), but no significant differences were found among the three fat-suppression methods in the low tumor load BM group. The higher BMPC% was within BM, the less conspicuous the focal lesion was on fat-suppressed MRI. The most effective protocol for detecting focal lesions was FS- T2 FSE. In the high tumor load BM group, no significant differences in lesion conspicuity were identified among the three fat-suppression techniques.