Dual-energy CT of the bone marrow in multiple myeloma: diagnostic accuracy for quantitative differentiation of infiltration patterns

• Published in: European radiology Vol. 28; no. 12; pp. 5083 - 5090
• Main Authors: Kosmala, Aleksander, Weng, Andreas Max, Krauss, Bernhard, Knop, Stefan, Bley, Thorsten Alexander, Petritsch, Bernhard
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.12.2018; Springer Nature B.V
• Subjects: Adult; Aged; Aged, 80 and over; Attenuation; Axial skeleton; Bone imaging; Bone marrow; Bone Marrow - diagnostic imaging; Calcium; Computed Tomography; Diagnostic Radiology; Diagnostic systems; Disorders; Female; Humans; Imaging; Infiltration; Informed consent; Internal Medicine; Interventional Radiology; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; Male; Mathematical models; Medical diagnosis; Medical imaging; Medicine; Medicine & Public Health; Middle Aged; Multiple myeloma; Multiple Myeloma - diagnosis; Neuroradiology; NMR; Nuclear magnetic resonance; Patients; Radiology; Reproducibility of Results; ROC Curve; Skeleton; Tomography, X-Ray Computed - methods; Ultrasound; Bone marrow; Tomography; Dual-energy scanned projection radiography; X-ray computed; Multiple myeloma
• ISSN: 0938-7994; 1432-1084; 1432-1084
• DOI: 10.1007/s00330-018-5537-5

Objectives Bone marrow imaging patterns in patients with multiple myeloma possess prognostic and potential therapeutic relevance. We aim to evaluate whether different magnetic resonance imaging (MRI) patterns also result in different bone marrow dual-energy computed tomography (DECT) virtual non-calcium (VNCa) attenuation values. Methods The institutional review board approved this study. Written informed consent was obtained from all participants. 53 patients with plasma cell disorders (24 with normal imaging pattern, 24 with focal infiltration, 5 with diffuse infiltration) and 21 control subjects sequentially underwent DECT and MRI of the axial skeleton. MRI served as reference standard for imaging pattern assessment. Bone marrow VNCa attenuation numbers were obtained according to pattern allocation. Generalised estimating equations and a receiver operating characteristic (ROC) analysis were performed. Results Mean VNCa attenuation numbers in patients with normal, focal and diffuse imaging patterns were – 65.8 HU, 3.3 HU and – 13.3 HU, respectively. We found significant differences between diffuse vs. normal ( p < 0.001), diffuse vs. focal ( p = 0.002) and normal vs. focal ( p < 0.001) patterns. A cut-off of – 35.7 HU showed a sensitivity of 100% (24/24) and specificity of 97% (116/120) for the identification of a diffuse pattern vs. normal pattern, with an area under the ROC curve of 0.997. Conclusions Bone marrow VNCa attenuation numbers of various imaging patterns in patients with plasma cell disorders differ significantly and a diffuse imaging pattern can be determined confidently using DECT, when ROIs are carefully selected on the basis of MRI findings. Key Points • DECT allows for imaging pattern allocation similar to MRI. • Bone marrow VNCa attenuation numbers differ significantly depending on the imaging pattern. • A diffuse imaging pattern can be determined confidently using DECT.