Whole-body low-dose multidetector row-CT in the diagnosis of multiple myeloma: an alternative to conventional radiography

Abstract

Objective: The goal of this study was to establish the feasibility of a low-dose whole-body multidetector row-CT (MDCT) protocol in the diagnosis of multiple myeloma (MM), as an alternative to conventional X-ray imaging, which is currently still state-of-the-art in these patients, with emphasis on the comparison of image resolution on axial and multiplanar reformatted (MPR) scans and reduction of radiation dose.

Material and methods: 100 patients with known MM, or monoclonal gammopathy of unknown significance (MGUS) underwent unenhanced whole-body MDCT on a 16-slice scanner in a randomised fashion, using a 16 x 1.5 mm collimation and four different energy parameters (40, 50, 60 and 70 mAs). Three different reconstruction algorithms were used in every patient (B40f, B50f and B60f kernel). CT scans were reviewed independently by two radiologists, with regard to correct classification into one of the three known MM stages, and recognition of fracture risk. Thereafter, axial and MPR images were evaluated in consensus by both readers, with respect to image resolution. Diagnosis of osteolytic lesions was performed on the basis of axial and multiplanar reformatted images, whereas the assessment of spinal misalignment and fracture was done only on MPR images. The distribution of image resolution categories (very good, good, sufficient, insufficient for diagnosis) was evaluated depending on following parameters: current time product, patient's weight, bone density and reconstruction algorithm. The effective radiation dose was determined with the aid of an anthropomorphic Alderson Rando-Phantom, using a tube current time product of 40 mAs, and then extrapolating it on all current time products applied in this study on a commercially available software program WinDose (Institute of Medical Physics, Erlangen, Germany).

Results: In all 100 patients, image resolution was diagnostic, regardless of scanning parameters, enabling correct classification of multiple myeloma patients. Image quality of MPR images was either equal or inferior to correspondent axial images in the delineation of smaller lytic lesions, because of the use of non-isotropic voxel size. However, they proved accurate in diagnosing fracture and spine misalignment. A strong dependency of image resolution on bone density was observed, with reduced quality in patients with either diffuse skeleton infiltration or concurrent osteoporosis. Spatial resolution was also dependent on the reconstruction algorithm and energy level (mAs) used, as well as on patient's weight, but their influence was low within the given ranges. A middle-frequency reconstruction algorithm (B50f kernel) proved beneficial for all energy protocols. The interobserver agreement was excellent (kappa = 0.95) for classification of MM-patients. Effective radiation dose of MDCT calculated at a tube current time product of 40 mAs was 1.7-fold higher than the mean radiation dose of conventional X-ray (4.1 mSv versus 2.4 mSv).

Discussion: Our study shows that whole-body low-dose MDCT investigational protocols are appropriate for the diagnosis of lytic bone changes and for assessment of fracture risk in multiple myeloma patients, representing a serious alternative to current standards.