Diagnostic performance of computed tomography and diffusion-weighted imaging as first-line imaging modality according to the International Myeloma Working Group (IMWG) imaging algorithm for monoclonal plasma cell disorders

Abstract

Background The latest International Myeloma Working Group (IMWG) guideline recommends low-dose whole-body (WB) computed tomography (CT) as the first-line imaging technique for the initial diagnosis of plasma cell disorders. Purpose To evaluate diagnostic performances of CT and diffusion-weighted imaging (DWI) as the first-line imaging modalities and assess misclassification rates obtained following the guideline. Material and Methods Two independent radiologists analyzed CT (acquired as PET/CT) and DWI (3-T; b-values = 50 and 900 s/mm2) of patients newly diagnosed with plasma cell disorder, categorizing the number of bone lesions. Diagnostic performance of CT and DWI was compared using the McNemar test, and misclassification rates were calculated with a consensus WB-MRI reading as the reference standard. Differences in lesion number categories were assessed using marginal homogeneity and kappa statistics. Results Of 56 patients (36 men; mean age = 63.5 years), 39 had myeloma lesions. DWI showed slightly higher sensitivity for detecting myeloma lesions (97.4%) than CT (84.6%–92.3%; P > 0.05). CT showed significantly higher specificity (88.2%) than DWI (52.9%–58.8%; P<0.05). CT had a higher additional study requirement rate than DWI (7.7%–15.4% vs. 2.6%), but a lower unnecessary treatment rate (11.8% vs. 41.2%–47.1%). Both readers showed significant differences in categorization of the number of lesions on CT compared with the reference standard ( P < 0.001), and one reader showed a significant difference on DWI ( P = 0.006 and 0.098). Conclusion CT interpreted according to the IMWG guideline is a diagnostically effective first-line modality with relatively high sensitivity and specificity. DWI alone may not be an acceptable first-line imaging modality because of low specificity.