Indra C. Pieters van den Bos

This systematic review of studies compared magnetic resonance imaging (MRI), (18) F-fluorodeoxyglucose positron emission tomography (FDG-PET), FDG-PET with computerized tomography (PET-CT) and CT with whole body X-Ray (WBXR) or (whole body) CT in order to provide evidence-based diagnostic guidelines in multiple myeloma bone disease. A comprehensive search of 3 bibliographic databases was performed; methodological quality was assessed using Quality Assessment of Diagnostic Accuracy Studies (QUADAS) criteria (score 1-14). Data from 32 directly comparative studies were extracted. The mean QUADAS score was 7·1 (3-11), with quality hampered mainly by a poor description of selection and execution criteria. All index tests had a higher detection rate when compared to WBXR, with up to 80% more lesions detected by the newer imaging techniques; MRI (1·12-1·82) CT (1·04-1·33), PET (1·00-1·58) and PET-CT (1·27-1·45). However, the modern imaging techniques detected fewer lesions in the skull and ribs. In a direct comparison CT and MRI performed equally with respect to detection rate and sensitivity. This systematic review supports the International Myeloma Working Group guidelines, which recommend that WBCT can replace WBXR. In our opinion, the equal performance of MRI also indicates that it is a valuable alternative. As lesions of the skull and ribs are underdiagnosed by modern imaging techniques we advise additional X-rays of these regions. The consequences of this approach are discussed.

PURPOSE: To evaluate magnetic resonance (MR) imaging morphologic- and signal intensity abnormalities of deep infiltrating endometriosis (DIE) of the bowel wall and to assess its value in predicting depth and extent of bowel wall infiltration. MATERIALS AND METHODS: This single-center study was performed in a tertiary referral center for endometriosis. All patients (n = 28) who underwent segmental bowel resection (2004-2010) were retrospectively studied. MR images were analyzed by two experienced readers independently (number of lesions, location, size, signal intensity, and depth of bowel wall infiltration) and this was correlated with histopathology. RESULTS: The sensitivity, specificity, positive and negative predictive values, and accuracy for diagnosis of endometriosis infiltrating the muscular layer of the bowel were 100%, 75%, 96%, 100%, and 96%, respectively. The inter-rater agreement was 0.84. "Fan shaped" configurations with hypointensity on T2- and T1-weighted imaging were characteristic for thickening of indigenous smooth muscle and smooth muscle hyperplasia at histopathology, as a consequence of infiltration by endometriosis. Thickening of the (sub)mucosa corresponded to edema with or without infiltration of endometriosis. CONCLUSION: MR imaging at 1.5 Tesla is useful to predict muscular infiltration of the bowel in endometriosis, whereas it is of limited value in diagnosis of (sub)mucosal infiltration.