Frédéric E Lecouvet1, Vassiliki Pasoglou2, Sandy Van Nieuwenhove2, Thomas Van Haver2, Quentin de Broqueville2, Vincent Denolin3, Perrine Triqueneaux2, Bertrand Tombal2, Nicolas Michoux2. 1. Department of Radiology an Medical Imaging, Centre du Cancer and Institut de Recherche Expérimentale et Clinique (IREC), Cliniques Universitaires Saint-Luc, Université Catholique de Louvain (UCLouvain), Avenue Hippocrate 10/2942, B-1200, Brussels, Belgium. frederic.lecouvet@uclouvain.be. 2. Department of Radiology an Medical Imaging, Centre du Cancer and Institut de Recherche Expérimentale et Clinique (IREC), Cliniques Universitaires Saint-Luc, Université Catholique de Louvain (UCLouvain), Avenue Hippocrate 10/2942, B-1200, Brussels, Belgium. 3. Philips Medical Systems International BV, Veenpluis 4-6, 5684 PC, Best, The Netherlands.
Abstract
PURPOSE: To compare 3D T1-weighted fast spin echo (FSE) and 3D T1-weighted gradient echo (GE) mDixon as morphologic sequences to complement diffusion-weighted imaging (DWI) for the metastatic screening in prostate cancer (PCa) patients. MATERIALS AND METHODS: Thirty PCa patients at high risk of metastases prospectively underwent both a 3D T1 FSE (14 min) and a rapid 3D T1 GEmDixon (1 min 20 s) sequences within a WB-MRI protocol. Two readers assessed the diagnostic performance of the FSE/Fat/in-phase (IP)/IP+Fat sequences in detecting bone and node metastases. The reference standard was established by a panel of four physicians on the basis of all baseline and follow-up imaging, biological and clinical information. The reproducibility of readings, predictive accuracy (Acc) from ROC curves analysis, and contrast-to-reference ratio (CRR) in lesions were assessed for each sequence. RESULTS: In bone and lymph nodes (per-region analysis), reproducibility was at least good for all sequences/readers, except for nodes in the common iliac/inguinal regions. In bone (per-organ analysis), Acc of FSE was superior to that of mDixon (difference + 4%, p < 0.0083). In nodes (per-organ analysis), Acc of Fat was superior to that of other sequences (difference + 4% to + 6% depending on reader, p < 0.0083). In the per-patient analysis, Acc of FSE was superior to that of mDixon (difference + 4% to + 6% depending on sequence, p < 0.0083). Fat images had higher CRR compared with FSE in the thoracic spine, the bony pelvis and lymph node metastases (p < 0.025). CONCLUSION: 3D T1 GEmDixon may replace 3D T1 FSE to complement DWI in WB-MRI for metastatic screening in PCa. It demonstrates an Acc ranging from + 4% to + 6% (nodes) to - 4% to - 6% (bone and patient staging) compared with FSE and considerably reduces the examination time, offering the perspective of acquiring WB-MRI examinations in less than 20 min. KEY POINTS: • The replacement of 3D T1 FSE by the 3D T1 GE mDixon as morphologic sequence to complement DWI drastically reduces the acquisition time of WB-MRI studies. • The 3D T1 GE mDixon sequence offers similar reproducibility of image readings compared with that of the 3D T1 FSE. • Differences in diagnostic accuracy are limited (+ 4%/+ 6% in favor of mDixon to detect node metastases; + 4%/+ 6% in favor of FSE to detect bone metastases/metastatic disease in a patient).
PURPOSE: To compare 3D T1-weighted fast spin echo (FSE) and 3D T1-weighted gradient echo (GE) mDixon as morphologic sequences to complement diffusion-weighted imaging (DWI) for the metastatic screening in prostate cancer (PCa) patients. MATERIALS AND METHODS: Thirty PCapatients at high risk of metastases prospectively underwent both a 3D T1 FSE (14 min) and a rapid 3D T1 GEmDixon (1 min 20 s) sequences within a WB-MRI protocol. Two readers assessed the diagnostic performance of the FSE/Fat/in-phase (IP)/IP+Fat sequences in detecting bone and node metastases. The reference standard was established by a panel of four physicians on the basis of all baseline and follow-up imaging, biological and clinical information. The reproducibility of readings, predictive accuracy (Acc) from ROC curves analysis, and contrast-to-reference ratio (CRR) in lesions were assessed for each sequence. RESULTS: In bone and lymph nodes (per-region analysis), reproducibility was at least good for all sequences/readers, except for nodes in the common iliac/inguinal regions. In bone (per-organ analysis), Acc of FSE was superior to that of mDixon (difference + 4%, p < 0.0083). In nodes (per-organ analysis), Acc of Fat was superior to that of other sequences (difference + 4% to + 6% depending on reader, p < 0.0083). In the per-patient analysis, Acc of FSE was superior to that of mDixon (difference + 4% to + 6% depending on sequence, p < 0.0083). Fat images had higher CRR compared with FSE in the thoracic spine, the bony pelvis and lymph node metastases (p < 0.025). CONCLUSION: 3D T1 GEmDixon may replace 3D T1 FSE to complement DWI in WB-MRI for metastatic screening in PCa. It demonstrates an Acc ranging from + 4% to + 6% (nodes) to - 4% to - 6% (bone and patient staging) compared with FSE and considerably reduces the examination time, offering the perspective of acquiring WB-MRI examinations in less than 20 min. KEY POINTS: • The replacement of 3D T1 FSE by the 3D T1 GE mDixon as morphologic sequence to complement DWI drastically reduces the acquisition time of WB-MRI studies. • The 3D T1 GE mDixon sequence offers similar reproducibility of image readings compared with that of the 3D T1 FSE. • Differences in diagnostic accuracy are limited (+ 4%/+ 6% in favor of mDixon to detect node metastases; + 4%/+ 6% in favor of FSE to detect bone metastases/metastatic disease in a patient).
Entities:
Keywords:
Bones; Magnetic resonance imaging; Neoplasm metastasis; Prostate; Whole-body imaging
Authors: Benedikt J Schwaiger; Charlotte Schneider; Sophia Kronthaler; Florian T Gassert; Christof Böhm; Daniela Pfeiffer; Thomas Baum; Jan S Kirschke; Dimitrios C Karampinos; Marcus R Makowski; Klaus Woertler; Markus Wurm; Alexandra S Gersing Journal: Eur Radiol Date: 2021-01-14 Impact factor: 5.315
Authors: Frederic E Lecouvet; Marie-Christiane Vekemans; Thomas Van Den Berghe; Koenraad Verstraete; Thomas Kirchgesner; Souad Acid; Jacques Malghem; Joris Wuts; Jens Hillengass; Vincent Vandecaveye; François Jamar; Olivier Gheysens; Bruno C Vande Berg Journal: Skeletal Radiol Date: 2021-08-07 Impact factor: 2.199