Contributions
Abstract: P945
Type: Poster Sessions
Abstract Category: Therapy - Tools for detecting therapeutic response
Background: Brain atrophy is an important feature of multiple sclerosis (MS) and atrophy of the deep grey matter (DGM) structures has been shown to be strongly related to clinical and cognitive decline1. However, its accurate measurement is hampered by technical differences2. We aim to evaluate the reliability of atrophy measurements using different vendors.
Methods: We scanned 21 MS patients at baseline and year-1. Participants underwent same-day scan-rescan 3DT1-w MRI on three 3T scanners (GE, Philips, Toshiba). All images were lesion-filled by LST and segmented by FreeSurfer's longitudinal pipeline. Within and between-scanner agreements were assessed using intra-class correlation coefficient (ICC). Repeated-measures ANOVA was used to assess systematic differences.
Results: Cross-sectional within-scanner agreement at baseline was high for total white (WM) and grey matter (GM) (ICC>0.9), and for all DGM structures in all scanners (>0.8). Cross-sectional between-scanner agreement was also good for total GM and WM (consistency ICC>0.9) as well as DGM structures (>0.75) except for nucleus accumbens on Philips-Toshiba (ICC=0.52). Between-scanner volume differences for most DGM structures were significant (p< 0.05). Within-scanner reproducibility of longitudinal volume change measurement was good for total GM (ICC>0.75), moderate for total WM (< 0.4< ICC< 0.75), and heterogeneous but mostly poor for DGM structures. Longitudinal between-scanner agreement was again worse for the total WM than total GM, lowest being ICC< 0.35 for Philips-Toshiba. This comparison for DGM structures was more disappointing revealing extremely low ICCs for most structures. Statistical analysis of the longitudinal data illustrated significant systematic differences between-scanner; e.g., Philips-Toshiba bilateral thalamus.
Discussion: Cross-sectional volume estimation by FreeSurfer is reproducible within and between scanners, even for the DGM structures. However, this reproducibility drops remarkably in longitudinal analyses, especially for DGM. In part this may be related to the small changes in our patient group. Interestingly, even with good reproducibility, systematic between-scanner differences were seen in both cross-sectional and longitudinal analysis. Therefore, interpretation of the longitudinal volume-change measurements needs more caution and advancements in techniques are warranted.
References: 1Houtchens et al., Neurology, 2007. 2H. Amiri et al., NeuroImage: Clinical, 2018.
Disclosure: Houshang Amiri was sponsored by a research grant from Novartis Pharma.
Sophie Renckens: nothing to disclose.
Iman Brouwer was sponsored by research grants provided by Novartis and Teva.
Stephanie Bosschaert: nothing to disclose.
Rick van Tuijl: nothing to disclose.
Marloes Hagens: nothing to disclose.
Joep Killestein has accepted speaker and consultancy fees from Merck, Biogen, Teva, Genzyme, Roche and Novartis.
Frederik Barkhof is supported by the NIHR UCLH biomedical research centre, serves as editorial board member of Brain, European Radiology, Neurology, Multiple Sclerosis Journal and Radiology and has accepted consulting fees from Bayer-Schering Pharma, Biogen-IDEC, TEVA, Merck-Serono, Novartis, Roche, Jansen Research, Genzyme-Sanofi, IXICO Ltd, GeNeuro, Apitope Ltd and speaker fees from Biogen-IDEC and IXICO.
Hugo Vrenken has received research grants from Novartis, Teva and MerckSerono, and consulting fees from MerckSerono; all funds were paid directly to his institution.
Abstract: P945
Type: Poster Sessions
Abstract Category: Therapy - Tools for detecting therapeutic response
Background: Brain atrophy is an important feature of multiple sclerosis (MS) and atrophy of the deep grey matter (DGM) structures has been shown to be strongly related to clinical and cognitive decline1. However, its accurate measurement is hampered by technical differences2. We aim to evaluate the reliability of atrophy measurements using different vendors.
Methods: We scanned 21 MS patients at baseline and year-1. Participants underwent same-day scan-rescan 3DT1-w MRI on three 3T scanners (GE, Philips, Toshiba). All images were lesion-filled by LST and segmented by FreeSurfer's longitudinal pipeline. Within and between-scanner agreements were assessed using intra-class correlation coefficient (ICC). Repeated-measures ANOVA was used to assess systematic differences.
Results: Cross-sectional within-scanner agreement at baseline was high for total white (WM) and grey matter (GM) (ICC>0.9), and for all DGM structures in all scanners (>0.8). Cross-sectional between-scanner agreement was also good for total GM and WM (consistency ICC>0.9) as well as DGM structures (>0.75) except for nucleus accumbens on Philips-Toshiba (ICC=0.52). Between-scanner volume differences for most DGM structures were significant (p< 0.05). Within-scanner reproducibility of longitudinal volume change measurement was good for total GM (ICC>0.75), moderate for total WM (< 0.4< ICC< 0.75), and heterogeneous but mostly poor for DGM structures. Longitudinal between-scanner agreement was again worse for the total WM than total GM, lowest being ICC< 0.35 for Philips-Toshiba. This comparison for DGM structures was more disappointing revealing extremely low ICCs for most structures. Statistical analysis of the longitudinal data illustrated significant systematic differences between-scanner; e.g., Philips-Toshiba bilateral thalamus.
Discussion: Cross-sectional volume estimation by FreeSurfer is reproducible within and between scanners, even for the DGM structures. However, this reproducibility drops remarkably in longitudinal analyses, especially for DGM. In part this may be related to the small changes in our patient group. Interestingly, even with good reproducibility, systematic between-scanner differences were seen in both cross-sectional and longitudinal analysis. Therefore, interpretation of the longitudinal volume-change measurements needs more caution and advancements in techniques are warranted.
References: 1Houtchens et al., Neurology, 2007. 2H. Amiri et al., NeuroImage: Clinical, 2018.
Disclosure: Houshang Amiri was sponsored by a research grant from Novartis Pharma.
Sophie Renckens: nothing to disclose.
Iman Brouwer was sponsored by research grants provided by Novartis and Teva.
Stephanie Bosschaert: nothing to disclose.
Rick van Tuijl: nothing to disclose.
Marloes Hagens: nothing to disclose.
Joep Killestein has accepted speaker and consultancy fees from Merck, Biogen, Teva, Genzyme, Roche and Novartis.
Frederik Barkhof is supported by the NIHR UCLH biomedical research centre, serves as editorial board member of Brain, European Radiology, Neurology, Multiple Sclerosis Journal and Radiology and has accepted consulting fees from Bayer-Schering Pharma, Biogen-IDEC, TEVA, Merck-Serono, Novartis, Roche, Jansen Research, Genzyme-Sanofi, IXICO Ltd, GeNeuro, Apitope Ltd and speaker fees from Biogen-IDEC and IXICO.
Hugo Vrenken has received research grants from Novartis, Teva and MerckSerono, and consulting fees from MerckSerono; all funds were paid directly to his institution.