Contributions
Abstract: P503
Type: Poster
Abstract Category: Pathology and pathogenesis of MS - Imaging
Background: Despite their abundance in multiple sclerosis (MS), conventional magnetic resonance imaging (MRI) techniques often miss grey matter (GM) lesions. We previously demonstrated that the transverse relaxation (R2*) of gradient echo MRI signal is sensitive to GM injury in MS. Our prior study of relapsing-remitting (RRMS), secondary-progressive (SPMS) and primary progressive (PPMS) patients analyzed regional R2* values in cortical GM, finding R2* values to correlate better with cognition than cortical volumes. Here, we analyzed quantitative R2* in two deep GM structures involved in memory.
Objective: We hypothesized that R2* of the hippocampus and amygdala would correlate with cognitive tests. Hippocampus was examined due to its importance for memory and amygdala because of its physical and neural connections to hippocampus.
Methods: Human studies committee approval was obtained. All subjects gave informed consent. 26 RRMS, 12 PPMS, 25 SPMS and 26 healthy controls (HC) underwent MRI to calculate quantitative R2* parameters. Paced auditory addition task (PASAT - auditory working memory) and symbol digit modalities test (SDMT - visual working memory) were done the same day. All volumes and R2* values were age-corrected. Volume was normalized to skull size. Due to known gender differences in the amygdala, its volume and R2* were compared between MS and HC stratified by sex.
Results: Hippocampal R2* (p< .03) and volume (p< .04) were lower in MS than HC. In amygdala, MS females had lower R2* than female HC in left amygdala (p=.02), whereas R2* on the right and amygdala volumes were not different between MS and HC. In all patients, amygdala and hippocampus volumes correlated with one another (r=.54 for left hemisphere, r=.69 for right), whereas R2* of hippocampus and amygdala were mildly correlated only in right hemisphere (r=.35). In RRMS, hippocampus (r=.4) and amygdala (r=.4) volumes correlated with SDMT, but not with PASAT. R2* did not correlate with SDMT or PASAT in either structure in RRMS. In SPMS, R2* of hippocampus and amygdala significantly correlated with PASAT and SDMT (r= 0.4 for all). Volumes did not correlate with test scores in SPMS.
Conclusions: Hippocampal R2* and volume were reduced in MS vs. HC. Hippocampus and amygdala volumes correlated with SDMT in RRMS, while R2* in hippocampus and amygdala correlated with both SDMT and PASAT in SPMS. These data suggest that R2* reflects injury in hippocampus and amygdala in SPMS patients.
Disclosure:
Jie Wen has nothing to disclose.
Anne H. Cross has done consulting for AbbVie, Biogen, EMD Serono, Genentech/Roche, Genzyme/Sanofi, Mallinckrodt, Novartis, and Teva.
Dmitriy A. Yablonskiy has nothing to disclose.
Funding: Funded by the National MS Society USA and the Conrad N. Hilton Foundation.
Jie Wen was a post-doctoral fellow of the NMSS during these studies.
Abstract: P503
Type: Poster
Abstract Category: Pathology and pathogenesis of MS - Imaging
Background: Despite their abundance in multiple sclerosis (MS), conventional magnetic resonance imaging (MRI) techniques often miss grey matter (GM) lesions. We previously demonstrated that the transverse relaxation (R2*) of gradient echo MRI signal is sensitive to GM injury in MS. Our prior study of relapsing-remitting (RRMS), secondary-progressive (SPMS) and primary progressive (PPMS) patients analyzed regional R2* values in cortical GM, finding R2* values to correlate better with cognition than cortical volumes. Here, we analyzed quantitative R2* in two deep GM structures involved in memory.
Objective: We hypothesized that R2* of the hippocampus and amygdala would correlate with cognitive tests. Hippocampus was examined due to its importance for memory and amygdala because of its physical and neural connections to hippocampus.
Methods: Human studies committee approval was obtained. All subjects gave informed consent. 26 RRMS, 12 PPMS, 25 SPMS and 26 healthy controls (HC) underwent MRI to calculate quantitative R2* parameters. Paced auditory addition task (PASAT - auditory working memory) and symbol digit modalities test (SDMT - visual working memory) were done the same day. All volumes and R2* values were age-corrected. Volume was normalized to skull size. Due to known gender differences in the amygdala, its volume and R2* were compared between MS and HC stratified by sex.
Results: Hippocampal R2* (p< .03) and volume (p< .04) were lower in MS than HC. In amygdala, MS females had lower R2* than female HC in left amygdala (p=.02), whereas R2* on the right and amygdala volumes were not different between MS and HC. In all patients, amygdala and hippocampus volumes correlated with one another (r=.54 for left hemisphere, r=.69 for right), whereas R2* of hippocampus and amygdala were mildly correlated only in right hemisphere (r=.35). In RRMS, hippocampus (r=.4) and amygdala (r=.4) volumes correlated with SDMT, but not with PASAT. R2* did not correlate with SDMT or PASAT in either structure in RRMS. In SPMS, R2* of hippocampus and amygdala significantly correlated with PASAT and SDMT (r= 0.4 for all). Volumes did not correlate with test scores in SPMS.
Conclusions: Hippocampal R2* and volume were reduced in MS vs. HC. Hippocampus and amygdala volumes correlated with SDMT in RRMS, while R2* in hippocampus and amygdala correlated with both SDMT and PASAT in SPMS. These data suggest that R2* reflects injury in hippocampus and amygdala in SPMS patients.
Disclosure:
Jie Wen has nothing to disclose.
Anne H. Cross has done consulting for AbbVie, Biogen, EMD Serono, Genentech/Roche, Genzyme/Sanofi, Mallinckrodt, Novartis, and Teva.
Dmitriy A. Yablonskiy has nothing to disclose.
Funding: Funded by the National MS Society USA and the Conrad N. Hilton Foundation.
Jie Wen was a post-doctoral fellow of the NMSS during these studies.