1. Diffusion-Tensor Imaging: Frontal Executive Function in Vascular Cognitive Impairment Stephen Salloway, MD Stephen Correia, PhD Paul Malloy, PhD William Heindel, PhD David Laidlaw, PhD 19 April 2005 CTBR Symposium 2005

3. Goal To develop diffusion-tensor imaging as an imaging biomarker of white matter integrity in aging and dementia

4. Research Focus Frontal Systems Disruption ↓ Changes in Executive Cognition and Behavior ↓ Functional Disability/Conversion to Dementia

5. Frontal Systems: Subcortical-thalamic connections dorsomedial Striatum Globus Pallidus/ Substantia Nigra Frontal Cortex Thalamus The prefrontal cortex is connected to the striatum and thalamus in parallel but separate circuits that help regulate behavior The prefrontal cortex is connected to the striatum and thalamus in parallel but separate circuits that help regulate behavior Topographic mapping of caudate and thalamus Topographic mapping of caudate and thalamus Subcortical white matter connections Subcortical white matter connections Long tracts Long tracts Cortical U-fibers Cortical U-fibers Injury anywhere in a circuit can produce a major deficit and small small subcortical lesions can mimic large cortical lesions Injury anywhere in a circuit can produce a major deficit and small small subcortical lesions can mimic large cortical lesions

6. Frontal Systems Function Processing speed Processing speed Mental flexibility Mental flexibility Planning Planning Sequencing Sequencing Decision-making Decision-making Working memory Working memory Behavioral regulation, self-monitoring Behavioral regulation, self-monitoring Motivation, drive, interest Motivation, drive, interest

7. White Matter Changes in Aging Volume loss Volume loss Greater than grey matter loss Greater than grey matter loss Greater in frontal lobes Greater in frontal lobes Loss of myelin Loss of myelin Wallerian degeneration Wallerian degeneration Subcortical ischemic vascular changes Subcortical ischemic vascular changes Selective vulnerability of frontal regions Selective vulnerability of frontal regions Increased interstitial fluid Increased interstitial fluid Peters, A. (2002) J. Neurocyt. 31: 581-93; Jernigan et al. (2001) Neurobiol Aging 22(4): 581-94; Guttman et al. (1998) Neurology 50(4): 972-8

8. Subcortical Hyperintensities None Mild ModerateSevere Malloy & Correia, The Clinical Neuropsychologist, in press

9. Vascular Cognitive Impairment Cognitive impairment due to cerebrovascular disease Cognitive impairment due to cerebrovascular disease Subcortical Ischemic Vascular Disease (SIVD) Subcortical Ischemic Vascular Disease (SIVD) Most common form Most common form Increases with age & cardiovascular risk factors Increases with age & cardiovascular risk factors Features of SIVD: Features of SIVD: Impaired executive function/mental flexibility Impaired executive function/mental flexibility Cognitive slowing Cognitive slowing Apathy & depression Apathy & depression

10. Diffusion-Tensor Imaging MRI technique that provides in-vivo characterization of 3D white matter microstructure. MRI technique that provides in-vivo characterization of 3D white matter microstructure. Measures magnitude and direction of water diffusion in biological tissue in 3D. Measures magnitude and direction of water diffusion in biological tissue in 3D. More sensitive to white matter changes than conventional MRI sequences. More sensitive to white matter changes than conventional MRI sequences. Detects changes in normal-appearing white matter (NAWM) that correlate w/cognition Detects changes in normal-appearing white matter (NAWM) that correlate w/cognition

11. DTI Basics Rosenbloom M, et al. (July 2004). NIAA pubs; http://www.niaaa.nih.gov/publications/arh27-2/146-152.htm

12. DTI Basics Measures water diffusion in at least 6 directions – we use 12 for better resolution Measures water diffusion in at least 6 directions – we use 12 for better resolution 1.5T magnet or greater capable of diffusion encoding 1.5T magnet or greater capable of diffusion encoding Echo-planar imaging (fast acquisition) Echo-planar imaging (fast acquisition) Collecting small voxels, scanning takes about 14 minutes Collecting small voxels, scanning takes about 14 minutes Off-line post-processing (Laidlaw lab) Off-line post-processing (Laidlaw lab) Image analysis: Butler Hospital Quantitative Imaging Lab Image analysis: Butler Hospital Quantitative Imaging Lab

13. DTI – Tractography Bammer, 2003

14. DTI Scalar Parameters Trace: Magnitude of diffusion in a voxel. Trace: Magnitude of diffusion in a voxel. Increases in damaged white matter Increases in damaged white matter Fractional Anisotropy (FA): Measure of directionally- restricted diffusion. Fractional Anisotropy (FA): Measure of directionally- restricted diffusion. Decreases in damaged white matter Decreases in damaged white matter Rosenbloom M, et al. (July 2004). NIAA pubs; http://www.niaaa.nih.gov/publications/arh27-2/146-152.htm

15. Prior Studies of DTI DTI in Aging: DTI in Aging: Anterior – posterior gradient of DTI changes. (e.g., Pfefferbaum, 2000) Anterior – posterior gradient of DTI changes. (e.g., Pfefferbaum, 2000) Correlations w/executive function. (e.g.; O’Sullivan, 2001, Madden 2004) Correlations w/executive function. (e.g.; O’Sullivan, 2001, Madden 2004) DTI in SIVD: DTI in SIVD: DTI abnormalities in normal appearing white matter (NAWM) DTI abnormalities in normal appearing white matter (NAWM) Those DTI changes more strongly correlated w/executive function than DTI in SH. (O’Sullivan 2004) Those DTI changes more strongly correlated w/executive function than DTI in SH. (O’Sullivan 2004)

16. AIMS To determine the integrity of anterior vs. posterior normal appearing white matter (NAWM) in VCI vs. controls using DTI. To determine the integrity of anterior vs. posterior normal appearing white matter (NAWM) in VCI vs. controls using DTI. H: VCI will show poorer anterior & posterior NAWM integrity than controls on DTI H: VCI will show poorer anterior & posterior NAWM integrity than controls on DTI To determine the association between anterior DTI parameters in NAWM and executive function and processing speed. To determine the association between anterior DTI parameters in NAWM and executive function and processing speed. H: DTI parameters in anterior NAWM will correlate with executive functioning and processing speed H: DTI parameters in anterior NAWM will correlate with executive functioning and processing speed

17. Subjects 6 VCI (scanned 9) 6 VCI (scanned 9) SIVD (sporadic or genetic forms [CADASIL]) SIVD (sporadic or genetic forms [CADASIL]) Impaired executive function &/or memory Impaired executive function &/or memory MMSE > 24 MMSE > 24 No cortical strokes No cortical strokes No dementia No dementia 6 Normal elderly controls (NEC) (scanned 10) 6 Normal elderly controls (NEC) (scanned 10)

18. Method Structural MRI, DTI Structural MRI, DTI Primary imaging outcomes: Primary imaging outcomes: DTI: FA & Trace DTI: FA & Trace Structural: parenchymal & SH volumes Structural: parenchymal & SH volumes Cognitive battery Cognitive battery Executive function Executive function DRS I/P, TMT-B, COWA DRS I/P, TMT-B, COWA Processing Speed Processing Speed TMT-A, Symbol-digit TMT-A, Symbol-digit Analysis: ANOVA, chi-square, correlation Analysis: ANOVA, chi-square, correlation

19. Results NEC(n=6)VCI (n = 6) p Age (yrs) 64.83 ± 7.89 62.13 ± 11.40 ns Education (yrs) 15.17 ± 4.12 14.00 ± 4.00 ns # Female 24ns MMSE 28.83 ± 1.62 28.17 ± 1.33 ns Correia S (2005) 33 rd INS meeting; St. Louis, MO Brennan-Krohn, T (2004) ISMRM Workshop; Boston, MA Correia, S (2004) 9 th ICAD; Philadelphia, PA Laidlaw, L (2004) 12 th ISMRM meeting; Kyoto, Japan

20. DTI Analysis – ROI Placement

21. Imaging Results VCI group had higher estimated SH volume (p =.040) VCI group had higher estimated SH volume (p =.040) Estimated parenchymal volume not different across groups (p =.378) Estimated parenchymal volume not different across groups (p =.378) DTI: No significant differences in FA or Trace in SH DTI: No significant differences in FA or Trace in SH

22. DTI – Trace Trace units: x 10 -3 mm 2 /s p =.033

23. DTI – FA p =.033 p = nsp <.001

24. Processing Speed p =.052 p = ns

25. Executive Measures p = ns

26. Executive Measures p =.003

27. Correlations DTI in NAWM: DTI in NAWM: TMT A & B: TMT A & B: Inversely correlated with anterior & posterior FA Inversely correlated with anterior & posterior FA Positively correlated with anterior & posterior Trace Positively correlated with anterior & posterior Trace all p <.05 all p <.05 DTI in SH: DTI in SH: TMT A &B not correlated with FA or Trace (p > 0.5) TMT A &B not correlated with FA or Trace (p > 0.5) SH Volume: SH Volume: TMT A & B correlate inversely with SH volume (r = -.70, p <.05) TMT A & B correlate inversely with SH volume (r = -.70, p <.05)

28. Conclusions Patients with VCI preform more poorly than NEC on tests of processing speed and executive functioning. Patients with VCI preform more poorly than NEC on tests of processing speed and executive functioning. VCI alters anterior and posterior NAWM. VCI alters anterior and posterior NAWM. Processing speed and executive functioning correlate with DTI parameters in NAWM but not in SH. Processing speed and executive functioning correlate with DTI parameters in NAWM but not in SH. DTI in NAWM may predict executive/processing speed better than SH volume DTI in NAWM may predict executive/processing speed better than SH volume

30. Future Directions Successful R03 to study further Successful R03 to study further Continue data collection and analysis Continue data collection and analysis Compare with amnestic MCI & AD Compare with amnestic MCI & AD Progression of white matter changes as imaging biomarker in clinical trials Progression of white matter changes as imaging biomarker in clinical trials DTI in cerebrovascular risk factors (HTN, DM). DTI in cerebrovascular risk factors (HTN, DM). Tractography Tractography

31. Fiber Clustering Courtesy of Song Zhang

32. Courtesy of Stephanie Lee, Laidlaw Lab, Brown University

35. Acknowledgments Stephen Correia Paul Malloy David Laidlaw Song Zhang Thea Brennan-Krohn Erin Schlicting Jerome Sanes Lynn Fanella David Tate Stephanie Lee

36. Support Center for Translational Brain Research NIA AG020498-02 Alzheimer’s Association NIRG-03-6195 Start-MH Grant NIMH K08MH01487W The Human Brain Project (NIBIB & NIMH) Ittleson Fund at Brown P20 NCRR15578-01 Brown VP for Research Seed Funds

37. THANK YOU CTBR!

38. Threshold to Dementia Cognitive Continuum Functional Continuum Normal Mild Cognitive Impairment Dementia Adapted from Petersen

39. MCI not a unitary construct Petersen, 2003

40. Why study frontal systems in VCI? Independent probes of frontal systems: Independent probes of frontal systems: Tests of executive function & processing speed Tests of executive function & processing speed DTI DTI The combination may help identify patients at greatest risk for dementia The combination may help identify patients at greatest risk for dementia

41. Image Analysis Skull stripping and parenchymal volume estimation

42. Image Analysis SH volume: Performed on pseudo-3D FLAIR images SH thresholding following skull stripping w/operator correction Sum of all voxels with intensity levels within SH threshold range

43. Regression Exploratory regression Exploratory regression IV: SH volume, anterior & posterior FA, Trace IV: SH volume, anterior & posterior FA, Trace DV: TMT A & B DV: TMT A & B Results Results Posterior FA significantly predicted TMT A & B; SH volume, anterior FA did not. Posterior FA significantly predicted TMT A & B; SH volume, anterior FA did not. Trace: overall model significant only. Trace: overall model significant only.

44. DTI Results Trace units: 10 -3 mm 2 /s Between subjects effect: p =.032 Within subjects effect: p =.025 Interaction effect: p =.057

45. DTI Results Between subjects effect: p =.012 Within subjects effect: p =.058 Interaction effect: p <.001

46. DTI Acquisition Siemens Symphony 1.5T Siemens Symphony 1.5T 3 acquisitions with offset in slice direction by 0.0mm, 1.7 mm and 3.4 mm, 5mm thick slices 3 acquisitions with offset in slice direction by 0.0mm, 1.7 mm and 3.4 mm, 5mm thick slices 0.1mm inter-slice spacing, 30 slices per acquisition 0.1mm inter-slice spacing, 30 slices per acquisition matrix = 128 mm x128 mm; FOV = 21.7cm x 21.7cm, in- plane sample spacing was 0.85 mm matrix = 128 mm x128 mm; FOV = 21.7cm x 21.7cm, in- plane sample spacing was 0.85 mm TR=7200, TE=156 TR=7200, TE=156 b values: (0, 500, 1000 mm 2 /s) or (0, 1000 mm 2 /s) b values: (0, 500, 1000 mm 2 /s) or (0, 1000 mm 2 /s) 12 non-collinear directions, 12 non-collinear directions, The first three datasets were interleaved and zero-filled in the slice direction to form a fourth dataset with resulting inter-slice distance of 0.85 mm. The first three datasets were interleaved and zero-filled in the slice direction to form a fourth dataset with resulting inter-slice distance of 0.85 mm. FA and Trace maps derived. FA and Trace maps derived.

47. Additional MRI Acquisitions 3D T1 volume (MPRAGE) for volumetric analysis 3D T1 volume (MPRAGE) for volumetric analysis 3 interleaved FLAIR acquisitions concatenated into a pseudo 3D volume for assessment of SH volume 3 interleaved FLAIR acquisitions concatenated into a pseudo 3D volume for assessment of SH volume Voxel dimensions on MPRAGE & pseudo FLAIR match DTI. Voxel dimensions on MPRAGE & pseudo FLAIR match DTI.