Pattern of cerebral hypoperfusion in Alzheimer's disease and mild cognitive impairment measured with arterial spin-labeling MR imaging: Initial experience

Nathan A. Johnson, Geon‐Ho Jahng, Michael W. Weiner, Bruce L. Miller, Helena C. Chui, William J. Jagust, Maria Luisa Gorno‐Tempini, Norbert SchuffInternational Congress Series 2006 · DOI

TL;DR

This study demonstrates that arterial spin-labeling (ASL) MRI can effectively detect regional cerebral hypoperfusion in Alzheimer's disease (AD) and mild cognitive impairment (MCI) patients, mirroring findings from established PET and SPECT techniques. Crucially, ASL MRI offers a non-invasive, radiation-free alternative that can identify these early functional deficits, even when accounting for brain atrophy.

What Problem Does This Paper Address?

Alzheimer's disease (AD) is a devastating neurodegenerative disorder characterized by progressive cognitive decline. Early detection and monitoring of AD are critical for developing and testing potential therapies. Current methods for assessing brain function, such as fluorodeoxyglucose (FDG) positron emission tomography (PET) and hexamethylpropyleneamine oxime (HMPAO) single photon emission computed tomography (SPECT), have shown that AD is associated with reduced cerebral metabolism or blood flow, particularly in the temporo-parietal and posterior cingulate cortices. However, these techniques involve radiation exposure and/or the use of contrast agents.

Mild cognitive impairment (MCI) is a transitional stage between normal aging and dementia, with individuals with MCI having a higher risk of developing AD. Identifying functional changes in MCI is crucial for understanding disease progression and for early intervention. While PET and SPECT have been used to study MCI, there's a need for non-invasive methods to characterize these early functional deficits. This paper addresses the need for a non-invasive, radiation-free imaging technique that can accurately detect regional cerebral hypoperfusion in both AD and MCI, and importantly, distinguish these functional changes from structural atrophy.

Methods

The researchers employed a pulsed arterial spin-labeling (ASL) magnetic resonance (MR) imaging technique called DIPLOMA. This method uses the body's own arterial blood water as an endogenous tracer to measure brain perfusion. Key aspects of their methodology include:

  • ASL Imaging: A specific ASL sequence (DIPLOMA) was used to label arterial blood water. This involves a series of inversion pulses to tag and control arterial blood, followed by image acquisition.
  • Image Registration and Normalization: Perfusion images were coregistered to structural T1-weighted MR images to align functional and anatomical data. This was followed by spatial normalization to a study-specific template to allow for group comparisons.
  • Partial Volume Effect (PVE) Correction: A significant innovation in this study was the correction for partial volume effects. This addresses the issue where voxels in the ASL image may contain a mixture of gray matter, white matter, and cerebrospinal fluid, which can confound perfusion measurements, especially in the presence of brain atrophy. They developed a method to estimate the proportion of each tissue type within each voxel and used this information to correct the perfusion data.
  • Statistical Analysis: A voxel-by-voxel analysis of covariance (ANCOVA) was performed to compare perfusion between groups (AD vs. CN, MCI vs. CN, AD vs. MCI). This analysis was adjusted for age and, in some cases, for global cerebral perfusion to account for intersubject variability. They also performed a conjunction analysis to assess group effects independent of sex differences.

Datasets

The study utilized data from three groups of participants:

  • Alzheimer's Disease (AD) group: 20 subjects (13 men, 7 women; mean age 72.9 years).
  • Mild Cognitive Impairment (MCI) group: 18 subjects (9 men, 9 women; mean age 73.3 years).
  • Cognitively Normal (CN) control group: 23 subjects (10 men, 13 women; mean age 72.9 years).

Participants were recruited from two academic dementia centers. All subjects underwent structural MR imaging and ASL MR imaging.

Key Insights and Ideas

  • ASL MRI as a Viable Alternative: The study successfully demonstrates that ASL MRI can reliably detect regional hypoperfusion in AD, mirroring findings from established PET and SPECT studies. This offers a promising non-invasive, radiation-free alternative for clinical assessment.
  • Hypoperfusion in Early Stages: The findings suggest that hypoperfusion is present not only in established AD but also in MCI, indicating that ASL MRI can detect functional changes even in the early stages of the disease process.
  • Independence from Atrophy: A critical insight is that the observed hypoperfusion patterns in AD are largely independent of underlying cortical gray matter atrophy. This is a significant advantage over purely structural imaging and highlights the ability of ASL MRI to capture true functional deficits.
  • PVE Correction Matters: The study emphasizes the importance of correcting for partial volume effects, especially given the relatively large voxel size in ASL imaging and the presence of atrophy in AD. Their PVE correction method helped to refine the hypoperfusion findings.
  • MCI Hypoperfusion Overlap with AD: The observation that hypoperfusion in MCI patients was localized to regions similar to those most affected in AD patients supports the hypothesis that MCI represents an early stage of AD pathology.

Key Results

  • AD vs. CN: The AD group showed significant regional hypoperfusion compared to the CN group in the bilateral posterior cingulate gyri, bilateral precuneus, and right inferior parietal lobe. Significant hypoperfusion was also observed in bilateral superior and middle frontal gyri.
  • AD vs. CN (adjusted for global perfusion and PVEs): Even after accounting for global perfusion and PVEs, significant hypoperfusion remained in the AD group, particularly in the right parietal lobe and right middle frontal gyrus. This reinforces the finding that these deficits are not solely due to atrophy.
  • MCI vs. CN: While no significant hypoperfusion was found in the MCI group initially, a lower voxel-level threshold (P<0.01) revealed significant hypoperfusion in the right inferior parietal lobe. This region was similar to the most affected areas in the AD group.
  • AD vs. MCI: When comparing AD to MCI, significant hypoperfusion was found in the AD group in the cingulate gyri and portions of the inferior parietal lobes, but these regions did not overlap with the most affected areas in the MCI group when compared to CN.

Limitations and Future Work

  • Limited Coverage: The study acknowledges a significant limitation: data were acquired only for the superior cerebral cortex due to technical constraints. This excluded major portions of the temporal lobes and inferior frontal lobes, which are known to be affected in AD. Future work should aim to include these regions.
  • Transit Time and Relaxation Times: The study did not measure transit time and T1 relaxation times, which could potentially influence perfusion measurements and mimic hypoperfusion. Future studies could incorporate quantitative ASL techniques that account for these factors.
  • MCI Heterogeneity: The MCI group was heterogeneous, likely including individuals who will and will not convert to AD. Future research with larger sample sizes and longer follow-up periods is needed to better understand the prognostic value of ASL findings in MCI.
  • Early Detection Potential: The authors suggest that ASL MRI is a useful tool for the functional characterization of AD and may be valuable for early detection. Future longitudinal studies are needed to confirm this and to track disease progression.

Summary updated

BibTeX
@inproceedings{johnson2006pattern,
  author = {Nathan A. Johnson and Geon‐Ho Jahng and Michael W. Weiner and Bruce L. Miller and Helena C. Chui and William J. Jagust and Maria Luisa Gorno‐Tempini and Norbert Schuff},
  title = {{Pattern of cerebral hypoperfusion in Alzheimer's disease and mild cognitive impairment measured with arterial spin-labeling MR imaging: Initial experience}},
  booktitle = {International Congress Series},
  year = {2006},
  doi = {10.1016/j.ics.2006.04.009},
}
Figure