Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print May 9, 2006, 10.1148/radiol.2401050739
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2401050739v1
240/1/177    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Petrella, J. R.
Right arrow Articles by Doraiswamy, P. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Petrella, J. R.
Right arrow Articles by Doraiswamy, P. M.
(Radiology 2006;240:177-186.)
© RSNA, 2006

Neuroradiology

Mild Cognitive Impairment: Evaluation with 4-T Functional MR Imaging1

Jeffrey R. Petrella, MD, Sriyesh Krishnan, BA, Melissa J. Slavin, PhD, Thanh-Thu T. Tran, BS, Lakshmi Murty, BS and P. Murali Doraiswamy, MD

1 From the Department of Radiology, Brain Imaging and Analysis Center (J.R.P., S.K., M.J.S., T.T.T.T., L.M.) and the Departments of Psychiatry and Medicine (Geriatrics) (P.M.D.), Duke University Medical Center, Box 3808, Durham, NC 27710-3808. Received May 1, 2005; revision requested June 30; revision received August 15; accepted September 13; final version accepted October 10. Supported by National Institute on Aging grant R01AG019728. Address correspondence to J.R.P. (e-mail: jeffrey.petrella{at}duke.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Purpose: To prospectively assess abnormalities in brain activation patterns during encoding and retrieval in subjects with mild cognitive impairment by using 4-T functional magnetic resonance (MR) imaging.

Materials and Methods: The institutional review board approved this HIPAA-compliant study; all subjects gave written informed consent. Twenty patients with mild cognitive impairment (12 men, eight women; mean age, 75.0 years ± 7.6 [standard deviation]) and 20 elderly control subjects (nine men, 11 women; mean age, 71.2 years ± 4.5) underwent functional MR imaging at 4 T during a novel-versus-familiar face-name encoding-retrieval task. The magnitude of blood oxygen level–dependent brain responses across the entire brain were compared within and between subjects with mild cognitive impairment and control subjects by using a voxelwise random-effects model. A one-sample t test was used for within-group analysis; an analysis-of-covariance model (with age as a covariate) was used for between-group analysis.

Results: Brain regions activated by the task (prefrontal, medial temporal, and parietal regions) during encoding were similar to those activated during retrieval, with larger areas activated during retrieval. Subjects with mild cognitive impairment showed decreased magnitude of activation in bilateral frontal cortex regions (during encoding and retrieval), the left hippocampus (during retrieval), and the left cerebellum (during encoding) compared with magnitude of activation in control subjects (P < .001). Patients with mild cognitive impairment showed increased activation in the posterior frontal lobes (during retrieval) (P < .001). Lower hippocampal activation during retrieval was the most significant correlate of clinical severity of memory loss in mild cognitive impairment (P < .001).

Conclusion: A difference exists in the response of brain regions underlying encoding and retrieval in mild cognitive impairment. Memory deficits in mild cognitive impairment may be linked to functional alterations in several specific brain regions both inside and outside the medial temporal lobe.

© RSNA, 2006


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Accumulating research suggests that Alzheimer disease may have an extended latent preclinical and prodromal period; hence, studies of people at risk have taken on great urgency (1–7). One condition that has taken on increasing importance is mild cognitive impairment because it may represent a transitional stage between normal aging and early Alzheimer disease (3–5). Recent pathologic data have shown strong support for the notion that mild cognitive impairment represents the earliest clinical manifestation of conditions responsible for age-related dementia (3). Patients with amnestic mild cognitive impairment have deficits in delayed recall but do not meet clinical criteria for dementia or Alzheimer disease (4). Longitudinal studies have shown that approximately 40% of patients with mild cognitive impairment progress to clinically diagnosed dementia during a 5-year period (4). Consequently, patients with mild cognitive impairment are a population of great interest to study vulnerability (prognostic) markers as well as to test interventions for delaying the onset of dementia (5,6).

The advent of functional magnetic resonance (MR) imaging has advanced cognitive neuroscience, particularly our ability to study human memory mechanisms in vivo (8,9). Results of functional MR imaging studies in healthy subjects have identified specific regions of the frontal cortex that are involved in both memory formation (encoding) and retrieval and have supported the notion that these regions may jointly participate with the medial temporal lobe (MTL) in human memory processing (10,11). Functional MR imaging also has been used to study age-related changes in memory in healthy volunteers (12,13), and age-related decreases in working memory retrieval have shown correlation with a reduction in dorsolateral prefrontal cortex activation (12). Although most functional MR imaging cognitive studies in the literature have been performed with 1.5-T units, there is some evidence that functional MR imaging at higher field strengths enables more sensitivity in the detection of cortical activation during cognitive and motor tasks. For example, in one study of 10 healthy subjects who performed cognitive tasks that required motor decisions (14), 3-T functional MR imaging, compared with 1.5-T MR imaging, enabled the detection of additional areas of cortical activation.

To our knowledge, there have been only three previously published controlled functional MR imaging studies of memory processing in patients with mild cognitive impairment (15–17). In the first study (performed at 3 T), investigators reported reduced activation in MTL regions during a visual memory task involving complex scenes in nine patients with mild cognitive impairment compared with control subjects (15). In that study, the regions of interest were restricted to the MTL. In the second study (16), performed at 1.5 T, investigators reported that enhancement of cholinergic activity with donepezil hydrochloride (Aricept; Eisai, Teaneck, NJ), a cholinesterase inhibitor, resulted in increased activation in frontal regions in nine patients with mild cognitive impairment but not in controls. In the third functional MR imaging study (17), performed at 1.5 T, investigators examined the adaptation to repeated stimuli in 12 patients with mild cognitive impairment. These investigators also restricted their analyses to the MTL. In addition, there have been two uncontrolled studies of patients with mild cognitive impairment and two controlled studies of elderly subjects with reduced or declining memory (all performed at 1.5 T) (18–21). The purpose of our study was to prospectively assess abnormalities in brain activation patterns during both encoding and retrieval in subjects with mild cognitive impairment by using 4-T functional MR imaging.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Subjects
The study was approved by our institutional review board and was conducted in compliance with the Health Insurance Portability and Accountability Act. Subjects who met entry criteria were consecutively recruited from the local community by means of advertisements and referrals. Twenty subjects with mild cognitive impairment of the amnestic subtype (12 men and eight women; mean age, 75.0 years ± 7.6 [standard deviation]) and 20 elderly control subjects with normal memory (nine men and 11 women; mean age, 71.2 years ± 4.5) were studied (Table 1). All subjects gave written informed consent before any testing or neuropsychologic evaluation. Subjects with mild cognitive impairment who were enrolled in this study were free of any conditions that would preclude informed decision making, as outlined in the inclusion and exclusion criteria, and, therefore, were capable of providing informed consent.


View this table:
[in this window]
[in a new window]

  		Table 1. Demographic and Clinical Characteristics in 20 Subjects with Mild Cognitive Impairment and 20 Control Subjects

 
Entry Criteria
All subjects were fluent in English, had at least 8 years of formal education, and were willing to undergo functional MR imaging. Subjects underwent a clinical interview and a focused neurologic and physical examination with a physician (P.M.D., with 15 years of experience in the evaluation of patients with memory disorders), basic laboratory testing (complete blood cell count, vitamin B12 determination, thyroid function tests, and blood chemistry analysis), and demographic inventory (age, sex, educational level). Subjects then underwent a detailed screening neuropsychologic battery by two individuals (M.J.S. and T.T.T.T., with 5 and 2 years of experience in neuropsychologic testing, respectively). The screening neuropsychologic battery consisted of the California Verbal Learning Test–II (22); logical memory and visual reproduction tests from the Wechsler Memory Scale–III (23); the Mini-Mental State Examination (24); assignment of a Clinical Dementia Rating scale score (25), which is an interview with the patient and an informant (usually a spouse or other family member) conducted by a certified rater (M.J.S. or T.T.T.T.); assignment of a Beck Depression Inventory–II score (26); and assignment of a Hachinski vascular dementia score (27).

Inclusion criteria for subjects with mild cognitive impairment.—Subjects with mild cognitive impairment were included in the study if they (a) had had a recent history of symptomatic worsening in memory (supported by the informant); (b) had objective memory impairment (at least 1 standard deviation below normal), as evidenced by their performance on the Calfifornia Verbal Learning Test–II and on the logical memory and visual reproduction subsets of the Wechsler Memory Scale–III; (c) had normal or near-normal performance on a global cognitive test as defined by an Mini-Mental State Examination score of more than 24; (d) had a global rating on the clinical dementia rating scale of 0.5 (questionable dementia), with a rating of at least 0.5 for the memory score; (e) did not meet National Institute of Neurological Disorders and Stroke–Alzheimer's Disease and Related Disorders Association (28) or Diagnostic and Statistical Manual of Mental Disorders, fourth edition (29), criteria for dementia; (f) had normal or near-normal independent function (as reported by an informant); and (g) had an absence of other factors that may better explain memory loss (eg, current major depression) (5,7).

Inclusion criteria for control subjects.—Control subjects were included in the study if they (a) had a Mini-Mental State Examination score of more than 28; (b) did not meet National Institute of Neurological Disorders and Stroke–Alzheimer's Disease and Related Disorders Association (28) or Diagnostic and Statistical Manual of Mental Disorders, fourth edition (29), criteria for dementia; (c) had normal or near-normal independent function (self-reported); (d) had a normal memory; and (e) had a global score of 0 according to the Clinical Dementia Rating scale.

Exclusion Criteria
Subjects were excluded if (a) they had uncontrolled depression or other psychiatric illness; (b) they were taking psychoactive medications known to substantially affect memory; (c) they had standard contraindications to MR imaging, including size incompatibility with the imaging unit (the 4-T unit has a bore size of 55 cm), metal implants, or cardiac pacemakers; (d) there were technical difficulties that prevented the completion of successful anatomic imaging and/or at least two of three functional MR imaging sessions; (d) there was excessive motion during the functional MR examination in excess of 5 mm as determined by means of center-of-mass plots; and (e) behavioral responses could not be adequately monitored while the subject was in the MR unit, as evidenced by a nonresponse rate of more than 50%.

Functional MR Imaging Stimuli and Encoding and Retrieval Tasks
In our study, we used a variation of the face-name associative memory encoding task developed by Sperling et al (30) for use in young adults that was later applied to elderly subjects and patients with Alzheimer disease (31). In addition to the learning (encoding) task, our experimental design included a recall (retrieval) task for which a response was required while the subject was within the MR unit. Subjects were required to encode and later retrieve face-name associations of two conditions—novel face-name pairs and familiar face-name pairs—that were presented within a blocked experimental design. Sixty novel and two familiar face-name pairs, drawn from the AR Face Database (32), were presented within a block design during a period of 6 minutes 50 seconds per session for a total of three sessions (Fig 1). Behavioral responses were monitored by means of a fiberoptic button box within the MR unit. For control purposes, subjects were also told to provide a button response during the encoding period to indicate whether they thought the name matched the face.


Figure 1
View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1a: Example of functional MR imaging behavioral task. The experimental paradigm was divided into three sessions, with each session including two sets of four blocks separated by a brief fixation cross. Both sets of four blocks included encoding of 10 novel face-name pairs, encoding of two familiar face-name pairs performed five times, retrieval of 10 novel face-name pairs, and retrieval of two familiar face-name pairs performed five times. The activation during the familiar condition for encoding or retrieval provided the control for the novel condition of that same task. (a) Schematic of one session of the behavioral paradigm. Each block represents 10 trials. FE = familiar encode, FR = familiar retrieval, NE = novel encode, NR = novel retrieval, 10s = 10 seconds, 50s = 50 seconds. There is no intertrial interval. (b) Example of a picture used for an encoding trial. (c) Example of a picture used for a retrieval trial. (b And c reprinted, with permission, from reference 32.)

 

Figure 1
View larger version (62K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1b: Example of functional MR imaging behavioral task. The experimental paradigm was divided into three sessions, with each session including two sets of four blocks separated by a brief fixation cross. Both sets of four blocks included encoding of 10 novel face-name pairs, encoding of two familiar face-name pairs performed five times, retrieval of 10 novel face-name pairs, and retrieval of two familiar face-name pairs performed five times. The activation during the familiar condition for encoding or retrieval provided the control for the novel condition of that same task. (a) Schematic of one session of the behavioral paradigm. Each block represents 10 trials. FE = familiar encode, FR = familiar retrieval, NE = novel encode, NR = novel retrieval, 10s = 10 seconds, 50s = 50 seconds. There is no intertrial interval. (b) Example of a picture used for an encoding trial. (c) Example of a picture used for a retrieval trial. (b And c reprinted, with permission, from reference 32.)

 

Figure 1
View larger version (44K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1c: Example of functional MR imaging behavioral task. The experimental paradigm was divided into three sessions, with each session including two sets of four blocks separated by a brief fixation cross. Both sets of four blocks included encoding of 10 novel face-name pairs, encoding of two familiar face-name pairs performed five times, retrieval of 10 novel face-name pairs, and retrieval of two familiar face-name pairs performed five times. The activation during the familiar condition for encoding or retrieval provided the control for the novel condition of that same task. (a) Schematic of one session of the behavioral paradigm. Each block represents 10 trials. FE = familiar encode, FR = familiar retrieval, NE = novel encode, NR = novel retrieval, 10s = 10 seconds, 50s = 50 seconds. There is no intertrial interval. (b) Example of a picture used for an encoding trial. (c) Example of a picture used for a retrieval trial. (b And c reprinted, with permission, from reference 32.)

 
Anatomic and Functional Whole-Brain Imaging
Imaging was performed with a 4-T unit (GE Medical Systems, Milwaukee, Wis). Transverse T2-weighted spin-echo images (3000/80 [repetition time msec/echo time msec], 256 x 256 matrix, 3.75-mm-thick sections with no intersection gap, 240-mm field of view) were obtained through the brain for diagnostic purposes to assess intracranial abnormalities. Anatomic images, consisting of 44 contiguous 3.75-mm-thick coronal sections, were acquired for normalization of the functional images by using inversion-recovery prepared three-dimensional spoiled gradient-recalled acquisition in the steady state high-spatial-resolution T1-weighted MR imaging (12.2/5.4/500 [repetition time msec/echo time msec/inversion time msec], 20° flip angle, 256 x 256 matrix, 240-mm field of view). Functional images, which consisted of a time series of 164 T2*-weighted isotropic image volumes, were obtained with inverse spiral echo-planar imaging (2500/31, 60° flip angle, 64 x 64 matrix, 240-mm field of view) from the same 44 continuous coronal section locations used for the anatomic series during each of the three functional sessions per subject.

Functional MR Image Analysis
All anatomic and functional MR images were initially screened for quality control purposes by using center-of-mass plots to detect excessive motion or section acquisition errors. Anatomic images were screened for intracranial abnormalities by a board-certified neuroradiologist (J.R.P.) with 12 years of experience in the interpretation of brain MR examinations. Subsequent image processing was performed by using statistical parametric mapping software (Statistical Parametric Mapping 2; Wellcome Department of Imaging Neuroscience, the Institute of Neurology, University College London, National Hospital for Neurology & Neurosurgery, London, England) (33). Preprocessing of data for each subject consisted of section timing and motion correction, normalization to the Montreal Neurologic Institute template, and spatial smoothing with an 8-mm Gaussian kernel. Two contrast maps of the novel versus familiar condition—one for encoding and one for retrieval—were created for each subject by using the general linear model approach in the software. These contrast (or activation magnitude) maps represent voxelwise differences in signal intensity magnitude between the novel and familiar conditions across the entire brain and were used for all further functional MR imaging analyses.

Statistical Analysis
The two groups were compared for differences in demographic and clinical variables by using software (SPSS, version 12.2, 2004; SPSS, Chicago, Ill) and two-tailed unpaired Student t tests (for age; education; Mini-Mental State Examination, California Verbal Learning Test–II, Wechsler Memory Scale–III, Beck Depression Inventory–II, and Hachinski vascular dementia scores; behavioral performance; and number of nonrecorded responses) or {chi}2 analysis (for sex and handedness). A P value of less than .05 was considered to indicate a statistically significant difference.

Activation magnitude was assessed in a voxelwise manner across the entire brain both within and between subjects with mild cognitive impairment and control subjects. Within-group and between-group statistical activation maps for control subjects and those with mild cognitive impairment were created from the contrast map of the individual subject by using a random-effects model in the statistical mapping software. A one-sample t test was used for the within-group analysis, and an analysis of covariance model (with age as a covariate) was used for the between-group analysis because patients with mild cognitive impairment were marginally older than control subjects. To determine whether possible between-group differences in activation magnitude were more than merely a reflection of task performance, a between-group analysis also was performed by using both age and performance as confounders in the analysis of covariance model. We also performed an exploratory analysis for correlation between activation magnitude and memory loss severity in subjects with mild cognitive impairment (by using delayed-recall scores from the California Verbal Learning Test–II). Cluster significance levels for all maps were set at a P value of .001 or less (uncorrected) with a 10-voxel cluster threshold, values comparable to or more rigorous than those used in prior studies in similar populations (16,19) in which the focus was on the whole brain.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Clinical Measures and Behavioral Performance on Task
The groups did not significantly differ in terms of sex, years of education, or handedness (Table 1). Because of the logistics of recruiting consecutive eligible subjects, those with mild cognitive impairment tended to be marginally older (P = .065) than control subjects, and, hence, we adjusted for age effects in all between-group functional MR imaging analyses. As expected, subjects with mild cognitive impairment had significantly lower scores on the Mini-Mental State Examination, delayed-recall part of the California Verbal Learning Test–II, and Wechsler Memory Scale–III (P < .001). There was no statistically significant difference in Beck Depression Inventory–II or Hachinski vascular dementia scores.

With regard to behavioral performance during the functional MR imaging task, the percentage of correct responses for novel face-name pairs was lower for subjects with mild cognitive impairment than for control subjects (P < .01). The number of nonresponses between the two groups did not significantly differ, with a recorded response rate of more than 92% (55.3 of 60 responses) in either group.

Within-Group Analyses in Control Subjects and Subjects with Mild Cognitive Impairment
The task resulted in significant (from zero) activation magnitude in specific brain regions, especially in the prefrontal cortex (Figs 2, 3), parietal lobe, and MTL (P ≤ .001).


Figure 2
View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a: Task-related areas of activation in all control subjects. Activation maps obtained with transverse functional MR imaging during (a) encoding and (b) retrieval created by using Statistical Parametric Mapping 2 software with an analysis of variance (P = .001, uncorrected threshold level for statistical significance; minimal cluster size, 10 voxels). Anatomic images are from the Statistical Parametric Mapping 2 software T1-weighted single-subject canonical brain. Threshold activation is displayed as a color overlay of t values.

 

Figure 2
View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b: Task-related areas of activation in all control subjects. Activation maps obtained with transverse functional MR imaging during (a) encoding and (b) retrieval created by using Statistical Parametric Mapping 2 software with an analysis of variance (P = .001, uncorrected threshold level for statistical significance; minimal cluster size, 10 voxels). Anatomic images are from the Statistical Parametric Mapping 2 software T1-weighted single-subject canonical brain. Threshold activation is displayed as a color overlay of t values.

 

Figure 3
View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a: Task-related areas of activation in subjects with mild cognitive impairment. Activation maps obtained with transverse functional MR imaging during (a) encoding and (b) retrieval created by using Statistical Parametric Mapping 2 software with analysis of variance (P = .001, uncorrected threshold level for statistical significance; minimal cluster size, 10 voxels). Anatomic images are from the Statistical Parametric Mapping 2 software T1-weighted single-subject canonical brain. Threshold activation is displayed as a color overlay of t values.

 

Figure 3
View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b: Task-related areas of activation in subjects with mild cognitive impairment. Activation maps obtained with transverse functional MR imaging during (a) encoding and (b) retrieval created by using Statistical Parametric Mapping 2 software with analysis of variance (P = .001, uncorrected threshold level for statistical significance; minimal cluster size, 10 voxels). Anatomic images are from the Statistical Parametric Mapping 2 software T1-weighted single-subject canonical brain. Threshold activation is displayed as a color overlay of t values.

 
Comparison of Activation Magnitude in Control Subjects and Subjects with Mild Cognitive Impairment
During encoding, the activation magnitude in the bilateral frontal lobes and left cerebellum was decreased in subjects with mild cognitive impairment, compared with that in control subjects (P ≤ .001) (Table 2, Fig 4). The largest clusters that showed decreased activation magnitude in subjects with mild cognitive impairment, compared with that in control subjects, were in the frontal gyri. There were no areas of significantly increased activation magnitude in subjects with mild cognitive impairment relative to control subjects during encoding. In subjects with mild cognitive impairment versus control subjects, there was also a decreased activation magnitude in the bilateral frontal lobes, most notably on the right side, and in the left hippocampus during retrieval (P ≤ .001). In subjects with mild cognitive impairment versus control subjects, there was increased activation magnitude in the posterior frontal lobes, notably in the right precentral gyrus (Table 3).


View this table:
[in this window]
[in a new window]

  		Table 2. Differences in Activation Magnitude during Memory Encoding between Subjects with Mild Cognitive Impairment and Control Subjects

 

Figure 4
View larger version (77K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4: Images show comparison of activation (regions with reduced activation) in subjects with mild cognitive impairment and control subjects. Brain surface renderings illustrate clusters of significantly reduced activation magnitude in subjects with mild cognitive impairment compared with control subjects during encoding (top row) and retrieval (bottom row). See text for details. Activations shown meet the P = .001 (uncorrected) threshold level for statistical significance. The minimal cluster size was 10 voxels. Surface display is from the Statistical Parametric Mapping 2 software T1-weighted single-subject canonical brain. Threshold activation is displayed as a volume rendering in red, where greater intensity of color denotes greater proximity to the brain surface. Surface views of the brain are as follows (from left to right): right lateral surface, left lateral surface, superior surface, and inferior surface.

 

View this table:
[in this window]
[in a new window]

  		Table 3. Differences in Activation Magnitude during Memory Retrieval between Subjects with Mild Cognitive Impairment and Control Subjects

 
Comparison of activation magnitude in control subjects and those with mild cognitive impairment after adjusting for accuracy of behavioral performance resulted in similar findings (not shown). During encoding, subjects with mild cognitive impairment versus control subjects showed a reduced activation magnitude in regions of the left prefrontal cortex, specifically in the left superior and inferior frontal gyri (P ≤ .001). During retrieval, subjects with mild cognitive impairment showed decreased activation magnitude in the right prefrontal cortex, as well as in the left medial frontal gyrus (P ≤ .001). Persistent significantly increased activation magnitude was noted during retrieval in small clusters in the right precentral gyrus and left insula (P ≤ .001).

Correlation between Functional MR Imaging Activation and Severity of Memory Loss in Mild Cognitive Impairment
Although clinical memory performance (measured with the delayed-recall score on the California Verbal Learning Test–II) correlated with activation magnitude in several clusters, the largest cluster that showed a significant correlation was in the left hippocampus (Talairach x, y, z coordinates: –26, –35, 5) during retrieval (r = 0.75, P < .001) (Fig 5).


Figure 5
View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5: Graph shows positive correlation between activation magnitude in the left hippocampal cluster (Talairach x, y, and z coordinates = –26, –35, and 5, respectively) during retrieval and the delayed recall score from the California Verbal Learning Test–II in patients with mild cognitive impairment. Regression equation is y = 0.289x – 1.579, and r = 0.75 (P < .001).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 References
 
To our knowledge, ours is the first study with 4-T functional MR imaging in which cerebral activation changes were examined across the whole brain during both encoding and retrieval phases of memory in both healthy elderly subjects and a population with well-defined mild cognitive impairment. Two main findings emerged from this study—one pertaining to brain regions activated by associative encoding and retrieval in healthy elderly subjects and the other pertaining to differences in activation between subjects who are experiencing normal aging and those who have mild cognitive impairment.

The results of our study confirmed that both prefrontal and medial temporal regions are activated by associative memory tasks in healthy elderly subjects. This is consistent with the view that memory function is dependent on joint participation of frontal and MTL structures (10,11). In addition, we found significant activation in the parietal lobe and cerebellum, and this finding confirms results of a previous study with 1.5-T MR imaging that involved healthy elderly subjects (34). Taken together, these data add to our knowledge of memory networks in the normal brain during aging.

Cerebral activation during the retrieval phase of a memory task has been studied in healthy volunteers but has not been as well studied in the population with mild cognitive impairment (in whom most previous studies have focused on encoding). Functional MR imaging studies of retrieval offer the advantage of directly monitoring subject responses during the functional MR imaging examination itself rather than afterward. We demonstrated that the pattern of functional MR imaging activation during retrieval is remarkably similar to that of encoding but with slightly more expanded areas of activation. Researchers in previously performed studies in both young subjects and older subjects with normal memory have shown that an explicit retrieval task shares much of the same functional anatomy as an encoding task but is associated with the recruitment of additional brain areas, including the anterior prefrontal cortex (10,35). With functional MR imaging at 4 T, we extended these findings to healthy elderly subjects and to subjects with mild cognitive impairment. Furthermore, we showed that such recruitment of additional brain areas during memory retrieval (compared with encoding) cannot be attributed solely to response preparation because we also controlled for response preparation during encoding by requiring the subject to push a button.

The results of our study also helped pinpoint several differences in functional MR imaging activation patterns during encoding and retrieval between subjects with mild cognitive impairment and control subjects. In general, subjects with mild cognitive impairment showed reduced activation magnitude in specific regions. The greatest reductions during encoding were seen in the frontal lobes (dorsolateral prefrontal cortex), and the greatest reductions during retrieval were seen in both the frontal lobes and the MTL (hippocampus). Initial views of amnestic mild cognitive impairment held that it was an "isolated" MTL disease. Taken together, our data from 4-T functional MR imaging and results of prior studies (at lower field strengths) help confirm that one or more MTL regions (hippocampus, fusiform gyrus, entorhinal cortex) are functionally altered in subjects with mild cognitive impairment. In our study, the largest activated brain region that correlated with delayed-recall memory performance in subjects with mild cognitive impairment was a cluster in the hippocampus. Our findings, however, also extend these views by showing that the functional anatomy of subjects with mild cognitive impairment may also involve frontal lobe dysfunction.

We speculate that areas of reduced functional MR imaging activation in subjects with mild cognitive impairment represent regions in which axonal or synaptic function has been disrupted by the earliest pathologic changes of Alzheimer disease and that regions of increased activation may represent compensatory changes. This finding is supported by recent postmortem data (3) that more than two-thirds of patients in whom mild cognitive impairment was clinically diagnosed show neocortical and/or extra-MTL involvement with neurofibrillary tangles (Braak stages III–VI) (3). Findings in a prior functional MR imaging study at 1.5-T in nine subjects with mild cognitive impairment also suggest reduced frontal lobe activation, and findings in studies in at-risk subjects suggested presumed less efficient, but compensatory, activation increases (16,35,36). Collectively, these data suggest that the notion of mild cognitive impairment as a "pure" entorhinal cortical disease and/or hippocampal disease is overly simplistic and that brain dysfunction and adaptation in subjects with mild cognitive impairment may be more widespread than previously understood. Most prior studies of functional MR imaging performed in subjects with mild cognitive impairment were cross-sectional analyses of small samples of subjects who underwent imaging at 1.5 T. Our sample size of 20 control subjects and 20 subjects with mild cognitive impairment is larger, and our study is the only one to examine both encoding and retrieval phases of memory at 4-T functional MR imaging. Clearly, however, even larger studies of patients with mild cognitive impairment that focus on the early diagnostic value of functional alterations both inside and outside the MTL are warranted, and high-field-strength (greater than 1.5 T) functional MR imaging is one tool for such studies.

Despite the numerous advantages of the use of a higher field strength in terms of sensitivity to cortical activation during cognitive and motor tasks, greater sensitivity to susceptibility artifacts that are near air-bone interfaces could present a serious disadvantage, especially in areas such as the inferior frontal lobe and MTL. In the left hippocampus of subjects with mild cognitive impairment, despite the positive correlation between activation magnitude and behavioral performance, the activation magnitude itself is highly variable (Fig 5). One source for the variable signal response in this region, with no response or a negative response in a considerable number of subjects, may be poor signal-to-noise ratio related to the degree of petrous bone aeration and susceptibility-induced signal loss in the overlying MTL. The signal-to-noise ratio in MTL structures such as the hippocampus and amygdala is inherently low during cognitive tasks such as memory and may be even worse at high field strengths (37). Thus, in a given subject, there may be a number of areas in which activation magnitude is modeled as a negative number. Of course, it is also possible that a given area may demonstrate a true negative activation magnitude due to a "steal effect" from the reallocation of cognitive processing resources, where there is decreased blood flow and oxygenation during the cognitive challenge compared with the resting state (38). Such negative activations, or deactivations, may be responsible for between-group differences in areas in which there are no detectable activations displayed on the within-group statistical maps.

Our study had limitations. In this study, a number of potential confounders may have contributed to a bias in between-group differences; these confounders included age, which has been shown to alter the shape of the functional MR imaging hemodynamic response. Because there was a trend for our subjects with mild cognitive impairment to be marginally older, on average 3.8 years older, than the control subjects, we used age as a covariate in the between-group whole-brain analysis. Although educational levels were similar between groups, there is no guarantee that lifestyle factors thought to slow progression to Alzheimer disease, such as exercise and mental stimulation (reading, crossword puzzles, etc), were equally distributed. Another possible confounder includes behavioral performance during the functional MR imaging task. Results of prior functional MR imaging studies have been inconsistent with regard to controlling this effect, and the effects of mild cognitive impairment and performance could potentially be collinear. We dealt with this problem by reporting findings both before and after adding task performance (measured by means of the percentage of novel face-name pairs that were correctly identified) as a covariate. To minimize the potential effects of atrophy on the normalization process, we applied an interpolation technique that did not change the signal intensity of the warped functional images regardless of whether a particular area of brain was contracted or expanded to match the template. We also chose to conduct whole-brain analyses because we believed the results of such analyses would convey a broader picture of the functional neurologic anatomy of subjects with mild cognitive impairment, both inside and outside the MTL. In our study, there were some apparent laterality effects that we cannot fully explain. For example, we found significant group differences in the left but not right hippocampus. Our task had a verbal component, and one possibility is that the left side is preferentially activated during such tasks (39).

In summary, to our knowledge, our study of functional MR imaging at 4-T is the first to examine cerebral activation changes across the entire brain during both encoding and retrieval phases of memory in a clinically defined population with mild cognitive impairment and in elderly control subjects. Findings in our study indicate that memory deficits in mild cognitive impairment may be linked to functional alterations in several specific brain regions both inside and outside the MTL. Such changes may represent early neuronal dysfunction due to disruption of the connections between the MTL and multimodal association areas, as well as functional adaptation. Results of functional MR imaging studies in which investigators directly examine the connectivity between temporal and extratemporal multimodality association areas, as well as results of longitudinal studies involving larger samples, may yield further insight into the early pathophysiologic findings in Alzheimer disease and its prodromal states.


   ADVANCES IN KNOWLEDGE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
References
 


   FOOTNOTES
 

Abbreviations: MTL = medial temporal lobe

Authors stated no financial relationship to disclose.

Author contributions: Guarantors of integrity of entire study, J.R.P., P.M.D.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, J.R.P., S.K., M.J.S., L.M., P.M.D.; clinical studies, J.R.P., M.J.S., T.T.T.T., L.M., P.M.D.; statistical analysis, J.R.P., S.K., M.J.S., L.M.; and manuscript editing, J.R.P., M.J.S., P.M.D.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 

  1. Coleman P, Federoff H, Kurlan R. A focus on the synapse for neuroprotection in Alzheimer disease and other dementias. Neurology 2004;63:1155–1162.[Abstract/Free Full Text]
  2. Petrella JR, Coleman RE, Doraiswamy PM. Neuroimaging and early diagnosis of Alzheimer disease: a look to the future. Radiology 2003;226:315–336.[Abstract/Free Full Text]
  3. Bennett DA, Schneider JA, Bienias JL, Evans DA, Wilson RS. Mild cognitive impairment is related to Alzheimer disease pathology and cerebral infarctions. Neurology 2005;64:834–841.[Abstract/Free Full Text]
  4. Petersen RC. Mild cognitive impairment: transition between aging and Alzheimer's disease. Neurologia 2000;15:93–101.[CrossRef][Medline]
  5. Petersen RC. Mild cognitive impairment clinical trials. Nat Rev Drug Discov 2003;2:646–653.[CrossRef][Medline]
  6. Doraiswamy PM. Interventions for mild cognitive impairment and Alzheimer disease: new strategies, new hope. Am J Geriatr Psychiatry 2003;11:120–122.[Free Full Text]
  7. Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild cognitive impairment: clinical characterization and outcome. Arch Neurol 1999;56:303–308.[Abstract/Free Full Text]
  8. Ungerleider LG. Functional brain imaging studies of cortical mechanisms for memory. Science 1995;270:769–775.[Abstract/Free Full Text]
  9. Buckner RL, Koutstaal W. Functional neuroimaging studies of encoding, priming, and explicit memory retrieval. Proc Natl Acad Sci U S A 1998;95:891–898.[Abstract/Free Full Text]
  10. Buckner RL, Kelley WM, Petersen SE. Frontal cortex contributes to human memory formation. Nat Neurosci 1999;2:311–314.[CrossRef][Medline]
  11. Wagner AD, Schacter DL, Rotte M, et al. Building memories: remembering and forgetting of verbal experiences as predicted by brain activity. Science 1998;281:1188–1191.[Abstract/Free Full Text]
  12. Rypma B, D'Esposito M. Isolating the neural mechanisms of age-related changes in human working memory. Nat Neurosci 2000;3:509–515.[CrossRef][Medline]
  13. Gazzaley A, D'Esposito M. The contribution of functional brain imaging to our understanding of cognitive aging. Sci Aging Knowledge Environ 2003;2003(4):PE2.[Medline]
  14. Hoenig K, Kuhl CK, Scheef L. Functional 3.0-T MR assessment of higher cognitive function: are there advantages over 1.5-T imaging? Radiology 2005;234:860–868.[Abstract/Free Full Text]
  15. Machulda M, Ward H, Borowski B, et al. Comparison of memory fMRI response among normal, MCI, and Alzheimer's patients. Neurology 2003;61:500–506.[Abstract/Free Full Text]
  16. Saykin AJ, Wishart HA, Rabin LA, et al. Cholinergic enhancement of frontal lobe activity in mild cognitive impairment. Brain 2004;127:1574–1583.[Abstract/Free Full Text]
  17. Johnson SC, Baxter LC, Susskind-Wilder L, Connor DJ, Sabbagh MN, Caselli RJ. Hippocampal adaptation to face repetition in healthy elderly and mild cognitive impairment. Neuropsychologia 2004;42:980–989.[CrossRef][Medline]
  18. Dickerson BC, Salat DH, Bates JF, et al. Medial temporal lobe function and structure in mild cognitive impairment. Ann Neurol 2004;56:27–35.[CrossRef][Medline]
  19. Goekoop R, Rombouts SA, Jonker C, et al. Challenging the cholinergic system in mild cognitive impairment: a pharmacological fMRI study. NeuroImage 2004;23:1450–1459.[CrossRef][Medline]
  20. Daselaar SM, Veltman DJ, Rombouts SA, Raaijmakers JG, Jonker C. Neuroanatomical correlates of episodic encoding and retrieval in young and elderly subjects. Brain 2003;126:43–56.[Abstract/Free Full Text]
  21. Small SA, Perera GM, DeLaPaz R, Mayeux R, Stern Y, Hem E. Differential regional dysfunction of the hippocampal formation among elderly with memory decline and Alzheimer's disease. Ann Neurol 1999;45:466–472.[CrossRef][Medline]
  22. Delis DC, Kramer JH, Kaplan E, Ober BA. The California Verbal Learning Test. New York, NY: Psychological Corporation, 1987.
  23. Wechsler D. Wechsler memory scale. 3rd ed. New York, NY: Psychological Corporation, 1997.
  24. Folstein MF, Folstein SE, McHugh PR. "Mini mental state": a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189–198.[CrossRef][Medline]
  25. Morris JC. The clinical dementia rating (CDR): current version and scoring rules. Neurology 1993;43:2412–2414.[Free Full Text]
  26. Beck AT, Steer RA, Brown GK. Manual for the Beck Depression Inventory-II. San Antonio, Tex: Psychological Corporation, 1996.
  27. Hachinski V, Lassen N, Marshall J. Multi-infarct dementia: a cause of mental deterioration in the elderly. Lancet 1974;2:207–210.[CrossRef][Medline]
  28. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer's disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer's Disease. Neurology 1984;34:939–944.[Abstract/Free Full Text]
  29. American Psychiatric Association. Diagnostic and statistical manual of mental disorders (DSM-IV). 4th ed. Washington, DC: American Psychiatric Association, 1994.
  30. Sperling RA, Bates JF, Cocchiarella AJ, Schacter DL, Rosen BR, Albert MS. Encoding novel face-name associations: a functional MRI study. Hum Brain Mapp 2001;14:129–139.[CrossRef][Medline]
  31. Sperling RA, Bates JF, Chua EF, et al. FMRI studies of associative encoding in young and elderly controls and mild Alzheimer's disease. J Neurol Neurosurg Psychiatry 2003;74:44–50.[Abstract/Free Full Text]
  32. Martinez AM, Benavente R. The AR face database. Computer Vision Center technical report no. 24. West Lafayette, Ind: Purdue University, 1998. http://rv11.ecn.purdue.edu/~aleix/ar.html. Accessed January 15, 2002.
  33. Frackowiak RSJ, Friston KJ, Frith CD, Dolan RJ, Mazziotta JC. Human brain function. San Diego, Calif: Academic Press, 1997.
  34. Rombouts SA, Barkhof F, Veltman DJ, et al. Functional MR imaging in Alzheimer's disease during memory encoding. AJNR Am J Neuroradiol 2000;21:1869–1875.[Abstract/Free Full Text]
  35. Bookheimer SY, Strojwas MH, Cohen MS, et al. Patterns of brain activation in people at risk for Alzheimer's disease. N Engl J Med 2000;343:450–456.[Abstract/Free Full Text]
  36. Bondi MW, Houston WS, Eyler LT, Brown GG. fMRI evidence of compensatory mechanisms in older adults at genetic risk for Alzheimer disease. Neurology 2005;64:501–508.[Abstract/Free Full Text]
  37. Greicius MD, Krasnow B, Boyett-Anderson JM, et al. Regional analysis of hippocampal activation during memory encoding and retrieval: fMRI study. Hippocampus 2003;13:164–174.[CrossRef][Medline]
  38. Lustig C, Snyder AZ, Bhakta M, et al. Functional deactivations: change with age and dementia of the Alzheimer type. Proc Natl Acad Sci U S A 2003;100:14504–14509.[Abstract/Free Full Text]
  39. Kelley WM, Miezin FM, McDermott KB, et al. Hemispheric specialization in human dorsal frontal cortex and medial temporal lobe for verbal and nonverbal memory encoding. Neuron 1998;20:927–936.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 AJGPHome page
A. L. Jefferson, L. K. Byerly, S. Vanderhill, S. Lambe, S. Wong, A. Ozonoff, and J. H. Karlawish
Characterization of Activities of Daily Living in Individuals With Mild Cognitive Impairment
Am J Geriatr Psychiatry, May 1, 2008; 16(5): 375 - 383.
[Abstract] [Full Text] [PDF]

Home page
 NeurologyHome page
G. Xu, P. G. Antuono, J. Jones, Y. Xu, G. Wu, D. Ward, and S. -J. Li
Perfusion fMRI detects deficits in regional CBF during memory-encoding tasks in MCI subjects
Neurology, October 23, 2007; 69(17): 1650 - 1656.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
J. R. Petrella, L. Wang, S. Krishnan, M. J. Slavin, S. E. Prince, T.-T. T. Tran, and P. M. Doraiswamy
Cortical Deactivation in Mild Cognitive Impairment: High-Field-Strength Functional MR Imaging
Radiology, October 1, 2007; 245(1): 224 - 235.
[Abstract] [Full Text] [PDF]

Home page
 NeurologyHome page
G. E. Smith, V. S. Pankratz, S. Negash, M. M. Machulda, R. C. Petersen, B. F. Boeve, D. S. Knopman, J. A. Lucas, T. J. Ferman, N. Graff-Radford, et al.
A plateau in pre-Alzheimer memory decline: Evidence for compensatory mechanisms?
Neurology, July 10, 2007; 69(2): 133 - 139.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2401050739v1
240/1/177    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Petrella, J. R.
Right arrow Articles by Doraiswamy, P. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Petrella, J. R.
Right arrow Articles by Doraiswamy, P. M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE