Matthias H. Tabert

Hippocampal and entorhinal atrophy in mild cognitive impairment: prediction of Alzheimer disease.

Objective: To evaluate the utility of MRI hippocampal and entorhinal cortex atrophy in predicting conversion from mild cognitive impairment (MCI) to Alzheimer disease (AD). Methods: Baseline brain MRI was performed in 139 patients with MCI, broadly defined, and 63 healthy controls followed for an average of 5 years (range 1 to 9 years). Results: Hippocampal and entorhinal cortex volumes were each largest in controls, intermediate in MCI nonconverters, and smallest in MCI converters to AD (37 of 139 patients converted to AD). In separate Cox proportional hazards models, covarying for intracranial volume, smaller hippocampal volume (risk ratio [RR] 3.62, 95% CI 1.93 to 6.80, p < 0.0001), and entorhinal cortex volume (RR 2.43, 95% CI 1.56 to 3.79, p < 0.0001) each predicted time to conversion to AD. Similar results were obtained for hippocampal and entorhinal cortex volume in patients with MCI with Mini-Mental State Examination (MMSE) scores ≥ 27 out of 30 (21% converted to AD) and in the subset of patients with amnestic MCI (35% converted to AD). In the total patient sample, when both hippocampal and entorhinal volume were entered into an age-stratified Cox model with sex, MMSE, education, and intracranial volume, smaller hippocampal volume (RR 2.21, 95% CI 1.14 to 4.29, p < 0.02) and entorhinal cortex volume (RR 2.48, 95% CI 1.54 to 3.97, p < 0.0002) predicted time to conversion to AD. Similar results were obtained in a Cox model that also included Selective Reminding Test (SRT) delayed recall and Wechsler Adult Intelligence Scale-Revised (WAIS-R) Digit Symbol as predictors. Based on logistic regression models in the 3-year follow-up sample, for a fixed specificity of 80%, the sensitivities for MCI conversion to AD were as follows: age 43.3%, MMSE 43.3%, age + MMSE 63.7%, age + MMSE + SRT delayed recall + WAIS-R Digit Symbol 80.6% (79.6% correctly classified), hippocampus + entorhinal cortex 66.7%, age + MMSE + hippocampus + entorhinal cortex 76.7% (85% correctly classified), age + MMSE + SRT delayed recall + WAIS-R Digit Symbol + hippocampus + entorhinal cortex 83.3% (86.8% correctly classified). Conclusions: Smaller hippocampal and entorhinal cortex volumes each contribute to the prediction of conversion to Alzheimer disease. Age and cognitive variables also contribute to prediction, and the added value of hippocampal and entorhinal cortex volumes is small. Nonetheless, combining these MRI volumes with age and cognitive measures leads to high levels of predictive accuracy that may have potential clinical application.

Functional deficits in patients with mild cognitive impairment Prediction of AD

ObjectiveTo evaluate the predictive utility of self-reported and informant-reported functional deficits in patients with mild cognitive impairment (MCI) for the follow-up diagnosis of probable AD. MethodsThe Pfeffer Functional Activities Questionnaire (FAQ) and Lawton Instrumental Activities of Daily Living (IADL) Scale were administered at baseline. Patients were followed at 6-month intervals, and matched normal control subjects (NC) were followed annually. ResultsSelf-reported deficits were higher for patients with MCI than for NC. At baseline, self- and informant-reported functional deficits were significantly greater for patients who converted to AD on follow-up evaluation than for patients who did not convert, even after controlling for age, education, and modified Mini-Mental State Examination scores. While converters showed significantly more informant- than self-reported deficits at baseline, nonconverters showed the reverse pattern. Survival analyses further revealed that informant-reported deficits (but not self-reported deficits) and a discrepancy score indicating greater informant- than self-reported functional deficits significantly predicted the development of AD. The discrepancy index showed high specificity and sensitivity for progression to AD within 2 years. ConclusionsThese findings indicate that in patients with MCI, the patient’s lack of awareness of functional deficits identified by informants strongly predicts a future diagnosis of AD. If replicated, these findings suggest that clinicians evaluating MCI patients should obtain both self-reports and informant reports of functional deficits to help in prediction of long-term outcome.

Combining early markers strongly predicts conversion from mild cognitive impairment to Alzheimer's disease.

BACKGROUNDnThe utility of combining early markers to predict conversion from mild cognitive impairment (MCI) to Alzheimers Disease (AD) remains uncertain.nnnMETHODSnIncluded in the study were 148 outpatients with MCI, broadly defined, followed at 6-month intervals. Hypothesized baseline predictors for follow-up conversion to AD (entire sample: 39/148 converters) were cognitive test performance, informant report of functional impairment, apolipoprotein E genotype, olfactory identification deficit, and magnetic resonance imaging (MRI) hippocampal and entorhinal cortex volumes.nnnRESULTSnIn the 3-year follow-up patient sample (33/126 converters), five of eight hypothesized predictors were selected by backward and stepwise logistic regression: Pfeffer Functional Activities Questionnaire (FAQ; informant report of functioning), University of Pennsylvania Smell Identification Test (UPSIT; olfactory identification), Selective Reminding Test (SRT) immediate recall (verbal memory), MRI hippocampal volume, and MRI entorhinal cortex volume. For 10% false positives (90% specificity), this five-predictor combination showed 85.2% sensitivity, combining age and Mini-Mental State Examination (MMSE) showed 39.4% sensitivity; combining age, MMSE, and the three clinical predictors (SRT immediate recall, FAQ, and UPSIT) showed 81.3% sensitivity. Area under ROC curve was greater for the five-predictor combination (.948) than age plus MMSE (.821; p = .0009) and remained high in subsamples with MMSE > or = 27/30 and amnestic MCI.nnnCONCLUSIONSnThe five-predictor combination strongly predicted conversion to AD and was markedly superior to combining age and MMSE. Combining the clinically administered measures also led to strong predictive accuracy. If independently replicated, the findings have potential utility for early detection of AD.

A 10-item smell identification scale related to risk for Alzheimer's disease

University of Pennsylvania Smell Identification Test data from control subjects (n = 63), patients with mild cognitive impairment (n = 147), and patients with Alzheimers disease (n = 100) were analyzed to derive an optimal subset of items related to risk for Alzheimers disease (ie, healthy through mild cognitive impairment to early and moderate disease stages). The derived 10‐item scale performed comparably with the University of Pennsylvania Smell Identification Test in classifying subjects, and it strongly predicted conversion to Alzheimers disease on follow‐up evaluation in patients with mild cognitive impairment. Independent replication is needed to validate these findings. Ann Neurol 2005;58:155–160

Covariance PET patterns in early Alzheimer's disease and subjects with cognitive impairment but no dementia: utility in group discrimination and correlations with functional performance.

Although multivariate analytic techniques might identify diagnostic patterns that are not captured by univariate methods, they have rarely been used to study the neural correlates of Alzheimers disease (AD) or cognitive impairment. Nonquantitative H2(15)O PET scans were acquired during rest in 17 probable AD subjects selected for mild severity [mean-modified Mini Mental Status Examination (mMMS) 46/57; SD 5.1], 16 control subjects (mMMS 54; SD 2.5) and 23 subjects with minimal to mild cognitive impairment but no dementia (mMMS 53; SD 2.8). Expert clinical reading had low success in discriminating AD and controls. There were no significant mean flow differences among groups in traditional univariate SPM Noxel-wise analyses or region of interest (ROI) analyses. A covariance pattern was identified whose mean expression was significantly higher in the AD as compared to controls (P = 0.03; sensitivity 76-94%; specificity 63-81%). Sites of increased concomitant flow included insula, cuneus, pulvinar, lingual, fusiform, superior occipital and parahippocampal gyri, whereas decreased concomitant flow was found in cingulate, inferior parietal lobule, middle and inferior frontal, supramarginal and precentral gyri. The covariance analysis-derived pattern was then prospectively applied to the cognitively impaired subjects: as compared to subjects with Clinical Dementia Rating (CDR) = 0, subjects with CDR = 0.5 had significantly higher mean covariance pattern expression (P = 0.009). Expression of this pattern correlated inversely with Selective Reminding Test total recall (r = -0.401, P = 0.002), delayed recall (r = -0.351, P = 0.008) and mMMS scores (r = -0.401, P = 0.002) in all three groups combined. We conclude that patients with AD may differentially express resting cerebral blood flow covariance patterns even at very early disease stages. Significant alterations in expression of resting flow covariance patterns occur even for subjects with cognitive impairment. Expression of covariance patterns correlates with cognitive and functional performance measures, holding promise for meaningful associations with underlying biopathological processes.

Olfactory identification deficits and MCI in a multi-ethnic elderly community sample

Odor identification deficits occur in Alzheimers disease (AD) and mild cognitive impairment (MCI), and predict clinical conversion from MCI to AD. In an epidemiologic study conducted in a multi-ethnic community elderly sample (average 80 years old), the University of Pennsylvania Smell Identification Test (UPSIT, range 0-40) was administered to 1092 non-demented subjects. Women (mean 26.6, S.D. 6.6) scored higher than men (mean 24.4, S.D. 7.4, p<.02), and ethnic differences were not significant after controlling for age and education. UPSIT scores correlated inversely with age (r=-0.24, p<.0001) and positively with Selective Reminding Test immediate recall (r=0.33), delayed recall (r=0.28), category fluency (r=0.28) and the 15-item Boston Naming Test (r=0.23), all ps<.0001. In a sub-sample in which MRI was done, UPSIT scores showed a significant correlation with hippocampal volume (n=571, r=0.16, p<.001) but not entorhinal cortex volume nor total number of white matter hyperintensities. In ANOVA, UPSIT scores differed (p<.0001) as a function of MCI classification: no MCI (mean 26.6, S.D. 6.8), non-amnestic MCI (mean 24.4, S.D. 7.2), and amnestic MCI (mean 23.5, S.D. 6.7). The difference between amnestic MCI and no MCI remained significant after controlling for relevant covariates. These findings indicate that the predictive utility of olfactory identification deficits for decline from no MCI to MCI and AD needs to be assessed in longitudinal studies of elderly community samples.

The impact of anxiety on conversion from mild cognitive impairment to Alzheimer's disease

To compare state and trait anxiety in mild cognitive impairment (MCI) patients and matched control subjects, and to assess the impact of these variables in predicting conversion to Alzheimers disease.

Investigating hemodynamic response variability at the group level using basis functions

Introduced is a general framework for performing group-level analyses of fMRI data using any basis set of two functions (i.e., the canonical hemodynamic response function and its first derivative) to model the hemodynamic response to neural activity. The approach allows for flexible implementation of physiologically based restrictions on the results. Information from both basis functions is used at the group level and the limitations avoid physiologically ambiguous or implausible results. This allows for investigation of specific BOLD activity such as hemodynamic responses peaking within a specified temporal range (i.e., 4-5 s). The general nature of the presented approach allows for applications using basis sets specifically designed to investigate various physiologic phenomena, i.e., age-related variability in poststimulus undershoot, hemodynamic responses measured with cerebral blood flow imaging, or subject-specific basis sets. An example using data from a group of healthy young participants demonstrates the methods and the specific steps to study poststimulus variability are discussed. The approach is completely implemented within the general linear model and requires minimal programmatic calculations.

Validation and optimization of statistical approaches for modeling odorant-induced fMRI signal changes in olfactory-related brain areas.

Recent neuroimaging studies have converged to show that odorant-induced responses to prolonged stimulation in primary olfactory cortex (POC) are characterized by a rapidly habituating time course. Different statistical approaches have effectively modeled this time course. One approach explicitly modeled rapid habituation using an exponentially decaying reference waveform that decreased to baseline levels within 30 to 40 s. A second approach modeled an early transient response by simply shortening the odorant ON period to be less than the actual stimulation period (i.e., 9 of 40 s). The goal of the current study was to validate, compare, and optimize these methodological approaches by applying them to an olfactory fMRI block-design dataset from 10 healthy young subjects presented with odorants for 12 s (ON), alternating with 30 s of clear air (OFF). Both approaches significantly improved sensitivity to odorant-induced signal changes in POC relative to a square-wave model based on the actual stimulation period. Our findings further demonstrate that the optimal model fit to the data was achieved by shortening the odorant ON period to approximately 6 s. These results suggest that sensitivity to odorant-induced POC activity in block-design experiments can be optimized by modeling an early phasic response followed by a precipitous rather than specific exponential decrease to baseline levels. Notably, whole brain voxel-wise analyses further established that modeling rapid habituation in this way is not only sensitive, but also highly specific to odorant-induced activation in a well-established network of olfactory-related brain areas.

PET Network Abnormalities and Cognitive Decline in Patients with Mild Cognitive Impairment

Temporoparietal and posterior cingulate metabolism deficits characterize patients with Alzheimers disease (AD). A H215O resting PET scan covariance pattern, derived by using multivariate techniques, was previously shown to discriminate 17 mild AD patients from 16 healthy controls. This AD covariance pattern revealed hypoperfusion in bilateral inferior parietal lobule and cingulate; and left middle frontal, inferior frontal, precentral, and supramarginal gyri. The AD pattern also revealed hyperperfusion in bilateral insula, lingual gyri, and cuneus; left fusiform and superior occipital gyri; and right parahippocampal gyrus and pulvinar. In an independent sample of 23 outpatients with mild cognitive impairment (MCI) followed at 6-month intervals, the AD pattern score was evaluated as a predictor of cognitive decline. In this MCI sample, an H215O resting PET scan was carried out at baseline. Mean duration of follow-up was 48.8 (SD 15.5) months, during which time six of 23 MCI patients converted to AD. In generalized estimating equations (GEE) analyses, controlling for age, sex, education, and baseline neuropsychological scores, increased AD pattern score was associated with greater decline in each neuropsychological test score over time (Mini Mental State Exam, Selective Reminding Test delayed recall, Animal Naming, WAIS-R digit symbol; Ps<0.01–0.001). In summary, a resting PET covariance pattern previously reported to discriminate AD patients from control subjects was applied prospectively to an independent sample of MCI patients and found to predict cognitive decline. Independent replication in larger samples is needed before clinical application can be considered.