Search tips
Search criteria 


Logo of neurologyNeurologyAmerican Academy of Neurology
Neurology. 2012 May 29; 78(22): 1769–1776.
PMCID: PMC3359586

Family history of Alzheimer disease predicts hippocampal atrophy in healthy middle-aged adults

O.C. Okonkwo, PhD, G. Xu, MD, PhD, N.M. Dowling, PhD, B.B. Bendlin, PhD, A. LaRue, PhD, B.P. Hermann, PhD, R. Koscik, PhD, E. Jonaitis, PhD, H.A. Rowley, MD, C.M. Carlsson, MD, MS, S. Asthana, MD, M.A. Sager, MD, and S.C. Johnson, PhDcorresponding author



To evaluate the longitudinal influence of family history (FH) of Alzheimer disease (AD) and apolipoprotein E ϵ4 allele (APOE4) on brain atrophy and cognitive decline over 4 years among asymptomatic middle-aged individuals.


Participants were cognitively healthy adults with (FH+) (n = 60) and without (FH−) (n = 48) a FH of AD (mean age at baseline 54 years) enrolled in the Wisconsin Registry for Alzheimer's Prevention. They underwent APOE genotyping, cognitive testing, and an MRI scan at baseline and 4 years later. A covariate-adjusted voxel-based analysis interrogated gray matter (GM) modulated probability maps at the 4-year follow-up visit as a function of FH and APOE4. We also examined the influence of parent of origin on GM atrophy. Parallel analyses investigated the effects of FH and APOE4 on cognitive decline.


Neither FH nor APOE4 had an effect on regional GM or cognition at baseline. Longitudinally, a FH × APOE4 interaction was found in the right posterior hippocampus, which was driven by a significant difference between the FH+ and FH− subjects who were APOE4−. In addition, a significant FH main effect was observed in the left posterior hippocampus. No significant APOE4 main effects were detected. Persons with a maternal history of AD were just as likely as those with a paternal history of AD to experience posterior hippocampal atrophy. There was no longitudinal decline in cognition within the cohort.


Over a 4-year interval, asymptomatic middle-aged adults with FH of AD exhibit significant atrophy in the posterior hippocampi in the absence of measurable cognitive changes. This result provides further evidence that detectable disease-related neuroanatomic changes do occur early in the AD pathologic cascade.

A greater understanding of brain changes that occur before symptomatic Alzheimer disease (AD) is imperative for the accurate prediction of subsequent clinical symptoms and would greatly inform the design of enriched treatment trials targeting the earliest phases of AD before extensive neurodegenerative changes.1 Although apolipoprotein E ϵ4 allele (APOE4) has been the strongest genetic risk factor for sporadic or late-onset AD,2 it has become increasingly clear that parental family history (FH) of AD3 not only confers additional risk but also may account for as much or more variance in preclinical brain changes than APOE4.49

For several years, our group has been following a cohort of cognitively healthy middle-aged adults with a FH of AD known as the Wisconsin Registry for Alzheimer's Prevention (WRAP).10 Initial cross-sectional analyses found evidence for FH-related differences in brain activation,1113 white matter fiber integrity,14 and list-learning efficiency15 that were independent of APOE4. In the present article, we extend these prior findings by investigating the longitudinal effect of FH and APOE4 on brain morphometry within this cohort, with a particular focus on temporoparietal regions that are now known to be vulnerable to AD pathology.16,17 In addition, because it has recently been shown that the sex of the AD-affected parent may be influential, with maternal FH being more deleterious,5,7,8,18 we conducted supplementary analyses to determine whether any observed FH effects were possibly driven by maternal FH. In parallel analyses, we also investigated the effect of FH and APOE4 on cognitive decline.



A total of 108 subjects participated in this study. Eighty-nine subjects were consecutively recruited from the larger WRAP10 cohort, a longitudinal registry of adults who were cognitively healthy and between the ages of 40 and 65 at study entry, and the remaining 19 subjects were consecutively recruited from the community. Of these 108 subjects, 60 persons (mean baseline age 53.40 years; SD 6.24) had at least one parent with a reported diagnosis of AD (FH+ group) and the other 48 individuals (mean baseline age 54.25 years; SD 6.54) reported no family history of AD (FH− group). All 60 FH+ subjects were from the larger WRAP cohort, whereas the 48 FH− subjects were either from the WRAP cohort (n = 29) or the community (n = 19). Among the 60 FH+ subjects, 44 persons had a maternal history of AD (mFH+), 13 had a paternal history (pFH+), and 3 had both parents afflicted with AD.

To verify the diagnosis of AD in the parent, parental medical records were obtained (including autopsy reports when available) and reviewed by a multidisciplinary diagnostic consensus panel. In the majority of cases, the parent's clinical workup and diagnosis of AD had occurred at the University of Wisconsin Memory Clinics. Parental reported onset of AD was, on average, at age 73. None of the families included had any of the known autosomal dominant mutations. Absence of FH of AD was verified through detailed medical history surveys and phone interview with the participants. Inclusion in the FH− group required that the father survive to at least age 70 and the mother to age 75 without incurring a formal diagnosis of dementia or exhibiting cognitive deterioration.

At baseline and approximately 4 years later, all subjects underwent a comprehensive neuropsychological evaluation and a high-resolution 3-dimensional T1-weighted anatomic MRI scan. The cognitive tests were administered per standard protocol10,19 and included the following measures: the Mini-Mental State Examination, the Reading subtest of the Wide Range Achievement Test, third edition, the Rey Auditory Verbal Learning Test, the Trail Making Test, the Controlled Oral Word Association Test, the Boston Naming Test, and the Digit Span subtest of the Wechsler Adult Intelligence Scale, third edition. The MRI scans were acquired in the axial plane on a GE Signa 3.0-T scanner with a standard transmit-receive quadrature coil using the following parameters: inversion time/echo time/repetition time = 600 msec/5 msec/8.4 msec, flip angle = 10°, slice thickness = 1.2 mm, field of view = 240 mm, and matrix size = 256 × 256. All scans were reviewed by a neuroradiologist for potential abnormalities. At baseline, subjects also underwent APOE genotyping using PCR sequencing.

All neurocognitive and MRI procedures were executed using identical parameters at both baseline and 4-year study visits. Study exclusion criteria included MRI contraindications, major neurologic disorder (e.g., head trauma with loss of consciousness, neoplasms, and seizure disorders), current major psychiatric disease (e.g., schizophrenia), or abnormal MRI findings (e.g., ventriculomegaly).

Standard protocol approvals, registrations, and patient consents.

The University of Wisconsin Institutional Review Board approved all study procedures and each subject provided signed informed consent before participation.

Image processing.

The MRI images were processed using the newseg option in SPM8 (, which combines segmentation of the original anatomic image into 6 tissue classes/probability maps: gray matter (GM), white matter, CSF, skull, fat tissue, and image background; normalization of the segmented GM map to the Montreal Neurological Institute (MNI) template; and correction for image intensity nonuniformities within a unified, iterative framework. A modulation step was then used, which scaled each voxel in the final GM map by the amount of contraction/expansion required to warp the image to the MNI template. The resulting GM modulated probability map provides a measure of GM volume at the local (i.e., voxel) level. The normalized maps were then smoothed using an 8-mm isotropic Gaussian kernel, yielding images with a voxel size of 1.5 × 1.5 × 1.5 mm.

To focus our analyses on brain regions known to be implicated in AD pathogenesis and reduce the risk of false-positive errors, we 1) imposed an a priori anatomic mask that included the bilateral posterior cingulate, hippocampus, parahippocampal gyrus, and amygdala using the WFU_PickAtlas toolbox,20 (figure 1) and and2)2) thresholded the GM maps at 0.1 to minimize inclusion of white matter and CSF in our analysis.

Figure 1
Anatomic search region examined in the study
Figure 2
Effects of family history (FH) and apolipoprotein E4 ϵ4 allele (APOE4) status on gray matter (GM) atrophy

Statistical analysis.

Neuropsychological and demographic data were analyzed in SPSS 20.0 (IBM Corp., Armonk, NY). For the analyses of demographic measures, a FH of AD was cross-tabulated with APOE4 status to yield 4 groups: negative FH and APOE4-negative (FH−APOE4−); negative FH and APOE4-positive (FH−APOE4+); positive FH and APOE4-negative (FH+APOE4−); and positive FH and APOE4-positive (FH+APOE4+). Categorical variables were analyzed using χ2 tests, whereas continuous variables were analyzed using 1-way analyses of variance.

The analysis of neuropsychological measures was done using a series of 2 × 2 factorial analyses of covariance (ANCOVAs) with FH and APOE4 status as fixed factors. For baseline models, cognitive test scores were the dependent variables and age was entered as a covariate. For the longitudinal models, test scores at the 4-year follow-up visit served as dependent measures with baseline age and the respective baseline test scores serving as covariates.

We evaluated the effects of family history of AD and APOE4 status on GM cross-sectionally and longitudinally using a voxel-based morphometry framework implemented in SPM8. For the cross-sectional analysis, we entered participants' baseline GM modulated probability voxel-wise maps into a 2 × 2 factorial ANCOVA with FH and APOE4 status as fixed factors and age and sex as covariates. To assess for longitudinal change in GM as a function of FH and APOE4 status, we performed a 2 × 2 factorial ANCOVA on participants' 4-year follow-up scans using the Biological Parametric Mapping toolbox of SPM8,21 with FH and APOE4 status as fixed factors and age, sex, and the baseline GM modulated probability voxel-wise map as covariates.

To determine whether maternal history of AD had a differential effect on GM atrophy than paternal history of AD,5,8 we repeated the longitudinal analysis described above but this time using a one-way ANCOVA design to assess whether there was a parent of origin effect on participants' 4-year follow-up scans, while adjusting for age, sex, baseline voxel-wise GM, and APOE4 status (see Results section for specific a priori contrasts tested). For all the imaging analyses described above, only clusters with a minimum of 188 continuous voxels and pvoxel < 0.005 were deemed significant. This a priori threshold was derived via Monte Carlo simulations (AlphaSim, AFNI,


Demographic findings.

Table 1 shows the results of group comparisons on demographic variables. The groups were statistically equated on all measures, except age: the FH−APOE4+ group was, on average, younger than the other groups.

Table 1
Demographics of study participants at baseline

Neuropsychological findings.

Results of the analyses of neuropsychological measures are displayed in table 2. There were no significant FH or APOE4 main effects and no significant FH × APOE4 interactions at baseline or at the 4-year follow-up assessment.

Table 2
Participants' performance scores on neuropsychological tests at baseline and 4-year follow-upa

Neuroimaging findings.

Our cross-sectional analyses did not reveal any significant FH or APOE4 main effects or FH × APOE4 interactions on baseline GM.

In the longitudinal analyses, we found a significant FH × APOE4 interaction at the right posterior hippocampus (figure 2, cyan color and bar graph) with a cluster of 201 voxels and maximum T of 3.63 (pvoxel < 0.0001) at x,y,z [36,−39,−5]. Follow-up simple effects analyses found that, among persons who were APOE4+, there was no significant FH effect, whereas among persons who were APOE4−, there was a significant FH effect (i.e., FH+ subjects showed more atrophy than FH− subjects) at the right posterior hippocampus with a cluster of 193 voxels and maximum T of 4.80 (pvoxel < 0.0001) at x,y,z [34,−36,−5].

In addition to the above interaction effect, we found a main effect of FH (wherein FH+ subjects exhibited more GM atrophy than FH− subjects) at the left posterior hippocampus (figure 2, red color) with a cluster of 286 voxels and maximum T of 3.65 (Pvoxel < 0.0001) and at x,y,z coordinate [−26,−37,−6]. No APOE4 effects were observed. Of note, for the main effects analyses, in addition to limiting our search region to the AD-related areas defined by our anatomic mask, we also imposed a statistical mask that excluded from consideration those voxels that were identified in the FH × APOE4 status analysis. This guarded against the statistical misstep of testing for main effects in the same regions where interactions were observed.

For our examination of potential parent of origin effects, we performed the following a priori contrasts: 1) no family history of AD vs maternal history of AD or paternal history of AD, 2) no family history of AD vs maternal history of AD, 3) no family history of AD vs paternal history of AD, and 4) maternal history of AD vs paternal history of AD. Consistent with the FH findings from the 2 × 2 analysis (cf. preceding paragraph), contrast 1 revealed that subjects who were either mFH+ or pFH+ harbored greater posterior hippocampal atrophy bilaterally than FH− subjects. For the right posterior hippocampus, cluster size = 911 voxels, maximum T = 4.20 (pvoxel < 0.0001), and x,y,z coordinate = [34,−27,−11] (see bar graph in figure 3). For the left posterior hippocampus, cluster size = 466 voxels, maximum T = 4.12 (pvoxel < 0.0001), and x,y,z coordinate = [−32,−34,−6].

Figure 3
Effects of parent of origin on gray matter (GM) atrophy

Contrast 2 revealed that mFH+ subjects had greater bilateral posterior hippocampal atrophy than FH− subjects (figure 3, green color). For the left posterior hippocampus, cluster size = 268 voxels, maximum T = 3.55 (pvoxel < 0.0001), and x,y,z coordinate = [−30,−36,−6]. For the right posterior hippocampus, cluster size = 907 voxels, maximum T = 3.50 (pvoxel < 0.0001), and x,y,z coordinate = [32,−25,−11]. Similarly, contrast 3 revealed that pFH+ subjects experienced greater bilateral posterior hippocampal atrophy than FH− subjects (figure 3, red color). For the right posterior hippocampus, cluster size = 425 voxels, maximum T = 3.79 (pvoxel < 0.0001), and x,y,z coordinate = [34,−25,−12]. For the left posterior hippocampus, cluster size = 301 voxels, maximum T = 3.50 (pvoxel < 0.0001), and x,y,z coordinate = [−34,−28,−9]. Contrast 4 did not reveal any regions where mFH+ subjects had greater GM atrophy than pFH+ subjects or vice versa.

Exploratory whole-brain analyses.

In addition to the targeted longitudinal neuroimaging analyses reported above, we also conducted exploratory longitudinal whole-brain analyses to determine whether there were any FH, APOE, or parent of origin-related effects outside of our anatomic search region. The findings from our focused analyses were essentially replicated, one notable exception being an APOE main effect in the right middle temporal gyrus (for details, see appendix e-1 and table e-1 on the Neurology® web site at


In this study, we found that whereas neither FH nor APOE4 appeared to affect cerebral GM at baseline, FH exerted both independent and interactive (with APOE4) effects on longitudinal atrophy of brain regions susceptible to AD pathology. Specifically, there was a significant longitudinal FH × APOE4 interaction at the right posterior hippocampus. Further interrogation of this interaction revealed that FH was associated with increased atrophy of the right posterior hippocampus among persons who were APOE4−, but that this effect was absent (at the prespecified threshold) among persons who were APOE4+. In addition to this interactive effect, FH was independently associated with greater atrophy of the left posterior hippocampus. In contrast, we did not find any associations between APOE4 status and GM atrophy within our search region. It is noteworthy that the observed FH-associated GM atrophy was found in the absence of any detectable decline on standard measures of cognitive functioning in this cohort during the same 4-year interval.

Prior studies from our group and other centers have shown that FH is associated with AD-related abnormalities in cerebral structure and function even after adjustment for the potentially confounding effect of APOE4. For example, recent cross-sectional18 and longitudinal5 studies have found evidence for GM volume reductions in AD-susceptible brain regions, such as the precuneus, among cognitively intact FH+ individuals relative to FH− peers. Our study extends these findings5,18 by using a considerably larger sample, implementing a longer follow-up, and, more importantly, showing that FH-related hippocampal atrophy is present at a relatively young age (our cohort's mean baseline age was 54 years). The relative consistency of these FH findings across diverse neuroimaging modalities4,79,11 suggests that FH is a prominent, even if nonspecific, risk factor for AD perhaps embodying an array of genetic and environmental indices that remain to be fully elucidated.10,22

Contrary to prior reports suggesting that maternal history of AD might confer a greater risk of AD-related cerebral abnormalities than paternal history of AD,5,7,8,18,23 we found that pFH+ subjects were just as likely as mFH+ subjects to experience atrophy of the posterior hippocampus. This observation suggests that, in our cohort, the FH effect was not driven by a particular parent of origin. The reasons for this discrepancy in study findings are not entirely clear. One possibility has to do with the age of the cohorts being studied. The participants in our study were, on average, considerably younger (in some cases by as much as 20 years) than the participants in these other studies. It might be that the effect of parent of origin only becomes detectable in older age, given that advanced age itself is a well-established risk factor for AD.22

Although we believe that our finding of FH-associated hippocampal atrophy is bona fide and probably a signal for preclinical AD, we also acknowledge that the question of hippocampal atrophy in the context of normal cognitive aging continues to be actively investigated with some reports suggesting that, even in the absence of neurodegenerative diseases, there is veritable shrinkage of the hippocampus as people age.2426 It is unlikely that the differential hippocampal atrophy we observed between FH− and FH+ subjects is merely a proxy for naturally occurring age-associated hippocampal shrinkage because our FH+ and FH− groups were essentially matched on age (mean age difference = 0.85 year, p = 0.493), and, furthermore, we adjusted for age in our statistical model, thereby accounting for whatever contribution it may have made to the observed atrophy. In addition, data from other groups have indicated that normal aging is not necessarily accompanied by hippocampal atrophy2730 and that the observation of age-associated hippocampal atrophy in some studies might reflect either the contamination of such studies with participants harboring diseases that are potentially inimical to the hippocampus31 or the differential inclusion of persons with preexisting, developmentally determined, smaller hippocampi.32

Our study had some limitations. First, parental history of AD was mostly established via clinical diagnosis in consensus diagnostic conferences using established criteria. Therefore, we cannot entirely rule out the possibility that the FH+ group may have contained people whose parents did not have AD. Similarly, negative parental history of AD was determined by self-report which may not be entirely reliable. However, the net effect of either form of misclassification would have been a diminution of our findings, as opposed to an augmentation. Another potential limitation is the racial homogeneity of our cohort. Although this reflects the demographic of the locale in which the study was conducted, it limits our ability to generalize our findings to non-Caucasian samples. We also acknowledge the modest size of the FH−APOE4+ group. However, this reflects the true state of affairs given that the joint probability of being APOE4+ and having no parental history of AD is small. Last, whereas the regression approach to longitudinal analyses has been shown to confer specific advantages over change score approaches,3336 it is possible that a somewhat different pattern of findings might have emerged had we used analytical approaches that estimate atrophy by computing change between follow up and baseline scans, such as tensor-based morphometry37 or boundary shift integral.38

The present study demonstrates that relatively young cognitively healthy persons with a parental FH of AD experience detectable shrinkage of brain regions that are implicated in the pathology of AD and that this atrophy occurs before measurable decline in cognitive function. This is a notable finding because current models of the AD pathologic cascade suggest that brain atrophy is a later event, occurring closer to the manifestation of clinical AD,39 whereas we show that, at least in at-risk individuals, atrophy of vulnerable brain regions exist much earlier (our cohort's mean age at baseline was 54 years). This and other studies reviewed above suggest that the AD pathophysiologic cascade may be expressed differently in enriched risk groups. Continued follow-up of our FH+ subjects would be helpful in determining whether those who subsequently experience clinical decline are identifiable by the observed atrophy of the posterior hippocampus. This would greatly boost the potential utility of such early posterior hippocampal atrophy as an endophenotype of preclinical AD.1 Furthermore, it would be useful to characterize the correspondence between such morphometric changes and other brain measures such as amyloid load and metabolic function within this relatively young cohort.

Supplementary Material

Data Supplement:



Alzheimer disease
analysis of covariance
family history
gray matter
maternal family history
Montreal Neurological Institute
paternal family history
Wisconsin Registry for Alzheimer's Prevention


Supplemental data at


Drafting/revising the manuscript for content: Drs. Okonkwo, Xu, Dowling, Bendlin, LaRue, Hermann, Koscik, Jonaitis, Rowley, Carlsson, Asthana, Sager, and Johnson. Study concept or design: Drs. Xu, Dowling, Asthana, Sager, and Johnson. Analysis or interpretation of data: Drs. Okonkwo, Xu, Dowling, Hermann, Koscik, Jonaitis, Carlsson, and Johnson. Acquisition of data: Drs. Bendlin, LaRue, Hermann, Rowley, Sager, and Johnson. Statistical analysis: Drs. Okonkwo and Xu. Study supervision or coordination: Drs. Carlsson and Johnson. Obtaining funding: Drs. Asthana, Sager, and Johnson.


The authors report no disclosures relevant to the manuscript. Go to for full disclosures.


1. Sperling RA, Aisen PS, Beckett LA, et al. Toward defining the preclinical stages of Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement 2011; 7: 280– 292 [PMC free article] [PubMed]
2. Bertram L, Tanzi RE. Thirty years of Alzheimer's disease genetics: the implications of systematic meta-analyses. Nat Rev Neurosci 2008; 9: 768– 778 [PubMed]
3. Cupples LA, Farrer LA, Sadovnick AD, Relkin N, Whitehouse P, Green RC. Estimating risk curves for first-degree relatives of patients with Alzheimer's disease: the REVEAL study. Genet Med 2004; 6: 192– 196 [PubMed]
4. Donix M, Burggren AC, Suthana NA, et al. Family history of Alzheimer's disease and hippocampal structure in healthy people. Am J Psychiatry 2010; 167: 1399– 1406 [PMC free article] [PubMed]
5. Honea RA, Swerdlow RH, Vidoni ED, Burns JM. Progressive regional atrophy in normal adults with a maternal history of Alzheimer disease. Neurology 2011; 76: 822– 829 [PMC free article] [PubMed]
6. Debette S, Wolf PA, Beiser A, et al. Association of parental dementia with cognitive and brain MRI measures in middle-aged adults. Neurology 2009; 73: 2071– 2078 [PMC free article] [PubMed]
7. Mosconi L, Mistur R, Switalski R, et al. Declining brain glucose metabolism in normal individuals with a maternal history of Alzheimer disease. Neurology 2009; 72: 513– 520 [PMC free article] [PubMed]
8. Mosconi L, Rinne JO, Tsui WH, et al. Increased fibrillar amyloid-β burden in normal individuals with a family history of late-onset Alzheimer's. Proc Natl Acad Sci USA 2010; 107: 5949– 5954 [PubMed]
9. Xiong C, Roe CM, Buckles V, et al. Role of family history for Alzheimer biomarker abnormalities in the adult children study. Arch Neurol 2011; 68: 1313– 1319 [PMC free article] [PubMed]
10. Sager MA, Hermann B, La Rue A. Middle-aged children of persons with Alzheimer's disease: APOE genotypes and cognitive function in the Wisconsin Registry for Alzheimer's Prevention. J Geriatr Psychiatry Neurol 2005; 18: 245– 249 [PubMed]
11. Johnson SC, Schmitz TW, Trivedi MA, et al. The influence of Alzheimer disease family history and apolipoprotein E ϵ4 on mesial temporal lobe activation. J Neurosci 2006; 26: 6069– 6076 [PMC free article] [PubMed]
12. Johnson SC, Ries ML, Hess TM, et al. Effect of Alzheimer disease risk on brain function during self-appraisal in healthy middle-aged adults. Arch Gen Psychiatry 2007; 64: 1163– 1171 [PMC free article] [PubMed]
13. Xu G, McLaren DG, Ries ML, et al. The influence of parental history of Alzheimer's disease and apolipoprotein E ϵ4 on the BOLD signal during recognition memory. Brain 2009; 132: 383– 391 [PMC free article] [PubMed]
14. Bendlin BB, Ries ML, Canu E, et al. White matter is altered with parental family history of Alzheimer's disease. Alzheimers Dement 2010; 6: 394– 403 [PMC free article] [PubMed]
15. La Rue A, Hermann B, Jones JE, Johnson S, Asthana S, Sager MA. Effect of parental family history of Alzheimer's disease on serial position profiles. Alzheimers Dement 2008; 4: 285– 290 [PMC free article] [PubMed]
16. Buckner RL, Snyder AZ, Shannon BJ, et al. Molecular, structural, and functional characterization of Alzheimer's disease: evidence for a relationship between default activity, amyloid, and memory. J Neurosci 2005; 25: 7709– 7717 [PubMed]
17. Reiman EM, Caselli RJ, Yun LS, et al. Preclinical evidence of Alzheimer's disease in persons homozygous for the ϵ4 allele for apolipoprotein E. N Engl J Med 1996; 334: 752– 758 [PubMed]
18. Honea RA, Swerdlow RH, Vidoni ED, Goodwin J, Burns JM. Reduced gray matter volume in normal adults with a maternal family history of Alzheimer disease. Neurology 2010; 74: 113– 120 [PMC free article] [PubMed]
19. Lezak MD, Howieson DB, Loring DW. Neuropsychological Assessment, 4th ed. New York: Oxford University Press; 2004.
20. Maldjian JA, Laurienti PJ, Kraft RA, Burdette JH. An automated method for neuroanatomic and cytoarchitectonic atlas-based interrogation of fMRI data sets. Neuroimage 2003; 19: 1233– 1239 [PubMed]
21. Casanova R, Srikanth R, Baer A, et al. Biological parametric mapping: a statistical toolbox for multimodality brain image analysis. Neuroimage 2007; 34: 137– 143 [PMC free article] [PubMed]
22. Jarvik LF, Blazer D. Children of Alzheimer patients: an overview. J Geriatr Psychiatry Neurol 2005; 18: 181– 186 [PubMed]
23. Mosconi L, Glodzik L, Mistur R, et al. Oxidative stress and amyloid-β pathology in normal individuals with a maternal history of Alzheimer's. Biol Psychiatry 2010; 68: 913– 921 [PMC free article] [PubMed]
24. Walhovd KB, Westlye LT, Amlien I, et al. Consistent neuroanatomical age-related volume differences across multiple samples. Neurobiol Aging 2011; 32: 916– 932 [PMC free article] [PubMed]
25. Raz N, Rodrigue KM, Head D, Kennedy KM, Acker JD. Differential aging of the medial temporal lobe: a study of a five-year change. Neurology 2004; 62: 433– 438 [PubMed]
26. De Leon MJ, George AE, Golomb J, et al. Frequency of hippocampal formation atrophy in normal aging and Alzheimer's disease. Neurobiol Aging 1997; 18: 1– 11 [PubMed]
27. Sullivan EV, Marsh L, Pfefferbaum A. Preservation of hippocampal volume throughout adulthood in healthy men and women. Neurobiol Aging 2005; 26: 1093– 1098 [PubMed]
28. Raz N, Gunning FM, Head D, et al. Selective aging of the human cerebral cortex observed in vivo: differential vulnerability of the prefrontal gray matter. Cereb Cortex 1997; 7: 268– 282 [PubMed]
29. Good CD, Johnsrude IS, Ashburner J, Henson RN, Friston KJ, Frackowiak RS. A voxel-based morphometric study of ageing in 465 normal adult human brains. Neuroimage 2001; 14: 21– 36 [PubMed]
30. DeCarli C, Murphy DG, Gillette JA, et al. Lack of age-related differences in temporal lobe volume of very healthy adults. AJNR Am J Neuroradiol 1994; 15: 689– 696 [PubMed]
31. Raz N, Lindenberger U, Rodrigue KM, et al. Regional brain changes in aging healthy adults: general trends, individual differences and modifiers. Cereb Cortex 2005; 15: 1676– 1689 [PubMed]
32. Lupien SJ, Evans A, Lord C, et al. Hippocampal volume is as variable in young as in older adults: implications for the notion of hippocampal atrophy in humans. Neuroimage 2007; 34: 479– 485 [PubMed]
33. Senn S. Change from baseline and analysis of covariance revisited. Stat Med 2006; 25: 4334– 4344 [PubMed]
34. Vickers AJ, Altman DG. Statistics notes: analysing controlled trials with baseline and follow up measurements. BMJ 2001; 323: 1123– 1124 [PMC free article] [PubMed]
35. Cohen J, Cohen P, West SG, Aiken LS. Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences, 3rd ed. Mahwah, NJ: Lawrence Erlbaum Associates; 2003.
36. Frison L, Pocock SJ. Repeated measures in clinical trials: analysis using mean summary statistics and its implications for design. Stat Med 1992; 11: 1685– 1704 [PubMed]
37. Ashburner J, Friston KJ. Voxel-based morphometry: the methods. Neuroimage 2000; 11: 805– 821 [PubMed]
38. Barnes J, Boyes RG, Lewis EB, et al. Automatic calculation of hippocampal atrophy rates using a hippocampal template and the boundary shift integral. Neurobiol Aging 2007; 28: 1657– 1663 [PubMed]
39. Jack CR, Jr, Knopman DS, Jagust WJ, et al. Hypothetical model of dynamic biomarkers of the Alzheimer's pathological cascade. Lancet Neurol 2010; 9: 119– 128 [PMC free article] [PubMed]

Articles from Neurology are provided here courtesy of American Academy of Neurology