The histological validation of post mortem magnetic resonance imaging-determined hippocampal volume in Alzheimer's disease

For 11 AD cases and four normal elderly controls, post mortem volumes of the hippocampal subdivisions were calculated by using magnetic resonance imaging and histological sections. After at least six weeks of fixation in formalin, brains were examined on a 1.5-T Philips Gyroscan imager producing T₁-weighted coronal images with a 3-mm slice thickness. Brains were then processed and embedded in paraffin. Serial coronal sections, 3 mm apart and stained with Cresyl Violet, were used for the planimetry and unbiased estimation of the total numbers of neurons in the hippocampal subdivisions. For all 15 cases, magnetic resonance imaging- and histology-based measurements were performed along the whole rostrocaudal extent of the hippocampal formation and included three subvolumes: (i) the hippocampus (CA1–CA4 and the dentate gyrus); (ii) hippocampus/subiculum; and (iii) hippocampus/parahippocampal gyrus. After controlling for shrinkage, strong correlations were found between magnetic resonance imaging and histological measurements for the hippocampus (r=0.97, P<0.001), hippocampus/subiculum (r=0.95, P<0.001) and hippocampus/parahippocampal gyrus (r=0.89, P<0.001). We also calculated the total number of neurons in the hippocampus and hippocampus/subiculum subvolumes. Strong correlations between the magnetic resonance imaging subvolumes and neuronal counts were found for the hippocampus (r=0.90, P<0.001) and the hippocampus/subiculum subvolume (r=0.84, P<0.001).We conclude that very accurate volumetric measurements of the whole hippocampal formation can be obtained by using a magnetic resonance imaging protocol. Moreover, the strong correlations between magnetic resonance imaging-based hippocampal volumes and neuronal numbers suggest the anatomical validity of magnetic resonance imaging volume measurements.     

Dual pathways connecting the dorsolateral prefrontal cortex with the hippocampal formation and parahippocampal cortex in the rhesus monkey

Anterograde and retrograde tracing methods including autoradiography, horseradish peroxidase histochemistry and fluorescent dye transport were used to demonstrate that the dorsolateral prefrontal cortex is connected with the hippocampal formation and associated cortical regions by two distinct pathways. Fibers forming a lateral pathway travel in the fronto-occipital fasciculus and connect the dorsolateral prefrontal cortex with the fundus of the rhinal sulcus, posterior subdivisions of the parahippocampal gyrus, and the presubiculum. A larger medial pathway forms in the cingulum bundle and terminates in the most caudal part of the presubiculum, as well as in adjacent transitional cortices. These cortices form a caudomedial promontory that is located between the posterior cingulate and prestriate areas. In all allo- and mesocortical targets of prefrontal cortex, labeled terminals form banding patterns reminiscent of the columnar organization of afferent fiber columns in neocortex. The same cytoarchitectonic areas that receive prefrontal afferents issue reciprocal projections. The largest source is the caudomedial lobule including its presubicular portion. Neurons in the parahippocampal gyrus and adjacent presubiculum also are retrogradely labeled following implants of horseradish peroxidase or injection of fluorescent dyes into prefrontal cortex. In addition, subicular neurons project to the prefrontal cortex although the subiculum does not appear to receive prefrontal afferent input. These findings emphasize that multiple channels of communication link the dorsolateral prefrontal cortex and the hippocampus via the parahippocampal gyrus, subiculum, presubiculum and adjacent transitional cortices. We speculate that each of these prefrontal projections may carry highly specific information into the hippocampus, whereas the reciprocal projections may allow retrieval by prefrontal cortex of memories stored in the hippocampus.     

Age-related metabolite changes and volume loss in the hippocampus by magnetic resonance spectroscopy and imaging

Magnetic resonance imaging (MRI) studies have produced controversial results concerning the correlation of hippocampal volume loss with increasing age. The goals in this study were: 1) to test whether levels of N-acetyl aspartate (NAA, a neuron marker) change in the hippocampus during normal aging and 2) to determine the relationship between hippocampal NAA and volume changes. Proton magnetic resonance spectroscopic imaging (1H MRSI) and MRI were used to measure hippocampal metabolites and volumes in 24 healthy adults from 36 to 85 years of age. NAA/Cho decreased by 24% (r = 0.53, p = 0.01) and NAA/Cr by 26% (r = 0.61, p < 0.005) over the age range studied, whereas Cho/Cr remained stable, implying diminished NAA levels. Hippocampal volume shrank by 20% (r = 0.64, p < 0.05). In summary, aging effects must be considered in 1H MRSI brain studies. Furthermore, because NAA is considered a marker of neurons, these results provide stronger support for neuron loss in the aging hippocampus than volume measurements by MRI alone.     

A CYP46 T/C SNP modulates parahippocampal and hippocampal morphology in young subjects

There is evidence that brain cholesterol metabolism modulates the vulnerability for Alzheimer's disease (AD). Previous data showed that brain β-amyloid load in elderly subjects with the CYP46 (cholesterol 24S-hydroxylase) TT-positive genotype was higher than in CYP46 TT-negative elderly subjects. We investigated effects of the CYP46 T/C polymorphism on parahippocampal and hippocampal grey matter (GM) morphology in 81 young subjects using structural magnetic resonance imaging based morphometry. We found that young TT-homozygotes exhibited smallest and CC-homozygotes largest parahippocampal and hippocampal GM volumes with the volumes of the CT-heterozygotes ranging in between. Parahippocampal and hippocampal volumes were positively correlated with delayed memory performance in C-carriers and negatively with immediate memory performance in TT-homozygotes. It has been shown that the brain cholesterol metabolism in general modulates dendrite outgrowth, synaptogenesis, and neuron survival, and it was suggested that CYP46 indirectly influences β-amyloid metabolism. CYP46 C-carriers are privileged both in terms of β-amyloid metabolism and in terms of brain reserve due to their larger parahippocampal and hippocampal structures. The exact cellular mechanisms that translate the CYP46 allelic variation into volumetric brain differences in the parahippocampal gyrus and hippocampus are still unknown and need to be further investigated.     

Morphological brain changes associated with Schneider's first-rank symptoms in schizophrenia: a MRI study

BACKGROUND: Schneider's first-rank symptoms involve an alienated feature of the sense of one's own mental or physical activity. To clarify the brain morphological basis for the production of these symptoms, volumes of the frontal and medial temporal regions and their clinical correlates were examined in patients with schizophrenia. METHOD: Twenty-two patients with schizophrenia and 44 age- and gender-matched healthy control subjects were included. All patients were in their psychotic episodes with definite Schneiderian symptoms, rated by using the Scale for Assessment of Positive Symptoms. Volumetric measurements of high-resolution magnetic resonance imaging were performed in the prefrontal area, cingulate gyrus, and precentral gyrus, and the medial temporal structures such as the amygdala, hippocampus, and parahippocampal gyrus. RESULTS: Patients had significantly decreased volumes in the cingulate gray matter and the amygdala compared to controls. In the patient group, Schneiderian symptom severity showed significant inverse correlations with volumes of the right posterior cingulate gray matter and of the left anterior parahippocampal gyrus. CONCLUSIONS: Schneiderian symptoms may be associated with morphological abnormalities in the limbic-paralimbic regions such as the cingulate gyrus and parahippocampal gyrus, which possibly serve the self-monitoring function and the coherent storage and reactivation of information.     

Association between the brain-derived neurotrophic factor Val66Met polymorphism and brain morphology in a Japanese sample of schizophrenia and healthy comparisons

Magnetic resonance imaging was used to investigate the relation between the brain-derived neurotrophic factor (BDNF) Val66Met polymorphism and volumetric measurements for the medial temporal lobe structures (amygdala, hippocampus, and parahippocampal gyrus) and prefrontal sub-regions (the superior frontal gyrus, middle frontal gyrus, inferior frontal gyrus, ventral medial prefrontal cortex, orbitofrontal cortex, and straight gyrus) in a Japanese sample of 33 schizophrenia patients and 29 healthy subjects. For the controls, the Met carriers had significantly smaller parahippocampal and left superior frontal gyri than the Val homozygotes. The schizophrenia patients carrying the Met allele had a significantly smaller right parahippocampal gyrus than those with the Val/Val genotype, but the genotype did not affect the prefrontal regions in schizophrenia patients. These findings might reflect different genotypic effects of BDNF on brain morphology in schizophrenia patients and healthy controls, implicating the possible role of the brain morphology as an endophenotype for future genetic studies in schizophrenia.     

2型糖尿病患者海马体脑功能连接的磁共振成像研究

目的:应用功能磁共振成像技术研究2 型糖尿病（T2D）患者静息状态下海马功能连接的变化。方法:采集27 例 T2D患者和32例正常人的脑功能磁共振成像信号,选择海马的4个分区作为种子点计算功能连接,对海马与其他脑区的 功能连接强度与临床指标进行相关性分析。结果:与正常人相比,T2D患者的右前海马与梭状回、枕中回之间的功能连接 减弱;左前海马与梭状回功能连接增强;右后海马与距状回功能连接减弱;左后海马与舌回功能连接增强。结论:T2D患 者相关脑区功能连接的改变可能与视觉相关认知功能的损伤有关,这为理解T2D患者海马体的功能提供线索。     

Fasting plasma insulin and the default mode network in women at risk for Alzheimer's disease

Brain imaging studies in Alzheimer's disease research have demonstrated structural and functional perturbations in the hippocampus and default mode network (DMN). Additional evidence suggests risk for pathological brain aging in association with insulin resistance (IR). This study piloted investigation of associations of IR with DMN-hippocampal functional connectivity among postmenopausal women at risk for Alzheimer's disease. Twenty middle-aged women underwent resting state functional magnetic resonance imaging. Subjects were dichotomized relative to fasting plasma insulin levels (i.e., > 8 μIU/mL [n = 10] and < 8 μIU/mL [n = 10]), and functional connectivity analysis contrasted their respective blood oxygen level-dependent signal correlation between DMN and hippocampal regions. Higher-insulin women had significantly reduced positive associations between the medial prefrontal cortex and bilateral parahippocampal regions extending to the right hippocampus, and conversely, between the left and right hippocampus and medial prefrontal cortex. Neuropsychological data (all within normal ranges) also showed significant differences with respect to executive functioning and global intelligence. The results provide further evidence of deleterious effects of IR on the hippocampus and cognition. Further imaging studies of the IR-related perturbations in DMN-hippocampal functional connectivity are needed.     

Preservation of hippocampal volume throughout adulthood in healthy men and women

To address controversies regarding the effect of age on the hippocampus, volumes of hippocampus and a comparison structure, temporal cortex, were measured on magnetic resonance imaging (MRI) in 84 healthy men and 44 healthy women (20-85 years). Neither men nor women showed significant correlations between hippocampal volumes and age, despite significant age-related decline in temporal volumes. Absence of hippocampus age relationships endured when restricting analyses to older individuals (> or =50 years) and considering menopause and hormone replacement therapy.     

Association between major depressive symptoms in heart failure and impaired regional cerebral blood flow in the medial temporal region: a study using 99m Tc-HMPAO single photon emission computerized tomography (SPECT)

BACKGROUND AND PURPOSE: Depressive symptoms are frequently associated with heart failure (HF), but the brain mechanisms underlying such association are unclear. We hypothesized that the presence of major depressive disorder (MDD) emerging after the onset of HF would be associated with regional cerebral blood flow (rCBF) abnormalities in medial temporal regions previously implicated in primary MDD, namely the hippocampus and parahippocampal gyrus. METHOD: Using 99mTc-SPECT, we measured rCBF in 17 elderly MDD-HF patients, 17 non-depressed HF patients, and 18 healthy controls, matched for demographic variables. Group differences were investigated with Statistical Parametric Mapping. RESULTS: Significant rCBF reductions in MDD-HF patients relative to both non-depressed HF patients and healthy controls were detected in the left anterior parahippocampal gyrus and hippocampus (ANOVA, p=0.008 corrected for multiple comparisons) and the right posterior hippocampus and parahippocampal gyrus (p=0.005 corrected). In the overall HF group, there was a negative correlation between the severity of depressive symptoms and rCBF in the right posterior hippocampal/parahippocampal region (p=0.045 corrected). CONCLUSIONS: These findings are consistent with the notion that the medial temporal region is vulnerable to brain perfusion deficits associated with HF, and provide evidence that such functional deficits may be specifically implicated in the pathophysiology of MDD associated with HF.     

The relationship between the Wechsler Memory Scale-Revised scores and whole-brain structure in patients with schizophrenia and healthy individuals

Introduction: The Wechsler Memory Scale (WMS) is a standardised battery for assessing memory functions. We aimed to investigate the relationship between all WMS scores, including subtests, and whole-brain structure in a relatively large sample.

Methods: Participants were 93 patients with schizophrenia and 117 healthy individuals, all right-handed and of Japanese ethnicity, and matched for age and sex. Their memory functions were assessed using the WMS-Revised (WMS-R). Their grey and white matter structure was analyzed using voxel-based morphometry and diffusion tensor imaging.

Results: Verbal memory score correlated positively with volumes of the left parahippocampal gyrus and hippocampus, while general memory score correlated positively with volumes of the left parahippocampal and fusiform gyri and hippocampus (p?<?0.05, corrected), while there was no correlation with white matter fractional anisotropy values in healthy individuals. No correlation was observed between any WMS-R score and grey or white matter structure in patients.

Conclusions: Using whole-brain structural magnetic resonance imaging, we found several significant correlations between WMS-R scores and grey matter volume in the brains of healthy individuals, while no correlation was found in those of patients with schizophrenia.     

Laterality effect on emotional faces processing: ALE meta-analysis of evidence

Recognizing emotion from facial expressions draws on diverse psychological processes implemented in a large array of neural structures. Two major theories of cerebral lateralization of emotional perception have been proposed: (i) the Right-Hemisphere Hypothesis (RHH) and (ii) the Valence-Specific Hypothesis (VSH). To test these lateralization models we conducted a large voxel-based meta-analysis of current functional magnetic resonance imaging (fMRI) studies employing emotional faces paradigms in healthy volunteers. Two independent researchers conducted separate comprehensive PUBMED (1990–May 2008) searches to find all functional magnetic resonance imaging studies using a variant of the emotional faces paradigm in healthy subjects. Out of the 551 originally identified studies, 105 studies met inclusion criteria. The overall database consisted of 1785 brain coordinates which yield an overall sample of 1600 healthy subjects. We found no support for the hypothesis of overall right-lateralization of emotional processing. Conversely, across all emotional conditions the parahippocampal gyrus and amygdala, fusiform gyrus, lingual gyrus, precuneus, inferior and middle occipital gyrus, posterior cingulated, middle temporal gyrus, inferior frontal and superior frontal gyri were activated bilaterally (p = 0.001). There was a valence-specific lateralization of brain response during negative emotions processing in the left amygdala (p = 0.001). Significant interactions between the approach and avoidance dimensions and prefrontal response were observed (p = 0.001).     

Abnormalities of the left temporal lobe and thought disorder in schizophrenia. A quantitative magnetic resonance imaging study.

BACKGROUND. Data from postmortem, CT, and magnetic resonance imaging (MRI) studies indicate that patients with schizophrenia may have anatomical abnormalities of the left temporal lobe, but it is unclear whether these abnormalities are related to the thought disorder characteristic of schizophrenia. METHODS. We used new MRI neuroimaging techniques to derive (without knowledge of the diagnosis) volume measurements and three-dimensional reconstructions of temporal-lobe structures in vivo in 15 right-handed men with chronic schizophrenia and 15 matched controls. RESULTS. As compared with the controls, the patients had significant reductions in the volume of gray matter in the left anterior hippocampus-amygdala (by 19 percent [95 percent confidence interval, 3 to 36 percent]), the left parahippocampal gyrus (by 13 percent [95 percent confidence interval, 3 to 23 percent], vs. 8 percent on the right), and the left superior temporal gyrus (by 15 percent [95 percent confidence interval, 5 to 25 percent]). The volume of the left posterior superior temporal gyrus correlated with the score on the thought-disorder index in the 13 patients evaluated (r = -0.81, P = 0.001). None of these regional volume decreases was accompanied by a decrease in the volume of the overall brain or temporal lobe. The volume of gray matter in a control region (the superior frontal gyrus) was essentially the same in the patients and controls. CONCLUSIONS. Schizophrenia involves localized reductions in the gray matter of the left temporal lobe. The degree of thought disorder is related to the size of the reduction in volume of the left posterior superior temporal gyrus.     

Distribution of CCK binding sites in the human hippocampal formation and their alteration in schizophrenia: a post-mortem autoradiographic study.

The distribution of cholecystokinin binding sites has been visualized and quantified by quantitative autoradiography in the human hippocampus from post-mortem brains of 11 controls and 11 schizophrenics. CCK receptors were localized to subiculum and parahippocampal gyrus. In the cortical areas there was a particularly dense lamination of receptors. In the schizophrenic material a similar overall pattern was seen, but there were significant losses of receptors in CA1 subiculum and cortex. These findings confirm the distribution of CCK receptors in the retrohippocampal areas in man and also provide further support for earlier homogenate studies which have also shown a loss of CCK binding sites in schizophrenia. This effect was localized primarily to parahippocampal gyrus suggesting that CCK plays some role in the genesis of developmental abnormalities in this region.     

Distribution of neuropeptides in the limbic system of the rat: the hippocampus

The distribution of several neuropeptides (vasoactive intestinal polypeptide, cholecystokinin octapeptide, substance P, neurotensin, methionine-enkephalin and somatostatin) in the hippocampal formation has been studied with immunocytochemical techniques. Numerous vasoactive intestinal polypeptide, cholecystokinin-octapeptide and somatostatin-positive cell bodies were found within the hippocampus and subiculum. Neurotensin-positive cell bodies were found within the subiculum, but no substance P or methionine-enkephalin-containing cell bodies were seen in either hippocampus proper or subiculum. Vasoactive intestinal polypeptide and cholecystokinin-octapeptide-positive cell bodies were predominantly located in the stratum moleculare and stratum radiatum of CA 1-2 regions and dentate gyrus, whilst somatostatin-containing cell bodies were found mainly in the stratum oriens. Nerve fibres containing each of the six peptides were found within the hippocampus. Large numbers of vasoactive intestinal polypeptide, cholecystokinin-octapeptide and somatostatin fibres innervated the pyramidal and granule cell layers, with smaller numbers in the stratum radiatum and fewer still in the stratum moleculare and stratum oriens. Other than a moderately dense neurotensin-positive fibre plexus in the dorsal subiculum, fewer neurotensin, substance P and methionine-enkephalin fibres were present. However, when present, these three peptides had a distribution restricted to a region close to the pyramidal layer in the CA 2/3 region and to the stratum moleculare of the CA 1 region. Peptide-containing fibres were identified entering or leaving the hippocampus in three ways, via (i) the fornix (all six peptides), (ii) the dorsal subiculum (neurotensin-positive fibres projecting to the cingulate cortex: somatostatin, vasoactive intestinal polypeptide, and cholecystokinin-octapeptide present in fibres running between the dorsal subiculum and occipito-parietal cortex) and (iii) the ventral subiculum (vasoactive intestinal polypeptide, cholecystokinin-octapeptide and somatostatin in fibres running between entorhinal cortex and hippocampus, and all six peptides in fibres crossing the amygdalo-hippocampal border). These findings indicate a major distinction between those peptides (vasoactive intestinal polypeptide, cholecystokinin-octapeptide, somatostatin, neurotensin) which are found in cell bodies intrinsic to the hippocampal formation and those peptides (substance P, methionine-enkephalin) which are found only in hippocampal afferents.(ABSTRACT TRUNCATED AT 400 WORDS)     

Atrophy in the parahippocampal gyrus as an early biomarker of Alzheimer��s disease

The main aim of the present study was to compare volume differences in the hippocampus and parahippocampal gyrus as biomarkers of Alzheimer's disease (AD). Based on the previous findings, we hypothesized that there would be significant volume differences between cases of healthy aging, amnestic mild cognitive impairment (aMCI), and mild AD. Furthermore, we hypothesized that there would be larger volume differences in the parahippocampal gyrus than in the hippocampus. In addition, we investigated differences between the anterior, middle, and posterior parts of both structures. We studied three groups of participants: 18 healthy participants without memory decline, 18 patients with aMCI, and 18 patients with mild AD. 3 T T1-weighted MRI scans were acquired and gray matter volumes of the anterior, middle, and posterior parts of both the hippocampus and parahippocampal gyrus were measured using a manual tracing approach. Volumes of both the hippocampus and parahippocampal gyrus were significantly different between the groups in the following order: healthy>aMCI>AD. Volume differences between the groups were relatively larger in the parahippocampal gyrus than in the hippocampus, in particular, when we compared healthy with aMCI. No substantial differences were found between the anterior, middle, and posterior parts of both structures. Our results suggest that parahippocampal volume discriminates better than hippocampal volume between cases of healthy aging, aMCI, and mild AD, in particular, in the early phase of the disease. The present results stress the importance of parahippocampal atrophy as an early biomarker of AD.     

Effect of temporal lobe structure volume on memory in elderly depressed patients

Objective

To compare the volume of the hippocampus and parahippocampal gyrus in elderly individuals with and without depressive disorders, and to determine whether the volumes of these regions correlate with scores on memory tests.

Method

Clinical and demographic differences, as well as differences in regional gray matter volumes, were assessed in 48 elderly patients with depressive disorders and 31 control subjects. Brain (structural MRI) scans were processed using statistical parametric mapping and voxel-based morphometry. Cognitive tests were administered to subjects in both groups.

Results

There were no between-group gray matter volume differences in the hippocampus or parahippocampal gyrus. In the elderly depressed group only, the volume of the left parahippocampal gyrus correlated with scores on the delayed naming portion of the visual-verbal learning test. There were also significant direct correlations in depressed subjects between the volumes of the left hippocampus, right and left parahippocampal gyrus and immediate recall scores on verbal episodic memory tests and visual learning tests. In the control group, there were direct correlations only between overall cognitive performance (as assessed with the MMSE) and the volume of right hippocampus, and between the total score on the visual-verbal learning test and the volume of the right and left parahippocampal gyrus.

Conclusions

These findings highlight different patterns of relationship between cognitive performance and volumes of medial temporal structures in depressed individuals and healthy elderly subjects. The direct correlation between delayed visual-verbal memory recall scores with left parahippocampal volumes specifically in elderly depressed individuals provides support to the view that depression in elderly populations may be a risk factor for dementia.     

功能磁共振在研究老年抑郁症脑功能及结构中的应用

目的:基于核磁共振下高分辨率结构成像及功能扫描评价老年抑郁症患者的脑功能及结构。方法:选取2015年1月至2016年7月于我院就诊的抑郁症老年患者50例为研究组,招募50例老年健康者为对照组,所有入组者均进行核磁共振下高分辨率结构像及功能扫描,采用MRIcro软件转化,应用REST、SPM 8软件进行图像结果分析。结果:抑郁组的汉密尔顿抑郁量表(HAMD)评分为(26.7±1.6)分,对照组为(4.7±1.7)分,两组比较差异具有统计学意义(t=66.636,P0.001)。抑郁组患者中左侧的梭状回、颞上回、楔前叶、顶下小叶、中央后回、中央旁小叶,以及右侧的枕中回、顶下小叶、额上回的ReHo较健康组显著减低(P0.05),抑郁组患者的左侧颞下回、额下回、楔前叶,右侧内侧扣带回、梭状回、颞下回、楔叶的GMV较健康组显著减低(P0.05)。结论:老年抑郁症患者中脑功能与结构异常的区域没有重叠,异常的脑功能与结构对于抑郁症神经病理机制的作用相互独立。     

Selective cytotoxic lesions of the retrohippocampal region produce a mild deficit in social recognition memory

Although a number of studies have implicated the hippocampal formation in social recognition memory in the rat, a recent study in this laboratory has demonstrated that selective cytotoxic lesions, confined to the hippocampus proper (encompassing the four CA subfields and the dentate gyrus), are without effect on this behaviour. This finding suggests that the hippocampus proper does not subserve social recognition memory in the rat, but does not preclude the possibility that other areas of the hippocampal formation, such as the entorhinal cortex or subiculum, could support this form of learning. The present study addressed this issue by examining the effects of selective cytotoxic retrohippocampal (RHR) lesions (including both the entorhinal cortex and subiculum) on social recognition memory in the rat. RHR lesions produced a mild social recognition memory impairment, although lesioned animals still displayed a reduction in investigation time between the first and second exposure to the juvenile. This result is consistent with other studies which have implicated the retrohippocampal or parahippocampal area in olfactory recognition memory processes. It also suggests, however, that other areas, out with the retrohippocampal region, are also likely to play an important role in social recognition memory.     

Increased fMRI responses during encoding in mild cognitive impairment

Structural and functional magnetic resonance imaging (fMRI) was performed on 21 healthy elderly controls, 14 subjects with mild cognitive impairment (MCI) and 15 patients with mild Alzheimer's disease (AD) to investigate changes in fMRI activation in relation to underlying structural atrophy. The fMRI paradigm consisted of associative encoding of novel picture-word pairs. Structural analysis of the brain was performed using voxel-based morphometry (VBM) and hippocampal volumetry. Compared to controls, the MCI subjects exhibited increased fMRI responses in the posterior hippocampal, parahippocampal and fusiform regions, while VBM revealed more atrophy in MCI in the anterior parts of the left hippocampus. Furthermore, the hippocampal volume and parahippocampal activation were negatively correlated in MCI, but not in controls or in AD. We suggest that the increased fMRI activation in MCI in the posterior medial temporal and closely connected fusiform regions is compensatory due to the incipient atrophy in the anterior medial temporal lobe.