Mammillary Body Volumes Research Articles

Mammillary bodies and fornix fibers are injured in heart failure

Cognitive abnormalities, including memory deficits, are common in heart failure (HF). Brain structures, including the hippocampus, fornix, and thalamus participate in memory processing, and most show structural injury and functional deficits in HF. The mammillary bodies and fornix play essential roles in spatial and working memory processing, interact with other structures, and may also be injured in HF. We assessed mammillary body volumes and cross-sectional fornix areas in 17 HF and 50 control subjects using high-resolution T1-weighted magnetic resonance images. Mammillary body volumes and fornix cross-sectional areas were significantly reduced bilaterally in HF, and these differences remained after controlling age, gender, and intracranial volume. Mammillary body and fornix injury may contribute to the compromised spatial and working memory deficits in HF. Pathological processes eliciting the damage may include injury accompanying hypoxic/ischemic processes in pathologic HF perfusion and breathing, and thiamine deficiency accompanying diuretic use and nutritional mal-absorption in the condition.

A disproportionate role for the fornix and mammillary bodies in recall versus recognition memory

Uncovering the functional relationship between temporal lobe amnesia and diencephalic amnesia depends on determining the role of the fornix, the major interlinking fiber tract. In this study relating fornix volume with memory, we made magnetic resonance imaging-based volume estimates of 13 brain structures in 38 individuals with surgically removed colloid cysts. Fornix status was assessed directly by overall volume and indirectly by mammillary body volume (which atrophies after fornix damage). Mammillary body volume significantly correlated with 13 out of 14 tests of episodic memory recall, but correlated poorly with recognition memory. Furthermore, as the volumes of the left fornix and the left mammillary bodies decreased, the difference between recall and recognition scores increased. No other structure was consistently associated with memory. These findings support models of diencephalic memory mechanisms that require hippocampal inputs for recall, but not for key elements of recognition.

Reduced mammillary body volume in patients with obstructive sleep apnea

Obstructive sleep apnea (OSA) patients show compromised emotional and cognitive functions, including anterograde memory deficits. While some memory inadequacies in OSA may result from earlier-described structural deficits in the hippocampus, mammillary body injury also could contribute, since these structures receive projections from the hippocampus via the fornix, project heavily to the anterior thalamus, and have been implicated in other conditions with memory deficiencies, such as Korsakoff's syndrome. However, volume loss in mammillary bodies has not been reported in OSA, likely a consequence of logistic difficulties in size assessment. We evaluated mammillary body volumes in 43 OSA (mean age+/-S.D., 46.9+/-9.2 years; mean apnea-hypopnea-index+/-S.D., 31.2+/-19.9 events/h) and 66 control subjects (age, 47.3+/-8.9 years). Two high-resolution T1-weighted image volumes were collected on a 3.0 T magnetic resonance scanner, averaged to improve signal-to-noise, and reoriented (without warping) into a common space. Brain sections containing both mammillary bodies were oversampled, and the bodies were manually traced and volumes calculated. OSA patients showed significantly reduced left, right, and combined mammillary body volumes compared with control subjects, after partitioning for age, gender, and head size (multivariate linear model, p<0.05). Left-side mammillary bodies showed greater volume reduction than the right side. Diminished mammillary body volume in OSA patients may be associated with memory and spatial orientation deficits found in the syndrome. The mechanisms contributing to the volume loss are unclear, but may relate to hypoxic/ischemic processes, possibly assisted by nutritional deficiencies in the syndrome.

78. Fornix and mammillary body volumes in individuals with schizophrenia

Bilateral volume reductions in the hippocampus and declarative memory impairment are both consistent findings among individuals with schizophrenia. The hippocampus and its primary inputs are involved in declarative memory formation. There is also evidence, though less conclusive, that the hippocampal efferent pathway and a primary synaptic target, i.e., the fornix and the mammillary bodies, play a role in learning and memory. Thus, memory deficits and associated abnormalities in various components of the medial temporal lobe system may be key characteristics of schizophrenia. MRI studies have examined hippocampal and parahippocampal volumes, but not the volumes of the fornix and the mammillary bodies in schizophrenics. Subjects included 44 individuals with DSM-IV schizophrenia and 44 normal controls for the volumetric analysis of the fornix. For a subset of the sample (20 patients and 26 normal controls), for whom the structures were clearly measurable, analysis of the mammillary bodies was also completed. Sixty contiguous 3 millimeter coronal slices were acquired on a 1.5 Tesla magnet. A new method was utilized to measure the width of the fornix and estimate the area of the mammillary bodies. Findings showed no differences in either of these measures of the fornix or mammillary bodies between schizophrenics and normal controls. We have found hippocampal and parahippocampal volume reductions in these individuals with schizophrenia (see Seidman et al., this issue). Thus, it appears that abnormalities are limited to those components of the “medial temporal lobe memory system.” Additional analyses will correlate brain volume and memory performance in these patients.

In Vivo Mammillary Body Volume Deficits in Amnesic and Nonamnesic Alcoholics

Neuropathological studies use the presence of mammillary body (MB) pathology as a cardinal, diagnostic feature of Wernicke's encephalopathy (WE) in neuropsychiatric diseases, most notably alcoholism. Although Korsakoffs Syndrome (KS), which is marked behaviorally by dense global amnesia, is a typical sequela of WE, it remains controversial whether these two conditions necessarily co-occur and whether MB pathology is therefore a diagnostic requisite for KS. We investigated these issues by examining, in vivo, 24 nonamnesic alcoholics (ALC), 5 amnesic alcoholics (KS), and 51 normal controls with three-dimensional MRI and memory testing. MB volume was determined from successive, 1 mm thick slices. The ALC group had significantly smaller MB volumes bilaterally (mean = 54.5 +/- 22.0 mm3) than controls (mean = 66.3 +/- 17.1 mm3), and the KS group had even smaller MB volumes than the ALC group (mean = 20.7 +/- 14.8 mm3). Only 2 ALC patients met historical clinical criteria for past WE, and their MB volumes were well within range of the remaining 22 ALC patients. Although all five KS patients met historical clinical criteria for WE, three KS did not have accompanying dementia and had the same degree of MB volume loss as the ALC; the remaining two KS had accompanying dementia and MB volumes half the volume of the ALC group and of KS patients without dementia. These findings provide volumetric in vivo evidence that: (1) MB volume deficits do occur in alcoholics without amnesia, although these deficits are not present in ail such alcoholics; (2) greater MB volume deficits are present in alcoholics with clinically detectable amnesia or dementia; (3) MB shrinkage is related to severity of cognitive and memory dysfunction, which suggests a continuum of MB pathology in chronic alcoholism to KS; and (4) the presence of WE in all of the KS patients and in the two ALC patients with the greatest long-term declarative memory deficit supports the possibility of an additional and unique pathology distinguishing nonamnesic and amnesic alcoholism.

MRI extrahippocampal volumes and visual memory: correlations independent of MRI hippocampal volumes in temporal lobe epilepsy patients.

Limbic system atrophy and memory dysfunction are common in patients with temporal lobe epilepsy (TLE). However, the relationship between extrahippocampal limbic structures and memory functioning within TLE has received little attention. The present study examined associations of MRI volumetric measurements of the mammillary body, fornix, amygdala, and hippocampus to measures of episodic verbal and visual memory. The Logical Memory and Visual Reproduction subtests from the Wechsler Memory Scale were administered to 47 unilateral TLE patients (25 right, 22 left). Normalized right and left MRI volumes were determined for each patient by cursor tracing 1.5 mm slices from 3D-MRI. Significant associations were found between left hippocampal volume and the immediate, delayed, and percent retention scores of the Logical Memory Test; between the left mammillary body volume and the Logical Memory Test delayed and percent retention scores; immediate Visual Reproduction performance was significantly related to the right and left amygdala volumes, and right mammillary body volume; only the right amygdala and right mammillary body volume were associated with the delayed Visual Reproduction trial. However, neither right nor left hippocampal volumes were related to visual memory performance. Multiple limbic system structural volumes were independently associated with verbal and nonverbal memory performance. Results suggest that visual memory, as measured by the Visual Reproduction Test, may be uniquely associated with extrahippocampal volumes in patients with TLE.

Mammillary Body and Cerebellar Shrinkage in Chronic Alcoholics with and without Amnesia

Mammillary body and cerebellar atrophy have been described as postmorten neuropathologic markers of Korsakoff's syndrome. This study examined whether shrinkage in the mammillary bodies and cerebellum is present consistently in amnesic chronic alcoholics during life and whether the degree of abnormality in these patients differs from that in nonamnesic alcoholic and healthy controls. The severity of shrinkage in the mammillary bodies, cerebellar hemispheres, and cerebellar vermis visualizable on MRI scans was rated on a three-point scale in 33 chronic nonamnesic alcoholics, 9 amnesic alcoholics, and 20 healthy controls. Although both alcoholic groups showed significant mammillary body and cerebellar shrinkage relative to controls, the two patient groups did not differ from each other. Furthermore, four of eight amnesic patients in our sample did not demonstrate clinically significant mammillary body atrophy. These results suggest that alcoholism is associated with mammillary body and cerebellar tissue volume loss but do not provide evidence that these markers distinguish accurately between amnesic and nonamnesic patients. In addition, they suggest that visualizable mammillary body atrophy is not necessary for the development of amnesia in alcoholic patients.