All About Multiple Sclerosis

More MS news articles for May 2004

Neuroimaging and Diagnosis Highlights

May, 2004
Rohit Bakshi, MD

Neuroimaging was featured at many of the scientific sessions at the Annual Meeting of the American Academy of Neurology (AAN). These included key presentations on the use of emerging imaging technologies, such as functional MRI, spectroscopy, and positron emission tomography. Following are the highlights of the neuroimaging presentations given at the meeting.

Multiple Sclerosis

Magnetic resonance imaging (MRI) can localize the sites of cortical activation associated with cognitive, motor, and sensory tasks. The most commonly used method detects regional tissue changes in venous oxygenation by blood oxygen-level-dependent (BOLD) imaging, also known as functional MRI (fMRI). This technique has opened a new window for defining the circuitry and activation sites in a variety of normal and disease states, and is a powerful tool to noninvasively map the human brain.

Cognitive impairment affects approximately half of patients with multiple sclerosis (MS) and significantly contributes to a lower quality of life. Treatments for this cognitive impairment are incomplete, raising the need to develop a better understanding of its pathogenesis. Ranjeva and colleagues,[1] from Marseille, France, used fMRI to study cognitive ability and cortical function in patients at the earliest stage of MS. Neuropsychologic function was assessed by the Paced Auditory Serial Addition Test (PASAT), a clinically relevant measure of attentional memory impairment, which is commonly impaired in the disease. The fMRI experiment with PASAT as the paradigm was performed in 18 patients with the first attack of demyelination, also known as a clinically isolated syndrome suggestive of MS. Eighteen healthy subjects served as controls. In addition to investigating cortical function, the level of tissue damage was assessed by magnetization transfer ratio (MTR) imaging of the normal-appearing white matter and gray matter.

Patients with MS had lower PASAT scores than controls (P = .004). In the MS group, the PASAT scores correlated with MTR in normal-appearing white matter (r = .657, P = .0068) but not in gray matter. Abnormal scores were seen in half of the MS patients. A normal score on the PASAT was associated with larger activations in bilateral prefrontal areas vs controls. These larger activations were related to the extent of damage in normal-appearing white matter and gray matter assessed by MTR (all r about -.50, P < .05).

Several interesting findings in this study are worthy of comment. First, this study shows that cognitive dysfunction begins very early in the disease course and is present commonly at the time of the first attack of MS. Second, the fMRI component of this study shows that cortical reorganization begins at this early stage and likely serves to provide compensatory activation for the performance of tasks requiring working memory and attention. Finally, the data suggest that MTR is a sensitive technique to detect the early occurrence of occult diffuse tissue damage, which is relevant for the development of cognitive dysfunction. Further studies are necessary to determine the predictive value of fMRI or MTR derangements in this early stage toward long-term clinical outcome as well as the unique variance in cognitive function predicted by these measures vs other MRI surrogates, such as brain atrophy or spectroscopic abnormalities.

New Therapeutic Targets

Current therapies for MS focus on the inflammatory aspect of the disease process and, in particular, are aimed at limiting immune-mediated damage. Unfortunately, these treatments are only partially effective, possibly because secondary, non-immune-related mediators of damage play a role in the pathophysiology of MS. Such secondary factors occurring in the central nervous system provide potential new targets for therapies. Excitatory amino acids have been implicated as one such secondary factor contributing to neurotoxicity and other tissue damage in both human and animal models of MS.

Pelletier and associates,[2] from San Francisco and Menlo Park, California, measured the levels of the amino acids glutamate and glutamine in addition to other metabolites -- myo-inositol and N-acetyl-aspartate (NAA) -- with magnetic resonance spectroscopy (MRS) in patients with MS. Glutamate is a key amino acid implicated in the excitotoxicity paradigm thought to affect a host of neurologic disorders. The study was performed with high-field (3T) MRI in 17 patients with MS and 10 healthy controls. One of the unique and robust features of this study was the use of a new MRS pulse sequence developed by the investigators. The pulse sequence was designed to isolate the glutamate peak with spectral data in 64 increments of 2.5 ms starting at TE = 35 ms (TR = 2000 ms). Another key aspect to the study was the absolute quantification of peaks with phantoms and correction for T1 effects. MRS data were obtained from normal-appearing white matter and gray matter regions, in the periphery of gadolinium (Gad)-enhancing lesions or chronic T1 hypointense lesions. Gad lesions had significantly elevated glutamate as compared with normal-appearing white matter and chronic lesions. By contrast, myo-inositol was elevated in Gad lesions, chronic lesions, and normal-appearing white matter as compared with control white matter.

These data indicated that MRS with 3T MRI can noninvasively detect excitatory amino acid increases in acute white matter lesions of patients with MS, lending further support to the hypothesis of excitotoxicity in the pathophysiology of MS. Glial activation, suggested by elevated myo-inositol is excessive in widespread brain areas in patients with MS. Further studies are warranted to examine the longitudinal relationship between glutamate increases and disease progression and treatment responses.


Cerebral embolism is a common and potentially preventable cause of stroke. The mechanisms contributing to brain embolism continue to unfold. The transesophageal echocardiogram (TEE) is a minimally invasive sonographic imaging tool that allows visualization of the heart and aorta with a particular relevance to uncovering embolic sources. However, TEE is often not performed in patients with an acute stroke syndrome that is suggestive of a small-vessel-disease mechanism, such as a lacunar or small subcortical infarction.

Rabinstein and colleagues,[3] from Miami, Florida, performed a retrospective study of 214 patients with acute stroke who underwent TEE. Common among these patients was that TEE was ordered after they had undergone nonrevealing carotid ultrasonography and transthoracic echocardiography. A relevant TEE abnormality was defined by the presence of any of the following: (1) left atrium or left atrial appendage thrombus; (2) large (> 4 mm), ulcerated, or mobile aortic plaque; (3) patent foramen ovale with or without atrial septal aneurysm; and (4) left-sided cardiac valve vegetations.

TEE demonstrated a surprisingly high yield for relevant abnormalities, which were present in 81 patients (38%). Patients with these abnormalities were significantly older than those without abnormalities, but did not differ in terms of comorbid conditions. The most common lesions were patent foramen ovale (22%) and aortic plaques (19%). Most importantly, a confirmed, acute stroke syndrome thought to be related to small-vessel disease did not predict a negative TEE on univariate or multivariate analysis. After adjusting for a host of covariates, such as age and traditional vascular risk factors, the odds ratio of having a negative TEE for a patient with lacunar syndrome was 1.08 (P = .93); the odds ratio for a patient with a small subcortical stroke was 1.18 (P = .55); and the odds ratio for a patient with both stroke subtypes was 2.96 (P = .07).

This study indicates that the stroke subtype defined by clinical and neuroimaging correlation in the acute stage is not reliable for excluding the presence of cardioaortic sources of embolism detectable by TEE. The early identification of such sources is likely important for the secondary prevention of subsequent strokes. Further studies are warranted to prospectively and longitudinally test the hypotheses of the high yield and clinical relevance of these TEE abnormalities in these subgroups of stroke patients.


The term "dolichoectasia" is derived from the Greek words dolichos (long) and ectasic (distended). The disorder is defined as either elongation or distention of the arterial system, which often occur together. The most commonly affected site is the vertebrobasilar system. The basilar artery is considered elongated if it extends lateral to the clivus or dorsum sellae or if it bifurcates superior to the suprasellar cistern;[4] ectasia of the basilar artery is defined as a diameter greater than 4.5 mm.[4] Vertebrobasilar dolichoectasia may lead to neurologic symptoms by a variety of mechanisms, including vertebrobasilar insufficiency, direct compression of neural structures, hydrocephalus, dissection, and hemorrhage. The most common cranial neuropathies related to vertebrobasilar dolichoectasia are hemifacial spasm and trigeminal neuralgia. The etiology of the disease is disputed and may include atherosclerosis and inherited factors. Predisposing traits include male sex, hypertension, aging, and smoking.

Michel and colleagues,[5] from Utrecht, The Netherlands; Boston, Massachusetts; Lugano, Switzerland; and Lausanne, Switzerland, reviewed the medical records of 5 major hospitals over a 5-year period and found 6 patients with subarachnoid hemorrhage (SAH) caused by rupture of a dolichoectatic basilar artery. The median age was 70 (range, 57-81 years). Five of the 6 patients had been previously symptomatic from brainstem compression or ischemia. For the sixth patient, SAH was the first symptom. Symptoms were rapidly progressive preceding the rupture. Chronic hypertension and ischemic white matter disease were present in all patients. This study shows that vertebrobasilar dolichoectasia may lead to fatal SAH. Patients typically have stigmata of vascular disease and symptoms of aneurysm growth or clot formation before SAH. These findings raise the need to develop a better understanding of the pathophysiology of dolichoectasia and how the disease can be prevented or treated.


Arteriovenous malformations (AVMs) are congenital anomalies of blood vessel development, characterized by persistent communications between arterial and venous channels without intermediary capillaries. The most common form of cerebral AVMs occurs in the parenchyma pathologically constituting a tangled cluster of tortuous and dilated blood vessels with dysplastic brain parenchyma, calcification, and chronic hemorrhage. Unruptured AVMs appear on MRI studies as a nidus of flow voids (fast-flowing vessels) with no associated mass effect. The surrounding brain tissue is usually normal in signal intensity.[4] On post-Gad images, enhancement of the slow-flowing veins in the nidus or peripheral draining veins may be seen.[4]

Stapf and associates,[6] from New York, New York, assessed demographic and structural risk factors for seizures in 634 consecutive patients with AVMs. They stratified the event leading to the diagnosis of AVM as epileptic (any focal or generalized seizure activity) or nonepileptic (hemorrhage, neurologic sign/symptom, or asymptomatic/other). Statistical models tested the predictive value of demographic (age and sex) and structural factors (ie, AVM nidus size, brain location, and venous-drainage pattern) on the risk of epilepsy. In total, 182 (29%) patients with AVMs presented with seizures. Multivariate analysis indicated that independent predictors for seizures included AVM size; superficial venous drainage; and frontal, temporal, parietal, and borderzone locations, whereas female sex had a protective effect. Superficial venous drainage, the temporal lobe, and borderzone location were the strongest predictors for seizures. This study provides valuable information from a large, well-characterized patient series on the factors associated with epilepsy in patients with brain AVMs. Such findings may assist in clinical decision making and risk stratification.

Parkinson's Disease

Parkinsonism is a syndrome defined clinically by an acquired, bradykinetic rigid state. Neurodegenerative causes are divided into 2 major categories: idiopathic Parkinson's disease and Parkinson's plus syndromes. Parkinson's plus syndromes include olivopontocerebellar atrophy, striatonigral degeneration/multiple system atrophy, Shy-Drager's syndrome, progressive supranuclear palsy, and corticobasal ganglionic degeneration.

Eckert and colleagues,[7] from Magdeburg, Germany, and New York, New York, used fluorodeoxyglucose positron emission tomography (FDG PET) of the brain to differentiate among patients with Parkinson's disease, multiple system atrophy, progressive supranuclear palsy, corticobasal ganglionic degeneration, and healthy subjects. The investigators studied 422 patients during a 6-year period. Most patients did not exhibit clinical signs for a firm diagnosis at the time of FDG PET. For 150 of the patients, a likely diagnosis was reached by an independent, blind review by 2 movement disorder specialists (average follow-up time, 1.2 years).

Using case-based, statistical parametric mapping analysis of PET images, a correct diagnosis was obtained in 95% of Parkinson's disease patients, 96% of multiple system atrophy patients, 83% of progressive supranuclear palsy patients, 83% of corticobasal ganglionic degeneration patients, and 87% healthy subjects. Visual analysis of the images correctly classified 92% of Parkinson's disease patients, 76% of multiple system atrophy patients, 56% of progressive supranuclear palsy patients, 83% of corticobasal ganglionic degeneration patients, and 91% of healthy subjects. A correct imaging diagnosis was obtained in 91% of all subjects with statistical parametric mapping and in 84% by visual PET analysis. This study shows that imaging with FDG PET can be used for the differentiation among causes of parkinsonism. Automated computer analysis may be helpful for situations in which trained readers are not readily available.


  1. Ranjeva JP, Audoin B, Van Au Duong M, et al. MR study of the influence of tissue damage and cortical reorganization on PASAT performance at the earliest stage of multiple sclerosis. Program and abstracts of the 56th Annual Meeting of the American Academy of Neurology; April 24-May 1, 2004; San Francisco, California. Abstract [S10.001].
  2. Pelletier D, Srinivasan R, Sailasuta N, Nelson SJ, Hurd RE. Excessive levels of glutamate and glial activity in multiple sclerosis measured by high field MR spectroscopy. Program and abstracts of the 56th Annual Meeting of the American Academy of Neurology; April 24-May 1, 2004; San Francisco, California. Abstract [S46.006].
  3. Rabinstein AA, Chirinos J, Fernandez FR, Zambrano JP. Is TEE useful in patients with small subcortical strokes? Program and abstracts of the 56th Annual Meeting of the American Academy of Neurology; April 24-May 1, 2004; San Francisco, California. Abstract [S16.005].
  4. Bakshi R, Ketonen L. MRI of the brain in clinical neurology. In: Joynt RJ, Griggs RC, eds. Baker and Joynt's Clinical Neurology on CD-ROM. Philadelphia, Pa: Lippincott, Williams & Wilkins; 2004.
  5. Michel P, Rinkel GJ, Thaler D, Sztajzel R, Staedler C, Bogousslavsky J. Subarachnoid haemorrhage due to ruptured dolichoectatic intracranial arteries: 6 cases. Program and abstracts of the 56th Annual Meeting of the American Academy of Neurology; April 24-May 1, 2004; San Francisco, California. Abstract [S13.004].
  6. Stapf C, Labovitz DL, Sciacca RR, Mast H, Mohr JP. Risk determinants for seizures in patients presenting with brain arteriovenous malformation. Program and abstracts of the 56th Annual Meeting of the American Academy of Neurology; April 24-May 1, 2004; San Francisco, California. Abstract [S19.003].
  7. Eckert T, Barnes A, Frucht S, Dhawan V, Gordon MF, Eidelberg D. The usefulness of FDG-PET in the differential diagnosis of Parkinsonian disorders -- an evidence based medicine approach. Program and abstracts of the 56th Annual Meeting of the American Academy of Neurology; April 24-May 1, 2004; San Francisco, California. Abstract [P06.137].

Copyright © 2004, Medscape