Thin-section Magnetic Resonance Imaging Research Articles

Oculomotor Palsy as the Initial Finding in Neurofibromatosis Type 2

A 15-year-old boy with oculomotor palsy in the left eye as the initial finding of neurofibromatosis type 2 (NF2) was presented. The eye motility examination revealed restriction in adduction, elevation and depression of the left eye and mild ptosis of the left eye. Cranial magnetic resonance imaging (MRI) showed a 3 cm sized mass in the cerebellar vermis that showed features of a schwannoma; bilateral lesions in the acoustic canals and bilateral optic nerve buckling: these findings were consistent with a diagnosis of NF2. Further thin-section MRI revealed a mass in the left cavernous sinus segment of the oculomotor nerve and a thickening of the right oculomotor nerve.

Magnetic resonance imaging features in two Chinese family with congenital fibrosis of extraocular muscles

The aim of this article was to investigate the structural basis of ocular motility and visual abnormalities in humans with congenital fibrosis of the extraocular muscles (CFEOM). 17 volunteers from 2 CFEOM pedigrees Clinical ophthalmic and motility examed and 18 normal control subjects were correlated with thin-sectioned magnetic resonance imaging (MRI) across the orbit and the brain-stem level. Subjects with CFEOM had severe bilateral blepharoptosis, limited supraduction, and variable ophthalmoplegia. In affected subjects, MRI demonstrated atrophy of the levator palpebrae superioris, all EOMs, and the optic nerves, and small or absent orbital motor nerves. The oculomotor nerve was most severely hypoplastic, but the abducens was also affected. Subjects with CFEOM exhibited subclinical but highly significant reduction from normal in mean optic nerve size (P < 0.05). There are also some difference between the two CFEOM pedigrees. These findings suggest that neuronal disease is primary in CFEOM, with myopathy arising secondary to abnormal innervation and the oculomotor nucleus and trochlear nucleus of the abnormalities defects.

Utility of High-Resolution MR Imaging in Demonstrating Transmural Pathologic Changes in Crohn Disease

Magnetic resonance (MR) imaging has emerged as an imaging modality that can be used to help diagnose and evaluate Crohn disease of the small and large bowel. MR imaging has high diagnostic accuracy in the detection of Crohn disease, and high-resolution thin-section MR images can demonstrate transmural pathologic changes of Crohn disease from the level of the mucosa to that of the mesentery. High-resolution MR image data also may be used to construct high-quality multiplanar and endoluminal views that may provide additional diagnostic information. Knowledge of the MR imaging findings of Crohn disease and how they correlate with the pathologic features of the disease is important to facilitate accurate diagnosis and detect complications.

Die Rolle der Magnetresonanztomographie für die Indikationsstellung zur präoperativen Therapie beim Rektumkarzinom

Traditionally, the decision to apply preoperative treatment for rectal cancer patients has been based on the T- and N-category. Recently, the radial distance of the tumor to the circumferential resection margin (CRM) has been identified as an important risk factor for local failure. By magnetic resonance imaging (MRI) this distance can be measured preoperatively with high reliability. Thus, selected groups have started to limit the indication for preoperative therapy to tumors extending to - or growing within 1 mm from - the mesorectal fascia (CRM+). Pros and cons of this selected approach for preoperative treatment and first clinical results are presented. Prerequisites are the availability of modern high-resolution thin-section MRI technology as well as strict quality control of MRI and surgical quality of total mesorectal excision (TME). By selecting patients with CRM-positive tumors on MRI for preoperative therapy, only approximately 35% patients will require preoperative radiotherapy (RT) or radiochemotherapy (RCT). However, with histopathologic work-up of the resected specimen after primary surgery, the indication for postoperative RCT is given for a rather large percentage of patients, i.e., for pCRM+ (5-10%), intramesorectal or intramural excision (30-40%), pN+ (30-40%). Postoperative RCT, however, is significantly less effective and more toxic than preoperative RCT. A further point of concern is the assertion that patients, in whom a CRM-negative status is achieved by surgery alone, do not benefit from additional RT. Data of the Dutch TME trial and the British MRC (Medical Research Council) CR07 trial, however, suggest the reverse. To omit preoperative RT/RCT for CRM-negative tumors on MRI needs to be further investigated in prospective clinical trials. The German guidelines for the treatment of colorectal cancer 2008 continue to indicate preoperative RT/RCT based on the T- and N-category.

Congenital Monocular Elevation Deficiency

The pathophysiology of monocular elevation deficiency is poorly understood. The goal of this study was to determine the appearance of the extraocular muscles and the oculomotor nerve. Observational case series. Six patients with monocular elevation deficiency. Ophthalmologic examination and thin-sectioned magnetic resonance imaging (MRI) at the brainstem level as well as across the orbit in 6 patients were performed. Ocular alignment and movement, extraocular muscles, and the oculomotor nerve on MRI. One out of 6 patients with monocular elevation deficiency showed focal thickening of the inferior rectus muscle near the orbital apex. The 5 remaining patients showed normal extraocular muscles and the oculomotor nerves on MRI. Focal thickening of the inferior rectus muscle may partially explain the cause of restricted gaze. In addition, the finding of normal oculomotor nerves might support an underlying deficit in the unilateral center for upgaze as the etiology of monocular elevation deficiency.

Commentary: Rectal Cancer—An Evolution of Treatment

The treatment of rectal cancer has evolved substantially over the past two decades. In the past, rectal cancer was frequently treated with surgery alone, which resulted in high rates of local failure, with significant patient morbidity and mortality. Sentinel trials in the 1980s to early 1990s demonstrated that adjuvant chemoradiotherapy resulted in lower rates of local failure and superior survival versus resection alone [1, 2]. This observation led to the adoption of adjuvant chemoradiotherapy as standard treatment for patients with stage II–III disease. Subsequently, proponents of total mesorectal excision (TME) reported that local regional failure rates were 10% with this approach, bringing the role of radiation therapy into question. However, a large, randomized Dutch trial comparing neoadjuvant radiation therapy alone followed by TME with TME alone showed that despite the superior surgical technique, radiation therapy resulted in a significantly lower rate of local failure. The 2-year local failure rate was 2.4% for patients undergoing preoperative radiation therapy and TME versus 8.2% for patients undergoing TME only (p .001) [3]. The sequencing of radiation therapy relative to surgery has been controversial for several decades. Advocates of preoperative therapy argued that this approach had many benefits, including higher rates of sphincter preservation and better radiation tolerance. In Germany, a large randomized trial (CAO/ARO/AIO-94) comparing preoperative with postoperative chemoradiotherapy confirmed that a preoperative approach resulted in superior treatment compliance, less acute and late toxicity, and a significantly higher rate of local regional control, again simply by altering the sequence of chemoradiotherapy in relationship to surgery [4]. This has led to a new standard of care in the U.S. and Europe in the treatment of stage II–III and selected stage IV rectal cancer patients. The primary disadvantage of preoperative chemoradiotherapy is the potential to irradiate patients with stage I (T1-2 N0) disease. When treated with TME alone, these patients have low rates of local recurrence and any benefit of radiation therapy is limited. Therefore, meticulous pretreatment staging with endoscopic ultrasound (EUS) and/or magnetic resonance imaging (MRI) is important. Despite this, there is still the potential to “upstage” patients who have early-stage disease. This is exemplified in the adjuvant arm of the German rectal trial where 18% of patients staged to have stage II or greater disease by EUS preoperatively were found to have pathologic stage I disease [4]. The integration of thin-section MRI into the pretherapy evaluation of rectal cancer patients likely represents a further refinement in staging accuracy and may potentially spare stage I patients from neoadjuvant therapy. In the hands of experienced endoscopists and radiologists, preoperative EUS and MRI appear to strongly correlate with pathologic stage. As discussed by Drs. Glynne-Jones and Harrison [5], contemporary randomized trials have demonstrated that neoadjuvant chemoradiotherapy produces superior tumor downstaging, pathologic complete response rates, and local control relative to preoperative radiation therapy alone. Therefore, it is logical to assume that, in patients with lowlying rectal cancers, sphincter preservation through downstaging may be achieved in some patients with neoadjuvant therapy, particularly with the knowledge that “historically” large distal margins may not be necessary to obtain durable

Thin-section diffusion-weighted magnetic resonance imaging of the brain with parallel imaging

Thin-section diffusion-weighted imaging (DWI) is known to improve lesion detectability, with long imaging time as a drawback. Parallel imaging (PI) is a technique that takes advantage of spatial sensitivity information inherent in an array of multiple-receiver surface coils to partially replace time-consuming spatial encoding and reduce imaging time. To prospectively evaluate a 3-mm-thin-section DWI technique combined with PI by means of qualitative and quantitative measurements. 30 patients underwent conventional echo-planar (EPI) DWI (5-mm section thickness, 1-mm intersection gap) without parallel imaging, and thin-section EPI-DWI with PI (3-mm section thickness, 0-mm intersection gap) for a b value of 1000 s/mm(2), with an imaging time of 40 and 80 s, respectively. Signal-to-noise ratio (SNR), relative signal intensity (rSI), and apparent diffusion coefficient (ADC) values were measured over a lesion-free cerebral region on both series by two radiologists. A quality score was assigned for each set of images to assess the image quality. When a brain lesion was present, contrast-to-noise ratio (CNR) and corresponding ADC were also measured. Student t-tests were used for statistical analysis. Mean SNR values of the normal brain were 33.61+/-4.35 and 32.98+/-7.19 for conventional and thin-slice DWI (P>0.05), respectively. Relative signal intensities were significantly higher on thin-section DWI (P<0.05). Mean ADCs of the brain obtained by both techniques were comparable (P>0.05). Quality scores and overall lesion CNR were found to be higher in thin-section DWI with parallel imaging. A thin-section technique combined with PI improves rSI, CNR, and image quality without compromising SNR and ADC measurements in an acceptable imaging time.

Clinical impact of magnetic resonance imaging on Gamma Knife surgery for brain metastases

Stereotactic radiosurgery is beneficial for patients with a limited number of small brain metastases. Increased numbers of brain metastases, not infrequently at unreachable locations, are identified using thin-section magnetic resonance (MR) imaging on the day of Gamma Knife surgery (GKS). To improve patient selection and design better treatment strategies, a retrospective study was conducted to determine factors that may contribute to detecting additional brain metastases on the day of GKS. A total of 100 patients with brain metastases who underwent GKS between October 2003 and May 2006 at the University of California Davis Medical Center were included in the present study. Patients were categorized by age, sex, Karnofsky Performance Scale score, status of systemic disease, histological characteristics of the primary tumor, and whether they received previous whole-brain radiotherapy (WBRT). The number of lesions identified by diagnostic MR imaging at referral, by thin-section double-contrast MR imaging on the day of GKS, and the actual lesions treated by GKS were recorded. The diagnostic MR images were categorized in terms of section thickness and time interval before GKS. The characteristics of this patient population match well with the general GKS practice. Fifty-six had been treated with WBRT. On average, patients presented with 2.2 +/- 1.7 lesions, a number based on their original diagnostic MR imaging, had 3.6 +/- 3.4 lesions identified on the thin-section treatment MR imaging (p < 0.05), and underwent treatment of 3.1 +/- 2.6 lesions on the day of GKS. Significantly, treatment was compromised in 21 patients, in whom not all additional lesions could be treated with the initial headframe placement. Analysis shows that a significantly greater number of lesions were detected using thin-section MR imaging on the day of GKS in patients who had undergone thick-section diagnostic MR imaging, did not receive WBRT, and had progressive systemic disease. To optimize treatment planning and minimize additional treatment, the number of metastases needs to be determined accurately before frame placement, ideally by performing thin-section MR imaging, as used on the day of GKS.

Nasal dermoids

Midline congenital nasal lesions are rare, occurring in one out of every 20,000-40,000 births. Of these midline lesions, nasal dermoids are the most common. This review centers on diagnosis of nasal dermoids, the role of imaging in diagnosis and surgical planning and the various approaches to surgical management of these lesions. Multiplanar, high-resolution thin section magnetic resonance imaging allows for excellent soft tissue detail, particularly when intracranial extension is expected. Open rhinoplasty is favored by many authors for excision of dermoids. Transnasal endoscopic excision of nasal dermoids has been reported but is not recommended for dermoids extending into or beyond the falx cerebri. Imaging of the midface and brain is essential for accurate diagnosis, assessment for any intracranial extension and appropriate surgical planning. Any surgical approach for removal of nasal dermoid cysts should permit adequate access, allow repair of the skull base and cerebrospinal fluid leak, facilitate nasal reconstruction and result in acceptable cosmesis. The head and neck surgeon should be able to consider various surgical approaches to manage these lesions.

Imaging of the Superior Rectus in Superior Rectus Overaction after Retrobulbar Anesthesia

Vertical diplopia after cataract surgery caused by an overaction of an extraocular muscle is more common when the superior rectus muscle is involved, whereas contracture is more common when the inferior rectus muscle is involved. However, no documented imaging has been presented. The aim of this report was to study the superior rectus in such a patient with magnetic resonance imaging (MRI). Observational case report. Ophthalmologic examination and thin-sectioned MRI across the superior rectus muscle were performed in a patient with superior rectus overaction after cataract surgery. Ocular alignment, ocular movement, and the superior rectus muscle on MRI. Magnetic resonance imaging disclosed focal thickening of the superior rectus muscle near the orbital apex in a patient who showed superior rectus overaction after retrobulbar anesthesia for cataract extraction. The focal thickening of the superior rectus muscle in this patient is consistent with the theory that segmental contracture leads to overactive muscles after retrobulbar anesthesia.

Development and Evaluation of Phantom for Verification of Section Thickness at Thin-Section MR Imaging

The purpose of this study was to construct and evaluate a phantom for measurement of magnetic resonance (MR) imaging sections with thicknesses of 0.3-0.6 mm. The phantom consisted of 304 0.1-mm-thick disks, each of which had a single notch filled with a diluted solution of gadopentetate dimeglumine. Qualitative and quantitative evaluations of T1-weighted images with section thicknesses of 0.3-0.6 mm were performed. A computer simulation was performed to evaluate the influence of different examination angles. At quantitative evaluation, the difference between nominal and measured section thicknesses ranged from 11% (nominal section thickness, 0.6 mm) to 18% (nominal section thickness, 0.4 mm). Computer simulations revealed that the phantom's positioning along the z-axis had a negligible influence at quantitative evaluation.

Hypoplastic Oculomotor Nerve and Absent Abducens Nerve in Congenital Fibrosis Syndrome and Synergistic Divergence with Magnetic Resonance Imaging

There is only one neuropathologic report of congenital fibrosis of the extraocular muscles (CFEOM), and none of synergistic divergence. The aim of this report was to study the oculomotor nerve and the abducens nerve in 2 such patients with magnetic resonance imaging. Observational case reports. Ophthalmologic examination and thin-sectioned magnetic resonance imaging across the brainstem level were performed in 2 patients with CFEOM and synergistic divergence. To confirm the accuracy of the procedure, we compared the results obtained with those of a control group of 40 individuals using the same technique. The oculomotor nerve and the abducens nerve on magnetic resonance imaging. Magnetic resonance imaging disclosed bilateral hypoplasia of the oculomotor nerve in both patients, and absence of the abducens nerve on the affected side of synergistic divergence. The oculomotor and abducens nerves were observed in 80 of 80 eyes (100%) screened as controls. In both patients with CFEOM and synergistic divergence, the oculomotor nerve was hypoplastic bilaterally, and the abducens nerve was absent on the side exhibiting synergistic divergence.

Normal Abduction in a Patient with Duplicated Abducens Nerve

To our knowledge, there has been no report of ophthalmologic findings related with a duplicated abducens nerve in the ophthalmic literature. This study reports such findings. An ophthalmologic examination was performed in one patient with a duplicated abducens nerve, revealed with thin-sectioned magnetic resonance imaging (MRI) across the brainstem level. The MRI disclosed a duplicated left abducens nerve. The patient was orthotropic in five cardinal positions, and her ductions and versions were full. One patient with a duplicated abducens nerve showed orthotropia and normal ocular movement.

Presence of the abducens nerve according to the type of Duane's retraction syndrome

To analyze whether the presence of the abducens nerve might depend upon the type of Duane's retraction syndrome (DRS). Observational case series. Sixteen patients with type 1 DRS, 2 patients with type 2 DRS, and 5 patients with type 3 DRS. Ophthalmologic examination and thin-sectioned magnetic resonance imaging (MRI) at the brain stem level were performed in 23 DRS patients. To confirm the accuracy of the procedure, we compared the results obtained with those of a control group of 30 individuals using the same technique. Examination of ductions and versions and the presence of the abducens nerve on MRI. The abducens nerve on the affected side could not be observed using MRI in 18 (100%) of 18 eyes (16 patients) with type 1 DRS and in 3 of 5 eyes with type 3 DRS. Conversely, the abducens nerve was observed in 2 of 2 eyes with type 2 DRS and 2 of 5 eyes with type 3 DRS. The abducens nerve was observed in 60 (100%) of 60 eyes screened as controls. The abducens nerve on the affected side was absent in type 1 DRS patients and some type 3 DRS patients, but present in type 2 DRS patients as well as in some type 3 DRS patients. In terms of the presence or absence of the abducens nerve, type 1 and type 2 DRS were homogenous, and type 3 DRS was heterogenous.

Lumps and Bumps on the Head in Children: Use of CT and MR Imaging in Solving the Clinical Diagnostic Dilemma

Lumps and bumps of the scalp are a common presenting complaint in children and often pose a diagnostic dilemma. These lesions can be difficult to image, with evaluation confounded by their small size. However, accuracy in diagnosis is critical because the diagnostic and therapeutic implications can vary significantly. The clinical examination can be helpful in developing the differential diagnosis and the imaging strategy. Often, however, a single imaging study is insufficient, and the radiologist finds it necessary to image with more than one modality to correctly diagnose a lesion and provide adequate information for the surgeon. Radiography and ultrasonography are often the initial screening diagnostic tests, followed by magnetic resonance (MR) imaging or computed tomography (CT) for more detail. Multidetector thin-section CT and thin-section MR imaging with surface coils are beneficial in the work-up of these small lesions of the head and neck. The use of newer MR imaging sequences such as heavily T2-weighted single-shot turbo spin-echo imaging and diffusion-weighted imaging can improve the characterization of difficult lesions. Familiarity with the variety of new imaging tools and techniques that are available can help characterize pediatric head and neck lesions and guide clinical management.

Magnetic resonance imaging of hepatocellular carcinoma

Hepatocellular carcinoma (HCC), the most common primary hepatic malignancy, usually develops in patients with cirrhosis, growing sequentially from low-grade dysplastic nodules to frank malignant HCC. Its recognition is critical because curative treatment and prognosis require early diagnosis. Survival in patients with HCC relates directly to the number, size, and extent of lesions at diagnosis. Imaging of HCC is complicated because the tumor has a varied imaging appearance and frequently coexists with other cirrhotic nodules. Magnetic resonance imaging (MRI), the best available diagnostic technique, offers good contrast resolution and diagnostic sensitivity ranging from 33% to 77%. The main difficulty is not in diagnosing large tumors, but rather small tumors (<2 cm), because of considerable overlap on imaging between benign (regenerative), borderline (dysplastic), and malignant nodules. Increasing degrees of histological malignancy are associated with increasing arterialization and loss of portal blood supply; therefore, recognition of HCC requires dynamic imaging with gadolinium-enhanced T1-weighted sequence. Typically, HCC is a focal lesion with high signal intensity on T2-weighted images, variable signal intensity on T1-weighted images, intense arterial phase enhancement after gadolinium injection, and isointensity or hypointensity at the portal venous phase. The sensitivity of MRI for detecting small lesions is low, and improvement is still needed. Newer contrast agents, higher field strength (3 Tesla) imaging, and perfusion and diffusion MRI techniques possibly will provide greater sensitivity and specificity for detecting small HCCs in the future.

Magnetic resonance imaging in Duane's retraction syndrome.

The aim of this study was to visualize the subarachnoid portion of the nervus abducens by magnetic resonance imaging and to analyze whether aplasia of the nervus abducens is an etiologic factor in Duane's retraction syndrome. We performed thin-sectioned magnetic resonance imaging across the brainstem level in 8 cases (11 eyes) that were clinically diagnosed as Duane's retraction syndrome. The same test was applied to 8 healthy control subjects to verify the accuracy of this technique. The nervus abducens on the affected side could not be observed in 6 (54.5%) of 11 eyes (8 cases) that were clinically diagnosed as having Duane's retraction syndrome. The nervus abducens was observed in 15 (94%) of 16 eyes that were screened as the control group. The results showed that aplasia of the nervus abducens, although an important etiologic factor, is not the only factor responsible for the diagnosis of Duane's retraction syndrome.

Neuroma as a cause of oculomotor palsy in infancy and early childhood

Introduction: Third nerve palsy in infancy and early childhood is rare. We present 5 children with isolated monocular oculomotor palsy originally believed to be idiopathic but subsequently documented to be secondary to a presumed neuroma of the third nerve, distal to the mesencephalon. Methods: Retrospective case review of patient charts from 1995 through 1999 was performed. Results: Mean age at initial presentation was 14.4 months (range, 2-36 months). Two patients were male. Initial manifestation was an incomplete paresis in all 5 cases; 2 patients presented initially with exotropia, 2 with ptosis, and 3 with anisocoria. The pupil was ultimately involved in all cases. Progression to total paresis occurred in 4 of 5 patients over a mean of 13 months (range, 4-22 months). Findings of initial neuroimaging were believed to be normal in all cases but further thin-section gadolinium-enhanced magnetic resonance imaging studies, including coronal views, revealed enlargement and/or enhancement of the subarachnoid or intracavernous portion of the oculomotor nerve. Aberrant regeneration did not develop in any case throughout follow-up (mean, 49.4 months). Visual acuity was decreased on the ipsilateral side in all cases, consistent with amblyopia. No patient had neurofibromatosis. Conclusions: Suspicion for neuroma should be high in infants and young children with acquired, isolated, apparently idiopathic third nerve palsy. Serial gadolinium-enhanced, 1- to 2-mm thin-section magnetic resonance imaging with coronal views is essential for proper diagnosis. (J AAPOS 2001;5:9-12)

Portal vein and its tributaries: evaluation with thin-section three-dimensional contrast-enhanced dynamic fat-suppressed MR imaging.

To evaluate the visibility of the main portal vein (MPV) and its tributaries in healthy subjects at thin-section three-dimensional (3D) contrast material-enhanced dynamic fat-suppressed magnetic resonance (MR) imaging and to determine whether this technique provides useful information in the evaluation of patients with cirrhosis. Seventy-two patients (37 control subjects, 35 patients with cirrhosis) underwent imaging with a high-performance-gradient (25 mT/m) system. In the 37 subjects in the control group, the MPV was visualized in 37; splenic vein (SV), in 37; superior mesenteric vein (SMV), in 37; inferior mesenteric vein (IMV), in 35; posterior superior pancreaticoduodenal vein (PSPDV), in 35; gastrocolic trunk (GT), in 34; right gastroepiploic vein (RGEV), in 31; right colic vein, in 30; anterior superior pancreaticoduodenal vein, in 22; middle colic vein (MCV), in 29; and first jejunal vein (FJV), in 36. Satisfactory visualization (mean ratings of 2 or higher) was achieved in the MPV, SV, SMV, IMV, PSPDV, GT, RGEV, and FJV in the control group. Mean diameters of the SV, SMV, MCV, and FJV were significantly larger in the cirrhosis group than in the control group (P <.001, P <.001, P =.048, and P =.002, respectively). Thin-section 3D contrast-enhanced dynamic fat-suppressed MR imaging can facilitate precise visualization of the MPV and its tributaries. Dilatation of the tributaries may be a nonspecific secondary finding that is suggestive of cirrhosis.

Brain atrophy in relapsing-remitting multiple sclerosis and secondary progressive multiple sclerosis: longitudinal quantitative analysis.

To determine annual rates of volumetric changes in the whole-brain parenchyma of patients with relapsing-remitting and secondary progressive multiple sclerosis (MS) and test the hypothesis that these changes correlate with clinical disability. A computer-assisted segmentation technique with thin-section magnetic resonance (MR) imaging was used in 36 patients with MS (27 relapsing-remitting, nine secondary progressive) and in 20 control subjects to quantify brain and cerebrospinal fluid volumes. To determine the degree of brain atrophy, the percentage brain parenchyma volume (PBV) relative to that of intracranial contents was calculated. At the beginning of the study, the PBV was smaller in the MS group than in the control group (P = .007); brain parenchyma volumes were similar. The median rate of brain volume loss was 17.3 mL per year in patients with relapsing-remitting MS and 23.6 mL per year in those with secondary progressive MS. There was a negative correlation between brain atrophy and Expanded Disability Status Scale (EDSS) score in patients with secondary progressive MS (r = -0.69, P = .004) and no correlation in patients with relapsing-remitting MS. T2 lesion volume did not correlate with brain atrophy in either group. The correlation between brain atrophy and EDSS score was better in patients with secondary progressive MS than in those with relapsing-remitting MS.