Rate of Ventricular Enlargement in Multiple Sclerosis: A Nine-Year Magnetic Resonance Imaging Follow-up Study

Transcription

1 Acta Radiologica ISSN: (Print) (Online) Journal homepage: Rate of Ventricular Enlargement in Multiple Sclerosis: A Nine-Year Magnetic Resonance Imaging Follow-up Study J. Martola, L. Stawiarz, S. Fredrikson, J. Hillert, J. Bergström, O. Flodmark, P. Aspelin & M. Kristoffersen Wiberg To cite this article: J. Martola, L. Stawiarz, S. Fredrikson, J. Hillert, J. Bergström, O. Flodmark, P. Aspelin & M. Kristoffersen Wiberg (2008) Rate of Ventricular Enlargement in Multiple Sclerosis: A Nine-Year Magnetic Resonance Imaging Follow-up Study, Acta Radiologica, 49:5, To link to this article: Published online: 09 Jul Submit your article to this journal Article views: 255 View related articles Citing articles: 2 View citing articles Full Terms & Conditions of access and use can be found at Download by: [ ] Date: 05 December 2017, At: 08:59

2 ORIGINAL ARTICLE ACTA RADIOLOGICA Rate of Ventricular Enlargement in Multiple Sclerosis: A Nine-Year Magnetic Resonance Imaging Follow-up Study J. MARTOLA, L. STAWIARZ, S. FREDRIKSON, J. HILLERT, J. BERGSTRO M, O. FLODMARK, P. ASPELIN & M. KRISTOFFERSEN WIBERG Division of Radiology, Department of Clinical Science, Intervention, and Technology, Karolinska Institute, Stockholm, Sweden; Department of Neuroradiology, Karolinska University Hospital, Stockholm, Sweden; Division of Neurology, Department of Clinical Neuroscience, Karolinska Institute, Huddinge, Stockholm, Sweden; Medical Statistics Unit, Department of Learning, Informatics, Management, and Ethics (LIME), Karolinska Institute, Stockholm, Sweden Downloaded by [ ] at 08:59 05 December 2017 Martola J, Stawiarz L, Fredrikson S, Hillert J, Bergstro m J, Flodmark O, Aspelin P, Kristoffersen Wiberg M. Rate of ventricular enlargement in multiple sclerosis: a nineyear magnetic resonance imaging follow-up study. Acta Radiol 2008;49: Background: In multiple sclerosis (MS), brain atrophy assessed by linear measurements of ventricular widths has been reported to be well correlated with three-dimensional (3D) measurements. Therefore, serial linear measurements with no need for advanced 3D evaluation may be proven to be robust markers of irreversible, destructive changes. Purpose: To evaluate the rate of supratentorial ventricular enlargement representing four decades of disease span. Material and Methods: 37 MS patients with disease duration at baseline ranging from 1 to 33 years were included. The mean time of the individual magnetic resonance imaging (MRI) follow-up was 9.25 years (range years). Enlargement rate of the third and lateral ventricles was studied over time by applying three linear measurements on axial 5-mm T1-weighted MRI images. Results: Progression of supratentorial ventricular widths during 9 years follow-up was found. The mean annual width increase of the third ventricle was 0.20 mm (PB0.001, 95% confidence interval [CI] ), for the frontal horn width 0.32 mm (PB0.001, 95% CI ), and increase of the intercaudate distance was 0.26 mm (PB0.001, 95% CI ). The association between these three measurements and disability status persisted at the time of follow-up. Conclusion: We found uniform ventricular enlargement progression during four decades of disease span, suggesting unchanging total brain atrophy progression over time. Key words: Atrophy progression; disability; MRI; multiple sclerosis; ventricular widening Juha Martola, Division of Radiology, Department of Clinical Science, Intervention, and Technology, Karolinska Institute, SE Stockholm, Sweden (tel , fax , . juha.martola@ki.se) Accepted for publication February 23, 2008 Multiple sclerosis (MS) is a complex inflammatory disease of the brain and spinal cord (13) manifested by the appearance of lesions and early-starting, irreversible atrophy (4). Brain atrophy is regarded as a possible predictor of a patient s change in disability (2, 58). Ventricular enlargement reflects a loss of brain parenchyma (9, 10). MS-related ventricular enlargement within 1 year of disease duration has been shown in a study of clinically isolated syndrome (CIS) (11), but the rate of ventricle enlargement in the long term (over a decade) and DOI / its relation to different MS courses has not been investigated. The starting point for prospective, longitudinal magnetic resonance imaging (MRI) studies usually focuses on the early years of disease. In an 8-year longitudinal brain volume study of relapsing-remitting MS (RRMS) patients, Fisher et al. reported significant but nonlinear atrophy progression based on three time points with a mean annual decrease in brain parenchymal fraction (BPF) of 0.4% (7). Dalton et al. reported in their 1-year longitudinal study significant ventricular # 2008 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS)

3 enlargement in CIS, RRMS, and secondary-progressive MS (SPMS), being more pronounced in the SPMS patients than in the RRMS patients (11). Contrary to that, a 15-month longitudinal study by Pagani et al. showed that RRMS patients had significantly higher progression of ventricular enlargement than SPMS and primary-progressive MS (PPMS) patients (12). RRMS patients convert to SPMS within a median time of 16 years (13, 14). Therefore, it is of interest to study whether the ventricular enlargement rates of the RRMS and SPMS subgroups differ over decades of disease course. Such comparisons may confirm a dissociation between persistent clinical deterioration and changes in rates of brain atrophy. Brain atrophy is not a specific pathology developing exclusively in MS. It is a well-known pathology associated with many neurological diseases, such as dementia (15), Alzheimer s disease (16), stroke (17), Huntington s disease (18), schizophrenia (19), etc. Moreover, brain volume shrinkage, affecting the whole brain or only a limited part of the brain, is a part of normal aging and is analyzed in a number of publications (2022). The effect of brain volume change during aging in healthy controls has to be taken into account and particularly discussed in longitudinal studies of different diseases (brain pathologies). Several studies have addressed this issue of atrophy. Scahill et al. reported that normal aging can explain 0.32% of annual whole-brain atrophy and 650 mm 3 of annual ventricular enlargement (20), accelerating after the age of 70 years. Liu et al. showed annual total brain volume change from 0.2% for age B34 years, 0.6% for age B55 years, and up to 1.4% for age 55 years (21). Thompson et al., in their study of Alzheimer s disease, reported % annual cerebral volume reduction in normal controls with a mean age at baseline of years (3.4 years follow-up; n14) (16). Narr et al., in their study of schizophrenia, reported a strong linear increase of almost 1% in cerebrospinal fluid spaces during aging in normal subjects with a mean age of years (range 2058 years; n 28) (19). We followed patients with MRI during a time period of 9.25 years (mean). The widespread range of disease duration at baseline (133 years) made it possible to assess four decades of disease span (23). The primary goal of the study was to evaluate longitudinally the widths of the third and lateral ventricles, reflecting brain atrophy progression. Secondly, we studied the correlation between atrophy rate and changes in disability as expressed by the Expanded Disability Status Score (EDSS) (24) MRI Follow-up of Rate of Ventricular Enlargement in MS 571 and MS Severity Score (MSSS) (25). Thirdly, we investigated the possible correlation between ventricular size and disability. Our fourth aim was to study the association between atrophy development rates and gender, MS course, age at MS onset, disease duration, and interferon treatment. Material and Methods The patient cohort consisted of 45 consecutively selected MS patients from the outpatient ward at the Department of Neurology, Karolinska University Hospital in Stockholm, Sweden, in All patients gave their informed consent. At baseline, patients had a broad range of disease duration (133 years) and a mean age of years (range 2465 years). Forty-five patients underwent MRI investigations in At the time of inclusion, all 37 patients (Table 1) that it was possible to track to final follow-up MRI exams in had clinically definitive MS (CDMS) according to the Poser criteria (26) and MS diagnosis according to the revised McDonald criteria (27). Sixteen of these 37 follow-up patients underwent additional MRI scanning during follow-up in (24 examinations in total). Twenty-four of the 37 follow-up patients had received interferon treatment during the follow-up period. There was no randomized selection or information available regarding compliance to medication. Both clinical and radiological followup examinations were performed at the Karolinska University Hospital in and The study was approved by the ethics committee at the Karolinska Institute. Clinical disability was evaluated by neurologists specialized in MS using EDSS (24), and was additionally adjusted for Table 1. Cohort characteristics at baseline. Gender* Females Males Interferon* Treatment Non-treatment MS course* RRMS SPMS PPMS Age** * Percentage of total (sample size). ** Mean (1 standard deviation, sample size) % (n26) 29.73% (n11) 64.86% (n24) 35.14% (n13) 43.2% (n16) 46% (n17) 10.8% (n4) 42 (10.29, n37)

4 572 J. Martola et al. disease duration and expressed as MSSS (25). Six patients converted from RRMS to SPMS during follow-up and were included in the SPMS group for statistical analyses. There were 16 RRMS, 17 SPMS, and four PPMS patients at the end of the study (Table 1). Six of eight patients lost to followup (of the original 45 patients) were found to be distributed on both sides of the median MSSS, indicating a small risk of selection bias (23). Of the eight patients lost to follow-up, one emigrated, one had postoperative MRI contraindicating metal clips, two died, and the remaining four refused to participate or had no contact with the MS clinics. Six patients improved in disability. The improvement measured in EDSS had a range of during the 9-year follow-up (23). Four of these patients had different EDSS at the first and second measurements in We used a mean EDSS value of these two EDSS in 1995 and 1996 in the statistical analyses. The second EDSS was not evaluated for two patients in We have not included any normal healthy controls in our MRI examinations because of a different primary design of the study in , not planned for such long-term follow-up. Thus, use of healthy controls afterward in had no meaning. Therefore, the results had to be interpreted in comparison with other studies of brain volume change in normal aging serving as controls. Magnetic resonance imaging The MRI examinations were performed on 1.5- Tesla MR systems: at baseline, in , on a Signa Advantage scanner (General Electric Medical Systems, Milwaukee, Wisc., USA) and, in , on a Magnetom Vision scanner (Siemens, Erlangen, Germany), both with T1-weighted 5-mm axial images using a standard T/R birdcage head coil. In , the MRI examinations were performed twice (range of 17 months) using a spinecho pulse sequence (repetition time [TR]/echo time [TE] 640/13 ms, flip angle [FA] 908, no. of excitations [NEX] 2, field of view [FOV] , matrix size , no gap between slices). In , MRI scans were performed with a turbo spin-echo pulse sequence (TR/TE 570/14 ms, FA 908, NEX 2, FOV , matrix size , distance between slices 1.5 mm). In order to improve the analysis, we included an additional 24 MRI scans of 16 patients (of our study cohort consisting of 37 patients) examined between 1996 and Twenty-one of them were performed on the same MRI scanner (Siemens Magnetom Vision) using the same protocol as mentioned above. The remaining three MRI examinations were performed on a 1.5T Siemens Symphony scanner (Siemens, Erlangen, Germany) with T1-weighted, 5-mm axial images (TR/TE 525/15 ms, FA 908, NEX 2, distance between slices 1.5 mm, FOV , matrix size ). Image analysis MRI images were analyzed by a radiologist blinded to clinical outcome data, using a PACS workstation (Sectra, Stockholm, Sweden). Three serial linear axial distances of the third and lateral ventricles were measured on an axial two-dimensional (2D) single-image slice with a distance tool after visual selection of the corresponding slice position for each examination. Measurements were defined as frontal horn width (FHW), intercaudate nucleus distance (ICD), and third ventricle width (3rd VW). These different measurements are illustrated in Fig. 1. Fig. 1. Axial measurements: frontal horn width (FHW), intercaudate distance (ICD), and third ventricle width (3rd VW).

5 Statistical analysis Statistical analyses were performed using SAS software version (SAS Institute Inc., Cary, N.C., USA) and LogXact 7.0 (Cytel Inc., Cambridge, Mass., USA). Graphical presentations were done using Statistica 7.0 software (StatSoft Inc., Tulsa, Okla., USA). All analyses were performed using time expressed in months, although all results are presented as annual progression. Primary analysis Linear mixed-effects models were used to evaluate the mean atrophy rate of the 3rd VW, FRW, and ICD per month of disease duration. A regression line, over duration, was fitted for each patient, estimating the intercept and slope for every individual. The intercept represents the level of ventricular measurement when duration is equal to zero. The slope describes the rate with which the ventricle increases per month, and thus the rate of atrophy. The mean of the individual slopes (rate of atrophy) was then used to make conclusions on the population of MS patients regarding the three different ventricular enlargement measurements (28). This approach is called a random coefficient model and is applied within the framework of the linear mixed-effects model. The individual regression lines need not be linear, and a test of quadric relationships was performed in all analyses. The quadric term represents a test of negative or positive accelerated atrophy over disease duration. When performing the analyses, the effects of disease duration, age, gender, and MS course assignment were controlled for. Due to previous reports of atrophy differences between MS course subgroups (9, 29), that variable was included in all estimated models, regardless of P value. The parameters of the models were estimated using the restricted maximum likelihood estimation (REML). The Satterthwaite approximation was applied to produce accurate Ftests for the model parameters. The assumption of normally distributed errors was assessed using normal probability plots. Cook s distance and various residual plots were used to detect possible influential observations on the estimated parameters of the fitted models. Furthermore, principal component analysis (procedure factor in SAS) was performed to investigate the intracorrelation among the progression of 3rd VW, ICD, and FHW. To decide whether the different ventricular variables formed a cluster, a correlation of at least 0.7 between a component and a variable was used. A scree plot and the amount of MRI Follow-up of Rate of Ventricular Enlargement in MS 573 explained variance served as guidelines to decide how many components should be retained in the final results. Secondary analysis Individual changes in EDSS and MSSS, between the start and end of the study, were classified as increase or decrease/unchanged. Due to the small sample size, the effect of the individual ventricular atrophy rates (see previous section) on the odds of an increase in EDSS or MSSS was analyzed using exact logistic regression. The three ventricular atrophy variables were analyzed in separate models due to high intracorrelation between these measurements. Spearman rank correlations were calculated on the original EDSS/MSSS ordinal values to describe any association between EDSS/MSSS and the three ventricle measurements, at baseline and at the end of the study. Results Primary outcomes Main analysis (random coefficient models). Analyzing 40 years of disease span, we found considerable enlargement of the supratentorial ventricles. The annual mean increase for each ventricular measurement was 0.20 mm for 3rd VW (Table 2, Figs. 2 and 3), 0.26 mm for ICD (Table 2, Fig. 4), and 0.32 mm for FHW (Table 2, Fig. 5). The effect of age at onset was statistically significant in all ventricular measurement analyses (Table 2). In other words, older patients had larger ventricles at baseline than younger ones. There was no statistical evidence of negative or positive acceleration of the atrophy in the 3rd VW, ICD, or FHW over disease duration (test of quadric effect, N.S.). This suggested a linear progression of ventricular enlargement over four decades of disease span. A significant effect of gender and interferon on the three ventricular measurements could not be demonstrated. A few notable differences could, however, be shown between MS groups, although they were not statistically significant (ICD: PPMS vs. SPMS, RRMS vs. SPMS; FHW: PPMS vs. SPMS, RRMS vs. SPMS; Table 2). Combined results (principal component analysis). The principle component analysis (PCA) suggested a one-component solution (Table 3). The first component explained 71% of the variance in the data. The next component explained 17% of the

6 574 J. Martola et al. Table 2. Restricted Maximum Likelihood (REML) parameter estimates for the radiological outcomes. 95% CI Effect Estimate SE (estimate) Lower Upper P value Width of third ventricle, 3rd VW (mm) Disease duration, months* B0.001 Age at onset, years* PPMS vs. RRMS** PPMS vs. SPMS** RRMS vs. SPMS** Intercaudate distance, ICD (mm) Disease duration, months* B0.001 Age at onset, years* PPMS vs. RRMS** PPMS vs. SPMS** RRMS vs. SPMS** Downloaded by [ ] at 08:59 05 December 2017 Frontal horn width, FHW (mm) Disease duration, months* B0.001 Age at onset, years* PPMS vs. RRMS** PPMS vs. SPMS** RRMS vs. SPMS** * Mean slope. ** Post-hoc tests of the differences in means. variance, not explained by the first component. All three variables of progression had correlations above 0.70 with the first component. The progression variables had fairly low correlations regarding the other components (Table 3). Our results suggest that the annual progressions of the width of the third ventricle, intercaudate distance, and the width of the frontal horn behave in the same manner. Secondary outcomes A 1-mm change in annual atrophy of the third ventricle width showed a and 2.43-times higher odds of an increase in EDSS and MSSS, respectively. An annual change of 1 mm in frontal horn width yielded a and 1.95-times higher odds of an increase in EDSS and MSSS, respectively. There was no considerable effect of atrophy of the intercaudate distance on the odds of an increase in EDSS or MSSS. There was no significant effect of baseline disease duration on the odds of an increase in EDSS or MSSS (Table 4) in any of the performed analyses. Ventricular size correlated positively with EDSS and MSSS. We found a positive correlation between the atrophy rate measured during 10 years with EDSS and MSSS. Spearman rank correlations showed low to moderate association between the ventricular measurements and EDSS/MSSS, at baseline and at the end of the study (Table 5). Discussion This 9.25-year longitudinal study, covering 40 years of disease span, demonstrates for the first time a relationship between clinical deterioration and supratentorial ventricular enlargement. A complex inflammatory process leads to irreversible brain atrophy in multiple sclerosis. Ventricular enlargement reflects loss of volume of several structures such as deep gray matter consisting of thalami, basal ganglia, and periventricular white matter, all of which form the walls of the third and lateral ventricles. There are only a few published longitudinal studies of atrophy in MS (3, 7, 30, 31), and it is still unknown when and at what rate focal brain atrophy develops over longer periods of time in MS. The relation between the rate of ventricular enlargement and disability progression is unclear. Furthermore, there may not be a direct temporal relationship between ventricular enlargement and clinical deterioration, i.e., underlying pathological processes occurring at one stage (directly causing clinical deterioration) might become visible in the form of ventricular enlargement at a later stage. Therefore, a direct correlation between morphologic changes and clinical changes might be less obvious. TURNER and coworkers reported that linear measurements correlated to 3D ventricle volumes

7 MRI Follow-up of Rate of Ventricular Enlargement in MS 575 Fig. 3. Mean predicted progression of third ventricle width (3rd VW) with individual follow-up values. The lines represent modelpredicted 3rd VW for RRMS (bold dashed line), PPMS (dashed line), and SPMS (solid line). Downloaded by [ ] at 08:59 05 December 2017 Fig. 4. Mean predicted progression of intercaudate distance (ICD) with individual follow-up values. The lines represent model-predicted ICD for RRMS (bold dashed line), PPMS (dashed line), and SPMS (solid line). (32). SIMON et al. reported significant progression of ventricular diameter in RRMS patients when performing linear measurements over 1 and 2 years with MRI (3). Furthermore, greater increase of disability over 1 and 2 years was associated with more pronounced third ventricle enlargement. Software-based 3D volumetric evaluation is, however, still of limited value in routine clinical assessment. Fig. 2. MRI images show widening of the third ventricle. Note also the widening of basal cisterns and sulci during follow-up (to the left, baseline; to the right, at end of study). Disease duration intervals: case 1 (first row) had a duration of 313 years; case 2 (second row) had a duration of 1322 years; case 3 (third row) had a duration of 2231 years; case 4 (fourth row) had a duration of 3342 years.

8 576 J. Martola et al. Fig. 5. Mean predicted progression of frontal horn width (FHW) with individual follow-up values. The lines represent model-predicted FHW for RRMS (bold dashed line), PPMS (dashed line), and SPMS (solid line). Table 3. Spearman correlation between ventricle measurements and disability (expressed by EDSS/MSSS), at baseline and at the end of the study. EDSS MSSS Baseline End Baseline End Third ventricle width Frontal horn width Intercaudate distance This is the longest follow-up study investigating supratentorial ventricular enlargement in MS to date. It demonstrates a uniform, significant progression of ventricular widening during a 10-year follow-up, spread over four decades of disease span (Fig. 2). In a 5-year longitudinal study of PPMS patients, Ingle et al. also reported that greater rates of reduction in brain volume were associated with longer disease duration (31). We found an unchanged rate of supratentorial ventricular widening, reflecting four decades of disease span. Our results are in accordance with those of FISHER et al. (7), presented in their 8-year follow-up study of RRMS patients. They showed significant, but nonlinear atrophy progression, based on three time points with a mean annual decrease in BPFof 0.4% (7). This longitudinal study was performed in association with the Interferon-b-1a clinical trial. We did not find any significant sign of declining atrophy rate as Fisher et al. did; although this decline, if it exists, would not be so pronounced in 1D measurements as it is in 3D measurements, because of a cubic relation between 3D (volumetric) and 1D (linear) measurements. In our study, the rate of ventricular widening was independent of age at onset and MS course. Our results did not indicate an interaction effect between disease duration and age at onset. This suggests an equal progression rate of ventricle enlargement between individuals at different age at onset, but at a higher level if disease debut occurred later in life. If the patient had disease onset in later years of life, her/his ventricles were already to some extent enlarged, but the further rate of ventricular enlargement was similar for all patients independent of age at onset. In a study of a normal population (range years), SCAHILL et al. found the mean annual whole-brain volume atrophy rate to be 0.32% with a 650-mm 3 annual enlargement of the ventricles (20). No evidence suggesting a significant acceleration in the rate of ventricular enlargement with increasing age could be shown before the age of 70 years (20). They did not find any significant effect of sex on the rates for any of the brain structures. This is in accordance with our results. We did not find any significant change of ventricular enlargement Table 4. Exact logistic regression parameter estimates of the atrophy outcomes of EDSS and MSSS: effect variables, annual progression rate, and disease duration. Covariate Odds ratio: annual progression rate 95% CI lower 95% CI upper P value Odds ratio: disease duration 95% CI lower 95% CI upper P value EDSS Third ventricle width (mm), 3 rd VW Intercaudate distance (mm), ICD Frontal horn width (mm), FHW MSSS Third ventricle width (mm), 3rd VW Intercaudate distance (mm), ICD Frontal horn width (mm), FHW

9 Table 5. Principal components pattern coefficients and the amount of explained variance. Component Eigenvalues Variance explained 71% 17% 12% Factor loadings Third ventricle width Intercaudate distance Frontal horn width rates over 40 years of disease span for our patients representing a range of ages between 24 and 65 years. We did not find any significant differences in atrophy rates. This is also in accordance with Fox et al. and Kalkers et al. who compared RRMS, PPMS, and SPMS, and found that brain atrophy rate is independent of MS course (9, 29). The corpus callosum forms the roof of the third and lateral ventricles, and is associated with enlargement of these supratentorial ventricles. In our recent 9-year follow-up study of corpus callosum area (CCA), we found a constant progression of non-age-related CCA decrease (23). Furthermore, there was a persisting strong correlation between CCA and EDSS, at baseline and at the end of the study. There were no significant differences between MS course and the rate of CCA decrease. In contrast to the CCA study, we report here worsening odds ratios for EDSS, with a large increase in FHW and a widening of the 3rd VW. In our study, the supratentorial ventricular widths had a significant, persisting association with disability status. This is in accordance with a 5-year longitudinal study of PPMS by INGLE et al., who found that increasing disability was associated with ventricular volume (31). Furthermore, INGLE et al. subdivided the cohort into clinically stable and clinically deteriorating groups, and found no correlation with either ventricle enlargement or brain atrophy rates (31). However, change in spinal cord area correlated with change in EDSS. KALKERS et al. found that annualized EDSS change did not trcorrelate with any annualized markers of brain atrophy (29). We found that higher annual increase in FHW and a widening of the n3rd VW were positively correlated with disability increase. In a four-year longitudinal study, TURNER et al. reported that an increase in width of the lateral ventricles was related to change in EDSS (10). Simon et al. reported significant progression of ventricular diameter in RRMS patients when performing linear measurements over 1 and 2 years with MRI (3). Furthermore, greater disability MRI Follow-up of Rate of Ventricular Enlargement in MS 577 increase over 1 and 2 years was associated with more pronounced third ventricle enlargement. There are several limitations to our study. The first, previously discussed limitation to our study is the lack of normal healthy controls (not included in ). Therefore, the results had to be compared with the other studies of brain volume change in normal aging. Normal controls for direct comparison would have been ideal. Unfortunately, they were not included at the baseline of our study in when we focused on active inflammation and contrast-enhancing lesions. Some limitations related to the use of disability scales are also necessary to consider. EDSS is known to overemphasize motor function and largely lacks input of cognitive assessment, i.e., it may assess spinal processes rather than cerebral processes. Another problem in using EDSS is its lack of linearity (33). If we had better measurements of disability, it might be possible to address the question of whether brain atrophy, as measured 0by MRI with morphometric methods, could actually reflect clinical disability. The MS Severity Score (MSSS) has been developed to classify patients or groups of patients into severity groups, taking into account duration of the disease. Therefore, we used the MSSS to assess relative severity at baseline and at 10 years followup rather than uncorrected EDSS. Our measurements could to some extent be biased by technical reasons such as slice location and the partial volume effects of 5-mm-thick slices. However, the strong intercorrelation of the changes in the three measurements supports our conclusion that there is significant longitudinal supratentorial ventricular enlargement. The RRMS and SPMS groups were comparable in size. However, the small amount of patients in the PPMS group makes the conclusion no difference in annual ventricular enlargement rate between MS courses less valid for that group. In conclusion, we used three axial linear measurements of ventricular size and found a strong correlation between progression of supratentorial ventricular enlargement and the duration of the disease. This measurement of cerebral atrophy was independent of MS course and suggested a uniform volume reduction over four decades of disease span. The rates of both third ventricle widening and frontal horn widening were associated with clinical deterioration. The significant association between the three ventricle enlargement measurements and disability status persisted during follow-up. Serial linear ventricle diameter measurements, which do not require advanced 3D software

10 578 J. Martola et al. assessment, may prove to be useful in monitoring irreversible and destructive changes in the MS brain, giving additional information for the monitoring of treatment. References 1. Bermel RA, Bakshi R. The measurement and clinical relevance of brain atrophy in multiple sclerosis. Lancet Neurol 2006;5: Liu C, Edwards S, Gong Q, Roberts N, Blumhardt LD. Three dimensional MRI estimates of brain and spinal cord atrophy in multiple sclerosis. J Neurol Neurosurg Psychiatry 1999;66: Simon JH, Jacobs LD, Campion MK, Rudick RA, Cookfair DL, Herndon RM, et al. A longitudinal study of brain atrophy in relapsing multiple sclerosis. The Multiple Sclerosis Collaborative Research Group (MSCRG). Neurology 1999;53: Brex PA, Jenkins R, Fox NC, Crum WR, O Riordan JI, Plant GT, et al. Detection of ventricular enlargement in patients at the earliest clinical stage of MS. Neurology 2000;54: Bakshi R, Benedict RH, Bermel RA, Jacobs L. Regional brain atrophy is associated with physical disability in multiple sclerosis: semiquantitative magnetic resonance imaging and relationship to clinical findings. J Neuroimaging 2001;11: Bakshi R, Dandamudi VS, Neema M, De C, Bermel RA. Measurement of brain and spinal cord atrophy by magnetic resonance imaging as a tool to monitor multiple sclerosis. J Neuroimaging 2005;15: Fisher E, Rudick RA, Simon JH, Cutter G, Baier M, Lee JC, et al. Eight-year follow-up study of brain atrophy in patients with MS. Neurology 2002;59: Zivadinov R, Bakshi R. Central nervous system atrophy and clinical status in multiple sclerosis. J Neuroimaging 2004;14: Fox NC, Jenkins R, Leary SM, Stevenson VL, Losseff NA, Crum WR, et al. Progressive cerebral atrophy in MS: a serial study using registered, volumetric MRI. Neurology 2000;54: Turner B, Lin X, Calmon G, Roberts N, Blumhardt LD. Cerebral atrophy and disability in relapsing-remitting and secondary progressive multiple sclerosis over four years. Mult Scler 2003;9: Dalton CM, Miszkiel KA, O?Connor PW, Plant GT, Rice GP, Miller DH. Ventricular enlargement in MS: one-year change at various stages of disease. Neurology 2006;66: Pagani E, Rocca MA, Gallo A, Rovaris M, Martinelli V, Comi G, et al. Regional brain atrophy evolves differently in patients with multiple sclerosis according to clinical phenotype. Am J Neuroradiol 2005;26: Eriksson M, Andersen O, Runmarker B. Long-term follow up of patients with clinically isolated syndromes, relapsing-remitting and secondary progressive multiple sclerosis. Mult Scler 2003;9: Runmarker B. Andersen O. Prognostic factors in a multiple sclerosis incidence cohort with twenty-five years of follow-up. Brain 1993;116: Bastos-Leite AJ, van der Flier WM, van Straaten EC, Staekenborg SS, Scheltens P, Barkhof F. The contribution of medial temporal lobe atrophy and vascular pathology to cognitive impairment in vascular dementia. Stroke 2007;38: Thompson PM, Hayashi KM, de Zubicaray G, Janke Al, Rose SE, Semple J, et al. Dynamics of gray matter loss in Alzheimer s disease. J Neurosci 2003;23: Mark VW, Taub E, Perkins C, Gauthier LV, Uswatte G, Ogorek J. Poststroke cerebral peduncular atrophy correlates with a measure of corticospinal tract injury in the cerebral hemisphere. Am J Neuroradiol 2008;29: Kassubek J, Juengling FD, Kioschies T, Henkel K, Karitzky J, Kramer B, et al. Topography of cerebral atrophy in early Huntington s disease: a voxel based morphometric MRI study. J Neurol Neurosurg Psychiatry 2004;75: Narr KL, Sharma T, Woods RP, Thompson PM, Sowell ER, Rex D, et al. Increases in regional subarachnoid CSFwithout apparent cortical gray matter deficits in schizophrenia: modulating effects of sex and age. Am J Psychiatry 2003;160: Scahill RI, Frost C, Jenkins R, Whitwell JL, Rossor MN, Fox NC. A longitudinal study of brain volume changes in normal aging using serial registered magnetic resonance imaging. Arch Neurol 2003;60: Liu RS, Lemieux L, Bell GS, Sisodiya SM, Shorvon SD, Sander JW, et al. A longitudinal study of brain morphometrics using quantitative magnetic resonance imaging and difference image analysis. Neuroimage 2003;20: Ge Y, Grossman RI, Babb JS, Rabin ML, Mannon LJ, Kolson DL. Age-related total gray matter and white matter changes in normal adult brain. Part I: volumetric MR imaging analysis. Am J Neuroradiol 2002;23: Martola J, Stawiarz L, Fredrikson S, Hillert J, Bergstro m J, Flodmark O, et al. Progression of non-agerelated callosal brain atrophy in multiple sclerosis: a 9-year longitudinal MRI study representing four decades of disease development. J Neurol Neurosurg Psychiatry 2007;78: Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 1983;33: Roxburgh RH, Seaman SR, Masterman T, Hensiek EA, Sawcer SJ, Vukusic S, et al. Multiple Sclerosis Severity Score: using disability and disease duration to rate disease severity. Neurology 2005;64: Poser CM, Paty DW, Scheinberg L, McDonald WI, Davis FA, Ebers GC, et al. New diagnostic criteria for multiple sclerosis: guidelines for research protocols. Ann Neurol 1983;13: Polman CH, Reingold SC, Edan G, Filippi M, Hartung HP, Kappos L, et al. Diagnostic criteria for multiple sclerosis: 2005 revisions to the McDonald Criteria. Ann Neurol 2005;58:8406.

11 MRI Follow-up of Rate of Ventricular Enlargement in MS Fitzmaurice GM, Laird NM, Ware JH. Applied longitudinal analysis. New Jersey: John Wiley & Sons; p Kalkers NF, Ameziane N, Bot JC, Minneboo A, Polman CH, Barkhof F. Longitudinal brain volume measurement in multiple sclerosis: rate of brain atrophy is independent of the disease subtype. Arch Neurol 2002;59: Ingle GT, Stevenson VL, Miller DH, Leary SM, Rovaris M, Barkhof F, et al. Two-year follow-up study of primary and transitional progressive multiple sclerosis. Mult Scler 2002;8: Ingle GT, Stevenson VL, Miller DH, Thompson AJ. Primary progressive multiple sclerosis: a 5-year clinical and MR study. Brain 2003;126: Turner B, Ramli N, Blumhardt LD, Jaspan T. Ventricular enlargement in multiple sclerosis: a comparison of three-dimensional and linear MRI estimates. Neuroradiology 2001;43: Hobart J, Freeman J, Thompson A. Kurtzke scales revisited: the application of psychometric methods to clinical intuition. Brain 2000;123: