ARTICLE

Multiple Sclerosis 2006; 12: 428
436

Memory performance in multiple sclerosis patients correlates with central brain atrophy H Hildebrandt1,2, HK Hahn3, JA Kraus1, A Schulte-Herbru ¨ggen1,2, B Schwarze4 and 1 G Schwendemann Objective To assess whole brain and central brain atrophy as well as their differential relation to memory, cognitive performance, fatigue, depression and quality of life in patients with relapsingremitting multiple sclerosis (RRMS). Methods A 3D flow compensated gradient recalled T1-weighted MRI was acquired in 45 RRMS patients. An automated analysis tool was used to calculate brain parenchymal fraction (BPF) and ventricular brain fraction (VF). All patients were assessed with neuropsychological tests focusing on memory and self-rating scales for depression, fatigue and quality of life. Age corrected partial correlations between brain atrophy, motor performance, psychological scales and test scores were calculated. Results BPF correlated moderately (0.3 5/rB/0.5) with duration of symptoms and disease, the Expanded Disability Status Scale (EDSS), the upper extremity motor performance, and with mental aspects of quality of life. VF correlated moderately with EDSS, upper and lower extremity motor performance and memory functions. Neither BPF nor VF correlated with fatigue and depression. Results of several cognitive tests correlated moderately with depression and fatigue, the Paced Auditory Serial Addition Test (PASAT) showing the largest correlation. Conclusions Memory performance shows a correlation with relative ventricular size in RRMS patients, indicating the strategic location of the ventricle system along the structures of the limbic system and its vulnerability in MS. The PASAT and several other cognitive tests show moderate correlations with depression and fatigue, arguing for an inter relation between the cognitive functioning and the emotional state of patients. However, this relation is independent of measurable brain atrophy. Multiple Sclerosis 2006; 12: 428
436. www.multiplesclerosisjournal.com Key words: brain atrophy; depression; memory; MRI; multiple sclerosis; quality of life

Introduction Previous studies revealed weak to moderate correlations between brain atrophy and cognitive functions in multiple sclerosis (MS) (for a review see Benedict et al . [1]). However, the methods to estimate brain atrophy varied considerably (from manual measurement of ventricular size in one slice

of a CT scan to semi-automated programs for the analysis of high resolution structural MRI). In addition, authors focused on different regions of interest: some analysed third ventricle width [2], others analysed bicaudate ratio [3], or corpus callosum thickness [4], and their relation to cognitive functions. More recent studies used high resolution structural MRI and encompassing mea-

1

Department of Neurology, Klinikum Bremen-Ost, Zu ¨ richer Str. 40, 28325 Bremen, Germany University of Oldenburg, Institute for Psychology, 26111 Oldenburg, Germany 3 MeVis, Center for Medical Diagnostic Systems and Visualization, Universita ¨tsallee 29, 28359 Bremen, Germany 4 Department of Radiology, Klinikum Bremen-Ost, Zu ¨ richer Str. 40, 28325 Bremen, Germany Address for correspondence: Helmut Hildebrandt, Ph.D., Department of Neurology, Klinikum Bremen-Ost, Zu ¨ richer Str. 40, 28325 Bremen, Germany. E-mail: [email protected] Received 9 April 2005; accepted 2 August 2005 2

– 2006 Edward Arnold (Publishers) Ltd

10.1191/1352458506ms1286oa

Memory performance and brain atrophy in MS sures of atrophy as brain parenchymal fraction (BPF; ie, relation of the brain volume to the intracranial volume) or ventricular to brain volume fraction (VF). Kalkers et al . [5], investigated the relation between brain atrophy and the Paced Auditory Serial Addition Test (PASAT), a subtask of the Multiple Sclerosis Functional Composite Score (MSFC), as a test of cognition [6,7], and found a weak correlation. Edwards et al . [8], and Christodoulou et al . [9], used cognitive screening batteries and could also show an association between brain atrophy and cognition. If separately analysed, the correlation between brain atrophy and cognition in MS was always stronger than between lesion load and cognition [1]. Using an extended neuropsychological testing battery, modern high resolution MRI and semiautomated methods to calculate brain atrophy, a recent study of Benedict et al . [10], replicated the finding of Rao et al . [2], that the width of the third ventricle is correlated with memory performance. They also found that, in particular, the width of the third ventricle is associated with performance in the PASAT and the Symbol Digit Matching Test (SDMT). The BPF contributed only to cognitive performance when the size of the third ventricle was excluded from analysis. In this case, BPF had a significant correlation with the PASAT and the SDMT, but not with verbal memory functions. Christodoulou et al . [9], but not Deloire et al . [11], found results similar to those of Benedict et al. [10]. The contradictory result of Deloire et al . [11], might be explained by the fact that they focused primarily on magnetization transfer imaging. Therefore, this group derived the additional data on brain atrophy from relatively coarse slices (5 mm). Moreover, these slices were acquired in an axial and not 3D orientation. Such a measurement of brain atrophy may have led to less precise neuroanatomical data. The strategic role of the third ventricle’s size for memory functions has also been shown in neurodegenerative diseases [12
14], and for patients with long-term alcohol abuse [15]. From a neuroanatomical view, the central localization of the third ventricle along the wall of the medial and anterior nuclei of the thalamus, playing a major role in the Papez circuit, renders such data plausible. But exactly the same line of reasoning demonstrates that the second ventricles are located next to areas important for memory functions: its temporal horn is located adjacent to hippocampal structures, the fornix runs through the dorsal parts of the second ventricles, and the cholinergic septal nuclei form the most caudal bounds of its frontal horn. Therefore, it appears likely that ventricular size, in general, is related to memory function in MS patients. www.multiplesclerosisjournal.com

429

Our study focuses on central and whole atrophy of the brain by means of VF and BPF in relapsingremitting (RR) MS patients and their differential impact on neurocognitive functions, such as cognitive speed, verbal and performance intelligence, working memory, object alternation (OA) and verbal memory functions. We also correlate the data of brain atrophy with clinical grading of disease progression, depression, fatigue and quality of life. Our primary goal is to extend and replicate the findings of Benedict et al . [10], and Christodoulou et al . [9], concerning the correlation between memory and cognitive functions and ventricle size in RRMS patients, but also to look for correlations between motor performance, cognitive performance, and subjective well-being in this patient group.

Methods Patients Forty-five patients with RRMS, diagnosed according to the McDonald criteria [16], participated in our study (Table 1). All patients were formerly inpatients of the Department of Neurology of the Klinikum Bremen-Ost/Germany and treated for an acute relapse. Neuropsychological testing was carried out at least four weeks after the end of the treatment with methylprednisolone (1000 mg i.v. for five days). All patients received treatment with either interferon beta or glatirameracetate. For our study, exclusion criteria were (1) an Expanded Disability Status Scale (EDSS) [17] score /6; (2) a current or past medical illness or a psychiatric disorder according to the Diagnostic and Statistical Manual of Mental Disorders, fourth edition; and (3) substance abuse. All patients gave their informed consent to participate in this study, which was approved by the ethical board of the Bremer Physicians Society. Of the 45 patients, 29 were female and 16 male, with a mean age of 38.9 years (range: 23
63 years). Seventeen patients had attended a ‘Gymnasium’ Table 1 Characteristics of the patients’ group (29 female and 26 male) Mean Age (years) EDSS MSFC z-score Disease duration (years) Symptoms duration (years)

Maximum Minimum

38.9 63 2.6 6 /0.0645 0.8739 5.02 20 8.8 29

23 0 /5.3278 0.16 0.16

EDSS, Expanded Disability Status Scale; MSFC, Multiple Sclerosis Functional Composite Score.

Multiple Sclerosis 2006; 12: 428
436

430

H Hildebrandt et al.

(13 years school education), a further 24 patients had at least a ‘Realschule’ (10 years school education) and only four patients had B/10 years of school education. Disease duration was two years in the mean with a range from two months to 20 years. Clinical symptoms were noticed in the mean for 8.8 years with a range of two months to 29 years. The mean EDSS score was 2.6, with a maximum of 6 and a minimum of 0.0. The MSFC was 0.0645, with a minimum of /5.327 and a maximum of 0.8739.

interface) and extracerebral CSF (CSF
bone interface) [19]. MBV defines BPF as brain parenchymal volume comprising GM and WM divided by the intracranial volume, which is computed as the sum of brain and CSF volumes. VF is calculated by dividing the ventricular volume by the brain parenchymal volume. The cerebral ventricular volume was measured using a simpler histogram model comprising three tissue types (WM, GM and CSF) plus mixed Gaussians [21].

Clinical investigation MRI protocol and analysis Some 160 transversal oriented 3D flow compensated gradient recalled T1-weighted images (repetition time/echo time: 25.0/4.6 ms) were obtained using a 1.5 Tesla Philips Medical Systems Gyroscan Intera Scanner. Images were acquired with a 256+256 matrix over a 260 mm field of view (FOV) and reconstructed with a rectangular FOV of 85%. Slice thickness was 1 mm. Scan time was 8 and a half minutes.

Calculation of BPF and VF BPF and VF were calculated by MeVisLab Brain Volumetry (MBV), a fast and robust method relying on skull stripping and histogram analysis only. More precisely, the following four subsequent steps were performed: (1) interactive definition of a cuboid ROI on three orthogonal and synchronized 2D views; (2) automatic resampling to an isotropic grid (spacing 0.9/0.9 /0.9 mm3) using a Mitchell filter for x , y , and z directions; (3) skull stripping using the marker-driven 3D Interactive Watershed Transform (IWT), which has been described in detail [18]; and (4) automatic histogram analysis for the 3D region defined by the IWT result. The application of MBV in MS was described by Lukas et al. [19]. The histogram analysis is based on a model consisting of four Gaussian distributions for pure tissue types (white matter (WM), grey matter (GM), cerebrospinal fluid (CSF), and bone/air), as well as dedicated partial volume distributions for mixed tissue types, so-called mixed Gaussians. The histogram model corresponds to the assumptions formulated by Santago and Gage [20], ie, uniform partial volume effects from each pair of tissues, and is fitted to the histogram by minimizing least square deviations. We used a fourth class to cover air and bone tissue, which is required since the IWT when applied to the original data (interpreted as depth information) includes most of the partial volume regions at the outer brain surface (GM
CSF Multiple Sclerosis 2006; 12: 428
436

All patients were assessed by an experienced neurologist and scored for the EDSS. The motor scores of the MSFC (9-Hole Peg Test, NHPT, and Timed Walk Test; TWT) were obtained by the psychologist blinded to the assessed clinical data, who also performed the neuropsychological investigation. To calculate the MSFC score, z scores were created for NHPT, PASAT and TWT. These z scores were obtained by using data from the National Multiple Sclerosis Society task force database, which includes a wide range of patients with MS [22]. According to published recommendations, the inability to perform the TWT was set to a score of /13.7 (one patient was unable to perform the TWT at the time of the investigation). The composite score was then calculated using the published formula [22].

Neuropsychological investigation To assess verbal intelligence, we used the subscales similarities and information from the Wechsler Adult Intelligence Scale (WAIS) [23], as prescribed by the short form of the WAIS [24]. Performance IQ was scored using the Block Design Test and Picture Completion. The Alertness Test of the ‘Testbatterie zur Aufmerksamkeitspru ¨ fung’ (TAP) [25], was used to measure cognitive speed, both for non-cued and a cued condition. In this test, the participant has to press a key a soon as a cross appears at the centre of a screen. In each of four sessions, 20 trials are performed. In two of these sessions, the participant is informed by an acoustic cue that the cross will appear, in the other two sessions, no warning signal is given. The test also provides a phasic alertness fraction between response times without warning cues minus response times with acoustic cue divided by the sum of both response times to assess for the ability to sustain attention [26]. The PASAT from the MSFC [22], served as a measure of higher order attention and working memory [27]. The PASAT (3-second presentation rate) requires the participant to monitor audiotaped www.multiplesclerosisjournal.com

Memory performance and brain atrophy in MS digits while adding each consecutive digit to the preceding digit. We used the OA task from the TAP [25], to investigate cognitive flexibility. In this task, a letter and a number are presented simultaneously on a computer screen for 100 trials. One stimulus appears on the right, the other on the left side; sides vary randomly. A left and right response button is available to the participant. The subject has to press alternately on the side of the letter and the number and the stimulus pair is shown until the reaction of the subject follows. The response times and number of errors are used for analysis. Verbal memory was assessed by using a German translation of the California Verbal Learning Test following the normative data for age and education published by Niemann et al . [28]. Depression was self-rated by Beck’s Depression Scale (BDS) [29], fatigue by the Fatigue Severity Scale (FSS) [30], and quality of life by the 12 question short form of the SF 36 health questionnaire [31]. For the latter instrument, we calculated a SF12 Bodily Score (SF12 BS), which focuses on the subjective rating of motor performance, and a SF12 Mental Score (SF12 MS), which focuses on mental well-being, as defined in the manual [31]. Statistical analysis Statistical evaluations were performed with the SPSS 12 software package. Due to the exploratory nature of our investigation, level of significance was set to 0.05 for all statistical evaluations. To avoid effects of aging, we used partial correlations to analyse the relations between brain atrophy, functional impairment and self-rating scores. Following the proposal of Cohen [32], for interpreting r values, we interpreted coefficients 0.1 5/r B/0.3 as small, B/0.5 as medium or moderate, and /0.5 as large correlations.

Results Results differed considerably between subjects depending on age, disease progression and social or intellectual status. Verbal and performance subscales of the WAIS ranged from superior to 1 standard deviation (SD) below age-controlled population mean. In the following, a score below 1 SD of the normative data will be considered as impairment; accordingly 9% of the patients were considered impaired either in verbal or in performance IQ. Both PASAT and OA revealed a considerable rate of impairment (33 and 38% impaired patients, respectively). Cognitive speed was also either accurate or extremely slowed (between 36 and 33% of the patients were impaired depending www.multiplesclerosisjournal.com

431

on the kind of measurement). In the case of verbal learning, some subjects were able to reproduce all items of the CVLT in long-delay free recall, whereas others had a zero recall performance, the rate of impairment was 20%. The same pattern was found for the affective scales: some patients were without any depressive symptoms or fatigue, whereas others suffered extremely in these respects. In respect of the BDS, we took a cut-off score of ‘16’ as showing some symptoms of a depression and ‘20’ for a definite depression [33]. This resulted in 24% of the patients with some symptoms of a depression and 7% with a definite depression. Twenty-three patients had an FSS of 4.5 and above (51% of the total group). Twenty-one patients (47%) had a score in the SF12 BS below 1 SD of the population mean and 16 (36%) below 1 SD of the SF12 MS.

Brain atrophy, motor performance, cognitive performance and quality of life Table 2 summarizes the correlations between brain atrophy on the one hand and clinical status, cognitive and motor performance, and quality of life on the other hand, sorted by size. BPF and VF were highly intercorrelated (r / /0.7167, P/0.000). BPF also correlated with duration of disease (r / /0.4697, P/0.001) and symptoms (r / /0.3727, P /0.01). Concerning motor functioning, it correlated with the upper extremity motor score of the MSFC (NHPT r /0.3488, P/0.02) and with the EDSS score (r / /0.3236, P/0.03). There was also a significant correlation with the SF12 MS (r / /0.3282, P/0.03). The VF correlated with the EDSS score (r / /0.4697, P/0.001), and with both motor scores of the MSFC (TWT: r / /0.3301, P/0.03; NHPT: r /0.4646, P /0.001). In contrast to BPF, VF correlated with several CVLT parameters (learning curve: r / /0.3423, P/0.02; sum of learned items: r /0.3375, P /0.03; short-term cued recall: r / /0.3674, P/0.01; long-delay free recall: r / /0.3979, P/0.007). But VF was not correlated with the BDS and the FSS, or with the SF12 MS.

Motor performance, cognitive performance and quality of life EDSS score and performance in the TWT and the NHPT were correlated (r/ /0.3191, P /0.035; r / /0.6152, P/0.000). Neither EDSS nor TWT and NHPT correlated with the BDS, the FSS and the SF12 MS, but SF12 BS correlated with EDSS (r / /0.5135, P B/0.001) and NHPT (r / /0.3282, P/0.029). The EDSS score correlated moderately with the memory scores of the CVLT (sum of learned Multiple Sclerosis 2006; 12: 428
436

432

H Hildebrandt et al.

Table 2 Age-corrected partial correlation with VF and BPF sorted for size Brain parenchymal fraction Ventricular fraction

Coeff

P

EDSS NHPT z-score CVLT LDFR CVLT SDCR CVLT LDCR CVLT leaning curve CVLT true recognition CVLT learning sum TWT z-score Object alternation hits SF12 bodily score Disease duration CVLT SDFR Duration of symptoms Object alternation response time SF12 mental score WAIS Perfomance IQ Alertness with acoustic cueing CVLT proactive interference decrease Phasic alertness fraction Alertness without cueing WAIS Verbal IQ Fatigue severity scale PASAT z-score Beck’s Depression Scale

0.46966 /0.46455 /0.39789 /0.36744 /0.36012 /0.34232 /0.33752 /0.33752 /0.33013 /0.27366 /0.26696 0.26173 /0.25153 0.23916 0.20789 0.18454 /0.17901 0.11000 0.10537 /0.09137 0.07393 /0.04606 0.03154 /0.02807 0.01575

0.00130 0.00149 0.00748 0.01414 0.01634 0.02294 0.02505 0.02505 0.02863 0.07226 0.07981 0.08613 0.09956 0.11793 0.17569 0.23045 0.24629 0.47721 0.49603 0.55528 0.63341 0.76605 0.83895 0.85645 0.91918

Disease duration Duration of symptoms NHPT z-score SF12 mental score EDSS WAIS Performance IQ CVLT SDCR CVLT LDCR CVLT LDFR SF12 bodily score CVLT learning curve TWT z-score CVLT true recognition CVLT SDFR CVLT learning sum Phasic alertness fraction Beck’s Depression Scale CVLT proactive interference decrease Object alternation hits Alertness without cueing Fatigue severity scale WAIS Verbal IQ Alertness with acoustic cueing Object alternation response time PASAT z-score

Coeff

P

/0.46971 /0.37268 0.34885 /0.32822 /0.32359 0.28423 0.24287 0.22056 0.19830 0.19781 0.19098 0.16753 0.15454 0.15380 0.13305 0.12366 0.11856 /0.11271 0.09234 0.07552 0.06726 0.05516 0.03115 0.01556 0.01122

0.00130 0.01272 0.02030 0.02962 0.03214 0.06139 0.11217 0.15023 0.19692 0.19807 0.21431 0.27706 0.31653 0.31889 0.38923 0.42386 0.44337 0.46635 0.55106 0.62612 0.66444 0.72373 0.84090 0.92016 0.94239

Values in bold and italics indicate significant correlations (PB/0.05). EDSS, Expanded Disability Status Scale; MSFC, Multiple Sclerosis Functional Composite Score; NHPT, Nine Hole Peg Test; TWT, Timed Walk Test; PASAT, Paced Auditory Serial Addition Test; CVLT, California Verbal Learning Test; LD, long delay; SD, short delay; FR, free recall; CR, cued recall; WAIS, Wechsler Adult Intelligence Scale.

items: r / /0.4919, P B/0.001; short-delay free: r/ /0.4776, P/0.001 and cued recall: r / /0.4525, P B/0.002; long-delay free: r / /0.4663, P/0.001 and cued recall: r/ /0.4815, P B/0.001; recognition: r/ /0.5001, P B/0.001), with the Block design test (r / /0.4074, P/0.006), and with hits in the OA task (r / /0.3387, P/0.024), but not with the other three subtest of the WAIS, the performance in the PASAT, and with attention. The TWT correlated with long-term free (r /0.5661, P B/0.001) and semantic cued recall (r /0.5324, P B/0.001) of the CVLT, with response times of the OA (r /0.4761, P/0.001), but with none of the other cognitive or emotional scores. In the case of the NHPT score, again all of the CVLT scores correlated with motor performance (sum of learned items: r /0.4033, P/0.007, short-delay free: r/0.3273, P/0.03 and cued recall: r /0.4861, P B/0.008; long-delay free: r /0.3930, P /0.008 and cued recall: r /0.4456, P B/0.002; recognition: r/0.44552, P /0.002). There was also a correlation with the PASAT (r /0.3232, P /0.035), with response time in the OA task (r / /0.3608, P/0.016), the cued attention test (r / /0.3163, P/0.036) and the attention test without cue (r/0.3111, P/0.039). Multiple Sclerosis 2006; 12: 428
436

Quality of life and cognitive performance The SF12 MS showed only one but large correlation, which concerned the BDS (r / /0.5165, P /0.000). The FSS correlated with the PASAT (r / /0.3843, P/0.008), attention (without acoustic cue: r /0.3115, P/0.035; with acoustic cue: r /0.3288, P/0.026) and with hits in the OA task (r / /0.3383, P /0.021). The BDS score correlated with the verbal intelligence tasks from the WAIS (general knowledge: r / /0.3117, P /0.035; similarities: r / /0.3940, P/0.007). There was also a considerable correlation with performance in the PASAT (r / /0.4402, P /0.002), with the OA task (response time: r / /0.4002, P /0.006; hits: r / /0.4455, P/0.002), with attention (without acoustic cue: r/0.3483, P/0.018; with acoustic cue: r /0.2954, P/0.046) and with long-delay free recall of the CVLT (r / /0.3154, P/0.033).

Discussion In this study, we focused on the relation between two different markers of brain atrophy and the motor, cognitive and emotional performance of www.multiplesclerosisjournal.com

Memory performance and brain atrophy in MS RRMS patients. We also analysed the relation between cognition and emotional well being in these patients. Of the two markers for brain atrophy (BPF and VF), only VF correlated with cognitive variables, whereas BPF correlated with only EDSS, NHPT and with the SF12 MS. Moreover, the correlation of the VF concerned exclusively memory performance and not executive functions, working memory and attention. Therefore, we did not only replicate the results of Benedict et al . [10], who also found a superior role of the width of the third brain ventricle for cognitive functions and especially for memory functions, but also extended them to the whole ventricular size. The central role of the ventricular size for disease progression was also shown by Simon et al . [34], who found a high correlation with inflammation, and by Dalton et al . [35], who found that ventricular size predicted whether or not patients with clinically isolated symptoms developed RRMS. Focusing on memory, the specific correlation with VF may be explained by alterations of the adjacent limbic structures. Hippocampus, enthorhinal cortex, thalamus, fornix and septal forebrain, ie, with the exception of the cingulate gyrus all major structures involved in memory, are located along the walls of the second and third ventricle. Widening of the ventricular size may indicate cell loss in limbic brain structures or injury to the limbic system by increased water diffusion [36]. Neuronal degeneration has been observed for MS patients in the thalamus [37,38]. The strategic position of the ventricle system for measuring neuroanatomical aspects of memory function has also been shown in several other neurodegenerative [12
14], and neurotoxic diseases [15]. Early in disease progression, MS leads to an enlargement of the ventricles [35]. Our study and the study by Amato et al . [39], demonstrate that such degeneration correlates with a subtle decrease in memory performance, because in mean our patients did not score below the age-controlled population norms. In addition, they performed better than the patients of Benedict et al . [10], and were clinically less impaired. However, there was no correlation between executive functions or working memory and brain atrophy. Previous studies measuring brain atrophy revealed a link between the PASAT and the BPF [1], but unlike our study, not all corrected the analysis for age [5]. In two former studies, we showed that in MS patients, working memory may be preserved for a long time, especially for the RR course of MS [40,41]. Archibald et al . [42], did not find a correlation between working memory performance and brain atrophy either. But, the same studies of our group [40,41], demonstrated already for RRMS www.multiplesclerosisjournal.com

433

patients a small, but significant impairment in OA, ie, in action control, an impairment, which, however, becomes more pronounced when MS progresses into its secondary chronic variant. Most of the patients in our study were in the very early stages of the disease. More than half of our patients (n/26) had an EDSS score of 2 or less, and an additional 11 had a score of 2.5 or 3. This fact, together with the inclusion of patients with a secondary chronic course in the previous studies [5,10], may explain why we were unable to replicate correlations of performance in the PASAT with data on brain atrophy. At least one recent study showed that an investigation of regional BPF focusing on the frontal lobe [43,44], might be more promising in detecting such a correlation. In our study, clinical and motor impairment correlated weakly or moderately with cognitive performance. Several other investigations have shown that clinical progression measured by the EDSS is not firmly related to PASAT [45,46]. Of all cognitive functions, memory performance was of specific importance in correlating with EDSS, endorsing the view of an early detrimental impact of the disease process on this function, as has been argued in several reviews [47
51]. There was also a correlation between the OA task and the EDSS, NHPT and the TWT. In a previous study, we demonstrated that the OA task is sensitive for cognitive impairment in RRMS when memory and attention performance remain intact [41]. None of the clinical or motor grading scales correlated with the self-estimated level of depression, fatigue and quality of life. Similar results have been reported by several other groups [6,52
54]. It shows that feelings of suffering or mental wellbeing in the early state of the disease are independent of the at that point weak motor impairments, but correlate highly with depression [55]. In later stages of the disease, brain atrophy and motor impairment seem to become more important for quality of life [56
59]. Several cognitive tests, but neither BPF nor VF, correlated with fatigue and depression experienced by patients. A similar lack of correlation between depression and diffuse brain atrophy has also been shown by Mohr et al . [60], and Feinstein et al. [61]. As for cognitive functions, it may be possible that the investigation between regional brain atrophy and mood would yield more promising results [1,61]. The relation between depression and cognitive dysfunction that we found, was also a result of the study by Cutajar et al. [62]. Together with the OA task, the PASAT
which has become one subtest of the MSFC
showed the strongest correlation with both fatigue and depression. Therefore, monitoring disease progression with the MSFC also reflects, in some respect, the affective Multiple Sclerosis 2006; 12: 428
436

434

H Hildebrandt et al.

response and subjective representation of MS, as shown in other studies [63]. Because the OA task also correlated with quality of life aspects and with clinical grading of disease progression, future studies should look for the competitive relation of this test and the PASAT in monitoring disease activity. In general, this study shows that subclinical alterations of memory functions may be present very early in RRMS and that these alterations correlate with relative ventricular size. Memory performance was also related to clinical disease grading with the EDSS, TWT, NHPT, executive functions and, to some degree, with depression. Therefore, it was the most sensitive and general neurocognitive marker in this study, relating at the same time to biological, clinical and emotional aspects of RRMS. However, neither brain atrophy nor clinical disease grading correlated with fatigue, depression and mental quality of life, showing at least in the early stages, a high interindividual variety of subjective disease representation, which cannot be reduced to a biological level alone. Finally, there are some short-comings to our study, which should be mentioned. First, we only included list learning for memory assessment to keep investigation time within reasonable time limits. Future studies may focus more exclusively on memory functions and should also include prose recall and prospective memory, which are of considerable importance in every day life. Second, we relied only on the population norms of the neuropsychological tests and did not include a control group. Third, lesion load was not assessed separately from brain atrophy. Although several studies showed that compared to brain atrophy, lesion load plays a minor role for cognitive functions [64,65], a separate analysis would have endorsed the quality of this study. Therefore, further studies including the analysis of lesion load and comparing MS patients directly to controls are necessary to extend our knowledge of brain atrophy and its relation to the cognitive performance of patients with RRMS.

Acknowledgements This study was partially sponsored by a research grant from Biogen Idec Germany, Sanofi-Aventis Inc., and Serono Inc.

References 1. Benedict RH, Carone DA, Bakshi R. Correlating brain atrophy with cognitive dysfunction, mood disturbances, and personality disorder in multiple sclerosis. J Neuroimaging 2004a; 14: 36S
45S.

Multiple Sclerosis 2006; 12: 428
436

2. Rao SM, Glatt S, Hammeke TA, McQuillen MP, Khatri BO, Rhodes AM et al . Chronic progressive multiple sclerosis. Relationship between cerebral ventricular size and neuropsychological impairment. Arch Neurol 1985; 42: 678
82. 3. Bermel RA, Bakshi R, Tjoa C, Puli SR, Jacobs L. Bicaudate ratio as a magnetic resonance imaging marker of brain atrophy in multiple sclerosis. Arch Neurol 2002; 59: 275
80. 4. Rao SM, Bernardin L, Leo GJ, Ellington L, Ryan SB, Burg LS. Cerebral disconnection in multiple sclerosis. Relationship to atrophy of the corpus callosum. Arch Neurol 1989; 46: 918
20. 5. Kalkers NF, Bergers L, de Groot V, Lazeron RHC, van Walderveen MA, Uitdehaag BM et al . Concurrent validity of the MS Functional Composite using MRI as a biological disease marker. Neurology 2001b; 56: 215
19. 6. Miller DM, Rudick RA, Cutter G, Baier M, Fischer JS. Clinical significance of the multiple sclerosis functional composite. Arch Neurol 2000; 57: 1319
23. 7. Cutter GR, Baier ML, Rudick RA, Cookfair DL, Fisher JS, Petkau J et al . Development of a multiple sclerosis functional composite as a clinical trial outcome measure. Brain 1999; 122: 871
82. 8. Edwards SG, Liu C, Blumhardt LD. Cognitive correlates of supra tentorial atrophy on MRI in multiple sclerosis. Acta Neurol Scand 2001; 104: 214
23. 9. Christodoulou C, Krupp LB, Liang Z, Huang W, Melville P, Roque C et al. Cognitive performance and MR markers of cerebral injury in cognitively impaired MS patients. Neurology 2003; 60: 1793
98. 10. Benedict RH, Weinstock-Guttman B, Fishman I, Sharma J, Tjoa CW, Bakshi R. Prediction of neuropsychological impairment in multiple sclerosis: comparison of conventional magnetic resonance imaging measures of atrophy and lesion burden. Arch Neurol 2004; 61: 226
30. 11. Deloire MSA, Salort E, Bonnet M, Arimone Y, Boudineau M, Amieva H et al . Cognitive impairment as marker of diffuse brain abnormalities in early relapsing remitting multiple sclerosis. J Neurol Neurosurg Psychiatry 2005; 76: 519
26. 12. Pennanen C, Testa C, Laakso MP, Hallikainen M, Helkala E-L, Hanninen T et al. A voxel based morphometry study on mild cognitive impairment. J Neurol Neurosurg Psychiatry 2005; 76: 11
14. 13. Kantarci K, Xu YC, Shiung MM, O’Brien PC, Cha RH, Smith GE et al. Comparative diagnostic utility of different MR modalities in mild cognitive impairment and Alzheimer’s disease. Dement Geriatr Cogn Disord 2002; 14: 198
207. 14. Visser PJ, Verhey FRJ, Hofman PAM, Scheltens P, Jolles J. Medial temporal lobe atrophy predicts Alzheimer’s disease in patients with minor cognitive impairment. J Neurol Neurosurg Psychiatry 2002; 72: 491
97. 15. Visser PJ, Krabbendam L, Verhey FRJ, Hofman PAM, Verhoeven WMA. Brain correlates of memory dysfunction in alcoholic Korsakoff’s syndrome. J Neurol Neurosurg Psychiatry 1999; 67: 774
78. 16. McDonald WI, Compston A, Edan G, Goodkin D, Hartung HD, Lublin FD et al . Recommended diagnostic criteria for multiple sclerosis: guidelines from the international panel on the diagnosis of multiple sclerosis. Annu Neurol 2001; 50: 121
27. 17. Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 1983; 33: 1444
52. 18. Hahn HK, Peitgen HO. IWT
Interactive Watershed Transform: a hierarchical method for efficient interactive

www.multiplesclerosisjournal.com

Memory performance and brain atrophy in MS

19.

20.

21.

22.

23. 24. 25. 26.

27.

28.

29.

30.

31.

32. 33.

34.

35.

36.

37.

and automated segmentation of multidimensional grayscale images. In Medical imaging: image processing. Proc. SPIE 5032 . 2003: 643
53. Lukas C, Hahn HK, Bellenberg B, Rexilius J, Schmid G, Schimrigk SK et al . Sensitivity and reproducibility of a new fast 3D segmentation technique for clinical MR-based brain volumetry in multiple sclerosis. Neuroradiology 2004; 46: 906
15. Santago P, Gage HD. Quantification of MR brain images by mixture density and partial volume modeling. IEEE Trans Med Imaging 1993; 12: 566
74. Hahn HK, Millar WS, Klinghammer O, Durkin MS, Tulipano PK, Peitgen HO. A reliable and efficient method for cerebral ventricular volumetry in pediatric neuroimaging. Methods Inf Med 2004; 43: 376
82. Fischer JS, Jak AJ, Kinker JE, Cutter G, Rudick RA. Administration and scoring manual for the multiple sclerosis functional composite measure (MSFC) . National Multiple Sclerosis Society, 2001. Wechsler D. WAIS-R
Wechsler Adult Intelligence Scale
revised . The Psychological Corporation, 1981. Dahl G. WIP, Handbuch zum Reduzierten Wechsler-Intelligenztest . Bodenheim, 1986. Zimmermann P, Fimm B. Testbatterie zur aufmerksamkeitspru¨fung . PsyTest, 1992. Hildebrandt H, Schwendemann G. Tonische und phasische Alertness
ein oder zwei rechtshemispha ¨rische kortiko-subkortikale Netzwerke? Aktuelle Neurol 2001; 28: 219
27. Lengenfelder J, Chiaravalloti ND, Ricker JH, DeLuca J. Deciphering components of impaired working memory in multiple sclerosis. Cogn Behav Neurol 2003; 16: 28
39. ¨ ne-Otto AI, Willmes Niemann H, Sturm W, Tho von Hinkeldey K. Deutsche Normierung des California Verbal Learning Tests . Leipzig, 2002. Beck AT, Hautzinger M, Bailer M, Worall H, Keller F. BDI-Beck-Depressions-Inventar , second edition. Hogrefe, 1985. Krupp LB, LaRocca NG, Muir-Nash J, Steinberg AD. The fatigue severity scale. Arch Neurol 1989; 46: 1121
23. Bullinger M, Kirchberger J. Der SF-12. In Bullinger M, Kirchberger J eds. Der SF-36-Fragebogen zum Gesundheitszustand . Hogrefe Verlag, 1998: 65
72. Cohen J. Statistical power analysis for the behavioral sciences , second edition. Lawrence Erlbaum, 1998. Pinto A, Francis G. Cognitive correlates of depressive symptoms in hospitalized adolescents. Adolescence 1993; 28: 661
72. Simon JH, Jacobs LD, Campion MK, Rudick RA, Cookfair DL, Herndon RM et al . A longitudinal study of brain atrophy in relapsing multiple sclerosis. Neurology 1999; 53: 139
48. Dalton CM, Brex PA, Jenkins R, Fox NC, Miszkiel KA, Crum WR et al . Progressive ventricular enlargement in patients with clinically isolated syndromes is associated with the early development of multiple sclerosis. J Neurol Neurosurg Psychiatry 2002; 73: 141
47. Fabiano AJ, Sharma J, Weinstock-Guttman B, Munschauer FE 3rd, Benedict RH, Zivadinov R et al . Thalamic involvement in multiple sclerosis: a diffusion-weighted magnetic-resonance imaging study. J Neuroimaging 2003; 13: 307
14. Wylezinska M, Cifelli A, Jezzard P, Palace J, Alecci M, Matthews PM. Thalamic neurodegeneration in relapsing-remitting multiple sclerosis. Neurology 2003; 24: 1949
54.

www.multiplesclerosisjournal.com

435

38. Cifelli A, Arridge M, Jezzard P, Esiri MM, Palace J, Matthews PM. Thalamic neurodegeneration in multiple sclerosis. Annu Neurology 2002; 52: 650
53. 39. Amato MP, Bartolozzi ML, Zipoli V, Portaccio E, Mortilla M. Neocortical volume decrease in relapsingremitting MS patients with mild cognitive impairment. Neurology 2004; 63: 89
93. ¨tze C, Brokate B, Kro ¨ ger B, Schwen40. Kraus JA, Schu demann G, Hildebrandt H. Discriminant analysis of the cognitive performance profile of MS patients differentiates their clinical course. J Neurol 2005, 252: 808
13. 41. Hildebrandt H, Brokate B, Lanz M, Ternes T, Timm C. Exekutivfunktionsleistungen bei patienten mit Multipler Sklerose. Aktuelle Neurol 2003; 30: 118
26. 42. Archibald CJ, Wei X, Scott JN, Wallace CJ, Zhang Y, Metz LM et al. Posterior fossa lesion volume and slowed information processing in multiple sclerosis. Brain 2004; 127: 1526
34. 43. Locatelli L, Zivadinov R, Grop A, Zorzon M. Frontal parenchymal atrophy measures in multiple sclerosis. Mult Scler 2004; 10: 562
68. 44. Sperling RA, Guttmann CRG, Hohol MJ, Warjfield SK, Jakab M, Parente M et al . Regional magnetic resonance imaging lesion burden and cognitive function in multiple sclerosis. Arch Neurol 2001; 58: 115
21. 45. Ozakbas S, Cagiran I, Ormeci B, Idiman E. Correlations between Multiple Sclerosis Functional Composite, Expanded Disability Status Scale and health-related quality of life during and after treatment of relapses in patients with multiple sclerosis. J Neurol Sci 2004; 15: 3
7. 46. Kalkers NF, Bergers E, Castelijns JA, van Walderveen M, Bot JC, Ader HJ et al. Optimizing the association between disability and biological markers in MS. Neurology 2000; 57: 1253
58. 47. Hildebrandt H, Schwendemann G. Kognitive Beeintra ¨chtigungen bei Multipler Sklerose. In Zieger A, Scho ¨ nle P eds. Neurorehabilitation . Hippokampus Verlag, 2004: 237
78. 48. Zakzanis KK. Distinct neurocognitive profiles in multiple sclerosis subtypes. Arch Clin Neuropsychol 2000; 15: 115
36. 49. Brassington JC, Marsh NV. Neuropsychological aspects of multiple sclerosis. Neuropsychol Rev 1998; 8: 43
77. 50. Wishart H, Sharpe D. Neuropsychological aspects of multiple sclerosis: a quantitative review. J Clin Exp Neuropsychol 1997; 19: 810
24. 51. Prosiegel M, Michael C. Neuropsychology and multiple sclerosis: diagnostic and rehabilitative approaches. J Neurol Sci 1993; 115 (Suppl): S51
54. 52. Isaksson AK, Ahlstrom G, Gunnarsson LG. Quality of life and impairment in patients with multiple sclerosis. J Neurol Neurosurg Psychiatry 2005; 76: 64
69. 53. Hoogervorst ELJ, van Winsen LML, Eikelenboom MJ, Kalkers NF, Uitdehaag BM, Polman CH. Comparisons of patient self-report, neurologic examination, and functional impairment in MS. Neurology 2001; 56: 934
37. 54. Rothwell PM, McDowell Z, Wong CK, Dorma PJ. Doctors and patients don’t agree: cross sectional study of patients’ and doctors’ perceptions and assessments of disability in multiple sclerosis. BMJ 1997; 314: 1580
83. 55. Lobentanz IS, Asenbaum S, Vass K, Sauter C, ¨ sch G, Kollegger H et al . Factors influencing quality Klo of life in multiple sclerosis patients: disability, depressive mood, fatigue and sleep quality. Acta Neurol Scand 2004; 110: 6
13.

Multiple Sclerosis 2006; 12: 428
436

436

H Hildebrandt et al.

56. Haupts M, Elias G, Hardt C, Langenbahn H, Obert ¨ hlau D et al. Lebensqualita¨t bei Patienten mit H, Po schubfo ¨ rmiger MS in Deutschland. Nervenarzt 2003; 74: 144
50. 57. Janssens ACJW, van Doorn PA, de Boer JB, van der Meche FGA, Passchier J, Hintzen RQ. Impact of recently diagnosed multiple sclerosis on quality of life, anxiety, depression and distress of patients and partners. Acta Neurol Scand 2003; 108: 389
95. ´ n J, Morales JM, Rivera-Navarro J. 58. Benito-Leo Health-related quality of life and its relationship to cognitive and emotional functioning in multiple sclerosis patients. Eur J Neurol 2002; 9: 497
502. 59. Janardhan V, Bakshi R. Quality of life and its relationship to brain lesions and atrophy on magnetic resonance images in 60 patients with multiple sclerosis. Arch Neurol 2000; 57: 1485
91. 60. Mohr DC, Epstein L, Luks TL, Goodkin D, Cox D, Golberg A et al. Brain lesion volume and neuropsychological function predict efficacy of treatment for depres-

Multiple Sclerosis 2006; 12: 428
436

61.

62.

63.

64.

65.

sion in multiple sclerosis. J Consul Clin Psychol 2004; 71: 1017
24. Feinstein A, Roy P, Lobaugh N, Feinstein K, O’Connor P, Black S. Structural brain abnormalities in multiple sclerosis patients with major depression. Neurology 2004; 62: 586
90. Cutajar R, Ferriani E, Scandellari C, Sabattini L, Trocino C, Marchello LP et al . Cognitive function and quality of life in multiple sclerosis patients. J Neurovirol 2000; 6: S186
90. Miller DM, Rudick RA, Baier M, Cutter G, Doughtery DS, Weinstock B. Factors that predict HealthRelated Quality of Life in patients with relapsing/remitting multiple sclerosis. Mult Scler 2003; 9: 1
5. Paolillo A, Piattella MC, Pantano P, Di Legge S, Caramia F, Russo P et al . The relationship between inflammation and atrophy in clinically isolated syndromes suggestive of multiple sclerosis. J Neurol 2004; 251: 432
39. Barkhof F. The clinico-radiological paradox in multiple sclerosis revisited. Curr Opin Neurol 2002; 15: 239
45.

www.multiplesclerosisjournal.com