Blood Brain Barrier Disruption is More Severe in Neuromyelitis Optica than in Multiple Sclerosis and Correlates with Clinical Disability

Size: px

Start display at page:

Download "Blood Brain Barrier Disruption is More Severe in Neuromyelitis Optica than in Multiple Sclerosis and Correlates with Clinical Disability"

Valentine Goodwin
6 years ago
Views:

1 The Journal of International Medical Research 2012; 40: Blood Brain Barrier Disruption is More Severe in Neuromyelitis Optica than in Multiple Sclerosis and Correlates with Clinical Disability Y TOMIZAWA 1, K YOKOYAMA 1, S SAIKI 1, T TAKAHASHI 2, J MATSUOKA 3 AND N HATTORI 1 1 Department of Neurology, and 3 Clinical Research Centre, Juntendo University Graduate School of Medicine, Tokyo, Japan; 2 Department of Neurology, Tohoku University Graduate School of Medicine, Sendai, Japan OBJECTIVES: This study evaluated blood brain barrier (BBB) integrity, using blood and cerebrospinal fluid (CSF) markers, and assessed the practicality of these markers in the differential diagnosis of neuromyelitis optica (NMO) and multiple sclerosis (MS). METHODS: This was a retrospective observational study of consecutive patients presenting with acute phase NMO or MS (first attack or relapse). Haematological tests (including antiaquaporin-4 antibody levels) and CSF parameters (using primary component analyses) were undertaken; the correlation between BBB permeability and disease severity (by Expanded Disability Status Scale [EDSS] score) was examined. RESULTS: Levels of several markers of BBB permeability were higher in patients with NMO (n = 21) than in those with MS (n = 52). The CSF : serum albumin ratio (AR) was the one of the main differentiators of NMO and MS. Additionally, there was a significant correlation between AR and clinical severity for NMO but not for MS. CONCLUSIONS: Markers of BBB permeability were significantly higher in NMO patients than in MS patients. AR was the best marker for differentiating NMO and MS. Thus, measurement of BBB disruption markers (such as AR) might help to differentiate the diagnosis of NMO and MS in acute clinical settings. KEY WORDS: NEUROMYELITIS OPTICA; MULTIPLE SCLEROSIS; BLOOD BRAIN BARRIER (BBB); NEUROMYELITIS OPTICA-IMMUNOGLOBULIN-G; ANTIAQUAPORIN-4 ANTIBODY Introduction Neuromyelitis optica (NMO) is characterized by optic neuritis and longitudinally extensive transverse myelitis. It is often misdiagnosed as multiple sclerosis (MS), however. Serum levels of autoantibody against aquaporin-4 (AQP4-Ab) or NMOimmunoglobulin (Ig)G are sensitive and specific markers for NMO, 1,2 which enable this disease to be distinguished from MS. Their measurement is difficult and timeconsuming to perform in the clinic, however. Several studies have demonstrated that binding of AQP4-Ab/NMO-IgG with astrocytic AQP4 initiates multiple potentially neuropathogenic mechanisms including complement activation, water, potassium and glutamate homeostasis, plasma protein 1483

2 and granulocyte influx, antibody-dependent cell-mediated cytotoxicity, and increased blood brain barrier (BBB) permeability. 3 Breakdown of the BBB, associated with perivascular oedema and inflammatory cell infiltration, is a hallmark of both NMO and MS. 4,5 Consequently, the evaluation of these features has been insufficient to differentiate between BBB dysfunction in NMO and clinical MS. Likewise, the association between BBB breakdown and clinical disability in patients with MS or NMO has not been examined. Thus, the present study used blood tests and cerebrospinal fluid (CSF) analyses to evaluate the significance of BBB permeability in diagnosing patients with NMO and MS. Patients and methods STUDY POPULATION This study included consecutive patients with NMO or MS who presented to the Department of Neurology, Juntendo University Hospital, Tokyo, Japan, between January 1998 and December 2008, because they were experiencing an acute-phase attack (including those experiencing a first attack, or those experiencing a relapse, in the case of patients with established disease). The acute phase was defined as the period within 1 month from the development of an acute attack. For MS patients, a single attack does not fulfil the diagnostic criteria; for NMO patients, the number of attacks is not included in the diagnostic criteria. Diagnosis of NMO was retrospective, based on the revised Wingerchuk s criteria. 6 Serum AQP4-Ab levels were used in place of NMO-IgG seropositive status to diagnose NMO, because both are considered to have the same diagnostic value. 7 9 MS was diagnosed retrospectively, based on the revised McDonald criteria. 10 As patients with MS sometimes fulfil the diagnostic criteria for both NMO and MS, those who also met the diagnostic criteria for NMO were excluded from MS diagnosis and were, therefore, considered to be NMO patients. Patients with clinical signs of coexisting infectious disease were excluded, as were those with primaryprogressive MS, as it was not possible to detect an acute phase for these patients. Patients were followed until December This study was approved by the Ethics Committee of Juntendo University School of Medicine, Tokyo, Japan (No. 205), and written informed consent was obtained from each patient for the use of their AQP4-Ab and CSF data. BLOOD AND CSF ANALYSES Evaluations of CSF and blood samples were performed for all patients during the acute phase of the disease. Patients did not receive intravenous (i.v.) corticosteroid infusion therapy, plasma exchange or i.v. Ig before blood and CSF samples were taken. The CSF samples were taken by routine lumbar puncture. Blood samples were taken and processed at the biochemical laboratory of Juntendo University Hospital, where serum was prepared within 30 min of sampling by centrifugation at 1200 g for 5 min. CSF and serum samples for everything other than analysis of AQP4-Ab levels were analysed immediately or stored at 4 C prior to routine laboratory testing. The serum AQP4-Ab assay was performed as detailed elsewhere, 7 at the Department of Neurology, Tohoku University Graduate School of Medicine, Sendai, Japan. Serum samples for the AQP4-Ab assay only were frozen immediately and transported to Tohoku University on dry ice; all samples were assayed within 1 week. Samples were not blinded, but were assayed without prior knowledge of clinical information. 1484

3 The CSF cell count, total protein level and IgG index were examined externally at the Japanese Special Reference Laboratory (Tokyo, Japan), according to standard laboratory practices. Methods described elsewhere were used to assess the following: CSF myelin basic protein (MBP); 11 Tourtellotte s albumin leakage rate (TAL) 12 and CSF : serum albumin ratio (AR). 12,13 As AR is age dependent, 14,15 the individual agerelated AR (Q diff ) was determined according to Brettschneider et al. 14 as Q diff = AR (4 + age/15) AR, Q diff and TAL were used as surrogate markers of BBB permeability. Although the measureable range of CSF MBP levels is deemed to be between 40 and 2000 pg/ml, if MBP levels were > 2000 pg/ml they were considered as 2000 pg/ml. Oligoclonal IgG bands (OCBs) were measured by isoelectric focusing at Mitsubishi Kagaku Bio-Clinical Laboratories, Tokyo, Japan. ASSESSMENT OF CLINICAL SYMPTOMS All acute phase attacks were assessed clinically and radiologically, and classified as one of four types in terms of distribution: hemisphere; brainstem; spinal cord; optic nerve. The clinical condition of each NMO and MS patient was evaluated, based on the criteria of the Expanded Disability Status Scale (EDSS), 16 by a neurologist at the patient s last outpatient clinic visit up to December STATISTICAL ANALYSES Categorical variables were compared using Fisher s exact test. Continuous variables were compared using a Student s two-sample t-test, or Mann Whitney U-test for nonparametric data. Spearman s rank correlation coefficients were calculated to test the relationship between laboratory data and EDSS score. The result was considered statistically significant if the P-value was < 0.05 (two-tailed). These statistical calculations were performed using SYSTAT 11.0 Japanese version (HULINKS, Tokyo, Japan). To evaluate clinical markers that differentiated NMO from MS, multivariate analysis was carried out using a principal component analysis (PCA) to correct for multiplicity. Using PCA, the relationships between clinical markers were analysed and the identification and grouping of key variables was performed. Principal components were adopted to surpass 70% of the cumulative contribution ratio. From the principal components, eigenvectors and a loading plot for each component were created and analysed, and factors that were useful to differentiate between NMO and MS were examined. PCA was performed using JMP statistical software, version (SAS Institute, Cary, NC, USA). Results A total of 73 patients were examined: 21 NMO patients (one male and 20 females) and 52 MS patients (15 males and 37 females). Patient characteristics are shown in Table 1. Mean age, percentage of females, disease duration, EDSS score and percentage of patients positive for AQP4-Ab were all significantly higher in the NMO group than in the MS group. Conversely, the number of patients with positive OCBs was significantly lower in the NMO group than in the MS group (P < 0.05 for all comparisons; Table 1). The CSF characteristics are summarized in Table 2. There were no statistically significant differences in lesion distribution in the central nervous system between NMO and MS patients. CSF cell count, CSF total protein, CSF MBP, AR, Q diff, and TAL, were all significantly higher in the NMO group than in the MS group, whereas the IgG index was 1485

4 TABLE 1: Baseline characteristics at study entry for patients with neuromyelitis optica (NMO) or multiple sclerosis (MS) who were evaluated for blood brain barrier integrity during the acute phase following acute attack (first attack or relapse) Characteristics NMO (n = 21) MS (n = 52) Statistical significance a Females 20 (95.2) 37 (71.2) P = Age, years 51.7 ± ± 11.5 P < Age at onset, years 35.6 ± ± 10.7 NS Disease duration, years 13.8 ± ± 6.31 P = Annual relapse rate 1.09 ± ± 0.50 NS EDSS score 5.83 ± ± 2.02 P < Positive OCB b 4/20 (20.0) 24/43 (55.8) P = Positive for AQP4-Ab 20/20 (100.0) 0/43 (0.0) P < Data presented as mean ± SD or n (%) of patients. a Categorical variables compared using Fisher s exact test; continuous variables compared using Student s twosample t-test or Mann Whitney U-test for nonparametric data. b OCB status for one NMO patient and nine MS patients was not examined. EDSS, Expanded Disability Status Scale; OCB, oligoclonal bands; AQP4-Ab, antiaquaporin-4 antibodies; NS, not statistically significant (P 0.05). TABLE 2: Cerebrospinal fluid (CSF) and blood analyses for patients with neuromyelitis optica (NMO) or multiple sclerosis (MS) who were evaluated for blood brain barrier integrity during the acute phase following acute attack (first attack or relapse) NMO MS Statistical Parameter n = 21 n = 52 significance a EDSS score during acute attack, nadir 5.40 ± ± 1.26 P < Distribution of lesions during acute attack Hemisphere 3 (14.3) 5 (9.6) NS Brainstem 3 (14.3) 17 (32.7) NS Spinal cord 12 (57.1) 21 (40.4) NS Optic nerve 3 (14.3) 9 (17.3) NS CSF cell count, cell No./µl 52.8 ± ± 6.94 P = CSF total protein, mg/dl 98.0 ± ± 10.5 P = CSF myelin basic protein, pg/ml 900 ± ± 346 P < CSF : serum albumin ratio (AR), ± ± 1.71 P = Age-related AR (Q diff ), ± ± 37.1 P = TAL rate, mg/day 197 ± ± 37.1 P = IgG index ± ± P = Data presented as mean ± SD or n (%) of patients. a Categorical variables compared using Fisher s exact test; continuous variables compared using Student s twosample t-test or Mann Whitney U-test for nonparametric data; NS, not statistically significant (P 0.05). EDSS, Expanded Disability Status Scale; TAL, Tourtellotte s albumin leakage, Ig, immunoglobulin. significantly higher in the MS group than in the NMO group (P < 0.05 for all comparisons). The cumulative contribution ratio of four components of the total variance exceeded 70% and reached 71.5%; thus, PCA analysis was conducted. Table 3 gives the eigenvectors for components 1 4 and Fig

5 shows the factor loading plot for components 1 4. From Fig. 1, the first principal component was the most valuable to differentiate between NMO and MS. As a result of PCA, AR had the highest eigenvector in the first principal component. Figure 2 shows scatter plots of EDSS during relapse (nadir) and AR of NMO and MS patients. Disease severity for NMO, but not for MS, was statistically significantly correlated with AR (Spearman s rank correlation coefficients: r = [P = 0.028] and r = [not significant], respectively). Discussion Early diagnosis and treatment are important to reduce morbidity from NMO. Differentiating between the diagnosis of NMO and MS is not difficult when a patient presents with typical symptoms such as acute transverse myelitis, with magnetic resonance imaging (MRI) evidence of a spinal cord lesion extending over three vertebral segments or more, and severe optic neuritis. Many cases can, however, be difficult to diagnose if such symptoms are not present. Although serum NMO- IgG/AQP4-Ab has been established as a useful diagnostic marker, the test is available at a limited number of facilities and is expensive. Analysis of the CSF, which is an easy routine examination, has not previously been fully evaluated for its significance in NMO diagnosis. 4,17,18 The present study recruited patients with NMO and MS who had baseline characteristics that were typical of those expected in people with these diseases. 1,8 However, an apparent difference between NMO and MS, in terms of BBB disruption was found. PCA was used to analyse this association further, and the first principal component primarily comprised markers of BBB permeability (AR, TAL), CSF total protein, CSF cell count and EDSS score. Levels of these parameters increase in inflammatory states. Thus, the first principal component was considered to be a comprehensive indicator of inflammation 19 and was able to distinguish between NMO TABLE 3: Eigenvectors for each component in a principal component analysis to evaluate clinical markers that differentiated between neuromyelitis optica (NMO) or multiple sclerosis (MS) Variable Component 1 Component 2 Component 3 Component 4 Age NMO or MS Disease duration EDSS score (final visit) EDSS score (nadir) CSF cell count CSF total protein CSF myelin basic protein IgG index CSF : serum albumin ratio (AR) Age-related AR (Q diff ) TAL rate EDSS, Expanded Disability Status Scale; CSF, cerebrospinal fluid; Ig, immunoglobulin; TAL, Tourtellotte s albumin leakage. 1487

6 1.0 Component Dx Idx Idx Cell TAL MBP AR Q diff Peak EDSS Dur Age Idx Component Dx Q diff Cell Dur Age Peak EDSS MBP TAL AR Age Dur Q diff Cell Peak Dx EDSS AR MBP TAL Component Idx EDSS EDSS Idx EDSS Idx Age Peak Age Peak Age Dx Cell AR Peak Dx Cell 0 AR Dx Cell Dur Dur AR Q TAL TAL Dur diff MBP Q TAL diff MBP Q diff Component 1 Component 2 Component 3 FIGURE 1: Factor loading plot for Components 1 4 for the following variables: age, cerebrospinal fluid (CSF) cell count (Cell); CSF myelin basic protein (MBP); CSF total protein (); CSF : serum albumin ratio (AR); disease duration (Dur); neuromyelitis optica or multiple sclerosis disease (Dx); Expanded Disability Status Scale score at final visit (EDSS); EDSS at nadir (Peak); IgG index (Idx); age-related AR (Q diff ); ; Tourtellotte s albumin leakage rate (TAL) and MS, indicating that the severity of inflammation may be the main difference between these diseases. Overall, AR had the greatest eigenvector in the first principal component, and clinical severity assessed by EDSS and AR showed a mild correlation in NMO that was not seen in MS. Although AR can be measured easily, this test is inadequately applied in daily medical practice. A correlation between EDSS score and glial fibrillary acidic protein (GFAP) levels in CSF has been reported elsewhere. 20 It would be interesting to know whether there is a relationship between CSF GFAP levels and BBB disruption markers, such as AR. The AR was considered especially important for clinical assessment because it is an indicator of BBB disruption, and the BBB is considered to be one of the main targets of AQP4-Ab. 21 The BBB consists of capillary 1488

7 NMO MS 60 AR ( 10 3 ) EDSS score (nadir) during relapse FIGURE 2: Scatter plots showing Expanded Disability Status Scale (EDSS) scores during the acute phase following acute attack (first attack or relapse [nadir]) and cerebrospinal fluid : serum albumin ratios (AR) for neuromyelitis optica (NMO; n = 21) and multiple sclerosis (MS; n = 52) patients. EDSS score was statistically significantly correlated with AR for NMO but not for MS; Spearman s rank correlation coefficients were r = (P = 0.028) and r = (not significant), respectively endothelial cells attached to each other by tight junctions, basal membrane and astrocyte endfeet, where AQP4 is highly localized. 22 Antibody antigen reactions, which are derived from AQP4-Ab-induced dysfunction of AQP4, lead to insufficient water elimination, increased BBB permeability and insufficient glutamate homeostasis. 21 Several radiological studies have shown that optic nerve and spinal cord lesions have gadolinium enhancement in NMO, with the exception of large hemispheric acute disseminating encephalomyelitis lesions Likewise, cranial MRI examinations with intravenous gadolinium infusion, which have been undertaken in AQP4-Ab-seropositive NMO cases, have shown multiple patchy enhancing lesions with blurred margins in the brain parenchyma. 26 Although NMO- IgG-positive serum induces a significant increase in BBB permeability in BBB models in vitro, 3 it remains unclear whether the phenomenon is primary or secondary. The present study showed that BBB was more disrupted in patients with NMO than in those with MS, and also showed a correlation between the level of the BBB disruption and disease severity. Taken together, it could be concluded that BBB disruption is a common characteristic of NMO, and that increased BBB permeability (as measured by AR) could 1489

8 be a useful marker for NMO and for the differential diagnosis of this condition from MS. Acknowledgement This study was supported in part by a Grant from Takeda Scientific Foundation, Osaka, Japan. Conflicts of interest The authors had no conflicts of interest to declare in relation to this article. Received for publication 12 December 2011 Accepted subject to revision 29 December 2011 Revised accepted 27 June 2012 Copyright Field House Publishing LLP References 1 Lennon VA, Wingerchuk DM, Kryzer TJ, et al: A serum autoantibody marker of neuromyelitis optica: distinction from multiple sclerosis. Lancet 2004; 364: Lennon VA, Kryzer TJ, Pittock SJ, et al: IgG marker of optic spinal multiple sclerosis binds to the aquaporin-4 water channel. J Exp Med 2005; 202: Vincent T, Saikali P, Cayrol R, et al: Functional consequences of neuromyelitis optica-igg astrocyte interactions on blood brain barrier permeability and granulocyte recruitment. J Immunol 2008; 181: Bergamaschi R: Importance of cerebrospinal fluid examination in differential diagnosis of Devic s neuromyelitis optica by multiple sclerosis. Neurol Sci 2003; 24: Wingerchuk DM, Lucchinetti CF: Comparative immunopathogenesis of acute disseminated encephalomyelitis, neuromyelitis optica, and multiple sclerosis. Curr Opin Neurol 2007; 20: Wingerchuk DM, Lennon VA, Pittock SJ, et al: Revised diagnostic criteria for neuromyelitis optica. Neurology 2006; 66: Takahashi T, Fujihara K, Nakashima I, et al: Anti-aquaporin-4 antibody is involved in the pathogenesis of NMO: a study on antibody titre. Brain 2007; 130: Matsuoka T, Matsushita T, Kawano Y, et al: Heterogeneity of aquaporin-4 autoimmunity and spinal cord lesions in multiple sclerosis in Japanese. Brain 2007; 130: Sellner J, Boggild M, Clanet M, et al: EFNS guidelines on diagnosis and management of neuromyelitis optica. Eur J Neurol 2010; 17: Polman CH, Reingold SC, Edan G, et al: Diagnostic criteria for multiple sclerosis: 2005 revisions to the McDonald Criteria. Ann Neurol 2005; 58: Ohta M, Ohta K, Ma J, et al: Clinical and analytical evaluation of an enzyme immunoassay for myelin basic protein in cerebrospinal fluid. Clin Chem 2000; 46: Syndulko K, Tourtellotte WW, Conrad AJ, et al: Trans-blood brain barrier albumin leakage and comparisons of intrathecal IgG synthesis calculations in multiple sclerosis patients. Multiple Sclerosis Study Group, Alpha Interferon Study Group, and Azathioprine Study Group. J Neuroimmunol 1993; 46: Link H, Tibbling G: Principles of albumin and IgG analyses in neurological disorders. III. Evaluation of IgG synthesis within the central nervous system in multiple sclerosis. Scand J Clin Lab Invest 1977; 37: Brettschneider J, Claus A, Kassubek J, et al: Isolated blood cerebrospinal fluid barrier dysfunction: prevalence and associated diseases. J Neurol 2005; 252: Rapoport SI, Schapiro MB, May C: Reduced brain delivery of homovanillic acid to cerebrospinal fluid during human aging. Arch Neurol 2004; 61: Kurtzke JF: Rating neurologic impairment in multiple sclerosis: an Expanded Disability Status Scale (EDSS). Neurology 1983; 33: Ghezzi A, Bergamaschi R, Martinelli V, et al: Clinical characteristics, course and prognosis of relapsing Devic s neuromyelitis optica. J Neurol 2004; 251: Zaffaroni M, Italian Devic s Study Group: Cerebrospinal fluid findings in Devic s neuromyelitis optica. Neurol Sci 2004; 25(suppl 4): S368 S Jolliffe IT (ed): Principal Component Analysis, 2nd edn. Berlin: Springer, 2002; pp Takano R, Misu T, Takahashi T, et al: Astrocytic damage is far more severe than demyelination in NMO: a clinical CSF biomarker study. Neurology 2010; 75: Hinson SR, McKeon A, Lennon VA: Neurological autoimmunity targeting aquaporin-4. Neuroscience 2010; 168: Tait MJ, Saadoun S, Bell BA, et al: Water movements in the brain: role of aquaporins. Trends Neurosci 2008; 31: Wingerchuk DM, Weinshenker BG: Neuromyelitis optica. Curr Treat Options Neurol 2008; 10:

9 24 Eichel R, Meiner Z, Abramsky O, et al: Acute disseminating encephalomyelitis in neuromyelitis optica: closing the floodgates. Arch Neurol 2008; 65: Saiki S, Ueno Y, Moritani T, et al: Extensive hemispheric lesions with radiological evidence of blood brain barrier integrity in a patient with neuromyelitis optica. J Neurol Sci 2009; 284: Ito S, Mori M, Makino T, et al: Cloud-like enhancement is a magnetic resonance imaging abnormality specific to neuromyelitis optica. Ann Neurol 2009; 66: Author s address for correspondence Dr Kazumasa Yokoyama Department of Neurology, Juntendo University Graduate School of Medicine, Hongo, Bunkyo-ku, Tokyo , Japan. kazumasa@juntendo.ac.jp 1491

ORIGINAL CONTRIBUTION

ORIGINAL CONTRIBUTION Markedly Elevated Soluble Intercellular Adhesion Molecule 1, Soluble Vascular Cell Adhesion Molecule 1 Levels, and Blood-Brain Barrier Breakdown in Neuromyelitis Optica Akiyuki Uzawa,