J Neurol (2014) 261:277–282 DOI 10.1007/s00415-013-7181-y
ORIGINAL COMMUNICATION
Prevalence of cortical superficial siderosis in patients with cognitive impairment Frank Arne Wollenweber • Katharina Buerger • Claudia Mueller Birgit Ertl-Wagner • Rainer Malik • Martin Dichgans • Jennifer Linn • Christian Opherk
•
Received: 24 September 2013 / Revised: 29 October 2013 / Accepted: 30 October 2013 / Published online: 13 November 2013 Ó Springer-Verlag Berlin Heidelberg 2013
Abstract Cortical superficial siderosis (cSS) is a magnetic resonance imaging marker of cerebral amyloid angiopathy (CAA) and can be its sole imaging sign. cSS has further been identified as a risk marker for future intracranial hemorrhage. Although uncommon in the general population, cSS may be much more prevalent in high risk populations for amyloid pathology. We aimed to determine the frequency of cSS in patients with cognitive impairment presenting to a memory clinic. We prospectively evaluated consecutive patients presenting to our memory clinic between April 2011 and April 2013. Subjects received neuropsychological testing using the Consortium to Establish a Registry for Alzheimer’s Disease battery (CERAD-NP). Two hundred and twelve patients
with documented cognitive impairment further underwent a standardized 3T-MR-imaging protocol with T2*-weighted gradient-echo sequences for detection of cSS. Thirteen of 212 patients (6.1 %) displayed cSS. In seven of them (54 %) cSS was the only imaging sign of CAA. Patients with cSS did not differ from patients without cSS with regard to medical history, age or cardiovascular risk profile. Subjects with cSS performed worse in the mini-mental state examination (p = 0.001), showed more white matter hyperintensities (p = 0.005) and more often had microbleeds (p = 0.001) compared to those without cSS. cSS is common in patients with cognitive impairment. It is associated with lower cognitive scores, white matter hyperintensities and microbleeds and can be the only imaging sign for CAA in this patient group.
J. Linn and C. Opherk contributed equally to this work.
Keywords Cerebral amyloid angiopathy Superficial siderosis MRI Cognitive decline
F. A. Wollenweber (&) K. Buerger C. Mueller R. Malik M. Dichgans C. Opherk Institute for Stroke and Dementia Research, Ludwig-Maximilians-University, Marchioninistr. 15, 81377 Munich, Germany e-mail: [email protected] B. Ertl-Wagner Institute of Clinical Radiology, Ludwig-Maximilians-University, Munich, Germany M. Dichgans Munich Cluster for Systems Neurology (SyNergy), Ludwig-Maximilians-University, Munich, Germany J. Linn Department of Neuroradiology, Ludwig-Maximilians-University, Munich, Germany C. Opherk Department of Neurology, SLK-Kliniken Heilbronn, Heilbronn, Germany
Introduction The diagnosis of cerebral amyloid angiopathy (CAA) is typically based on magnetic resonance imaging (MRI) markers such as residues of macro-hemorrhages and presence of lobar cerebral microbleeds (MB) using the Boston criteria [1]. Recently, ‘‘cortical superficial siderosis’’ (cSS) was proposed as an additional imaging marker [2] and confirmed to be highly prevalent in CAA patients [3, 4]. cSS is best detected on T2*-weighted gradient-echo sequences as linear residues of blood in the superficial layers of the cerebral cortex (Fig. 1a, b) and generally assumed to reflect recurrent bleedings into the subarachnoid space [2]. With an estimated prevalence of about 0.7 %, the frequency of cSS in the elderly general population is low [5]. However, in
123
278
J Neurol (2014) 261:277–282
Study population One thousand and seventy consecutive patients presenting to a specialized memory clinic of a tertiary level hospital between April 2011 and April 2013 were screened for study inclusion. All patients were assessed by structured medical history and received a complete physical and neurologic examination together with detailed cognitive assessment, laboratory tests, and measurements of blood pressure and body weight. Three hundred and thirty-two (31 %) of the 1,070 patients had a normal cognitive testing as described below and were, therefore, not eligible. Another 66 (6 %) patients fulfilled one of the exclusion criteria. Of patients fulfilling the inclusion criteria, 441 did not undergo standardized MRI because imaging had recently (\6 months) been performed by an outside institution or because the patient declined MRI. Two hundred and thirty-one patients underwent standardized MRI. Nineteen of these were excluded because of insufficient image quality, e.g. due to motion artifacts, resulting in a final study cohort of 212 patients. Fig. 1 Imaging signs in patients with cortical superficial siderosis presenting to a memory clinic. T2*-weighted gradient-echo sequences: a MCI patient with focal occipital cortical superficial siderosis (cSS) and a single lobar microbleed (insert) b Dementia patient with cSS (insert) but without microbleeds c Dementia patient with cSS (not shown) and multiple lobar microbleeds d Fluid attenuated inversion recovery (FLAIR) sequence: patient with cSS (not shown) and without cardiovascular risk factors showing extensive white matter hyperintensities
selected patient groups such as patients with cognitive impairment, prevalence rates may be much higher. cSS has recently shown to be a predictor for future intracranial hemorrhages, frequently at the location of pre-existing cSS [6]. Thus, the detection of cSS might have clinical implications, e.g. with regard to antithrombotic treatment. In the current study we aimed to investigate the prevalence of cSS in the setting of a memory clinic, as those patients are more likely to suffer from amyloid pathology. We further sought to determine whether patients with cSS differ from patients without cSS with regard to medical history, cardiovascular risk profile, cognitive measures and other imaging markers of CAA.
Methods The study has been approved by the appropriate ethics committee and have therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. All persons gave their informed consent prior to their inclusion in the study.
123
Inclusion and exclusion criteria Inclusion criteria were age older than 50 years and a diagnosis of either mild cognitive impairment (MCI) based on the Petersen criteria [7], or dementia according to ICD10 criteria. Exclusion criteria were severe other neurologic or psychiatric conditions that could interfere with the cognitive status (e.g. brain tumor, advanced stages of multiple sclerosis or Parkinson’s disease) and contraindications for MRI such as pacemaker or severe claustrophobia. Neuropsychological assessment Subjects were screened by the mini-mental state examination (MMSE) [8] and went on to receive detailed neuropsychological assessment if the MMSE score was above ten. Detailed testing included the Consortium to Establish a Registry for Alzheimer’s Disease battery (CERAD-NP) [9] with the following subtests: ‘‘Word List Learning/Recall and Recognition’’, ‘‘Boston Naming Test’’, ‘‘Figure Copy and Recall’’, ‘‘Semantic and Phonemic Fluency’’, ‘‘Trail Making Test Part A and B’’. The Geriatric Depression Scale (GDS) [10] was used to screen for depressive symptoms. Diagnoses were made by an interdisciplinary team including a neurologist, psychiatrist, and neuropsychologist. Assessment of cardiovascular risk factors Hypertension (HTN) was diagnosed if the patient either had a history of arterial HTN or was taking
J Neurol (2014) 261:277–282
antihypertensive drugs. Obesity was defined by a body mass index [30. Dyslipidemia was defined by a LDLcholesterol[130 mg/dl or a known history of dyslipidemia together with a prescription of lipid-lowering drugs. Diabetes mellitus was defined by history, a documented serum HbA1c [6.5 % or anti-diabetic medication. Neuroimaging acquisition and analysis MRI was performed on a 3 Tesla system (Magnetom Verio, Siemens Healthcare, Erlangen, Germany) using the following acquisition protocol that was optimized for the detection of small-vessel-related lesions [11]: (1) axial diffusion-weighted sequence [echo time (TE) = 110 ms, repetition time (TR) = 5,900 ms, slice thickness (ST) = 4 mm, matrix = 192 9 192], (2) axial fluidattenuated inversion recovery (FLAIR)-sequence [TE = 94 ms, TR = 7,000 ms, inversion time (TI) = 2,210 ms, ST = 3 mm, matrix = 256 9 2,204], (3) axial T2*weighted gradient-echo sequence (TE = 20 ms, TR = 1,020 ms, ST = 3 mm, matrix = 256 9 154), (4) axial T2-weighted turbo-spin-echo sequence (TE = 94 ms, TR = 6,560 ms, ST = 3 mm, matrix = 320 9 280), (5) sagittal three-dimensional magnetization prepared rapid acquisition gradient-echo (MPRAGE) T1-weighted sequence (TE = 4.76 ms, TR = 11 ms, ST = 1 mm, matrix = 256 9 256) with coronal and axial multi-planar reconstructions. Image analysis was performed by two experienced neuroradiologists in consensus using a standardized evaluation sheet. Readers were blinded to all clinical and patient information data. All T2*-weighted gradient-echo images were systematically assessed for evidence of cSS, intracerebral hemorrhage (ICH, defined as parenchymal hemorrhages [5 mm in diameter), and microbleeds (MB, defined as parenchymal hemorrhages B5 mm in diameter) [11]. cSS was defined as profoundly hypointense linear structures within the subarachnoid space or in the superficial layers of the cerebral cortex on T2*-weighted gradient-echo images (Fig. 1a, b). Its presence and its extent were noted and classified as focal (restricted to B3 adjacent sulci) or disseminated. In addition, we recorded the presence, number and location of ICH and MB. FLAIR images were used to identify acute ICH and subarachnoid hemorrhages and to distinguish them from old ICH and chronic cSS. The presence and extent of WMH was rated using the Wahlund Scale [12]. Statistical analysis Cardiovascular risk factors were assessed by constructing a 2 9 2 contingency table and were tested using Fisher’s exact test. Neuroimaging parameters were evaluated using either Fisher’s exact test (variables with two levels) or a v2
279
test (variables with more than two levels). Neuropsychological test results were evaluated using a WilcoxonMann–Whitney test to compare patients with and without cSS. The predictive value of neuropsychological tests (predictor variable) on cSS status (dependent variable) was evaluated using logistic regression using age and sex as covariates. A significance level of p \ 0.05 was applied in all analyses.
Results The demographic and clinical characteristics of the 212 patients included in the analysis are provided in Table 1. Thirteen patients (6.1 %) showed cSS, which was focal in seven (54 %) and disseminated in six (46 %) subjects. The majority (92 %, n = 12) of patients with cSS had no history of ICH. Overall, 128 patients (60 %) met the Petersen criteria for MCI [7], and 84 (40 %) were diagnosed as being demented according to ICD-10 criteria. The proportion of patients with dementia was similar in the cSS group compared to the group without cSS (p = 0.41). Patients with cSS did not differ significantly from those without cSS with respect to age, history of stroke, cardiovascular risk factors, and intake of antithrombotic drugs, while there was a slight preponderance of males in the group with cSS (p = 0.04, Table 1). cSS was associated with lower MMSE scores (p = 0.001), more severe WMH as measured by the Wahlund-score (p = 0.005), and with the presence of MB (p = 0.0001), which were also more often multiple [[10 MB (p = 0.02), Table 2]. However, there was no significant difference in MB localization (lobar versus deep versus cerebellar) in patients with cSS compared to those without cSS (Table 2). The CERAD-NP revealed no significant differences between the two groups nor were there differences in depressive symptoms as determined by the GDS.
Discussion The main findings of this study are that (1) cSS is common in patients with cognitive impairment; (2) it can be the only imaging sign for presumed CAA in this patient group; and (3) cSS may independently be associated with poor overall cognitive performance. Thus, cSS may be a valuable imaging marker in this group of patients. The prevalence of cSS in our sample of patients with cognitive impairment was 6.1 %, which is notably higher than the prevalence of cSS in the general population as reported from the Rotterdam Scan Study [5]. Apart from
123
280
J Neurol (2014) 261:277–282
Table 1 Clinical characteristics of patients with and without cortical superficial siderosis Total
Cortical superficial siderosis Not present
Present 13
p*
Number of patients
212
199
Age in years, mean (SD)
74 (7.5)
74 (7.7)
73 (3.2)
0.32
Female, n (%)
123 (58)
119 (60)
4 (31)
0.04
Dementia, n (%)
84 (40)
78 (39)
6 (46)
0.41
History of ICH, n (%)
3 (1)
2 (1)
1 (7)
0.17
History of ischemic stroke, n (%)
33 (16)
30 (15)
3 (23)
0.43
Hypertension
135 (63)
126 (63)
9 (69)
0.77
Diabetes mellitus
24 (11)
24 (12)
0 (0)
0.37
Dyslipidemia
91 (43)
88 (44)
3 (23)
0.15
Obesity
24 (11)
24 (12)
0 (0)
0.37
Cardiovascular risk factors, n (%)
Cognitive testing MMSE, median (IQR)
26 (23–28)
CERAD-NP composite Z-Score, mean (SD)
-1.29 (0.88)
26 (23–28)
-1.27 (0.88)
23 (7–27) à
-1.48 (0.78)
0.001 à
0.29
Medication, n (%) Antiplatelet drugs
61 (29)
57 (29)
4 (30)
0.81
Oral anticoagulation
15 (7)
13 (6)
2 (15)
0.23
Antihypertensive drugs
103 (49)
98 (49)
5 (38)
0.57
Imaging markers Wahlund Scale, mean (SD)
1.2 (0.7)
1.1 (0.7)
1.8 (0.8)
0.005
Microbleeds present, n (%)
25 (12)
19 (10)
6 (46)
0.0001
ICH intracerebral hemorrhage, IQR Interquartile range, SD Standard deviation, CERAD-NP consortium to establish a registry of Alzheimer’s disease-neuropsychological battery *p values \ 0.05 are indicated in bold, missing n = 6, àmissing n = 3
Table 2 Relationship between the characteristics of microbleeds and cortical superficial siderosis Total
Number of patients with microbleeds Multiple ([10) microbleeds, n (%) Lobar microbleeds, n (%) Deep microbleeds, n (%) Cerebellar microbleeds, n (%)
Cortical superficial siderosis
p*
Not present
Present
25
19
6
7 (28) 18 (72) 1 (4) 8 (32)
2 (10) 12 (63)
5 (83) 6 (100)
0.02 0.10
1 (5) 6 (32)
0 (0) 2 (33)
0.7 0.6
*p values \ 0.05 are indicated in bold
technical aspects such as differences in image acquisition and image analysis, the enrichment of cSS in our sample might relate to the following factors: First, prevalence rates
123
of CAA increase with age [13] and the mean age in our sample was 74 years, which is slightly higher than the mean age of subjects in the Rotterdam Scan Study (70 years) [5]. Second, our sample was enriched for patients with Alzheimer’s disease (AD) and CAA pathology is common in AD brains [14]. We did not systematically assess amyloid pathology by amyloid imaging or cerebrospinal fluid markers and the number of cases with cSS was too small to perform meaningful analyses in diagnostic subgroups. However, this should not affect our conclusions. Third, CAA may cause cognitive deficits independently from AD pathology especially in more advanced stages and we may have also enriched for patients with isolated CAA [15]. Because of the restricted sample size intergroup comparisons between patients with and without cSS must be interpreted as exploratory analyses. Nevertheless, several interesting aspects can be derived from our data: The clinical diagnosis of CAA typically requires a history of
J Neurol (2014) 261:277–282
lobar ICH as part of the Boston criteria [1]. Of note, however, is that the majority (92 %) of our subjects with cSS had no known history of ICH. As expected, we found MB and WMH to be more frequent in patients with cSS compared to those without (Fig. 1c, d), which underlines the close relationship between cSS and presumed CAA. Interestingly, in about half of our patients with cSS we did not detect any MB. Thus, it can be speculated that cSS may reflect the first sign of CAA in those cases as previously described in the literature [4]. Nevertheless, this hypothesis needs to be proven in histopathological studies. A high prevalence of cSS and, therefore, presumed CAA in patients with cognitive impairment may have implications for risk prediction and clinical management: Patients with CAA-related ICH carry a high risk for recurrent intracranial hemorrhage [16] and CAA has been identified as an important cause of warfarin-associated bleedings [17]. Further, a recent retrospective analysis of 51 patients with cSS identified cSS as a risk factor for future intracranial hemorrhage, especially at the location of preexisting siderosis [6]. Although prospective data in patients with isolated cSS not fulfilling the Boston criteria for CAA are currently lacking, these patients potentially also carry an increased risk for intracranial hemorrhage. In the future, cSS may, therefore, help to identify cognitively impaired patients who are at specific risks and might benefit from specific treatments. We found patients with cSS to have lower MMSE scores than those without. This broadly agrees with observations from a recent longitudinal study [15] that found cognitive deficits especially for perceptual speed in histologically proven CAA cases independent from AD pathology. Interestingly, our study did not show a specific cognitive profile in cSS patients. This might relate to the small number of subjects with cSS (n = 13) and to the fact that severely affected patients were not able to perform the full cognitive battery. Additional studies are needed to determine whether patients with cSS have a distinct cognitive profile. This study has several strengths including the prospective design with standardized cognitive testing and standardized MRI using independent experienced raters and established scales. Limitations include the moderate sample size and a potential selection bias related to patients that were not included because they had already received an outside MRI. Also, we did not determine APO-E status. In conclusion, cSS is common in cognitively impaired patients. Our preliminary analyses suggest an association of cSS with lower cognitive scores, WMH and microbleeds. Further, cSS may reflect the only imaging sign for CAA in a subgroup of patients. Additional studies are needed to explore the clinical value of cSS as an imaging marker.
281 Conflicts of interest of interest.
The authors declare that they have no conflict
References 1. Knudsen KA, Rosand J, Karluk D, Greenberg SM (2001) Clinical diagnosis of cerebral amyloid angiopathy: validation of the Boston criteria. Neurology 56:537–539. doi:10.1212/WNL.56.4. 537 2. Linn J, Bruckmann H (2010) Superficial siderosis in cerebral amyloid angiopathy. AJNR Am J Neuroradiol 31:E29. doi:10. 3174/ajnr.A1913 3. Charidimou A, Jager RH, Fox Z, Peeters A, Vandermeeren Y, Laloux P, Baron JC, Werring DJ (2013) Prevalence and mechanisms of cortical superficial siderosis in cerebral amyloid angiopathy. Neurology 81:626–632. doi:10.1212/WNL.0b013e3182a08f2c 4. Linn J, Halpin A, Demaerel P, Ruhland J, Giese AD, Dichgans M, van Buchem MA, Bruckmann H, Greenberg SM (2010) Prevalence of superficial siderosis in patients with cerebral amyloid angiopathy. Neurology 74:1346–1350. doi:10.1212/ WNL. 0b013e3181dad605 5. Vernooij MW, Ikram MA, Hofman A, Krestin GP, Breteler MM, van der Lugt A (2009) Superficial siderosis in the general population. Neurology 73:202–205. doi:10.1212/WNL.0b013e3181ae7c5e 6. Linn J, Wollenweber FA, Lummel N, Bochmann K, Pfefferkorn T, Gschwendtner A, Bruckmann H, Dichgans M, Opherk C (2013) Superficial siderosis is a warning sign for future intracranial hemorrhage. J Neurol 260:176–181. doi:10.1007/s00415012-6610-7 7. Petersen RC (2004) Mild cognitive impairment as a diagnostic entity. J Intern Med 256:183–194. doi:10.1111/j.1365-2796.2004. 01388.x 8. Folstein MF, Folstein SE, McHugh PR (1975) ‘‘Mini-mental state’’. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12:189–198. doi:10. 1016/0022-3956(75)90026-6 9. Morris JC, Heyman A, Mohs RC, Hughes JP, van Belle G, Fillenbaum G, Mellits ED, Clark C (1989) The consortium to establish a registry for Alzheimer’s disease (CERAD). Part I. Clinical and neuropsychological assessment of Alzheimer’s disease. Neurology 39:1159–1165. doi:10.1212/WNL.39.9.1159 10. Yesavage JA, Brink TL, Rose TL, Lum O, Huang V, Adey M, Leirer VO (1982) Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res 17:37–49. doi:10.1016/0022-3956(82)90033-4 11. Wardlaw JM, Smith EE, Biessels GJ, Cordonnier C, Fazekas F, Frayne R, Lindley RI, O’Brien JT, Barkhof F, Benavente OR, Black SE, Brayne C, Breteler M, Chabriat H, Decarli C, de Leeuw FE, Doubal F, Duering M, Fox NC, Greenberg S, Hachinski V, Kilimann I, Mok V, Oostenbrugge R, Pantoni L, Speck O, Stephan BC, Teipel S, Viswanathan A, Werring D, Chen C, Smith C, van Buchem M, Norrving B, Gorelick PB, Dichgans M, nEuroimaging STfRVco (2013) Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lan Neurol 12:822–838. doi:10.1016/ S1474-4422(13)70124-8 12. Wahlund LO, Agartz I, Almqvist O, Basun H, Forssell L, Saaf J, Wetterberg L (1990) The brain in healthy aged individuals: MR imaging. Radiology 174:675–679. doi:10.1148/radiology.174.3. 2305048 13. Charidimou A, Gang Q, Werring DJ (2012) Sporadic cerebral amyloid angiopathy revisited: recent insights into pathophysiology
123
282 and clinical spectrum. J Neurol Neurosurg Psychiatr 83:124–137. doi:10.1136/jnnp-2011-301308 14. Ellis RJ, Olichney JM, Thal LJ, Mirra SS, Morris JC, Beekly D, Heyman A (1996) Cerebral amyloid angiopathy in the brains of patients with Alzheimer’s disease: the CERAD experience, part XV. Neurology 46:1592–1596. doi:10.1212/WNL.46.6.1592 15. Arvanitakis Z, Leurgans SE, Wang Z, Wilson RS, Bennett DA, Schneider JA (2011) Cerebral amyloid angiopathy pathology and cognitive domains in older persons. Ann Neurol 69:320–327. doi:10.1002/ana.22112
123
J Neurol (2014) 261:277–282 16. O’Donnell HC, Rosand J, Knudsen KA, Furie KL, Segal AZ, Chiu RI, Ikeda D, Greenberg SM (2000) Apolipoprotein E genotype and the risk of recurrent lobar intracerebral hemorrhage. New Engl J Med 342:240–245. doi:10.1056/NEJM20000 1273420403 17. Rosand J, Hylek EM, O’Donnell HC, Greenberg SM (2000) Warfarin-associated hemorrhage and cerebral amyloid angiopathy: a genetic and pathologic study. Neurology 55:947–951. doi:10.1212/WNL.55.7.947
ORIGINAL COMMUNICATION
Prevalence of cortical superficial siderosis in patients with cognitive impairment Frank Arne Wollenweber • Katharina Buerger • Claudia Mueller Birgit Ertl-Wagner • Rainer Malik • Martin Dichgans • Jennifer Linn • Christian Opherk
•
Received: 24 September 2013 / Revised: 29 October 2013 / Accepted: 30 October 2013 / Published online: 13 November 2013 Ó Springer-Verlag Berlin Heidelberg 2013
Abstract Cortical superficial siderosis (cSS) is a magnetic resonance imaging marker of cerebral amyloid angiopathy (CAA) and can be its sole imaging sign. cSS has further been identified as a risk marker for future intracranial hemorrhage. Although uncommon in the general population, cSS may be much more prevalent in high risk populations for amyloid pathology. We aimed to determine the frequency of cSS in patients with cognitive impairment presenting to a memory clinic. We prospectively evaluated consecutive patients presenting to our memory clinic between April 2011 and April 2013. Subjects received neuropsychological testing using the Consortium to Establish a Registry for Alzheimer’s Disease battery (CERAD-NP). Two hundred and twelve patients
with documented cognitive impairment further underwent a standardized 3T-MR-imaging protocol with T2*-weighted gradient-echo sequences for detection of cSS. Thirteen of 212 patients (6.1 %) displayed cSS. In seven of them (54 %) cSS was the only imaging sign of CAA. Patients with cSS did not differ from patients without cSS with regard to medical history, age or cardiovascular risk profile. Subjects with cSS performed worse in the mini-mental state examination (p = 0.001), showed more white matter hyperintensities (p = 0.005) and more often had microbleeds (p = 0.001) compared to those without cSS. cSS is common in patients with cognitive impairment. It is associated with lower cognitive scores, white matter hyperintensities and microbleeds and can be the only imaging sign for CAA in this patient group.
J. Linn and C. Opherk contributed equally to this work.
Keywords Cerebral amyloid angiopathy Superficial siderosis MRI Cognitive decline
F. A. Wollenweber (&) K. Buerger C. Mueller R. Malik M. Dichgans C. Opherk Institute for Stroke and Dementia Research, Ludwig-Maximilians-University, Marchioninistr. 15, 81377 Munich, Germany e-mail: [email protected] B. Ertl-Wagner Institute of Clinical Radiology, Ludwig-Maximilians-University, Munich, Germany M. Dichgans Munich Cluster for Systems Neurology (SyNergy), Ludwig-Maximilians-University, Munich, Germany J. Linn Department of Neuroradiology, Ludwig-Maximilians-University, Munich, Germany C. Opherk Department of Neurology, SLK-Kliniken Heilbronn, Heilbronn, Germany
Introduction The diagnosis of cerebral amyloid angiopathy (CAA) is typically based on magnetic resonance imaging (MRI) markers such as residues of macro-hemorrhages and presence of lobar cerebral microbleeds (MB) using the Boston criteria [1]. Recently, ‘‘cortical superficial siderosis’’ (cSS) was proposed as an additional imaging marker [2] and confirmed to be highly prevalent in CAA patients [3, 4]. cSS is best detected on T2*-weighted gradient-echo sequences as linear residues of blood in the superficial layers of the cerebral cortex (Fig. 1a, b) and generally assumed to reflect recurrent bleedings into the subarachnoid space [2]. With an estimated prevalence of about 0.7 %, the frequency of cSS in the elderly general population is low [5]. However, in
123
278
J Neurol (2014) 261:277–282
Study population One thousand and seventy consecutive patients presenting to a specialized memory clinic of a tertiary level hospital between April 2011 and April 2013 were screened for study inclusion. All patients were assessed by structured medical history and received a complete physical and neurologic examination together with detailed cognitive assessment, laboratory tests, and measurements of blood pressure and body weight. Three hundred and thirty-two (31 %) of the 1,070 patients had a normal cognitive testing as described below and were, therefore, not eligible. Another 66 (6 %) patients fulfilled one of the exclusion criteria. Of patients fulfilling the inclusion criteria, 441 did not undergo standardized MRI because imaging had recently (\6 months) been performed by an outside institution or because the patient declined MRI. Two hundred and thirty-one patients underwent standardized MRI. Nineteen of these were excluded because of insufficient image quality, e.g. due to motion artifacts, resulting in a final study cohort of 212 patients. Fig. 1 Imaging signs in patients with cortical superficial siderosis presenting to a memory clinic. T2*-weighted gradient-echo sequences: a MCI patient with focal occipital cortical superficial siderosis (cSS) and a single lobar microbleed (insert) b Dementia patient with cSS (insert) but without microbleeds c Dementia patient with cSS (not shown) and multiple lobar microbleeds d Fluid attenuated inversion recovery (FLAIR) sequence: patient with cSS (not shown) and without cardiovascular risk factors showing extensive white matter hyperintensities
selected patient groups such as patients with cognitive impairment, prevalence rates may be much higher. cSS has recently shown to be a predictor for future intracranial hemorrhages, frequently at the location of pre-existing cSS [6]. Thus, the detection of cSS might have clinical implications, e.g. with regard to antithrombotic treatment. In the current study we aimed to investigate the prevalence of cSS in the setting of a memory clinic, as those patients are more likely to suffer from amyloid pathology. We further sought to determine whether patients with cSS differ from patients without cSS with regard to medical history, cardiovascular risk profile, cognitive measures and other imaging markers of CAA.
Methods The study has been approved by the appropriate ethics committee and have therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. All persons gave their informed consent prior to their inclusion in the study.
123
Inclusion and exclusion criteria Inclusion criteria were age older than 50 years and a diagnosis of either mild cognitive impairment (MCI) based on the Petersen criteria [7], or dementia according to ICD10 criteria. Exclusion criteria were severe other neurologic or psychiatric conditions that could interfere with the cognitive status (e.g. brain tumor, advanced stages of multiple sclerosis or Parkinson’s disease) and contraindications for MRI such as pacemaker or severe claustrophobia. Neuropsychological assessment Subjects were screened by the mini-mental state examination (MMSE) [8] and went on to receive detailed neuropsychological assessment if the MMSE score was above ten. Detailed testing included the Consortium to Establish a Registry for Alzheimer’s Disease battery (CERAD-NP) [9] with the following subtests: ‘‘Word List Learning/Recall and Recognition’’, ‘‘Boston Naming Test’’, ‘‘Figure Copy and Recall’’, ‘‘Semantic and Phonemic Fluency’’, ‘‘Trail Making Test Part A and B’’. The Geriatric Depression Scale (GDS) [10] was used to screen for depressive symptoms. Diagnoses were made by an interdisciplinary team including a neurologist, psychiatrist, and neuropsychologist. Assessment of cardiovascular risk factors Hypertension (HTN) was diagnosed if the patient either had a history of arterial HTN or was taking
J Neurol (2014) 261:277–282
antihypertensive drugs. Obesity was defined by a body mass index [30. Dyslipidemia was defined by a LDLcholesterol[130 mg/dl or a known history of dyslipidemia together with a prescription of lipid-lowering drugs. Diabetes mellitus was defined by history, a documented serum HbA1c [6.5 % or anti-diabetic medication. Neuroimaging acquisition and analysis MRI was performed on a 3 Tesla system (Magnetom Verio, Siemens Healthcare, Erlangen, Germany) using the following acquisition protocol that was optimized for the detection of small-vessel-related lesions [11]: (1) axial diffusion-weighted sequence [echo time (TE) = 110 ms, repetition time (TR) = 5,900 ms, slice thickness (ST) = 4 mm, matrix = 192 9 192], (2) axial fluidattenuated inversion recovery (FLAIR)-sequence [TE = 94 ms, TR = 7,000 ms, inversion time (TI) = 2,210 ms, ST = 3 mm, matrix = 256 9 2,204], (3) axial T2*weighted gradient-echo sequence (TE = 20 ms, TR = 1,020 ms, ST = 3 mm, matrix = 256 9 154), (4) axial T2-weighted turbo-spin-echo sequence (TE = 94 ms, TR = 6,560 ms, ST = 3 mm, matrix = 320 9 280), (5) sagittal three-dimensional magnetization prepared rapid acquisition gradient-echo (MPRAGE) T1-weighted sequence (TE = 4.76 ms, TR = 11 ms, ST = 1 mm, matrix = 256 9 256) with coronal and axial multi-planar reconstructions. Image analysis was performed by two experienced neuroradiologists in consensus using a standardized evaluation sheet. Readers were blinded to all clinical and patient information data. All T2*-weighted gradient-echo images were systematically assessed for evidence of cSS, intracerebral hemorrhage (ICH, defined as parenchymal hemorrhages [5 mm in diameter), and microbleeds (MB, defined as parenchymal hemorrhages B5 mm in diameter) [11]. cSS was defined as profoundly hypointense linear structures within the subarachnoid space or in the superficial layers of the cerebral cortex on T2*-weighted gradient-echo images (Fig. 1a, b). Its presence and its extent were noted and classified as focal (restricted to B3 adjacent sulci) or disseminated. In addition, we recorded the presence, number and location of ICH and MB. FLAIR images were used to identify acute ICH and subarachnoid hemorrhages and to distinguish them from old ICH and chronic cSS. The presence and extent of WMH was rated using the Wahlund Scale [12]. Statistical analysis Cardiovascular risk factors were assessed by constructing a 2 9 2 contingency table and were tested using Fisher’s exact test. Neuroimaging parameters were evaluated using either Fisher’s exact test (variables with two levels) or a v2
279
test (variables with more than two levels). Neuropsychological test results were evaluated using a WilcoxonMann–Whitney test to compare patients with and without cSS. The predictive value of neuropsychological tests (predictor variable) on cSS status (dependent variable) was evaluated using logistic regression using age and sex as covariates. A significance level of p \ 0.05 was applied in all analyses.
Results The demographic and clinical characteristics of the 212 patients included in the analysis are provided in Table 1. Thirteen patients (6.1 %) showed cSS, which was focal in seven (54 %) and disseminated in six (46 %) subjects. The majority (92 %, n = 12) of patients with cSS had no history of ICH. Overall, 128 patients (60 %) met the Petersen criteria for MCI [7], and 84 (40 %) were diagnosed as being demented according to ICD-10 criteria. The proportion of patients with dementia was similar in the cSS group compared to the group without cSS (p = 0.41). Patients with cSS did not differ significantly from those without cSS with respect to age, history of stroke, cardiovascular risk factors, and intake of antithrombotic drugs, while there was a slight preponderance of males in the group with cSS (p = 0.04, Table 1). cSS was associated with lower MMSE scores (p = 0.001), more severe WMH as measured by the Wahlund-score (p = 0.005), and with the presence of MB (p = 0.0001), which were also more often multiple [[10 MB (p = 0.02), Table 2]. However, there was no significant difference in MB localization (lobar versus deep versus cerebellar) in patients with cSS compared to those without cSS (Table 2). The CERAD-NP revealed no significant differences between the two groups nor were there differences in depressive symptoms as determined by the GDS.
Discussion The main findings of this study are that (1) cSS is common in patients with cognitive impairment; (2) it can be the only imaging sign for presumed CAA in this patient group; and (3) cSS may independently be associated with poor overall cognitive performance. Thus, cSS may be a valuable imaging marker in this group of patients. The prevalence of cSS in our sample of patients with cognitive impairment was 6.1 %, which is notably higher than the prevalence of cSS in the general population as reported from the Rotterdam Scan Study [5]. Apart from
123
280
J Neurol (2014) 261:277–282
Table 1 Clinical characteristics of patients with and without cortical superficial siderosis Total
Cortical superficial siderosis Not present
Present 13
p*
Number of patients
212
199
Age in years, mean (SD)
74 (7.5)
74 (7.7)
73 (3.2)
0.32
Female, n (%)
123 (58)
119 (60)
4 (31)
0.04
Dementia, n (%)
84 (40)
78 (39)
6 (46)
0.41
History of ICH, n (%)
3 (1)
2 (1)
1 (7)
0.17
History of ischemic stroke, n (%)
33 (16)
30 (15)
3 (23)
0.43
Hypertension
135 (63)
126 (63)
9 (69)
0.77
Diabetes mellitus
24 (11)
24 (12)
0 (0)
0.37
Dyslipidemia
91 (43)
88 (44)
3 (23)
0.15
Obesity
24 (11)
24 (12)
0 (0)
0.37
Cardiovascular risk factors, n (%)
Cognitive testing MMSE, median (IQR)
26 (23–28)
CERAD-NP composite Z-Score, mean (SD)
-1.29 (0.88)
26 (23–28)
-1.27 (0.88)
23 (7–27) à
-1.48 (0.78)
0.001 à
0.29
Medication, n (%) Antiplatelet drugs
61 (29)
57 (29)
4 (30)
0.81
Oral anticoagulation
15 (7)
13 (6)
2 (15)
0.23
Antihypertensive drugs
103 (49)
98 (49)
5 (38)
0.57
Imaging markers Wahlund Scale, mean (SD)
1.2 (0.7)
1.1 (0.7)
1.8 (0.8)
0.005
Microbleeds present, n (%)
25 (12)
19 (10)
6 (46)
0.0001
ICH intracerebral hemorrhage, IQR Interquartile range, SD Standard deviation, CERAD-NP consortium to establish a registry of Alzheimer’s disease-neuropsychological battery *p values \ 0.05 are indicated in bold, missing n = 6, àmissing n = 3
Table 2 Relationship between the characteristics of microbleeds and cortical superficial siderosis Total
Number of patients with microbleeds Multiple ([10) microbleeds, n (%) Lobar microbleeds, n (%) Deep microbleeds, n (%) Cerebellar microbleeds, n (%)
Cortical superficial siderosis
p*
Not present
Present
25
19
6
7 (28) 18 (72) 1 (4) 8 (32)
2 (10) 12 (63)
5 (83) 6 (100)
0.02 0.10
1 (5) 6 (32)
0 (0) 2 (33)
0.7 0.6
*p values \ 0.05 are indicated in bold
technical aspects such as differences in image acquisition and image analysis, the enrichment of cSS in our sample might relate to the following factors: First, prevalence rates
123
of CAA increase with age [13] and the mean age in our sample was 74 years, which is slightly higher than the mean age of subjects in the Rotterdam Scan Study (70 years) [5]. Second, our sample was enriched for patients with Alzheimer’s disease (AD) and CAA pathology is common in AD brains [14]. We did not systematically assess amyloid pathology by amyloid imaging or cerebrospinal fluid markers and the number of cases with cSS was too small to perform meaningful analyses in diagnostic subgroups. However, this should not affect our conclusions. Third, CAA may cause cognitive deficits independently from AD pathology especially in more advanced stages and we may have also enriched for patients with isolated CAA [15]. Because of the restricted sample size intergroup comparisons between patients with and without cSS must be interpreted as exploratory analyses. Nevertheless, several interesting aspects can be derived from our data: The clinical diagnosis of CAA typically requires a history of
J Neurol (2014) 261:277–282
lobar ICH as part of the Boston criteria [1]. Of note, however, is that the majority (92 %) of our subjects with cSS had no known history of ICH. As expected, we found MB and WMH to be more frequent in patients with cSS compared to those without (Fig. 1c, d), which underlines the close relationship between cSS and presumed CAA. Interestingly, in about half of our patients with cSS we did not detect any MB. Thus, it can be speculated that cSS may reflect the first sign of CAA in those cases as previously described in the literature [4]. Nevertheless, this hypothesis needs to be proven in histopathological studies. A high prevalence of cSS and, therefore, presumed CAA in patients with cognitive impairment may have implications for risk prediction and clinical management: Patients with CAA-related ICH carry a high risk for recurrent intracranial hemorrhage [16] and CAA has been identified as an important cause of warfarin-associated bleedings [17]. Further, a recent retrospective analysis of 51 patients with cSS identified cSS as a risk factor for future intracranial hemorrhage, especially at the location of preexisting siderosis [6]. Although prospective data in patients with isolated cSS not fulfilling the Boston criteria for CAA are currently lacking, these patients potentially also carry an increased risk for intracranial hemorrhage. In the future, cSS may, therefore, help to identify cognitively impaired patients who are at specific risks and might benefit from specific treatments. We found patients with cSS to have lower MMSE scores than those without. This broadly agrees with observations from a recent longitudinal study [15] that found cognitive deficits especially for perceptual speed in histologically proven CAA cases independent from AD pathology. Interestingly, our study did not show a specific cognitive profile in cSS patients. This might relate to the small number of subjects with cSS (n = 13) and to the fact that severely affected patients were not able to perform the full cognitive battery. Additional studies are needed to determine whether patients with cSS have a distinct cognitive profile. This study has several strengths including the prospective design with standardized cognitive testing and standardized MRI using independent experienced raters and established scales. Limitations include the moderate sample size and a potential selection bias related to patients that were not included because they had already received an outside MRI. Also, we did not determine APO-E status. In conclusion, cSS is common in cognitively impaired patients. Our preliminary analyses suggest an association of cSS with lower cognitive scores, WMH and microbleeds. Further, cSS may reflect the only imaging sign for CAA in a subgroup of patients. Additional studies are needed to explore the clinical value of cSS as an imaging marker.
281 Conflicts of interest of interest.
The authors declare that they have no conflict
References 1. Knudsen KA, Rosand J, Karluk D, Greenberg SM (2001) Clinical diagnosis of cerebral amyloid angiopathy: validation of the Boston criteria. Neurology 56:537–539. doi:10.1212/WNL.56.4. 537 2. Linn J, Bruckmann H (2010) Superficial siderosis in cerebral amyloid angiopathy. AJNR Am J Neuroradiol 31:E29. doi:10. 3174/ajnr.A1913 3. Charidimou A, Jager RH, Fox Z, Peeters A, Vandermeeren Y, Laloux P, Baron JC, Werring DJ (2013) Prevalence and mechanisms of cortical superficial siderosis in cerebral amyloid angiopathy. Neurology 81:626–632. doi:10.1212/WNL.0b013e3182a08f2c 4. Linn J, Halpin A, Demaerel P, Ruhland J, Giese AD, Dichgans M, van Buchem MA, Bruckmann H, Greenberg SM (2010) Prevalence of superficial siderosis in patients with cerebral amyloid angiopathy. Neurology 74:1346–1350. doi:10.1212/ WNL. 0b013e3181dad605 5. Vernooij MW, Ikram MA, Hofman A, Krestin GP, Breteler MM, van der Lugt A (2009) Superficial siderosis in the general population. Neurology 73:202–205. doi:10.1212/WNL.0b013e3181ae7c5e 6. Linn J, Wollenweber FA, Lummel N, Bochmann K, Pfefferkorn T, Gschwendtner A, Bruckmann H, Dichgans M, Opherk C (2013) Superficial siderosis is a warning sign for future intracranial hemorrhage. J Neurol 260:176–181. doi:10.1007/s00415012-6610-7 7. Petersen RC (2004) Mild cognitive impairment as a diagnostic entity. J Intern Med 256:183–194. doi:10.1111/j.1365-2796.2004. 01388.x 8. Folstein MF, Folstein SE, McHugh PR (1975) ‘‘Mini-mental state’’. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12:189–198. doi:10. 1016/0022-3956(75)90026-6 9. Morris JC, Heyman A, Mohs RC, Hughes JP, van Belle G, Fillenbaum G, Mellits ED, Clark C (1989) The consortium to establish a registry for Alzheimer’s disease (CERAD). Part I. Clinical and neuropsychological assessment of Alzheimer’s disease. Neurology 39:1159–1165. doi:10.1212/WNL.39.9.1159 10. Yesavage JA, Brink TL, Rose TL, Lum O, Huang V, Adey M, Leirer VO (1982) Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res 17:37–49. doi:10.1016/0022-3956(82)90033-4 11. Wardlaw JM, Smith EE, Biessels GJ, Cordonnier C, Fazekas F, Frayne R, Lindley RI, O’Brien JT, Barkhof F, Benavente OR, Black SE, Brayne C, Breteler M, Chabriat H, Decarli C, de Leeuw FE, Doubal F, Duering M, Fox NC, Greenberg S, Hachinski V, Kilimann I, Mok V, Oostenbrugge R, Pantoni L, Speck O, Stephan BC, Teipel S, Viswanathan A, Werring D, Chen C, Smith C, van Buchem M, Norrving B, Gorelick PB, Dichgans M, nEuroimaging STfRVco (2013) Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lan Neurol 12:822–838. doi:10.1016/ S1474-4422(13)70124-8 12. Wahlund LO, Agartz I, Almqvist O, Basun H, Forssell L, Saaf J, Wetterberg L (1990) The brain in healthy aged individuals: MR imaging. Radiology 174:675–679. doi:10.1148/radiology.174.3. 2305048 13. Charidimou A, Gang Q, Werring DJ (2012) Sporadic cerebral amyloid angiopathy revisited: recent insights into pathophysiology
123
282 and clinical spectrum. J Neurol Neurosurg Psychiatr 83:124–137. doi:10.1136/jnnp-2011-301308 14. Ellis RJ, Olichney JM, Thal LJ, Mirra SS, Morris JC, Beekly D, Heyman A (1996) Cerebral amyloid angiopathy in the brains of patients with Alzheimer’s disease: the CERAD experience, part XV. Neurology 46:1592–1596. doi:10.1212/WNL.46.6.1592 15. Arvanitakis Z, Leurgans SE, Wang Z, Wilson RS, Bennett DA, Schneider JA (2011) Cerebral amyloid angiopathy pathology and cognitive domains in older persons. Ann Neurol 69:320–327. doi:10.1002/ana.22112
123
J Neurol (2014) 261:277–282 16. O’Donnell HC, Rosand J, Knudsen KA, Furie KL, Segal AZ, Chiu RI, Ikeda D, Greenberg SM (2000) Apolipoprotein E genotype and the risk of recurrent lobar intracerebral hemorrhage. New Engl J Med 342:240–245. doi:10.1056/NEJM20000 1273420403 17. Rosand J, Hylek EM, O’Donnell HC, Greenberg SM (2000) Warfarin-associated hemorrhage and cerebral amyloid angiopathy: a genetic and pathologic study. Neurology 55:947–951. doi:10.1212/WNL.55.7.947
