Association between Microbleeds Observed on T2*-weighted Magnetic Resonance Images and Dysphagia in Patients with Acute Supratentorial Cerebral Hemorrhage Shinichiro Maeshima, MD, PhD,*1 Aiko Osawa, MD, PhD,*1 Fumitaka Yamane, MD, PhD,† Shoichiro Ishihara, MD, PhD,† and Norio Tanahashi, MD, PhD‡
Background: Approximately one-half of all patients with acute intracerebral hemorrhage (ICH) develop dysphagia that can lead to pneumonia or fatal outcomes; however, there are no reports about the relationship between swallowing function and cerebral microbleeds (CMBs) in these patients. Therefore, we investigated whether CMBs were associated with dysphagia in patients with ICH. Methods: A total of 100 patients (mean age, 65.1 6 14.1 years; range, 36-95 years) with acute supratentorial ICH were included in this study. CMBs were detected on T2*-weighted magnetic resonance imaging performed during admission using a 1.5-T scanner (mean duration from onset to magnetic resonance imaging, 12.6 6 8.4 days). We assessed swallowing function by using bedside swallowing assessments on admission and by monitoring the mode of nutritional intake at discharge in relation to the number and location of CMBs. Results: CMBs were detected in 60 of the 100 patients (60%) and were related to dysphagia. The number of CMBs and dysphagia were not related, but a significant relationship was observed in the presence of bilateral hematomas and CMBs. Conclusions: In addition to assessing just the presence or absence of CMBs, clinicians should identify the presence of bilateral or deep lesions in patients with ICH from the perspective of swallowing dysfunction. Key Words: Cerebral hemorrhage—cerebral microbleeds—dysphagia—neuroradiology— rehabilitation. Ó 2014 by National Stroke Association
From the *Department of Rehabilitation Medicine, Saitama Medical University International Medical Center, Hidaka, Saitama, Japan; †Department of Endovascular Neurosurgery, Saitama Medical University International Medical Center, Hidaka, Saitama, Japan; and ‡Department of Neurology, Saitama Medical University International Medical Center, Hidaka, Saitama, Japan. Received April 22, 2014; revision received May 19, 2014; accepted May 24, 2014. Address correspondence to Shinichiro Maeshima, MD, PhD, Nanakuri Sanatorium, 424-1 Oodori-cho, Tsu, Mie 514-1295, Japan. E-mail: [email protected]. 1 Present affiliation: Shinichiro Maeshima: Department of Rehabilitation Medicine II, School of Medicine, Fujita Health University; Aiko Osawa: Department of Rehabilitation Medicine, National Hospital for Geriatrics Medicine, National Center for Geriatrics and Gerontology, Obu, Aichi, Japan. 1052-3057/$ - see front matter Ó 2014 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2014.05.029
2458
Background Dysphagia frequently is observed in patients who have suffered a stroke. It has been detected in 29%-81% cases in the acute phase1,2 and is a known risk factor for complications such as aspiration pneumonia, dehydration, malnutrition, and suffocation.3,4 Lesions located in the brainstem, cerebral cortex, and subcortical structures such as the basal ganglia may be associated with dysphagia5,6; however, in clinical settings, it can be difficult to identify the number and location of the lesions responsible for dysphagia in recurrent cases and in cases with multiple lesions. Within the central nervous system, T2*-weighted (T2*WI) magnetic resonance imaging (MRI) is commonly used to detect hemosiderin deposits; these manifest as small spots with attenuated signals and are referred to
Journal of Stroke and Cerebrovascular Diseases, Vol. 23, No. 9 (October), 2014: pp 2458-2463
CEREBRAL MICROBLEEDS AND DYSPHAGIA
2459
improvement. Oral intake was subsequently initiated when normal BSA findings were obtained, and the number of patients who could consume a regular diet at discharge was counted. Computed Tomography/MRI
Figure 1. Cerebral microbleeds in patients with intracerebral hemorrhage Gradient-echo T2*-weighted magnetic resonance imaging is used to detect hemosiderin deposits; these manifest as small spots with attenuated signals and are referred to as cerebral microbleeds in patients with stroke.
as cerebral microbleeds (CMBs) in patients with stroke (Fig 1). Histopathologic analysis indicates that these hemosiderin deposits belong to red blood cells that have presumably leaked out of small brain vessels.7 CMBs are found in a relatively high number of patients with spontanous intracerebral hemorrhage (ICH).8,9 Because they have been considered as asymptomatic counterparts of ICH, few studies on the relationship between clinical symptoms and CMBs have been conducted. Some recent studies, however, have suggested that CMBs may contribute to clinical deficits.10-12 Therefore, we investigated whether CMBs identified via the use of T2*WI MRI were associated with dysphagia in patients with acute ICH.
Methods Procedures Swallowing Assessment and Nutritional Intake For the initial evaluation of bedside swallowing, the repetitive saliva swallowing test (RSST)13 and modified water swallowing test (MWST)14 were conducted by speech pathologists. These can be excellent screening methods for swallowing function and do not require the use of videofluoroscopy; the results are highly correlated with videofluoroscopy assessments.13,14 In addition, these bedside swallowing assessments (BSAs) have been strongly recommended by the Japanese Guidelines for the Management of Stroke.15 In this study, patients in the acute stroke phase were considered to have normal swallowing function if speech pathologists observed $3 swallows within 30 seconds during RSST and if they were assigned $4 points during MWST. On the other hand, patients were diagnosed with impaired swallowing if speech pathologists observed #2 swallows within 30 seconds during RSST and/or if they were assigned #3 points during MWST. When the BSA results were normal, oral intake was initiated. If a patient experienced difficulties during these bedside tests, the speech pathologists re-evaluated swallowing function when the patient demonstrated an
For each patient in this study, ICH was verified on admission by the use of plain computed tomography, and the hematoma volume was estimated using the ABC/2 formula (cm3).16 MRI was performed using a 1.5-T scanner (MAGNETOM Avanto; Siemens Japan, Tokyo). Gradient-recalled echo (GRE) MRI was performed in the axial plane using the following parameters: repetition time (TR), 729 ms; echo time (TE), 18 ms; flip angle, 20 ; section thickness, 7 mm; gap width, 1.4 mm; matrix, 203 3 320; and field of view, 20.8 3 23 cm. Conventional MRI techniques such as axial T1-weighted imaging (TR, 460 ms; TE, 8.4 ms), axial fluidattenuated inversion recovery imaging (TR, 8000 ms; inversion time, 2500 ms; TE, 88 ms), or axial fast spin-echo T2*WI (TR, 5000 ms; TE, 72 ms) were also performed with the same section thickness and matrix. CMBs were defined as rounded areas of signal loss on GRE MRI (diameter, 2-10 mm). Two investigators blinded to clinical data reviewed the presence, number, and location of CMBs. Symmetrical hypointensities in the globus pallidus caused by calcification and flow void artifacts of the pial vessels were carefully excluded by the investigators. In this study, unilateral hematomas without CMBs or unilateral hematomas with unilateral CMBs on the same side were defined as unilateral lesions, whereas unilateral hematomas with CMBs on the contralateral or both sides were defined as bilateral lesions. The severity of CMBs was graded as mild (total number of CMBs, 1-5), moderate (6-15), or severe (.15). The locations of CMBs were categorized according to cerebral regions: lobar (cortical gray and subcortical or periventricular white matter), deep (deep gray matter such as basal ganglia and thalamus and the white matter of the corpus callosum, internal, external, and extreme capsules), and infratentorial (brainstem and cerebellum).17 We examined the number of CMBs using T2*WI MRI and assessed the relationships with age, gender, hematoma volume, Canadian Neurological Scale (CNS) score, Mini-Mental State Examination (MMSE) score, vascular risk factors, length of hospitalization, pneumonia, swallowing function (BSA findings), and mode of nutritional intake at discharge.
Patients This was an observational clinical study. We retrospectively analyzed data from 353 patients with ICH who were consecutively referred to our rehabilitation unit at the Saitama Medical University International Medical Center between January 2008 and December 2011. Subjects
2460
S. MAESHIMA ET AL.
Figure 2. Patient flow diagram. A total of 223 MRIs were performed; 14 patients were excluded because T2*WI MRI was not available and 123 were excluded because of the presence of cavernoma, arteriovenous malformation, infratentorial lesion, a history of dementia, or cerebrovascular disease that may have caused an abnormal state. Eventually, 100 patients were included. MRI, magnetic resonance imaging.
were excluded from this study if they refused to participate or if there was a contraindication such as the presence of certain implantable devices, intracranial aneurysm clips, cardiac pacemakers, or valvular prostheses. A total of 223 MRIs were performed; 56 subjects were excluded because of a history of previous stroke; neuropsychiatric disease, including dementia; or surgical treatment. A total of 57 subjects were excluded because of the presence of an infratentorial lesion or bilateral lesions or because they were unconscious. Five patients were excluded because of the presence of cavernoma, dural arteriovenous fistula, amyloid angiopathy, or Moyamoya disease that may have caused an abnormal state. Five subjects were excluded because T2*WI MRI was not available. Eventually, 100 patients were included in the study (Fig 2). We received approval from the hospital’s institutional review board committee and obtained informed consent from each patient after providing a complete explanation of the test.
present on the right side in 30 patients and on the left side in 70. The ICH lesions were observed at several locations (Fig 3): putamen (n 5 39), thalamus (n 5 34), and subcortex (n 5 27). The following vascular risk factors were present: hypertension (blood pressure . 140/90 mm Hg or treatment history) in 82 patients (82.0%), diabetes mellitus (glycated hemoglobin .5.8% or treatment history) in 12 (12.0%), and dyslipidemia (total cholesterol . 220 mg/dL, triglycerides . 150 mg/dL, or high-density lipoprotein cholesterol , 35 mg/dL or treatment history) in 5 (5.0%). CMBs were detected in 60 of the 100 patients (60.0%). The prevalence and grade of CMBs are shown in
Statistical Analysis Unpaired t-tests and c2 tests were used to compare the incidence of demographic findings and ICH. Significance was set at P , .05. All statistical analyses were performed using the JMP version 8.0 for Macintosh software (2009; SAS Institute, Cary, NC).
Results The patients included 61 males and 39 females aged 65.1 6 14.1 years (range, 36-95 years). Hematomas were
Figure 3. Prevalence of microbleeds in patients with supratentorial intracerebral hemorrhage CMBs are detected in the brains of 60 of 100 patients (60.0%). Among the 60 patients, the severity of CMBs was graded as mild (total number of CMBs, 1-5) in 28, moderate in 21 (6-15), and severe in 11 (.15). CMBs, cerebral microbleeds.
CEREBRAL MICROBLEEDS AND DYSPHAGIA
2461
Table 1. Demographic, clinical, and morphologic variables for patients with and without CMBs among those with intracerebral hemorrhage CMBs present Patients, n (M/F) Number of CMBs Lobar, mean (SD) Deep brain, mean (SD) Infratentorial, mean (SD) Age, y (SD) CNS, /11.5 (SD) MMSE, /30 (SD) Hematoma volume, mL (SD) Hypertension, n (%) Diabetes mellitus, n (%) Hyperlipidemia, n (%) Abnormal BSA on admission, n (%) Regular diet at discharge, n (%) Pneumonia, n (%) Return to home, n (%) Length of hospitalization, days (SD)
60 (37/23) 7.5 (2.5) 4.6 (2.1) 3.5 (6.3) 67.2 (12.0) 6.0 (3.5) 13.9 (9.5) 11.4 (11.4) 52 (86.7) 6 (10.0) 2 (3.3) 39 (65.0) 20 (33.3) 14 (23.3) 14 (23.3) 30.1 (17.0)
CMBs absent 40 (24/16) – – – 61.9 (16.3) 6.7 (3.6) 16.6 (9.9) 15.4 (13.0) 30 (75.0) 6 (15.0) 3 (7.5) 23 (57.5) 14 (35.0) 7 (17.5) 12 (30.0) 23.3 (10.6)
P value .86 .25 .03 .08 .06 .30 .18 .10 .14 .45 .36 .45 .86 .48 .46 .03
Abbreviations: BSA, bedside swallowing assessment; CMBs, cerebral microbleeds; CNS, Canadian Neurological Scale; F, female; M, male; MMSE, Mini-Mental State Examination.
Figure 3. Of the 100 patients, 53 showed bilateral hemispheric lesions and 47 showed unilateral hemispheric lesions. The time between stroke onset and MRI was 12.8 6 8.1 days (range, 1-47 days). There were no significant differences in age, gender, CNS, MMSE, hematoma volume, length of hospitalization, destination after discharge, and swallowing function between patients with and without CMBs (Table 1).
Although there were significant differences in MMSE, BSA, length of hospitalization, and the number of patients who returned home between patients with and without bilateral lesions, there were no significant differences in age, gender, CNS, hematoma volume, and oral intake at discharge between the two groups (Table 2). There was a significant difference in the prevalence of hypertension and diabetes mellitus between patients
Table 2. Demographic, clinical, and morphologic variables for patients with bilateral and unilateral lesions among those with intracerebral hemorrhage
Patients, n (M/F) Number of CMBs Lobar, mean (SD) Deep brain, mean (SD) Infratentorial, mean (SD) Age, y (SD) CNS, /11.5 (SD) MMSE, /30 (SD) Hematoma volume, mL (SD) Hypertension, n (%) Diabetes mellitus, n (%) Hyperlipidemia, n (%) Abnormal BSA findings on admission, n (%) Regular diet at discharge, n (%) Pneumonia, n (%) Return to home, n (%) Duration of hospitalization, days (SD)
Bilateral
Unilateral
53 (32/21)
47 (29/18)
8.5 (2.7) 5.1 (.88) 3.9 (6.6) 67.5 (12.5) 5.6 (3.5) 12.7 (9.4) 12.3 (12.0) 46 (86.8) 4 (7.6) 2 (3.8) 38 (71.7) 14 (26.4) 14 (26.4) 9 (17.0) 31.7 (17.4)
.1 (4.6) .39 (.8) .17 (.38) 62.3 (15.3) 7.0 (3.5) 17.6 (9.4) 13.8 (12.6) 36 (76.6) 8 (17.0) 3 (6.4) 24 (51.1) 20 (42.6) 7 (14.9) 17 (36.2) 22.5 (10.1)
P value .89 .1191 ,.0001 .02 .06 .05 .01 .39 .18 .14 .55 .03 .09 .15 .03 .002
Abbreviations: BSA, bedside swallowing assessment; CMBs, cerebral microbleeds; CNS, Canadian Neurological Scale; F, female; M, male; MMSE, Mini-Mental State Examination.
S. MAESHIMA ET AL.
2462
with and without infratentorial lesions; however, there were no significant differences in age, gender, CNS, MMSE, hematoma volume, length of hospitalization, destination after discharge, and swallowing function and oral intake at discharge between patients with and without infratentorial lesions.
Discussion The prevalence of CMBs has been reported previously to range from 4.5% to 6.8% in healthy, elderly individuals.18-20 This prevalence increases with age and with advancements in imaging modalities for CMB detection.21 In patients with symptomatic intracranial hemorrhage, the prevalence of CMBs reportedly ranges from 54% to 71%.19,22-25 In this study, the prevalence of CMBs was 60% in patients with supratentorial ICH, which was similar to the prevalence reported previously.8,9 A large percentage of stroke patients experience swallowing difficulties in the acute phase, and the existence and severity of these difficulties can significantly influence the patient’s destination after discharge.3 Our previous study demonstrated that 90% patients who exhibited no abnormalities in the first BSA could consume a regular diet when discharged from the hospital.26 On the other hand, only 30% patients with suspected aspiration in the first BSA could consume a regular diet at the time of discharge. Although this study was limited to ICH results, 65.0% patients with CMBs and 57.5% patients without CMBs showed abnormal BSAs at the initial evaluation; approximately 33% patients with CMBs and 35.0% without could consume a regular diet at discharge. Regardless of the presence of CMBs, the prevalence of dysphagia was similar to that reported in previous studies1,2; however, previous studies did not identify the number and location of CMBs or report that dysphagia could be caused by a single hematoma. In this study, we documented the correlation of hematomas with CMBs, either unilateral or bilateral. We also investigated the relationship with dysphagia. The possibility that CMBs can contribute to clinical deficits has been supported by previous studies on patients with ICH.10-12 In our study, we examined the relationship between swallowing function and CMBs and reported the probable effects on outcomes. Some investigators have evaluated the different lesion sites that can be responsible for dysphagia. Swallowing is mediated by a multilayered neural network involving the cerebral cortex, subcortical structures, and brainstem; damage to these structures at any level can lead to dysphagia. Daniels and Foundas27 found that lesions in a specific location and in the periventricular white matter were associated with an increased risk of aspiration. Our study results demonstrated a significant difference in the number of CMBs in deep brain lesions between pa-
tients with unilateral lesions and those with bilateral lesions. This was probably because patients with bilateral lesions were more likely to exhibit a larger number of CMBs and damage to the subcortical networks compared with patients with unilateral lesions. Patients with multiple lesions have exhibited a tendency to have severe dysphagia.2,28 CMBs, as detected by T2*WI MRI, can have negative symptomatic effects on the swallowing function of patients with acute cerebral hemorrhage. Therefore, clinicians should be aware of the presence of bilateral or deep lesions; they should not simply assess the presence or absence of CMBs. Neuropathologic analyses of CMBs have generally an association between these lesions and surrounding tissue damage,7,20 suggesting a potential mechanism for brain dysfunction. There have been no previous studies on lesions in dysphagia patients that have attempted to describe the CMBs. A noteworthy finding of this study was that CMBs, particularly bilateral CMBs, had the potential to cause dysphagia. The CMBs visualized on T2*WI MRI also were significantly related to the presence of dysphagia. Naturally, dysphagia may be caused by many factors, and we did not perform a multivariate analysis of dysphagia in this study. Therefore, studies should be performed in the future to identify the possible variables affecting swallowing function. Taken together, the findings of our study indicate that examinations should be performed with clinical considerations for the presence and locations of CMBs, even in patients with unilateral cerebral hemorrhage.
References 1. Barer DH. The natural history and functional consequences of dysphagia after hemispheric stroke. J Neurol Neurosurg Psychiatry 1989;52:236-241. 2. Meng NH, Wang TG, Lien IN. Dysphagia in patient with brainstem stroke: incident and outcome. Am J Phys Med Rehabil 2000;79:170-175. 3. Smithard DG, O’Neill PA, Park C, et al. Complications and outcome after acute stroke: dose dysphagia matter? Stroke 1996;27:1200-1204. 4. Ickenstein GW, Stein J, Ambrosi D, et al. Predictors of survival after severe dysphagic stroke. J Neurol 2005; 252:1510-1516. 5. Cola MG, Daniels SK, Corey DM, et al. Relevance of subcortical stroke in dysphagia. Stroke 2010; 41:482-486. 6. Steinhagen V, Grossmann A, Benecke R, et al. Swallowing disturbance pattern relates to brain lesion location in acute stroke patients. Stroke 2009;40:1903-1906. 7. Fazekas F, Kleinert R, Roob G, et al. Histopathological analysis of foci of signal loss on gradient-echo T2*-weighted MR images in patients with spontaneous intracerebral hemorrhage: evidence of microangiopathyrelated microbleeds. AJNR Am J Neuroradiol 1999; 20:637-642. 8. Kleinig TJ. Associations and implications of cerebral microbleeds. J Clin Neurosci 2013;20:919-927.
CEREBRAL MICROBLEEDS AND DYSPHAGIA 9. Lee SH, Bae HJ, Kwon SJ, et al. Cerebral microbleeds are regionally associated with intracerebral hemorrhage. Neurology 2004;62:72-76. 10. Werring DJ, Frazer DW, Coward LJ, et al. Cognitive dysfunction in patients with cerebral microbleeds on T2*weighted gradient-echo MRI. Brain 2004;127:2265-2275. 11. Yakushiji Y, Nishiyama M, Yakushiji S, et al. Brain microbleeds and global cognitive function in adults without neurological disorder. Stroke 2008;39:3323-3328. 12. de Laat KF, van den Berg HA, van Norden AG, et al. Microbleeds are independently related to gait disturbances in elderly individuals with cerebral small vessel disease. Stroke 2011;42:494-497. 13. Oguchi K, Saitoh E, Baba M, et al. The repetitive saliva swallowing test (RSST) as a screening test of functional dysphagia (2) Validity of RSST. Jpn J Rehabil Med 2000; 38:383-388. 14. Tohara H, Saitoh E, Mays K, et al. Three tests for predicting aspiration without videofluorography. Dysphagia 2003; 18:126-134. 15. Shinohara Y, Yanagihara T, Abe K, et al. VII. Rehabilitation. J Stroke Cerebrovasc Dis 2011;20(4 Suppl):S145-180. 16. Kothari RU, Brott T, Broderick JP, et al. The ABCs of measuring intracerebral hemorrhage volumes. Stroke 1996;27:1304-1305. 17. Gregoire SM, Chaudhary UJ, Brown MM, et al. The Microbleed Anatomical Rating Scale (MARS): reliability of a tool to map brain microbleeds. Neurology 2009; 73:1759-1766. 18. Roob G, Schmidt R, Kapeller P, et al. MRI evidence of past cerebral microbleeds in a healthy elderly population. Neurology 1999;52:991-994. 19. Kinoshita T, Okudera T, Tamura H, et al. Assessment of lacunar hemorrhage associated with hypertensive stroke
2463
20.
21.
22.
23.
24.
25.
26.
27. 28.
by echo-planar gradient-echo T2*-weighted MRI. Stroke 2000;31:1646-1650. Jeerakathil T, Wolf PA, Beiser A, et al. Cerebral microbleeds: prevalence and associations with cardiovascular risk factors in the Framingham Study. Stroke 2004; 35:1831-1835. Greenberg SM, Vernooij MW, Cordonnier C, et al, Microbleed Study Group. Cerebral microbleeds: a guide to detection and interpretation. Lancet Neurol 2009; 8:165-174. Tanaka A, Ueno Y, Nakayama Y, et al. Small chronic hemorrhages and ischemic lesions in association with spontaneous intracerebral hematomas. Stroke 1999; 30:1637-1642. Roob G, Lechner A, Schmidt R, et al. Frequency and location of microbleeds in patients with primary intracerebral hemorrhage. Stroke 2000;31:2665-2669. Kato H, Izumiyama M, Izumiyama K, et al. Silent cerebral microbleeds on T2*-weighted MRI: correlation with stroke subtype, stroke recurrence, and leukoaraiosis. Stroke 2002;33:1536-1540. Jeong SW, Jung KH, Chu K, et al. Clinical and radiologic differences between primary intracerebral hemorrhage with and without microbleeds on gradient-echo magnetic resonance images. Arch Neurol 2004; 61:905-909. Maeshima S, Osawa A, Miyazaki Y, et al. Influences of dysphagia to a short-term outcome in patients with acute stroke. Am J Phys Med Rehabil 2011;90:316-320. Daniels SK, Foundas AL. The role of the insular cortex in swallowing. Dysphagia 1997;12:146-156. Ertekin C, Aydogdu I, Tarlaci S, et al. Mechanisms of dysphagia in suprabulbar palsy with lacunar infarct. Stroke 2000;31:1370-1376.
Background: Approximately one-half of all patients with acute intracerebral hemorrhage (ICH) develop dysphagia that can lead to pneumonia or fatal outcomes; however, there are no reports about the relationship between swallowing function and cerebral microbleeds (CMBs) in these patients. Therefore, we investigated whether CMBs were associated with dysphagia in patients with ICH. Methods: A total of 100 patients (mean age, 65.1 6 14.1 years; range, 36-95 years) with acute supratentorial ICH were included in this study. CMBs were detected on T2*-weighted magnetic resonance imaging performed during admission using a 1.5-T scanner (mean duration from onset to magnetic resonance imaging, 12.6 6 8.4 days). We assessed swallowing function by using bedside swallowing assessments on admission and by monitoring the mode of nutritional intake at discharge in relation to the number and location of CMBs. Results: CMBs were detected in 60 of the 100 patients (60%) and were related to dysphagia. The number of CMBs and dysphagia were not related, but a significant relationship was observed in the presence of bilateral hematomas and CMBs. Conclusions: In addition to assessing just the presence or absence of CMBs, clinicians should identify the presence of bilateral or deep lesions in patients with ICH from the perspective of swallowing dysfunction. Key Words: Cerebral hemorrhage—cerebral microbleeds—dysphagia—neuroradiology— rehabilitation. Ó 2014 by National Stroke Association
From the *Department of Rehabilitation Medicine, Saitama Medical University International Medical Center, Hidaka, Saitama, Japan; †Department of Endovascular Neurosurgery, Saitama Medical University International Medical Center, Hidaka, Saitama, Japan; and ‡Department of Neurology, Saitama Medical University International Medical Center, Hidaka, Saitama, Japan. Received April 22, 2014; revision received May 19, 2014; accepted May 24, 2014. Address correspondence to Shinichiro Maeshima, MD, PhD, Nanakuri Sanatorium, 424-1 Oodori-cho, Tsu, Mie 514-1295, Japan. E-mail: [email protected]. 1 Present affiliation: Shinichiro Maeshima: Department of Rehabilitation Medicine II, School of Medicine, Fujita Health University; Aiko Osawa: Department of Rehabilitation Medicine, National Hospital for Geriatrics Medicine, National Center for Geriatrics and Gerontology, Obu, Aichi, Japan. 1052-3057/$ - see front matter Ó 2014 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2014.05.029
2458
Background Dysphagia frequently is observed in patients who have suffered a stroke. It has been detected in 29%-81% cases in the acute phase1,2 and is a known risk factor for complications such as aspiration pneumonia, dehydration, malnutrition, and suffocation.3,4 Lesions located in the brainstem, cerebral cortex, and subcortical structures such as the basal ganglia may be associated with dysphagia5,6; however, in clinical settings, it can be difficult to identify the number and location of the lesions responsible for dysphagia in recurrent cases and in cases with multiple lesions. Within the central nervous system, T2*-weighted (T2*WI) magnetic resonance imaging (MRI) is commonly used to detect hemosiderin deposits; these manifest as small spots with attenuated signals and are referred to
Journal of Stroke and Cerebrovascular Diseases, Vol. 23, No. 9 (October), 2014: pp 2458-2463
CEREBRAL MICROBLEEDS AND DYSPHAGIA
2459
improvement. Oral intake was subsequently initiated when normal BSA findings were obtained, and the number of patients who could consume a regular diet at discharge was counted. Computed Tomography/MRI
Figure 1. Cerebral microbleeds in patients with intracerebral hemorrhage Gradient-echo T2*-weighted magnetic resonance imaging is used to detect hemosiderin deposits; these manifest as small spots with attenuated signals and are referred to as cerebral microbleeds in patients with stroke.
as cerebral microbleeds (CMBs) in patients with stroke (Fig 1). Histopathologic analysis indicates that these hemosiderin deposits belong to red blood cells that have presumably leaked out of small brain vessels.7 CMBs are found in a relatively high number of patients with spontanous intracerebral hemorrhage (ICH).8,9 Because they have been considered as asymptomatic counterparts of ICH, few studies on the relationship between clinical symptoms and CMBs have been conducted. Some recent studies, however, have suggested that CMBs may contribute to clinical deficits.10-12 Therefore, we investigated whether CMBs identified via the use of T2*WI MRI were associated with dysphagia in patients with acute ICH.
Methods Procedures Swallowing Assessment and Nutritional Intake For the initial evaluation of bedside swallowing, the repetitive saliva swallowing test (RSST)13 and modified water swallowing test (MWST)14 were conducted by speech pathologists. These can be excellent screening methods for swallowing function and do not require the use of videofluoroscopy; the results are highly correlated with videofluoroscopy assessments.13,14 In addition, these bedside swallowing assessments (BSAs) have been strongly recommended by the Japanese Guidelines for the Management of Stroke.15 In this study, patients in the acute stroke phase were considered to have normal swallowing function if speech pathologists observed $3 swallows within 30 seconds during RSST and if they were assigned $4 points during MWST. On the other hand, patients were diagnosed with impaired swallowing if speech pathologists observed #2 swallows within 30 seconds during RSST and/or if they were assigned #3 points during MWST. When the BSA results were normal, oral intake was initiated. If a patient experienced difficulties during these bedside tests, the speech pathologists re-evaluated swallowing function when the patient demonstrated an
For each patient in this study, ICH was verified on admission by the use of plain computed tomography, and the hematoma volume was estimated using the ABC/2 formula (cm3).16 MRI was performed using a 1.5-T scanner (MAGNETOM Avanto; Siemens Japan, Tokyo). Gradient-recalled echo (GRE) MRI was performed in the axial plane using the following parameters: repetition time (TR), 729 ms; echo time (TE), 18 ms; flip angle, 20 ; section thickness, 7 mm; gap width, 1.4 mm; matrix, 203 3 320; and field of view, 20.8 3 23 cm. Conventional MRI techniques such as axial T1-weighted imaging (TR, 460 ms; TE, 8.4 ms), axial fluidattenuated inversion recovery imaging (TR, 8000 ms; inversion time, 2500 ms; TE, 88 ms), or axial fast spin-echo T2*WI (TR, 5000 ms; TE, 72 ms) were also performed with the same section thickness and matrix. CMBs were defined as rounded areas of signal loss on GRE MRI (diameter, 2-10 mm). Two investigators blinded to clinical data reviewed the presence, number, and location of CMBs. Symmetrical hypointensities in the globus pallidus caused by calcification and flow void artifacts of the pial vessels were carefully excluded by the investigators. In this study, unilateral hematomas without CMBs or unilateral hematomas with unilateral CMBs on the same side were defined as unilateral lesions, whereas unilateral hematomas with CMBs on the contralateral or both sides were defined as bilateral lesions. The severity of CMBs was graded as mild (total number of CMBs, 1-5), moderate (6-15), or severe (.15). The locations of CMBs were categorized according to cerebral regions: lobar (cortical gray and subcortical or periventricular white matter), deep (deep gray matter such as basal ganglia and thalamus and the white matter of the corpus callosum, internal, external, and extreme capsules), and infratentorial (brainstem and cerebellum).17 We examined the number of CMBs using T2*WI MRI and assessed the relationships with age, gender, hematoma volume, Canadian Neurological Scale (CNS) score, Mini-Mental State Examination (MMSE) score, vascular risk factors, length of hospitalization, pneumonia, swallowing function (BSA findings), and mode of nutritional intake at discharge.
Patients This was an observational clinical study. We retrospectively analyzed data from 353 patients with ICH who were consecutively referred to our rehabilitation unit at the Saitama Medical University International Medical Center between January 2008 and December 2011. Subjects
2460
S. MAESHIMA ET AL.
Figure 2. Patient flow diagram. A total of 223 MRIs were performed; 14 patients were excluded because T2*WI MRI was not available and 123 were excluded because of the presence of cavernoma, arteriovenous malformation, infratentorial lesion, a history of dementia, or cerebrovascular disease that may have caused an abnormal state. Eventually, 100 patients were included. MRI, magnetic resonance imaging.
were excluded from this study if they refused to participate or if there was a contraindication such as the presence of certain implantable devices, intracranial aneurysm clips, cardiac pacemakers, or valvular prostheses. A total of 223 MRIs were performed; 56 subjects were excluded because of a history of previous stroke; neuropsychiatric disease, including dementia; or surgical treatment. A total of 57 subjects were excluded because of the presence of an infratentorial lesion or bilateral lesions or because they were unconscious. Five patients were excluded because of the presence of cavernoma, dural arteriovenous fistula, amyloid angiopathy, or Moyamoya disease that may have caused an abnormal state. Five subjects were excluded because T2*WI MRI was not available. Eventually, 100 patients were included in the study (Fig 2). We received approval from the hospital’s institutional review board committee and obtained informed consent from each patient after providing a complete explanation of the test.
present on the right side in 30 patients and on the left side in 70. The ICH lesions were observed at several locations (Fig 3): putamen (n 5 39), thalamus (n 5 34), and subcortex (n 5 27). The following vascular risk factors were present: hypertension (blood pressure . 140/90 mm Hg or treatment history) in 82 patients (82.0%), diabetes mellitus (glycated hemoglobin .5.8% or treatment history) in 12 (12.0%), and dyslipidemia (total cholesterol . 220 mg/dL, triglycerides . 150 mg/dL, or high-density lipoprotein cholesterol , 35 mg/dL or treatment history) in 5 (5.0%). CMBs were detected in 60 of the 100 patients (60.0%). The prevalence and grade of CMBs are shown in
Statistical Analysis Unpaired t-tests and c2 tests were used to compare the incidence of demographic findings and ICH. Significance was set at P , .05. All statistical analyses were performed using the JMP version 8.0 for Macintosh software (2009; SAS Institute, Cary, NC).
Results The patients included 61 males and 39 females aged 65.1 6 14.1 years (range, 36-95 years). Hematomas were
Figure 3. Prevalence of microbleeds in patients with supratentorial intracerebral hemorrhage CMBs are detected in the brains of 60 of 100 patients (60.0%). Among the 60 patients, the severity of CMBs was graded as mild (total number of CMBs, 1-5) in 28, moderate in 21 (6-15), and severe in 11 (.15). CMBs, cerebral microbleeds.
CEREBRAL MICROBLEEDS AND DYSPHAGIA
2461
Table 1. Demographic, clinical, and morphologic variables for patients with and without CMBs among those with intracerebral hemorrhage CMBs present Patients, n (M/F) Number of CMBs Lobar, mean (SD) Deep brain, mean (SD) Infratentorial, mean (SD) Age, y (SD) CNS, /11.5 (SD) MMSE, /30 (SD) Hematoma volume, mL (SD) Hypertension, n (%) Diabetes mellitus, n (%) Hyperlipidemia, n (%) Abnormal BSA on admission, n (%) Regular diet at discharge, n (%) Pneumonia, n (%) Return to home, n (%) Length of hospitalization, days (SD)
60 (37/23) 7.5 (2.5) 4.6 (2.1) 3.5 (6.3) 67.2 (12.0) 6.0 (3.5) 13.9 (9.5) 11.4 (11.4) 52 (86.7) 6 (10.0) 2 (3.3) 39 (65.0) 20 (33.3) 14 (23.3) 14 (23.3) 30.1 (17.0)
CMBs absent 40 (24/16) – – – 61.9 (16.3) 6.7 (3.6) 16.6 (9.9) 15.4 (13.0) 30 (75.0) 6 (15.0) 3 (7.5) 23 (57.5) 14 (35.0) 7 (17.5) 12 (30.0) 23.3 (10.6)
P value .86 .25 .03 .08 .06 .30 .18 .10 .14 .45 .36 .45 .86 .48 .46 .03
Abbreviations: BSA, bedside swallowing assessment; CMBs, cerebral microbleeds; CNS, Canadian Neurological Scale; F, female; M, male; MMSE, Mini-Mental State Examination.
Figure 3. Of the 100 patients, 53 showed bilateral hemispheric lesions and 47 showed unilateral hemispheric lesions. The time between stroke onset and MRI was 12.8 6 8.1 days (range, 1-47 days). There were no significant differences in age, gender, CNS, MMSE, hematoma volume, length of hospitalization, destination after discharge, and swallowing function between patients with and without CMBs (Table 1).
Although there were significant differences in MMSE, BSA, length of hospitalization, and the number of patients who returned home between patients with and without bilateral lesions, there were no significant differences in age, gender, CNS, hematoma volume, and oral intake at discharge between the two groups (Table 2). There was a significant difference in the prevalence of hypertension and diabetes mellitus between patients
Table 2. Demographic, clinical, and morphologic variables for patients with bilateral and unilateral lesions among those with intracerebral hemorrhage
Patients, n (M/F) Number of CMBs Lobar, mean (SD) Deep brain, mean (SD) Infratentorial, mean (SD) Age, y (SD) CNS, /11.5 (SD) MMSE, /30 (SD) Hematoma volume, mL (SD) Hypertension, n (%) Diabetes mellitus, n (%) Hyperlipidemia, n (%) Abnormal BSA findings on admission, n (%) Regular diet at discharge, n (%) Pneumonia, n (%) Return to home, n (%) Duration of hospitalization, days (SD)
Bilateral
Unilateral
53 (32/21)
47 (29/18)
8.5 (2.7) 5.1 (.88) 3.9 (6.6) 67.5 (12.5) 5.6 (3.5) 12.7 (9.4) 12.3 (12.0) 46 (86.8) 4 (7.6) 2 (3.8) 38 (71.7) 14 (26.4) 14 (26.4) 9 (17.0) 31.7 (17.4)
.1 (4.6) .39 (.8) .17 (.38) 62.3 (15.3) 7.0 (3.5) 17.6 (9.4) 13.8 (12.6) 36 (76.6) 8 (17.0) 3 (6.4) 24 (51.1) 20 (42.6) 7 (14.9) 17 (36.2) 22.5 (10.1)
P value .89 .1191 ,.0001 .02 .06 .05 .01 .39 .18 .14 .55 .03 .09 .15 .03 .002
Abbreviations: BSA, bedside swallowing assessment; CMBs, cerebral microbleeds; CNS, Canadian Neurological Scale; F, female; M, male; MMSE, Mini-Mental State Examination.
S. MAESHIMA ET AL.
2462
with and without infratentorial lesions; however, there were no significant differences in age, gender, CNS, MMSE, hematoma volume, length of hospitalization, destination after discharge, and swallowing function and oral intake at discharge between patients with and without infratentorial lesions.
Discussion The prevalence of CMBs has been reported previously to range from 4.5% to 6.8% in healthy, elderly individuals.18-20 This prevalence increases with age and with advancements in imaging modalities for CMB detection.21 In patients with symptomatic intracranial hemorrhage, the prevalence of CMBs reportedly ranges from 54% to 71%.19,22-25 In this study, the prevalence of CMBs was 60% in patients with supratentorial ICH, which was similar to the prevalence reported previously.8,9 A large percentage of stroke patients experience swallowing difficulties in the acute phase, and the existence and severity of these difficulties can significantly influence the patient’s destination after discharge.3 Our previous study demonstrated that 90% patients who exhibited no abnormalities in the first BSA could consume a regular diet when discharged from the hospital.26 On the other hand, only 30% patients with suspected aspiration in the first BSA could consume a regular diet at the time of discharge. Although this study was limited to ICH results, 65.0% patients with CMBs and 57.5% patients without CMBs showed abnormal BSAs at the initial evaluation; approximately 33% patients with CMBs and 35.0% without could consume a regular diet at discharge. Regardless of the presence of CMBs, the prevalence of dysphagia was similar to that reported in previous studies1,2; however, previous studies did not identify the number and location of CMBs or report that dysphagia could be caused by a single hematoma. In this study, we documented the correlation of hematomas with CMBs, either unilateral or bilateral. We also investigated the relationship with dysphagia. The possibility that CMBs can contribute to clinical deficits has been supported by previous studies on patients with ICH.10-12 In our study, we examined the relationship between swallowing function and CMBs and reported the probable effects on outcomes. Some investigators have evaluated the different lesion sites that can be responsible for dysphagia. Swallowing is mediated by a multilayered neural network involving the cerebral cortex, subcortical structures, and brainstem; damage to these structures at any level can lead to dysphagia. Daniels and Foundas27 found that lesions in a specific location and in the periventricular white matter were associated with an increased risk of aspiration. Our study results demonstrated a significant difference in the number of CMBs in deep brain lesions between pa-
tients with unilateral lesions and those with bilateral lesions. This was probably because patients with bilateral lesions were more likely to exhibit a larger number of CMBs and damage to the subcortical networks compared with patients with unilateral lesions. Patients with multiple lesions have exhibited a tendency to have severe dysphagia.2,28 CMBs, as detected by T2*WI MRI, can have negative symptomatic effects on the swallowing function of patients with acute cerebral hemorrhage. Therefore, clinicians should be aware of the presence of bilateral or deep lesions; they should not simply assess the presence or absence of CMBs. Neuropathologic analyses of CMBs have generally an association between these lesions and surrounding tissue damage,7,20 suggesting a potential mechanism for brain dysfunction. There have been no previous studies on lesions in dysphagia patients that have attempted to describe the CMBs. A noteworthy finding of this study was that CMBs, particularly bilateral CMBs, had the potential to cause dysphagia. The CMBs visualized on T2*WI MRI also were significantly related to the presence of dysphagia. Naturally, dysphagia may be caused by many factors, and we did not perform a multivariate analysis of dysphagia in this study. Therefore, studies should be performed in the future to identify the possible variables affecting swallowing function. Taken together, the findings of our study indicate that examinations should be performed with clinical considerations for the presence and locations of CMBs, even in patients with unilateral cerebral hemorrhage.
References 1. Barer DH. The natural history and functional consequences of dysphagia after hemispheric stroke. J Neurol Neurosurg Psychiatry 1989;52:236-241. 2. Meng NH, Wang TG, Lien IN. Dysphagia in patient with brainstem stroke: incident and outcome. Am J Phys Med Rehabil 2000;79:170-175. 3. Smithard DG, O’Neill PA, Park C, et al. Complications and outcome after acute stroke: dose dysphagia matter? Stroke 1996;27:1200-1204. 4. Ickenstein GW, Stein J, Ambrosi D, et al. Predictors of survival after severe dysphagic stroke. J Neurol 2005; 252:1510-1516. 5. Cola MG, Daniels SK, Corey DM, et al. Relevance of subcortical stroke in dysphagia. Stroke 2010; 41:482-486. 6. Steinhagen V, Grossmann A, Benecke R, et al. Swallowing disturbance pattern relates to brain lesion location in acute stroke patients. Stroke 2009;40:1903-1906. 7. Fazekas F, Kleinert R, Roob G, et al. Histopathological analysis of foci of signal loss on gradient-echo T2*-weighted MR images in patients with spontaneous intracerebral hemorrhage: evidence of microangiopathyrelated microbleeds. AJNR Am J Neuroradiol 1999; 20:637-642. 8. Kleinig TJ. Associations and implications of cerebral microbleeds. J Clin Neurosci 2013;20:919-927.
CEREBRAL MICROBLEEDS AND DYSPHAGIA 9. Lee SH, Bae HJ, Kwon SJ, et al. Cerebral microbleeds are regionally associated with intracerebral hemorrhage. Neurology 2004;62:72-76. 10. Werring DJ, Frazer DW, Coward LJ, et al. Cognitive dysfunction in patients with cerebral microbleeds on T2*weighted gradient-echo MRI. Brain 2004;127:2265-2275. 11. Yakushiji Y, Nishiyama M, Yakushiji S, et al. Brain microbleeds and global cognitive function in adults without neurological disorder. Stroke 2008;39:3323-3328. 12. de Laat KF, van den Berg HA, van Norden AG, et al. Microbleeds are independently related to gait disturbances in elderly individuals with cerebral small vessel disease. Stroke 2011;42:494-497. 13. Oguchi K, Saitoh E, Baba M, et al. The repetitive saliva swallowing test (RSST) as a screening test of functional dysphagia (2) Validity of RSST. Jpn J Rehabil Med 2000; 38:383-388. 14. Tohara H, Saitoh E, Mays K, et al. Three tests for predicting aspiration without videofluorography. Dysphagia 2003; 18:126-134. 15. Shinohara Y, Yanagihara T, Abe K, et al. VII. Rehabilitation. J Stroke Cerebrovasc Dis 2011;20(4 Suppl):S145-180. 16. Kothari RU, Brott T, Broderick JP, et al. The ABCs of measuring intracerebral hemorrhage volumes. Stroke 1996;27:1304-1305. 17. Gregoire SM, Chaudhary UJ, Brown MM, et al. The Microbleed Anatomical Rating Scale (MARS): reliability of a tool to map brain microbleeds. Neurology 2009; 73:1759-1766. 18. Roob G, Schmidt R, Kapeller P, et al. MRI evidence of past cerebral microbleeds in a healthy elderly population. Neurology 1999;52:991-994. 19. Kinoshita T, Okudera T, Tamura H, et al. Assessment of lacunar hemorrhage associated with hypertensive stroke
2463
20.
21.
22.
23.
24.
25.
26.
27. 28.
by echo-planar gradient-echo T2*-weighted MRI. Stroke 2000;31:1646-1650. Jeerakathil T, Wolf PA, Beiser A, et al. Cerebral microbleeds: prevalence and associations with cardiovascular risk factors in the Framingham Study. Stroke 2004; 35:1831-1835. Greenberg SM, Vernooij MW, Cordonnier C, et al, Microbleed Study Group. Cerebral microbleeds: a guide to detection and interpretation. Lancet Neurol 2009; 8:165-174. Tanaka A, Ueno Y, Nakayama Y, et al. Small chronic hemorrhages and ischemic lesions in association with spontaneous intracerebral hematomas. Stroke 1999; 30:1637-1642. Roob G, Lechner A, Schmidt R, et al. Frequency and location of microbleeds in patients with primary intracerebral hemorrhage. Stroke 2000;31:2665-2669. Kato H, Izumiyama M, Izumiyama K, et al. Silent cerebral microbleeds on T2*-weighted MRI: correlation with stroke subtype, stroke recurrence, and leukoaraiosis. Stroke 2002;33:1536-1540. Jeong SW, Jung KH, Chu K, et al. Clinical and radiologic differences between primary intracerebral hemorrhage with and without microbleeds on gradient-echo magnetic resonance images. Arch Neurol 2004; 61:905-909. Maeshima S, Osawa A, Miyazaki Y, et al. Influences of dysphagia to a short-term outcome in patients with acute stroke. Am J Phys Med Rehabil 2011;90:316-320. Daniels SK, Foundas AL. The role of the insular cortex in swallowing. Dysphagia 1997;12:146-156. Ertekin C, Aydogdu I, Tarlaci S, et al. Mechanisms of dysphagia in suprabulbar palsy with lacunar infarct. Stroke 2000;31:1370-1376.
