- Email: [email protected]
Clinical Review of 37 Patients with Medullary Infarction
Clinical Review of 37 Patients with Medullary Infarction Takuya Fukuoka, MD,* Hidetaka Takeda, MD,* Tomohisa Dembo, MD,* Harumitu Nagoya, MD,* Yuji Kato, MD,* Ichiro Deguchi, MD,* Hajime Maruyama, MD,* Yohsuke Horiuchi, MD,* Akira Uchino, MD,† Susumu Yamazaki, MD,* and Norio Tanahashi, MD*
Background: Clinical features of medullary infarction were compared between patients with lateral medullary infarction and medial medullary infarction Methods: Thirty-seven patients with medullary infarction (29 with lateral medullary infarction and 8 with medial medullary infarction) who were admitted to our center between April 1, 2007 and March 31, 2010 were examined. Background factors, neurologic signs and symptoms, imaging findings, cause of disease, and outcomes were assessed for patients with lateral and those with medial medullary infarction. Results: Examination of the clinical symptoms and neurologic findings suggested that among patients with medial medullary infarction, few demonstrated all of the symptoms of Dejerine syndrome at onset, and many had lesions that were difficult to locate based only on neurologic findings. Both lateral and medial medullary infarction were frequently caused by atherothrombosis. However, cerebral artery dissection was observed in 31% of patients with lateral medullary infarction and 12.5% of those with medial medullary infarction. In 13% of patients with lateral and 37% of patients with medial medullary infarction, magnetic resonance imaging diffusion-weighted images on the day of onset did not show abnormalities, and the second set of diffusion-weighted images confirmed infarction lesions. For lateral medullary infarction, a more rostral lesion location was correlated with a poorer 90-day outcome. For medial medullary infarction, a more dorsal lesion location was correlated with a poorer 90-day outcome. Conclusions: The diagnosis rate of medullary infarction using imaging examinations at onset—particularly medial medullary infarction—is not necessarily high. The imaging examinations need to be repeated for patients who are suspected to have medullary infarction based on neurologic signs and symptoms. Key Words: MRI—LMI—MMI. Crown Copyright Ó 2012 Published by Elsevier Inc. on behalf of National Stroke Association. All rights reserved.
The diagnosis of medullary infarction has become relatively easy with advances in magnetic resonance imaging (MRI) and other diagnostic imaging technologies. Lateral medullary infarction (LMI) and medial medullary infarc-
tion (MMI) are uncommon disorders, accounting for approximately #2.5% and #1%, respectively, of all cases of cerebral infarction.1,2 LMI is known as Wallenberg syndrome and presents with facial sensory disturbance,
From the *Departments of Neurology and Cerebrovascular Medicine; and †Diagnostic Radiology, Saitama Medical University International Medical Center, Saitama, Japan. Received November 2, 2010; revision received January 12, 2011; accepted January 19, 2011. Address correspondence to Takuya Fukuoka, MD, Department of Neurology and Cerebrovascular Medicine, Saitama International
Medical Center, Saitama Medical University, 1397-1 Yamane, Hidaka, Saitama 350-1298, Japan. E-mail: [email protected]. 1052-3057/$ - see front matter Crown Copyright Ó 2012 Published by Elsevier Inc. on behalf of National Stroke Association. All rights reserved. doi:10.1016/j.jstrokecerebrovasdis.2011.01.008
594
Journal of Stroke and Cerebrovascular Diseases, Vol. 21, No. 7 (October), 2012: pp 594-599
CLINICAL REVIEW OF MEDULLARY INFARCTION
cerebellar ataxia, Horner syndrome, and paralysis of the soft palate and pharynx on the side of the lesion. The symptoms on the contralateral side include disturbance of pain and temperature sense in the trunk and the upper and lower limbs. In MMI, which is known as Dejerine syndrome, atrophy and paralysis of the tongue develop on the side of the lesion, while impaired tactile and deep sensation of the limbs and trunk and hemiplegia, both excluding the face, are observed on the contralateral side. However, these symptoms of LMI and MMI are unlikely to appear all at once. In particular, MMI may present with only those features indicative of a supratentorial disorder or provide falsenegative results on diffusion-weighted images (DWIs) of MRI early in the development of medullary lesions.3-6 In the present study, the imaging findings and clinical symptoms of patients with medullary infarction were reviewed, and the MRI and magnetic resonance angiography (MRA) findings, cause, and outcomes were examined.
Methods Of the 955 patients with ischemic stroke who were admitted to our center between April 1, 2007 and March 31, 2010, 37 (3.8%) had medullary infarction. These 37 patients, including 29 (3.0%) with LMI and 8 (0.8%) with MMI, were examined in this study. MRI examinations were conducted using an Achieva (1.5-T; Philips Electronics Japan, Tokyo, Japan) and a MAGNETOM Avant (1.5-T; Siemens Japan, Tokyo, Japan). DWIs were taken using the spin echo echo planar imaging (EPI) method using the Achieva (repetition time [TR]/echo time [TE] 3829 msec/65 msec; slice thickness 5 mm; field of view [FOV] 230 mm 3 207 mm; matrix 144 3 100.8) or the Avant (TR/TE 4200 msec/81 msec; slice thickness 5 mm; FOV 230 mm 3 230 mm; matrix 128 3 102). MRAwas taken using the 3-dimensional time of flight (3D-TOF) method using the Achieva (TR/TE 19 msec/6.9 msec; FOV 200 mm 3 180 mm) or the Avant (TR/TE 22 msec/7 msec; FOV 220 mm 3 176 mm). Images were obtained in 5-mm slices parallel to the orbitomeatal (OM) line, and infarction was classified into upper, central, and lower medulla oblongata. Based on reports by Currier et al,1 Kim,2 and Vuilleumier et al,3 the upper, central, and lower medulla oblongata were defined as the area where the inferior cerebellar peduncle is raised posterolaterally, the area where the nodular surface attributed to the inferior olivary nucleus is observed, and the area where the bulge disappears and the cross section becomes round, respectively. For MMI, in addition to the classification by the levels in the medulla oblongata, infarction was also categorized into ventral (including the pyramid), central (including the medial lemniscus), and dorsal (including the medial longitudinal fasciculus).4,5 All patients were transported to our center by ambulance, and MRI was performed on the day they developed infarction. When
595
brainstem infarction was suspected at admission but there were no abnormal findings on MRI, MRI was repeated between day 3 and day 7 after onset. Carotid arterial ultrasound examination was performed for all patients, and 3-dimensional computed tomographic angiography (3D-CTA) and cerebral angiography were also performed for patients suspected of having cerebrovascular dissection. The background of the patients, neurologic signs and symptoms on admission, MRI and MRA findings, cause of disease, and 90-day outcomes were examined for the LMI group and the MMI group. Causes of medullary infarction were divided into 3 types: atherothrombosis, cardioembolism, and others (including arterial dissection). To diagnose cerebral arterial dissection, the criteria issued by the Working Group for Cerebral Artery Dissection were used.6 Functional outcome was evaluated using the modified Rankin Scale (mRS) score at 90 days. Patients who were transferred to the rehabilitation hospital were evaluated by interview. Statistical evaluation of data was performed with PASW Statistics 18 (SPSS; Chicago, IL). The Chi-square test was used for analyzing differences in patients’ baseline characteristics and risk factors between the 2 groups. The Student t test was used for analyzing the frequency of negative MRI and time from onset to MRI examination between the 2 groups. The Wilcoxon rank sum test was used for analyzing the relationship between the MRI findings and the National Institutes of Health Stroke Scale (NIHSS) score on admission, level of infarction, and mRS score at 90 days. P , .05 was considered statistically significant. All data are presented as mean 6 SD. This study was approved by the Ethics Committee at Saitama Medical University International Medical Center.
Results The clinical features of the 29 patients with LMI and the 8 patients with MMI are summarized in Tables 1 and 2, respectively, and mRS scores at 90 days after onset for patients with LMI or MMI are shown by the levels of infarction in Table 3. The mean age of the LMI group was 60 6 15 years (range 18-83), and that of the MMI group was 62 6 10 years (range 44-68). The LMI group consisted of 24 men and 5 women, and the MMI group was comprised of 7 men and 1 woman. As risk factors, 26 (89%) patients in the LMI group had hypertension, 8 (27%) had diabetes, 6 (20%) had dyslipidemia, 2 (6.8%) had atrial fibrillation, and 2 (6.8%) had a history of smoking, while 6 (75%) patients in the MMI group had hypertension, 3 (37%) had diabetes, 1 (12%) each had dyslipidemia and atrial fibrillation, and 2 (25%) had a history of smoking. There were no significant differences in risk factors between the 2 groups (P 5 .800). The neurologic signs and symptoms at the first examination included, in the LMI group, dizziness/
596
Table 1. List of cases of lateral medullary infraction
Age/ Patient Sex
Risk factor
Headache, posterior Dissociated cervical Horner sensory Cerebellar pain Dysphagia Dysarthria syndrome loss symptoms Cause Dizziness
Paralysis
Time from Level of Lesion onset to medullary initial MRI (h) infraction MRI
1 1 1 1 –
1 2 1 1 2
1 1 1 1 2
1 1 1 1 2
1 2 1 1 2
1 2 1 2 2
1 1 1 1 1
2 2 2 2 2
5.3 5.1 2 6 18.3
Upper Upper Upper Upper Upper
AD AT AT AT
1 1 1 2
1 2 2 2
1 1 2 1
1 1 2 1
2 1 1 2
1 1 1 2
2 2 2 2
2 2 2 2
19 23.3 10.3 21.6
AT AT AT CE AT
2 2 1 1 1
2 2 2 2 2
2 1 2 1 1
2 1 1 1 1
2 2 2 1 1
1 1 2 1 1
2 2 2 1 1
15 16 17 18 19 20 21 22
4/M5 84/F 55/M 66/F 37/M 46/F 40/M 64/M
DL, smoking HT HT HT, DM HT HT, DL, DM HT HT
AD CE AD AT AD AT AD AT
2 1 1 1 2 1 2 1
1 2 2 2 2 1 1 2
2 1 1 2 2 2 1 1
2 2 1 2 2 2 1 1
2 1 2 2 2 2 1 2
1 2 2 1 1 2 1 1
2 1 1 1 1 1 2 2
23 24
74/M HT, AF 62/M HT, DL, DM, smoking 38/M HT, DM 69/M HT 69/M HT, DL 62/M HT 56/F HT, DL, DM
CE AT
1 1
2 2
1 1
1 1
2 1
2 1
2 1
2 2 2 2 Left upper and lower limits 2 2 2 2 2 2 2 Right (old) 2 2
AD AT AT AD AT
2 2 2 1 1
1 2 2 1 1
1 2 1 1 2
1 2 1 2 2
2 2 1 1 2
1 1 1 1 2
2 2 1 1 1
2 2 2 2 2
71/M 71/M 57/M 18/M 59/M
6 7 8 9
39/M 59/M 63/M 83/M
10 11 12 13 14
25 26 27 28 29
Presence or mRS at 90 absence of NIHSS at days after dissection admission onset
Left VA Right PICA Right VA Left VA Left VA
Left VA 2 Right VA Left VA 2
4 2 5 1 3
2 1 3 2 2
Upper Upper Upper Upper
Right VA Left PICA Right VA BA
Right VA 2 2 2
2 6 2 6
2 4 1 5
23.3 9.1 8.1 11 3.1
Central Central Central Central Central
Left VA Left VA Left VA Left VA Right VA
2 2 2 2 2
5 3 2 4 3
2 0 1 1 4
3.3 5 11.3 11 23.3 10 10 24
Central Central Central Central Central Central Central Central
Left VA Right VA Right VA Left VA Left VA Left PICA Right VA Right VA
Left VA 2 Right VA 2 Right VA 2 Right VA 2
1 5 4 3 5 1 1 5
1 4 2 1 1 1 3 2
5 24
Central Central
Left VA Right VA
2 2
2 3
1 3
12 12 10 25.8 9
Central Central Central Central Lower
Both VA Left VA Left VA Left VA Right VA
Left VA 2 2 Left VA 2
3 1 1 2 1
1 1 2 3 1
2
2
2
2
Abbreviations: AD, arterial dissection; AF; atrial fibrillation; ASD, atrial septal defect; AT, atherothrombosis; BA, basilar artery; CABG, coronary artery bypass graft; CE, cardiac embolism; DCM, dilated cardiomyopathy; DL, dyslipidemia; DM, diabetes mellitus; F, female; HT, hypertension; M, male; MI, myocardial infraction; MRA, magnetic resonance angiography; MRI, magnetic resonance imaging; mRS, modified Rankin Scale; NIHSS, National Institutes of Health Stroke Scale; PICA, posterior inferior cerebellar artery; SAS, sleep apnea syndrome; VA, vertebral artery.
T. FUKUOKA ET AL.
AD AT AD AD AT
76/M 77/M 62/M 74/M 53/F
HT, DL HT HT None HT, DM, DL, DCM SAS HT, DM, ASD HT HT, DM, CABG HT, DM, MI HT, CABG DVT AF, HT HT
1 2 3 4 5
Vessel poorly visualized on MRA
2 Central Central Central Central (1 lower) 5 12 11 11 2 2 2 1 L 2 R L 2 1 1 1 1 1 1 2 68/M 58/M 53/M 68/M 5 6 7 8
HT, DM HT Smoking DM, HT, DL, AF, ASD
AT AT AT AT
2 1 1 2
L L R and L L
2 Central (1 upper) 2 1 L 1 2 44/M 4
HT, migraine
AD
1
L
2 Upper 4 1 2 1 L 2 1 AT 3
HT 65/F
1 2
597 Abbreviations: AF; atrial fibrillation; ASD, atrial septal defect; AT, atherothrombosis; BA, basilar artery; DL, dyslipidemia; DM, diabetes mellitus; F, female; HT, hypertension; L, left; M, male; MRI, magnetic resonance imaging; mRS, modified Rankin Scale; NIHSS, National Institutes of Health Stroke Scale; PICA, posterior inferior cerebellar artery; R, right; VA, vertebral artery.
1 1 1 4 4 4 10 6 BA RVA 2 R PICA
3 4 R VA (dissection)
1 5 R VA
1 1 4 5 R VA R VA
Ventral Ventral 1 central Ventral 1 central 1 dorsal Ventral 1 central 1 dorsal Ventral Ventral Ventral Ventral 1 central 1 dorsal Upper Upper 14 8 2 2 L L 1 2 2 2 AT AT 77/M 61/M
Patient
HT, DM Smoking
Age/ Sex
Risk factor
Cause
2 2
Paralysis
Dysarthria
Deep sensation Dizziness, vomiting
L L.
Level of medullary infraction Deviation of tongue
Time from onset to MRI (h) Headache, posterior cervical pain
Table 2. List of cases of medical medullary infarction
Presence or absence of lesion in initial MRI
Location of infraction in horizontal section
Occluded vessel
NIHSS at admission
mRS at 90 days after onset
CLINICAL REVIEW OF MEDULLARY INFARCTION
lightheadedness in 19 patients (65%), headache in 10 (34%), dysphagia in 19 (65%), dysarthria in 18 (62%), Horner syndrome in 12 (41%), dissociated sensory loss in 19 (65%), hemiplegia in 1 (3.4%), and cerebellar symptoms in 16 (55%). In the LMI group, dysphagia was more common among patients with infarction in the upper medulla oblongata. It was observed in 7 of 9 (78%) patients with infarction in this area, whereas it was reported in 9 of 18 (50%) patients with lesions in the central medulla oblongata. In the MMI group, neurologic signs and symptoms at the first examination consisted of headache in 3 patients (37%), dizziness and vomiting in 4 (50%), deviation of the tongue in 3 (37%), impaired deep sensation in 6 (75%), and hemiplegia not involving the face in 8 (including one with quadriplegia; 100%). The causes of infarction were, in the LMI group, atherothrombosis in 17 patients (58%), embolism caused by cardiac sources in 3 (10.3%), and arterial dissection in 9 (31%), and in the MMI group, atherothrombosis in 7 (87%) and arterial dissection in 1 (12.5%). The frequency of arterial dissection was significantly higher in the LMI group than in the MMI group (P 5 .007). In the LMI group, the occluded or narrowed vessels confirmed by MRA were the vertebral artery (VA) in 25 patients, the posterior inferior cerebellar artery (PICA) in 3, and the basilar artery (BA) in 1. The lesions were located on the right side in 11 patients and on the left side in 18. In 4 patients, the second MRI-DWI detected infarction lesions for the first time. The time from onset to MRI examination was 12.4 6 7.3 hours in the LMI group. The time from onset to MRI examination was significantly shorter in the initial MRI-negative group (7.5 6 2.3 hours) than in the initial MRI-positive group (13.2 6 7.8 hours; P 5 .022). There was no significant correlation between NIHSS score at admission and MRI findings (positive or negative MRI) in the LMI group (P 5 .138). In the MMI group, the occluded or narrowed vessels were the VA in 5 patients, the PICA in 1, and the BA in 1. Lesions were on the right side in 6 patients and on the left side in 1 patient. Lesions were observed for the first time on the second MRI in 3 patients. The time from onset to MRI examination was 8.3 6 4.0 hours in the LMI group. The time from onset to MRI examination was significantly shorter in the initial MRI-negative group (3.6 6 1.5 hours) than in the initial MRI-positive group (11.1 6 2.1 hours; P 5 .022). There was no significant correlation between NIHSS score at admission and MRI findings (positive or negative MRI) in the MMI group (P 5 .337). The mRS scores at 90 days after onset (90-day outcome) were, in the LMI group, 0 in 1 patient, 1 in 12, 2 in 8, 3 in 4, 4 in 3, and 5 in 1, and in the MMI group, while no patient had a score of 0, 2, or 5—the scores were 1 in 6 patients, 3 in 1, and 4 in 1. As shown in Table 3, the frequency of good recovery (mRS 0-1) was 22% in the upper group (2/9), 52% in the central group (10/19), and 100% in the
T. FUKUOKA ET AL.
598
Table 3. Modified Rankin scale score 90 days after onset by level of infraction Level of infarction mRS at 90 days after onset LMI
MMI
Infarction location Upper Central Lower Upper Central Ventral Ventral 1 central Ventral 1 central 1 dorsal
0
1
2
3
4
5
9 cases 19 cases 1 case 3 cases 5 cases 4 cases 1 case
None 1 case None None None None None
2 cases 9 cases 1 case 3 cases 3 cases 4 cases 1 case
4 cases 4 cases None None None None None
1 case 3 cases None None 1 case None None
1 case 1 case None None 1 case None None
1 case None None None None None None
3 cases
None
1 case
None
1 case
1 case
None
Abbreviations: LMI, lateral medullary infarction; MMI, medial medullary infraction; mRS, modified Rankin Scale.
lower group (1/1; P 5 .073). The outcome of LMI with more rostral lesions tended to be poor. Of cases of MMI, none had infarction lesions confined to the lower area. The frequency of good recovery (mRS 0-1) was 100% in the upper group (3/3) and 60% in the central group (3/5; P 5 .237). The frequency of good recovery (mRS 0-1) was 100% in the ventral group (4/4), 100% in the ventral and central group (1/1), and 33% in the ventral and central and dorsal group (P 5 .068). The outcome of MMI with more dorsal lesions tended to be poor.
Discussion Of the patients with medullary infarction examined in the present study, the mean age at onset was 60 6 15 years (range 18-83) for LMI and 62 6 10 years (range 44-68) for MMI; they were relatively young compared to patients with cerebral infarction in general. Other investigators also reported a younger age of onset for these conditions. LMI and MMI developed at an average age of 57 6 11.9 years and 62 6 10 years, respectively, according to Kim2 and Kim and Han,5 and 60.7 6 12.4 years and 65.0 6 12.3 years, respectively, according to Kameda et al.7 Kim et al8 reported that, of the symptoms of LMI, infarction in the rostral medulla oblongata in particular often presented with dysphagia, hoarseness, and facial palsy, and that the more caudal the lesions were, the more frequent the symptoms of dizziness/lightheadedness, nystagmus, and ataxic gait were. In the present study, dysphagia was frequently observed in patients with lesions in the upper medulla oblongata, which was probably because the nucleus ambiguus is located in this area. Kameda et al7 reported headache in 47% of patients with LMI. The incidence of headache in the present study was as high as 34%, which is in line with the findings of Kameda et al.7 The relatively high incidence of headache appeared to be associated with the fact that some of our patients had cerebrovascular dissection, which will be discussed later.
For MMI, Kim et al5 reported hemiplegia in 91% of the patients they examined, sensory disturbance in 73%, and vertigo/dizziness in 59%, and Kameda et al7 reported headache in 13% of their patients, dizziness in 56%, vomiting in 44%, deviation of the tongue in 30%,sensory disturbance in 68%, and hemiplegia in 93%. Bassetti et al9 reported, in their examination of 7 patients with MMI, dizziness/vomiting in 71% of patients, hyperalgesia in 85%, headache in 57%, dysarthria in 42%, and disturbance of consciousness in 28%, which were similar to the distributions of the initial symptoms in previous reports. In the present study, paralysis (except facial) was seen in all patients, impaired deep sensation was seen in 6 patients, and deviation of the tongue was seen in only 3 patients. The symptoms of Dejerine syndrome were not common. Some patients were diagnosed with MMI for the first time based on imaging findings. We considered that deviation of the tongue and impaired deep sensation were important as symptoms indicative of MMI. On diagnostic imaging, MRI-DWI on hospitalization did not detect culprit lesions, and the second MRI confirmed infarction lesions in 4 of 29 patients with LMI and 3 of 8 patients with MMI. Pedraza et al10 suggested that DWI is less sensitive during the first 24 hours after the onset of cerebral infarction, and that false-negative results may occur, especially in small infarctions in the posterior circulation. In their study of 139 patients with cerebral infarction who underwent MRI within 48 hours after onset, Oppenheim et al11 examined 8 patients (5.8%) for whom DWI produced falsenegative results. False-negative results occurred in 19% of patients with infarction in the posterior circulation and in 31% of patients with vertebrobasilar infarction. The mean size of the infarction lesions was 0.19 6 0.16 cm.3 As for the causes of false-negative results, they suggested lesion location, time to imaging, the small size of lesions that does not allow image visualization on DWI, inadequate signalto-noise (S/N) ratio in the early phase of infarction, and magnetic artifact produced during EPI that causes poor visualization of the brainstem. Wang et al12 reported a patient
CLINICAL REVIEW OF MEDULLARY INFARCTION
with cerebral infarction in the area of the middle cerebral artery whose DWI taken 27 hours after onset was negative, but the DWI 7 days after onset was positive. They speculated that false-negative results occur when there is a mismatch between diffusion and perfusion—that is, when there is sufficient reperfusion. Kitis et al13 confirmed 1 false-negative case among 13 cases (7.6%) of LMI. In the present study, the first DWI was negative in 4 of 29 patients (13%) with LMI and 3 of 8 patients (37%) with MMI, revealing a higher incidence of false-negatives among patients with MMI. In the present study, the time of imaging was related to false-negative results. In addition, we performed MRI with a slice thickness of 5 mm, and the interslice gap was 1.5 mm. Therefore, some early infarction lesions may have been hidden in these interslice gaps, and this contributed to the false-negative results. Therefore, careful neurologic examination is still considered important for the detection of medullary infarction. Cerebral artery dissection is known as a cause of LMI and MMI. Vertebral artery dissection (VAD) in general is reported to develop in 1 to 1.5 per 100,000 people.14 The incidence of LMI caused by VAD was 15% (20/123 patients) and 29% (31/107 patients) according to Kim2 and Kameda et al,7 respectively. Hosoya et al15 reported that, among 93 patients with LMI, cerebral artery dissection was determined to be the cause in 23 patients (24.7%) and suspected as a cause in an additional 27 patients. Including the suspected cases, the incidence of LMI caused by cerebral artery dissection was 53.7% (50/93 patients). In the present study, based on strict diagnostic criteria, 9 of 29 patients (31%) with LMI had cerebral arterial dissection. The incidence based on definite diagnosis (31%) was higher than the incidences in previous reports. The incidence of MMI caused by VAD was reported to be 2.3% (2/86 patients), 21% (4/19 patients), and 18% (2/11 patients) by Kim,2 Kameda et al,7 and Toyoda et al,16 respectively. In the present study, 1 of 8 patients (12.5%) had MMI caused by VAD, which was less common than for LMI caused by VAD. The 90-day outcome of LMI tended to be poorer in patients with more rostrally located lesions. This is probably because even without quadriplegia, persisting dysphagia affects activities of daily living in general as a prognostic factor for LMI. When we compared the outcome of LMI located at different levels of the medulla oblongata, the outcome was better for infarction in the lower medulla oblongata than for infarction in the upper or central medulla oblongata. For MMI, Kim and Han5 suggested that severe hemiplegia and age were the factors associated with poor prognosis, and that rostral and ventral locations of MMI lesions were associated with the incidence of dyskinesia. Although the infarction lesion was confined to the lower medulla oblongata in none of the 8 patients with MMI in our study, severe cases were more common among those with caudal (infarction in the central medulla oblongata) and ventral lesions. Our examination of medullary infarction revealed that both LMI and MMI were often caused by atherothrombo-
599
sis, and that VAD was frequently observed in patients with LMI. Careful consideration of neurologic findings is essential, because MRI-DWI on the day of onset may not show abnormalities, so that diagnosis of infarction in the early phase can be difficult. Because MMI rarely presents with typical symptoms, and because its clinical symptoms and neurologic findings are diverse, MMI was not detected at onset in many cases in our study.
References 1. Currier RD, Giles CL, Dejong RN. Some comments on Wallenberg’s lateral medullary syndrome. Neurology 1961;11:778-791. 2. Kim JS. Pure lateral medullary infarction: Clinicalradiological correlation of 130 acute, consecutive patients. Brain 2003;126:1864-1872. 3. Vuilleumier P, Bogousslavsky J, Regli F. Infarction of the lower brainstem. Clinical, aetiological and MRItopographical correlations. Brain 1995;118:1013-1025. 4. Parent A. Carpenter’s human neuroanatomy. 9th ed. Baltimore: Williams & Wilkins, 1996:421-458. 5. Kim JS, Han YS. Medial medullary infarction: Clinical, imaging, and outcome study in 86 consecutive patients. Stroke 2009;40:3221-3225. 6. Takagi M. Guideline for diagnosis and treatment of cerebral artery dissection. In: Guidelines for treatment of stroke in younger patients, Research Grant for Cardiovascular Diseases 12shi-2, National Multicenter Study of Diagnosis, Treatment and Preventive Strategy for Stroke in Younger Generations. Osaka, Japan: Cerebrovascular Division, Department of Medicine, National Cardiovascular Center, 2003:85-89. 7. Kameda W, Kawanami T, Kurita K, et al. Lateral and medial medullary infarction: A comparative analysis of 214 patients. Stroke 2004;35:694-699. 8. Kim JS, Lee JH, Suh DC, et al. Spectrum of lateral medullary syndrome. Correlation between clinical findings and magnetic resonance imaging in 33 subjects. Stroke 1994; 25:1405-1410. 9. Bassetti C, Bogousslavsky J, Mattle H, et al. Medial medullary stroke: Report of seven patients and review of the literature. Neurology 1997;48:882-890. 10. Pedraza S, Osuna MT, Davalos A, et al. False negative diffusion in acute ischemic stroke. Rev Neurol 2002;34:1127-1129. 11. Oppenheim C, Stanescu R, Dormont D, et al. False-negative diffusion-weighted MR findings in acute ischemic stroke. AJNR Am J Neuroradiol 2000;21:1434-1440. 12. Wang W, Goldstein S, Scheuer ML, et al. Acute stroke syndrome with fixed neurological deficit and falsenegative diffusion-weighted imaging. J Neuroimaging 2003;13:158-161. 13. Kitis O, Calli C, Yunten N, et al. Wallenberg’s lateral medullary syndrome: Diffusion-weighted imaging findings. Acta Radiol 2004;45:78-84. 14. Schievink WI. Spontaneous dissection of the carotid and vertebral arteries. N Engl J Med 2001;344:898-906. 15. Hosoya T, Nagahata M, Yamaguchi K. Prevalence of vertebral artery dissection in Wallenberg syndrome: Neuroradiological analysis of 93 patients in the Tohoku District, Japan. Radiat Med 1996;14:241-246. 16. Toyoda K, Imamura T, Saku Y, et al. Medial medullary infarction: Analyses of eleven patients. Neurology 1996; 47:1141-1147.
Background: Clinical features of medullary infarction were compared between patients with lateral medullary infarction and medial medullary infarction Methods: Thirty-seven patients with medullary infarction (29 with lateral medullary infarction and 8 with medial medullary infarction) who were admitted to our center between April 1, 2007 and March 31, 2010 were examined. Background factors, neurologic signs and symptoms, imaging findings, cause of disease, and outcomes were assessed for patients with lateral and those with medial medullary infarction. Results: Examination of the clinical symptoms and neurologic findings suggested that among patients with medial medullary infarction, few demonstrated all of the symptoms of Dejerine syndrome at onset, and many had lesions that were difficult to locate based only on neurologic findings. Both lateral and medial medullary infarction were frequently caused by atherothrombosis. However, cerebral artery dissection was observed in 31% of patients with lateral medullary infarction and 12.5% of those with medial medullary infarction. In 13% of patients with lateral and 37% of patients with medial medullary infarction, magnetic resonance imaging diffusion-weighted images on the day of onset did not show abnormalities, and the second set of diffusion-weighted images confirmed infarction lesions. For lateral medullary infarction, a more rostral lesion location was correlated with a poorer 90-day outcome. For medial medullary infarction, a more dorsal lesion location was correlated with a poorer 90-day outcome. Conclusions: The diagnosis rate of medullary infarction using imaging examinations at onset—particularly medial medullary infarction—is not necessarily high. The imaging examinations need to be repeated for patients who are suspected to have medullary infarction based on neurologic signs and symptoms. Key Words: MRI—LMI—MMI. Crown Copyright Ó 2012 Published by Elsevier Inc. on behalf of National Stroke Association. All rights reserved.
The diagnosis of medullary infarction has become relatively easy with advances in magnetic resonance imaging (MRI) and other diagnostic imaging technologies. Lateral medullary infarction (LMI) and medial medullary infarc-
tion (MMI) are uncommon disorders, accounting for approximately #2.5% and #1%, respectively, of all cases of cerebral infarction.1,2 LMI is known as Wallenberg syndrome and presents with facial sensory disturbance,
From the *Departments of Neurology and Cerebrovascular Medicine; and †Diagnostic Radiology, Saitama Medical University International Medical Center, Saitama, Japan. Received November 2, 2010; revision received January 12, 2011; accepted January 19, 2011. Address correspondence to Takuya Fukuoka, MD, Department of Neurology and Cerebrovascular Medicine, Saitama International
Medical Center, Saitama Medical University, 1397-1 Yamane, Hidaka, Saitama 350-1298, Japan. E-mail: [email protected]. 1052-3057/$ - see front matter Crown Copyright Ó 2012 Published by Elsevier Inc. on behalf of National Stroke Association. All rights reserved. doi:10.1016/j.jstrokecerebrovasdis.2011.01.008
594
Journal of Stroke and Cerebrovascular Diseases, Vol. 21, No. 7 (October), 2012: pp 594-599
CLINICAL REVIEW OF MEDULLARY INFARCTION
cerebellar ataxia, Horner syndrome, and paralysis of the soft palate and pharynx on the side of the lesion. The symptoms on the contralateral side include disturbance of pain and temperature sense in the trunk and the upper and lower limbs. In MMI, which is known as Dejerine syndrome, atrophy and paralysis of the tongue develop on the side of the lesion, while impaired tactile and deep sensation of the limbs and trunk and hemiplegia, both excluding the face, are observed on the contralateral side. However, these symptoms of LMI and MMI are unlikely to appear all at once. In particular, MMI may present with only those features indicative of a supratentorial disorder or provide falsenegative results on diffusion-weighted images (DWIs) of MRI early in the development of medullary lesions.3-6 In the present study, the imaging findings and clinical symptoms of patients with medullary infarction were reviewed, and the MRI and magnetic resonance angiography (MRA) findings, cause, and outcomes were examined.
Methods Of the 955 patients with ischemic stroke who were admitted to our center between April 1, 2007 and March 31, 2010, 37 (3.8%) had medullary infarction. These 37 patients, including 29 (3.0%) with LMI and 8 (0.8%) with MMI, were examined in this study. MRI examinations were conducted using an Achieva (1.5-T; Philips Electronics Japan, Tokyo, Japan) and a MAGNETOM Avant (1.5-T; Siemens Japan, Tokyo, Japan). DWIs were taken using the spin echo echo planar imaging (EPI) method using the Achieva (repetition time [TR]/echo time [TE] 3829 msec/65 msec; slice thickness 5 mm; field of view [FOV] 230 mm 3 207 mm; matrix 144 3 100.8) or the Avant (TR/TE 4200 msec/81 msec; slice thickness 5 mm; FOV 230 mm 3 230 mm; matrix 128 3 102). MRAwas taken using the 3-dimensional time of flight (3D-TOF) method using the Achieva (TR/TE 19 msec/6.9 msec; FOV 200 mm 3 180 mm) or the Avant (TR/TE 22 msec/7 msec; FOV 220 mm 3 176 mm). Images were obtained in 5-mm slices parallel to the orbitomeatal (OM) line, and infarction was classified into upper, central, and lower medulla oblongata. Based on reports by Currier et al,1 Kim,2 and Vuilleumier et al,3 the upper, central, and lower medulla oblongata were defined as the area where the inferior cerebellar peduncle is raised posterolaterally, the area where the nodular surface attributed to the inferior olivary nucleus is observed, and the area where the bulge disappears and the cross section becomes round, respectively. For MMI, in addition to the classification by the levels in the medulla oblongata, infarction was also categorized into ventral (including the pyramid), central (including the medial lemniscus), and dorsal (including the medial longitudinal fasciculus).4,5 All patients were transported to our center by ambulance, and MRI was performed on the day they developed infarction. When
595
brainstem infarction was suspected at admission but there were no abnormal findings on MRI, MRI was repeated between day 3 and day 7 after onset. Carotid arterial ultrasound examination was performed for all patients, and 3-dimensional computed tomographic angiography (3D-CTA) and cerebral angiography were also performed for patients suspected of having cerebrovascular dissection. The background of the patients, neurologic signs and symptoms on admission, MRI and MRA findings, cause of disease, and 90-day outcomes were examined for the LMI group and the MMI group. Causes of medullary infarction were divided into 3 types: atherothrombosis, cardioembolism, and others (including arterial dissection). To diagnose cerebral arterial dissection, the criteria issued by the Working Group for Cerebral Artery Dissection were used.6 Functional outcome was evaluated using the modified Rankin Scale (mRS) score at 90 days. Patients who were transferred to the rehabilitation hospital were evaluated by interview. Statistical evaluation of data was performed with PASW Statistics 18 (SPSS; Chicago, IL). The Chi-square test was used for analyzing differences in patients’ baseline characteristics and risk factors between the 2 groups. The Student t test was used for analyzing the frequency of negative MRI and time from onset to MRI examination between the 2 groups. The Wilcoxon rank sum test was used for analyzing the relationship between the MRI findings and the National Institutes of Health Stroke Scale (NIHSS) score on admission, level of infarction, and mRS score at 90 days. P , .05 was considered statistically significant. All data are presented as mean 6 SD. This study was approved by the Ethics Committee at Saitama Medical University International Medical Center.
Results The clinical features of the 29 patients with LMI and the 8 patients with MMI are summarized in Tables 1 and 2, respectively, and mRS scores at 90 days after onset for patients with LMI or MMI are shown by the levels of infarction in Table 3. The mean age of the LMI group was 60 6 15 years (range 18-83), and that of the MMI group was 62 6 10 years (range 44-68). The LMI group consisted of 24 men and 5 women, and the MMI group was comprised of 7 men and 1 woman. As risk factors, 26 (89%) patients in the LMI group had hypertension, 8 (27%) had diabetes, 6 (20%) had dyslipidemia, 2 (6.8%) had atrial fibrillation, and 2 (6.8%) had a history of smoking, while 6 (75%) patients in the MMI group had hypertension, 3 (37%) had diabetes, 1 (12%) each had dyslipidemia and atrial fibrillation, and 2 (25%) had a history of smoking. There were no significant differences in risk factors between the 2 groups (P 5 .800). The neurologic signs and symptoms at the first examination included, in the LMI group, dizziness/
596
Table 1. List of cases of lateral medullary infraction
Age/ Patient Sex
Risk factor
Headache, posterior Dissociated cervical Horner sensory Cerebellar pain Dysphagia Dysarthria syndrome loss symptoms Cause Dizziness
Paralysis
Time from Level of Lesion onset to medullary initial MRI (h) infraction MRI
1 1 1 1 –
1 2 1 1 2
1 1 1 1 2
1 1 1 1 2
1 2 1 1 2
1 2 1 2 2
1 1 1 1 1
2 2 2 2 2
5.3 5.1 2 6 18.3
Upper Upper Upper Upper Upper
AD AT AT AT
1 1 1 2
1 2 2 2
1 1 2 1
1 1 2 1
2 1 1 2
1 1 1 2
2 2 2 2
2 2 2 2
19 23.3 10.3 21.6
AT AT AT CE AT
2 2 1 1 1
2 2 2 2 2
2 1 2 1 1
2 1 1 1 1
2 2 2 1 1
1 1 2 1 1
2 2 2 1 1
15 16 17 18 19 20 21 22
4/M5 84/F 55/M 66/F 37/M 46/F 40/M 64/M
DL, smoking HT HT HT, DM HT HT, DL, DM HT HT
AD CE AD AT AD AT AD AT
2 1 1 1 2 1 2 1
1 2 2 2 2 1 1 2
2 1 1 2 2 2 1 1
2 2 1 2 2 2 1 1
2 1 2 2 2 2 1 2
1 2 2 1 1 2 1 1
2 1 1 1 1 1 2 2
23 24
74/M HT, AF 62/M HT, DL, DM, smoking 38/M HT, DM 69/M HT 69/M HT, DL 62/M HT 56/F HT, DL, DM
CE AT
1 1
2 2
1 1
1 1
2 1
2 1
2 1
2 2 2 2 Left upper and lower limits 2 2 2 2 2 2 2 Right (old) 2 2
AD AT AT AD AT
2 2 2 1 1
1 2 2 1 1
1 2 1 1 2
1 2 1 2 2
2 2 1 1 2
1 1 1 1 2
2 2 1 1 1
2 2 2 2 2
71/M 71/M 57/M 18/M 59/M
6 7 8 9
39/M 59/M 63/M 83/M
10 11 12 13 14
25 26 27 28 29
Presence or mRS at 90 absence of NIHSS at days after dissection admission onset
Left VA Right PICA Right VA Left VA Left VA
Left VA 2 Right VA Left VA 2
4 2 5 1 3
2 1 3 2 2
Upper Upper Upper Upper
Right VA Left PICA Right VA BA
Right VA 2 2 2
2 6 2 6
2 4 1 5
23.3 9.1 8.1 11 3.1
Central Central Central Central Central
Left VA Left VA Left VA Left VA Right VA
2 2 2 2 2
5 3 2 4 3
2 0 1 1 4
3.3 5 11.3 11 23.3 10 10 24
Central Central Central Central Central Central Central Central
Left VA Right VA Right VA Left VA Left VA Left PICA Right VA Right VA
Left VA 2 Right VA 2 Right VA 2 Right VA 2
1 5 4 3 5 1 1 5
1 4 2 1 1 1 3 2
5 24
Central Central
Left VA Right VA
2 2
2 3
1 3
12 12 10 25.8 9
Central Central Central Central Lower
Both VA Left VA Left VA Left VA Right VA
Left VA 2 2 Left VA 2
3 1 1 2 1
1 1 2 3 1
2
2
2
2
Abbreviations: AD, arterial dissection; AF; atrial fibrillation; ASD, atrial septal defect; AT, atherothrombosis; BA, basilar artery; CABG, coronary artery bypass graft; CE, cardiac embolism; DCM, dilated cardiomyopathy; DL, dyslipidemia; DM, diabetes mellitus; F, female; HT, hypertension; M, male; MI, myocardial infraction; MRA, magnetic resonance angiography; MRI, magnetic resonance imaging; mRS, modified Rankin Scale; NIHSS, National Institutes of Health Stroke Scale; PICA, posterior inferior cerebellar artery; SAS, sleep apnea syndrome; VA, vertebral artery.
T. FUKUOKA ET AL.
AD AT AD AD AT
76/M 77/M 62/M 74/M 53/F
HT, DL HT HT None HT, DM, DL, DCM SAS HT, DM, ASD HT HT, DM, CABG HT, DM, MI HT, CABG DVT AF, HT HT
1 2 3 4 5
Vessel poorly visualized on MRA
2 Central Central Central Central (1 lower) 5 12 11 11 2 2 2 1 L 2 R L 2 1 1 1 1 1 1 2 68/M 58/M 53/M 68/M 5 6 7 8
HT, DM HT Smoking DM, HT, DL, AF, ASD
AT AT AT AT
2 1 1 2
L L R and L L
2 Central (1 upper) 2 1 L 1 2 44/M 4
HT, migraine
AD
1
L
2 Upper 4 1 2 1 L 2 1 AT 3
HT 65/F
1 2
597 Abbreviations: AF; atrial fibrillation; ASD, atrial septal defect; AT, atherothrombosis; BA, basilar artery; DL, dyslipidemia; DM, diabetes mellitus; F, female; HT, hypertension; L, left; M, male; MRI, magnetic resonance imaging; mRS, modified Rankin Scale; NIHSS, National Institutes of Health Stroke Scale; PICA, posterior inferior cerebellar artery; R, right; VA, vertebral artery.
1 1 1 4 4 4 10 6 BA RVA 2 R PICA
3 4 R VA (dissection)
1 5 R VA
1 1 4 5 R VA R VA
Ventral Ventral 1 central Ventral 1 central 1 dorsal Ventral 1 central 1 dorsal Ventral Ventral Ventral Ventral 1 central 1 dorsal Upper Upper 14 8 2 2 L L 1 2 2 2 AT AT 77/M 61/M
Patient
HT, DM Smoking
Age/ Sex
Risk factor
Cause
2 2
Paralysis
Dysarthria
Deep sensation Dizziness, vomiting
L L.
Level of medullary infraction Deviation of tongue
Time from onset to MRI (h) Headache, posterior cervical pain
Table 2. List of cases of medical medullary infarction
Presence or absence of lesion in initial MRI
Location of infraction in horizontal section
Occluded vessel
NIHSS at admission
mRS at 90 days after onset
CLINICAL REVIEW OF MEDULLARY INFARCTION
lightheadedness in 19 patients (65%), headache in 10 (34%), dysphagia in 19 (65%), dysarthria in 18 (62%), Horner syndrome in 12 (41%), dissociated sensory loss in 19 (65%), hemiplegia in 1 (3.4%), and cerebellar symptoms in 16 (55%). In the LMI group, dysphagia was more common among patients with infarction in the upper medulla oblongata. It was observed in 7 of 9 (78%) patients with infarction in this area, whereas it was reported in 9 of 18 (50%) patients with lesions in the central medulla oblongata. In the MMI group, neurologic signs and symptoms at the first examination consisted of headache in 3 patients (37%), dizziness and vomiting in 4 (50%), deviation of the tongue in 3 (37%), impaired deep sensation in 6 (75%), and hemiplegia not involving the face in 8 (including one with quadriplegia; 100%). The causes of infarction were, in the LMI group, atherothrombosis in 17 patients (58%), embolism caused by cardiac sources in 3 (10.3%), and arterial dissection in 9 (31%), and in the MMI group, atherothrombosis in 7 (87%) and arterial dissection in 1 (12.5%). The frequency of arterial dissection was significantly higher in the LMI group than in the MMI group (P 5 .007). In the LMI group, the occluded or narrowed vessels confirmed by MRA were the vertebral artery (VA) in 25 patients, the posterior inferior cerebellar artery (PICA) in 3, and the basilar artery (BA) in 1. The lesions were located on the right side in 11 patients and on the left side in 18. In 4 patients, the second MRI-DWI detected infarction lesions for the first time. The time from onset to MRI examination was 12.4 6 7.3 hours in the LMI group. The time from onset to MRI examination was significantly shorter in the initial MRI-negative group (7.5 6 2.3 hours) than in the initial MRI-positive group (13.2 6 7.8 hours; P 5 .022). There was no significant correlation between NIHSS score at admission and MRI findings (positive or negative MRI) in the LMI group (P 5 .138). In the MMI group, the occluded or narrowed vessels were the VA in 5 patients, the PICA in 1, and the BA in 1. Lesions were on the right side in 6 patients and on the left side in 1 patient. Lesions were observed for the first time on the second MRI in 3 patients. The time from onset to MRI examination was 8.3 6 4.0 hours in the LMI group. The time from onset to MRI examination was significantly shorter in the initial MRI-negative group (3.6 6 1.5 hours) than in the initial MRI-positive group (11.1 6 2.1 hours; P 5 .022). There was no significant correlation between NIHSS score at admission and MRI findings (positive or negative MRI) in the MMI group (P 5 .337). The mRS scores at 90 days after onset (90-day outcome) were, in the LMI group, 0 in 1 patient, 1 in 12, 2 in 8, 3 in 4, 4 in 3, and 5 in 1, and in the MMI group, while no patient had a score of 0, 2, or 5—the scores were 1 in 6 patients, 3 in 1, and 4 in 1. As shown in Table 3, the frequency of good recovery (mRS 0-1) was 22% in the upper group (2/9), 52% in the central group (10/19), and 100% in the
T. FUKUOKA ET AL.
598
Table 3. Modified Rankin scale score 90 days after onset by level of infraction Level of infarction mRS at 90 days after onset LMI
MMI
Infarction location Upper Central Lower Upper Central Ventral Ventral 1 central Ventral 1 central 1 dorsal
0
1
2
3
4
5
9 cases 19 cases 1 case 3 cases 5 cases 4 cases 1 case
None 1 case None None None None None
2 cases 9 cases 1 case 3 cases 3 cases 4 cases 1 case
4 cases 4 cases None None None None None
1 case 3 cases None None 1 case None None
1 case 1 case None None 1 case None None
1 case None None None None None None
3 cases
None
1 case
None
1 case
1 case
None
Abbreviations: LMI, lateral medullary infarction; MMI, medial medullary infraction; mRS, modified Rankin Scale.
lower group (1/1; P 5 .073). The outcome of LMI with more rostral lesions tended to be poor. Of cases of MMI, none had infarction lesions confined to the lower area. The frequency of good recovery (mRS 0-1) was 100% in the upper group (3/3) and 60% in the central group (3/5; P 5 .237). The frequency of good recovery (mRS 0-1) was 100% in the ventral group (4/4), 100% in the ventral and central group (1/1), and 33% in the ventral and central and dorsal group (P 5 .068). The outcome of MMI with more dorsal lesions tended to be poor.
Discussion Of the patients with medullary infarction examined in the present study, the mean age at onset was 60 6 15 years (range 18-83) for LMI and 62 6 10 years (range 44-68) for MMI; they were relatively young compared to patients with cerebral infarction in general. Other investigators also reported a younger age of onset for these conditions. LMI and MMI developed at an average age of 57 6 11.9 years and 62 6 10 years, respectively, according to Kim2 and Kim and Han,5 and 60.7 6 12.4 years and 65.0 6 12.3 years, respectively, according to Kameda et al.7 Kim et al8 reported that, of the symptoms of LMI, infarction in the rostral medulla oblongata in particular often presented with dysphagia, hoarseness, and facial palsy, and that the more caudal the lesions were, the more frequent the symptoms of dizziness/lightheadedness, nystagmus, and ataxic gait were. In the present study, dysphagia was frequently observed in patients with lesions in the upper medulla oblongata, which was probably because the nucleus ambiguus is located in this area. Kameda et al7 reported headache in 47% of patients with LMI. The incidence of headache in the present study was as high as 34%, which is in line with the findings of Kameda et al.7 The relatively high incidence of headache appeared to be associated with the fact that some of our patients had cerebrovascular dissection, which will be discussed later.
For MMI, Kim et al5 reported hemiplegia in 91% of the patients they examined, sensory disturbance in 73%, and vertigo/dizziness in 59%, and Kameda et al7 reported headache in 13% of their patients, dizziness in 56%, vomiting in 44%, deviation of the tongue in 30%,sensory disturbance in 68%, and hemiplegia in 93%. Bassetti et al9 reported, in their examination of 7 patients with MMI, dizziness/vomiting in 71% of patients, hyperalgesia in 85%, headache in 57%, dysarthria in 42%, and disturbance of consciousness in 28%, which were similar to the distributions of the initial symptoms in previous reports. In the present study, paralysis (except facial) was seen in all patients, impaired deep sensation was seen in 6 patients, and deviation of the tongue was seen in only 3 patients. The symptoms of Dejerine syndrome were not common. Some patients were diagnosed with MMI for the first time based on imaging findings. We considered that deviation of the tongue and impaired deep sensation were important as symptoms indicative of MMI. On diagnostic imaging, MRI-DWI on hospitalization did not detect culprit lesions, and the second MRI confirmed infarction lesions in 4 of 29 patients with LMI and 3 of 8 patients with MMI. Pedraza et al10 suggested that DWI is less sensitive during the first 24 hours after the onset of cerebral infarction, and that false-negative results may occur, especially in small infarctions in the posterior circulation. In their study of 139 patients with cerebral infarction who underwent MRI within 48 hours after onset, Oppenheim et al11 examined 8 patients (5.8%) for whom DWI produced falsenegative results. False-negative results occurred in 19% of patients with infarction in the posterior circulation and in 31% of patients with vertebrobasilar infarction. The mean size of the infarction lesions was 0.19 6 0.16 cm.3 As for the causes of false-negative results, they suggested lesion location, time to imaging, the small size of lesions that does not allow image visualization on DWI, inadequate signalto-noise (S/N) ratio in the early phase of infarction, and magnetic artifact produced during EPI that causes poor visualization of the brainstem. Wang et al12 reported a patient
CLINICAL REVIEW OF MEDULLARY INFARCTION
with cerebral infarction in the area of the middle cerebral artery whose DWI taken 27 hours after onset was negative, but the DWI 7 days after onset was positive. They speculated that false-negative results occur when there is a mismatch between diffusion and perfusion—that is, when there is sufficient reperfusion. Kitis et al13 confirmed 1 false-negative case among 13 cases (7.6%) of LMI. In the present study, the first DWI was negative in 4 of 29 patients (13%) with LMI and 3 of 8 patients (37%) with MMI, revealing a higher incidence of false-negatives among patients with MMI. In the present study, the time of imaging was related to false-negative results. In addition, we performed MRI with a slice thickness of 5 mm, and the interslice gap was 1.5 mm. Therefore, some early infarction lesions may have been hidden in these interslice gaps, and this contributed to the false-negative results. Therefore, careful neurologic examination is still considered important for the detection of medullary infarction. Cerebral artery dissection is known as a cause of LMI and MMI. Vertebral artery dissection (VAD) in general is reported to develop in 1 to 1.5 per 100,000 people.14 The incidence of LMI caused by VAD was 15% (20/123 patients) and 29% (31/107 patients) according to Kim2 and Kameda et al,7 respectively. Hosoya et al15 reported that, among 93 patients with LMI, cerebral artery dissection was determined to be the cause in 23 patients (24.7%) and suspected as a cause in an additional 27 patients. Including the suspected cases, the incidence of LMI caused by cerebral artery dissection was 53.7% (50/93 patients). In the present study, based on strict diagnostic criteria, 9 of 29 patients (31%) with LMI had cerebral arterial dissection. The incidence based on definite diagnosis (31%) was higher than the incidences in previous reports. The incidence of MMI caused by VAD was reported to be 2.3% (2/86 patients), 21% (4/19 patients), and 18% (2/11 patients) by Kim,2 Kameda et al,7 and Toyoda et al,16 respectively. In the present study, 1 of 8 patients (12.5%) had MMI caused by VAD, which was less common than for LMI caused by VAD. The 90-day outcome of LMI tended to be poorer in patients with more rostrally located lesions. This is probably because even without quadriplegia, persisting dysphagia affects activities of daily living in general as a prognostic factor for LMI. When we compared the outcome of LMI located at different levels of the medulla oblongata, the outcome was better for infarction in the lower medulla oblongata than for infarction in the upper or central medulla oblongata. For MMI, Kim and Han5 suggested that severe hemiplegia and age were the factors associated with poor prognosis, and that rostral and ventral locations of MMI lesions were associated with the incidence of dyskinesia. Although the infarction lesion was confined to the lower medulla oblongata in none of the 8 patients with MMI in our study, severe cases were more common among those with caudal (infarction in the central medulla oblongata) and ventral lesions. Our examination of medullary infarction revealed that both LMI and MMI were often caused by atherothrombo-
599
sis, and that VAD was frequently observed in patients with LMI. Careful consideration of neurologic findings is essential, because MRI-DWI on the day of onset may not show abnormalities, so that diagnosis of infarction in the early phase can be difficult. Because MMI rarely presents with typical symptoms, and because its clinical symptoms and neurologic findings are diverse, MMI was not detected at onset in many cases in our study.
References 1. Currier RD, Giles CL, Dejong RN. Some comments on Wallenberg’s lateral medullary syndrome. Neurology 1961;11:778-791. 2. Kim JS. Pure lateral medullary infarction: Clinicalradiological correlation of 130 acute, consecutive patients. Brain 2003;126:1864-1872. 3. Vuilleumier P, Bogousslavsky J, Regli F. Infarction of the lower brainstem. Clinical, aetiological and MRItopographical correlations. Brain 1995;118:1013-1025. 4. Parent A. Carpenter’s human neuroanatomy. 9th ed. Baltimore: Williams & Wilkins, 1996:421-458. 5. Kim JS, Han YS. Medial medullary infarction: Clinical, imaging, and outcome study in 86 consecutive patients. Stroke 2009;40:3221-3225. 6. Takagi M. Guideline for diagnosis and treatment of cerebral artery dissection. In: Guidelines for treatment of stroke in younger patients, Research Grant for Cardiovascular Diseases 12shi-2, National Multicenter Study of Diagnosis, Treatment and Preventive Strategy for Stroke in Younger Generations. Osaka, Japan: Cerebrovascular Division, Department of Medicine, National Cardiovascular Center, 2003:85-89. 7. Kameda W, Kawanami T, Kurita K, et al. Lateral and medial medullary infarction: A comparative analysis of 214 patients. Stroke 2004;35:694-699. 8. Kim JS, Lee JH, Suh DC, et al. Spectrum of lateral medullary syndrome. Correlation between clinical findings and magnetic resonance imaging in 33 subjects. Stroke 1994; 25:1405-1410. 9. Bassetti C, Bogousslavsky J, Mattle H, et al. Medial medullary stroke: Report of seven patients and review of the literature. Neurology 1997;48:882-890. 10. Pedraza S, Osuna MT, Davalos A, et al. False negative diffusion in acute ischemic stroke. Rev Neurol 2002;34:1127-1129. 11. Oppenheim C, Stanescu R, Dormont D, et al. False-negative diffusion-weighted MR findings in acute ischemic stroke. AJNR Am J Neuroradiol 2000;21:1434-1440. 12. Wang W, Goldstein S, Scheuer ML, et al. Acute stroke syndrome with fixed neurological deficit and falsenegative diffusion-weighted imaging. J Neuroimaging 2003;13:158-161. 13. Kitis O, Calli C, Yunten N, et al. Wallenberg’s lateral medullary syndrome: Diffusion-weighted imaging findings. Acta Radiol 2004;45:78-84. 14. Schievink WI. Spontaneous dissection of the carotid and vertebral arteries. N Engl J Med 2001;344:898-906. 15. Hosoya T, Nagahata M, Yamaguchi K. Prevalence of vertebral artery dissection in Wallenberg syndrome: Neuroradiological analysis of 93 patients in the Tohoku District, Japan. Radiat Med 1996;14:241-246. 16. Toyoda K, Imamura T, Saku Y, et al. Medial medullary infarction: Analyses of eleven patients. Neurology 1996; 47:1141-1147.