Brain Advance Access originally published online on May 21, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Brain, Vol. 126, No. 8, 1864-1872,
August 2003
© 2003 Guarantors of Brain
doi: 10.1093/brain/awg169
Pure lateral medullary infarction: clinical–radiological correlation of 130 acute, consecutive patients
Department of Neurology, University of Ulsan, Asan Medical Center, Seoul, South Korea
Correspondence to: Jong S. Kim, MD, Department of Neurology, Asan Medical Center, Song-Pa PO Box 145, Seoul 138-600, South Korea E-mail: jongskim{at}amc.seoul.kr
Received July 4, 2002. Revised October 23, 2002. Second revision March 13, 2003. Accepted March 24, 2003.
 |
Summary |
|---|
 Top Summary IntroductionSubjects and methodsResultsDiscussionReferences |
|---|
Although there have been attempts to make clinical–MRI correlation in patients with lateral medullary infarction (LMI), studies with a large number of patients are unavailable. In this study, clinical features, MRI findings and angiogram results of 130 acute, consecutive patients with pure LMI were studied and correlated. MRI-identified lesions were classified rostro-caudally as rostral, middle and caudal, and horizontally as typical, ventral, large, lateral and dorsal. The distribution of horizontal subtypes was significantly different (P <0.001) among three rostro-caudal lesions in that rostral lesions tend to be ventral types and caudal lesions are lateral types. Patients with rostrally located lesions had dysphagia, facial paresis (P < 0.01, each) and dysarthria (P < 0.05) significantly more often, and severe gait ataxia and headache (P < 0.05, each) less often than those with caudal lesions. The frequencies of dysphagia (P < 0.01), dysarthria (P < 0.01) and bilateral trigeminal sensory pattern (P < 0.05) were significantly different among horizontal subtypes in that these symptoms were frequent in patients with ‘large type’ as compared with those with lateral type lesions. Angiograms performed in 123 patients showed vertebral artery (VA) disease in 67% and posterior inferior cerebellar artery (PICA) disease in 10%. The presumed pathogenetic mechanisms included large vessel infarction in 50%, arterial dissection in 15%, small vessel infarction in 13% and cardiac embolism in 5%. Dissection occurred more often in patients with caudal (versus rostral) lesions (P < 0.01), whereas dorsal type infarcts (versus other types) were related more often to cardiogenic embolism and normal angiogram findings (P < 0.05, each). Patients with isolated PICA disease (versus those with VA disease) more often had cardiogenic embolism (P < 0.05) and less often had dissection (P < 0.01). It is concluded that rostro-caudal and horizontal classification of MRI helps us to understand the clinical and, partly, the aetiopathogenetic aspect of the heterogeneous LMI syndrome.
Keywords: medulla; cerebrovascular disease; lateral medullary infarction; MRI; angiogram
Abbreviations: LMI= lateral medullary infarction; PICA = posterior inferior cerebral artery; VA = vertebral artery
|
Introduction |
|---|
|
TopSummaryIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
Since the first description of Wallenberg’s syndrome >100 years ago (Wallenberg, 1895,
1901), clinical (Merritt and Finland, 1930; Lewis et al., 1952; Peterman and Siekert, 1960; Currier et al., 1961; Grant, 1966; Hörnsten, 1974; Norrving and Cronqvist, 1991) and pathological (Merritt and Finland, 1953; Fisher et al., 1961; Gillilan, 1964) findings of lateral medullary infarction (LMI) have been reported occasionally. However, pre-mortem lesion identification became possible only after MRI was introduced (Bogousslavsky et al., 1986; Ross et al., 1986).
Recent clinical (Sacco et al., 1993) or clinical–radiological correlation studies (Kim et al., 1994; Vuilleumier et al., 1995; Kim et al., 1997) using MRI have rapidly expanded our understanding of LMI syndromes (Caplan, 1996). However, the number of patients included in these reports was too small to perform a reliable clinical–MRI correlation study. Furthermore, in the previous studies, patients with concomitant cerebellar infarction were included. Because symptoms caused by cerebellar involvement were likely to be confused, the previous results may not reflect the true features of pure LMI syndrome. Therefore, a larger study using pure LMI patients based on MRI data is required.
In the present study, 130 consecutive LMI patients were studied in whom MRI showed appropriate lesions in the lateral medulla but not in the cerebellum. Clinical, MRI findings and aetiopathogenesis were studied and correlated.
|
Subjects and methods |
|---|
|
TopSummaryIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
Between March 1994 and February 2002, I examined 222 consecutive patients at the Asan Medical Center who were suspected of having LMI; they had at least two of the following symptoms/signs: sudden vertigo, nausea/vomiting, gait ataxia, Horner sign, dysphagia, hoarseness and spinothalamic sensory symptoms/signs.
Three patients did not undergo MRI, in two of them due to their poor clinical condition. In the remaining 219 patients, axial T2 [(TR/TE) repetition time/echo time 2500 ms/80 ms], proton density and gadolinium-enhanced T1-weighted scans were performed in a horizontal plane at 3 mm intervals from the medulla to the midbrain. A sagittal T1-weighted image was also obtained. We excluded the following 89 patients: (i) 51 who had concomitant infarction outside of the medulla (middle cerebral artery territory in one, the pons in five, the medial medulla in three and the cerebellum in 42); (ii) 11 in whom MRI lesions were not visualized; and (iii) 25 who were examined by us >7 days after the onset of symptoms. In the remaining patients, 17 had experienced previous strokes (n = 13) or transient ischaemic attacks (n = 4). Among them, two patients who had residual gait ataxia and sensory symptoms, respectively, were excluded as symptomatic interpretation of the patients was difficult. However, three with minor residual hemiparesis and one with visual field defect were included. Thus, 130 patients became the subject of this study. The majority of them were seen within 3 days after the onset. All the patients were examined by the author himself, with specific signs and symptoms of LMI recorded prospectively, except for seven in whom a retrospective chart review was carried out.
The mode of onset was categorized as sudden and non-sudden. Non-sudden onset was defined as stepwise or progressive development of symptoms/signs for longer than 1 day. Among the symptoms/signs of LMI, vertigo was defined as a whirling, rotational sensation of the self or environment. Nystagmus was considered to be present only when it was seen on forward gaze. Dysphagia was designated as ‘severe’ when patients required a naso-gastric tube for feeding and ‘mild’ when they had difficulty in feeding without requiring the tube. Gait ataxia was designated as ‘severe’ when the patient was unable to stand or walk alone (Kim et al., 1994).
Risk factors for stroke such as hypertension, diabetes mellitus, current cigarette smoking and heart disease were recorded. An electrocardiogram was performed in all patients, and those who were <50 years of age or without conventional vascular risk factors underwent transthoracic and transoesophagal echocardiograms. One hundred and twenty-three patients (95%) underwent angiography (conventional angiogram in 52, MR angiogram in 71).
The patients’ MRI findings (T2-weighted axial image) were assessed retrospectively and classified by two neurologists who were blinded to the clinical information. Rostro-caudally, the lesions were categorized as: the rostral medulla characterized by massive bulging of the dorso-lateral area due to the restiform body; the middle medulla by bulging of the lateral surface due to the inferior olive; and the caudal medulla by a relatively round shape without bulging of the lateral surface (Kim et al., 1994). Horizontally, the lesions were classified with a slight modification of previous publications (Currier et al., 1961; Kim et al., 1994; Vuilleumier et al., 1995). Diagonal band-shaped lesions sparing the most dorsolateral portion were the most common and were therefore designated as ‘typical type’. Similarly shaped, but more ventrally situated lesions involving some portion of the inferior olive and sparing relatively large portions of the dorsolateral area were classified as ‘ventral’ type. Large lesions extending ventrally so as to involve some portion of the olivary nucleus and dorsally to involve most (or all) of the dorsolateral area were classified as ‘large type’. Lesions restricted to the most dorsal or dorsolateral portion were classified as ‘dorsal type’. Some lesions, usually at the caudal medulla, were restricted to the lateral, superficial area without extending dorsally and were classified as ‘lateral type’ (Fig. 1). Other lesions not classifiable were designated as ‘unclassifiable type’.
|
For this article, we arbitrarily defined the aetiopathogenesis with a slight modification of previous criteria (Kim et al., 1998
): (i) Dissection: patients with (a) an obvious history of recent (within 1 week) head/neck trauma or sudden neck rotation (chiropractic manipulation, golf practicing, yoga, etc.); (b) concurrent neck or occipital pain; (c) no evidence of atherosclerotic vascular changes on angiogram; and (d) angiogram findings consistent with dissection: fusiform aneurysm, double lumen sign or an elongated, tapered stenosis/occlusion (Shimoji et al., 1984; Yamaura et al., 1990). (ii) Large vessel disease: patients with (a) at least one conventional risk factor for atherosclerosis; (b) angiogram findings consistent with atherosclerotic vascular lesion; and (c) no evidence of dissection. (iii) Cardiogenic embolism: patients with (a) concurrent emboligenic heart disease (atrial fibrillation, prosthetic valve, sick sinus syndrome, valvular disease, cardiomyopathy, recent myocardial infarction); (b) young patients (age <50 years) who had patent foramen ovale with right to left shunt; (c) no angiographic features of atherosclerosis; and (d) no hypertension or diabetes mellitus. (iv) Small vessel disease: patients with (a) hypertension; (b) no emboligenic heart disease; and (c) normal angiogram findings. (v) Unknown: patients (a) who did not undergo MRI angiogram or conventional angiogram; and (b) ambiguous cases such as those with concomitant presence of emboligenic heart disease and atherosclerotic vascular changes
For statistics, we used 
2 tests for the comparison of symptoms between rostral and caudal groups, and among different horizontal subtypes. Fisher’s exact test was used when the number of the cells was small. 2 test and correspondence analysis were done for the correlation between vertical and horizontal lesions of the medulla. All statistical tests were performed with the use of the SPSS program (version 10.0); P values <0.05 were regarded as indicating significance.
|
Results |
|---|
|
TopSummaryIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
Demographic features and risk factors
There were 90 men and 40 women, aged from 28 to 84 years (mean 57 ± 11.9 years). Risk factors included hypertension in 83 (64%), diabetes mellitus in 32 (25%), current cigarette smoking in 32 (25%) and atrial fibrillation in six (5%). Six patients had a past history of coronary heart disease. Two patients underwent radiation therapy on the neck for a laryngeal cancer, 4 and 5 years before, respectively. One had concurrent hyperthyroidism and one had a history of active migraine that occurred at the time of stroke. Fifteen patients did not have any conventional risk factors.
Clinical findings
The patients’ neurological symptoms/signs are summarized in Table 1. The onset was sudden in 97 patients (75%) and non-sudden in 40 (25%) patients. Among the patients with non-sudden onset, the first symptoms/signs were usually headache, vertigo, dizziness or gait ataxia, whereas sensory symptoms/signs, dysphagia, hoarseness and hiccups tended to occur later. The patients’ symptoms/signs were divided into very common (
90%), moderately common (50–70%) and less common (
40%). Very common symptoms/signs included sensory symptoms/signs, gait ataxia, dizziness and Horner sign. Sensory symptoms/signs were the most frequent manifestation, occurring in 125 patients (96%). The patterns included ipsilateral trigeminal–contralateral limb/body (ipsilateral trigeminal pattern) in 34 (26%), contralateral trigeminal–contralateral limb/body (contralateral trigeminal pattern) in 32 (25%), bilateral trigeminal–contralateral limb/body (bilateral trigeminal pattern) in 18 (14%), limb/body involvement without trigeminal involvement (isolated limb/body pattern) in 27 (21%) and trigeminal involvement without limb/body involvement (isolated trigeminal pattern) in 13 (10%). Sensory gradient or level was observed in 28 patients (22%); symptoms were more severe in the lower extremities in 20, and more severe in the upper extremities in eight patients. Although selective impairment of spinothalamic sensory modality (pinprick and temperature) was usual, 15 patients (12%) also had decreased vibration sense in the hypalgic areas, usually to a mild degree. In five patients (4%), sensory symptoms/signs were not present.
|
Moderately common symptoms/signs included dysphagia, hoarseness, vertigo, nystagmus, limb ataxia, nausea/vomiting and headache. Forty percent of the patients required a naso-gastric tube. The nystagmus was horizontal or horizontal–rotational, to the side opposite to the lesions, and usually became more prominent on looking downward. Headaches generally started in the beginning or a few days before the onset of other symptoms/signs and subsided in several days or weeks. They occurred most often in the ipsilateral occipital area, followed by the frontal region. They were described as dull in 49, aching in nine, throbbing in eight, and paroxysmal in one patient. The degree of severity was mild in 32, moderate in 27 and severe in eight patients. Less common symptoms/signs included skew deviation, diplopia, dysarthria, facial paresis and gaze deviation. During the clinical course, seven (5%) patients developed respiratory difficulties along with aspiration pneumonia. One of them died in the acute stage.
To elucidate closely related symptoms/signs, 2 tests were performed on each pair of symptoms/signs of LMI, except for very frequent ones (dizziness, gait ataxia, Horner sign and sensory symptoms). The results are summarized in Table 2. Among diverse symptoms/signs of LMI, vertigo, diplopia, nystagmus, skew deviation, nausea/vomiting and severe gait ataxia tended to co-occur, whereas dysphagia, hoarseness, dysarthria and hiccups tendeds to occur simultaneously in one patient. Limb ataxia and headache were not related to any of the symptoms/signs.
|
Classification of LMI according to MRI findings
MRI showed that the medullary lesions were located in the right in 72 and in the left in 58 patients. Rostro-caudally, they were located in the rostral medulla only in 21, middle medulla only in 26, and caudal medulla only in 27 patients. Fourteen patients had lesions in the rostral + middle medulla, 31 had lesions in the middle + caudal medulla, and 11 had lesions in the rostral + middle + caudal medulla. Thus, rostro-caudally, the lesions were categorized as 197 lesions (46 rostral, 82 middle and 69 caudal lesions).
Horizontally, the lesions were classified as defined above. One patient had a caudal lesion presenting with two MRI cuts with different types, typical and lateral. Therefore, 198 horizontal subtypes were finally identified, which are summarized in Table 3. Four patients were considered unclassifiable: one had a small central lesion, one had fragmentary lesions and two had lesions with unclear margins. The distribution of horizontal subtypes was significantly (P < 0.001) different among three rostro-caudal levels; rostral lesions tended to be ventral or dorsal type, middle lesions of typical or large type, while caudal lesions tended to be of lateral type (Fig. 2).
|
|
Clinical–MRI correlation
Rostral–caudal correlation
For clinical–topographical correlation, we re-defined a ‘rostral group’ as the sum of the patients with lesions at the rostral medulla and those having lesions at both the rostral and middle medulla. The ‘caudal group’ was defined as the sum of the patients with lesions at the caudal medulla and those with lesions at both the middle and caudal medulla. The differences in clinical presentations between the rostral and caudal group patients are summarized in Table 1. Rostral group patients had dysphagia, dysarthria, facial paresis and a bilateral trigeminal sensory pattern significantly more often. They had severe gait ataxia, headache, isolated limb/body sensory pattern and sensory gradient worse in the leg than in the arm significantly less often than caudal group patients.
Horizontal correlation
Some of the patients had lesions at more than one rostro-caudal level, with different horizontal subtypes at each level. These patients were excluded for further analysis: for instance, those having a typical type lesion in the middle medulla and a ventral type lesion in the rostral medulla. However, a patient was designated as having a ‘large’ type lesion if they had a large type lesion in any level of medulla regardless of the types shown at other rostro-caudal levels.
The differences in clinical manifestations among different horizontal subtypes are presented in Table 4. The frequencies of dysphagia, dysarthria, Horner sign and bilateral trigeminal sensory pattern were significantly different among horizontal subtypes; patients with ‘large type’ lesions tended to have frequent dysphagia, hoarseness, dysarthria and bilateral trigeminal sensory pattern, whereas these symptoms were uncommon in those with lateral lesions. Horner sign tended to be uncommon in patients with dorsal lesions.
|
Patients with rostral group lesions had dysphagia more often, and those with lateral type less often than those with caudal group lesions. Therefore, I used the data to elucidate whether this difference in the frequency of dysphagia is related to rostral–caudal or horizontal differences. I found that only one (8%) of the patients with caudal lesions with either large (n = 2) or typical (n = 10) types had severe dysphagia, whereas this symptom occurred in all the patients with rostral lesions with either large (n = 5) or typical (n = 6) types. The difference was statistically significant (P < 0.01), suggesting that the frequency of severe dysphagia is related to rostro-caudal rather than horizontal differences.
Angiographic findings and presumed aetiopathogenesis
Among the 123 patients in whom an angiogram was performed, the results were abnormal in 95 (77%). There were 83 patients (67%) with vertebral artery (VA) disease and 12 (10%) with isolated posterior inferior cerebellar artery (PICA) disease. Among the patients with VA disease, 33 had distal VA disease (occlusion 23, stenosis 21), 34 had whole VA disease (occlusion 24, stenosis 10), while five had proximal VA disease with intact distal VA. Among the patients with isolated PICA disease, five had occlusion and seven had stenosis. The presumed pathogenic mechanisms of infarction were large vessel infarction in 65 (50%); arterial dissection in 20 (15%); small vessel infarction in 17 (13%); cardiac embolism in seven (5%); moya-moya disease in one; and unknown in 20 (15%) patients. They were significantly different between patients with rostral group lesions and those with caudal group lesions (Table 1).
I attempted to see whether there were rostro-caudal or horizontal differences among the frequency of angiographic vascular lesions (PICA disease, distal VA disease, whole VA disease, normal angiogram findings). I found that patients with PICA disease had typical lesions more often (versus non-typical lesions, P < 0.05) and that those with normal angiographic findings had dorsal type lesions more often (versus non-dorsal type lesions, P < 0.05). Patients with presumed dissection were found to have caudal group lesions more often (versus rostral group lesions, P < 0.01) and dorsal type lesions less often (versus non-dorsal type lesions, P < 0.01). Patients with cardiogenic embolism and those with normal angiographic findings were found more often to have dorsal type lesions (versus non-dorsal type lesions, P < 0.05 for each). Patients with isolated PICA disease (versus VA disease) were found to have cardiogenic embolism more often (P < 0.05) and dissection less often (P < 0.01).
|
Discussion |
|---|
|
TopSummaryIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
This is the largest collection of LMIs documented by MRI. Although the symptoms/signs of the patients were comparable with previous studies, the frequency of each symptom/sign was generally higher than in previous studies using MRI (Sacco et al., 1993
). This may be partly because we excluded the chronic patients and partly because most of the patients were evaluated prospectively by the author himself, thus minimizing missing values. Exceptions were the frequencies of limb ataxia and dysarthria, which were lower than those reported in previous studies (Norrving and Cronqvist, 1991; Sacco et al., 1993). This seems to be related to the exclusion of patients with cerebellar infarction. According to my unpublished data, LMI patients with concomitant cerebellar involvement had dysarthia and limb ataxia more often than those without, probably because these symptoms are related to cerebellar lesions.
Symptoms such as vertigo, skew deviation, nausea/vomiting, diplopia and severe gait ataxia were closely related, suggesting that these are all caused by an involvement of the vestibular nuclei or vestibular–cerebellar connections (Currier et al., 1961; Grant, 1966; Dieterich and Brandt, 1992). On the other hand, patients with dysphagia had hoarseness and dysarthria significantly more often than those without, illustrating that these symptoms are related to involvement of the nucleus ambiguus (Currier et al., 1961). Although the neural substrate for hiccups remains unknown, our data showed that it was closely associated with hoarseness or dysphagia, suggesting that involvement of the nucleus ambiguus or adjacent areas regulating respiration may be responsible for the generation of hiccups. A previous study (Currier et al., 1961) also reported that hiccups were common in patients with abnormal muscle functions innervated by the nucleus ambiguus.
In the present study, I attempted to classify the LMI lesions according to MRI findings, rostro-caudally and horizontally. Although similar trials have been carried out previously, the number of the patients was too small for statistical analysis (Kim et al., 1994; Vuilleumier et al., 1995), and rostro-caudal differences were not considered (Currier et al., 1961; Vuilleumier et al., 1995). The horizontally classified types in this study included typical, ventral, large, lateral and dorsal, which correspond to ‘typical’, ‘ventral’, ‘dorsal–ventral’, ‘superficial’ and ‘dorsal’ lesions of Currier et al. (1961), and to ‘dorsolateral’, ‘mid-lateral’, ‘large inferodorsolateral’, ‘inferolateral’ and ‘dorsal’ infarcts of Vuilleumier et al. (1995), respectively. Consistent with previous studies (Kim et al., 1994; Vuilleumier et al., 1995), rostral lesions tended to have ventral subtypes while caudal lesions were usually of lateral type. The horizontal subtype differences according to rostro-caudal levels may be related to the anatomical course of the VA: a pair of vertebral arteries located adjacent to the lateral surface of the caudal medulla ascend ventrorostrally to fuse into the basilar artery at the ponto-medullary junction.
There were symptomatic differences between patients with rostral lesions and those with caudal lesions, illustrating that rostro-caudal differences must be considered in the clinical–MRI correlation. In patients with caudal infarcts, the less frequent presence of a bilateral trigeminal sensory pattern, and more frequent isolated limb/body sensory pattern and sensory gradient worse in the leg than those with rostral lesions, are probably related to the fact that the lesions are usually situated lateral-superficially in the former group. Thus, the lesions may easily spare the trigeminal sensory fibres and tend selectively to involve the outer part of the sensory lammellation where sensations from the leg are represented (Matsumoto et al., 1988; Kim et al., 1997) (Fig. 1).
The more frequent occurrence of dysphagia and dysarthria in rostral group patients than in caudal group patients was a finding consistent with previous studies (Kim et al., 1994, 2000). This may be caused by the fact that the caudal lesions usually do not extend deeply enough to involve the nucleus ambiguus. Indeed, patients with large type lesions were found to have frequent dysphagia and hoarseness compared with those with lateral type lesions. However, this was an unlikely explanation because the patients with caudal lesions of typical or large types rarely developed severe dysphagia. Thus, the rostro-caudal difference in the frequency of dysphagia seems to be explained by the anatomical characteristics, i.e. the caudal area of the nucleus ambiguus is not directly related to pharyngeal muscles (Grant, 1966; Carpenter and Sutin, 1983).
However, although hoarseness was closely related to dysphagia, it was not more frequent in patients with rostral than those with caudal lesions. Facial palsy, usually mild, may be caused by involvement of either the facial nerve fascicles (Fisher and Tapia, 1987) or aberrant, looping corticobulbar fibres (Kuypers, 1958). Considering our data, these structures seem to be more easily involved in rostral than caudal lesions. On the other hand, headache was more common in patients with caudal infarcts than those with rostral lesions. Because headache was not related to any symptoms/signs of LMI, it is unlikely to be caused by an involvement of a particular region of the medulla. Rather, this observation appears to be related to the fact that dissection was more common in patients with caudal group lesions than those with rostral lesions.
Although statistical significance was reached only for a bilateral trigeminal pattern, horizontal classification seems to be related to sensory patterns: typical lesion–ipsilateral trigeminal, large lesion–bilateral trigeminal, ventral lesion–contralateral trigeminal, lateral lesion–isolated limb/body and sensory gradient worse in the leg, and dorsal lesion–isolated trigeminal patterns (Table 4). This should be related to anatomical characteristics of the medulla: the spinothalamic tract located most lateral-superficially, ascending secondary trigeminal tract medially and descending trigeminal tract posteriorly (Kim et al., 1997) (Fig. 1).
We performed angiograms in 95% of the patients, and found that VA disease (67%) was much more frequent than isolated PICA disease (10%). However, an interpretation of the results should be made with caution because more than half of our patients were evaluated by MR angiography, which is less effective in the evaluation of the PICA than conventional angiography. Therefore, the prevalence of isolated PICA disease may have been underestimated. In addition, the degree of stenosis was not assessed precisely in our study. Despite these limitations, the frequency of vascular lesions shown was similar to those reported previously. Norrving and Cronqvist (1991) reported the occurrence of VA disease in 72% and PICA disease in 9%, while Sacco et al. (1993) reported that 73% of the LMI patients had angiographic/sonographic evidence of VA disease. Fisher et al. (1961) summarized pathological results of 42 autopsy-proven LMI patients, and found 27 (64%) had VA disease (including VA + PICA disease) and six (14%) had PICA diseases.
Although large vessel infarction (50%) was the most frequent cause of stroke, dissection (15%) was relatively common in our study. The frequency was similar to a recent series from Switzerland (14%) (Vuilleumier et al, 1995), and lower than the results of Sacco et al. (1993) where five out of 15 patients (33%) who underwent angiogram had evidence of vascular dissection. However, the prevalence of dissection cannot be made with certainty unless typical angiographic findings are presented, which actually are a rare occurrence in patients with clinically suspected dissection (Houser et al., 1984). The uncommon occurrence of cardiogenic embolism (5%) compared with other series (Vuilleumier et al., 1995) may be caused partly by our exclusion of patients with concomitant cerebellar infarction and partly by ethnic differences. In Korea, cardiogenic embolism is a cause of stroke in only 7% of the stroke population (Korean Neurological Association, 1993).
As expected, isolated PICA occlusion/stenosis was related more frequently to cardiogenic embolism and less often to dissection than other stroke mechanisms. Dissection was seen more often in patients with caudal than rostral lesions, possibly because it usually occurs at C1, C2 levels below the branching of the PICA (Schievink et al., 1994). Although a dorsal part of the medulla has been considered to be supplied solely by the PICA (Durvernoy, 1978; Amarenco and Hauw, 1989; Tatu et al., 1996), none of the patients with dorsal type lesions had isolated PICA disease. However, patients with dorsal type lesions had normal angiogram findings and emboligenic heart diseases more often than those with other types of lesions, suggesting that the occluded medullary branch of the PICA from a cardiac embolism may have been recanalized or undetected by anogiogram studies. A previous study also reported that dorsal type lesions were related to cardiogenic embolism (Vuilleumier et al., 1995). Aside from the observation that isolated PICA disease was associated more often with typical type lesions, the result of a vascular lesion–MRI topography correlation study was generally negative and does not provide an answer as to how the heterogeneous morphology of the LMI lesions is produced. A previous study using a conventional angiogram (Kim et al., 1998) suggested that the morphology of the MRI lesions is related to the rostro-caudal differences in vascular lesions, the speed of vascular disease progression and the degree of collateralization in the setting of extremely variable VA and PICA branches (Gillilan, 1964).
Regarding the prognosis, patients with large, rostral lesions tended to have severe dysphagia, aspiration pneumonia and prolonged admission. Nevertheless, our patients had an excellent prognosis, in that only one (0.8%) died during hospitalization. In the literature, the hospital mortality of LMI patients ranges from 2.9 to 11.6% (Lewis et al., 1952; Peterman and Siekert, 1960; Norrving and Cronqvist, 1991). Aside from improved respiratory care, the reason for the good prognosis in our results may be explained in several ways. First, we excluded patients with concomitant strokes outside the medulla, who were likely to have a poorer prognosis than those with small infarcts restricted to the medulla. Secondly, we excluded three patients who did not undergo MRI, two of them due to poor clinical condition. One of the two died in the acute stage. Finally, in our MRI-based study, patients with very small lesions showing fragmentary symptoms were included. These patients with an extremely benign clinical course would not have been included in previous studies where a diagnosis was dependent on clinical criteria only.
|
References |
|---|
|
TopSummaryIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
Amarenco P, Hauw JJ. Anatomie des artères cérébelleuses. Rev Neurol (Paris) 1989; 145: 267–76.[Medline]
Bogousslavsky J, Fox AJ, Barnett HJM, Hachinski VC, Vinitski S, Carey LS. Clinico-topographic correlation of small vertebrobasilar infarct using magnetic resonance imaging. Stroke 1986; 17: 929–38.
Caplan LR. Posterior circulation disease: clinical findings, diagnosis, and management. Cambridge (MA): Blackwell Science; 1996. p. 262–323.
Carpenter MB, Sutin J. Human neuroanatomy. 8th edn. Baltimore: Williams & Wilkins; 1983. p. 315–57.
Currier RD, Giles CL, DeJong RN. Some comments on Wallenberg’s lateral medullary syndrome. Neurology 1961; 1: 778–92.
Dieterich M, Brandt T. Wallenberg’s syndrome: lateropulsion, cyclorotation, and subjective visual vertical in thirty-six patients. Ann Neurol 1992; 31: 399–408.[CrossRef][Web of Science][Medline]
Durvernoy HM. Human brainstem vessels. Berlin: Springer-Verlag; 1978.
Fisher CM, Tapia J. Lateral medullary infarction extending to the lower pons. J Neurol Neurosurg Psychiatry 1987; 50: 620–4.
Fisher CM, Karnes WE, Kubik CS. Lateral medullary infarction: the pattern of vascular occlusion. J Neuropathol Exp Neurol 1961; 20: 323–79.[Web of Science][Medline]
Gillilan LA. The correlation of the blood supply to the human brain stem with clinical brain stem lesions. J Neuropathol Exp Neurol 1964; 23: 78–108.[Web of Science][Medline]
Grant G. Infarction localization in a case of Wallenberg’s syndrome. A neuroanatomical investigation with comments on structures responsible for nystagmus, impairment of taste and deglutition. J Hirnforsch 1966; 8: 419–30.[Medline]
Hörnsten G. Wallenberg’s syndrome. Acta Neurol Scand 1974; 50: 434–46.[Web of Science][Medline]
Houser OW, Mokri B, Sundt TM, Baker HL Jr, Reese DF. Spontaneous cervical cephalic arterial dissection and its residuum: angiographic spectrum. AJNR Am J Neuroradiol 1984; 5: 27–34.[Abstract]
Kim JS, Lee JH, Suh DC, Lee MC. Spectrum of lateral medullary syndrome: correlation between clinical findings and magnetic resonance imaging in 33 subjects. Stroke 1994; 25: 1405–10.[Abstract]
Kim JS, Lee JH, Lee MC. Patterns of sensory dysfunction in lateral medullary infarction: clinical–MRI correlation. Neurology 1997; 49: 1557–63.
Kim JS, Lee JS, Choi CG. Patterns of lateral medullary infarction: vascular lesion–magnetic resonance imaging correlation of 34 cases. Stroke 1998; 29: 645–52.
Kim H, Chung C-S, Lee K-H, Robbins J. Aspiration subsequent to a pure medullary infarction. Arch Neurol 2000; 57: 478–83.
Korean Neurological Association. Epidemiology of cerebrovascular disease in Korea. J Korean Med Sci 1993; 8: 281–9.[Medline]
Kuypers HGJM. Cortico-bulbar connections to the pons and lower brainstem in man: an anatomical study. Brain 1958; 81: 364–88.
Lewis GN, Littman A, Foley EF. The syndrome of thrombosis of the posterior inferior cerebellar artery: a report of 28 cases. Ann Intern Med 1952; 36: 592–602.[Web of Science][Medline]
Matsumoto S, Okuda B, Imai T, Kameyama M. A sensory level on the trunk in lower lateral brainstem lesions. Neurology 1988; 38: 1515–9.
Merritt H, Finland M. Vascular lesions of the hindbrain (lateral medullary syndrome). Brain 1930; 53: 290–305.
Norrving B, Cronqvist S. Lateral medullary infarction: prognosis in an unselected series. Neurology 1991; 41: 244–248.
Peterman AF, Siekert RG. The lateral medullary (Wallenberg) syndrome: clinical features and prognosis. Med Clin North Am 1960; 44: 887–96.[Web of Science][Medline]
Ross MA, Biller J, Adams HP, Dunn V. Magnetic resonance imaging in Wallenberg’s lateral medullary syndrome. Stroke 1986; 17: 542–5.
Sacco RL, Freddo L, Bello JA, Odel JG, Onesti ST, Mohr JP. Wallenberg’s lateral medullary syndrome: clinical–magnetic resonance imaging correlations. Arch Neurol 1993; 50: 609–14.
Schievink WI, Mokri B, O’Fallon WM. Recurrent spontaneous cervical-artery dissection. N Engl J Med 1994; 330: 393–7.
Shimoji T, Bando K, Nakajima K, Ito K. Dissecting aneurysm of the vertebral artery: report of seven cases and angiographic findings. J Neurosurg 1984; 61: 1038–46.[Web of Science][Medline]
Tatu L, Moulin T, Bogousslavsky J, Duvernoy H. Arterial territories of human brain: brainstem and cerebellum. Neurology 1996; 47: 1125–35.
Vuilleumier P, Bogousslavsky J, Barth F. Infarction of the lower brainstem. Clinical, aetiological and MRI–topographical correlations. Brain 1995; 118: 1013–25.
Wallenberg A. Acute bulbar affection (Embolie der art. cerebellar post. inf. sinistra?) Arch Psychiatr Nervenkr 1895; 27: 504–40.[CrossRef]
Wallenberg A. Anatomischer befund in einem als ‘acute bulbar affection (Embolie der art. cerebellar post. inf. sinistra?) beschreibenen falle. Arch Psychiatr Nervenkr 1901; 34: 923–59.[CrossRef][Web of Science]
Yamaura A, Watanabe Y, Saeki N. Dissecting aneurysms of the intracranial vertebral artery. J Neurosurg 1990; 72: 183–8.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  N. R Chopra, D. A Jones, N. Gadi, and F. Huwez Sudden onset of pain in the eye and blurring of vision BMJ, January 21, 2009; 338(jan21_2): a3111 - a3111. [Full Text]
|
|
|
|
 |
|
 J. S. Lee, S. W. Yong, O. Y. Bang, Y. S. Shin, B. M. Kim, and S. Y. Kim Comparison of Spontaneous Intracranial Vertebral Artery Dissection With Large Artery Disease Arch Neurol, December 1, 2006; 63(12): 1738 - 1744. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F Vuillier, L Tatu, E Dietsch, E Medeiros, and T Moulin Pontomedullary sulcus infarct: a variant of lateral medullary syndrome. J. Neurol. Neurosurg. Psychiatry, November 1, 2006; 77(11): 1276 - 1278. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kwon, J. H. Lee, and J. S. Kim Dysphagia in unilateral medullary infarction: Lateral vs medial lesions Neurology, September 13, 2005; 65(5): 714 - 718. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Thomke, J. J. Marx, G. D. Iannetti, G. Cruccu, S. Fitzek, P. P. Urban, P. Stoeter, M. Dieterich, and H. C. Hopf A topodiagnostic investigation on body lateropulsion in medullary infarcts Neurology, February 22, 2005; 64(4): 716 - 718. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||