Sensory Sequelae of Medullary Infarction
Differences Between Lateral and Medial Medullary Syndrome
Background and Purpose—A comparison between long-term sensory sequelae of lateral medullary infarction (LMI) and medial medullary infarction (MMI) has never been made.
Methods—We studied 55 patients with medullary infarction (41 with LMI and 14 with MMI) who were followed up for >6 months. We examined and interviewed the patients with the use of a structured format regarding the most important complaints, functional disabilities, and the presence of sensory symptoms. The nature and the intensity of sensory symptoms were assessed with the modified McGill-Melzack Pain Questionnaire and the visual analog scale, respectively.
Results—There were 43 men and 12 women, with an average age of 59 years. Mean follow-up period was 21 months. The sensory symptoms were the most important residual sequelae in LMI patients and the second most important in MMI patients. In LMI patients, the severity of residual sensory symptoms was significantly related to the initial severity of objective sensory deficits (P<0.05). Sensory symptoms were most often described by LMI patients as numbness (39%), burning (35%), and cold (22%) in the face, and cold (38%), numbness (29%), and burning (27%) in the body/limbs, whereas they were described as numbness (60%), squeezing (30%) and cold (10%), but never as burning, in their body/limbs by MMI patients. LMI patients significantly (P<0.05) more often cited a cold environment as an aggravating factor for the sensory symptoms than did the MMI patients without spinothalamic sensory impairment. The subjective sensory symptoms were frequently of a delayed onset (up to 6 months) in LMI patients, whereas they usually started immediately after the onset in MMI patients.
Conclusions—Our study shows that sensory symptoms are major sequelae in both LMI and MMI patients. However, the nature, the mode of onset, and aggravating factors are different between the 2 groups, which probably is related to a selective involvement of the spinothalamic tract by the former and the medial lemniscus by the latter. We suggest that the mechanisms for the central poststroke pain or paresthesia may differ according to the site of damages on the sensory tracts (spinothalamic tract versus medial lemniscal tract).
Recently, clinical and etio-pathophysiological aspects of medullary infarction have been increasingly published.1 2 3 4 5 6 7 These studies illustrate that sensory dysfunction is one of the major symptoms of medullary stroke, both lateral medullary infarction (LMI) and medial medullary infarction (MMI). Nevertheless, long-term sequelae of the patients with medullary infarction were rarely studied,8 9 10 and the residual sensory symptoms in these patients have never been properly investigated. A comparison between the sensory sequelae of LMI and MMI patients would be of interest, considering that this attempt may allow us to evaluate the consequences of damages on the 2 distinctly different sensory pathways, the spinothalamic tract and the medial lemniscal tract, each of which is involved separately by LMI and MMI. Therefore, in the present study, we attempted to compare the long-term sequelae of the patients with LMI with those with MMI, with special reference to their sensory manifestations.
Subjects and Methods
At the outpatient clinic of the Asan Medical Center, we examined 64 consecutive patients with clinically suspected medullary infarction (LMI and MMI) between April 1997 and October 1998. All patients underwent MRI that identified the medullary lesion compatible with the clinical symptoms. The single exception was a patient who had typical symptoms and signs of LMI, and had an MRI-identified lesion in the cerebellum but not in the medulla.
Of these patients, we excluded following: 1 presenting with clinical features of both LMI and MMI, with MRI-identified infarcts occurring in both the medial and lateral medullary areas; 2 who had experienced previous infarcts; 1 who developed a recurrent infarction (at the midbrain and the thalamus) 4 months after the LMI; and 5 patients whose follow-up period was <6 months. Thus, the remaining subjects were 55 patients who experienced a single episode of either lateral or medial medullary infarction and were examined >6 months after the onset. Of these patients, 45 had been admitted to the Asan Medical Center in the acute stage (<5 days, mostly <1 day after the onset) and had been examined by J.S.K. Others (n=10; patients 2, 13, 14, 21, 26, 30, 32, 41, 48, and 52) were transferred to our hospital or to the clinic of J.S.K. from other hospitals in the subacute stage, and detailed initial neurological findings were therefore unknown.
At the time of the examination and interview, the neurological examination was performed by one of the authors (J.S.K.). Pain perception was tested with the use of a pinprick, temperature with a cold tuning fork, and vibration sense with a 128-Hz vibrator. The deficit was categorized as “mild”, “moderate” and “severe” when the sensory perception was >70%, 30% to 70%, and <30%, respectively, of the normal side. Trigeminal sensation was compared with the sensation of an intact limb because it may be impaired bilaterally in patients with LMI.6 Position sensation of the fingers and toes was assessed as “mild” when a patient correctly identified the up-and-down movements, but with a feeling of “unclear” sensation; “moderate” when there was an occasional error; and “severe” when most of the movements could not be identified.
In addition, the patients were interviewed with a standardized questionnaire by another author (S.C.-K.), who was unaware of the stroke subtypes and neurological examination results. The questionnaire included following items.
Chief complaints: What are your 2 most troublesome current complaints?
Assessment of patient by Barthel Index score.
Sensory symptoms: Do you have any sensory symptoms? If yes, how would you describe them? Select any of the followings that best (or if there are >1, second best) characterize your sensory symptoms. The examples were those modified from the McGill-Melzack Pain Questionnaire,11 as follows: burning, including sensations of “heat”, “warmth” “flushing”; cold, including sensations of “cool” and “freezing”; numbness, including sensations of “tingling”, “tightness” and “being covered by something”; heaviness, including a sensation of “dullness” and “aching”; pricking, including a sensation of “shooting”; lancinating, squeezing, or throbbing; and others.
The severity of the major sensory symptom was assessed with the use of a visual analog scale (10-point scale: 1, slight; 10, most severe).
What are the aggravating factors for the sensory symptoms: cold, heat, body movements, bright light, feeding, hunger, psychological stress, rest, others?
Were the sensory symptoms continuous or fluctuating/ intermittent?
When did the sensory symptoms develop? Because the patients often were not able to pinpoint the exact date, this was categorized as immediately after the stroke, <1 month, ≥1 and <3 months, ≥3 and <6 months, and ≥6 months.
For statistical analyses, we used the Student t test (for numerical variables) and χ2 test (for nominal scale variables). The Fisher exact test was used when the number of the subjects was small, and ANOVA was used when there were >2 variables. All statistical tests were 2-tailed and performed with the use of a PC-SAS package (version 6.10; SAS Institute Inc); a value of P<0.05 was regarded as indicating significance.
Demography and Initial Neurological Features
There were 43 men and 12 women, aged 35 to 86 (mean 59) years. There were no differences in age, gender, or side of lesion between LMI and MMI patients.
In the LMI patients who were examined at the acute stage (n=33), the initial neurological symptoms/signs were vertigo/dizziness (88%), gait ataxia (88%), Horner sign (88%), nystagmus (70%), nausea/vomiting (64%), dysphagia (61%), and hoarseness (39%). Sensory abnormalities were noted in the face in 61% and in the body/limbs in 79% of the patients. Patterns of initial sensory dysfunction analyzed according to previous criteria6 were crossed pattern (ipsilateral trigeminal–contralateral hemibody/limbs) in 13 patients, contralateral trigeminal pattern in 7, and bilateral trigeminal pattern in 4. Each of the patterns could be attributed to an involvement of the descending trigeminal tract, the ascending secondary trigeminal tract, or both, respectively.6 Isolated hemibody/limbs sensory involvement sparing the face was noted in 8 patients, and 1 patient exhibited no sensory abnormalities.
Among the 12 patients with MMI who were examined at the acute stage of the stroke, neurological symptoms/signs included hemiparesis in 11, hemisensory disturbances in 10 (decreased vibration and/or position sense in 8 and paresthesia only in 2), and lingual paresis in 2 patients. Four patients (patients 43, 44, 47, and 49) had transient and mild sensory symptoms in the face as well. In 2 patients (patients 42 and 44), pinprick and temperature sensations were also mildly impaired in their limbs.
Sequelae of the Patients: General View
The patients were studied at 6 to 40 (mean 21) months after the onset of stroke. There were no differences in the follow-up periods between LMI and MMI patients. The functional ability as measured by Barthel Index was significantly worse in patients with MMI than in those with LMI (Table 2⇑). Only 1 patient with LMI (patient 4) and 1 with MMI (patient 55) reported that they had no residual neurological problems. As shown in Table 2⇑, the most important complaint of the LMI patients was a sensory symptom that was followed by dizziness/sense of imbalance and dysphagia. The order of frequency of these 3 important complaints was the same as the items described by the patients as the second most troublesome complaints. Other less-frequent complaints included excessive sweating, hoarseness, and dyspnea on exertion (without cardiac or pulmonary disease). In MMI patients, motor dysfunction was reported as the most troublesome complaint, followed by sensory symptoms. Although the order of frequency was reversed in the items described as the second most important complaint, muscle weakness generally outweighed sensory symptoms as their residual symptoms. The degree of motor dysfunction in patients on the date of interview is presented in Table 1⇑.
Subjective Sensory Symptoms
In LMI patients, 23 patients (56%) had subjective sensory symptoms on the face when they were interviewed: on the side ipsilateral to the lesion in 15, contralateral to the lesion in 7, and bilaterally in 1. The symptoms were described most often as “numb”, “burning” and “cold” (Table 2⇑). “Lancinating,” “pricking,” “squeezing,” and “electrical” sensations were described less frequently. Some patients described the sensation of a foreign body in the eye (patients 5 and 6) and a “salt-and-pepper”–like feeling on the face (patients 11 and 18). Although the symptom was usually most marked in the periorbital area, it was most severe in the perinasal region in patients 6 and 15. The degree of severity varied from 1 to 8 on a visual analog scale. On the other hand, only one MMI patient (patient 49) had a sensation of “numbness” on the face.
Thirty-four LMI patients (83%) had residual sensory symptoms on the body/limbs. The symptoms were restricted to the regions below certain levels of the trunk/leg in 9 patients (patients 7, 10, 13, 19, 21, 32, 33, 38, and 40) or to the arm/upper trunk in 2 patients (patients 9 and 41). The symptoms were most often described as “cold,” “numb,” and “burning,” in that order of frequency, while many additionally described their symptoms as “numb,” “cold,” and “burning” (Table 2⇑). “Squeezing,” “electrical,” “heavy,” and “pricking” were less frequently described. Ten MMI patients (71%) had residual sensory symptoms that were usually described as “numb,” “squeezing,” and less frequently, “cold.” None described their symptoms as “burning.” A few additionally described their symptoms as “numb,” “heavy,” or “cold.” In both subgroups, the severity of the symptoms varied from 1 to 8 on the visual analog scale. The severity of the symptoms in the body/limb was not different between the LMI and MMI patients.
The sensory symptoms were usually continuous, but 4 patients reported them as fluctuating or intermittent. Aggravating factors for the sensory symptoms, presented in Table 2⇑, included a cold environment, body movements, and psychological stress. Four patients with LMI reported “heat” and 2 reported “rest” as aggravating factors, which were never described by MMI patients. Other less-common aggravating factors included feeding, hunger, and exposure to a bright light. The frequency of the 3 most important aggravating factors (a cold environment, body movements, and psychological stress) was not different between LMI and MMI patients (Fisher exact test). However, when MMI patients with spinothalamic sensory impairment (n=2) were excluded, a cold environment was found to be a significantly more frequent aggravating factor in LMI than in MMI patients (P<0.05).
Regarding the onset of the sensory symptoms, the sensory symptoms caused by MMI, if they are present, developed immediately after the onset of stroke except for patient 41. However, in the patients with LMI the onset of sensory symptoms varied between immediately after the onset to 6 months after the stroke (Table 1⇑). The sensory symptoms remained stationary in the majority of the patients. However, 4 patients (patients 18, 21, 24, and 43) reported that the symptomatic severity had been decreasing until the day of interview. None reported that it had been increasing.
Objective Sensory Deficit
On the date of the examination and interview, an abnormal perception of sensation was noted in the body/limb in 88% (n=36; 3 had increased perception) and in the face in 44% (n=18; 2 had increased perception) of the LMI patients. The degree of the sensory deficit was presented in Table 1⇑. In patients with MMI, a sensory deficit in the body/limb was seen in 5 patients (36%).
Relation of the Sensory Symptoms to Other Factors
In patients with LMI and those with LMI and MMI combined, we found that the severity of subjective sensory symptoms in the face or body/limbs, as measured by a visual analog scale, were not different between men and women, young (<59 years) and old (≥59 years) patients, or those with right- and left-sided lesions. To see whether the degree of the subjective symptoms were related to the objective sensory deficit, we divided the severity of the current sensory deficit into 2 groups (group 1, none or mild, and group 2, moderate or severe) and compared the severity of the subjective sensory symptoms between the 2 groups. We found that there were no significant differences. However, when we compared the severity of the subjective sensory symptoms in the body/limbs between the patients with initially mild or no deficits (group 3) and those with initially moderate or severe deficits (group 4) among the patients who had been examined from the acute stage (n=33), it was found that the severity of the subjective sensory complaint was significantly (P<0.05) greater in group 4 than group 3 patients. The severity of subjective symptoms was not related to the nature of complaint when the 3 most important descriptions (numbness, cold, and burning) were compared.
In this study, we systematically examined the patients with medullary infarction and compared the sequelae of LMI with those of MMI. We found that almost all patients had neurological sequelae, in which sensory symptoms/signs were the most (LMI) and the second-most (MMI) important. However, it must be realized at this point that although we included consecutive patients at our clinic, general prognosis of the medullary stroke cannot be assessed in the setting of the present design. The patients who had died or remained bedridden were excluded. A previous study9 reported that approximately 12% of the patients with LMI died in the acute stage. Our own experience showed that 2 of 90 patients (2%) died during hospitalization. On the other hand, the patients with very mild symptoms may not have visited the outpatient clinic. These considerations, along with the fact that our hospital is a tertiary, teaching hospital, indicate that our study population does not precisely represent general medullary stroke patients. However, the main aim of the present study was not to investigate the general prognosis of medullary stroke but instead to assess the sensory sequelae that may differ between LMI and MMI.
We found that the 3 main neurological sequelae of LMI were sensory symptoms, dizziness/sense of imbalance, and dysphagia, in that order of frequency. Our results are at odds with a recent study by Nelles et al,10 in which residual sequelae of LMI patients were disequilibrium/dizziness in 69%, dysphagia in 23%, and numbness only in 15%. However, the number of patients in their study (n=13) was too small to draw a reliable conclusion. Although it is possible that the patients referred to our hospital at the subacute stage (n=10) may have had more severe sensory symptoms, exclusion of them from our subject population also revealed an identical order of frequency in their main complaints. Our results are thus in agreement with a previous report8 in which as many as 33 of 35 LMI patients had residual neurological sequelae, which included sensory symptoms in 16, unsteady gait in 10, and dysphagia in 8.
The long-term sequelae of the MMI have not been studied previously, due probably to the rarity of MMI.5 12 In the era of MRI, unilateral MMI with benign prognosis has been found to be much more frequent than the bilateral type with grave prognosis.3 Indeed, all our patients with MMI had a unilateral lesion. Nevertheless, the functional disability measured by the Barthel Index was distinctly more severe in patients with MMI than in those with LMI due to relatively severe residual hemiparesis. However, we found that sensory symptoms were also an important sequela in these patients.
It has been well documented that painful sensory symptoms can develop after stroke in patients with a nonthalamic stroke.13 14 15 16 We think that medullary strokes occupy a significant portion of nonthalamic central poststroke pain (CPSP) syndrome. This has not yet been properly recognized because previous authors have considered “brain stem stroke” as a single entity16 17 when it actually is quite heterogeneous. Only recently, MacGowan et al18 reported that 25% of LMI patients developed CPSP. In our series, as many as 83% of LMI patients had residual sensory symptoms in the body/limbs and 53% in the face. The differences are due to differing criteria for the sensory symptoms. While MacGowan et al defined CPSP as “chronic, spontaneously irritating sensation,” we included all patients who were complaining of sensory symptoms. If we had defined “pain” as a sensory symptom more severe than grade 5 or 6, a result similar to theirs would have been produced. Actually, however, a majority of our patients did not describe their symptoms as “pain,” even if their symptoms were severe. A clear-cut definition of poststroke “pain” is therefore often difficult to make, and we would propose that “poststroke paresthesia” might be a more appropriate description for poststroke sensory sequelae. As discussed earlier, we found that a considerable portion of MMI patients also had sensory sequelae that were often very severe. This observation is intriguing in that the importance of MMI as a cause of CPSP has not been previously recognized and also that CPSP has previously been thought to be related mainly, though not exclusively, to the spinothalamic rather than lemniscal sensory disturbances.13 19 20 21
We have observed several distinct differences in the sensory sequelae between LMI and MMI (Table 2⇑). First, as expected, the face was almost always spared in MMI patients, even though 4 of them had transient facial sensory symptoms/signs in the acute stage. On the other hand, more than half of the LMI patients had sensory symptoms on the face, on the side either ipsilateral or contralateral to the lesion, which illustrates that sensory sequelae or CPSP can occur after damage on the secondary ascending as well as the descending trigeminal tracts. Second, the sensory symptoms were differently described. Although patients in both groups frequently reported their symptoms as “numb” or “tingling,” those with LMI described them as “burning” or “cold” much more frequently than the MMI patients. Third, the LMI patients more often reported a cold environment or heat whereas MMI patients tended to designate body movements as aggravating factors for the symptoms. Finally, the subjective sensory symptoms tended to start immediately after the onset of stroke in MMI patients, whereas there often was a delay of up to 6 months in patients with LMI.
The pathogenic mechanism of CPSP still remains obscure, but development of hyperexcitation secondary to partially damaged sensory tracts or damage to central inhibitor pathways has been suggested as potential mechanisms.14 In our study, all the lesions except for 1 were identified by MRI, and it would be reasonable to assume that the spinothalamic tract was primarily injured by LMI and the lemniscal fibers by MMI. The frequent description of “burning” or “cold” in LMI but not MMI patients implies that an involvement of the spinothalamic but not the lemniscal tract is related to this type of CPSP. We may therefore speculate that the CPSP after the spinothalamic tract injury is related to uninhibited firing of certain central neurons through excessive feedback mechanism following partial damage.17 The observation that “cold” and “heat” were provoking factors more often in LMI than in MMI patients also appears to be consistent with this hypothesis. In our results, the severity of the subjective sensory complaint was related to the degree of the initial sensory deficit, which suggests that the severity of the initial spinothalamic tract injury may be a marker for the strength of the feedback inputs.
How and where the hypersensitive phenomenon develops remain unknown, but activation of certain parts of the thalamus has been demonstrated by electrophysiological22 23 or blood flow24 25 studies. One of the possible mediators for the hypersensitivity phenomenon could be the spinoreticulothalamic system.26 27 Taster28 previously provided evidence that the spinothalamic and the adjacent spinoreticulothalamic tracts are interrelated in such a way that deafferentation of the former renders the normally nonexcitable reticulothalamic system responsive to stimulation, thus provoking painful sensations. Therefore, development of CPSP after a spinothalamic tract injury may at least in part depend on the relatively intact reticulothalamic system.29 Indeed, MacGowan et al18 previously reported that LMI lesions medially extending as to involve the reticular formation rarely produced CPSP. Disagreeing with their result, our 4 patients (patients 1, 15, 16, and 39) who initially had bilateral facial sensory abnormalities associated with large medullary lesions encompassing the ventromedial as well as the posterolateral area6 did suffer from persistent sensory symptoms. Nevertheless, considering that the reticulothalamic tract cannot be discretely localized by MRI, our findings are not necessarily at odds with the “reticular formation” hypothesis.
Regarding the sensory sequelae of MMI, there has been evidence that the lemniscal tract has certain components which inhibit the spinothalamic sensory functions,30 31 32 possibly through the reticulothalamic system.27 It may therefore be speculated that lemniscal tract injury in MMI disinhibits the spinothalamic system via the reticulothalamic system, ultimately producing hypersensitivity of the spinothalamic sensations. The delayed onset of CPSP in LMI patients may reflect the time required for the partial recovery of the injured spinothalamic fibers (eg, through sprouting and receptor hypersensitivity), a process not required in MMI patients.
In summary, our data illustrate that sensory symptoms are important sequelae in both LMI and MMI patients but that they have distinctly different manifestations. The difference is probably related to selective involvement of the spinothalamic tract by LMI and the medial lemniscal tract by MMI. Our observation therefore suggests that pathogenic mechanisms for CPSP differ according to the injured sensory tract. Unfortunately, however, our data have limitations, in that there are referral bias and quantitative sensory testings were not performed. Therefore, further prospective studies are required for a better understanding of the poststroke sensory sequelae and highly complex mechanism of central pain.33
- Received June 29, 1999.
- Revision received September 7, 1999.
- Accepted September 7, 1999.
- Copyright © 1999 by American Heart Association
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;25:1405–1410.
Kim JS, Kim HG, Chung CS. Medial medullary syndrome: Report of 18 new patients and a review of the literature. Stroke. 1995;26:1548–1552.
Vuilleumier P, Bogousslavsky J, Barth F. Infarction of the lower brainstem: Clinical, aetiological and MRI-topographical correlations. Brain. 1995;118:1013–1026.
Bassetti C, Bogousslavsky J, Mattle H, Bernasconi A. Medial medullary stroke: report of seven patients and review of the literature. Neurology. 1997;48:882–890.
Kim JS, Lee JH, Lee MC. Patterns of sensory dysfunction in lateral medullary infarction: clinical-MRI correlation. Neurology. 1997;49:1557–1563.
Kim JS, Lee JH, Choi CG. Patterns of lateral medullary infarction: vascular lesion- magnetic resonance imaging correlation of 34 cases. Stroke. 1998;29:645–652.
Norving B, Cronqvist S. Lateral medullary infarction; prognosis in an unselected series. Neurology. 1991;41:244–248.
Nelles G, Contois KA, Valente SL, Higgins JL, Jacobs DH, Kaplan JD, Pessin MS. Recovery following lateral medullary infarction. Neurology. 1998;50:1418–1422.
Toyoda K, Imamura T. Saku Y, Oita J, Ibayashi S, Minematsu K, Yamaguchi T, Fujishima M. Medial medullary infarction: analyses of eleven patients. Neurology. 1996;47:1141–1147.
Schott GD. From thalamic syndrome to central poststroke pain. J Neurol Neurosurg Psychiatry. 1995;61:560–564.
Bowsher D, Leijon G, Thuomas K-Å. Central poststroke pain: correlation of MRI with clinical pain characteristics and sensory abnormalities. Neurology. 1998;51:1352–1358.
MacGowan GJL, Janal MN, Clark WC, Wharton RN, Lazar RM, Sacco RL, Mohr JP Central poststroke pain and Wallenberg’s lateral medullary infarction: frequency, character, and determinants in 63 patients. Neurology. 1997;49:120–125.
Bowsher D. Central pain: Clinical and physiological characteristics. J Neurol Neurosurg Psychiatry. 1996;61:62–69.
Peyron R, García-Larrea L, Grégoire MC, Convers P, Lavenne F, Veyre L, Froment JC, Mauguirère F, Michel D, Laurent B. Allodynia after lateral medullary (Wallenberg) infarct: a PET study. Brain. 1998;121:345–356.
Bowsher D. Termination of the central pain pathway in man: the conscious appreciation of pain. Brain. 1957;80:606–622.
Tasker RR, de Carvalho G, Dostrovsky JO. The history of central pain syndrome, with observations concerning pathophysiology and treatment. In: Casy KL, ed. Pain and Central Nervous System Disease: The Central Pain Syndromes. New York, NY: Raven Press Publishers; 1991:31–58.
Nathan PW, Smith MC. Dysesthésie après cordotomie. Med Hygiene. 1984;42:1788–1790.
Melzack R, Wall PD. Pain mechanisms: a new theory. Science. 1965;150:971–979.
Gonzales GR. Central pain: Diagnosis and treatment strategies. Neurology. 1995;45(suppl 9):S11–S16.