| OCR Text |
Show Neuro-ophthalmological . Findings with NeuropathologIcal Correlation in Bilateral Thalamic-Mesencephalic Infarction* FREDERICK E. LEPORE, M.D. VITO GULLI, M.D. DOUGLAS C. MILLER, M.D., Ph.D. Abstract A 91-year-old man presented with the sudden onset of coma, vertical gue paresis, and transient disturbance of horizontal gue, with fixed but unequal pupils. Both cranial computed tomography and postmortem study documented infarction of the paramedian thalamus bilaterally and of the ventral mesencephalon, accounted for by presumilbly embolic occlusion of a single artery. This artery arose as an unpaired perforating artery originating from the proximal segment of one posterior cerebral ilrtery and thus conformed to the villiant anatomic configuration assodated with such infarcts as described by Percheron. Thromboembolic occlusion of the rostral basilar artery produces a variety of visual, oculomotor, and behavioral dysfunctions often termed "top of the basilar syndrome".1A subset of patients with this general syndrome has been described in whom the ischemic injury is limited to the rostral brainstem, specifically excluding the cerebral hemispheric territories of the posterior cerebral arteries. 2•3 These infarcts have been attributed by Segarra2 and Castaigne et al.3 to occlusive disease of the proximal segment of the posterior cerebral artery, i.e., that segment extending from the rostral basilar ar- From the Departments of Neurology and Pathology (D.C.M.), University of Medicine and Dentistry of New Jersey-Rutgers Medical School, Academic Health Science Center, New Brunswick, New Jersey. • Presented at the 17th annual Frank B. Walsh Society meeting on February 22, 1985, in Baltimore, Maryland. Write for reprints to: Frederick E. Lepore, M.D., Department of Neurology, University of Medicine and Dentistry of New Jersey-Rutgers Medical School, Academic Health Science Center (CN-19), New Brunswick, NJ 08903. 224 tery bifurcation to the junction with the posterior communicating artery. The paramedian portio~s of th~ thala~usand the ventromedial midbram receive their blood supply from perforating branche~ that arise from this first portion of the postenor cereb~al artery, which, unlike the rest of the postenor cerebral artery, is derived embryologically from the basilar artery and not the internal carotid artery4; this most proximal segment has been termed the "basilar communicating artery" by Percheron. 5 Castaigne et al. 3 and Percheron5 described the variations in the number and distribution of these perforating arteries and of the infarcts that result from occlusion of the proximal posterior cerebral artery or these branches. One important, if rare, variant in this arterial anatomy consists of a single artery arising from one posterior cerebral artery and supplying both sides of the medial thalamus and the rostral ventromedial midbrain; occlusion of this single artery was reported to result in bilateral infarction, in a relatively stereotypical distribution, of the medial thalamus and the rostral mesencephalon. We have recently studied a patient with an infarct of this type. The patient had neuroophthahnological signs not previously weU-described, which could be correlated with the neuropathological findings at postmortem. Case Report A 91-year-old man was found in an unresponsive state on June 17, 1984. The patient had been in good health except for cervical arthritis and a cardiac condition requiring digoxin. On admission, he moved and moaned spontaneously but would not follow verbal commands. Respirations were spontaneous, and his blood pressure was 130/60. The neck was rigid, and no bruits were auscultated. Funduscopic Journal of Clinical Neuro-ophthalmo" examination was normal. There were bilateral ptosis and exotropia with no spontaneous eye movements. No Bell's phenomenon was apparent. Pupils were 5 mm 00 and 4 mm OS, with no reaction to light (Fig. 1). Cold caloric testing with 50 ml of ice water elicited a trace of abduction 00 on right stimulation and no response on left stimulation. Corneal and gag reflexes were intact. There was a slight right facial droop. Motor examination showed vigorous movement of all limbs to noxious stimuli but no stereotypical posturing. Deep tendon reflexes were absent in the legs and there were bilateral Babinski responses. A lumbar puncture performed the following day revealed 1 Iymphocyte/mm) and 173 erythrocytes/ mm) of cerebrospinal fluid (CSF). CSF protein level was 51 mg/dl and glucose level was 121 mgtdl. The results of the examination of June 20 were unchanged. An electroencephalogram revealed diffuse bilateral 7-8 Hz 10 ....V activity superimposed on bilateral frontal-central 1-2 Hz polymorphic slow waves. Increased latency between waves III and Vof the brainstem auditory evoked response bilaterally was felt to imply dysfunction above the level of the medulla on both sides. Cranial computed tomography (Fig. 2) revealed areas of hypodensity in the paramedian thalamus bilaterally, the right cerebellum, and the midbrain. Seven days later the patient remained unresponsive, breathing spontaneously. Simultaneous bilateral ice-water caloric testing elicited tonic conjugate deviation to the right and icewater caloric stimulation on the left elicited tonic abduction OS and no movement 00. Nonreactive anisocoria (00 > OS) was unchanged, and corneal reflexes were still present. His feet withdrew appropriately from a pin and the bilateral Babinski responses persisted. On June 29, somatosensory evoked responses Lepore et al. showed bilateral (right greater than left) prolonged interpeak latency between waves N13 and N20, thought to indicate dysfunction between the cuneate nucleus and the somesthetic cortex. The patient was found to be pulseless and apneic on July 2 and was pronounced dead. An autopsy was performed, but the permit restricted the examination to evaluation of the brain only. At autopsy, three discrete cerebral infarcts were found. The first and largest involved both sides of the paramedian thalamus and the ventromedial midbrain, extending more caudally and posteriorly on the right than on the left. The second lesion was in the right cerebellum, where it involved cortex, white matter, and a portion of dentate nucleus in the distribution of supply of the right superior cerebellar artery. The smallest lesion was a virtually microscopic infarct in the pontine tegmentum. In the thalamic lesion there was involvement of the dorsomedial, centromedian, intralaminar, and ventroposterolateral thalamic nuclear groups. In the midbrain, the Edinger-Westphal nuclei, both oculomotor nuclear complexes, and both red nuclei were damaged, but there was sparing of the trochlear nuclei and the paramedian pontine reticular formation; there was also partial sparing (but hence partial destruction) of the medial longitudinal fasciculi, the cerebral peduncles, and both substantiae nigrae (Fig. 3). The age of the infarcts was estimated histologically to be between 10 days and 2 weeks, since they were characterized by cavitation with abundant macrophages and had only a few hypertrophic astrocytes in the borders of the lesions. No significant atherosclerosis of the cerebral vessels was seen. Examination of the circle of Willis showed that both posterior cerebral vessels had the adult pattern, i.e., they arose primarily from the basilar artery with only small posterior com- Figure 1. Photograph of the patient's eyes. The eyelids are manually elevated by the examiner because of the bilateral ptosis. The pupils are nonre.active, with an anisocoria (right greater than left). There is also exotropia. December 1985 22S Bilateral Thalamic-Mesencephalic Infarct la) Ibl Figure 2. Computed tomographic scans. a: Infarct in the midbrain is shown as a paramedian mesencephalic hypodensity. b: There are discrete bilateral paramedian thalamic lucencies. municating arteries connecting them to the internal carotid arteries. There was a distinct absence of small perforating branches at the top of the basilar or from the proximal segments of the posterior cerebral arteries. Instead, there was a single, relatively large-caliber artery arising from the proximal left posterior cerebral artery to supply both sides of the thalamus and the rostral midbrain (Fig. 4). Discussion Although our p~tient had infarcts affecting one cerebellar hemisphere and a small portion of the pons, his neurological picture, in the main, can be attributed, as with previously reported similar cases,3 to infarction of the para- 226 median thalami and midbrain, i.e., the territOty supplied by the single perforating vessel arising from the prOXimal segment of the posterior cerebral artery. The proximal posterior cerebral artery gives rise to perforating branches in three different patterns as described by Percheron5 : type I is a symmetrical variant of (usually) two or more perforating branches from each posterior cerebral artery; type II is asymmetrical, with a single perforating artery arising from one posterior cerebral artery without a contralateral counterpart; and type III consists of an arterial arcade emanating from an artery bridging between the two posterior cerebral arteries (Fig. 5). The arrangement in our patient is therefore type II, with a single trunk from the left posterior cerebral artery dividing distally to supply both sides of the thalamus as well as the rostral ventromedial midbrain. Castaigne et al.3 stated that, in all cases of this type II arterial pattern, "the thalamic infarct is always bilateral," as indeed it was in our case. The absence of cerebral atherosclerosis or an occluded vessel in our patient at autopsy suggests an embolic etiology for the infarcts. Having traversed the vertebral arteries, an embolus passing through the distally tapering basilar artery (and incidentally sending a fragment to occlude one superior cerebellar artery) would have lodged in the proximal left posterior cerebral artery or its single branch. I Given the limitations imposed upon the autopsy, no source for such an embolus was found; the history of a cardiac condition requiring digoxin certainly raises the possibility of a cardiac mural thrombus. The salient findings of behavioral disturbance and abnormal ocular motility, whether resulting from paramedian thalamic and midbrain infarcts or midbrain infarcts alone, have been stressed by investigators studying proximal posterior cerebral artery ("basilar communicating artery") ischemia.l.2.3.6.7 The behavioral abnormalities have included drowsiness, somnolent mutism,l coma, and, rarely, peduncular hallucinosis. I Disturbances of ocular motility can present as paresis of vertical gaze, wrongway o.r downward gaze deviation, convergence paresIS or spasm, lightning-like eye move'ments, skew deviation, convergence-retractiol\ nyst~gm~s, internuclear or total ophthalm~ plegt~, thrrd and fourth nerve palsies, eyelid reo tr~cho~ ,or ptosi.s, anisocoria, corectopia, or mldp~Sltion to dilated poorly reactive pupils.s Cast~lgne et a1.3 reported 28 cases with paramedian thalamic and midbrain infarcts, of which 7 resembled our case in that there were bU~teral thalamic infarcts and no associated postenor cerebral artery territory cerebral herni- Journal of Clinical Neuro-ophthalmology Lepore et aI. Figure 3. Histological section of the rostral midbrain. The paramedian mesencephalon is infarcted on both sides, with the lesion encompassing the red nuclei, the third nerve nuclei, and the Edinger-Westphal nuclei. (Luxol fast blue-hematoxylin and eosin stain, x 12). spheric disease. All seven patients had total ophthalmoplegia or third nerve palsy(ies). Five also had pyramidal tract signs, five had abnormal movements, and all seven had alterations of mental status.3 Castaigne et al. also reviewed the literature, and found nine cases with bilateral thalamidrnidbrain infarction among 23 reported cases of paramedian thalamic infarcts (fable 5 in ref. 3). In aU but one of these nine cases there were "associated ischemic lesions," which makes clinicopathological correlation somewhat problematical. Our patient's depressed level of consciousness can be attributed to destruction of the rostral medial reticular formation. Although the vertical gaze disturbance (i.e., no downward eye movements elicited by bilateral simultaneous cold caloric testing) may be due to bilateral third nerve nuclear involvement or to su-pranuclear midbrain damage, the nonreactive anisocoria and the evolution of a horizontal gaze disturbance are not as readi~ explicated. Following Warwick's scheme, Daroff1o has proposed that bilateral internal ophthalmoplegia may result from a nuclear third nerve lesion in which the Edinger-Westphal nucleus is damaged. It is difficult, however, to envision a lesion of the Edinger- Westphal nucleus sufficiently complete that it abolishes the light reflex yet somehow still spares enough parasympathetic innervation to cause relative left miosis. Although "nuclear anisocoria" has been observed in cases of midbrain metastasis with involvement of the Edinger- Westphal nucleus, Il·12 an alternative explanation would be interruption of the sympathetic pathways descending through the prerubral field and the capsule of the red nucleus. 13 Figure 4. Blood supply to the thalamus and rostral midbrain. The basilar artery here gives rise to the proximal segments of the posterior cerebral arteries; paramedian perforating arteries are conspicuously absent, except for a single large trunk (arrow) emerging from the left posterior cerebral artery, which subsequently bifurcates to supply both sides. December 1985 227 Bilateral Thalamic-Mesencephalic Infarct logical consequences of th~ anatomical .variant but raises important questions concerrung the anatomy and physiology subserving certain aspects of ocular motility and pupillary response mechanisms. Acknowledgment The authors gratefuJJy acknowledge the assistance of Dr. Betty Marchant for preparing the drawings for Fig. 5. Figure s. Schematic representations of the three variants of the paramedian thalamic-mesencephalic arterial supply. as described by Percherons (redrawn from Castaigne et al. 3). As is dearly shown in Fig. 4. our patient had the type 11 arterial anatomy. The unusual evolution of our patient's horizontal gaze disturbance from a left "one-and-ahalf syndrome" (ipsilateral conjugate gaze paresis and internuclear ophthalmoplegia) to a right internuclear ophthalmoplegia may be partially accounted for by destruction of the rostral portion of the medial longitudinal fasciculi where they abut the third nerve nuclear complexes. The evanescent left conjugate gaze palsy with no demonstrable lesion of the paramedian pontine reticular formation raises the possibility of midbrain influences on horizontal gaze mechanisms. Zackon and Sharpe14 demonstrated this phenomenon in two patients with unilateral metastatic deposits in the oculomotor nuclear complex and mesencephalic reticular formation. These patients manifested (a) ipsilateral paralysis of adduction, (b) paresis of contralateral saccades, and (c) paresis of ipsilateral smooth pursuit. Interruption of saccadic and smooth pursuit pathways as they descended through the mesencephalic reticular formation was the presumed pathophysiological basis for these horizontal gaze disturbances. Similar mechanisms, or even selective brainstem diaschisis, 14 merit consideration in the analysis of the deficits in our patient. In summary, our patient demonstrates the profound neuro-ophthalmological and behavioral disturbances caused by an embolus into a rare but described anatomically variant artery supplying the rostral midbrain and both paramedian thalami. The case illustrates the patho- 228 References 1. Caplan, L. R.: "Top of the basilar" syndrome. Neurology 30: 72-79, 1980. 2. Segarra, J. M.: Cerebral vascular disease and behavior: 1. The syndrome of the mesencephalic artery. Arch. Neurol. 22: 408-418, 1970. 3. Castaigne, P., Lhermitte, F., Buge, A., et al.: Paramedian thalamic and midbrain infarcts: clinical and neuropathological study. Ann. Neurol. 10: 127-148, 1981. 4. Padget, D. H.: The circle of Willis. Its embryology and anatomy. In Intracranial Artmol Aneurysms, W. E. Dandy, Ed. Comstock, New York, 1944. 5. Percheron, G.: Les arteres du thaJamus humain, II. Arteres et territoires thalamiques paramedians de I'artere basilaire communicante. RnJ. Neural. (Paris) 132: 309-324, 1976. 6. Biller. J., Shapiro, R., Evans, L., et al.: Oculomotor nuclear complex infarction. Oinical and radiological correlation. Arch. Neurol. 41: 985987, 1984. 7. Graff-Radford, N. R., Damasio, H., Yamada, T., et al.: Nonhaemorrhagic thalamic infarction. Clinical, neuropsychological, and electrophysiological findings in four anatomical groups defined by computerized tomography. Brain 108: 485-516, 1985. 8. Miller, N. R.: Walsh and Hoyt's Clinicol Neuro-ophthalmology. ed. 4, vol 2. Williams and Wilkins, Baltimore, 1985. 9. Warwick, R.: The ocular parasympathetic nerve supply and its mesencephalic sources. J. Anat. 88: 71-93, 1955. 10. Daroff, R. B.: Ocular motor manifestations of brainstem and cerebellar dysfunction. In Neurtr Ophthalmology: Symposium of the University of Miami and the Bascom Palmer Eye Institute, J. L. Smith, Ed. Huffman, Hallandale, Florida, 1m. 11. Stevenson, G. c., and Hoyt, W. F.: Metastasis to midbrain from mammary carcinoma. J.A.M.A. 186: 514-516, 1963. 12. Keane, J. R., Zaias, B., and ltabashi, H.: Levatorsparing oculomotor nerve palsy caused by a solitary midbrain metastasis. Arch. Neurol. 41: 210212, 1984. 13. Carmel, P. W.: Sympathetic deficits following thalamotomy. Arch. Neurol. 18: 378-387, 1968. 14. Zackon, D. H., and Sharpe, J. A.: Midbrain paresis of horizontal gaze. Ann. Neural. 16: 495504, 1984. Journal of Clinical Neuro-ophthalmology |
