Hemianesthesia and Aphasia An Anatomical and Behavioral Study Bradley T. Hyman, MD, PhD, Daniel Tranel, PhD \s=b\ A 61-year-old right-handed man with a history of lacunar cerebrovascular disease and hypertension had the sudden onset of right-sided numbness and difficulty speaking. Neurologic evaluation revealed a dense right hemianesthesia that included the face, trunk, arm, and leg. Neuropsychological examination documented a conduction aphasia, which resolved nearly completely several months later. Computed tomographic and magnetic resonance imaging studies showed a lesion in the left hemisphere that involved the posterior insula and disrupted thalamocortical connections but entirely spared the thalamus proper. We suggest that the combination of hemianesthesia and aphasia indicates a white matter lesion subjacent to inferior parietal and posterior temporal cortices. {Arch Neurol. 1989;46:816-819) 1965, Fisher1 reported 26 cases of "pure sensory stroke," in which there was loss of all sensory modali¬ ties on the face, arm, and/or leg. The evidence indicated that these strokes occurred primarily in patients with hypertension and were lacunar infarc¬ tions secondary to disease in the pen¬ etrating arteries (100- to 400-um diameter) that originate from the major vessels at the base of the brain. This interpretation was consistent with the autopsy results of one of Fisher's original cases,2 which con¬ firmed a lacunar infarction in the ventroposterior nucleus of the thala¬ mus. Thalamic infarctions on the left side can also produce aphasia (GraffRadford and Damasio3). Theoretical¬ ly, a strategically placed lesion in the Accepted for publication November 1, 1988. From the Department of Neurology, Division of Behavioral Neurology and Cognitive Neuroscience, University of Iowa College of Medicine, Iowa City. Reprint requests to Department of Neurology, University of Iowa Hospitals and Clinics, Iowa City, IA 52242 (Dr Tranel). left thalamus could produce concur¬ rent manifestation of both aphasia and a hemisensory loss. We have recently treated a patient who had the intriguing symptom com¬ bination of hemianesthesia and apha¬ sia following a stroke in the left hemi¬ sphere. Neuroimaging studies showed that rather than a thalamic lesion this patient had a posterior insular lesion and disruption of thalamocortical connections. Here, we describe the language profile, clinical course, and anatomic findings in this patient. REPORT OF A CASE Patient History The patient is a 61-year-old man with a 20-year history of diabetes and a 5-year history of hypertension. Five years prior to the current investigation, he presented with the sudden onset of a dense rightsided "pure motor hemiplegia" involving the arm and leg, which was attributed to a left subcortical lacunar infarction. Carotid Doppler ultrasonogram and echocardiogram were unremarkable, and he recov¬ ered well, with only a mild residual rightsided weakness. He had no further neuro¬ logic problems until 5 years later when, 2 weeks before Christmas, he was unable to control his feet for several minutes. One week later, the patient was briefly unable to control his right hand. On Christmas morning, he had the sudden onset of rightsided numbness and weakness associated with the feeling that he could not say the words he wanted. The weakness resolved, but the numbness and speech difficulty persisted. On admission to the neurology service at University of Iowa Hospitals, Iowa City, the patient had a dense, right-sided hemi¬ anesthesia to pain, touch, vibration, and proprioception. Taste was preserved. Strength was full, except for a mild pronator drift of his right arm. Reflexes were brisk on the right side. Plantar reflexes were flexor bilaterally. Speech was fluent but had multiple phonemic paraphasias. Comprehension of conversational speech was normal. Carotid pulses were 2/2 with bruits bilaterally. Fundi showed atrioventricular nicking but no embolie material. Blood pressure and pulse rate, heart sounds, and femoral pulse rates were nor- Downloaded From: http://archneur.jamanetwork.com/ by a New York University User on 06/14/2015 mal. Visual fields were full. Laboratory evaluation showed a choles¬ terol value of 6.7 mmol/L and a glucose value of 20.8 mmol/L. The patient's elec¬ trocardiogram was suggestive of previous inferior wall ischemia, without any changes indicative of acute ischemia. Echocardiogram and chest roentgenogram were unremarkable. Somatosensory evoked responses were abnormal due to a poorly defined N60 wave on the left side. Brain-stem auditory evoked response was normal. The patient was entered into an experi¬ mental protocol that used heparinoid anticoagulation.4 His clinical course remained stable while undergoing this anticoagulation regimen for 7 days. Carotid Doppler ultrasonogram showed 100% occlusion of the left internal carotid artery and 16% to 49% occlusion of the right internal carotid artery. Cerebral angiogram revealed near¬ ly complete stenosis of the proximal inter¬ nal carotid artery on the left side. After angiography, anticoagulation therapy was continued using heparin. The patient underwent carotid endarterectomy 21 days after admission. At the time of discharge, he suffered from a dense, unremitting hemianesthesia and was only able to recog¬ nize deep pain. Neuropsychological Examination Acute Epoch.—The patient is fully right handed (+100 on the Geschwind-Oldfield questionnaire), completed high school, and was in the military until his retirement 17 years ago. The first neuropsychological evaluation was conducted during the acute/periacute epoch, between the fifth and 11th days after onset. He was fully alert, attentive, and cooperative. The social and paralinguistic components of his communication skills (eg, turn taking and eye contact) were preserved. He ini¬ tiated speech without difficulty and could produce utterances of eight words or more. He had some word-finding difficulties, and very mild dysarthria, but the melodic con¬ tour of his speech was well preserved. Volume was normal, and syntax was intact. The patient's speech was dominated by phonemic paraphasias, present in virtually every utterance. Successive phonemic approximations to target words (conduites d'approche) were common (eg, "... and the kool... sool... tool... stool, fallis ing..."), especially in confrontation-nam¬ ing responses. Semantic paraphasias were less frequent. Despite the phonemic errors, his speech was informative and compre¬ hensible. There were far fewer paraphasic mistakes in automatic, overlearned phrases. Linguistic functioning was formally assessed at the acute epoch with selected subtests of the Multilingual Aphasia Examination5 and the Boston Diagnostic Aphasia Examination.6 Results are pre¬ sented in the Table. Visual and tactile naming were typified by phonemic errors, ie, "trianger" for "triangle" and "rectan¬ gular" for "rectangle." Recognition of the correct name in a four-alternative multiple choice format was flawless. Sentence repe¬ tition was severely defective, as phonemic errors emerged when sentence length extended beyond three or four words. How¬ ever, he was able to produce automatized sequences accurately (eg, the days of the week, months of the year, and the alpha¬ bet). Aural and reading comprehension of sin¬ gle words and short phrases were intact. However, the patient had difficulty with more complex material (Token test, sen¬ tences, and paragraphs). Comprehension deflcits were most evident in the category of "closed-class" words including preposi¬ tions (on, with) and conjunctions (and, or). Paralexias were common. In writing with his preferred hand (right), the patient had interfering motor deficits, but he com¬ plained that it "felt funny" when he was writing, an obvious reference to the dense sensory loss in his right hand and arm. He no tended to stray slightly in the vertical plane, writing above or below the line of writing, but the praxic components of his writing were otherwise normal. He made few mistakes when writing to copy. There were two paragraphic errors in three long sentences. In contrast, writing to dictation was very impaired, with multiple paragraphias, omissions, and misspellings. Spontaneous production of sentences and prose material was linguistically accurate. The patient's gestural praxis was awkward with the right hand and arm but intact with the left hand and arm. Nonverbal cognitive abilities, including visual memory, visuospatial judgment, and visuoperceptive discrimination, were well preserved. The patient was able to recall correctly the past five presidents of the United States and recognized pictures of them. He was fully oriented to time, place, and personal information. Recall of recent news events was normal. He showed nor¬ mal learning of ongoing events in his envi¬ ronment. Chronic Epoch.—Four and one-half months following his stroke, the patient had a persistent dense sensory loss on the right side of his body (face, arm, and leg). However, in a follow-up neuropsychologi¬ cal examination performed between the fifth and sixth months after the onset of stroke, he demonstrated excellent recovery of speech and linguistic functions. The most notable feature of his speech was the continued presence of frequent phonemic paraphasias, with numerous literal trans¬ positions. Otherwise, his speech was flu¬ ent, well articulated, and melodic; volume and syntax were normal. Paralinguistic communication was intact. Results of formal linguistic assessment are presented in the Table. The patient performed in the normal range in nearly every area. Four major assessment proto¬ cols, including the Multilingual Aphasia Examination, the Boston Diagnostic Apha¬ sia Examination, the Philadelphia Com¬ prehension Battery for Aphasia, and Curtiss-Yamada Comprehensive Language Evaluation, were conducted, and the results indicated that the patient had nor¬ mal semantics, syntax, and morphologic findings, both receptively and expressive¬ ly· The patient did, however, show several notable deficits. In dichotic listening to word pairs, he had major difficulty with words heard from the right ear (score, 52 of 110), while performing at a near-normal level in words heard from the left ear (score, 90 of 110). He showed defective finger localization and recognition on the right hand. Tactile naming was impaired on the right side. There were no deficits in the left hand. These impairments are referable to the severe primary sensory compromise on the right side of his body. Calculation abilities, right-to-left discrim¬ ination, and writing of letters, numbers, and sentences (to copy or dictation and spontaneously) were normal. The patient performed in the average range on tests of verbal and nonverbal intellect (scores on the Wechsler Adult Intelligence Scale Revised-are as follows: Verbal IQ, 107; Performance IQ, 95), within expectations based on educational and Neuropsychological Evaluation of Speech and Language Result or Score Area Tested Speech Language Visual namingt Tactile namingt Right hand Left hand Responsive namingt Sentence repetition! Automatized sequences^ Aural comprehension Words and phrasest Token testt Reading comprehension Words and phrasest Sentences and paragraphs} Writing with preferred hand Copy Dictation Spontaneous Spellingt Oral Written Gestural praxis Right hand Left hand Verbal associative fluencyt * Acute Epoch Fluent; multiple phonemic paraphasias; slight dysarthria; normal volume and syntax Chronic Epoch Fluent; frequent phonemic paraphas^ normal volume and syntax 42/60 (5th) 60/60 (intact) 3/16 (defective) 8/16 (defective) 8/16 (defective) 16/16 (intact) 6/14 (0) 8/8 (intact) 30/30 (intact) 12/14 (intact) 8/8 (intact) 17/18 (41st) 20/44 (0) 18/18 (intact) 38/44 (9th) 18/18 (intact) 5/10 (defective) 18/18 (intact) 10/10 (intact) Mild impairment Severe impairment Normal Normal Mild impairment Normal Not tested Not tested 10/13 (40th) 11/13 (70th) Mild impairment Normal Normal Normal 12(0) 17 (3rd) 24/30 (defective) Parentheses contain percentile score or interpretation. tFrom Benton6 (Multilingual Aphasia Examination). tFrom Goodglass and Kaplan6 (Boston Diagnostic Aphasia Examination). Downloaded From: http://archneur.jamanetwork.com/ by a New York University User on 06/14/2015 occupational background. Immediate and delayed verbal and visual memory, assess¬ ed with the Wechsler Memory Scale, Rec¬ ognition Memory Test of Warrington, Benton Visual Retention Test, and recall of the Rey-Osterrieth Complex Figure, were nor¬ mal. Visuoperceptive discrimination, vi¬ suospatial judgment, and visuoconstructive abilities remained well preserved. Neuroimaging Results Computed tomographic studies were performed in the acute, periacute, and chronic epochs, and magnetic resonance imaging was performed 4 days after the infarction (Figure). These studies showed that the thalamus was entirely spared. Subjacent to the posterior temporal and inferior parietal cortices, there was a lesion in the white matter that extended to the trigone. The posterior insula was also damaged. COMMENT A patient with stroke presented with the unusual combination of a dense hemianesthesia and aphasia, without concomitant motor or visual dysfunction. The anatomic localiza¬ tion of this combination of signs leads to consideration of the more common syndrome of "pure sensory stroke." Fisher1-2·7·8 and Groothius et al9 argued that the syndrome of sensory stroke could be due to lesions anywhere along the sensory pathways, but that the thalamus was the most likely place. The brain stem and internal capsule contain tightly packed senso¬ ry fibers, but in close approximation to the corticospinal tract. Lesions in these areas may produce pure sensory stroke, but more likely lead to a sensorimotor stroke.10·11 The thalamic radi¬ ations are part of the large, continu¬ ous fan of fibers that reciprocally connect the thalamus and cortex. One case of a lacunar infarction in the thalamic radiations leading to a "pure sensory syndrome" has been de¬ scribed, but the patient was not apha- sic.12 Finally, a lesion in the primary sensory cortices could, obviously, produce a sensory deficit,13 but, in this setting, the lesion is usually associ¬ ated with compromise of the territory of a large vessel leading to dysfunc¬ tion in sensory, motor, and visual sys¬ tems, as well as cognitive defects. In our patient, there were several reasons to believe that the thalamus was not directly involved in the infarction. The severity and unremit¬ ting nature of the sensory deficit are atypical for a thalamic lesion. Also, although certain lesion loci in the thalamus, especially involving the ventrolateral and anteroventral nu¬ clei, can lead to aphasia, the language defects in our patient are not charac¬ teristic of either thalamic aphasia3·14 or the "atypical" aphasia associated with basal ganglia infarctions.15·16 Instead, the aphasia most closely resembled a conduction aphasia, with disproportionate impairment of repe¬ tition and multiple phonemic para¬ phasias.6 Finally, somatosensory evoked potentials of our patient had normal N13 waves but a poorly defined N60 wave on the left side. In pure sensory Transverse magnetic resonance sections (T2-weighted image) obtained four days after the infarction. The lower two cuts show the full preservation of the thalamus. There is an area of abnormal signal in the left hemisphere. It involves the white matter subjacent to the inferior parietal and posterior temporal cortices, including the thalamocortical connections between thalamus and parietal cortices. The posterior part of the insula is also involved, as can be seen on the middle cut. Downloaded From: http://archneur.jamanetwork.com/ by a New York University User on 06/14/2015 stroke, somatosensory evoked potentials have been reported to be entirely normal17 or to have prolonged N2 latencies.18 The neuroimaging data in our case clearly indicate that the thalamus was, in fact, entirely spared. The posi¬ tion of the lesion in white matter is such that it interrupts connections between the thalamus and parietal cortices. The posterior insula is also involved, allowing for the possibility of damage to the second sensory area and/or its connections. This cortex is located on the superior bank of the lateral sulcus within the postcentral gyrus and extends onto the insula. The persistent hemianesthesia in our patient may be attributable to the combined effects of the white matter and insular lesions, although either lesion might suffice. Along with the case reported by Rosenberg and Koller,12 this case con¬ firms Fisher's hypothesis that hemi¬ anesthesia can occur as a result of interruption of bidirectional connec¬ tions between the thalamus and pari¬ etal cortices. The conduction aphasia in our patient is in keeping with both the white matter localization, which disrupts a specific set of projections from auditory to motor regions," and with the posterior insular involve¬ ment.20 It is likely that the complete sparing of the inferior parietal lobule and the intactness of Wernicke's area account for the excellent language recovery shown by this patient. Our indicates that hemianesthesia and aphasia can be viewed as a symp¬ tom cluster caused by disruption of both thalamocortical and auditorycase motor projections. This investigation has been supported by pro¬ ject grant Nos. NS 19632 and NS 26609, National Institute of Neurological and Communicative Disorders and Stroke, Bethesda, Md. We thank Hanna Damasio, MD, Iowa City, Iowa for anatomic interpretation of the neuroi¬ maging data; Harold P. Adams, Jr, MD, Iowa City, for referring the patient; and Antonio Damasio, MD, PhD, Iowa City, for helpful edito¬ rial comments. References 1. Fisher CM. Pure sensory stroke involving face, arm, and leg. Neurology. 1965;15:76-80. 2. Fisher CM. Thalamic pure sensory stroke: a pathologic study. Neurology. 1978;28:1141-1144. 3. Graff-Radford N, Damasio AR. Disturbances of speech and language associated with thalamic dysfunction. Semin Neurol. 1984;4:162\x=req-\ 168. 4. Biller J, Massey EW, Marler JR, et al. A dose escalation study of ORG 10172 (low molecular weight heparinoid) in stroke. Neurology. In press. 5. Benton AL. Multilingual Aphasia Examination. Iowa City, Iowa: Department of Neurology, University of Iowa College of Medicine; 1976. 6. Goodglass H, Kaplan E. The Assessment of Aphasia and Related Disorders. Philadelphia, Pa: Lea & Febiger; 1983. 7. Fisher CM. Pure sensory stroke and allied conditions. Stroke. 1982;13:434-447. 8. Fisher CM. Lacunar strokes and infarcts: a review. Neurology. 1982;32:871-876. 9. Groothius DR, Duncan GW, Fisher CM. The human thalamocortical sensory path in the internal capsule: evidence from a small capsular hemorrhage causing a pure sensory stroke. Ann Neurol. 1977;2:328-331. 10. Mohr JP, Kase CS, Mecker RJ, Fisher CM. Sensorimotor stroke due to thalamocapsular ischemia. Arch Neurol. 1977;34:739-741. 11. Tuttle PV, Reinmuth OM. Midbrain hemorrhage producing pure sensory stroke. Arch Neurol. 1984;41:794-795. 12. Rosenberg NL, Koller R. Computerized tomography and pure sensory stroke. Neurology. 1981;31:217-220. 13. Derouesne C, Mas J-L, Bolgert F, Cas- taigne P. Pure sensory stroke caused by small cortical infarct in the middle cerebral artery territory. Stroke. 1984;14:660-662. 14. Gorelick PB, Hier DB, Benerento L, Levitt S, Tan W. Aphasia after left thalamic infarction. Arch Neurol. 1984;41:1296-1298. 15. Damasio AR, Damasio H, Rizzo M, Varney N, Gersh F. Aphasia with nonhemorrhagic lesions in the basal ganglia and internal capsule. Arch Neurol. 1982;39:15-20. 16. Naeser M, Alexander MP, Helm-Esta-$ brooks N, Levine HL, Laughlin SA, Geschwind N. Aphasia with predominantly subcortical lesion sites. Neurology. 1982;39:2-14. 17. Robinson RK, Richey ET, Kase CS, Mohr JP. Somatosensory evoked potentials in pure sensory stroke and related conditions. Stroke. 1986;16:818-823. 18. Nelson FW, Grotta JC, Goldie WD. Somatosensory evoked potentials (SEPs) in pure sensory stroke (PSS). Stroke. 1986;17:778. 19. Damasio H. Anatomical and neuroimaging contributions to the study of aphasia. In: Good-$ glass H, ed. Handbook of Neuropsychology: Language. New York, NY: Elsevier Science Publishing Co Inc. In press. 20. Damasio H, Damasio AR. The anatomical basis of conduction aphasia. Brain. 1980;103:337\x=req-\ 350. An Unusual Case of Carnitine Palmitoyl Transferase Deficiency Albert P. Galdi, MD, Jonathan B. Clark, MD \s=b\ A 36-year-old man presented with episodic exertional dyspnea, fluctuating exercise intolerance, and myoglobinuria. He never experienced cramps or myalgias. Subsequent evaluation revealed carnitine palmitoyl transferase deficiency. The unusual features of this case suggested that carnitine palmitoyl transferase deficiency may have a more diverse clinical picture than previously described. {Arch Neurol. 1989;46:819-820) 7" e describe a patient with carnitine palmitoyl transferase deficiency who presented with episodic dyspnea. Cramps and exertional myalgias were never experienced by the patient. This article emphasizes that patients with Accepted for publication May 16, 1988. From the Department of Neurology, Uniformed Services University of the Health Sciences (Dr Galdi), and the Department of Neurology, National Naval Medical Center (Dr Clark), Bethesda, Md. Presented at the Annual American Association of Electromyography and Electrodiagnosis, San Antonio, Tex, October 1987. The opinions or assertions contained herein are the private ones of the authors and are not to be construed as official or as necessarily reflecting the views of the Department of Defense or the Uniformed Services University of the Health Sciences. Reprint requests to Department of Neurology, Uniformed Services University of the Health Sciences, 4301 Jones Bridge Rd, Bethesda, MD 20814\x=req-\ 4799 (Dr Galdi). carnitine palmitoyl transferase defi¬ ciency may have fluctuations in muscle performance and because of this may be misdiagnosed early in the course of their disease. REPORT OF A CASE A 36-year-old white man was admitted for evaluation of intermittent dyspnea and myoglobinuria. The patient first noted nonexertional dyspnea at the age of 33 years, occurring primarily at the end of the day and lasting several hours. During this time, however, he was able to run 10-km races without dif¬ ficulty, and he had previously run mara¬ thons. At the age of 34 years, he noted ex¬ ercise intolerance, and he went from being able to run 9.6 km/d to being overcome by dyspnea after two flights of stairs. He was hospitalized; results of his exercise stress test were normal, and cardiac and pulmo¬ nary angiograms were also normal. A pul¬ monary function test revealed a mildly de¬ creased forced expiratory volume in 1 sec¬ ond and forced vital capacity, suggesting a possible neuromuscular problem. Findings from a neurologic examination and routine muscle biopsy were normal and negative, respectively, and an electromyogram with repetitive nerve stimulation studies was normal. His symptoms subsided spontane¬ ously, and he resumed his normal running schedule. Six months later, exertional dyspnea re¬ turned, and on two occasions, myoglobin¬ uria and acute renal failure developed fol¬ lowing minimum exertion. He gradually Downloaded From: http://archneur.jamanetwork.com/ by a New York University User on 06/14/2015 recovered and was admitted to the National Naval Medical Center, Bethesda, Md, for further evaluation. The patient was otherwise healthy, and he denied alcohol or drug use. His only medication was levothyroxine sodium (Synthroid). There was no family history of neurologic or neuromuscular diseases. Re¬ sults of general physical and neurologic ex¬ aminations were normal, except for mild proximal weakness of the lower extremi¬ ties. Laboratory findings, including serum creatine kinase, carnitine, cholesterol, and triglycéride levels, were normal, as were findings from thyroid function studies. A urine screening for narcotics, barbiturates, and amphetamines was negative. A stan¬ dard forearm ischemie exercise test1 re¬ vealed a fourfold rise in the venous lactate level and a fivefold rise in the venous am¬ monia level following 90 seconds of isch¬ emie exercise. The patient was fasted to as¬ sess his ability to produce ketone bodies, and his urine tested positive for ketone bodies 20 hours into the fast. Serum creat¬ ine kinase levels did not change signifi¬ cantly during the fast. A repeated biopsy specimen obtained from the left vastus lateralis muscle revealed normal histologie and histochemical features. There was no evidence of increased muscle glycogen or lipid deposits. Electron microscopy of the tissue disclosed morphologically normal mitochondria. Biochemical analysis indi¬ cated a total absence of carnitine palmitoyl transferase (CPT) activity. Muscle concen¬ trations of phosphorylase and carnitine were normal.