Normal Memory After Damage to Medial Thalamus Mark Kritchevsky, MD; Neill R. Graff-Radford, MBBCh, MRCP; Antonio R. Damasio, MD, PhD \s=b\ We studied two patients with nonhemorrhagic infarcts of the thalamus and assessed their cognitive functions comprehensively using standardized neuropsychological probes. Neither patient had any discernible memory impairment for verbal or nonverbal material. Analysis of magnetic resonance images with a ste- reotaxic method revealed that one subject had a right-sided lesion involving about 15% of the dorsomedial nucleus (DM). The other had bilateral lesions that affected about 15% of the left DM and less than 5% of the right DM. The mamillothalamic tract appeared intact in both patients. Considering that medial thalamic lesions commonly cause amnesia in human beings as well as nonhuman primates, there are two possible reasons, alone or in combination, that may explain why these patients failed to have amnesia: the amount of DM damage was less than required to cause amnesia; or the amnesia related to thalamic lesions requires damage to a second structure, such as the mamillothalamic tract or the anterior nucleus. (Arch Neurol 1987;44:959-962) structures, and, in par¬ rjiencephalic ticular, the dorsomedial nucleus of the thalamus (DM), appear to play an important role in memory. Victor et al1 found damage in the DM of all patients with amnesia in the setting of Wernicke-Korsakoff disease. Is¬ chemie infarction of the medial thala¬ mus has also been associated with amnesia. Left- or right-sided lesions impair memory for verbal or nonver¬ bal material, respectively.26 The wellknown patient NA, an example of traumatic thalamic damage, has amnesia for verbal material in associ¬ ation with a left medial thalamic lesion.7 Bilateral damage of DM has been associated with amnesia, both in human beings6'8·9 and in nonhuman primates.1013 Here we describe two patients who did not develop any memory defect after DM damage. We also describe a stereotaxic method for the analysis of thalamic lesions identified on mag¬ netic resonance imaging (MRI) scans. Accepted for publication April 13, 1987. From the Department of Neurology, Division of Behavioral Neurology, University of Iowa College of Medicine, Iowa City. Dr Kritchevsky is now with the Veterans Administration Medical Center, San Diego. Reprint requests to Neurology Service (127), Veterans Administration Medical Center, 3350 La Jolla Village Dr, San Diego, CA 92161 (Dr Kritchevsky). REPORT OF CASES Case 1.—A 32-year-old right-handed man, while driving home from work on Jan 10,1986, noted the sudden onset of vertical double vision and difficulty with judgment of depth. On leaving the car, he had diffi¬ culty walking. His wife noticed a mild weakness of the face on the left side and slurred speech. During admission to a local hospital the neurologic signs were con¬ firmed, but results of all laboratory inves¬ tigations were normal, including a com¬ puted tomographic (CT) scan of the head with and without contrast, which was per¬ formed one day after onset. Examination at the University of Iowa Hospital (Iowa City) on Jan 20 showed some flattening of the left nasolabial fold, a mild left pronator drift, slowness of rapid alternating movements of the left fingers, and moderate difficulty with tan¬ dem gait. Results of a complete blood cell count, erythrocyte sedimentation rate, VDRL, blood chemistry studies, urinalysis, coagulation profile, electrocardiogram, chest roentgenogram, and echocardiogram were normal. An MRI scan showed a lesion in the medial right thalamus (see anatomic description below). The aortic arch, right carotid, and right vertebral angiograms were normal. No neurologic deficits were found on a follow-up examination on May 27, 1986. Case 2.—A 25-year-old right-handed female nurse awoke at 6 am on Jan 9,1986, and had a transient episode of numbness of the left hand. Shortly after dressing she felt nauseated and lightheaded. She left her apartment to call a neighbor, but lost consciousness and fell in the hallway. About 20 minutes later, she awoke sponta¬ neously and returned to her apartment, noting that her vision was blurred and double, that she was unable to look down, and that her gait was unbalanced. Again, she went for help, and was then taken to the University of Iowa Hospital. On arrival, the patient was sleepy but oriented to person, place, and time. She responded to commands slowly. She incor¬ rectly stated that her husband was in Chicago. She was unable to look up or down, had right-sided facial paralysis, lim¬ ited movement of the right side of the palate and pharynx, and rightward devia¬ tion of the tongue on protrusion. Moderate weakness of her right arm was noted. Two hours after admission, she was alert and had become normal on neurologic exami¬ nation, except for a slightly unsteady gait. The patient had discontinued taking oral contraceptives 4Vfe weeks previously, after having taken them for three years. Results of a complete blood cell count, erythrocyte sedimentation rate, VDRL, antinuclear antibody test, blood chemistry studies, coagulation profile, electrocardio¬ gram, chest roentgenogram, and echocar¬ diogram were normal. Head CT without contrast (Jan 9) was normal. An MRI brain scan (Jan 10) showed a 1-cm lesion in the left medial thalamus and a smaller lesion in the right medial thalamus (see anatomic description below). The aortic arch, left carotid, and right vertebral angiograms were normal. At a follow-up examination on Jan 24, 1986, she had no neurologic signs and was subjectively back to baseline. METHODS Neuropsychologic Studies Patient 1 was studied at 11 days and 4.5 months following stroke. Patient 2 was studied at four and 18 days after stroke. The procedures used are listed in Table 1. Neuroanatomical Studies An MRI scanner (Picker International Superconducting 0.5 tesla) was used to obtain 5-mm thick horizontal and coronal Tl-weighted inversion recovery images. The method used for lesion localization was a modification of that described by Archer et al14 and Graff-Radford et al.15 If the lesion was seen on one horizontal and one coronal scan only, the following technique was used: 1. On the midline sagittal MR cut (Fig 1), the computer was instructed to place lines H and C. These corresponded, respectively, to the horizontal and coronal cuts in which the lesion was seen. Point O was the inter¬ section of these two lines and corresponded to the center of the lesion. The position of the anterior commissure (AC) was marked inferior to the junction of the anterior corpus callosum and fornix. The position of Table 1.—Neuropsychologic Studies Wechsler Adult Intelligence Scale, Revised29 Temporal Orientation Test30 Rey auditory verbal learning test (AVLT)31 Wechsler memory scale (WMS)32 logical memory, paired associates, and visual reproduction subtests Twenty-minute delayed recall of WMS subtests Release from proactive inhibition (adapted from Wickens)33 Rey-Osterrieth or Taylor figure copy and 20-minute delayed recall31*395' Kimura recurring figures test34,35 Benton visual retention test (administration A, form C)31·?447 Recall of past presidents Boston famous faces test36 Autobiographic reconstruction Wisconsin Card Sorting Test37 Serial hand positions38· p423 Verbal associative fluency (controlled oral word association)39 Design fluency40 Multilingual and Boston Diagnostic Aphasia Examination Subtests39·41 Benton-Van Allen Facial Discrimination Test30 Judgment of line orientation30 Three-dimensional block construction30 Hooper Visual Organization Test31.p357 Right-left discrimination30 Dichotic listening Downloaded From: http://archneur.jamanetwork.com/ by a University of Pennsylvania User on 06/12/2015 Fig 2.—Lesion of patient 1 plotted on sagittal atlas plate 5 mm from midline. Lesion involved 29% of area of right dorsomedial nucleus in this plate. It did not extend to mamillothalamic tract or anterior nucleus. MD indicates dor¬ somedial nucleus; PF, parafascicular nucleus; A, anterior nucleus; IML, internal medullary lamina; MTT, mamillothalamic tract; and MB, mamillary body. (Adapted from Schaltenbrand and Wahren atlas.16) angle between lines C and AC-PC was measured. 2. On the coronal and horizontal MR cuts, which contained the lesion, the medi¬ al and lateral limits of the infarct were measured. These distances were multiplied by a conversion factor (CP) to account for the different size of the atlas and patient brains. The CF equals the length in milli¬ meters of the AC-PC line in the atlas, divided by the length in millimeters of the AC-PC line on the scan. The distances from the midline of the sagittal plates of the Schaltenbrand and Wahren atlas16 were then reviewed. Three plates were chosen to correspond as nearly as possible to the medial aspect, midline, and lateral aspect of the lesion. Returning to the MR cuts, the anterior, posterior, superior, and inferior limits of the infarct were measured at the distances corresponding to the three atlas plates chosen. 3. The following procedure was used for each of the three sagittal atlas plates deter¬ mined above (Fig 2). After tracing paper was placed over the plate, the AC-PC line was drawn. Points M and O were then marked using the distances calculated in the first section (multiplied by CF and by the atlas magnification factor of 4). The coronal and horizontal planes, lines C and H, were drawn using the angle between lines C and AC-PC as measured in the first section. The anterior, posterior, inferior, and superior limits of the lesion were marked using the distances calculated in the second section (multiplied by CF and by the atlas magnification factor of 4). A line was drawn through these points and taken to represent the extent of the lesion. 4. For each sagittal plane, a digitizer was used to calculate the area of each thalamic nucleus and the area of the nucle¬ us involved by the lesion. By comparing the ratio of these areas for the three sagittal planes, a good estimate of the volume of were measured. The Fig 1.—Midline sagittal magnetic resonance cut from patient 1. Lines H andC denote locations of horizontal and coronal cuts on which lesion was seen. Point O corresponded to center of lesion. AC and PC denote locations of anterior and posterior commissures. Point M identifies intersection of AC-PC line with perpendicular line through point O. Table 2.—Memory Test Scores of Two Subjects With Thalamic Infarctions* Patient 2 Patient 1 Memory Tests 11 d 4.5 mo 4d 18 d Verbal anterograde memory Rey auditory verbal learning test Trials 1 through 5 7,10,13,13,14 6,11,13,12,14 8,11,14,15,14_NA _Delay_13_14_12_NA Wechsler memory scale logical memory Immediate/delay_NA_12, 8/13. 8_NA_8,11/11,NA Wechsler memory scale paired associates Immediate/delay_NA_12, 14, 14/14_NA_6,12,12,/12 Release from proactive _inhibition_NA_Present_NA_Present Nonverbal anterograde memory Wechsler memory scale visual reproduction Immediate/delay_NA_13/12_NA_NA Benton visual retention test Number correct/errors 8/2 NA 8/3 A Complex figure Copy/delay_34/16.5_31/20_NA_34/21 Kimura recurring figures score_NA_61_NA_NA *NA indicates not administered. All delay trials were 20 minutes. tAll scores were normal; remote memory was also normal (see text). the posterior commissure was marked at the superior point of the midbrain aque¬ duct. The anterior commissure-posterior commissure (AC-PC) line was then drawn. Another line, perpendicular to the AC-PC line, was drawn through point 0. The intersection of these lines was point M. The distances from 0 to M and from PC to M Downloaded From: http://archneur.jamanetwork.com/ by a University of Pennsylvania User on 06/12/2015 Fig 3.—Coronal magnetic resonance and axial cuts from patient 1. The lesion involved right medial thalamus. the nucleus involved by the infarction was obtained. If the lesion was seen on more than one horizontal or coronal MR scan, a similar technique was used so that on each sagittal atlas plate the extent of the lesion was drawn based on six or more points. RESULTS Neuropsychologic Findings Patient 1 had a low average verbal IQ, consistent with his education and occupation. Performance IQ was bor¬ derline at the initial test session, but improved to a low average level at a follow-up session. Memory test results are summarized in Table 2. Results of tests of orientation, verbal and non¬ verbal short-term memory, recall of past presidents, and autobiographic reconstruction were normal. He had a normal performance on the Wisconsin Card Sorting Test, was easily able to repeat serial hand positions, and had normal verbal associative fluency. His performance on a test of design fluen¬ cy was not examined at the first test session, but was abnormal at followup testing. His language was normal. His performance on tests of visuoperceptive and visuoconstructive abilities was impaired only for three-dimen¬ sional block construction. Visuocon¬ structive abilities were normal at fol¬ low-up. Dichotic listening was nor¬ mal. Patient 2 had an average perfor¬ mance on the revised Wechsler Adult Intelligence Scale subtests of verbal intellect (information, 12; digit span, 12; arithmetic, 10) and of nonverbal intellect (block design, 13). This was consistent with her education and occupation. Her performance on tests of orientation, short-term verbal and nonverbal memory, recall of past presidents, and autobiographic recon- Fig 4.—Coronal magnetic resonance and axial cuts from patient 2. Lesion involved bilateral medial thalamus. struction was normal. She recognized 20 faces of the Boston famous faces test, and identified one other with cuing, a normal score for her age and background. Her performance on tests of verbal associative fluency, language, facial discrimination, and dichotic listening was also normal. Anatomical Findings Patient 1 had a lesion limited to the right thalamus (Fig 3). It involved about 15% of the DM nucleus, about 5% of the parafascicular nucleus, and less than 5% of the midline nuclei. The mamillothalamic tract and the anterior nucleus were spared. Patient 2 had a bilateral lesion (Fig 4). On the left, it involved about 15% of DM, and less than 5% of the ante¬ rior, parafascicular, lateral posterior, and ventral posterior nuclei. There was also some infarction of the inter¬ nal medullary lamina. The mamillo¬ thalamic tract was spared. The lesion on the right involved less than 5% of DM and less than 5% of the parafasci¬ cular nucleus. COMMENT Both patients had medial thalamic damage but neither had an impair¬ ment of memory. Patient 1 had no evidence of amnesia for nonverbal material, in contrast to the patient with a similar right medial thalamic lesion reported by Speedie and Heilman.5 Patient 2 had no evidence of amnesia for verbal or nonverbal material, unlike patients with bilate¬ ral medial thalamic lesions of Korsa¬ koff's syndrome1 or bilateral ischemie infarctions.6'8·9 Even if we only took into account her large left-sided lesion, and ignored the right thalamic damage, she was still remarkably dif¬ ferent from patients with left medial thalamic ischemie infarction,24·6 or patient NA,7 all of whom had verbal amnesia. The source of this discrepan¬ cy resides, in all likelihood, in the diverse anatomic placement of lesions in our cases and in those previously reported. The following discussion is aimed at clarifying such differences. Although damage to DM has been traditionally implicated in amnesia, previously published evidence sug¬ gests that damage to DM alone might not be sufficient to cause amnesia. Thus, Mair et al17 have suggested that the amnesia of patients with Korsa¬ koff's syndrome is caused by damage to medial thalamic structures only when there are associated lesions of the mamillary bodies. Von Cramon et al6 localized CT lesions stereotactically in patients with unilateral or bilateral ischemie medial thalamic infarcts. Two patients, one with a left-sided lesion and the other with bilateral lesions, had no amnesia. Their lesions were restricted to DM and the internal medullary lamina. The lesions of amnesic patients involved DM, white matter of the internal medullary lamina, and the mamillothalamic tract. The anterior thalamic nucleus is another structure that might be re¬ lated to memory. It could be damaged by an infarct extending more anteri¬ orly in the thalamus. The very small amount of involvement seen in patient 2 was not sufficient to cause amnesia. However, more extensive damage to this nuclear complex might add to the effect of a DM lesion to produce amne¬ sia. Patient NA might also have damage to structures other than DM. His pen¬ etrating injury could have affected other structures important to memo- Downloaded From: http://archneur.jamanetwork.com/ by a University of Pennsylvania User on 06/12/2015 including basal forebrain nuclei pathways, anterior thalamic nucleus, fornix, mamillary bodies, and ry and mamillothalamic tract. Nonhuman primate studies are also consistent with the explanation that DM lesions must be combined with damage to another structure to produce amnesia. Aggleton and Mishkin"·12 studied monkeys with bilateral anterior DM lesions. Although these lesions produced amnesia, the degree of amnesia was considerably greater when the lesions were combined with damage to the mamillothalamic tract and anterior nucleus. An alternative explanation for the lack of amnesia in our patients is that there could be a critical amount of DM damage below which amnesia does not develop. Zola-Morgan and Squire13 studied monkeys with bilateral poste¬ riorly placed DM lesions involving 32% to 44% of the nucleus. They found that the monkeys had a degree of amnesia comparable to that of Aggleton and Mishkin's12 monkeys, which had lesions of anterior DM combined with mamillothalamic tract. The effect of smaller posterior lesions has not been reported. Our patients had relatively small lesions in terms of percent of DM involved. Thus, it may be that lesions of DM which do not extend anteriorly must involve a critical amount of posterior DM to produce amnesia. It is possible that damage to some critical part of DM is required to produce amnesia. On this point, the literature offers no help. All human and nonhuman primates with amne¬ sia caused by medial thalamic lesions seem to have anterior lesions with associated damage to mamillothal¬ amic tract or large posterior DM lesions. It is unlikely that young age and lack of previous neurologic disease are the reasons for our patients' intact memory. Nonetheless, we considered that possibility and reviewed pub¬ lished cases of amnesia following medial thalamic ischemie infarc¬ tion.2 6'8·91828 We found that an uncom¬ mon number of these patients were young. Including the cases we have seen, 35% (9/26) of the patients with clinically diagnosed, and neuroradiologically confirmed, medial thalamic infarcts were between 20 and 40 years old. A final alternative explanation would be that our patients might have had an exceptional anatomic varia¬ tion due to unusual brain develop¬ ment. Not only would it be unlikely to find two such cases consecutively, but there was nothing in the patients' histories or current observation to suggest unusual brain development. Our interpretation, then, is that lesions of DM probably cause amnesia when they damage a second structure such as the mamillothalamic tract, or when they involve a critical amount of DM. Our patients failed to develop amnesia because their lesions did not meet either criteria. This work was supported in part by National Institutes of Health, Bethesda, Md, grant POI NS 19632 to A.R.D. This work was completed while the first author was on sabbatical leave from the Vete¬ rans Administration Medical Center, San Diego, and the Department of Neurosciences, School of Medicine, University of California, San Diego. We wish to thank Stuart Zola-Morgan, PhD, for helpful comments regarding the study. We thank Barbara Reader for assistance in the prep¬ aration of the manuscript, as well as the Medical Media Production Service at the San Diego Vet¬ erans Administration Medical Center and the University of Iowa Hospital and Clinics, Iowa City, for technical assistance. References 1. Victor M, Adams RD, Collins GH: The Wer- nicke-Korsakoff Syndrome. Philadelphia, FA Davis Co Publishers, 1971. 2. Michel D, Laurent B, Foyatier N, et al: Infarctus thalamique param\l=e'\diangauche: Etude de la memoire et du language. Rev Neurol 1982;138:533-550. 3. Speedie LJ, Heilman KM: Amnestic disturbance following infarction of the left dorsomedial nucleus of the thalamus. Neuropsychologia 1982;20:597-604. 4. Goldenberg G, Wimmer A, Maly J: Amnesic syndrome with a unilateral thalamic lesion: A report. J Neurol 1983;229:79-86. 5. Speedie LJ, Heilman KM: Anterograde memory deficits for visuospatial material after infarction of the right thalamus. Arch Neurol case 1983;40:183-186. 6. Von Cramon DY, Hebel N, Schuri U: A contribution to the anatomical basis of thalamic amnesia. Brain 1985;108:993-1008. 7. Squire LR, Moore RY: Dorsal thalamic lesion in a noted case of human memory dysfunction. Ann Neurol 1979;6:503-506. 8. Guberman A, Stuss D: The syndrome of bilateral paramedial thalamic infarction. Neu- rology 1983;33:540-546. 9. Winocur G, Oxbury S, Roberts R, et al: Amnesia in a patient with bilateral lesions to the thalamus. Neuropsychologia 1984;22:123-143. 10. Isseroff A, Rosvold HE, Galkin TW, et al: Spatial memory impairments following damage to the mediodorsal nucleus of the thalamus in rhesus monkeys. Brain Res 1982;232:97-113. 11. Aggleton JP, Mishkin M: Memory impairments following restricted medial thalamic lesions in monkeys. Exp Brain Res 1983;52:199\x=req-\ 209. 12. Aggleton JP, Mishkin M: Visual recognition impairment following medial thalamic lesions in monkeys. Neuropsychologia 1983; 21:189-197. 13. Zola-Morgan S, Squire LR: Amnesia in monkeys after lesions of the mediodorsal nucleus of the thalamus. Ann Neurol 1985;17:558-564. 14. Archer CR, Ilinsky IA, Goldfader PR, et al: Aphasia in thalamic stroke: CT stereotactic localization. J Comput Assist Tomogr 1981;5:427\x=req-\ 432. 15. Graff-Radford NR, Schelper RL, Ilinsky IA, et al: Computed tomographic and postmortem study of a nonhemorrhagic thalamic infarction. Arch Neurol 1985;42:761-763. 16. Schaltenbrand G, Wahren W: Atlas for Stereotaxy of the Human Brain. Chicago, Year Book Medical Publishers Inc, 1977. 17. Mair WGP, Warrington EK, Weiskrantz L: Memory disorder in Korsakoff's psychosis: A neuropathological and neuropsychological investigation of two cases. Brain 1979;102:749-783. 18. Halmagyi GM, Evans WA, Hallinan JM: Failure of downward gaze: The site and nature of the lesion. Arch Neurol 1978;35:22-26. 19. Mills RP, Swanson PD: Vertical oculomotor apraxia and memory loss. Ann Neurol 1978;4:149-153. 20. Von Cramon D, Eilert P: Ein Beitrag zum amnestischen Syndrom des Menschen. Nervenarzt 1979;50:643-648. 21. Schott B, Maugui\l=e'\reF, Laurent B, et al: L'amn\l=e'\siethalamique. Rev Neurol 1980;136:117\x=req-\ 130. 22. Barbizet J, Degos JD, Louarn F, et al: Amn\l=e'\siepar l'\l=e'\sionisch\l=e'\miquebi-thalamique. Rev Neurol 1981;137:415-424. 23. Petit H, Rousseaux M, Clarisse J, et al: Troubles oculo-c\l=e'\phalomoteurset infarctus thalamo-sous-thalamique bilat\l=e'\ral.Rev Neurol 1981;137:709-722. 24. Pierrot-Deseilligny CH, Chain F, Gray F, et al: Parinaud's syndrome: Electro-oculographic and anatomical analyses of six vascular cases with deductions about vertical gaze organization in the premotor structures. Brain 1982;105:667\x=req-\ 696. 25. K\l=o"\mpfD, Oppermann J, K\l=o"\nigF et al: Vertikale Blickparese und thalamische Demenz. Nervenarzt 1984;55:625-636. 26. Karabelas G, Kalfakis N, Kasvikis I, et al: Unusual features in a case of bilateral paramedian thalamic infarction. J Neurol Neurosurg Psychiatry 1985;48:186. 27. Swanson RA, Schmidley JW: Amnestic syndrome and vertical gaze palsy: Early detection of bilateral thalamic infarction by CT and NMR. Stroke 1985;16:823-827. 28. Graff-Radford NR, Damasio H, Yamada T, et al: Nonhaemorrhagic thalamic infarction: Clinical, neuropsychological, and electrophysiological findings in four anatomical groups defined by computerized tomography. Brain 1985;108:485-516. 29. Wechsler D: Wechsler Adult Intelligence Scale-Revised. New York, Harcourt Brace Jovanovich Publishers, 1981. 30. Benton AL, Hamsher K deS, Varney NR, et al: Contributions to Neuropsychological Assessment: A Clinical Manual. New York, Oxford University Press Inc, 1983. 31. Lezak MD: Neuropsychological Assessment, ed 2. New York, Oxford University Press Inc, 1983. 32. Wechsler D: A standardized memory scale for clinical use. J Psychol 1945;19:87-95. 33. Wickins DD: Characteristics of word encoding, in Melton AW, Martin E (eds): Coding Processes in Human Memory. Washington, DC, Winston, 1972, pp 191-215. 34. Kimura D: Right temporal-lobe damage: Perception of unfamiliar stimuli after damage. Arch Neurol 1963;8:264-271. 35. Rixecker H, Hartje W: Kimura's Recurring Figures Test: A normative study. J Clin Psychol 1980;36:465-467. 36. Albert MS, Butters N, Levin J: Temporal gradients in the retrograde amnesia of patients with alcoholic Korsakoff's disease. Arch Neurol 1979;36:211-216. 37. Heaton RK: A Manual for the Wisconsin Card Sorting Test. Odessa, Fla, Psychological Assessment Resources Inc, 1981. 38. Luria AR: Higher Cortical Functions in Man, ed 2. New York, Basic Books Inc Publishers, 1980. 39. Benton AL, Hamsher K deS: Multilingual Aphasia Examination: Manual of Instructions. Iowa City, AJA Associates Inc, 1983. 40. Jones-Gotman M, Milner B: Design fluency: The intervention of nonsense drawings after focal cortical lesions. Neuropsychologia 1977;15:653-674. 41. Goodglass H, Kaplan E: The Assessment of Aphasia and Related Disorders, ed 2. Philadelphia, Lea & Febiger, 1983. Downloaded From: http://archneur.jamanetwork.com/ by a University of Pennsylvania User on 06/12/2015