Recovery of Anterograde Amnesia in a Case of Craniopharyngioma

OBSERVATION Recovery of Anterograde Amnesia in a Case of Craniopharyngioma

Ron C. Kupers, PhD; Audrey Fortin, PhD; Jens Astrup, MD; Albert Gjedde, MD; Maurice Ptito, PhD, DM

Background: Studies of the amnesic syndrome have in- months after tumor removal. A postsurgical magnetic reso- dicated that telencephalic and diencephalic structures are nance image did not show evidence of damage to the hy- critical components of the memory system. The exact role pothalamus, thalamus, hippocampus, or MBs. of the mammillary bodies (MBs) in human memory re- mains elusive, since few cases of selective MB damage have Main Outcome Measures: Changes in brain imag- been reported. ing data and results of neuropsychological testing.

Objective: To study a case of severe anterograde am- Results: After tumor removal, the patient showed a com- nesia due to a third-ventricle craniopharyngioma with plete recovery of memory functions. Performance on the severe MB compression. associative memory test was at chance level before surgery and dramatically improved postoperatively. Results Design: Case report. of the preoperative positron emission tomographic study showed no activity in memory-related structures. In con- Setting: Neurosurgery clinic of an academic hospital. trast, a significant blood flow increase occurred in the anterior thalamic nuclei postoperatively. Patient: A 53-year old woman who developed severe anterograde amnesia due to a third-ventricle craniopha- Conclusions: These behavioral and brain imaging data ryngioma strongly compressing the MBs and, to a lesser stress the importance of the MBs in this patient’s amne- extent, the right hippocampus. sia. Our data further suggest that the clinical prognosis of decompressing the mammillothalamic tract is excel- Interventions: Surgical excision of the tumor and neu- lent, even in cases of massive compression. ropsychological testing and positron emission tomography during an associative memory test before and 2 Arch Neurol. 2004;61:1948-1952

T IS GENERALLY ACCEPTED THAT compression of the right hippocampus and the critical components of the the MBs. Tumor removal abolished the memory system are not only the functional lesion and led to a complete re- medial temporal lobe (MTL) but covery of memory functions. also diencephalic stuctures such Ias the dorsomedial and anterior thalamic REPORT OF A CASE nuclei, fornix, septal nuclei, and the mammillary bodies (MBs), as well as the fron- A 53-year-old, right-handed woman with tal lobes.1-3 Memory loss associated with no neurologic antecedents developed acute thalamic lesions and related structures is onset of memory problems in July 1999 that described as “diencephalic amnesia”4 and turned rapidly into a full anterograde am- to a certain extent resembles amnesia in Author Affiliations: PET 5 nesic syndrome. The episode that prompted Center (Drs Kupers, Gjedde, cases of MTL lesions or Korsakoff syn- 6 her husband to take her to the hospital, be- and Ptito), Center for drome. There is controversy regarding sides her continuous forgetting of his re- Functionally Integrative which of these structures, the dorsome- quests, was preparation for their holidays. Neuroscience (Drs Kupers and dial thalamus or the MBs (or a combina- When the husband arrived home from Gjedde), and Department of tion of both), are involved in the amnesic work, the patient had completely forgot- Neurosurgery (Dr Astrup), syndrome.7 In humans, damage re- ten that they were leaving the next day and Aarhus University and Aarhus stricted to the MBs7-9 or the mammillo- nothing had been prepared. University Hospitals, Aarhus, 4 Denmark; and Ecole thalamic tract produces severe and per- The patient arrived at the neurologic d’Optométrie, Université de manent memory impairments. ward of the hospital in September 1999. Montréal, Montréal, Québec We studied a patient who developed se- She was alert, oriented for place, person, (Drs Fortin and Ptito). vere anterograde amnesia due to tumor and time, and did not show signs of de-

(REPRINTED) ARCH NEUROL / VOL 61, DEC 2004 WWW.ARCHNEUROL.COM 1948

Figure 1. Preoperative and postoperative magnetic resonance images (MRIs) and 18F-fluorodeoxyglucose positron emission tomography (FDG-PET), showing (from left to right) horizontal, sagittal, and coronal sections. A, Preoperative views of the craniopharyngioma from MRIs (upper row) and FDG-PET (lower row). Note the compression of the right hippocampus and mammillary bodies. B, Postoperative MRIs (upper row) and FDG-PET (lower row) showing complete removal of the tumor and normal perfusion of the brain. The left part of the brain is shown to the left. The PET data acquisition (45 minutes) started 1 hour after an intravenous bolus injection of 365 MBq (9.9 mCi) of FDG.

mentia or speech impairment. A T1-weighted magnetic the third ventricle to the level of the foramina Monroi, resonance image of the brain showed a suprasellar cra- and elevating and compressing the MBs by 2.5 cm, as well niopharyngioma with a diameter of 3.5 cm extending into as the cerebral peduncles. The tumor did not reach the

(REPRINTED) ARCH NEUROL / VOL 61, DEC 2004 WWW.ARCHNEUROL.COM 1949

©2004 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/29/2021 Table 1. ROI Analysis of FDG Scans* Table 2. Preoperative and Postoperative Neuropsychological Test Results* FDG Uptake, rCMRGLc Preoperative Postoperative Region of Interest Preoperative Postoperative WAIS–Revised, mean score Hippocampus Verbal IQ (similarities) 16 (10 ± 5†) ND Right 83.2 ± 12.9 78.5 ± 7.9 Performance IQ (block design) 15 (10 ± 5†) 15 Left 83.5 ± 11.9 81.8 ± 10.6 Memory Asymmetry index 0.004 0.041 Verbal Text Abbreviations: FDG, 18F-fluorodeoxyglucose; rCMRGLc, regional cerebral First immediate recall 7/18 12/18 metabolic rate of glucose. Second recall 14/18 17/18 The asymmetry index was calculated for quantitative FDG–positron * Delayed recall 4/18‡ 17/18 emission tomography (PET) metabolism, according to the following formula: [(Left−right region of interest [ROI] activity)/(left+right ROI activity)]/2. The 10-Word list ROIs for left and right hippocampus were drawn on coronal preoperative and First immediate recall 4/10‡ 7/10 postoperative magnetic resonance images, following the guidelines by After 5 recitations 7/10‡ 10/10 Watson and colleagues.10 The FDG-PET scans were coregistered to the Delayed recall 2/10‡ 10/10 magnetic resonance images by means of Automated Image Registration 3 Words (explicit memory), (AIR 3.0) software (Roger Woods, Department of Neurology, University of mean No. California, Los Angeles, UCLA School of Medicine). First immediate recall 3 3 Recall after 5 min 0‡ 3 Recall after 1 h 0‡ 3 Nonverbal left hippocampus but was impinging on the mesial sur- 3 Objects/3 hiding places face of the right hippocampus by 4 mm. The tumor was Recall after 10 min 0/3‡ 3/3 predominantly cystic but had a small solid part in rela- Recall after 1 h 0/3‡ 3/3 tion to the pituitary stalk in the suprasellar cistern. The 4 Visual figures temporal horns of the lateral ventricles were in their nor- Copy (mistakes) 9/16‡ 14/16 Immediate recall 8/16‡ 14/16 mal place on both sides, and no structural abnormali- Delayed recall 3/16‡ 16/16 ties were seen in temporal and frontal lobes, amygdalae, 12 Drawing items thalamus, or fornix (Figure 1A, upper row). This was Immediate recall 6/12‡ 9/12 confirmed by the 18F-fluorodeoxyglucose positron emis- Delayed recall 0/12‡ ND sion tomography (FDG-PET) results (Figure 1A, lower Object naming row) showing normal FDG uptake in these regions. (implicit learning) Immediate recall 7/30‡ 27/30 On October 14, 1999, the cystic tumor was surgi- Language, mean score cally removed de visu without injuring the surrounding Sentence repetition 45 ND pituitary gland or hypothalamus. A histologic examina- (Benton t score) tion confirmed the diagnosis of cystic craniopharyn- Object naming test 30 30 gioma. Two days after surgery, the patient developed poly- Problem-solving uria because of diabetes insipidus, which was treated with and attention tests Symbol-number test, 30 43 desmopressin acetate. The patient’s memory problems mean score disappeared completely, and she returned to her nor- Trail-Making Test, part A, s 35 (35 ± 11†) 30 mal occupations. Trail-Making Test, part B, s 88 (78 ± 24†) 73 A magnetic resonance image obtained 1 month after the surgery confirmed the total removal of the tumor Abbreviations: ND, not done; WAIS, Wechsler Adult Intelligence Scale. *Data are given as mean score/total possible score unless otherwise (Figure 1B, upper row). An FDG-PET scan taken at the indicated. same time indicated normal brain metabolism in the MTL †Mean±SD for normal age-matched controls. and upper brainstem (Figure 1B, lower row). A region- ‡Significantly low score. of-interest analysis did not show differences in FDG uptake in right hippocampus before and after surgery, as evidenced by unaltered asymmetry indexes of hippo- house, and she had to learn the specific face-house pair- campal activity (Table 1). This was confirmed when pre- ings. After 4 presentations of the series, pairs of stimuli operative FDG scans were subtracted from postopera- were presented in which only 50% were correctly paired, tive FDG scans (data not shown). and she had to indicate whether the pairings were cor- Formal neuropsychological testing was performed be- rect. Twelve sequential measurements (2 conditions re- fore and 2 months after surgery (Table 2). To elucidate peated 6 times each) of regional cerebral blood flow (rCBF) the mechanisms involved in memory loss and recovery, a were acquired. Visual stimuli were displayed on a large PET study was performed preoperatively and postopera- monitor placed in front of the subject at a rate of 1 pair tively. The patient gave informed consent and the proto- every 3 seconds. In the baseline condition, pairs of either col was approved by the Ethics Committee of Aarhus 2 houses or 2 faces were presented, and the subject had County, Denmark. We used a nonspatial associative learn- to indicate whether the images of each pair were the same ing test (adapted from Henke et al11). One hour before the (50% of the pairs) or different. In the test condition, pairs start of the PET study, the subject was presented a pic- of previously learned face-house stimuli (50% of the pairs) ture series of 10 faces, each associated with a particular were intermixed with new face-house combinations, and

(REPRINTED) ARCH NEUROL / VOL 61, DEC 2004 WWW.ARCHNEUROL.COM 1950

t = 6.5

t = 3.0

Figure 2. Preoperative and postoperative positron emission tomographic activations in the paired associate task. A, Before surgery. Significant elevation in regional cerebral blood flow was not observed in any memory-related structures. B, After surgery. Significant activation sites were found in the anterior thalamic nuclei. The color scale to the right shows the range of t-statistical values ranging from t=3.0 to t=6.5.

the patient responded verbally as to whether the pairs were memory tested with abstract design was below the refer- correct. Baseline and test conditions were performed in a ence range in the first assessment. Postoperatively, learn- random order. A PET camera (ECAT HR 47+; Siemens/ ing, recalling, and recognition results were all in the ref- CTI, Knoxville, Tenn) was used to measure changes in erence range for verbal and visuospatial memory. Both rCBF after intravenous injection of 500 MBq (13.5 mCi) preoperatively and postoperatively, the patient per- of water labeled with oxygen 15. Presentation of the vi- formed well on parts A and B of the Trail-Making Test, a sual stimuli started 10 seconds before the injection and test associated with sustained attention. Although there continued throughout the 60-second scanning window. was a moderate improvement in performance, the differ- The PET images were coregistered to the subject’s mag- ence between preoperative and postoperative results dif- netic resonance images and transformed into Talairach fered by less than 1 SD. space. After a voxel-by-voxel subtraction of the baseline During the preoperative PET study, the patient per- condition from the test condition, a statistical parametric formed at chance level on the nonspatial associative map of the rCBF changes was created by means of a t sta- memory test (55% correct responses). Postoperatively, tistic.12 Regions significantly activated were those that ex- performance increased dramatically (98% correct re- ceeded a t threshold of 3.6 (PϽ.05 corrected for multiple sponses) and was at the level of performance in age- comparisons). matched controls. Preoperatively, significant rCBF in- During the neuropsychological examination, the pa- creases occurred in the anterior cingulate and cerebellum tient was alert and behaved in a socially appropriate man- during recall (Figure 2A). No rCBF increases were seen ner. She was aware of her memory problems and was apolo- in hippocampus or thalamus. Postoperatively, signifi- getic about them. She was verbally fluent and used an cant loci of activation were observed in the cerebellum, elaborate vocabulary. The test results indicated a high level the frontal lobe, and the anterior portion of the thala- of intellectual functioning consistent with her education mus (Figure 2B). (bachelor’s degree) and professional activities (teacher). Verbal subtest scores showed excellent abstraction abili- ties. Nonverbal subtest scores indicated superior construc- COMMENT tion aptitudes. Subjective observations gathered during the interviews did not give evidence of concrete thinking or This case showed full recovery of severe anterograde perseveration tendencies. Preoperative and postopera- amnesia after removal of a craniopharyngioma strongly tive language test scores for naming familiar objects were compressing the MBs and, to a lesser extent, the right in the normal range. However, implicit learning of the same hippocampus. Few cases of anterograde amnesia fol- objects was weak preoperatively but improved signifi- lowing selective damage of the MBs have been de- cantly after surgery. The patient also performed poorly in scribed. In some cases the memory loss was permanent, another implicit learning task, in which she had to recall either because the MBs were irreversibly damaged8,9 or 3 real objects in 3 hiding places in the testing room. After because of a combined lesion of the medial thalamus surgery, she easily recalled the objects and the hiding places. and MBs.9,13 In a patient with a tumor invading the MB, Preoperative tests of verbal memory functions showed defi- partial memory improvements were observed after tu- cits in the recall of a short text and in learning and recall- mor removal.7 Our patient represents, to our knowl- ing 10 unrelated words. Learning and immediate recall of edge, the first reported case of full recovery from an- a 3-word list was just adequate. However, 5 minutes later terograde amnesia after surgical decompression of the she could not remember any of the words. Visuospatial MBs and right hippocampus.

(REPRINTED) ARCH NEUROL / VOL 61, DEC 2004 WWW.ARCHNEUROL.COM 1951

©2004 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/29/2021 Lesions of the MTL and medial diencephalic struc- Funding/Support: This study was supported by the Dan- tures may induce anterograde amnesia.14 The spectrum of ish National Research Foundation, Copenhagen. symptoms observed in disconnection syndromes de- Acknowledgment: We are indebted to Lise Ehlers for the pends on the extent of the afferent input to a specific brain neuropsychological evaluation, Pedro Rosa-Neto, MD, region.15 Because the hippocampus has major projec- PhD (PET Center), for critical comments, and Carsten tions both to the limbic system and into the MBs via the Gyldensted, MD, PhD (Department of Radiology), for the fornix, the question arises as to the differential memory anatomic description of the lesion. effects of one or the other structures. Transection or damage of the fornix produces selective anterograde amnesia.16,17 Anterograde amnesia has also been observed in pa- REFERENCES tients with thalamic damage.18,19 This often seems to involve 2 the MBs or the mammillothalamic tract. The thalamic re- 1. Buckner RL, Wheeler ME. The cognitive neuroscience of remembering. Nat Rev cipient zones of the mammillary tract (anterior nuclei)20 Neurosci. 2001;2:624-634. project to the anterior cingulate,21 which in turn sends 2. Van der Werf YD, Witter MP, Uylings HBM, Jolles J. Neuropsychology of infarction in the thalamus: a review. Neuropsychologia. 2000;38:613-627. axons to the MTL and prefrontal cortex. The latter struc- 3. Fletcher PC, Henson RN. Frontal lobes and human memory: insights from func- ture has a role in executive functions that are affected by tional neuroimaging. Brain. 2001;124:849-881. thalamic lesions. These data suggest that the thalamus is 4. Graff-Radford NR, Tranel D, Van Hoesen GW, Brandt JP. Diencephalic amnesia. involved in memory and that the mammillothalamic tract Brain. 1990;113:1-25. thereby plays an important role.2 Several lines of evi- 5. Aggleton JP, Saunders RC. The relationships between temporal lobe and diencephalic structures implicated in anterograde amnesia. Memory. 1997;5:49- dence also point to the importance of the mammillotha- 71. lamic tract in the anterograde amnesia in our patient. Mag- 6. Mair WG, Warrington EK, Weiskrantz L. Memory disorders in Korsakoff psycho- netic resonance images indicated a strong compression of sis: a neuropathological and neuropsychological investigation of two cases. Brain. the MBs and a milder compression of the right hippocam- 1979;102:749-783. 7. Tanaka Y, Miyazawa Y, Akaoka F, Yamada T. Amnesia following damage to the pus with no damage to the surrounding structures as evi- mammillary bodies. Neurology. 1997;48:160-165. denced by FDG-PET. 8. Dusoir H, Kapur N, Byrnes DP, McKinstry S, Hoare D. The role of diencephalic Since both the MBs and the right hippocampus were pathology in human memory disorder: evidence from a penetrating paranasal compressed, it is difficult to disentangle their respective brain injury. Brain. 1990;113:1695-1706. roles in the production of the amnesic syndrome. We have 9. Hildebrandt H, Muller S, Bussmann-Mork B, Goebel S, Eilers N. Are some memory deficits unique to lesions of the mammillary bodies? J Clin Exp Neuropsychol. a few arguments favoring a role for the MBs. First, the 2001;23:490-501. lack of asymmetry between verbal and nonverbal memory 10. Watson C, Andermann F, Gloor P, et al. Anatomic basis of amygdaloid and hip- impairment argues against a functional role of the right pocampal volume measurement by magnetic resonance imaging. Neurology. 1992; hippocampus.22-24 Second, our region-of-interest analy- 42:1743-1750. 11. Henke K, Buck A, Weber B, Wieser HG. Human hippocampus establishes asso- sis showed that FDG uptake in the right hippocampus ciations in memory. Hippocampus. 1997;7:249-256. before and after surgery was basically unchanged. How- 12. Worsley KJ, Evans AC, Marrett S, Neelin P. A three-dimensional statistical analy- ever, these arguments do not completely rule out a po- sis for CBF activation studies in human brain. J Cereb Blood Flow Metab. 1992; tential role of the right hippocampus, since there is also 12:900-918. evidence against the material-specific hypothesis25 and 13. Kapur N, Thompson S, Cook P, Lang D, Brice J. Anterograde but not retrograde memory loss following combined mammillary body and medial thalamic lesions. we obtained only 1 FDG scan preoperatively and post- Neuropsychologia. 1996;34:1-8. operatively. Postsurgically, our patient regained full 14. Gabrieli JD. Cognitive neuroscience of human memory. Annu Rev Psychol. 1998; memory functions. At the same time, increased activa- 49:87-115. tion was observed in the anterior thalamic nuclei and the 15. Geschwind N, Kaplan E. A human cerebral deconnection syndrome: a prelimi- nary report: 1962. Neurology. 1998;50:1201-1212. supplementary motor area and pre–supplementary mo- 16. D’Esposito M, Verfaelli M, Alexander MP, Karz DI. Amnesia following traumatic 11 tor area, consistent with results by others. Together, the bilateral fornix lesions. Neurology. 1995;45:1546-1550. data suggest that the amnesic syndrome was secondary 17. Moudgil SS, Azzouz M, Al-Azzaz A, Haut M, Gutmann L. Amnesia due to fornix to a massive mammillothalamic tract compression. Our infarction. Stroke. 2000;31:1418-1421. case further suggests that the clinical prognosis of de- 18. Squire LR, Moore RY. Dorsal thalamic lesions in a noted case of human memory disfunction. Ann Neurol. 1979;6:503-506. compressing the mammillothalamic tract is excellent, even 19. Guberman A, Stuss D. The syndrome of bilateral paramedian thalamic infarction. in cases of massive compression. Neurology. 1983;33:540-546. 20. Aggleton JP. Anatomy of memory. In: Yanagihara T, Peterson RC, eds. Memory Accepted for Publication: February 2, 2004. Disorders: Research and Clinical Practice. New York, NY: Marcel Dekker; 1991: 23-62. Correspondence: Ron C. Kupers, PhD, Center for Func- 21. Musil SY, Olson CR. Organization of cortical and subcortical projections to an- tionally Integrative Neuroscience, Aarhus University Hos- terior cingulate cortex in the cat. J Comp Neurol. 1988;272:203-218. pital, Nørrebrogade 44, Building 30, DK-8000 Aarhus, 22. Jones-Gotman M, Zatorre RJ, Olivier A, et al. Learning and retention of words Denmark ([email protected]). and designs following excision from medial or lateral temporal-lobe structures. Author Contributions: Study concept and design: Neuropsychologia. 1997;35:963-973. 23. Moscovitch DA, McAndrews MP. Material-specific deficits in “remembering” in Kupers, Astrup, and Ptito. Acquisition of data: Kupers, patients with unilateral temporal lobe epilepsy and excisions. Neuropsychologia. Fortin, Astrup, Gjedde, and Ptito. Analysis and interpre- 2002;40:1335-1342. tation of data: Kupers and Ptito. Drafting of the manu- 24. Golby AJ, Poldrack RA, Brewer JB, et al. Material-specific lateralization in the script: Kupers and Ptito. Critical revision of the manu- medial temporal lobe and prefrontal cortex during memory encoding. Brain. 2001; 124:1841-1854. script for important intellectual content: Kupers, Astrup, 25. Griffith HR, Perlman SB, Woodard AR, et al. Preoperative FDG-PET temporal lobe Gjedde, and Ptito. Obtained funding: Gjedde. Administra- hypometabolism and verbal memory after temporal lobectomy. Neurology. 2000; tive, technical, and material support: Astrup and Gjedde. 54:1161-1165.

(REPRINTED) ARCH NEUROL / VOL 61, DEC 2004 WWW.ARCHNEUROL.COM 1952