APRAXIA AFTER A SUPERIOR PARIETAL LESION I Kenneth M. Heilman, Leslie Gonzalez Rothi, Linda Mack, Todd Feinberg and Robert T. Watson (Department of Neurology, College of Medicine, University of Florida, Veterans Administration Medical Center, Gainesville) When right-handed patients with left hemisphere lesions are asked to pantomime or imitate transitive movements (e.g., how one uses scissors), they may make spatial and temporal errors or may use body parts as the object. Often these patients have right hemiparesis. However, because their left hand is not weak and they can make precise movements with their left hand (e.g., picking up a dime from a flat surface using a pincer grasp), these movement abnormalities cannot be attributed to an elemental sensorimotor defect. This disorder of skilled movements is termed "apraxia" (Geschwind, 1965) and is thought to reflect either a destruction of the representations (space-time engrams) that program skilled movements or a disconnection of these engrams from motor areas (Heilman, 1979). Because ideomotor apraxia in right-handers is associated with left hemisphere lesions, the representations (space-time engrams) are thought to be stored in the left hemisphere (Liepmann and Maas, 1907). Since the movement representation is contained in one hemisphere (the left), patients with a left temporoparietal lesion, if not hemiparetic, may have bilateral ideomotor apraxia. The left limbs are thought to be apraxic because the left hemisphere lesion either destroyed or disconnected the engrams needed to program the right motor areas. Apraxia limited to the left side is rare but may be induced by callosal lesions (Watson and Heilman, 1984). Recently we examined a righthanded woman who had a right hemisphere lesion and apraxia limited to her left side. The lesion included portions of the superior parietal lobule (Brodmann's areas 5 and 7). In humans damage to these areas is unusual, and little is known about their functions. Analysis of this woman's motor deficit may help elucidate some of the functions of these areas. 1 Presented before the American Academy of Neurology, Dallas, Texas, April 1985, and was pubĀ· lished in abstract form in Neurology, 35:119, 1985. Cortex (1986) 22, 141-150 142 Kenneth M. Heilman et al. CASE REPORT A 20-year-old . right-handed female college student noted the sudden onset of a severe right-sided retroorbital headache with nausea and vomiting. She was examined in a local emergency room and was treated with meperidi.J;le for what was thought to be a migraine headache. She then returned home. The symptoms persisted, so she returned to the hospital and was admitted. Because after three days her condition generally worsened, she was transferred to the University of Florida Medical Center. The patient was well developed and well nourished, and her vital signs were stable. She was slightly lethargic and had mild meningism. The rest of her general physical examination was unremarkable. Although she was right-handed, she did report that rather than hitting tennis balls backhanded, she would switch the racquet to her left hand. She also reported that although she consistently wrote with her right hand, writing with her left hand was quite legible. She was oriented to person, place, year and month, but not day or date. Her speech, reading and writing were normal. She recalled all of three objects after five minutes of distraction, but calculations were mildly impaired. Cranial nerve findings were remarkable only for left inferior quadrantanopia. She had a mild motor drift of her left arm, otherwise her strength was normal. Stereognosis and graphesthesia were diminished, and there was tactile extinction to double simultaneous stimulation of the left arm. Mild left-sided neglect was also present. A computed tomogram (CT) performed three days after the onset of symptoms disclosed a large right high convexity parietal intraparenchymal hematoma without intraventricular blood. Four-vessel cerebral angiography failed to show an arteriovenous malformation, only an avascular mass consistent with the hemorrhage. She was treated with phenytoin and dexamethasone. An electroencephalogram showed mild diffuse slowing on the left with polymorphic slowing on the right. A CT nine months after admission showed complete resolution of .the hematoma. There was however a lesion of the right superior and inferior parietal lobes (see Figure 1). Findings from a vasculitis work-up were negative, and all routine laboratory tests were unremarkable. The neuropsychological evaluation performed one month after the onset of her symptoms revealed a VIQ of 114, a PIQ of 86, and FSIQ of 100 as measured by the WAIS-R. She had no difficulty discriminating faces, facial affect or emotional prosody. However, she had marked visuoconstructive impairment as measured on block-design performance and drawing of complex figures. Verbal memory was unimpaired, but figural memory, especially involving a large visuospatial element, showed deficits. She was given 10 pantomimes to perform, and although her right hand performance was flawless, she erred on eight pantomimes with her left hand. All buccofacial commands were carried out correctly. She could also recognize the examiner's pantomimes. She was discharged to her home and readmitted to the hospital for follow-up examination two weeks later. After she was discharged, the headaches decreased, but she complained of persistent difficulty using her left hand. Re-examination showed that her quadrantanopia was unchanged. There was no weakness of drift. Position sense in her fingers, wrist and elbow was normal. Graphesthesia and stereognosis findings were also normal. There was no evidence of neglect or extinction or of optic ataxia. She could pick up a dime with her left hand using a pincer grasp (Lawrence and Kuypers, 1968). Apraxia and superior parietal lesion 143 Fig. 1 - CT scan taken nine months after admission demonstrating damage to the right superior and inferior parietal lobes. Tests of Praxis The patient's ability to pantomime transitive movements with her eyes open and closed was examined. The right hand performance was flawless, but her left arm was apraxic. She made more errors with her eyes closed (10/12 possibile) than with them open (3/12 possible). In addition, her use of actual objects with her left hand was as poor as her pantomimes and was worse with eyes closed (8/ 12 possible) than with eyes open (3/12 possibile). To classify her errors, we videotaped her performance. After she watched the examiner perform a transitive movement, she closed her eyes and performed the movement. The videotaping was done while she was making the same movement with her left or right hand. There were 12 trials for each hand. The videotapes were viewed by three observers who attempted to classify her errors as temporal or spatial, or both. Only when two of three observers classified a movement as abnormal was it considered an error. She made only two minor temporal errors. By far, the most common errors were spatial, and these significantly degraded the accuracy of her performance. The examiners divided her spatial errors into three types - postural, amplitude and external configuration. Although our patient made only one major postural error (1/12 errors) she made nine major external configuration errors. By configuration errors we are referring to the movement of the limb in space and the manner in which this movement relates to the object being used. For example, when asked to use a screwdriver, she correctly postured her 144 Kenneth M. Heilman et al. hand as if grasping a screwdriver, but instead of twisting her arm at the elbow, she made an arc with the tip of the pretended screwdriver by erroneously moving her arm at the shoulder. The patient's praxis and ability to make reaching movements were examined again 140 days after the onset of symptoms. She was given 12 transitive pantomimes to perform. When she was allowed to observe her hands and arms pantomiming the use of objects, she made three errors with her left forelimb and was flawless with her right. Using her left hand and arm with eyes closed, she made errors on six of the 12 pantomimes. Although she continued to make errors with her left forelimb, her movements at 140 days after onset more closely approximated the normal pantomime. In this testing session we also asked her to fixate on a midline object and to perform left arm movements in her left visual field. She made errors on four of the 12 pantomimes (similar to her free-field performance), and her errors did not improve when she moved her left hand to the right visual field. With her eyes open the patient was asked to make seven intransitive gestures (e.g., wave bye-bye) with each hand. Although her right hand performance was flawless, she made errors on three gestures with her left hand. With her eyes closed, she was given 12 objects to name with each hand. We made a videotape of her hands to learn whether there were differences in the way she explored the objects. She explored and named all objects correctly with either hand. Test of Limb Projection and Reaching Movements In addition to testing praxis, we tested her reaching. Although she did not have evidence of optic ataxia when previously tested, we wanted to retest her within a visual field (Levine, Kaufman and Mohr, 1978). We drew 12 dots (1 mm) to the right and 12 dots (1 mm) to the left of a midline fixation point. Although the dots were randomly placed, the two sides were symmetrical. The patient was instructed to fixate on a midline point, and after being shown the target dot she was to move her hand to the target. Once she touched the board she was told not to move her finger. The examiner marked the location where she touched the board. There were four blocks of 12 trials, two with each hand given in a counterbalanced order. When performing this task, the patient made no ataxic movements. The distance between where her finger landed and the target was measured for each response. The performances of the right and left hands did not differ, which provided evidence that she reached accurately within a visual field. Using a similar method, we tested for somatesthetic ataxia. The patient was blindfolded, and her index finger was placed on one of the dots. She then had to touch her nose and return to the same spot. As before, we had four blocks of 12 trials counterbalanced across hands. We measured errors as we did on the visual task and found no differences between the left and right hands. Last, we performed a visual-somatesthetic reaching task in which the patient was to look at a dot on top of a poster board and, without seeing her hand under the poster board, was to place her finger directly below the target. The procedure was otherwise similar to the previously discussed reaching tasks. In this task the performance with the left hand was poorer than that with the right. The mean error of her left hand for 24 trials was 17 mm. The mean error for her right hand was 9 mm. When we compared her performance for each symmetrical target Apraxia and superior parietaiiesion 145 position, her left hand performed worse on 18/24 trials. Because the left hand may normally perform worse than the right, we tested five normal control subjects. They had mean errors of 11 mm with their right and left hands. Our patient's left hand performance was worse than that of any of the normal subjects. DISCUSSION We have postulated that right-handers have representation of movements (space-time engrams) for some skilled acts in their dominant inferior parietal lobe (Heilman, 1979) and that these engrams are responsible for providing the motor areas in both hemispheres with the critical information necessary for programming these skilled acts. Therefore, with a left hemisphere lesion there is bilateral ideomotor apraxia (if the right arm is not hemiparetic), and with callosal lesions there is a left-sided ideomotor apraxia. The errors made by this patient seem different from those that accompany ideomotor apraxia after left hemisphere lesions in areas such as the inferior parietal lobe (IPL) or with callosal lesions. Our patient made no body-part-as-object errors. In addition, Haaland (1984) demonstrated that patients with ideomotor apraxia make apraxic errors mainly when performing transitive movements. Our patient made errors when performing both transitive and intransitive movements. Although she did not have some of the classic signs of ideomotor apraxia, she did have a striking disorder of complex movements that were limited to her left arm and not explained by weakness, sensory loss, ataxia or lack of comprehension. We therefore call her disorder apraxia for want of a better term. The underlying nature of the disorder remains speculative. She was right-handed, but because before her illness' she could carry out certain highly skilled activities (e.g., writing) with her left hand is suggestive that she may have had bilateral space-time engrams. Perhaps, unlike the brain of strongly right-handed persons, hers was organized so that each hemisphere was responsible for programming the skilled movements of the opposite hand. Therefore, with a right hemisphere IPL lesion she had apraxia of her left hand, and if she had had a left IPL lesion, she may have had apraxia of only the right hand. That she was more apraxic with eyes closed than open is consistent with this hypothesis. The proprioceptive feedback from the hand goes primarily to the contralateral hemisphere. However, visual feedback goes to both hemispheres. Unlike the acts performed with eyes closed, which may have been controlled entirely by the right hemisphere, in the eyes-open condition, the left hand movements could be monitored by the intact left hemisphere's space-time 146 Kenneth M. Heilman et al. engrams, which could then provide information to the right hemispheric motor systems. Because she often used proximaljoint movements in place of the distal movements also supports this view. Proximal muscle groups have more ipsilateral control than do distal muscle groups (Lawrence and Kuypers, 1968), and our patient may have been programming these proximal left-sided muscles with her left hemisphere but may have been unable to program the distal muscles with her left hemisphere. The major piece of evidence against this hypothesis is the absence of prior reports of unilateral (e.g., left hand) apraxia in mixed-handers with right hemisphere lesions and the nature of the errors. The absence of such reports however may be related to other factors: not all neurologists test their patients for left hand apraxia, many patients are hemiparetic and cannot be tested, and testing patients with their eyes closed may be critical for demonstrating this defect. The second hypothesis to explain why our patient was apraxic may relate to lesion localization. The lesion involved Brodmann's area 7 as well as the IPL; area 5 may also have been involved. We examined another patient who had a cerebral infarction of the right IPL, not including Brodmann's areas 5 or 7. Like our apraxic patient, the control patient was asked to pantomime 12 transitive movements with his left and right hands. His performance with both hands was flawless. He was also asked to imitate pantomimes, with eyes open and closed; once again, his performance was flawless. Although we must be cautious about interpreting the results from one control subject, a comparison between their lesion loci suggests that the critical site for inducing our patient's apraxia is the superior parietal lobe (SPL) (Brodmann's areas 5 and 7). In humans, the functions of areas 5 and 7 are not well known, but the pattern of connectivity of these areas suggests that they may be critical in programming complex movements. In monkeys it has been demonstrated that area 5 projects directly to the forelimb area of the motor cortex (Jones, Coulter and Hendry, 1978). Area 5 also has distinct projections to convexity premotor cortex and supplementary motor area (P. Strick, personal communication, 1985), which in tum project to motor cortex. Although area 5 cells respond to proprioceptive input evoked by passive joint movement (Sakata, Takaoka, Kawarasaki and Shibutani, 1973), Mountcastle, Lynch, Georgopoulos, Sakata and Acuna (1975) showed that when a monkey reached for an object, activity in area 5 neurons increased even more than when the arm was passively manipulated. Mountcastle and co-workers proposed that areas 5 and 7 may be involved in the motor command for movements in space. That the activity of these area 5 cells is related to motor command is supprted by stimulation and recording studies. In monkeys, stimulation of area 5 may induce complex movement of the forelimbs (Humphrey, 1979). Kalaska, Apraxia and superior parietal lesion 147 Caminiti and Georgopoulos (1983) recorded from area 5 neurons during arm movements in monkeys and demonstrated orderly changes in neuronal discharge that varied with the direction of movement. Bioulac and Larriarre (1979) examined deafferented limbs and showed that many cells in area 5 remained active during active movement. The observations of Bioulac and Lamarre (1979) suggest that in addition to the cells activated by proprioceptive feedback, cells in area 5 are being activated by a central source. The premotor and motor cortex have reciprocal connections with area 5, and corollary feedback from these motor areas (Kalaska et al., 1983) may be responsible for the central activation of area 5. Area 5 may be important in limb guidance because of its comparator functions (Kalaska et al., 1983); that is, it compares efferent copy from motor and premotor areas with proprioceptive feedback and corrects mismatches. The observation that our patient's movements deteriorated when she closed her eyes suggests that she depended, at least in part, on visual feedback to perform the correct movements. The dependence on visual feedback in the face of normal prioprioception provides support for the postulate that the proprioceptive comparator system was impaired. Although our discussion has primarily centered on area 5, area 7 was also clearly involved in our apraxic patient but not in our control patient. Area 7 not only has reciprocal connections with area 5 and somatosensory cortex but also receives input from visual and auditory association areas. In monkeys, area 7 also receives input from the region of the superior temporal sulcus, a region of a sensory convergence. Area 7, like area 5, projects to premotor cortex. Electrophysiological studies provide evidence that area 7 is important in motor control. For example, Hyvarinen and Poranen (1974) as well as Mountcastle et al. (1975) found neurons in area 7 that show increased activity during arm projection and hand manipulation. Monkeys and humans with area 7 lesions may misreach (Hartje and Ettlinger, 1973; Levine et al., 1978). In general, nonballistic movement programs used in reaching and arm projection may be broken down to three major types: First, there are those that require "on-line" feedback of the results of action and are therefore called "closed loop." Monkeys with lesions of the posterior parietal lobe, which includes area 7, show visual closed-loop misreaching. This misreaching is confined to the contralateral limb and is independent of the location of the target in space (Hartje and Ettlinger, 1973). Area 7 in humans is part of the SPL. Levine et al. (1978) described a patient who had a right SPL tumor, and like the monkey, the patient misreached with his left hand in all parts of space. With closed-loop visual reaching, our patient, however, was as accurate reaching with her left arm 148 Kenneth M. Heilman et al. into left space as she was with her right arm into right space. Therefore, the mechanism that programs the motor system to reach to a (external) visuospatial coordinate in our patient appears intact. The second type of reaching movement requires that the. subject without vision return to a specific spatial position. In this type of movement, the final position of a movement is set in advance and does not depend on on-line proprioceptive feedback. Our somatesthetic localization task demonstrated that our patient could perform these types of movements. The third type of reaching or arm projection movement cannot be accomplished by visual on-line feedback; and unlike the second type of movement, the final position is not given. This type of movement requires transcoding of information into a somatesthetic spatial coordinate system. For example, in the task in which subjects see a dot and must point to the location of the dot but cannot see their hand, they do not have on-line feedback and do not know the final position. The subjects therefore must transcode the visual spatial coordinates into somatesthetic spatial coordinates that provide the premotor and motor areas with the critical information needed to program the accurate movements. Hyvarinen (1982) has postulated that the function of the posterior parietal cortex is to permute visual representations into a somatic coordinate system for reacting movements. He proposes that area 7 is responsible for the visual part of the permutation and area 5 transforms these visual coordinates in a somatic reference system. Based on Hyvarinen's postulate, lesions in area 7 should destroy this coordinate-reaching system and produce misreaching when these systems (visual and somatesthetic) have to be used in concert. Since our patient had a unilateral lesion of this area, she may have had a unilateral defect in this type of visuospatial-somatesthetic-spatial transcoding. Pantomimes and gestures are unlike the visual-somatesthetic task in that there is no visual target. Pantomimes and gestures depend on internal representations. These internal representations (space-time engrams) have been postulated to be in the dominant IPL (Brodmann's areas 39 and 40) (Heilman, 1979). Because our patient's right hand performance was normal, we know that these engrams were intact but, and especially with the eyes closed, had to be transcoded from a spatial-temporal code into a somatesthetic motor code. We propose that each SPL is not only responsible for transcoding from a visual-spatial code into a somatestheticspatial code but is also critical for transcoding space-time engrams into a somatesthetic-spatial code. This somatesthetic code is then responsible for activating the appropriate premotor and motor neurons so that the arm projects correctly into space. Apraxia and superior parietal lesion 149 ABSTRACT We studied a patient who had unilateral (left) limb apraxia associated with a lesion of her right superior (Brodmann areas 5 and 7) and inferior parietal lobes. In monkeys areas 5 and 7 contain "hand-manipulation neurons" that are thought to be important in proprioceptive guidance. Therefore, we tested our patient with her eyes open and closed. Her left hand performance with pantonime, imitation, and use of actual objects dramatically deteriorated in the absence of visual guidance. In addition, although the patient did not have optic ataxia and had normal proprioception, she had difficulty using her left forelimb in tasks that require transcoding from a visual to a somatesthetic spatial coordinate system. We propose that each superior parietal lobe is not only responsible for transcoding from a visual-spatial code into a somatesthetic-spatial code but is also critical for transcoding spatial-temporal representations of skilled movement into a somatesthetic-spatial code. Acknowledgment. This study was supported by National Institutes of Health grant NS20204 and the Medical Research Service of the Veterans Administration. REFERENCES BIOULAC, B., and LAMARRE, Y. Activity of post central cortical neurons of the monkey during conditioned movements of a deafferented limb. Brain Research, 171: 427-437, 1979. GESCHWIND, N. Disconnexion syndromes in animals and man. Brain, 88: 585-644, 1965. HAALAND, K.Y., and FLAHERTY, D. The different types of limb apraxia errors made by patients with left vs. right hemisphere damage. Brain and Cognition, 3: 370-384, 1984. ' HARTJE, W., and ETILINGER, G. Reaching in light and dark after unilateral posterior , parietal ablations in the monkey. Cortex, 9: 346-354, 1973. HEILMAN, K.M. Apraxia. In K.M. Heilman and E. Valenstein (Eds.), Clinical Neuropsychology. New York: Oxford University Press, 1979, pp. 59-185. HUMPHREY, D.R On the cortical control of visually directed reaching: contributions by nonprecentral motor areas. In R.E. Talbott and D.R Humphrey (Eds.), Posture and Movement. New York: Raven Press, 1979, p. 57. HYVARINEN, J. Posterior parietal lobe of the primate brain. Physiological Review, 62: lO60-1129, 1982. HYVARINEN, J., and PORANEN, A. Function of the parietal associative area 7 as revealed from cellular discharge in alert monkeys. Brain, 97: 673-692, 1974. JONES, E.G., COULTER, J.D., and HENDRY, S.H.C. Intracortical connectivity of architectonic fields in the somatic sensory, motor and parietal cortex of monkeys. Journal of Comparative Neurology, 181: 291-348, 1978. KALASKA, J.F., CAMINITI, R, and GEORGOPOULOS A. Cortical mechanisms related to the direction of two dimensional arm movements: relations of parietal area 5 and comparison with motor cortex. Brain Research, 51: 247-260, 1983. LAWRENCE, D.G., and KUYPERS, H.G.J.M. The functional organization of the motor system in the monkey. Brain, 91: 1-26, 1968. LEVINE, D.N., KAUFMAN, K.J., and MOHR, J.P. Inaccurate reaching associated with a superior parietal lobe tumor. Neurology, 28: 556-561, 1978. LIEPMANN, H., and MAAs, O. Fall von linksseitger Agraphie und Apraxie bei Rechtsseitiger Lahmung. Zeitschrift fur Psychologie und Neurologie, 10: 214-227, 1907. 150 Kenneth M. Heilman et al. MOUNTCASTLE, V.B., LYNCH, J.C., GEORGOPOULOS, A., SAKATA, R., and ACUNA, C. Posterior parietal association cortex of the monkey: command functions for operations within extrapersonal space. Journal of Neurophysiology, 38: 871-908, 1975. SAKATA, H., TAKAOKA, Y., KAWARASAKl, A., and SHIBUTANI,-H. Somatosensory properties of neurons in superior parietal cortex (area 5) of the rhesus monkey. Brain Research, 64: '85-102, 1973. WATSON, R.T., and HEILMAN, K.M. Callosal apraxia. Brain, 106: 391-403, 1983. Kenneth M. Heilman, M:D., Box 1-236, Department of Neurology, College of Medicine, University of Florida, Gainesville, FL 32610, U.S.A.