600 CLINICAL NOTES Adverse Effect of Dopamine Agonist Therapy in a Patient With Motor-Intentional Neglect Anna M. Barrett, MD, Gregory f! Crucian, PhD, Ronald L. Schwartz, MD, Kenneth M. Heilman, MD ABSTRACT. Barrett AM, Crucian GP, Schwartz RL. Adverse effect of dopamine agonist therapy in a patient with motor-intentional neglect. Arch Phys Med Rehabil1999;80:600603. Studies in animals and humans report dopamine agonists can improve neglect. Because dopamine deficit reduces intention to act, it has been suspected the dopamine agonist bromocriptine would improve deficient hemispatial intention. Thus, the effect of bromocriptine on line bisection was examined in a patient with neglect and failure of the action-intention system. The 5%year-old patient had left-sided neglect from a right cerebral infarction involving both cortical and subcortical (striatal) structures. It was determined that neglect on a line bisection task was attributable to a motor-intentional bias by testing under congruous and incongruous video monitoring. Testing sessions were held before starting bromocriptine, on 20mg/d, and after stopping bromocriptine. The patient’s ipsilesional bias increased on bromocriptine, and improved when bromocriptine was stopped. Bromocriptine may worsen neglect if putamenal receptors are damaged. Dopamine agonists may activate the normal hemisphere, increasing an intentional bias. Clinicians using dopaminergic pharmacotherapy should assesspatients for this possible adverse effect. 0 1999 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation U NILATERAL NEGLECT, defined as a failure to report, respond, or orient to a stimulus in the side of space opposite a brain lesion, often follows right-sided brain damage, and can be induced by dysfunction in cortical and subcortical areas.la2Fleet and associates3noted that the brain structures related to this syndrome are heavily dopamine innervated. In clinical recovery of neglect in animals, dopamine pathway activity increases.4 Dopamine agonists, including amphetamine,5 bromocriptine and apomorphine,3s6-9and metbylphenidate,9 may improve neglect after brain injury in animals and humans. We studied a patient with akinesia, abulia, and left-sided neglect after a right hemisphere stroke. We wished to confirm our clinical suspicion that his neglect reflected motorFrom the Department of Neurology, College of Medicine, University of Florida (Drs. Barrett, Crocian, Heilman), and the Depar?ment of Neurology, University of Florida Health Science Center (Dr. Schwartz), Jacksonville; and the Veterans Affairs Medical Center, Gainesville (Drs. Barrett, Heilman), FL. Submitted for publication July 21, 1998. Accepted in revised form December 7, 1998. Suooorted bv Deoartment of Veterans Affairs Merit Review arant 114-30-4813. Presented atthe i9th annual meeting of the American Academy of Neurology, April 17, 1997, Boston, MA, and published in abstract form in Neurology 1997;48:A374. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprint requests to Anna M. Barrett, MD, POB 100236, Department of Neurology, University of Florida College of Medicine, Gainesville, FL 32610. 0 1999 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003-9993/99/8005-5133$3.00 Arch Phys Med Rehabil Vol 80, May 1999 intentional rather than sensory-attentional bias. Since dopamine therapy may improve neglect and deficits of motor activation (akinesia),loJ1 we also wanted to learn if bromocriptine would influence motor-intentional bias more than sensory-attentional bias. We had surprising results. CASE REPORT A 58-year-old, right-handed man with hypertension and coronary artery disease was transferred to the Veteran’s Affairs Medical Center after right middle cerebral artery (MCA) territory infarction was confirmed by computed tomography. Carotid dopplers revealed complete right internal carotid occlusion. Eleven days poststroke, a mental status examination revealed frontal-executive dysfunction: visual distractibility, motor impersistence, defective motor response inhibition on crossed response tasks, and generalized hypokinesia. He had reduced verbal fluency (another feature of frontal lobe dysfunction), with reduced speech output, simplified grammar, and abnormal performance on a word generation task (23 words total for the letters F, A, and S).12The patient showed unilateral neglect. He was aware of his condition, but not of the extent of his disability. He exhibited visual and auditory allesthesia, orienting rightward when spoken to from the left side and searching for movement on the right when a moving visual stimulus was presented on the left side. He extinguished left-sided stimuli with double simultaneous stimulation in the auditory, visual, and tactile modalities. He erred 37mm rightward when bisecting a 200-mm line and cancelled only a small number of right-sided targets when presented with a target array spread over a page.13 He omitted left-sided features when drawing, from memory, a flower, a house, and a person. Production of emotional facial expressions and emotional speech prosody to verbal command was poor, but he accurately named others’ emotional facial expressions and emotional speech prosody. Orientation, three-object memory, common object naming, reading and writing, calculations, finger naming, limb praxis, and right/left orientation were normal. On general neurologic examination, the patient had no left visual field deficit. His left face, arm, and leg were weak, with increased tone. Graphesthesis and stereognosis were impaired in the left hand. Left-sided deep-tendon reflexes were increased and the left plantar response was extensor. Magnetic resonance imaging 1 month poststroke revealed cortical and subcortical infarction (fig 1). Because of frontallobe dysfunction on neuropsychologic testing, we believed that the patient would benefit from bromocriptine for abulia and akinesia. We also suspected that his line bisection error reflected a motor bias, and embarked on evaluating bromocriptine’s effect on his neglect. METHODS Apparatus Both motor-intentional and sensory-attentional biases can induce errors on line bisection. The components of motorintentional and sensory-attentional bias in a line bisection task BROMOCRIPTINE Fig 1. T&weighted be hn lerintense, magnetic consistent resonance with middle imaging cerebral WORSENS MOTOR NEGLECT, 601 Barrett of the brain 1 month after patient’s stroke. Right artery infarction. Subcortical regions, including frontotemporal the striatum, cortical areas are also involved. can be seen to were dissociated by having the patient bisect lines under conditions of congruous and incongruous video monitoring.14 The subject sat at a clear Plexiglas table 21 inches from the screen of a 12-inch, closed-circuit television (TV), centered with respect to his midsaggital plane. Lines that were 23cm long were similarly centered in the workspace on the table (fig 2). A camera directed on the worktable displayed the line at actual size on the TV screen, and a curtain between the patient and the workspace on the Plexiglas table prevented him from directly viewing the action of his hand on the line. Therefore, to carry out the line bisection he had to watch his hand and the line on the TV screen. Care was taken to maintain the patient’s head and eyes in a midline position during this testing. The borders of the paper on which the lines were printed were just beyond the margins of the TV screen, precluding any nonstimulus spatial cues. In the direct condition, the video camera was positioned above the table. Stimuli right and left, as displayed on the screen, were congruent with the line and movements in the workspace. In the indirect condition, the camera was below the Plexiglas table, so that the camera perspective was changed by 180”. In this condition, visual right and left on the TV screen were incongruous with right and left in the workspace. Thus, in the indirect condition, if the patient moved his hand to the right side of the workspace, his hand would appear to move toward the left side of the TV screen and vice versa. Procedure Informed consent was obtained from the patient before beginning the test protocol. Five lines were presented for bisection on a table in front of the subject (free field), 2 weeks poststroke. Video testing was administered before beginning bromocriptine (session 1, 4 weeks poststroke), at peak dose (session 2, 6 weeks poststroke), and after stopping bromocriptine (session 3, 8 weeks poststroke). Bromocriptine was given at the same time twice daily starting at 5mg orally on day 36 Fig 2. Video apparatus used in the current experiment (after Na and coworkersY. The camera above the worksoace is used in the direct condition (right workspace appears on right side of screen) and the camera below is used in the indirect condition (right workspace appears on left side of screen). Arch Phys Med Rehabil Vol 80, May 1999 602 BROMOCRIPTINE WORSENS poststroke. Peak dose (20mg/d) occurred on day 39. Peak dose of 1Omgtwice daily was continued until day 42, when session2 began. On day 43, 1Omg of bromocriptine was given, and on day 44, Smg. Bromocriptine was discontinued completely on day 45. The patient bisected 20 lines 23cm long in the direct and indirect conditions for each session. Measures of motor strength, tactile perception/extinction, and emotional prosody were performed as comparison tasks. Left deltoid, biceps, triceps, fingers, iliopsoas, quadriceps, and foot dorsiflexor strength was tested using the Medical Research Council scale. Tactile perception/extinction was assessedwith 30 trials of touching one or both hands with von Frei hairs. The patient’s response was audiotaped and later scored. For emotional prosody production, the patient read 12 neutral sentences with happy, sad, angry, or neutral emotional intonations. His responses were audiotaped and later scored by two blinded independent raters. RESULTS The patient erred a mean 37mm right of midline (SD 15.9) on the free field line bisections 2 weeks poststroke. In the premeditation video session (session l), he erred on direct condition video testing 31.77mm to the right (SD 11.3), and 32.67mm to the right in the indirect condition (SD 16.6). Performance was unchanged between direct and indirect conditions (p = .81, t test, two-tailed). These results are consistent with pure motor-intentional biasI When tested on 20mg bromocriptine (session 2), the patient erred a mean 55.35mm right of midline (SD 26.5). After bromocriptine was stopped (session 3), he erred a mean 16.15mn-1 right of midline (SD 10.7). Performance over the three video testing sessions differed significantly (repeatedmeasures ANOVA, p < .Ol). The patient’s bias on the line bisection task was similar in the direct and indirect conditions during treatment (session 2 means 55.35mm [SD 26.51 and 68mm [SD 24.51, p = .125, t test, two-tailed) and after treatment (session 3 means 16.15mm [SD 10.71 and 20.57mm [SD 19.41, p = .37, t test, two-tailed). This implies that his bias continued to be similar in character and consistent with pure motor-intentional deficit. Figure 3 illustrates mean line bisection errors in free field 2 weeks poststroke, and for the direct video testing condition over sessions 1 to 3. Line bisection errors decrease with time, presumably as neglect spontaneously recovers. This recovery curve (dotted line) contrasts with performance on bromocriptine (session 2, third data point). Comparison tasks remained stable or improved over the three testing sessions. Extinction of left-sided tactile stimuli to double simultaneous stimulation occurred in all trials in all three testing sessions. Strength improved in the left deltoid (session 1,215; sessions2 and 3,315) and biceps (session 1,015; sessions 2 and 3, l/5). Emotional prosodic output was poor throughout (session 1, l/12 correct; session 2, O/12; session 3, 2/12). Ward staff charted that on bromocriptine the patient was “brighter,” initiated speech, was “more eager” to participate in ward activities, and was more kinetic. This suggests that abulia and akinesia improved. After bromocriptine was discontinued, however, the patient was transferred to another facility and it is unknown whether social initiative and kinesis again declined off medication. DISCUSSION Neglect may spontaneously improve,15,16but many patients with neglect remain disabled.17 Bromocriptine pharmacoArch Phys Med Rehabil Vol 80. May 1999 MOTOR NEGLECT, Barrett 50 40 30 20 10 0 Fig 3. Error on the line bisection task over time. Y-axis represents mean error in millimeters-deviation to right of center of 23cm line-and X-axis represents time after onset of symptoms. For first data point 2 weeks postonset, mean represents five trials in free field. Data points at 4, 6, and 8 weeks represent 20 trials, direct condition performance on video apparatus. Hatched bars represent sessions not on bromocriptine therapy, and black line and bar represent performance while taking 20mg bromocriptine. The patient’s performance improved from 2 to 4 weeks postonset of symptoms and then worsened on bromocriptine. This difference is statistically significant (p < .Ol). At 8 weeks, after bromocriptine was discontinued, performance improved to a point consistent with what may occur during natural recovery (dotted line). therapy is safe,‘* and neglect improved with this treatment in several reports.3,7-9This patient, however, who had “pure” motor-intentional neglect with a rightward bias on video testing, erred more toward the right on line bisection when on bromocriptine than when off bromocriptine. No worsening occurred in tactile extinction, strength, or emotional prosodic output. Thus, it appears bromocriptine can specifically increase an action-intentional spatial bias. The patient’s paradoxical response to medication may be related to the locus of his lesion. Magnetic resonance imaging showed putamenal hyperintensity, suggesting ischemic injury. Right-sided putamenal receptor damage could render bromocriptine ineffective on right subcortical (basal ganglia) motor systems. Under these circumstances, bromocriptine could still act on the left striatum, thereby increasing a motor-intentional bias. Our results could also be related to altered receptor response accompanying recovery.19 We could only test this hypothesis by later readministering bromocriptine, and we decided against this. The risk of repeated worsening of neglect, which might have immediate functional consequences and might even limit the amount of recovery ultimately achievedzO was unacceptable. The patient’s performance deteriorated on line bisection during bromocriptine treatment, but social interaction and kinesis improved, consistent with its reported benefit for abuliaG21This combination of increased intentional neglect and improved social behavior reinforces the importance of following all functions influenced by a therapy, even those not being primarily treated. Although a large controlled study of dopamine agonist therapy in neglect is needed, group studies may still obscure atypical responses in individuals. Based on our results, we would suggest that clinicians exercise caution in pharmacotherapy of neglect, or in administering bromocriptine or other dopamine agonists for any indication when a patient has symptoms of unilateral neglect. Subjects should be assessedin all appropriate domains before treatment, when on medication, and after medication has been withdrawn. If improvement on medication occurs, this will better distinguish medication effect BROMOCRIPTINE WORSENS from spontaneous recovery. Treatment can then be reinstituted if a positive response is obtained. Acknowledgments: We are grateful to the patient for his participation, and to Anastasia M. Raymer, PhD, Manho Kim, MD, David Beversdorf, MD, John Hughes, MD, Jean Cibula, MD, and Jeffrey Anderson, MA, for their help in testing and rating comparison tasks. We thank Hema Nallamshetty for helping us prepare the references. We are grateful for the suggestions of the anonymous reviewers who read a preliminary version of this manuscript. References 1. Heilman KM, Watson RT, Valenstein E. Neglect and related disorders. In: Heilman KM, Valenstein E, editors. Clinical neuropsychology. New York: Oxford University Press; 1993. p. 279336. 2. Mesulam M-M. Attention, confusional states, and neglect. In: Mesulam M-M, editor. Principles of behavioral neurology. Philadelphia (PA): FADavis; 1985, p. 125-68. 3. Fleet WS, Valenstein E, Watson RT, Heilman KM. Dopamine agonist therapy for neglect in humans. Neurology 1987;37: 1765-70. 4. Deuel RK, Collins RC. Recovery from unilateral neglect. Exp Neurol 1983;81:733-48. 5. Feeney DM, Sutton RL. Pharmacotherapy for recovery of function after brain injury. Crit Rev Neurobiol 1987;3:135-97. 6. Corwin JV. Kanter S. Watson RT. Heilman KM. Valenstein E. Hashimoto’ A. Apomdrphine has a therapeutic effect on neglect produced by unilateral dorsomedial prefrontal cortex lesions in rats. Exp Neurol 1986;94:683-98. management of 7. McNeny R, Zasler ND. Neuropharmacologic hemi-inattention after brain injury. Neurol Rehabil 1991;1:72-8. a. Whyte .I. Neurologic disorders of attention and arousal: assessment and treatment. Arch Phys Med Rehabil 1992;73:1094-103. MOTOR NEGLECT, 603 Barrett 9. Hurford P, Stringer AY, Jann B. Neuropharmacologic treatment of hemineglect: a case report comparing bromocriptine and methylphenidate. Arch Phys Med Rehabil 1998;79:346-9. 10. Ross ED, Stewart RM, Akinetic mutism from hypothalamic damage: successful treatment with dopamine agonists. Neurology 1981;31:1435-9. 11. Anderson B. Relief of akinetic mutism from obstructive hydrocephalus using bromocriptine and ephedrine. J Neurosurg 1992;76: 152-5. 12. Benton AL, Hamsher KdeS. Multilingual aphasia examination. Iowa City (IA): AJAAssociates; 1989. 13. Albert ML. A simple test of visual neglect. Neurology 1973;23: 658-64. 14. Na DL, Adair JC, Williamson DJ, Schwartz RL, Haws B, Heilman KM. Dissociation of sensory-attentional from motor-intentional neglect. J Neurol Neurosurg Psychiatry 1998;64:331-8. 15. Campbell DC, Oxbmy JM. Recovery from unilateral visuo-spatial neglect? Cortex 1976/12:303-12. 16. Wade DT. Wood VA. Hewer RL. Recoverv of cognitive function soon after stroke: a ‘study of visual neglect, attemion span and verbal recall. J Neurol Neurosurg Psychiatry 1988;51:10-3. 17. Diller L, Weinberg J. Hemi-inattention in rehabilitation: the evolution of a rational remediation program. In: Weinstein EA, Friedland RR, editors. Advances in neurology, Vol 18. New York: Raven Press; 1997. p. 63-82. 18. Miiller U, von Cramon DY. The therapeutic potential of bromocriptine in neuropsychological rehabilitation of patients with acquired brain damage. Prog Neuropsychopharmacol Biol Psychiatry 1994; l&1103-20. 19. Avemo A, Antelman S, Ungerstedt U. Rotational behavior after unilateral frontal cortex lesions in the rat [abstract]. Acta Physiol Stand Suppl 1973;396:77. 20. Jones TA, Schallert T. Use-dependent growth of pyramidal neurons after neocortical damage. J Neurosci 1994;14:2140-52. 21. Barrett K. Treating organic abulia with bromocriptine and lisuride: four case studies. J Neurol Neurosurg Psychiatry 1991;54:718-21. Arch Phys Med Rehabil Vol 80, May 1999