BRAIN AND LANGUAGE 33, 31 l-322 (1988) A Long-Term Follow-Up Case Study of Crossed Aphasia Assessed by Single-Photon Emission Tomography (SPECT), Language, and Neuropsychological Testing DELAINA WALKER-BATSON,*” JOHN S. WENDT,? MICHAEL D. DEVOUS, SR.,$ MARY M. BARTON,§ AND FREDERICK J. BONTES *Department of Communication Sciences, Texas Woman’s University; tDepartment Neurology, University of Texas Health Science Center and Dallas Veteran’s Administration Medical Center; SNuclear Medicine Center, University of Texas Health Science Center; and IProgram in Human Development, University of Texas at Dallas of A 6%year-old man with well-defined crossed aphasia secondary to right cerebral infarction IO years previously was studied for current language and cognitive abilities and regional cerebral blood flow (rCBF) during cognitive activation measured by single-photon emission tomography (SPECT). Reversed hemispheric lateralization was demonstrated by qualitative aspects of the patient’s constructional deficits, dominant parietal lobe signs, and absence of the neglect syndrome. Language activation procedures during SPECT produced focal increases in rCBF to both frontal lobes with a phoneme detection task and to right temporal and parietal lobes with a math task. The authors stress the complexities of assessing brain/language mechanisms in vivo and demonstrate variabilities in rCBF during language activation dependent on task selection. Q 1988 Academic PXSS. 1~. Crossed aphasia (aphasia in a strongly right-handed individual secondary to a right hemisphere lesion) is rare, occurring in only l-2% of brainlesioned patients. Follow-up studies in the long term of language function in crossed aphasia are sparse (Haaland & Miranda, 1982). Some authors (Brown & Hecaen, 1976) have suggested that because the majority of crossed aphasic individuals have a seemingly greater resolution of language ’ To whom requests for reprints should be addressed, at Department of Communication Sciences, Texas Woman’s University, Box 23775, Denton, TX 76204. This study was supported in part by Institutional research Grant 29111 from Texas Woman’s University and in part by NIH Grant HL29577 to the Nuclear Medicine Center, University of Texas Health Science Center. The authors thank Dr. Margaret Naeser for comments on an earlier version of this manuscript. 311 0093-934X/88 $3.00 Copyright B 1988 by Academic Press, Inc. All rights of reproduction in any form reserved. 312 WALKER-BATSON ET AL. deficits than the right-handed aphasic patient with left infarction, language function in the right hemisphere is less strongly lateralized. Recent studies of crossed aphasics (Assal, Perentes, & Deruaz, 1981; Cart-, Jacobson, & Boller, 1981; Haaland & Miranda, 1982; Henderson, 1983), however, have cast doubt on the diffuse laterality hypothesis. The evolution of aphasia in right-handed individuals with left hemisphere lesions has been suggested to involve the intact right hemisphere (Cummings, Benson, Walsh, & Levine, 1979; Knopman, Rubens, Selnes, Klassen, & Meyer, 1984; Kinsbourne, 1971; Pettit, 1979; Tikofsky, Collier, Hellman, Saxena, Zielonka, Krohn, & Gresch, 1985). Cummings et al. (1979) suggested that transfer of function to the nondominant hemisphere may underlie the return of some language function. Cumming’s conclusions were taken from a single case study based on the severity of structural damage to the left hemisphere documented on CT. No physiological assessments of brain function were available. Two recent blood flow studies employing both two-dimensional (with language activation) (Knopman et al., 1984) and tomographic techniques (resting state only) (Tikofsky et al., 1985) also interpreted their findings in terms of shift of function hypothesis. Problems with two-dimensional blood flow techniques have been elaborated previously (Ackerman 1984) and resting state regional cerebral blood flow (rCBF) has not been predictive of recovery in several aphasia recovery studies (Demmeurisse, Verhas, Capon, & Paternot, 1983; Nagata, Yunoki, Kabe, Suzuki, & Araki, 1986). Aside from their rarity, studies of crossed aphasic patients have lacked adequate subject definition. Henderson (1983) points out the confounding variables that must be taken into account when assessing brain/language mechanisms in crossed aphasia. Many crossed aphasics reported in the literature had lesions other than a discrete region of infarction; many were bilingual and did not have adequate language testing. Henderson (1983) suggests that individuals who are studied for crossed aphasia should be strongly right handed, be monolingual, be without history of childhood brain damage, and have a CT scan or pathologically confirmed hemisphere infarction. We have had the opportunity for a detailed follow-up evaluation of a well-defined crossed aphasia patient 10 years after right cerebral infarction. Hospital records and physician report indicate an initial severe aphasic disturbance for a period of 18 months in all aspects of language functioning (reading, writing, speaking, and comprehending). Although the patient is not fully recovered, he currently presents with language skills adequate for social communication and the ability to comprehend written material enough to read for pleasure. In order to determine the possible neural basis for his improvement, we assessed rCBF using single-photon emission tomography (SPECT) during language activation tasks documented to require linguistic processing CROSSED APHASIA in the left hemisphere we present a detailed cognitive abilities. ASSESSED BY SPECT 313 of normal right-handed individuals. In addition, account of our patient’s current language and CASE REPORT The patient is a 65year-old, strongly right-handed man (laterality quotient 100 on the Edinburgh-Oldfield Index) (Oldfield, 1971) who completed 3 years of college. Ten years previously, he suffered an extensive thrombosis of the right middle cerebral artery verified by arteriography. A neurologic report 6 weeks postinsult indicated that the patient was completely paralyzed on the left side and exhibited a severe aphasia. No formal language testing results are available, but the neurological evaluation noted severe difficulties in all areas of language functioning-comprehension, reading, speaking, and writing. Ten months after the stroke a follow-up neurological evaluation reported improved comprehension with the ability to follow some commands. The patient was able to sign his name to checks in a legible manner. Paralysis of the left extremities persisted but sensation to painful stimulation had returned. Again, no formal language assessment is available, but his spouse stated that language abilities remained severely impaired with very limited verbal output. Ten years after the stroke we found that the patient was fully alert, oriented times 3, and appeared to be very intelligent. He was generally able to follow commands, although he exhibited some difficulty with “right,” “up,” and “down.” Further language evaluation is discussed below. Visual fields and extraocular movements were intact, there was numbness and weakness over the left side of the face, and there was a left hemiparesis and hemisensory loss with the arm more severely involved than the leg. There was no evidence of the neglect syndrome. Language and Neuropsychological Evaluation The spontaneous speech of this patient did not fit into a fluent/nonfluent continuum. On ratings of melody, phrase length, and articulatory agility he resembled a patient with fluent aphasia, yet his language was agrammatic with few verbs or function words. An independent rater judged the spontaneous speech sample with the comment that the patient did a beautiful job of stringing together content words. On the Boston Diagnostic Aphasia Examination (BDAE, Goodglass & Kaplan, 1972) (Table 1) the patient presented the profile of a chronic Broca’s aphasia except for the unusual aspect of his productive speech. Confrontation naming of single words was normal, and responsive naming was slightly impaired, but fluency in controlled association (animal naming) was deficient with the patient naming only seven animals in a I-min time interval. Single word repetition was adequate; however, phrase repetition was severely impaired. Reading was normal for symbol and word dis- 314 WALKER-BATSON TABLE BOSTON DIAGNOSTIC APHASIA Spontaneous speech No. of words per phrase (7) Grammatical form (7) Articulatory agility (7) Comprehension BDAE Z score (+ 1) Body party (20) Single word discrimination (72) Commands (15) Complex ideational material (12) Repetition Single words (IO) High-probability phrases (8) Low-probability phrases (8) Naming Pictures (105) Objects (18) Letters (18) Silent reading sentence comprehension Writing narrative sentences Praxis Buccofacial Music Singing (2) Rhythm (2) ET AL. 1 EXAMINATION (BDAE) TEST SCORES Fluent, but content words 7 3 6 -.2 12 72 2 12 10 2 1 (10) 104 18 18 9 Incomplete but some comprehensible phrases Moderate 2 2 Note. Numbers in parentheses are highest possible score for each item. crimination, word recognition, and reading sentences and paragraphs. There was some difficulty with body-part identification and difficulty in following commands with the patient scoring only 3 out of a possible 13 points on this subtest. Writing tasks showed most letters to be adequately formed with spontaneously written material being more legible than that written to dictation. Writing as well as oral expression was characterized by the limited use of verbs or relational terms and few paraphasic errors. An overall severity rating of three was attained on the BDAE. The F’rutting Pragmatic Protocol (Brutting & Gallagher, 1983) revealed adequate pragmatic skills with the exception of the fluency dimension. Checks for aprosodia with the method suggested by Ross (1981) demonstrated intact exression and comprehension of the affective aspects of language. The patient was able to sing melodies from old musicals without difficulty. On the Wechsler Adult Intelligence Scale (WAS) the patient received a scaled score of 8 on the Block Design subtest (Wechsler, 1955). His block arrangements resembled those observed with individuals who have CROSSED APHASIA ASSESSED BY SPECT 315 FIG. 1. Drawings from our subject with a model (top) and to command (bottom). left hemisphere lesions where the outer configurations of the blocks are maintained but the internal features are distorted (Weintraub, 1979). Drawings to command were deficient, as demonstrated on constructional tasks; the patient improved with practice, however, when allowed to copy from a model (Strub & Black, 1977) (Fig. 1). The nonlanguage battery of the BDAE revealed severe buccofacial apraxia, right-left confusion, and finger agnosia. On tasks of nonverbal reasoning the patient showed inconsistent abilities. He performed well on the Picture Arrangement subtest and the Picture Completion subtests of the WAIS with scaled scores of 12 and II, respectively. However, he attained a raw score of only 20 on the Raven’s Progressive Matrices (Raven, 1958). The patient did well on overlearned material, earning a scaled score of 12 on the WAIS Information subscale and a 10 on the Vocabulary subtest scale. Proverb interpretations from the neurologic mental status examination were concrete. Radiologic Findings A CT scan (Fig. 2) was obtained from a GE 8800 scanner. Slice thickness was 10 mm. The CT scan revealed a large area of infarction in the territory of the right middle cerebral artery, including areas homologous to Broca’s and Wemicke’s areas, extensive involvement of subcortical white matter, and portions of the basal ganglia. There appeared to be a small isolated area of cortical sparing in Wemicke’s area on one slice but more than 316 WALKER-BATSON ET AL. FIG. 2. CT scan of our subject showing a large area of infarction in the territory of the right middle cerebral artery including areas homologous to Broca’s and Wernicke’s areas with involvement of the white matter and portions of the basal ganglia. The left hemisphere showed no abnormalities. half of Wernicke’s area was damaged. The left hemisphere showed no abnormalities. rCBF was studied by means of SPECT, employing Xenon 133 inhalation and washout, a technique which has been described previously (Stokely, Sveinsdottir, & Lassen, 1980; Wallesch, Henriksen, Kornhuber, & Paulson, 1985). Figure 3 depicts the rCBF orientation. Figure 4 represents our CROSSED APHASIA FIG. FIG. 4. Regional cerebral answering silently questions task. ASSESSED BY SPECT 317 3. rCBF orientation. blood flow during to the Information resting state, doing mental arithmetic task, subscale of the WAIS, and during phoneme 318 WALKER-BATSON TABLE BL~~DFLOW Frontal Parietal Temporal Central Occipital RATIOS NORMALIZED ET AL. 2 TO HEMISPHERIC VALUES (CBF ML/~OO G/MN) Resting L/R Math L/R Weschler L/R Phoneme L/R .94/.92 1.21/.46 1.03/.58 .94/1.04 .94/1.58 .92/.88 1.26/.51 1.031.70 .86/.89 1.00/1.79 .94/.88 1.28/.31 .94/.69 1.02/1.14 1.25/.43 1.04/.50 1.081.96 .98/l .58 .95/.91 1.01/1.72 patient’s rCBF measured on four separate occasions: during a resting state, while answering silently the questions to the Information subscale of the WAIS, during a phoneme detection task, and during a mental arithmetic task. Slice 2 shown in Fig. 3 was 6 cm above and parallel to the canthomeatal line. Slice thickness was 2 cm. During the math activation the subject was told to answer the questions in his head. One of the experimenters stood at the subject’s feet and asked simple serial math problems, i.e., 8 minus 2 plus 3 minus 1, etc., for the 4 min required for the Xenon 133 inhalation and washout. An appropriate interval between questions was allowed for a response. The questions were presented continually for 4 min. Because our subject performed so well on the WAIS Information subscale, we were interested in possible neural mechanisms for performing this task. The questions were posed to the subject in the same manner as the math activation task and the subject was told to answer silently in his head. The phoneme detection task was developed in our laboratory for linguistic activation. The phonemes /p/, /b/, /k/, /g/, /t/, and /d/ were produced randomly every 4 set paired with the central vowel /a/. The subject was asked to lift his left index finger each time he heard a syllable beginning with /b/. The experimenter kept track of the accuracy of the responses. Our subject correctly identified 95% of the /b/ syllables during the 4-min activation period. The SPECT study obtained in the resting state showed a greatly diminished blood flow through the territory of the right middle cerebral artery, correlating with the CT scan. Blood flow in the left hemisphere was normal. Regional cerebral blood flow values determined from SPECT slice 2 and normalized to hemispheric values are presented in Table 2. To determine the effects of the cognitive activation tasks, activation ratios relative to resting values were determined by the formula shown in Table 3. The three different activation tasks were compared for left and right hemispheres separately. From the activation ratios shown in Table 3 it can be seen that there were regional increases to the right parietal and temporal lobes during the math activation, increases in left CROSSED APHASIA ASSESSED TABLE ACTIVATION 319 BY SPECT 3 RATIOS” Right WAIS Phoneme -4 -33 19 -8 0 24 -1 -14 -13 9 central and right temporal lobes during the WAIS Information increases to both frontal lobes during the phoneme detection task, and task. Math Frontal Parietal Temporal Central Occipital -2 4 0 -9 6 Left WAIS 0 6 -9 15 4 Phoneme Math 19 3 4 1 I -4 11 21 - 14 13 L?Activation ratios are defined as region flow AR = activated flow ratio = flow ratio RR = resting flow ratio hemisphere flow Activation ratio = AR - RR RR x loo DISCUSSION This is the 13th patient reported in the literature to satisfy Henderson’s criteria for study of patients with crossed aphasia. This patient was strongly right handed, had no relatives who were left handed, was monolingual, and had no history of childhood brain damage. Onset of the language disorder was associated with a left hemiparesis and sensory loss but no unilateral neglect syndrome. Ten years later the neurologic examination disclosed a persistent left-sided sensorimotor deficit with no abnormalities referable to the left hemisphere. Clinical findings were supported by a CT scan showing extensive infarction in the territory of the right middle cerebral artery and by a SPECT scan showing greatly diminished blood flow in the distribution of the right middle cerebral artery. No radiologic abnormalities were detected in the left hemisphere. This patient appears to present a picture of reversal of hemispheric functioning. On the nonlanguage battery of the BDAE this patient had deficits which were typical of individuals who suffer dominant parietal lobe lesions. The constructional deficits of our patient were typical of subjects who have left hemisphere lesions when the left hemisphere is the language hemisphere (Gainotti & Tiacci, 1979; Piercy, Hecaen, & Ajuriaguerra, 1960). Our patient’s drawings were simplified, improved with practice, and were better in copying than in free drawings (Fig. 1). On the Block Design subtest of the WAIS our patient made errors typical of individuals with left language-dominant lesions (Weintraub, 1979). Geschwind and Galaburda (1984) recently suggested that when aphasia 320 WALKER-BATSON ET AL. occurs in right handers from right-sided lesions it is likely to be accompanied by typical right hemisphere spatial and emotional alterations. This was not the case with our subject. He demonstrated none of the problems of affect or prosody suggested by Ross (1981); his constructional deficits were qualitatively like left dominant language individuals, and there was no evidence of unilateral neglect. Regional cerebral blood flow determinations were used to assess possible brain/language mechanisms by several activation procedures. The data in Tables 2 and 3 indicate task-specific increases in rCBF for math activation, WAIS subscale, and phoneme detection tasks. While the WAIS activation could represent an attentional or alerting response, both the phoneme detection task (bilateral frontal increases) and the math activation (right parietal and temporal increases) represent differential language processing in our subject. Luria (1963) postulated three explanations for the return of neurologic function after brain injury: (1) recovery of injured tissue, (2) shift of function to homologous areas of the other hemisphere, and (3) reorganization in which entirely different neuronal structures assume impaired functions. Most previous aphasia recovery studies have intepreted their results based on Luria’s shift of function hypothesis. Nicuum and associates (1986a,b), in a recent electrophysiological study of recovery in aphasia, exercise caution in interpreting results in favor of the shift of function hypothesis. We agree. These researchers found that lesion site and size and task complexity interact to determine whether the shift of function phenomena occurs. The problem of task complexity is demonstrated in our single case study in that different neuromechanisms were employed during the two language activation tasks. The phoneme detection task could be interpreted in terms of the shift of function hypothesis and the math activation task in terms of the reorganization theory. We feel that the data from our single case study are insufficient to uphold or dispel either hypothesis. Additional subjects need to be followed before any generalizations can be made regarding trends in focal uptake during cognitive activation procedures. With the opportunity to study patients in the acute stages of CVA along with increasing sensitivity of the newer imaging techniques plus careful task selection, data regarding neural bases for recovery in aphasic patients will improve. SPECT determination of rCBF has the potential to become a useful tool in the study of recovery from aphasia. REFERENCES Ackerman, R. H. 1984. Of cerebral blood flow, stroke and SPECT. Stroke, 15, l-4. Assal, G., Perentes, E., & Deruaz, J. P. 1981. Crossed aphasia in a right-handed patient. Postmortem findings. Archives of Neurology, 38, 455-458. Brown, J. W., & Hecaen, H. 1976. Lateralization and language representation: Observations CROSSED APHASIA ASSESSED BY SPECT on aphasia in children, left-handers, and “anomalous” 321 dextrals. Neurology, 26, 183- 189. Carr, M. S., Jacobson, T., & Boller, F. 1981. Cross aphasia: Analysis of four cases. Brain and Language, 14, 190-202. Cummings, J., Benson, D. F., Walsh, M. J., & Levine, H. L. 1979. Left-to-right transfer of language dominance: A case study. Neurology, 29, 1547-1550. Demmeurisse, G., Verhas, M., Capon, A., & Paternot, J. 1983. Lack of evolution of the cerebral blood flow during clinical recovery of a stroke. Srroke, 14, 77-81. Gainotti, G., & Tiacci, C. 1979. Patterns of drawing disability in right and left hemispheric patients. Neuropsychologia, 8, 379-434. Geschwind, N., & Galaburda, A. M. 1984. Cerebral lateralization. Biological mechanisms, associations, and pathology. I. A hypothesis and a program for research. Archives of Neurology, 42, 428-450. Goodglass, H., & Kaplan, E. 1972. Assessmenf of aphasia and related disorders. Philadelphia: Lea & Febiger. Haaland, K. Y., & Miranda, F. 1982. Psychometric and CT scan measurements in a case of crossed aphasia in a dextral. Brain and Language, 17, 240-260. Henderson, V. 1983. Speech fluency in crossed aphasia. Bruin, 106, 837-859. Kinsbourne, M. 1971. The minor cerebral hemisphere as a source of aphasic speech. Archives of Neurology, 25, 302-306. Knopman, D. S., Rubens, A. B., Seines, D., Klassen, A. C., & Meyer, M. W. 1984. Mechanisms of recovery from aphasia: Evidence from serial Xenon-133 cerebral blood flow studies. Annals of Neurology, 15, 530-535. Luria, A. R. 1963. Restoration offunction uffer brain injury. Oxford: Pergamon (original work published 1948). Nagata, K., Yunoki, K., Kabe, S., Suzuki, A., & Araki, G. 1986. Regional cerebral blood flow correlates of aphasia outcome in cerebral hemorrhage and cerebral infarction. Stroke, 17, 417-423. Niccum, N., Speaks, C., Rubens, A. B., & Knopman, D. S. 1986a. Longitudinal dichotic listening patterns for aphasic patients. II. Relationship with lesion variables. Brain and Language, 28, 289-303. Niccum, N., Speaks, C., Rubens, A. B., & Knopman, D. S. 1986b. Longitudinal dichotic listening patterns for aphasic patients. III. Relationship to language and memory variables. Brain and Language 28, 303-318. Oldfield, R. C. 1971. The assessment and analysis of handedness: The Edinburgh Inventory. Neuropsychologia, 9, 97-l 13. Pettit, J. M. 1979. Cerebral dominance in aphasia recovery. Bruin and Language, 7, 191200. Piercy, M., Hecaen, H., & Ajuriaguerra, J. 1960. Construction apraxia associated with unilateral cerebral lesions-Left and right sided cases compared. Brain, 83, 225-242. Prutting, C. A., & Gallagher, T. M. 1983. Pragmatic assessment and intervenfion issues in language. San Diego: College Hill Press. Raven, J. C. 1958. Standard progressive matrices. Los Angeles: Western Psychological Services. Ross, E. E. 1981. The aprosodias. Functional-anatomic organization of the affective components of language in the right hemisphere. Archives of Neurology, 38, 561-569. Stokely, E. M., Sveinsdottir, E., & Lassen, N. A. 1980. A single photon dynamic computer assisted tomograph (DCAT) for imaging brain function in multiple cross-sections. Journal of Computer Assisted Tomography, 4, 230-240. Strub, R. L., & Black, F. W. 1977. The mental status examination in neurology. Philadelphia: Davis. Tikofsky, R. S., Collier, B. D., Hellman, R. S., Saxena, V. K., Zielonka, J. S., Krohn, 322 WALKER-BATSON ET AL. L., & Gresch, A. 1985. Cerebral blood flow patterns determined by SPECT I-123 iodoamphetamine (IMP) imaging and WAB AQs in chronic aphasics: A preliminary report. Proceedings of the Academy of Aphasia, Pittsburg. Wallesch, C. W., Hemiksen, L., Komhuber, H. H., & Paulson, 0. B. 1985. Observations on regional cerebral blood flow in cortical and subcortical structures during language production in normal man. Brain and Language, 25, 224-233. Wechsler, D. 1955. The Wechsler adult intelligence scale. New York: The Psychological Corporation. Weintraub, S. 1979. The clinical neuropsychological examination: Its contribution to the assessment of the brain-damaged patient. Neurology of Behavior. Cambridge, MA: Harvard Medical School.