- Email: [email protected]
FMRI reveals functional cortex in a case of inconclusive Wada testing
Clinical Neurology and Neurosurgery 107 (2005) 147–151
Case report
FMRI reveals functional cortex in a case of inconclusive Wada testing Rupert Lanzenbergera,b,c , Gerald Wiestb , Alexander Geisslera,b,c , Markus Bartha,d , Helmut Ringld , Christian W¨oberb , Andreas Gartusa,b,c , Christoph Baumgartnerb , Roland Beisteinera,b,c,∗ a
Study Group Clinical fMRI at the Departments of Neurology and Radiology, Medical University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria b Department of Neurology, Medical University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria c Ludwig Boltzmann Institute for Functional Brain Topography, Medical University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria d Department of Radiology, Medical University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria Received 4 August 2003; received in revised form 22 May 2004; accepted 3 June 2004
Abstract Objectives: The intracarotid amobarbital test (Wada test) currently represents the gold standard for preoperative lateralization of hemispheric dominance. Here, we report an epileptic patient with a longstanding extended lesion of the left hemisphere showing absence of motor and speech dysfunction with left carotid amobarbital injection, but tetraplegia and speech arrest with right carotid injection interpreted as a neuroplastic shift of motor and language functions to the right hemisphere. In contrast to the Wada results, motor functional magnetic resonance imaging (fMRI) showed a strong left hemispheric activation with right hand movements. Methods: Right and left hand motor fMRI was performed. FMRI results and neurophysiological information obtained by motor and sensory evoked potential measurements were compared with the Wada test results. Results: Initial interpretation of neuroplastic shifts of intrinsic left hemisphere functions to the right brain was revised after fMRI results which were confirmed by motor and sensory evoked potentials. Conclusion: As motor inactivation usually is thought to be the most robust feature of the Wada test, this case demonstrates that fMRI may reveal residual functional cortex in cases of inconclusive Wada results. © 2004 Elsevier B.V. All rights reserved. Keywords: Wada test; fMRI; Lateralization; Preoperative diagnostic; Epilepsy
1. Introduction A large amount of literature exists for preoperative language or memory lateralization by the Wada test although several critics describe lack of international standardization as well as limited sensitivity and validity of this procedure [1,2]. In recently published studies comparing Wada test and functional magnetic resonance imaging, congruent results and dissociations were reported [3,4]. In addition, some studies claimed higher sensitivity and specificity for the fMRI ∗
Corresponding author. Tel.: +43 1 40400 3117; fax: +43 1 40400 3141. E-mail address: [email protected] (R. Beisteiner).
0303-8467/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.clineuro.2004.06.006
technique [5] and others demonstrated a gain of complementary information by fMRI [6]. It was shown that this additional localizatory information improved patient care in the sense of changed operative approaches or less invasive preoperative diagnostics [7]. Comparable discrepancies between Wada and fMRI results have not yet been reported for the usually distinctive and robust motor lateralization. Here, we report the first such case showing a clear discrepancy between the induction of arm paresis by a standard Wada test and functional lateralization of motor arm function by fMRI in a patient with a longstanding extensive left telencephalic lesion of presumably inflammatory origin.
148
R. Lanzenberger et al. / Clinical Neurology and Neurosurgery 107 (2005) 147–151
2. Case report
3. Methods and results
The 37 year old female patient K.T. suffered a left hemispheric meningoencephalitis complicated by abscess formations 3.5 months after birth. Since that time, the patient developed various epileptic attacks consisting of epigastric auras, oral and manual automatisms and generalized tonic clonic seizures. Response to various medications (e.g., carbamazepine) was not satisfactory, therefore the patient was examined for possible neurosurgical intervention. Various investigations including ictal and interictal SPECT as well as EEG video monitoring, yielded the left temporal lobe as the most likely origin of her seizures. High resolution MR showed an extended left central lesion of 5 cm × 5 cm × 4 cm, involving frontal and temporal areas including the hippocampus (Fig. 1, right). Clinically, the patient presented with a mild right-sided hemiparesis with reduced but exertable force, reduced fine motor skills of her right hand and a mildly limping gait. After completing all examinations as described below, a left antero-medial temporal resection was performed with postoperative cessation of epileptic attacks and no deterioration of the presurgical neurological status.
3.1. Wada testing Patient K.T. underwent Wada testing for presurgical evaluation of language and memory representation. The Wada test was performed according to a modified protocol used at other epilepsy centers [8,9]. Immediately after the injection of 125 mg of amobarbital in the internal carotid artery, speech (asking for the patient for his name, naming objects) and motor (elevation of the arm, hand grip with clenching the fist) functions were assessed. Injection was done with arms held up in supination to assess pronator drift (for catheter positions see angiograms, Fig. 1, left). Thereafter, we administered four blocks of four items comprising one picture of an object, one abstract design, one written concrete and one written non-concrete word (total of 16 items). After the presentation of each block, speech and motor functions were tested again. Besides fist clenching no other movement tasks were required of the subject during the Wada test. Once speech and motor functions had completely recovered, the patient’s memory was tested by a forced choice recognition paradigm
Fig. 1. Left: Angiograms taken during Wada testing of the patient for catheter placement control. Upper image shows left internal carotid area. Note: cross-flow from left to right in the central region. Right: Anatomical MR images (radiological convention) showing an extended left central lesion of 5 cm × 5 cm × 4 cm, involving frontal and temporal areas including the hippocampus.
R. Lanzenberger et al. / Clinical Neurology and Neurosurgery 107 (2005) 147–151
where each item was presented again together with matched distractor items. We calculated hits minus weighted false alarms and considered a difference score of more than one as significant. After injecting the left carotid artery, counting was not arrested and speech was fluent throughout the test. In addition, no change in motor function of the mildly preparetic right arm was observed and the arm was kept elevated with continuing strong grip of the hand. In order to exclude false negative results the catheter position in the left carotid artery was re-checked and testing was repeated after 30 min, yielding the same results. Using the same amobarbital ampulla for right carotid testing about 60 min later, caused almost simultaneously speech arrest and left hemiplegia followed by tetraplegia and loss of consciousness within about 1 min. The whole procedure was recorded on video tape, concomitant EEG recording was not available. 3.2. FMRI The fMRI measurements were carried out using a GEEPI sequence (3T BRUKER Medspec scanner (BRUKER
149
Biospin, Ettlingen, Germany), TE/TR = 55.5/4000 ms, 128 × 128 matrix, voxel size 1.8 mm × 1.8 mm × 3 mm, 25 axial slices). We used a block design with 20 s block length and seven alternating on and off conditions (four rest and three movement blocks) establishing a single experimental run of 140 s duration (7 × 20 s). In total, seven experimental runs with rhythmical fist clenching were performed per hand. Due to the preexisting paresis, right hand movements were considerably slower and somewhat more irregular compared to the left hand movements. FMRI risk maps were generated as described earlier [10–12]. Briefly, the technique consists of optimized head fixation by an individually produced plaster cast helmet which allows repeated measurements over long time periods and calculation of activation reliability values. Here, only pixels showing an activation reliability >50% (correlation coefficient >0.4) were accepted for analysis of primary motor cortex (M1) hand representation activation, thereby extracting the most robust and valid results. With right hand movements 5 pixels fulfilled these criteria for contralateral M1 and no pixel for ipsilateral M1. With left hand movements 12 pixels were found active contralaterally and one pixel ipsilaterally (see Fig. 2). For control
Fig. 2. Slices showing MI activation corresponding to the central sulcus (see inversed omega structure) for left and right hand movements. Right hemisphere is shown left. The first three images from left show original functional EPI images analyzed with the SPM standard technique. The right sided image shows a functional risk map overlaid on an anatomical image and corresponding to slice 3. It displays only most reliably activated voxels at original resolution (yellow = reliability of 70–100%, orange = about 60%). Arrows indicate MI activations. Note stronger activation corresponding to movements of the healthy left hand, no ipsilateral activation with right hand movements and small ipsilateral MI activation with left hand movements (slice 3). Note that seemingly posterior shift of left hemisphere activation with right hand movements is due to slice position and EPI distortions. Both hands locate main activation around the contralateral central sulcus (depicted by white arrows).
150
R. Lanzenberger et al. / Clinical Neurology and Neurosurgery 107 (2005) 147–151
purposes an SPM99 analysis was performed as a widely used quasi standard method. Image smoothing using an isotropic Gaussian kernel of 6 mm-FWHM, scaling for global intensity normalization and hemodynamic response function low pass filtering were applied for generation of T-value maps (P < 0.001, corrected P < 0.05). The same lateralization results were found with both analysis methods (Fig. 2). 3.3. Motor evoked potentials (MEP) Right cerebral stimulation of the motor hand area at rest yielded two responses out of two stimulations with a minimum latency of 21.9 ms and a maximum amplitude of 5.2 mV. Left cerebral stimulation yielded one response out of two stimulations with 21.2 ms latency and 0.459 mV amplitude. 3.4. Somatosensory evoked potentials (SSEP) SSEP latencies were completely normal after left median nerve stimulation (N9 = 10.2 ms, N13 = 12.7 ms, P15 = 17.3 ms, N20 = 19.7 ms, P25 = 22.5 ms) but were increased after right stimulation (N9 = 9.5 ms, N13 = 12.8 ms, P15 = 21.7 ms, N20 = 24.0 ms, P25 = 30.3 ms). 3.5. Neurosurgical intervention and postoperative outcome Total left amygdalo-hippocampectomy was used to treat the intractable seizures of the patient. A follow-up of more than 10 months shows excellent results, the patient suffers no seizures since the operation. There were no permanent complications and no post-operative dysphasia. Intraoperative electrical cortical stimulation was not possible in this case.
4. Discussion This case illustrates a clear fMRI activation of residual motor cortex within a pathological hemisphere, where the standard deactivation technique (Wada test) fails to produce a deficit. It adds two important points to the existing literature. The first demonstration of a motor lateralization failure with a Wada test as widely performed and the first demonstration of a major difference concerning detection of viable motor cortex between a standard Wada test and fMRI. Considering the longstanding extended lesion of the left hemisphere and the occurrence of speech arrest and tetraplegia with right carotid artery injection, the Wada test results were initially interpreted as a neuroplastic shift of intrinsic left hemispheric functions to the right brain, including essential motor functions – a status already reported in the literature [13]. Descriptions of coma states after injecting the healthy right hemisphere in patients with left frontal structural lesions also exist [14]. Thus, interpretation of Wada results was difficult and consequently the unexpected lack of left hemisphere
motor effects was reinvestigated by fMRI. Right hand fMRI results showed a clear left dominance of the primary motor cortex hand area and left hand fMRI showed right hemisphere dominance. The extent of fMRI activation was larger with left hand movements, and only the left hand generated ipsilateral results. Both facts correspond to the faster movements and the stronger grip of the healthy left hand, because activation intensity and suprathreshold ipsilateral motor results depend on movement frequency and force. Since it is not known, how fMRI performs in such difficult situations, we tried to confirm the fMRI results by motor and somatosensory evoked potentials. MEP demonstrated detectable residual motor cortex function and motor pathways in the left hemisphere and in consistence, SSEP indicated damage but no disruption of sensory pathways. All noninvasive evidence taken together strongly suggests that important eloquent cortex exists in the diseased left hemisphere. Taking into account lower MEP amplitudes and smaller fMRI foci, a more diffuse and variable sub-threshold representation can not be excluded for the left hemisphere. We suggest two reasons for the ambiguous Wada results. First, angiographic results (Fig. 1, left) showed a larger left anterior cerebral and left pericallosal artery with an increased area of blood supply and a left to right crossflow in the central region obviously taking over part of the territory of the right anterior cerebral artery, which correspondingly appeared hypoplastic. Therefore the pronounced effects of the anesthesia during the right hemisphere injection could simply reflect the cumulative effects of the residual anesthesia exceeding a tolerance threshold and causing global loss of function and consciousness, whereas the initial 125 mg doses may not have reached the necessary threshold to cause unilateral anesthesia during the first two left hemisphere trials. Note that a pharmacological left hemisphere problem may be excluded since the right hemisphere injection was done with residual amobarbital from the second left try (taken from the same 500 mg ampulla). In this context a limitation of this study might be that we do not use routine EEG recordings during the Wada test any more because the electrodes might interfere with the interpretation of the angiograms. Therefore, it is not clear whether the left hemisphere was ever fully anesthetized during the task. However, in our patient a marked continuous slowing was evident on baseline interictal EEG already which would have compromised the interpretation of an EEG recorded during the Wada test procedure anyhow. Second, the required dose of amobarbital may need to be higher than 125 mg. Although a higher dose could be tried, this procedure increases the risk for possible side effects (e.g., breathing cessation). Altogether, the data indicate a failure of left hemisphere inactivation with a standard Wada procedure as widely used in epileptic patients and an abnormal intracerebral vascularization seems to be the most likely cause. Without detailed angiographic analysis such situations are not predictable and as a consequence one should bear in mind that the Wada technique may fail to inactivate a hemisphere, although usual injection techniques and dosages are used. Of course one should be aware that it is also quite possible to
R. Lanzenberger et al. / Clinical Neurology and Neurosurgery 107 (2005) 147–151
have a cluster of activation present during an fMRI that is not essential for normal motor functioning. Then the left hemisphere activations found with right hand movements would have little to do with the actual performance of the motor task. Given the fact that only left hemisphere fMRI activations were detected with right hand movements in our patient and that these results were congruent with two different data analysis techniques (see Fig. 2) a complete miss of essential motor activations (now shifted to another region or hemisphere) but simultaneous detection of non essential motor areas seems unlikely in our case. Although a definite explanation of this novel observation is not possible and clinical consequences were limited here, we may conclude that in cases with conspicuous Wada results, fMRI may help to reveal residual functional cortex. Acknowledgement We would like to acknowledge important general support by Prof. Dr. Lueder Deecke, Head Ludwig Boltzmann Institute for Functional Brain Topography, Department of Clinical Neurology and Prof. Siegfried Trattnig, Medical director, MR Centre of Excellence, Department of Radiology, Medical University of Vienna. The authors are grateful to Nick Tahamtan and Denny Milakara for technical assistance. This study was supported by the Austrian Science Foundation (P15102). References [1] Hunter KE, Blaxton TA, Bookheimer SY, et al. (15)O water positron emission tomography in language localization: a study comparing positron emission tomography visual and computerized region of interest analysis with the Wada test. Ann Neurol 1999;45:662–5.
151
[2] Meador KJ, Loring DW. The Wada test. Controversies, concerns, and insights. Neurology 1999;52:1535–6. [3] Bazin B, Cohen L, Lehericy S, et al. Study of hemispheric lateralization of the language regions by functional MRI: validation with the Wada test. Rev Neurol 2000;156:145–8. [4] Lehericy S, Cohen L, Bazin B, et al. Functional MR evaluation of temporal and frontal language dominance compared with the Wada test. Neurology 2000;54:1625–33. [5] Brockway JP. Two functional magnetic resonance imaging f(MRI) tasks that may replace the gold-standard, Wada testing, for language lateralization while giving additional localization information. Brain Cogn 2000;43:57–9. [6] Killgore WD, Glosser G, Casasanto DJ, et al. Functional MRI and the Wada test provide complementary information for predicting post-operative seizure control. Seizure 1999;8:450–5. [7] Spreer J, Quiske A, Altenmuller DM, et al. Unsuspected atypical hemispheric dominance for language as determined by fMRI. Epilepsia 2001;42:957–9. [8] Lancman ME, Benbadis S, Geller E, Morris HH. Sensitivity and specificity of asymmetric recall on Wada test to predict outcome after temporal lobectomy. Neurology 1998;50:455–9. [9] Jokeit H, Ebner A, Holthausen H, Markowitsch HJ, Moch A, Pannek H, Schulz R, Tuxhorn I. Individual prediction of change in delayed recall of prose passages after left-sided anterior temporal lobectomy. Neurology 1997;49:481–7. [10] Beisteiner R, Lanzenberger R, Novak K, et al. Improvement of presurgical patient evaluation by generation of functional magnetic resonance risk maps. Neurosci Lett 2000;290:13–6. [11] Beisteiner R, Windischberger C, Lanzenberger R, et al. Finger somatotopy in human motor cortex. Neuroimage 2001;13:1016–26. [12] Beisteiner R. Indikationen, Probleme und Ergebnisse der funktionellen MRT im Kindesalter. P¨adiatrische Praxis 2004;64(2):285–98. [13] Staudt M, Pieper T, Grodd W, et al. Functional MRI in a 6-year-old boy with unilateral cortical malformation: Concordant representation of both hands in the unaffected hemisphere. Neuropediatrics 2001;32:159–61. [14] Lee GP, Loring DW, Meador KJ, Flanigin HF, Brooks BS. Severe behavioral complications following intracarotid sodium amobarbital injection: implications for hemispheric asymmetry of emotion. Neurology 1988;38:1233–6.
Case report
FMRI reveals functional cortex in a case of inconclusive Wada testing Rupert Lanzenbergera,b,c , Gerald Wiestb , Alexander Geisslera,b,c , Markus Bartha,d , Helmut Ringld , Christian W¨oberb , Andreas Gartusa,b,c , Christoph Baumgartnerb , Roland Beisteinera,b,c,∗ a
Study Group Clinical fMRI at the Departments of Neurology and Radiology, Medical University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria b Department of Neurology, Medical University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria c Ludwig Boltzmann Institute for Functional Brain Topography, Medical University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria d Department of Radiology, Medical University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria Received 4 August 2003; received in revised form 22 May 2004; accepted 3 June 2004
Abstract Objectives: The intracarotid amobarbital test (Wada test) currently represents the gold standard for preoperative lateralization of hemispheric dominance. Here, we report an epileptic patient with a longstanding extended lesion of the left hemisphere showing absence of motor and speech dysfunction with left carotid amobarbital injection, but tetraplegia and speech arrest with right carotid injection interpreted as a neuroplastic shift of motor and language functions to the right hemisphere. In contrast to the Wada results, motor functional magnetic resonance imaging (fMRI) showed a strong left hemispheric activation with right hand movements. Methods: Right and left hand motor fMRI was performed. FMRI results and neurophysiological information obtained by motor and sensory evoked potential measurements were compared with the Wada test results. Results: Initial interpretation of neuroplastic shifts of intrinsic left hemisphere functions to the right brain was revised after fMRI results which were confirmed by motor and sensory evoked potentials. Conclusion: As motor inactivation usually is thought to be the most robust feature of the Wada test, this case demonstrates that fMRI may reveal residual functional cortex in cases of inconclusive Wada results. © 2004 Elsevier B.V. All rights reserved. Keywords: Wada test; fMRI; Lateralization; Preoperative diagnostic; Epilepsy
1. Introduction A large amount of literature exists for preoperative language or memory lateralization by the Wada test although several critics describe lack of international standardization as well as limited sensitivity and validity of this procedure [1,2]. In recently published studies comparing Wada test and functional magnetic resonance imaging, congruent results and dissociations were reported [3,4]. In addition, some studies claimed higher sensitivity and specificity for the fMRI ∗
Corresponding author. Tel.: +43 1 40400 3117; fax: +43 1 40400 3141. E-mail address: [email protected] (R. Beisteiner).
0303-8467/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.clineuro.2004.06.006
technique [5] and others demonstrated a gain of complementary information by fMRI [6]. It was shown that this additional localizatory information improved patient care in the sense of changed operative approaches or less invasive preoperative diagnostics [7]. Comparable discrepancies between Wada and fMRI results have not yet been reported for the usually distinctive and robust motor lateralization. Here, we report the first such case showing a clear discrepancy between the induction of arm paresis by a standard Wada test and functional lateralization of motor arm function by fMRI in a patient with a longstanding extensive left telencephalic lesion of presumably inflammatory origin.
148
R. Lanzenberger et al. / Clinical Neurology and Neurosurgery 107 (2005) 147–151
2. Case report
3. Methods and results
The 37 year old female patient K.T. suffered a left hemispheric meningoencephalitis complicated by abscess formations 3.5 months after birth. Since that time, the patient developed various epileptic attacks consisting of epigastric auras, oral and manual automatisms and generalized tonic clonic seizures. Response to various medications (e.g., carbamazepine) was not satisfactory, therefore the patient was examined for possible neurosurgical intervention. Various investigations including ictal and interictal SPECT as well as EEG video monitoring, yielded the left temporal lobe as the most likely origin of her seizures. High resolution MR showed an extended left central lesion of 5 cm × 5 cm × 4 cm, involving frontal and temporal areas including the hippocampus (Fig. 1, right). Clinically, the patient presented with a mild right-sided hemiparesis with reduced but exertable force, reduced fine motor skills of her right hand and a mildly limping gait. After completing all examinations as described below, a left antero-medial temporal resection was performed with postoperative cessation of epileptic attacks and no deterioration of the presurgical neurological status.
3.1. Wada testing Patient K.T. underwent Wada testing for presurgical evaluation of language and memory representation. The Wada test was performed according to a modified protocol used at other epilepsy centers [8,9]. Immediately after the injection of 125 mg of amobarbital in the internal carotid artery, speech (asking for the patient for his name, naming objects) and motor (elevation of the arm, hand grip with clenching the fist) functions were assessed. Injection was done with arms held up in supination to assess pronator drift (for catheter positions see angiograms, Fig. 1, left). Thereafter, we administered four blocks of four items comprising one picture of an object, one abstract design, one written concrete and one written non-concrete word (total of 16 items). After the presentation of each block, speech and motor functions were tested again. Besides fist clenching no other movement tasks were required of the subject during the Wada test. Once speech and motor functions had completely recovered, the patient’s memory was tested by a forced choice recognition paradigm
Fig. 1. Left: Angiograms taken during Wada testing of the patient for catheter placement control. Upper image shows left internal carotid area. Note: cross-flow from left to right in the central region. Right: Anatomical MR images (radiological convention) showing an extended left central lesion of 5 cm × 5 cm × 4 cm, involving frontal and temporal areas including the hippocampus.
R. Lanzenberger et al. / Clinical Neurology and Neurosurgery 107 (2005) 147–151
where each item was presented again together with matched distractor items. We calculated hits minus weighted false alarms and considered a difference score of more than one as significant. After injecting the left carotid artery, counting was not arrested and speech was fluent throughout the test. In addition, no change in motor function of the mildly preparetic right arm was observed and the arm was kept elevated with continuing strong grip of the hand. In order to exclude false negative results the catheter position in the left carotid artery was re-checked and testing was repeated after 30 min, yielding the same results. Using the same amobarbital ampulla for right carotid testing about 60 min later, caused almost simultaneously speech arrest and left hemiplegia followed by tetraplegia and loss of consciousness within about 1 min. The whole procedure was recorded on video tape, concomitant EEG recording was not available. 3.2. FMRI The fMRI measurements were carried out using a GEEPI sequence (3T BRUKER Medspec scanner (BRUKER
149
Biospin, Ettlingen, Germany), TE/TR = 55.5/4000 ms, 128 × 128 matrix, voxel size 1.8 mm × 1.8 mm × 3 mm, 25 axial slices). We used a block design with 20 s block length and seven alternating on and off conditions (four rest and three movement blocks) establishing a single experimental run of 140 s duration (7 × 20 s). In total, seven experimental runs with rhythmical fist clenching were performed per hand. Due to the preexisting paresis, right hand movements were considerably slower and somewhat more irregular compared to the left hand movements. FMRI risk maps were generated as described earlier [10–12]. Briefly, the technique consists of optimized head fixation by an individually produced plaster cast helmet which allows repeated measurements over long time periods and calculation of activation reliability values. Here, only pixels showing an activation reliability >50% (correlation coefficient >0.4) were accepted for analysis of primary motor cortex (M1) hand representation activation, thereby extracting the most robust and valid results. With right hand movements 5 pixels fulfilled these criteria for contralateral M1 and no pixel for ipsilateral M1. With left hand movements 12 pixels were found active contralaterally and one pixel ipsilaterally (see Fig. 2). For control
Fig. 2. Slices showing MI activation corresponding to the central sulcus (see inversed omega structure) for left and right hand movements. Right hemisphere is shown left. The first three images from left show original functional EPI images analyzed with the SPM standard technique. The right sided image shows a functional risk map overlaid on an anatomical image and corresponding to slice 3. It displays only most reliably activated voxels at original resolution (yellow = reliability of 70–100%, orange = about 60%). Arrows indicate MI activations. Note stronger activation corresponding to movements of the healthy left hand, no ipsilateral activation with right hand movements and small ipsilateral MI activation with left hand movements (slice 3). Note that seemingly posterior shift of left hemisphere activation with right hand movements is due to slice position and EPI distortions. Both hands locate main activation around the contralateral central sulcus (depicted by white arrows).
150
R. Lanzenberger et al. / Clinical Neurology and Neurosurgery 107 (2005) 147–151
purposes an SPM99 analysis was performed as a widely used quasi standard method. Image smoothing using an isotropic Gaussian kernel of 6 mm-FWHM, scaling for global intensity normalization and hemodynamic response function low pass filtering were applied for generation of T-value maps (P < 0.001, corrected P < 0.05). The same lateralization results were found with both analysis methods (Fig. 2). 3.3. Motor evoked potentials (MEP) Right cerebral stimulation of the motor hand area at rest yielded two responses out of two stimulations with a minimum latency of 21.9 ms and a maximum amplitude of 5.2 mV. Left cerebral stimulation yielded one response out of two stimulations with 21.2 ms latency and 0.459 mV amplitude. 3.4. Somatosensory evoked potentials (SSEP) SSEP latencies were completely normal after left median nerve stimulation (N9 = 10.2 ms, N13 = 12.7 ms, P15 = 17.3 ms, N20 = 19.7 ms, P25 = 22.5 ms) but were increased after right stimulation (N9 = 9.5 ms, N13 = 12.8 ms, P15 = 21.7 ms, N20 = 24.0 ms, P25 = 30.3 ms). 3.5. Neurosurgical intervention and postoperative outcome Total left amygdalo-hippocampectomy was used to treat the intractable seizures of the patient. A follow-up of more than 10 months shows excellent results, the patient suffers no seizures since the operation. There were no permanent complications and no post-operative dysphasia. Intraoperative electrical cortical stimulation was not possible in this case.
4. Discussion This case illustrates a clear fMRI activation of residual motor cortex within a pathological hemisphere, where the standard deactivation technique (Wada test) fails to produce a deficit. It adds two important points to the existing literature. The first demonstration of a motor lateralization failure with a Wada test as widely performed and the first demonstration of a major difference concerning detection of viable motor cortex between a standard Wada test and fMRI. Considering the longstanding extended lesion of the left hemisphere and the occurrence of speech arrest and tetraplegia with right carotid artery injection, the Wada test results were initially interpreted as a neuroplastic shift of intrinsic left hemispheric functions to the right brain, including essential motor functions – a status already reported in the literature [13]. Descriptions of coma states after injecting the healthy right hemisphere in patients with left frontal structural lesions also exist [14]. Thus, interpretation of Wada results was difficult and consequently the unexpected lack of left hemisphere
motor effects was reinvestigated by fMRI. Right hand fMRI results showed a clear left dominance of the primary motor cortex hand area and left hand fMRI showed right hemisphere dominance. The extent of fMRI activation was larger with left hand movements, and only the left hand generated ipsilateral results. Both facts correspond to the faster movements and the stronger grip of the healthy left hand, because activation intensity and suprathreshold ipsilateral motor results depend on movement frequency and force. Since it is not known, how fMRI performs in such difficult situations, we tried to confirm the fMRI results by motor and somatosensory evoked potentials. MEP demonstrated detectable residual motor cortex function and motor pathways in the left hemisphere and in consistence, SSEP indicated damage but no disruption of sensory pathways. All noninvasive evidence taken together strongly suggests that important eloquent cortex exists in the diseased left hemisphere. Taking into account lower MEP amplitudes and smaller fMRI foci, a more diffuse and variable sub-threshold representation can not be excluded for the left hemisphere. We suggest two reasons for the ambiguous Wada results. First, angiographic results (Fig. 1, left) showed a larger left anterior cerebral and left pericallosal artery with an increased area of blood supply and a left to right crossflow in the central region obviously taking over part of the territory of the right anterior cerebral artery, which correspondingly appeared hypoplastic. Therefore the pronounced effects of the anesthesia during the right hemisphere injection could simply reflect the cumulative effects of the residual anesthesia exceeding a tolerance threshold and causing global loss of function and consciousness, whereas the initial 125 mg doses may not have reached the necessary threshold to cause unilateral anesthesia during the first two left hemisphere trials. Note that a pharmacological left hemisphere problem may be excluded since the right hemisphere injection was done with residual amobarbital from the second left try (taken from the same 500 mg ampulla). In this context a limitation of this study might be that we do not use routine EEG recordings during the Wada test any more because the electrodes might interfere with the interpretation of the angiograms. Therefore, it is not clear whether the left hemisphere was ever fully anesthetized during the task. However, in our patient a marked continuous slowing was evident on baseline interictal EEG already which would have compromised the interpretation of an EEG recorded during the Wada test procedure anyhow. Second, the required dose of amobarbital may need to be higher than 125 mg. Although a higher dose could be tried, this procedure increases the risk for possible side effects (e.g., breathing cessation). Altogether, the data indicate a failure of left hemisphere inactivation with a standard Wada procedure as widely used in epileptic patients and an abnormal intracerebral vascularization seems to be the most likely cause. Without detailed angiographic analysis such situations are not predictable and as a consequence one should bear in mind that the Wada technique may fail to inactivate a hemisphere, although usual injection techniques and dosages are used. Of course one should be aware that it is also quite possible to
R. Lanzenberger et al. / Clinical Neurology and Neurosurgery 107 (2005) 147–151
have a cluster of activation present during an fMRI that is not essential for normal motor functioning. Then the left hemisphere activations found with right hand movements would have little to do with the actual performance of the motor task. Given the fact that only left hemisphere fMRI activations were detected with right hand movements in our patient and that these results were congruent with two different data analysis techniques (see Fig. 2) a complete miss of essential motor activations (now shifted to another region or hemisphere) but simultaneous detection of non essential motor areas seems unlikely in our case. Although a definite explanation of this novel observation is not possible and clinical consequences were limited here, we may conclude that in cases with conspicuous Wada results, fMRI may help to reveal residual functional cortex. Acknowledgement We would like to acknowledge important general support by Prof. Dr. Lueder Deecke, Head Ludwig Boltzmann Institute for Functional Brain Topography, Department of Clinical Neurology and Prof. Siegfried Trattnig, Medical director, MR Centre of Excellence, Department of Radiology, Medical University of Vienna. The authors are grateful to Nick Tahamtan and Denny Milakara for technical assistance. This study was supported by the Austrian Science Foundation (P15102). References [1] Hunter KE, Blaxton TA, Bookheimer SY, et al. (15)O water positron emission tomography in language localization: a study comparing positron emission tomography visual and computerized region of interest analysis with the Wada test. Ann Neurol 1999;45:662–5.
151
[2] Meador KJ, Loring DW. The Wada test. Controversies, concerns, and insights. Neurology 1999;52:1535–6. [3] Bazin B, Cohen L, Lehericy S, et al. Study of hemispheric lateralization of the language regions by functional MRI: validation with the Wada test. Rev Neurol 2000;156:145–8. [4] Lehericy S, Cohen L, Bazin B, et al. Functional MR evaluation of temporal and frontal language dominance compared with the Wada test. Neurology 2000;54:1625–33. [5] Brockway JP. Two functional magnetic resonance imaging f(MRI) tasks that may replace the gold-standard, Wada testing, for language lateralization while giving additional localization information. Brain Cogn 2000;43:57–9. [6] Killgore WD, Glosser G, Casasanto DJ, et al. Functional MRI and the Wada test provide complementary information for predicting post-operative seizure control. Seizure 1999;8:450–5. [7] Spreer J, Quiske A, Altenmuller DM, et al. Unsuspected atypical hemispheric dominance for language as determined by fMRI. Epilepsia 2001;42:957–9. [8] Lancman ME, Benbadis S, Geller E, Morris HH. Sensitivity and specificity of asymmetric recall on Wada test to predict outcome after temporal lobectomy. Neurology 1998;50:455–9. [9] Jokeit H, Ebner A, Holthausen H, Markowitsch HJ, Moch A, Pannek H, Schulz R, Tuxhorn I. Individual prediction of change in delayed recall of prose passages after left-sided anterior temporal lobectomy. Neurology 1997;49:481–7. [10] Beisteiner R, Lanzenberger R, Novak K, et al. Improvement of presurgical patient evaluation by generation of functional magnetic resonance risk maps. Neurosci Lett 2000;290:13–6. [11] Beisteiner R, Windischberger C, Lanzenberger R, et al. Finger somatotopy in human motor cortex. Neuroimage 2001;13:1016–26. [12] Beisteiner R. Indikationen, Probleme und Ergebnisse der funktionellen MRT im Kindesalter. P¨adiatrische Praxis 2004;64(2):285–98. [13] Staudt M, Pieper T, Grodd W, et al. Functional MRI in a 6-year-old boy with unilateral cortical malformation: Concordant representation of both hands in the unaffected hemisphere. Neuropediatrics 2001;32:159–61. [14] Lee GP, Loring DW, Meador KJ, Flanigin HF, Brooks BS. Severe behavioral complications following intracarotid sodium amobarbital injection: implications for hemispheric asymmetry of emotion. Neurology 1988;38:1233–6.