Article Text

Download PDFPDF

Validity of language lateralisation by unilateral intracarotid Wada test
  1. A Schulze-Bonhage1,
  2. A Quiske1
  1. 1Epilepsy Centre, University Clinic of Freiburg, Germany
  1. Correspondence to:
 Dr Andreas Schulze-Bonhage
  1. T Loddenkemper2,3,
  2. D S Dinner2,
  3. E Wyllie2
  1. 2Department of Neurology, The Cleveland Clinic Foundation, Cleveland, Ohio, USA
  2. 3Department of Pediatrics, The Cleveland Clinic Foundation

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    The case report by Loddenkemper et al1 is most important as it poses the question of whether unilateral Wada testing is sufficient for language lateralisation. It is certainly possible that similar complications resulting from epilepsy surgery based on Wada tests are underreported. The implications of erroneous language lateralisation by unilateral amobarbital injection, which is increasingly used in several epilepsy centres including our own, would be far reaching.

    There are, however, some caveats regarding the interpretation of the Wada test results reported. The first two refer to the assessment of cortical inactivation in this patient with pre-existing damage to the motor cortex; the third refers to the extent of brain areas inactivated by the intracarotid Wada procedure.

    (1) Little is known about the relative susceptibility to the effect of amobarbital of language areas compared with damaged motor cortex. Thus it may be that a low dose of amytal (which 100 mg is) does impair a partially damaged motor cortex when language areas still maintain sufficient function to solve the tasks tested during the Wada procedure. In our experience, higher doses (140 to 200 mg) are often needed in adults to induce a complete contralateral hemiplegia. In order to avoid a possible risk of incomplete inactivation of the middle cerebral artery (MCA) territory and of language areas in particular, it may be advantageous to choose higher injection dosages.

    (2) There is a striking discrepancy between well preserved proximal arm power (4-/5) after hemispherectomy as compared with complete proximal hemiplegia (0/5) reported during the Wada procedure which requires additional explanation. This may indicate that the lack of voluntary movement observed during the Wada procedure does not result from a complete inactivation of the contralateral motor cortex but may be related to contralateral motor neglect.

    (3) With regard to clinical history and magnetic resonance imaging, there is some indication that this patient may have had reorganisation of the language areas following the initial brain injury at the age of five years. Whereas the Wada test was interpreted as providing evidence for an interhemispheric shift to the right hemisphere, an intrahemispheric transfer of expressive functions to posterior temporal areas may have taken place. As the vascular supply of this region often is not restricted to MCA branches, incomplete inactivation of this area may occur.2 The results of the procedure could thus be explained by the incomplete inactivation of posterior left hemispheric language areas. It would be interesting to know if the authors coinjected a SPECT tracer3 in order to define the territory perfused by amobarbital.


    Authors’ reply

    We appreciate the above comments and are delighted that our case has sparked a discussion about the validity of the Wada test. The intention of our report was to question the use of the (unilateral) intracarotid amobarbital test (IAT) in selected cases, and we are pleased that Dr Schulze-Bonhage and Dr Quiske recognise the potential importance of this case. We would like to address their comments in detail.

    Comment 1: No clinical data are available on the effect of amobarbital on the damaged motor cortex. Thus the effectiveness of an initial high dose of amobarbital, as well as incomplete inactivation of different MCA areas, remains an interesting speculation. Although we do not know whether higher amobarbital doses are more effective in cases like ours, we know that higher doses may increase the probability of drug related side effects such as encephalopathy and impaired consciousness, particularly in patients with pre-existing left hemispheric lesions.1 Additionally, amobarbital may have a stronger effect in older patients,2 supporting the use of a lower initial dose in our 55 year old patient. Furthermore, the recent replacement of amobarbital by methohexital, owing to a shortage of amobarbital in many centres, may subject patients to a dose dependent increase in seizure frequency during Wada testing when there is a high initial dose of methohexital.3 Clinical examination of the patient remains the most important factor in the careful titration of the amobarbital dose, in order to prevent adverse effects during its administration.

    Comment 2: We thank the authors for highlighting the potentially interesting finding of the acute worsening of unilateral motor function during the Wada test. As documented, we would like to point out that the proximal strength of 4-/5 was only achieved after eight months, following vigorous rehabilitation. The discrepancy from the results during the Wada test arises because Dr Schulze-Bonhage and Dr Quiske are comparing the examination during the acute amobarbital inactivation with chronic findings eight months after recovery from hemispherectomy. The patient was completely hemiplegic immediately after hemispherectomy and then gradually regained the reported outcome strength of 4-/5. Thus the amobarbital test mimicked the immediate postoperative outcome well. Proximal weakness has been reported during IATs,4 and the documented gradual recovery of strength over 145 seconds is unlikely in patients with motor neglect.5

    Comment 3: IATs are often undertaken in patients undergoing epilepsy surgery. However, ipsilateral language areas are only rarely resected, especially if the test is positive. Even if the test is negative, the resected area does not necessarily include typical language areas. Therefore, little information on the validity of the IAT is available. IAT validation studies in predicting aphasia may be divided into four categories: true negatives, true positives, false negatives, and false positives. A false negative is someone who passed the IAT but nonetheless became aphasic. The incongruity of the clinically inactivated motor area and the target area (language) is a potential problem within the routine Wada test itself, which may be highlighted by cases like ours, and which is possibly exacerbated by an underlying subcortical lesion. Intrahemispheric reorganisation or interhemispheric language reorganisation is a possible explanation for atypical language representation in our case. The possibility of residual function and intrahemispheric reorganisation in our case is supported by intact visual fields before hemispherectomy, as previously discussed.5

    Another issue related to the validity of the IAT is the distribution of the amobarbital. Conventional angiography, digital subtraction angiography (DSA), and SPECT have been compared in terms of perfusion patterns during IAT. With regard to posterior cerebral artery (PCA) filling in 46 patients, SPECT showed filling in 22%, DSA showed filling in 37%, and conventional angiography showed contrast in 50% of patients.6 Based on these data, a conventional angiography before the IAT (as carried out and reported in our case) can rule out perfusion of the PCA. Based on this study, no HMPAO distribution would have been observed in the PCA territory. SPECT is therefore unnecessary.

    To our knowledge, we have reported the first case in which a patient continued to speak during the Wada test and then became aphasic after ipsilateral hemispherectomy. Aphasia after hemispherectomy was not anticipated. We agree with Drs Schulze-Bonhage and Quiske that it is possible that similar complications resulting from epilepsy surgery based on Wada tests are underreported. The teaching point of our case report is that a negative unilateral routine Wada test result for language may be misleading in an older patient with a predominantly subcortical lesion that was acquired after the primary development of language function. We encourage other epilepsy centres to reflect on their current practice of unilateral Wada testing for presurgical language lateralisation in selected cases and possibly replace or supplement the Wada test by additional language lateralisation techniques.

    We thank Dr Schulze-Bonhage and Dr Quiske for their comments.