Shift-back of right into left hemisphere language dominance after control of epileptic seizures: Evidence for epilepsy driven functional cerebral organization

Abstract

Atypical, i.e. right hemisphere language dominance is frequently observed in early onset left hemisphere epilepsies. In left mesial temporal lobe epilepsy, where eloquent cortex is not directly involved, it is a matter of debate, to which degree atypical language dominance is driven not only by morphological lesions but also by epileptic dysfunction, and whether atypical dominance is hardwired or not. Taking this as the background this study evaluated the hypothesis that epilepsy driven atypical dominancy might be reversible when seizures are successfully controlled. This was evaluated in patients with left mesial temporal lobe epilepsy, who were atypically language dominant by means of language fMRI before surgery, and became seizure free after left selective amygdalo-hippocampectomy. Three out of 53 consecutive atypically dominant patients with chronic epilepsy fulfilled these criteria. Postoperative follow-up language fMRI indicated reversal of right into left dominance in one patient going along with unexpected losses in verbal memory performance. The two other patients experienced unchanged or even enhancement of the pre-existing dominance pattern, going along with consistent postoperative performance changes in cognition. The data thus provide supporting evidence that atypical language dominance can indeed be functionally driven and moreover that in at least some patients, right hemispheric language can shift-back to the left hemisphere when the driving factor, i.e. seizures, becomes successfully controlled. The results have clinical implications for outcome prediction after brain surgery in atypically dominant patients with epilepsy. However, further research in larger groups of atypically dominant patients is required to identify the conditions under which atypical dominance becomes hardwired and when not.

Introduction

Right hemisphere language dominance is a very rare condition in healthy subjects (∼6%) (Knecht et al., 2000, Springer et al., 1999) but is more prevalent after left hemispheric brain damage if this was acquired early in life in a time window until puberty (Rasmussen and Milner, 1977, Staudt et al., 2002). In the presence of early brain damage, atypical language dominance represents the attempt of the maturing and still functionally plastic brain to preserve or restitute language function, which obviously has superior relevance in human phylo- and ontogenetic development. In patients who suffer from early onset left hemispheric epilepsy up to 40% display right hemispheric language functions to some degree (Helmstaedter et al., 1997). This in itself fascinating phenomenon gains additional interest because of the recent suggestion from clinical observations, that a shift in language dominance may not only be induced by damage of brain tissue but also by epileptic dysfunction (Helmstaedter et al., 1997, Janszky et al., 2003, Regard et al., 1994). Correlation of atypical dominance with epileptic activity and the fact that lesions alone do not suffice to explain atypical language dominance in epilepsy patients gave rise to the interesting hypothesis that atypical dominance – at least in some patients – might not be hardwired but principally reversible if epilepsy, i.e. seizures, becomes successfully controlled (Gleissner et al., 2002). Patients with left mesial hippocampal sclerosis as the sole pathology appear particularly interesting in this respect since it is remarkable that such a circumscribed structural lesion in primarily memory processing structures should cause the brain to shift language to the right hemisphere (Helmstaedter et al., 1994, Knecht, 2004). First empiric support for the back-shift hypothesis of epilepsy driven right hemisphere language to the left hemisphere was inferred indirectly from behavioural assessment (Gleissner et al., 2002, Helmstaedter et al., 2004). This hypothesis was based on the observation of a postoperative reversal of the so called suppression or crowding effect, which describes the suppression or sacrifice of originally right hemisphere functions in favour of a right hemisphere restitution of language after left hemispheric damage (Lansdell, 1969). Direct evidence for the reversal of atypical language dominance could not yet be obtained since the gold standard for language lateralization (separate intracarotidal anaesthesia of the left/right hemisphere) is invasive and thus not repeatable after surgery. However, now that functional magnetic resonance imaging (fMRI) allows for reliable and repeated non-invasive assessment of language dominance, direct evaluation of the back-shift hypothesis is possible (Fernández et al., 2002). For this purpose patients were selected according to the following criteria: left mesial temporal lobe epilepsy (TLE) with hippocampal sclerosis as the sole epileptogenic pathology, left selective epilepsy surgery (amygdalo-hippocampectomy/SAH) not affecting language relevant cortex, no persisting seizures postoperatively, and significant right hemisphere language involvement before surgery as determined by preoperative fMRI.