Abstracts

There is a group of patients who have medically refractory epilepsy that either have no structural abnormality on MRI, have mutiple lesions, extremely large lesions or a lesion in eloquent cortex where the risks of surgical resection are very significant. Appropriate patient selection and investigation, with the ability to co-register functional modalities to an magnetic resonance imaging (MRI) and use image-guided surgery, has allowed a group of these patients to become either seizure-free, or have seizure reduction. Seizure outcome, imaging modalities used and complications following sugery in patients is discussed.
Between April 1999 and October 200, 116 operations were performed on 109 patients with medically refractory epilepsy. Patients were selected to have multimodality image-guided surgery primarily on the basis that they had a non-localising or non-lesional MRI, or where there were: (a) no lesion detected on MRI; (b) multiple lesions detected; or (c) a very large single lesion, such that it could not be completely resected without the risk of significant postoperative morbidity. A fourth group had a single lesion in eloquent cortex, where it was anticipated that there was a significant risk of postoperative morbidity. Imaging modalities such as positron emission tomography (PET), fluid attenuated inversion recovery (FLAIR) MRI and subtraction single positron emission computerised tomography (SPECT) co-registered to MRI (SISCOM) were used to identify epileptogenic foci, and then co-registered to an MRI. These images were then downloaded onto the image guidedsystem (IGS). Intracranial grid electrodes were used where lesions were in eloquent cortex, were co-registered to MRI, downloaded and used intraoperatively to minimise electrode position error.
Thirtyfive (30%) procedures on 22 (20%) patients fulfilled the criteria for inclusion. The mean age was 33 years (range 17-46), with a mean follow-up of 27 months (range 14-41). Multimodality co-registrations used were: 10 PET, 8 subdural electrode grids, 4 SISCOM and 1 FLAIR MRI. Seizure outcome was excellent in 17 (77%) patients and non-excellent in 5 (23%), and favourable in 19 (86%) patients and non-favourable in 3 (14%). Six (27%) patients had a transient neurological deficit, 1 (5%) permanent major and 3 (15%) permanent minor. Five (24%) patients had a transient psychiatric problem postoperatively
Multimodality image-guided surgery offers a new perspective in epilepsy surgery. Functional imaging modalities often lateralised and localised a seizure focus but did not provide enough anatomical information to confidently and safely resect the epileptogenic zone.The co-registration of these modalities to a volumetric MRI and their incorporation into the IGS has allowed surgeons to offer surgery to a significant number of patients previously considered inoperable with outcomes comparable to that seen in patients with more straightforward lesional epilepsy.