Surgical treatment of drug-resistant epilepsy, University Hospital of Strasbourg

Discussion/conclusion Two types of medical devices are used in the management of epilepsy surgery: the left vagus nerve stimulation (8900 € paid besides hospital benefits) and the electrodes of stereo-electroencephalography (SEEG) in possible prior to conventional surgery. Having introduced the use of SEEG electrodes in our institution in 2004, for epilepsy surgery an average of 11 electrodes (eight to 15 electrodes) is required at an average cost of € 4284 per patient. These electrodes represent 58% of average tariff, of the disease-related groups found among patients (€ 7418). In our study, epilepsy surgery appears to be of significant clinical benefit. However, the cost-effectiveness of conventional surgery for epilepsy post-SEEG must be reevaluated to confirm the long-term interest. Résumé Objectif Le traitement chirurgical de l’épilepsie pharmacorésistante constitue le dernier recours vers la réduction, voire la guérison totale de l’épilepsie. Les objectifs de ce travail étaient, d’une part, d’étudier les différents profils et les complications postopératoires des patients pris en charge dans notre établissement et, d’autre part, d’estimer le coût des explorations par les électrodes de stéréo-électroencéphalographie (SEEG). Méthode Une étude rétrospective a été menée sur les patients opérés entre 2003 et 2008 et suivis par le service de neurologie. Résultats Au total, 104 dossiers médicaux ont été analysés. La moyenne d’âge des patients lors de l’intervention était de 34 ans (extrêmes : 10–70 ans) et la durée moyenne de l’épilepsie avant l’opération de 21 ans (extrêmes : 3–56 ans). Dans 80 % des cas, l’épilepsie rencontrée était de type temporal, principalement traitée par une résection temporale (72 %). Un an après la chirurgie, 60 % des patients étaient libres de crises (classe I d’Engel), 62 % des patients ont pu reprendre leur activité professionnelle ou scolaire et 26 % des patients avaient des complications postopératoires permanentes, mineures et attendues chez la majorité d’entre eux ; aucun décès n’a été rapporté. Discussion/conclusion Deux types de dispositifs médicaux existent pour la prise en charge de la chirurgie de l’épilepsie : la stimulation du nerf vague gauche (8900 € pris en charge en sus des prestations d’hospitalisation) et les électrodes de stéréo-électroencéphalographie (SEEG) en préalable possible à la chirurgie conventionnelle. Ayant débuté dans notre établissement en 2004, l’utilisation des électrodes de SEEG pour la chirurgie de l’épilepsie nécessite en moyenne 11 électrodes (de huit à 15 électrodes) pour un coût moyen de 4284 € par patient. Ces électrodes représentent 58 % du tarif moyen des groupes homogènes de séjours (GHS) retrouvés chez les patients (7418 €). Selon notre étude, la chirurgie de l’épilepsie semble présenter un bénéfice clinique important. Néanmoins, le rapport coût–efficacité de la chirurgie conventionnelle de l’épilepsie post-SEEG doit être réévalué à long terme pour en confirmer l’intérêt. Mots clés Chirurgie de l’épilepsie Épilepsie pharmacorésistante Complications postopératoires Coût–efficacité Électrodes de stéréo-électroencéphalographie Keywords Epilepsy surgery Drug-resistant epilepsy Postoperative complications Cost-effective stereo-electroencephalographic electrodes Introduction Approximately 30% of epilepsies are refractory to current drug treatments. Drug-resistant epilepsy (DRE) is defined as the persistence of epileptic seizures of any kind, sufficiently frequent and/or deleterious, despite the regular use of an antiepileptic drug, properly prescribed a priori, for at least two years, in compliant patients [1] . There are two surgical approaches: left vagus nerve stimulation (VNS) and conventional surgery or classical (palliative or curative). In the conventional surgical approach, one distinguishes the so-called curative or surgical excision (lobectomy, cortectomy, with or without lesionectomy cortectomy) which aims to cure epilepsy, or by removing the epileptogenic tissue and reconstructive surgery (hemispherotomy, callosotomy), proposed in patients not eligible for curative surgery, which aims to reduce the frequency or severity of some attacks, interrupting some connections. For its part, vagal stimulation is an adjunctive treatment of epilepsy based on the electrical stimulation of the vagus nerve, which requires surgical implantation of stimulating material. Medical devices used in the vagal stimulation will be charged extra, apart from hospital benefits. The operability of a patient is defined on the basis of information from the scalp video EEG, neuropsychological testing and brain imaging. Deep intracerebral stereo-electroencephalographic (SEEG) electrodes are used if preliminary investigations fail to demonstrate the topography and extent of the epileptogenic zone. The objectives of this study were to investigate the different profiles of patients cared for with DRE in our institution, to evaluate the postoperative complications in the series and to estimate the average cost of the medical devices used in exploration using SEEG, including that such costs are not inclusive but in addition to hospitalization benefits. Material and methods The survey conducted in the Neurology Department of the institution in all patients (adults and children) supported (conventional surgery or vagus nerve stimulation) for DRE, between January 2003 and April 2008, was completed in April 2009 with the compendium of recent follow-up data at the end of one year. We developed a data collection sheet, which has been validated by the medical team from Neurology. The collection and analysis of survey data were done by the pharmacy resident, after consulting the medical records and information recorded in the medicalisation program information systems (PMSI-MCO). The parameters studied included: age, gender, type of epilepsy, number of crises before and after surgery, drugs, surgical technique used, identified benefits of surgery and postoperative complications observed one year after surgery ( table I ). Of the surgical techniques, diagnostic procedures (implantation of electrodes to record video-stereo-EEG) and therapeutic procedures (lobectomy, cortectomy, hemispherotomy, lesionectomy and vagal stimulation) were differentiated. The benefit of epilepsy surgery was evaluated according to Engel's classification, which quantifies the level of seizure control after surgery as follows: patients become seizure-free (class I), with seizures decreased by more than 90% (class II) or more than 50% (class III) or with little or no change in seizure frequency (class IV) [2] . Cost of SEEG explorations These investigations require the use of several electrodes, which varies from one patient to another depending on the extent of the epileptogenic zone suspected earlier by noninvasive investigations. The average cost of electrodes in patients operated upon after an assessment by the stereo-electroencephalographic method was calculated taking into account the number of electrodes and the number of electrode contacts used for each patient. For the years 2004 to 2008, the rate of homogeneous groups living (GHS) provided by the Department of Medical Information of the establishment has been considered. The average rate of GHS in our institution was estimated by calculating the average cost of GHS in which patients previously identified were located. Results Patients Totally, the medical records of 104 patients were analyzed. The characteristics of surgical patients were: 55 men and 49 women (sex ratio M/F = 1.12) comprising 95 adults and 9 children, with a mean age at intervention of 34 years (range: 10–70 years) and an average time between the onset of seizures and surgical treatment of 21 years (range 3–56 years). (fig. 1) shows the main types of epilepsy encountered in surgical patients. The median age of first seizure was 4 years (range: 8 months and 7 years) in children and 12 years; (range: 6 months and 48 years) in adults. The average monthly number of seizures before surgery was 21 (range: 2 to 60) in children and 9 (range:1 to 60) in adults. Drug treatment Various antiepileptic (AE) treatments used prior to surgery were distributed among 104 patients between 11% for monotherapy, 44% for combination therapy and 45% for multitherapy. Mostly third generation antiepileptic drugs were used in combination therapy: levetiracetam ( n = 45), topiramate ( n = 36), oxcarbamazepine ( n = 34) and lamotrigine ( n = 30). In monotherapy, carbamazepine was the most commonly used drug ( n = 7). Diagnostic procedures with implantation of implantable medical devices Of the 104 patients studied, 18 received an VNS and 86 were operated. Of the 86 patients operated upon, only 23 (27%) underwent an SEEG exploration with an average of 11 electrodes (eight to 15 electrodes) and five to 15 contacts per electrode. Surgical treatment of epilepsy The conventional surgical techniques were applied as follows. Curative techniques Temporal lobectomy (72%, n = 75), which involves the removal of the anterior temporal lobes including the amygdala, the hippocampus and cerebral cortex. It also includes the anterior temporal lobectomy and amygdalohippocampectomy. Extratemporal cortectomy (5%, n = 5), which consists of a resection of the epileptogenic extratemporal cortex. Lesionectomy (4%, n = 4) which involves the removal of a brain lesion (cavernoma, hamartoma, cortical malformation…). Palliative techniques Hemispherotomy (1%, n = 1), which consists of the complete disconnection of the epileptic hemisphere, leaving its blood supply in place. This technique is sometimes curative. Cortical stimulation (1%, n = 1) by inducing a depotentialisation (or depression) in the long-term epileptogenic network by recruiting endogenous inhibitory mechanisms, via the application of a brief low-frequency stimulus. Vagal stimulation Vagal stimulation (17%, n = 18), which principally consists of intermittent stimulation of the vagus nerve from a pacemaker implanted in the left subclavian region. It is a palliative technique. In total, 81% of patients in our institution have received care with curative and palliative techniques in 19%. Benefit of surgery after 1 year According to Engel's classification, a total of 60% ( n = 62/104) of patients who received conventional surgery or vagal stimulation were seizure-free(class I) after surgery ( (fig. 2) ). Drug treatment after surgery/vagal stimulation The results for the antiepileptic drug treatment after surgery showed that 25% of 104 patients were on monotherapy, 32% on combination therapy, 35% on multitherapy and 8% on no antiepileptic treatment. Combination therapy involved 89% ( n = 92/104) of patients prior to surgery/stimulation; therefore it dropped to 67% ( n = 70/104) a year later ( p < 0.001, α = 0.05). Psychosocial effect All types supported combined (surgery or vagal stimulation), 62% of patients were satisfied with the surgery. They were able to resume their work or school, one year after surgery. Side effects of surgery after 1 year Of the 18 patients stimulated, undesirable effects (dysphonia, cough) due to vagal stimulation were recorded in 10 of them (55%). These effects have not been included in the complications of care. They depend on the intensity of stimulation, and vary from one patient to another. Of the 86 patients operated upon by conventional surgery during the study period, 54 patients had transient adverse events (32 headache, eight visual hallucinations, seven mood disorders, four aseptic meningitis and three incision pain) and 22 patients had permanent complications (13 quadrantanopias, seven depressive syndromes and two focal ischemia by arterial spasm). Then 28 patients had many complications whereas 10 patients had no complications during the postoperative period. The postoperative complications are represented according to the surgical technique used in table II . The permanent complication observed was quadrantanopia superior temporal lobe contralateral to surgery, which patients had been warned of, prior to surgery. This minor reduction of the visual field does not usually interfere in daily life. Exceptionally, two patients suffered from focal ischemia by arterial spasm, causing a hemiplegia in one, a hemianopia in the other. Cost of SEEG explorations The presurgical evaluation with SEEG electrodes for epilepsy surgery has not been used until 2004, among adults. The average cost of SEEG electrodes was € 4,284 per patient. The cost of SEEG exploration was included in the GHS label corresponding to “craniotomy without any trauma, age over 17 years without comorbidities” for which costs differed from year to year (€ 8880 in 2004; € 7396 in 2005; € 6695 in 2006; € 7072 in 2007 and € 6844 in 2008). Taking into account the year of SEEG exploration per patient, the average tariff of GHS was € 7418. The SEEG electrodes represented 58% of the average costs of GHS found in surgical patients. Discussion Our study is retrospective, therefore it was not always easy to find information in the patients’ medical records. However, first, this study has explored the different profiles and postoperative complications of the patients cared for DRE in our institution and then, it compared these results with those found in the literature. Secondly, the proportion of the cost of medical devices used in SEEG exploration in our institution was estimated over the considered rate of reimbursement of Diagnosis-related groups (DRG). According to this study, our series of patients comprised mainly adults with a sex ratio of nearly 1. The majority of surgical patients (80%) suffering from temporal lobe epilepsy were mostly treated by temporal resection (72%), which concurs with the literature data [3,4] . One year after surgical intervention (surgery or stimulation), the benefit was clear as the results of this analysis showed that 60% ( n = 62/104) of patients were seizure-free (class I). This result is comparable to that observed (between 48 and 84%) in several studies in the literature [5–8] . Complications following conventional surgery remain largely transient (65%, n = 54/86) and no deaths were reported in perioperative or following complications from surgery. Note that the percentage of patients (14%, n = 13/86) who experienced the quadrantanopia complication, the main permanent complication, was much lower than that found in the literature (50%) [9] . This was due possibly to the surgeon's operating technique, which enabled him to spare the optic radiations. However, according to published studies, the mortality rate of patients free of seizures after surgery is close to that of the general population [10,11] . Indeed, epilepsy surgery cancels the risk of excess mortality due to unexplained sudden death in epilepsy (Sudden Unexpected Death in Epilepsy [SUDEP]) found to occur in drug-resistant patients [12] . Moreover, even if the presurgical evaluation with intracerebral electrodes in SEEG is sometimes necessary, it is reserved for cases where noninvasive explorations were inconclusive by assessing the risk/benefit ratio of electrode implantation. The cost of SEEG electrodes is important because it represents a large proportion (58%) of the average amount of GHS despite other payable benefits in the GHS. Finally, the benefit of surgery/stimulation from this study is evident as enhancing the quality of life for the majority of patients operated/stimulated, with 62% of them having returned to work or school after the operation. Besides, a prospective French study evaluated the cost-effectiveness of epilepsy surgery by comparing long-term changes before and after surgery between both groups of patients (medical arm and surgical arm), surgery becomes cost-effective between 7 and 8 years after surgery [13] . Conclusion The epilepsy surgery is the only hope to cure patients with drug-resistant epilepsy. Patients operated upon at our institution include mainly adults with equal percentage of men and women. However, the survey confirms that in our institution, as in other published studies, temporal resection is mainly used. In our series, the high percentage of patients free of seizures after surgery/stimulation indicates that they have been effective in patients previously selected. Sometimes, this may require the use of SEEG electrodes in cases where noninvasive explorations failed to clearly identify the epileptogenic zone. These explorations have an obvious cost and are not supported, in addition to hospital benefits. Further, the postoperative complications are mostly transient, and quality of life of the patients was much improved in our series. In our study, epilepsy surgery appears to be of significant clinical benefit. However, the cost-effectiveness of conventional surgery for epilepsy post-SEEG needs to be reevaluated to confirm the long-term interest. Conflict of interest statement None. References [1] P. Jallon Epidemiology of drug-resistant partial epilepsy Rev Neurol 160 1 2004 22 30 [2] J. Engel A Proposed Diagnostic Scheme for People with epileptic seizures and with epilepsy: report of the ILAE Task Force on classification and terminology Epilepsia 42 2001 796 803 [3] H. Silander S. Blom K. Malmgren I. Rosén P. Uvebrant Surgical Treatment for epilepsy: a retrospective Swedish Multicenter Study Neurol Act Scand 95 6 1997 321 330 [4] B. Devaux F. Chassoux M. Guenot Epilepsy surgery in France: evaluation of the activity Neurosurgery 54 3 2008 453 465 [5] M. Al-Kaylani P. Konrad B. Lazenby B. Blumenkopf B. Abou-Khalil Seizure freedom off antiepileptic drug temporal lobe epilepsy after surgery Seizure 16 2007 95 98 [6] A.A. Cohen-Gadol B.G. Wilhelmi F. Collingnon Long-term outcome of epilepsy surgery among 399 patients with non-lesional mesial temporal seizure foci including lobe sclerosis J Neurosurgery 104 2006 513 524 [7] J. Engel Jr. S. Wiebe J. French Practice parameter: temporal lobe and localized neocortical resection for epilepsy Epilepsia 44 2003 741 751 [8] S.C. Schachter Vagus nerve stimulation therapy summary: five years after FDA approval Neurology 59 4 2002 15 20 [9] S. Chabardès L. Minotti S. Hamelin Disconnecting the temporal lobe in drug-resistant temporal lobe epilepsy: techniques, complications and results Neurosurgery 54 3 2008 297 302 [10] R. Love Two hit hypothesis for temporal lobe epilepsy Lancet Neurol 4 8 2005 458 [11] A.M. Siegel G.D. Cascino F.B. Meyer Intractable partial epilepsy for reoperation: predictive factors and outcome operative Neurology 58 3 2002 96 [12] P. Ryvlin A. Montavont Does epilepsy surgery reduces the mortality of it drug-resistant partial epilepsy? Neurosurgery 54 3 2008 282 286 [13] M. Picot D. Neveu P. Kahane Medical and economic evaluation of surgery for drug-resistant partial epilepsy in adults. Cost-effectiveness: preliminary results Neurol Rev 160 2004 354 367