Complications of epilepsy surgery: A systematic review of focal surgical resections and invasive EEG monitoring

Transcription

1 FULL-LENGTH ORIGINAL RESEARCH Complications of epilepsy surgery: A systematic review of focal surgical resections and invasive EEG monitoring *Walter J. Hader, Jose Tellez-Zenteno, * Amy Metcalfe, Lisbeth Hernandez-Ronquillo, * Samuel Wiebe, Churl-Su Kwon, and * Nathalie Jette *Department of Clinical Neurosciences, University of Calgary, Calgary, Alberta, Canada; Division of Neurology, University of Saskatchewan, Saskatoon, Saskatchewan, Canada; Department of Community Health Sciences, University of Calgary, Calgary, Alberta, Canada; Hotchkiss Brain Institute and Institute for Public Health, Calgary, Alberta, Canada; and Department of Neurosurgery, Massachusetts General Hospital, Boston, Massachusetts, U.S.A. SUMMARY Purpose: Underutilization of epilepsy surgery remains a major problem and is in part due to physicians misconceptions about the risks associated with epilepsy surgery. The purpose of this study was to systematically review the literature on complications of focal epilepsy surgery. Methods: A literature search was conducted using PubMed and Embase to identify studies examining epilepsy surgery complications. Abstract and full text review, along with data extraction, was done in duplicate. Minor medical and neurologic complications were defined as those that resolved completely within 3 months of surgery, whereas major complications persisted beyond that time frame. Descriptive statistics were used to report complication proportions. Key Findings: Invasive monitoring: Minor complications were reported in 7.7% of patients, whereas major complications were reported in only 0.6% of patients undergoing invasive monitoring. Resective surgery: Minor and major medical complications were reported in 5.1% and 1.5% of patients respectively, most common being cerebrospinal fluid (CSF) leak. Minor neurologic complications occurred in 10.9% of patients and were twice as frequent in children (11.2% vs. 5.5%). Minor visual field defects were most common (12.9%). Major neurologic complications were noted in 4.7% of patients, with the most common being major visual field defects (2.1% overall). Perioperative mortality was uncommon after epilepsy surgery, occurring in only 0.4% of temporal lobe patients (1.2% extratemporal). Significance: The majority of complications after epilepsy surgery are minor or temporary as they tend to resolve completely. Major permanent neurologic complications remain uncommon. Mortality as a result of epilepsy surgery in the modern era is rare. KEY WORDS: Surgical risk, Epilepsy, Mortality, Neurologic deficits. Epilepsy surgery is a widely accepted therapeutic option for selected patients with drug-resistant focal epilepsy. It has been found in two randomized controlled trials to be safe and more effective than medical treatment alone in patients with temporal lobe epilepsy (Wiebe et al., 2001; Engel et al., 2012). Based on this single randomized trial and a systematic review of outcomes of numerous retrospective reports on temporal lobe resections for epilepsy, the Quality Standards Subcommittee of the American Academy of Neurology (AAN) recommended that referral to an epilepsy surgery center should be strongly considered for appropriate candidates (Engel et al., 2003). Accepted February 24, 2013; Early View publication April 3, Address correspondence to Walter J. Hader, Division of Neurosurgery, Department of Clinical Neurosciences, Foothills Medical Centre, 12th floor, th Street NW, Calgary, AB T2N 2T9, Canada. walter.hader@albertahealthservices.ca Wiley Periodicals, Inc International League Against Epilepsy Despite the establishment of clear guidelines recommending referral of appropriate candidates to epilepsy surgery centers (Engel et al., 2003), underutilization of epilepsy surgery (de Flon et al., 2010; Haneef et al., 2010) remains problematic. Reasons that have been cited for the lack of referrals of appropriate candidates for epilepsy surgery include uncertainty in referring physicians about when a patient has become drug resistant and a lack of knowledge regarding the efficacy and safety of epilepsy surgery (Sirven, 2010). Attitudes toward epilepsy surgery in patients with medically intractable epilepsy are often negative, and epilepsy surgery is often considered by many to be last ditch and experimental and not performed routinely enough to consider (Swarztrauber et al., 2003). Potential surgical candidates with intractable seizures continue to be enrolled in new antiepileptic drug (AED) trials where it has been shown that patients are not informed about the evidence for surgical effectiveness (Janszky et al., 2010). 840

2 841 Complications after Epilepsy Surgery It has been proposed that an inherent bias against brain surgery may exist (Sirven, 2010) due to concerns about the risks of epilepsy surgery. In a survey of referral patterns of patients who underwent epilepsy surgery (Benbadis et al., 2003), 14% of patients had been advised by their primary neurologists not to have surgery, primarily because of its potential neurologic complications. Self-referral of these patients for epilepsy surgery evaluation was necessary in order for them to eventually undergo appropriate surgical intervention. A survey of Swedish neurologists involved in epilepsy care revealed that almost one third had never referred a patient for an epilepsy surgery evaluation (Kumlien & Mattsson, 2010). In a similar study of referring neurologists, one third of those who had referred patients reported that serious complications had resulted from surgery in their patients (Hakimi et al., 2008). It is possible that misconceptions about the perceived risks associated with epilepsy surgery act as a deterrent for referring physicians and patients alike, and may significantly contribute to the underutilization of epilepsy surgery evaluations and epilepsy surgery. Accurate characterization of the risks of epilepsy surgery in addition to the benefits of improved seizure control are essential to foster promotion of epilepsy surgery, not only as an effective but safe alternative treatment modality for patients with intractable epilepsy. The purpose of this study was to systematically review the epilepsy surgery literature to determine the frequency and nature of specific reported complications associated with the most common epilepsy procedures performed in patients with focal epilepsy. Both resective procedures with the goal of removal of the primary epileptogenic focus and cessation of seizures, and long-term invasive monitoring procedures aimed at defining the epileptic focus, were considered in this analysis. Methods Literature search and study selection A literature search to identify all epilepsy surgery series was conducted using PubMed, Embase, and the Cochrane database until June 2008 as part of a larger RAND appropriateness and necessity study of epilepsy surgery. The full search strategy (June 2008) is outlined in Appendix 1. Inclusion criteria for the initial search were partial epilepsy, focal resections (e.g., lesionectomy, lobectomy, corticectomy, selective amygdalo-hippocampectomy), English language, and sample size 20. Only studies reporting seizure outcomes were included. Complications had to be clearly defined and assessed as an outcome to be included in the study. Initial exclusion criteria were hypothalamic hamartoma, multiple subpial transections, hemispherectomies, callosotomies, other palliative procedures, stimulation studies, neonates, and abstracts not published as full papers. Additional exclusion criteria specific to this systematic review were studies published before 1990 (pre-mri imaging era) and epilepsy surgery reports involving predominance of lesion resection alone such as arteriovenous malformations, cavernomas, and tumors. All abstracts and full-text articles were reviewed by two physicians (one epileptologist, one epilepsy neurosurgeon) and disagreement was resolved by discussion. Hand searching of included articles to ensure no studies were missed was also performed. Data extraction and statistical analysis Data extraction for all included studies was performed in duplicate by two independent reviewers using a standardized data abstraction form. Study location, age of the study population, surgical site, and complications (see below) were extracted. Reported proportions of complications were derived by dividing the total number of patients with a complication (from all studies) by the total number of patients in all studies that reported on the complication of interest. Wilson s method was used to calculate a 95% confidence interval around these estimates. All analysis were done using Stata IC version 10 (Statacorp LP, College Station, Texas, U.S.A.). Complications: definitions Medical complications Minor medical complications referred to cerebrospinal fluid (CSF) leak, intracranial/extracranial infection (superficial or deep), aseptic meningitis, deep vein thrombosis/ pulmonary embolus (DVT/PE), pneumonia, intracranial hematomas, metabolic disturbances (Table S1). Electrode complications referred to electrode placement or functional issues. Major medical complications included hydrocephalus and deep infections, such as intracerebral and epidural abscesses, requiring intervention for drainage and or bone flap removal with eventual cranioplasty. Neurologic complications Minor neurologic or psychiatric complications were considered temporary and included those that resolved completely within 3 months of the specific surgical procedure completed (Table S1). Major neurologic complications persisted beyond that time frame. Minor or major neurologic deficits involving cranial nerves, dysphasia, memory disturbances (subjective), hemiparesis (all or part of a limb), and psychiatric complications were recorded as well as status epilepticus and death as a direct result of the procedure. Visual field defects as a complication were separated into minor field (quadrant or less) and major field (hemianopia). Results Literature review The initial search identified 5,061 abstracts, with 162 reviewed in full text (Fig. S1). Ultimately, 76 articles

3 842 W. J. Hader et al. met all eligibility criteria for this review (Data S1). Reasons for exclusions included 33 review papers, 25 papers with no details of complications, 10 papers with predominance of lesion removal alone, 6 papers with <20 patients, 5 with duplicate data, and 7 with the majority of surgery completed in the pre-imaging era. All included articles were surgical series. Twenty articles provided data on children, 20 provided data on adults, 32 provided data on children and adults, and 4 articles did not specify the age range of included patients. Forty-one articles provided data on temporal resections, 14 on extratemporal resections, and 21 on both temporal and extratemporal resections. A total of 14 articles that reported complications of invasive monitoring either alone or as part of surgical series involving resections were included for analysis. Invasive monitoring Various methods of invasive monitoring were described including subdural strip and depth electrode implantation, commonly completed by burr hole or twist drill, and subdural grid implantation, by way of a craniotomy. Complications were reported as minor or major only and generally not separated into medical or neurologic. The reported frequency of minor complications associated with invasive monitoring was 7.7%, with CSF leak being the most common complication followed by electrode complications with a frequency of 1.8%. The reported frequency of a major or permanent complication of invasive monitoring was 0.6%. The overall frequency of minor complications associated with invasive monitoring was higher in the pediatric population (23.5% vs. 4.1%) where subdural grid implantation via craniotomy was more commonly utilized in the process of invasive monitoring investigations. Permanent complications of invasive monitoring in adults, however, were quite low (0.3%) where subdural and depth electrode placement commonly occurred through twist drill or burr holes. Resective surgery medical complications Overall, minor medical complications were reported in 5.1% of patients, whereas major medical complications were identified in 1.5%. The most common minor medical complication was CSF leak noted in 8.5% of patients overall followed by aseptic meningitis 3.6%, bacterial infection 3.0%, and intracranial hematomas 2.5%. CSF leaks (14.3% vs. 4.3%), aseptic meningitis (5.8% vs. 3.4%), intracranial hematomas (4.0% vs. 2.0%), and infections (3.9% vs. 1.9%) were all more commonly reported in children than in adults (Table 1). Location of resection (Table 1) did not alter the frequency of most medical complications. However, patients were twice as likely to develop an intracranial hematoma after extratemporal surgery compared to temporal lobe surgery alone (3.1% vs. 1.5%). Additional medical complications were identified 1% or less of the time including DVT/PE, extracranial infections, and hydrocephalus, although the frequency of hydrocephalus in children was higher than in adults (2.4% vs. 1.3%). Resective surgery neurologic complications Overall, minor neurologic complications occurred in 10.9% of patients, whereas major complications were identified in 4.7% of patients. Minor neurologic complications were twice as frequent in children (11.2% vs. 5.5%; Table 2), whereas major neurologic complications were more common after extratemporal resections (6.5% vs. 4.1%; Table 2) and in the pediatric age group (5.1% vs. 3.3%). The most common neurologic complication after resective epilepsy surgery was a minor visual field deficit (one quadrant or less) seen in 12.9% of patients, the majority of which were not evident to the patient. These were more than twice as likely to be identified in patients after temporal lobe resections compared to extratemporal resections (17.9% vs. 7.2%). Major visual field deficits (hemianopia), identified in 2.1% of patients, occurred at approximately equal proportions in temporal and extratemporal resections. Minor and major cranial nerve deficits were identified in 2.1% and 0.4% of patients, respectively. Minor or temporary dysphasia was seen in 3.7% of patients and major dysphasia was identified in 0.8%, which was similar regardless of resection location. Minor or temporary hemiparesis, including all or part of a limb, was identified in 3.3% of all patients; however, it was much more commonly associated with extratemporal compared to temporal resections (7.9% vs. 1.8%) and after resections in the pediatric age group (6.0% vs. 1.9%). Major or permanent hemiparesis was identified in 1.8% of patients overall, similar in frequency after temporal or extratemporal lobe resections (1.8% vs. 2.3%) but more commonly associated with resection in the pediatric age group (3.3% vs. 1.0%). Minor psychiatric complications were reported in 5.5% of patients, more commonly after temporal than after extratemporal resections (5.8% vs. 1.6%). Major psychiatric complications were reported in 1.9% of patients after temporal lobe operations only. Status epilepticus was reported rarely, after either temporal or extratemporal resections (0.2%, 2 of 941 patients). Similarly, perioperative mortality was reported in only 0.6% patients (1.2% extratemporal vs. 0.4% temporal). Discussion A variety of medical and neurologic complications may occur after epilepsy surgery, associated with both invasive monitoring for localization of the epilepsy focus or after definitive resective procedures for removal of the epileptogenic focus. Apprehension to consider epilepsy surgery by both patients and referring physicians as a result of the perceived risks associated with surgery may be a factor in deterring appropriate candidates from pursuing epilepsy

4 843 Complications after Epilepsy Surgery Table 1. Reported frequency of medical complications: overall, by age group and location of resection Type of complication Patient group Total number of patients Number reporting complication Frequency (%, 95% CI) Minor medical complications Total 4, ( ) Pediatric only ( ) Adults only 1, ( ) Temporal 2, ( ) Extratemporal ( ) Major medical complications Total 2, ( ) Pediatric only ( ) Adults only ( ) Temporal 1, ( ) Extratemporal ( ) Infection Total 4, ( ) Pediatric only ( ) Adults only 1, ( ) Temporal 2, ( ) Extratemporal ( ) Aseptic meningitis Total ( ) Pediatric only ( ) Adults only ( ) Temporal ( ) Extratemporal ( ) Deep vein thrombosis/pulmonary embolus Total 1, ( ) Pediatric only ( ) Adults only ( ) Temporal 1, ( ) Extratemporal ( ) Intracranial hematoma Total 2, ( ) Pediatric only ( ) Adults only ( ) Temporal 1, ( ) Extratemporal ( ) Pneumonia Total ( ) Pediatric only ( ) Adults only ( ) Temporal ( ) Extratemporal ( ) Cerebrospinal fluid leak Total ( ) Pediatric only ( ) Adults only ( ) Temporal ( ) Extratemporal ( ) Hydrocephalus Total 1, ( ) Pediatric only ( ) Adults only ( ) Temporal ( ) Extratemporal ( ) surgery evaluation. Our study provides a detailed understanding of these risks for patients and physicians to ensure an informed decision can be made about epilepsy surgery, and to remove the fallacy that this type of surgery is very dangerous. Failure to offer epilepsy surgery for appropriately selected patients may withhold a treatment capable of providing seizure freedom, improved quality-of-life, and a reduction in the likelihood of premature mortality associated with ongoing intractable epilepsy for the majority of patients (Wiebe et al., 2001). Our systematic review demonstrates that the majority of complications reported after epilepsy surgery are considered minor, as the effects are temporary and resolution is complete. The reported frequency and nature of specific complications of epilepsy surgery vary depending on the type of surgery (invasive vs. resective), the location of resection (temporal vs. extratemporal), and the age of the patient (pediatric or adult). This is most evident with invasive monitoring in which self-limited minor complications were far more prevalent than permanent complications, which occurred in <1% of patients. Pediatric patients were reported to experience a greater number of minor complications associated with invasive monitoring, which may be due in part because the preferred method of monitoring was

5 844 W. J. Hader et al. Table 2. Reported frequency of neurologic complications: overall, by age group and location of resection Type of complication Patient groups Total number of patients Number reporting complication Frequency (%, 95% CI) Minor neurologic complications Total 5, ( ) Pediatric only 1, ( ) Adults only 1, ( ) Temporal 3, ( ) Extratemporal ( ) Major neurologic complications Total 5, ( ) Pediatric only ( ) Adults only 1, ( ) Temporal 3, ( ) Extratemporal 1, ( ) Minor cranial nerve Total 2, ( ) Pediatric only ( ) Adults only ( ) Temporal 2, ( ) Extratemporal ( ) Major cranial nerve Total 1, ( ) Pediatric only ( ) Adults only (0 2.5) Temporal 1, ( ) Extratemporal ( ) Minor field Total 1, ( ) Pediatric only ( ) Adults only ( ) Temporal 1, ( ) Extratemporal ( ) Major field Total 3, ( ) Pediatric only ( ) Adults only ( ) Temporal 2, ( ) Extratemporal ( ) Minor dysphasia Total 4, ( ) Pediatric only ( ) Adults only 1, ( ) Temporal 2, ( ) Extratemporal ( ) Major dysphasia Total 2, ( ) Pediatric only ( ) Adults only ( ) Temporal 1, ( ) Extratemporal ( ) Minor memory Total 1, ( ) Pediatric only 0 Adults only ( ) Temporal ( ) Extratemporal ( ) Major memory Total 1, ( ) Pediatric only 0 Adults only ( ) Temporal 1, ( ) Extratemporal ( ) Minor hemiparesis Total 4, ( ) Pediatric only ( ) Adults only ( ) Temporal 2, ( ) Extratemporal ( ) Major hemiparesis Total 4, ( ) Pediatric only ( ) Adults only 1, ( ) Temporal 2, ( ) Continued

6 845 Complications after Epilepsy Surgery Table 2. Continued. Type of complication Patient groups Total number of patients Number reporting complication Frequency (%, 95% CI) Extratemporal ( ) Minor psychiatric Total 1, ( ) Temporal 1, ( ) Extratemporal ( ) Major psychiatric Total ( ) Temporal ( ) Extratemporal 0 0 through craniotomy and subdural grid implantation as opposed to subdural strip or depth electrode monitoring completed through burr holes, which is more frequently used in adults. Assessing the frequency of permanent complications related to invasive monitoring alone in children was difficult because most go on to have resections immediately after implantation, and subsequent complications may be difficult to attribute to the invasive procedure itself as opposed to the cortical resection (Onal et al., 2003). It has been reported that an increased reluctance to refer patients for evaluation of epilepsy surgery exists when there is no lesion on MRI (Berg, 2011). This may be in part due to the greater need for invasive monitoring and perceived risks required with an additional surgical procedure in this population. Although patients with normal MRIs fare less well after epilepsy surgery than those with obvious epileptic lesions, the risks of invasive monitoring remains low and seizure freedom after surgery is still possible in up to one half of patients after nonlesional temporal lobe surgery (Tellez-Zenteno et al., 2005), whereas only 8% of patients with ongoing medical management become seizure free (Wiebe et al., 2001). The most feared and important complications that result from epilepsy surgery are the neurologic complications affecting visual, motor, speech, and memory systems, which do not resolve and therefore have the potential to affect longterm quality-of-life. Postoperative neurologic deficits have been reported in a systematic review of social outcomes to be highly associated with dissatisfaction after epilepsy surgery (Macrodimitris et al., 2011), even in patients who are seizure free after surgery. Permanent neurologic complications of definitive resective epilepsy surgical procedures are often closely associated with the location of specific resective procedures. Temporary or minor hemiparesis was far more frequently reported after extratemporal resections and in the pediatric age group, where a greater frequency of epilepsy resections occur outside of the temporal lobe. Permanent or major hemiparesis and major visual field deficits (hemianopia), as well as certain medical complications including hematomas and aseptic meningitis were more common in children and may reflect the necessity of large extratemporal multilobar resections required in pediatric patients with catastrophic epilepsies. Minor field, cranial nerve deficits, subjective memory disturbances, and psychiatric complications were more common after temporal lobe resections. Mortality after epilepsy surgery was rarely reported in adults, but was more common in the pediatric population and may be attributable to the large multilobar resective procedures performed in the very young age group. Complication proportions for the most common surgically treated epilepsy syndrome in adults, temporal lobe epilepsy, are low, whereas the majority of patients undergoing temporal lobe resection can attain freedom from seizures and associated significant improvement in quality-of-life (Wiebe et al., 2001). The most frequently reported complications were expected minor field deficits, confined to the superior quadrant alone, and in most cases not reported by the patient. Permanent complications resulting in either contralateral weakness or major field loss in the form of a homonymous hemianopsia, capable of affecting one s ability to drive after successful epilepsy surgery, rarely occur after resections involving the temporal lobe alone. These results are consistent with the reported frequency of in hospital morbidity following temporal lobe surgery determined from a population-based analysis from 1998 to 2003 across multiple American centers, which revealed overall postoperative morbidity, medical, and neurologic of 8% with no mortality (McClelland et al., 2011). Medical complications, similarly reported to be low after temporal lobe surgery, is due in part to the fact that the majority of patients undergoing temporal lobe surgery have one or less comorbidities that have been shown to be associated with reduced perioperative morbidity (McClelland et al., 2011). Although this systematic review represents the most comprehensive assessment of reported frequencies of complications after epilepsy surgery, several limitations exist. The review was limited to English language alone and most reports are retrospective case series and often reflect the experience of a single surgeon at large volume centers. Although large unselected series with multiple surgeons were included which may reflect a more generalizable experience, complication proportions are also likely dependant on the skills of the surgeon and epilepsy center. Methods of reporting complications varied considerably and reported frequencies in this review were determined from studies of varying sample sizes across multiple age groups and encompassing varying locations of resections. A variety of definitions of whether a complication is minor or major, transient

7 846 W. J. Hader et al. or permanent, some even considering complications to be permanent only after 1 year post surgery, adds to confusion for interpretation of different studies. Deep venous thrombosis/pulmonary embolus while reported as a minor complication in the majority of studies, may justifiably be considered a more serious complication because of its lifethreatening potential. Few reports of complications were reported prospectively in a standardized manner by independent observers and as a result, complication proportions may vary from that determined by this systematic review. In a recent report of a prospective trial of early temporal lobe surgery with only 15 surgical patients, evidence of ischemia was identified on postsurgical MRIs in three patients (one clinical), and one patient required shunting for postsurgical hydrocephalus (Engel et al., 2012) suggesting possible higher proportions of complications may occur than this systematic review when evaluated in a standardized manner. In addition, a variety of approaches to resections involving the temporal lobe structures alone are performed. These include selective amygdalohippocampectomy from transsylvian, transcortical, and subtemporal approaches, and anterior temporal lobectomy with varying extent of neocortical resections and performed under general or local anesthesia particularly on the dominant side, which may make direct comparisons more difficult. Actual risks of surgical complications for adult patients undergoing epilepsy surgery, whose ages range from as young as the late second to the seventh decade, may vary depending on the age. Studies of epilepsy surgery patients at the older extreme of ages routinely were excluded from this study because of small numbers. We have included studies only after 1990, when the use of structural MRI in preoperative investigations of epilepsy surgical patients became more widespread to reflect the practice that exists today in determining surgical candidacy. A comparison of complications of a more common neurosurgical procedure, carotid endarterectomy for high-grade carotid stenosis, previously having been evaluated in numerous large randomized trials, resulted in permanent neurologic complications of stroke or death in <6% of patients. Despite significant results favoring carotid endarterectomy over medical treatment alone, recommendations for surgical treatment of carotid stenosis was felt to be advisable only when acceptable rates of neurologic complications of the performing surgeons are low, that is, that being <6%. This is comparable to the overall risk of a major complication from epilepsy surgery irrespective of location of the resection or the age of the patient and greater than the 4% risk of major complication associated with temporal lobe surgery in adults. Cumulative results of several large randomized trials have demonstrated that the number of patients needed to be operated on to prevent one disabling stroke or death (NNT) was 15 in patients with high grade stenosis, whereas only two patients are needed to be treated with temporal lobe epilepsy surgery to render one patient free of disabling seizures at 1 year (Wiebe et al., 2001). When the risks and outcome of epilepsy surgery are considered with other neurosurgical procedures such as carotid endarterectomy, the low overall risk of epilepsy surgery with more favorable surgical outcome is clear. Conclusions A variety of complications may occur after epilepsy surgery. However, the majority of these complications result in only temporary impairment, as the effects tend to completely resolve over time. Permanent neurologic complications associated with the most common epilepsy surgical procedure, temporal lobe resection, are low. Variations in frequency of certain neurologic complications based on location of resection and the age of patients are reported. The results of this study reinforce the safety of the most common epilepsy surgical procedures and should alleviate some of the fears and misconceptions regarding epilepsy surgery. It is hoped that these findings will increase the likelihood of appropriate referrals to specialized epilepsy centers for surgical evaluation. Acknowledgments This study was in part supported by a Clinician Scientist award to N. Jette from the American Epilepsy Society (AES) and the Milken Family Foundation and operating grants and or funds from Alberta Innovates Health Solutions (AI-HS), Alberta Health Services, the Canadian Institutes of Health Research (CIHR), the Hotchkiss Brain Institute, and the University of Calgary. N. Jette holds a salary award from AI-HS and a Canada Research Chair in Neurological Health Services Research. S. Wiebe holds the Hopewell Professorship of Clinical Neurosciences Research at the University of Calgary. JF Tellez-Zenteno holds a salary award from the Royal University Hospital Foundation in Saskatoon trough the Mudjadik Thyssen Mining Professorship in Neurosciences and also receives operating funds from the University of Saskatchewan. A. Metcalfe held a doctoral award from CIHR in Genetics (Ethics, Law, and Society) and CIHR strategic training grant studentships in Maternal Fetal Newborn Health and Genetics, Child Development, and Health during this study. Disclosure None of the authors have any conflict of interest to disclose. We confirm that we have read the Journal s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. References Benbadis SR, Heriaud L, Tatum WO, Vale FL. (2003) Epilepsy surgery, delays and referral patterns-are all your epilepsy patients controlled? Seizure 12: Berg AT. (2011) Epilepsy: efficacy of epilepsy surgery: what are the questions today? Nat Rev Neurol 7: de Flon P, Kumlien E, Reuterwall C, Mattsson P. (2010) Empirical evidence of underutilization of referrals for epilepsy surgery evaluation. Eur J Neurol 17: Engel J Jr, Wiebe S, French J, Sperling M, Williamson P, Spencer D, Gumnit R, Zahn C, Westbrook E, Enos B. (2003) Practice parameter: temporal lobe and localized neocortical resections for epilepsy. Epilepsia 44:

8 847 Complications after Epilepsy Surgery Engel J Jr, McDermott MP, Wiebe S, Langfitt JT, Stern JM, Dewar S, Sperling MR, Gardiner I, Erba G, Fried I, Jacobs M, Vinters HV, Mintzer S, Kieburtz K. (2012) Early surgical therapy for drugresistant temporal lobe epilepsy: a randomized trial. JAMA 307: Hakimi AS, Spanaki MV, Schuh LA, Smith BJ, Schultz L. (2008) A survey of neurologists views on epilepsy surgery and medically refractory epilepsy. Epilepsy Behav 13: Haneef Z, Stern J, Dewar S, Engel J Jr. (2010) Referral pattern for epilepsy surgery after evidence-based recommendations: a retrospective study. Neurology 75: Janszky J, Kovacs N, Gyimesi C, Fogarasi A, Doczi T, Wiebe S. (2010) Epilepsy surgery, antiepileptic drug trials, and the role of evidence. Epilepsia 51: Kumlien E, Mattsson P. (2010) Attitudes towards epilepsy surgery: a nationwide survey among Swedish neurologists. Seizure 19: Macrodimitris S, Sherman EM, Williams TS, Bigras C, Wiebe S. (2011) Measuring patient satisfaction following epilepsy surgery. Epilepsia 52: McClelland S III, Guo H, Okuyemi KS. (2011) Population-based analysis of morbidity and mortality following surgery for intractable temporal lobe epilepsy in the United States. Arch Neurol 68: Onal C, Otsubo H, Araki T, Chitoku S, Ochi A, Weiss S, Elliott I, Snead OC III, Rutka JT, Logan W. (2003) Complications of invasive subdural grid monitoring in children with epilepsy. J Neurosurg 98: Sirven JI. (2010) The silent gap between epilepsy surgery evaluations and clinical practice guidelines. Eur J Neurol 17: Swarztrauber K, Dewar S, Engel J Jr. (2003) Patient attitudes about treatments for intractable epilepsy. Epilepsy Behav 4: Tellez-Zenteno JF, Dhar R, Wiebe S. (2005) Long-term seizure outcomes following epilepsy surgery: a systematic review and meta-analysis. Brain 128: Wiebe S, Blume WT, Girvin JP, Eliasziw M. (2001) A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 345: Appendix 1. Search Strategy *General search refers to partial epilepsy and epilepsy surgery search strategy (aimed particularly at identifying all surgical series). The review search was broader and included all review articles as well (rather than only original research articles). 1 epilep*.mp. 2 (surger* or surgi*).mp. [mp=title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer name] 3 (epilepsy and surgery).mp. [mp=title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer name] 4 (incidence or mortality or follow-up studies or prognosis or prognos* or predict* or course or outcome or psychology or quality of life or memory or survival analysis or seizure* or utilization or cost or efficacy or complications or effectiveness or sudep).mp. [mp=title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer name] 5 (1 and 2) or and 5 7 (randomized controlled trial or random* or (double and blind) or placebo or drug therapy or therapeutic or cohort studies or risk or (odds and ratio) or (relative and risk) or case control or case control studies or clinical trial or random allocation or case series or decision analysis or economic).mp. [mp=title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer name] 8 6 and 7 limit 8 to (human and English) Supporting Information Additional Supporting Information may be found in the online version of this article: Figure S1. Systematic literature review search results. Table S1. Complications of epilepsy surgery: medical and neurologic. Data S1. Systematic review references.