The SUDEP Risk Inventory: Association with postictal generalized EEG suppression

Epilepsy Research 117 (2015) 82–84

Contents lists available at www.sciencedirect.com

Epilepsy Research journal homepage: www.elsevier.com/locate/epilepsyres

Short communication

The SUDEP Risk Inventory: Association with postictal generalized EEG suppression Brian D. Moseley a,∗ , Christopher M. DeGiorgio b,1 a b

Department of Neurology and Rehabilitation Medicine, University of Cincinnati, Cincinnati, OH, USA Department of Neurology, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA

a r t i c l e

i n f o

Article history: Received 23 June 2015 Received in revised form 18 August 2015 Accepted 7 September 2015 Available online 9 September 2015 Keywords: Sudden unexpected death in epilepsy SUDEP Risk Factor Inventory PGES Epilepsy

a b s t r a c t To help identify patients at greatest risk for sudden unexpected death in epilepsy (SUDEP), screening inventories like the SUDEP-7 Inventory can be useful. In this study, we examined the strength of association between this inventory’s risk factors and postictal generalized EEG suppression (PGES), a biomarker of SUDEP risk. We reanalyzed data from an epilepsy monitoring unit study of 37 children. We performed a 2 by 2 contingency table analysis to determine the association between “yes” responses on the inventory questions and PGES following >=1 seizure. Having a history of >3 generalized tonic–clonic seizures (GTCS) in the past year had the strongest association with PGES (Pearson chi-square p < 0.001, Cramer’s V = 0.75). Having >=1 GTCS in the past year was also strongly associated with PGES (Pearson chi-square p < 0.001, Cramer’s V = 0.636). Histories of >50 seizures of any type/month (Pearson chi-square p = 0.14, Cramer’s V = 0.241) and intellectual disability (Pearson chi-square p = 0.04, Cramer’s V = 0.337) were not as robustly associated with PGES. Current use of >=3 AEDs had the weakest association with PGES (Pearson chi-square p = 0.66, Cramer’s V = 0.072). Given that all study patients had >=1 seizure per year and epilepsy durations <30 years, the strength of association with these questions and PGES could not be analyzed. © 2015 Elsevier B.V. All rights reserved.

1. Introduction One of the most dreaded consequences of epileptic seizures is sudden unexpected death in epilepsy (SUDEP). This phenomenon is defined as the sudden and unexpected, non-traumatic and nondrowning death in a patient with epileptic seizures (Nashef et al., 2012). The estimated incidence of SUDEP for all people with epilepsy is 0.35–1.8 per 1000 patient-years (Ficker et al., 1998; Hughes, 2009). However, the prevalence of SUDEP in drug resistant epilepsy is much higher, ranging from 3 to 9/1000 patient-years (Tomson et al., 2005). Despite almost 90% of patients wishing to be informed about SUDEP, the communication of such risk by the treating neurologist is low (Xu et al., 2015). In a survey of US and Canadian neurologists, only 6.8% reported discussing SUDEP with nearly all (>90%) patients. Almost twice as many respondents (11.6%) reported never discussing it (Friedman et al., 2014). Part of the hesitancy may stem

∗ Corresponding author at: 260 Stetson Street, Suite 2300, Cincinnati, OH 452670525, USA. E-mail addresses: [email protected] (B.D. Moseley), [email protected] (C.M. DeGiorgio). 1 Address: 710 Westwood Plaza, Suite 1250, Los Angeles, CA 90095, USA. http://dx.doi.org/10.1016/j.eplepsyres.2015.09.006 0920-1211/© 2015 Elsevier B.V. All rights reserved.

from a lack of objective biomarkers and inventories which accurately stratify risk. Electrophysiological markers associated with a potentially increased risk of dying suddenly and unexpectedly exist, yet are not currently part of the practice lexicon for most clinicians. These include postictal generalized electroencephalogram (EEG) suppression (PGES) and reduced heart rate variability (HRV) (Lhatoo et al., 2010; Lotufo et al., 2012). However, such markers can be difficult to readily obtain, often requiring prolonged electrocardiogram (ECG) monitoring and the recording of seizures with EEG. Fortunately, risk inventories based on historical factors are now becoming available. The first inventory to be introduced was the SUDEP-7 Inventory, a checklist of seven factors associated with increased risk of SUDEP. These historical risk factors were previously identified in prospective cohort studies (Walczak et al., 2001). It has been investigated in a cohort of patients with severe drug resistant epilepsy. Higher scores were inversely correlated with reduced root mean square of successive differences (RMSSD), a measure of high-frequency HRV and the integrity of vagus nerve mediated cardiac autonomic control (DeGiorgio et al., 2010). However, more recent studies have questioned the validity of some of the risk factors in this inventory, particularly the need for antiepileptic drug (AED) polytherapy (Hesdorffer et al., 2012). In the present study, we examined the strength of association between

B.D. Moseley, C.M. DeGiorgio / Epilepsy Research 117 (2015) 82–84

83

Table 1 Revised SUDEP-7 Risk Factor Inventory. SUDEP risk factor

Definition

ODDS ratio

Weighting LOGE X

1 2 3 4 5 6 7

More than 3 tonic–clonic seizures in last year 1 or more tonic–clonic seizures in last year (if risk factor 1 selected, score as 0) 1 or more seizures of any type over the last 12 months (if risk factor 4 selected, score as 0) >50 seizures of any type per month over the last 12 months Duration of Epilepsy >=30 years Current use of 3 or more antiepileptic drugs Mental Retardation, I.Q. <70, or too impaired to test

8.1 2.4 2.2, 3.8, 4.6 11.5 13.9 4.0 5.0

0 or 2 0 or 1 0 or 1 0 or 2 0 or 3 0 or 1 0 or 2

Table 2 Strength of association between individual SUDEP-7 Inventory questions and PGES. SUDEP-7 Question

Number answering yes with PGES

Number answering no with PGES

Pearson Chi-square (p) value – any PGES

Cramer’s V value – any PGES

Pearson Chi-square (p) value – PGES 30 s

Cramer’s V PGES 30 s

Pearson Chi-square (p) value – PGES 45 s

Cramer’s V PGES 45 s

Pearson Chi-square (p) value – PGES 60 s

Cramer’s V PGES 60 s

1. More than 3 GTCS in last year? 2. One or more GTCS in last year? 3. One or more seizures of any type in the last 12 months? 4. Greater than 50 seizures of any type per month over the last 12 months? 5. Duration of epilepsy >=30 years? 6. Current use of 3 or more AEDs? 7. Intellectual disability?

10/12 (83.3%)

2/23 (8.7%)

<0.001

0.753

<0.001

0.618

0.015

0.402

0.036

0.345

11/17 (64.7%)

1/20 (5%)

<0.001

0.636

0.001

0.538

0.10

0.270

0.12

0.259

12/37 (32.4%)

0/0

N/A

N/A

N/A

N/A

N/A

N/A

N/A

N/A

0/4 (0%)

12/33 (36.4%)

0.14

0.241

0.20

0.212

0.40

0.138

0.61

0.083

0/0

12/37 (32.4%)

N/A

N/A

N/A

N/A

N/A

N/A

N/A

N/A

3/11 (27.3%)

9/26 (34.6%)

0.66

0.072

0.98

0.004

0.59

0.089

0.52

0.106

2/15 (13.3%)

10/22 (45.5%)

0.04

0.337

0.12

0.255

0.047

0.326

0.51

0.197

AEDs, antiepileptic drugs; GTCS, generalized tonic–clonic seizures; N/A, not applicable; PGES, postictal generalized electroencephalogram suppression; SUDEP, sudden unexpected death in epilepsy.

the individual risk factors in the SUDEP-7 Inventory and PGES, a potential biomarker of SUDEP risk (Lhatoo et al., 2010). We sought to determine the relative strength of association of each risk factor with PGES.

The protocol was approved by the applicable institutional review board.

3. Results 2. Materials and methods We performed a reanalysis of clinical and electrophysiological data obtained from an epilepsy monitoring unit (EMU) study of 37 children with focal dyscognitive and primary/secondarily generalized tonic–clonic seizures (GTCS) (Moseley et al., 2013). In that study, the medical records of all children were reviewed to calculate SUDEP-7 Inventory scores. For the present study, we also utilized the Revised SUDEP-7 Risk Factor Inventory (which was created following a re-evaluation of the scoring strategy by this study’s senior author, see Table 1). All children in the original study were evaluated with continuous 30-channel scalp EEG. The International 10–20 system was used for electrode placement. PGES was defined as the absence of EEG activity in all leads (viewed at <=10 ␮V amplitude) following seizure offset. Children with at least one recorded seizure with various durations of PGES (any PGES, >=30, >=45, and >=60 s) were identified. Data entry and statistical analysis were performed using IBM SPSS Statistics Version 22 (IBM, Armonk, NY, U.S.A.). We performed a 2 by 2 contingency table analysis to determine the strength of association between “yes” responses on each of the inventory questions and the presence of PGES of various durations (any, 30, 45, and 60 s) following at least one recorded seizure. Pearson chi-square and Cramer’s V values were calculated. P-values <0.05 were considered statistically significant.

A history of >=3 GTCS in the past year had the strongest association with the recording of any duration of PGES (Pearson chi-square p < 0.001, Cramer’s V = 0.753). This remained significant when examining PGES durations of >=30 s (p < 0.001), >=45 s (p = 0.015), and >=60 s (p = 0.036, see Table 2). Having >=1 GTCS in the past year was also strongly associated with PGES of any duration (Pearson chi-square p < 0.001, Cramer’s V = 0.636). This association remained significant when PGES lasted >=30 s (p = 0.001) but not when lasting >=45 s (p = 0.1) or >=60 s (p = 0.12, see Table 2) Histories of >50 seizures of any type per month (Pearson chi-square p = 0.14, Cramer’s V = 0.241) and intellectual disability (Pearson chisquare p = 0.04, Cramer’s V = 0.337) were not as robustly associated with PGES of any duration. Such associations did not improve when PGES durations ranging from >=30 s to >=60 s were examined (see Table 2) Current use of >=3 AEDs had the weakest association with PGES of any duration (Pearson chi-square p = 0.66, Cramer’s V = 0.072), with even lower Cramer’s V values calculated as PGES duration increased (see Table 2) Although this improved when examining the combination of currently using >=3 AEDs and having >=3 GTCS in the past year, the strength of association with any duration of PGES still did not reach statistical significance (Pearson chi-square p = 0.099, Cramer’s V = 0.345). Given that all study patients had at least one seizure per year, the strength of association with this question and PGES could not be analyzed. Given that all study subjects were <=18 years of age, we could not analyze data regarding epilepsy duration >=30 years.

84

B.D. Moseley, C.M. DeGiorgio / Epilepsy Research 117 (2015) 82–84

Given the weak association of PGES with use of >=3 AEDs, we changed the inventory to a revised SUDEP-6 Inventory (which excluded this question). We subsequently utilized the revised inventory to recalculate risk factor scores. The difference between inventory scores of those with PGES versus those without PGES became more significant with the revised SUDEP-6 inventory (3.92 ± 1.38 for those with PGES versus 2.44 ± 1.33 for those without, independent samples T test p = 0.006) than with the original SUDEP-7 Inventory (4.17 ± 1.34 versus 2.76 ± 1.39, p = 0.007) or the Revised SUDEP-7 Risk Factor Inventory (3.33 ± 1.07 versus 2.84 ± 1.28, p = 0.231). 4. Discussion We discovered that PGES is highly correlated with a revised SUDEP-6 Inventory. Furthermore, our analysis revealed that having >=3 GTCS in the last year had the highest strength of association with PGES of all SUDEP-7 Inventory factors. Other factors such as intellectual disability and more than 50 seizures per year did not have as strong an association with PGES. This argues that historical risk factor inventories for SUDEP should take GTCS frequency into account. Previous studies have identified GTCS, particularly when uncontrolled, as key factors contributing to sudden unexpected death (Walczak et al., 2001; Morentin and Alcaraz, 2002; Hughes, 2009; Hesdorffer et al., 2012). Our results stress the need to counsel patients with frequent GTCS about their higher risk of SUDEP and institute appropriate measures to reduce such risk. This could include more aggressive AED management, earlier referral for resective epilepsy surgery evaluations, and alternative therapies (e.g. dietary and devices) when resective surgery is not possible. Conversely, the low strength of association between use of >=3 AEDs and PGES argues this question is of limited value in estimating SUDEP risk. Previously, some studies had suggested that AED polytherapy as well as specific AEDs (carbamazepine and lamotrigine) might be associated with increased risk (Walczak et al., 2001; Walczak, 2003; Aurlien et al., 2012). However, a more recent reanalysis of 3 case-control studies of SUDEP found that individual AEDs and number of AEDs were not associated with increased risk when adjusted for GTCS frequency (Hesdorffer et al., 2012). This suggests AED polytherapy is merely a surrogate marker of intractability. It is likely more important to achieve the best seizure control possible (including through the prescription of 3 or more AEDs if needed) than to worry about individual or combinations of AEDs when attempting to lower a patient’s risk of mortality. Our study was not without limitations. Given that our study involved data gathered in a pediatric EMU (where patients were <=18 years of age), we could not analyze data regarding epilepsy duration >=30 years. Therefore, we cannot measure the strength of association with this SUDEP risk inventory variable and PGES. Studies in adult patients with variable durations of epilepsy (including >=30 years) will be needed to further determine the usefulness of this question in revised risk inventories. In addition, the association of PGES with sudden unexpected death is not without controversy. Although some studies have found a direct correlation between PGES (particularly prolonged PGES) and higher risk of SUDEP (Lhatoo et al., 2010), other studies have failed to reproduce such findings (Surges et al., 2011). Rather, these conflicting studies argue that PGES could be the final common pathophysiologic pathway in a vicious cycle of events (including reduced activity of pulmonary stretch receptors, increased carotid chemoreceptor sensitivity, and asystole) which culminates in death (So, 2008).

5. Conclusions Based on the strong association with PGES, we conclude that SUDEP risk factor inventories should emphasize the frequency and intractability of GTCS. We believe changing the SUDEP-7 inventory to a new inventory which excludes >=3 AEDs is indicated to help clinicians more accurately convey such risk to their patients. 6. Conflict of interest Dr. Moseley has received previous research support from the Mayo Clinic Department of Neurology Research Fund. He has performed previous consulting work for Nonin Medical, Inc. He serves on an advisory board for UCB Pharma. He also serves on speakers bureaus for UCB Pharma, Cyberonics, and Eisai. Dr. DeGiorgio an inventor of trigeminal nerve stimulation and has an equity interest in NeuroSigma which is unrelated to this work. Dr. DeGiorgio received NIH/NCCAM grant support for work related to the SUDEP risk inventory. This study was not sponsored by a third party. Acknowledgements The authors have no acknowledgements. References Aurlien, D., Larsen, J.P., Gjerstad, L., Tauboll, E., 2012. Increased risk of sudden unexpected death in epilepsy in females using lamotrigine: a nested, case–control study. Epilepsia 53, 258–266. DeGiorgio, C.M., Miller, P., Meymandi, S., Chin, A., Epps, J., Gordon, S., Gornbein, J., Harper, R.M., 2010. RMSSD, a measure of vagus-mediated heart rate variability, is associated with risk factors for SUDEP: the SUDEP-7 Inventory. Epilepsy Behav. 19, 78–81. Ficker, D.M., So, E.L., Shen, W.K., Annegers, J.F., O’Brien, P.C., Cascino, G.D., Belau, P.G., 1998. Population-based study of the incidence of sudden unexplained death in epilepsy. Neurology 51, 1270–1274. Friedman, D., Donner, E.J., Stephens, D., Wright, C., Devinsky, O., 2014. Sudden unexpected death in epilepsy: knowledge and experience among U.S. and Canadian neurologists. Epilepsy Behav. 35, 13–18. Hesdorffer, D.C., Tomson, T., Benn, E., Sander, J.W., Nilsson, L., Langan, Y., Walczak, T.S., Beghi, E., Brodie, M.J., Hauser, W.A., ILAE Commission on Epidemiology (Subcommission on Mortality), 2012. Do antiepileptic drugs or generalized tonic–clonic seizure frequency increase SUDEP risk? A combined analysis. Epilepsia 53, 249–252. Hughes, J.R., 2009. A review of sudden unexpected death in epilepsy: prediction of patients at risk. Epilepsy Behav. 14, 280–287. Lhatoo, S.D., Faulkner, H.J., Dembny, K., Trippick, K., Johnson, C., Bird, J.M., 2010. An electroclinical case–control study of sudden unexpected death in epilepsy. Ann. Neurol. 68, 787–796. Lotufo, P.A., Valiengo, L., Bensenor, I.M., Brunoni, A.R., 2012. A systematic review and meta-analysis of heart rate variability in epilepsy and antiepileptic drugs. Epilepsia 53, 272–282. Morentin, B., Alcaraz, R., 2002. Sudden unexpected death in epilepsy in children and adolescents. Rev. Neurol. 34, 462–465. Moseley, B.D., So, E., Wirrell, E.C., Nelson, C., Lee, R.W., Mandrekar, J., Britton, J.W., 2013. Characteristics of postictal generalized EEG suppression in children. Epilepsy Res. 106, 123–127. Nashef, L., So, E.L., Ryvlin, P., Tomson, T., 2012. Unifying the definitions of sudden unexpected death in epilepsy. Epilepsia 53, 227–233. So, E.L., 2008. What is known about the mechanisms underlying SUDEP? Epilepsia 49 (Suppl. 9), 93–98. Surges, R., Strzelczyk, A., Scott, C.A., Walker, M.C., Sander, J.W., 2011. Postictal generalized electroencephalographic suppression is associated with generalized seizures. Epilepsy Behav. 21, 271–274. Tomson, T., Walczak, T., Sillanpaa, M., Sander, J.W., 2005. Sudden unexpected death in epilepsy: a review of incidence and risk factors. Epilepsia 46 (Suppl. 11), 54–61. Walczak, T., 2003. Do antiepileptic drugs play a role in sudden unexpected death in epilepsy? Drug Saf. 26, 673–683. Walczak, T.S., Leppik, I.E., D’Amelio, M., Rarick, J., So, E., Ahman, P., Ruggles, K., Cascino, G.D., Annegers, J.F., Hauser, W.A., 2001. Incidence and risk factors in sudden unexpected death in epilepsy: a prospective cohort study. Neurology 56, 519–525. Xu, Z., Ayyappan, S., Seneviratne, U., 2015. Sudden unexpected death in epilepsy (SUDEP): what do patients think? Epilepsy Behav. 42, 29–34.