Cocaine Use As an Independent Predictor of Seizures After Aneurysmal Subarachnoid Hemorrhage

clinical article J Neurosurg 124:730–735, 2016

Cocaine use as an independent predictor of seizures after aneurysmal subarachnoid hemorrhage

Tiffany R. Chang, MD,1 Robert G. Kowalski, MBBCh, MS,2 J. Ricardo Carhuapoma, MD,2–4 Rafael J. Tamargo, MD,2,4 and Neeraj S. Naval, MD2–4

1Departments of Neurosurgery and Neurology, University of Texas Medical School, Houston, Texas; and Departments of 2Anesthesia and Critical Care Medicine, 3Neurology, and 4Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, Maryland

Objective Seizures are relatively common after aneurysmal subarachnoid hemorrhage (aSAH). Seizure prophylaxis is controversial and is often based on risk stratification; middle cerebral artery (MCA) aneurysms, associated intracerebral hemorrhage (ICH), poor neurological grade, increased clot thickness, and cerebral infarction are considered highest risk for seizures. The purpose of this study was to evaluate the impact of recent cocaine use on seizure incidence following aSAH. Methods Prospectively collected data from aSAH patients admitted to 2 institutional neuroscience critical care units between 1991 and 2009 were reviewed. The authors analyzed factors that potentially affected the incidence of seizures, including patient demographic characteristics, poor clinical grade (Hunt and Hess Grade IV or V), medical comorbidities, associated ICH, intraventricular hemorrhage (IVH), hydrocephalus, aneurysm location, surgical clipping and cocaine use. They further studied the impact of these factors on “early” and “late” seizures (defined, respectively, as occurring before and after clipping/coiling). Results Of 1134 aSAH patients studied, 182 (16%) had seizures; 81 patients (7.1%) had early and 127 (11.2%) late seizures, with 26 having both. The seizure rate was significantly higher in cocaine users (37 [26%] of 142 patients) than in non–cocaine users (151 [15.2%] of 992 patients, p = 0.001). Eighteen cocaine-positive patients (12.7%) had early seizures compared with 6.6% of cocaine-negative patients (p = 0.003); 27 cocaine users (19%) had late seizures compared with 10.5% non–cocaine users (p = 0.001). Factors that showed a significant association with increased risk for seizure (early or late) on univariate analysis included younger age (< 40 years) (p = 0.009), poor clinical grade (p = 0.029), associated ICH (p = 0.007), and MCA aneurysm location (p < 0.001); surgical clipping was associated with late seizures (p = 0.004). Following multivariate analysis, age < 40 years (OR 2.04, 95% CI 1.355–3.058, p = 0.001), poor clinical grade (OR 1.62, 95% CI 1.124–2.336, p = 0.01), ICH (OR 1.95, 95% CI 1.164–3.273, p = 0.011), MCA aneurysm location (OR 3.3, 95% CI 2.237–4.854, p < 0.001), and cocaine use (OR 2.06, 95% CI 1.330–3.175, p = 0.001) independently predicted seizures. Conclusions Cocaine use confers a higher seizure risk following aSAH and should be considered during risk stratification for seizure prophylaxis and close neuromonitoring. http://thejns.org/doi/abs/10.3171/2015.2.JNS142856 KEY WORDS subarachnoid hemorrhage; aneurysm; cocaine; seizure; seizure prophylaxis; antiepileptic medication; vascular disorders

ocaine abuse is a significant public health issue in among cocaine users.13,14,28 Among patients with aneurys- the United States. There are approximately 1.6 mil- mal SAH (aSAH), recent cocaine use has been reported in lion cocaine users in the US, with 639,000 new us- up to 33% of cases.10 Cocaine users with aSAH tend to be ersC per year.27 Cocaine use is a known risk factor for stroke, younger than other aSAH patients, and their aneurysms particularly among young patients without other vascular are more often located in the anterior circulation.3,19 Re- risk factors.4,13,14 While subarachnoid hemorrhage (SAH) cent cocaine use is independently associated with higher comprises only 5% of all strokes in the general popula- rates of aneurysm rerupture and a nearly 3-fold increase in tion,1 SAH accounts for roughly 20%–30% of strokes hospital mortality following aSAH.2

ABBREVIATIONS aSAH = aneurysmal SAH; EEG = electroencephalographic; GCS = Glasgow Coma Scale; ICH = intracerebral hemorrhage; IVH = intraventricular hemorrhage; MCA = middle cerebral artery; NCCU = neuroscience critical care unit; OR = odds ratio; SAH = subarachnoid hemorrhage. submitted December 17, 2014. accepted February 24, 2015. include when citing Published online August 28, 2015; DOI: 10.3171/2015.2.JNS142856.

730 J Neurosurg Volume 124 • March 2016 ©AANS, 2016

Unauthenticated | Downloaded 09/29/21 01:59 AM UTC Cocaine and seizures after subarachnoid hemorrhage

Cocaine abuse has also been historically associated characteristics, poor clinical Hunt and Hess grade (Grade with seizures. Seizures have been reported in 1%–10% IV or V), medical comorbidities, associated intracerebral of patients with acute cocaine use, but there have been hemorrhage (ICH) or intraventricular hemorrhage (IVH), conflicting reports regarding the true impact on seizure hydrocephalus, aneurysm location, surgical clipping, and incidence compared with the incidence in non–cocaine cocaine use. Furthermore, we investigated whether modi- users.26 In general, seizures are a relatively frequent oc- fication of the seizure prophylaxis regimen over the past currence after aSAH and may occur in up to 20% of pa- 2 decades—from phenytoin (1991–2006) to levetiracetam tients.15 In the International Subarachnoid Aneurysm Trial and/or phenytoin (2006–2009)—had any impact on seizure (ISAT), seizures were found to be more common follow- occurrence. Our methods for identifying cocaine-positive ing surgical clipping (14%) than after endovascular coil- versus cocaine-negative patients have been described in ing (8%).9 Other reported risk factors for seizure following detail in previous publications.2,20 Acute cocaine use was aSAH include middle cerebral artery (MCA) aneurysm identified based on positive urine toxicology or a history location, younger age, worse clinical grade, delayed ce- of cocaine use in the past 72 hours. As some patients pre- rebral ischemia, subarachnoid clot burden, and subdural sented a few days after the initial onset of headache, when hematoma.9,11 The relationship between cocaine use and the urine toxicology findings may have been falsely nega- seizures following aSAH has been poorly studied. tive, a positive history of cocaine use within 72 hours took We sought to further investigate the relationship be- precedence over negative results on urine drug screening tween cocaine use and seizures following aSAH by re- in those cases. We further studied the impact of each of viewing prospectively collected data from our institution these factors on early and late seizures. between 1991 and 2009. We examined the impact of cocaine use and other potential seizure risk factors on the Statistical Analysis incidence and timing of seizures after aSAH. The unpaired t-test was used when data were normally distributed, and nonparametric tests (Mann-Whitney U- Methods test, Kruskal-Wallis test) were used when data were not normally distributed or categorical. Dichotomous vari- Cases involving patients admitted to the 2 neurosci- ables were compared with outcome using the chi-square ence critical care units (NCCUs) at Johns Hopkins Medi- test. The Fisher exact test result was reported where appro- cal Institutions with a diagnosis of aSAH between 1991 priate. Data were analyzed with SPSS version 22.0 (IBM and 2009 were reviewed. Vascular neurosurgeons, neu- Inc.) to assess the potential impact of each of the admission rointerventional faculty, and neurointensivists with spe- factors on seizure incidence. The impact of various factors cialized qualifications in the care of patients with SAH was assessed using multiple logistic regression analysis, are common across both the campuses, located at Johns excluding factors that were deemed to be collinear. Hopkins Hospital and Johns Hopkins Bayview Medical Center. Treatment protocols for aSAH are similar between the 2 units. Data were collected prospectively in Results an institutional review board–approved database. Patients Data from 1134 patients with aSAH were included in with SAH due to trauma, angiogram-negative SAH of un- the analysis. A total of 182 (16%) of these patients had certain etiology, or SAH secondary to other causes such seizures. Of these, 81 (7.1%) had early seizures and 127 as arteriovenous malformations, dural arteriovenous fis- (11.2%) had late seizures; 26 patients had both early and tulas, and brain tumors were excluded from our analysis. late seizures. Patient characteristics are shown in Table The Johns Hopkins University institutional review board 1. Patients who had seizures were significantly younger approved the database collection and waived the need for (mean age 50 ± 14.7 years) than patients without seizures informed consent. (mean age 53 ± 13.8 years, p = 0.009). Compared with Patients were divided into 2 groups—those with any those without seizures, patients with seizures had a higher seizure at initial presentation and/or during their hospital- rate of cocaine use (20% vs 11%, p = 0.001), poor Hunt ization and those without. Seizure was defined based on 1) and Hess grade at admission (32% vs 24%, p = 0.029), as- a clinical diagnosis of seizure by the treating physician, or sociated ICH (13% vs 7%, p = 0.007), and MCA aneurysm 2) an electroencephalogram reading confirming seizure. location (30% vs 11%, p < 0.001). The two groups had Seizures were also categorized as early (before clipping or similar median GCS scores and rates of medical comor- coiling) or late (after clipping or coiling). Electroencepha- bidities, IVH, and hydrocephalus. lographic (EEG) studies were obtained on an individual Characteristics of patients with early and late seizures basis based on the clinical assessment of the treatment are shown in Table 2. Patients with early seizures had a team; routine EEG monitoring was not performed. Con- higher incidence of associated ICH (16% vs 7%, p = 0.005) tinuous EEG monitoring was available if clinically indi- and higher frequency of poor Hunt and Hess grade at ad- cated. All patients received antiepileptic prophylaxis with mission (35% vs 24%, p = 0.038) than those without early phenytoin and/or levetiracetam. This treatment was con- seizures. Among patients with late seizures, the incidence tinued for a minimum of 7 days, or longer at the discretion of ICH (11% vs 7%, p = 0.134) and frequency of poor Hunt of the treating physician. Seizures, if they occurred, were and Hess grade (31% vs 24%, p = 0.073) were not signifi- managed by the neurocritical care team. cantly different between the 2 groups. Based on our defi- We analyzed factors that potentially affected the over- nition, surgical clipping should only be expected to alter all incidence of seizures, including patient demographic the incidence of late (postprocedural) seizures. Surgical

J Neurosurg Volume 124 • March 2016 731

Unauthenticated | Downloaded 09/29/21 01:59 AM UTC T. R. Chang et al.

TABLE 1. Patient characteristics affecting seizures after aSAH* phenytoin prophylaxis regimens used for the last 3 years No Seizure Seizure of the studied database (16.6%, p > 0.05). Characteristic (n = 952) (n = 182) p Value Multivariate analysis was performed using factors found to be associated with seizures in univariate analyses Sex, female 688 (72) 125 (69) 0.325 (Table 3). Following multivariate analysis, age < 40 years Race (OR 2.04, 95% CI 1.355–3.058, p = 0.001), poor clinical White 559 (59) 102 (56) 0.794 grade (OR 1.62, 95% CI 1.124–2.336, p = 0.01), ICH (OR Black 358 (38) 73 (40) 1.95, 95% CI 1.164–3.273, p = 0.011), MCA location (OR Hispanic 13 (1) 4 (2) 3.3, 95% CI 2.237–4.854, p < 0.001) and cocaine use (OR 2.06, 95% CI 1.330–3.175, p = 0.001) independently pre- Asian 2 (1) 0 (0) dicted seizures after aSAH. Other 20 (2) 3 (2) Mean age in yrs 53.0 ± 13.8 50.0 ± 14.7 0.009 Cocaine Use Age <40 yrs 139 (15) 45 (25) 0.001 The incidence of seizures following aSAH in patients Cocaine use 105 (11) 37 (20) 0.001 with and without cocaine use is shown in Fig. 1. Cocaine GCS score at admission use was significantly associated with higher rates of over- Median 14 14 0.809 all, early, and late seizures. Of 142 cocaine users, 37 (26%) Range 3–15 3–15 had seizures compared with 151 (15.2%) of 992 non–cocaine users (p = 0.001). Eighteen cocaine-positive patients Poor admission HH 230 (24) 58 (32) 0.029 (12.7%) had early seizures compared with 6.6% cocaine- grade† negative patients (p = 0.003); 27 cocaine users (19%) had Comorbid condition late seizures compared with 10.5% non–cocaine users (p No risk factors 339 (36) 66 (36) 0.940 = 0.001). 1 risk factor 353 (37) 65 (36) Aneurysm rerupture occurred in 11 patients (7.7%) 2 or more risk factors 260 (27) 51 (28) with cocaine use and 27 patients (2.7%) without cocaine IVH 494 (52) 92 (50) 0.740 use. As there is a potential interaction between aneurysm rerupture and seizures, we further investigated this by ex- ICH 69 (7) 24 (13) 0.007 cluding all patients with aneurysm rerupture and repeating Hydrocephalus 503 (53) 86 (47) 0.167 our analysis. Cocaine users still had a higher incidence of Aneurysm location overall seizures (25.2%) compared with non–cocaine us- ACoA/ACA 358 (38) 69 (38) <0.001 ers (14.4%) after excluding those patients with rerupture MCA 106 (11) 54 (30) (p = 0.003). ICA 98 (10) 19 (10) VB/AICA/PICA/SCA 162 (17) 8 (4) Discussion PCoA 228 (24) 32 (18) We found that acute cocaine use is independently as- MCA vs other 106 (11) 54 (30) <0.001 sociated with seizures after aSAH. This is consistent Surgical intervention with the results of a recent retrospective review from the 18 Clipping 748 (79) 152 (83) 0.288 Nationwide Inpatient Sample. Murthy et al. reported a seizure incidence of 16.2% in cocaine users compared Coiling 132 (14) 18 (10) with 11.1% in non–cocaine users following aSAH.18 Due None 72 (8) 12 (7) to the nature of the database, seizure was poorly defined ACA = anterior cerebral artery; ACoA= anterior communicating artery; AICA = and subject to errors in reporting and coding. Our current anterior inferior cerebellar artery; HH = Hunt and Hess; ICA = internal carotid study confirms that prospectively identified seizures after artery; MCA = middle cerebral artery; PCoA = posterior communicating aSAH occur more commonly among cocaine users than artery; PICA = posterior inferior cerebellar artery; SCA = superior cerebellar in non–cocaine users at our institution. artery; VB = vertebral or basilar artery. Cocaine use is traditionally thought to increase the risk * Data are presented as numbers of cases (%) unless otherwise indicated. of seizures.24,25 There are a number of proposed mecha- Mean values are presented ± SD. † Grade IV or V. nisms by which cocaine may induce seizures. Cocaine de- creases GABA activity through alternations in the func- tion of ion channels, in particular sodium and calcium channels, leading to an increase in excitatory neurotrans- clipping was not significantly more frequent among pa- mission.7,22,29 Increased serotonin neurotransmission due tients with any seizure (83%) compared with seizure-free to decreased serotonin reuptake has also been implicated patients (79%, p = 0.288); however, clipping was used sig- in cocaine-induced seizures.17 The 5-HT2 receptors ap- nificantly more often in patients with late seizures (91%) pear to play a significant role in this effect. 5-HT2 recep- than in those with no seizures (79%, p = 0.004). There tor agonists and serotonin reuptake inhibitors have been was no significant difference in seizure incidence in the shown to increase cocaine-induced seizures in animal 2 NCCUs (15.9% vs 16.5%, p > 0.05). Also, rates of sei- models, while 5-HT2 antagonists appear to increase the zures were not significantly different when comparing seizure threshold.16,17,21 Lastly, chronic cocaine use may the “phenytoin era” (15.9%) with the levetiracetam and/or increase the body’s sensitivity to the convulsant effects of

732 J Neurosurg Volume 124 • March 2016

Unauthenticated | Downloaded 09/29/21 01:59 AM UTC Cocaine and seizures after subarachnoid hemorrhage

TABLE 2. Patient characteristics affecting early and late seizures after aSAH* No Seizure Early Seizure No Seizure Late Seizure Characteristic (n = 952) (n = 81) p Value (n = 952) (n = 127) p Value Female sex 688 (72) 59 (73) 0.912 688 (72) 84 (66) 0.151 Race White 559 (59) 48 (59) 0.534 559 (59) 71 (56) 0.824 Black 358 (38) 29 (36) 358 (38) 51 (40) Hispanic 13 (1) 3 (4) 13 (1) 3 (2) Asian 2 (1) 0 (0) 2 (1) 0 (0) Other 20 (2) 1 (1) 20 (2) 2 (2) Mean age in yrs 53.0 ± 13.8 49.5 ± 14.0 0.028 53.0 ± 13.8 50.5 ± 15.5 0.064 Age <40 yrs 139 (15) 20 (25) 0.016 139 (15) 31 (24) 0.004 Cocaine use 105 (11) 18 (22) 0.003 105 (11) 27 (21) 0.001 GCS score at admission Median 14 13 0.592 14 14 0.585 Range 3–15 3–15 3–15 3–15 Poor admission HH grade 230 (24) 28 (35) 0.038 230 (24) 40 (31) 0.073 Comorbid condition No risk factors 339 (36) 32 (39) 0.653 339 (36) 45 (35) 0.796 1 risk factor 353 (37) 26 (32) 353 (37) 44 (35) 2 or more risk factors 260 (27) 23 (28) 260 (27) 38 (30) IVH 494 (52) 36 (44) 0.198 494 (52) 69 (54) 0.605 ICH 69 (7) 13 (16) 0.005 69 (7) 14 (11) 0.134 Hydrocephalus 503 (53) 38 (47) 0.306 503 (53) 60 (47) 0.236 Aneurysm location ACoA/ACA 358 (38) 28 (35) <0.001 358 (38) 45 (35) <0.001 MCA 106 (11) 32 (39) 106 (11) 38 (30) ICA 98 (10) 10 (12) 98 (10) 11 (9) VB/AICA/PICA/SCA 162 (17) 4 (5) 162 (17) 4 (3) PCoA 228 (24) 7 (9) 228 (24) 29 (23) MCA vs other 106 (11) 32 (39) <0.001 106 (11) 38 (30) <0.001 Surgical intervention Clipped 748 (79) 60 (74) 0.309 748 (79) 115 (91) 0.004 Coiled 132 (14) 11 (14) 132 (14) 10 (8) None 72 (8) 10 (12) 72 (8) 2 (2) * Data are numbers of patients (%) unless otherwise indicated. Mean values are presented ± SD. the drug through kindling. Cocaine kindling is postulated contributing factors include direct toxicity of the blood to be secondary to modified NDMA receptor expression itself, inflammation, increased intracranial pressure, vaso- with repeated exposure, resulting in increased glutamate spasm, and cortical spreading depolarization.5,8,11 Regard- activity in the brain.12,23 less of the etiology, SAH patients appear to have a low- The underlying mechanism behind seizures after SAH ered seizure threshold. It is plausible the cocaine-induced remains unclear and is possibly multifactorial. Potential alterations in neuronal activity previously discussed have an additive or synergistic effect on the proconvulsive state TABLE 3. Multivariate analysis of factors impacting seizures seen after SAH. For example, spreading convulsions have after aneurysmal SAH been identified in patients with SAH and are attributed to 6 Factor OR p Value 95% CI reduced GABA activity. This would be further exacer- bated by both the GABA suppression and NMDA recep- Age <40 yrs 2.037 0.001 1.355–3.058 tor–mediated excitatory effects of cocaine. Cocaine 2.055 0.001 1.330–3.175 In a previous publication involving the same patient Poor admission HH grade 1.620 0.010 1.124–2.336 population, we reported that cocaine users have signifi- 2 ICH 1.952 0.011 1.164–3.273 cantly higher rates of aneurysm rerupture. After excluding MCA aneurysm 3.300 <0.001 2.237–4.854 the patients with rerupture, we were able to exclude this as the possible mechanism underlying the higher incidence

J Neurosurg Volume 124 • March 2016 733

Unauthenticated | Downloaded 09/29/21 01:59 AM UTC T. R. Chang et al.

stitutional protocol, all patients with aSAH are started on prophylactic antiepileptic therapy with phenytoin or levetiracetam at the time of admission, and this is not universal practice across the country. This could impact our overall seizure rates, as could the availability and utilization of EEG monitoring in more recent years. We did not, however, note a change in the reported rate of seizures over time with the availability of more frequent long-term EEG monitoring or with the use of newer antiepileptic drugs for seizure prophylaxis as reported above. Although there are many limitations, we believe our results identify cocaine use as an independent risk factor for seizures following aSAH. Cocaine use should be taken into consideration during risk stratification for seizure prophylaxis. A careful history of substance use should be obtained whenever possible, and it is reasonable Fig. 1. Incidence of seizures following aSAH in patients with and without to consider toxicology screening in all patients present- acute cocaine use. ing with aSAH. In addition, a higher level of suspicion for seizure should be maintained in patients identified to be of seizures in the cocaine population. Another potential cocaine positive at admission. Cocaine-positive patients confounding factor is the concurrent use of alcohol, other may warrant more frequent or continuous EEG monitor- illicit substances, or centrally acting medications that could ing, particularly patients with a poor and hence unreliable impact the seizure threshold. Other substances have the po- neurological examination. Further studies are needed to tential to either augment or diminish the convulsive effect determine whether cocaine users should receive a differ- of cocaine, depending on the substance and dose.16 Unfor- ent prophylactic agent, dose, or duration of treatment and tunately, we had insufficient data to control for this effect. whether different EEG monitoring practices should be ad- Surgical clipping was noted to be the modality used opted in this population. to secure the aneurysm in 91% of the patients who had late seizures; this was higher than the overall 84% surgi- Conclusions cal clipping rate for this aSAH population who had their Acute cocaine use confers a higher risk of seizures fol- aneurysms secured. This suggests a likely impact of the lowing aSAH. Cocaine use should be considered during choice of procedure on subsequent risk of seizures. The risk stratification for seizure prophylaxis and neuromoni- fact that surgical clipping did not have a significant im- toring in aSAH patients. pact on the overall incidence of seizures was a surprising finding at first review of the results; however on further reflection, given that a significant number of seizures oc- References curred prior to securing the aneurysm, the choice of pro- 1. Bederson JB, Connolly ES Jr, Batjer HH, Dacey RG, Dion cedure could not be held accountable for this early seizure JE, Diringer MN, et al: Guidelines for the management incidence given our definition for timing of seizures. The of aneurysmal subarachnoid hemorrhage: a statement for true impact of the procedure was really only seen in late healthcare professionals from a special writing group of seizures (Table 3), which was significant (p = 0.004). the Stroke Council, American Heart Association. Stroke 40:994–1025, 2009 Our study is limited by its retrospective nature and the 2. Chang TR, Kowalski RG, Caserta F, Carhuapoma JR, Ta single academic institution experience. The patient popu- margo RJ, Naval NS: Impact of acute cocaine use on aneurys- lation included in our analysis may not be representative mal subarachnoid hemorrhage. Stroke 44:1825–1829, 2013 of other centers around the country. A notable difference 3. Conway JE, Tamargo RJ: Cocaine use is an independent risk is the high rate of surgical clipping at our institution (over factor for cerebral vasospasm after aneurysmal subarachnoid 80%), which is disparate from what is practiced across hemorrhage. Stroke 32:2338–2343, 2001 most centers in the country. Given that craniotomy was 4. Daras M, Tuchman AJ, Koppel BS, Samkoff LM, Weitzner I, Marc J: Neurovascular complications of cocaine. Acta not performed more often in the cocaine cohort than in Neurol Scand 90:124–129, 1994 the non–cocaine group, we do not believe that craniotomy 5. De Reuck J, Goethals M, Vonck K, Van Maele G: Clinical in itself invalidates our findings as it relates to epilepto- predictors of late-onset seizures and epilepsy in patients with genic impact of cocaine. Our analysis also did hold true cerebrovascular disease. Eur Neurol 54:68–72, 2005 when only coil-treated patients were evaluated. Within the 6. Dreier JP, Major S, Pannek HW, Woitzik J, Scheel M, Wies subgroup of patients treated with coiling, cocaine use was enthal D, et al: Spreading convulsions, spreading depolariza- associated with a higher incidence of late seizures (12.5%) tion and epileptogenesis in human cerebral cortex. Brain compared with the incidence in patients without cocaine 135:259–275, 2012 7. Gasior M, Kaminski R, Witkin JM: Pharmacological use (6.3%); however, due to the relatively small number of modulation of GABA(B) receptors affects cocaine-induced aneurysms that were treated by this modality, this differ- seizures in mice. Psychopharmacology (Berl) 174:211–219, ence did not meet statistical significance. All patients were 2004 admitted to a specialized NCCU, which is not available at 8. Hart RG, Byer JA, Slaughter JR, Hewett JE, Easton JD: many centers in the United States. According to our in- Occurrence and implications of seizures in subarachnoid

734 J Neurosurg Volume 124 • March 2016

Unauthenticated | Downloaded 09/29/21 01:59 AM UTC Cocaine and seizures after subarachnoid hemorrhage

hemorrhage due to ruptured intracranial aneurysms. Neuro- tion in the SCN1A sodium channel gene on cocaine-induced surgery 8:417–421, 1981 seizure susceptibility in mice. Psychopharmacology (Berl) 9. Hart Y, Sneade M, Birks J, Rischmiller J, Kerr R, Molyneux 228:263–270, 2013 A: Epilepsy after subarachnoid hemorrhage: the frequency 23. Rolland B, Karila L, Geoffroy PA, Cottencin O: Shared of seizures after clip occlusion or coil embolization of a vulnerability between seizures and psychosis in cocaine ad- ruptured cerebral aneurysm: results from the International diction? Epilepsy Behav 22:596–598, 2011 Subarachnoid Aneurysm Trial. J Neurosurg 115:1159–1168, 24. Ropper AH, Samuels MA: Adams and Victor’s Principles 2011 of Neurology, ed 9. New York: McGraw-Hill Medical, 10. Howington JU, Kutz SC, Wilding GE, Awasthi D: Cocaine 2009 use as a predictor of outcome in aneurysmal subarachnoid 25. Rowland LP, Merritt HH: Merritt’s Neurology, ed 11. Phila- hemorrhage. J Neurosurg 99:271–275, 2003 delphia: Lippincott Williams & Wilkins, 2005 11. Ibrahim GM, Fallah A, Macdonald RL: Clinical, laboratory, 26. Sordo L, Indave BI, Degenhardt L, Barrio G, Kaye S, and radiographic predictors of the occurrence of seizures fol- Ruíz-Pérez I, et al: A systematic review of evidence on the lowing aneurysmal subarachnoid hemorrhage. J Neurosurg association between cocaine use and seizures. Drug Alcohol 119:347–352, 2013 Depend 133:795–804, 2013 12. Kaminski RM, Núñez-Taltavull JF, Budziszewska B, Lasoń 27. Substance Abuse and Mental Health Services Administra- W, Gasior M, Zapata A, et al: Effects of cocaine-kindling on tion: Results from the 2012 National Survey on Drug Use the expression of NMDA receptors and glutamate levels in and Health: Summary of National Findings. Rockville, mouse brain. Neurochem Res 36:146–152, 2011 MD: US Department of Health and Human Services, 2013 13. Klonoff DC, Andrews BT, Obana WG: Stroke associated 28. Toossi S, Hess CP, Hills NK, Josephson SA: Neurovascular with cocaine use. Arch Neurol 46:989–993, 1989 complications of cocaine use at a tertiary stroke center. J 14. Levine SR, Brust JC, Futrell N, Ho KL, Blake D, Millikan Stroke Cerebrovasc Dis 19:273–278, 2010 CH, et al: Cerebrovascular complications of the use of the 29. Ye JH, Ren J: Cocaine inhibition of GABA(A) current: role “crack” form of alkaloidal cocaine. N Engl J Med 323:699– of dephosphorylation. Crit Rev Neurobiol 18:85–94, 2006 704, 1990 15. Lin CL, Dumont AS, Lieu AS, Yen CP, Hwang SL, Kwan AL, et al: Characterization of perioperative seizures and epilepsy following aneurysmal subarachnoid hemorrhage. J Disclosure Neurosurg 99:978–985, 2003 16. Macêdo DS, Santos RS, Belchior LD, Neto MA, Vasconcelos The authors report no conflict of interest concerning the materi- SM, Lima VT, et al: Effect of anxiolytic, antidepressant, and als or methods used in this study or the findings specified in this antipsychotic drugs on cocaine-induced seizures and mortal- paper. ity. Epilepsy Behav 5:852–856, 2004 17. Morita K, Hamamoto M, Arai S, Kitayama S, Irifune M, Author Contributions Kawahara M, et al: Inhibition of serotonin transporters by Conception and design: all authors. Acquisition of data: Chang, cocaine and meprylcaine through 5-TH2C receptor stimula- Tamargo, Naval. Analysis and interpretation of data: all authors. tion facilitates their seizure activities. Brain Res 1057:153– Drafting the article: Chang, Naval. Critically revising the article: 160, 2005 all authors. Reviewed submitted version of manuscript: all 18. Murthy SB, Moradiya Y, Shah S, Naval NS: In-hospital authors. Approved the final version of the manuscript on behalf outcomes of aneurysmal subarachnoid hemorrhage associ- of all authors: Chang. Statistical analysis: Kowalski. Study super- ated with cocaine use in the USA. J Clin Neurosci 21:2088– vision: Naval. 2091, 2014 19. Nanda A, Vannemreddy PS, Polin RS, Willis BK: Intracra- Supplemental Information nial aneurysms and cocaine abuse: analysis of prognostic Previous Presentation indicators. Neurosurgery 46:1063–1069, 2000 20. Naval NS, Kowalski RG, Chang TR, Caserta F, Carhuapoma Portions of this work were presented in poster form at the 12th JR, Tamargo RJ: The SAH Score: a comprehensive commu- Annual Meeting of the Neurocritical Care Society, Seattle, nication tool. J Stroke Cerebrovasc Dis 23:902–909, 2014 Washington, September 11, 2014. 21. O’Dell LE, Kreifeldt MJ, George FR, Ritz MC: The role of serotonin(2) receptors in mediating cocaine-induced convul- Correspondence sions. Pharmacol Biochem Behav 65:677–681, 2000 Tiffany Chang, Department of Neurosurgery, University of Texas 22. Purcell RH, Papale LA, Makinson CD, Sawyer NT, Schro- at Houston, 6431 Fannin St., MSB 7.154, Houston, TX 77030. eder JP, Escayg A, et al: Effects of an epilepsy-causing muta- email: [email protected].

J Neurosurg Volume 124 • March 2016 735

Unauthenticated | Downloaded 09/29/21 01:59 AM UTC