Association of Aggression and Antiepileptic Drugs: Analysis Using the Japanese Adverse Drug Event Report (JADER) Database

Abstract

Aggression is the most common adverse effect of antiepileptic drugs (AEDs). This study aimed to investigate the association of aggression with AED use. The reporting odds ratio (ROR) from adverse event reports, submitted to the Japanese Adverse Drug Event Report database between 2004 and 2020, was used to calculate and investigate the association between AEDs and aggression. We also analyzed the association of aggression with the combined use of AEDs and the relationship between AED-associated aggression and patient characteristics. A total of 433 patients developed aggression. Significant aggression signals were detected for perampanel (crude ROR: 325.04, 95% confidence interval (CI): 118.48–752.58, p < 0.01), levetiracetam (crude ROR: 17.14, 95% CI: 10.33–26.90, p < 0.01), lacosamide (crude ROR: 16.90, 95% CI: 2.02–62.51, p < 0.01), lamotrigine (crude ROR: 15.98, 95% CI: 9.99–24.39, p < 0.01), valproate (crude ROR: 6.68, 95% CI: 4.27–10.02, p < 0.01), and carbamazepine (crude ROR: 2.47, 95% CI: 1.17–4.59, p < 0.01). The combined therapy with perampanel and levetiracetam had a significant aggression signal (adjusted ROR: 25.90, 95% CI: 1.14–59.10, p < 0.01). In addition, we found that aggression frequently occurred in patients <60 year (adjusted ROR: 2.88, 95% CI: 1.49–5.56, p < 0.01) treated with levetiracetam. These results may be useful for minimizing the risk of aggression during the treatment of AEDs.

INTRODUCTION

More than 45 million patients with active epilepsy worldwide require chronic treatment with antiepileptic drugs (AEDs) to control seizures.¹⁾ It has been estimated that up to 50% of AED users experience adverse reactions; it is essential to control these adverse effects for the successful treatment of epilepsy.²⁾ Psychiatric and behavioral adverse effects, such as aggression, are the most common adverse effects associated with AEDs (experienced by 10–22% of AED users).^3,4) Aggression can negatively impact the QOL, lead to poor adherence to AEDs, and cause difficulty in caring for such epileptic patients.^5–7) Epilepsy is very difficult to treat and often involves combination therapy with AEDs. However, the relationship between aggression and AED combinations is not clear. Therefore, knowledge of the possible factors and AED combinations associated with aggression may help manage and reduce the adverse effects.

Multiple AEDs have been developed in recent decades. Although new-generation AEDs have better tolerability profiles than conventional AEDs,⁴⁾ their novel pharmacological mechanisms of action may cause peculiar adverse effects.⁸⁾ Some studies have demonstrated that the use of new-generation AEDs is associated with psychiatric and behavioral adverse effects.⁶⁾ However, the relationship between aggression and new-generation AED combination is not clear among the psychiatric and behavioral adverse effects. In particular, levetiracetam is a broad-spectrum new-generation AED that is effective and safe and has been widely used as a treatment option for various seizure types.⁹⁾ In addition, combinations of levetiracetam are often prescribed to treat intractable epilepsy.¹⁰⁾ There are limited studies on the development of aggression when taking combinations of levetiracetam, and there is only one single-center cohort study.¹¹⁾ This indicates a lack of sufficient evidence for the relationship between aggression and combinations of levetiracetam with other AEDs, as well as patient characteristics.

The Japan Adverse Drug Report (JADER) is a database that collects information on adverse drug reaction cases reported voluntarily to the Pharmaceuticals and Medical Devices Agency (PMDA). Some previous studies have used JADER to investigate rare side effects.¹²⁾ To perform an analysis using a larger number of cases, JADER was used in this study.

In this study, we analyzed the JADER database to investigate the differences in the frequency of aggression with AEDs. We also examined the association of aggression with the combined use of levetiracetam and patient characteristics.

MATERIALS AND METHODS

Study Design

A retrospective, observational cohort study of patients enrolled in the JADER database was performed. First, we calculated the reporting odds ratios (RORs) and 95% confidence intervals (CIs) of aggression using AEDs and patient characteristics, using 594599 cases, which excluded 73130 missing data on sex and age from all 667729 cases in the JADER dataset, based on previous reports.^12,13) To analyze the onset of drug-induced aggression, we selected 12 AEDs (zonisamide, carbamazepine, phenobarbital, phenytoin, valproate, clobazam, lamotrigine, perampanel, levetiracetam, lacosamide, topiramate, and gabapentin) from the JADER dataset.

First, using 587017 cases, which excluded 7582 cases with concomitant use of AEDs, we investigated which single-AED was significantly associated with aggression to eliminate effects due to concurrent administration of other AEDs.¹⁴⁾ Second, we focused on combinations of levetiracetam with other AEDs, which are frequently used in cases of difficult-to-treat epilepsy.^10,11) We selected 2893 cases with levetiracetam use, excluding cases where three or more AEDs were administered together, and investigated which other AEDs and patient characteristics were significantly associated with aggression using a multivariate analysis with adjusted ROR.

We used the ICH Medical Dictionary for Regulatory Activities (MedDRA) v24.0 to extract the adverse events listed in the JADER database. Aggression was defined by preferred terms (PTs) included in the Standardized MedDRA Queries (SMQ), suggestive of hostility and aggression (SMQ: 20000142). All of the PTs included in SMQ 20000142 were reviewed against our knowledge. The following 11 types were specified: aggression (PT code: 10001488), anger (PT code: 10002368), antisocial behavior (PT code: 10002820), belligerence (PT code: 10004224), borderline personality (PT code: 10006034), homicidal ideation (PT code: 10049666), homicide (PT code: 10020364), hostility (PT code: 10020400), physical assault (PT code: 10034983), psychopathic personality (PT code: 10037218), and violence-related symptom (PT code: 10047426).

Data Extraction

The Japanese drug regulatory authority, the PMDA, collects cases with adverse events and makes them available to the public via the JADER database. The JADER database informs on post-marketing surveillance, spontaneous reports, and adverse drug reactions identified from clinical trials and obtained from descriptions of cases in medical publications. The database contains 667729 adverse events reported between April 2004 and December 2020. The relevant data were downloaded from the PMDA website (http://www.pmda.go.jp/). The JADER dataset consists of four tables containing the following information: 1) patient information, such as sex, age, and body weight; 2) patient drug information; 3) patient adverse events and outcomes; and 4) medical history and primary illness, inter alia. These four tables were integrated using the FUND E-Z Backup Archive (FUND E-Z Development Corporation, NY, U.S.A.).

Analysis of the ROR

The cases were classified (a–d), as follows:

(a) individuals who received the drugs and exhibited adverse events; b) individuals who received the drugs but did not exhibit adverse events; c) individuals who did not receive the drugs and exhibited adverse events; and d) individuals who did not receive the drugs and did not exhibit adverse events. The ROR was calculated using the equation below.¹⁵⁾ A signal was detected when the lower limits of the 95% CIs of the RORs exceeded one.

  

(1)

  

(2)

The RORs were expressed as point estimates with 95% CIs. The data were analyzed using Fisher’s exact test. The signal was considered positive if the lower limit of the 95% CI was >1, the p-value was <0.05, and the reported number was ≥2.

Analysis of Adjusted ROR

In this analysis, ROR allows for adjustments through logistic regression analysis and offers the advantage of covariate control. We calculated the adjusted ROR based on previous reports.¹⁶⁾ Reports were stratified by age as <60 year (y) and ≥60 y.¹⁷⁾ To construct the logistic model, sex (male), reporting year (defined as the year of reporting to PMDA), and stratified age groups were coded. We included the reporting year in the coding because we thought multiple AEDs might have been launched in recent years,⁸⁾ and might include reporting bias. The following logistic model was used for analysis:

  

(3)

where Y = reporting year, S = sex, A = stratified age group.

The ≥60 y group was used as a reference to calculate the RORs adjusted for age variations, and the sex = female group was used as a reference to calculate the RORs adjusted for sex variations; results with p < 0.05 were considered statistically significant.

Statistical Analyses

Statistical analyses were conducted using R software (version 3.3.1, R Foundation for Statistical Computing, Vienna, Austria. http://www.Rproject.org). Categorical variables were analyzed using Fisher’s exact test. The RORs and 95% CIs were calculated for each explanatory variable included in the multivariate model. Statistical significance was defined as p < 0.05. All data analyzed in this study were assessed using two or more independent analyses. All p-values less than 0.01 are expressed as p < 0.01.

RESULTS

Analysis of Possible Factors for Aggression

Of the 587017 cases reported between April 2004 and December 2020, 433 cases developed aggression. Significant aggression signals were detected for perampanel (crude ROR: 325.04, 95% CI: 118.48–752.58, p < 0.01), levetiracetam (crude ROR: 17.14, 95% CI: 10.33–26.90, p < 0.01), lacosamide (crude ROR: 16.90, 95% CI: 2.02–62.51, p < 0.01), lamotrigine (crude ROR: 15.98, 95% CI: 9.99–24.39, p < 0.01), valproate (crude ROR: 6.68, 95% CI: 4.27–10.02, p < 0.01), and carbamazepine (crude ROR: 2.47, 95% CI: 1.17–4.59, p < 0.01) (Table 1).

Table 1. Signals of Single-AED Induced Aggression

	Total	Case	Non-case	Ratio (%)	Crude ROR (95% CI)	p-Value
Total	587017	433	586584	0.074
Perampanel	37	7	30	18.92	325.04 (118.48–752.58)	<0.01
Levetiracetam	1673	20	1653	1.20	17.14 (10.33–26.90)	<0.01
Lacosamide	163	2	161	1.23	16.90 (2.02–62.51)	<0.01
Lamotrigine	2075	23	2052	1.11	15.98 (9.99–24.39)	<0.01
Valproate	5353	25	5328	0.47	6.68 (4.27–10.02)	<0.01
Carbamazepine	5575	10	5565	0.18	2.47 (1.17–4.59)	<0.01
Zonisamide	1395	2	1393	0.14	1.95 (0.23–7.10)	0.28
Gabapentin	768	1	767	0.13	1.77 (0.045–9.93)	0.43
Phenobarbital	1239	1	1238	0.08	1.09 (0.028–6.14)	0.6
Phenytoin	1622	1	1621	0.06	0.84 (0.021–4.68)	1.0
Topiramate	65	0	65	0	—	1.0
Clobazam	39	0	39	0	—	1.0

AED, antiepileptic drug; ROR, reporting odds ratio; 95% CI, 95% confidence interval; categorical variables were analyzed using Fisher’s exact test; RORs and 95% CIs were calculated for each explanatory variable included in the multivariate model; statistical significance was defined as p < 0.05.

Analysis of Aggression with the Combined Use of New-Generation AEDs

The adjusted RORs and 95% CIs of levetiracetam, which was used in combination with other AEDs, are summarized in Table 2. Of the 2893 cases that received levetiracetam and other AEDs, 48 developed aggression. The combined therapy with perampanel in patients receiving levetiracetam had a significant aggression signal (adjusted ROR: 25.90, 95% CI: 1.14–59.10, p < 0.01). Signals for aggression were evaluated with age as a factor, taking ≥60 y as a reference, and were as follows: age <60 y (adjusted ROR: 2.88, 95% CI: 1.49–5.56, p < 0.01), treated with levetiracetam.

Table 2. Signals of Combination-AED Induced Aggression with Levetiracetam

	Total	Case	Non-case	Ratio (%)	Crude ROR (95% CI)	Adjusted ROR (95% CI)	p-Value
Total	2893	48	2845	1.66
Reporting year	—	—	—	—	—	0.95 (0.82–1.10)	0.5
Male sex, n (%)	1585	32	1553	2.02	1.66 (0.88–3.26)	1.77 (0.95–3.30)	0.072
Age <60 y, n (%)	1418	33	1385	2.33	2.32 (1.22–4.61)	2.88 (1.49–5.56)	<0.01
Perampanel	63	12	51	19.05	18.19 (8.13–38.37)	25.90 (1.14–59.10)	<0.01
Carbamazepine	270	5	265	1.89	1.13 (0.35–2.89)	1.29 (0.47–3.57)	0.62
Phenobarbital	88	2	86	2.27	1.39 (0.16–5.49)	1.38 (0.31–6.13)	0.68
Valproate	311	7	304	2.25	1.43 (0.54–3.25)	1.47 (0.60–3.61)	0.41
Phenytoin	115	1	114	0.87	0.51 (0.013–3.04)	0.66 (0.086–5.11)	0.69
Zonisamide	91	1	90	1.10	0.65 (0.016–3.91)	0.73 (0.094–7.72)	0.77
Lacosamide	152	0	152	0	—	—	—
Lamotrigine	86	0	86	0	—	—	—
Clobazam	21	0	21	0	—	—	—
Gabapentin	12	0	12	0	—	—	—
Topiramate	11	0	11	0	—	—	—

ROR, reporting odds ratio; 95% CI, 95% confidence interval; categorical variables were analyzed using Fisher’s exact test; RORs and 95% CIs were calculated for each explanatory variable included in the multivariate model; statistical significance was defined as p < 0.05.

DISCUSSION

Aggression is common in patients with epilepsy who are treated with AEDs,^3,4) and is associated with a negative impact on QOL, poor adherence to AEDs, and difficulty in caring for such epileptic patients.^5–7) Epilepsy is very difficult to treat and often involves combination therapy of AEDs. However, the relationship between aggression and AED combinations is unclear because the number of cases used in a previous study was limited.¹¹⁾ Therefore, we quantified the aggression signals associated with single-AED therapy and the concomitant AED treatment using reporting-derived signals. The present study showed that the combined therapy with perampanel in patients receiving levetiracetam had a significant aggression signal. Furthermore, we demonstrated that aggression upon treatment with levetiracetam frequently occurred in patients <60 y old.

The single-AED signals associated with aggression identified six AEDs in total, including perampanel, levetiracetam, lacosamide, lamotrigine, valproate, and carbamazepine. The association of these six AEDs with aggression has been reported in previous studies, and the incidence of aggression ranged from about 0.1 to 10.0%, depending on the AED.¹⁸⁾ In particular, perampanel was a frequently reported cause of aggression (18.9%) in this study. Perampanel was only used in combination with other AEDs until recently. The aggression associated with perampanel may be overestimated because the number of reports of perampanel alone is still small. However, most systematic reviews also showed that perampanel was associated with higher rates of aggression when compared with the aggregate of the other AEDs.³⁾ Therefore, it may be necessary to monitor aggression when using perampanel.

The combined therapy with perampanel in patients receiving levetiracetam had a significant aggression signal in the present study. The incidence of aggression associated with the concomitant use of these AED may be due to the pharmacological action of the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) type glutamate receptor, which plays an essential role in aggression.¹⁹⁾ Previous studies have suggested that the antagonism of AMPA receptors has been linked to aggression observed with AED treatment.²⁰⁾ Perampanel is a non-competitive AMPA receptor antagonist. Levetiracetam is a synaptic vesicle glycoprotein 2A ligand with some activity toward AMPA-gated currents, which have been postulated to be associated with aggression.^21,22) Therefore, the combination of levetiracetam and perampanel may synergistically affect AMPA receptors and may trigger further aggression.

Aggression was associated with ages <60 y. It has been reported that the levels of the aforementioned neurotransmitters decrease with age.²²⁾ However, new-onset epilepsy in elderly patients is often associated with strokes, Alzheimer’s disease, dementia, or other medical co-morbidities.²³⁾ These conditions restrict activity levels, possibly reducing muscle mass and frailty.²⁴⁾ Therefore, the incidence of aggression associated with AEDs might be lower in elderly patients.

There were several limitations in this study. First, there was no evidence that the drug caused the reported event because the natural progression of the disease can cause confusion. The type of combination AEDs used may vary depending on the patient’s epilepsy type, and these factors may also affect AED-induced aggression. Second, the AED administration methods, such as intravenous or oral administration, were not evaluated in this study. The route of administration is a factor that may also affect AED-induced aggression, because the method of AED administration may vary depending on the patient’s conditions. Finally, cases in the JADER database are spontaneously reported, so there may be reporting bias. Only data on patients with side effects are available, and there is a concern that research data may target a patient group distinct from those generally present in actual clinical practice. Therefore, the results of this study need to be further investigated through new studies.

We investigated aggression associated with AEDs using the JADER database. We found that certain AEDs may be associated with aggression and specific patient characteristics and AED combinations. Our findings can be used to minimize the risk of aggression in patients being treated with AEDs.

Conflict of Interest

The authors declare no conflict of interest.

REFERENCES

• 1) Beghi E, Giussani G, Nichols E, et al. Global, regional, and national burden of epilepsy, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol., 18, 357–375 (2019).
• 2) Perucca P, Gilliam FG. Adverse effects of antiepileptic drugs. Lancet Neurol., 11, 792–802 (2012).
• 3) Steinhoff BJ, Klein P, Klitgaard H, Laloyaux C, Moseley BD, Ricchetti-Masterson K, Rosenow F, Sirven JI, Smith B, Stern JM, Toledo M, Zipfel PA, Villanueva V. Behavioral adverse events with Brivaracetam, levetiracetam, perampanel, and topiramate: a systematic review. Epilepsy Behav., 118, 107939 (2021).
• 4) Chen B, Choi H, Hirsch LJ, Katz A, Legge A, Buchsbaum R, Detyniecki K. Psychiatric and behavioral side effects of antiepileptic drugs in adults with epilepsy. Epilepsy Behav., 76, 24–31 (2017).
• 5) Schmitz B, Montouris G, Schäuble B, Caleo S. Assessing the unmet treatment need in partial-onset epilepsy: looking beyond seizure control. Epilepsia, 51, 2231–2240 (2010).
• 6) Perucca P, Carter J, Vahle V, Gilliam FG. Adverse antiepileptic drug effects: toward a clinically and neurobiologically relevant taxonomy. Neurology, 72, 1223–1229 (2009).
• 7) Kwan P, Brodie MJ. Early identification of refractory epilepsy. N. Engl. J. Med., 342, 314–319 (2000).
• 8) Kamitaki BK, Minacapelli CD, Zhang P, Wachuku C, Gupta K, Catalano C, Rustgi V. Drug-induced liver injury associated with antiseizure medications from the FDA Adverse Event Reporting System (FAERS). Epilepsy Behav., 117, 107832 (2021).
• 9) Carreño M. Levetiracetam. Drugs Today (Barc), 43, 769–794 (2007).
• 10) Kaminski RM, Matagne A, Patsalos PN, Klitgaard H. Benefit of combination therapy in epilepsy: a review of the preclinical evidence with levetiracetam. Epilepsia, 50, 387–397 (2009).
• 11) Hasegawa N, Tohyama J. Differences in levetiracetam and perampanel treatment-related irritability in patients with epilepsy. Epilepsy Behav., 116, 107644 (2021).
• 12) Ishida T, Kawada K, Morisawa S, Jobu K, Morita Y, Miyamura M. Risk factors for pseudoaldosteronism with yokukansan use: analysis using the Japanese Adverse Drug Report (JADER) database. Biol. Pharm. Bull., 43, 1570–1576 (2020).
• 13) van Puijenbroek EP, Bate A, Leufkens HG, Lindquist M, Orre R, Egberts AC. A comparison of measures of disproportionality for signal detection in spontaneous reporting systems for adverse drug reactions. Pharmacoepidemiol. Drug Saf., 11, 3–10 (2002).
• 14) Anzai T, Takahashi K, Watanabe M. Adverse reaction reports of neuroleptic malignant syndrome induced by atypical antipsychotic agents in the Japanese Adverse Drug Event Report (JADER) database. Psychiatry Clin. Neurosci., 73, 27–33 (2019).
• 15) Rothman KJ, Lanes S, Sacks ST. The reporting odds ratio and its advantages over the proportional reporting ratio. Pharmacoepidemiol. Drug Saf., 13, 519–523 (2004).
• 16) Suzuki Y, Suzuki H, Umetsu R, Uranishi H, Abe J, Nishibata Y, Sekiya Y, Miyamura N, Hara H, Tsuchiya T, Kinosada Y, Nakamura M. Analysis of the interaction between clopidogrel, aspirin, and proton pump inhibitors using the FDA adverse event reporting system database. Biol. Pharm. Bull., 38, 680–686 (2015).
• 17) Shimada K, Hasegawa S, Nakao S, Mukai R, Sasaoka S, Ueda N, Kato Y, Abe J, Mori T, Yoshimura T, Kinosada Y, Nakamura M. Adverse reaction profiles of hemorrhagic adverse reactions caused by direct oral anticoagulants analyzed using the Food and Drug Administration Adverse Event Reporting System (FAERS) database and the Japanese Adverse Drug Event Report (JADER) database. Int. J. Med. Sci., 16, 1295–1303 (2019).
• 18) Brodie MJ, Besag F, Ettinger AB, Mula M, Gobbi G, Comai S, Aldenkamp AP, Steinhoff BJ. Epilepsy, antiepileptic drugs, and aggression: an evidence-based review. Pharmacol. Rev., 68, 563–602 (2016).
• 19) Coccaro EF, Lee R, Vezina P. Cerebrospinal fluid glutamate concentration correlates with impulsive aggression in human subjects. J. Psychiatr. Res., 47, 1247–1253 (2013).
• 20) Hansen CC, Ljung H, Brodtkorb E, Reimers A. Mechanisms underlying aggressive behavior induced by antiepileptic drugs: focus on topiramate, levetiracetam, and perampanel. Behav. Neurol., 2018, 2064027 (2018).
• 21) Custer KL, Austin NS, Sullivan JM, Bajjalieh SM. Synaptic vesicle protein 2 enhances release probability at quiescent synapses. J. Neurosci., 26, 1303–1313 (2006).
• 22) Chang L, Jiang CS, Ernst T. Effects of age and sex on brain glutamate and other metabolites. Magn. Reson. Imaging, 27, 142–145 (2009).
• 23) Annegers JF, Rocca WA, Hauser WA. Causes of epilepsy: contributions of the Rochester epidemiology project. Mayo Clin. Proc., 71, 570–575 (1996).
• 24) Bortz WM 2nd. A conceptual framework of frailty: a review. J. Gerontol. A Biol. Sci. Med. Sci., 57, M283–M288 (2002).