- Email: [email protected]
Prospective study on the withdrawal and reinstitution of antiepileptic drugs among seizure-free patients in west China
Journal of Clinical Neuroscience 21 (2014) 997–1001
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Clinical Study
Prospective study on the withdrawal and reinstitution of antiepileptic drugs among seizure-free patients in west China Wei Li a,b, Yang Si a, Xue-mei Zou a, Dong-mei An a, Hui Yang b, Dong Zhou a,⇑ a b
Department of Neurology, West China Hospital, Sichuan University, Chengdu, Sichuan 610041, China Department of Neurosurgery, Xinqiao Hospital, The Third Military Medical University, Chongqing, China
a r t i c l e
i n f o
Article history: Received 15 April 2013 Accepted 30 September 2013
Keywords: Antiepileptic drugs Epilepsy Recurrence Reinstitution Risk factor Withdrawal
a b s t r a c t This study explored the relapse rates and risk factors for seizure recurrence after discontinuing antiepileptic drug (AED) therapy among seizure-free patients in west China, and explored whether to reinstitute AED immediately after a single seizure after AED withdrawal. Patients with epilepsy who were seizurefree for at least 2 years and decided to gradually stop AED therapy were followed up every 3 months for seizure relapse. Patients who experienced their first seizure after drug withdrawal were divided into two groups according to their willingness to reinstitute AED therapy, and were followed up until their second seizure. In the mean 29.35 months of follow-up, 37 patients (37/162, 22.8%) suffered at least one seizure after withdrawal. The cumulative probability of seizure recurrence was 16% at 12 months and 20.2% at 24 months. AED response time >1 year and multiple types of seizure were identified as risk factors for seizure recurrence. Eight patients (8/32, 25%) suffered a second seizure within 1 year after the first whether or not they reinstituted AED immediately. There were no significant demographic or clinical differences between patients who reinstituted AED therapy and those who did not. The epilepsy recurrence rate after AED withdrawal is relatively low, with a relatively slow tapering process. Patients with long AED response times and/or multiple types of seizures have a higher risk of seizure recurrence. The first seizure after drug withdrawal is not an indication for immediate AED reinstitution, but may be recommended after a second seizure. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Epilepsy is one of the most common chronic neurological disorders. About 70 million patients have epilepsy worldwide and 50.4/ 100,000 new cases occur every year [1]. Epilepsy causes serious physiologic, psychological, and economic burden [2]. Fortunately, approximately 70% of patients with recent-onset epilepsy become seizure-free with reasonable drug therapy [3]. The next consideration is whether these antiepileptic drugs (AED) should be withdrawn once the patient becomes seizure-free. Considering the side effects and huge cost of medication, AED should be withdrawn once they become unnecessary. The main barrier to this is the fear of relapse after discontinuing AED. In recent decades, many studies have focused on the risk of seizure recurrence after AED withdrawal, which varies from 12% to 67% [4], depending on the research method and population. Studies have also identified some relapse risk factors, including age at seizure onset, type and severity of epilepsy, duration of active epilepsy, period of seizure freedom on AED, multiple AED, abnormal ⇑ Corresponding author. Tel.: +86 139 8000 8088; fax: +86 028 8542 2893. E-mail address: [email protected] (D. Zhou). http://dx.doi.org/10.1016/j.jocn.2013.09.019 0967-5868/Ó 2013 Elsevier Ltd. All rights reserved.
electroencephalography (EEG) before drug withdrawal, presence of recognised epileptogenic lesions in neuroimaging, and certain epilepsy syndromes, such as juvenile myoclonic epilepsy [5–13]. Furthermore, after AED withdrawal, most patients choose to reinstitute therapy (restart AED or increase the dosage) immediately after a seizure occurs, and most achieve seizure control again [14,15]. However, some experts exclude single seizures as relapses: thus, they deem management unnecessary [16]. Additionally, some patients refuse to reinstitute AED due to their side effects. Few studies have focused on AED withdrawal in patients in west China. This study aimed to determine the incidence and risk factors of seizure recurrence after discontinuing AED among seizure-free patients in west China. We also sought to determine whether AED should be reinstituted immediately after a single seizure following AED withdrawal. 2. Methods 2.1. Patients The study population consisted of epilepsy patients from the Department of Neurology at West China Hospital between March
998
W. Li et al. / Journal of Clinical Neuroscience 21 (2014) 997–1001
2006 and June 2011. All patients in the study were candidates for AED withdrawal who were followed up every 3 months until October 2012. A total of 172 patients fulfilled the inclusion and exclusion criteria. The inclusion criteria were as follows: (1) diagnosis of epilepsy, defined as at least two unprovoked seizures at least 24 hours apart; (2) patients remained seizure-free for at least 24 consecutive months during AED therapy [9,11,17,18]; and (3) patients expressed a desire to gradually discontinue AED therapy and agreed to return for regular follow-up. The exclusion criteria were as follows: (1) patients with psychogenic seizures or uncertain diagnosis of epilepsy; and (2) patients who underwent epilepsy surgery.
spectively studied 162 patients. The characteristics of the subjects are shown in Table 1. Most patients were male and only 51 patients had adult onset (>18 years old at onset). The average time of followup was more than 2 years (29.35 months), and most seizure-free periods before withdrawal exceeded 3 years (42.33 months). The most common type of seizure was generalised seizures. Abnormal neurological imaging findings included cerebral dysplasia, calcifications, cysts, encephalomalacia, and hippocampal degeneration, whereas abnormal EEG consisted of spiked, sharp, or slow waves. Most patients achieved long-term seizure control with AED monotherapy, but only after more than 1 year of treatment.
3.2. Seizure recurrence 2.2. Methods All information was recorded on case-record forms by trained extractors. The demographic data and risk factors were collected as follows: sex; age at drug withdrawal; age at seizure onset; period between the start of AED and the last seizure before withdrawal (AED response time); seizure-free period before AED withdrawal; duration of follow-up after AED withdrawal; AED tapering off period (taper period); findings on brain CT scan and/ or MRI; EEG before drug withdrawal; seizure type (classified as generalised, partial, or multiple types based on history); number of episodes (frequency multiplied by duration of active seizures); and number of AED administered for long-term seizure control. The AED doses were slowly tapered off by one-quarter every 3 months in most patients [8]. Those taking one AED were tapered off for at least 6 months and those taking multiple AED were tapered off for 12 months. All patients were followed up every 3 months until the end of the study period. Patients who could not return to the hospital for follow-up were followed-up by telephone. Patients who experienced their first seizure after AED withdrawal were assigned to one of two groups according to their willingness to reinstitute AED therapy. The patients were again followed up for another 12 months or until their second seizure. All patients who suffered two or more seizures after withdrawal were advised to reinstitute AED therapy immediately. 2.3. Statistics The Statistical Package for the Social Sciences version 17.0 was used to analyse the data (SPSS, Chicago, IL, USA). Survival analysis was performed to determine seizure recurrence. Survival time was defined as the period between AED withdrawal and seizure recurrence or end of follow-up. The censored value was defined as the first seizure after withdrawal or being seizure-free at the end of follow-up. A survival curve was constructed to describe the recurrence rate after AED withdrawal. A log-rank test was used for univariate analysis and a Cox proportional hazard model was used for multivariate analysis to identify the risk factors for seizure recurrence. A chi-squared test (Fisher’s exact test) was used to distinguish the differences in the incidence of second seizures between groups that reinstituted AED and those that did not. All statistical tests were two-tailed. Differences between groups were considered significant at p < 0.05. 3. Results 3.1. Patient characteristics A total of 172 patients were enrolled from March 2006 to June 2011 in this study. Ten patients (5.8%) were excluded from the analysis because of incomplete follow-up information, meaning we pro-
A total of 37 patients (22.8%) experienced at least one seizure after AED withdrawal, with a mean follow-up time of 29.35 months. The details are shown in Table 2. The precipitating factors for seizure included staying up late, drinking wine, missing an AED dose, fever, and drinking coffee. Most patients (73%) chose to reinstitute the AED therapy immediately after their first seizure. A survival curve is presented in Figure 1. Most patients (70.3%) relapsed within 12 months
Table 1 Characteristics of epilepsy patients undergoing antiepileptic drug withdrawal Characteristic
Number of patients (%)/ Mean ± SD (range)
Patients Male Age at drug withdrawal, years Age at seizure onset, years Adult onset AED response time, months AED response time >1 year Seizure-free period, months Seizure-free period <3 years Duration of follow-up, months
162 82 (50.6%) 24.17 ± 13.27 16.79 ± 13.20 51 (31.5%) 18.41 ± 26.61 92 (56.8%) 42.33 ± 17.92 26 (16%) 29.35 ± 13.10
Taper period, months With one AED With multiple AED
10.65 ± 4.93 (4–30) 16.71 ± 6.60 (8–34)
Seizure type Generalised Partial Multiple types
105 (64.8%) 33 (20.4%) 24 (14.8%)
Number of AED One Multiple Abnormal findings on neuroimaging Abnormal EEG before drug withdrawal
111 (67.3%) 55 (32.7%) 20 (12.3%) 30 (18.5%)
(4–79) (1–74) (0–138) (24–168) (15–79)
AED = antiepileptic drug, EEG = electroencephalography, SD = standard deviation.
Table 2 Characteristics of epilepsy patients who relapsed after antiepileptic drug withdrawal
Patients Male Adult onset AED response time >1 year Seizure-free period <3 year Multiple AED Multiple types of seizure Abnormal findings in neuroimaging Abnormal EEG before withdrawal With precipitating factor AED reinstitution
Patients, n
Percent
37 17 14 20 5 17 9 5 6 13 27
100.0 45.9 37.8 54.1 13.5 45.9 24.3 13.5 16.2 35.1 73.0
AED = antiepileptic drug, EEG = electroencephalography.
W. Li et al. / Journal of Clinical Neuroscience 21 (2014) 997–1001
999
Fig. 1. Cumulative probability of remaining seizure-free after antiepileptic drug withdrawal.
Table 3 Univariate analysis by log rank test of risk factors for seizure recurrence
Male Adult onset AED response time >1 year Seizure-free period <3 year Multiple AED Multiple types of seizure Abnormal findings in neuroimaging Abnormal EEG before withdrawal
after AED withdrawal. The cumulative probability of seizure relapse was 16% at 12 months and 20.2% at 24 months.
Patients, n
Relapsed, n
Relapse rate (%)
82 51 57 26 55 24 20
17 14 20 5 17 9 5
20.7 27.5 35.1 19.2 30.9 37.5 25.0
0.636 0.158 0.006 0.381 0.256 0.019 0.710
30
6
20.0
0.693
AED = antiepileptic drug, EEG = electroencephalography.
p value
3.3. Risk factors Univariate analysis was performed to determine the risk factors for seizure recurrence. We identified two risk factors, namely, AED response time >1 year and multiple types of seizures (Table 3). Figure 2 shows the cumulative probabilities of remaining seizure-free with different AED response times, whereas Figure 3 shows them with different seizure types. Multivariate analysis was also performed (Table 4). The difference between varying AED response times reached statistical significance (p < 0.05), with a hazard ratio of 2.096.
Fig. 2. Cumulative probabilities of remaining seizure-free for patients with different antiepileptic drug response times. AED = antiepileptic drug.
1000
W. Li et al. / Journal of Clinical Neuroscience 21 (2014) 997–1001
Table 4 Multivariate analysis by Cox proportional hazard model of risk factors for seizure recurrence HR Adult onset AED response time >1 year Seizure-free period <3 years Multiple AED Multiple types of seizure
0.627 2.096 0.848 1.281 1.455
95% CI for HR
p value
0.315–1.248 1.026–4.279 0.315–2.281 0.598–2.740 0.623–3.398
0.184 0.042 0.744 0.524 0.386
AED = antiepileptic drug, CI = confidence interval, HR = hazard ratio.
3.4. Second seizure after withdrawal Only 32 patients completed follow-up after their first seizure relapse. Among them, eight suffered their second seizure within 1 year. Given the small sample size, a chi-squared test (Fisher’s exact test) was used to analyze the differences between patients who reinstituted AED therapy and those who did not. The difference was not statistically significant (p > 0.05; Table 5). No other statistically significant differences were found between the groups. 4. Discussion After achieving prolonged seizure-free intervals through AED therapy, most patients consider stopping drug therapy. As reported, the relapse rate after AED withdrawal varies from 12% to 67%. In our study, 37 patients experienced at least one seizure after withdrawal within the mean follow-up time of 29.35 months. The total recurrence rate was 22.8% and most relapses happened within 1 year after withdrawal. The cumulative probability of seizure recurrence was 16% at 12 months and 20.2% at 24 months. This result is similar to that of another study [19], but lower than those of others [5,8,9,11,12]. The characteristics of our study population may have contributed to the lower incidence. Most of the patients were children at the time of seizure onset, which confers a better prognosis after AED withdrawal [5,22]. And the proportion of patients with partial seizures, which is related to a high relapse risk [5,8,9], was only 20.4%. Although the minimum seizure-free period in the study was 24 months, 84% of patients were seizure-free for more than 3 years. The mean seizure-free period before AED withdrawal
Table 5 Chi-squared test (Fisher’s exact test) of second seizure incidence after antiepileptic drug withdrawal
Adult onset AED response time >1 year Seizure free <3 year Multiple AED Multiple types of seizure With precipitating factor AED reinstitution
Patients with single seizure (n = 24)
Patients with multiple seizures (n = 8)
Exact significance (two-sided)
9 11 3 11 7 8 17
4 6 1 4 2 2 6
0.684 0.229 1.000 1.000 1.000 1.000 1.000
AED = antiepileptic drug.
was 42.33 months, which is relatively long and is associated with lower seizure recurrence [8,22]. In addition, most patients achieved long-term seizure control with monotherapy, which is consistent with the findings of Kwan and Brodie [3], and these patients had a low relapse rate. Our slow tapering off process also has likely caused this difference. We advised our patients to stop the AED slowly, over at least 6 months for one AED and 12 months for multiple AED. Our average taper period was 10.65 months among patients taking one AED and 16.71 months for patients taking multiple AED. This tapering off process allows patients to adapt to the withdrawal and minimises withdrawal reactions. Although Tennison et al. [20] and Serra et al. [21] found that the taper period of AED did not influence the risk of seizure recurrence, most of their taper periods were less than 9 months, which is shorter than in the current study. Therefore, we believe that longer taper periods may be related to our relatively lower relapse rates. Many studies have identified risk factors for seizure relapse. The age at seizure onset, type and severity of epilepsy, duration of active epilepsy, period of seizure freedom on AED, multiple AED, abnormal EEG before drug withdrawal, presence of recognised epileptogenic lesions on neurological imaging, and certain epilepsy syndromes, such as juvenile myoclonic epilepsy, are all reported factors [5–13]. Our findings suggest that higher relapse risk relates to longer AED response time and to multiple seizure types. Univariate analysis and multivariate analysis identified long AED response times (AED response time >1 year) as a risk factor for seizure
Fig. 3. Cumulative probabilities of remaining seizure-free for patients with different seizure types.
W. Li et al. / Journal of Clinical Neuroscience 21 (2014) 997–1001
recurrence, with a hazard ratio of 2.096. This indicates its predictive value for seizure recurrence, with patients who have long AED response times having double risk of relapse after AED withdrawal. Long AED response times indicate the relative severity of the disease and/or that the drugs are inappropriate. This also emphasizes the importance of controlling seizures as soon as possible. Univariate analysis also identified multiple seizure types as a risk factor for seizure recurrence. This also reflects disease severity and difficulty with control of the disease, which has been previously reported [8,9,11]. Therefore in patients with long AED response times and/or multiple seizure types, physicians should be careful when deciding whether to stop AED, because of the higher risks of seizure relapse after drug withdrawal. Within the limitations of this study, we did not find any other risk factors related to sex, adult onset epilepsy, short seizure-free periods, partial seizures, multiple AED, abnormal findings on neurological imaging, or abnormal EEG before drug withdrawal. Considering the debate on whether to reinstitute AED immediately after the first seizure after drug withdrawal, we also explored the influence of AED reinstitution on the incidence of second seizures. Although most patients (73%) chose to reinstitute the AED immediately after the first seizure, eight patients still suffered subsequent seizures within the 12 month follow-up period. The present study failed to show any statistically significant difference in second seizure incidence between patients who reinstituted AED therapy and those who did not. This may indicate that a single seizure after withdrawal is not an indication for immediate AED reinstitution, which supports the findings of Sillanpää and Schmidt [16]. A single seizure may be an occasional response to withdrawal, similar to reactions to the withdrawal of alcohol or antipsychotics. Our study has several limitations. The primary limitation is the small number of patients studied. Definitive conclusions require a larger sample size, especially from multicentre randomised control trials. The short follow-up time (mean: 29.35 months; range: 15– 79 months) is another limitation because longer follow-up periods more accurately reflect the prognosis of AED withdrawal. In addition, because all of the patients were seizure-free for many years, some of the baseline epilepsy information may have been affected by recall bias. Therefore, a database established from the onset of epilepsy to long-term follow-up would be helpful. These factors will be considered in subsequent studies.
Conflicts of Interest/Disclosures The authors declare that they have no financial or other conflicts of interest in relation to this research and its publication.
1001
Acknowledgments We would like to thank all colleagues who took part in this study. References [1] Ngugi AK, Kariuki SM, Bottomley C, et al. Estimation of the burden of active and lifetime epilepsy: a meta-analytic approach. Epilepsia 2010;51:883–90. [2] Radhakrishnan K. Challenges in the management of epilepsy in resource-poor countries. Nat Rev Neurol 2009;5:323–30. [3] Kwan P, Brodie MJ. Early identification of refractory epilepsy. N Engl J Med 2000;342:314–9. [4] Shih JJ, Ochoa JG. A systematic review of antiepileptic drug initiation and withdrawal. Neurologist 2009;15:122–31. [5] Berg AT, Shinnar S. Relapse following discontinuation of antiepileptic drugs: a meta-analysis. Neurology 1994;44:601–8. [6] Verrotti A, Trotta D, Salladini C, et al. Risk factors for recurrence of epilepsy and withdrawal of antiepileptic therapy: a practical approach. Ann Med 2003;35:207–15. [7] Aktekin B, Dogan EA, Oguz Y, et al. Withdrawal of antiepileptic drugs in adult patients free of seizures for 4 years: a prospective study. Epilepsy Behav 2006;8:616–9. [8] Specchio LM, Tramacere L, La Neve A, et al. Discontinuing antiepileptic drugs in patients who are seizure-free on monotherapy. J Neurol Neurosurg Psychiatry 2002;72:22–5. [9] Specchio LM, Beghi E. Should antiepileptic drugs be withdrawn in seizure-free patients? CNS Drugs 2004;18:201–12. [10] Ohta H, Ohtsuka Y, Tsuda T, et al. Prognosis after withdrawal of antiepileptic drugs in childhood-onset cryptogenic localization-related epilepsies. Brain Dev 2004;26:19–25. [11] Quality Standards Subcommittee of the American Academy of Neurology. Practice parameter: a guideline for discontinuing antiepileptic drugs in seizure-free patients—summary statement. Neurology 1996;47:600–2. [12] Medical Research Council Antiepileptic Drug Withdrawal Group. Randomized study of antiepileptic drug withdrawal in patients in remission. Lancet 1991;337:1175–80. [13] Cardoso TA, Coan AC, Kobayashi E, et al. Hippocampal abnormalities and seizure recurrence after antiepileptic drug withdrawal. Neurology 2006;67:134–6. [14] Schmidt D, Löscher W. Uncontrolled epilepsy following discontinuation of antiepileptic drugs in seizure-free patients: a review of current clinical experience. Acta Neurol Scand 2005;111:291–300. [15] Camfield P, Camfield C. The frequency of intractable seizures after stopping AED in seizure-free children with epilepsy. Neurology 2005;64:973–5. [16] Sillanpää M, Schmidt D. Prognosis of seizure recurrence after stopping antiepileptic drugs in seizure-free patients: a long-term population-based study of childhood-onset epilepsy. Epilepsy Behav. 2006;8:713–9. [17] Camfield P, Camfield C. When is it safe to discontinue AED treatment? Epilepsia 2008;49:25–8. [18] Sirven J, Sperling MR, Wingerchuk DM. Early versus late antiepileptic drug withdrawal for people with epilepsy in remission. The Cochrane Library 2010;2010:CD001902. [19] Lossius MI, Hessen E, Mowinckel P, et al. Consequences of antiepileptic drug withdrawal: a randomized, double-blind study. Epilepsia 2008;49:455–63. [20] Tennison M, Greenwood R, Lewis D, et al. Discontinuing antiepileptic drugs in children with epilepsy. A comparison of a six-week and a nine-month taper period. N Engl J Med 1994;330:1407–10. [21] Serra JG, Montenegro MA, Guerreiro MM. Antiepileptic drug withdrawal in childhood: does the duration of tapering off matter for seizure recurrence? J Child Neurol 2005;20:624–6. [22] Berg AT, Shinnar S, Levy SR, et al. Two-years remission and subsequent relapse in children with newly diagnosed epilepsy. Epilepsia 2001;42:1553–62.
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Clinical Study
Prospective study on the withdrawal and reinstitution of antiepileptic drugs among seizure-free patients in west China Wei Li a,b, Yang Si a, Xue-mei Zou a, Dong-mei An a, Hui Yang b, Dong Zhou a,⇑ a b
Department of Neurology, West China Hospital, Sichuan University, Chengdu, Sichuan 610041, China Department of Neurosurgery, Xinqiao Hospital, The Third Military Medical University, Chongqing, China
a r t i c l e
i n f o
Article history: Received 15 April 2013 Accepted 30 September 2013
Keywords: Antiepileptic drugs Epilepsy Recurrence Reinstitution Risk factor Withdrawal
a b s t r a c t This study explored the relapse rates and risk factors for seizure recurrence after discontinuing antiepileptic drug (AED) therapy among seizure-free patients in west China, and explored whether to reinstitute AED immediately after a single seizure after AED withdrawal. Patients with epilepsy who were seizurefree for at least 2 years and decided to gradually stop AED therapy were followed up every 3 months for seizure relapse. Patients who experienced their first seizure after drug withdrawal were divided into two groups according to their willingness to reinstitute AED therapy, and were followed up until their second seizure. In the mean 29.35 months of follow-up, 37 patients (37/162, 22.8%) suffered at least one seizure after withdrawal. The cumulative probability of seizure recurrence was 16% at 12 months and 20.2% at 24 months. AED response time >1 year and multiple types of seizure were identified as risk factors for seizure recurrence. Eight patients (8/32, 25%) suffered a second seizure within 1 year after the first whether or not they reinstituted AED immediately. There were no significant demographic or clinical differences between patients who reinstituted AED therapy and those who did not. The epilepsy recurrence rate after AED withdrawal is relatively low, with a relatively slow tapering process. Patients with long AED response times and/or multiple types of seizures have a higher risk of seizure recurrence. The first seizure after drug withdrawal is not an indication for immediate AED reinstitution, but may be recommended after a second seizure. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Epilepsy is one of the most common chronic neurological disorders. About 70 million patients have epilepsy worldwide and 50.4/ 100,000 new cases occur every year [1]. Epilepsy causes serious physiologic, psychological, and economic burden [2]. Fortunately, approximately 70% of patients with recent-onset epilepsy become seizure-free with reasonable drug therapy [3]. The next consideration is whether these antiepileptic drugs (AED) should be withdrawn once the patient becomes seizure-free. Considering the side effects and huge cost of medication, AED should be withdrawn once they become unnecessary. The main barrier to this is the fear of relapse after discontinuing AED. In recent decades, many studies have focused on the risk of seizure recurrence after AED withdrawal, which varies from 12% to 67% [4], depending on the research method and population. Studies have also identified some relapse risk factors, including age at seizure onset, type and severity of epilepsy, duration of active epilepsy, period of seizure freedom on AED, multiple AED, abnormal ⇑ Corresponding author. Tel.: +86 139 8000 8088; fax: +86 028 8542 2893. E-mail address: [email protected] (D. Zhou). http://dx.doi.org/10.1016/j.jocn.2013.09.019 0967-5868/Ó 2013 Elsevier Ltd. All rights reserved.
electroencephalography (EEG) before drug withdrawal, presence of recognised epileptogenic lesions in neuroimaging, and certain epilepsy syndromes, such as juvenile myoclonic epilepsy [5–13]. Furthermore, after AED withdrawal, most patients choose to reinstitute therapy (restart AED or increase the dosage) immediately after a seizure occurs, and most achieve seizure control again [14,15]. However, some experts exclude single seizures as relapses: thus, they deem management unnecessary [16]. Additionally, some patients refuse to reinstitute AED due to their side effects. Few studies have focused on AED withdrawal in patients in west China. This study aimed to determine the incidence and risk factors of seizure recurrence after discontinuing AED among seizure-free patients in west China. We also sought to determine whether AED should be reinstituted immediately after a single seizure following AED withdrawal. 2. Methods 2.1. Patients The study population consisted of epilepsy patients from the Department of Neurology at West China Hospital between March
998
W. Li et al. / Journal of Clinical Neuroscience 21 (2014) 997–1001
2006 and June 2011. All patients in the study were candidates for AED withdrawal who were followed up every 3 months until October 2012. A total of 172 patients fulfilled the inclusion and exclusion criteria. The inclusion criteria were as follows: (1) diagnosis of epilepsy, defined as at least two unprovoked seizures at least 24 hours apart; (2) patients remained seizure-free for at least 24 consecutive months during AED therapy [9,11,17,18]; and (3) patients expressed a desire to gradually discontinue AED therapy and agreed to return for regular follow-up. The exclusion criteria were as follows: (1) patients with psychogenic seizures or uncertain diagnosis of epilepsy; and (2) patients who underwent epilepsy surgery.
spectively studied 162 patients. The characteristics of the subjects are shown in Table 1. Most patients were male and only 51 patients had adult onset (>18 years old at onset). The average time of followup was more than 2 years (29.35 months), and most seizure-free periods before withdrawal exceeded 3 years (42.33 months). The most common type of seizure was generalised seizures. Abnormal neurological imaging findings included cerebral dysplasia, calcifications, cysts, encephalomalacia, and hippocampal degeneration, whereas abnormal EEG consisted of spiked, sharp, or slow waves. Most patients achieved long-term seizure control with AED monotherapy, but only after more than 1 year of treatment.
3.2. Seizure recurrence 2.2. Methods All information was recorded on case-record forms by trained extractors. The demographic data and risk factors were collected as follows: sex; age at drug withdrawal; age at seizure onset; period between the start of AED and the last seizure before withdrawal (AED response time); seizure-free period before AED withdrawal; duration of follow-up after AED withdrawal; AED tapering off period (taper period); findings on brain CT scan and/ or MRI; EEG before drug withdrawal; seizure type (classified as generalised, partial, or multiple types based on history); number of episodes (frequency multiplied by duration of active seizures); and number of AED administered for long-term seizure control. The AED doses were slowly tapered off by one-quarter every 3 months in most patients [8]. Those taking one AED were tapered off for at least 6 months and those taking multiple AED were tapered off for 12 months. All patients were followed up every 3 months until the end of the study period. Patients who could not return to the hospital for follow-up were followed-up by telephone. Patients who experienced their first seizure after AED withdrawal were assigned to one of two groups according to their willingness to reinstitute AED therapy. The patients were again followed up for another 12 months or until their second seizure. All patients who suffered two or more seizures after withdrawal were advised to reinstitute AED therapy immediately. 2.3. Statistics The Statistical Package for the Social Sciences version 17.0 was used to analyse the data (SPSS, Chicago, IL, USA). Survival analysis was performed to determine seizure recurrence. Survival time was defined as the period between AED withdrawal and seizure recurrence or end of follow-up. The censored value was defined as the first seizure after withdrawal or being seizure-free at the end of follow-up. A survival curve was constructed to describe the recurrence rate after AED withdrawal. A log-rank test was used for univariate analysis and a Cox proportional hazard model was used for multivariate analysis to identify the risk factors for seizure recurrence. A chi-squared test (Fisher’s exact test) was used to distinguish the differences in the incidence of second seizures between groups that reinstituted AED and those that did not. All statistical tests were two-tailed. Differences between groups were considered significant at p < 0.05. 3. Results 3.1. Patient characteristics A total of 172 patients were enrolled from March 2006 to June 2011 in this study. Ten patients (5.8%) were excluded from the analysis because of incomplete follow-up information, meaning we pro-
A total of 37 patients (22.8%) experienced at least one seizure after AED withdrawal, with a mean follow-up time of 29.35 months. The details are shown in Table 2. The precipitating factors for seizure included staying up late, drinking wine, missing an AED dose, fever, and drinking coffee. Most patients (73%) chose to reinstitute the AED therapy immediately after their first seizure. A survival curve is presented in Figure 1. Most patients (70.3%) relapsed within 12 months
Table 1 Characteristics of epilepsy patients undergoing antiepileptic drug withdrawal Characteristic
Number of patients (%)/ Mean ± SD (range)
Patients Male Age at drug withdrawal, years Age at seizure onset, years Adult onset AED response time, months AED response time >1 year Seizure-free period, months Seizure-free period <3 years Duration of follow-up, months
162 82 (50.6%) 24.17 ± 13.27 16.79 ± 13.20 51 (31.5%) 18.41 ± 26.61 92 (56.8%) 42.33 ± 17.92 26 (16%) 29.35 ± 13.10
Taper period, months With one AED With multiple AED
10.65 ± 4.93 (4–30) 16.71 ± 6.60 (8–34)
Seizure type Generalised Partial Multiple types
105 (64.8%) 33 (20.4%) 24 (14.8%)
Number of AED One Multiple Abnormal findings on neuroimaging Abnormal EEG before drug withdrawal
111 (67.3%) 55 (32.7%) 20 (12.3%) 30 (18.5%)
(4–79) (1–74) (0–138) (24–168) (15–79)
AED = antiepileptic drug, EEG = electroencephalography, SD = standard deviation.
Table 2 Characteristics of epilepsy patients who relapsed after antiepileptic drug withdrawal
Patients Male Adult onset AED response time >1 year Seizure-free period <3 year Multiple AED Multiple types of seizure Abnormal findings in neuroimaging Abnormal EEG before withdrawal With precipitating factor AED reinstitution
Patients, n
Percent
37 17 14 20 5 17 9 5 6 13 27
100.0 45.9 37.8 54.1 13.5 45.9 24.3 13.5 16.2 35.1 73.0
AED = antiepileptic drug, EEG = electroencephalography.
W. Li et al. / Journal of Clinical Neuroscience 21 (2014) 997–1001
999
Fig. 1. Cumulative probability of remaining seizure-free after antiepileptic drug withdrawal.
Table 3 Univariate analysis by log rank test of risk factors for seizure recurrence
Male Adult onset AED response time >1 year Seizure-free period <3 year Multiple AED Multiple types of seizure Abnormal findings in neuroimaging Abnormal EEG before withdrawal
after AED withdrawal. The cumulative probability of seizure relapse was 16% at 12 months and 20.2% at 24 months.
Patients, n
Relapsed, n
Relapse rate (%)
82 51 57 26 55 24 20
17 14 20 5 17 9 5
20.7 27.5 35.1 19.2 30.9 37.5 25.0
0.636 0.158 0.006 0.381 0.256 0.019 0.710
30
6
20.0
0.693
AED = antiepileptic drug, EEG = electroencephalography.
p value
3.3. Risk factors Univariate analysis was performed to determine the risk factors for seizure recurrence. We identified two risk factors, namely, AED response time >1 year and multiple types of seizures (Table 3). Figure 2 shows the cumulative probabilities of remaining seizure-free with different AED response times, whereas Figure 3 shows them with different seizure types. Multivariate analysis was also performed (Table 4). The difference between varying AED response times reached statistical significance (p < 0.05), with a hazard ratio of 2.096.
Fig. 2. Cumulative probabilities of remaining seizure-free for patients with different antiepileptic drug response times. AED = antiepileptic drug.
1000
W. Li et al. / Journal of Clinical Neuroscience 21 (2014) 997–1001
Table 4 Multivariate analysis by Cox proportional hazard model of risk factors for seizure recurrence HR Adult onset AED response time >1 year Seizure-free period <3 years Multiple AED Multiple types of seizure
0.627 2.096 0.848 1.281 1.455
95% CI for HR
p value
0.315–1.248 1.026–4.279 0.315–2.281 0.598–2.740 0.623–3.398
0.184 0.042 0.744 0.524 0.386
AED = antiepileptic drug, CI = confidence interval, HR = hazard ratio.
3.4. Second seizure after withdrawal Only 32 patients completed follow-up after their first seizure relapse. Among them, eight suffered their second seizure within 1 year. Given the small sample size, a chi-squared test (Fisher’s exact test) was used to analyze the differences between patients who reinstituted AED therapy and those who did not. The difference was not statistically significant (p > 0.05; Table 5). No other statistically significant differences were found between the groups. 4. Discussion After achieving prolonged seizure-free intervals through AED therapy, most patients consider stopping drug therapy. As reported, the relapse rate after AED withdrawal varies from 12% to 67%. In our study, 37 patients experienced at least one seizure after withdrawal within the mean follow-up time of 29.35 months. The total recurrence rate was 22.8% and most relapses happened within 1 year after withdrawal. The cumulative probability of seizure recurrence was 16% at 12 months and 20.2% at 24 months. This result is similar to that of another study [19], but lower than those of others [5,8,9,11,12]. The characteristics of our study population may have contributed to the lower incidence. Most of the patients were children at the time of seizure onset, which confers a better prognosis after AED withdrawal [5,22]. And the proportion of patients with partial seizures, which is related to a high relapse risk [5,8,9], was only 20.4%. Although the minimum seizure-free period in the study was 24 months, 84% of patients were seizure-free for more than 3 years. The mean seizure-free period before AED withdrawal
Table 5 Chi-squared test (Fisher’s exact test) of second seizure incidence after antiepileptic drug withdrawal
Adult onset AED response time >1 year Seizure free <3 year Multiple AED Multiple types of seizure With precipitating factor AED reinstitution
Patients with single seizure (n = 24)
Patients with multiple seizures (n = 8)
Exact significance (two-sided)
9 11 3 11 7 8 17
4 6 1 4 2 2 6
0.684 0.229 1.000 1.000 1.000 1.000 1.000
AED = antiepileptic drug.
was 42.33 months, which is relatively long and is associated with lower seizure recurrence [8,22]. In addition, most patients achieved long-term seizure control with monotherapy, which is consistent with the findings of Kwan and Brodie [3], and these patients had a low relapse rate. Our slow tapering off process also has likely caused this difference. We advised our patients to stop the AED slowly, over at least 6 months for one AED and 12 months for multiple AED. Our average taper period was 10.65 months among patients taking one AED and 16.71 months for patients taking multiple AED. This tapering off process allows patients to adapt to the withdrawal and minimises withdrawal reactions. Although Tennison et al. [20] and Serra et al. [21] found that the taper period of AED did not influence the risk of seizure recurrence, most of their taper periods were less than 9 months, which is shorter than in the current study. Therefore, we believe that longer taper periods may be related to our relatively lower relapse rates. Many studies have identified risk factors for seizure relapse. The age at seizure onset, type and severity of epilepsy, duration of active epilepsy, period of seizure freedom on AED, multiple AED, abnormal EEG before drug withdrawal, presence of recognised epileptogenic lesions on neurological imaging, and certain epilepsy syndromes, such as juvenile myoclonic epilepsy, are all reported factors [5–13]. Our findings suggest that higher relapse risk relates to longer AED response time and to multiple seizure types. Univariate analysis and multivariate analysis identified long AED response times (AED response time >1 year) as a risk factor for seizure
Fig. 3. Cumulative probabilities of remaining seizure-free for patients with different seizure types.
W. Li et al. / Journal of Clinical Neuroscience 21 (2014) 997–1001
recurrence, with a hazard ratio of 2.096. This indicates its predictive value for seizure recurrence, with patients who have long AED response times having double risk of relapse after AED withdrawal. Long AED response times indicate the relative severity of the disease and/or that the drugs are inappropriate. This also emphasizes the importance of controlling seizures as soon as possible. Univariate analysis also identified multiple seizure types as a risk factor for seizure recurrence. This also reflects disease severity and difficulty with control of the disease, which has been previously reported [8,9,11]. Therefore in patients with long AED response times and/or multiple seizure types, physicians should be careful when deciding whether to stop AED, because of the higher risks of seizure relapse after drug withdrawal. Within the limitations of this study, we did not find any other risk factors related to sex, adult onset epilepsy, short seizure-free periods, partial seizures, multiple AED, abnormal findings on neurological imaging, or abnormal EEG before drug withdrawal. Considering the debate on whether to reinstitute AED immediately after the first seizure after drug withdrawal, we also explored the influence of AED reinstitution on the incidence of second seizures. Although most patients (73%) chose to reinstitute the AED immediately after the first seizure, eight patients still suffered subsequent seizures within the 12 month follow-up period. The present study failed to show any statistically significant difference in second seizure incidence between patients who reinstituted AED therapy and those who did not. This may indicate that a single seizure after withdrawal is not an indication for immediate AED reinstitution, which supports the findings of Sillanpää and Schmidt [16]. A single seizure may be an occasional response to withdrawal, similar to reactions to the withdrawal of alcohol or antipsychotics. Our study has several limitations. The primary limitation is the small number of patients studied. Definitive conclusions require a larger sample size, especially from multicentre randomised control trials. The short follow-up time (mean: 29.35 months; range: 15– 79 months) is another limitation because longer follow-up periods more accurately reflect the prognosis of AED withdrawal. In addition, because all of the patients were seizure-free for many years, some of the baseline epilepsy information may have been affected by recall bias. Therefore, a database established from the onset of epilepsy to long-term follow-up would be helpful. These factors will be considered in subsequent studies.
Conflicts of Interest/Disclosures The authors declare that they have no financial or other conflicts of interest in relation to this research and its publication.
1001
Acknowledgments We would like to thank all colleagues who took part in this study. References [1] Ngugi AK, Kariuki SM, Bottomley C, et al. Estimation of the burden of active and lifetime epilepsy: a meta-analytic approach. Epilepsia 2010;51:883–90. [2] Radhakrishnan K. Challenges in the management of epilepsy in resource-poor countries. Nat Rev Neurol 2009;5:323–30. [3] Kwan P, Brodie MJ. Early identification of refractory epilepsy. N Engl J Med 2000;342:314–9. [4] Shih JJ, Ochoa JG. A systematic review of antiepileptic drug initiation and withdrawal. Neurologist 2009;15:122–31. [5] Berg AT, Shinnar S. Relapse following discontinuation of antiepileptic drugs: a meta-analysis. Neurology 1994;44:601–8. [6] Verrotti A, Trotta D, Salladini C, et al. Risk factors for recurrence of epilepsy and withdrawal of antiepileptic therapy: a practical approach. Ann Med 2003;35:207–15. [7] Aktekin B, Dogan EA, Oguz Y, et al. Withdrawal of antiepileptic drugs in adult patients free of seizures for 4 years: a prospective study. Epilepsy Behav 2006;8:616–9. [8] Specchio LM, Tramacere L, La Neve A, et al. Discontinuing antiepileptic drugs in patients who are seizure-free on monotherapy. J Neurol Neurosurg Psychiatry 2002;72:22–5. [9] Specchio LM, Beghi E. Should antiepileptic drugs be withdrawn in seizure-free patients? CNS Drugs 2004;18:201–12. [10] Ohta H, Ohtsuka Y, Tsuda T, et al. Prognosis after withdrawal of antiepileptic drugs in childhood-onset cryptogenic localization-related epilepsies. Brain Dev 2004;26:19–25. [11] Quality Standards Subcommittee of the American Academy of Neurology. Practice parameter: a guideline for discontinuing antiepileptic drugs in seizure-free patients—summary statement. Neurology 1996;47:600–2. [12] Medical Research Council Antiepileptic Drug Withdrawal Group. Randomized study of antiepileptic drug withdrawal in patients in remission. Lancet 1991;337:1175–80. [13] Cardoso TA, Coan AC, Kobayashi E, et al. Hippocampal abnormalities and seizure recurrence after antiepileptic drug withdrawal. Neurology 2006;67:134–6. [14] Schmidt D, Löscher W. Uncontrolled epilepsy following discontinuation of antiepileptic drugs in seizure-free patients: a review of current clinical experience. Acta Neurol Scand 2005;111:291–300. [15] Camfield P, Camfield C. The frequency of intractable seizures after stopping AED in seizure-free children with epilepsy. Neurology 2005;64:973–5. [16] Sillanpää M, Schmidt D. Prognosis of seizure recurrence after stopping antiepileptic drugs in seizure-free patients: a long-term population-based study of childhood-onset epilepsy. Epilepsy Behav. 2006;8:713–9. [17] Camfield P, Camfield C. When is it safe to discontinue AED treatment? Epilepsia 2008;49:25–8. [18] Sirven J, Sperling MR, Wingerchuk DM. Early versus late antiepileptic drug withdrawal for people with epilepsy in remission. The Cochrane Library 2010;2010:CD001902. [19] Lossius MI, Hessen E, Mowinckel P, et al. Consequences of antiepileptic drug withdrawal: a randomized, double-blind study. Epilepsia 2008;49:455–63. [20] Tennison M, Greenwood R, Lewis D, et al. Discontinuing antiepileptic drugs in children with epilepsy. A comparison of a six-week and a nine-month taper period. N Engl J Med 1994;330:1407–10. [21] Serra JG, Montenegro MA, Guerreiro MM. Antiepileptic drug withdrawal in childhood: does the duration of tapering off matter for seizure recurrence? J Child Neurol 2005;20:624–6. [22] Berg AT, Shinnar S, Levy SR, et al. Two-years remission and subsequent relapse in children with newly diagnosed epilepsy. Epilepsia 2001;42:1553–62.