Antiepileptic drug withdrawal: literature review

Predicting efficacy of antiseizure medication treatment with machine learning algorithms in North Indian population

PURPOSE

This study aimed to develop a classifier using supervised machine learning to effectively assess the impact of clinical, demographical, and biochemical factors in accurately predicting the antiseizure medications (ASMs) treatment response in people with epilepsy (PWE).

METHODS

Data was collected from 786 PWE at the Outpatient Department of Neurology, Institute of Human Behavior and Allied Sciences (IHBAS), New Delhi, India from 2005 to 2015. Patients were followed up at the 2nd, 4th, 8th, and 12th month over the span of 1 year for the drugs being administered and their dosage, the serum drug levels, the frequency of seizure control, drug efficacy, the adverse drug reactions (ADRs), and their compliance to ASMs. Several features, including demographic details, medical history, and auxiliary examinations electroencephalogram (EEG) or Computed Tomography (CT) were chosen to discern between patients with distinct remission outcomes. Remission outcomes were categorized into 'good responder (GR)' and 'poor responder (PR)' based on the number of seizures experienced by the patients over the study duration. Our dataset was utilized to train seven classical machine learning algorithms i.e Extreme Gradient Boost (XGB), K-Nearest Neighbor (KNN), Support Vector Classifier (SVC), Decision Tree (DT), Random Forest (RF), Naïve Bayes (NB) and Logistic Regression (LR) to construct classification models.

RESULTS

Our research findings indicate that 1) among the seven algorithms examined, XGB and SVC demonstrated superior predictive performances of ASM treatment outcomes with an accuracy of 0.66 each and ROC-AUC scores of 0.67 (XGB) and 0.66 (SVC) in distinguishing between PR and GR patients. 2) The most influential factor in discerning PR to GR patients is a family history of seizures (no), education (literate) and multitherapy with Chi-square (χ2) values of 12.1539, 8.7232 and 13.620 respectively and odds ratio (OR) of 2.2671, 0.4467, and 1.9453 each. 3). Furthermore, our surrogate analysis revealed that the null hypothesis for both XGB and SVC was rejected at a 100 % confidence level, underscoring the significance of their predictive performance. These findings underscore the robustness and reliability of XGB and SVC in our predictive modelling framework.

SIGNIFICANCE

Utilizing XG Boost and SVC-based machine learning classifier, we successfully forecasted the likelihood of a patient's response to ASM treatment, categorizing them as either PR or GR, post-completion of standard epilepsy examinations. The classifier's predictions were found to be statistically significant, suggesting their potential utility in improving treatment strategies, particularly in the personalized selection of ASM regimens for individual epilepsy patients.

Analysis of the short-term outcomes and risk factors of seizure relapse in patients with gliomas after antiepileptic drugs withdrawal

OBJECTIVE

The optimal timing for glioma patients to stop taking antiepileptic drugs (AEDs) and the risk factors of seizure relapse have not been determined. Here, we explored the short-term outcomes and risk factors of seizure relapse in glioma patients after withdrawal of AEDs.

METHODS

91 patients with gliomas who had no seizures at least 2 years after surgery were enrolled in the study. The patients were followed up for 1 year or until the relapse of seizure after AEDs withdrawal. The risk factors of seizure relapse were analyzed by univariate and multivariate analysis. The optimal discrimination point was determined by plotting a receiver operating characteristic (ROC) curve to explore the relationship between the number of risk factors and seizure relapse.

RESULTS

28 patients (30.8%) relapsed during the follow-up period while 63 patients (69.2%) remained seizure-free. Of the 28 relapsed patients, 20 (71.4%) relapsed within the first 6 months after the AEDs withdrawal. Multivariate analyses revealed that subtotal resection (p = 0.026), IDH1 mutation (p = 0.019), and combined use of AEDs (p = 0.037) were independent risk factors for seizure relapse in glioma patients. ROC curve based on the seizure relapse showed that the sensitivity was 0.821 and 1-specificity was 0.238, corresponding to 1.5 independent risk factors for each patient.

CONCLUSION

To obtain a favorable outcome for glioma patients with preoperative seizures, only patients with less than two independent risk factors for seizure relapse should consider discontinuing AEDs.

Seizure recurrence risk in persons with epilepsy undergoing antiepileptic drug tapering

OBJECTIVES

Antiepileptic drug (AED) tapering in persons with epilepsy (PWE) after 2-3 years of seizure freedom is still debatable because of the risk of seizure recurrence. Tapering patterns have wide variability and could impact seizure recurrence; this study aimed to find out the correlation between them.

MATERIAL AND METHODS

This prospective, observational independent assessor study enrolled PWE undergoing AED tapering in a tertiary care hospital. Data collected included demography, seizure history, AED treatment, and investigational findings. Tapering pattern was assessed based on seizure-free period and AED dose before onset of tapering, dose reduction percentage and frequency, duration of tapering, and follow-up. These variables were compared among the PWE with seizure recurrence and no seizure recurrence.

RESULTS

Among 408 enrolled PWE, 181 were on AED monotherapy: levetiracetam (73), valproate (45), carbamazepine (44), phenytoin (16), and clobazam (3). With a minimum 19 (maximum 41 months) follow-up, seizure recurrence was reported in 119 (29.2%) PWE. The seizure recurrence was not significantly different in-between mono and polytherapy groups; however, among monotherapy groups seizure recurrence was significantly higher (P = .023) in valproate (35.5%) followed by levetiracetam (28.8%) group. Parameters having significant association with seizure recurrence were duration of epilepsy (P = .03), frequency of seizures before control (P = .002), history of previously failed tapering (P = .04), and history of smoking/alcoholic/tobacco intake (P = .003).

CONCLUSIONS

There is a wide variation in AEDs tapering pattern and seizure recurrence risk can be minimized by considering the risk factors like history of smoking/alcoholic/tobacco, longer duration of epilepsy, frequency of seizures before control, and previously failed tapering.

Prediction of antiepileptic drug treatment outcomes of patients with newly diagnosed epilepsy by machine learning

OBJECTIVE

The objective of this study was to build a supervised machine learning-based classifier, which can accurately predict the outcomes of antiepileptic drug (AED) treatment of patients with newly diagnosed epilepsy.

METHODS

We collected information from 287 patients with newly diagnosed epilepsy between 2009 and 2017 at the Second Affiliated Hospital of Zhejiang University. Patients were prospectively followed up for at least 3 years. A number of features, including demographic features, medical history, and auxiliary examinations (electroencephalogram [EEG] and magnetic resonance imaging [MRI]) are selected to distinguish patients with different remission outcomes. Seizure outcomes classified as remission and never remission. In addition, remission is further divided into early remission and late remission. Five classical machine learning algorithms, i.e., Decision Tree, Random Forest, Support Vector Machine, XGBoost, and Logistic Regression, are selected and trained by our dataset to get classification models.

RESULTS

Our study shows that 1) compared with the other four algorithms, the XGBoost algorithm based machine learning model achieves the best prediction performance of the AED treatment outcomes between remission and never remission patients with an F1 score of 0.947 and an area under the curve (AUC) value of 0.979; 2) The best discriminative factor for remission and never remission patients is higher number of seizures before treatment (>3); 3) XGBoost-based machine learning model also offers the best prediction between early remission and later remission patients, with an F1 score of 0.836 and an AUC value of 0.918; 4) multiple seizure type has the highest dependence to the categories of early and late remission patients.

SIGNIFICANCES

Our XGBoost-based machine learning classifier accurately predicts the most probable AED treatment outcome of a patient after he/she finishes all the standard examinations for the epilepsy disease. The classifier's prediction result could help disease guide counseling and eventually improve treatment strategies.

Long-term outcome of seizure-free patients and risk factors of relapse following antiepileptic drug withdrawal

OBJECTIVE

This study aimed to investigate the long-term outcome in seizure-free patients and to explore the risk factors of seizure relapse following antiepileptic drug (AED) withdrawal.

METHODS

This study included 161 patients who had been seizure-free for at least 2 years. These patients were monitored over a follow-up period of at least 3 years or until seizure relapse after AED withdrawal. Patients were grouped into a seizure-free group and a relapse group. Risk factors of seizure relapse were analyzed.

RESULTS

During the follow-up period, 72 patients (44.7%) relapsed while 89 patients (55.3%) did not. The average time of relapse was 19.12 ± 27.17 months after the initiation of AED withdrawal, and majority of cases occurred within the first 24 months of AED withdrawal (73.6%). Univariate analysis showed that the likelihood of relapse was higher in patients with uncontrolled seizures beyond the first 6 months of AED therapy initiation, patients with a history of perinatal injury, patients with multiple seizure types, and patients who had been treated with a combination of AEDs. Multiple logistic regression analysis revealed that uncontrolled seizures beyond the first 6 months of AED treatment and a history of perinatal injury were independent risk factors of seizure relapse.

CONCLUSION

In the majority of cases, epilepsy relapse occurred within the first two years after the initiation of AED withdrawal. The independent risk factors of seizure relapse were uncontrolled seizures beyond the first 6 months of AED therapy and cases with a history of perinatal injury.

Early versus late antiepileptic drug withdrawal for people with epilepsy in remission

BACKGROUND

Epilepsy is a chronic neurological disorder which affects millions of people around the world. Antiepileptic drugs (AED) are the main interventions used to prevent seizures and control epilepsy. Although effective in most cases, AEDs are related to long-term adverse effects, such as cognitive and behavioural alterations. Thus when epilepsy is in remission, it may be in the individual's best interest to discontinue medication. However, the optimal timing of AED discontinuation is still unknown.This is an updated version of the original Cochrane review published in Issue 3, 2001.

OBJECTIVES

(1) To quantify and compare risk of seizure recurrence, status epilepticus and mortality after early and late AED discontinuation in adult and pediatric epilepsy patients.(2) To assess which variables modify the risk of seizure recurrence.(3) To define a subpopulation in which early AED discontinuation is safe.

SEARCH METHODS

We searched the Cochrane Epilepsy Group Specialised Register (June 2014); CENTRAL (Issue 5, The Cochrane Library, May 2014); MEDLINE (1946 to June 2014); CINAHL (23 June 2014); Scopus (1823 to June 2014); ClinicalTrials.gov (23 June 2014); and WHO International Clinical Trials Registry Platform (23 June 2014). We also checked the reference lists of studies found through the electronic searches.

SELECTION CRITERIA

Randomised controlled trials that evaluate withdrawal of AEDs after varying periods of seizure remission in adults and children with epilepsy. Included studies compared an early AED discontinuation time (defined as a period of remission of seizures of less than two years) versus a late AED discontinuation time (defined as a period of remission of seizures of more than two years).

DATA COLLECTION AND ANALYSIS

Two authors independently extracted data and assessed trial quality. Risk ratio (RR) with 95% confidence interval (CI) was calculated for each trial. Summary RRs and 95% CIs for dichotomous data were calculated using a fixed-effect model. A test of statistical heterogeneity was conducted for each pooled risk ratio calculation. Each included study underwent a 'Risk of bias' assessment, based on the Cochrane Handbook recommendations, and we examined the overall quality of information through the GRADE system, presented in two 'Summary of Findings' tables.

MAIN RESULTS

Five trials were included in this review, representing 924 randomised children with epilepsy, all under 16 years of age at randomisation, with a median follow-up of 5.6 years. No eligible trial evaluated adults or assessed mortality or status epilepticus as outcomes. The pooled risk ratio for seizure relapse after AED withdrawal was 1.34 (95% CI 1.13 to 1.59, P = 0.0007). Conforming to this estimate, the number needed to harm, that is expose an individual to a higher risk of seizure relapse because of early withdrawal of AED, is 8 (95% CI 5 to 20). Early discontinuation was associated with greater relapse rates in people with partial seizures with a pooled risk ratio of 1.51 (95% CI 0.97 to 2.35, P = 0.07). Absence type epilepsy showed a lower risk of relapse. Variables associated with higher risk of seizure relapse were abnormal EEG findings (pooled RR 1.44, 95% CI 1.13 to 1.83, P = 0.003), especially epileptiform activity (RR 2.58, 95% CI 2.03 to 3.28, P < 0.0001); epilepsy onset before 2 years or after 10 years of age; history of status epilepticus; intellectual disability (IQ < 70); and high seizure frequency before and during treatment. Gender and family history did not show any significant influence over seizure relapse. Overall, the included trials were classified as low or unclear risk of bias where methodological information was not reported and could not be provided by original study authors.

AUTHORS' CONCLUSIONS

There is evidence to support waiting for at least two seizure-free years before discontinuing AEDs in children, particularly if individuals have an abnormal EEG or partial seizures, or both. There is insufficient evidence to establish when to withdraw AEDs in children with generalised seizures. There is no evidence to guide the timing of withdrawal of AEDs in seizure-free adults. Further high-quality randomised controlled trials are needed, particularly recruiting adults and recruiting those with generalised seizure types, to identify the optimal timing of AED withdrawal and risk factors predictive of relapse.

Adjusting to a seizure-free "new normal" life following discontinuation of antiepileptic drugs during adolescence

This qualitative study sought to understand how children in adolescence adjust to their newly acquired normal life without epilepsy, following discontinuation of antiepileptic drugs during this dynamic period of growth and development. Three major themes with subthemes were identified: 1) setting the body and mind free; 2) engaging in self-regulation; and 3) protection by significant others. A sense of relief from constraints related to treatment schedules, special diets, and avoiding seizure-provoking activities was expressed by all participants. Freedom from side effects of the antiepileptic drugs improved life at home and school. Most of the participants said that they were not worried about seizure recurrence but would use caution against a possible relapse. Family members also must adjust to a new lifestyle. Medical staff needs to provide support and adequate care to adolescents during their period of identity adjustment following antiepileptic drug discontinuation.

[EEG--when and how?]

BACKGROUND

Electroencephalography (EEG) is used for studying functional disorders of the brain. We provide an overview of the areas where EEG is most strongly indicated and the clinical benefit of the examination.

METHOD

This article is based on articles from personal literature archives and literature searches in PubMed, textbook on methods in clinical neurophysiology, the work of one of the authors on a new European standard for EEG and the authors' clinical experience.

RESULTS

EEG is primarily indicated for assessing suspected epileptic seizures and encephalopathies. Standard EEG, sleep-deprived EEG and long-term EEG examinations are used in diagnosing epilepsy. EEG will identify epileptiform activity in about 90% of patients with epilepsy. Epileptiform EEG activity in these patients will provide support for the diagnosis of epilepsy, while the absence of such activity will not exclude the diagnosis. EEG is also important for classifying the type of epileptic syndrome. In cases of encephalopathy, pathology will take the form of an increase in slow activity on the EEG and will support the diagnosis, but as a rule the EEG finding is non-specific. Findings in some cases may point to the underlying aetiology.

INTERPRETATION

Electroencephalograms must be interpreted by an experienced clinical neurophysiologist or neurologist with specialist expertise in EEG interpretation, and must always be evaluated in the light of the clinical picture.

Antiepileptic drug withdrawal in childhood epilepsy: what are the risk factors associated with seizure relapse?

BACKGROUND

In recent years, several studies have been conducted to determine the risk of relapse after antiepileptic drug (AED) withdrawal: there is no general agreement on criteria that can predict safe discontinuation or seizure recurrence.

AIMS

To evaluate prospectively the relapse rate of epilepsy associated to AED withdrawal in epileptic children and to determine the risk factors of seizure recurrence.

METHODS

One hundred-sixty-eight children with epilepsy who were seizure-free for at least 2 years were enrolled and all children were proposed to stop AED treatment and were followed. In all children electroencephalograms (EEGs) were performed before the withdrawal of AEDs and after discontinuation of the treatment.

RESULTS

A total of 48 (28.6%) children relapsed; half of these patients relapsed while reducing the AED dose and the other half after the AED was withdrawn; after 2 years without AEDs, the risk of relapse was very low. Data evaluated by multivariable analysis showed that the children receiving polytherapy before AED withdrawal, having a history of febrile seizures and suffering from multiple seizure types relapsed more frequently. The presence of abnormal post-withdrawal EEG recordings was associated with a higher risk of seizure recurrence.

CONCLUSIONS

Epileptic children, after a seizure-free period of 2 years, have a low risk of seizure recurrence. The potential risk factors of relapse, are multiple seizure types previous polytherapy, history of febrile seizures and abnormalities in post-withdrawal EEG.

Acute anti-emetic withdrawal associated with a hemorrhagic cerebellar arteriovenous malformation

We present a 67-year-old right-handed male with a brachium pontis arteriovenous malformation on continuous anti-emetic therapy who demonstrated acute withdrawal symptoms after the abrupt discontinuation of ondansetron, a 5-HT(3) receptor antagonist. Removal of anti-emetic therapy led to the development of extreme flushing and tremor, but subsequent return of ondansetron resulted in the resolution of these symptoms. This is the first clinical report demonstrating acute withdrawal from an anti-emetic agent and we further highlight the need for future studies evaluating not only arterial supply with pressure gradients and anatomical location, but also the association with periventricular venous drainage, venous drainage stenosis, and mass effect from venous stasis as this may contribute partly to the sensitivity of the serotonergic receptors seen here.

Withdrawal of antiepileptic drugs in adult patients free of seizures for 4 years: a prospective study

We aimed to assess the relapse rate of epilepsy, prospectively attributable to antiepileptic drug (AED) withdrawal in seizure-free patients and to determine the risk factors for seizure recurrence. Seventy-nine patients with epilepsy who were seizure-free for at least 4 years were enrolled into the study. The AEDs were tapered by one-sixth every 2 months. The EEG and clinical examination were performed at the beginning; at each visit during discontinuation and 2, 6, 12, 24, and 36 months after the complete drug withdrawal. For each patient, records were obtained of the main demographic and clinical variables. A total of 49 patients completed the discontinuation programme. Twenty-eight patients (57%) relapsed while 21 of those (42.8%) did not suffer a relapse at the end of the study period. In patients discontinuing treatment, the probability of relapse was 21.4% during the tapering period (especially in the last months), 28.6% at 1 month, 14.3% at 3 months, 3.6% at 6 months, 7.1% at 12 months, 17.8% at 24 months, and 7.1% at 36 months. The age at onset of epilepsy and the duration of active disease were found to affect the risk of relapse. Although drug withdrawal could be considered in adult patients free of seizures for 4 years, the final decision should be tailored to the patient's clinical, emotional, and socio-cultural profile.

The natural history of epilepsy: an epidemiological view

Better information of the natural history of epilepsy has important implications for understanding the underlying neurobiology, evaluating treatment strategies, and planning healthcare resources. The traditional pessimistic view has been dispelled by results from modern community based prospective studies, showing that over 60% of newly diagnosed patients will enter remission upon treatment. Recent outcome studies suggest that medical intractability may be predicted after failure of two antiepileptic drugs. Poor prognostic factors include a high initial seizure density, symptomatic aetiology, and presence of structural cerebral abnormalities, all of which can be identified early on. Among patients who have entered remission, many will remain seizure-free after antiepileptic drug treatment is withdrawn, suggesting that the underlying seizure generating factor has remitted. Whether some of these patients have entered remission "spontaneously" is contentious because, with effective pharmacotherapy for epilepsy in use for over 100 years, the natural history of untreated epilepsy is largely unknown. Circumstantial evidence, mostly arising from resource poor countries where antiepileptic drug treatment is not readily available, indicates that spontaneous remission may occur in up to 30% of cases. Observations from these complementary sources suggest that, at the population level, prognosis of newly diagnosed epilepsy may be broadly categorised into three groups: remission without treatment, remission with treatment only, and persistent seizures despite treatment. As understanding of the prognostic factors improves, the potential of a "prognostic group specific" management approach should be explored so that effective treatments may be used in a more rational and targeted fashion.

The impact of early versus late anticonvulsant reduction after ketogenic diet initiation

The ideal timing of anticonvulsant reduction for a child started on the ketogenic diet is unclear. The records of 81 children started on the ketogenic diet consecutively over a 2-year period were reviewed for a 6-month period after diet initiation. During their first 6 months on the ketogenic diet, medications were tapered in 53 of 81 (65%) patients, with 30 of 53 (57%) considered "early" (tapered during diet initiation or within the first month afterward). No differences were seen between the early and late groups with respect to percentage with > 90% seizure reduction at 3 months (47% vs 48%), diet duration (1.1 vs 0.9 years), percentage remaining on the ketogenic diet to date (73% vs 65%), or improved alertness (63% vs 57%). Nine of fifty-three (17%) had transient seizure increases during medication tapering, with no correlation to the timing in which this occurred; however, five were taking benzodiazepines or barbiturates. All had > 50% seizure reduction at 3 months despite the transient worsening. Early reduction of anticonvulsants in children initiated on the ketogenic diet appears to be safe and well tolerated. However, it offers no specific advantage compared with a later taper.