17 When and How to Stop Antiepileptic Drugs

John D. Hixson

Department of Neurology, University of California San Francisco, San Francisco, CA, USA

Introduction

When a patient is first diagnosed with epilepsy, the primary aim is to determine the appropriate treatment and stop further seizures. However, once seizure freedom is achieved, the eventual question becomes: how long should a patient remain on antiepileptic therapy? For other medical conditions, this question is more straightforward; antihypertensive therapy is generally considered a long-term commitment, while antibiotics have an intrinsically finite treatment plan. For epilepsy, the time frame is less clear. Some patients on antiepileptic drugs (AEDs) certainly benefit from prevented seizure recurrence, but other epilepsy patients may have entered a period of remission, even without medication. From the perspective of the healthcare provider, assigning a particular patient to a given category is difficult.

This chapter provides a summary of the evidence and resultant recommendations on this subject. A number of studies have examined this topic in imperfect conditions, using population-based methodology in both adults and children. This chapter will not review these studies exhaustively, but will summarize the primary approach to the concept of an AED withdrawal trial.

In general, for all patients with epilepsy who have been successfully treated with an AED, the seizure recurrence risk following medication withdrawal is significant, primarily in the first year following discontinuation. The overall seizure risk does not return to the population norm. However, depending on the situation, a withdrawal trial may be reasonable in many cases. Individual risk factors should be considered and discussed with each patient.

When to stop an AED

Most studies that have examined this question in children and adult populations have had methodological shortcomings. Heterogeneity of epilepsy classification, variance in followup lengths, and different study designs prevent a definitive set of guidelines. However, some very clear trends emerge from the studies. First, even after a patient achieves seizure freedom, the risk of seizure recurrence after stopping an AED does not return to the healthy population norm. Second, certain risk factors predict an even higher rate of seizure recurrence following AED withdrawal. Thus, the best clinical practice is to have a detailed conversation with each patient about his or her relative risk (RR), focusing on the individual’s fear of seizure recurrence versus the effect of continued therapy on quality of life and overall health.

Much of the practice of AED withdrawal trials has been guided by an American Academy of Neurology (AAN) practice parameter issued in the mid-1990s. This analysis examined 17 studies available at that time that dealt with AED withdrawal; only one study was a randomized controlled trial. The practice parameter noted pooled recurrence rates of 31.2% in children and 39.4% in adults. Across the different studies, the time of seizure freedom prior to AED withdrawal ranged from 1–2 years in children to 2–5 years in adults.

Since the publication of this AAN practice parameter, other studies have examined AED withdrawal in adults with epilepsy, demonstrating recurrence rates of 41–52%, depending on the length of followup. These investigations used a range of seizure freedom periods of 2–4 years, likely based on the guidance of the AAN practice parameter. It should also be noted that these studies suggested that the highest risk of recurrence occurred relatively early, within 6 months of the withdrawal period.

EVIDENCE AT A GLANCE

In 2008, Lossius and colleagues reported a prospective randomized trial to explore the risks and benefits of AED withdrawal in adults. Although the primary aim of the study was to examine the effect of AED withdrawal on cognitive function and quality of life, it also provides evidence on seizure recurrence risk. Patients were required to be seizure-free for more than 2 years and were on AED monotherapy. The randomized withdrawal group exhibited an RR of 2.46 compared with the nonwithdrawal group (15% vs. 7%) in the initial 12-month followup. With longer followup, the withdrawal group demonstrated an increased relapse rate of 27%. The withdrawal group did show cognitive improvements on neuropsychological testing, supporting the rationale for the withdrawal attempt. The findings of this unique and important randomized, controlled adult study are largely consistent with the growing body of evidence concerning AED withdrawal.

In contrast, a randomized controlled trial of AED withdrawal in children that had been seizure-free for only 6 or 12 months, by Peters and colleagues, demonstrated relatively high recurrence rates of 55% and 49%, generally consistent with other studies performed since that time. Based on this study and others in children, a Camfield review in 2008 suggested a seizure-free period of 1–2 years as a reasonable pediatric benchmark for a trial of AED withdrawal. Notably, this standard for duration of seizure freedom is lower than that of the adult population, possibly because some pediatric epilepsy syndromes are known to naturally remit over time. Thus, it is imperative to consider age when making this important decision.

It is also worth mentioning a special set of individuals that hopefully achieve seizure freedom: postsurgical epilepsy patients. By definition, these patients have risk factors for seizure recurrence (history of medication resistance, longer epilepsy duration), and the evidence of good long-term postsurgical seizure freedom rates depends on maintenance of AED therapy. However, several studies do support attempting AED withdrawal after 1–2 years of postsurgical seizure freedom. Although recurrence risk increases with duration of postoperative followup in most studies, the evidence suggests that this is not directly caused by AED withdrawal. Thus, it is reasonable to discuss this option with postsurgical patients while providing appropriate education and guidance after 1–2 years of complete seizure freedom.

It is important to recognize that a predefined period of seizure freedom cannot be broadly applied to all epilepsy patients. Intuitively, the length of the observation period depends on the pretreatment seizure frequency, which may differ widely among patients. The “rule of three” approach can be applied to this relationship: to be reasonably confident that a given treatment has improved seizure control, a span of at least three times the preintervention seizure interval is required. For example, in someone with one seizure every 3 months, seizure freedom for 9 months would be needed for confidence that control had truly been achieved. For those with very infrequent seizures, this “rule” may be less meaningful; however, it serves as a reminder that the necessary period of seizure freedom prior to considering AED withdrawal depends on the individual situation.

Only gold members can continue reading. Log In or Register to continue