Keywords
Epilepsy - Anti-epileptic drug withdrawal - Neurocysticercosis - Epilepsy surgery
- Cortical venous sinus thrombosis
1
Introduction
Seizure freedom is achieved in about two thirds of the patients who are treated with
anti-epileptic drugs (AEDs) in new onset epilepsy.[1] AED treatments suppress the seizures and some unknown phenomena “resolves” the tendency
to throw a seizure.[2] Nevertheless, the most important consideration in such patients is whether to continue
AEDs or stop them. Decision to withdraw anti-epileptic drugs in patients with epilepsy
in remission requires a good clinical judgment and detailed discussion with the patient
and family members. There are no established guidelines concerning this issue. The
Medical Research Council (MRC) study was the first randomized trial that tried to
answer this issue and since then there have been many studies and reviews about when
and how to stops AEDs but the controversies still persist. In this review, we shall
highlight on some important issues with regard to stopping AEDs:
-
Why to stop AEDs
-
Risk of recurrence on withdrawal of AEDs
-
When to stop AEDs
-
How to withdraw AEDs
-
Special situations
2
Why to stop AEDs
AEDs are associated with significant systemic and neurocognitive side effects. Teratogenic
effects of AEDs are well known and decision to stop AEDs in females of child bearing
age group with controlled seizures needs to be the primary concern of the treating
physician. Most AEDs are partially or completely eliminated by hepatic metabolism.
AEDs have enzyme inducing or inhibiting properties of their own and hence, can alter
the clearance rates of other drugs e.g. carbamazepine induces the metabolism of oral
contraceptives through hepatic enzyme induction and lowers the efficacy of oral contraceptives.[3] Thus, potentially harmful interactions should be considered in patients who are
taking other drugs in addition to the AEDs.
Epilepsy treatment requires AEDs to be taken regularly, mostly in a twice daily dosage
for prolonged periods. The cost related to epilepsy treatment comprises direct costs
like hospital admissions, pharmacological therapies, consultation charges and indirect
costs like absenteeism and unemployment. Das et al showed that 90% of the patients
who discontinued AEDs after 1 year expressed their inability to continue treatment
due to low annual income and comparatively large amount of their income being spent
for the cost of treatment.[4] With the introduction of newer AEDs, the economic burden for epilepsy management
has increased even further. Haroon et al compared the monthly costs of old and new
AEDs prescribed and found that the cost of lamotrigine, levetiracetam and lacosamide
was approximately more than 10 times as compared with the mean monthly cost of the
traditional anti-epileptics.[5] Due to this financial burden, the first question asked to the attending physician
with “control” of seizures is if the AEDs can be stopped.
Social stigma attached with epilepsy is a well-known phenomenon, especially in developing
countries like India. In a study by Das et al, 130 out of 1450 patients with epilepsy
had marital disharmony due to divorce or separation. Women with epilepsy discontinued
treatment due to the misunderstanding with their husbands and family members that
the disease may be transmitted to the offspring or baby may be physically and mentally
abnormal. This eventually creates an enormous economic burden.[4]
Prolonged AED treatment impairs the quality of life of the epileptic patients. Nabukenya
et al showed that the health related quality of life (HRQOL) mean score among patients
on AEDs was low, thus suggesting their poor physical, psychological and mental functioning
and poor emotional wellbeing.[6] Lossius et al in a double blinded, randomized study found an improvement in the
neuropsychological functioning upon drug withdrawal in the form of an improved ability
to perform activities demanding rapid cognitive performance and complex motor coordination.[7] Similarly, the MRC (UK) study reported an improved feeling of wellbeing after stopping
AEDs, thus signifying an improvement in the cognitive function after stopping AEDs.[8]
Mood disorders are the most frequent psychiatric comorbidity in patients with epilepsy
with a prevalence rate for depressive disorders in the order of 20%–22%.[9] Andersohn et al found that the use of newer AEDs with a high potential of causing
depression increases the risk of self-harm/suicidal behavior by three times in patients
of epilepsy.[10]
Thus, we can infer that withdrawing AEDs in patients, who have achieved seizure freedom,
has obvious pharmacologic, financial and social implications. Majority of patients
wish to discontinue AEDs at the first opportunity.
Risk of seizure recurrence on stopping AEDs
3
Risk of seizure recurrence on stopping AEDs
Recurrence of seizure is the most dreaded effect of stopping AEDs. About 50% recurrences
are seen during the first 6 months of stopping therapy.[2] In a prospective, randomized study conducted by the MRC, 1013 patients participated,
out of which, 59% patients who were randomized to the withdrawal group and 22% in
the continuing therapy group had recurrence of seizures at the end of two years. Longer
seizure free periods at randomization significantly reduced the risk while the number
of AEDs at randomization and history of tonic-clonic seizures significantly increased
the risk of recurrence.[8] Archana et al showed an overall risk of seizure recurrence of 31% over a period
of 18 months after stopping AEDs.[11] As per a study done by Camfield et al in children, about 1% developed medically
refractory epilepsy upon stopping medications.[12] The psychological impact of recurrence of seizures is detrimental to the quality
of life of the patients and hence, in this scenario, patients usually prefer to continue
AEDs.
What are the factors that may help to predict the chances of recurrence of seizures
on AED withdrawal? The guidelines published by the American Academy of Neurology (AAN)
in 1996 listed 4 primary characteristics that need to be considered before AED withdrawal:
a seizure free period of 2–5 years, single type of partial or generalized seizure,
normal neurological examination and intelligence quotient and electroencephalogram
(EEG) normalized with treatment.[13] Olmez et al studied the risk of recurrence after drug withdrawal in childhood epilepsy
and found that post withdrawal EEG abnormalities were significantly associated with
seizure recurrence.[14] Su et al examined the role of EEG abnormalities at the time of, during and 1 year
after AED withdrawal and found that patients with epileptiform EEG abnormalities within
1 year after AED withdrawal have an increased risk of seizure relapse.[15] Specchio et al recruited 330 patients, out of which, 225 discontinued treatment
after a seizure free period of at least two years. They observed that the cumulative
risk of a relapse in those who discontinued therapy was 2.9 times higher than that
of patients continuing treatment. The factors affecting the risk of relapse were the
duration of active disease and the number of years of seizure remission while on treatment.
The 24-month risk of relapse was 0% in idiopathic partial epilepsies whereas it was
higher for symptomatic partial epilepsies, cryptogenic partial epilepsies, idiopathic
generalized epilepsies, and symptomatic or cryptogenic generalized epilepsies.[16]
Idiopathic generalized epilepsies (IGE) account for 20% of all epilepsies and refer
to a diverse group of epileptic seizures and syndromes, which usually have a genetic
basis. The common syndromic varieties of IGE are benign neonatal and infantile idiopathic
generalized epilepsies, childhood absence epilepsy (CAE), juvenile absence epilepsy
(JAE), juvenile myoclonic epilepsy (JME) and epilepsy with primary generalized tonic-clonic
seizures.[17] There are not many studies that have focused on the issues of risk of seizure recurrence
after AED withdrawal in patients with different types of IGE. Pavlovic et al studied
the risk of seizure recurrence after AED withdrawal in 59 patients with IGE syndrome.
They observed that the risk of relapse after AED withdrawal was maximum in JME (100%)
and least in CAE (6%). EEG worsening and multiple seizure types were the significant
factors associated with relapse after AED withdrawal.[18] Similarly, in a study by Murakami et al, AED withdrawal was studied in children
with cryptogenic, symptomatic and idiopathic epilepsies. Of the 304 patients included
in the study, 18% had IGE syndrome. The authors observed that after a seizure free
period of three years, the relapse rate was 6% in childhood absence epilepsy (CAE),
25% in juvenile absence epilepsy (JAE) and 100% in juvenile myoclonic epilepsy (JME)
after stopping AEDs.[19] Withdrawal of AEDs in IGE syndrome can thus result in definite recurrences in JME
to occasional recurrences in CAE.
Benign epilepsy with centro-temporal spikes (BECTS) is considered to be the most common
childhood epilepsy syndrome, accounting for 8–20% of pediatric patients with epilepsy.[20] It is characterized by brief, simple, orofacial partial seizures (paresthesias and
tonic or clonic activity of the lower face, often spreading to the ipsilateral arm,
associated with drooling and anarthria), often occurring during sleep or on awakening
and associated with a slight male preponderance. This is a self-remitting syndrome
with more than 99% patients achieving remission by the age of 18 years. The risk of
recurrence in these patients on stopping AEDs is very low.[21]
A number of other predictors have been found to be associated with increased risk
of relapse ([Table 1]).
Table 1
Factors predicting increased risk of recurrence on AED withdrawal (Ref).
|
Onset of epilepsy in adolescent and adulthood[14]
[15]
[18]
|
|
Use of multiple AEDs[3]
[15]
|
|
Low IQ status[13]
[15]
|
|
Abnormal neurological examination[13]
[15]
[18]
|
|
Continuing ‘epileptiform’ discharges while on treatment[13]–[15]
[18]
|
|
Worsening ‘epileptiform’ discharges after AED discontinuation[14]
[15]
[18]
|
|
Epileptogenic lesion on neuroimaging[14]
|
|
Syndromic epilepsies like juvenile myoclonic epilepsy, Lennox-Gastaut syndrome, West
syndrome, progressive myoclonic epilepsies, Tuberous sclerosis[42]
|
Thus, we can infer that the risk of recurrence of seizures is high in the immediate
post withdrawal period but it gradually reduces after the first 2 years. The factors
that predict the chances of recurrence of seizures, especially EEG abnormalities,
should be considered on an individual basis by the treating physician and the decision
to withdraw anti-epileptics should be well discussed with the patient and caregivers.
4
When to stop AEDs
There is no specified time period as to when AEDs need to be tapered in adults and
children. Camfield et al, in their study on children found that out of 260 patients
who were seizure free for 2.8 ± 1.4 years, AEDs were tapered over a period of 6 weeks
and 66% (171) remained seizure free over a follow up period of 3.7 ± 2.3 years. Only
5 patients, who were seizure free for more than 2 years, developed intractable epilepsy
on stopping AEDs (<1%). Hence, they proposed a seizure free period of 2 years prior
to AED withdrawal.[12] Similarly, Lossius et al conducted a randomized trial in adult population where
79 patients were in the withdrawal arm and 81 in the continued treatment arm with
a seizure free duration of 2 years. Within 12 months, 7% of the non-withdrawers and
15% of the withdrawers experienced seizure relapse (p = 0.095).[7]
Berg et al, in a meta-analysis reviewed more than 20 studies related to AED withdrawal
in both children and adults and found that the rate of seizure recurrence to be 25%
at the end of 1 year and 29% at the end of 2 years of seizure free period.[22] AAN guidelines published in 1996 advocated a seizure free period of 2–5 years prior
to drug withdrawal.[13]
It can be summarized that there is no general consensus on an optimum seizure free
period prior to AED withdrawal. However, in children, a seizure free period of two
years appears justifiable though the same may not be true for adults, where the semiology
of the seizure, neurological examination and imaging/EEG findings need to be considered
prior to drug withdrawal.
5
How to withdraw AEDs
One of the main goals of the treating epileptologist while withdrawing AEDs in a patient
who has attained seizure freedom is to prevent the recurrence of seizures. Abrupt
withdrawal of AEDs is not recommended due to the risk of recurrence. The time period
over which these drugs should be tapered and stopped is again a matter of much debate
with no definite consensus.[3]
Tennison et al studied the effect of relatively rapid tapering (over 6 weeks) to slow
tapering over 9 month period in 149 children. They found that the length of the taper
period did not significantly influence the seizure recurrence. Their results showed
a higher recurrence of seizure in the first one or two years after rapid withdrawal
of AED, compared to slow withdrawal. The factors that significantly affected the risk
of recurrence during the tapering period were the presence of mental retardation and
EEG abnormalities at the time of tapering.[23] In a similar study by Serra et al, AEDs were tapered over a period of either 1 or
6 months in 57 children. The results were similar to the previous study with no significant
difference in seizure recurrence in the two groups.[24] In a Cochrane review of randomized trials comparing slow versus rapid tapering of
AEDs, the authors were not able to draw any significant conclusions.[25]
In the MRC study, mean age at entry into the trial was 26–27 years while mean age
at seizure onset was 13–14 years. Those randomized to slow withdrawal had AED withdrawn
with dosage decrements every 4 weeks with the aim to extend the withdrawal to a minimum
of 6 months. AEDs were withdrawn sequentially in the patients, who were on polytherapy.[8] Guidelines issued by the Italian League Against Epilepsy recommend slow discontinuation
of the anti-epileptic drugs with tapering period tailored to the patient's needs and
preferences.[26]
A special note has to be mentioned with regards to the use of benzodiazepines and
barbiturates. These drugs have a tendency to provoke withdrawal seizures and hence,
require a longer tapering period compared to other AEDs.[27]
In summary, withdrawal and stoppage of AEDs depends on clinical profile of the patient,
which can be safely achieved over the course of several weeks. If seizures recur on
tapering dosage, pre seizure dosage can be resumed and continued.
6
Special situations
6.1
Drug withdrawal after epilepsy surgery
One of the main goals of epilepsy surgery in medically refractory cases is to reduce
or stop the anti-epileptic drugs. Epilepsy is resolved in only about 20% of the patients
following surgery, thereby indicating a protective role of AEDs in the post-operative
period.[28]
[29] There are no established guidelines on AED withdrawal after surgery. Schiller et al
evaluated the frequency of seizure recurrence associated with AED withdrawal in patients
undergoing successful epilepsy surgery. Out of 210 patients who were seizure free
for more than 1 year, 22 of the 84 patients who had stopped treatment had a seizure
recurrence. Among these, reinstitution of AED treatment resulted in seizure control
in 20 patients. The authors postulated that the surgical procedures in these patients
resulted in subtotal excision of epileptogenic zone and hence, a combination of surgery
and AEDs was useful. They found that seizure recurrence in the patients was unrelated
to the duration of the seizure free postoperative AED treatment.[30] Park et al studied 283 patients who underwent AED withdrawal after neocortical epilepsy
surgery including 100 patients with temporal lobe epilepsy. Seizures recurred in 78/147
patients where AED withdrawal was attempted. Multivariate analysis revealed that early
drug tapering (<9 months post-surgery), normal MRI results, seizure before reduction,
and longer epilepsy duration were associated with seizure recurrence.[29] The seizure outcome following extra temporal resections is generally less favorable
as compared to temporal lobe resections. Menon et al studied 106 patients undergoing
AED withdrawal after epilepsy surgery with a median time of starting AED tapering
being 5 months (range – 3–124 months). Ninety four patients had a seizure recurrence
but interestingly, there was statistically no significant difference in the timing
of starting the AED tapering in patients who had seizure recurrence versus who did
not have.[31]
Hence, as to when tapering of AEDs needs to be done following epilepsy surgery, is
not well-defined and it has to be treating clinician's decision based on multiple
factors. There is an increased chance of recurrence of seizures in patients who have
undergone extra temporal epilepsy surgery, so AED withdrawal has to be done more cautiously
in this group of patients.
6.2
Drug withdrawal in solitary cysticercal granuloma
Neurocysticercosis (NCC) contributes substantially to the burden of epilepsy. The
geographic distribution of cysticercosis is wide, with high prevalence reported from
India, Sub-Saharan Africa Mexico, Central and South America.[32] The clinical and radiologic manifestations of NCC vary depending upon the number,
location and the stage of the cysticerci in the brain. A form of NCC most commonly
seen in clinical practice as a cause of seizures is solitary cysticercal granuloma
(SCG). It is a benign form of NCC where seizures are relatively well controlled with
a single AED. An important issue to be addressed in this context is how long to continue
AED in a patient with SCG. Three open labeled trials have compared the administration
of AEDs for a short duration (6 months) vs a longer duration (12–24 months) in individuals
with SCG and seizures.[33]
[34]
[35] An expert group meeting from India comparing these trials concluded that there was
no additional benefit of the longer duration AED administration in individuals in
whom the SCG had resolved. However, in those individuals where there is a calcific
residue, the risk of seizure recurrence was significantly high. It was concluded that
the AEDs have to be continued until the lesion (granuloma) is actively degenerating
(i.e., appears as an enhancing lesion on imaging studies) and the decision to withdraw
AED is to be taken once the complete resolution of the granuloma is demonstrated on
follow-up imaging studies.[36] Verma et al randomized 206 subjects with SCG in two groups, one where patients were
treated for 6 months and other where treatment was given for 2 years. They found that
in patients where there was a complete disappearance of the lesion, there was statistically
no significant difference in the recurrence of seizures. However, in patients having
residual calcification, the short-term withdrawal group had a statistically significant
recurrence of seizures (42% vs 21%, p < 0.05).[33]
We can infer that a longer duration of therapy is not warranted in patients having
total resolution of lesion. But in patients with residual calcific lesions, chances
of seizure recurrence are high and withdrawal should be attempted with great caution.
6.3
Drug withdrawal in seizures associated with cortical venous thrombosis
Cerebral venous thrombosis (CVT) is a rare type of cerebrovascular disease accounting
for 0.5% of all stroke cases. The most common symptoms and signs are headache, seizures,
focal neurological deficits and altered consciousness which can present in isolation
or in association with other symptoms.[37]
Ferro et al analyzed the risk factors for seizures as a presenting symptom or early
in the course (first two weeks) of CVT in 624 patients. 40% patients had presenting
seizures and 6% had early seizures.[38] In an earlier study, Ferro et al found that out of 91 patients with CVT, 29 patients
had seizures as a presenting feature (32%). Early symptomatic seizures were more frequent
in patients with neurological deficits and imaging abnormalities on MRI/CT brain in
the form of edema, ischemia or hemorrhage. The authors concluded that there is a moderate
risk of seizure recurrence early in the course and during the first year after CVT.[39] In a retrospective analysis of 77 patients with CVT over a mean of 77 months, Preter
et al observed that out of 28 patients who had seizures at the presentation, only
4 had recurrence (14%). In 3 out of these 4 patients, seizures occurred in the first
year after CVT. The authors recommended that with a low risk of seizure recurrence,
it seems appropriate to maintain anticonvulsant therapy for a year and to taper off
gradually thereafter.[40]
As per the guidelines issued by the European Federation of Neurological Societies
(EFNS), treatment with AED in the patients of CVT for 1 year may be reasonable for
patients with early seizures and hemorrhagic lesions on brain scan on admission, whereas
AED therapy may be tapered off gradually after the acute stage in patients without
these risk factors.[41]
7
Conclusions
Anti-epileptic drugs are the mainstay in the management of epilepsy. The exact duration
of AED treatment in patients who are in remission is a matter of controversy and not
yet established. The prognosis on AED withdrawal will vary according to the type of
epilepsy syndromes and certain factors that are not firmly determined. In patients
where seizures are adequately controlled, AED withdrawal should be encouraged because
of substantial side effects, particularly in cognitive function. The duration of a
two year seizure free period is well defined in children based on the available literature
but in adults, this time period is not very definite though tapering should be attempted
after due considerations. The duration of tapering is again a questionable issue but
slow withdrawal should be attempted with sequential tapering in cases of polytherapy.
The process of drug tapering should be discussed with the patient and caregivers in
great detail and the risk of recurrence should be told to the patient well in advance.
In the absence of well established guidelines, the process of AED tapering should
be individualized.
Conflicts of interest
All authors have none to declare.
