|
|
| |||||||||||||||||||||||||||||||
Background The optimal regimen for discontinuing antiepileptic medications in children with epilepsy is unknown.
Methods We randomly assigned 149 children to either a six-week or a nine-month period of drug tapering, after which therapy was discontinued. Each group was composed of patients who had been seizure-free for either two or four years before drug tapering was begun. Most patients were receiving one antiepileptic drug; none were taking more than two. The children were evaluated periodically during and after the taper period. Sixteen patients were lost to follow-up before the beginning of the taper period. Proportional-hazards regression analysis was used to assess the risk of seizure recurrence among the remaining 133 patients.
Results Seizures recurred in 53 patients (40 percent). The mean duration of follow-up was 39 months (range, 11 to 105) for the patients who did not have a recurrence of seizures. Neither the length of the taper period (six weeks vs. nine months, P = 0.38) nor the length of time the patients were free of seizures before the taper period was begun (two years vs. four years, P = 0.20) significantly influenced the risk of seizure recurrence. The presence of mental retardation (relative risk, 3.1; 95 percent confidence interval, 1.5 to 6.2) or spikes in the electroencephalogram at the time of tapering (relative risk, 1.9; 95 percent confidence interval, 1.0 to 3.4) increased the risk of seizure recurrence.
Conclusions The risk of seizure recurrence during drug tapering and after the discontinuation of antiepileptic drug therapy in children with epilepsy is not different whether the medications are tapered over a six-week or a nine-month period.
In this study we compared a six-week (relatively short) period and a nine-month (relatively long) period of drug tapering in a group of children with epilepsy who had had no seizures for either two or four years.
Methods
Design and Recruitment
The study design was prospective, randomized, and unblinded. The study protocol was approved by the review boards for the study of human subjects of the two participating medical schools, and informed consent was obtained for each child. All children receiving care at the pediatric epilepsy clinics at the two institutions who had had no seizures for approximately 18 months were eligible for the study. Patients who had had a single seizure or only febrile seizures were excluded, as were patients with neonatal seizures or infantile spasms, since the natural history of these conditions is unique.
The eligible children were randomly assigned by two coin tosses to have their dosage of antiepileptic medications tapered to zero over a period of six weeks or nine months after a seizure-free period of two or four years. A previously constructed table of the four possible combinations of heads and tails was then consulted. The result of the first toss determined the rate of tapering (six weeks or nine months), and the result of the second determined whether the period would begin after the patient had been seizure-free for two years or four years.
An electroencephalogram was obtained just before the start of the taper period. When the patient had reached the targeted seizure-free interval, the medications were tapered according to a standard scheme in which the taper period was divided into three equal parts (two weeks for the six-week group and three months for the nine-month group). The dose was reduced by 25 percent during each interval. The actual dose was as close to the calculated dose as was feasible with commercially available drug preparations. If the patient was receiving more than one drug, the drugs were tapered sequentially with the use of the same six-week or nine-month period for each. The dose of the barbiturates was tapered last.
In all analyses, follow-up was begun at the start of drug tapering, since that was the time at which the risk of seizure recurrence due to a reduction in the dose or the discontinuation of therapy began. Follow-up information during and after the taper period was obtained by telephone, at clinic visits, or both. All patients were followed for at least 11 months after the beginning of the taper period or until their seizures recurred.
The patients were considered to be compliant with the rate of drug tapering if the process was completed within two weeks of the specified target date for those in the six-week group or within four weeks in the nine-month group. To be in compliance with the timing of the start of the taper period, the patients had to begin the process within three months of their scheduled start date.
Seizures were classified according to the definitions of the International League against Epilepsy17 on the basis of history, electroencephalographic findings, and imaging criteria.
Statistical Analysis
A survival analysis of the data was performed with seizure recurrence as the hazard and the length of time to seizure recurrence (failure time) measured from the start of the taper period. Proportional-hazards regression analysis was used to compare the two taper groups (six weeks vs. nine months) and the two groups classified according to the length of time the patients were free of seizures before tapering was begun (two years vs. four years). Kaplan-Meier plots of the data were constructed. Differences in selected attributes between study groups were assessed with Pearson's chi-square test. All statistical tests were two-tailed.
A secondary, exploratory analysis of the effect of a number of additional factors, other than the rate of tapering or the duration of therapy before the taper period, on the risk of seizure recurrence was performed. Proportional-hazards analysis was also used to calculate the relative risks for these selected variables after adjustment for any differences in the rate or timing of drug tapering.
Results
Of the 150 children who were considered for the study, 1 declined to participate. The outcome of the 149 patients who initially entered the study is shown in Table 1. Sixteen patients were lost to follow-up before the start of the taper period. The primary analysis was carried out on the 133 patients who remained in the study at the start of the taper period. The results of a secondary analysis of the 111 patients who correctly completed the taper protocol were similar. The mean duration of follow-up was 39 months for the 80 patients who did not have a recurrence of seizures (range, 11 to 105). Overall, seizures recurred in 53 patients (40 percent).
|
The number of patients who were lost to follow-up or had a recurrence of seizures before the beginning of the planned taper period was similar in the two taper groups (Table 1). More patients (or their parents) in the nine-month group than in the six-week group later decided not to proceed with the taper process.
The characteristics of the 133 patients who completed the protocol are shown in Table 2. The mean duration of follow-up after the start of the taper period for patients who did not have a recurrence of seizures was slightly longer in the six-week group than the nine-month group (44 months [range, 11 to 105] vs. 33 months [range, 11 to 91], P = 0.07).
|
|
Compliance with the scheduled start of the taper period was less reliable. Ten of 72 patients who had been seizure-free for two years before drug tapering was begun (14 percent) and 25 of 61 patients who had been seizure-free for four years (41 percent) did not begin the taper period at the scheduled time (P<0.001). The most common deviations from the schedule were longer-than-expected treatment in the two-year group and shorter-than-expected treatment in the four-year group. Both groups were similar with respect to the characteristics listed in Table 2. The mean duration of follow-up for patients who did not have a recurrence of seizures during drug tapering was similar in the two-year and four-year groups (42 and 36 months, respectively; P = 0.24). Figure 2 shows the proportion of patients remaining seizure-free after the start of the taper period in the two groups. There was a trend toward a greater risk of seizure recurrence in the group that had been seizure-free for two years before drug tapering was begun, although the curves were not significantly different (P = 0.20). Similarly, in a chi-square analysis, more patients in the two-year group than in the four-year group had a recurrence of seizures (34 of 72 patients [47 percent] vs. 19 of 61 patients [31 percent], P = 0.06).
|
The relative risks of seizure recurrence for 12 variables on which information was available are shown in Table 3. The presence of mental retardation and the presence of spikes in the electroencephalogram obtained just before the start of the taper period increased the likelihood of a recurrence of seizures.
|
We found an overall risk of seizure recurrence (40 percent) similar to that in other studies of both adults and children in whom antiepileptic drugs were discontinued over periods ranging from 1 to 18 months1,3,4,5,6,10,11,12,13,14,15. The similar success rates reported in these studies indirectly support the contention that the duration of the taper period is not a major determinant of the risk of seizure recurrence, at least in the range of periods studied.
Although the abrupt discontinuation of antiepileptic drug therapy has been advocated,2 there is evidence to support the benefit of gradual tapering, especially in the case of barbiturates or benzodiazepines18,19,20,21,22,23,24,25,26,27,28,29. The type of medication being tapered did not affect the risk of seizure recurrence in our study, suggesting that six weeks is an adequate taper period, even for barbiturates.
The question of the optimal rate of tapering has rarely been directly addressed. In three nonrandomized studies,2,8,30 there was no correlation between the risk of seizure recurrence and the length of the taper period if abrupt discontinuation was excluded. In a randomized study of the discontinuation of antiepileptic drugs, the risk of seizure recurrence was increased if the drugs were discontinued over a period of less than six months31. Several aspects of that study may explain the discrepancy between their findings and our findings. First, the restrictive definition of epilepsy used in that study probably led to the selection of patients with more severe seizures. Second, multiple medications were stopped simultaneously rather than sequentially. Finally, the sizes of the treatment groups differed greatly, calling into question the effectiveness of the study's randomization process, so that unidentified confounding factors may explain the poorer outcome with the shorter taper periods.
Other studies have reached conflicting conclusions regarding the effect on the risk of seizure recurrence of the duration of the seizure-free period before drug tapering is initiated3,6,8. We found a trend toward a lower risk of seizure recurrence in the group that had been seizure-free for four years before drug tapering was begun, but the shorter follow-up after the start of the taper period in this group might have biased the study in their favor. On the other hand, the risk of recurrence in this group might have been even lower had more patients waited the prescribed four years before starting the taper period.
Some compromise, individually determined for each patient, must be reached between the possibly lower risk of seizure recurrence with a longer duration of treatment and the cumulative costs of longer treatment regimens before tapering is begun. These include financial, social, and psychological costs as well as the direct adverse effects of medication.
No general agreement has been reached on the factors that affect the risk of seizure recurrence when antiepileptic drugs are tapered1,3,4,5,6,7,8,11,12,13,31,32,33,34,35. In our study, only the presence of mental retardation or spikes on the electroencephalogram obtained just before the start of the taper period increased the risk of seizure recurrence.
There are several advantages to a shorter taper period. Medication expenses will be lower. The length of time during which a patient must be restricted from activities such as driving may be reduced. Likewise, the outcome of the attempt to discontinue treatment may be known sooner, shortening the period of adjustment for the patient.
On the basis of our findings of a similar risk of seizure recurrence with a six-week and a nine-month taper period, we recommend that antiepileptic medications be tapered over a period of six weeks when a decision is made to discontinue treatment of children with epilepsy.
Supported in part by grants from the General Clinical Research Center (RR 000046) and the National Institutes of Health (R01-DE-06541).
Source Information
From the Departments of Neurology (M. Tennison, R.G.) and Pediatrics (M. Tennison, R.G.) and the Curriculum in Neurobiology (R.G.), University of North Carolina at Chapel Hill; the Department of Pediatrics, Duke University, Durham, N.C. (D.L.); and Burroughs Wellcome Company, Research Triangle Park, N.C. (M. Thorn).
Address reprint requests to Dr. Tennison at CB# 7025, UNC-CH, Chapel Hill, NC 27599-7025.
References
| |||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
