SEARCH

SEARCH BY CITATION

Keywords:

  • Childhood absence epilepsy;
  • Absence seizures;
  • Levetiracetam;
  • Placebo;
  • Randomized controlled trial

Summary

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Appendix

Purpose: To evaluate the potential efficacy of levetiracetam as an antiabsence agent in children and adolescents with newly diagnosed childhood or juvenile absence epilepsy.

Methods: Patients were randomized in a 2:1 ratio to receive de novo monotherapy with levetiracetam (up to 30 mg/kg/day) or placebo for 2 weeks under double-blind conditions. Responder status (primary end point) was defined as freedom from clinical seizures on days 13 and 14 and from electroencephalographic (EEG) seizures during a standard EEG recording with hyperventilation and intermittent photic stimulation on day 14. The double-blind phase was followed by an open-label follow-up.

Key Findings: Nine of 38 patients (23.7%) were responders in the levetiracetam group, compared with one of 21 (4.8%) in the placebo group (p = 0.08). Seven of 38 patients (18.4%) were free from clinical and EEG seizures during the last 4 days of the trial (including 24-h EEG monitoring on day 14) compared with none of the patients treated with placebo (p = 0.04). Seventeen patients remained seizure-free on levetiracetam after 1 year follow-up. Of the 41 patients who discontinued levetiracetam due to lack of efficacy (n = 39) or adverse events (n = 2), 34 became seizure-free on other treatments.

Significance: Although superiority to placebo just failed to reach statistical significance for the primary end point, the overall findings are consistent with levetiracetam having modest efficacy against absence seizures. Further controlled trials exploring larger doses and an active comparator are required to determine the role of levetiracetam in the treatment of absence epilepsy.

The antiepileptic drugs (AEDs) most commonly used to treat typical absence seizures include ethosuximide, valproic acid, and lamotrigine (Posner, 2006; Wheless et al., 2007). None of these can be considered ideal (Posner & Panayiotopoulos, 2008; Arzimanoglou et al., 2010). Ethosuximide has unsurpassed efficacy against absence seizures, but it is ineffective against generalized tonic–clonic seizures, which often coexist in patients with idiopathic generalized epilepsies (IGEs) (Gören & Onat, 2007; Vining, 2010). Valproic acid is highly effective against absences and virtually all other seizure types, but it has significant side effects, including attentional deficits, weight gain, liver toxicity, and teratogenicity (Perucca, 2002; Glauser et al., 2010). Lamotrigine protects against absence and tonic–clonic seizures, but it is less efficacious than ethosuximide and valproic acid against absences (Coppola et al., 2004; Glauser et al., 2010), requires a long titration, and may cause serious cutaneous hypersensitivity reactions (Guberman et al., 1999).

Among other AEDs, levetiracetam has several features that make it an attractive candidate for the treatment of absence seizures. Preclinically it has been found to reduce the frequency and the duration of spike-wave epileptic discharges in established models of absence seizures, such as the Genetic Absence Epilepsy Rat from Strasbourg (GAERS) (Gower et al., 1995; Dedeurwaerdere et al., 2005; Kaminski et al., 2008) and the WAG/Rij rat (Bouwman & van Rijn, 2004; Russo et al., 2010), as well as in less specific models such as the DBA/2J mouse (Marrosu et al., 2007). Clinically, levetiracetam shows a favorable tolerability profile both in adults and children (Abou-Khalil, 2008), and its efficacy has been demonstrated by randomized controlled trials (RCTs) in partial seizures (Otoul et al., 2005; Brodie et al., 2007), primary generalized tonic–clonic seizures (Berkovic et al., 2007; Rosenfeld et al., 2009), and myoclonic seizures associated with juvenile myoclonic epilepsy (Noachtar et al., 2008). The potential usefulness of levetiracetam in the treatment of absence seizures is suggested by single case reports (Cavitt & Privitera, 2004; Lagae et al., 2005; Altenmüller et al., 2008), small case series (Koukkari & Guarino, 2004; Di Bonaventura et al., 2005) and a recent open-label uncontrolled trial in 21 patients with childhood and juvenile absence epilepsy (Verrotti et al., 2008). However, these observations need to be confirmed by adequately controlled studies.

Because absence seizures carry virtually no risk of causing direct serious harm and a treatment effect in this seizure type can be demonstrated within a very short period (at least with AEDs, which can be up-titrated rapidly), use of a brief course of placebo as sole treatment can be ethically justifiable in children with absence epilepsy who have no additional seizure types (Trudeau et al., 1996; Frank et al., 1999; Perucca, 2008). Based on this background, we conducted a short-term randomized, placebo-controlled study to determine whether levetiracetam is efficacious in controlling typical absence seizures in patients with newly diagnosed childhood or juvenile absence epilepsy.

Methods

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Appendix

Experimental design

The study was carried out in a total of 11 specialized epilepsy and child neurology centers in Italy (Appendix) according to a randomized, double-blind parallel-group placebo-controlled design, followed by an open-label long-term extension. The protocol was approved by the ethics committee of the promoting academic institution (IRCCS National Neurological Institute Casimiro Mondino Foundation, Pavia) and other participating centers. Written informed consent were obtained from the parents and, for children able to understand the implications of the research, also from study participants. The protocol of the study was registered in the national registry of clinical trials of the Italian Dug Agency (trial number code EudraCT 2005-003520-18 for the double-blind study and EudraCT 2005-003522-26 for the open-label extension). The first patient was enrolled on October 17, 2006 and the last patient completed double-blind treatment on November 20, 2008. The last assessment for the open-label extension phase took place on December 3, 2009.

Eligibility criteria

Subjects were enrolled according to the following inclusion criteria: (1) age between 4 and 16 years; (2) a recent diagnosis of childhood or juvenile absence epilepsy, as defined by the International League Against Epilepsy (ILAE) criteria (Commission, 1989); (3) electroencephalographic (EEG) evidence of regular, synchronous and symmetrical spike-wave paroxysmal discharges with a frequency of about 3 Hz and duration of at least 4 s, occurring spontaneously or during hyperventilation (a 2–3-min hyperventilation test, followed 2 min later by a second hyperventilation if paroxysms were not detected in the first test) or intermittent photic stimulation (IPS; 5 s stimulation at each frequency in the 10–30 Hz range, eye closed); (4) a history of clinically evident spontaneously occurring absence seizures impacting on functional abilities; and (5) written informed consent.

Exclusion criteria were (1) a history of generalized tonic–clonic seizures; (2) clinical or EEG findings inconsistent with a diagnosis of childhood or juvenile absence epilepsy; (3) previous treatment with AEDs (except for earlier treatments for other indications such as febrile seizures, or brief exposures to other AEDs prior to diagnostic assessment); (4) intellectual disability; (5) clinically significant hepatic or renal disorders; (6) history of hypersensitivity reactions to study products or structurally related substances; and (7) any condition that, in the investigator’s judgment, was expected to impact negatively on subjects’ health or study procedures.

Baseline assessment and randomization procedure

At baseline, all eligible subjects underwent a medical examination and safety blood chemistry and hematology tests. Information on seizure frequency during the previous 3 days was obtained from the parents using a semiquantitative scale. Thereafter, a standard EEG recording with hyperventilation and IPS was obtained, followed by a 24 h ambulatory EEG.

Subjects exhibiting the specified EEG paroxysms during at least one of these EEG assessments were randomized in a 2:1 ratio to receive levetiracetam or a placebo with identical appearance and taste. Random assignment was carried out using computer-generated random numbers with blocks of 6, to maintain blinding for subsequent assignments when the code of each patient was disclosed at the end of the study (see subsequent text).

Treatments and dosing schedules

Depending on subject’s age, body weight, and preference, either a liquid formulation (100 mg/ml solution) or 500 mg tablets were used, with corresponding placebos.

Levetiracetam dosage was 10 mg/kg/day for 3 days (days 1–3), followed by 15 mg/kg/day for the next 4 days (days 4–7). If tolerability was acceptable and clinical seizures occurred at any time between 1 h after the dose increase on day 4 and end of day 7, or if epileptic discharges were detected during the EEG assessment on day 7 (see below), dosage was further increased to 20 mg/kg/day on day 8 and, if well tolerated, to 30 mg/kg/day on day 11 and maintained until day 14. If no clinical or EEG seizures were detected on days 4–7, but clinical seizures re-emerged during days 8, 9, or 10, dosage was increased to 20 mg/kg/day on the day of seizure emergence and, if well tolerated, increased again after 3 days to 30 mg/kg/day and maintained until day 14. If no clinical or EEG seizures were detected during the double-blind period (or if seizures were detected after day 10 only), the dose was maintained from day 4 to day 14 at 15 mg/kg/day. If a tonic–clonic seizure occurred during the 14-day period, the subject was required to exit the study immediately, because it was not considered ethical to continue such subjects on a treatment that could potentially be placebo.

Treatments were administered in the morning and in the evening in two equally divided doses. For tablets, the dose that provided the closest approximate to the nominal mg/kg dose was used.

After completion of evaluations on day 15 (or earlier for patients exiting the trial prematurely), clinical investigators were required to contact the central randomization center and to provide details on outcome data up to that time. The randomization code was then broken and the patient allowed to continue on an extension open-label study. In the extension study, treatments could be modified as clinically indicated and placebo-treated patients were given the option of receiving levetiracetam.

Evaluations

Efficacy evaluations during the 14-day double-blind period included the recording of clinical seizures by parents on a semiquantitative scale and two EEG assessments, one on day 7 and one on day 14. On day 7, a standard EEG recording with hyperventilation and IPS was conducted as described for baseline. If no epileptic EEG discharges were detected during the standard EEG, an additional EEG recording of 3-h duration with nap was obtained. On day 14, the standard EEG with IPS and hyperventilation was repeated, followed by a 24-h ambulatory EEG.

Tolerability assessments included recording of adverse events and medical examinations on study visits on days 7 and 14. If considered clinically indicated, safety laboratory tests could be repeated on day 14.

In the extension study, further evaluations were conducted as considered indicated by the attending physician and generally included standard EEG in addition to clinical assessments. Information about treatment changes, seizure freedom status, and adverse events were collected for up to 1 year during follow-up.

Outcome measures

The primary end point was the percentage of patients free from clinical seizures on days 13 and 14 and from EEG seizures during the standard EEG with hyperventilation and IPS on day 14. An EEG seizure was defined as a spike-wave discharge of at least 4-s duration.

Secondary end points included: (1) percentage of patients free from clinical and EEG seizures lasting ≥4 s on days 11–14 (including the 24-h ambulatory EEG); (2) percentage of patients free from clinical and EEG seizures lasting ≥4 s on days 4–7 (including the 3-h recording with nap); (3) percentage of patients free from clinical seizures from day 1 to day 14 and from EEG seizures lasting ≥4 s during standard EEGs on days 7 and 14; (4) percentage change (vs. baseline) in number and total duration of EEG seizures and spike-wave discharges lasting ≥4 s during the 24-h EEG on day 14; (5) percentage of patients with at least 50% reduction (vs. baseline) in the total duration of EEG seizures lasting ≥4 s during the 24-h EEG on day 14; and (6) safety and tolerability data.

Sample size consideration and statistical analysis

Assuming a 40% greater responder rate for the primary end point on active medication than on placebo, and a 10% responder rate on placebo, a sample size of 51 patients was calculated to provide 80% power to identify a difference with a p-value of 0.05. To account for potential dropouts, the target sample size was set at 60 subjects (40 levetiracetam, 20 placebo).

In agreement with the study protocol, differences in responder rates (primary end point) between groups were assessed by using the two-tailed Fisher’s exact test with the significance level set at 0.05. Analysis was conducted on the intent-to-treat (ITT) population, defined as subjects who took at least one dose. In this analysis, subjects who did not complete the 14-day double-blind period were considered nonresponders.

For secondary end points, between-group comparisons were made exploratively by using the two-tailed Fisher’s exact test for percentages, and the two-tailed Student t-test for percent changes versus baseline. Secondary analysis were conducted on both the ITT and the per protocol (PP) population (patients who completed 14-day double-blind treatment). A secondary analysis also included a between-group comparison of the primary end point in the PP population.

Results

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Appendix

Clinical characteristics and disposition of subjects

Of a total of 71 patients screened, 59 (27 male, 32 female, age 4–15 years) met eligibility criteria and were randomized (38 to levetiracetam and 21 to placebo) (Fig. 1). As shown in Table 1, the main demographic and clinical characteristics were comparable in the two groups. All subjects had spontaneously occurring absences confirmed during the standard EEG and/or the ambulatory 24-h EEG recording at baseline.

image

Figure 1.   Disposition of subjects who participated in the study.

Download figure to PowerPoint

Table 1.   Characteristics of the subjects enrolled in the study
 Levetiracetam group (n = 38)Placebo group (n = 21)
  1. n, number of subjects; n.a. not applicable.

  2. an = 37 because one child discontinued treatment prematurely.

Gender distribution23 F 15 M9 F 12 M
Age (years, mean ± SD and range)8.7 ± 2.2 (4.9–13.0)7.9 ± 3.0 (4.0–15.0)
Body weight (kg, mean ± SD and range)33.7 ± 12.4 (20.0–64.5)34.5 ± 14.9 (17.0–64.0)
Syndromic diagnosis (n)
 Childhood absence epilepsy3420
 Juvenile absence epilepsy 4 1
 Not established/not available 0 0
Seizure frequency at pretreatment (n)
 >10 per day2618
 6–10 per day 6 0
 1–5 per day 4 2
 1–6 per week 2 1
Mean levetiracetam dose (mg/day, mean ± SD)
 Day 3335.5 ± 114.6n.a
 Day 7510.4 ± 192.3n.a
 Day 11878.4 ± 367.5n.a
 Day 14913.0 ± 373.5an.a

Of the subjects randomized to levetiracetam, 37 completed the final double-blind assessment. In one subject the final assessment on double-blind treatment was conducted a day earlier than scheduled because the child had developed a widespread urticarioid reaction, which was subsequently found to be due to a streptococcal infection. This was considered as a minor violation and the subject was included among the completers. One child withdrew prematurely on day 12, when her parent stopped treatment due to appearance of drowsiness and dizziness, which eventually resolved on day 14. Among mean completers, levetiracetam dose [standard deviation (SD)] was 15.3 (1.9) mg/kg on day 7 and 28.5 (4.8) mg/kg on day 14.

Of the subjects randomized to placebo, all completed double-blind treatment.

Efficacy end points

To meet the primary efficacy end point (responder status), subjects had to have no clinical seizures on days 13 and 14, and no EEG seizures during the standard EEG on day 14. A total of 9 of 38 patients (23.7%) were responders in the levetiracetam group, compared with 1 of 21 (4.8%) in the placebo group. The difference just failed to reach statistical significance (p = 0.08, Fisher’s exact test). In the secondary analysis conducted in the PP population, results were very similar, with a responder rate of 21.6% (8 of 37) in the levetiracetam group and 4.8% (1 of 21) in the placebo group (p = 0.13).

The results of the additional efficacy analyses (ITT) are summarized in Table 2. Seven of 38 patients (18.4%) were completely free from clinical and EEG seizures during the last 4 days of the trial (including 24-h EEG monitoring on day 14) compared with none of the 21 placebo-treated patients, a statistically significant difference (p = 0.04). Only four subjects (all on levetiracetam) were seizure-free during the last 4 days of the first treatment week, when levetiracetam dose could not exceed 15 mg/kg/day, and only two subjects (both on levetiracetam) were seizure-free for the entire period from day 1 to day 14. Changes in number and duration of epileptic discharges during the 24-h EEG on day 14 did not differ significantly between groups.

Table 2.   Secondary efficacy end points (ITT analysis)
 Levetiracetam group (n = 38)Placebo group (n = 21)p-Value
Patients free from clinical and EEG seizures on days 11–147/38 (18.4%)0/21 (0%)0.043
Patients free from clinical and EEG seizures on days 4–74/38 (10.5%)0/21 (0%)0.286
Patients free from clinical and EEG seizures from day 1 to day 14 (excluding prolonged EEG recordings on days 7 and 24)2/38 (5.2%)0/21 (0%)0.534
Patients with at least 50% reduction (vs. baseline) in total duration of EEG seizures during the 24-h EEG on day 1412/38 (31.6%)3/21 (14.3%)0.214
Percentage change in number of EEG discharges during 24 h EEG on day 14 versus baseline (mean ± SD)−10 ± 80%−7 ± 33%0.856
Percentage change in duration of EEG discharges during 24 h EEG on day 14 versus baseline (mean ± SD)3 ± 98%−9 ± 31%0.592

Results of additional statistical analysis on the PP population were very similar to those for the ITT population (data not shown), which was expected because the two populations were almost identical.

Although the study was not powered to test for factors influencing clinical response to levetiracetam, a number of variables were evaluated descriptively. Three of the 4 patients with juvenile absence epilepsy were responders, compared with only 6 of the 34 patients with childhood absence epilepsy. Compared with nonresponders, responders tended to be older (10.5 ± 2.1 vs. 8.1 ± 2.0 years, means ± SD), to have fewer epileptic discharges at the baseline EEG (2.8 ± 2.6 vs. 3.6 ± 4.6) and 24-h ambulatory EEG (40.1 ± 28.6 vs. 75.9 ± 61.3), and to have a shorter total duration of discharges (495 ± 600 vs. 737 ± 568 s) at the baseline 24-h ambulatory EEG. The overlap between responders and nonresponders was, however, considerable. Among the nine responders, seven had epileptic discharges during hyperventilation and two during IPS at the baseline EEG. Among the 29 nonresponders, 28 had epileptic discharges during hyperventilation and five during IPS.

Tolerability and safety data

No serious adverse events were recorded. Treatments were generally well tolerated. Seven subjects on levetiracetam (18%) and three on placebo (14%) reported adverse events. Adverse events considered by the investigator to be at least possibly related to treatment occurred in three children, all randomized to levetiracetam. These consisted of somnolence and irritability (n = 1), irritability and dysphoria (n = 1), and dizziness and drowsiness in the only child who discontinued treatment prematurely.

Long-term open-label follow-up

Of the 38 patients randomized to levetiracetam, 12 continued on levetiracetam monotherapy (30–50 mg/kg/day) and were seizure-free for at least 267 days at the last follow-up. This group included four patients who did not achieve seizure-free status during the double-blind phase, but became free from seizures during follow-up (in three cases following a dose increase). Of the remaining 26 patients, 24 discontinued levetiracetam (highest dose tested 70 mg/kg/day, range 15–70 mg/kg/day) 14–64 days after the onset of treatment due to persistence of seizures. One child with uncontrolled seizures discontinued after 13 days due to drowsiness and dizziness, and another child who was seizure-free discontinued after 22 days due to a drug-unrelated urticarioid reaction. Of the 26 patients who discontinued levetiracetam, 24 were seizure free at the last follow-up at least 6 months later, and of these 21 were on valproate and 3 on ethosuximide. Two children, both treated with valproate in combination with lamotrigine (n = 1) or zonisamide (n = 1) still had uncontrolled absences.

Of the 21 patients randomized to placebo, one had a generalized tonic–clonic seizure on day 15 and was immediately started on valproate, without further seizures during follow-up. All other patients were started on levetiracetam, and 5 of these were seizure-free on levetiracetam (20–30 mg/kg/day) at the last follow-up after 365–602 days of active treatment. Of the 15 patients who did not respond to levetiracetam, 10 were seizure-free at the last assessment at least 6 months later while on treatment with valproate (n = 6), ethosuximide (n = 3), or a combination of valproic acid and ethosuximide (n = 1). Of the five who remained uncontrolled, two were receiving ethosuximide, two valproate, and one a combination of lamotrigine and levetiracetam.

Overall, of the initial cohort of 59 patients, 17 were seizure free on levetiracetam at the last follow-up, and 34 were seizure-free on other treatments, which included valproate (n = 27), ethosuximide (n = 6), and the combination of valproate with ethosuximide (n = 1). Adverse events considered to be treatment-related were recorded during follow-up in five children. These included somnolence in a child treated with levetiracetam, vomiting (n = 1), nausea and abdominal pain (n = 1), and skin rash (n = 1) in three children treated with ethosuximide, and hair loss in a child treated with valproate (the same child had previously reported weight gain and irritability when treated with levetiracetam).

Discussion

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Appendix

Despite extensive evidence of levetiracetam possessing antiabsence effects in animal models (Gower et al., 1995; Bouwman & van Rijn, 2004; Dedeurwaerdere et al., 2005; Marrosu et al., 2007; Kaminski et al., 2008; Russo et al., 2010), there is surprisingly little information on its potential efficacy in patients with absence seizures. In the first published report, Cavitt and Privitera (2004) found levetiracetam (4,000 mg/day) effective in decreasing the frequency of absence-like seizures and generalized 3.5-Hz spike-wave and polyspike-wave EEG discharges in a 34-year-old woman with refractory generalized epilepsy. The effect was reversible after discontinuation of the drug, and was restored after reintroduction. Efficacy of levetiracetam against absence seizures, including absence status epilepticus (Altenmüller et al., 2008), was suggested by subsequent case reports and very small case series (Koukkari & Guarino, 2004; Lagae et al., 2005), but these publications may be affected by reporting bias and do not provide estimates of responder rates. In a recent open-label prospective trial, 9 (42.9%) of 21 children with newly diagnosed childhood or juvenile absence epilepsy treated with levetiracetam (31–70 mg/kg/day) were reported to be free from clinical and EEG seizures at the 6-month assessment (Verrotti et al., 2008). However, no details were given about the duration of EEG recordings, and the lack of a control group makes these results difficult to interpret.

In the past, a placebo-controlled monotherapy design to assess efficacy against absence seizures has been applied in two trials with gabapentin (Trudeau et al., 1996) and one with lamotrigine (Frank et al., 1999). Because gabapentin was not found to be effective and lamotrigine was tested in an enriched group of patients who had previously responded to the drug, these trials do not provide sufficient information to optimize choice of end points and calculation of sample size. Because in routine clinical practice, response to antiabsence drugs is generally monitored by recording parental reports and standard EEGs, these measures were selected as primary end point, using as assessment the last 2 days of the double-blind phase during, which participants were generally on the highest dose tested. Sample size was based on the assumption of a low response on placebo (Trudeau et al., 1996; Frank et al., 1999) and a 50% responder rate to active medication, which is comparable to the 53% responder rate reported after 1 month on valproic acid in a recent randomized trial (Coppola et al., 2004), although in the latter trial response was defined more conservatively as lack of clinically observed seizures since the previous visit and lack of recorded seizures during a 24-h EEG and a video-EEG session with hyperventilation. Because of the need to minimize exposure to placebo, the trial had a short duration, and high levetiracetam doses could not be achieved, although the doses reached were within the range recommended for children with other seizure types (French & Tonner, 2009).

Based on the primary end point selected, levetiracetam was associated with a responder rate of 23.7% compared with 4.8% for placebo. Although the difference failed to reach the threshold of statistical significance, the probability of levetiracetam being superior to placebo based on these data was still >90%. An antiabsence effect of levetiracetam is further supported by the observation that the proportion of patients free from clinical and EEG seizures, including a final 24 h EEG, was significantly greater in the levetiracetam than in the placebo group. At least in a subgroup of patients, efficacy appeared to be maintained long-term because 17 of the 58 patients (29%) in whom levetiracetam was tested remained on the drug for at least 1 year during open-label follow-up and were seizure-free at the last assessment. Our sample size was insufficient to ascertain prognostic factors associated with a good response, although the trend for the few subjects with juvenile myoclonic epilepsy to respond better compared with those with childhood absence epilepsy is interesting and should be investigated in future studies.

Because of expected difficulties with recruitment and the complexity of maintaining double-blind conditions, we elected not to include an established active control in the trial. As a result, how levetiracetam would compare with other drugs assessed under the same conditions is not known. In a recent open-label randomized trial in a total of 38 children with absence epilepsy, seizure freedom rates at 3 months, defined more conservatively than in our trial, were 63.1% for valproate (20–25 mg/kg/day) and 36.8% for lamotrigine (2–11.5 mg/kg/day) (Coppola et al., 2004). At 12 months, responder rates increased to 68.4% for valproic acid (20–30 mg/kg/day) and 52.6% for lamotrigine (2–11 mg/kg/day). In a more recent double-blind trial of 16–20 weeks restricted to patients with childhood absence epilepsy, freedom from intolerable adverse events and from seizures occurring spontaneously or during a 1-h video-EEG session was reported in 58% of children with valproate, 53% with ethosuximide, and 29% with lamotrigine (Glauser et al., 2010). Although these data suggest a lower responder rate with levetiracetam than with other antiabsence drugs, comparisons should be interpreted cautiously due to differences in design, study population, and, most notably, duration of treatment. With respect to doses, the short double-blind period in our study allowed testing of levetiracetam only at doses up to 30 mg/kg/day, which are considerably lower than the median dose (50 mg/kg/day, range 31–79) used in the uncontrolled trial by Verrotti et al. (2008). During long-term open label follow-up, however, some of our patients with uncontrolled seizures also did receive higher doses, including nine patients who continued to have seizures after receiving doses of 60–70 mg/kg/day. Although open-label data should be interpreted cautiously, 34 of 41 patients (83%) in whom levetiracetam had failed subsequently achieved seizure freedom on other treatments, including 27 treated with valproic acid.

In conclusion, our trial provides suggestive evidence that levetiracetam is superior to placebo in suppressing absence seizures in patients with newly diagnosed childhood or juvenile absence epilepsy. Although responder rates were modest, efficacy may have been underestimated by the short duration of treatment and the use of relatively low doses. Further controlled trials, including preferably an established active comparator, appear to be justified to ascertain whether levetiracetam could have a useful role in the de novo treatment of absence epilepsy.

While further data are awaited, ethosuximide should probably be considered as the agent of choice for the treatment of childhood absence epilepsy, due to recent evidence that it is associated with lesser risk of cognitive impairment compared with valproic acid (Glauser et al., 2010). Valproic acid, however, is a preferable choice in subjects at higher risk of developing tonic–clonic seizures, for example, in juvenile absence epilepsy (Arzimanoglou et al., 2010). In patients with typical absence seizures not controlled with ethosuximide and/or valproic acid monotherapy, treatment options include the combination of both agents, or the introduction of lamotrigine or levetiracetam. Factors to be considered in drug selection include the risk of associated seizure types, and the tolerability profile of the individual drugs. In certain circumstances, as in female adolescents in whom the teratogenicity of valproic acid is a special concern (Tomson et al., 2004), lamotrigine and levetiracetam may also be considered for first-line therapy, because expected safety advantages may outweigh the drawback of their probably inferior antiabsence effect.

Acknowledgments

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Appendix

We wish to thank Dr. Emanuela Gerosa for running the central randomization office, Dr. Sara Poledri for help with the clinical and EEG assessments, and Dr. Cristian Brocchieri for assistance with data entry. This work was supported by a grant from UCB S.p.A., Pianezza, Italy.

Disclosure

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Appendix

Emilio Perucca received speaker’s or consultancy fees and/or research grants from the manufacturers of carbamazepine and oxcarbazepine (Novartis); carisbamate and topiramate (Johnson & Johnson); eslicarbazepine acetate (BIAL, Sepracor); ethosuximide, gabapentin, phenytoin, and pregabalin (Pfizer); lamotrigine (GSK); brivaracetam, lacosamide, and levetiracetam (UCB Pharma); retigabine (Valeant); tiagabine, valproic acid, and vigabatrin (Sanofi-Aventis); and rufinamide and zonisamide (Eisai). Giuseppe Capovilla received speaker’s or consultancy fees and/or research grants from the manufacturers of carbamazepine and oxcarbazepine (Novartis); carisbamate and topiramate (Johnson & Johnson); ethosuximide, gabapentin, phenytoin, and pregabalin (Pfizer); lamotrigine (GSK); brivaracetam, lacosamide, and levetiracetam (UCB Pharma); tiagabine, valproic acid, and vigabatrin (Sanofi-Aventis); and rufinamide and zonisamide (Eisai). Giangennaro Coppola received speaker’s and/or research grants from the manufacturers of topiramate (Johnson & Johnson); levetiracetam (UCB Pharma); and rufinamide (Eisai). The remaining authors have no conflicts of interest to declare.

We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

References

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Appendix

Appendix

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Appendix
List of trial centers and clinical investigators at each site
CentersInvestigators
Child Neuropsychiatry Unit, Bologna University, BolognaEmilio Franzoni, Ilaria Cecconi, Caterina Garone, Luisa Iero, Valentina Marchiani, Filomena Moscano, Grazia Salerno
Child Neuropsychiatry Division, University of Cagliari, CagliariDario Pruna, Melania Falchi
Institute of Neurology, Magna Graecia University, CatanzaroAntonio Gambardella, Roberta Ambrosio, Eleonora Colosimo, Angelo Labate
Department of Child Neuropsychiatry, C Poma Hospital, MantuaGiuseppe Capovilla, Paolo Avantaggiato, Francesca Beccaria, Alessandra Montanini
Clinic of Child Neuropsychiatry, Second University of Naples, NaplesGiangennaro Coppola, Salvatore Arcieri, Francesca Operto
Clinical Neurophysiology, Department of Pediatrics, PaduaClementina Boniver, Sara Poledri, Marina Saladini, Marilena Vecchi
Department of Child Neurology and Psychiatry, IRCCS National Neurological Institute Casimiro Mondino Foundation, PaviaPierangelo Veggiotti, Enrico Alfei, Antonietta Citterio, Cinzia Fattore, Emilio Perucca, Federica Teutonico
Department of Neuroscience, Child Neurology Unit, Tor Vergata University of Rome, RomePaolo Curatolo, Caterina Cerminara, Luigi D’Argenzio
Child Neuropsychiatry Unit, Department of Pediatric and Adolescence Sciences, University of Turin, TurinGiorgio Capizzi, Ilaria Pieri, Elena Rainò, Roberta Vittorini
Department of Child Neuropsychiatry, IRCSS Burlo Garofolo, TriesteMarco Carrozzi, Paola Costa, Raffaella Devescovi, Aldo Skabar
Unit of Child Neuropsychiatry, University of Verona, VeronaBernardo Dalla Bernardina, Francesca Darra, Elena Fontana, Elena Fiorini