Address correspondence to Friedhelm C. Schmitt, Department of Neurology, University of Magdeburg, Leipziger Str. 44, 39120 Magdeburg, Germany. E-mail: firstname.lastname@example.org
Purpose: This study investigates immediate efficacy and safety of intravenous application of de novo lacosamide (LCM) as add-on therapy in patients with pharmacoresistant focal epilepsy.
Methods: During presurgical video–electroencephalography (EEG) monitoring, 17 adult inpatients received LCM infusion (200 mg every 12 h for 2–3 days) followed by oral formulation with the same regimen. Before and after intravenous application of LCM, seizures and interictal epileptiform discharges (IEDs) recorded with continuous video-EEG monitoring were analyzed, and an assessment of adverse events (AEs) was performed daily. To evaluate the midterm tolerability and efficacy, follow-up visits were conducted 1 and 3 months after discharge from hospital.
Key Findings: In the acute phase, intravenous initiation of LCM was well tolerated with few mild or moderate AEs (3 of 17, 17.6%). A significant reduction of seizure frequency in the treatment phase as compared to mean seizure frequency in the 2-day baseline phase was achieved (p < 0.05 for the first treatment day, and p < 0.005 for the second treatment day). On the first treatment day, 61.5% of the patients were seizure free, and 84.6% on the second treatment day. IED reduction after intravenous application of LCM was not significant. After 1 month, the 50% responder rate was 46.6% and after the 3-month period, 42.8%.
Significance: Our data suggest that rapid intravenous initiation of de novo LCM is safe and may protect against seizures in a rapid and midterm time window.
Currently, >20 antiepileptic drugs (AEDs) with different mechanisms of action are available. Most of them can neither be rapidly titrated without severe adverse effects (AEs) nor have a verifiable immediate efficacy. Lacosamide (LCM) has linear pharmacokinetics (Horstmann et al., 2002), minimal protein binding (Horstmann et al., 2002; Thomas et al., 2006), minimal interactions with other AEDs or other drugs (Chung, 2010) and is available in both oral and intravenous formulations. Previous studies showed that intravenous LCM as a short-term parenteral replacement for oral therapy was well tolerated (Biton et al., 2008; Krauss et al., 2010). In addition, the nature and frequency of AEs in these studies were consistent with the AEs most commonly reported for oral administration of LCM (dizziness, headache, nausea, and diplopia) (Biton et al., 2008; Krauss et al., 2010).
The setting of video-EEG monitoring can provide a precise evaluation of the efficacy and safety of AEDs (Stefan et al., 2001), and proof-of-principle studies have been performed in this setting, for example, for felbamate (Bourgeois et al., 1993; Devinsky et al., 1995), losigamone (Stefan et al., 2001), topiramate (Wang et al., 2002), and levetiracetam (Stefan et al., 2006). However, AED trials in the setting of video-EEG monitoring incorporate several problems: during a predefined study design the ethical dilemma could arise that continuation of video-EEG monitoring remains necessary due to the protocol, even though sufficient electroclinical data for diagnostic purposes had already been collected (Schmidt, 2006). Furthermore, because of the short observation period, it has been questioned whether the presurgical setting alone sufficiently contributes to relevant clinical information (Mohanraj & Brodie, 2003), for example, in terms of midterm efficacy and tolerability of the newly administered agent.
In consideration of the optimal pharmacokinetic properties of intravenous LCM and its putative benefit for the patients who need fast anticonvulsant protection after AED dose reduction, we prospectively investigated the acute efficacy and tolerability of de novo application of intravenous LCM. Concerning an acute electroencephalic effect, the change of interictal epileptiform discharges (IEDs) before after the LCM application was determined. To enhance informative value in the clinical context, we added a midterm follow-up of 1 and 3 months.
The present study was conducted from June 1, 2009 to November 30, 2011. During this time, 93 consecutive patients with pharmacoresistant focal epilepsies were admitted to our hospital for presurgical video-EEG monitoring. Exclusion criteria for the study were the following: age younger than 18 years, inability to give full consent, pregnancy, previous application of LCM, presence of atrioventricular block or other cardiac dysrhythmia in the baseline electrocardiography (ECG) and increased serum creatinine level in clinical laboratory test. According to these criteria, 17 patients could be included in the study.
The primary outcome parameter was acute efficacy and tolerability of rapid loading of intravenous LCM. As a secondary outcome parameter, the midterm efficacy and tolerability was assessed at 1- and 3-month visits.
During the outpatient clinic visit, we obtained patient history and the clinical, EEG and magnetic resonance imaging (MRI) data to assess whether presurgical video-EEG monitoring would be useful for the patient. Criteria included pharmacoresistance to two adequately dosed AEDs and clinical, electroencephalographic, or morphologic findings suggesting focal onset of seizures. The seizure frequency in the last 3 months before entry into this study was recorded from the patient’s diaries on the day of admission for the video-EEG monitoring. AEDs with negligible efficacy or apparent side effects were gradually tapered off in order to provoke seizures (Marks et al., 1991). The withdrawal period lasted from 2 to 10 days and LCM was never administered without AED comedication. All patients consented to AED withdrawal, its potential risks, and the off-label use of rapid intravenous LCM. To evaluate acute efficacy, we defined a 4-day investigational period with a 2-day baseline period and followed by 2-day treatment period during which the patients received the intravenous administration of LCM twice daily (Fig. 1). Time of initiation of LCM infusion was variable and decided on clinical grounds. After sufficient electroclinical information was obtained, the patients received intravenous LCM infusion (200 mg every 12 h) for 2–3 days, followed by oral LCM with the same regimen. Infusion time was 15 min in 10 patients, 30 min in 5 patients, and 60 min in 2 patients. There was no dose increase of concomitant AEDs or addition of new AED during the video-EEG monitoring or during the 3-month follow-up period. The patients had follow-up visits 1 and 3 months after discharge from the hospital, usually in the outpatient clinic or, if necessary, by telephone interview.
Data acquisition: seizure frequency, seizure duration, frequency of interictal epileptiform discharges, and adverse events
During the whole investigation period, the EEG data were acquired using simultaneous Neurofile XP digital video EEG system (ITmed, Usingen, Germany), with 31 EEG channels and one ECG channel, at a 256-Hz sampling rate. The seizure frequency was defined as the number of seizures recorded in 24 h and the duration of the seizures as the clinic duration in seconds. All patients were asked to report the beginning of a seizure by pressing an alarm button. If the patients were unable to press the button, clinical onset was defined by post hoc analysis of the video. Clinical laboratory and 12-lead ECG test were performed upon admission. Blood pressure was measured once daily. A daily assessment of AEs and a clinical screening for cerebellar symptoms was carried out once daily after LCM application. Seizure frequency and tolerability were assessed in the two follow-up visits, and LCM dosage or the concomitant AED was adjusted if the patients had AEs.
The EEG was retrospectively analyzed for IED frequency. During the 2-day baseline period and the treatment periods 1 and 2, a 2.5-min EEG segment was stored every hour and then visually analyzed for occurrence of IEDs. Criteria for an epileptiform discharge were defined according to the literature (Gloor, 1977). Spikes and sharp waves were manually counted during the 2-day baseline period and the treatment periods 1 and 2, retrospectively. The sum of the IEDs was divided by the sum of minutes of visually analyzed EEG segments. The frequency of IEDs of the 2-day baseline periods was compared with the treatment periods 1 and 2. EEG segments with considerable artifacts were not regarded for analysis.
The mean of seizure number in the 2-day baseline phase was compared with the seizure number on both the first and the second treatment days using Mann-Whitney U test. We assumed a nonparametric, skewed data distribution. The AEs and the vital signs were summarized for all patients. The sum of the seizure durations and mean values of them were calculated and compared by Mann-Whitney U test. For comparison of IEDs frequency before and after LCM infusion the paired, two-tailed Student’s t-test was used. The analyses were performed by using IBM SPSS Statistics version 19.0 for Windows (IBM Corporation, Armonk, NY, U.S.A.).
A total of 17 patients with refractory partial epilepsy were enrolled in this study. The mean seizure number in the baseline period (2 days) was 2.1 ± 2.7/24 h. Four patients were excluded from the analysis of the immediate efficacy of intravenous administration of LCM, because they had no seizures during the baseline period. They were still enrolled for the part of evaluation of safety and midterm efficacy. None of them had seizures after LCM application. The demographic data and baseline characteristics are shown in Table 1. Eight of 17 patients had temporal lobe epilepsy (left, 4; right, 1; left and right, 3), and nine patients had extratemporal epilepsy.
Table 1. Patient characteristics
Total (n = 17)
Mean ± SD
40.2 ± 10.9
Seizure history (year)
Mean ± SD
22.5 ± 9.2
Number of previous AEDs
Mean ± SD
3.4 ± 1.6
Number of AEDs at entry
Mean ± SD
1.8 ± 0.7
Seizure number per month (n)
Mean ± SD
16.3 ± 37.1
During the treatment period, 3 of the 17 patients (17.6%) reported AEs. Two of them experienced transient AEs (dizziness/headache) and recovered spontaneously within 48 h (15- and 30-min infusion time). The third patient (30-min infusion time) reported diplopia after each LCM infusion and recovered after a dose reduction from 500–300 mg/day of the concomitant AED lamotrigine (LTG). There were no relevant changes in laboratory values, mean heart rate, or blood pressure. During the treatment period and at the time of discharge, no cerebellar symptoms or other focal signs were detected in the clinical screening.
Despite active recruiting, 2 (11.8%) of 17 patients were lost at the 1-month follow-up visit. Four (26.7%) of 15 patients reported AEs at 1-month follow-up, in one patient LCM application had to be discontinued, and in three patients AEs subsided either spontaneously or with dose reduction. At the 3-month follow-up, three patients (21.4%) had AEs; in one patient LCM had to be discontinued, and the two remaining patients had a dose adjustment of LCM or the concomitant AED LTG. Two patients with neurotoxic AEs during the intravenous LCM application and the two patients in whom LCM had to be discontinued had a classical sodium channel blocker (LTG) in the concomitant AED (see Table 2).
Table 2. Incidence of adverse events
Adverse event rate (%)
LTG, lamotrigine; LEV, levetiracetam; VPA, valproic acid; GBP, gabapentin.
Changes of seizure frequency and duration
Thirteen patients experienced a total of 54 seizures during the 2-day baseline period. Eight (61.5%) of 13 patients were seizure-free on the first treatment day, and 11 (84.6%) on the second treatment day (Fig. 2A). The mean seizure frequency in the baseline phase with 2.1 seizures per 24 h was significantly reduced to 1.0 seizure per 24 h on the first (p < 0.05) and to 0.8 seizures per 24 h on the second treatment day (p < 0.005) (Fig. 2B). No significant difference of the seizure duration between the baseline and treatment phase was observed. However, the sum of seizure duration decreased significantly, the mean value from 93.2 s (±69.4) in the baseline phase to 30.8 s (±46.6) on the first treatment day (p < 0.05), and 9.4 s (±28.9) on the second treatment day (p = 0.001). A significant reduction of seizure frequency (≥50% reduction in comparison with the seizure frequency in the 3 months before entry into this study) was achieved in 46.6% (7 of 15) of the patients in the 1-month observation period and 42.8% (6 of 14) of the patients in the 3-month observation period.
Frequency of interictal epileptiform discharges
From the 17 patients, 10 patients (58.8%) had interictal spikes or sharp waves. None had IEDs exclusively in the treatment period 1 or 2. There was a reduction of 0.45 (± [standard deviation] 0.54) IEDs per minute during the 2-day baseline period to 0.31 (±0.36) IEDs per minute, which did not reach statistical significance (p = 0.079).
We investigated systematically the effect of de novo intravenous administration of LCM in the clinical setting of a presurgical follow-up. The main findings were the following:
1 There was a pronounced acute antiseizure effect after intravenous LCM infusion.
2 The de novo intravenous rapid loading of LCM was well tolerated, with few transient mild or moderate AEs.
In 58.8% of our patients interictal epileptiform activity was noted, which was in line with other reports on focal epilepsies (Knake et al., 2006). After rapid loading of LCM there was a tendency toward reduction of interictal epileptiform activity, which did not reach significance. A significant reduction of IED frequency could have been interpreted as a result of an acute antiepileptic effect of LCM. However, in this case, the limited number of patients would not have allow a detailed subanalysis in view of the known complex interplay of occurrence of seizures and epileptiform discharges (Gotman & Koffer, 1989; Janszky et al., 2005), which also depends on different factors such as seizure location, sleep-wake cycle (Malow et al., 1998), seizure frequency, and epilepsy duration (Janszky et al., 2005).
In a previous study (Krauss et al., 2010), the frequency of the AEs (19 of 97, 19.6%) was similar to the presented results (17.6%). In our study, more patients complained about AEs after switching from intravenous to oral therapy: After the intravenous application of LCM, AEs increased from 17.6% to 26.7% at the 1-month follow-up and to 21.4% at the 3-month follow-up. Since there was no dose increase of the concomitant AEDs during the study, three reasons for the lower rate of AEs during the LCM infusion in comparison to the follow-up visits could be hypothesized. First, the patients had less physical activity than in their normal life, since our patients spent most of the time in bed during the video-EEG monitoring. However, the daily clinical screening should have detected significant symptomatic AEs. Secondly, the route of LCM application differs, so that either the gastric route itself (e.g., via a topic mode of action) or the >20 components of the tablets could be responsible for the increased frequency of AEs during the midterm follow-up. Thirdly, the incidence of side effects could be a function of the time of drug intake itself and not of titration rate or administration route.
Our data suggest that rapid intravenous initiation of LCM is well tolerated. Intravenous administration of LCM is not only advantageous for patients who are unable to take oral medications, but also helpful for patients who need a rapid effective protection against seizures. Clinical settings suitable for intravenous administration of LCM could be seizures in clusters, status epilepticus (SE), and prolonged or increased seizure activity due to withdrawal of AEDs, as in the setting of a presurgical workup. Concerning refractory SE, several studies (Kellinghaus et al., 2009; Höfler et al., 2011) or case reports (Albers et al., 2011) have shown that intravenous LCM seems to be a successful third-line treatment. A systemic study to evaluate the treatment for SE is difficult because it will always remain uncertain whether the new agent, the previously applied first- and second-line AEDs, or a spontaneous remission are responsible for the termination of the SE. In addition, in comparison with self-limited seizures, refractory SE is an epileptic entity, in which additional pathophysiologic factors may complicate therapeutic interventions. An investigation of an immediate efficacy of AEDs in the setting of presurgical video-EEG monitoring, by contrast, provides more accurate results, which can serve as proof-of-principle data for acute efficacy of antiseizure agents.
In a double-blind trial with >200 patients the responder after 3 months (38.3%) of 400 mg orally administered LCM (Chung et al., 2010) was comparable with the results of our study (42.8%), supporting an overall clinical utility of de novo intravenous LCM application beyond acute efficacy and tolerability. AEs seemed to be more frequent in patients with a combination of LCM and classical sodium channel blockers (in our cases predominantly LTG) as shown in previous studies (Hillenbrand et al., 2011). Eleven (64.7%) of the 17 patients had sodium blockers as part of their comedication. It could be hypothesized that the midterm efficacy could be higher, if the proportion of patients with sodium-blocker comedication were lower. However, the small number of patients prevents a more detailed analysis of the role of concomitant AEDs.
Certainly, the limited number of patients constitutes a caveat in this study. In addition, we cannot provide plausible data, which reveals whether or by which pharmacologic mechanism rapid loading of LCM is enhanced by intravenous as opposed to oral administration. Because the bioavailability of oral and intravenous administration is virtually identical (Kropeit et al., 2005), missing serum levels of LCM in our study seem to be negligible. Only a double-blind comparative study of rapid oral versus intravenous LCM initiation will properly address this issue.
In summary, our data suggest that de novo intravenous administration of LCM at a therapeutic dose is a feasible add-on therapy for patients in clinical situations with increased seizure activity, such as presurgical evaluation and putatively refractory SE. This holds true both during acute administration and for midterm follow-up. Further studies are needed to evaluate safety, tolerability and efficacy of the rapid loading of LCM.
The following authors have received travel grants or honoraria for advisory board participation and lecturing from various pharmaceutical companies: W. Li (Eisai GmbH, Pfizer Pharma, UCB Pharma), H. Stefan (Cerbomed Cyberonics, Desitin Arzneimittel, Eisai GmbH, Electa, Janssen-Cilag, GlaxoSmithKline, Pfizer Pharma, UCB Pharma), J. Matzen (Eisai GmbH, UCB Pharma), and F.C. Schmitt (Cyberonics, Desitin Arzneimittel, Eisai GmbH, Janssen-Cilag, GlaxoSmithKline, Medtronics, Pfizer Pharma, UCB Pharma). In addition, the following grants have been issued: H. Stefan (Deutsche Forschungsgemeinschaft [German Research Foundation], Erlanger Leistungsbezogene Anschubfinanzierung und Nachwuchsförderung [Perfomance-related Start-up Financing and Promotion of Young Academics by the Erlangen University], and Sander Foundation), S. Rampp (Deutsche Forschungsgemeinschaft [German Research Foundation] and American Clinical Neurophysiology Society), and H.J. Heinze (Deutsche Forschungsgemeinschaft [German Research Foundation], Bundesministerium für Bildung und Forschung [Federal Ministry of Education and Research, Germany], Helmholtz Society, Volkswagen Foundation). There is no conflict of interest with the content of this article. 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.