Lacosamide as a new treatment option in status epilepticus


Address correspondence to Eugen Trinka, Universitätsklinik für Neurologie, Christian-Doppler-Klinik, Paracelsus Medical University, Ignaz-Harrer-Str. 79, A-5020 Salzburg, Austria. E-mail:


Status epilepticus is among the most common neurologic emergencies, with a mortality rate of up to 20%. The most important therapeutic goal is fast, effective, and well-tolerated cessation of status epilepticus. Intravenous phenytoin/fosphenytoin, phenobarbital, or valproate is the current standard treatment after failure of benzodiazepines. Lacosamide as a new antiepileptic drug has been available as an intravenous solution since 2009. To date, PubMed lists 19 studies (10 single case reports and 9 case series), reporting a total of 136 episodes of refractory status epilepticus (50% nonconvulsive status epilepticus, 31% focal status epilepticus, and 19% convulsive status epilepticus) treated with lacosamide. The most often used bolus dose was 200–400 mg over 3–5 min. The overall success rate was 56% (76/136). Adverse events (AEs) were reported in 25% (34/136) of patients: mild sedation in 25 cases, 1 patient with possible angioedema, 2 with allergic skin reaction, 4 with hypotension, and 1 with pruritus. One patient developed a third-degree atrioventricular (AV) block and paroxysmal asystole. Overall, the rate of AEs was low. Current evidence on the use of intravenous lacosamide in acute seizures and status epilepticus is restricted to retrospective case reports and case series (class IV). Further prospective studies to inform clinicians are necessary.

Status epilepticus (SE) is a life-threatening condition that requires urgent and rigorous treatment with antiepileptic drugs (AEDs). More than 60% of generalized convulsive SE will be controlled by first-line treatment with intra-venous (IV) lorazepam (LZP) (Leppik et al., 1983; Treiman et al., 1998; Alldredge et al., 2001) or intramuscular midazolam (MDZ) (Silbergleit et al., 2012), irrespective of its cause. At least 40% of all patients will need further treatment with IV AEDs. Despite its known disadvantages (such as cardiac arrhythmias or hypotension), IV phenytoin (PHE) is frequently used to control the seizures, assuming that treatment benefits outweigh the well-known risks (Brophy et al., 2012).

Lacosamide (LCM) became available in Europe in September 2009 and in the United States in October 2009 as an IV solution based on bioequivalence to the oral formulation (Biton et al., 2008; Krauss et al., 2010). LCM is a functionalized amino acid with anticonvulsant properties. Its mechanism of action is the selective enhancement of sodium channel slow inactivation (Beyreuther et al., 2007). LCM was effective in different rodent seizure models for generalized and complex partial seizures as well as for SE (Beyreuther et al., 2007; Stöhr et al., 2007).

Based on experimental evidence (Stöhr et al., 2007; Wasterlain et al., 2011), safety studies in healthy individuals and patients (Ben-Menachem et al., 2007; Biton et al., 2008; Krauss et al., 2010), as well as retrospective case series with SE and seizure clusters in humans (Kellinghaus et al., 2009; Tilz et al., 2010; Turpin-Fenoll et al., 2010; Albers et al., 2011; Chen et al., 2011; Goodwin et al., 2011; Granda-Mendez et al., 2011; Höfler et al., 2011; Kellinghaus et al., 2011; Koubeissi et al., 2011; Krause et al., 2011; LaRoche & Shivdat-Nanhoe, 2011; Parkerson et al., 2011; Rantsch et al., 2011; Shiloh-Malawsky et al., 2011; Torres-Cano et al., 2011; Cherry et al., 2012; Jain & Harvey, 2012; Mnatsakanyan et al., 2012), IV LCM may be a useful alternative in emergency treatment of seizures with IV AEDs.

The aim of the present review is to analyze all published studies on IV LCM in the treatment of SE and acute recurrent seizures.


Identification of studies and search strategy

All 19 studies were identified by electronic search of PubMed (January 2009–May 2012) using the search terms: “lacosamide” AND “intravenous” AND “status epilepticus.” Only studies published as a full article were eligible if they reported the use of IV LCM in the treatment of SE.

Mechanism of action of LCM

LCM is a functionalized amino acid with anticonvulsive properties (Andurkar et al., 1999; Hovinga, 2003; Duncan & Kohn, 2005). LCM selectively enhances slow inactivation of sodium channels, with no effect on fast inactivation (Errington et al., 2006). This mechanism is in contrast to other AEDs such as carbamazepine (CBZ) or PHE, which do not enhance slow inactivation. LCM shifts the voltage dependence of inactivation to more hyperpolarized membrane potentials and thereby promotes the transition of sodium channels into the slow inactivated state (Beyreuther et al., 2007).

Pharmacokinetics of LCM

Oral LCM is rapidly absorbed, and plasma concentrations are dose proportional. With an elimination half-life of 13 h, a twice-daily dosing is possible. Steady state is reached within 2–3 days (Horstmann et al., 2002; Hovinga, 2003; Cawello et al., 2004). LCM has a low protein binding (<15% bound) and no effect on the plasma concentrations of concomitantly administered AEDs (Kropeit et al., 2004; Jatuzis et al., 2005). A small proportion of LCM is metabolized to an O-desmethyl metabolite that is not pharmacologically active in in vivo models (Doty et al., 2007; Cross & Curran, 2009). LCM and its major metabolites are eliminated renally.

Because of its high water solubility, LCM was also developed as an IV solution, in parallel with the oral formulations. The isotonic solution contains 10 mg/ml and has a pH of 3.5–5. The solution is stable at room temperature and can be administered without dilution (Biton et al., 2008). However, it can also be diluted in sodium chloride 0.9%, dextrose 5%, or lactated Ringer's solution, if needed. In a phase 1 trial with 24 healthy volunteers, bioequivalence of short-time infusion was demonstrated as a replacement therapy for oral LCM at doses of 200 mg twice daily and IV infusions at 30 and 60 min (Kropeit et al., 2004).

Krauss et al. (2010) reported a multicenter, open-label, inpatient, dose-escalation trial with 160 patients from ongoing open-label, long-term trials who were taking stable doses of oral LCM and up to three concomitant AEDs. The patients received IV LCM dosed over progressively shorter infusion durations: 30, 15, and 10 min for 2–5 days. Most patients received IV LCM 200–600 mg/day; around 4% (7/160) received 700–800 mg/day in 15 min.

The adverse event (AE) rate was low, the most common (10% or less) being headache, dizziness, diplopia, and somnolence. The occurrence of AEs did not correlate with shorter infusion time, but with higher doses of LCM. Two patients had cardiac AEs: one patient treated with beta-blockers had a reversible bradycardia of 26 bpm during a 15-min infusion on day 2 of treatment with 300 mg/day IV LCM; the second patient had a nonrecurring prolonged QTcB interval, without clinical symptoms or signs, on day 4 of treatment with 100 mg/day IV LCM by 15-min infusion. It should be noted that all patients were long-term responders to LCM in the open-label extension trials, which enriches the study population with those patients who tolerate LCM well (Krauss et al., 2010).

Another multicenter double-blind, double-dummy, randomized, inpatient trial supported the safety and utility of IV LCM as replacement for oral LCM. Intravenous LCM was administered as 60- or 30-min twice-daily infusions and showed a safety and tolerability profile similar to that of oral LCM. Treatment-emergent AEs were reported by 16 of 60 patients—including dizziness, headache, back pain, somnolence, and injection-site pain. All AEs were considered mild or moderate in intensity, and no AE led to discontinuation (Biton et al., 2008).

In a safety and tolerability trial with 25 subjects, IV LCM was applied in three progressively increasing doses (200, 300, and 400 mg). Optimal tolerance was found with IV loading doses of 200 and 300 mg administered over 15 min; a higher frequency of dose-related AEs (dizziness, somnolence, nausea, and diplopia) was observed with a loading dose of 400 mg (Fountain et al., 2011).

Experimental basis for the use of LCM in SE

In an experimental model of limbic self-sustaining SE induced by perforant path stimulation, rats were treated with IV LCM (3, 10, 30, or 50 mg/kg) either after 10 min (early) or after 40 min (late). Compared with the late treatment, LCM given after 10 min showed a significant dose-dependent reduction of acute SE seizure activity (Wasterlain et al., 2011). In other models, such as maximal electroshock test in mice, LCM protected mice and rats against tonic-extension seizures and prevented seizure spread (Swinyard et al., 1952; Borowicz et al., 1997). In the 6-Hz psychomotor seizure test, LCM demonstrated full efficacy in contrast to other AEDs such as LTG, PHE, and CBZ (Barton et al., 2001). However, LCM was not effective in chemoconvulsant-induced seizures induced by the subcutaneous bolus injection of pentylenetetrazole (Stöhr et al., 2007).

Clinical studies in SE

There are 19 published articles on the use of IV LCM in SE and acute repetitive seizures, including 10 single case reports (Kellinghaus et al., 2009; Tilz et al., 2010; Turpin-Fenoll et al., 2010; Chen et al., 2011; Granda-Mendez et al., 2011; Krause et al., 2011; LaRoche & Shivdat-Nanhoe, 2011; Parkerson et al., 2011; Shiloh-Malawsky et al., 2011; Torres-Cano et al., 2011; see Table 1) and 9 retrospective case series (Albers et al., 2011; Goodwin et al., 2011; Höfler et al., 2011; Kellinghaus et al., 2011; Koubeissi et al., 2011; Rantsch et al., 2011; Cherry et al., 2012; Jain & Harvey, 2012; Mnatsakanyan et al., 2012; see Table 2).

Table 1. Overview of all case reports of SE treated with IV LCM
StudySE (n)Age (y)Why? Indication?When? Conditions?How?OutcomeAdverse events
SE typeSE etiologySE specific etiologyAEDs before LCM (mg)BDZOrder of LCMTime from SE onset to LCM initiationLCM doseLCM route
  1. AED, antiepileptic drug; BDZ, benzodiazepines; CLN, clonazepam; CSE, convulsive status epilepticus; DZP, diazepam; EPC, epilepsia partialis continua; FBM, felbamate; FOS, fosphenytoin; GBM, glioblastoma multiforme; IV, intravenous; IVIG, intravenous immunoglobulin; LCM, lacosamide; LEV, levetiracetam; LTG, lamotrigine; LZP, lorazepam; MDZ, midazolam; NCSE, nonconvulsive status epilepticus; OXC, oxcarbazepine; PB, phenobarbital; PEG tube, Percutaneous gastrostomy tube; PHE, phenytoin; PO, orally; RSE, resistant status epilepticus; SE, status epilepticus; SESA, subacute encephalopathy and seizures in alcoholics; TPM, topiramate; VPA, valproic acid.

Kellinghaus et al. (2009)142NCSERemote symptomaticStrokeIV DZP (2.5); IV LZP (6)Y3rd≥1.5 h200 mg (within 3–5 min)IVResponder after 5 minPossible skin itching and mild erythema
Tilz et al. (2010)138CSERemote symptomaticPerinatal hypoxia with resulting infantile cerebral palsyIV DZP (22.5); IV etomidate (12.5); IV MDZ (5); enteral LZP (4); enteral LEV (1,500)Y6th (with etomidate)Not reported300 mgPEG-tube (Syrup)Responder after 30 minN
Turpin-Fenoll et al. (2010)172NCSERemote symptomaticStrokeLEV (3,000), VPA, CLN, OXC (dose and route not reported)Y5thNot reported25 mg twice dailyPOResponder 1 week after LCM was startedN
Parkerson et al. (2011)1775Focal SERemote symptomaticGBMNot reportedYNot reportedNot reported100 mg twice dailyIVResponderN
Chen et al. (2011)157Focal SE/EPCRemote symptomaticTraumatic brain injuryLZP (2); FOS (1,000); LEV (3,000); LZP (2); VPA (loading 20 mg/kg, maintain 10 mg/kg every 6 h); LEV daily dose increased to 4,000 mg, PHE administration within a serum level of 20–30 μg/mlY4th4 days50 mg twice per day for 5 days, then uptitration to 400 mg/day within 8 daysPOResponderN
Krause et al. (2011)189NCSEAcute symptomaticNot reportedIV LZP (2); Day 1:IV LEV (2,000) Day 2: IV LEV (3,000)Y3rdNot reportedTwo 400-mg bolus doses within 6 hIVResponderThird-degree AV block and asystolia
Granda-Mendez et al. (2011)181NCSEAcute symptomaticStrokeIV CLN (2); IV VPA (40 mg/kg) and VPA 800 mg/day over continuous infusionY3rdNot reported100 mg/dayIVResponderN
LaRoche & Shivdat-Nanhoe (2011)161NCSEAcute symptomaticSESALZP; FOS; VPA; doses not reportedY4thNot reportedNot reportedIVResponderN
Torres-Cano et al. (2011)149NCSERemote symptomaticPolymicrogyria and nodular heterotopiaDZP (5); VPA (500 + 500 mg/12 h); LEV (1,000); CLN (1 g/12 h); LTG (200 mg/2 h); propofol and MDZ per perfusor; PHEY7th15 days200 mg/12 hIVResponderN
Shiloh-Malawsky et al. (2011)18CSE→NCSEAcute symptomaticUnclearLZP; PHE; PB; VPA; LEV; FBM; TPM; propofol; corticosteroids; IVIG; plasmapheresisY8th10 weeks25 mg twice dailyEnteralResponder; 5 days after LCM was started, SE was resolvedN
Table 2. Overview of all case series of SE treated with IV LCM
StudyEpisodes of SE (n)Age (y)Why? Indication?When? Conditions?HOW?Outcome,% (n)Adverse events (n)
Episodes of SE by type (n)SE etiology,% (n) ptsSE specific etiology, n/NAEDs before LCM, % (n) ptsBDZ, % (n) ptsOrder of LCMTime from SE onset to LCM initiationLCM dose (mg)LCM route
  1. AED, antiepileptic drug; AVM, arteriovenous malformation; BDZ, benzodiazepines; CLN, clonazepam; CSE, convulsive status epilepticus; EPC, epilepsia partialis continua; FOS, fosphenytoin; GBM, glioblastoma multiforme; HSV, herpes simplex virus; ICH, intracerebral hemorrhage; LCM, lacosamide; LEV, levetiracetam; LZP, lorazepam; MDZ, midazolam; NCSE, nonconvulsive status epilepticus; NK, not known; PB, phenobarbital; PGB, pregabalin; PHE, phenytoin; SDH, subdural hematoma; SE, status epilepticus; TPM, topiramate; VPA, valproic acid; VPS, ventriculoperitoneal shunt.

Kellinghaus et al. (2011)39Median 62 (range, 18–90)



17 focal SE

26% (10/39) Acute symptomatic; 23% (9/39) Remote symptomatic (new-onset epilepsy); 49% (19/39) Remote symptomatic (preexisting epilepsy); 3% (1/39) Others/NKNot reported95% (37/39) BDZ; 85% (33/39) LEV; 36% (14/39) PHE 13% (5/39) other AEDs; 10% (4/39) Anesthesia95% (37/39)13% (5/39) pts LCM as 1st/2nd; 49% (19/39) pts LCM as 3rd; 38% (15/39) pts LCM as 4th/laterMedian latency 30 (range, 0.5–1,440) hMedian bolus dose 400 (range, 200–400; mean dose 1st day 424 (range, 200–600)IV100% (5/5) LCM as 1st/2nd; 95% (18/19) LCM as 3rd; 73%(11/15) LCM as 4th/later1 Allergic skin reaction; 25 Sedation; 4 Hypotension
Koubeissi et al. (2011)4Median 65 (range, 53–79)4/4 NCSE75% (3/4) Acute symptomatic; 25% (1/4) Preexisting epilepsy

2/4 Hemorrhage

1/4 Meningioma

1/4 Temporal encephalomalacia

100% (4/4) LZP; 50% (2/4) FOS; 100% (4/4) LEV; 25% (1/4) PGB; 25% (1/4) VPA100% (4/4)Median4 (range, 3–5)Median 34 (range, 3–50) hMedian initial dose 100 (range, 50–100)IV100% (4/4)N
Goodwin et al. (2011)9Median 63 (range, 47–89)

6/9 NCSE

3/9 CSE

4 Preexisting epilepsy

5 Acute symptomatic

2/9 rapid neurodegenerative disease; 1/9 Rosai–Dorfman disease; 1/9 GBM and stroke; 1/9 Aneurysms; 1/9 Stroke; 1/9 Meningioma; 1/9 VPS infection; 1/9 No underlying neurologic disease100% (9/9) LEV, 78% (7/9) PHE, 22% (2/9) Pentobarbital coma100% (9/9)Median 3 (range, 2–5)Median 2 (range, 0–14) daysMedian loading dose 200 (range, 100–300)IV100% (9/9) Nonresponder2 Episodes of angioedema;
Albers et al. (2011)7Median 63 (range, 33–83)7/7 focal SE42% (3/7) Acute symptomatic; 29% (2/7) Remote symptomatic; 29% (2/7) Pre-existing epilepsy3/7 SDH; 1/7 History of stroke; 1/7 Alcohol abuse and hypertension; 1/7 ICH; 1/7 Infantile brain damage29% (2/7) VPA; 100% (7/7) LEV; 4/7 propofol; 14% (1/7) PHE; 42% (3/7) TPM; 42% (3/7) MDZ42% (3/7)Median 4 (range, 2–5)Not reportedMedian initial dose 400 (range, 400)IV100% (7/7) ResponderN
Höfler et al. (2011)31Median 67 (range, 22–95)11/31 CSE; 10/31 NCSE; 10/31 focal SE19% (6/31) Acute symptomatic; 35% (11/31) Remote symptomatic; 46% (14/31) Preexisting epilepsy1/31 NK; 5/31 Traumatic lesions; 1/31 Cortical dysplasia; 9/31 Vascular lesions; 3/31 Tumor; 1/31 Encephalitis, 11/31 Others68% (21/31) LEV; 94% (29/31) BDZ; 10% (3/31) PHE; 16% (5/31) VPA; 3% (1/31) Anesthesia94% (29/31)7% (2/31) pts LCM 1st; 19% (6/31) pts LCM 2nd; 48% (15/31) pts LCM 3rd; 26% (8/31) pts LCM 4th or laterNot listedMedian 200 (range, 200–400)IV

Overall 81% (25/31) Responders

LCM 1st 2/2, LCM 2nd 6/6, LCM 3rd 11/15, LCM 4th or later 6/8

In group SE no AEs
Cherry et al. (2012)13Median 51.1 (range, 24–80)7/13 NCSE; 4/13 focal SE; 2/13 CSE9/13 Preexisting epilepsyNot reported92% (12/13) LZP; 85%(11/13) LEV; 69% (9/13) FOS; 31% (4/13) VPA; 31% (4/13) PB92% (12/13)Mean 3.2 (range, 1–7)Mean 39.5 (range 6.5–74) hMean 180.8 (range, 100–400); Infusion rate, mean 3.33 (range, 1.67–6.67) mg/minIV38% (5/13) Complete seizure cessation; 54% (7/13) ≥50% Reduction in seizure frequency3/13 Episodes of significant hypotension; 1/13 Fever without source; 1/13 Elevated liver function tests
Jain & Harvey (2012)3Median 16 (range, 12–17)3/3 refractory tonic SE3/3 Pre-existing epilepsy1/3 Lissencephaly; 1/3 Rett syndrome; 1/3 Generalized epilepsy3/3 LEV; 2/3 VPA; 3/3 CLZ; 1/3 PB100% (3/3)Median 6 (range, 5–7)Median 28 (range, 8–29) hMedian 100 (range, 50–200)IV100% (3/3)1/3 Chorea; 1/3 Oculogyric crisis
Mnatsakanyan et al. (2012)10Median 60.5 (range, 16–90)10/10 NCSE60% (6/10) Acute symptomatic; 40% (4/10) Preexisting epilepsy1/10 NK; 2/10 Asystole; 2/10 Tumor; 1/10 HSV encephalitis; 1/10 Frontal AVM resection; 1/10 Metastases; 1/10 Stroke; 1/10 Posterior reversible encephalopathy syndrome40% (4/10) BDZ; 80% (8/8) PHE; 90% (9/10) LEV; 30% (3/10) VPA; 10% (1/10) Propofol40% (4/10)Median 4 (range, 2–8)Not reportedMedian loading dose 200–300 within 30 minIV70% (7/10) ResponderN

Single case reports

Kellinghaus et al. (2009) published a case of a 42-year-old woman who had preexisting epilepsy after a severe cardioembolic stroke in the territory of the left middle cerebral artery in 1993. After SE in 2006 with vomiting, worsening of her preexisting aphasia, hemiparesis, and nystagmus, she received levetiracetam (LEV) 2,000 mg/day. Under this therapy, she developed psychotic symptoms; therapy was therefore changed, first to lamotrigine (LTG) and then to gabapentin (GBP). Due to noncompliance with AED dosing, four further episodes of SE followed. In October 2008, the patient had her sixth SE with her habitual clinical semiology. Previous medication was LTG 200 mg and GBP 1,200 mg/day. In the emergency situation, she received IV diazepam (DZP) 2.5 mg. Because of rhythmic slowing and spiking on the left posterior temporal area, IV LZP 6 mg was administered, but without success, and the patient developed respiratory problems. Because of the enzyme-inducing effect of valproic acid (VPA) and previous treatment with LTG, VPA was not used; LEV was also avoided, because of her previous psychiatric episode. Instead, IV LCM was administered as a bolus of 200 mg within 3–5 min; epileptic activity on electroencephalography (EEG) ceased about 5 min later. After 6 days, the patient developed mild pruritus.

Tilz et al. (2010) described a 38-year-old man who had preexisting epilepsy with infantile cerebral palsy. Diagnosis on admission was a series of complex partial seizures, which evolved into generalized tonic–clonic SE. PHE and VPA had been ineffective in the past. Initially, the patient received IV DZP 22.5 mg, IV etomidate 12.5 mg, and IV MDZ 5 mg within 45 min. Because of ongoing SE, LZP 4 mg and LEV 1,500 mg were administered via percutaneous gastric fistula within 30 min. Because of the persistent SE, treatment with a 150-mg LCM minced tablet powder was given via percutaneous gastrostomy tube, following which the seizures were interrupted; a second dose of 150 mg LCM syrup stopped SE after 30 min. Lower doses of AEDs were used because of the patient's low body weight.

Turpin-Fenoll et al. (2010) presented a 72-year-old patient with a nonconvulsive SE (NCSE) 2 weeks after his first tonic–clonic seizure. As PHE had previously caused bradycardia and respiratory failure, it was avoided as first-line therapy. The patient received IV LCM after failed therapy with LEV (3,000 mg/day), oxcarbazepine, VPA, and clobazam. One week after NCSE had started, the seizures were stopped with a dose of LCM 50 mg/day. Because of the low dose of LCM, the authors discussed a remote effect of previously given drugs.

Parkerson et al. (2011) reported on 17 patients who received LCM (13 intravenously) who were undergoing prolonged video-EEG monitoring. Three patients had no seizures, but periodic lateralized epileptiform discharges plus (PLEDs-plus) or generalized periodic discharges. Nine patients had repetitive focal seizures and EEG changes (like PLEDs or PLEDs-plus). One 75-year-old male patient had focal SE resulting from a glioblastoma multiforme and was taking LEV and PHE at admission. He received a benzodiazepine as first-line therapy, then IV LCM 200 mg/day, and the focal SE was stopped. There were no AEs reported; in particular, no change of PR intervals prior to LCM or at the end of infusion.

Chen et al. (2011) reported a patient with epilepsia partialis continua: a 57-year-old man with right-sided hemiparesis with cognitive and language impairment after a brain injury at age 25, who was admitted because of involuntary jerking of the right arm and leg. EEG showed a continuous left posterior slowing and epileptiform rhythms, with hyperintense areas in diffusion-weighted imaging and fluid-attenuated inversion recovery (FLAIR) sequences observed on magnetic resonance imaging (MRI). The patient received IV LZP 2 mg, IV fosphenytoin (FOS) 1,000 mg, and IV LEV 3,000 mg, but clinical status and EEG remained unchanged. Furthermore, IV LZP 2 mg, IV VPA (loading dose 20 mg/kg), increase of IV LEV daily dose from 3,000 to 4,000 mg, and PHE administration within a serum level of 20–30 μg/ml achieved no seizure control. After 4 days of refractory SE, an oral dose of LCM 100 mg/day was started and all other AEDs except LEV were tapered down. On day 4, the SE was stopped clinically and electrographically. This low dose of LCM was claimed to be effective, but of course SE may have resolved unrelated to LCM.

Krause et al. (2011) described an 89-year-old woman who was admitted to the hospital with an NCSE. She had a history of arterial hypertension, heart insufficiency, and hypothyroidism, and had been treated with a beta-blocker and amlodipine. Laboratory tests showed a potassium level of 2.5 mEq/L, so supplementation was started. For SE, the patient initially received IV LZP 2 mg; SE continued, and she developed respiratory problems. On day 1, IV LEV 2,000 mg was administered as a bolus within 60 min, followed by 3,000 mg on day 2. Because of persistent SE, IV LCM was administered in two 400-mg bolus doses within 6 h, following which the patient developed third-degree AV block and asystole three times, with duration of ≤10 s. After infusion 30 min later with 500 ml hydroxyethyl starch, sinus rhythm with first-degree AV block returned; the PQ interval normalized over the course of the next day. Possible reasons for the occurrence of AV block included the low potassium level, negative effects of premedication on AV conduction, and the low glomerular filtration rate. The authors concluded that high doses of LCM should be avoided in patients at risk for AV conduction delay.

One 61-year-old patient with NCSE during the course of subacute encephalopathy with seizures was fully controlled by IV LCM after failure of LZP, FOS, and VPA. Dose and duration of SE before LCM were not reported. No AEs were noted (LaRoche & Shivdat-Nanhoe, 2011).

Shiloh-Malawsky et al. (2011) reported an 8-year-old boy with convulsive seizures 2 days after a febrile illness. Because of frequent focal seizures, he received PHE, phenobarbital (PB), VPA, and LEV, but without success; thus he had up to 300 focal-onset electrographic seizures per day under a combination of six AEDs (PHE, LEV, PB, topiramate [TPM], felbamate [FBM], and VPA). He also received propofol, pentobarbital, MDZ, and ketamine, until a burst-suppression EEG pattern, for up to 1 week and a ketogenic diet for 6 weeks. Doses for AEDs and anesthetics were not reported. Because of a possible autoimmune etiology, the patient was treated with high-dose corticosteroids, IV immunoglobulin, and plasmapheresis, but again without effect. Ten weeks after refractory SE (RSE), LCM 25 mg was administered enterally twice daily; seizure frequency decreased significantly within 3 days, with seizures ceasing altogether after 5 days. It is possible that the response after 3 days of therapy with low dose LCM may have been nothing to do with the LCM.

Granda-Mendez et al. (2011) presented the case of an 81-year-old woman with an acute symptomatic SE after an ischemic infarction of the left/right posterior cerebral artery. Initially she received IV clonazepam (CLN) 2 mg, followed by IV VPA at a dose of 40 mg/kg, and then 800 mg/day continuous infusion. Because of comorbidities (chronic renal failure), PHE and LEV were avoided and the patient received IV LCM 100 mg/day. No AEs were reported.

Torres-Cano et al. (2011) reported a 49-year-old woman with an NCSE. She had symptomatic epilepsy due to polymicrogyria and nodular heterotopia. The premedication was VPA 500 mg twice daily, CLN 1 mg twice daily, and LEV 500 mg twice daily. In the emergency department she received IV DZP 5 mg, IV VPA 500 mg, and IV LEV 500 mg before being admitted to the intensive care unit, where she received another IV LEV 500 mg over 12 h, IV CLN (1 g/12 h), and IV VPA (500 mg/12 h). Because of generalized convulsions, the patient received propofol and MDZ by perfusor. After 10 days, the SE persisted; treatment was IV CLN 1 g/12 h, IV LTG 200 mg/12 h, and PHE. On day 15, IV LCM 200 mg/12 h was added, and SE resolved at day 17.

Retrospective case series

Kellinghaus et al. (2011) reported the pooled experience of four centers in Germany, Austria, and Switzerland. Altogether, 39 patients receiving IV LEV for the treatment of SE (six convulsive SE [CSE], 17 NCSE, and 16 focal SE) refractory to benzodiazepines were included. Acute symptomatic etiology was found in 10 patients (26%), remote symptomatic in 9 (23%), and preexisting epilepsy in 19 (49%). Specific etiologies were not reported for individual patients. LCM was successfully used as first or second drug in 3 of 5 patients, as third drug in 11 of 19 patients, and as fourth or later drug in 3 of 15 patients. Median bolus dose was 400 mg (range 200–400). In 44% of patients (17/39), SE was controlled by LCM; only mild AEs were reported (1 allergic skin reaction, 25 sedation, 4 hypotension). Of the 39 patients, 37 received benzodiazepines as first-line treatment. When LCM was administered early (as first, second, or third drug) it was successful in 60%, whereas only 20% were treated successfully in later stages of SE.

Rantsch et al. (2011) reported on nine patients with 10 episodes of SE (eight NCSE and two focal SE). The etiology was remote in six patients, and acute symptomatic in three. Median administration of LCM was as the sixth drug (range 4th–12th), with an initial dose of 50–100 mg. In 2 (20%) of 10 episodes of SE, LCM was the last drug. AEs were not reported. Three patients died during the observation period due to complications of their underlying disease. In this study, the investigators found a highly significant (r = 0.94, p < 0.00005) correlation between age and outcome: patients aged <60 years returned to baseline, whereas those aged >75 years died. However, this is an uncontrolled small series and these data must be interpreted with caution.

Goodwin et al. (2011) described nine patients with RSE, which was defined as seizure lasting >30 min, or multiple seizures without return of consciousness, and a failure of at least two medications to abort SE. Six of their patients had NCSE; three had CSE. Four had a preexisting epilepsy and five were acute symptomatic. Patients had three (range 2–5) AEDs prior to LCM, and all received benzodiazepines as first-line treatment. Response to LCM was defined for the patients in burst suppression as an absence of electrographic seizure activity for 24 h; for patients not in burst suppression, response was defined as electrographic seizure resolution ≤4 h after LCM administration. None of their patients had their SE controlled with IV LCM, and only one patient had an improvement of his EEG, without clinical improvement. Two patients had a decrease of seizure frequency with PHE and LCM. Two patients died during the observation period due to their underlying disease. The median initial dose was 200 mg (range 100–300). Two patients had angioedema during LCM treatment.

Koubeissi et al. (2011) reported on four patients with NCSE. Etiology was acute symptomatic in three patients; one had a preexisting epilepsy. One patient had a temporal encephalomalacia, one had a meningioma, and two had an intracerebral hemorrhage; they all received benzodiazepines as first-line drugs. LCM was on median the fourth AED (range 3rd–5th) and the initial dose was in the range of 50–100 mg. All patients were responders and no AEs were observed. Positive response or “seizure freedom” was defined as a reduction of seizure frequency from two to three seizures per hour to one seizure every 2 h.

Another retrospective study (Albers et al., 2011) described seven patients treated with IV LCM for focal SE. Three had acute symptomatic, two remote symptomatic, and two a preexisting epilepsy. Three patients had intracerebral subdural hematomas; one had a history of ischemic stroke, one patient a history of infantile brain damage, and one patient a history of intracerebral bleeding. One patient had alcohol abuse and hypertension. LCM was used on median as fourth drug (range 2nd–5th). Initial loading was 400 mg/day. Response to LCM was defined as resolution of EEG status within 24 h. In all patients, SE was resolved and no serious AEs were reported. Four of the seven patients had no initial treatment with benzodiazepines, whereas one received this only during the course of treatment.

Cherry et al. (2012) reported on 24 patients: 10 with 13 episodes of RSE (defined as SE with failed response to ≥2 drugs), and 14 with isolated seizures. Specific etiologies for SE were not reported in detail. Among 13 episodes of RSE, 7 were nonconvulsive, 4 focal, and 2 convulsive. The median use of LCM was as third drug (range 1st–7th). Concurrent AEDs were LZP in 92% (12/13), LEV in 85% (11/13), FOS in 69% (9/13), VPA in 31% (4/13), PB in 22% (6/13), “other” in 19% (5/13), and sedatives or anesthetics in 33% (9/13). Complete seizure cessation was achieved in 5 of 13 episodes (38%) of SE, and seizure reduction of >50% in 7 of 13 episodes (54%). The median loading dose was 180.8 mg (range 100–400 mg) and mean maintenance dose 361.5 mg/day. Three episodes of significant hypotension, and one fever with unknown origin, were observed. One patient had elevated liver function test leading to discontinuation of LCM.

Our own group (Höfler et al., 2011) reported on 48 patients: 17 (35%) with seizure clusters (SCs) and 31 (65%) with SE. The etiology was acute symptomatic in six (19%) and remote symptomatic in 11 (36%) patients, whereas 14 patients (45%) had a preexisting epilepsy. Altogether, 32% had NCSE, 36% had a convulsive, and 32% had a focal SE. Median initial bolus dose was 200 mg (range 200–400 mg). LCM was used as first drug in two patients (7%); in both, SE was controlled. In six patients (19%), LCM was used as second drug; again, all of these patients responded. In 15 patients (48%), LCM was administered as the third drug, and in 11 (73%) of these SE was stopped. Eight patients (26%) received LCM as fourth or later drug, and six of these (75%) were responders. Overall, we observed cessation of SE in 81% (25/31). No serious AEs were identified.

Mnatsakanyan et al. (2012) identified 10 patients with NCSE, of whom 6 had an acute symptomatic SE. Specific etiologies were found in 9 of 10 patients: glioma resection, herpes simplex virus encephalitis, frontal arteriovenous malformation, multifocal glioma, brain metastases, stroke, posterior reversible encephalopathy syndrome, and anoxic brain injury due to asystole. The median use of LCM was as fifth drug (range 3rd–9th), including initial treatment with LZP. The median loading dose was 200–300 mg within 30 min, with a maintenance dose in the range of 100–200 mg every 12 h. SE resolved in seven patients; AEs were not reported. In two of three nonresponders, the specific etiology was anoxic brain injury.

Jain & Harvey (2012) described three patients (ages 12–17 years) with symptomatic generalized epilepsy with intel-lectual disability and refractory tonic SE. LCM bolus doses (2–2.5 mg/kg; 50–200 mg in the first 2 h) were used after three or more standard AEDs had been ineffective. In two patients, side effects were reported (chorea and oculogyric crisis). Duration of SE prior to LCM was 8–29 h. In two patients, there was an immediate clinical response to LCM; the other had resolution of SE after the second dose.

Route of Administration of LCM

Single case reports

The route of administration was oral in one patient (Chen et al., 2011), enteral in three (Tilz et al., 2010; Turpin-Fenoll et al., 2010; Shiloh-Malawsky et al., 2011), and IV in the remaining six patients (Granda-Mendez et al., 2011; Kellinghaus et al., 2011; Krause et al., 2011; LaRoche & Shivdat-Nanhoe, 2011; Parkerson et al., 2011; Torres-Cano et al., 2011).

Case series

We divided the case series into two subgroups, according to the number of included patients: Group I included all studies with a study population of 3–29 patients (Albers et al., 2011; Goodwin et al., 2011; Koubeissi et al., 2011; Rantsch et al., 2011; Cherry et al., 2012; Jain & Harvey, 2012; Mnatsakanyan et al., 2012), whereas Group II included studies with >30 patients (Höfler et al., 2011; Kellinghaus et al., 2011).

In both groups, all episodes (126/126) of SE were treated with IV LCM (Albers et al., 2011; Goodwin et al., 2011; Höfler et al., 2011; Kellinghaus et al., 2011; Koubeissi et al., 2011; Rantsch et al., 2011; Cherry et al., 2012; Jain & Harvey, 2012; Mnatsakanyan et al., 2012).

Type and Outcome of SE; Dose and Order of LCM

Single case reports

Seven of 10 reported SE were NCSE (Kellinghaus et al., 2009; Turpin-Fenoll et al., 2010; Granda-Mendez et al., 2011; Krause et al., 2011; LaRoche & Shivdat-Nanhoe, 2011; Shiloh-Malawsky et al., 2011; Torres-Cano et al., 2011), one a CSE (Tilz et al., 2010), and two a focal SE (Chen et al., 2011; Parkerson et al., 2011). All 10 SE reported by single case reports stopped after LCM treatment (Kellinghaus et al., 2009; Tilz et al., 2010; Turpin-Fenoll et al., 2010; Chen et al., 2011; Granda-Mendez et al., 2011; Krause et al., 2011; LaRoche & Shivdat-Nanhoe, 2011; Parkerson et al., 2011; Shiloh-Malawsky et al., 2011; Torres-Cano et al., 2011). In one of 10 case reports, the initial dose was not specified (LaRoche & Shivdat-Nanhoe, 2011); the median initial dose of the other nine case reports was 100 mg (range 50–400 mg). The lowest doses of 25 mg twice daily were applied by enteral route for a 72-year-old patient and an 8-year-old patient (Turpin-Fenoll et al., 2010; Shiloh-Malawsky et al., 2011).

Order of LCM in the treatment algorithm: median use of LCM was as fourth drug (range 2nd–8th) (Kellinghaus et al., 2009; Tilz et al., 2010; Turpin-Fenoll et al., 2010; Chen et al., 2011; Granda-Mendez et al., 2011; Krause et al., 2011; Shiloh-Malawsky et al., 2011; Torres-Cano et al., 2011). In 2 of 10 case reports this information was missing (LaRoche & Shivdat-Nanhoe, 2011; Parkerson et al., 2011). One 8-year-old patient received LCM as eighth drug, following treatment with LZP, PHE, PB, VPA, LEV, FBM, TPM, and other therapies (propofol, corticosteroids, IV immunoglobulin, and plasmapheresis). One patient received LCM as the second drug (Kellinghaus et al., 2009).

Case series

In group I, there were 63% (35/56) NCSE, 23% (13/56) focal SE, and 14% (9/56) CSE. Half (28/56) were responders: among these, 43% (12/28) had NCSE, 29% (8/28) focal SE, and 11% (3/28) CSE. Comparing the different types, there was more focal SE in the responder group (eight patients, vs. one nonresponder) but no notable differences for NCSE (12 vs. 16 responders/nonresponders) or CSE (three responders, three nonresponders). In one study of 13 patients, the type of SE was not specified (Cherry et al., 2012).

In group I the outcome in terms of seizure control varied enormously, between 0% and 100%: 0% (1/7 studies: nine patients with NCSE; Goodwin et al., 2011), 20% (1/7 studies: 2/10 patients; one NCSE, one focal SE; Rantsch et al., 2011), 38% (1/7 studies: 5/13 patients; types of SE of non-responders were not reported; Cherry et al., 2012), 70% (1/7 studies: 7/10 patients with NCSE; Mnatsakanyan et al., 2012), and up to 100% (3/7 studies: four patients with NCSE; Koubeissi et al., 2011; and seven patients with partial SE; Albers et al., 2011). In the studies with 100% responders, the loading dose ranged from 50 mg (Koubeissi et al., 2011; median 100 mg [range 50–100 mg]; four NCSE) to 400 mg (Albers et al., 2011: four had focal SE). In the remaining four studies, the initial dose of LCM varied from 50 mg (Rantsch et al., 2011: 50–100 mg; eight NCSE, two focal SE) to a maximum of 400 mg (Cherry et al., 2012; mean initial dose 180.8 [range 100–400]; Mnatsakanyan et al., 2012: 200 mg LCM/12 h). In the study with 100% nonresponders to LCM (six NCSE, three CSE), the median initial dose was 200 mg (range 100–300 mg) (Goodwin et al., 2011). Overall, the initial dose of LCM responders (Albers et al., 2011; Koubeissi et al., 2011; Jain & Harvey, 2012; Mnatsakanyan et al., 2012) versus nonresponders (Goodwin et al., 2011; Rantsch et al., 2011; Cherry et al., 2012) did not differ.

In group II, there were 39% (27/70) NCSE, 37% (26/70) focal SE, and 24% (17/70) CSE. The responder rate varied from 44% (17/39) (Kellinghaus et al., 2011) to 81% (25/31) (Höfler et al., 2011) of patients with SE. In the collaborative study of Germany, Austria, and Switzerland (Kellinghaus et al., 2011), the median initial dose of LCM was 400 mg (range 200–400 mg). In our group's study, the median initial dose of LCM was 200 mg (range 200–400 mg). In both studies, median use of LCM was as third drug (Höfler et al., 2011; Kellinghaus et al., 2011).

The loading dose of LCM was higher in group II (200–400 mg) than in group I (50–400 mg).

Titration Time

Given the retrospective nature of the studies and variations in reporting strategies, it is difficult to evaluate titration times: this information was supplied only in case reports, not in case series.

Single case reports

Chen et al. (2011) used a titration time of 8 days to reach a daily dose of 400 mg in a case of partial SE. In the case of the 8-year-old patient (Shiloh-Malawsky et al., 2011), LCM was administered at a dose of 25 mg twice daily, with resolution of SE after 5 days. In the case reported by Kellinghaus et al. (2009) IV LCM was administered as a 200-mg bolus within 3–5 min, followed by oral LCM 100 mg twice daily from day 2. In the study described by Torres-Cano et al. (2011), LCM was given at a dose of 200 mg/12 h. All patients reported by case reports were responders to IV LCM.

Duration of SE until LCM Therapy

Single case reports

Duration of SE until LCM application was detailed in 5 of 10 case reports (Kellinghaus et al., 2009; Chen et al., 2011; Krause et al., 2011; Shiloh-Malawsky et al., 2011; Torres-Cano et al., 2011). The longest duration of SE—10 weeks (CSE evolving in a NCSE) until LCM was used—was described by Shiloh-Malawsky et al. (2011). SE appeared to resolve with LCM treatment. The shortest duration of SE before LCM was 1.5 h (Kellinghaus et al., 2009). All reported patients were responders.

Case series

In group I, duration of SE until LCM therapy was reported in four of seven case series (Goodwin et al., 2011; Koubeissi et al., 2011; Cherry et al., 2012; Jain & Harvey, 2012). Goodwin et al. had a median time span of 48 h (range 0–336 h); Koubeissi et al., 3–50 h. Cherry et al. had a median duration of 39.5 h (range 6.5–74 h), whereas Jain & Harvey reported 8–29 h duration of SE prior to LCM.

In group II, duration of SE until LCM therapy was reported only in one case series (Kellinghaus et al., 2011). Patients had a median latency of 30 h (range 0.5–1,440 h) in SE before they received LCM. However, both studies in group II could demonstrate a clear order effect: if LCM was administered early, it was successful in 60% of cases, whereas the success rate was only 20% in more refractory patients (Höfler et al., 2011; Kellinghaus et al., 2011).

Definition of Resistant SE

All contemporary protocols use a staged approach for treatment of SE (Trinka, 2007; Kurthen et al., 2008; Shorvon et al., 2008; Meierkord et al., 2010). Treatment in stage 1 (“early SE”) includes benzodiazepines. After the failure of benzodiazepines, the next level will be stage 2 (“established SE”), the most commonly used therapies being IV AEDs such as PHE, PB, or VPA. If seizures continue despite this treatment, the SE is classed as stage 3 (“refractory SE”), at which point general anesthesia is recommended (Trinka, 2007; Shorvon et al., 2008).

Single case reports

All reported 10 patients with SE treated with LCM initially received benzodiazepines (Kellinghaus et al., 2009; Tilz et al., 2010; Turpin-Fenoll et al., 2010; Chen et al., 2011; Granda-Mendez et al., 2011; Krause et al., 2011; LaRoche & Shivdat-Nanhoe, 2011; Shiloh-Malawsky et al., 2011; Torres-Cano et al., 2011).

Case series

In 16% (13/126) of episodes of SE, benzodiazepines were not given before initiation of LCM therapy (Albers et al., 2011; Höfler et al., 2011; Kellinghaus et al., 2011; Rantsch et al., 2011; Cherry et al., 2012); in 20% (16/126) of episodes of SE, LCM was given immediately after benzodiazepines (Kellinghaus et al., 2009; Höfler et al., 2011). Because dose information was missing regarding AEDs received before LCM, a detailed analysis of prior treatment was not possible.


In summary, 136 episodes of SE treated with LCM were reported. The type of SE was an NCSE in 50% (69/136 episodes), focal in 31% (40/135), and convulsive in 19% (26/135). Fifty-one percent (70/136) had preexisting epilepsy. The overall success rate was 56% of SE (76/136). AEs were reported in 25% (34/136) of patients: sedation in 25 cases, 1 patient with possible angioedema, 2 with allergic skin reaction, 4 with hypotension, and 1 with pruritus. One patient developed a third-degree AV block and paroxysmal asystole (Krause et al., 2011). Overall, the rate of AEs was low.

None of the published studies had a prospective design, and all were uncontrolled. Study design, heterogeneity of population, and variations in the definition of SE may have contributed to bias, in addition to the publication bias. The varying—and sometimes missing—definition of response to IV LCM also weakens the conclusions. Although LCM is a promising drug in the treatment of acute seizures and SE, several issues remain to be clarified: What is the right dose? What is the maximum tolerated rate of infusion? How does LCM penetrate across the blood–brain barrier in patients with SE? Based on the current limited evidence LCM may have a potential as second-line alternative to standard anticonvulsants in treatment of established SE, in case of contraindications to phenytoin or valproic acid. The available evidence of efficacy is limited and randomized controlled data are needed.


The authors thank Alison Terry for editorial work.


J. Höfler has received speaker's honoraria from UCB, and travel grants from UCB, Eisai, and Gerot E. Trinka has acted as a paid consultant to Eisai, Biogen Idec, Medtronics, Bial, and UCB. He has received research funding from UCB, Biogen-Idec, Sanofi-Aventis, Red-Bull and speakers' honoraria from Bial, Cyberonics, Desitin Pharma, Eisai, Gerot, Böhringer-Ingelheim, Sanofi, Medis, and UCB. 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.