Atrioventricular block following lacosamide intoxication

Atrioventricular block following lacosamide intoxication

Lars U. Krause, Kai O. Brodowski, Christoph Kellinghaus

Dept. of Neurology, Klinikum Osnabrück, Osnabrück, Germany

correspondence to:

Dr. Christoph Kellinghaus

Dept. of Neurology

Klinikum Osnabrück

Am Finkenhügel 1

49076 Osnabrück

Germany

Tel.: +49-541-405-6501

Fax: +49-541-405-6599

e-mail: [email protected]

Abstract:

Lacosamide (LCM) is a novel anticonvulsant that modulates voltage-dependent sodium

channels. Although it is known to cause a slight, dose-dependent prolongation of the P-R-

interval in the ECG, a third-degree atrioventricular (AV) block has been described as adverse

event only in few patients participating in diabetic neuropathic pain studies, and in no

epilepsy patient. We report a 89-year old patient with decreased renal function and co-

medication of two other negative dromotropic agents who accidentally received two

intravenous boli of 400mg LCM within six hours. She had a normal P-Q-interval before and

after the first dose of LCM and developed a reversible complete AV-block approximately 30

minutes after the second bolus. We conclude that particular caution has to be exercised when

applying very high doses of LCM in patients with significant cardial and renal risk factors.

Key words: lacosamide - cardial side effect – status epilepticus - intravenous

Introduction:

Lacosamide (LCM) is a functional amino acid that has been developed as anticonvulsant as

well as pain modulating agent. LCM is approved by the U.S. and European licensing

authorities for add-on therapy in patients with partial and secondarily generalized seizures in

2008. Since LCM has high water solubility[1], an intravenous (i.v.) formulation has been

developed and approved early on and does not seem to have side effects different from the

oral formulation. I.v. LCM has been used successfully for treatment of status epilepticus[2].

One of the modes of action of LCM is enhancement of the slow inactivation of the voltage-

gated sodium channel, thus reducing the ability of neurons to sustain prolonged firing

bursts[1]. As in other anticonvulsant drugs influencing the VGSC, a mild, dose-dependent

prolongation of the atrioventricular (AV) conduction time has been observed, resulting in a

mean prolongation of the PR-interval of 4.2 to 12.3ms at a dose of 400mg/d, compared to

baseline[1]. However, this prolongation is within the range of the prolongation caused by

moderate doses of lamotrigine[3], pregabalin[4] or carbamazepine[5], and its clinical

relevance is not clear. Only in 4 of 941 patients of the epilepsy studies [6,7,8]did this PR-

change lead to drop out. In the studies for approval of LCM for treatment of painful

neuropathy, the rate of treatment-emergent first-grade atrioventricular blocks was similar in

the active groups as well as in the placebo groups[1]. After approval, there has been one

report of atrial fibrillation correlated with LCM administration[9], but it is unclear whether

prolongation of the AV conduction time did play a role in this case. We report a patient who

suffered from a reversible and complete block of the AV conduction after receiving a high

dose of intravenous LCM for treatment of status epilepticus.

Case report:

We report the case of a 89-year-old female patient suffering from arterial hypertension, heart

insufficiency, hypothyreosis and chronic back pain. Her medication consisted in metoprolol

95mg per day, amlodipine 5mg per day, molsidomin 4mg per day, torasemide 10mg per day,

levothyroxine 100µg per day, tramadol 200mg per day, metamizol 1000mg per day, ass

100mg per day and ramipril 7.5mg per day. In the early morning of the admission day, she

had been found somnolent and without moving in her nursing home. Because she had been

able to put on her clothes independently, sudden onset of the symptoms could be assumed.

After initial assessment at the local hospital she was transferred to our stroke unit. Here a

cranial CT scan followed by CT-arteriography could rule out the initial suspicion of basilar

artery occlusion. The subsequent EEG showed rhythmic sharp waves with left temporal

maximum and continous slowing in the left frontoparietal region. We diagnosed a non-

convulsive left-hemispheric status epilepticus and started anticonvulsant treatment with

lorazepam (LZP) 2mg i.v. which did not alter her clinical and electroencephalographic state

but resulted in decrease of her respiration rate. Thus, further application of benzodiazepines

did not seem suitable. We decided against therapy with phenytoin (PHT) because of the

patient’s cardiac disease. Levetiracetam (LEV) was used instead, with a bolus of 2000mg i.v.

within 60 minutes on day one followed by a maintenance dose of 3000mg per day from the

following day. The patient remained on the intermediate care unit with continuous monitoring

of ECG, blood pressure, oxygen saturation and respiration rate. On day one, no further

anticonvulsants were applied. The patient remained somnolent, her voluntary motor functions

were reduced, her rare efforts of vocalization remained unintelligible. A few times, some

short-lasting convulsions of the right facial musculature occurred. Laboratory tests at

admission revealed low serum potassium level (2.5 mval/l), so potassium was substituted by

infusion. On the following morning, serum potassium had increased to 3.0mval/l. ECG at

admission was found to be normal without any sign of atrioventricular conduction delay.

On the day following admission, EEG still showed a left-hemispheral status epilepticus, and

the state of the patient had not improved. We decided to add lacosamide (LCM) as additional

anticonvulsant. LCM was administered as bolus of 400mg i.v. within 5-10 minutes

approximately at noon. In the following hours, no abnormalities in ECG (see figure 1a), blood

pressure and respiration were documented. Due to an ambiguous chart note, the patient

received a second bolus of LCM (400mg i.v. within 5-10 minutes) approximately six hours

after the first bolus. Fifteen minutes after the second administration of LCM was finished, a

third degree AV block occurred. During a period of 15 minutes, the patient suffered from

asystolia three times, up to ten seconds in each case (see figure 1b). Each time, asystolia

resulted in a fall of systolic blood pressure to approximately 60mmHg. Intravascular volume

was increased by infusion of hydroxyethyl starch (500ml), but no further therapy was

necessary. Thirty minutes after the onset of the AV block, sinus rhythm of normal frequency

reappeared, but ECG showed ongoing AV block of first degree (PQ-interval almost 215ms)

(see figure 1c) until PQ-interval normalized on the following day (see figure 1d) After the

intermittent atrioventricular block, metoprolol was discontinued, and LCM was not

administered again. The remaining medication was continued. No further cardiac arrhythmias

occurred.

During the following days, a variety of additional anticonvulsive agents were applied

(valproate, clobazam, carbamazepine, topiramate), but despite these efforts, the patients

condition remained unchanged. EEG showed a persisting status epilepticus, predominant

continous polyspike-wave-complexes in the left temporal area and occasional generalized

sharp waves. On day 8, aspiration pneumonia occurred and was treated with moxifloxazin

400mg per day. At the sixteenth day of treatment, respiration became increasingly difficult.

Because of the age and condition of the patient, we did not induce artificial ventilation and

intensive care treatment but applied palliative care. The patient died on day 19 after

admission.

The cause of the status epilepticus remained uncertain until late in the course of the disease.

Bacterial or viral meningoencephalitis had been ruled out by examination of the cerebrospinal

fluid (CSF) immediately after admission (normal cell count with only mild increase of total

protein [724mg/l], negative PCR of herpes virus and varizella zoster virus). MRI showed a

small left parietal area of mild hyperintensity in the diffusion-weighted images, but no clear

correlation in the apparent diffusion coefficient (ADC) map. Fluid attenuated inversion

recovery (FLAIR) sequences showed a mild hyperintensity in the left mesial temporal

structures, but no abnormalities in the T1-weighted, contrast enhanced sequences. Finally,

there was evidence for elevated levels of NMDA-receptor-antibodies (assay by IFA in serum,

IgA 1:100, IgM 1:320, IgG negative), a constellation suspicious of autoimmune encephalitis.

In addition, the patient´s relatives had observed behavioral abnormalities and mild psychotic

symptoms during the previous weeks. However, due to the severe pneumonia and the switch

to a palliative treatment strategy, an immunosuppressive therapy had not been initiated.

Discussion:

Administration of a high dose of intravenous LCM was followed by a reversible block of AV

conduction in a multimorbid patient suffering from status epilepticus. Although a dose-

dependent mild prolongation of the PR-interval is already known from pre-approval studies,

we think that in this case several risk factors accumulated and finally resulted in this incident.

First, the patient was already receiving two drugs with negative effects on AV conduction,

namely metoprolol [10] and amlodipine [11]. Second, the patient had a significantly low

serum potassium level that also significantly increases atrioventricular conduction time by

hyperpolarization of the resting membrane potential of the nodal cells [12]. Third, the

glomerular filtration rate was estimated only 45ml/kg/d. Given the dose-dependency of the

negative dromotropic effect of LCM in the setting of significant risk factors for AV

conduction delay and LCM accumulation, the second bolus of LCM might have turned the

vulnerable equilibrium of AV conduction. The reversibility of the AV block after cessation of

LCM and the normalization of the PQ latency within 24 hours argue against the causation of

the event by other triggers. Thus, high doses of LCM should be avoided in patients at risk for

atrioventricular conduction delay. However, the first bolus of LCM did not result in AV

conduction delay even in this high risk patient. Therefore, AV conduction problems seem to

occur only in most extreme circumstances.

References:

1. Kellinghaus C. Lacosamide as treatment for partial epilepsy: mechanisms of action,

pharmacology, effects, and safety. Ther Clin Risk Manag 2009;5:757-766.

2. Kellinghaus C, Berning S, Immisch I, Larch J, Rosenow F, Rossetti AO, Tilz C,

Trinka E. Intravenous lacosamide for treatment of status epilepticus. Acta

Neurol Scand 2011;123:137-141.

3. Matsuo F, Bergen D, Faught E, Messenheimer JA, Dren AT, Rudd GDea. Placebo-

controlled studey of the efficacy and safety of lamotrigine in patients with

partial seizures. Neurology 1993;43:2284-2291.

4. Laville MA, de la Gastine B, Husson B, Le Boisselier R, Mosquet b, Coquerel A.

Should we care about pregabalin for elderly patients with a history of cardiac

dysrhythmia? Rev Med Interne 2008;29:152-154.

5. Kennebäck G, Bergfeldt L, Vallin H, Tomson T, Edhag O. Electrophysiologic effects

and clinical hazards of carbamazepine treatment for neurologic disorders in

patients with abnormalities of the cardiac conduction system. Am Heart J

1991;121:1421-1429.

6. Ben Menachem E, Biton V, Jatuzis D, Abou-Khalil B, Doty P, Rudd GD. Efficacy and

safety of oral lacosamide as adjunctive therapy in adults with partial-onset

seizures. Epilepsia 2007;48:1308-1317.

7. Chung S, Sperling M, Biton V, Krauss G, Hebert D, Rudd GD, Doty P, SP754 Study

Group. Lacosamide as adjunctive therapy for partial-onset seizures: a

randomized controlled trial. Epilepsia 2010;51:958-967.

8. Halasz P, Kalviainen R, Mazurkiewicz-Beldzinska M, Rosenow F, Doty P, Hebert D,

Sullivan T. Adjunctive lacosamide for partial-onset seizures: Efficacy and

safety results from a randomized controlled trial. Epilepsia 2009;50:443-453.

9. Degiorgio CM. Atrial flutter/atrial fibrillation associated with lacosamide for partial

seizures. Epilepsy Behav 2010;18:322-324.

10. Prystowsky EN. The effects of slow channel blockers and beta blockers on

atrioventricular nodal conduction. J Clin Pharmacol 1988;28:6-21.

11. Dunlap ED, Plowden JS, Lathrop DA, Millard RW. Hemodynamic and

electrophysiologic effects of amlodipine, a new calcium channel blocker. Am J

Cardiol 1989;64:71-77.

12. Ruiz-Ceretti E,Ponce Zumino A. Atrioventricular block in low potassium and K

dependence of the nodal resting potential. Can J Physiol Pharmacol

1986;64:531-538.

Legend:

Figure 1: Traces from ECG-Monitoring. Lead II (Einthoven)

a) ECG approximately 3 hours after infusion of the first bolus (400mg) of intravenous LCM.

Normal PQ-interval.

b) ECG approximately 30 minutes after infusion of the second bolus (400mg) of intravenous

LCM. Arrows point to P-Waves without conduction to ventricular action (third degree AV-

block)

c) ECG approximately 90 minutes after infusion of the second bolus of intravenous LCM.

Prolonged PQ-interval (first degree AV-block)

d) ECG on the day following LCM administration. Normal PQ-interval.