Defining a Place for Lacosamide in the Intensive Care Unit

Commentary

Randomized Trial of Lacosamide vs Fosphenytoin for Nonconvulsive Seizures.

Husain AM, Lee JW, Kolls BJ, Hirsch LJ, Halford JJ, Gupta PK, Minazad Y, Jones JM, LaRoche SM, Herman ST, Swisher CB, Sinha SR, Palade A, Dobrowski KE, Gallentine WB, Hahn CD, Gerard EE, Bhapkar M, Lokhnygina Y, Westover MB; for the Critical Care EEG Monitoring Research Consortium [published online ahead of print May 7, 2018]. Ann Neurol doi: 10.1002/ana.25249.

OBJECTIVE: The optimal treatment of nonconvulsive seizures in critically ill patients is uncertain. We evaluated the comparative effectiveness of antiseizure drugs (ASD) lacosamide (LCM) and fosphenytoin (fPHT) in this population. METHODS: The Treatment of Recurrent Electrographic Nonconvulsive Seizures (TRENdS) study was a non-inferiority, prospective, multicenter, randomized treatment trial of patients diagnosed with NCS by continuous electroencephalography (cEEG). Treatment was randomized to IV LCM 400 mg or IV fPHT 20 mg PE/kg. The primary endpoint was absence of electrographic seizures for 24 hours as determined by one blinded EEG reviewer. The frequency with which NCS control was achieved in each arm was compared, and the 90% confidence interval (CI) was determined. Non-inferiority of LCM to fPHT was to be concluded if the lower bound of the CI for relative risk was above 0.8. RESULTS: Seventy-four subjects were enrolled (37 LCM, 37 fPHT) between August 21, 2012 and December 20, 2013. The mean age was 63.6 years; 38 were women. Seizures were controlled in 19 of 30 (63.3%) subjects in the LCM arm and 16 of 32 (50%) subjects in the fPHT arm. LCM was non-inferior to fPHT (p = 0.02), with a risk ratio of 1.27 (90% CI = 0.88; 1.83). Treatment emergent adverse events (TEAE) were similar in both arms, occurring in 9/35 (25.7%) of LCM and 9/37 (24.3%) of fPHT subjects (p = 1.0). INTERPRETATION: LCM was non-inferior to fPHT in controlling NCS, and TEAE were comparable. LCM can be considered an alternative to fPHT in the treatment of ieNCS detected on cEEG.

Lacosamide (LCM) has become popular as an intravenous (IV) treatment for seizures. It may be perceived as safer and less prone to drug interactions than alternatives, such as phenytoin (PHT) and valproate (1). However, there has been little solid evidence supporting its use in critically ill persons. Data from the TRENdS study indicate that LCM is an acceptable choice. Results from this study indicate that LCM is noninferior to PHT. However, it is essential to understand the specific conditions of this trial and the conclusions that can be drawn from the results.

TRENdS was a multicenter randomized study comparing the efficacy of intravenous fPHT 20-mg/kg PHT equivalents with intravenous LCM 400 mg for nonconvulsive seizures (NCS) in intensive care unit patients. The primary outcome measure was continuous EEG-verified seizure control. The electroencephalographers, but not the treating physicians, were blinded to the the drug used. Provisions were made for additional dosing, maintenance dosing, and crossover to the other drug in case of failure.

What was meant by NCS in this study? The NCS afflicting these patients were either generalized or focal by EEG pattern. This was not a trial of nonconvulsive status epilepticus (NCSE). The mean baseline seizure time for these patients was only approximately 2 min/hr and those with over 50% ictal time on EEG were excluded. Neither was it a trial of convulsive status epilepticus that had converted to NCS (2), because a convulsion within the previous 24 hours was disqualifying. NCS in this trial did not necessarily mean subclinical, although most of the seizures were purely electrographic. Critically ill patients with seizures can exhibit a variety of motor movements other than full convulsions, some of which are seizures and some of which have no EEG correlate, often myoclonic twitches or brief tonic movements (3).

To put the TRENdS trial results in perspective, it is essential to understand the concept of a “noninferiority trial.” This is not intuitively obvious. Trials of this type have become increasingly popular in recent years because they are believed to circumvent the ethical issues of randomization to placebo. However, their interpretation depends upon some of the following a priori assumptions (4): 1) at least one of the treatments is assumed to be effective based on historic data, and 2) a noninferiority margin can be chosen that is clinically meaningful. This margin is based on a clinical judgment, which takes into account such things as the seriousness of the outcome and the historical results of treatment for the condition. In TRENdS, the noninferiority margin was chosen to be 20%. In other situations this margin would be inappropriate. For example, an acceptable noninferiority margin for the difference in crash rates between propeller planes and jets would be far less than 1% because of the seriousness and rarity of the outcome and the large number of samples. If the difference in outcomes between the two arms of a trial is less than this predefined margin (that is, a one-sided p value for the odds ratio exceeds 0.05) then it is concluded that neither of the treatments is inferior. Note that this does not mean that they are “equivalent.” To prove that they were truly equivalent would require an infinite number of subjects, so “noninferior” actually means “for practical purposes.” Confusing? As Diamond and Kaul (4) put it in an excellent review, “Anyone not confused by this is just not thinking clearly.”

In the light of these considerations the TRENdS trial design was appropriate. There is little doubt that PHT is effective for seizures, although specific trials for NCS in persons with altered awareness are lacking. Treatments for NCS that have a success rate within 20% of one another are worth considering, because other factors, such as expected side effects, ease of use, and costs, might well override that degree of difference. Furthermore, if one treatment does not work, an alternative could be tried, perhaps without a disastrous consequence (unlike as in the plane crash analogy).

There are limitations to the TRENdS study. Enrollment was slow, not because NCS are rare in intensive care units, but for a variety of procedural issues such as 24-hour availability of coordinators. Approximately two-thirds of patients had already been treated with levetiracetam and many also with lorazepam, so the primary conclusion applies mostly to patients who failed these drugs.

It is useful to know that LCM is effective in this circumstance. There are good reasons not to choose PHT for some patients, such as drug interactions and nonlinear pharmacokinetics (5), especially if the drug is needed for maintenance after the acute illness. Other drugs affecting voltage-gated sodium channels, such as the dibenzazepines, are not available in IV form. A corollary finding of this study was that the LCM loading dose of 400 mg was well tolerated. The commonly prescribed dose of 200 mg is probably too low as an IV-loading dose, so the TRENdS result is likely to prompt an important revision of clinical protocols.