Q: What approach do you recommend in patients with epilepsy who fail the first AED? 

When starting a patient on therapy, it is important to select an agent that not only will likely work for the patient’s type of seizure, but also can be easily withdrawn or supplemented with another AED. Studies show that the majority of patients are not going to be on the first AED that was prescribed to treat their epilepsy 2 years after starting treatment. In light of this, it is reasonable to anticipate the possibility of adding and/or sequencing drugs with different mechanisms of action (MOAs). We often see patients whose first and second AEDs were sodium channel blockers that did not work, and, yet, they were put on a third sodium channel blocker. While it is true that even AEDs in the same class may differ from one another, it makes more sense in such cases to consider replacing that AED with another agent that works through a different MOA or combining AEDs with different MOAs. Thus, you have to be thinking ahead, from the outset. There are abundant data showing that combining AEDs with the same MOA can worsen the side-effect profile. Combining 2 sodium channel blockers dramatically enhances side effects, which is why, for example, the prescribing information for eslicarbazepine warns against combinations with oxcarbazepine.

Of course, cost is also a consideration, and costs are not limited to drug costs. Newer AEDs that reduce the side-effect burden and simplify the dosing schedule may result in improved adherence, which, in turn, may lead to lower health care utilization and costs. Unfortunately, many insurance companies still “silo” costs and simply see a newer drug that is more expensive. They do not necessarily consider that, if we can use an AED that helps to keep the patient out of the hospital and from having to visit the doctor as often, the global costs to the system may actually be less than with the older drugs. 

The next steps in a case of new-onset epilepsy when the dose of the first AED has been optimized but the patient is still having seizures depend on each specific scenario. Numerous factors come into play (eg, the patient’s seizure type, the level of efficacy with the current AED, the side effects associated with the AED), and there is no 1-size-fits-all approach. With monotherapy, the patient is much less likely to experience side effects than with polytherapy. Sometimes, discontinuing a patient’s first therapy and converting them to a new monotherapy is attributed to the first therapy not only being ineffective, but also causing side effects. If a patient is tolerating the first AED but it has not been fully effective, I will often add a second AED, get the second AED to a good level, and watch the patient for 3 to 6 months on the new regimen. If the individual is seizure free, then it becomes a question of whether to whittle away at the first AED that never made them seizure free in the first place, and this depends on patient-specific factors. Whether rational polytherapy improves efficacy is a controversial topic. Lamotrigine and divalproex sodium result in increased lamotrigine levels, leading to potentially higher efficacy. But that also increases the risk of side effects such as rash and, rarely, Stevens-Johnson syndrome. I try to practice rational polytherapy using different MOAs to avoid the worsening of side effects. Even if the evidence for improved efficacy is not extremely robust, rational polytherapy can make the situation better when fewer side effects leads to improved patient compliance.

In the specific scenario involving the failure of a single AED, we would not yet consider that patient to be drug refractory. There may be factors that would guide you toward either the use of adjunctive therapy or the conversion to a different AED as monotherapy; however, studies generally have not shown a difference in efficacy between these 2 approaches. A patient is considered to have drug-resistant or refractory epilepsy once they have failed therapy with 2 different AEDs. At this point, you try to confirm the diagnosis of epilepsy, the seizure type, and appropriate AED and dosing for the seizure type. This is determined definitively through prolonged electroencephalogram (EEG) video monitoring, where we record the seizures on video and EEG to determine whether we are, in fact, dealing with epilepsy and then the type of epilepsy. Is it a primary generalized epilepsy, a focal epilepsy, or a multifocal type of epilepsy? On the basis of those results, we can usually identify appropriate treatment options, which vary by epilepsy type. For a smaller percentage of patients, monitoring is unsuccessful and we are unable to record an event. In these cases, we live with diagnostic uncertainty. Some of the older AEDs, such as phenytoin and carbamazepine, may work well for focal epilepsy but not for primary generalized epilepsy; in fact, they may aggravate some forms of primary generalized epilepsy. The broad-spectrum agents, such as levetiracetam and perampanel, work well for both focal epilepsy and primary (idiopathic, genetic) generalized epilepsy.