Principles of epilepsy diagnosis and management Flashcards
1.
Epidemiology of epilepsy
Approximately 3.1 % of the US population (about 9 millions) suffers from epilepsy. About 5 % of the US population (about 15 millions) will have a seizure at some time in their lives and only just a bit more than half of those will progress into epilepsy. Incidence rate for epilepsy is about 44 per 100,000 in the US. This rate is 61 for first unprovoked seizures, 39 for acute symptomatic seizures, and 100 for all seizures including seizures from epileptic patients Hauser et al. [1]. The incidence of epilepsy tends to be higher in males and is highest at or after age 75. The most common known etiology is cerebrovascular disease at 11 %, followed by neurologic deficits from birth, mental retardation, or cerebral palsy at 8 % Hauser et al.
Provoked seizures
This term is often misinterpreted as all seizures are provoked. However, in epilepsy, the term “provoked seizure” often implies the presence of a simple fixable cause for the seizure. The term “provoked seizure” is mostly used for a selected group of seizures. These are the seizures triggered by some reversible process, such as electrolyte abnormality, severe sleep deprivation, or a medi- cation adverse effect. A provoked seizure is pre- ventable after reversing the provocation. In contrast, intractable epilepsies are often caused either by a cortical malformation, hippocampal sclerosis, or some other structural abnormalities such as acute cerebral insults. These are called symptomatic sei- zures (related to a structural abnormality) rather than provoked seizures. Even when we cannot identify a clear etiology, we assume there is always a reason for seizures including genetic causes.
Unprovoked seizure
This is defined as a seizure with no clearly identified provocation or etiology at the initial work up.
Work up of first unprovoked seizure
Appropriate work up for the first unprovoked seizure:
There is level B evidence supporting the use of EEG and brain imaging mainly MRI for the initial work-up of a first unprovoked seizure.
MRI is usually preferred over CT scans because of its higher resolution for soft tissue and malforma- tions. In one pooled analysis of 928 subjects with brain MRI after first seizure, 15 % were found to have brain MRI abnormalities. In the same study, 51 % of 1766 patients (pooled analysis) had an abnormal initial routine EEG. An additional 35 % were identified to have abnormalities on the second sleep-deprived EEG. An abnormal EEG predicts a higher recurrence rate, and a normal EEG predicts a lower recurrence rate but does not rule out epilepsy.
On average, about 50 % of individuals clinically diagnosed with a seizure have a normal first routine EEG. Laboratory testing, such as complete blood counts, blood glucose, and electrolyte panels (particularly sodium), cerebrospinal fluid analysis, and toxi- cology screening may be helpful as determined by the specific clinical scenario based on history, physical, and neurologic examination, but there are insufficient data to support or refute recom- mending any of these tests for the routine evalu- ation of adults presenting with an apparent first unprovoked seizure (Krumholz et al.).
In one study by King et al., brain MRI abnormalities were seen in 14 % (38 of 277) of patients with a first unprovoked seizure. Tumors were the most commonly identified etiology for seizures followed by developmental anomalies then hippocampal pathologies and vascular malformations King et al. [2]. In another similar study done 15 years later, 23 % (177 of 764) of patients had MRI abnormalities. This time developmental anomalies were equally common as tumors. The higher yield was attributed to the use of 3T MRIs as opposed to 1.5T in the original study Hakami et al. [3].
Seizure recurrence after first unprovoked seizure and onward, general
The recurrence rate after the first unprovoked seizure averages around 40 % within the first two years. This was concluded based on three large pivotal studies:
Meta-analysis (USA) in 1991 which found a 36 % recurrence by the two years.
FIRST Trial (Italy) in 1993 found a 51 % recurrence by two years.
UK (MESS Trial) in 2005 found 39 % recurrence by two years.
The recurrence rates were higher years after a remote symptomatic and after having two or more seizures (Tables 15.1 and 15.2).
Treatment after the first seizure decreases the recurrence rate by 30–60 % but does not eliminate the risk completely (Table 15.3).
Seizure recurrence after the first unpro- voked seizure according to etiologic factors and EEG findings Berg and Shinnar
Etiology
Idiopathic 32 %
Symptomatic 57 %
EEG
Normal 27 %
Epileptiform 58 %
Etiology + EEG
Idiopathic + Normal 24 %
Symptomatic + Abnormal EEG 65 %
Focal onset seizures had higher risk of
recurrence than generalized-onset seizures. A first seizure occurring during sleep also had higher risk for recurrence than a seizure occur- ring during wakefulness. The other risk factors for seizure recurrence included status epilepticus as first seizure, abnormal interictal neurological examination, abnormal brain imaging, multiple or clustered seizures at onset, and strong family history of seizures.
Treatment decision after first seizure
This subject has always been controversial as scholars usually differ in their practices of treat- ing or not treating the first seizure.
In general, about 60 % of the patients who had their first seizure will never have another one. In this case, it is appropriate to give the facts to the patient and let them make their decisions. If the patient prefers not to be treated and understands the risk of receiving no treatment, it is perfectly accept- able not to treat. Some patients will insist on being treated since they do not want to risk another seizure. In this case, treatment is offered for these patients provided that they understand the seizure recurrence risk is decreased but not necessarily eliminated with treatment. After an acute or remote symptomatic (cerebral insult) seizure in a clinically unstable patient, treatment is usually indicated without waiting for a second seizure.
Choice of AED
There are several available factors that could help us make an educated decision of what AED would be best to start with. The spectrum of a particular AED (narrow or broad) is one of these factors (Table 15.4). In general, narrow spectrum AEDs are reserved for seizures with known. When seizure classification is unclear, it is wiser to initiate a broad spectrum agent AED. The reason for this approach is that most narrow spectrum AEDs will not benefit patients with generalized-onset seizures or generalized
Acute: within one week, remote: after one week
Seizure rate
Untreated (%) Treated (%)
Italian FIRST study 1993 (n = 397)
6 months 41 17
24 months 51 25
UK MESS study 2005 (n = 1443)
6 months 26 18
24 months 39 32
epilepsies. On the contrary, some narrow spec- trum AEDs could cause worsening of seizures if given to a patient with generalized-onset seizures (see Table 15.4).
Mechanism of action (MOA) is another factor that may influence the selection of an AED (Table 15.5). However, the role of MOA is very limited, and the existing data do not support consideration of MOA as a major criterion in choosing an AED. Many AEDs have multiple MOAs, and some have unknown MOAs. This makes it even more difficult to make a judgment based on MOA alone [5, 6]. According to a study by Deckers CL et al., there is some evidence showing that AED polytherapy based on MOA may enhance effectiveness. In particular, com- bining a sodium channel blocker with a drug enhancing GABAergic inhibitor could be an advantageous combination. Combining two GABA mimetic drugs or combining an AMPA antagonist with an NMDA antagonist may enhance efficacy, but tolerability may be reduced with this combination.
Other factors that influence AED selection
Seizure type/classification
* Epilepsy syndrome
* Established drug efficacy for a particular sei-
zure type or epilepsy syndrome
* Safety
* Tolerability profile
* Co-morbidities (weight, cognition, psychi-
atric, other)
* Side-effect profile
* Metabolic status (renal, hepatic)
* Drug-drug interactions
* Drug formulations
* Ease of administration/titration
* Pregnancy or contraception
* Prior allergies and cross reactivity
* Cost
* Availability
Broad vs narrow spectrum AEDs
Available AEDs
Available AEDs: Currently, there are about 25 FDA-approved AEDs in the USA. The graph below lists AEDs in chronological order (Fig. 18.1).
The classic AEDs include Phenobarbital, Pheny- toin, Primidone, Ethosuximide, Benzodiazepines, Carbamazepine, and Sodium Valproate. There were no new AEDs during the period of 1978 (approval of VPA) until 1993 (approval of Fel- bamate). The AEDs which came after 1993 are considered to be newer AEDs. The newer AEDs have shown comparable efficacy to older ones. However, newer AEDS offer numerous advan- tages over the older AEDs. These include minimal drug-drug interactions, better pharmacokinetic profiles, and minimal or no long-term side effects. They are often associated with less teratogenicity and less immediate side effects. Most importantly, they usually do not need routine blood level monitoring because of low protein binding and negligible hepatic induction or inhibition proper- ties. The availability of generic formulations of some of the newer AEDs helped boost their use. AED selection is still determined on a case-by-case basis but overall, the newer AEDs mostly replaced the older ones as first choice AEDs.
MOA of AEDs
Commonly used abbreviations, brand names, and side-effects of AEDs
Studies re: AED selelction for particular seizure types, syndromes, etc.
In one study, Mattson et al. compared CBZ, PHB, PHT, or PRM in partial seizures and secondarily generalized tonic-clonic seizures SGTCS). This was a multi-center double-blind trial for initiation monotherapy. A total of 622 adult epilepsy patients were recruited. Treatment success was highest with CBZ or PHT, inter- mediate with PHB, and lowest with PRM (Mattson et al. 1985). In another multi-center double-blind study, VPA was compared to CBZ for the treatment of complex partial seizures and SGTCS in adult patients. Both drugs were equally effective in controlling SGTCS. For complex partial seizures, outcome measures favored CBZ which also had less adverse effects (Mattson et al. 1992). Glauser et al. compared ESX, VPA, and LTG monotherapies for the treatment of childhood absence epilepsy in a double-blind randomized controlled trial. This study was done on 453 children with newly diagnosed absence epilepsy. ESX and VPA had similar efficacy which was better than LTG. ESX had fewer attentional dysfunction than VPA. ESX was re-established as the first choice drug in the childhood absence epilepsy despite the availability of multiple newer AEDs. No prospective double-blind controlled study showed superior efficacy of newer drugs com- pared to the older ones. Several studies showed better tolerability and less discontinuation rates with newer AEDs. In one study, PGB had similar tolerability but inferior efficacy to LTG for the treatment of newly diagnosed partial seizures in adults Kwan et al. [7]. Another study demon- strated PGB to be non-inferior to LTG in the treatment of refractory partial seizures. In a meta-analysis by Costa J. et al., clinical compa- rability of the newer AEDs in refractory partial epilepsy was analyzed. A total of 62 randomized controlled trials comparing a new AED to a placebo as adjunctive therapy, and 8 randomized controlled trials comparing a new AED to another AED as add were reviewed. In this meta-analysis, LEV shined as the AED with high responder rate as well as low withdrawal rate. Looking at the results overall, TPM and LEV had the highest responder rates. GBP and TGB had the lowest responder rates. Withdrawal rate was highest with OXC and TPM and lowest with GBP and LEV Costa et al. [8]. Table 15.8 summarizes ideal AED(s) selection based on seizure type and comorbidities.
Ideal AED selection
Drug resistant epilepsy
Drug resistant epilepsy is defined as failure to achieve seizure freedom after adequate trials of two appropriately chosen AEDs used as thera- peutic levels in monotherapy or combination therapy (2010 consensus document by ILAE, Kwan et al. 2010). Overall, about one third of all epilepsies prove to be drug resistant. Among all epilepsies, symptomatic or cryptogenic epilepsies have higher rate of drug resistance compared to idiopathic epilepsies Kwan et al. [9]. Drug resistance also tends to be higher in patients who had more than 20 seizures prior to starting treatment. When the initial AED fails, adding a second AED or switching to another AED did not differ statistically; however, common practice favors the combination therapy Kwan et al. [9].
AED rash cross-sensitivity rates
AEDs inducers and inhibitors
