Universitätsklinik für Neurologie
The problem of seizures and epilepsy in the elderly is growing along with the ageing population. Although epilepsy is generally perceived as a condition of the young, the highest incidence rates are observed in the elderly population,(1) and mis- and underdiagnoses are common.(2) As people grow older, they become more likely to have concurrent medical illnesses, which require multiple medications. This leads to increasingly challenging drug treatments, since comedication, combined with altered metabolism and excretion, can result in increased susceptibility to neurotoxic side-effects and adverse drug reactions.(3) Moreover, aged people are particularly vulnerable to the sequels of seizures, including physical injury and reduced independence. Despite these issues, access to specialised epilepsy services becomes more difficult with advancing age.(4,5)
Antiepileptic drug treatment in elderly people
The use of antiepileptic drugs (AEDs) in the elderly can be challenging, since multiple pharmacokinetic and pharmacodynamic parameters are altered in the ageing individual.(6) These changes probably affect slowly absorbed AEDs, particularly those administered as solid-dosage forms or extended-release formulations, or drugs absorbed by active transporters, such as gabapentin. Older individuals experience a gradual decrease in serum albumin and increased alpha(1)-acid glycoprotein (AAG) concentrations.(7) This may result in a free (pharmacologically active) fraction of highly albumin-bound AEDs (eg, phenytoin or valproate) higher than expected. This phenomenon should be taken into account when interpreting the serum levels of these drugs in the elderly, as therapeutic and toxic effects may be seen at total drug concentrations lower than usual.(7) In elderly patients, the measurement of unbound concentration is encouraged when altered protein binding is suspected, or when response does not correlate with total drug concentration.
The reduction in liver mass is the age-related change that has the greatest effect on pharmacokinetics, resulting in decreased drug metabolising capacity and diminished kidney function, which leads to decreased clearance.(8) Decrease in drug-metabolising capacity results in slower unbound clearance, which, in turn, will cause an increase in unbound concentration and, if there is no change in protein binding, total drug concentration. Renal function decreases by approximately 10% with each decade of age, starting at age 40.(8) Creatinine clearance, which is a reliable marker of glomerular filtration, correlates well with unbound renal clearance of drugs eliminated by the kidneys. In our institution, we measure creatinine clearance in addition to performing whole-blood cell count, liver function test and amylase, lipase and clotting tests in every patient older than 60 years before starting an AED treatment.
In addition to pharmacokinetic changes, older patients appear to be more sensitive to the central nervous system (CNS) and systemic adverse effects of AEDs, especially on cognition.(9,10) However, controlled data are scarce and contradictory.(11)
Standard antiepileptic drugs
Standard AEDs, which are still the mainstay of
pharmacological treatment in the elderly, are all sodium channel inhibitors.(12) However, due to the CNS-related adverse effects of barbiturates, the use of these drugs is not advised in these patients.
Due to its nonlinear pharmacokinetics, phenytoin is difficult to use, and serum drug concentrations (SDCs) are frequently disproportionately outside the usual therapeutic range in elderly patients.(13,14) Phenytoin metabolism, mainly by CYP 2C9, is capacity-limited and decreases with age, resulting in decreased clearance in elderly patients. An 84% lower clearance of unbound phenytoin was observed in 34 elderly patients, compared with 26 younger adults, at steady-state conditions.(14) These findings indicate the need for a smaller initial dose (no more than 50mg) and smaller dosage increments than is usually recommended for younger adults. Long-term phenytoin treatment may decrease folate levels, and ultimately lead to megaloblastic anaemia,(15) and in debilitated elderly people folate substitution is often required.
Phenytoin may cause ataxia and dizziness, especially at high levels, leading to imbalance and repetitive falls. A study investigating the risk factors for nonvertebral fractures in elderly women found that phenytoin, despite relatively low usage, was the only drug with a significant effect.(16)
Carbamazepine and its active epoxide metabolite are bound to both AAG and, to a lesser extent, to albumin.(7) Unlike albumin, AAG increases with advanced age, which may lead to an increase of total SDCs without any change in the unbound (pharmacologically active) fraction. Measuring the unbound fraction may be helpful, although this measurement is not performed routinely. In three studies of elderly patients, carbamazepine clearance during long-term therapy was 23–40% lower than in younger patients,(17) suggesting the need for dosage adjustments in elderly patients. Dose-related adverse effects include dizziness and diplopia, which place elderly patients at risk for falls.
Valproate is a highly protein-bound drug;(18) thus, valproic acid may either displace other highly protein-bound drugs (eg, salicylic acid) or be displaced by them, which results in elevated unbound valproate concentrations. With increasing age and lower protein binding, the free unbound fraction increases and the clearance of the unbound fraction decreases by 65%, compared with younger patients; however, the decreased unbound clearance is fully compensated by lower protein binding, so that total valproate SDCs are similar to those in younger adults.(19) Consequently, dose reductions of up to 50% are recommended in the elderly patients. If there is discord between total SDC and clinical response, the measurement of the free unbound fraction is recommended.(20)
Newer antiepileptic drugs
Newer AEDs generally offer a better pharmacokinetic profile and a more favourable safety profile than standard AEDs. However, careful postmarketing evaluations led to the recognition of severe, at times fatal, adverse effects of some newer drugs, such as felbamate and vigabatrin, leading to restricted use; treatment of elderly patients with these drugs is not advised.
Although the pharmacokinetics of gabapentin are not altered with age, clearance is reduced with impaired renal function (creatinine clearance <60ml/min).(21) Therefore, a potentially lower starting dose, based on renal function, and a slower titration, lower than normal target dosages, are recommended in elderly patients. The most common side-effects in the elderly are somnolence, dizziness, ataxia and peripheral oedema.(22) Some patients may also experience excitatory CNS effects with hypomanic mood changes.(23)
Lamotrigine is moderately bound to plasma proteins and is primarily eliminated by glucuronidation in the liver; 90% of the renally excreted drug is in the form of inactive glucuronide conjugates.(24) Unlike oxidative drug metabolism, conjugative metabolism appears to remain unchanged with increasing age. Valproate inhibits glucuronidation of lamotrigine, decreasing its clearance by approximately 50%,(25) indicating the need for lower starting doses (maximum 12.5mg/two weeks). Conversely, common enzyme inducers, such as carbamazepine and phenytoin, increase lamotrigine elimination, resulting in reduced half-life;(26) this has also been observed with a concomitant administration of acetaminophen.(27)
A large study comparing carbamazepine and lamotrigine in 150 elderly patients found that patients receiving carbamazepine were more than twice as likely to discontinue the drug due to adverse effects (skin rash: 3% with lamotrigine vs 19% with carbamazepine).(28)
Other newer antiepileptic drugs
Topiramate has an appropriate efficacy profile for use in older patients: it is both metabolised in the liver and excreted unchanged in the urine. However, there may be a reduction of its clearance with advanced age.(29) Topiramate does have substantial negative effects on cognition, such as impaired concentration, difficulties in word finding or verbal fluency,(30) which may be even more severe in the elderly. However, experience with elderly patients is limited.
Levetiracetam, a newly licensed drug, is primarily excreted renally as unchanged drug, but a small fraction is hydrolysed in the liver, a process that does not involve cytochrome P450 enzymes. The drug accumulates with renal insufficiency; therefore, dose reductions are recommended in elderly patients with reduced renal capacity. The most commonly reported side-effects are asthenia, somnolence and dizziness, but no studies with older patients are yet available.
- Epilepsia 1992;33 Suppl 4:S6-14.
- Epilepsia 2002;43:165-9.
- Acta Neurol Scand 2003;180 Suppl:33-6.
- Epilepsia 2002;43 Suppl 8:S53.
- J R Soc Med 1995;88:686-9.
- Antiepileptic drugs. Philadelphia: Lippincott Williams & Wilkins; 2002. p. 149-58.
- Clin Pharmacokinet 1987;12:41-72.
- J Gerontol 1976;31:155-63.
- Clin Pharmacol Ther 1999;65:630-9.
- Epilepsy Res 1995;21:149-57.
- Epilepsia 2001;42:764-71.
- Epilepsia 1999;40:1471-83.
- Ann Pharmacother 1997;31:279-84.
- Clin Pharmacol Ther 1982;31:301-4.
- Antiepileptic drugs. New York: Raven Press; 1995. p. 329-38.
- Am J Epidemiol 1999;149:1002-9.
- Epilepsia 2003;44:923-9.
- Ther Drug Monit 1986;8:411-5.
- Clin Pharmacol Ther 1985;37:697-700.
- Ann Pharmacother 1993;27:32-5.
- Epilepsia 1999;40;474-9.
- JAMA 1998; 280:1837-42.
- Seizure 2000;9:505-8.
- Clin Pharmacol Ther 1987;13:191-203.
- Clin Pharmacol Ther 1996;60:145-56.
- Br J Clin Pharmacol 1991;32:658.
- Clin Pharmacol Ther 1990;48;346-55.
- Epilepsy Res 1999;37:81-7.
- Epilepsia 1999;40 Suppl 7:S105.
- Epilepsia 2003;44:339-47.