In developed countries, demographic trends project continued increases in the number of older people in the population. In the United States, there were 36.8 million adults over the age of 65 in 2005.¹ By 2030, the U.S. Department of Health and Human Services predicts that this figure will have increased to 71.5 million and account for roughly 20% of the population. In this older age group, seizures and epilepsy are the third most common neurological condition, behind only stroke and dementia. Although the primary goals of treatment—freedom from seizures, absence of adverse drug effects, and maintenance of a high quality of life—are the same for all patients with epilepsy, several issues specific to the elderly population need to be considered in approaching diagnosis and making treatment decisions. Diagnosis can be challenging because of the many comorbid conditions that are common in the elderly. These comorbidities, along with the medications prescribed to treat them, must be considered carefully when devising treatment strategies. Dealing with these issues will only become more challenging in the coming years, as the number of people over the age of 65 increases steadily.

Epidemiology

The incidence of both acute symptomatic seizures and unprovoked seizures or epilepsy is highest in people over the age of 65.²^–⁴ In a British population study, nearly 25% of newly identified seizures occurred in patients 60 years of age or older.⁵ The incidence continues to rise with increasing age: In those older than 75, the incidence is five times that of younger adults. In the United States, about 50,000 new cases of epilepsy occur each year in this age group.⁶

In people over 60 years of age, the prevalence of unprovoked seizures and epilepsy is at least 1%, and it is even higher, 1.2 to 1.5%, in people over 75 (Figure 13-1). This is a prevalence rate about twice that seen in younger adults.^10,¹¹ In a study of 1, 130, 155 U.S. veterans who were at least 65 years of age, 1.8% had epilepsy.^12,¹³ In specific populations at risk, such as nursing home residents, who have significantly higher rates of comorbid conditions associated with epilepsy, including dementia and stroke, prevalence rates as high as 3 to 9% have been reported.¹⁴^–¹⁶

Figure 13–1 Incidence of unprovoked seizures through lifespan.^3,^4,^7,⁸

Reprinted with permission from Cloyd J, Hauser W, Towne A, et al. Epidemiological and medical aspects of epilepsy in the elderly. Epilepsy Res. 2006;68 Suppl 1:S39.

Etiology

ACUTE SYMPTOMATIC SEIZURES

Because brain injuries caused by stroke, head injury, infections, neoplasms, and metabolic disturbances (hypoglycemia, hyponatremia, uremia) are so common in the elderly, it is not surprising that the incidence of acute symptomatic seizures is likewise high in this age group.⁶ Cerebrovascular disease accounts for 40% to 50% of acute symptomatic seizures, metabolic disturbances for 10 to 15%, and acute head trauma, brain infections, neoplasms and toxins/alcohol each for 5 to 10%^2,^6,¹⁷ (Figure 13-2).

Figure 13–2 Causes of acute symptomatic seizures in the elderly.

UNPROVOKED SEIZURES AND EPILEPSY

The high frequency of unprovoked seizures and epilepsy in elderly persons can also be attributed in large part to the high prevalence of a history of stroke, brain tumor, and head trauma in older age groups. Even so, about 50% of cases remain cryptogenic (Figure 13-3), although this is a significantly lower percentage than that found in younger age groups, where about 70% of cases have no identifiable cause.^3,¹⁷

Figure 13–3 Etiologies of epilepsy in the elderly.

Cerebrovascular disease is the most frequently identified antecedent of epilepsy, accounting for 30 to 40% of all new cases.^6,¹⁸ Individuals with cerebrovascular disease are 20 times more likely to develop epilepsy than the general population.¹⁹ About 15% of stroke survivors will develop unprovoked seizures within the first 5 years, and the elevated risk continues for many years thereafter.

Seizures may, in fact, be the first presentation of unrecognized cerebrovascular disease. This is especially true if previous strokes had occurred in clinically silent brain areas. In a review of new onset seizures in 4709 patients older than age 60, none of whom had a known history of cerebrovascular disease, dementia, tumor, or alcohol abuse, Cleary and colleagues found that the relative risk of stroke was nearly three times that of controls.²⁰ Hypertension is also an independent risk factor for unprovoked seizures,²¹ although hypertension may be a surrogate for the progressive arterial changes associated with lacunes, other small strokes, and periventricular white matter lesions.

Although the incidence of head trauma is highest in adolescents and young adults, a relative peak of occurrence in the elderly is largely attributable to falls. Approximately one-third of people over the age of 65 living at home, and about half of nursing home residents, have at least one fall each year.^22,²³ Trauma with loss of consciousness is associated with a threefold increased risk of epilepsy.⁶ With more severe head injuries, risk is even higher. Overall, head trauma accounts for 2 to 3% of new cases of epilepsy in the elderly.

Although the incidence of central nervous system infections is highest in childhood, a second peak occurs in the elderly. Survivors of central nervous system (CNS) infections have a threefold risk of developing epilepsy, and a history of infections accounts for 2 to 3% of cases.²⁴

Alzheimer’s and other neurodegenerative diseases are associated with a 5- to 10-fold increase in epilepsy compared to the general population.^6,²⁵ Unprovoked seizures may be seen in as many as 8 to 15% of patients with Alzheimer’s disease.^25,²⁶

The incidence of intracranial neoplasms increases with age, and these are strongly associated with epilepsy. However, given the progressive nature of most brain tumors, it is often difficult to distinguish acute symptomatic seizures from unprovoked seizures or epilepsy.

Clinical Presentation

As would be expected from the etiological profile, the great majority of seizures in the elderly, more than 70%, are of partial onset.^18,^27,²⁸ Even generalized tonic-clonic seizures without obvious focal features are likely to have a localized or regional onset, as it is very unusual for idiopathic generalized epilepsy syndromes to present at this age. In any event, diagnosis of epilepsy in the elderly can be difficult and may be delayed. In a recent Veteran’s Administration cooperative study of epilepsy in elderly patients, epilepsy was an initial diagnostic consideration in 73% of patients that were eventually diagnosed with epilepsy.¹⁸ In a subset of 151 patients from this study, Spitz and colleagues found that the delay from initial symptoms to diagnosis was 2.3 years, with a median time of 1 year. Only 37% of patients in this subgroup were correctly diagnosed at the time of initial evaluation. Two-thirds of patients with generalized tonic-clonic (GTC) seizures, but only one-fourth of patients with complex partial seizures, were correctly diagnosed.²⁹

Delay in making the correct diagnosis likely derives from three characteristics of seizures in this population:

1. The manifestations of seizures in the elderly may be quite different from those of younger patients.

2. Epileptic symptoms in the elderly are often attributed inaccurately to other disorders common in this age group.

3. Adequate descriptions of the seizures are often lacking.

Because the lesions associated with epilepsy in elderly people can involve any area of the brain, extratemporal neocortical epilepsies are more common than in younger adults. Consequently, seizures may manifest with a wide variety of sensory, visual, cognitive, and behavioral phenomena that are frequently atypical for the physician’s experience and thus more difficult to recognize as having an epileptic basis. Classic auras are less common in the elderly, and initial manifestations of partial seizures, such as dizziness, a vague feeling related to the head, memory loss, or confusion, may be interpreted as nonspecific symptoms due to any number of possible causes. Thus, given the high frequency with which metabolic derangements, cerebrovascular events, and dementia occur in the elderly, clinical manifestation of partial seizures, as well as of postictal states, may be incorrectly attributed to these other conditions. For example, postical aphasia or hemiparesis may be diagnosed as ischemic events.³⁰ Finally, although published data are lacking, it is said that generalized tonic-clonic seizures, the easiest type of seizure to recognize, occur with less frequency in older age groups.

Falls in the elderly are common and a frequent cause of admission to a hospital.²³ However, seizures are rarely considered an important etiology, especially early in the diagnostic evaluation. More often, falls are attributed to cardiovascular, cerebrovascular, or arthritic etiologies. Confusion or memory loss, which may be ictal or postical in nature, is often first considered to be a manifestation of dementia, metabolic abnormalities (e.g., dehydration) or a head injury related to a fall.

Difficulty in diagnosis is compounded by the increase in social isolation in the elderly population. In 2006, the U.S. Department of Health and Human Services found that 30% of noninstitutionalized people over 65 years of age, totaling more than 10 million people, lived alone.¹ As such, elderly patients are often brought to the hospital by emergency medical services without anyone available to provide information through direct observation of an episode.

NONEPILEPTIC PAROXYSMAL PHENOMENA

As at other ages, elderly persons can have episodic paroxysmal phenomena that mimic seizures but are nonepileptic (Table 13-1). In a study of 94 patients at least 60 years of age (mean 70 years) who were referred for video-electroencephalogram (EEG) monitoring for evaluation of paroxysmal episodes, 27 (29%) had nonepileptic events, including 13 with psychogenic seizures.³¹ The majority of these patients had been taking antiepileptic drugs for presumed epileptic seizures. Such findings underscore the need for video-EEG monitoring in patients whose seizures are atypical or have not responded to treatment with antiepileptic drugs (AEDs).

TABLE 13–1 Paroxysmal Phenomena That May Mimic Seizures in Elderly Patients

Syncope

Transient ischemic attacks

Transient global amnesia

Migraine

Drop attacks

Myoclonus

Confusional episodes due to medication interactions or overmedication

Hypoglycemia

Electrolyte disturbances/dehydration

REM behavior disorder

Nonepileptic psychogenic seizures

Risk of Recurrence

Physicians should have a low threshold for suspecting seizures in the elderly because of the increased incidence and the higher rate of seizure recurrence in this population.^3,⁴

In patients presenting with a first unprovoked seizure, the overall risk of recurrence with at least 2 years of follow-up is about 38% (range 25 to 52%).³²^–³⁵ Factors that increase the risk of recurrence include abnormal EEG findings and presence of an underlying definable etiology.

Data regarding specific effects of older age on risk of recurrence are inconclusive, and there are no population-based studies. However, when seizures can be attributed to a recognized underlying neurologic condition (that is, they are symptomatic of an acquired pathogenic mechanism), risk of recurrence is roughly double that of a cryptogenic first seizure.³³ In selected populations, the recurrence rate may be especially high. For example, seizures recurred in more than 80% of a small series of patients with a remote history of stroke.³⁶

As the majority of newly diagnosed unprovoked seizures in the elderly are symptomatic (that is, they have a known antecedent cause), it is reasonable to assume that most elderly people have a high risk of seizure recurrence. Thus, in contrast to accepted treatment paradigms in younger adults, we believe that it is reasonable to begin anticonvulsant medication after the first unprovoked seizure in someone over 65, even in the absence of an underlying demonstrable lesion.

Treatment Considerations

Although it often stated that epilepsy is more easily controlled in elderly patients than in younger ones,^37,³⁸ it is not clear that this is accurate, as no well-controlled large studies specifically address this issue. Data from two recent studies looking at tolerability of anticonvulsant drugs in the elderly showed that seizure-free rates are similar to those seen in the adult population at large. In a randomized trial that compared response in the elderly to treatment regimens using modest dosages of carbamazepine (dosage range 200 to 800 mg daily) and lamotrigine (dosage range 75 to 300 mg daily), Brodie and colleagues found that only 33% remained seizure free during the final 16 weeks of the study.³⁹ Although a high rate of medication withdrawal likely contributed to the relatively low rate of seizure control, these findings suggest that seizures are, in fact, not necessarily easily controlled in this population. Similarly, in the VA cooperative study #428, which randomized elderly patients with new-onset epilepsy to treatment with gabapentin, lamotrigine, or carbamazepine, 53% of patients who remained on treatment were seizure free at 12 months.⁴⁰ These data are not substantially different from seizure control rates in the general population, as illustrated by the study by Kwan and Brodie of 470 previously untreated patients with epilepsy, ranging in age from 9 to 93.⁴¹ They found that 47% of patients became seizure free on the first drug, and a total of 61% became seizure free with the second or third monotherapy agent.

PHARMACOKINETIC CHANGES

Significant physiological changes occur in elderly persons that affect drug pharmacokinetics, and these must be considered in determining dosing regimens. However, the degree of these changes and the extent to which they occur at a particular age are not reliably predictable due to the great degree of normal variability and the extent to which disease-related changes may be present.

Absorption

Absorption of drugs depends on dissolution of their particular formulations. This, in turn, is mainly related to gastric acid secretion, which often declines in the elderly.⁴² In addition, to varying degrees, gastric emptying slows, intestinal transit time increases, mesenteric blood flow decreases, and the intestinal absorptive surface may decrease.⁴³ All these changes contribute to variable and often unpredictable absorption of different drugs. Overall, this combination of factors typically results in a diminished ability to absorb antiepileptic drugs, which reduces their bioavailability.

Protein Binding

In healthy adults, serum albumin concentration decreases only slightly with age.⁴⁵ However, in elderly patients with acute systemic and neurological illnesses, serum albumin levels can decline significantly. Suboptimal nutrition may exacerbate this. With reductions in serum albumin concentrations, the free fraction of highly protein-bound medications can increase substantially, sometimes resulting in prominent adverse effects despite little or no change in the total serum level.

AEDS that are highly protein bound include tiagabine, phenytoin, valproate, diazepam, clonazepam, clobazam, and to a lesser extent, carbamazepine.⁴⁶

Hepatic Clearance

Hepatic mass and blood flow both decline with age. As a result, liver volume is about 25% lower in people age 65 compared to young adults.⁴⁷ The degree to which the various hepatic enzymes change with increasing age is not known definitively.⁴⁸ The majority of drugs metabolized in the liver utilize the cytochrome P450 system. With increasing age, this system is believed to decline in function, though to a variable and unpredictable degree. The hepatic glucoronidation conjugation process is thought to be less affected by age.

AEDs metabolized primarily by the cytochrome P450 system include phenytoin, phenobarbital, primidone, carbamazepine, ethosuximide, oxcarbazepine, and tiagabine. Felbamate, topiramate, valproate, and zonisamide are also partially metabolized via this pathway.⁴⁶ AEDs that primarily undergo conjugation include lamotrigine, zonisamide, and valproate.⁴⁶

Renal Clearance

The most consistent age-related change affecting pharmacokinetics is a decline in renal function related to reduction in renal mass and loss of glomeruli. This results in lower glomerular filtration rate (GFR) and a reduced ability to handle renally excreted medications and toxins. On average, GFR declines by about 50% between the third and eighth decades of life.⁴⁹ However, this is variable, and about one-third of people may not experience declines of this degree.⁵⁰ Because muscle mass, the source of serum creatinine, also decreases with age, changes in serum creatinine levels frequently do not parallel the decline in GFR. The validity of common formulas used to estimate GFR has been questioned, even in otherwise healthy elderly patients.⁵¹

AEDs that are excreted primarily by the kidney include gabapentin, levetiracetam, vigabatrin, and pregabalin. Felbamate, zonisamide, and topiramate are also partially renally excreted.⁴⁶

Table 13-2 summarizes the effect of older age on AED clearance.

TABLE 13–2 Average Changes in Apparent Oral Clearance of Older and Newer AEDs in Elderly Patients (Interindividual variation may be considerable in relation to age and other factors.)

Drug	Effect of Old Age on Drug Clearance
Carbamazepine	Decrease by 25–40%⁵²
Felbamate	Decrease by 10–20%⁵³
Gabapentin	Decrease by about 30–50%⁵⁴
Lamotrigine	Decrease by about 35%⁵⁵
Levetiracetam	Decrease by about 20–40%⁵⁶
Oxcarbazepine	Decrease by 25–35%⁵⁷
Phenobarbital	Decrease by about 20%⁵⁸
Phenytoin	Decrease by about 25%⁵⁹
Tiagabine	Decrease by about 30%⁶⁰
Topiramate	Decrease by 20%⁶¹
Valproic acid	Decrease by about 40%⁶²
Vigabatrin	Decrease by 50–85%⁶³
Zonisamide	No data

Reprinted with permission from Perucca E, Berlowitz D, Birnbaum A, et al. Pharmacological and clinical aspects of antiepileptic drug use in the elderly. Epilepsy Res. 2006;68[Suppl 1]:S49-S63.

PHARMACODYNAMIC CHANGES

In addition to age-related pharmokinetic changes, it is likely that actions of AEDs at the cellular level are also altered in the elderly. Although such changes have not been well characterized, the appearance of adverse central nervous system effects (e.g., drowsiness, unsteadiness) at unbound serum levels that rarely produce similar toxicity in younger adults is probably one reflection of this. As a result, it is important to anticipate a much narrower therapeutic window for AEDs in the elderly population (Figure 13-4).

Figure 13–4 Effect of age on therapeutic ranges. The elderly typically have a narrower therapeutic window, the range between the lowest effective concentration and the maximal tolerated concentration.^64,⁶⁵

Reprinted with permission from Bergey GK. Initial treatment of epilepsy: special issues in treating the elderly. Neurology. 2004;63[10 Suppl 4]:S40-S48.

CHOOSING AN AED

As nearly all of the currently used AEDs have similar efficacy in treating partial and secondarily generalized seizures, choice of medication depends mainly on tolerability and the presence of comorbid conditions. Nearly 90% of elderly patients in the community take at least one prescription drug,⁶⁶ and nursing home residents take an average of five.⁶⁷ Serum levels of AEDs that are metabolized by the cytochrome p450 enzymes are most likely to be affected, if other medications administered concurrently inhibit this pathway. Common examples include H₂-blockers (cimetidine), macrolide antibiotics (erythromycin, clarithromycin), antifungal agents (ketoconazole, fluconazole), and isoniazid. Ticlopidine increases levels of carbamazepine and phenytoin. Thus, the possibility of therapeutically significant drug interactions must be a major consideration when selecting an AED.

Detailed information concerning drug interactions can be found in an extensive review by Patsalos and Perucca.⁶⁸

Older-Generation AEDs

Although phenytoin, phenobarbital, primidone, valproate, and carbamazepine are effective in localization-related epilepsy,^69,⁷⁰ tolerability concerns have generally displaced them as drugs of first choice in elderly patients.

Phenobarbital and primidone have significant sedative and cognitive effects and the first VA Cooperative Trial ⁷⁰ showed them to be much less well tolerated than carbamazepine and phenytoin. They are also potent inducers of hepatic enzymes, and this reduces the efficacy of many drugs administered concurrently to treat comorbid conditions.

Phenytoin also has troublesome features. The transition from first-order to zero-order kinetics at therapeutic dosages often leads to widely variable serum levels, even with small changes in dosing. This greatly increases the likelihood of toxicity. Development of imbalance and ataxia at modest dosages increases the risk for falls and consequent fractures. And as already noted, medication interactions are common and often problematic. When phenytoin must be used, development of toxicity can be minimized by starting at no more than 100 mg BID and making subsequent dosage adjustments in increments of 30 mg.

Drug interactions are also common with valproate because it is both a potent inhibitor of hepatic enzymes and also highly protein bound. Occasionally, encephalopathy occurs due to hyperammonemia, which can occur in the absence of hepatic enzyme abnormalities.⁷¹

Carbamazepine is another inducer of hepatic enzymes, and thus drug interactions must be anticipated. In addition, hyponatremia may be more common and less well tolerated in elderly patients taking carbamazepine.⁷²

Newer-Generation AEDs

Two prospective trials have shown that lamotrigine and gabapentin are better tolerated than carbamazepine.

In a recent VA cooperative study, patients 65 or older with newly diagnosed epilepsy were randomized to treatment with gabapentin 1500 mg per day, lamotrigine 150 mg per day, or carbamazepine 600 mg per day.⁴⁰ The primary outcome measure was early termination, most often due to adverse effects, which occurred in 64.5% of patients taking carbamazepine in contrast to 51% of patients taking gabapentin and 44% of patients taking lamotrigine. Seizure-free rates and time to first seizure did not differ significantly among the three medications.

An earlier study by Brodie and colleagues, in which newly diagnosed elderly patients with epilepsy were randomized to lamotrigine or carbamazepine, had similar findings: 42% of patients on carbamazepine dropped out due to adverse events compared to only 18% of patients on lamotrigine.³⁹ Somnolence occurred in 29% of patients on carbamazepine versus only 12% of patients on lamotrigine. It was also of interest that only 3% of patients on lamotrigine developed allergic rash compared to 19% of those taking carbamazepine.

In general, the newer generation AEDs are less sedating and overall have more favorable adverse effect profiles than the older generation AEDs. Details of these comparisons were reviewed extensively in the 2004 AAN guidelines on the efficacy and tolerability of the newer antiepileptic drugs.⁷³

Lamotrigine has a generally favorable adverse effect profile. It is nonsedating and does not produce significant cognitive dysfunction. It is not an inducer of hepatic enzymes, nor is it highly protein bound, factors important in minimizing the possibility of drug interactions. However, enzyme-inducing medications and hormone replacements can reduce blood levels significantly. Valproate increases lamotrigine levels two- to threefold, and caution is necessary when using the two together. One disadvantage of lamotrigine is that therapeutic dosages cannot be achieved rapidly, in part because of the increased risk of potentially severe allergic reactions. Gradual dosage titration reduces this risk to one that is no greater than that seen with older AEDs.

Levetiracetam also has a favorable adverse effect profile. It is nonsedating and does not produce significant cognitive dysfunction. It is renally excreted, is not highly protein bound, and does not induce hepatic enzymes. As a result, drug interactions are rare. About 7% of patients experience irritability and behavioral side effects.⁷⁴ A major advantage of levetiracetam is that therapeutic efficacy can be achieved at starting dosages.

Gabapentin and pregabalin likewise have therapeutic efficacy at starting dosages. Both drugs are renally excreted, are not significantly protein bound, and do not induce hepatic enzymes.

Oxcarbazepine is metabolized in the liver, but it is only a weak inducer of hepatic enzymes. Drug interactions are minimal. Like carbamazepine, hyponatremia is seen more often in elderly patients, and concomitant use of diuretics should be cautious.⁷⁵

Zonisamide and topiramate are generally well tolerated, but both are associated with higher incidences of cognitive side effects. For each, metabolism is partially renal and partially hepatic; neither induces hepatic enzymes. Because of the risk of adverse cognitive effects, zonisamide and topiramate should be titrated slowly. As both are weak carbonic anhydrase inhibitors, they are associated with a small increase in the risk for renal stones. This requires ensuring adequate hydration in elderly patients who take these drugs.

To summarize, lamotrigine, levetiracetam, gabapentin, pregabalin, zonisamide, and topiramate have the least potential for interaction with medications used to treat common comorbid conditions in the elderly. However, because zonisamide and topiramate are more likely to produce cognitive side effects than the other drugs, and they also increase the risk of renal stones, we prefer using lamotrigine, levetiracetam, gabapentin, or pregabalin as initial therapy in elderly patients with seizures.

EPILEPSY SURGERY

Neurologists have generally been reluctant to consider resective surgery in elderly patients with epilepsy because of the higher incidence of associated comorbidities, as well as a perceived increase in complication rates. In addition, a long duration of epilepsy has been reported to be negative predictor for achieving long-term seizure control.⁷⁶ However, in appropriate patients, one might equally argue that the same comorbid conditions and their consequent effects on drug absorption, clearance, rate of adverse effects, and interactions justify earlier consideration of potentially curative epilepsy surgery.

Reports on the efficacy of surgery in elderly patients come from a relatively small series of patients.⁷⁷^–⁷⁹ Grivas and colleagues recently published the outcome of temporal lobe resection in 52 patients older than 50 (range 50 to 71, mean 56) years.⁸⁰ At follow-up of at least 12 months (mean duration 33 months), 71% remained seizure free (Engel class 1). This seizure-free rate is virtually identical to that seen in a comparable cohort of 321 patients less than 50 years (72% seizure free) operated on in the same time period. Notably, 11 of the 52 patients were older than 60, and outcome was equally good in this subgroup. Only one patient showed no improvement in postoperative seizure control (2% Engel class 4). Surgical complications were somewhat higher (7.7%) in this older cohort, and there was also a higher rate (3.8%, 2 patients) of permanent mild neurologic morbidity, compared to younger patients. No patient had severe permanent neurologic morbidity, and there were no mortalities. These data suggest that surgery should not be excluded as a treatment option simply on the basis of age.

Epilepsy Morbidity

STATUS EPILEPTICUS

The incidence of status epilepticus in persons older than 60 years of age is at least twice, and possibly greater than 10 times, that seen in younger adults. The highest rates occur in patients older than 70.⁸¹^–⁸³ In a series of 171 elderly patients with status epilepticus, more than half (56%) had no previous history of seizures.⁸⁴ Cerebrovascular disease is the most common cause, accounting for as many as 60% of cases.^81,^85,⁸⁶ Other causes include hypoxia, metabolic derangement, alcohol, tumor, infection, trauma, and dementia. The mortality rate associated with status epilepticus is 38% in patients over 60 years of age, but nearly 50% in those over 80. In contrast, the overall mortality rate for status epilepticus is 22%.^84,⁸⁵ The increased mortality rates reflect the elderly population’s susceptibility to systemic illnesses (e.g., pneumonia, organ failure), as well as the often life-threatening or progressive nature of the underlying cause of the status epilepticus. Along with advancing age, etiology is a strong determinant of mortality: The highest rates occur from acute symptomatic causes, such as ischemic stroke, intracranial hemorrhage, and anoxic brain injury.^87,⁸⁸

Nonconvulsive status epilepticus (NCSE) represents up to one-fourth of cases in the elderly. As these patients often present with varying degrees of confusion, behavioral change, altered consciousness, and sometimes subtle motor activity, diagnosis can be challenging. It is particularly important to consider this diagnosis when evaluating elderly patients with new-onset encephalopathy.

There are no established protocols for treating status epilepticus that are age specific, but the same pharmacokinetic considerations discussed earlier in this chapter must be considered. Pharmacodynamic changes and the presence of comorbid conditions in elderly patients increase their susceptibility to CNS and respiratory depression from benzodiazepines and barbiturates, as well as to the cardiovascular effects of intravenous phenytoin and fosphenytoin.

BONE HEALTH, FALLS, AND FRACTURES

The combination of age-related diminished bone health and the increased risk of falls results in a high rate of fractures in the elderly population. Epilepsy confers an additional risk of falls and consequent fractures, as do the deleterious effects of older generation AEDs on bone health.

Patients with epilepsy have an approximately twofold increase in fractures compared to controls.⁸⁹ In a study of more than 8000 community-dwelling women older than 65, those taking AEDs were 75% more likely to experience a fall.⁹⁰ Although part of this increase relates to the seizures themselves causing falls, loss of balance due to the increased sensitivity of the aging brain to adverse CNS effects of AEDs also plays a role.

In adults, peak bone mineral density is attained between 20 and 30 years of age. After this, bone mineral density gradually declines, but it becomes most pronounced in women following onset of menopause. Older-generation AEDs, including phenytoin, carbamazepine, primidone, and phenobarbital, are commonly associated with altered bone metabolism and decreased bone density because they are potent inducers of the hepatic cytochrome P450 enzyme system.⁹¹ Valproate, an inhibitor of the P450 system, was initially believed not to affect bone health. However, it has recently been shown to be associated with reductions in bone mineral density. Polytherapy has a higher risk of abnormalities in bone metabolism than monotherapy. Multiple mechanisms have been postulated to support the association, but none is entirely satisfactory.⁹²

To date, the newer generation AEDs, including lamotrigine, gabapentin, and levetiracetam, have not been associated with deleterious effects on bone mineral metabolism and bone mineral density.