This site welcomes original publications, review articles, case records in the field of neurology, psychiatry, neuroradiology, neuropathology, and neurosurgery
The author: Professor Yasser Metwally
http://yassermetwally.com
INTRODUCTION
August 23, 2010 — Scleroderma is characterized by widespread microvascular changes and diffuse tissue fibrosis affecting predominantly the skin, gastrointestinal tract, lungs, heart, an kidneys. In the skin there is atrophy of the dermis and dermal appendages, loss of elastic tissue, and a marked increase in dermal collagen. This may be paralleled by interstitial fibrosis of the viscera. Arterioles, capillaries, and sometimes small arteries undergo hyalinization with marked endothelial proliferation, intimal, and adventitial fibrosis, with resultant end-organ hypoperfusion.
The incidence of systemic sclerosis is 17 cases/million/year (approximately 40% of the incidence of SLE). Women are affected three times as often as men. African-Americans are more frequently and severely affected than whites. Persons of all ages may develop the disease, but the peak incidence is in the fourth through sixth decades of life. There is a distinctly higher incidence of the disease among coal, gold, and uranium miners, stone masons, and other workers exposed to high quantities of silica. The most characteristic clinical feature is acrosclerosis, which occurs in 95% of patients, and in one third of patients may involve more proximal parts of the limbs and face. Disease involving the proximal limbs and torso is classed as diffuse and is more likely to be associated with visceral pathology. Dermal sclerosis limited to digital extremities, now termed "limited disease," was once referred to as the CREST syndrome: calcinosis, Raynaud’s phenomenon, esophageal dysmotility, sclerodactyly, and telangiectasias. Visceral involvement is often subclinical until late in the disease course, at which point patients may experience nausea, vomiting, anorexia, constipation, diarrhea, and malabsorption; respiratory dysfunction caused by interstitial pulmonary fibrosis and pulmonary hypertension; cardiac conduction abnormalities and heart failure; and scleroderma renal crisis, which is characterized by abrupt onset of malignant, high-renin hypertension with rapidly progressive renal failure. Systemic necrotizing vasculitis develops in I % of patients, most often in those with limited disease (CREST syndrome) in conjunction with Sjogren’s syndrome.
The most common neurologic manifestation of scleroderma is myopathy, which occurs in 17% of patients. 27 In 80% of those affected, this is a relatively indolent disorder marked by muscle fibrosis without inflammation, minimal proximal weakness, and serum muscle enzyme elevations up to twice normal levels. In 20% of patients, the myopathy meets criteria for polymyositis. [1] In many of these patients, there is concurrent cardiac involvement and a substantial risk of congestive heart failure and sudden death. [2]
Other neurologic manifestations are rare. The most common of these are cranial nerve palsies, most involving the fifth (more than 200 cases reported), less often the seventh, ninth, eighth, fourth, and rarely the sixth, tenth, and twelfth cranial nerves. Isolated trigeminal sensory loss is reported in 4% of patients with scleroderma but is actually more common in patients with undifferentiated collagen vascular syndromes or mixed connective tissue disease (see below). It is most often bilateral, affects all sensory modalities, most often affects the V2 and V3 distributions (but rarely the muscles of mastication), and is frequently associated with pain, which may be throbbing, aching, scalding, burning, or lancinating. The inside of the mouth is frequently affected. The nature of all the various cranial neuropathies of scleroderma suggests peripheral neural involvement, but the trigeminal neuropathy is most likely to reflect gasserian ganglionitis.
Other neurologic manifestations of scleroderma include rare symptomatic peripheral neuropathy (carpal tunnel syndrome, sensorimotor polyneuropathy). Mononeuritis multiplex is reported in 1.3% of patients with the CREST syndrome. [3] Autonomic neuropathy is relatively common, and symptomatic orthostatic hypotension is reported in up to 9% of patients. [4] Disseminated cerebral arteritis and myelopathy are reported.
The diagnosis of scleroderma is strictly clinical. Laboratory abnormalities include modest elevation of the erythrocyte sedimentation rate (ESR), low titer rheumatoid factor, elevation of serum globulins, and antinuclear antibody, typically in a speckled or nucleolar pattern. A number of drugs show some value in palliating the disease, but treatment remains unsatisfactory. No drug is effective in treating the cranial neuropathies, although a question remains As to whether the duration of trials has been long enough. Treatment of the inflammatory myositis is similar to that of polymyositis. [5]
References
August 23, 2010 — Sjogren’s syndrome (SS) is characterized by two components of the triad of keratoconjunctivitis sicca, xerostomia, and another connective tissue diseases. It may affect 3% of the population, mainly women. When secondary (about 50% of cases), it is most often associated with RA. It is caused by lymphoid infiltrates of major and minor salivary glands and the lacrimal gland. In many cases, the lymphoid invasion generalizes to involve exocrine tissues throughout the body (with sequelae of serous otitis caused by eustachian dysfunction, recurrent bronchial infections, severe dryness of the genital mucosa, atrophic gastritis with achlorhydria, and pancreatic hyposecretion), and rarely, even nonexocrine tissues, including renal interstitium and muscle. Lymphoid infiltrates sometimes become massive ("pseudolymphoma"), and there is an increased risk of lymphoreticular neoplasia. Many patients have Raynaud’s phenomenon, and up to 50% experience arthralgias or arthritis, even in the absence of associated RA. Vasculitis is uncommon, and most consistently involves the skin, less often muscles and nerves, and is most likely to occur in the setting of SS with RA. It accounts for many of the most serious neurologic manifestations. As in RA, it is typically a chronic, smoldering process.
Neuromuscular complications may affect up to 10% of patients and generally resemble the extraspinal manifestations of RA, even in the absence of associated RA. A myopathy characterized by patchy, focal inflammatory infiltrates is common but rarely symptomatic, and polymyositis is rare. Several forms of neuropathy have been described, affecting up to 10% of patients, including a slowly progressive sensory polyneuropathy, usually minimally disabling sensorimotor polyneuropathy, and occasionally more rapidly progressive mononeuritis multiplex. A number of patients have been described with a sensory neuronopathy (resembling that described as a remote effect of neoplasia), which is caused by inflammation of selected dorsal root ganglia. [1,2,3] Over weeks to years, these patients develop progressive sensory loss, affecting all modalities but particularly vibratory and position senses, which ultimately becomes complete. It is patchy and often asymmetrical, at times involving only one side of the body. Pain and paresthesias may be severe, and patients may end up with profound and disabling sensory ataxia. Many patients exhibit severe autonomic impairment (presumably reflecting autonomic ganglionitis), with Adie’s pupils, orthostatic hypotension, and generalized anhydrosis. Motor function is typically spared.
The most common CNS disorder in SS is a trigeminal sensory neuropathy identical to that observed in scleroderma and MCTD, most likely caused by a gasserian ganglionitis, often associated with more widespread sensory neuronopathy, but also occurring in isolation. Most series suggest that other CNS manifestations are rare and very rarely related to SS. There are well-documented cases of transverse myelitis; seizures, stroke, encephalopathy, and myelopathy related to CNS vasculitis; and dementia, stupor, or coma related to meningoencephalitis. Alexander et al have reported the occurrence of a variety of CNS manifestations, including a multiple sclerosis-like syndrome, in 25% of their patients with primary SS; [4,5] however, in the 20 years since their reports first appeared, these findings have not been replicated.
Major laboratory manifestations of SS include a mild normochromic, normocytic anemia, an elevated ESR, polyclonal hypergammaglobulinemia, RF (52% of cases without RA, 99% of cases with RA), ANAs (70%), anti-Ro(SSA) (60%) and anti-La(SSB) (50%). Tests used to confirm the diagnosis included Schirmer’s test of lacrimal secretion, slit lamp examination for the characteristic punctate or filamentary keratitis, sialography, salivary gland scintiscanning, and minor salivary gland biopsy. Some of the neurologic features of SS have a significant differential diagnosis. Trigeminal neuropathy may be symptomatic of tumors at the base of the brain. The sensory neuronopathy of SS needs to be differentiated from the paraneoplastic variety most often associated with small cell lung carcinomas. The latter is more reliably relentless, progressive, typically subacute, more consistently marked by small fiber sensory loss, often associated with oculomotor and cerebellar involvement, and characterized by the presence of specific antiganglion antibodies (e.g., anti Hu) and by CSF lymphocytosis and increased protein. SS sensory neuronopathy is more often associated with autonomic dysfunction and an inflammatory infiltrate on sural nerve biopsy specimens. [6,7,8] Idiopathic sensory neuronopathies are commonly associated with antibodies to particular antigens, such as GDlb gangliosides and myelin-associated glycoprotein. [9]
Cyclophosphamide and prednisone are the treatments of choice for the vasculitis of SS. The treatment of neurologic manifestations produced by other mechanisms is unclear. The meningoencephalitis may respond well to corticosteroids. Treatment of the neuronopathies with a variety of agents, including cortico steroids and cytotoxic agents, has been discouraging, although it is possible that treatment has not been sufficiently prolonged to be successful, and that there is limited capacity for recovery once neuronopathy is sufficiently advanced to provoke treatment. Molina et al reported a favorable response to intravenous immunoglobulin in conjunction with corticosteroids. [9]
July 21, 2010 — Any hypothesis of pathogenesis of Idiopathic Intracranial Hypertension should explain the following observations of patients with the disorder:
1. High rate of occurrence in obese women during the childbearing years
2. Reduced conductance to CSF outflow [1]
3. Normal ventricular size; no hydrocephalus [2]
4. No histologic evidence of cerebral edema. [3]
Changes in cerebral hemodynamics (ie, increased cerebral blood volume and decreased cerebral blood flow) have been reported. [4] However, others have found no significant changes in these factors. [5] The most popular hypothesis is that Idiopathic Intracranial Hypertension is a syndrome of reduced CSF absorption. Decreased conductance to CSF outflow may be caused by dysfunction of the absorptive mechanism of the arachnoid granulations or possibly through the extracranial lymphatics. [6] This latter mechanism of an alternative route of drainage along extracranial and spinal nerve roots to the extracranial lymphatics, proposed by Miles Johnston and coworkers, [6] may be an important factor in the mechanism of Idiopathic Intracranial Hypertension, because this route may account for a substantial percentage of CSF absorption.
So, regardless of the outflow mechanism, if outflow resistance is increased then intracranial pressure must increase for CSF to be absorbed. Although interstitial and intracellular edema have been reported in brain biopsy specimens, [7] a study with current methods of analysis has concluded that the histologic features of the brain parenchyma are normal and the findings from the initial report are artifactual. [3]
July 19, 2010 — Internuclear ophthalmoplegia (INO) refers to disruption of rapid, coordinated, horizontal saccades by slowed or limited adduction. [3, 4, 5, 6] Conjugate adduction in normal horizontal saccades is facilitated by a subset of interneurons within the abducens nucleus. These fibers cross the midline to travel through the contralateral medial longitudinal fasciculus (MLF) from the pons to the medial rectus subnucleus of the ocular motor complex in the midbrain. The MLF is highly myelinated to support the rapid neural transmission necessary for abduction of one eye and adduction of the fellow eye to be nearly synchronous. [7] Even slight impairment of the transmission speeds through the MLF produces symptoms by compromising this synchronicity, causing ocular misalignment during horizontal saccades. Unlike other myelinated tracts in which slight impairment may produce no overt clinical deficits, the system for coordinated horizontal saccades is extremely sensitive to transmission speeds, making INO a frequent manifestation of MS.
Video 1. Internuclear ophthalmoplegia
The adduction deficit in INO may manifest as slowing during the horizontal duction (adduction lag, ultimately with a full excursion of the eye) or as incomplete adduction producing an incomitant exotropia (Fig. 1). Normally, a rapid horizontal saccade is produced by pulse discharges originating in the paramedian pontine reticular formation (PPRF). Eccentric gaze following the saccade is maintained by a step function based on inputs from the medial vestibular nucleus and the nucleus prepositus hypoglossus. The step function is derived from velocity information by the process of neural integration, and it serves to maintain gaze holding by overcoming the elastic forces of orbital tissues. Demyelination of the MLF may have greater impact upon the high frequency discharges necessary to produce the rapid saccadic pulse, whereas there may be sparing of the lower frequency discharges required for the step function that ultimately determines the range of adduction. [3] Alternatively, the range of adduction may be spared because of pathways apart from the MLF that possibly mediate a full adducting excursion. [8]
Figure 1. A 27-year-old woman developed horizontal diplopia and oscillopsia. Examination revealed bilateral INO, greater on left gaze. (A) Primary position (0.0s); (B) adduction lag of the right eye on a rapid left saccade (0.10s); (C) near-complete adduction of the right eye at the end of the saccade (0.20s). (D) Axial FLAIR MRI through the pons, revealing hyper-intensity of the MLF bilaterally (arrow). (Click to enlarge figure)
Despite deficient adduction during horizontal saccades, the normal function of the medial rectus in INO can typically be demonstrated by testing convergence of the eyes (Fig. 2). Convergence is mediated by separate inputs to the medial rectus subnucleus that are distinct from the inputs arriving via the MLF. The dissociation between limited adduction on horizontal saccades and spared adduction during convergence highlights the supranuclear nature of the adduction deficit in INO, with intact nuclear and infranuclear components of adduction. In some cases, however, dysfunction of the MLF is sufficiently rostral that the medial rectus subnucleus itself is impaired; therefore, impaired convergence will accompany impaired adduction on horizontal saccades. In this setting, referred to as the anterior INO of Cogan, the distinction is blurred between INO and partial third nerve palsy. [9]
Figure 2. Bilateral INO. A 47-year-old woman presented with horizontal diplopia. Examination revealed large-angle exotropia and bilateral INO. (A) Limited adduction of the left eye on right gaze. (B) Limited adduction of the right eye on left gaze. (C) Spared convergence of the eyes. (D) Axial FLAIR MRI revealed numerous areas of white matter hyper-intensity (for example, arrow). (E) However, no signal abnormality was detected in the region of the MLF in the pons or midbrain. (F) Improvement of INO after 2 months. Improved right eye adduction. (G) Full left gaze. (Click to enlarge figure)
Patients with unilateral INO typically do not have significant exotropia in primary gaze, likely because of intact convergence tone. In contrast, bilateral MLF lesions often cause exotropia (see Fig. 2). This clinical presentation is described as the WEBINO (wall-eyed bilateral INO) syndrome.
The misalignment produced by INO may cause a variety of ophthalmic symptoms, including visual blurring, diplopia, loss of stereopsis, and asthenopia (eye fatigue). [10] Normally there is cortical sensory suppression during saccades to eliminate blur from retinal slippage, but in INO visual blurring may occur because this mechanism fails to fully suppress inputs from the eye with slowed saccades. [3] Visual symptoms are often proportional to the degree of INO, and patients with mild INO may be essentially asymptomatic. Because use-related fatigue and Uhtoff’s phenomenon (worsening symptoms with elevated body temperature) are common in patients with MS, the symptoms caused by INO may fluctuate over the course of the day.
Video 2. Internuclear ophthalmoplegia
Many patients with INO have normal horizontal pursuit, optokinetic, and vestibulo-ocular responses. [11] These functions may be preserved because they are mediated by lower-frequency neural signals with transmission that is spared despite demyelination of the MLF, or because they are also mediated by alternate connections between the abducens and oculomotor nuclei.
INO is often associated with a dissociated horizontal nystagmus most prominent in the abducting eye (Fig. 4). The slow phase of nystagmus is opposite the direction of attempted gaze, with quick saccades in the direction of attempted gaze. The nystagmus has a unique slow phase with an exponentially decaying wave form. [3] With greater excursions, the amplitude or frequency of the nystagmus may increase. [3]
Several mechanisms may account for the abducting nystagmus in INO, and the explanations are not necessarily mutually exclusive. One possibility is that there is a central adaptive response to reduce visual blurring. To attempt to overcome adduction weakness, a compensatory increased saccadic pulse and step could occur (and would affect both eyes by Hering’s law of dual innervation). [3, 12] Although the adaptive response may improve the adduction of the paretic eye, it would disturb the abduction of the non-paretic eye in two ways. First, amplification of the pulse would lead to saccadic hypermetria; second, pulse-step mismatch would lead to slow post-saccadic drift with exponential decay. According to this account, the phenomenon of abducting nystagmus in unilateral INO is expected to be greatest if patients habitually fixate with the paretic eye, leading to higher demands for central adaptation. On the other hand, if patients fixate with the non-paretic eye, the abducting nystagmus may not be present. [3] In keeping with these predictions, Zee and colleagues [13] demonstrated that in some (but not all) patients with INO, prolonged patching (1–5days) of the paretic eye reduces the abducting nystagmus, whereas patching of the non-paretic eye increases it. On the other hand, temporary patching of one eye (or the performance of horizontal saccades in total darkness) does not mitigate the abducting nystagmus, suggesting that the nystagmus is generated not by online target-position error signals but by a stored, long-term adaptive mechanism. [3]
A central adaptive mechanism, however, does not fully account for abducting nystagmus, because not all patients demonstrate the predicted changes following patching. [13] An alternate explanation proposes the disruption of inhibitory fibers that travel in the MLF and are postulated to cross in the midbrain to arrive at the antagonist medial rectus of the contralateral eye. [14] By this account, impaired inhibition of these medial rectus motoneurons reduces the abducting step function in that eye and causes a slow movement back from the abducted position. A corrective abducting saccade follows, and the repeating cycle generates abducting nystagmus. This explanation, however, would predict hypometric abducting saccades, rather than the hypermetric abducting saccades commonly seen. [3]
Yet another explanation for abducting nystagmus in INO is that injury to additional structures outside the MLF may directly lead to an asymmetric gaze-holding disturbance, which would manifest with greater severity in the non-paretic eye. By this account, however, the abducting nystagmus would have a typical saw-tooth waveform (in which the slow component has constant velocity directly relating to insufficient gaze-holding mechanisms), rather than the exponentially decaying waveform that is seen. [3]
In severe INO, the affected eye may demonstrate abduction slowing in addition to adduction slowing. A potential explanation for reduced abduction velocity is that normal abduction depends upon appropriate inhibition of the antagonist medial rectus of the same eye, which could be compromised by an MLF lesion. [15] An alternative explanation, however, is that patients with INO and abduction slowing in fact have more extensive pontine lesions not limited to the MLF, but also potentially involving the abducens nucleus, fascicle, or other structures. [6, 16]
Impaired horizontal saccades are not the only manifestation of an MLF lesion. The MLF also contains fibers mediating many vertical eye movements (pursuit, vestibular, and otolithic pathways); corresponding impairments of vertical gaze therefore frequently accompany INO. [11, 17, 18] Impaired vertical pursuit may manifest as “staircasing” ductions interrupted by horizontal movements. Patients with bilateral INO may have marked impairment of vertical gaze holding, resulting in primary-position or gaze-evoked vertical nystagmus. In contrast to the exponentially decaying slow waveform of abducting nystagmus, vertical nystagmus in INO has a typical saw-tooth pattern caused by insufficient gaze-holding mechanisms. Impairment of the utricular pathways within the MLF may additionally lead to vertical misalignment of the eyes, in the form of skew deviation or the full ocular tilt reaction.
The precise measurement of eye movements, using methods, such as infrared oculography, allows highly accurate detection and quantification of INO. Furthermore, these methods serve as a gold standard by which the accuracy of the bedside examination can be assessed.19 Using this method, Frohman and colleagues found that severe INO was accurately detected by virtually all physician observers (regardless of level of training), but that milder INO was missed by many physicians other than trained neuro-ophthalmologists.
Various metrics from eye movement recordings have been used to quantify INO. These include the versional dysconjugacy index (VDI), which compares the peak velocities for abduction in one eye to adduction in the other. [20, 21] This measure has the benefit of cancelling intra- and interindividual variations of absolute saccade velocities (caused by fatigue, for example). The VDI has also been assessed by a Z score and histogram analysis, which is a statistical method to better distinguish normal and abnormal results. [22] VDI measures use velocity rather than final amplitude because the extent of final amplitude in INO is often normal. The first-pass amplitude, on the other hand, evaluates the ratio of abducting and adducting eye position at the time that the abducting eye has initially completed its saccade. [23] Finally, recent studies have employed a phase-plane analysis, which plots eye velocity directly as a function of position, removing the effects of temporal variation that arise, for example, from onset latency. [24] Quantified measures of INO may provide a useful way to index the clinical effects of fatigue and Uhtoff’s phenomenon, and ultimately may provide a method to objectively assess potential symptomatic treatments. [25]
Patients with INO frequently have a corresponding abnormality in the pons or midbrain that is detectable by MRI. [26] Frohman and colleagues studied 58 subjects with MS and INO and found that the sensitivity of proton density imaging, T2-weighted imaging, and fluid-attenuated inversion recovery (FLAIR) imaging was 100%, 88%, and 48%, respectively. It is not clear from this study, however, how the severity of INO relates to these MRI findings; cases of mild INO may have a higher rate of normal imaging. Furthermore, because patients with MS without INO were not included in this study, the exact specificity of these MRI abnormalities is not known. In some cases, MRI signal abnormality in this region may not have a clinical correlate. Another MRI measure that has been studied in INO is diffusion tensor imaging (DTI), in which the spatial constraints of water diffusion allow assessment of the integrity of white-matter tracts. [27] Fox and colleagues found a modest correlation between INO severity (graded by VDI) and mean white matter diffusivity in the MLF, showing that DTI measures may serve as a surrogate marker of brain-tissue integrity.
June 24, 2010 — The initial evaluation of elderly patients with seizures and epilepsy is similar to that of younger patients. [3] A thorough history should be obtained, eliciting information from family members or caretakers, whenever possible. Because multiple prescription drug use is common in elderly patients, a careful medication history should be elicited. Some medications may cause seizures [2], whereas others may exert a permissive effect by unmasking potentially epileptogenic foci in an elderly brain. Trauma should always be considered, even if a history of such is lacking. Elderly patients have an increased propensity for subdural hematomas and may not recall minor head trauma. A physical examination, serum electrolyte determination, electrocardiogram (ECG), neuroimaging study, and electroencephalogram (EEG) should be obtained in all patients. Additional studies are not routine and should be performed only as dictated by the results of the previous evaluations. These may include lumbar puncture, bacterial or viral cultures, toxicologic screens, or studies appropriate for cerebrovascular evaluation. Although the studies reviewed previously raise the issue of whether late-onset epilepsy is a signal of cerebrovascular disease, this question remains unanswered. Available data do not seem to justify routine evaluation for stroke (e.g., carotid Doppler studies, echocardiograms, procoagulant studies).
Several EEG changes are evident in old age. Brief runs of temporal 8- to 10-Hz activities and episodic temporal theta activities, often present only on the left, become increasingly common after 50 years of age. [1] Although some clinicians consider these findings normal age-related variants, others suggest that these patterns may be markers for undiagnosed cerebrovascular insufficiency. Kellaway [1] noted that in patients with cerebrovascular insufficiency, temporal slow activity developed at an earlier age and higher incidence than in age-matched controls. However, these patterns have not been studied adequately, and their clinical significance remains uncertain. There is no evidence that they are related to epilepsy or potentially epileptogenic tissue.
In a retrospective study of poststroke seizures, when the EEG showed periodic lateralized epileptiform discharges (PLEDS) within I week of stroke onset, seizures always developed. [4] Seizures were also more likely to develop if the EEG revealed sharp waves or other epileptiform discharges within 1 week of stroke onset. Nonepileptiform EEG patterns (normal, diffuse slowing, focal slowing) remained the most common findings among patients in whom poststroke seizures eventually developed.
Gupta and colleagues [18] found that 10% of patients with early-onset and 12% with late-onset poststroke seizures had epileptiform discharges or PLEDs on EEG during a first or subsequent hospitalization. Another study found that of those patients with poststroke seizures who had EEGS, 11% showed epileptiform discharges or PLEDs. [6] In Dam and colleagues review of late-onset epilepsy [5], 8 of 31 patients (26%) with epilepsy secondary to stroke were noted to have paroxysmal features (not further defined) on an EEG study. Clinical history continues to form the cornerstone of diagnosis in the group of patients with epilepsy secondary to stroke.
EEG changes in epilepsy in the elderly
PROGNOSIS IN ELDERLY PATIENTS WITH SEIZURES AND EPILEPSY
No prospective studies have specifically addressed the question of recurrence risk following a first unprovoked seizure in patients older than 60 years. Luhdorf and colleagues [7] retrospectively studied the prognosis of seizures and epilepsy in a population of patients aged 60 years or more who were admitted to a hospital. The study group included 151 patients with new-onset seizures and 88 patients with established epilepsy. At least I month, but generally more than 12 months of follow-up observation were available for analysis. Seventy-five percent of patients with new-onset seizures had at least one recurrence and were treated with antiepileptic drugs. Of those previously untreated patients observed for at least 12 months, 62% became seizure-free and 26% had fewer than three seizures per year. Patients with seizure recurrence in the first year following initial admission tended to have more seizures overall in comparison with patients who were seizure-free in the first year. Only 47% of patients with established epilepsy remained seizure-free in the study period. The results of this retrospective hospital-based study were subject to multiple sources of bias and, thus, cannot be used to draw firm conclusions regarding the prognosis for seizure recurrence or epilepsy control in the elderly. In a recent review of elderly (aged 75 years or more) epileptic patients observed in the neurology clinics of Columbia-Presbyterian Medical Center, 45% averaged zero seizures per year, 40% sustained 1 to 10 seizures per year, and 15% had more than 10 seizures per year.
Few studies have evaluated the effects of seizures and epilepsy on activities of daily living or longevity in the elderly. Analysis of data obtained from the study population just described led Luhdorf and colleagues [7] to conclude that there is an increased mortality rate in patients older than 60 years with seizures and epilepsy. This was true for several subgroups: (1) those with brain tumors or strokes, (2) those with established epilepsy (a population described as institutional patients with chronic epilepsy ), and (3) those not previously treated with antiepileptic drugs (primarily patients with new-onset seizures). In contrast, the number of deaths in patients with seizures of unknown cause did not exceed that in the population norm. This suggests that the increased risk of death in patients with late-onset seizures is primarily related to seizure etiology. Controlled studies are needed to evaluate the effect of seizures and epilepsy on morbidity and mortality in the elderly.
