Other recent neurologic advances include new drug therapy for epilepsy, treatment of multiple sclerosis, the high prevalence of parkinsonism in elderly persons, further support for the belief that fear can cause death, a new understanding of hypertensive encephalopathy, and the recognition that the restless legs syndrome is a common and often unrecognized problem.

Stroke

Three studies have given new impetus to efforts to decrease the rate of death and disability caused by stroke. The first study addresses prevention; the second, treatment; and the third, new techniques for imaging the brain after a stroke.

Surgery Was Superior in Asymptomatic Patients with High-Grade Carotid Stenosis

Endarterectomy for asymptomatic carotid artery stenosis. Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. JAMA. 1996; 273:1421-8.

Many strokes are caused by a high-grade (80% to 90%) stenosis located in the portion of the internal carotid artery that is directly adjacent to the common carotid artery. Patients who have stenosis and such symptoms as transient ischemic attacks fare better with carotid endarterectomy (if the procedure is done by surgeons whose work is associated with low complication rates) than with medical therapy [1, 2]. The Asymptomatic Carotid Atherosclerosis Study showed that surgery also appears to be superior in asymptomatic patients, at least in those who have a high-grade stenosis of the internal carotid artery.

As the data from this study are reviewed, however, readers should keep in mind the clinical questions that pertain to carotid artery stenosis. Patients need answers to these questions.

First, what are the characteristics of patients with such lesions? These patients are about 70 years of age, and they almost certainly have coronary artery disease. Many have had coronary artery bypass grafting and still have hypertension, diabetes, and peripheral artery disease. Therefore, stroke does not necessarily present the greatest risk for death that these patients face.

Second, will a patient's high-grade stenosis become occluded? The answer depends primarily on the severity of the other diseases. The stenosis will become occluded, apparently, if the patient lives long enough.

Third, if the artery becomes occluded, will the patient have a stroke? The answer is unclear, but autopsy studies show that about 50% of patients found to have complete carotid artery occlusions have no symptoms, presumably because the brain is protected by collateral circulation.

Fourth, if a stroke does occur, will it be severe? This is the critical question to an elderly patient, who may fear disability more than death. The simple, honest answer is that we don't know.

Finally, should an asymptomatic patient with a high-grade stenosis of the internal carotid artery be referred to a surgeon for carotid endarterectomy? To answer this question, the Asymptomatic Carotid Atherosclerosis Study examined whether adding carotid endarterectomy to aggressive medical management could reduce the incidence of stroke in patients with asymptomatic carotid artery stenosis.

The study was a randomized, controlled trial done at 39 centers. Between 1987 and 1993, 1662 patients with stenosis greater than 60% were randomly assigned to receive either medical therapy (325 mg of aspirin daily plus counseling about risk factors) or medical therapy plus carotid endarterectomy. Outcome measures were stroke in the area of distribution of the artery or any stroke or death that occurred within the 30-day perioperative period.

The median follow-up of 2.7 years yielded 4657 patient-years of observation. On the basis of these data, the researchers estimated that the 5-year aggregate risk for ipsilateral stroke and any perioperative stroke or death was 5.1% in patients in the surgery group and 11.0% in patients in the medical therapy group (relative risk reduction, 53% [95% CI, 22% to 72%]; number of patients needed to treat with surgery to prevent one ipsilateral stroke, 17) (Figure 1).

View larger version (27K): [in this window] [in a new window]   Figure 1. Proportion of patients without end point at a given time during follow-up, by treatment group, using Kaplan-Meier estimation method. TIA = transient ischemic attack; solid line = patients in the medical therapy group; dashed line = patients in the surgery group. Reproduced with permission from JAMA. 1995; 273:1421-8.

The clinical conclusion of this study is that surgery is generally preferable if the surgeon's work is associated with a rate of serious complications of less than 3%. However, only the best surgeons can claim such low rates, and it is therefore imperative that the primary physician know who these surgeons are. Moreover, because of the other medical conditions that these patients have, the overall outcomes are not as impressive as the study outcomes. At 5 years, 25.5% of patients in the medical therapy group would be expected to have had any stroke or to have died; the proportion of patients in the surgery group would be 20.7% (risk reduction, 19% [CI, –8% to 39%]) (Figure 1).

Low-Dose Tissue Plasminogen Activator Improved Survival and Reduced Disability

Tissue plasminogen activator for acute ischemic stroke. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. N Engl J Med. 1995; 333:1581-7.

Strokes caused by carotid artery stenosis occur in 400 000 Americans each year. Until now, the only treatment for these patients has been heparin, which has had, at best, mixed results in halting the progression of strokes. It was initially thought that thrombolytic therapy would be as efficacious for stroke as it was for acute myocardial infarction, but cerebral hemorrhage was soon identified as a major complication of the relatively high-dose thrombolysis that was given to patients with heart disease. However, results of preliminary studies of lower doses of tissue plasminogen activator (t-PA) given carefully to patients who had had a stroke were more promising. The randomized trial by the National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group sought to determine whether lower-dose t-PA given within 3 hours of symptom onset reduces the extent of stroke and the severity of symptoms.

Part 1 of the study, which enrolled 291 patients who had had a stroke, examined whether patients who received t-PA were clinically improved compared with patients who received placebo. The end points were an improvement of 4 points over baseline in the score of the National Institutes of Health Stroke Scale [3] (a 42-point scale that classifies deficits into 11 categories) or resolution of the neurologic deficit within 24 hours. Part 2 of the study which enrolled 333 patients, assessed the clinical outcome 3 months after stroke occurred by using four validated scales (including the National Institutes of Health Stroke Scale) (Figure 2).

View larger version (46K): [in this window] [in a new window]   Figure 2. Outcome at 3 months in part 2 of the study, according to treatment. A favorable outcome was indicated by scores of 1 or less on the National Institutes of Health Stroke Scale (NIHSS), 95 or 100 on the Barthel index, 1 or less on the modified Rankin scale, and 1 on the Glasgow outcome scale. Values do not total 100% because of rounding. t-PA = tissue plasminogen activator. Reproduced with permission from N Engl J Med. 1995; 333:1581-7.

Outcomes at 24 hours did not differ between groups. However, at 3 months, patients had improvement on all scales (odds ratio, 1.7 [CI, 1.2 to 2.6.]) Patients who received t-PA were 30% more likely to have minimal or no disability at 3 months. Symptomatic intercerebral hemorrhage occurred within 36 hours after stroke onset in 6.4% of t-PA recipients and in 0.6% of placebo recipients. However, mortality at 3 months was 17% in the t-PA group and 21% in the placebo group.

This study confirms that t-PA has a role in treating stroke. It is a dangerous drug, however, and must be given carefully. First, it must be administered soon after a stroke occurs. All patients in this study received t-PA within 3 hours of symptom onset, and all had computed tomography to rule out hemorrhagic stroke. Patients were excluded from this study if they had had serious head trauma in the 3 months before the stroke occurred; if they had had surgery in the 14 days before the stroke occurred; if they were receiving anticoagulant therapy; or if they had a history of cerebral, gastrointestinal, or urinary tract bleeding. Second, patients should have relatively normal blood pressure. Patients with systolic blood pressure greater than 185 mm Hg or diastolic blood pressure greater than 110 mm Hg were excluded. Third, the t-PA dose must be low: 0.9 mg/kg of body weight to a maximum of 90 mg/kg. Cardiac doses of t-PA are usually 100 mg for all patients. In summary, t-PA may be efficacious, but most clinicians and hospitals probably are not yet prepared to administer it routinely.

Multiple Imaging Studies Often Added Nothing to Stroke Diagnosis

Salerno SM, Landry FJ, Schick JD, Schoomaker EB. The effect of multiple neuroimaging studies on classification, treatment, and outcome of acute ischemic stroke. Ann Intern Med. 1996; 124(1 pt 1):21-6.

Several new methods for imaging the brain have recently come into clinical use: magnetic resonance angiography, spiral computed tomography, diffusion-weighted imaging, and single-photon emission computed tomography. In most institutions, however, the mainstays of brain imaging remain computed tomography and magnetic resonance imaging. These studies are certainly the most commonly used in patients who have symptoms of stroke.

In many institutions, patients suspected of having had a stroke receive two imaging studies; the second one is done 24 to 48 hours after the first. Physicians who endorse this practice cite several factors in support of it. First, although computed tomographic scans are usually easy to obtain and are excellent at ruling out hemorrhagic strokes, they are insensitive for the detection of early ischemic strokes. Computed tomography is more sensitive when done 1 or 2 days after the stroke occurs; thus, a follow-up scan is usually obtained to confirm the ischemic stroke. Some data also suggest that magnetic resonance imaging is superior for detecting lacunar strokes, multiple lesions, and posterior fossa lesions. Salerno and colleagues examined the effect of serial neuroimaging studies on the therapy and outcome of patients with a diagnosis of acute stroke.

The study was a retrospective case series of 206 adult patients (mean age, 66 years) who were hospitalized at a tertiary care teaching hospital between 1990 and 1993. The researchers classified the strokes that these patients had into one of five categories (large-vessel, small-vessel, cardioembolic, other, or unknown cause) using criteria based on results of the history, physical examination, ancillary tests, and computed tomography or magnetic resonance imaging. Strokes were reclassified after the results of additional neuroimaging studies were reviewed. The major end points were type of therapy and patient outcomes.

The additional imaging studies changed the classification of stroke in 20% of the 140 patients who had at least two studies (Table 1). All changes were from the unknown cause category to another category. In 93.2% of patients, therapy was initiated before a second study was done. Of the patients who had one imaging study, 70.1% went home, 24.0% went to a nursing home, and 5.9% died. Of the patients who had more than one study, 73.3% went home, 24.4% went to nursing homes, and 2.2% died.

Salerno and colleagues concluded that in cases in which a stroke can be classified using clinical criteria and a single imaging study, a second imaging study is not needed. The second study appears to help only when the cause of the stroke symptoms cannot be determined by a history, physical examination, and one imaging study. In this study, cause could not be established without the second study in about half of the patients. Even after additional studies were done, the eventual outcome appeared to be unchanged.

Summary

It is clear that the management of patients who have had stroke has become much more aggressive. A passive attitude toward prevention and treatment is no longer acceptable. Patients at risk deserve to be identified and referred if necessary, and many of those who have had a stroke can benefit from more aggressive therapy than the routinely used computed tomography and anticoagulation therapy.

Other Neurologic Disorders

Monotherapy Is the Goal of Epilepsy Treatment

Brodie MJ, Dichter MA. Antiepileptic drugs. N Engl J Med. 1996; 334:168-75.

The treatment of epilepsy has recently received much attention. New agents are available, but some have serious side effects. Rather than focus on the new, internists should remain up to date on some of the older treatments and the ways in which their use may have changed in recent years.

About 0.4% to 0.8% of the U.S. population has seizures. The incidence of new-onset seizures is highest in the very young and the very old. In about 70% of persons who have seizures, the seizures have no identifiable cause. Drug therapy is usually initiated after a patient has had more than one unprovoked seizure in 1 year. In general, an adult who has one seizure faces a 30% to 70% chance of having recurrent episodes.

In their review of the treatment of seizures, Brodie and Dichter correctly emphasize that the goal of treatment should be "the restoration of a normal life through complete control of seizures with the use of a single drug that has no side effects." Monotherapy is the preferred route; although therapy with several agents may control seizures, it results in a greater number of adverse effects, such as drowsiness, hirsutism, gingival hypertrophy, impotence, and, ultimately, noncompliance. When monotherapy fails, neurologic consultation is probably the best next step.

Because the choice of drug depends on the type of seizure, proper classification is imperative. The classification scheme of the International League against Epilepsy is hundreds of words long, but seizures generally can be classified into the categories shown in Figure 3.

View larger version (19K): [in this window] [in a new window]   Figure 3. Categories of seizures and general rank of drugs of choice.

In general, tonic-clonic seizures are treated first with valproic acid and then with carbamazepine and phenytoin. This strategy may seem to conflict with the traditional approach, but the adverse effects of valproic acid are less severe than those of other agents. The side effects of valproic acid are usually dose related and consist of tremor, appetite stimulation, hair loss, and irregular menstrual cycles.

In patients who have partial seizures, carbamazepine is the drug of choice, followed by phenytoin and valproic acid. Common dose-related side effects of these three drugs are diplopia, headache, dizziness, and nausea. Unusual effects include a morbilliform rash, which occurs in about 10% of patients, and more severe rashes, including the Stevens-Johnson syndrome.

Ethosuximide is used to treat seizures because of its superior effect against absence seizures. Phenobarbital and primidone (which is metabolized to phenobarbital) are rarely used now because of their negative effects on cognition, mood, and behavior.

The effect of therapy is judged primarily by the amelioration of seizures and the drug's adverse effects. The use of serum drug levels to monitor therapy is important for two reasons. First, these levels are a fair indicator of proper dosing when therapy is being initiated. Second, using these levels is sometimes an effective way to monitor compliance. If seizures recur and serum levels are very low, physicians can directly discuss the issue of compliance with the patient.

Finally, Brodie and Dichter discuss the consensus opinion on when and how to discontinue therapy. Drug therapy can eventually be discontinued in about 60% of patients. Seizure experts usually wait until 2 to 5 seizure-free years have passed before beginning to discontinue therapy and then taper the drug dose during a period of 2 to 6 months.

Although patients who have uncomplicated seizures can be well managed by primary care physicians and generalist neurologists, a few need to be referred to specialized seizure centers. Most of these patients cannot tolerate the common drugs or have seizures that are not controlled with a single drug. Seizure centers can provide expertise in several ways. They can ensure that the seizure is properly classified, often through the use of video electroencephalography. Not all seizures fit easily into the classification system presented in Figure 3, and many patients actually have pseudoseizures, which an untrained witness may not recognize. Epilepsy centers may also be able to simplify polypharmacy; prescribe newer, less-used agents; and assess the possible usefulness of surgical treatment.

Interferon for Multiple Sclerosis Has Limited Usefulness

Lublin FD, Whitaker JN, Eidelman BH, Miller AE, Arnason BG, Burks JS. Management of patients receiving interferon ß-1b for multiple sclerosis: report of a consensus conference. Neurology. 1996; 46:12-8.

In 1993, the Food and Drug Administration approved the use of interferon-ß 1b for treating ambulatory patients with relapsing-remitting multiple sclerosis. This approval came after publication of results of a clinical trial that showed that interferon-ß 1b reduced the exacerbation rate by 31% [4]. The study sample consisted of ambulatory patients (18 to 50 years of age) who had relapsing-remitting multiple sclerosis and who had had at least two acute exacerbations of the disease in a 2-year period. A companion article [5] showed that the results of magnetic resonance imaging improved in patients who received interferon-ß 1b.

Soon after these articles were published, a subcommittee of the American Academy of Neurology stated that interferon-ß 1b should also be used in older patients, patients who have relapsing-progressing disease, and patients who are no longer ambulatory.

The consensus conference (the proceedings of which are summarized by Lublin and colleagues) was held in Washington, D.C., in 1994. The conference panelists reaffirmed the role of interferon-ß 1b as approved by the Food and Drug Administration but were more conservative about using the drug for other patients. Citing the available data, the panelists pointed out that higher doses may be needed in sicker or older patients. Patients must be made aware that increased spasticity commonly occurs with interferon-ß 1b. Finally, the panelists agreed that patients with progressive or very severe multiple sclerosis should receive treatment through research protocols.

In conclusion, interferon-ß 1b does have a place in clinical practice, but its use should be limited to relatively young, ambulatory patients who have relapsing-remitting disease. More data are needed to determine whether the drug will be beneficial in patients who do not have these characteristics. Referral to research centers is therefore critical. (Since this Update was presented at the American College of Physicians' 1996 Annual Session, the Food and Drug Administration has approved interferon-ß 1a for use in multiple sclerosis. The indications for this drug are the same as those for interferon-ß 1b, but interferon-ß 1a is more convenient because it can be administered intramuscularly once a week.)

Parkinsonism Is Common and Is Associated with Death

Bennett DA, Beckett LA, Murray AM, Shannon KM, Goetz CG, Pilgrim DM, et al. Prevalence of parkinsonian signs and associated mortality in a community population of older people. N Engl J Med. 1996; 334:71-6.

Signs of parkinsonism are often seen in elderly persons visiting primary care physicians. Until now, the actual prevalence of these signs has not been measured. Further, the implications of parkinsonian signs have often been minimized because their association with more severe morbidity and mortality has not been studied.

Bennett and colleagues studied a stratified random sample of 467 residents of East Boston who were at least 65 years of age. Examiners were trained to recognize four categories of parkinsonian signs: bradykinesia, gait disturbance, rigidity, and tremor. The examiners then administered structured neurologic examinations. Actual "idiopathic" Parkinson disease was not studied because it could not be distinguished from other causes of parkinsonian symptoms.

Of the persons examined, 159 (34%) had parkinsonism. Seven persons could not be classified, and the remaining 301 had no signs. The estimated prevalences of parkinsonism were 14.9% in persons 65 to 74 years of age, 29.5% in those 75 to 84 years of age, and 52.4% in those 85 years of age and older. After adjustment for age and sex, the overall risk for death was 2.0 (CI, 1.6 to 2.6) in persons with parkinsonian signs compared with persons with no signs. Gait disturbance was most highly associated with a higher risk for death.

The results of this study should be a wake-up call for clinicians to pay attention to early signs of parkinsonism. Clinicians should try to alleviate these signs by reviewing lists of medications and minimizing the risks for falls (from gait disturbances) and pulmonary emboli (from the sedentary lifestyle caused by bradykinesia). Drug therapy used to alleviate the signs themselves should be initiated with great care. As with most drug therapy in the elderly, dosing regimens should be started slowly and kept low. If Bennett and colleagues' data prove valid, instituting therapy only after full-blown Parkinson disease evolves may hasten a patient's death.

Deaths Occurring during an Earthquake Supported Possibility of "Voodoo Death"

Leor J, Poole WK, Kloner RA. Sudden cardiac death triggered by an earthquake. N Engl J Med. 1996; 334:413-9.

The belief that fear can cause death—so-called "voodoo death"—is ancient. Proving under controlled conditions that such a cause exists is nearly impossible, not to mention unethical. However, natural events can allow study of voodoo death. Leor and colleagues took advantage of one such natural event.

On 17 January 1994, one of the strongest earthquakes ever recorded in a major North American city occurred in the Los Angeles area. The earthquake was named the Northridge earthquake because its epicenter was in nearby Northridge, California. After the earthquake, Leor and coworkers reviewed the records of the Los Angeles County coroner's office to count the number of deaths that occurred in the week before the earthquake, on the day of the event, and during the 6 days afterwards. They repeated this process for the same periods in 1991 through 1993.

In the week before the earthquake, a mean (±SD) of 4.6 ± 2.1 deaths from atherosclerotic cardiovascular disease occurred each day. On the day of the earthquake, 24 such deaths were recorded. Sixteen of these deaths occurred within 1 hour after the first tremor occurred. Three of these deaths were related to physical exertion. During the 6 days after the earthquake, the death rate decreased to 2.7 ± 1.2 deaths per day.

The Northridge earthquake appeared to be a trigger for sudden cardiac death unrelated to physical exertion. This conclusion can easily be carried into a discussion of teleologic and neurologic issues. The sympathetic storm created by major stress was probably a valuable survival reflex in prehistoric humans. If a potentially lethal danger was present, a huge catecholamine flux would provide the fight-or-flight response. One or two persons might die suddenly because of this autonomic storm, but most of the inhabitants of a village might be able to flee. A reflex so valuable in ancient times may be a worrisome cause of death in the modern world.

Hypertensive Encephalopathy May Not Be Rare

Hinchey J, Chaves C, Appignani B, Breen J, Pao L, Wang A, et al. A reversible posterior leukoencephalopathy syndrome. N Engl J Med. 1996; 334:494-500.

Most physicians were taught to believe that hypertensive encephalopathy is seen only in patients with malignant hypertension. However, new data, combined with knowledge about cerebral blood flow, suggest that this syndrome may be much more common than previously believed. It may also offer special insight into toxemia in pregnant women.

Hinchey and colleagues noted that some patients admitted to their hospitals in Paris and Boston for acute illness had a reversible syndrome of headaches, altered mental status, seizures, and loss of vision associated with findings of encephalopathy of the posterior portion of the white matter of the brain. The researchers sought to identify other patients with this syndrome by reviewing the computed tomography and magnetic resonance imaging logs of their hospitals. The logs indicated that between 1988 and 1994, 15 patients had the symptoms described above.

Seven of these patients were receiving immunosuppressive therapy after transplantation or for aplastic anemia when the acute syndrome developed. One patient was receiving interferon for melanoma, three had eclampsia, and four had acute hypertensive encephalopathy associated with renal disease. The clinical findings included headaches, vomiting, confusion, seizures, and cortical blindness. With antihypertensive treatment, all symptoms resolved and results of imaging studies returned to normal.

All 7 patients were receiving immunosuppression therapy or had hypertension, but the hypertension was not necessarily "malignant." It has been established that in most patients, cerebral blood flow is fairly constant over a mean arterial blood pressure of about 50 to 150 mm Hg (Figure 4). These limits can shift, however, as is seen in neonates, whose cerebral blood flow decreases when mean arterial pressure is in the so-called "normal" range. Moreover, magnesium seems to shift the relation between blood pressure and cerebral blood flow back toward normal. Thus, the reason magnesium sulfate works so well to reverse the encephalopathy of eclampsia can be understood on a physiologic level.

View larger version (8K): [in this window] [in a new window]   Figure 4. The association between mean arterial blood pressure and cerebral blood flow. Cerebral flow usually reaches a plateau at a mean arterial blood pressure of 50 to 150 mm Hg, but the association is not fixed.

The Restless Legs Syndrome: An Often-Missed Diagnosis

Walters AS, Hickey K, Maltzman J, Verrico T, Joseph D, Hening W, et al. A questionnaire study of 138 patients with restless legs syndrome: the ‘Night-Walkers’ survey. Neurology. 1996; 46:92-5.

The restless legs syndrome is characterized by three features: 1) Patients need to move their legs because of discomfort that is often described as a creeping, crawling, tingling, or painful sensation; 2) patients seek relief by walking, turning in bed, rubbing the legs, or stretching and flexing the legs; and 3) the leg discomfort is worse at rest and at night. The syndrome is common; prevalence studies show that it affects 10% to 15% of the U.S. population. The course of the syndrome has been unknown until now, although the syndrome has been assumed to be a condition of middle or older age.

Walters and colleagues verified the diagnosis of the restless legs syndrome in 105 persons who belonged to a nationwide support group for the syndrome. They then administered telephone surveys to each patient. The answers were compared with those of 33 patients with the syndrome who had been identified in a neurology center.

In both groups, symptoms had developed before age 20 years in approximately one third of the patients. For the younger group, symptoms were typically labeled "psychogenic" by physicians, and diagnoses ranged from growing pains to attention deficit disorder. Older patients commonly received such diagnoses as skin irritation or arthritis. More than 50% of patients had at least one first-degree relative with similar symptoms.

In the original description of the syndrome [6], Ekbom noted that some cases were associated with iron deficiency anemia. In other patients, the problem seems to be related to electrolyte disturbances (particularly hypomagnesemia) or renal disease. Once an underlying cause has been excluded, the symptoms may be effectively treated with a benzodiazepine before bedtime or with levodopa—carbidopa. Valproic acid may also be effective.

Dr. Roberts (Series Editor): York Health System, York, PA 17403.