Migraine Headache

Patients with migraine headaches often have difficulty explaining their suffering. In the physician's office, the typical patient places the heel of the hand in one orbit and the fingers up over the forehead, all in the first division of the fifth cranial nerve. Migraine pain is actually pain that is referred into the first division of the trigeminal nerve, and it is probably due to an abnormality in the way in which the central nervous system processes nociceptive information.

Promotility Agent Was Effective in Treatment

Tfelt-Hansen P, Henry P, Mulder LJ, et al. The effectiveness of combined oral lysine acetylsalicylate and metoclopramide compared with oral sumatriptan for migraine. Lancet. 1995; 346:923-6.

Sumatriptan has become a commonly used agent for migraine headaches. It is expensive, however, and patients often delay taking a drug that costs as much as $35 a dose. The injectable form is difficult for some patients to take, and the nasal form is poorly absorbed and leaves a bitter aftertaste. The oral drug is poorly absorbed because migraine attacks also cause gastric stasis and therefore decrease absorption of all oral drugs.

In this clinical trial, Tfelt-Hansen and colleagues assessed the effectiveness of salicylates (the class of drug most commonly used for migraine) plus a promotility agent. They randomly assigned 421 patients with migraine to receive metoclopramide, 10 mg, followed by lysine acetylsalicylate (equivalent to 900 mg of aspirin); oral sumatriptan (100 mg); or placebo. In 57% of patients receiving the combined regimen, 53% of those receiving sumatriptan, and 24% of those receiving placebo, the intensity of headache decreased from severe to mild or absent.

This study makes a critical point about the treatment of migraine: Simple oral medication alone is relatively ineffective. A promotility agent plus aspirin performed at least as well as the most expensive drug available. Another effective drug is indomethacin, which is given by suppository and crosses the blood-brain barrier. This is an effective and commonly used agent in Europe but has not become popular in the United States [1]. Indomethacin suppositories contain 50 mg of the drug. Because many headaches recur, I instruct patients to cut a 50-mg indomethacin suppository into thirds and insert them every 10 to 15 minutes as necessary. For patients who decline to use suppositories, a basic oral analgesic following a dose of metoclopramide is a reasonable second choice that is much less expensive than most alternatives. Two new drugs, zolmitriptan and naratriptan, are now available, but they offer little advantage over previously available medications because they are available only in oral preparations.

Divalproex Sodium Seemed To Be Effective Prophylaxis

Silberstein SD. Divalproex sodium in headache: literature review and clinical guidelines. Headache. 1996; 36:547-55.

For patients with severe or frequent migraine attacks, propranolol and amitriptyline are frequently used as prophylaxis. Many other agents have been shown to decrease the frequency and intensity of migraine attacks (Table 1).

In this nonsystematic review, Silberstein cited all clinical studies of divalproex sodium (sodium valproate plus valproic acid) used as prophylaxis against migraine. Valproic acid, a standard antiseizure medication, is presumed to exert its effects by facilitating the action of the neurotransmitter -aminobutyric acid (GABA). In reviewing four small, randomized clinical trials, Silberstein found that divalproex sodium greatly improved the headache indices in 75% to 87% of patients. In general, headache frequency decreased by about 50% in patients receiving divalproex sodium and by about 10% to 15% in patients receiving placebo. Divalproex sodium seemed to be slightly more effective than propranolol.

Although the studies were small and larger ones are needed, valproic acid, generally prescribed as divalproex sodium, seems to be a good choice for migraine prophylaxis. It appears to be at least as effective as and better tolerated than ß-blockers or tricyclic antidepressants. However, valproic acid may cause weight gain and hair loss and should not be used by women who may become pregnant. This contraindication could be a serious clinical problem because many patients with migraine are young women.

Stroke

Stroke remains a leading cause of death and disability in the United States. Recent advances in treatment include thrombolytic therapy given within 3 hours after onset of ischemic stroke. Early studies also indicate that blockage of glutamate, an excitatory neurotransmitter, reduces the sensitivity of central neurons to ischemia; however, no glutamate-blocking drugs are clinically available. These therapeutic advances have reinvigorated clinical investigation of stroke.

Transesophageal Echocardiography Outperformed Transthoracic Echocardiography

Rauh R, Fischereder M, Spengel FA. Transesophageal echocardiography in patients with focal cerebral ischemia of unknown cause. Stroke. 1996; 27:691-4.

The onset of stroke due to occlusive vascular disease, or thrombosis, is stepwise or progressive ("stuttering" onset). This type of stroke often occurs while the patient is sleeping. Embolic strokes usually are sudden in onset and produce maximum neurologic deficits at the outset. A patient presenting with stroke usually undergoes computed tomography to rule out bleeding and undergoes carotid Doppler studies to detect severe stenoses, which can be repaired surgically. However, when patients have neither a hemorrhagic stroke nor evidence of carotid stenosis, the rest of the diagnostic evaluation is controversial.

Rauh and colleagues assessed the usefulness of transesophageal echocardiography in patients with cerebral ischemia of unknown cause. Thirty consecutive patients who had completed strokes were enrolled. All patients were in sinus rhythm, had no cardiac disease known to be linked to the occurrence of cardiogenic emboli, had minimal carotid artery stenosis, and were evaluated by transthoracic and transesophageal echocardiography.

Transthoracic echocardiography yielded abnormal findings in 16 patients, but no embolic source was found. Transesophageal echocardiography identified thrombus in the left atrial appendage in 3 patients, atrial septal aneurysm in 2, and patent foramen ovale in 7. It also disclosed aortic plaques in 19 patients. The use of transesophageal echocardiography prompted a change of therapy in 3 of the 30 patients.

Transesophageal echocardiography is thus the test of choice for evaluating patients suspected of having embolic strokes. Compared with dual-helical computed tomography, an expensive test that is not always available, transesophageal echocardiography has a sensitivity of 87% and specificity of 82%. Thus, it is 3 to 6 times more accurate than transthoracic echocardiography. It is a safe test; only 1 person died among 10 000 tests, and this patient had undiagnosed esophageal cancer. Because of its ability to detect patent foramen ovale, the technique also precludes the need for venography or other studies to look for deep venous thrombi.

Many patients do not require transthoracic or transesophageal echocardiography. Patients who will receive anticoagulants (such as those with atrial fibrillation or dilated cardiomyopathy) do not need this study. Similarly, if a decision has been made not to use anticoagulation, no study is necessary. As with all tests, transesophageal echocardiography should be used only in patients whose therapy may be altered as a result of the test.

Evidence Linked Alzheimer Disease to Hemorrhagic Stroke

Greenberg SM, Briggs ME, Hyman BT, et al. Apolipoprotein E 4 is associated with the presence and earlier onset of hemorrhage in cerebral amyloid angiopathy. Stroke. 1996; 27:1333-7.

Intracerebral hemorrhage has historically resulted from hypertension. In such strokes, the source of bleeding is small vessels in the putamen and other areas in the deep brain. Despite aggressive treatment of hypertension, however, cerebral hemorrhage has remained nearly as common as in the past. The reason is the increased frequency of cortical hemorrhages. This type of hemorrhage may result from extension of a hypertensive hemorrhage, trauma, hemorrhagic transformation of an ischemic infarction, hemorrhage of a tumor, or cerebral vascular malformation.

A growing cause of hemorrhage in the cortex or gray or white matter is amyloid angiopathy, which leads to small and sometimes multiple hemorrhages. Risk factors include old age or a personal or family history of Alzheimer disease. Because apolipoprotein E 4 is associated with a family risk for Alzheimer disease, Greenberg and colleagues sought to determine whether this allele is associated with amyloid angiopathy. They identified 27 patients who presented with cerebral lobar hemorrhage and had a high probability of amyloid angiopathy. These patients were compared with 1899 elderly patients from a population-based sample and 18 patients with deep cerebral hemorrhage that was characteristic of a hypertensive mechanism.

Patients with multiple lobar hemorrhage had a twofold increase in frequency of the apolipoprotein E 4 allele compared with the general population sample. Patients who carried the 4 allele and sustained lobar hemorrhages were more than 5 years younger than those without the allele who sustained lobar hemorrhages.

The clinical message from this study is not completely clear, but the results should cause clinicians to at least consider age and Alzheimer disease as risk factors for adverse effects of warfarin. However, for most patients who are at risk because of thromboembolic disease, including stroke, the benefits of warfarin still far outweigh its risks. At this stage of our knowledge, the routine use of a test for the apolipoprotein E 4 allele is not warranted.

Screening for Asymptomatic Bruit Had Poor Cost-Effectiveness

Lee TT, Solomon NA, Heidenreich PA, et al. Cost-effectiveness of screening for carotid stenosis in asymptomatic persons. Ann Intern Med. 1997; 126:337-46.

A common clinical scenario is an asymptomatic bruit. In the Asymptomatic Carotid Artery Study (ACAS), a 5-year randomized clinical trial of patients with asymptomatic bruit, those with stenoses greater than 60% had a 53% reduction in the risk for ipsilateral stroke if they underwent endarterectomy done by skilled surgeons compared with patients treated with anticoagulants only [2].

Lee and colleagues performed a cost-effectiveness analysis on patients in clinical trials who matched the characteristics of those in ACAS. The analysis used the following assumptions: Screening would be done by ultrasonography, and angiography would be done on patients with at least 60% stenosis on ultrasonography. Because stroke occurs more often in men, the analysis included men only. Patients with clinical evidence of carotid artery disease were excluded from screening programs.

The authors found that under these conditions, identification of patients who had asymptomatic carotid artery stenosis of at least 60% and would then undergo surgery would cost $120 000 per quality-adjusted life-year. Under the most favorable assumptions (in which the screening test was perfect and the prevalence of disease among persons screened was 40%), the marginal cost would decrease to $50 000 per quality-adjusted life-year. For hypertension, cost estimates for screening and treatment are about $12 000 to $43 000 per year of life saved (1993 U.S. dollars), and coronary artery bypass grafting (CABG) for patients with left main coronary artery disease costs $6300 to $7000 per quality-adjusted life-year (1991 U.S. dollars). Many experts believe that interventions costing more than $100 000 per quality-adjusted life-year are not cost-effective. However, this is an arbitrary value judgment that could conflict with a physician's ethical responsibility to the individual patient rather than the economic benefit of society at large.

Much of the cost is incurred because the typical patient with carotid artery disease tends to be an approximately 70-year-old man with coronary artery disease. These results may therefore underscore what clinicians already know intuitively: that patients with many comorbid conditions may not be good candidates for endarterectomy. However, clinicians do deal with individual patients; thus, despite these results, patients who are relatively healthy and who have asymptomatic carotid artery disease should undergo endarterectomy, as demonstrated by ACAS. No available technology predicts when a stroke will occur in patients with carotid artery stenosis. However, one can assume that if no other disease kills a patient with carotid stenosis, a stroke will eventually occur if surgery is not done.

Neuroimaging

Most patients who present to emergency departments with symptoms of stroke undergo noncontrast computed tomography to rule out bleeding as a cause. Once this test is done, many physicians do no further brain imaging studies; instead, they focus on treating the underlying causes of thrombotic or embolic stroke. With the advent of more intensive therapies that are effective only in a narrow window of time, new imaging techniques are becoming useful.

Diffusion-Weighted Magnetic Resonance Imaging Detected Early Ischemia

Lutsep HL, Albers GW, DeCrespigny A, et al. Clinical utility of diffusion-weighted magnetic resonance imaging in the assessment of ischemic stroke. Ann Neurol. 1997; 41:574-80.

Since the introduction of thrombolytic therapy for acute stroke, advocacy groups such as the American Heart Association have promoted the concept of "brain attack" to the public. Analogous to heart attack symptoms, brain attack symptoms should prompt a patient to go to an emergency department as rapidly as possible. Thrombolytic therapy is most efficacious if given within 3 hours of symptom onset [3]. Unfortunately, patients often arrive at the hospital aphasic, so the exact time of symptom onset cannot be determined. In these cases, thrombolysis is not offered because the risks might outweigh the benefits.

Diffusion-weighted magnetic resonance imaging (MRI) detects the diffusion of water molecules associated with cytotoxic edema. It has been shown that diffusion-weighted MRI can detect brain ischemia before standard MRI can (Figure 1) [4]. In a retrospective analysis, Lutsep and colleagues reviewed 103 diffusion-weighted MRI studies of patients with stroke symptoms to determine the clinical usefulness of the test.

View larger version (127K): [in this window] [in a new window]   Figure 1. A patient with early cerebral ischemia had an almost normal magnetic resonance imaging scan (top), but diffusion-weighted magnetic resonance imaging done simultaneously (bottom) showed obvious evidence of ischemia.

Diffusion-weighted MRI was done within 24 hours of stroke in one third of patients. It aided in diagnosis or management in 8% of patients. Unfortunately, because the earliest study was performed 5 hours after onset of symptoms, this study could not determine whether diffusion-weighted MRI will help to identify patients in the 3-hour window in which thrombolysis is most beneficial.

Diffusion-weighted MRI is not yet ready to become part of routine practice, but it probably will be soon. It holds great promise for identifying patients who will benefit most from thrombolytic therapy.

Degenerative Diseases

Most neurologic advances in the past 2 years have been made in the area of parkinsonism and other neurodegenerative conditions. Surgical approaches have become a clinical option for patients with severe Parkinson disease. Three therapeutic options are now available for patients with multiple sclerosis.

Surgical Treatments Emerge for Symptom Control in Parkinson Disease

Lang AE, Lozano AM, Montgomery E, et al. Posteroventral medial pallidotomy in advanced Parkinson's disease. N Engl J Med. 1997; 337:1036-42.

Parkinson disease occurs most commonly in persons 45 to 65 years of age. Degeneration of the substantia nigra causes a depletion of dopamine and an imbalance between dopamine and other neurotransmitters, such as acetylcholine, that are normally present in the corpus striatum. Medical treatment attempts to correct this imbalance by blocking the effect of acetylcholine with anticholinergic drugs or by using levodopa, the precursor of dopamine. In a poignant account of living with Parkinson disease, the author describes the results of pallidotomy, a surgical therapy [5].

Pallidotomy is a stereotactic operation that attempts to destroy the motor portion of the globus pallidum, a part of the basal ganglia that secretes neurotransmitters that oppose dopamine. Lang and colleagues studied 40 patients with severe Parkinson disease who underwent pallidotomy. Patients had either disabling immobility or dyskinesias from levodopa therapy. Approximately half of the patients who had been dependent on assistance in activities of daily living before surgery became independent for 6 months afterward. However, this improvement was not sustained over a 2-year period. Until prospective clinical trials are conducted, this study presents the best evidence that pallidotomy is an effective short-term treatment for patients with severe motor symptoms due to Parkinson disease or from the complications of pharmacologic treatment.

Another surgical intervention for parkinsonism is deep-brain stimulation [6]. During this procedure, high-frequency electrodes are placed in the globus pallidus. The electrodes imitate a lesion but do not kill the cells. Patients with these stimulators can turn them off and on to accomplish specific motor tasks. For example, by placing a magnet over a battery pack placed in the chest wall, a patient with a severe tremor can stop the tremor long enough to sign a check. During sleep, the device is kept off to avoid permanent brain damage. No formal clinical trials have yet been done, but the technique seems promising.

The third surgical treatment is brain tissue transplantation. Surgeons can take cells from the substantia nigra of a fetus and stereotactically place them into the striatum of a patient with parkinsonism. These cells are not rejected, and they secrete dopamine. However, the technique has two major problems. The first is ethical: Is it right to use cells from a fetus, even a dead fetus, to treat patients who are at least middle-aged? If the answer to that question is "yes," we then face the problem of insufficient numbers of fetuses to treat a disease as common as parkinsonism-the prevalence is about 15% in persons 65 to 74 years of age and more than 50% in those 85 years of age and older [7]. The ultimate solution, short of preventing the disease, may come from genetic engineering by which dopaminergic cells might be grown in culture.

Options for Treatment of Multiple Sclerosis Increase

Rudick RA, Cohen JA, Weinstock-Guttman B, et al. Management of multiple sclerosis. N Engl J Med. 1997; 337:1604-11.

Multiple sclerosis is a common disease in young people marked by physical decline that usually occurs over 30 to 40 years. Before 1984, diagnosis was based on clinical signs and symptoms, but MRI has become a test with reasonably good specificity. Most patients have the relapsing-remitting form of disease, in which they are stable for long periods and then have episodes of acute exacerbations.

In this review, Rudick and colleagues discussed four therapeutic options, mainly for relapsing-remitting multiple sclerosis. For exacerbations, short-term high-dose corticosteroid therapy significantly accelerates recovery from acute symptoms. In patients with relapsing-remitting disease, interferon-ß1a, interferon-ß1b, and glatiramer acetate (a mixture of polypeptides) each produced about a one-third reduction in relapses compared with placebo.

Despite these advances, many questions about treatment remain: When should therapy be started? How long should therapy be continued? What are the mechanisms of action of the drugs? What are the long-term benefits? Answers to these questions are still guesses.

Neurologic Aspects of Medical Disease

Many clinical problems in neurology are complications of medical diseases or treatments. For example, hyponatremia, which can produce neurologic symptoms, frequently occurs in patients who have recently had surgery. The cause of this phenomenon was reviewed in the recent Update in Nephrology [8].

Major Events Occurred in More Than 6% of Patients Having Routine Coronary Artery Bypass Grafting

Roach GW, Kanchuger M, Mangano CM, et al. Adverse cerebral outcomes after coronary bypass surgery. Multicenter Study of Perioperative Ischemia Research Group and the Ischemia Research and Education Foundation Investigators. N Engl J Med. 1996; 335:1857-63.

More than 800 000 myocardial revascularization operations are performed annually throughout the world. One of the most common complications is neurologic dysfunction, including encephalopathy and stroke. Many observational studies have reported on a wide range of these complications, but the actual neurologic risk has been unknown.

Roach and colleagues sought to determine the incidence and predictors of major perioperative adverse neurologic events and the use of resources associated with these events. In a multicenter prospective study, they followed 2108 patients undergoing elective CABG to monitor the development of two types of neurologic complication: focal injury, stupor, or coma at discharge (type 1) and deterioration of intellectual function, memory deficit, or seizure (type 2).

These major adverse neurologic complications occurred in 6.1% of patients: type 1 complications in 3.1% of patients (including 8 patients who died of stroke), and type 2 complications in 3.0%. Patients with neurologic symptoms had about an eightfold higher risk for death, a more than twofold greater length of stay, and a fivefold higher rate of discharge to skilled nursing facilities compared with those without neurologic complications. Risk factors for complications are shown in Table 2.

Some decrease in neuropsychological function has been reported in up to one third of patients placed on extracorporeal circulation for any reason [9]. Roach and colleagues' study is worrisome because it brings into sharp focus the substantial rate of major neurologic complications sustained by patients undergoing CABG. Because CABG is so common, internists should use extra caution in referring patients with risks for these complications, which may trade one disease for another and could result in substantial suffering.

Incidence of Haemophilus influenzae Meningitis Decreased 93%

Schuchat A, Robinson K, Wenger JD, et al. Bacterial meningitis in the United States in 1995. Active Surveillance Team. N Engl J Med. 1997; 337:970-6.

Until recently, bacterial meningitis occurred in 10 000 to 20 000 Americans each year [10]. Five pathogens were responsible for more than 80% of all cases: Haemophilus influenzae, Streptococcus pneumoniae, Neisseria meningitidis, group B streptococci, and Listeria monocytogenes. In the 1980s, bacterial meningitis was primarily a disease of infants, with H. influenzae accounting for 70% of all cases. By 5 years of age, about 1 in 200 children had had H. influenzae meningitis.

By 1990, routine vaccination of infants against H. influenzae type b had become commonplace. This surveillance team measured the effect that the vaccine has had on bacterial meningitis by comparing data from 1986 with data from 1995. Summary results are shown in Table 3.

Since the introduction of the vaccine, the number of cases of meningitis reported to have been caused by the five major pathogens decreased from 12 920 in 1986 to 5755 in 1995. Nearly all of that decline has been due to the reduction in H. influenzae-related disease among children. The median age at which bacterial meningitis occurs increased from 15 months in 1986 to 25 years in 1996, again because of the 93% reduction in the number of cases of H. influenzae meningitis.

In the United States, bacterial meningitis is now primarily a disease of adults. This is not true for most of the rest of the world. In the United States, however, the ranks of the most common causes of meningitis have changed radically; clinicians should take this into account when choosing empirical therapy.

Seizure Disorders

The management of seizure disorder was relatively straightforward until the 1990s. In the past 4 years, the U.S. Food and Drug Administration has approved five new drugs for use against partial seizures. The most recent is tiagabine. Other antiepileptic drugs and their relative costs are listed in Table 4.

Newer Drugs Were Equally Effective among Themselves

Marson AG, Kadir ZA, Chadwick DW. New anti-epileptic drugs: a systematic review of their efficacy and tolerability. BMJ. 1996; 313:1169-74.

At least eight drugs are available for treating seizure disorders. The obvious question is, Which are most efficacious? Marson and colleagues evaluated the efficacy and tolerability of the newly developed antiepileptic drugs gabapentin, lamotrigine, tiagabine, topiramate, vigabatrin, and zonisamide in patients with refractory partial epilepsy. They systematically reviewed 20 published and 8 unpublished randomized, controlled trials of add-on treatment in 3883 patients.

All drugs were found to be superior to placebo in suppressing seizures, but none was more effective than any other. The results lead to two conclusions. First, older drugs-valproate and phenytoin-are still the best studied and are the first-line drugs to use against partial seizures. Second, new drugs with little or no improved efficacy should be used with great caution. Felbamate was the first of the newer drugs approved in the United States for use against partial seizures. When felbamate was finally released in 1993, the U.S. Food and Drug Administration was criticized for being too slow in approving the new antiepileptic agents, many of which were being sold in Europe. These criticisms were made despite the report that felbamate was associated with agranulocytosis at a rate 1000 times that of carbamazepine. Curiously, physicians in practice have remained hesitant to use carbamazepine because of early case reports of agranulocytosis, even though it is now a well-studied drug. However, sales of newer drugs, including felbamate, have done well even among general internists.

This is not to say that new drugs should not be used, but it seems only reasonable to favor drugs that have been well tested and have been shown to be effective and well tolerated. For the most part, the older antiepileptic agents fit these criteria. The new drugs do have a place; in most cases, however, only specialists in seizure disorders should be using them, and then only in patients in whom older drugs have not controlled the seizures.

To sum up, the drugs of choice for partial seizures remain valproate and phenytoin. For generalized seizures, valproate should be the first choice, followed by phenytoin or carbamazepine.

Treatment of Status Epilepticus Is Becoming Simpler

Lowenstein DH, Alldredge BK. Status epilepticus. N Engl J Med. 1998; 338:970-6.

Status epilepticus (continuing or rapidly recurring seizures) occurs in up to 150 000 persons per year, killing about 55 000 of them. Up to one third of adults with new-onset seizures present with status epilepticus. In general, single seizures last for about 60 seconds, and few last more than 120 seconds. Most practitioners know that therapy becomes urgent after about 20 minutes of seizure activity. Their estimates are hindered, however, because most medical histories come from the patients' families, who vastly overestimate the length of a seizure.

The main goal of Lowenstein and Alldredge's review was to develop an approach to the management of status epilepticus. The message was clear: Treatment is getting simpler. After ensuring that life support measures are taken and glucose and thiamine given, one should begin with intravenous phenytoin, 20 mg/kg of body weight at a rate of 50 mg/min. Fosphenytoin is now available and may be administered at the same dose as phenytoin. Therapeutic blood levels may be reached sooner with fosphenytoin, but the major advantage of the drug is that, unlike phenytoin, it does not result in local tissue damage at the infusion site; this complication is caused by the diluent in which phenytoin is dissolved. The greater expense of fosphenytoin, however, may outweigh the benefits of the drug. Many advocate using lorazepam or another benzodiazepine, but there are good reasons to avoid the latter: Phenytoin is an excellent anticonvulsant, and benzodiazepines sedate consciousness and depress respiration. (A common joke among neurologists is "If you're nervous enough to feel you have to use diazepam, then it is probably more helpful for the patient if you give it to yourself.") If a benzodiazepine is used, lorazepam is a reasonable choice because its half-life is relatively short. Phenobarbital is given for seizures that continue after the phenytoin infusion. If phenobarbital fails, general anesthesia should be administered.

Status epilepticus is life-threatening, and, like most life-threatening emergencies, it is a crisis that responds well if simple maneuvers are carried out in a calm, deliberate manner.