NeurologyBMJ 1999; 319 doi: https://doi.org/10.1136/bmj.319.7206.362 (Published 07 August 1999) Cite this as: BMJ 1999;319:362
- A J Larner, senior registrar,
- S F Farmer, consultant ()
- Correspondence to: S F Farmer, National Hospital for Neurology and Neurosurgery, Queen Square, London WC1N 3BG
Most neurological problems are dealt with by general practitioners and hospital physicians, not by neurologists.1 Neurological disorders account for 10%-20% of acute hospital admissions. Around 10% of the adult population consult their general practitioner each year with neurological symptoms, but of these less than 10% are referred to hospital clinics. Developments in the management of neurological disorders are therefore relevant to doctors without specialist neurological training.
We identified references by regular reading of general medical and neurological journals, from searching the electronic literature (Medline, BIDS), and through discussion with general practitioners and neurological colleagues with specialist interests. The final selection of papers was partly subjective.
The international stroke trial and the Chinese acute stroke trial, each concerning around 20 000 patients, examined antithrombotic therapy (aspirin, heparin) given within 48 hours of acute ischaemic stroke.2 3 Both found aspirin to be associated with about 10 fewer deaths or recurrent strokes in the first 4 weeks for each 1000 patients treated, but with slightly more haemorrhagic strokes. The international stroke trial reported no benefit from subcutaneous heparin (5000 or 12 500 IU twice daily) given with or without aspirin. Hence it was concluded that aspirin should be started as soon as possible after the onset of an acute ischaemic stroke.2 3 Whether aspirin use is “acute treatment” or simply early secondary prevention remains debatable.
No clinical indicators reliably differentiate ischaemic from haemorrhagic stroke. The recommendation that aspirin be started only after appropriate brain imaging in patients requiring admission to hospital will place a huge burden on acute neurological services (over 100 000 people have a first stroke in England and Wales each year). The issue is still more problematic for thrombolytic therapy (tissue type plasminogen activator, streptokinase, urokinase). An overview of previous trials indicated significant excesses of early and total deaths, and of symptomatic and fatal intracranial haemorrhages, after acute thrombolytic therapy, but a significant reduction in death or dependency in patients randomised to treatment within 3 hours of stroke onset.4 To identify the small number of patients likely to benefit from thrombolysis, early hospital admission and prompt investigation will be necessary. This may be achieved in dedicated stroke units, which have been shown to produce long term reductions in death, dependency, and need for institutional care.5
Carotid artery stenosis is an important predisposing factor for cerebrovascular ischaemic events, the risk increasing with the severity of stenosis and the presence of symptoms. For severe (more than 70% narrowing) symptomatic stenosis, carotid endarterectomy is recommended. For severe symptom free stenosis, optimal management has yet to be defined: a meta-analysis of trials6 showed only a small absolute benefit from surgery in reducing the odds of ipsilateral stroke; hence the procedure cannot be routinely recommended. For mild to moderate symptomatic stenosis (less than 70% narrowing), antiplatelet therapy with aspirin or dipyridamole, or both, is recommended. Persistent symptoms may necessitate use of other treatments such as ticlopidine (named patient basis only); clopidogrel, which reduces the relative risk for further ischaemic events slightly more than aspirin7; or anticoagulation with warfarin.
Given within 48 hours of ischaemic stroke, aspirin reduces risk of death and recurrent stroke
Thrombolysis for acute ischaemic stroke is most effective if delivered within 3 hours of stroke onset
Stroke units reduce death, dependency, and need for institutional care after stroke
Carotid endarterectomy is currently recommended only for severe and symptomatic carotid stenoses
Most individuals with newly diagnosed epilepsy enter prolonged seizure remission and have an excellent prognosis, but seizures remain refractory in 20%-30%.8 Improved evaluation of such patients with magnetic resonance imaging and telemetry may identify the structural and functional abnormalities that give rise to seizures. Up to 75% of patients with refractory partial epilepsy show evidence of abnormalities on magnetic resonance imaging,9 some of which are amenable to surgery.
The first line drugs for epilepsy monotherapy remain carbamazepine and sodium valproate; phenytoin is now less used, and although lamotrigine has a monotherapy licence its place has still to be defined. Several new “add on agents” have been licensed in recent years including vigabatrin, gabapentin, lamotrigine, and topiramate. An overview of trials in patients with refractory partial seizures suggests no major differences between these agents in either efficacy or tolerability.10 Severe visual field defects have recently been reported after prolonged use of vigabatrin, prompting the development of guidelines for monitoring vision.11 Vagal stimulation remains an experimental approach to seizure control.12
Population based studies show that patients with epilepsy have an increased risk of death compared with age and sex matched controls.13 Some of these deaths are related to epilepsy itself, for example, as a consequence of accidents, but others are unexplained. The category of “sudden unexpected death in epilepsy” (SUDEP) has recently been introduced to encompass all such deaths, which are more common in individuals with refractory epilepsy (about 1 per 200 patients per year).13 Many of these deaths may be related to unwitnessed seizures, possibly associated with ventricular fibrillation, asystole, respiratory arrest, or neurologically mediated pulmonary oedema. A proportion of cases of sudden unexpected death in epilespy may therefore be preventable with improved seizure control.
Evaluation of patients with refractory partial epilepsy with magnetic resonance imaging and telemetry may yield diagnostic information in up to 75%
Second line, “add on,” agents for refractory partial seizures show no significant differences in efficacy or tolerability
Guidelines for monitoring of visual fields in patients receiving long term vigabatrin have been issued
Sudden unexplained death in epilepsy (SUDEP) may be preventable with better seizure control
Inteferon betas reduce relapse rate in relapsing-remitting multiple sclerosis by about one third in patients experiencing two disabling relapses every 2 years, and may have a small effect in delaying disability progression
Magnetic resonance imaging can help predict which patients with clinically isolated syndromes will progress to multiple sclerosis
Liaison between primary and secondary healthcare services is essential for the appropriate management of established disability in multiple sclerosis
Interferon betas (interferon beta-1b, Betaferon; interferon beta-1a, Avonex, Rebif) have been shown to reduce relapse rate in relapsing-remitting (non-progressive) multiple sclerosis by about one third.14–16 The Association of British Neurologists recommends (guidelines, June 1999) interferon beta be prescribed for ambulant patients with at least two definite relapses in the previous 2 years followed by recovery (may or may not be complete). Whether reduction in relapse rate reduces or prevents later disability is not known; some evidence has been presented in favour.15 16 The uncertainty reflects the difficulties of treatment trials for multiple sclerosis owing to the variable clinical course of the disease, the validation of disability measurements, and discrepancies between secondary outcome measures such as evidence of changes on magnetic resonance imaging, and clinical behaviour.17 Interferon beta-1b has been reported to delay progression (for 9-12 months in a study period of 2-3 years) in secondarily progressive multiple sclerosis of moderate severity (minimum walking distance of 20 metres with assistance)18 and has been licensed for this indication. The Association of British Neurologists has produced guidelines to aid therapeutic decision making in secondary progressive multiple sclerosis, but these may be altered in the light of a recent large study (SPECTRIMS; to date presented only in abstract), which reported no significant effect of interferon beta-1a in delaying disability in secondary progressive multiple sclerosis. Sorting out this complex area, with profound financial implications, will be a high priority for NICE.
Trials are also in progress to ascertain whether interferon betas are useful in primary progressive multiple sclerosis, or in delaying or preventing the onset of disseminated disease after clinically isolated syndromes such as optic neuritis and transverse myelitis Defining which of these patients will progress to widespread disease may be facilitated by magnetic resonance imaging; clinically silent lesions are predictive of the long term risk of subsequent development of multiple sclerosis.19
Other possible treatments for multiple sclerosis include copolymer-120 and pulsed intravenous immunoglobulin,21 which, like interferon betas, reduce the relapse rate. However, the place for these immunomodulatory agents remains to be defined Intravenous methylprednisolone may hasten recovery from acute relapses but has no effect in the long term. A recent trial suggested intravenous methylprednisolone is no better than equivalent oral doses of methylprednisolone for acute relapses.22 If so, then switching from intravenous to oral steroids for acute relapses may make considerable savings.
In advanced multiple sclerosis, management of established disability may be the primary role of the neurologist. Bladder dysfunction usually consists of combined detrusor hyperreflexia and incomplete emptying volumes of less than 100 ml of urine remaining in the bladder after micturition are managed with oxybutinin or detrusitol; volumes greater than 100 ml require clean, intermittent, self catheterisation.23 Sexual dysfunction (erectile failure) may be helped with the phosphodiesterase inhibitor sildenafil citrate (Viagra), or more established agents such as yohimbine or other α adrenoreceptor blockers Management of limb spasticity requires a multidisciplinary approach ensuring correct posture, prevention of skin ulceration from pressure, management of bladder and bowel dysfunction, as well as pharmacological measures. Tizanidine, an α2 adrenoreceptor agonist, has recently been licensed in the United Kingdom as an antispastic agent.24
The management of idiopathic Parkinson's disease is still centred around the use of levodopa preparations since they produce the most effective relief of symptoms in most patients. Distinguishing idiopathic Parkinson's disease from other parkinsonian syndromes (for example, multiple system atrophy, Steele-Richardson-Olszewski syndrome) is crucial as the latter differ in prognosis and management. With long term treatment with levodopa, response fluctuations develop in around 10% of patients with idiopathic Parkinson's disease per year Delaying the use of levodopa in early disease, particularly in young patients (under 50 years), is therefore desirable.
Levodopa sparing agents that may be used as monotherapy in early Parkinson's disease include amantadine, anticholinergics (particularly if tremor is predominant), selegiline, and dopamine agonists. Bromocriptine alone improves about 50% of patients during the first year of treatment but there is a gradual loss of benefit thereafter, with only 10% responding at 5 years; other dopamine agonists have similar effects. Selegiline delays the need for levodopa by a mean of 8 months. One study found increased mortality in patients taking levodopa in combination with selegilne.25
Accurate diagnosis of parkinsonian syndromes is important for appropriate management
Use of levodopa sparing agents may be desirable in the treatment of early onset Parkinson's disease
The late stages of Parkinson's disease necessitate the use of polypharmacy to manage unpredictable response fluctuations to levodopa, which are an almost invariable feature after several years' treatment
Functional neurosurgery may improve some of the features of Parkinson's disease and treatment complications in selected patients
Once levodopa is started it may be several years before response fluctuations develop. These are usually predictable initially, such as end of dose or “wearing off” effects. Strategies to ameliorate these problems include dose fractionation, long acting levodopa preparations (particularly helpful for nocturnal immobility), and addition of dopamine agonists.26 Three new dopamine agonists are now available: ropinirole, cabergoline, and pramipexole.27–29 It is not yet known whether these agents confer significant therapeutic advantages over bromocriptine and pergolide as add on therapy. Add on use of catechol-O-methyltransferase inhibitors (tolcapone, entacapone) may increase total “on” time by about 20%,30 but tolcapone has been withdrawn in the United Kingdom because of hepatotoxicity.
Sudden unpredictable changes between periods of mobility (“on”) with severe levodopa induced involuntary movements (dyskinesias) and disabling parkinsonism (“off”), the “on-off phenomenon”, present a major management problem. Various strategies may be tried to minimise the severity of dyskinesias31—for example, adjusting the timing of levodopa intake, optimising levodopa absorption, and addition of other antiparkinsonian agents such as dopamine agonists and amantadine thus permitting levodopa dose reduction. Increased use of the dopamine agonist apomorphine, given by subcutaneous injection or infusion, may rescue patients from severe sudden off periods, and improve overall mobility.
Renewed interest has been shown in surgery for the symptoms of Parkinson's disease, levodopa induced dyskinesias, and disorders of movement such as tremor and dystonia through the use of stereotactically placed lesions or chronic electrical stimulation with indwelling electrodes.32 Thalamotomy or thalamic stimulation helps tremor and often rigidity but not akinesia. Unilateral pallidotomy (figure) significantly improves contralateral dyskinesias; bilateral pallidotomy may, in addition, improve parkinsonian symptoms but carries increased risks.33 Bilateral subthalamic nucleus stimulation34 or subthalamotomy has produced dramatic improvements in parkinsonism, sufficient to allow large reductions in the levodopa dose and thus improvement in levodopa induced dyskinesias. Appropriate patient selection is a key factor in its successful use, and it will be available only in a few specialist centres.
The definition of a new variant of Creutzfeldt-Jakob disease35 has attracted huge media attention, because it seems to be caused by the same strain of prion protein that causes bovine spongiform encephalopathy, and hence may have been transmitted to humans through contaminated food. However, this uniformly fatal condition remains extremely rare, with only 41 cases reported to date in the United Kingdom.
For Alzheimer's disease, the commonest cause of dementia, two cholinesterase inhibitors, donepezil (Aricept)36 and rivastigmine (Exelon),37 are now licensed for the symptomatic treatment of mild to moderate cognitive impairment (minimental state examination score of 10-26) Trials suggest that in some patients these agents improve cognition, global function, and behavioural function, but there are as yet no data as to whether they delay deterioration or improve outcome Currently it is recommended that these drugs are only commenced on the advice of a specialist. Despite advances in understanding the pathological basis of Alzheimer's disease, centred around amyloid β peptides, this has not resulted in new treatments although many are in development.38
Since the launch of sumatriptan, several triptans have become available for the treatment of acute migraine.39 They differ in pharmacodynamic properties such as time of onset and duration of action. Direct comparisons have not been performed and it is not known whether the differences claimed will have significant impact on clinical practice. Preventive treatments include pizotifen and amitriptyline, but the strongest evidence of efficacy is available for β blockers and sodium valproate.39
The glutamate antagonist riluzole (Rilutek) remains the only licensed treatment for motor neurone disease. Risk of death or tracheostomy was lower with 100 mg riluzole than placebo in limb or bulbar onset disease,40 but it is debatable whether this translates into an improved quality of remaining life.
Intravenous immunoglobulin has been used in many neurological disorders,21 41 but its place remains to be defined; many of these applications are empirical and the long term effects of intravenous immunoglobulin are unknown. Trials in acute idiopathic neuropathy (Guillain-Barré syndrome) suggest that intravenous immunglobulin is equivalent to plasma exchange in reducing disability at 4 weeks, but that the combination of intravenous immunoglobulin and plasma exchange offers no significant additional advantage.42 Likewise, in myasthenia gravis, intravenous immunglobulin seems as efficacious as plasma exchange,43 and it may also be useful in chronic inflammatory demyelinating polyneuropathy in relapse.44
Dementia, rare disorders
Cholinesterase inhibitors are of symptomatic benefit in about 20% of patients with mild to moderate Alzheimer's disease
Some peripheral neuropathies are treatable may show profound benefit with intravenous immunoglobulin treatment
Intravenous immunglobulin may be helpful in the acute treatment of Guillain-Barré syndrome and myasthenia gravis
In the past 5 years, new treatments have become available for neurological disorders previously considered untreatable (multiple sclerosis, Alzheimer's disease, motor neurone disease). Although the high cost of these treatments has sometimes led to antagonism between purchasers and doctors wanting to prescribe, these small, initial, therapeutic inroads have focused on the broader needs of patients with neurological disorders, engendering specialist clinics, specialist nurses, and fostering liaison with community services and patient interest groups.
We thank Drs DJ Thomas, RJ Butterworth, M Prevett, P Rudge, D Kidd, CAH Fisher, S Fearn, and S Addison, and Professors NP Quinn and MN Rossor for commenting on earlier drafts of the article.
Competing interests None declared.