Stroke in the developing world

Abstract

Stroke is the third most common cause of death in most western populations after coronary-heart disease and cancer1Warlow CP Dennis MS van Gijn J et al.Reducing the burden of stroke and improving the publich health.in: Warlow CP Dennis MS van Gijn J Stroke. 10th edn. A practical guide to management. Blackwell Science Ltd, Oxford1996: 632-649Google Scholar It is thus the commonest life-threatening neurological disorder, and the resulting disability is the most important single cause of severe disability among western people living in their own homes2Martin J Meltzer H Elliot D OPCS survey of disability in Great Britain Report I: the prevalence of disability among adults. Her Majesty's Stationery Office, London1988Google Scholar The burden of stroke on patients, their families, and society is, correctly, publicised in most developed countries. Stroke in the developing world is less well documented, and some data are not retrievable because of the language barrier or the limited dissemination of data collected. However, more than two-thirds of the world's population live in developing countries, and one of the consensus statements from the Asia-Pacific Consensus Forum on Stroke Management predicts that “In the next 30 years the burden of stroke will grow most in developing countries rather than in developed countries”3Asia Pacific Consensus Forum on stroke management. Stroke (in press).Google Scholar Thus stroke in developing countries ought to be examined further—in particular, the similarities and differences in stroke between developed and developing countries. The epidemiology of stroke in the developed countries is well established (see page 1).4Davis PH Hachinski V Epidemiology of cerebrovascular disease.in: Anderson DW Neuroepidemiology: a contribute to Bruce Schoenberg. CRC Press, Boston1991: 27-53Google Scholar Information on incidence, prevalence, and mortality of stroke is extremely important in the assessment of priorities for dealing with this disorder, in the recognition of unusual patterns of disease occurrence, in the provision of clues to the cause of disease, and hence in the design of programmes for prevention and control. Such information is limited in the developing world—eg, in Asian countries, where more than half the world's population lives (table 1).Table 1Stroke epidemlological data for nine Asian countriesStroke IncidenceStroke prevalence (per 100 000)Crude death rates (per 100 000) (>1990)% stroke subtypeSINNANA50·310Annual reports. Registry of Births and Deaths, Singapore 1994Google ScholarSGH/TTSH*Singapore General Hospital/Tan Tock Seng Hospital14Venketasubramanian N Sadasivan B Tan AKY et al.Stroke patterns in a hospital-based stroke data bank.Cerebrovasc Dis. 1994; 4: 250Summary Full Text PDF Scopus (1) Google Scholar32%/25% (atherothrombotic)10%/10% (cardioembolic)24%/39%(lacunar)7%/0% (undetermined)27%/26% (haemorrhagic)MALNANANAKL Hospital, patients >60yr†Kuala Lumpur Hospital.15Hanip MR The survey of current management and outcome of patients admitted for first stroke at Kuala Lumpur Hospital.in: Proceeding of 1st MOH-AM (Ministry of Public Health Administration) Scientific Meeting, Kuala Lumpur 1996Google Scholar53% (infarction)33% (haemorrhagic)8%(lacunar)3% (SAH)3% (others)INDNA90-2228Dalal PM Stroke in young and elderly: risk factors and strategies for stroke prevention.J Assoc Physicians India. 1997; 45: 125-131Google ScholarNA83% (ischaemic)16Dalal PM Shah PM Airy RR Kikami BJ Cerebrovascular diseases in the West Central India; a repot on angiographic finding from a prospective study.BMJ. 1968; 3: 769-774Crossref PubMed Scopus (27) Google Scholar17% (haemorrhagic)TLDNA6909Viriyavejakul A Vannasaeng S Poungvarin N The epidemiology of cerebrovascular disease in Thailand.in: Sixth Asian and Oceanian Congress of Neurology. 10th edn. Asia Pacific Congress Series No 22. Excerpta Medica, Amsterdam1983: 10Google Scholar (age>20)1111Annual report of national health statistics. Statistics Department, Ministry of Public Health, Bangkok, Thailand 1996Google ScholarSiriraj Hospital17Poungvarin N Epidemiology of stroke.in: Poungvarin N Stroke. Roenkaew Kanpim, Bangkok1991: 11-37Google Scholar70% (ischaemic)30% (haemorrhagic)INDORural 51·65Surtideni L Hypertension as risk factor of stroke.in: Thesis for Master in Clinical Epidemiology. University of Indonesia, 1995Google ScholarNAFalling9% (TIA)18Misbach J Androni S Malik SM Stroke survey in Jakarta, research and development.Department of Health's Report. 1995; Google ScholarUrban NA4% (RIND)26% (haemorrhagic)61% (thrombosis)1% (SAH)2% (other)HK100-200NA521210th edn. Hospital Authority Statistical Report 1993/94. 1995Google Scholar70% (ischaemic)19Kay R Woo J Kreel L Wong HY Teoh R Nicholls MG Stroke subtypes among Chinese living in Hong Kong.in: The Shatin Stroke Registry. 10th edn. Neurology. 42. 1992: 985-987Google Scholar30% (haemorrhagic)PLPNANANA53%(ischaemia with infarct)20Proceedings from the 1993 Philippines Neurological Association Convention. 1993Google Scholar17% (ischaemia without infarct)30% (haemorrhage)TWN3306Hu HH Sheng WY Chu FL Lan C Chiang BN Incidence of stroke in Taiwan.Stroke. 1992; 23: 1237-1241Crossref PubMed Scopus (146) Google Scholar (age >35)1430-16406Hu HH Sheng WY Chu FL Lan C Chiang BN Incidence of stroke in Taiwan.Stroke. 1992; 23: 1237-1241Crossref PubMed Scopus (146) Google Scholar (age >35)65 (falling)13Pan WH Li LA Tsai MJ Temperature extremes and mortality from coronary heart disease and cerebral infraction in elderly Chinese.Lancet. 1995; 345: 353-355PubMed Scopus (213) Google ScholarShin-Hong WHS Hospital21Yip PK Jeng JS Lee TK et al.Subtypes of ischemic stroke: a hospital-based stroke registry in Taiwan (SCAN-IV).Stroke. 1997; 28: 88-94Crossref PubMed Scopus (141) Google Scholar63% (infarction)28% (haemorrhagic)6% (SAN)4% (other)KOR2807The Korean Neurological Association Epidemiology of cerebrovascular diseases in Korea: a collaborative study, 1989–1990.J Korean Med Sci. 1993; 8: 281-289Crossref PubMed Scopus (37) Google ScholarNA80·47The Korean Neurological Association Epidemiology of cerebrovascular diseases in Korea: a collaborative study, 1989–1990.J Korean Med Sci. 1993; 8: 281-289Crossref PubMed Scopus (37) Google Scholar49% (ischaemic)31% (haemorrhagic)18% (SAH)2% (other)NA=data not available. SIN=Singapore; MAL=Malaysia, IND=India; TLD=Thailand; INDO=Indonesia; HK=Hong Kong; PLP=Philippines; TWN=Taiwan; KOR=Korea; SAH-subarachnoid haemorrhage; TIA=transient ischaemic attack; RIND=reversible ischaemic neurological deficit.* Singapore General Hospital/Tan Tock Seng Hospital† Kuala Lumpur Hospital. Open table in a new tab NA=data not available. SIN=Singapore; MAL=Malaysia, IND=India; TLD=Thailand; INDO=Indonesia; HK=Hong Kong; PLP=Philippines; TWN=Taiwan; KOR=Korea; SAH-subarachnoid haemorrhage; TIA=transient ischaemic attack; RIND=reversible ischaemic neurological deficit. The most important consideration in epidemiological investigations is the accuracy of the diagnosis and the representativeness of the population surveyed. Diagnostic accuracy is a direct function of neurological expertise. The demography and the poverty of national health resources in this region contribute to the difficulty in conducting neuroepidemiological studies in the developing world (table 2).Table 2Demographic data and national health resources in nine Asian countriesPopulation (million)/%rural% age >60/65 yrNo of neurologists/neurosurgeonsNo of hospitals*Total includes public and private (private more than public in some countries); includes clinics for Korea.</fn> (teaching)No of hospital beds (stroke units)No of CT/MRI scannersSIN3·1/-9·7(>60)22Lim SH Tan CH Spectrum of neurological disorders in Singapore.Neurol J Southeast Asia. 1996; 1: 19-26Google Scholar21/1522Lim SH Tan CH Spectrum of neurological disorders in Singapore.Neurol J Southeast Asia. 1996; 1: 19-26Google Scholar25(5)10 446 (2)13/6MAL20·6/535·8 (>60)2310th edn. Annual report of national health statistics. Statistics Department, Ministry of Public Health, Kuala-Lumper, Malaysia1996Google Scholar15/162310th edn. Annual report of national health statistics. Statistics Department, Ministry of Public Health, Kuala-Lumper, Malaysia1996Google Scholar302 (6)42 500 (0)22/10IND943·7/737·0 (>60)24Annual report of national health statistics, India 1996–1997 Statistics Department, Tata Service, Ltd. Department of Economics and Statistics, Mumbai1997Google Scholar405/48324Annual report of national health statistics, India 1996–1997 Statistics Department, Tata Service, Ltd. Department of Economics and Statistics, Mumbai1997Google Scholar13 700 (146)810 000(5)510/203TLD61·0/648·4 (>60)11Annual report of national health statistics. Statistics Department, Ministry of Public Health, Bangkok, Thailand 1996Google Scholar150/15011Annual report of national health statistics. Statistics Department, Ministry of Public Health, Bangkok, Thailand 1996Google Scholar1043(12)90 740 (1)150/20INDO197·6/665·0 (>60)2510th edn. Annual report of national health statistics. Statistics Department, Ministry of Publicdh Health, Jarkata, Indonesia1995Google Scholar303/432510th edn. Annual report of national health statistics. Statistics Department, Ministry of Publicdh Health, Jarkata, Indonesia1995Google Scholar926 (30)110460(1)35/10HK6·4/59·0 (>60)1210th edn. Hospital Authority Statistical Report 1993/94. 1995Google Scholar30/411210th edn. Hospital Authority Statistical Report 1993/94. 1995Google Scholar41 (2)29 342 (3)NA/9PLP69·3/545·0 (>65)2610th edn. Annual report of national health statistics. Department of Health, Ministry of Public Health, Manila, Philippines1995Google Scholar99/402610th edn. Annual report of national health statistics. Department of Health, Ministry of Public Health, Manila, Philippines1995Google Scholar1111(7)37 571 (1)25/6TWN21·7/427·3 (>65)2710th edn. Annual report of national health statistics. Statistics Department, Ministry of Public Health, Taiwan1996Google Scholar414/1722710th edn. Annual report of national health statistics. Statistics Department, Ministry of Public Health, Taiwan1996Google Scholar596 (26)112 380 (7)264/38KOR45/225·5 (>65)2810th edn. Annual report of national health statistics. Medical Insurance Statistics Year Book, Korea Medical Insurance Cooperation, Seuol, Korea1994Google Scholar674/9232810th edn. Annual report of national health statistics. Medical Insurance Statistics Year Book, Korea Medical Insurance Cooperation, Seuol, Korea1994Google Scholar28 121 (252)NA(4)700/160NA=data not available, see table 1 for abbreviation for countries.* Total includes public and private (private more than public in some countries); includes clinics for Korea.</fn> Open table in a new tab NA=data not available, see table 1 for abbreviation for countries. Inaccuracy of death certification in rural areas is one reason for the unreliability29Viriyavejakul A Stroke in Asia: an epidemiological consideration.Clin Neuropharmacol. 1990; 13: S26-S33Crossref PubMed Scopus (29) Google Scholar of stroke mortality rate in the developing world. The death certificates are usually filled in by government officers on the basis of data from the deceased person's relatives, without verification of cause of death by a doctor. Moreover, the necropsy rate is usually low, ranging from 1–15%. What is needed is the degree of attention that has been paid in developed countries to the validity of cause-of-death statistics, relations between certificate entries and necropsy findings, multiple cause-of-death analyses, and programmes to improve the quality of death records.30Havlik RJ Rosenberg HM The quality and application of death records of older persons.in: Wallace RB Woolson RF The epidemiologic study of the elderly. Oxford University Press, New York1992: 262-280Google Scholar Risk factors for stroke in the developing world are usually similar to those in developed countries because most of the data have been collected on forms of standard design. Age and hypertension are the most important risk factors for stroke. Other factors such as diabetes mellitus, smoking, raised blood lipids, and obesity seem to be less important in stroke than in heart disease.31INCLEN Multicentre Collaborative GroupSocio-economic status and risk factors for cardiovascular disease: a multicentre collaborative study in the International Epidemiology Network (INCLEN).J Clin Epidemiol. 1994; 47: 1401-1409Summary Full Text PDF PubMed Scopus (62) Google Scholar, 32INCLEN Multicentre Collaborative GroupBody mass index and cardiovascular disease risk factors in seven Asian and five Latin American Centers: data from the International Clinical Epidemiology Network (INCLEN).Obes Research. 1996; 4: 221-228Crossref PubMed Scopus (40) Google Scholar The course and prognosis of stroke in the developing world is uncertain because of the lack of long-term follow-up data. Depending on the length of the follow-up, the percentage of stroke survivors in the developed countries with recurrence of stroke is 15–40%.33Marquardsen J Epidemiology of stroke in Europe.in: Barnett HJM Mohr JP Stein BM Yatsu FM Stroke. 10th edn. Pathophysiology, diagnosis and management. volume 1. Churchill Livingstone, New York1986: 31-43Google Scholar The likelihood of stroke recurrence in the developing world is expected to be even higher because compliance with treatment for control of risk factors and prophylaxis against stroke is poor. Atheroma seems to be an almost inevitable accompaniment of ageing in the developed countries. It is by far the commonest arterial disorder and, when complicated by thrombosis or embolism, is the most frequent, but by no means the only, cause of cerebral ischaemia and infarction. Extracranial atherothrombo-embolism is one of the major causes of ischaemic stroke in the developed countries, whereas intracranial small-vessel disease is the case in the developing world. This difference is probably due to the atherogenic diet taken in developed countries.34Warlow C Disorders of the cerebral circulation.in: Walton J Brain's diseases of the nervous system. 10th edn. Oxford University Press, Oxford1993: 197-268Google Scholar Hypertension is less well controlled in developing than in the developed world, so it is a major contributing factor to lipohyalinosis and microaneurysms (small-vessel arteriopathy), which can lead to small, deep haemorrhages as well as lacunar infarction, both of which can occur in the same patient. Causes of ischaemic stroke that are more prevalent in the developing than in the developed world include infective arterial disease (eg, syphilis, tuberculosis), cerebral embolism due to rheumatic heart disease or infective endocarditis, Takayasu's disease, snake bites, malaria, and cerebral venous sinus thrombosis.34Warlow C Disorders of the cerebral circulation.in: Walton J Brain's diseases of the nervous system. 10th edn. Oxford University Press, Oxford1993: 197-268Google Scholar Spontaneous intracerebral haemorrhage is more prevalent in developing countries than in developed countries because of poor control of hypertension, but the incidence of subarachnoid haemorrhage is similar worldwide. Some specific causes of intracerebral haemorrhage that are more prevalent in the developing countries include parasitic infestation (eg, gnathostomiasis, schistosomiasis, larva migrans), ruptured mycotic aneurysm, snake bite, scorpion bite, inflammatory vascular disease, and intracranial venous thrombosis.35Allen CMC Clinical diagnosis of the acute stroke syndrome.G J Med. 1983; 42: 515-523Google Scholar The diagnosis of whether or not a stroke has occurred is straightforward if there is a clear history of sudden onset of focal brain dysfunction, or if the symptoms were first noticed on waking, especially if the patient is aged over 50 and has vascular risk factors or disorders. Most of the history of stroke patients in the developed countries is usually clear-cut, and the likelihood of a computed tomographic (CT) scan showing anything other than an infarction or a haemorrhage is under 5%.35Allen CMC Clinical diagnosis of the acute stroke syndrome.G J Med. 1983; 42: 515-523Google Scholar However, in the developing countries, patients may not give clear histories, and CT scans are not widely available. Thus, the differential diagnosis of acute stroke may include head injury (ie, subdural haematoma), cerebral cysticercosis, malaria, tuberculous meningitis, cryptococcal meningitis, encephalitis, brain abscess, brain tumour, and parasitic infestation. A complete blood count may give a clue to the presence of infection, especially if there is eosinophilia (for parasitic diseases). Serum must be taken to test for HIV-1 infection in all young stroke patients because the prevalence of this infection in developing countries is high. Routine laboratory tests should include fasting blood sugar, lipid concentrations, and immunological tests for syphilis. A CT scan is essential for the definite diagnosis of stroke subtype (cerebral infarction or haemorrhage) as well as for identifying the site of the lesion and determining the extent of the disease. Unfortunately, CT scanners are expensive, and in developing countries they are affordable only in large hospitals. Where scans are not widely available, haemorrhage can be distinguished clinically from infarction with 90% accuracy36Poungvarin N Viriyavejakul A Komontri C Siriraj stroke score and validation study to distinguish supratentorial intracerebral haemorrhage from infarction.BMJ. 1991; 302: 1565-1567Crossref PubMed Scopus (119) Google Scholar by use of the Siriraj stroke score (SSS). This score is based on weights given to each of five clinical variables (level of consciousness, diastolic blood pressure, headache, vomiting, atheroma markers). It is useful for developing countries, but each country should adapt it for local use because the prevalence of haemorrhagic stroke differs from country to country. Where CT scanners are not widely available, a scan should be limited to cases in which the diagnosis is uncertain, the source of the disease is atypical, the stroke is in the cerebellum or thalamus, there is subarachnoid haemorrhage, the patient is young (aged under 45), and intracranial haemorrhage has to be excluded (eg, when anticoagulants/thrombolytic therapy is intended). Magnetic resonance imaging of the brain is superior to CT scanning when the lesion is small or located in posterior fossa, or when the onset is very acute. In developing countries, magnetic resonance imaging for stroke management is not cost-effective. In developed countries, routine lumbar puncture has no place in stroke management because of the danger of brain shift if there is a large intracerebral haematoma or oedematous infarct, but in developing countries, lumbar puncture may be indicated in some stroke patients to verify certain causes of stroke, such as parasitic infestation, or chronic meningitis (eg, syphilis, tuberculosis, or infective endocarditis). The wide variation in hospital admission rates for stroke in developed countries is presumably a reflection of the resources available to the general practitioner, the proximity and quality of a hospital with an available bed and that is prepared to take stroke patients, expectations of the patient and family, local or national guidelines, and cost. Some patients cannot be cared for at home because the stroke is so severe, they live alone, or domiciliary nursing support is inadequate; nursing and general care is the commonest, or even the sole, reason for hospital admission. By contrast, in developing countries hospital admission depends mainly on the severity of the stroke— the more severe the stroke, the better the chance of being admitted. Thus hospital data on stroke admission are usually biased towards the more serious or complicated cases. Home and traditional treatment of stroke is still accepted practice in the rural areas of most developing countries. The aims in the general management of acute stroke are good nursing care, maintenance of pulmonary and cardiovascular function, fluid, electrolyte, and nutritional balance, and avoidance of systemic complications. These goals are usually achieved in inpatients but the mortality and complication rates in developing countries are still higher than those in developed countries because of the admission bias towards severe cases. Expert stroke teams are rarely available in most developing countries, so patients there are unlikely to be treated urgently. The patients are usually cared for by general physicians, with only a minority of patients being under the care of a neurologist. There is evidence that management in a specialised stroke environment improves mortality and outcome in a cost-effective way.3Asia Pacific Consensus Forum on stroke management. Stroke (in press).Google Scholar Treatment for acute stroke in developing countries is generally symptomatic; thrombolytic and neuroprotective drugs are the exception rather than the rule, even though there is emerging evidence that thrombolysis is effective for ischaemic stroke if given within 3 h of the stroke.37Hacke W rtPA in acute ischemic stroke: European perspective.Neurology. 1997; 49: S60-S62Crossref PubMed Google Scholar, 38Morgenstern LB rtPA in acute ischemic stroke: the North American perspective.Neurology. 1997; 49: S60-S65PubMed Google Scholar Instead unproven medications such as cerbral vasodilators and “cerebroactive” drugs are intensively marketed and commonly prescribed for stroke patients in developing countries. Antiplatelet drugs are usually the first line of treatment if the stroke is not haemorrhagic. For Asians, low doses of aspirin (60–300 mg) are preferable to high doses because they are less likely to produce gastric side-effects yet have similar efficacy.39Poungvarin N Ketsa-ard K Low dose aspirin and its antithrombotic effect in ischaemic stroke patient.J Med Assoc Thai. 1989; 72: 421-426PubMed Google Scholar Anticoagulants are usually underprescribed in developing countries because of poor drug compliance and the need for frequent monitoring of blood coagulation. Removal of cerebral haematomas and wide craniotomy for brain decompression are the main neurosurgical procedures for stroke patients in developing countries. Embolectomy, endarterectomy, and extracranial-intracranial bypass surgery are done rarely. Stroke rehabilitation is the restoration of patients to their previous physical, mental, and social capability. The importance of “quality of life” after stroke is widely acknowledged in developed countries, but this approach is still lacking in developing countries because it requires a team effort involving many people as well as the patient and, importantly, the carer. The team must be well organised, enthusiastic, optimistic, and professional. It must also meet regularly to discuss the assessment, goals, interventions, and evaluation for each patient by using locally and culturally available resources, such as traditional massage as a substitute for physical therapy in rural areas where no physiotherapist is available. The best policy for combating stroke worldwide is prevention (see page 15)—primary prevention, in particular, since only about 15% of strokes are preceded by transient ischaemic attacks.40Dennis M Bamford J Sandercock P Warlow CP Incidence of transient ischaemic attacks in Oxfordshire, England.Stroke. 1989; 20: 333-339Crossref PubMed Scopus (167) Google Scholar The principles remain the same globally, so developed and developing countries should cooperate to motivate and educate the public and to raise awareness of stroke prevention. Governments and health planners in developing countries tend to underestimate the importance of stroke. To compound this difficulty 80% of the population in developing countries live in rural areas, a factor that, like lack of resources and cultural practices, limits access to stroke services. Other difficulties that developing countries face include the need for special considerations (eg, the fact that anticoagulant therapy may be more hazardous than in western populations and the need to address risk factors such as rheumatic heart disease and puerperal stroke). In these parts of the world top priority for resource allocation for stroke services should go to primary prevention of stroke, in particular the detection and management of hypertension, discouragement of smoking, diabetes control, and other lifestyle issues such as diet. To obtain such priority for stroke prevention awareness of stroke must be raised among health planners and governments. Another priority is education of the public and of health-care providers about the preventable nature of stroke, as well as about the warning symptoms of the disease and the need for a rapid response. Specific interventions were identified at the Asia Pacific Consensus Forum on Stroke Management (panel). The campaign against stroke should be underpinned by research into stroke prevention. The research should ideally be community based.PanelSpecific Interventions for improving stroke management in developing countries •Identify local individuals or teams within a defined community who are responsible for the implementation and delivery of stroke services—eg, rural and community health workers or stroke teams.•Identify or establish key national institutions or organisations which promote training and education of health professionals and disseminate information.•Establish a minimum data set to document and monitor key indicators of stroke at a national, regional, and local level.•Countries should demonstrate implementation of these strategies by the year 2010.From 3Asia Pacific Consensus Forum on stroke management. Stroke (in press).Google Scholar, with permission. •Identify local individuals or teams within a defined community who are responsible for the implementation and delivery of stroke services—eg, rural and community health workers or stroke teams.•Identify or establish key national institutions or organisations which promote training and education of health professionals and disseminate information.•Establish a minimum data set to document and monitor key indicators of stroke at a national, regional, and local level.•Countries should demonstrate implementation of these strategies by the year 2010. From 3Asia Pacific Consensus Forum on stroke management. Stroke (in press).Google Scholar, with permission.