Epidemiology and Etiology of Ischemic Stroke in Young Adults Aged 18 to 44 Years in Northern Sweden
Background and Purpose The aim of this study was to conduct a population-based epidemiological survey among young adults aged 18 to 44 years in Northern Sweden and furthermore to gain further insight into the etiology of ischemic stroke in this age group.
Methods Two studies were done. In the first part, epidemiological data were collected to calculate incidence and mortality from 1991 through 1994. This was based on the World Health Organization Northern Sweden MONICA register of acute stroke events. Eighty-eight first-ever ischemic stroke patients were identified during that period. In the second part, 107 consecutive patients aged 18 to 44 years with ischemic stroke referred to a university hospital were studied prospectively during a 5-year period and were extensively evaluated according to a standardized protocol. On the basis of modified Trial of ORG 10172 in Acute Stroke Treatment (TOAST) criteria, the patients were classified into eight subtypes of ischemic stroke.
Results The average population-based annual incidence rate for ischemic stroke (cases per 100 000 per year) was 11.3 (95% confidence interval, 6.7 to 16.1). The case-fatality rate was 5.7%. According to the modified TOAST criteria, a probable cause of ischemic stroke was identified in 36% and remained unexplained in 21% of cases. Spontaneous cervical arterial dissection was the leading probable etiology (13%). Patent foramen ovale or atrial septal aneurysm was a possible cause of stroke in 28% of cases. The percentages of ischemic stroke attributed to IgG anticardiolipin antibodies (4.7%), atherothrombotic vasculopathy (3.7%), oral contraceptive use (7%), and migraine (1%) were lower than reported in recent clinical series.
Conclusions The incidence rate for ischemic stroke was higher than previously reported from most countries in Western Europe. The higher incidence was not explained by a higher prevalence of premature atherosclerotic vasculopathy. Without the additional diagnostic information derived from advanced cardiac imaging, the proportion of indeterminate cases would have constituted 37% of the patients.
Ischemic stroke in young adults has been considered a relatively rare event, with fewer than 5%1 of all cerebral ischemic infarctions occurring below the age of 45 years, although more than 10% has been reported.2 The age-specific incidence of stroke among individuals in this young age group has been reported by community and hospital surveys from various geographic areas. However, hardly any population-based data have been reported.3
The causes of stroke among young adults are more diverse than in the elderly and require a thorough diagnostic workup. A major problem has thus been that previous studies addressing the etiology of ischemic cerebral infarction in the young have often been retrospective and nonstandardized, with highly variable investigational techniques and sets of diagnostic criteria being applied.4 In addition, advances in technology, including transesophageal investigation and new biochemical assays, have introduced new potential causes of ischemic stroke that still need to be substantiated. Finally, without the acknowledgment of cases not referred for investigation, a biased pattern of the causes of the disease may be present in hospital-based studies from third-level facility hospitals.
The objectives of the present study were to evaluate a true age-specific incidence and furthermore to delineate the causes of ischemic stroke in young adults. The study was thus undertaken in two parts. First, a prospective population-based epidemiological survey was conducted to provide data on the incidence and mortality rates of ischemic stroke in young adults from northern Sweden. Second, in a series of patients admitted to our university hospital, an accurate hierarchy of causes of ischemic stroke was established on the basis of the TOAST classification modified for ischemic stroke in young adults.5
Subjects and Methods
Acute stroke events in the two northernmost counties of our catchment area (Fig 1⇓) have been monitored since 1985 by the WHO Northern Sweden MONICA project.6 The case finding and validation of data quality have been described in detail earlier.7
The total population in the MONICA surveillance area was 518 669 on January 1, 1991, and 527 423 on December 31, 1994. The target population considered to be at risk included all residents aged 18 to 44 years in this area. They represented 37% (194 194 and 193 113, respectively) of the total population at the beginning and at the end of the epidemiological survey period.
Clinical information from discharge records on all subjects in the age range 25 to 45 years with ICD-9 codes 430 to 438 was screened and validated for acute stroke events that met the definition of ischemic stroke. Data from the group aged 18 to 25 years not originally computed in the MONICA register were obtained, adhering to the same guidelines. This part of the study thus contains patients referred to our university hospital (included in the etiological study; see below) as well as patients investigated only at local acute-care hospitals in Northern Sweden. The same inclusion and exclusion criteria as in the etiological study were used.
Northern Sweden is served by 13 local hospitals and a third-level university hospital (Fig 1⇑). The population in the present study consists of patients aged 18 to 44 years who were admitted to Umeå University Hospital between January 1991 and May 1996 as a result of ischemic stroke. The inclusion criteria were as follows: (1) first-ever completed ischemic stroke, defined as a rapidly developing focal neurological deficit with no apparent cause other than a vascular origin, that persisted beyond 24 hours in surviving patients; (2) age from 18 through 44 years; and (3) evaluation possible within 3 months after stroke onset. Patients were scheduled for follow-up at 4 and 12 months after admission. Exclusion criteria were as follows: ischemic stroke due to complications of subarachnoid hemorrhage, cardiac surgery, and malignancy in a terminal stage or occurring as an immediate consequence of trauma.
Clinical and Laboratory Evaluation
Medical history and information regarding cerebrovascular risk factors such as arterial hypertension, diabetes mellitus, smoking, alcohol use, illicit drug use, hyperlipidemia, oral contraceptive use, history of migraine, and occurrence of venous or arterial thrombosis in the family were obtained according to a standardized protocol. Hypertension was defined as systolic blood pressure >160 mm Hg and/or diastolic pressure >95 mm Hg on two different occasions measured outside of the acute phase of stroke or treatment with antihypertensive drugs during the last 2 weeks before recruitment; diagnosis of diabetes mellitus was documented by medical records or at recruitment according to the WHO criteria8 ; hypercholesterolemia was considered present if fasting blood level was ≥6.5 mmol/L at recruitment; and hypertriglyceridemia was present if fasting triglyceride levels were >2.2 mmol/L at recruitment. Current smoking was defined as smoking one or more cigarettes per day. Migraine was defined according to the criteria of the International Headache Society.9
Neuroimaging included CT and MRI of the brain, assessment of cerebral blood flow (single-photon emission CT with Tc–hexamethylpropyleneamine oxime), and cerebral angiography including posterior circulation angiography. Furthermore, duplex ultrasonography of the cervical arteries, chest roentgenography, electrocardiography, and 24-hour Holter electrocardiography recording was performed.
In reference to echocardiographic studies, the TTE studies were performed with the use of standard techniques. Views were obtained in the parasternal, apical, and subcostal windows. The TEE images were obtained with a 5-MHz monoplane transducer in 69 patients and with a multiplane transducer in 28 patients. All examinations were recorded on videotape and analyzed in a blinded manner off-line in a random order. ASA was diagnosed when the atrial septum appeared abnormally redundant and mobile and exhibited an excursion into the left or right atrium, or both, of >10 mm and a base of ≥10 mm. For PFO, the echocardiographic detection of interatrial right-to-left shunting was identified by color-flow Doppler or by the administration of 5 mL agitated saline in an antecubital vein. Two to four contrast injections were systematically performed in each patient, in the resting state and during provocative maneuvers (Valsalva and cough test) to transiently reverse the interatrial pressure gradient. The echocardiographic diagnosis of PFO was based on the appearance of at least three microcavitations, either spontaneously or after provocation maneuvers, into the left atrium, not later than four cycles after the appearance of the microcavitations in the right atrium. The presence of atheroma of any severity was noted when detected on echocardiography, occurring between the aortic valve and the origin of the left subclavian artery. Mitral valve prolapse was defined on TTE as mitral leaflet thickening and displacement beyond the plane of the mitral annulus and into the left atrium in the parasternal long-axis view during systole.
A detailed laboratory study was performed, including complete blood cell count; electrolytes; serum creatinine; amino-transferases; creatine kinase; urinary analysis; erythrocyte sedimentation rate; serum protein electrophoresis; concentrations of blood glucose and glycosylated hemoglobin; antinuclear antibodies and IgG aCLs; rheumatoid factor; complement factors (C3/C4); serological testing for syphilis, borreliosis, and viral infections, including HIV; serum cholesterol and triglyceride levels; LDL and HDL levels; lipoprotein(a); and prothrombin and activated partial thromboplastin times. Activated partial thromboplastin time was also used as a screening test for the presence of lupus anticoagulants. Levels of protein C, protein S, and antithrombin III were analyzed both in the acute phase and at least 4 months after first admission.
Classification of Subtypes
We used a modified stroke subtype classification for the etiology of ischemic stroke with the definitions based on the TOAST classification, accommodated and validated for stroke in the young.5 The patients were classified independently by two groups of paired investigators. A consensus approach was applied when necessary. The main diagnostic criteria are presented in Fig 2⇓.
Incidence rate was derived from the number of first episodes of ischemic stroke among the residents in the two counties during the study period. The average of the populations living in the area between January 1, 1991, and December 31, 1994, was the denominator of the equations for the incidence rates. Case-fatality rate was defined as the number of subjects who died within 28 days from the onset of stroke during the same period. Ninety-five percent confidence intervals were computed according to the Poisson variation for the number of events within the age groups. Fisher’s exact and χ2 tests were used when appropriate for statistical analysis. Probability was two tailed, and P<.05 was considered significant.
The study was approved by the Research Ethics Committee of Umeå University; data handling procedures were approved by the National Computer Data Inspection Board.
A total of 88 first-ever ischemic strokes in the age range 18 through 44 years from the MONICA surveillance area were recognized during the time period of January 1, 1991, to December 31, 1994. Seventy-one cases (81%) were primarily evaluated or referred from secondary-level care settings for further investigations at the university hospital. These patients are thus included in the etiologic study part. The additional 17 cases, all admitted to local acute-care hospitals, were identified from the MONICA project register. These additional cases had all undergone CT scan or necropsy. No further cases were recognized from death certificates only or from medical discharge records in the group aged 18 to 24 years.
The average annual incidence rate of first-ever ischemic stroke in the group aged 18 to 44 years was 11.3/100 000. The age- and sex-specific incidence rates and the 95% confidence intervals are shown in Table 1⇓. In both sexes the risk of stroke increased with age, but there was no statistically significant difference between men and women (χ2=3.03, P=.08).
Five patients died within 28 days, resulting in a case-fatality rate of 5.7%. Deaths resulted from ischemic cerebral edema and herniation in 3 patients and from the effects of severe brain stem cerebellar infarction in 1 patient. One patient died as a result of a myocardial infarction and subsequent congestive heart failure.
Data pertaining to the yearly average of ischemic stroke events in men and women in the group aged 25 to 74 years during the time period 1991 to 1993 were available from the MONICA register.7 On average, 562 first-ever ischemic stroke events occurred yearly during 1991 to 1993. Ischemic stroke in young adults thus represented 3.9% of all ischemic strokes in the group aged 18 to 74 years.
The distribution of the main diagnostic categories for the 71 patients evaluated at the university hospital during the epidemiological survey period is shown in Fig 2⇑. The mean age of the 17 patients (10 men, 7 women) not evaluated at our university hospital was 40.2±2.3 years (range, 27 to 44 years). Based on information from local hospital records on admission and on discharge, an apparent probable diagnosis could be assigned to 3 patients. Because of incomplete evaluation at the local- level hospitals, it was not possible to allocate the remaining 14 patients to a particular etiologic subtype.
In this part of the study, 63 men and 44 women referred to our hospital from January 1, 1991, through May 31, 1996, fulfilled the inclusion criteria. The mean age of the patients was 36.5±6.2 years (range, 19 to 44 years); age and sex distribution is shown in Table 2⇓. Unwillingness to participate in certain procedures, technical problems, pregnancy, and early death explained why scheduled investigations were not accomplished in all cases. Selective angiography of both carotids and at least one vertebral artery was performed in 95 patients (89%), and abnormalities related to clinical symptoms were found in 58 patients (61%). In addition, 3 patients underwent MRI angiography. All patients were investigated by CT, and 80 patients (75%) were investigated by MRI of the brain. The carotid territory was involved in 56%, the posterior circulation was affected in 41%, and multiple territories were involved in 3%. Six patients did not display any visible ischemic lesions on neuroimaging. One hundred five patients (98%) underwent either a TTE or TEE investigation.
After completion of the etiologic workup, assessment of a probable or possible etiology led to classification of the patients into one or more of eight diagnostic groups subdivided into higher-priority (I to IV) and lower-priority (V to VIII) diagnoses (Fig 2⇑). The distribution of diagnoses relating to the specific age groups of 18 to 34 years and 35 to 44 years is shown in Table 3⇓. The main finding was an increased occurrence of nonatherosclerotic vasculopathy in the older age group.
Risk factors were distributed as reported in Table 4⇓. There were no statistically significant differences between sexes with regard to atherosclerotic risk factors, whereas migraine history was more frequently associated with female sex (P=.041).
Ninety higher-priority diagnoses were identified in 76 patients. In 14 patients two higher-priority diagnoses coexisted. Atherosclerotic vasculopathy was diagnosed as the cause of cerebral infarction in 13 patients (11%). Atherogenic risk factors were present in all of these patients except 1. All patients with a possible diagnosis of atherothrombotic vasculopathy demonstrated only discrete plaque formation in the carotid arteries without any signs of flow abnormalities. In addition, TEE revealed a simple aortic arch atheroma in 3 patients.
The main cause for nonatherosclerotic vasculopathy was nontraumatic cervicocerebral arterial dissection. The carotid arteries were affected in 9 patients and the vertebral arteries in 10 patients. One patient with fibromuscular dysplasia as an underlying cause had bilateral carotid dissections. Isolated intracranial arteritis was diagnosed in 1 patient. Coexistent higher-priority diagnosis was found in 7 patients with probable or possible arterial dissection, 5 with a PFO/ASA, 1 with low positive IgG aCL titer, and 1 patient with a simple aortic arch atheroma, in which case priority in the final classification was given to the diagnosis of arterial dissection.
A cardioembolic etiology was presumed in 35 patients (33%). The most frequent abnormality was right to left cardiac shunts consistent with PFO, which became evident in 32 of 97 patients (33%) investigated with TEE. In one patient PFO was associated with pulmonary arteriovenous fistulas. Atrial septum aneurysm was detected in 9 patients (9%), isolated in 5 patients, and associated with PFO in 4 patients. Angiographic lesions compatible with intracranial embolic occlusion were detected in 16 of 30 patients (53%) with PFO/ASA as the possible cause of stroke. Coexistent higher-priority diagnoses was found in 6 patients with PFO/ASA, 1 with a low positive IgG aCL titer and 5 with a possible atherothrombotic vasculopathy, including 1 patient with a simple aortic arch atheroma, but in the final classification priority was given to the cardioembolic diagnosis. Additionally, 6 patients with PFO/ASA had coexisting lower-priority diagnoses (lacunar infarction [n=2], migraine-induced infarction [n=2], and oral contraceptive use [n=2]). Mitral valve prolapse was observed in 1 patient. In 4 patients a probable source of cardiac embolism was present (congenital heart disease [n=2], ischemic dilated cardiomyopthy with left ventricular thrombus and PFO [n=1], and atrial septum defect with left atrial thrombus [n=1]). There was no evidence of potential emboligenic arrhythmias from electrocardiography or Holter monitoring.
With respect to hematological causes of stroke, natural anticoagulant levels were determined for 102 patients at admission and for 97 patients at follow-up. One patient had an inherited protein S deficiency. Seven patients (7%) had low positive readings for IgG aCL. Other laboratory features such as ANA titers, VDRL tests, complement C4, thrombocytes, and aPPT were negative or within normal range in all IgG aCL–positive patients, and no valve abnormalities were detected by echocardiographic investigation. A history of heavy alcohol ingestion within the preceding 24 hours was elicited in 1 patient. Ischemic stroke occurred in the postpartum state in 1 patient.
Thirteen patients met the criteria for lacunar infarct, ie, a lacunar syndrome and small deep infarction compatible with small-artery disease. In 8 patients a coexistent higher-order diagnosis was present (possible cardiac embolism [n=2], possible atherothrombotic vasculopathy [n=5], and IgG aCL low-positive reading [n=1]).
Eighteen percent of men and 35% of women had a history of migraine, but only 1 woman fulfilled the criteria of the International Headache Society for a probable migraine-induced stroke.9 Migraine-induced stroke was possible in an additional 3 patients with higher-priority diagnosis (possible cardiac embolism [n=2] and probable arterial dissection [n=1]).
Oral contraceptive use was the likely cause of stroke in 3 female patients with additional risk factors (smoking and migraine [n=1], smoking [n=1], and hypertension [n=1]). Five women qualified for another diagnosis (possible cardiac embolism [n=2], probable arterial dissection [n=2], and possible atherothrombotic vasculopathy [n=1]).
The etiology of cerebral infarction was indeterminate in 22 patients (21%). The evaluation was “truly negative” except for 1 patient who did not have an angiography and 1 patient who was unable to endure TEE but had a normal TTE investigation. In this subgroup of patients, angiographic investigations were abnormal in 12 patients (52%) with unexplained intracranial occlusion of a major artery or a branch cortical artery. Overall, there was a preponderance of ischemic lesions in the vertebrobasilar system (57%). Fourteen patients (61%) with an indeterminate diagnosis had at least one risk factor for atherosclerosis (6 with one and 8 with two or more atherogenic risk factors).
The present study is one of the largest reported series of young adults with ischemic stroke investigated by a group of physicians at a single medical center and the first study of stroke in young individuals in which a clear population-based strategy for case finding has been combined with a very extensive diagnostic workup.
By international standards, the crude incidence rates for ischemic stroke in the present study are higher than those reported earlier from most countries in Western Europe3 10 11 12 (Table 5⇓) and similar to those among whites in Baltimore, Md.13 One study from Israel has provided information from this decade with an estimated incidence for ischemic stroke of 5/100 000 in the group aged 17 to 44 years.14 Our rates are only lower than the unusually high rates of stroke among males and females aged 15 to 40 years of age in Benghazi, Libya,15 and among blacks in Baltimore, Md.13 The reported 50% higher incidence rates for ischemic stroke in the elderly population of the Northern Sweden MONICA study compared with available data from the MONICA study of Gothenburg, Sweden,10 together with older limited data from the Stockholm region11 may suggest a south to north “stroke gradient” in Sweden. The explanation for such a possible geographic variation is currently unclear.
Similar to reports from other countries, incidence rates for both men and women were found to rise steeply after the age of 35 years.16 In our study this increase was mainly explained by an increase in the number of arterial dissections and cardioembolic cases, but it was influenced to only a minor degree by an increase in premature atherosclerosis (Table 3⇑).
A case-fatality rate of 5.7% in the present study is considerably lower, as expected, in comparison with elderly stroke patients but corresponds to case-fatality rates reported for similar age groups in epidemiological studies3 10 and case series.17 18
The topography of cerebral infarctions in young adults with ischemic stroke has rarely been detailed in previous studies. The proportion of patients with involvement of the vertebrobasilar territory has varied from 25% to 34%.2 19 The relatively high proportion of involvement of the vertebrobasilar territory in our study (41%) may at least partially be due to the extensive use of MRI investigations. Thus, quite a few cases with ischemic lesions in this territory, including cerebellar strokes, may have gone undetected in studies mainly relying on CT scanning.
The criteria for atherosclerotic disease have varied considerably in previous studies, and several studies included cases defined only by the coexistence of risk factors for atherosclerosis, which may explain why atherosclerosis has been considered to be the cause of stroke in 5% to 50% of patients younger than 50 years.17 20 Using the TOAST classification, we detected 3.7% of patients with a probable atherosclerotic vasculopathy; when similar criteria for probable atherosclerotic vasculopathy were used, the rate of atherosclerotic etiology in recent studies has varied from 5% to 23%.17 18 21 22 23
Several recent studies have demonstrated that the presence of plaques ≥4 mm in the aortic arch is an important new source of emboli to the brain in patients older than 60 years.24 However, this does not seem to be the case in a young stroke population. This is consistent with the results reported from a few previous studies including young as well as old stroke patients.25 26
The use of duplex ultrasonographic and arteriographic evaluation of precerebral and intracerebral vessels resulted in a diagnosis of cervicocerebral arterial dissection in approximately 20% of patients, a proportion considerably higher than reported in most earlier studies.17 20 27 Our routine use of posterior circulation angiography revealed that vertebrobasilar dissection was a common cause of stroke in this age group. This emphasizes that a diagnosis of arterial dissection should be considered in all cases of stroke in young adults and that a thorough angiographic evaluation, including the vertebrobasilar territory, is warranted in most cases.
The prevalence of PFO and ASA is increased in younger adults with stroke,28 29 particularly in patients with otherwise unexplained stroke.21 30 In those younger than 45 years, a prevalence of PFO within the same range as in our study (24% to 50%) has been reported from three previous studies.26 30 31 The mechanism underlying thromboembolic events in patients with interatrial septum abnormalities is not well known.21 32 Angiographic evidence of embolic intracranial arterial occlusions was present in 53% of our patients and gave some evidence of a nidus for thrombus formation. It is of interest to note that 50% of our patients with vertebral dissection had a PFO, an ASA, or both. It is thus important to emphasize that significant vascular pathology must be excluded before these cardiac abnormalities are accepted as the cause of stroke in each individual case.
In our stroke population, hereditary deficiency of natural anticoagulants (protein S, protein C, and antithrombin III deficiency) was very rarely encountered, which is in agreement with results reported by Adams et al33 but at variance with the findings reported from small or selected case series.34 35 Furthermore, the low frequency and low titers of IgG aCL in the present study imply that these antibodies do not account for a significant proportion of strokes in young people, at least not in all young stroke populations. These findings are in agreement with two recent larger prospective studies in which the relevance of aCL for ischemic stroke in unselected stroke populations has been questioned.36 37 However, in respect to other hematologic causes of stroke, it is possible that factor V Leiden gene point mutation could be of interest, although whether activated protein C resistance secondary to this mutation causes arterial vascular disorder is still controversial.38 39
The definition of migraine-induced stroke applied in studies conducted thus far has been inconsistent and probably explains why cerebral infarctions in the young attributed to “migrainous infarction” have varied between 1.2% and 25%.18 40 41 In the present study only 1% of the patients (one patient), based on the criteria of the International Headache Society,9 fulfilled the criteria for migrainous infarction, although the prevalence of migraine with or without aura was higher than might be expected from the background.42 43 Based on data from our epidemiological survey, the incidence of migraine-induced infarction meeting the criteria of the International Headache Society can be estimated at 0.14/100 000 per year and including possible migraine-induced infarction can be estimated at 0.7/100 000 per year in the group aged 18 to 44 years.
The risk increment for cardiovascular disease, including stroke, among users of oral contraceptives is currently a matter of controversy.44 45 46 By ruling out coexistent and more convincing etiologies such as cardioembolism and arterial dissection, we could attribute a probable pathogenetic role to oral contraceptive use in 7% of women.
We report a high incidence of stroke among young adults in Northern Sweden. This finding is not explained by conventional causes of stroke, particularly not by a high prevalence of premature atherosclerotic disease. Without the additional diagnostic information derived from TEE contrast investigation, the proportion of indeterminate cases would have constituted 37% of the patients. Nontraumatic arterial dissection was a leading cause of ischemic stroke in our study. New exciting data pointing toward the possibility that spontaneous cervical artery dissections may be due to an underlying connective tissue disorder warrant further investigations of this specific subgroup of patients.47
Selected Abbreviations and Acronyms
|ASA||=||atrial septal aneurysm|
|ICD-9||=||International Classification of Diseases, 9th Revision|
|MONICA||=||Monitoring Trends and Determinants in Cardiovascular Disease|
|PFO||=||patent foramen ovale|
|TOAST||=||Trial of ORG 10172 in Acute Stroke Treatment|
|WHO||=||World Health Organization|
This study was supported by the Swedish Medical Research Council (grant K97-19X-12237-01A to Dr Olsson), Karl-Oskar Hansson’s Foundation, the Swedish Society of Neurologically Disabled (NHR), 1987 Stroke Foundation, Norrlandsfonden, and the Swedish Heart and Lung Foundation.
- Received January 27, 1997.
- Revision received May 15, 1997.
- Accepted May 16, 1997.
- Copyright © 1997 by American Heart Association
Hart R, Sherman D, Miller V, Easton JD. Diagnosis and management of ischemic stroke: selected controversies. Curr Probl Cardiol. 1983;8:43-53.
Bogousslavsky J, Van Melle G, Regli F. The Lausanne Stroke Register: analysis of 1000 consecutive patients with first stroke. Stroke. 1988;19:1083-1092.
Nencini P, Inzitari D, Baruffi MC, Fratiglioni L, Gagliardi R, Benvenuti L, Buccheri AM, Cecchi L, Passigli A, Rosselli A, Amaducci L. Incidence of stroke in young adults in Florence, Italy. Stroke. 1988;19:977-981.
van der Berg JSP, Limburg M. Ischemic stroke in the young: influence of diagnostic criteria. Cerebrovasc Dis. 1993;3:227-230.
Johnson CJ, Kittner SJ, McCarter RJ, Sloan MA, Stern BJ, Buchholz D, Price TR. Interrater reliability of an etiologic classification of ischemic stroke. Stroke. 1995;26:46-51.
Stegmayr B, Asplund K, Wester PO. Trends in incidence, case-fatality rate, and severity of stroke in northern Sweden, 1985-1991. Stroke. 1994;25:1738-1745.
WHO Study Group on Diabetes Mellitus. World Health Organization Technical Report, Series 727. Geneva, Switzerland: World Health Organization; 1985.
Headache Classification Committee of the International Headache Society. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalgia. 1988;8(suppl 7):1-96.
Mettinger KL, Söderström CE, Allander E. Epidemiology of acute cerebrovascular disease before the age of 55 in the Stockholm county 1973-77, I: incidence and mortality rates. Stroke. 1984;15:795-800.
Kittner SJ, McCarter RJ, Sherwin RW, Sloan MA, Stern BJ, Johnson CJ, Buchholz D, Seipp MJ, Price TR. Black-white differences in stroke risk among young adults. Stroke. 1993;24(suppl I):I-13-I-15.
Rozenthul-Sorokin N, Ronen R, Tamir A, Geva H, Eldar R. Stroke in the young in Israel: incidence and outcomes. Stroke. 1996;27:838-841.
Leno C, Berciano J, Combarros O. A prospective study of stroke in young adults in Cantabria, Spain. Stroke. 1993;24:792-795.
Carolei A, Marini C, Ferranti E, Massimiliano P, Fieschi C. A prospective study of cerebral ischemia in the young: analysis of pathogenic determinants. Stroke. 1993;24:362-367.
Hart RG, Miller VT. Cerebral infarction in young adults: a practical approach. Stroke. 1983;14:110-114.
Cabanes L, Mas JL, Cohen A, Amarenco P, Cabanes PA, Oubary P, Chedru F, Guérin F, Bousser MG, de Recondo J. Atrial septal aneurysm and patent foramen ovale as risk factors for crytogenic stroke in patients less than 55 years of age. Stroke. 1993;24:1865-1873.
Alvarez J, Matias-Guiu J, Sumalia J, Molins M, Insa R, Molto’ JM, Martin R, Codina A, Martinez-Vazquez J. Ischemic stroke in young adults, I: analysis of the etiological subgroups. Acta Neurol Scand. 1988;80:28-34.
Neto JIS, Antonio CS, Fabio SRC, Americo CS. Cerebral infarction in patients aged 15 to 40 years. Stroke. 1996;27:2016-2019.
Jones EF, Kalman JM, Calafiore P, Tonkin AM, Donnan GA. Proximal aortic atheroma: an independent risk factor for cerebral ischemia. Stroke. 1995;26:218-224.
Leung DY, Black IW, Cranney GB, Walsh WF, Grimm RA, Stewart WJ, Thomas JD. Selection of patients for transesophageal echocardiography after stroke and systemic events: role of transthoracic echocardiography. Stroke. 1995;26:1820-1824.
Smoker WRK, Biller J, Hingtgen WL. Angiography of nonhemorrhagic cerebral infarction in young adults. Stroke. 1987;18:708-711.
Bogousslavsky J, Garazi S, Jeanreneaud X, Aebischer N, Van Melle G. Stroke recurrence in patients with patent foramen ovale: the Lausanne study. Neurology. 1996;46:1301-1305.
Job FP, Ringelstein EB, Grafen Y, Flachskampf FA, Doherty C, Stockmanns A, Hanrath P. Comparison of transcranial contrast Doppler sonography and transesophageal contrast echocardiography for the detection of patent foramen ovale in young stroke patients. Am J Cardiol. 1994;74:381-384.
Homma S, Di Tullio MR, Sacco RL, Mihalatos D, Li Mandri G, Mohr JP. Characteristics of patent foramen ovale associated with cryptogenic stroke: a biplane transesophageal echocardiographic study. Stroke. 1994;25:582-586.
Martinez HR, Rangel-Guerra RA, Marfil LJ. Ischemic stroke due to deficiency of coagulation inhibitors: report of 10 young adults. Stroke. 1993;24:19-25.
Barinagarrementeria F, Cantu-Brito C, Aurora P, Izaguirre R. Prothrombotic states in young people with idiopathic stroke. Stroke. 1994;25:287-290.
Montalban J, Rio J, Khamastha M, Davalos A, Codina M, Swana GT, Calcagnotto ME, Sumalla J, Mederer S, Gil A, Hughes GRV, Codina A. Value of immunologic testing in stroke patients: a prospective multicenter study. Stroke. 1994;25:2412-2415.
Bevan H, Sharma K, Bradley W. Stroke in young adults. Stroke. 1990;21:382-386.
Lipton RB, Stewart WF. Migraine in the United States: a review of epidemiology and health care use. Neurology. 1993;43(suppl 3):S6-S10.
Lidegaard O. Oral contraception and risk of a cerebral thromboembolic attack: results of a case-control study. BMJ. 1993;306:956-963.
Tzourio C, Tehindrazanarivelo A, Iglesias S, Alperovitch A, Chedru F, dÀnglejan C-J, Bousser M. Case-control study of migraine and risk of ischemic stroke in young women. BMJ. 1995;310:830-831.
Brandt T, Hausser I, Orbeck E, Muller-Kuppers M, Anton-Lamprecht I, Hacke W. Connective tissue disorder in patients with spontaneous cervical artery dissections. Cerebrovasc Dis. 1996;6(suppl):1-31. Abstract.