Cerebral Infarction in Patients Aged 15 to 40 Years
Background and Purpose Cerebral infarction among young adults has been little studied in Brazil. Most studies adopt a variable upper age limit and often use the term “young” to designate different age ranges. The objective of the present study was to determine possible differences in etiology of cerebral infarction in specific age subpopulations.
Methods We studied 106 patients aged 15 to 40 years and divided them into two subpopulations by age at time of ictus (ie, 15 to 29 years and 30 to 40 years). On the basis of Trial of ORG 10172 in Acute Stroke Treatment (TOAST) criteria, we classified the patients into the following five groups: (1) large-artery atherosclerosis, (2) small-vessel occlusion or lacunae, (3) cardioembolism, (4) other determined causes, and (5) undetermined causes. The proportions for each group were compared between the two age ranges to determine whether there was a significant difference in group distribution.
Results The proportions were different between the two age ranges, with a predominance of women and TOAST-criterion group 4 in the 15- to 29-year age range. In the 30- to 40-year age range, there was a predominance of men and of TOAST-criterion group 3.
Conclusions The proportions of the groups in the two age ranges were significantly different for lacunar infarctions and infarctions of other etiologies. This observation suggests that different age ranges present characteristically different distributions and therefore should be considered separately when investigated. The older age range already presents features that suggest a transition to the spectrum observed among the aged.
Cerebral infarctions frequently occur among the aged and in the past were considered to be rare among young individuals. Recently, Bogousslavsky and Pierre,1 in a study of 1638 cases of CI, identified 202 patients (12.3%) with first-ever ischemic stroke in patients younger than 45 years. Regardless of the percentage of cases occurring among young individuals, most authors today agree that among victims of CI, young people are not rare.1 2 3 4 5 6 7 8 9 10 11 12 13 14 Another common conclusion is the existence of a much broader spectrum of pathologies involved in CI etiology among young patients compared with patients older than 50 years. However, available studies do not agree about the upper age limit to consider a patient “young,” which varies between 30 and 50 years.1 2 3 4 5 6 7 8 9 10 11 12 13 14 This lack of definition of an age range, as well as the absence of standardization of criteria for the classification of different CI subtypes, impairs the comparison of results obtained in different studies.
The objective of the present study was to classify a population of patients with CI aged 15 to 40 years in two different manners: (1) by subgroup, according to etiology on the basis of the TOAST classification15 and (2) by age range (15 to 29 years and 30 to 40 years), according to age at ictus.
Subjects and Methods
We studied 106 patients aged 15 to 40 years with a diagnosis of CI admitted to the University Hospital of Ribeirão Preto, University of Sao Paulo, from January 1990 through February 1994. During this period, we had 2545 admissions with the diagnosis of ischemic stroke, and 270 of these (10.6%) were patients aged 15 to 40 years. All patients underwent a basic examination protocol that included skull CT; chest x-ray; electrocardiogram; Doppler echocardiogram; B-mode carotid ultrasonography; and laboratory tests such as creatinine, blood glucose, electrolytes, lipid profile (triglycerides, total cholesterol, and fractions), complete blood counts (counts of the three series, differential leukocyte counts, and hemacytometric indices), erythrocyte sedimentation rate, protein electrophoresis, antinuclear factor, autoantibodies (anti-SM, anti-SSA, and anti-RNA), anticardiolipin antibody, partial activated thromboplastin time, prothrombin time, lupus inhibitor, serology for syphilis (Wassermann and VDRL), and serology for Chagas' disease (immunofluorescence reaction and complement fixation test). Specific studies for the detection of natural anticoagulant deficiency, such as measurement of protein C, protein S, and antithrombin III, were carried out for patients with an undetermined diagnosis and when personal or family history indicated a prothrombotic disorder. Cerebral angiography was indicated in all cases in which etiology was not determined by noninvasive tests. Patients initially included in the evaluation protocol but considered to have been incompletely investigated after the study period were excluded from the study.
After investigation, we classified the patients according to the criteria proposed by TOAST15 into five subgroups: (1) large-artery atherosclerosis, (2) small-vessel occlusion or lacunes, (3) cardioembolism, (4) stroke of other determined cause, and (5) stroke of undetermined cause. The classifications were reviewed independently by two clinical neurologists of the cerebrovascular disease sector, and disagreements were discussed by all investigators, including an experienced neuroradiologist, until a consensual decision was reached.
After establishing the classification, we divided the patients into two subpopulations according to age at the time of CI: 15 to 29 years and 30 to 40 years. After determining the proportions of the subgroups in each age range and submitting the data to statistical analysis (χ2 test and confidence intervals), we looked for characteristics that would permit us to establish distinct subpopulations in these subgroups.
Of the 106 studied patients with CI, 55 were men and 51 were women, with 38 belonging to age range 1 (15 to 29 years) and 68 cases to age range 2 (30 to 40 years). The χ2 test showed a significant difference at the 5% level in the proportion of patients in age ranges 1 and 2. Women predominated in age range 1 with 26 cases (68.4%), and men predominated in age range 2 with 43 cases (63.2%), with a statistically significant difference by χ2 test (Fig 1⇓). When the confidence intervals were calculated, the proportions of men and women in age ranges 1 and 2 were found to be different.
We performed cerebral angiography in 60 patients for a total of 66 studies and a 75.8% rate of abnormal examinations. Transthoracic echocardiogram was performed in 96 patients, with 31.2% of the examinations demonstrating a potential source for cardioembolism. According to the TOAST15 classification, the 106 patients were distributed as follows: group 1, ATR (9 cases, 8.5%); group 2, LAC (13 cases, 12.3%); group 3, CEMB (30 cases, 28.3%); group 4, OTH (37 cases, 34.9%); and group 5, UND (17 cases, 16%). By comparing this distribution by 2×2 group pairing and using the χ2 test when the total number of observations was 30 or more (n≥30), we observed that (1) the proportion of ATR was significantly different from the proportion of CEMB and OTH; (2) the proportion of LAC was also significantly different from the proportion of CEMB and OTH; (3) the proportion of CEMB did not differ significantly from the proportion of OTH and UND; and (4) the proportion of OTH was significantly different from the proportion of UND.
ATR was diagnosed in 9 patients as the cause of CI. Only 1 patient was younger than 30 years old at the time of the ictus. All patients in this group presented at least two risk factors for ATR. Seven patients had arterial hypertension, and 7 were heavy smokers. Other risk factors detected were left ventricular hypertrophy in 5, heavy drinking in 4, and hypercholesterolemia in 3.
LACs were diagnosed in 13 patients, only 1 of whom was younger than 30 years. The most common clinical syndrome was pure motor hemiparesis in 11 cases. In all cases we ruled out potential sources of CEMB and significant (≥50%) atherosclerotic carotid lesions. Seven patients (54%) showed mild to severe left ventricle hypertrophy in the transthoracic echocardiogram. All patients in this group were hypertensive, and only 2 (15.4%) had diabetes mellitus.
We diagnosed cardioembolic infarctions in 30 patients. This type of infarction occurred at almost identical proportions in the two age ranges. The sources of CEMB detected are listed in Table 1⇓. With respect to etiology, we detected rheumatic disease in 10 (33.3%), mitral valve prolapse in 5 (16.7%), Chagas' heart disease in 4 (13.3%), dilated cardiomyopathy in 4 (13.3%), ischemic cardiopathy in 3 (10%), “idiopathic” segmental hypokinesis in 3 (10%), and aseptic endocarditis in 1 (3.3%).
The OTH group had the largest number of patients, ie, 37 (34.9%). Another important characteristic of this group compared with the others was the predominance of women over men, with a significant difference determined by χ2 test (Fig 2⇓). In this group we diagnosed more than 15 different causes of CI (Table 2⇓). The subgroup of hematologic causes (5 cases) included sickle cell anemia (2), protein C deficiency (2), and protein S deficiency (1). The 1 case of vasospasm secondary to subarachnoid hemorrhage was included because the primary subarachnoid hemorrhage was misdiagnosed and the CI was the first diagnosis.
The etiology of CI was undetermined in 17 patients (16%). Most of these cases (88.2%) were assigned to this group because the findings of the investigation did not permit us to assign them to any other group, and only 2 showed a fully negative investigation. It should be emphasized that 12 patients (70.6%) in this group had at least two risk factors for atheromatous disease, but neuroimaging examination did not confirm this diagnostic impression.
By subdividing the total population into two groups according to age at the time of ictus (ie, 15 to 29 years and 30 to 40 years) and then applying the TOAST classification, we obtained the following distribution for age range 1 (38 patients): ATR, 1 (2.6%); LAC, 1 (2.6%); CEMB, 10 (26.3%); OTH, 21 (55.3%); and UND, 5 (13.2%). In contrast, for age range 2 (68 patients), we obtained the following distribution: ATR, 8 (11.8%); LAC, 12 (17.65%); CEMB, 20 (29.4%); OTH, 16 (23.5%); and UND, 12 (17.65%) (Fig 3⇓).
By comparing the two age range groups on the basis of the confidence intervals, we observed that (1) the proportions of groups ATR, CEMB, and UND did not differ significantly between the two age ranges and (2) the proportions of LAC and OTH were significantly different between age ranges.
CI is not rare among young people, as was previously assumed in early reports. Our experience from the stroke registry of a University Hospital located in Ribeirão Preto, southwestern Brazil, corroborated this hypothesis, showing that 10.6% of 2445 admissions for ischemic stroke during a 4-year period were for patients aged 40 years or less. Most of our cases were in the 30- to 40-year age range, coinciding with the observation that the incidence of CI increases with age.
The etiologic spectrum is quite broad.1 2 3 4 5 6 7 8 9 10 11 12 13 14 Hart and Miller14 suggested that an “aggressive” investigation permits diagnostic clarification in as many as 90% of cases. They also propose that cerebral angiography be performed as early as possible in all cases not clarified by other noninvasive diagnostic tests. Lisovoski and Rousseaux17 studied 148 patients with early cerebral angiography, and this procedure revealed the cause in most cases. In our study, cerebral angiography was decisive for the diagnosis of several patients and revealed abnormalities in 75.8% of the studies performed. This proportion was practically identical to that reported by Smoker et al,18 who detected abnormalities in 76% of 93 young patients who underwent angiography. Another important diagnostic test for our series was transthoracic Doppler echocardiogram. Of the 96 patients examined, 30 (31.2%) presented potential sources of CEMB, results similar to those reported by Biller et al,19 who diagnosed cardioembolic sources in 27.1% of 96 patients. It should be pointed out that this expressive proportion was obtained without the aid of special techniques such as contrast echocardiography or transesophageal study, which are known to greatly increase the chance of identifying sources often inaccessible by conventional study (patent foramen ovale, atrial septum aneurysm, etc).20
According to most investigators, the number of atherothrombotic infarctions is small among young patients.1 3 4 5 6 7 The high incidence among older patients is in contrast to an almost negligible percentage among patients younger than 30 years.9 In our series, we detected 8.5% of patients with atherothrombotic infarction. Adams et al2 detected a 25% rate among 144 patients aged 15 to 45 years, and Bevan et al10 detected a 31% rate among patients aged 25 to 40 years, results that are significantly different from ours.
In our series, lacunar infarctions were significantly more numerous in the 30- to 40-year age range compared with the 15- to 29-year age range. Today, some authors still question the validity of the lacunar hypothesis.21 However, classifications that consider this group separately are currently in use and are almost universally accepted by many authors.22 The significant increase in the proportion of lacunar infarctions observed here from 30 to 40 years may suggest that degenerative arteriolar alterations occur earlier than expected in patients with severe systemic arterial hypertension.
In the group of cardioembolic infarctions, the subgroup with synthetic valve prostheses represented one fifth of the total cases and was the most numerous. In the series of Adams et al,2 this subgroup was also the most numerous, together with the group with rheumatic heart disease. In contrast, among the cases of Bogousslavsky and Pierre,1 synthetic valve prostheses were less important than mitral valve prolapse and patent foramen ovale. Considering cerebral embolisms as a whole for all ages, nonrheumatic atrial fibrillation is the most frequent cause,23 24 a fact that is not observed in young patients, among whom there is a predominance of rheumatic heart disease with or without atrial fibrillation, patent foramen ovale, and mitral valve prolapse.1 2 3 8 14 Rheumatic valvular disease and dilated cardiomyopathy also occurred in a significant number of patients. It should be pointed out that rheumatic heart disease was the most important cause, since all patients with a synthetic valve prosthesis had rheumatic lesions.
A finding peculiar to our study environment was Chagas' heart disease, diagnosed in 4 patients with emboligenic cardiopathies. In the series reported in the literature, this type of heart disease was cited only as a potential source1 2 3 4 5 6 7 8 9 10 11 12 13 14 but was not diagnosed in any patient. Thromboembolic phenomena are frequent in chronic chagasic cardiopathy.25 Oliveira et al26 reported a large incidence of intramural thrombus in chagasic patients with apical aneurysms without congestive heart failure. This observation is very important because it can explain ischemic stroke in patients without severe forms of dilated cardiomyopathy. In our cases, 2 had akinetic segments (apical aneurysms), one had PVM with mild dilatation of the left ventricle, and the last had severe dilated cardiomyopathy.
Group OTH represented the largest number of cases in our series and represented more than 50% of age range 1. Among the etiologies, particularly important were hematologic abnormalities, the most numerous subgroup in this group. According to Hart and Kanter,27 this proportion of hematologic abnormalities may be higher than 4% for all CIs in the young, and this observation is supported by our results (4.7%). Another frequent pathology in the OTH group was “arterial dissection,” corresponding to 3.8% of the total but remaining much behind the 20.8% rate reported by Bogousslavsky and Pierre.1 Moyamoya arteriopathy also presented a high incidence identical to that of dissection cases. This percentage was also similar to the 3.5% rate detected by Adams et al.2
The UND group was smaller for our younger patients than older persons, but this difference was not significant; our results (16%) are similar to those published in the series recently reported by Adams et al,28 who found 16.6% undetermined CIs among 329 patients aged 30 years and younger according to the TOAST criteria.
Group-by-group comparisons between different age ranges of young patients are very rare in the literature.1 2 3 4 5 6 7 8 9 10 11 12 13 14 28 Only Bogousslavsky and Pierre1 constructed a comparative table, and, even so, they did not apply the TOAST criteria. Some of our data were similar to those obtained by these authors,1 who also detected a predominance of the other etiology groups, immediately followed by the CEMB group. However, an important difference was the detection of a large number of lacunae and a significant decrease in the proportion of the OTH group in the 31- to 45-year age range, which suggests the existence of two different populations related to age limits.
CIs among young patients require extensive investigation and result from a broad spectrum of etiologic diagnoses. Our results indicate differences in proportion between the groups of the TOAST15 classification when the 15- to 29-year age range is compared with the 30- to 40-year age range. The LAC and OTH groups presented significantly different proportions as determined by confidence intervals. These findings suggest that these age ranges should be considered separately, since the first one, from 15 to 29 years, has a highly peculiar characteristic distribution that is typical of young or very young people (multiplicity of etiologies and cardioembolic stroke), whereas from 30 to 40 years of age, a pattern starts that is similar to that observed among aged patients.
Selected Abbreviations and Acronyms
|LAC||=||occlusion of small vessels or lacunar infarction|
|TOAST||=||Trial of ORG 10172 in Acute Stroke Treatment|
- Received June 3, 1996.
- Revision received July 30, 1996.
- Accepted July 30, 1996.
- Copyright © 1996 by American Heart Association
Abraham J, Shetty G, Jose CJ. Strokes in the young. Stroke. 1971;2:258-267.
Grindal AB, Cohen RJ, Saul RF, Taylor JR. Cerebral infarction in young adults. Stroke. 1978;9:39-42.
Bevan H, Sharma K, Bradley W. Stroke in young adults. Stroke. 1990;21:382-386.
Hart RG, Miller VT. Cerebral infarction in young adults: a practical approach. Stroke. 1983;14:110-114.
Adams HP Jr, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, Marsh EE III, and TOAST Investigators. Classification of subtype of acute ischemic stroke: definitions for use in a multicenter clinical trial. Stroke. 1993;24:35-41.
Lisovoski F, Rousseaux P. Cerebral infarction in young people: a study of 148 patients with early cerebral angiography. J Neurol Neurosurg Psychiatry. 1991;54:576-579.
Smoker WRK, Biller J, Hingtgen WL, Adams HP Jr, Toffol GJ. Angiography of nonhemorrhagic cerebral infarction in young adults. Stroke. 1987;18:708-711.
Biller J, Johnson MR, Adams HP Jr, Kerber RE, Toffol GJ, Butler MJ. Echocardiographic evaluation of young adults with nonhemorrhagic cerebral infarction. Stroke. 1986;17:608-612.
Millikan C, Futrell N. The fallacy of the lacune hypothesis. Stroke. 1990;21:1251-1257.
Chamorro A, Sacco RL, Mohr JP, Foulkes MA, Kase CS, Tatemichi TK, Wolf PA, Price TR, Hier DB. Clinical-computed tomographic correlations of lacunar infarction in the stroke data bank. Stroke. 1991;22:175-181.
Oliveira JSM, Mello De Oliveira JA, Frederigue U Jr, Lima Filho EC. Apical aneurysm of Chagas' heart disease. Br Heart J. 1981;46:432-437.
Hart RG, Kanter MC. Hematologic disorders and ischemic stroke. Stroke. 1990;21:111-121.