J Neurol (2012) 259:653–659
DOI 10.1007/s00415-011-6234-3
ORIGINAL COMMUNICATION
Ischemic stroke in young adults: classification and risk factors
Anastasios Chatzikonstantinou • Marc E. Wolf •
Michael G. Hennerici
Received: 5 August 2011 / Revised: 26 August 2011 / Accepted: 30 August 2011 / Published online: 17 September 2011
Ó Springer-Verlag 2011
Abstract Our aim was to analyze an important subgroup
represented by young adult patients (19–45 years) with
acute ischemic stroke according to stroke classification
(including the novel ASCO score), infarct types and risk
factors. All patients up to 45 years of age with an acute
ischemic stroke confirmed by MRI and treated in our stroke
unit from 2006 to 2009 were recruited for this study.
Patients were neurologically examined and underwent
thorough stroke work-up. One hundred four patients (58
women, 46 men) with a mean age of 38 ± 6.9 years were
evaluated. The mean NIHSS score (±SD) was 3 ± 5 on
admission and 1 ± 4 on discharge. The classification using
TOAST/ASCO (grade 1) was as follows: Macroangio-
pathic 10.6%/8.7%, cardiac origin 21.2%/10.6%, micro-
angiopathic 9.6%/9.6%, other causes 19.2%/13.5% and
undetermined 39.4%/19.2% (for A0S0C0O0). The most
common risk factors were smoking (55.2%), hypertension
(31.4%) and hyperlipidemia (27.6%). Twenty nine of 74
patients with TEE (39.2%) had a patent foramen ovale
(PFO). Hypoplastic posterior circulation (21.9%) and
migraine (21.0%) were also quite common. Young adult
ischemic stroke patients share many of the characteristic
risk factors with the general elderly ischemic stroke pop-
ulation. If regular work-up includes TEE, a high percentage
of young patients reveal comorbidities with PFO, hypo-
plasia of the posterior circulation and migraine. The ASCO
classification should be favored for a better classification of
coexisting stroke subtypes and lower number of undeter-
mined etiologies in this patient group.
A. Chatzikonstantinou (&)
M. E. Wolf
M. G. Hennerici
Department of Neurology, UniversitätsMedizin Mannheim,
University of Heidelberg, Theodor-Kutzer-Ufer 1-3,
68167 Mannheim, Germany
e-mail: anastasios.chatzikonstantinou@umm.de
Keywords Brain ischemia
Stroke in young

Risk factors
ASCO
TOAST
Introduction
Ischemic stroke in young adult patients between 18 and
45 years is relatively rare (accounting for 2–12% of cere-
bral infarcts), but often poses a particular challenge with
regard to the diagnosis of stroke subtypes [1–3]. This is not
only important for best secondary prevention recommen-
dations, but also because stroke etiologies and risk factors
are generally considered to be very different from those of
older patients [3–6]. Embolism due to cardiac abnormali-
ties, arterial dissection, migraine and specific conditions of
hypercoagulation rather than large or small atherosclerotic
arteriopathies are thought to be the most common causes of
stroke in the young population [5–7]. Although younger
patients usually receive complete stroke work-up, the fact
that a large number of strokes are of undetermined etiology
(usually [1/3 of all cases) is alarming [2, 8, 9]. One pos-
sible explanation is the continuous use of an outdated
stroke subtype classification, the Trial of Org 10172 in
Acute Stroke Treatment (TOAST) score, which was
introduced for a randomized acute stroke trial using hep-
arin in 1993 [10]. The aim of our study was to reevaluate
the actual risk factor pattern in young ischemic stroke
patients and to compare their etiologies using the more
descriptive ASCO score [11].
Patients and methods
One hundred four patients (58 women and 46 men) ranging
between 19 and 45 years of age (mean age 38 ± 6.9 years)
123
654 J Neurol (2012) 259:653–659

and consecutively admitted to our stroke unit with the Table 1 Patient characteristics and risk factors (n = 104 total
diagnosis of acute ischemic stroke were recruited for this population)
study. The diagnosis of ischemic stroke was based on Parameter Value
clinical features and MRI in all patients. Each patient
Age
underwent a clinical neurological examination [including
National Institute of Health Stroke Scale (NIHSS) scores] Mean (years ± SD) 38.4 ± 6.9
and received a thorough stroke work-up according to the Range (years) 19–45
guidelines of the European Stroke Organization, which was Women [n (%)] 58 (55.8%)
documented in our stroke registry in detail: Brain MRI and Thrombolysed patients [n (%)] 23 (22.1%)
MRA [including T1-, T2-, T2*-, FLAIR- and diffusion- 3 h time window 19 (18.3%)
weighted images (DWI) and time-of-flight (TOF) angiog- ECASS III time window 3 (2.9%)
raphy], extra- and intracranial ultrasound cerebrovascular DWI/PWI-Mismatch [4.5 h 1 (0.9%)
after symptom onset
imaging, ECG-monitoring and standard laboratory blood
NIHSS score (median ± SD)
tests. Most of the patients (74/104; 71.2%) also received
transesophageal echocardiograms (TEE). The rest refused On admission 3 (±5) (range 0–25)
to undergo TEE or could not be examined during the period On discharge 1 (±4) (range 0–18)
of hospitalization. In patients with no obvious stroke eti- Work-up [n (%)]
ology or risk factors, thorough coagulation and vasculitis MRI (incl. MRA) 104 (100%)
tests were performed. Infarct types and localization on MRI Duplex vascular imaging 104 (100%)
were recorded, as well as history of stroke and/or older ECG-Monitoring 104 (100%)
cerebral ischemia visible on MRI. Furthermore, patients’ TEE 74 (71.2%)
characteristics and risk factors [patent foramen ovale Infarct type [n (%)]
(PFO), mitral valve prolapse (MVP), atrial fibrillation, Territorial 49 (47.1%)
history of coronary and/or peripheral artery diseases, Lacunar 14 (13.5%)
arterial hypertension (history/treatment or hypertensive Other (multiple territories, emboligenic 41 (39.4%)
blood pressure on admission), hyperlipidemia (history/ distribution)
treatment or high fasting cholesterol levels), diabetes Infarct territory (circulation)
mellitus (history/treatment or high fasting glucose levels Anterior only 70 (67.3%)
and elevated HbA1c), obesity, smoking, history of Posterior only 26 (25.0%)
migraine and positive stroke family history] were recorded Both 8 (7.7%)
in our Stroke Unit database and analyzed. As an additional Risk factors [n (%)]
parameter, we chose to include hypoplastic vertebral Atrial fibrillation 3 (2.9%)
arteries (defined as lumen diameter B2 mm in MR-TOF) in Coronary heart disease 1 (0.9%)
the analysis [12]. To classify ischemic strokes according to Peripheral artery disease 2 (1.9%)
etiology, we used both the TOAST [10] and ASCO [11] Cerebral large artery disease 14 (13.5%)
scores and compared their results. Diabetes 7 (6.7%)
Arterial hypertension 33 (31.4%)
Hyperlipidemia 29 (27.6%)
Results Obesity 8 (7.6%)
Smoking 58 (55.2%)
Patient characteristics and ischemia types Migraine 22 (21.0%)
Previous stroke: symptomatic/asymptomatic 3 (2.9%)/10 (9.6%)
The patient characteristics are summarized in Table 1. Hypoplastic vertebral artery 23 (21.9%)
There were slightly more women [58 (55.2%)] than men Stroke family history 6 (5.7%)
[46 (43.8%)]. We were able to apply thrombolytic treat- Patent foramen ovale (all) 29 (27.9%)
ment to 23 (22.1%) patients, the majority of them (19/23; Arterial dissection (carotid/vertebral artery) 16 (15.4%)
82.6%) in the standard 3-h-window, the rest (3/23; 13%) Other cardiac disease 5 (4.8%)
received thrombolytic treatment in the extended 3–4.5-h- Hypercoagulable state 9 (8.6%)
window (according to the results and recommendations of Other (e.g., drugs, antiphospholipid 9 (8.6%)
the ECASS III study [13]) and in 1/23 (4.3%) using mis- syndrome, vasculitis)
match of DWI and perfusion-weighted images (PWI)

123
J Neurol (2012) 259:653–659
Fig. 1 Risk factors in all
patients (orange, n = 104) and
in patients with non-cardiac
stroke (blue, n = 83). AF atrial
fibrillation, CHD coronary heart
disease, PAD peripheral artery
disease, VA vertebral artery
Fig. 2 Patent foramen ovale. Pie chart: distribution of different PFO
grades (ASD aortic septal defect). Tables: the frequency of PFO in the
different etiology stroke subtype groups using TOAST (LA large
artery, CO cardiac origin, SV small vessel, OD other determined, UD
([4.5 h after onset of symptoms) [14]. The median NIHSS
score on admission was 3 ± 5 (range 0–25) and on dis-
charge 1 ± 4 (range 0–18). Most of the patients (49/104;
47.1%) suffered a territorial and 14/104 (13.5%) a lacunar
stroke, while 41/104 (39.4%) had other stroke morpholo-
gies, mostly of emboligenic distribution. In 70/104 patients
(67.3%), ischemia was located in the anterior circulation
only, in 26/104 patients (25.0%) in the posterior circula-
tion, while in 8/104 cases (7.7%) both vascular territories
were simultaneously affected.
Risk factors
Risk factors are presented in Table 1 and are visualized in
Fig. 1, which also shows the frequency of risk factors in all
patients compared to patients with non-cardiac stroke, in
which cardioembolic risk factors such as atrial fibrillation are
usually absent. Results were similar for these two groups;
the following numbers refer to the whole population.
Smoking (58; 55.8%) was the most common risk factor
in our population, followed by arterial hypertension (33;
31.7%), hyperlipidemia (29; 27.9%) and cerebral artery
655
undetermined) and ASCO (A atherothrombosis, S small vessel,
C cardiac origin, O other). On the right side are the results only for the
patients who received TEE
atherosclerosis (14; 13.5%). Classic atherosclerotic risk
factors (apart from smoking) were found in 59/104 (56.7%)
patients. While not being among the classic stroke risk fac-
tors and not considered by the TOAST or ASCO criteria,
migraine (22; 21.2%) was quite commonly found among our
young stroke patients. Hypoplastic vertebral arteries were
also a common finding (23; 22.1%). Atrial fibrillation was
found only in three patients (2.9%), all of whom had suffered
cardioembolic stroke. Arterial dissection was relatively
common and found in 16 (15.4%) of patients.
Patent foramen ovale
PFO was the particular focus of our attention, as most of
our patients were also examined using TEE, in which PFOs
are easy to detect. Figure 2 shows the corresponding
results. PFO was identified in 29/74 (27.9%) patients. Most
commonly (18/29; 62%), it was a PFO III° with or without
atrial septal defect (ASD). If we only consider patients with
TEE (74), which definitely reveals PFO, the percentage of
PFO increases to 39.2%. Three quarters of patients (15 of
20; 75%) with a cardioembolic stroke according to the
123
656
Fig. 3 Comparison of ASCO and TOAST classification results.
Comparison of TOAST and ASCO results for the different etiology
groups. A atherothrombosis/large artery, S small vessel disease,
C cardioembolic/cardiac origin, O other (determined). The numbers
TOAST classification and a TEE showed a PFO. Using the
ASCO classification, patients with certain (C1, n = 8) or
possible (C2; n = 11) cardioembolic stroke and TEE had a
PFO in a similarly high percentage (63.2%). A combina-
tion of PFO and other possible stroke etiologies was
present in 11/29 (37.9%) cases.
ASCO and TOAST comparison
We also compared how the two stroke etiology scores
(TOAST and the newer ASCO) classify strokes in this
young population. Figure 3 shows the comparison of the
two scores. In greater detail, the classification using
TOAST/ASCO (only grade 1) was as follows: Athero-
thrombotic (large artery) 10.6%/8.7%, cardiac origin
21.2%/10.6%, small vessel disease 9.6%/9.6% and other
causes 19.2%/13.5%. The TOAST score delivered the
result ‘‘undetermined’’ in 41 cases (39.4%). The corre-
sponding score in ASCO would be A0S0C0O0, because it
delivers no additional information, like the last TOAST
category. Only 20 patients (19.2%) received this classifi-
cation, about half as many as in TOAST.
Discussion
In this study, we describe the risk factors and etiopatho-
genetic classification in 104 young adult patients with
ischemic stroke aged up to and including 45 years. Age
range (19–45 years) and mean age (38.4 ± 6.9 years) were
very similar to other recent studies on young stroke patients
[5, 7, 9, 15, 16].
The rate of thrombolytic treatment was about 22% and
thus did not substantially differ from the rate in the general
ischemic stroke population in our stroke center. This seems
surprising at first, as one might expect younger patients to
123
J Neurol (2012) 259:653–659
1, 2 refer to the corresponding grade in ASCO [11]. The ASCO score
A0S0C0O0 was taken as the corresponding TOAST category
‘‘undetermined’’
be more sensitive and better informed than older ones. On
the other hand, thrombolysis rate is generally high in our
region, due to optimization of publicity, informing of the
general population and training of emergency personnel
[17]. In addition to this, many of the patients had no, or
only minor, neurological deficits on admission, thus not
requiring thrombolysis (58.7% of patients had a NIHSS
score \3 on admission).
Our young stroke population received a very thorough
work-up. In addition to MRI, MRA, vascular duplex
imaging and ECG monitoring, which are standard in acute
stroke management, TEE was also performed in over 70%
of our patients. This is considerably more than in most
other studies (5–60%) which have investigated stroke
causes and risk factors in young ischemic stroke patients
and may subsequently have influenced our results, espe-
cially concerning PFO [5, 7, 9, 15].
Risk factors
Cerebrovascular risk factors are generally thought to be
less important in younger patients than in older ones [3]. In
our population, the most common risk factor was smoking
(55.2%). Other studies on young stroke patients have also
found a high percentage of smokers; however, not as high
as in our study (37–44%), with the exception of the study
performed by Spengos et al., who found almost 60% of
smokers in their population of Greek stroke patients [5, 9,
15, 18, 19]. Hypertension and hyperlipidemia were also
quite common in these studies (22–56 and 17–59%,
respectively), in approximately the same levels as in our
study (31.4 and 27.6%, respectively). Diabetes was found
in 6.7% of our patients, i.e., with a similar frequency as in
other comparable studies (5–17%) [5, 9, 15, 18, 19].
Overall, 58.7% patients had one or more atherosclerotic
risk factors (not counting smoking), which is close to what
J Neurol (2012) 259:653–659
one might have expected from an elderly population.
Surprisingly, we found a high number of patients with
silent previous strokes as evidenced on MRI (10/104;
9.6%) and there were three patients (2.9%) with previous,
symptomatic strokes. This parameter was not always
included in previous studies. Putaala et al. [5], in their
study with 1008 young stroke patients with first-ever
ischemic stroke, found a similar percentage (12.5%) of
patients with previous silent infarcts. Spengos et al. [19]
found 13.4% of patients with a history of previous cere-
brovascular events in their population of 253 young stroke
patients, including, however, transient ischemic attacks.
Janssen et al. [18] found a very high percentage (47.9%)
with a history of vascular events; this, however, included
cardiovascular events as well.
Migraine may be a less-well documented and even
controversial risk factor for stroke, but an association
between migraine and stroke, especially in young patients,
has indeed been shown in the literature [6, 20–22]. Quite a
large percentage of our patients had a history of migraine
(21%). In studies with similar populations, percentages
varied between 13 and 20% [5, 18, 19]. The role of
hypoplastic posterior circulation, in particular affecting
both vertebral and the basilar arteries, in ischemic stroke is
not clear, whereas unilateral and sometimes bilateral ver-
tebral artery hypoplasia is not rare in the general population
[23]. Since this anomaly is probably congenital, it may play
an underestimated role in a population of younger patients
than in a general stroke population. Not many studies have
investigated this topic. A study by Park et al. [23] found
approximately 35% of a general ischemic stroke population
(n = 529) and 26.5% of 303 healthy people to have
hypoplastic vertebral arteries. We had 23 (21.9%) patients
with hypoplasia of either or, less frequently, both vertebral
arteries, which would signify that this feature does not
seem to be of particular importance in young stroke
patients.
PFO is present in about 25% of the general population
and many case–control studies have shown a statistical
association between PFO and stroke [6, 24]. Numbers on
the frequency of PFO in young ischemic stroke patients
vary in the literature. In the largest series with 1,008
patients, Putaala et al. found PFO in 10.5% of all patients,
while Spengos et al. in 8.3% and Janssen et al. in only 6.4%
[5, 18, 19]. However, in the last two studies, the minority of
patients received a TEE, which is the most sensitive
diagnostic test for the detection of PFO. The study with the
highest percentage of PFO (Putaala et al.) also had the
higher percentage of TEEs performed (60%). We found
27.9% of our population to have a PFO, which would be
about the same as the general population. However, most
(but not all) of our patients received a TEE. If we consider
only the 74 (71.2%) patients with TEE, the percentage of
657
PFO increases to 39.2%, which is surprisingly high. Per-
haps less surprising, the frequency of PFO is even higher in
patients with stroke of cardiac origin (63.2–75%, according
to the classification of stroke used––A1 and A2 grades in
ASCO and ‘‘cardiac origin’’ in TOAST). Of course, the
high percentage of PFO in ‘‘cardiac origin’’ strokes as
defined by TOAST is at least partially explained by the fact
that TOAST considers the presence of PFO as a criterion
for the ‘‘cardiac origin’’ classification, whereas ASCO
requires PFO to be associated with in situ thrombosis or
pulmonary embolism or deep vein thrombosis preceding
the brain infarction to classify the stroke as definitely
cardioembolic (grade 1).
These findings suggest that PFO may play a more
important role in young stroke adults than previously
thought, although the mechanism of stroke in patients with
PFO, as well as its significance, are not always clear.
Further studies with larger patient numbers and more
complete cardiac work-up will be needed to clarify this
issue.
Stroke classification
Most studies on ischemic stroke in young adults have used
the TOAST classification of stroke subtype [6]. While this
classification has some benefits and is quite simple, it does
not include risk factors in its criteria and in some cases
leaves much room for different interpretations. Further-
more, its last category ‘‘undetermined’’ offers no more
information apart from the fact that the cause of stroke
could not be determined with certainty.
Our results concerning the etiology of stroke using
TOAST were similar to other studies performed on young
stroke adults [5–7, 9, 19, 25, 26]. We found 10.6% of
patients to be in the large-vessel disease category
(6.7–26.7% in the literature), 21.2% in the cardiac origin
category (5.2–25% in the literature), 9.6% in the small-
vessel disease category (1.7–17.4% in the literature),
19.2% in the category of other defined causes (29–35% in
the literature) and 39.4% of undetermined etiology
(13.3–62.4% in the literature). As previously observed,
variations in the classification among these studies is
mainly explained by the different criteria used and by the
grade of completeness of stroke work-up [6].
The new ASCO classification was created with risk
factors in mind and with the aim of offering more infor-
mation in the appointed stroke subtypes [11]. To our
knowledge, this is the first series of young adults with
ischemic stroke to apply and compare both stroke subtype
classification systems. In the category of macroangiopathy
and other determined causes, both classification methods
delivered similar results. In the cardiac origin category, the
percentage of patients using ASCO grade 1 was about half
123
658
that of those using TOAST, but exactly the same in both
systems when including both grade 1 and 2 of ASCO. The
opposite was true in the category of microangiopathic
strokes, where TOAST and ASCO grade 1 delivered
exactly the same percentage of patients, whereas ASCO
grade 1 ? 2 showed a much higher result. These effects
may be explained by the very detailed and strict criteria
used in ASCO, as opposed to TOAST, at least in some
cases.
The last category ‘‘undetermined cause’’ was particularly
interesting, as cryptogenic strokes pose a particular chal-
lenge, especially regarding the choice of secondary preven-
tion. As already mentioned, most studies found about
one-third of strokes to be of undetermined cause (24–36%),
with some exceptions, e.g., Musolino et al. (13.3%) and Leys
et al. (62.4%), obviously depending on the completeness of
investigations [5, 7, 9, 19, 25, 26]. In our study, this per-
centage was 39.4%, slightly more than one-third of the
patients, despite quite a broad work-up. However, during
these extensive examinations, data pertaining to risk factors
and possible stroke causes were collected which are not
reflected in the TOAST classification. This data, however,
are found in the ASCO score. Even the score ‘‘A0S0C3O0’’,
for example, contains information (in this case that cardiac
problems exist), even if the cause of stroke is uncertain.
Therefore, we chose the score ‘‘A0S0C0O0’’ as the best
corresponding category to TOAST’s ‘‘undetermined cause’’.
Only about half of strokes then remained completely unde-
termined as compared to using TOAST (A0S0C0O0: 19.2%,
TOAST ‘‘undetermined cause’’: 39.4%).
The limitations of our study are the relatively small
patient number and the fact that most, but not all, patients
received a complete work-up including TEE, something
our study has in common with previous ones. However, we
believe we could demonstrate that young ischemic stroke
patients share some of the classic risk factors with elderly
stroke patients. In addition, provided corresponding
investigations including TEE studies are regularly per-
formed, a significantly increased percentage of young
stroke patients reveal comorbidities with PFO, migraine
and previous, usually asymptomatic, cerebrovascular
events. There are also arguments (more contained infor-
mation, less undetermined strokes) for preferring the
ASCO classification over the TOAST one for a better
classification of stroke subtypes in this patient group.
Conflict of interest None.
References
1. Correia M, Silva MR, Matos I, Magalhaes R, Lopes JC, Ferro JM,
Silva MC (2004) Prospective community-based study of stroke in
123
J Neurol (2012) 259:653–659
Northern Portugal: incidence and case fatality in rural and urban
populations. Stroke 35(9):2048–2053
2. Kristensen B, Malm J, Carlberg B, Stegmayr B, Backman C,
Fagerlund M, Olsson T (1997) Epidemiology and etiology of
ischemic stroke in young adults aged 18 to 44 years in northern
Sweden. Stroke 28(9):1702–1709
3. Bogousslavsky J, Pierre P (1992) Ischemic stroke in patients
under age 45. Neurol Clin 10(1):113–124
4. Lee TH, Hsu WC, Chen CJ, Chen ST (2002) Etiologic study of
young ischemic stroke in Taiwan. Stroke 33(8):1950–1955
5. Putaala J, Metso AJ, Metso TM, Konkola N, Kraemer Y, Haa-
paniemi E, Kaste M, Tatlisumak T (2009) Analysis of 1008
consecutive patients aged 15 to 49 with first-ever ischemic stroke:
the Helsinki young stroke registry. Stroke 40(4):1195–1203
6. Ferro JM, Massaro AR, Mas JL (2010) Aetiological diagnosis of
ischaemic stroke in young adults. Lancet Neurol 9(11):1085–1096
7. Varona JF, Guerra JM, Bermejo F, Molina JA, Gomez de la
Camara A (2007) Causes of ischemic stroke in young adults, and
evolution of the etiological diagnosis over the long term. Eur
Neurol 57(4):212–218
8. Carolei A, Marini C, Ferranti E, Frontoni M, Prencipe M, Fieschi
C (1993) A prospective study of cerebral ischemia in the young.
Analysis of pathogenic determinants. The national research
council study group. Stroke 24(3):362–367
9. Cerrato P, Grasso M, Imperiale D, Priano L, Baima C, Giraudo
M, Rizzuto A, Azzaro C, Lentini A, Bergamasco B (2004) Stroke
in young patients: etiopathogenesis and risk factors in different
age classes. Cerebrovasc Dis 18(2):154–159
10. Adams HPJ, Bendixen BH, Kappelle LJ, Biller J, Love BB,
Gordon DL, EEr Marsh (1993) Classification of subtype of acute
ischemic stroke. Definitions for use in a multicenter clinical trial.
TOAST. Trial of Org 10172 in Acute Stroke Treatment. Stroke
24(1):35–41
11. Amarenco P, Bogousslavsky J, Caplan LR, Donnan GA, Henn-
erici MG (2009) New approach to stroke subtyping: the A-S-C-O
(phenotypic) classification of stroke. Cerebrovasc Dis
27(5):502–508
12. Chaturvedi S, Lukovits TG, Chen W, Gorelick PB (1999)
Ischemia in the territory of a hypoplastic vertebrobasilar system.
Neurology 52(5):980–983
13. Hacke W, Kaste M, Bluhmki E, Brozman M, Davalos A, Guidetti
D, Larrue V, Lees KR, Medeghri Z, Machnig T, Schneider D, von
Kummer R, Wahlgren N, Toni D (2008) Thrombolysis with
alteplase 3 to 4.5 h after acute ischemic stroke. N Engl J Med
359(13):1317–1329
14. Kane I, Sandercock P, Wardlaw J (2007) Magnetic resonance
perfusion diffusion mismatch and thrombolysis in acute ischae-
mic stroke: a systematic review of the evidence to date. J Neurol
Neurosurg Psychiatry 78(5):485–491
15. Balci K, Utku U, Asil T, Celik Y (2011) Ischemic stroke in young
adults: risk factors, subtypes, and prognosis. Neurologist
17(1):16–20
16. Spengos K, Vemmos KN (2010) Etiology and outcome of car-
dioembolic stroke in young adults in Greece. Hellenic J Cardiol
51(2):127–132
17. Ellinger K, Koch C, Daffertshofer M, Behrens S, Luiz T (1999)
Das Mannheimer Schlaganfallprojekt. Eine Initiative zur Opti-
mierung der Versorgungsstrategie. Notfall and Rettungsmedizin
2:428–433
18. Janssen AW, de Leeuw FE, Janssen MC (2010) Risk factors for
ischemic stroke and transient ischemic attack in patients under
age 50. J Thromb Thrombolysis 31(1):85–91
19. Spengos K, Vemmos K (2010) Risk factors, etiology, and out-
come of first-ever ischemic stroke in young adults aged 15 to
45––the Athens young stroke registry. Eur J Neurol 17(11):1358–
1364
J Neurol (2012) 259:653–659
20. Camerlingo M, Romorini A, Ferrante C, Valente L, Moschini L
(2010) Migraine and cerebral infarction in young people. Neurol
Sci 31(3):293–297
21. Kurth T (2007) Migraine and ischaemic vascular events. Ceph-
alalgia 27(8):965–975
22. Wolf ME, Szabo K, Griebe M, Forster A, Gass A, Hennerici MG,
Kern R (2011) Clinical and MRI characteristics of acute
migrainous infarction. Neurology 76(22):1911–1917
23. Park JH, Kim JM, Roh JK (2007) Hypoplastic vertebral artery:
frequency and associations with ischaemic stroke territory.
J Neurol Neurosurg Psychiatry 78(9):954–958
659
24. Alsheikh-Ali AA, Thaler DE, Kent DM (2009) Patent foramen
ovale in cryptogenic stroke: incidental or pathogenic? Stroke
40(7):2349–2355
25. Leys D, Bandu L, Henon H, Lucas C, Mounier-Vehier F,
Rondepierre P, Godefroy O (2002) Clinical outcome in 287
consecutive young adults (15 to 45 years) with ischemic stroke.
Neurology 59(1):26–33
26. Musolino R, La Spina P, Granata A, Gallitto G, Leggiadro N, Carerj
S, Manganaro A, Tripodi F, Epifanio A, Gangemi S, Di Perri R
(2003) Ischaemic stroke in young people: a prospective and long-
term follow-up study. Cerebrovasc Dis 15(1–2):121–128
123