High Incidence of Diabetes after Stroke in Young Adults

and Risk of Recurrent Vascular Events: The FUTURE Study

Loes C. A. Rutten-Jacobs1., Pim A. J. Keurlings2., Renate M. Arntz1, Noortje A. M. Maaijwee1,
Henny C. Schoonderwaldt1, Lucille D. Dorresteijn3, Maureen J. van der Vlugt4, Ewoud J. van Dijk1,
Frank-Erik de Leeuw1*
1 Department of Neurology, Radboud University Medical Center, Donders Institute for Brain, Cognition and Behaviour, Nijmegen, The Netherlands, 2 Department of
Internal Medicine, Canisius Wilhelmina Hospital, Nijmegen, The Netherlands, 3 Department of Neurology, Medisch Spectrum Twente, Enschede, The Netherlands,
4 Department of Cardiology, Radboud University Medical Center, Nijmegen, The Netherlands

Abstract
Background: Diabetes diagnosed prior to stroke in young adults is strongly associated with recurrent vascular events. The
relevance of impaired fasting glucose (IFG) and incidence of diabetes after young stroke is unknown. We investigated the
long-term incidence of diabetes after young stroke and evaluated the association of diabetes and impaired fasting glucose
with recurrent vascular events.

Methods: This study was part of the FUTURE study. All consecutive patients between January 1, 1980, and November 1,
2010 with TIA or ischemic stroke, aged 18–50, were recruited. A follow-up assessment was performed in survivors between
November 1, 2009 and January 1, 2012 and included an evaluation for diabetes, fasting venous plasma glucose and
recurrent vascular events. The association of diabetes and IFG with recurrent vascular events was assessed by logistic
regression analysis, adjusted for age, sex and follow-up duration.

Results: 427 survivors without a medical history of diabetes were included in the present analysis (mean follow-up of 10.1
(SD 8.4) years; age 40.3 (SD 7.9) years). The incidence rate of diabetes was 7.9 per 1000 person-years and the prevalence of
IFG was 21.1%. Patients with diabetes and IFG were more likely to have experienced any vascular event than those with
normal fasting glucose values (OR 3.5 (95%CI 1.5–8.4) for diabetes and OR 2.5 (95%CI 1.3–4.8) for IFG).

Conclusions: Diabetes or IFG in young stroke survivors is frequent and is associated with recurrent vascular events. Regular
screening for IFG and diabetes in this population, yields potential for secondary prevention.

Citation: Rutten-Jacobs LCA, Keurlings PAJ, Arntz RM, Maaijwee NAM, Schoonderwaldt HC, et al. (2014) High Incidence of Diabetes after Stroke in Young Adults
and Risk of Recurrent Vascular Events: The FUTURE Study. PLoS ONE 9(1): e87171. doi:10.1371/journal.pone.0087171
Editor: Hugo ten Cate, Maastricht University Medical Center, Netherlands
Received October 14, 2013; Accepted December 18, 2013; Published January 23, 2014
Copyright: ß 2014 Rutten-Jacobs et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The authors have no support or funding to report.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: [email protected]
. These authors contributed equally to this work.

Introduction tolerance, conditions that precede diabetes, have been associated

with an increased risk of vascular events [8,9]. Moreover, more
Patients, who suffered a stroke at young age, are at high risk of than half of older stroke patients, who were not previously known
recurrent vascular events and death [1–3]. Because of the young to have diabetes, was diagnosed to have either impaired glucose
age of these patients, the initial stroke as well as possible recurrent tolerance or diabetes three months after stroke [10]. Analogous to
vascular events have a large impact on number of years lost to ill- these older stroke patients, young stroke patients without a medical
health, disability and early death. Previous studies reported that history of diabetes at the time of their index event may still develop
vascular risk factors are common in these young adults [4,5]. IFG or incident diabetes after their young stroke as well.
Secondary prevention measures targeting these vascular risk Particularly since regular monitoring of glucose levels after the
factors may diminish the risk of recurrent vascular events. acute phase of stroke in young adults without diabetes is seldom
However, risk factors that emerge after a young stroke often may performed. Glucose control in patients with IFG or incident
go undetected in many patients as current protocols and guidelines diabetes could be an important way to reduce risk of recurrent
only recommend screening of young stroke patients in the acute vascular events [11]. However, the incidence of diabetes and IFG
phase and only few months thereafter [6]. after stroke in young adults is currently unknown. Moreover, we
Risk of recurrent vascular events seems especially high in young are not aware of any study that investigates the association
stroke patients with a medical history of diabetes [7]. In both the between impaired fasting blood glucose and recurrent vascular
general population and in stroke patients over 65 years, also events in young stroke patients.
impaired fasting blood glucose (IFG) or impaired glucose

PLOS ONE | www.plosone.org 1 January 2014 | Volume 9 | Issue 1 | e87171

 Diabetes in Young Stroke and Vascular Events

Therefore, we first investigated the incidence of diabetes after a Diabetes and impaired fasting glucose
mean follow-up of 10 years in survivors of a young TIA or To answer the first research question, the incidence of diabetes
ischemic stroke. Secondly, we investigated whether impaired was the primary outcome measure, either diagnosed during follow-
fasting blood glucose and diabetes at follow-up were associated up or at the follow-up assessment.
with the occurrence of vascular events during follow-up. The detection of incident diabetes during follow-up was done by
a two step approach. First patients were asked whether diabetes
Methods was diagnosed during the follow-up period, by means of a
standardized structured questionnaire. If so, patients’ general
Patients and study design practitioner was contacted to verify the diagnosis systematically,
This study is a part of the ‘‘Follow-Up of Transient ischemic and to ascertain information about the plasma glucose level, type
attack and stroke patients and Unelucidated Risk factor Evalua- of diagnosed diabetes and initiated treatment.
tion’’ (FUTURE) study, a prospective cohort study of prognosis of Secondly, venous plasma samples were taken from all partic-
stroke in young adults [2,12]_ENREF_2.The Medical Review ipants at the follow-up assessment after overnight fasting to
Ethics Committee region Arnhem-Nijmegen approved the study. measure plasma glucose. Whenever glucose was $5.6 mmol/L,
In short, the FUTURE study comprised all consecutive patients the patient was sent to the general practitioner to obtain a second
aged 18 through 50 years with a TIA, ischemic stroke or fasting venous plasma glucose.
intracerebral hemorrhage admitted to the Radboud university
Incident diabetes was defined as: 1) treatment with antidiabetic
medical center from January 1, 1980 until November 1, 2010.
medication or a diagnosis of diabetes (confirmed by a physician)
Only patients with TIA or ischemic stroke without a medical
during the follow-up period or 2) two consecutive fasting venous
history of diabetes, who survived until the follow-up assessment,
plasma glucose levels of $7.0 mmol/L at the follow-up assess-
were included in the present study. Exclusion criteria were
ment.
cerebral venous sinus thrombosis and retinal infarct.
Regarding the second research question, secondary outcomes
To minimize bias resulting from changing diagnostic tech-
were the prevalence of diabetes or IFG and the occurrence of
niques, the World Health Organization definitions for TIA and
vascular events in relation to fasting blood glucose levels at the
stroke were used [13,14]. The definition of TIA included a rapidly
follow-up assessment. IFG was only assessed at the follow-up
evolving focal neurologic deficit, without positive phenomena such
assessment, defined as a fasting blood glucose of 5.6 mmol/L–
as twitches, jerks or myoclonus, with vascular cause only and
6.9 mmol/L.
persisting for a period of less than 24 hours. Stroke was defined as
focal neurologic deficit persisting for more than 24 hours. Stroke
was subdivided into ischemic and hemorrhagic stroke, on the basis Vascular events
of radiological findings. Patients were evaluated for recurrent vascular events by means
Patients were identified through a prospective registry of all of a standardized, structured questionnaire [3]. Whenever a
patients with young stroke that has been maintained at our centre, recurrent event was suspected, information retrieved was verified
beginning in 1978 [15], with a standardized data collection of and adjudicated by physicians from the appropriate specialty
baseline and clinical characteristics, including demographic data, (FEdL, EvD, MvdV).
stroke subtype and vascular risk factors [12]. Assessment of both A composite vascular event was defined as the combination of
the etiology (Trial of Org 10172 in Acute Stroke Treatment stroke (ischemic or hemorrhagic), myocardial infarction, and
[TOAST] classification) [16] and severity (National Institutes of cardiovascular procedures (coronary artery bypass grafting,
Health Stroke Scale [NIHSS]) [17] was performed retrospectively percutaneous transluminal coronary angioplasty, carotid endar-
in all cases on the basis of medical records, because these scales did terectomy or other peripheral arterial revascularization proce-
not exist when a substantial number of our patients experienced dures), whichever occurred first. Separate analyses were done for
their index event. In comparison to the original TOAST stroke and other arterial events.
classification [18], the presently used classification has an
additional category, ‘‘likely large-artery atherosclerosis’’ [16]. Statistical analysis
Atherothrombotic stroke is defined as patients with (1) an To answer the first research question, the incidence rate of
ipsilateral internal carotid stenosis .50% (in NASCET criteria), diabetes was calculated for stroke subtypes. To answer the second
or (2) an ipsilateral stenosis .50% of another intra/extracranial research question, fasting blood glucose values at the follow-up
artery, or (3) mobile thrombus in the aortic arch. Likely assessment were categorized into normal fasting blood glucose
atherothrombotic stroke is defined as patients with no evidence (,5.6 mmol/L), impaired fasting blood glucose (5.6 mmol/L–
of atherothrombotic stroke with (1) an ipsilateral internal carotid 6.9 mmol/L) and diabetes ($7.0 mmol/L or incident diabetes
stenosis ,50%, or (2) an ipsilateral stenosis ,50% of another during follow-up). Baseline characteristics were compared between
intra/extracranial artery, or (3) aortic arch plaques .4 mm in patients without diabetes or impaired fasting glucose and patients
thickness without a mobile component, or (4) a history of with diabetes or impaired fasting glucose using Student’s t test,
myocardial infarction or coronary revascularization, (5) a history Mann-Whitney U test or chi-square-test whenever appropriate.
of documented peripheral arterial disease, or (6) at least two risk Odds ratios were calculated for the association between fasting
factors for atherosclerotic disease: arterial hypertension (treated or blood glucose categories at the follow-up assessment and the
known blood pressure before stroke .135/85 mm Hg or occurrence during follow-up of the composite vascular event, other
hypertensive retinopathy), diabetes mellitus (treated or known arterial events and stroke separately, adjusted for age of the index
blood fasting glucose .7 mmol/dl), current smoking (or smoking stroke, sex, and follow-up duration.
stopped within the last 6 months), high cholesterol (treated or Analyses were done using IBM SPSS Statistics version 20. Two-
known low-density lipoprotein before the stroke .160 mg/dl). sided P values of less than 0.05 were considered to indicate
Patients alive were invited for follow-up assessment between statistical significance.
November 1, 2009 and January 1, 2012. Participants provided
written informed consent.

PLOS ONE | www.plosone.org 2 January 2014 | Volume 9 | Issue 1 | e87171

 Diabetes in Young Stroke and Vascular Events

Figure 1. Flowchart of the study population.

doi:10.1371/journal.pone.0087171.g001

Results the follow-up assessment were at baseline more frequently men,

had a higher age, a longer mean follow-up duration, a likely
427 patients completed follow-up assessment (Figure 1). Baseline atherothrombotic stroke and a medical history of hypertension.
characteristics are presented in table 1. There were no differences At follow-up, 12 patients with incident diabetes (35.3%) had
in baseline characteristics between participants and non-partici- experienced any vascular event (composite event) and 7 patients
pants (patients lost to follow-up, patients with no venipuncture or (20.6%) of them experienced more than one event; 4 patients
patients who refused), except for history of TIA (3.5% in (11.8%) had at least one stroke and 10 patients (29.4%) had
participants and 0.7% in nonparticipants). experienced at least one other arterial event. Among patients with
After a mean follow-up of 10.1 years (SD 8.4), diabetes was IFG at follow-up, 21 patients (25.3%) had experienced any
diagnosed in 11 TIA patients (7.1%) and 23 ischemic stroke vascular event and 6 patients (7.2%) of them experienced more
patients (8.5%), resulting in an incidence rate per 1000 person than one event; 10 patients (12.9%) had at least one stroke and 11
years of 7.9 and 7.8 respectively. Among those without diabetes at patients (13.3%) had experienced at least one other arterial event.
follow-up, 83 patients (21.1%) had an IFG (5.6–6.9 mmol/L) and Among patients with normal fasting blood glucose levels at follow-
310 patients (78.9%) had normal blood glucose values. up, 30 patients (9.7%) had experienced any vascular event and 6
Compared with patients without IFG or incident diabetes at the patients (1.9%) of them; 24 patients (7.7%) had experienced at
follow-up assessment, patients with incident diabetes were at least one stroke and 8 patients (2.6%) had experienced at least one
baseline more often older, had a longer mean follow-up duration, other arterial event. In all three fasting blood glucose groups, the
had a likely atherothrombotic stroke, a medical history of proportion of patients on antiplatelet medication at discharge did
hypertension, a medical history of smoking and a family history not differ between patients who experienced a recurrent vascular
of diabetes (Table 2). Compared with patients without IFG or event compared with patients who did not experience a recurrent
incident diabetes at the follow-up assessment, patients with IFG at vascular event during follow-up.

PLOS ONE | www.plosone.org 3 January 2014 | Volume 9 | Issue 1 | e87171

 Diabetes in Young Stroke and Vascular Events

Table 1. Baseline characteristics of patients.

Total TIA Ischemic stroke

n (% of total) 427 (100) 156 (36.5) 271 (63.5)

Mean age at event, years (SD) 40.3 (7.9) 41.3 (7.8) 39.9 (7.8)
Male 190 (44.5) 71 (45.5) 119 (43.9)
Median NIHSS at admission (IQR)* 2 (0–6) 0 (0–1) 4 (2–8)
Mean follow-up, years (SD) 10.1 (8.3) 8.9 (8.5) 10.9 (8.2)
TOAST
Atherothrombotic stroke 33 (7.7) 9 (5.8) 24 (8.9)
Likely atherothrombotic stroke 61 (14.3) 27 (17.3) 34 (12.5)
Cardioembolic stroke 44 (10.3) 15 (9.6) 29 (10.7)
Small vessel occlusion 41 (9.6) 7 (4.5) 34 (12.5)
Rare causes 66 (15.5) 16 (10.3) 50 (18.5)
Multiple causes 10 (2.3) 3 (1.9) 7 (2.6)
Unknown cause 172 (40.3) 79 (50.6) 93 (34.3)
Risk factors in medical history
Previous TIA 15 (3.5) 8 (5.1) 7 (2.6)
Previous stroke 6 (1.4) 2 (1.3) 4 (1.5)
Hypertension 101 (23.7) 46 (29.5) 55 (20.3)
Atrial fibrillation 6 (1.4) 2 (1.3) 4 (1.5)
Smoking{ 196 (46.8) 55 (35.9) 141 (53.0)
Excess alcohol consumption` 27 (6.3) 11 (7.1) 16 (5.9)
Family history of diabetes1 175 (41.4) 69 (45.1) 106 (39.3)

Abbreviations: TIA, transient ischemic attack; SD, standard deviation; NIHSS, National Institute of Health Stroke Scale; IQR, interquartile range; TOAST, Trial of Org 10172
in Acute Stroke Treatment.
Data are given as number (percentage) or otherwise stated
*Scores range from 0 to 42 with higher scores on the scale indicating worse stroke severity. 0.5% of NIHSS was missing.
{
Smoking was defined as smoking at least 1 cigarette a day in the year prior to the event. 1.9% of data on smoking was missing.
`
Excess alcohol consumption was defined as consuming more than 200 grams of pure alcohol per week
1
First degree family member. 0.9% of data on family history of diabetes was missing.
doi:10.1371/journal.pone.0087171.t001

After adjusting for age of index stroke, sex and follow-up To our knowledge, our study is the first to evaluate the
duration, patients with diabetes and IFG were more likely to have incidence of diabetes after stroke in young adults and to study the
experienced any vascular event during follow-up than those with association between fasting blood glucose values and recurrent
normal fasting blood glucose values (OR 3.5 (95%CI 1.5–8.4) for vascular events. Moreover, our study has the longest follow-up
diabetes and OR 2.5 (95%CI 1.3–4.8) for IFG). Risk for the period reported and one of the largest study populations in the
recurrence of stroke was not different for patients with incident field of young stroke. Collecting data all in one site allowed us to
diabetes and IFG compared with those with normal fasting blood collect baseline and follow-up information according to identical
glucose values (OR 1.2 (95%CI 0.4–4.0) for diabetes and OR 1.4 procedures in all patients thereby reducing the risk of information
(95%CI 0.6–3.3) for IFG). Risk of other arterial events was bias.
increased in patients with diabetes and IFG compared with those Our study has some limitations. First, it may be that not all cases
with normal fasting blood glucose levels (OR 8.4 (95% CI 2.7– of young stroke in our catchment area were included in our
26.4) for diabetes and (OR 3.6 (95%CI 1.3–9.6) for IFG). cohort, because our cohort is a single-center, hospital-based study,
rather than community-based. Only those patients who sustained
Discussion a fatal stroke, who were not admitted to our hospital, would not
have been included in our study. Patients who survive usually visit
We demonstrated that 8% of young stroke survivors developed a university medical center during the course of their disease. In
diabetes during a mean follow-up of 10 years after stroke, which is addition, there are no restrictions to be admitted to our hospital
more than two times higher than expected compared with persons and we included all consecutive cases admitted. We therefore
from a Dutch general practitioner registry with similar age and sex presume that our study population is a representative sample of
[19]. Moreover, we showed that among those patients without Dutch patients with young stroke, although formal data are
diabetes at the follow-up assessment, 21% had impaired fasting lacking to prove this generalizability.
blood glucose values. In our study, both patients with diabetes and Second, we investigated the association of IFG and diabetes
patients with IFG at the follow-up assessment were about three with recurrent vascular events during follow-up in a cross-sectional
times more likely to experience any vascular event during follow- analysis, on average 10 years after the index event in patients that
up than those with normal fasting blood glucose values. survived until the follow-up assessment.

PLOS ONE | www.plosone.org 4 January 2014 | Volume 9 | Issue 1 | e87171

 Diabetes in Young Stroke and Vascular Events

Table 2. Presence of baseline factors in patients with incident diabetes or impaired fasting glucose at follow-up.

No diabetes or IFG Diabetes p* IFG p{

n (% of total) 310 (72.6) 34 (8.0) 83 (19.4)

Mean age at event, years (SD) 39.2 (8.2) 44.5 (4.5) 0.002 42.8 (6.5) 0.001
Male 123 (39.7) 16 (47.1) 0.41 51 (61.4) ,0.001
Median NIHSS at admission (IQR)` 2 (0–6) 2 (1–4) 0.82 3 (0–6) 0.31
Mean follow-up, years (SD) 8.7 (7.8) 16.7 (8.0) ,0.001 12.6 (8.6) 0.001
TOAST
Atherothrombotic stroke 20 (6.5) 5 (14.7) 0.08 8 (9.6) 0.32
Likely atherothrombotic stroke 30 (9.7) 14 (41.2) ,0.001 17 (20.5) 0.007
Cardioembolic stroke 35 (11.3) 1 (2.9) 0.15 8 (9.6) 0.67
Small vessel occlusion 34 (11.0) 0 0.06 7 (8.4) 0.50
Rare causes 56 (18.1) 1 (2.9) 0.03 9 (10.8) 0.14
Multiple causes 7 (2.3) 1 (2.9) 1.00 2 (2.4) 1.00
Unknown cause 128 (41.3) 12 (35.3) 0.50 32 (38.6) 0.65
Risk factors in medical history
Previous TIA 10 (3.2) 1 (2.9) 1.00 4 (4.8) 0.51
Previous stroke 5 (1.6) 1 (2.9) 1.00 0 0.37
Hypertension 58 (18.7) 14 (41.2) 0.002 29 (34.9) 0.002
Smoking1 133 (43.6) 23 (67.6) 0.008 40 (50.0) 0.31
Excess alcohol consumption|| 19 (6.1) 3 (8.8) 0.71 5 (6.0) 0.97
Family history of diabetes" 124 (40.5) 20 (58.8) 0.04 31 (37.3) 0.60

Abbreviations: IFG, impaired fasting glucose; TIA, transient ischemic attack; SD, standard deviation; NIHSS, National Institute of Health Stroke Scale; IQR, interquartile
range; TOAST, Trial of Org 10172 in Acute Stroke Treatment.
Data are given as number (percentage) or otherwise stated
*p values refer to a comparison between patients with incident diabetes and patients with no IFG or diabetes
{
p values refer to a comparison between patients with IFG and patients with no IFG or diabetes
`
Scores range from 0 to 42 with higher scores on the scale indicating worse stroke severity. 0.4% of NIHSS was missing.
1
Smoking was defined as smoking at least 1 cigarette a day in the year prior to the event. 2.9% of data on smoking was missing.
||
Excess alcohol consumption was defined as consuming more than 200 grams of pure alcohol per week
"
First degree family member. 1.0% of data on family history of diabetes was missing.
doi:10.1371/journal.pone.0087171.t002

Thus the measurement of blood glucose values is done after a Fifth, secondary prevention might have influenced our results.
recurrent event occurred. This may have induced survivor bias. In our study about 90% of all patients used secondary preventive
IFG and diabetes may be associated with the severity of the medication at discharge. Consequently the shown risk of recurrent
recurrent event and as a consequence, patients with IFG and vascular events might be an underestimation attributable to the
diabetes may be underrepresented in survivors with recurrent use of this preventive medication. Sixth, as is reflected by the wide
events, which may have attenuated the association between IFG/ CIs, estimates for some subgroups that contain only a few patients
diabetes and recurrent events. might be unstable and should therefore be interpreted with
Furthermore, IFG was only measured at the follow-up caution.
assessment, whereas for diabetes also a diagnosis established So far, the only studies reporting on epidemiology of diabetes in
during the follow-up period was taken in account. Diabetes that young stroke patients restricted their reports to diabetes diagnosed
developed during follow-up might otherwise have been missed at prior to stroke. The proportion of patients with a medical history
the follow-up assessment due to initiated treatment. of diabetes varied widely in these studies, ranging from 2–12%
Third, some patients were lost to follow-up or refused to [7,20,21]. Our observed prevalence of diabetes based on the
participate, which potentially could have resulted in selection bias. medical history of 4.9% is in the middle of this range.
However, non-participants did not differ in baseline characteristic We showed in univariate analysis that incident diabetes after
from participants, making selection bias in this group unlikely. TIA or ischemic stroke was associated with age, likely athero-
Fourth, our study has a long inclusion period, during which thrombotic stroke and family history of diabetes, which are among
diagnostic equipment, acute treatment and secondary prevention well established risk factors for diabetes in the general population.
have improved. However, this is an unavoidable feature of a long- In addition, we showed that both patients with diabetes and
patients with IFG were far more likely to have experienced any
term follow-up study. Furthermore, the long follow-up period
arterial event during follow-up than those with normal fasting
might have resulted in recall bias with respect to vascular events.
blood glucose values. These results suggest an intimate relationship
However, this probably would have underestimated the associa-
in young stroke patients between pre-existent vulnerability to
tion between diabetes and recurrent vascular events, since the
atherosclerosis and incident diabetes, which is an atherogenic risk
incidence of diabetes was strongly related to the number of follow-
factor itself. However, it is also possible that diabetes was already
up years.
present but not revealed during the index event.

PLOS ONE | www.plosone.org 5 January 2014 | Volume 9 | Issue 1 | e87171

 Diabetes in Young Stroke and Vascular Events

Incident diabetes or IFG was not associated with recurrent after stroke in young adults may allow for early diagnoses of IFG
stroke. An explanation might be that diabetes needs to be present and diabetes and thereby provide a therapeutic window to lower
for many years to be a risk factor for recurrent stroke. This is in the risk of recurrent vascular events. Similar to the general
line with a previous study in young adults with ischemic stroke that population, young stroke patients with a higher age, having other
showed that among patients with type 1 diabetes, duration of vascular risk factors or a family history of diabetes, might benefit
diabetes was on average 10 years longer in those with recurrent the most from active screening.
stroke versus those without recurrent stroke [7]. Another To conclude, IFG and diabetes after stroke in young patients
explanation for the lack of association might be the possibility of may remain unnoticed in many patients. A regular screening for
index event bias [22]. In a study investigating recurrence, patients IFG and diabetes after young stroke, particularly in those with
are included based on the occurrence of the first event that is increasing age, having other vascular risk factors or a family
similar to the recurrent event. This has an effect on the distribution history of diabetes, yields potential for secondary prevention.
of risk factors in this selected population and the association of
these risk factors with the outcome of interest. Author Contributions
The high incidence of diabetes during our long follow-up
period, but also the high proportion of patients with IFG, Conceived and designed the experiments: LRJ HS LD FEdL EvD.
emphasizes that young stroke survivors remain vulnerable to the Performed the experiments: LRJ PK RA NM MvdV FEdL. Analyzed the
data: LRJ PK FEdL. Wrote the paper: LRJ PK FEdL EvD.
development of (risk factors for) vascular disease, even decades
after their initial stroke. Active screening for IFG and diabetes

References
1. Putaala J, Haapaniemi E, Metso AJ, Metso TM, Artto V, et al. (2010) Recurrent study: a prospective cohort study. Study rationale and protocol. BMC Neurol 11:
ischemic events in young adults after first-ever ischemic stroke. Ann Neurol 68: 109.
661–671. 13. Aho K, Harmsen P, Hatano S, Marquardsen J, Smirnov VE, et al. (1980)
2. Rutten-Jacobs LC, Arntz RM, Maaijwee NA, Schoonderwaldt HC, Dorresteijn Cerebrovascular disease in the community: results of a WHO collaborative
LD, et al. (2013) Long-term mortality after stroke among adults aged 18 to 50 study. Bull World Health Organ 58: 113–130.
years. JAMA 309: 1136–1144. 14. Hatano S (1976) Experience from a multicentre stroke register: a preliminary
3. Rutten-Jacobs LC, Maaijwee NA, Arntz RM, Schoonderwaldt HC, Dorresteijn report. Bull World Health Organ 54: 541–553.
LD, et al. (2013) Long-term risk of recurrent vascular events after young stroke: 15. Boers GH, Smals AG, Trijbels FJ, Fowler B, Bakkeren JA, et al. (1985)
The FUTURE study. Ann Neurol. Heterozygosity for homocystinuria in premature peripheral and cerebral
4. Putaala J, Yesilot N, Waje-Andreassen U, Pitkaniemi J, Vassilopoulou S, et al. occlusive arterial disease. N Engl J Med 313: 709–715.
(2012) Demographic and geographic vascular risk factor differences in European 16. Bousser MG, Amarenco P, Chamorro A, Fisher M, Ford I, et al. (2009)
young adults with ischemic stroke: the 15 cities young stroke study. Stroke 43: Rationale and design of a randomized, double-blind, parallel-group study of
2624–2630. terutroban 30 mg/day versus aspirin 100 mg/day in stroke patients: the
5. von Sarnowski B, Putaala J, Grittner U, Gaertner B, Schminke U, et al. (2013) prevention of cerebrovascular and cardiovascular events of ischemic origin with
Lifestyle risk factors for ischemic stroke and transient ischemic attack in young terutroban in patients with a history of ischemic stroke or transient ischemic
attack (PERFORM) study. Cerebrovasc Dis 27: 509–518.
adults in the Stroke in Young Fabry Patients study. Stroke 44: 119–125.
17. Brott T, Adams HP Jr., Olinger CP, Marler JR, Barsan WG, et al. (1989)
6. Davis SM, Donnan GA (2012) Clinical practice. Secondary prevention after
Measurements of acute cerebral infarction: a clinical examination scale. Stroke
ischemic stroke or transient ischemic attack. N Engl J Med 366: 1914–1922.
20: 864–870.
7. Putaala J, Liebkind R, Gordin D, Thorn LM, Haapaniemi E, et al. (2011)
18. Adams HP, Jr., Bendixen BH, Kappelle LJ, Biller J, Love BB, et al. (1993)
Diabetes mellitus and ischemic stroke in the young: clinical features and long- Classification of subtype of acute ischemic stroke. Definitions for use in a
term prognosis. Neurology 76: 1831–1837. multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke
8. Vermeer SE, Sandee W, Algra A, Koudstaal PJ, Kappelle LJ, et al. (2006) Treatment. Stroke 24: 35–41.
Impaired glucose tolerance increases stroke risk in nondiabetic patients with 19. Poos MJJC (2011) Diabetes mellitus: prevalence, incidence and death according
transient ischemic attack or minor ischemic stroke. Stroke 37: 1413–1417. to age and gender. Volksgezondheid Toekomst Verkenning, Nationaal Kompas
9. Ford ES, Zhao G, Li C (2010) Pre-diabetes and the risk for cardiovascular Volksgezondheid. Bilthoven: RIVM.
disease: a systematic review of the evidence. J Am Coll Cardiol 55: 1310–1317. 20. Leys D, Bandu L, Henon H, Lucas C, Mounier-Vehier F, et al. (2002) Clinical
10. Kernan WN, Viscoli CM, Inzucchi SE, Brass LM, Bravata DM, et al. (2005) outcome in 287 consecutive young adults (15 to 45 years) with ischemic stroke.
Prevalence of abnormal glucose tolerance following a transient ischemic attack Neurology 59: 26–33.
or ischemic stroke. Arch Intern Med 165: 227–233. 21. Waje-Andreassen U, Naess H, Thomassen L, Eide GE, Vedeler CA (2007)
11. Holman RR, Paul SK, Bethel MA, Matthews DR, Neil HA (2008) 10-year Arterial events after ischemic stroke at a young age: a cross-sectional long-term
follow-up of intensive glucose control in type 2 diabetes. N Engl J Med 359: follow-up of patients and controls in western Norway. Cerebrovasc Dis 24: 277–
1577–1589. 282.
12. Rutten-Jacobs LC, Maaijwee NA, Arntz RM, Van Alebeek ME, Schaapsmeer- 22. Dahabreh IJ, Kent DM (2011) Index event bias as an explanation for the
ders P, et al. (2011) Risk factors and prognosis of young stroke. The FUTURE paradoxes of recurrence risk research. JAMA 305: 822–823.

PLOS ONE | www.plosone.org 6 January 2014 | Volume 9 | Issue 1 | e87171

© 2014 Rutten-Jacobs et al. This is an open-access article distributed under the
terms of the Creative Commons Attribution License:
http://creativecommons.org/licenses/by/4.0/ (the “License”), which permits
unrestricted use, distribution, and reproduction in any medium, provided the
original author and source are credited. Notwithstanding the ProQuest Terms
and Conditions, you may use this content in accordance with the terms of the
License.