Reviews/Commentaries/ADA Statements

R E V I E W A R T I C L E

Diabetes, the Metabolic Syndrome, and

Ischemic Stroke
Epidemiology and possible mechanisms
ELLEN L. AIR, MD, PHD1 younger ages (25). The Greater Cincin-
BRETT M. KISSELA, MD2 nati–Northern Kentucky Stroke Study
(GCNKSS) found that the risk for isch-
emic stroke in white diabetic patients is
higher at every age-group compared with

S
troke affects more than 700,000 in- 50% of mortality in diabetes. Because of
dividuals each year; it is the third the overwhelming impact of CHD, the nondiabetic patients, with highest relative
largest cause of death and the largest impact of stroke has been relatively under- risk (RR) of 5.3 found in the 45- to 54-
cause of adult disability in the U.S. Diabe- appreciated. Thus, physicians, diabetes ed- year age-group. Among African Ameri-
tes is a major risk factor for the develop- ucators, and nurses are less equipped to cans, the highest risk was even greater (RR
ment of stroke, yet this risk is not realized educate patients. We therefore review the 9.9) and was found in the 35- to 44-year
or understood by patients with diabetes. relationship between diabetes and stroke. age-group. A substantial peak in stroke
This likely reflects a lack of understanding Given that more than one million risk is seen in the 45- to 64-year age-
within the medical community of how di- people are diagnosed with diabetes group in whites and in the 35- to 54-year
abetes confers this risk. We will explore yearly, a figure that is expected to rise, the age-group in African Americans (7).
the potential underlying mechanisms that impact of diabetes on the incidence of Although stroke is more common
lead to increased incidence of stroke stroke is of increasing importance. Dia- among diabetic patients, most studies re-
among diabetic patients. Beyond diabetes betic patients compose roughly 6.3% of port a significantly reduced rate of tran-
itself, the metabolic syndrome and its the U.S. population but account for 15– sient ischemic attacks (TIAs) in diabetic
components will also be discussed. The 27% of all incident strokes, based on patients compared with nondiabetic pa-
impact of diabetes and hyperglycemia on 2002 estimates (4,7–12). This is certainly tients. Diabetic patients are more likely to
stroke outcomes and a discussion of cur- an underestimation, as most studies clas- present with cerebral infarct, indicating
rent approaches to reduce stroke in this sify patients as having diabetes only if di- that ischemia in diabetic patients is less
high-risk population are included. Because agnosed before stroke. When considering likely to be reversible (7,26 –28). This
type 2 diabetes affects the vast majority of age-adjusted incidence rates, diabetic pa- presents a unique problem for preventing
those diagnosed with diabetes, it will be the tients are 2.9 times as likely to have a stroke in this population. TIAs can serve
primary focus of this discussion. stroke compared with nondiabetic pa- as a warning sign, providing a window of
tients, a disparity that is seen in multiple opportunity for medical intervention to
DEFINING THE PROBLEM — It racial/geographic groups (4,7,9,13–15). prevent a completed stroke. The relative
has been well documented that diabetes This is due specifically to an increase in lack of warning in diabetic patients re-
confers a significantly increased risk of the rate of ischemic stroke rather than quires that physicians, nurses, and educa-
stroke, as well as increased mortality fol- hemorrhagic stroke (7,16 –18). tors be aggressive about risk factor
lowing stroke (1–7). Stroke is a prevent- The heaviest burden of stroke for the intervention, as comprehensive programs
able disease with high personal and general population lies with older and mi- to reduce risk can be highly successful
societal cost. While great progress has nority groups (4,12,19 –22). Diabetes ap- (29). For those who do present with a
been made in understanding the link be- pears to amplify these nonmodifiable TIA, aggressive treatment is equally im-
tween diabetes and coronary heart disease risks, in part due to the increased preva- portant since diabetes has been shown to
(CHD), the literature on diabetes and lence of diabetes in these groups increase the risk of subsequent completed
stroke has been less enlightening. CHD is (7,23,24). Diabetes also confers an in- stroke (30).
a larger problem that accounts for 40 – creased risk for neurovascular disease at
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
CAUSE AND EFFECT? — Many at-
From the 1Department of Neurosurgery, University of Cincinnati College of Medicine, Cincinnati, Ohio; and tempts have been made to discern the un-
the 2Department of Neurology, University of Cincinnati College of Medicine, Cincinnati, Ohio.
Address correspondence and reprint requests to Brett M. Kissela, Department of Neurology, University of derlying mechanisms through which
Cincinnati College of Medicine, 231 Albert Sabin Way, ML 0525, Cincinnati, OH 45267-0525. E-mail: diabetes increases stroke risk. Such stud-
[email protected]. ies have largely taken cues from the car-
Received for publication 21 July 2006 and accepted in revised form 8 September 2007. diovascular literature in which diabetes
Published ahead of print at http://care.diabetesjournals.org on 11 September 2007. DOI: 10.2337/dc06-
1537.
and the associated components of the
Abbreviations: ARIC, Atherosclerosis Risk in Communities; CAD, coronary artery disease; CARDS, metabolic syndrome (i.e., hypertension
Collaborative AtoRvastatin Diabetes Study; CHD, coronary heart disease; CIMT, carotid intima-media thick- and hyperlipidemia) have been found to
ness; EPIC, European Prospective Investigation Into Cancer; GCNKSS, Greater Cincinnati–Northern Ken- contribute to cardiovascular disease de-
tucky Stroke Study; NHANES III, Third National Health and Nutrition Survey; TIA, transient ischemic velopment (31–33). This approach has
attack; UKPDS, UK Prevention in Diabetes Study; WHR, waist-to-hip ratio.
A table elsewhere in this issue shows conventional and Système International (SI) units and conversion been informative, yet the relationships
factors for many substances. between diabetes, the components of the
© 2007 by the American Diabetes Association. metabolic syndrome, and stroke are

DIABETES CARE, VOLUME 30, NUMBER 12, DECEMBER 2007 3131

Diabetes and ischemic stroke

clearly unique. Here, we discuss these in- tionship between hyperglycemia and summarize, a significant association be-
dividual relationships, highlighting the stroke incidence. Rather, it is apparent tween insulin resistance and stroke risk
differences between stroke and cardiac that diabetic patients have an increased has been found, but the magnitude of this
risk. risk of stroke regardless of their level of association is less than the association
metabolic control. seen with cardiovascular disease.
DIABETES VERSUS
HYPERGLYCEMIA — As in any dis- INSULIN RESISTANCE, THE HYPERTENSION — A m o n g t h e
cussion of diabetes and its sequelae, the METABOLIC SYNDROME, components of the metabolic syndrome,
fundamental question arises as to whether AND STROKE — Without substan- hypertension is the single most important
stroke risk is increased due to chronic hy- tive evidence that intensive glucose con- risk factor for the development of stroke.
perglycemia. Published studies provide trol reduces stroke risk, the focus has In this respect, stroke varies significantly
conflicting evidence. Lehto et al. (34) shifted to insulin resistance and its asso- from cardiac disease, where hypertension
studied 1,059 diabetic patients and corre- ciated metabolic syndrome. Type 2 diabe- is a lesser risk factor.
lated their baseline fasting glucose levels, tes, characterized by an inability to produce Evidence suggests that some of the in-
A1C, and duration of diabetes with stroke enough insulin to overcome insulin resis- creased risk of stroke among diabetic pa-
over 7 years of follow-up. All three factors tance, frequently coexists with a constella- tients is attributable to the increased
contributed significantly to increased risk tion of cardiovascular risk factors including prevalence of hypertension. The GCNKSS
of stroke, while fasting hyperglycemia hypertension, obesity, and hyperlipidemia. found that the prevalence of hypertension
(⬎13.4 mmol/l) remained significant af- Together, these have been termed the met- was 79% among diabetic patients and
ter accounting for other cardiovascular abolic syndrome (also known as syndrome 57% among nondiabetic patients (P ⬍
risk factors (odds ratio [OR] 2.6 [95% CI X or insulin-resistance syndrome). The role 0.0001) (7). A significant, though
1.5–3.8] compared with normoglycemia) that these factors have played individually, smaller, difference was found in the
(34). The Honolulu Heart Program re- as well as together, in the development of Copenhagen Stroke Study (48 vs. 30%,
ported similar results in nondiabetic pa- cardiovascular disease (40) has made them respectively, P ⬍ 0.0001) (10). Prospec-
tients when comparing the extremes the target of studies regarding stroke as well. tively, follow-up of diabetic patients in
(80th and 20th percentiles) of serum glu- the UKPDS found that the occurrence of
cose levels (RR for thromboembolic INSULIN RESISTANCE — Insulin vascular complications, including stroke,
stroke 1.4 [95% CI 1.1–1.8]) (16). A resistance, as measured by basal hyperin- were significantly associated with hyper-
Finnish cohort study measured A1C and sulinemia (or impaired glucose tolerance, tension (48). The converse relationship
fasting glucose in diabetic and nondia- which is equated to a state of insulin re- has also been seen. Among hypertensive
betic patients. In both groups, they found sistance) has been associated with coro- patients, diabetes is a significant predictor
a significant association between each nary artery disease (CAD) and subsequent of ischemic stroke (OR 3.76 [95% CI
measure of glucose control and stroke risk cardiovascular events (41– 44). Several 1.67– 8.46]) (49). Data from the ARIC
using multivariate analysis (35). More re- studies have evaluated whether an analo- Study suggest a similar increased risk
cent data from the Atherosclerosis Risk in gous relationship exists between insulin among diabetic patients with prehyper-
Communities (ARIC) Study reiterated resistance and stroke. In a retrospective tension, as compared with nondiabetic
this relationship, finding an increased RR study, impaired glucose tolerance was not patients, although the number of strokes
of stroke with increasing levels of A1C in associated with stroke (45). A prospective was insufficient to calculate an RR for
both diabetic and nondiabetic patients study of Japanese men found no relation- stroke alone (50). No study has included
(36). In contrast, the European Prospec- ship between insulin resistance and statistical modeling to specifically address
tive Investigation Into Cancer (EPIC)- stroke incidence (46). In contrast, the whether hypertension fully accounts for
Norfolk Study did not find a significant ARIC Study found an increase in RR for the increased risk of stoke in diabetic pa-
relationship between A1C and stroke risk ischemic stroke of 1.19 for every 50 tients. It appears that the two are syner-
until a threshold level was reached (37). pmol/l increase in basal insulin among gistic in increasing stroke risk and
The only clinical trial to date that has nondiabetic patients, supporting a role account for up to 40% of the population-
directly evaluated the effect of tight glu- for insulin resistance (2). This was similar attributable risk for all ischemic strokes
cose control on stroke is the UK Preven- to results from the elderly patient popu- (7). A number of studies have found anti-
tion in Diabetes Study (UKPDS). Type 2 lation of the Finnish cohort study that in- hypertensive treatment to reduce the inci-
diabetic patients in the intensive treat- cluded both diabetic patients and dence of cardiovascular events, including
ment group (average A1C 7.0%) had no nondiabetic patients (35). As with studies stroke, in those with diabetes (51–57), but
significant reduction in stroke incidence of insulin resistance and cardiovascular fewer studies have focused on stroke specif-
(P ⫽ 0.52) compared with those receiving disease, the association of insulin resis- ically. The Systolic Hypertension in Europe
traditional medical therapy (average A1C tance with stroke is attenuated by the ad- Trial specifically noted a 73% decrease in
7.9%), indicating that tight glucose con- justment for other cardiovascular risk stroke incidence in diabetic patients treated
trol is not sufficient to prevent excess factors (2,35,43,44). However, data from with antihypertensive medication. Stroke
strokes (38,39), though the study may the Third National Health and Nutrition incidence was decreased in nondiabetic pa-
not have been sufficiently powered to de- Survey (NHANES III) revealed a small, tients by 38% (58). Thus, diabetic patients
tect a stroke-specific relationship and/or but significant, independent association appear to benefit preferentially from antihy-
the intensive control may not have been between insulin resistance and stroke pertensive treatment.
“intensive enough” to substantially im- when other risk factors such as hyperten-
pact stroke incidence. sion and level of glycemic control were HYPERLIPIDEMIA — Hyperlipid-
To summarize, there is no clear rela- taken into account (OR 1.06) (47). To emia is one of the most important risk

3132 DIABETES CARE, VOLUME 30, NUMBER 12, DECEMBER 2007

 Air and Kissela

factors for CHD and CAD but a less im- to find a convincing association between obesity—is a significant risk factor for
portant risk factor for stroke. As with hy- BMI and risk for stroke (2,72). An associ- ischemic stroke (81). Regardless, the im-
pertension, diabetic patients who have ation has been noted in studies of specific pact that obesity has on the risk of
suffered a stroke are more likely to have subpopulations, such as middle-aged, diabetes, CAD, hypertension, and
hyperlipidemia than those without diabe- Korean, or nonsmoking Japanese men hyperlipidemia will confound studies
tes (16 vs. 8%, respectively, P ⬍ 0.0001 in (73–75). The Nurses’ Health Study re- that address the risk of stroke (71). It has
the GCNKSS) (7,10). It is currently not ported a significant association with BMI, been estimated that the reductions in di-
clear to what degree the increased preva- such that subjects with BMI 27–28.9 abetes, hypertension, and hyperlipidemia
lence of hyperlipidemia accounts for the kg/m2 had an RR of 1.8 (95% CI 1.2–2.6), associated with a 10% weight loss could
increased risk of stroke, especially as the subjects with BMI 29 –31.9 kg/m2 had an lead to reduction of stroke of up to 13 per
contribution of hyperlipidemia alone to RR of 1.9 (1.3–2.8), and subjects with 1,000 people (82).
stroke incidence is controversial (59 – BMI ⱖ32 kg/m2 had an RR of 2.4 (1.6 –
64). Subset analysis from large placebo- 3.5) compared with those with BMI ⬍25 MICROALBUMINURIA — The
controlled trials, such as the Helsinki kg/m2 (76). A less robust, but still signif- World Health Organization definition of
Heart Study and Scandinavian Simvastatin icant, association was found in the Wom- the metabolic syndrome also includes mi-
Survival Study, which evaluated choles- en’s Health Study (77). The Physician’s croalbuminuria (30 –300 mg/24 h) as a
terol reduction as primary or secondary Health Study found an RR of 1.95 (1.39 – final component. Microalbuminuria is a
prevention of cardiovascular disease, in- 2.72) for ischemic stroke for those with significant marker of cardiovascular dis-
dicate that diabetic patients may benefit BMI ⬎30 kg/m2 compared with those ease and is highly associated with hyper-
preferentially from treatment in stroke re- with BMI ⬍23 kg/m2. The risk increased tension (83,84). It is encountered in
duction. Recently reported results from by 6% for each unit increase in BMI, al- diabetic patients more than twice as often
the Heart Protection Study, in contrast, though it was attenuated when other car- as in nondiabetic patients (84) and may
did not support this difference, finding diovascular risk factors were taken into also contribute to the increased risk of
that risk reduction did not vary with dia- account (78). stroke. The largest population-based pro-
betic status (65). The Collaborative AtoR- While BMI has been commonly used spective study to evaluate microalbumin-
vastatin Diabetes Study (CARDS), which in the literature as an obesity measure, uria and stroke risk is the EPIC-Norfolk
expressly evaluated the contribution of many studies have shown it to poorly re- Study. Among 23,630 individuals aged
hyperlipidemia to stroke risk in the dia- flect the health impact of obesity. Rather, 40 –79 years over 7.2 years of follow-up,
betic population without known CAD, abdominal obesity has been more specif- microalbuminuria conferred a signifi-
was halted early due to a significant 48% ically associated with vascular disease and cantly increased risk of total and ischemic
reduction in the incidence of stroke other health complications (79). Waist-to- stroke in multivariate modeling (hazard
among the treatment group (66). CARDS, hip ratio (WHR), while highly correlated ratio [HR] 1.49 [95% CI 1.13–2.14] and
taken together with the Stroke Prevention with BMI, better represents abdominal 2.01 [1.29 –3.31], respectively) (85).
by Aggressive Reduction in Cholesterol obesity and therefore may provide addi- Data from the Heart Outcomes Preven-
Levels Study (67), is highly significant in tional information on stroke risk. Despite tion Evaluation Study implicate mi-
that statin treatment can now be recom- the lack of a relationship between stroke croalbuminuria as a factor in stroke
mended for stroke prevention even in pa- and BMI, the Northern Manhattan Stroke incidence among those with diabetes
tients who do not have cardiovascular Study did find a significant relationship (57). Treatment of nonhypertensive dia-
disease, regardless of diabetes status. between WHR and risk of stroke. Analysis betic patients with an ACE inhibitor, a
However, based on the CARDS results, it included 576 ischemic stroke patients class of medications known to reduce mi-
seems that patients with diabetes may sig- and 1,142 age-, sex-, and race/ethnicity- croalbuminuria (86 – 88), reduced stroke
nificantly benefit from statins, making it matched control subjects. Compared incidence by 32% despite a minimal de-
even more important that those with dia- with the first quartile, the third and fourth crease in blood pressure (57). These data
betes be considered for statin treatment as quartiles of WHR had an increased risk of support a role for microalbuminuria in
part of their stroke prevention regimen. stroke (third quartile: OR 2.4 [95% CI increasing the risk of ischemic stroke,
1.5–3.9]; fourth quartile: 3.0 [1.8 – 4.8]) which may not be entirely dependent on
OBESITY — Obesity contributes to after adjustment for other risk factors. its direct relationship with hypertension
more than 300,000 deaths per year and These findings were consistent across and other well-known stroke risk factors.
nearly doubles the risk of death from all both sexes and all race/ethnic groups, al-
causes (68 –70). Given its particular asso- though the effect of WHR was stronger THE METABOLIC
ciation with CAD, hypertension, and dia- among younger subjects (72). Direct SYNDROME — Each of the compo-
betes (71), investigators have attempted comparison of BMI versus WHR and nents of the metabolic syndrome is asso-
to discern the contribution that obesity stroke risk in 28,643 male health care ciated with higher stroke risk to various
makes to stroke incidence with variable professionals without previous cardiovas- degrees, as described above. As has been
results. Many studies utilize BMI (mea- cular or cerebrovascular disease yielded mentioned, analysis of individual factors
sured as weight divided by the square of similar results. RR for the first and fifth causes substantial adjustment of observed
height in meters), which provides a broad quintiles of WHR was 2.33 (95% CI 1.25– risk because of the interrelationship of
though nonspecific estimate of obesity, is 4.37), whereas that for the first and fifth these factors. Therefore, studying the
easily obtained from patient self-report or quintiles of BMI was 1.29 (0.73–2.27) metabolic syndrome as a whole may pro-
medical charts and is commonly used in (80). vide a better estimation of the true risk for
clinical practice. Both the ARIC and Taken together, these studies suggest ischemic stroke.
Northern Manhattan Stroke Study failed that obesity—in particular, abdominal The Botnia Study examined risk for

DIABETES CARE, VOLUME 30, NUMBER 12, DECEMBER 2007 3133

Diabetes and ischemic stroke

cardiovascular events and stroke con- role in the response to ischemic challenge (115). Thromboxane A2 is also elevated in
ferred by the metabolic syndrome in (96,97). diabetic patients, possibly contributing to
4,483 subjects. In a multiple logistic re- Only indirect evidence is available at hyperaggregation as well (116).
gression analysis, the metabolic syndrome the present time linking NO dysregula- The relative contribution of these
was a significant independent risk for tion and stroke. A recent study found a mechanisms to increased ischemic stroke
stroke (RR 2.3, P ⬍ 0.001 compared with decreased response of cerebrovascular risk in those with diabetes has not been
those without the metabolic syndrome). blood flow to NO synthase inhibition in specifically evaluated, although several
None of the individual components of the diabetic patients compared with nondia- studies have implicated these pathways in
metabolic syndrome contributed signifi- betic patients, although not enough pa- the general population. In both cross-
cantly to stroke risk (89). Similar results tients were enrolled to determine sectional and prospective studies, in-
were obtained from examination of more significance (98). In addition, parasym- creased tissue plasminogen activator
than 10,000 subjects in the NHANES III. pathetic neurons that secrete NO into the antigen and plasminogen activator inhib-
In logistic regression modeling, the meta- perivascular space have been docu- itor-1 levels have been significantly asso-
bolic syndrome was associated with in- mented to degenerate and eventually die ciated with ischemic stroke (117–119).
creased odds of stroke (OR 2.2 [95% CI in the absence of insulin signaling (99). Treatment with aspirin or clopidogrel tar-
1.5–3.2] compared with those without Numerous studies have found that HMG- gets platelet aggregation by inhibiting
the metabolic syndrome). After the meta- CoA reductase inhibitors (statins), which thromboxane A2 and ADP, respectively,
bolic syndrome was in the model, each upregulate NO synthesis in addition to and are now widely used in the secondary
individual component was also tested. their activity in stabilizing atherosclerotic prevention of stroke, as they significantly
Only hypertriglyceridemia entered as an plaques (100), significantly reduce the reduce the risk of recurrent stroke (120 –
additional factor with independent sig- risk of stroke (56,67,101–104). The dual 125). Several trials, such as the Clopi-
nificance, while hypertension and insu- actions of statins make it difficult to dis- dogrel for High Atherothrombotic Risk
lin resistance/diabetes trended toward tinguish which action exerts the greatest and Ischemic Stabilization, Management,
significance (90). A few studies have effect. However, the growing body of ev- and Avoidance (CHARISMA), Clopi-
evaluated the risk of stroke associated idence indicates that statins exert protec- dogrel versus Aspirin in Patients at Risk of
with the metabolic syndrome in the ab- tive effects against stroke independent of Ischemic Events (CAPRIE), and Manage-
sence of diabetes, revealing similar two- changes in cholesterol levels. ment of ATherothrombosis with Clopi-
fold increases (91,92). In the ARIC dogrel in High-risk patients (MATCH)
Study, both hypertension and low HDL HYPERCOAGUABILITY studies, evaluated whether diabetic pa-
cholesterol independently and signifi- CONFERRED BY DIABETES — tients derived more or less benefit from
cantly increased risk (92). Defects in endothelial function may be antiplatelet therapy in preventing recur-
The data presented above provide ev- further confounded by the hypercoagula- rent ischemic events with mixed results.
idence that the individual components of ble state of diabetic patients. Plasminogen As the reported end point in these studies
the metabolic syndrome significantly activator inhibitor-1 and antithrombin was a composite of all ischemic events
contribute to the incidence of ischemic III, which inhibit fibrinolysis, as well as and mortality, the specific impact of anti-
stroke. These components are more prev- tissue plasminogen activator antigen, a platelet therapy in diabetic patients on
alent among diabetic patients and may act marker of impaired fibrinolysis, consis- stroke is unclear (126 –128). Further in-
synergistically to promote increased risk tently have been found to be elevated in vestigation is required to determine the
of stroke. In addition, several studies sup- diabetic patients and in those with insulin relative importance of these mechanisms
port a significant relationship between the resistance (105–107). Some studies have in diabetic patients.
collective metabolic syndrome and isch- further suggested that coagulation fac-
emic stroke. tors, such as factor VII, factor VIII, and CAROTID INTIMA-MEDIA
The metabolic syndrome and diabe- von Willebrand factor, also rise with de- THICKNESS — Consideration has
tes have their association with insulin re- gree of insulin resistance (108,109). This also been given to the impact of the in-
sistance in common. At a cellular and upregulation is likely secondary to a creased incidence of atherosclerosis
molecular level, insulin resistance confers chronic inflammatory state induced by among those with diabetes and stroke in-
changes that are becoming recognized as diabetes, as several inflammatory markers cidence. Carotid intima-media thickness
increasingly important in the pathophys- (e.g., C-reactive protein, lipoprotein- (CIMT) has been found in a number of
iology of vascular disease, including associated phospholipase A2) have been studies to be increased with diabetes. The
stroke. correlated with increased thrombotic fac- Insulin Resistance Atherosclerosis Study
tors and stroke incidence (108,110 – found a significant increase in common
112). The promotion of thrombus carotid thickness in the setting of estab-
ENDOTHELIAL formation likely occurs via platelet hyper- lished diabetes as compared with those
DYSFUNCTION AND NITRIC reactivity. Studies of platelets from dia- with newly diagnosed diabetes (129). Al-
OXIDE — Both diabetic patients and betic patients have found increased though not to the same degree, impaired
those with impaired glucose tolerance aggregation in response to ADP (113), a glucose tolerance is also associated with
have decreased endothelium-dependent response that may be mediated by the up- increased CIMT (130). Diabetic patients
vasodilation (93–95), due to either de- regulation of GPIIb-IIIa receptors that oc- that have suffered a stroke have signifi-
creased nitric oxide (NO) production or curs in diabetic patients (114). Insulin cantly greater CIMT than both those with-
impaired NO metabolism (95). Normally, normally acts to inhibit platelet aggrega- out stroke and nondiabetic patients
NO exerts a protective effect against plate- tion in response to ADP; however, this (131,132). As hyperglycemia, regardless
let aggregation and plays an important action is attenuated in diabetic patients of diabetes duration, was directly related

3134 DIABETES CARE, VOLUME 30, NUMBER 12, DECEMBER 2007

 Air and Kissela

to CIMT, tight glucose control may yield among these is educating those with di- Sarti C, Vartiainen E: Diabetes mellitus
benefits on carotid disease (129). abetes as to their true risk of stroke. A as a risk factor for death from stroke:
significant barrier appears to be the in- prospective study of the middle-aged
SURVIVING STROKE — Despite congruence between the information Finnish population. Stroke 27:210 –215,
1996
the uncertainties of the pathogenesis of the medical community believes it is
4. Wolf PA, D’Agostino RB, Belanger AJ,
stroke in those with diabetes, the im- imparting to patients and the actual Kannel WB: Probability of stroke: a risk
pact of hyperglycemia and diabetes on level of knowledge demonstrated by pa- profile from the Framingham Study.
outcomes has been more consistently de- tients. Ninety percent of physicians report Stroke 22:312–318, 1991
fined. Hyperglycemia during the post- discussing the risk of cardiovascular dis- 5. Ho JE, Paultre F, Mosca L: Is diabetes
stroke period, regardless of diabetic ease and the importance of prevention, mellitus a cardiovascular disease risk
status, is associated with increased mor- although only one-half of patients report equivalent for fatal stroke in women?
bidity and mortality. Studies have gener- their physician had discussed risk factor Data from the Women’s Pooling Project.
ally found increased 30-day and 1-year modification (165). Recent data from the Stroke 34:2812–2816, 2003
mortality rates among hyperglycemic pa- REduction of Atherothrombosis for Con- 6. Davis BR, Vogt T, Frost PH, Burlando A,
tients (133–137), although increased tinued Health (REACH) registry corrobo- Cohen J, Wilson A, Brass LM, Frishman
W, Price T, Stamler J: Risk factors for
mortality was not seen in other studies rates the continued undertreatment of stroke and type of stroke in persons with
(7,138). Morbidity, as defined by func- cardiovascular risk factors (166). Frequent isolated systolic hypertension: Systolic
tional outcome and neurologic recov- and repeated patient advising regarding Hypertension in the Elderly Program
ery, is also worsened in the setting of cardiovascular and cerebrovascular Cooperative Research Group. Stroke 29:
hyperglycemia and diabetes (134,139 – complications of diabetes and warning 1333–1340, 1998
142). This holds true among those with signs is necessary to improve utilization 7. Kissela BM, Khoury J, Kleindorfer D,
only transient hyperglycemia, although of primary and secondary prevention Woo D, Schneider A, Alwell K, Miller R,
such individuals fare better than those measures. Ewing I, Moomaw CJ, Szaflarski JP,
with chronically elevated glucose levels The potential benefit of aggressive Gebel J, Shukla R, Broderick JP: Epide-
whether diagnosed pre- or poststroke multiple risk-reduction measures in those miology of ischemic stroke in patients
with diabetes: the Greater Cincinnati/
with diabetes (143,144). In imaging with diabetes has been highlighted by the Northern Kentucky Stroke Study. Diabe-
studies, the initial infarct size and in- Steno-2 Study. Intensive standardized tes Care 28:355–359, 2005
farct progression are greater in hyper- risk factor reduction, including 1) treat- 8. Centers for Disease Control and Preven-
glycemic patients (142,145–147). One ment of hyperglycemia, hypertension, tion: National Diabetes Fact Sheet: Gen-
recent study has found a decreased re- dyslipidemia, and microalbuminuria; 2) eral Information and National Estimates on
canalization rate following recombinant secondary prevention of cardiovascular Diabetes in the United States, 2002. At-
tissue plasminogen activator (rt-PA) ad- disease with aspirin; and 3) behavioral lanta, GA, Department of Health and
ministration in the presence of hyper- modification, resulted in significant re- Human Services, Centers for Disease
glycemia, although this was not seen in ductions in cardiovascular disease, in- Control and Prevention Atlanta, 2003
the pivotal National Institute of Neuro- cluding stroke (HR 0.47 [95% CI 0.24 – 9. Grau AJ, Weimar C, Buggle F, Heinrich
A, Goertler M, Neumaier S, Glahn J,
logical Diseases and Stroke rt-PA trial 0.73]). This effect was larger than that
Brandt T, Hacke W, Diener HC: Risk fac-
(139,148). Normalization of glucose lev- seen in studies that targeted treatment to tors, outcome, and treatment in sub-
els was associated with 4.6 times individual risk factors (29). Although the types of ischemic stroke: the German
decreased risk in mortality in one specific mechanisms that underlie the re- stroke data bank. Stroke 32:2559 –2566,
retrospective study, indicating the poten- lationship between diabetes, the meta- 2001
tially large impact that can be made with bolic syndrome, and stroke require 10. Jorgensen H, Nakayama H, Raaschou
aggressive medical management in these ongoing investigation to provide new HO, Olsen TS: Stroke in patients with
patients (149). methods for prevention and treatment, diabetes: the Copenhagen Stroke Study.
Diabetes is also one of the most con- these data underscore the strides that can Stroke 25:1977–1984, 1994
sistent predictors of recurrent stroke or be made with the tools at hand. 11. Lai SM, Alter M, Friday G, Sobel E: Prog-
stroke after TIA (150 –162). The in- nosis for survival after an initial stroke.
Stroke 26:2011–2015, 1995
creased risk of recurrent stroke due to di- 12. Kissela B, Schneider A, Kleindorfer D,
abetes ranges from 2.1 to 5.6 times that of References Khoury J, Miller R, Alwell K, Woo D,
nondiabetic patients (154,156) and is in- 1. Kothari V, Stevens RJ, Adler AI, Stratton Szaflarski J, Gebel J, Moomaw C, Pan-
dependent of glucose control during the IM, Manley SE, Neil HA, Holman RR: cioli A, Jauch E, Shukla R, Broderick J:
interstroke period (163). The significance UKPDS 60: risk of stroke in type 2 dia- Stroke in a biracial population: the ex-
of these findings is underscored by the betes estimated by the UK Prospective cess burden of stroke among blacks.
increased morbidity and mortality associ- Diabetes Study risk engine. Stroke 33: Stroke 35:426 – 431, 2004
ated with recurrent stroke (164). 1776 –1781, 2002 13. Self-reported heart disease and stroke
2. Folsom AR, Rasmussen ML, Chambless among adults with and without diabete-
CHALLENGES AHEAD — Diabetes LE, Howard G, Cooper LS, Schmidt MI, s–United States, 1999 –2001. MMWR
Heiss G: Prospective associations of fast- Morb Mortal Wkly Rep 52:1065–1070,
significantly increases the risk of incident ing insulin, body fat distribution, and di- 2003
stroke and stroke recurrence. The magni- abetes with risk of ischemic stroke: the 14. Almdal T, Scharling H, Jensen JS, Vest-
tude of this problem will continue to ex- Atherosclerosis Risk in Communities ergaard H: The independent effect of
pand as the prevalence of diabetes increases (ARIC) Study Investigators. Diabetes type 2 diabetes mellitus on ischemic
in the U.S., thus presenting numerous Care 1077–1083, 1999 heart disease, stroke, and death: a pop-
challenges for the future. Foremost 3. Tuomilehto J, Rastenyte D, Jousilahti P, ulation-based study of 13,000 men and

DIABETES CARE, VOLUME 30, NUMBER 12, DECEMBER 2007 3135

Diabetes and ischemic stroke

women with 20 years of follow-up. Arch stroke in patients with atherosclerotic 352:837– 853, 1998
Intern Med 164:1422–1426, 2004 disease of the great vessels: the role of 40. Uwaifo GI, Ratner RE: The roles of insu-
15. Ottenbacher KJ, Ostir GV, Peek MK, diabetes. Stroke 14:709 –712, 1983 lin resistance, hyperinsulinemia, and
Markides KS: Diabetes mellitus as a risk 28. Lithner F, Asplund K, Eriksson S, Hagg thiazolidinediones in cardiovascular dis-
factor for stroke incidence and mortality E, Strand T, Wester PO: Clinical charac- ease. Am J Med 115 (Suppl. 8A):12S–
in Mexican American older adults. J Ger- teristics in diabetic stroke patients. Dia- 19S, 2003
ontol A Biol Sci Med Sci 59:M640 –M645, betes Metab 14:15–19, 1988 41. Despres JP, Lamarche B, Mauriege P,
2004 29. Gaede P, Vedel P, Larsen N, Jensen GV, Cantin B, Dagenais GR, Moorjani S, Lu-
16. Abbott RD, Donahue RP, MacMahon Parving HH, Pedersen O: Multifactorial pien PJ: Hyperinsulinemia as an inde-
SW, Reed DM, Yano K: Diabetes and the intervention and cardiovascular disease pendent risk factor for ischemic heart
risk of stroke: the Honolulu Heart Pro- in patients with type 2 diabetes. N Engl disease. N Engl J Med 334:952–957,
gram. JAMA 257:949 –952, 1987 J Med 348:383–393, 2003 1996
17. Adams HP Jr, Putman SF, Kassell NF, 30. Johnston SC, Rothwell PM, Nguyen- 42. Pyorala M, Miettinen H, Halonen P,
Torner JC: Prevalence of diabetes melli- Huynh MN, Giles MF, Elkins JS, Bern- Laakso M, Pyorala K: Insulin resistance
tus among patients with subarachnoid stein AL, Sidney S: Validation and syndrome predicts the risk of coronary
hemorrhage. Arch Neurol 41:1033– refinement of scores to predict very early heart disease and stroke in healthy mid-
1035, 1984 stroke risk after transient ischaemic at- dle-aged men: the 22-year follow-up re-
18. Karapanayiotides T, Piechowski-Jozwiak tack. Lancet 369:283–292, 2007 sults of the Helsinki Policemen Study.
B, van Melle G, Bogousslavsky J, Devuyst 31. Mooradian AD: Cardiovascular disease Arterioscler Thromb Vasc Biol 20:538 –
G: Stroke patterns, etiology, and progno- in type 2 diabetes mellitus: current man- 544, 2000
sis in patients with diabetes mellitus. Neu- agement guidelines. Arch Intern Med 163: 43. Lakka HM, Lakka TA, Tuomilehto J, Siv-
rology 62:1558 –1562, 2004 33– 40, 2003 enius J, Salonen JT: Hyperinsulinemia
19. Brown RD, Whisnant JP, Sicks JD, 32. Bloomgarden ZT: Cardiovascular dis- and the risk of cardiovascular death and
O’Fallon WM, Wiebers DO: Stroke inci- ease in type 2 diabetes. Diabetes Care 22: acute coronary and cerebrovascular
dence, prevalence, and survival: secular 1739 –1744, 1999 events in men: the Kuopio Ischaemic
trends in Rochester, Minnesota, through 33. Grundy SM, Benjamin IJ, Burke GL, Heart Disease Risk Factor Study. Arch
1989. Stroke 27:373–380, 1996 Chait A, Eckel RH, Howard BV, Mitch Intern Med 160:1160 –1168, 2000
20. Sacco RL, Boden-Albala B, Gan R, Chen W, Smith SC Jr, Sowers JR: Diabetes and 44. Bonora E, Willeit J, Kiechl S, Oberhollen-
X, Kargman DE, Shea S, Paik MC, cardiovascular disease: a statement for zer F, Egger G, Bonadonna R, Muggeo M:
Hauser WA: Stroke incidence among healthcare professionals from the Amer- U-shaped and J-shaped relationships be-
white, black, and Hispanic residents of ican Heart Association. Circulation 100: tween serum insulin and coronary heart
an urban community: the Northern 1134 –1146, 1999 disease in the general population: the
Manhattan Stroke Study. Am J Epidemiol 34. Lehto S, Ronnemaa T, Pyorala K, Laakso Bruneck Study. Diabetes Care 21:221–
147:259 –268, 1998 M: Predictors of stroke in middle-aged 230, 1998
21. Broderick J, Brott T, Kothari R, Miller R, patients with non-insulin-dependent di- 45. Qureshi AI, Giles WH, Croft JB: Im-
Khoury J, Pancioli A, Gebel J, Mills D, abetes. Stroke 27:63– 68, 1996 paired glucose tolerance and the likeli-
Minneci L, Shukla R: The Greater Cin- 35. Kuusisto J, Mykkanen L, Pyorala K, hood of nonfatal stroke and myocardial
cinnati/Northern Kentucky Stroke Laakso M: Non-insulin-dependent dia- infarction: the Third National Health
Study: preliminary first-ever and total betes and its metabolic control are im- and Nutrition Examination Survey.
incidence rates of stroke among blacks. portant predictors of stroke in elderly Stroke 29:1329 –1332, 1998
Stroke 29:415– 421, 1998 subjects. Stroke 25:1157–1164, 1994 46. Adachi H, Hirai Y, Tsuruta M, Fujiura Y,
22. Morgenstern LB, Smith MA, Lisabeth 36. Selvin E, Coresh J, Shahar E, Zhang L, Imaizuml T: Is insulin resistance or dia-
LD, Risser JM, Uchino K, Garcia N, Steffes M, Sharrett AR: Glycaemia (hae- betes mellitus associated with stroke? An
Longwell PJ, McFarling DA, Akuwumi moglobin A1c) and incident ischaemic 18-year follow-up study. Diabetes Res
O, Al-Wabil A, Al-Senani F, Brown DL, stroke: the Atherosclerosis Risk in Com- Clin Pract 51:215–223, 2001
Moye LA: Excess stroke in Mexican munities (ARIC) Study. Lancet Neurol 47. Bravata DM, Wells CK, Kernan WN,
Americans compared with non-His- 4:821– 826, 2005 Concato J, Brass LM, Gulanski BI: Asso-
panic whites: the Brain Attack Surveil- 37. Myint PK, Sinha S, Wareham NJ, Bing- ciation between impaired insulin sensi-
lance in Corpus Christi Project. Am J ham SA, Luben RN, Welch AA, Khaw tivity and stroke. Neuroepidemiology 25:
Epidemiol 160:376 –383, 2004 KT: Glycated hemoglobin and risk of 69 –74, 2005
23. Harris MI: Diabetes in America: epide- stroke in people without known diabe- 48. Adler AI, Stratton IM, Neil HA, Yudkin
miology and scope of the problem. Dia- tes in the European Prospective Investi- JS, Matthews DR, Cull CA, Wright AD,
betes Care 21 (Suppl. 3):C11–C14, 1998 gation into Cancer (EPIC)-Norfolk Turner RC, Holman RR: Association of
24. Kannel WB, McGee DL: Diabetes and prospective population study: a thresh- systolic blood pressure with macrovas-
cardiovascular disease: the Framingham old relationship? Stroke 38:271–275, cular and microvascular complications
study. JAMA 241:2035–2038, 1979 2007 of type 2 diabetes (UKPDS 36): prospec-
25. Booth GL, Kapral MK, Fung K, Tu JV: 38. Effect of intensive blood-glucose control tive observational study. BMJ 321:412–
Relation between age and cardiovascular with metformin on complications in 419, 2000
disease in men and women with diabetes overweight patients with type 2 diabetes 49. Khealani BA, Syed NA, Maken S, Mapari
compared with non-diabetic people: a (UKPDS 34): UK Prospective Diabetes UU, Hameed B, Ali S, Qureshi R, Akhter
population-based retrospective cohort Study (UKPDS) Group. Lancet 352:854 – N, Hassan A, Sonawalla AB, Baig SM,
study. Lancet 368:29 –36, 2006 865, 1998 Wasay M: Predictors of ischemic versus
26. Fritz VU, Bilchik T, Levien LJ: Diabetes 39. Intensive blood-glucose control with hemorrhagic strokes in hypertensive pa-
as risk factor for transient ischaemic at- sulphonylureas or insulin compared tients. J Coll Physicians Surg Pak 15:22–
tacks as opposed to strokes. Eur J Vasc with conventional treatment and risk of 25, 2005
Surg 1:259 –262, 1987 complications in patients with type 2 di- 50. Kshirsagar AV, Carpenter M, Bang H,
27. Weinberger J, Biscarra V, Weisberg MK, abetes (UKPDS 33): UK Prospective Di- Wyatt SB, Colindres RE: Blood pressure
Jacobson JH: Factors contributing to abetes Study (UKPDS) Group. Lancet usually considered normal is associated

3136 DIABETES CARE, VOLUME 30, NUMBER 12, DECEMBER 2007

 Air and Kissela

with an elevated risk of cardiovascular coronary heart disease in men with ab- emic stroke: the Northern Manhattan
disease. Am J Med 119:133–141, 2006 normal glucose tolerance: the Honolulu Stroke Study. Stroke 34:1586 –1592,
51. Efficacy of atenolol and captopril in re- Heart Program. J Intern Med 234:471– 2003
ducing risk of macrovascular and mi- 478, 1993 73. Jood K, Jern C, Wilhelmsen L, Rosen-
crovascular complications in type 2 61. Kagan A, Popper JS, Rhoads GG: Factors gren A: Body mass index in mid-life is
diabetes: UKPDS 39: UK Prospective related to stroke incidence in Hawaii associated with a first stroke in men: a
Diabetes Study Group. BMJ 317:713– Japanese men: the Honolulu Heart prospective population study over 28
720, 1998 Study. Stroke 11:14 –21, 1980 years. Stroke 35:2764 –2769, 2004
52. Pahor M, Psaty BM, Alderman MH, 62. Shahar E, Chambless LE, Rosamond 74. Song YM, Sung J, Davey Smith G, Ebra-
Applegate WB, Williamson JD, Furberg WD, Boland LL, Ballantyne CM, McGov- him S: Body mass index and ischemic
CD: Therapeutic benefits of ACE inhib- ern PG, Sharrett AR: Plasma lipid profile and hemorrhagic stroke: a prospective
itors and other antihypertensive drugs in and incident ischemic stroke: the Ath- study in Korean men. Stroke 35:831–
patients with type 2 diabetes. Diabetes erosclerosis Risk in Communities (ARIC) 836, 2004
Care 23:888 – 892, 2000 study. Stroke 34:623– 631, 2003 75. Abbott RD, Behrens GR, Sharp DS, Ro-
53. Tatti P, Pahor M, Byington RP, Di Mauro 63. Harmsen P, Rosengren A, Tsipogianni A, driguez BL, Burchfiel CM, Ross GW,
P, Guarisco R, Strollo G, Strollo F: Out- Wilhelmsen L: Risk factors for stroke in Yano K, Curb JD: Body mass index and
come results of the Fosinopril Versus middle-aged men in Goteborg, Sweden. thromboembolic stroke in nonsmoking
Amlodipine Cardiovascular Events Ran- Stroke 21:223–229, 1990 men in older middle age: the Honolulu
domized Trial (FACET) in patients with 64. Cholesterol, diastolic blood pressure, Heart Program. Stroke 25:2370 –2376,
hypertension and NIDDM. Diabetes Care and stroke: 13,000 strokes in 450,000 1994
21:597– 603, 1998 people in 45 prospective cohorts: pro- 76. Rexrode KM, Hennekens CH, Willett
54. Hansson L, Lindholm LH, Niskanen L, spective studies collaboration. Lancet WC, Colditz GA, Stampfer MJ, Rich-
Lanke J, Hedner T, Niklason A, Luoman- 346:1647–1653, 1995 Edwards JW, Speizer FE, Manson JE: A
maki K, Dahlof B, de Faire U, Morlin C, 65. Collins R, Armitage J, Parish S, Sleigh P, prospective study of body mass index,
Karlberg BE, Wester PO, Bjorck JE: Ef- Peto R: MRC/BHF Heart Protection weight change, and risk of stroke in
fect of angiotensin-converting-enzyme Study of cholesterol-lowering with sim- women. JAMA 277:1539 –1545, 1997
inhibition compared with conventional vastatin in 5963 people with diabetes: a 77. Kurth T, Gaziano JM, Rexrode KM, Kase
therapy on cardiovascular morbidity randomised placebo-controlled trial. CS, Cook NR, Manson JE, Buring JE:
and mortality in hypertension: the Cap- Lancet 361:2005–2016, 2003 Prospective study of body mass index
topril Prevention Project (CAPPP) ran- 66. Colhoun HM, Betteridge DJ, Durrington and risk of stroke in apparently healthy
domised trial. Lancet 353:611– 616, PN, Hitman GA, Neil HA, Livingstone women. Circulation 111:1992–1998,
1999 SJ, Thomason MJ, Mackness MI, Charl- 2005
55. Estacio RO, Jeffers BW, Hiatt WR, Big- ton-Menys V, Fuller JH: Primary preven- 78. Kurth T, Gaziano JM, Berger K, Kase CS,
gerstaff SL, Gifford N, Schrier RW: The tion of cardiovascular disease with Rexrode KM, Cook NR, Buring JE, Man-
effect of nisoldipine as compared with atorvastatin in type 2 diabetes in the Col- son JE: Body mass index and the risk of
enalapril on cardiovascular outcomes in laborative Atorvastatin Diabetes Study stroke in men. Arch Intern Med 162:
patients with non-insulin-dependent di- (CARDS): multicentre randomised pla- 2557–2562, 2002
abetes and hypertension. N Engl J Med cebo-controlled trial. Lancet 364:685– 79. Megnien JL, Denarie N, Cocaul M, Si-
338:645– 652, 1998 696, 2004 mon A, Levenson J: Predictive value of
56. Effects of ramipril on cardiovascular and 67. Amarenco P, Bogousslavsky J, Callahan waist-to-hip ratio on cardiovascular risk
microvascular outcomes in people with A 3rd, Goldstein LB, Hennerici M, Ru- events. Int J Obes Relat Metab Disord 23:
diabetes mellitus: results of the HOPE dolph AE, Sillesen H, Simunovic L, Sza- 90 –97, 1999
study and MICRO-HOPE substudy: rek M, Welch KM, Zivin JA: High-dose 80. Walker SP, Rimm EB, Ascherio A, Kawa-
Heart Outcomes Prevention Evaluation atorvastatin after stroke or transient chi I, Stampfer MJ, Willett WC: Body
Study Investigators. Lancet 355:253– ischemic attack. N Engl J Med 355:549 – size and fat distribution as predictors of
259, 2000 559, 2006 stroke among US men. Am J Epidemiol
57. Bosch J, Yusuf S, Pogue J, Sleight P, Lonn 68. Centers for Disease Control (CDC): The 144:1143–1150, 1996
E, Rangoonwala B, Davies R, Ostergren Surgeon General’s 1989 report on re- 81. Lu M, Ye W, Adami HO, Weiderpass E:
J, Probstfield J: Use of ramipril in pre- ducing the health consequences of Prospective study of body size and risk
venting stroke: double blind random- smoking: 25 years of progress. MMWR for stroke amongst women below age
ised trial. BMJ 324:699 –702, 2002 Morb Mortal Wkly Rep 38 (Suppl. 2):1– 60. J Intern Med 260:442– 450, 2006
58. Tuomilehto J, Rastenyte D, Birkenhager 32, 1989 82. Oster G, Thompson D, Edelsberg J, Bird
WH, Thijs L, Antikainen R, Bulpitt CJ, 69. Mokdad AH, Marks JS, Stroup DF, Ger- AP, Colditz GA: Lifetime health and eco-
Fletcher AE, Forette F, Goldhaber A, berding JL: Correction: actual causes of nomic benefits of weight loss among
Palatini P, Sarti C, Fagard R: Effects of death in the United States, 2000. JAMA obese persons. Am J Public Health 89:
calcium-channel blockade in older pa- 293:293–294, 2005 1536 –1542, 1999
tients with diabetes and systolic hyper- 70. Mokdad AH, Marks JS, Stroup DF, Ger- 83. Ravera M, Ratto E, Vettoretti S, Viazzi F,
tension: Systolic Hypertension in berding JL: Actual causes of death in the Leoncini G, Parodi D, Tomolillo C, Del
Europe Trial Investigators. N Engl J Med United States, 2000. JAMA 291:1238 – Sette M, Maviglio N, Deferrari G, Pon-
340:677– 684, 1999 1245, 2004 tremoli R: Microalbuminuria and sub-
59. Iso H, Jacobs DR Jr, Wentworth D, Nea- 71. Wilson PW, D’Agostino RB, Sullivan L, clinical cerebrovascular damage in
ton JD, Cohen JD: Serum cholesterol lev- Parise H, Kannel WB: Overweight and essential hypertension. J Nephrol 15:
els and six-year mortality from stroke in obesity as determinants of cardiovascu- 519 –524, 2002
350,977 men screened for the multiple lar risk: the Framingham experience. 84. Gerstein HC, Mann JF, Yi Q, Zinman B,
risk factor intervention trial. N Engl J Med Arch Intern Med 162:1867–1872, 2002 Dinneen SF, Hoogwerf B, Halle JP,
320:904 –910, 1989 72. Suk SH, Sacco RL, Boden-Albala B, Young J, Rashkow A, Joyce C, Nawaz S,
60. Laws A, Marcus EB, Grove JS, Curb JD: Cheun JF, Pittman JG, Elkind MS, Paik Yusuf S: Albuminuria and risk of cardio-
Lipids and lipoproteins as risk factors for MC: Abdominal obesity and risk of isch- vascular events, death, and heart failure

DIABETES CARE, VOLUME 30, NUMBER 12, DECEMBER 2007 3137

Diabetes and ischemic stroke

in diabetic and nondiabetic individuals. Increased basal levels of plasma nitric Tofler GH, Singer DE, Murphy-Sheehy
JAMA 286:421– 426, 2001 oxide in type 2 diabetic subjects: rela- PM, Lipinska I, D’Agostino RB, Wilson
85. Yuyun MF, Khaw KT, Luben R, Welch tionship to microvascular complica- PW: Hyperinsulinemia, hyperglycemia,
A, Bingham S, Day NE, Wareham NJ: tions. J Diabetes Complications 15:135– and impaired hemostasis: the Framing-
Microalbuminuria and stroke in a British 143, 2001 ham Offspring Study. JAMA 283:221–
population: the European Prospective 96. McCarty MF: Up-regulation of endothe- 228, 2000
Investigation into Cancer in Norfolk lial nitric oxide activity as a central strat- 107. Davi G, Gennaro F, Spatola A, Catalano
(EPIC-Norfolk) population study. J In- egy for prevention of ischemic stroke: I, Averna M, Montalto G, Amato S, No-
tern Med 255:247–256, 2004 just say NO to stroke! Med Hypotheses tarbartolo A: Thrombin-antithrombin
86. Marre M, Lievre M, Chatellier G, Mann 55:386 – 403, 2000 III complexes in type II diabetes melli-
JF, Passa P, Menard J: Effects of low dose 97. Cosentino F, Rubattu S, Savoia C, Ven- tus. J Diabetes Complications 6:7–11, 1992
ramipril on cardiovascular and renal turelli V, Pagannonne E, Volpe M: En- 108. Romano M, Guagnano MT, Pacini G,
outcomes in patients with type 2 diabe- dothelial dysfunction and stroke. Vigneri S, Falco A, Marinopiccoli M,
tes and raised excretion of urinary albu- J Cardiovasc Pharmacol 38 (Suppl. 2): Manigrasso MR, Basili S, Davi G: Associ-
min: randomised, double blind, placebo S75–S78, 2001 ation of inflammation markers with
controlled trial (the DIABHYCAR 98. Nazir FS, Alem M, Small M, Connell JM, impaired insulin sensitivity and coag-
study). BMJ 328:495, 2004 Lees KR, Walters MR, Cleland SJ: ulative activation in obese healthy
87. Slowik A, Turaj W, Iskra T, Strojny J, Blunted response to systemic nitric ox- women. J Clin Endocrinol Metab 88:
Szczudlik A: Microalbuminuria in non- ide synthase inhibition in the cerebral 5321–5326, 2003
diabetic patients with acute ischemic circulation of patients with type 2 diabe- 109. Mertens I, Van Gaal LF: Obesity, haemo-
stroke: prevalence, clinical correlates, tes. Diabet Med 23:398 – 402, 2006 stasis and the fibrinolytic system. Obes
and prognostic significance. Cerebrovasc 99. Cellek S, Anderson PN, Foxwell NA: Ni- Rev 3:85–101, 2002
Dis 14:15–21, 2002 trergic neurodegeneration in cerebral ar- 110. Ballantyne CM, Hoogeveen RC, Bang H,
88. Lasaridis AN, Sarafidis PA: Diabetic ne- teries of streptozotocin-induced diabetic Coresh J, Folsom AR, Chambless LE,
phropathy and antihypertensive treat- rats: a new insight into diabetic stroke. Myerson M, Wu KK, Sharrett AR, Boer-
ment: what are the lessons from clinical Diabetes 54:212–219, 2005 winkle E: Lipoprotein-associated phos-
trials? Am J Hypertens 16:689 – 697, 100. Endres M, Laufs U, Huang Z, Nakamura pholipase A2, high-sensitivity C-reactive
2003 T, Huang P, Moskowitz MA, Liao JK: protein, and risk for incident ischemic
89. Isomaa B, Almgren P, Tuomi T, Forsen Stroke protection by 3-hydroxy-3- stroke in middle-aged men and women
B, Lahti K, Nissen M, Taskinen MR, methylglutaryl (HMG)-CoA reductase in the Atherosclerosis Risk in Commu-
Groop L: Cardiovascular morbidity and inhibitors mediated by endothelial nitric nities (ARIC) study. Arch Intern Med
mortality associated with the metabolic oxide synthase. Proc Natl Acad Sci U S A 165:2479 –2484, 2005
syndrome. Diabetes Care 24:683– 689, 95:8880 – 8885, 1998 111. Adeniyi A, Folsom AR, Brancati FL, Des-
2001 101. Gil-Nunez AC, Villanueva JA: Advan- vorieux M, Pankow JS, Taylor H: Inci-
90. Ninomiya JK, L’Italien G, Criqui MH, tages of lipid-lowering therapy in cere- dence and risk factors for cardiovascular
Whyte JL, Gamst A, Chen RS: Associa- bral ischemia: role of HMG-CoA disease in African Americans with diabe-
tion of the metabolic syndrome with his- reductase inhibitors. Cerebrovasc Dis 11 tes: the Atherosclerosis Risk in Commu-
tory of myocardial infarction and stroke (Suppl. 1):85–95, 2001 nities (ARIC). J Natl Med Assoc 94:1025–
in the Third National Health and Nutri- 102. Plehn JF, Davis BR, Sacks FM, Rouleau 1035, 2002
tion Examination Survey. Circulation JL, Pfeffer MA, Bernstein V, Cuddy TE, 112. Engstrom G, Stavenow L, Hedblad B,
109:42– 46, 2004 Moye LA, Piller LB, Rutherford J, Simp- Lind P, Eriksson KF, Janzon L,
91. Kurl S, Laukkanen JA, Niskanen L, Laak- son LM, Braunwald E: Reduction of Lindgarde F: Inflammation-sensitive
sonen D, Sivenius J, Nyyssonen K, Sa- stroke incidence after myocardial infarc- plasma proteins, diabetes, and mortality
lonen JT: Metabolic syndrome and the tion with pravastatin: the Cholesterol and incidence of myocardial infarction
risk of stroke in middle-aged men. Stroke and Recurrent Events (CARE) study: the and stroke: a population-based study.
37:806 – 811, 2006 Care Investigators. Circulation 99:216 – Diabetes 52:442– 447, 2003
92. McNeill AM, Rosamond WD, Girman 223, 1999 113. Mandal S, Sarode R, Dash S, Dash RJ:
CJ, Golden SH, Schmidt MI, East HE, 103. Blauw GJ, Lagaay AM, Smelt AH, West- Hyperaggregation of platelets detected
Ballantyne CM, Heiss G: The metabolic endorp RG: Stroke, statins, and choles- by whole blood platelet aggregometry in
syndrome and 11-year risk of incident terol: a meta-analysis of randomized, newly diagnosed noninsulin-dependent
cardiovascular disease in the Atheroscle- placebo-controlled, double-blind trials diabetes mellitus. Am J Clin Pathol 100:
rosis Risk in Communities study. Diabe- with HMG-CoA reductase inhibitors. 103–107, 1993
tes Care 28:385–390, 2005 Stroke 28:946 –950, 1997 114. Tschoepe D, Roesen P, Kaufmann L,
93. Hogikyan RV, Galecki AT, Pitt B, Halter 104. Collins R, Armitage J, Parish S, Sleight P, Schauseil S, Kehrel B, Ostermann H,
JB, Greene DA, Supiano MA: Specific Peto R: Effects of cholesterol-lowering Gries FA: Evidence for abnormal platelet
impairment of endothelium-dependent with simvastatin on stroke and other glycoprotein expression in diabetes mel-
vasodilation in subjects with type 2 dia- major vascular events in 20536 people litus. Eur J Clin Invest 20:166 –170, 1990
betes independent of obesity. J Clin En- with cerebrovascular disease or other 115. Trovati M, Mularoni EM, Burzacca S,
docrinol Metab 83:1946 –1952, 1998 high-risk conditions. Lancet 363:757– Ponziani MC, Massucco P, Mattiello L,
94. Steinberg HO, Chaker H, Leaming R, 767, 2004 Piretto V, Cavalot F, Anfossi G: Impaired
Johnson A, Brechtel G, Baron AD: Obe- 105. Aoki I, Shimoyama K, Aoki N, Homori insulin-induced platelet antiaggregating
sity/insulin resistance is associated with M, Yanagisawa A, Nakahara K, Kawai Y, effect in obesity and in obese NIDDM
endothelial dysfunction: implications Kitamura SI, Ishikawa K: Platelet-depen- patients. Diabetes 44:1318 –1322, 1995
for the syndrome of insulin resistance. dent thrombin generation in patients 116. Halushka PV, Rogers RC, Loadholt CB,
J Clin Invest 97:2601–2610, 1996 with diabetes mellitus: effects of glyce- Colwell JA: Increased platelet throm-
95. Maejima K, Nakano S, Himeno M, Tsuda mic control on coagulability in diabetes. boxane synthesis in diabetes mellitus.
S, Makiishi H, Ito T, Nakagawa A, Kigo- J Am Coll Cardiol 27:560 –566, 1996 J Lab Clin Med 97:87–96, 1981
shi T, Ishibashi T, Nishio M, Uchida K: 106. Meigs JB, Mittleman MA, Nathan DM, 117. Macko RF, Kittner SJ, Epstein A, Cox

3138 DIABETES CARE, VOLUME 30, NUMBER 12, DECEMBER 2007

 Air and Kissela

DK, Wozniak MA, Wityk RJ, Stern BJ, dogrel alone after recent ischaemic erick JP, Kwiatkowski TG, Fineberg SE:
Sloan MA, Sherwin R, Price TR, Mc- stroke or transient ischaemic attack in Admission glucose level and clinical out-
Carter RJ, Johnson CJ, Earley CJ, Buch- high-risk patients (MATCH): random- comes in the NINDS rt-PA Stroke Trial.
holz DW, Stolley PD: Elevated tissue ised, double-blind, placebo-controlled Neurology 59:669 – 674, 2002
plasminogen activator antigen and trial. Lancet 364:331–337, 2004 140. Bruno A, Biller J, Adams HP Jr, Clarke
stroke risk: the Stroke Prevention In 129. Wagenknecht LE, D’Agostino R Jr, Sav- WR, Woolson RF, Williams LS, Hansen
Young Women Study. Stroke 30:7–11, age PJ, O’Leary DH, Saad MF, Haffner MD: Acute blood glucose level and out-
1999 SM: Duration of diabetes and carotid come from ischemic stroke: Trial of
118. Lindgren A, Lindoff C, Norrving B, Ast- wall thickness: the Insulin Resistance ORG 10172 in Acute Stroke Treatment
edt B, Johansson BB: Tissue plasmino- Atherosclerosis Study (IRAS). Stroke 28: (TOAST) Investigators. Neurology 52:
gen activator and plasminogen activator 999 –1005, 1997 280 –284, 1999
inhibitor-1 in stroke patients. Stroke 27: 130. Brohall G, Oden A, Fagerberg B: Carotid 141. Kiers L, Davis SM, Larkins R, Hopper J,
1066 –1071, 1996 artery intima-media thickness in pa- Tress B, Rossiter SC, Carlin J, Ratnaike S:
119. Ridker PM, Hennekens CH, Stampfer tients with type 2 diabetes mellitus and Stroke topography and outcome in rela-
MJ, Manson JE, Vaughan DE: Prospec- impaired glucose tolerance: a systematic tion to hyperglycaemia and diabetes.
tive study of endogenous tissue plas- review. Diabet Med 23:609 – 616, 2006 J Neurol Neurosurg Psychiatry 55:
minogen activator and risk of stroke. 131. Matsumoto K, Sera Y, Nakamura H, 263–270, 1992
Lancet 343:940 –943, 1994 Ueki Y, Miyake S: Correlation between 142. Parsons MW, Barber PA, Desmond PM,
120. Teal PA: Recent clinical trial results with common carotid arterial wall thickness Baird TA, Darby DG, Byrnes G, Tress
antiplatelet therapy: implications in and ischemic stroke in patients with type BM, Davis SM: Acute hyperglycemia
stroke prevention. Cerebrovasc Dis 17 2 diabetes mellitus. Metabolism 51:244 – adversely affects stroke outcome: a
(Suppl. 3):6 –10, 2004 247, 2002 magnetic resonance imaging and spec-
121. Lalouschek W, Lang W, Mullner M: 132. Chlumsky J, Charvat J, Empana JP, troscopy study. Ann Neurol 52:20 –28,
Current strategies of secondary preven- Zureik M, Gariepy J, Courbon D, Dart- 2002
tion after a cerebrovascular event: the Vi- igues JF, Ritchie K, Tzourio C, Alpero- 143. Szczudlik A, Slowik A, Turaj W, Wyr-
enna stroke registry. Stroke 32:2860 – vitch A, Ducimetiere P, Brohall G, Oden wicz-Petkow U, Pera J, Dziedzic T,
2866, 2001 A, Fagerberg B, Tsivgoulis G, Vemmos Trabka-Janik E, Iskra T: Transient hy-
122. A comparison of two doses of aspirin (30 KN, Spengos K, Papamichael CM, Cim- perglycemia in ischemic stroke pa-
mg vs. 283 mg a day) in patients after a boneriu A, Zis V, Zakopoulos N, Ma- tients. J Neurol Sci 189:105–111, 2001
transient ischemic attack or minor isch- vrikakis M, Matsumoto K, Sera Y, 144. Topic E, Pavlicek I, Brinar V, Korsic M:
emic stroke: the Dutch TIA Trial Study Nakamura H, Ueki Y, Miyake S: Echo- Glycosylated haemoglobin in clarifica-
Group. N Engl J Med 325:1261–1266, cardiography and carotid sonography in tion of the origin of hyperglycaemia in
1991 diabetic patients after cerebrovascular acute cerebrovascular accident. Diabet
123. Swedish Aspirin Low-Dose Trial (SALT) attacks. J Int Med Res 34:689 – 694, 2006 Med 6:12–15, 1989
of 75 mg aspirin as secondary prophy- 133. Weir CJ, Murray GD, Dyker AG, Lees 145. de Falco FA, Sepe Visconti O, Fucci G,
laxis after cerebrovascular ischaemic KR: Is hyperglycaemia an independent Caruso G: Correlation between hyper-
events: the SALT Collaborative Group. predictor of poor outcome after acute glycemia and cerebral infarct size in pa-
Lancet 338:1345–1349, 1991 stroke? Results of a long-term follow up tients with stroke: a clinical and X-ray
124. A randomized trial of aspirin and sulfin- study. BMJ 314:1303–1306, 1997 computed tomography study in 104 pa-
pyrazone in threatened stroke: the Cana- 134. Pulsinelli WA, Levy DE, Sigsbee B, tients. Schweiz Arch Neurol Psychiatr 144:
dian Cooperative Study Group. N Engl Scherer P, Plum F: Increased damage af- 233–239, 1993
J Med 299:53–59, 1978 ter ischemic stroke in patients with hy- 146. Kushner M, Nencini P, Reivich M, Rango
125. Diener HC, Cunha L, Forbes C, Sivenius perglycemia with or without established M, Jamieson D, Fazekas F, Zimmerman
J, Smets P, Lowenthal A: European diabetes mellitus. Am J Med 74:540 – R, Chawluk J, Alavi A, Alves W: Relation
Stroke Prevention Study. 2. Dipyridam- 544, 1983 of hyperglycemia early in ischemic brain
ole and acetylsalicylic acid in the sec- 135. Toni D, Sacchetti ML, Argentino C, Gen- infarction to cerebral anatomy, metabo-
ondary prevention of stroke. J Neurol Sci tile M, Cavalletti C, Frontoni M, Fieschi lism, and clinical outcome. Ann Neurol
143:1–13, 1996 C: Does hyperglycaemia play a role on 28:129 –135, 1990
126. A randomised, blinded, trial of clopi- the outcome of acute ischaemic stroke 147. Baird TA, Parsons MW, Phanh T,
dogrel versus aspirin in patients at risk of patients? J Neurol 239:382–386, 1992 Butcher KS, Desmond PM, Tress BM,
ischaemic events (CAPRIE): CAPRIE 136. Moulin T, Tatu L, Crepin-Leblond T, Colman PG, Chambers BR, Davis SM:
Steering Committee. Lancet 348:1329 – Chavot D, Berges S, Rumbach T: The Be- Persistent poststroke hyperglycemia is
1339, 1996 sancon Stroke Registry: an acute stroke independently associated with infarct
127. Bhatt DL, Fox KA, Hacke W, Berger PB, registry of 2,500 consecutive patients. expansion and worse clinical outcome.
Black HR, Boden WE, Cacoub P, Cohen Eur Neurol 38:10 –20, 1997 Stroke 34:2208 –2214, 2003
EA, Creager MA, Easton JD, Flather MD, 137. Williams LS, Rotich J, Qi R, Fineberg N, 148. Ribo M, Molina C, Montaner J, Rubiera
Haffner SM, Hamm CW, Hankey GJ, Espay A, Bruno A, Fineberg SE, Tierney M, Delgado-Mederos R, Arenillas JF,
Johnston SC, Mak KH, Mas JL, Montale- WR: Effects of admission hyperglycemia Quintana M, Alvarez-Sabin J: Acute hy-
scot G, Pearson TA, Steg PG, Steinhubl on mortality and costs in acute ischemic perglycemia state is associated with
SR, Weber MA, Brennan DM, Fabry-Rib- stroke. Neurology 59:67–71, 2002 lower tPA-induced recanalization rates
audo L, Booth J, Topol EJ: Clopidogrel 138. Kronmal RA, Barzilay JI, Smith NL, Psaty in stroke patients. Stroke 36:1705–1709,
and aspirin versus aspirin alone for the BM, Kuller LH, Burke GL, Furberg C: 2005
prevention of atherothrombotic events. Mortality in pharmacologically treated 149. Gentile NT, Seftchick MW, Huynh T,
N Engl J Med 354:1706 –1717, 2006 older adults with diabetes: the Cardio- Kruus LK, Gaughan J: Decreased mortal-
128. Diener HC, Bogousslavsky J, Brass LM, vascular Health Study, 1989 –2001. ity by normalizing blood glucose after
Cimminiello C, Csiba L, Kaste M, Leys PLoS Med 3:e400, 2006 acute ischemic stroke. Acad Emerg Med
D, Matias-Guiu J, Rupprecht HJ: Aspirin 139. Bruno A, Levine SR, Frankel MR, Brott 13:174 –180, 2006
and clopidogrel compared with clopi- TG, Lin Y, Tilley BC, Lyden PD, Brod- 150. Hillen T, Coshall C, Tilling K, Rudd AG,

DIABETES CARE, VOLUME 30, NUMBER 12, DECEMBER 2007 3139

Diabetes and ischemic stroke

McGovern R, Wolfe CD: Cause of stroke 1998 246, 2004

recurrence is multifactorial: patterns, 156. Alter M, Sobel E, McCoy RL, Francis ME, 162. Berthet K, Neal BC, Chalmers JP, Mac-
risk factors, and outcomes of stroke re- Davanipour Z, Shofer F, Levitt LP, Mee- Mahon SW, Bousser MG, Colman SA,
currence in the South London Stroke han EF: Stroke in Lehigh Valley: risk fac- Woodward M: Reductions in the risks of
Register. Stroke 34:1457–1463, 2003 tors for recurrent stroke. Neurology 37: recurrent stroke in patients with and
151. Johnston SC, Sidney S, Bernstein AL, 503–507, 1987 without diabetes: the PROGRESS Trial.
Gress DR: A comparison of risk factors 157. Hier DB, Foulkes MA, Swiontoniowski Blood Press 13:7–13, 2004
for recurrent TIA and stroke in patients M, Sacco RL, Gorelick PB, Mohr JP, Price 163. Alter M, Lai SM, Friday G, Singh V, Ku-
diagnosed with TIA. Neurology 60:280 – TR, Wolf PA: Stroke recurrence within 2 mar VM, Sobel E: Stroke recurrence in
285, 2003 years after ischemic infarction. Stroke diabetics: does control of blood glucose
152. Staaf G, Lindgren A, Norrving B: Pure 22:155–161, 1991 reduce risk? Stroke 28:1153–1157, 1997
motor stroke from presumed lacunar in- 158. Sacco RL, Foulkes MA, Mohr JP, Wolf 164. Jorgensen HS, Nakayama H, Reith J,
farct: long-term prognosis for survival PA, Hier DB, Price TR: Determinants of Raaschou HO, Olsen TS: Stroke recur-
and risk of recurrent stroke. Stroke 32: early recurrence of cerebral infarction: rence: predictors, severity, and prog-
2592–2596, 2001 the Stroke Data Bank. Stroke 20:983– nosis: the Copenhagen Stroke Study.
153. Eriksson SE, Olsson JE: Survival and re- 989, 1989 Neurology 48:891– 895, 1997
current strokes in patients with different 159. Sacco RL, Shi T, Zamanillo MC, Karg- 165. American Diabetes Association/Ameri-
subtypes of stroke: a fourteen-year fol- man DE: Predictors of mortality and re- can College of Cardiology: The Diabetes-
low-up study. Cerebrovasc Dis 12:171– currence after hospitalized cerebral Heart Disease Link: Surveying Attitudes,
180, 2001 infarction in an urban community: the Knowledge and Risk. Report of commis-
154. Hankey GJ, Jamrozik K, Broadhurst RJ, Northern Manhattan Stroke Study. Neu- sioned RoperASW Survey, conducted
Forbes S, Burvill PW, Anderson CS, rology 44:626 – 634, 1994 Aug-Oct 2001. Released Feb 19, 2002.
Stewart-Wynne EG: Long-term risk of 160. Soda T, Nakayasu H, Maeda M, Kusumi The Make the Link! Initiative. www.
first recurrent stroke in the Perth Com- M, Kowa H, Awaki E, Saito J, Nakashima diabetes.org/makethelink. Accessed 10
munity Stroke Study. Stroke 29:2491– K: Stroke recurrence within the first year April 2007
2500, 1998 following cerebral infarction: Tottori 166. Bhatt DL, Steg PG, Ohman EM, Hirsch
155. Petty GW, Brown RD Jr, Whisnant JP, University Lacunar Infarction Prognosis AT, Ikeda Y, Mas JL, Goto S, Liau CS,
Sicks JD, O’Fallon WM, Wiebers DO: Study (TULIPS). Acta Neurol Scand 110: Richard AJ, Rother J, Wilson PW: Inter-
Survival and recurrence after first cere- 343–349, 2004 national prevalence, recognition, and
bral infarction: a population-based 161. Lee AH, Somerford PJ, Yau KK: Risk fac- treatment of cardiovascular risk factors
study in Rochester, Minnesota, 1975 tors for ischaemic stroke recurrence after in outpatients with atherothrombosis.
through 1989. Neurology 50:208 –216, hospitalisation. Med J Aust 181:244 – JAMA 295:180 –189, 2006

3140 DIABETES CARE, VOLUME 30, NUMBER 12, DECEMBER 2007