Hyperglycemia in Acute Stroke

Elevated blood glucose is common in the early phase of stroke. The prevalence of
hyperglycemia, defined as blood glucose level >6.0 mmol/L (108 mg/dL), has been
observed in two thirds of all ischemic stroke subtypes on admission and in at least
50% in each subtype including lacunar strokes.1 Extensive experimental evidence in
stroke models supports that hyperglycemia has adverse effects on tissue outcome, and
an association between blood glucose and functional outcome has been found in an
increasing number of clinical studies. Although no interventional stroke studies have
addressed the acute reversal of hyperglycemia, active lowering of elevated blood
glucose by rapidly acting insulin is recommended in most published guidelines, even
in nondiabetic patients (European Stroke Initiative [EUSI] guidelines >10 mmol/L,
American Stroke Association [ASA] guidelines >300 mg/dL).2

Causes of Acute Hyperglycemia

Although up to one third of acute stroke patients have either diagnosed or newly
diagnosed diabetes, probably a major proportion of patients have stress hyperglycemia
mediated partly by the release of cortisol and norepinephrine. It is also a manifestation
of relative insulin deficiency, which is associated with increased lipolysis. Even in
nondiabetic patients, stress hyperglycemia may be a marker of deficient glucose
regulation in individuals with insulin resistance and developing diabetes mellitus.

How Elevated Glucose Injures the Ischemic Brain

By provoking anaerobic metabolism, lactic acidosis, and free radical production,
hyperglycemia may exert direct membrane lipid peroxidation and cell lysis in
metabolically challenged tissue. Moderately and severely increased blood glucose has
been found to further the metabolic state and mitochondrial function in the area of
ischemic penumbra.3 Insulin resistance is a known risk factor for the onset of stroke
acting through a number of intermediate vascular disease risk factors (ie,
thrombophilia, endothelial dysfunction, and inflammation).4 The evolution of an acute
infarction may be expedited by the very same vascular factors, explaining why
ischemia time seems to fly faster with patients with diabetes or grave hyperglycemia.
Relative insulin deficiency liberates circulating free fatty acids, which, together with
hyperglycemia, reportedly diminishes vascular reactivity.5,6 Furthermore, lowering
glucose with insulin has been reported to reduce ischemic brain damage in an animal
model.7
The evolution of an infarction is accompanied by glutamate release mediating
repeated waves of spreading depression (SD), another mechanism believed to
propagate the necrosis of the penumbral tissue. Although hyperglycemia alone did not
trigger early-response genes in the cortical tissue of rats, in conjunction with induced
SD, the expression of c-fos and cyclooxygenase-2 were substantially increased.8 This
suggests that elevated glucose may trigger untoward intracellular biochemical
cascades also by altering early gene expression in metabolically challenged neurons.

The blood-brain barrier is well known to be vulnerable to hyperglycemia, presumably

through the liberation of lactic acid and free radicals. The recent experimental study
by Song et al in a rat model of collagenase-induced intracerebral hemorrhage (ICH)
adds that hyperglycemia aggravates edema formation in a zone surrounding cerebral
hemorrhages.9 The study also documented increased cell death measured by the
TUNEL staining. It is conceivable that hemorrhages are surrounded by a zone of
similarly challenged tissue as infarctions are, where the availability of glucose
influences the metabolic state.

Clinical Correlation of Hyperglycemia and Infarct

Progression
Although experimental studies have clarified several mechanisms by which
hyperglycemia influences the destiny of ischemic brain tissue, studies bridging the
gap between clinical stroke and experimental models have been scarce. Recent
advances in MRI techniques have permitted correlation of loss of penumbral tissue
with elevated blood glucose, which was linked to increased brain lactate production.10
Using a subcutaneous glucose sensor for continuous monitoring up to 72 hours, the
same group could reproduce the finding that the infarcts expanded more in
hyperglycemic patients, and that hyperglycemia was independently associated with
the infarct volume change.11 This suggests that elevated glucose not only reflects the
initial volume of infarcted tissue in the acute stage but is one of the true determinants
of early infarct progression in man.
Previous SectionNext Section

Prognosis and Hyperglycemia

Already ample literature has demonstrated that hyperglycemia on admission is
associated with worsened clinical outcome, as reviewed in a systematic overview of
33 studies.12 Glycemic control may be indicated also in nondiabetic patients, in which
stress hyperglycemia was associated with a 3-fold risk of fatal 30-day outcome and
1.4-fold risk of poor functional outcome, as compared with normoglycemic patients.
Good glycemic control seems warranted also in hemorrhagic stroke,9 although more
clinical information is needed in this area. At least 2 clinical trials have recently been
initiated to examine the efficacy of early insulin therapy in acute stroke.11,13 Still, there
is no evidence to prove that reversal of hyperglycemia improves the prognosis, as it
has been demonstrated to do in acute myocardial infarction and in critically ill
postsurgical patients.14,15

Hyperglycemia and Thrombolytic Therapy of Acute

Ischemic Stroke
In several thrombolysis trials, hyperglycemia has been found to be associated with
hemorrhagic events16 and was reconfirmed recently17 as well as in a re-analysis of the
NINDS rt-PA trial.18 In the latter study, an increase of admission glucose level was
independently associated with decreased odds for neurologic improvement (odds ratio

[OR]=0.76 per 100-mg/dL increase in admission glucose) and the OR for

symptomatic ICH was 1.75 per 100-mg/dL increase in admission glucose (95% CI
1.11 to 2.78, P=0.02). The relationship was weaker after excluding patients with ICH,
suggesting that admission hyperglycemia may exert its hazards in part through
hemorrhagic events. However, another recent study by Alvarez-Sabin et al found
admission glucose >140 mg/dL (OR 8.4, CI 1.8 to 40.0) to be the sole independent
predictor of poor functional outcome at 3 months in patients with recanalization
within 6 hours, even after excluding the patients with symptomatic ICH.19 The same
was not true for the patients who did not recanalize, which leads to speculation that
hyperglycemia might partially preclude the beneficial effect of rtPA and early
reperfusion.
Previous SectionNext Section

Conclusions
This recent evidence supports that acutely elevated, predominantly stress-related
hyperglycemia is associated with poor outcomes such as dependent state or
intracerebral hemorrhage. Through several different biochemical mechanisms,
elevated glucose in the setting of cerebrovascular insults probably accelerates the
course of ischemic injury, also in the boundary regions with milder perfusion deficit.
Although admission hyperglycemia has been clearly demonstrated to be a risk factor
for symptomatic hemorrhage and worsened outcome after thrombolytic therapy, there
is perhaps not enough evidence to withhold thrombolysis from hyperglycemic patients
within the 3-hour time window. However, restoration of normoglycemia as soon as
possible should be encouraged, although conclusive evidence of decreased risk with
this approach is lacking. Especially the nondiabetic patients may be at risk of further
brain damage if hyperglycemia prevails. The recent evidence summarized above and
in the Table urges corroboration in randomized controlled trials of the efficacy of
immediate glycemic control, and determination of where the level of target glucose
concentrations of the relatively different current target values in the published
guidelines (EUSI: <10 mmol/L, ASA: <300 mg/dL=16.63 mmol/L)2 should be set. In
the interim, we should fare well with adhering to good general stroke management,
including control of blood glucose, normalization of body temperature, fluid balance
and hemodynamics, or we may otherwise risk the favorable outcome even in the
patient with early recanalization.

Preexisting hyperglycemia worsens the clinical outcome of acute stroke. Nondiabetic

ischemic stroke patients with hyperglycemia have a 3-fold higher 30-day mortality
rate than do patients without hyperglycemia. In diabetic patients with ischemic stroke,
the 30-day mortality rate is 2-fold higher.[1]
With regard to hypoglycemia, the condition can mimic acute stroke or symptoms of
transient ischemic attack (TIA).[2, 3, 4]

Essential update: ACP guidelines raise glucose target

for ICU patients with hyperglycemia
In their new guidelines for the treatment of hyperglycemia in the intensive care unit,
the American College of Physicians (ACP) revised the target blood glucose level to
140-200 mg/dL from the previous goal of 80-110 mg/dL. The change is based on
evidence that in ICU patients, not only does treatment using intensive insulin therapy
and a normoglycemic target appear to lack a short-term mortality benefit, but a target
below 140 mg/dL can actually be harmful.[5, 6]

Signs and symptoms

Hyperglycemia in stroke

Patients may come to the attention of clinicians because of

preexisting diabetes mellitus
Diabetes may also be seen with other risk factors for stroke, such as
hypertension and hypercholesterolemia

High glycemic levels may also be seen in the setting of an acute

stroke without a history of diabetes, presumably due to a
sympathetic response to the infarct

Retinopathy, neuropathy, and peripheral vascular disease may be

found in patients with long-standing diabetes

Hypoglycemia in strokelike occurrences

In the literature, signs of an acute stroke, such as hemiplegia, aphasia, and cortical
blindness, have been reported with hypoglycemia.
In individuals presenting with low glycemic levels and strokelike symptoms, diabetes
mellitus may have been previously diagnosed, and recent changes in the doses of
hypoglycemic agents and insulin may have been instituted. In particular, aggressively
tight glucose control, either patient driven or clinician directed, may give rise to
chronic or recurrent episodes of hypoglycemia. However, if factitious hypoglycemia
is suspected, such behavior may have manifested earlier as similar episodes or other
factitious behaviors.
Symptoms caused by hypoglycemia can occur suddenly and fluctuate, suggesting a
vascular etiology.

Diagnosis
Laboratory studies
In the setting of acute stroke, obtaining the following is routine practice:

Serum glucose levels

Complete blood count (CBC)

Electrolyte values

Prothrombin time (PT)

Activated partial thromboplastin time (aPTT)

Imaging studies
Because hyperglycemia may accelerate the ischemic process in stroke, it is possible
that characteristic features of acute stroke will appear on computed tomography (CT)
or magnetic resonance imaging (MRI) scans sooner than they would in patients
without hyperglycemia.
If strokelike symptoms are a result of hypoglycemia, a CT scan of the head may
initially be normal. Later, in patients with severe hypoglycemia that is prolonged and
complicated by anoxic brain injury and coma, CT scanning of the brain may show
cortical atrophy (reflecting laminar necrosis).[7] If the hypoglycemia is transitory and
the clinical status of the patient returns to normal, follow-up CT-scan findings may
again be normal.

Management
Hyperglycemia
In terms of primary prevention, treatment of diabetes appears to reduce the incidence
of atherosclerotic complications.
Intensive approaches to multiple risk factors in stroke have been suggested, including
the following:

Reduction of low-density lipoprotein (LDL) - To below 100 mg/dL in

diabetic patients
Increase of high-density lipoprotein (HDL) - With fibrates if tolerated,
an effect that is especially beneficial in patients with insulin
resistance [8]

Tight glucose control

Hypertensive management

Patients with acute stroke and hyperglycemia are often kept NPO (nothing by mouth),
because of the complicating effects of feeding on the blood glucose level.

Typically, hyperglycemia in the setting of acute stroke is treated with subcutaneous

insulin on a sliding scale. Refractory hyperglycemia may require the use of
intravenous (IV) insulin; however, IV insulin increases the risk of hypoglycemia. The
safety and efficacy of IV insulin in the treatment of hyperglycemia in patients with
acute stroke are being determined by ongoing/planned clinical trials.
Bellolio et al analyzed the results of 7 trials involving 1296 participants (639 in the
intervention group and 657 in the control group) and concluded that the
administration of intravenous insulin to maintain serum glucose levels in the first
hours after an acute ischemic stroke did not provide any benefit in terms of function,
death, or improvement in final outcome.[9]
Ntaios et al designed an intravenous insulin protocol that controls acute poststroke
hyperglycemia but frequently leads to hypokalemia. Further study is therefore
required.[10]
Transition from acute therapy to the initiation of chronic therapy in hyperglycemia
depends on the conditions persistence or whether evidence of diabetes exists.
Hypoglycemia
When hypoglycemia is discovered, the glucose level must be brought expeditiously to
a normal level. IV fluids, such as dextrose 25% in water (D25W) or dextrose 50% in
water (D50W),[11] may be necessary. Treatment of hypoglycemia beyond the initial
therapy depends on the conditions underlying cause.
Neurologists typically do not treat patients with glucose-containing fluids without
coadministration of thiamine in order to avoid the possibility of precipitating acute
Wernicke encephalopathy or chronic Korsakoff psychosis.