**Hypoglycemia in Critically Ill Patients**

Cornelia Hoedemaekers and Johannes van der Hoeven

*Radboud University Nijmegen Medical Centre, Department of ICU Nijmegen, The Netherlands* 

### **1. Introduction**

76 Diabetes – Damages and Treatments

Virally, M, Laloi-Michelin, M, Médeau, V, Meas, T, Kévorkian, JP, Mouly, S, Guillausseau,

Vogel, A, Strassburg, CP, Obermayer-Straub, P, Brabant, G, Manns, MP. (2002). The genetic

Voulgari, C, Tentolouris, N. (2010). The performance of a glucose-ketone meter in the

room. *Diabetes technology & therapeutics*, Vol.12, No.7, July 2010, pp.529-535 Voulgari, C, Psallas, M, Kokkinos, A, Argiana, V, Katsilambros, N, Tentolouris, N. (2010).

Weinzimer, SA, Tamborlane, WV, Chase, HP, Garg, SK. (2004). Continuous glucose

Widom B, Simonson DC. (1992). Intermittent hypoglycemia impairs glucose counterregulation. *Diabetes*, Vol.41, No.12, (December 1992), pp.1597-1602 Witkowski, P, Zakai, SB, Rana, A, Sledzinski, Z, Hardy, MA. (2006). Pancreatic islet

Woodyatt, RT (1942). The American Diabetes Association. *Science*, Vol.7, No.96 (2484),

Woolley, SL, Smith, DR. (2006). Acute compartment syndrome secondary to diabetic muscle

Wroe, J. (2004). The second international DAWN summit: a call to action to improve

*transplantation,* Vol.11, No.2, pp.5-13, discussion 32-43

*Diabetes & Metabolism*, Vol.33, No.6, (December 2007), pp.466-468

Vol.80, No.4, (April 2002), pp.201-211

*complications* (August 2010)

(August 1942), pp.138

Vol. 13, No.2, (April 2006), pp.113-116

pp.95-100

pp.201 -208

PJ. (2007). Muscle infarction in a young woman with brittle type 1 diabetes.

background of autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy and its autoimmune disease components. *Journal of molecular medicine*,

diagnosis of diabetic ketoacidosis in patients with type 2 diabetes in the emergency

The association between cardiac autonomic neuropathy with metabolic and other factors in subjects with type 1 and type 2 diabetes. *Journal of diabetes and its* 

monitoring in type 1 diabetes. *Current diabetes reports*, Vol.4, No.2, (April 2004),

transplantation, what has been achieved since Edmonton break-through. *Annals of* 

infarction: case report and literature review. *European journal of emergency medicine*,

psychosocial care for people with diabetes. *Practical Diabetes International*, Vol.21,

Dysglycemia is common in critically ill patients. Both hyperglycemia and hypoglycemia are independent risk factors for increased morbidity and mortality. Hypoglycemia severity may even have a 'dose-response' relationship with increased mortality(Bagshaw et al., 2009). Acute hypoglycemia induces a systemic, counter-regulatory stress response that leads to an increase in blood norepinephrine, epinephrine, glucagon, growth hormone, and cortisol concentrations. Risk factors that are associated with the occurrence of hypoglycemia in ICU patients include severity of illness, strict glucose control, continuous veno-venous hemodialysis, decrease of nutrition without adjustment for insulin infusion, a prior diagnosis of diabetes mellitus, sepsis, and need for inotropic support (Arabi et al, 2009;Krinsley & Grover, 2007;Vriesendorp et al. ,2006).

The association between hyperglycemia and mortality seems population dependent, with the strongest association in patients in the cardiac, cardiothoracic and neurological ICU(Whitcomb et al., 2005). During acute illness, hyperglycemia might exert an even more deleterious effect on ICU patients without diabetes than among patients with diabetes(Capes et al., 2000;Krinsley 2006;Rady et al., 2005). Unlike nondiabetic patients, diabetic patients show no clear association between hyperglycemia during intensive care unit stay and mortality and markedly lower odds ratios of death at all levels of hyperglycemia. These findings suggest that, in critically patients with diabetes mellitus, hyperglycemia may have different biological and/or clinical implications(Egi et al., 2008).

Recently, variability of glucose concentrations has been identified as an additional factor that may contribute to the mortality and morbidity of dysglycemia. A retrospective evaluation of over 7000 patients identified glycemic variability, defined as the standard deviation of each patient's mean glucose level during ICU stay, as a stronger predictor of mortality than hyperglycemia(Egi et al., 2006). High glucose variability during ICU stay was associated with increased mortality in patients without diabetes, even after adjustment for severity of illness and mean glucose concentration(Krinsley 2009). In contrast, there was no independent association of glucose variability with mortality among patients with diabetes. Glucose variability contributes to ICU mortality and in-hospital death by increasing oxidative stress, neuronal damage, mitochondrial damage, and coagulation activation.

The relation between glucose concentration and outcome is complex and not linear. The interaction between glucose concentrations and outcome may arise from independent and synergistic domains of glycemic control including central tendency (such as mean and

Hypoglycemia in Critically Ill Patients 79

gluconeogenesis and glycogenolysis in the liver, but has no effect on the kidney. Cytokines such as IL-6 and TNF-α also contribute to the synthesis of glucose by regulation of key enzymes such as glucose 6-phosphatase (G6Pase) involved in the gluconeogenesis(Blumberg et al., 1995a;Blumberg et al., 1995b). Insulin suppresses glucose release in both the liver and kidney by direct enzyme activationdeactivation, as well as by reducing the availability of gluconeogenic substrates and actions on gluconeogenic activators. Glycogenolysis and reduced glycogen synthesis in the liver probably contribute

Insulin resistance is characterized by raised plasma levels of insulin, organ-specific alterations in glucose utilization, and impaired insulin-mediated uptake(Brealey & Singer 2009). The insulin resistance occurs at several levels and is mainly driven by the inflammatory response. TNF-α can activate protein kinase B, resulting in phosphorylation of the insulin receptor thereby reducing the glucose uptake(Fan et al., 1996;Ueki et al., 2004). In addition, a number of animal models measured a significant reduction in the number of

Under normal conditions, the human brain is an obligate glucose consumer and depends almost entirely on the availability of systemic glucose to maintain its normal metabolism. Glucose concentration in brain normally shows a linear relationship to blood concentrations with normal human blood glucose levels ranging between 70-128 mg/dL and the corresponding normal brain concentrations ranging from roughly 14.4-41.4 mg/dL

Autoradiography and PET have shown that the rate of glucose consumption differs between brain regions, with higher values in grey matter, and also varies with time, with active areas capturing more glucose compared to inactive areas(Raichle & Mintun, 2006). In critically ill neurological patients the metabolic demand of the brain is increased, resulting in a relative deficiency in cerebral extracellular glucose(Bergsneider et al., 1997;Hutchinson et al., 2009). Glucose is transported from the blood across the endothelial cells in the blood-brain barrier and across the plasma membranes of neurons and glia cells. Rapid breakdown of glucose by the brain creates a concentration gradient between the cerebrospinal fluid and the blood, resulting in a driving force of glucose towards the brain. Glucose is transported to the brain by facilitated glucose transport that is mediated by members of the glucose transporter (GLUT) protein family. Several GLUT isoforms have been identified in the brain: GLUT1 is highly expressed on endothelial cells at the blood-brain-barrier and in astrocytes, whereas

During normal activity, glucose is the predominant energy substrate for neurons. The metabolic coupling between neurons and astrocytes preserves energy homeostasis in the brain during increased neuronal activity. Activation of neurons is accompanied by an increase in local cerebral blood flow, thus increasing the delivery of energy substrates. Glucose metabolism itself is also tightly coupled to neuronal activity: activation of astrocytes by glutamate, that is released by neurons, results in increased production of lactate by astrocytes that can be used as fuel by neurons to meet their energy demand (the astrocyteneuron lactate shuffle)(Pellerin et al., 2007;Pellerin & Magistretti, 2004;Rothman et al., 1999). This astrocyte-neuron lactate shuffle also contributes to neurotransmitter recycling and

only to a small extent to the stress induced hyperglycemia.

**3. Glucose in critically ill neurological patients: Friend or foe?** 

GLUT3 is detected in neurons and GLUT5 in microglia (Vannucci et al., 1997).

insulin receptors(McCowen et al., 2001).

**3.1 Glucose is fuel for the brain** 

(Gruetter et al., 1998).

median glucose values), variability and the minimum glucose value (Mackenzie et al., 2011). Using these different metrics of glycemic control, a population-specific relationship between metrics of outcome was demonstrated in patients in a surgical, trauma, cardiac and neurological ICU. This relationship had a dose response component with an n-shape curve in neurological ICU patients, suggesting a survival advantage during hyperglycemia in these patients.

Strict glycemic control may improve morbidity and mortality in critically ill patients. Intensive glucose control is however associated with a higher incidence of hypoglycemia compared to conventionally treated patients. This increased risk of hypoglycemia may limit the use of strict glucose control in critically ill neurological patients, since hypoglycemia is a well known cause of secondary brain injury. In this review we will describe the different aspects of dysglycemia and glycemic control in critically ill patients, with a special emphasis on the critically ill neurological patients.

### **2. Glucose homeostasis in critically ill patients**

Glucose homeostasis is a physiologically well-balanced mechanism depending on coordinated and simultaneously ongoing processes involving insulin secretion by the pancreas, hepatic and renal glucose output and glucose uptake by splanchnic (liver and gut) and peripheral tissues. Cellular uptake of glucose occurs via insulin- or noninsulindependent mechanisms. The noninsulin-dependent pathway is the major mechanism of glucose uptake in the basal state, accounting for 75-85% of the total post-prandial glucose uptake and is mainly directed at the brain(Gottesman et al., 1983).

The brain plays a central role in the orchestration of the changes in blood glucose and the appropriate counterregulatory responses (reviewed in (Watts & Donovan, 2010)). Some neurons possess specialized mechanisms that allow them to act as glucosensors and alter their firing rates with fluctuating ambient glucose concentrations. These neurons are predominantly located in the hypothalamus and hindbrain. Important glucosensing elements are also present in the hepatic portal/mesenteric vein, gut, carotid body and oral cavity. Information from the glucosensing elements is processed by neurons in the hypothalamus and hindbrain. These neurons regulate the counterregulatory responses and provide direct input to the appropriate neuroendocrine motor and preganglionic neurons in the hypothalamus, hindbrain, and autonomic ganglia. Their output in turn controls and/or modulates effector cells in the adrenal medulla (chromaffin cells), anterior pituitary (corticotopes and somatotropes), and pancreatic islets (α- and β-cells).

Severe illness induces a stress response with alterations in the glucose metabolism including enhanced peripheral glucose uptake and utilization, hyperlactatemia, increased glucose production, depressed glycogenesis, glucose intolerance, and insulin resistance(Mizock, 1995). Stress hyperglycemia is caused by a highly complex interplay between counterregulatory hormones such as cortisol, epinephrine, norepinephrine and glucagon, leading to an increase in hepatic and renal glucose production and insulin resistance (reviewed in (Dungan et al., 2009)). The hepatic gluconeogenesis that is induced by glucagon, cortisol and epinephrine is the key source of glucose production. The kidney is responsible for approximately 20% of the glucose production under resting conditions(Meyer et al., 2002). During stress, however, cathecholamines can increase the renal contribution to glucose production up to 40% by increasing substrate availability and the gluconeogenic efficiency of the kidney(Meyer et al., 2003). Glucagon increases both

median glucose values), variability and the minimum glucose value (Mackenzie et al., 2011). Using these different metrics of glycemic control, a population-specific relationship between metrics of outcome was demonstrated in patients in a surgical, trauma, cardiac and neurological ICU. This relationship had a dose response component with an n-shape curve in neurological ICU patients, suggesting a survival advantage during hyperglycemia in

Strict glycemic control may improve morbidity and mortality in critically ill patients. Intensive glucose control is however associated with a higher incidence of hypoglycemia compared to conventionally treated patients. This increased risk of hypoglycemia may limit the use of strict glucose control in critically ill neurological patients, since hypoglycemia is a well known cause of secondary brain injury. In this review we will describe the different aspects of dysglycemia and glycemic control in critically ill patients, with a special emphasis

Glucose homeostasis is a physiologically well-balanced mechanism depending on coordinated and simultaneously ongoing processes involving insulin secretion by the pancreas, hepatic and renal glucose output and glucose uptake by splanchnic (liver and gut) and peripheral tissues. Cellular uptake of glucose occurs via insulin- or noninsulindependent mechanisms. The noninsulin-dependent pathway is the major mechanism of glucose uptake in the basal state, accounting for 75-85% of the total post-prandial glucose

The brain plays a central role in the orchestration of the changes in blood glucose and the appropriate counterregulatory responses (reviewed in (Watts & Donovan, 2010)). Some neurons possess specialized mechanisms that allow them to act as glucosensors and alter their firing rates with fluctuating ambient glucose concentrations. These neurons are predominantly located in the hypothalamus and hindbrain. Important glucosensing elements are also present in the hepatic portal/mesenteric vein, gut, carotid body and oral cavity. Information from the glucosensing elements is processed by neurons in the hypothalamus and hindbrain. These neurons regulate the counterregulatory responses and provide direct input to the appropriate neuroendocrine motor and preganglionic neurons in the hypothalamus, hindbrain, and autonomic ganglia. Their output in turn controls and/or modulates effector cells in the adrenal medulla (chromaffin cells), anterior pituitary

Severe illness induces a stress response with alterations in the glucose metabolism including enhanced peripheral glucose uptake and utilization, hyperlactatemia, increased glucose production, depressed glycogenesis, glucose intolerance, and insulin resistance(Mizock, 1995). Stress hyperglycemia is caused by a highly complex interplay between counterregulatory hormones such as cortisol, epinephrine, norepinephrine and glucagon, leading to an increase in hepatic and renal glucose production and insulin resistance (reviewed in (Dungan et al., 2009)). The hepatic gluconeogenesis that is induced by glucagon, cortisol and epinephrine is the key source of glucose production. The kidney is responsible for approximately 20% of the glucose production under resting conditions(Meyer et al., 2002). During stress, however, cathecholamines can increase the renal contribution to glucose production up to 40% by increasing substrate availability and the gluconeogenic efficiency of the kidney(Meyer et al., 2003). Glucagon increases both

these patients.

on the critically ill neurological patients.

**2. Glucose homeostasis in critically ill patients** 

uptake and is mainly directed at the brain(Gottesman et al., 1983).

(corticotopes and somatotropes), and pancreatic islets (α- and β-cells).

gluconeogenesis and glycogenolysis in the liver, but has no effect on the kidney. Cytokines such as IL-6 and TNF-α also contribute to the synthesis of glucose by regulation of key enzymes such as glucose 6-phosphatase (G6Pase) involved in the gluconeogenesis(Blumberg et al., 1995a;Blumberg et al., 1995b). Insulin suppresses glucose release in both the liver and kidney by direct enzyme activationdeactivation, as well as by reducing the availability of gluconeogenic substrates and actions on gluconeogenic activators. Glycogenolysis and reduced glycogen synthesis in the liver probably contribute only to a small extent to the stress induced hyperglycemia.

Insulin resistance is characterized by raised plasma levels of insulin, organ-specific alterations in glucose utilization, and impaired insulin-mediated uptake(Brealey & Singer 2009). The insulin resistance occurs at several levels and is mainly driven by the inflammatory response. TNF-α can activate protein kinase B, resulting in phosphorylation of the insulin receptor thereby reducing the glucose uptake(Fan et al., 1996;Ueki et al., 2004). In addition, a number of animal models measured a significant reduction in the number of insulin receptors(McCowen et al., 2001).

### **3. Glucose in critically ill neurological patients: Friend or foe?**

### **3.1 Glucose is fuel for the brain**

Under normal conditions, the human brain is an obligate glucose consumer and depends almost entirely on the availability of systemic glucose to maintain its normal metabolism. Glucose concentration in brain normally shows a linear relationship to blood concentrations with normal human blood glucose levels ranging between 70-128 mg/dL and the corresponding normal brain concentrations ranging from roughly 14.4-41.4 mg/dL (Gruetter et al., 1998).

Autoradiography and PET have shown that the rate of glucose consumption differs between brain regions, with higher values in grey matter, and also varies with time, with active areas capturing more glucose compared to inactive areas(Raichle & Mintun, 2006). In critically ill neurological patients the metabolic demand of the brain is increased, resulting in a relative deficiency in cerebral extracellular glucose(Bergsneider et al., 1997;Hutchinson et al., 2009).

Glucose is transported from the blood across the endothelial cells in the blood-brain barrier and across the plasma membranes of neurons and glia cells. Rapid breakdown of glucose by the brain creates a concentration gradient between the cerebrospinal fluid and the blood, resulting in a driving force of glucose towards the brain. Glucose is transported to the brain by facilitated glucose transport that is mediated by members of the glucose transporter (GLUT) protein family. Several GLUT isoforms have been identified in the brain: GLUT1 is highly expressed on endothelial cells at the blood-brain-barrier and in astrocytes, whereas GLUT3 is detected in neurons and GLUT5 in microglia (Vannucci et al., 1997).

During normal activity, glucose is the predominant energy substrate for neurons. The metabolic coupling between neurons and astrocytes preserves energy homeostasis in the brain during increased neuronal activity. Activation of neurons is accompanied by an increase in local cerebral blood flow, thus increasing the delivery of energy substrates. Glucose metabolism itself is also tightly coupled to neuronal activity: activation of astrocytes by glutamate, that is released by neurons, results in increased production of lactate by astrocytes that can be used as fuel by neurons to meet their energy demand (the astrocyteneuron lactate shuffle)(Pellerin et al., 2007;Pellerin & Magistretti, 2004;Rothman et al., 1999). This astrocyte-neuron lactate shuffle also contributes to neurotransmitter recycling and

Hypoglycemia in Critically Ill Patients 81

et al., 2009), intracranial hemorrhage (Fogelholm et al., 2005;Godoy et al., 2008;Godoy et al., 2009;Godoy & Di, 2007;Kimura et al., 2007;Passero et al., 2003) and subarachnoid hemorrhage (Badjatia et al., 2005;Frontera et al., 2006;Kruyt et al., 2008;Kruyt et al.,

Hyperglycemia is associated with many detrimental effects, including reduced immune function, increased inflammation and coagulation, and modulation of the endothelium. Plasma concentrations of pro-inflammatory cytokines are increased during hyperglycemia, while insulin reduces the pro-inflammatory cytokine response and restores the pro- and anti-inflammatory balance (Turina et al., 2005). Glucose increases basal TNF and IL-6 production in human monocytes *in-vitro* (Morohoshi et al., 1996). Similarly, a glucosedependent increased production of TNF by peripheral blood cells *in-vitro* after stimulation with lipopolysaccharide was measured, whereas glucose does not influence the production of the anti-inflammatory cytokine IL-10 (Hancu et al., 1998). *In-vivo* induced hypoglycaemia in hypoglycaemic human clamp models resulted in a down-modulation of lipopolysaccharide -induced TNF synthesis(de Galan et al., 2003). Hyperglycemia increases the expression of tissue factor, which has both proinflammatory and procoagulant functions(Brealey & Singer, 2009). Hyperglycemia induces endothelial dysfunction through several damaging pathways, including the polyol/sorbitol/aldose reductase pathway, the protein kinase C pathway, the accumulation of non-enzymatic glycation end products and by increased oxidative stress, ultimately leading to increased expression of endothial cytokines and adhesion molecules (van den Oever et al., 2010). Insulin infusion restores normoglycemia in critically ill patients and improves and restores host defence,

The deleterious effects of acute hyperglycemia on brain injury has been demonstrated in a large number of animal studies (reviewed in (Ergul et al., 2009)). Hyperglycemia increases infarct volume in focal models of ischemia and aggravates necrosis in global ischemia/reperfusion models. In addition, hyperglycemia contributes to the vascular damage during ischemia/reperfusion injury, resulting in increased hemorrhagic transformation during reperfusion(de Court et al., 1989;de Court et al., 1988). Marked bloodbrain barrier disruption with formation of brain edema has been found in hyperglycemic rats after temporary and permanent middle cerebral artery occlusion(Kamada et al., 2007). In diabetes adaptive protective mechanisms gradually develop, protecting the subject against acute hyperglycemia. Diabetes promotes neovascularization, remodeling and increases in vascular tone limiting cerebral perfusion(Ergul et al., 2009). Resulting hypoxia and/or metabolic changes mediate ischemic tolerance via neuronal preconditioning but decreases vascular ischemic tolerance leading to increased and accelerated hemorrhagic transformation and development of edema in the event of an ischemic event. Acute hyperglycemia also increases vascular tone and disrupts vascular integrity but in the absence of sufficient time to stimulate adaptive protective mechanisms, the magnitude of

Cerebral ischemia results in widespread activation of the systemic inflammatory system (Offner et al., 2006). Systemic inflammatory mediators such as cytokines and adhesion molecules can activate microglial cells and perivascular macrophages and contribute to irreversible brain ischemia(Bemeur et al., 2007). After the initial activation of the innate immune response, inflammatory cells from the periphery are mobilized and contribute to microvessel obstruction, edema formation, cellular necrosis and tissue infarction.

2009;Lanzino et al., 1993).

haemodynamics and coagulation abnormalities.

neuronal damage is greater.

restoration of neuronal membrane potentials. Recent studies indicate that the human brain has the capacity to support up to 10% of its energy metabolism with lactate. Lactate in plasma can cross the blood-brain barrier through monocarboxylate transporters(Simpson et al., 2007). Plasma lactate may become a significant source of fuel in conditions of increased plasma lactate levels or when blood glucose levels are reduced, accounting for up to 60% of the energy metabolism(Boumezbeur et al., 2010).

Glycogen metabolism is also under the control of metabolic coupling. Glycogen is predominantly localized in the peripheral astrocytic processes surrounding the neuronal elements and serves as an endogenous source of energy for these cells and for neurons during extreme energy failure. Neurochemical signals from neurons and astrocytes trigger glycogenolysis. In astrocytes, glucose derived from glycogenolysis is used for both oxidative metabolism and for production of lactate(Benarroch, 2010). This lactate serves as an energy substrate for oxidative metabolism in neurons.

Since the brain relies upon plasma glucose as its primary energy source, a reduced blood glucose concentration promptly induces a complex counter regulatory response aiming at recovery of plasma glucose concentrations. Glucagon and epinephrine are released after a small decrease in glucose concentration, followed by activation of the autonomic nervous system. This hypoglycemia induced systemic stress response is accompanied by an increasing cerebral blood flow and altered cerebral metabolism with increased glycogenolysis. Depending on the duration and severity, the effects of hypoglycemia span from mild changes in EEG signals to irreversible brain injury and coma. Not all neurons are equally sensitive to hypoglycemic injury. Neurons in the cerebral cortex and the hippocampus are affected preferentially, followed by neurons in the basal ganglia and the thalamus. Neurons in the brain stem, the cerebellum, and the spinal cord are generally spared, as are glial cells and white matter tracts (Auer et al., 1989). The hypoglycemic neuronal damage is not a direct result of energy failure but mainly caused by an excitotoxic amino acid mediated increase in intracellular calcium, production of reactive oxygen species and apoptosis (Suh et al., 2007).

In acute brain injury, the hypoglycemic threshold is lower compared to normal brain and even mild hypoglycemia can induce neuroglycopenia. In patients after traumatic brain injury, arterial glucose levels < 108 mg/dL resulted in decreased brain glucose concentrations with an increased cerebral uptake of glucose(Meierhans et al., 2010). Microdialysis markers of brain metabolic distress were significantly reduced at brain glucose concentrations > 18 mg/dL, reaching the lowest levels at arterial blood glucose levels between 108-162 mg/dL. From this study it was concluded that arterial blood glucose concentrations between 108-162 mg/dL were optimal in traumatic brain injury. In addition, low brain glucose concentrations are associated with recurrent, spontaneous, spreading depolarizations in pericontusional tissue, resulting in a further reduction in brain glucose concentration and an ongoing brain damage (Feuerstein et al., 2010;Parkin et al., 2005).

### **3.2 Glucose is toxic to the brain**

Stress-related hyperglycaemia, previously considered to be a protective physiological response to meet the increased demands of an injury, is associated with a poor outcome in critically ill patients. This association of poor outcome and hyperglycemia has been consistently confirmed across multiple studies and different disease entities such as traumatic brain injury (Jeremitsky et al., 2005;Lam et al., 1991;Rovlias & Kotsou, 2000;Salim

restoration of neuronal membrane potentials. Recent studies indicate that the human brain has the capacity to support up to 10% of its energy metabolism with lactate. Lactate in plasma can cross the blood-brain barrier through monocarboxylate transporters(Simpson et al., 2007). Plasma lactate may become a significant source of fuel in conditions of increased plasma lactate levels or when blood glucose levels are reduced, accounting for up to 60% of

Glycogen metabolism is also under the control of metabolic coupling. Glycogen is predominantly localized in the peripheral astrocytic processes surrounding the neuronal elements and serves as an endogenous source of energy for these cells and for neurons during extreme energy failure. Neurochemical signals from neurons and astrocytes trigger glycogenolysis. In astrocytes, glucose derived from glycogenolysis is used for both oxidative metabolism and for production of lactate(Benarroch, 2010). This lactate serves as an energy

Since the brain relies upon plasma glucose as its primary energy source, a reduced blood glucose concentration promptly induces a complex counter regulatory response aiming at recovery of plasma glucose concentrations. Glucagon and epinephrine are released after a small decrease in glucose concentration, followed by activation of the autonomic nervous system. This hypoglycemia induced systemic stress response is accompanied by an increasing cerebral blood flow and altered cerebral metabolism with increased glycogenolysis. Depending on the duration and severity, the effects of hypoglycemia span from mild changes in EEG signals to irreversible brain injury and coma. Not all neurons are equally sensitive to hypoglycemic injury. Neurons in the cerebral cortex and the hippocampus are affected preferentially, followed by neurons in the basal ganglia and the thalamus. Neurons in the brain stem, the cerebellum, and the spinal cord are generally spared, as are glial cells and white matter tracts (Auer et al., 1989). The hypoglycemic neuronal damage is not a direct result of energy failure but mainly caused by an excitotoxic amino acid mediated increase in intracellular calcium, production of reactive oxygen species

In acute brain injury, the hypoglycemic threshold is lower compared to normal brain and even mild hypoglycemia can induce neuroglycopenia. In patients after traumatic brain injury, arterial glucose levels < 108 mg/dL resulted in decreased brain glucose concentrations with an increased cerebral uptake of glucose(Meierhans et al., 2010). Microdialysis markers of brain metabolic distress were significantly reduced at brain glucose concentrations > 18 mg/dL, reaching the lowest levels at arterial blood glucose levels between 108-162 mg/dL. From this study it was concluded that arterial blood glucose concentrations between 108-162 mg/dL were optimal in traumatic brain injury. In addition, low brain glucose concentrations are associated with recurrent, spontaneous, spreading depolarizations in pericontusional tissue, resulting in a further reduction in brain glucose concentration and an ongoing brain damage (Feuerstein et al., 2010;Parkin et al., 2005).

Stress-related hyperglycaemia, previously considered to be a protective physiological response to meet the increased demands of an injury, is associated with a poor outcome in critically ill patients. This association of poor outcome and hyperglycemia has been consistently confirmed across multiple studies and different disease entities such as traumatic brain injury (Jeremitsky et al., 2005;Lam et al., 1991;Rovlias & Kotsou, 2000;Salim

the energy metabolism(Boumezbeur et al., 2010).

substrate for oxidative metabolism in neurons.

and apoptosis (Suh et al., 2007).

**3.2 Glucose is toxic to the brain** 

et al., 2009), intracranial hemorrhage (Fogelholm et al., 2005;Godoy et al., 2008;Godoy et al., 2009;Godoy & Di, 2007;Kimura et al., 2007;Passero et al., 2003) and subarachnoid hemorrhage (Badjatia et al., 2005;Frontera et al., 2006;Kruyt et al., 2008;Kruyt et al., 2009;Lanzino et al., 1993).

Hyperglycemia is associated with many detrimental effects, including reduced immune function, increased inflammation and coagulation, and modulation of the endothelium. Plasma concentrations of pro-inflammatory cytokines are increased during hyperglycemia, while insulin reduces the pro-inflammatory cytokine response and restores the pro- and anti-inflammatory balance (Turina et al., 2005). Glucose increases basal TNF and IL-6 production in human monocytes *in-vitro* (Morohoshi et al., 1996). Similarly, a glucosedependent increased production of TNF by peripheral blood cells *in-vitro* after stimulation with lipopolysaccharide was measured, whereas glucose does not influence the production of the anti-inflammatory cytokine IL-10 (Hancu et al., 1998). *In-vivo* induced hypoglycaemia in hypoglycaemic human clamp models resulted in a down-modulation of lipopolysaccharide -induced TNF synthesis(de Galan et al., 2003). Hyperglycemia increases the expression of tissue factor, which has both proinflammatory and procoagulant functions(Brealey & Singer, 2009). Hyperglycemia induces endothelial dysfunction through several damaging pathways, including the polyol/sorbitol/aldose reductase pathway, the protein kinase C pathway, the accumulation of non-enzymatic glycation end products and by increased oxidative stress, ultimately leading to increased expression of endothial cytokines and adhesion molecules (van den Oever et al., 2010). Insulin infusion restores normoglycemia in critically ill patients and improves and restores host defence, haemodynamics and coagulation abnormalities.

The deleterious effects of acute hyperglycemia on brain injury has been demonstrated in a large number of animal studies (reviewed in (Ergul et al., 2009)). Hyperglycemia increases infarct volume in focal models of ischemia and aggravates necrosis in global ischemia/reperfusion models. In addition, hyperglycemia contributes to the vascular damage during ischemia/reperfusion injury, resulting in increased hemorrhagic transformation during reperfusion(de Court et al., 1989;de Court et al., 1988). Marked bloodbrain barrier disruption with formation of brain edema has been found in hyperglycemic rats after temporary and permanent middle cerebral artery occlusion(Kamada et al., 2007). In diabetes adaptive protective mechanisms gradually develop, protecting the subject against acute hyperglycemia. Diabetes promotes neovascularization, remodeling and increases in vascular tone limiting cerebral perfusion(Ergul et al., 2009). Resulting hypoxia and/or metabolic changes mediate ischemic tolerance via neuronal preconditioning but decreases vascular ischemic tolerance leading to increased and accelerated hemorrhagic transformation and development of edema in the event of an ischemic event. Acute hyperglycemia also increases vascular tone and disrupts vascular integrity but in the absence of sufficient time to stimulate adaptive protective mechanisms, the magnitude of neuronal damage is greater.

Cerebral ischemia results in widespread activation of the systemic inflammatory system (Offner et al., 2006). Systemic inflammatory mediators such as cytokines and adhesion molecules can activate microglial cells and perivascular macrophages and contribute to irreversible brain ischemia(Bemeur et al., 2007). After the initial activation of the innate immune response, inflammatory cells from the periphery are mobilized and contribute to microvessel obstruction, edema formation, cellular necrosis and tissue infarction.

Hypoglycemia in Critically Ill Patients 83

After these landmark studies by Van den Berghe in 2001 and 2006, tight glycemic control was adopted as standard care in a large number of ICUs. However, subsequent randomized controlled multicentre trials were unable to replicate the results of these landmark trials. The Glucontrol study compared strict glucose control (blood glucose concentrations 80-110 mg/dL) to a target glucose between 140-180mg/dL, in an attempt to prevent the adverse effects of severe hyperglycemia, while reducing the risks of hypoglycemia(Preiser et al., 2009). This multicentre study was stopped early due to the high rate of protocol violations. Strict glucose control failed to induce any clinical benefit, but was associated with a higher incidence of hypoglycemia. The VISEP trial was a multicenter, two-by-two factorial trial, that randomly assigned patients with severe sepsis to receive either intensive insulin therapy to maintain euglycemia or conventional insulin therapy and either a low-molecularweight hydroxyethyl starch or modified Ringer's lactate for fluid resuscitation(Brunkhorst et al., 2008). After the first safety analysis, involving 488 patients intensive insulin therapy was terminated early by the data and safety monitoring board, owing to an increased number of hypoglycemic events (12.1%), as compared with conventional insulin therapy (2.1%). No differences in mortality and morbidity between the intensive and conventional treatments groups were found, but patients in the intensive-therapy group tended to have longer stays in the ICU than did patients in the conventional therapy group. The NICE-SUGAR trial was a large, international, randomized trial comparing intensive glucose control, with a target blood glucose range of 81 to 108 mg/dl to conventional glucose control, with a target < 180 mg/dl(Finfer et al. 2009). Intensive glucose control increased the absolute risk of death at 90 days by 2.6 percentage points; this represents a number needed to harm of 38. The difference in mortality remained significant after adjustment for potential confounders. Severe hypoglycemia was significantly more common with intensive glucose control. Given this lack of reproducible results in a heterogeneous group of ICU patients, and concerns over excessive hypoglycemia, extremely tight glucose control cannot be

The multicentre trials were unable to replicate the findings of the 2 Leuven trials and raised the possibility that intensive insulin therapy may even increase the risk of mortality and morbidity in ICU patients. The explanation for the disparate findings seems multifactorial. In the Leuven studies the rate of use of total parenteral nutrition was higher compared to the other studies. Intensive insulin therapy may increase mortality in patients receiving enteral nutrition, possibly related to the adverse effects associated with hypoglycemia(Marik & Preiser, 2010). In turn, high dose parenteral glucose administration in the absence of intensive insulin therapy results in hyperglycemia with associated organ failure and death. In the Leuven studies, adjustments of insulin dosage were exclusively based on blood glucose measured on arterial blood via a point-of-care blood gas/glucose analyzer, whereas other studies used samples obtained from different sites and measured with different devices. The conventional treatment differed among the studies(Gunst & Van den Berghe, 2010). In the Leuven studies higher glucose values were tolerated compared to the other trials and the beneficial effects of intensive insulin therapy in these studies may be obtained

Hyperglycemia at the time of brain injury is associated with increased morbidity and mortality. A planned subgroup analysis in patients with isolated brain injury of the first

considered standard of care in ICU patients.

by preventing excessive hyperglycemia.

**4.2 Intensive insulin therapy in brain injury** 

Hyperglycemia enhances neutrophil infiltration and increases cytokine expression in several animal models of cerebral ischemia and likely exacerbates the ischemic injury(Bemeur et al., 2005).

### **4. Glucose control in critically ill patients**

### **4.1 Intensive insulin therapy**

A number of randomized controlled trials have been performed on the effects of strict glucose control in critically ill patients. Two single centre trials were performed in Leuven by van den Berghe et al(Van den Berghe et al., 2001;Van den Berghe et al., 2006a), followed by a number of multicentre trials. The first Leuven trial compared maintenance of blood glucose levels between 80 and 110 mg/dl versus 180 and 200 mg/dl in critically ill patients in a surgical ICU(Van den Berghe et al., 2001). In this trial strict glucose control resulted in a 42% reduction in mortality compared with conventional treatment. Septic patients and patients with an ICU stay > 5 days showed the largest reduction in mortality. In the second trial from Leuven performed in medical ICU patients no mortality benefit was demonstrated in the overall intention to treat analysis(Van den Berghe et al., 2006a). A reduction in hospital mortality from 52.5 to 43.0 percent was found in patients treated with intensive insulin therapy who stayed in the ICU for 3 days or more. Among patients treated < 3 days in the ICU, mortality was higher in the insulin group compared to the conventional group. The incidence of hypoglycemia was higher in the intervention group compared to the conventional treatment in both trials. Pooling the two datasets of both Leuven trials revealed that intensive insulin therapy reduced morbidity and mortality in a mixed surgical/medical ICU population, especially when continued for at least 3 days, without causing harm to patients treated for < 3 days(Van den Berghe et al., 2006b). The subgroup of patients with a prior history of diabetes did not appear to benefit. Blood glucose maintained at < 110 mg/dl was more effective that at 110-150 mg/dl, but also carried the highest risk of hypoglycemia.

How strict control of blood glucose reduces morbidity and mortality is unknown, but the mechanism may be related either to a direct effect of normalization of hyperglycemia or to the concomitantly higher insulin levels. *Post hoc* multivariate logistic regression analysis of the study by van den Berghe *et al*. suggests that the lowered blood glucose level rather than the insulin dose is related to the reduction in mortality (Van den Berghe et al., 2003). Apart from glucose lowering, insulin has a number of nonglycemic metabolic effects that may be important in critical illness (reviewed in (Honiden & Inzucchi, 2010)). Insulin can modulate inflammation via the mannose binding lectin pathway, via NF-kB and via modulation of pro- and anti-inflammatory cytokines. It can reduce free fatty acids and reverse the state of dyslipidemia in critical illness, regulate apotosis, prevent endothelial dysfunction and hypercoagulation, decrease neutrophil chemotaxis and leukocyte adhesion and prevent excessive nitric oxide which may help regulate oxidative stress.

A retrospective analysis of the databases of the two Leuven trials assessed the effect of intensive insulin therapy on blood glucose amplitude variation and pattern irregularity in critically ill patients(Meyfroidt et al., 2010). The Leuven intensive insulin therapy strategy increased mean daily delta blood glucose while not affecting standard deviation blood glucose. Increased blood glucose amplitude variation and pattern irregularity were associated with mortality, irrespective of blood glucose level. In contrast, the reduced mortality observed with intensive insulin therapy in the Leuven trials could not be attributed to an effect on blood glucose amplitude variation.

Hyperglycemia enhances neutrophil infiltration and increases cytokine expression in several animal models of cerebral ischemia and likely exacerbates the ischemic injury(Bemeur et al.,

A number of randomized controlled trials have been performed on the effects of strict glucose control in critically ill patients. Two single centre trials were performed in Leuven by van den Berghe et al(Van den Berghe et al., 2001;Van den Berghe et al., 2006a), followed by a number of multicentre trials. The first Leuven trial compared maintenance of blood glucose levels between 80 and 110 mg/dl versus 180 and 200 mg/dl in critically ill patients in a surgical ICU(Van den Berghe et al., 2001). In this trial strict glucose control resulted in a 42% reduction in mortality compared with conventional treatment. Septic patients and patients with an ICU stay > 5 days showed the largest reduction in mortality. In the second trial from Leuven performed in medical ICU patients no mortality benefit was demonstrated in the overall intention to treat analysis(Van den Berghe et al., 2006a). A reduction in hospital mortality from 52.5 to 43.0 percent was found in patients treated with intensive insulin therapy who stayed in the ICU for 3 days or more. Among patients treated < 3 days in the ICU, mortality was higher in the insulin group compared to the conventional group. The incidence of hypoglycemia was higher in the intervention group compared to the conventional treatment in both trials. Pooling the two datasets of both Leuven trials revealed that intensive insulin therapy reduced morbidity and mortality in a mixed surgical/medical ICU population, especially when continued for at least 3 days, without causing harm to patients treated for < 3 days(Van den Berghe et al., 2006b). The subgroup of patients with a prior history of diabetes did not appear to benefit. Blood glucose maintained at < 110 mg/dl was more effective that at 110-150 mg/dl,

How strict control of blood glucose reduces morbidity and mortality is unknown, but the mechanism may be related either to a direct effect of normalization of hyperglycemia or to the concomitantly higher insulin levels. *Post hoc* multivariate logistic regression analysis of the study by van den Berghe *et al*. suggests that the lowered blood glucose level rather than the insulin dose is related to the reduction in mortality (Van den Berghe et al., 2003). Apart from glucose lowering, insulin has a number of nonglycemic metabolic effects that may be important in critical illness (reviewed in (Honiden & Inzucchi, 2010)). Insulin can modulate inflammation via the mannose binding lectin pathway, via NF-kB and via modulation of pro- and anti-inflammatory cytokines. It can reduce free fatty acids and reverse the state of dyslipidemia in critical illness, regulate apotosis, prevent endothelial dysfunction and hypercoagulation, decrease neutrophil chemotaxis and leukocyte adhesion and prevent

A retrospective analysis of the databases of the two Leuven trials assessed the effect of intensive insulin therapy on blood glucose amplitude variation and pattern irregularity in critically ill patients(Meyfroidt et al., 2010). The Leuven intensive insulin therapy strategy increased mean daily delta blood glucose while not affecting standard deviation blood glucose. Increased blood glucose amplitude variation and pattern irregularity were associated with mortality, irrespective of blood glucose level. In contrast, the reduced mortality observed with intensive insulin therapy in the Leuven trials could not be

2005).

**4. Glucose control in critically ill patients** 

but also carried the highest risk of hypoglycemia.

excessive nitric oxide which may help regulate oxidative stress.

attributed to an effect on blood glucose amplitude variation.

**4.1 Intensive insulin therapy** 

After these landmark studies by Van den Berghe in 2001 and 2006, tight glycemic control was adopted as standard care in a large number of ICUs. However, subsequent randomized controlled multicentre trials were unable to replicate the results of these landmark trials. The Glucontrol study compared strict glucose control (blood glucose concentrations 80-110 mg/dL) to a target glucose between 140-180mg/dL, in an attempt to prevent the adverse effects of severe hyperglycemia, while reducing the risks of hypoglycemia(Preiser et al., 2009). This multicentre study was stopped early due to the high rate of protocol violations. Strict glucose control failed to induce any clinical benefit, but was associated with a higher incidence of hypoglycemia. The VISEP trial was a multicenter, two-by-two factorial trial, that randomly assigned patients with severe sepsis to receive either intensive insulin therapy to maintain euglycemia or conventional insulin therapy and either a low-molecularweight hydroxyethyl starch or modified Ringer's lactate for fluid resuscitation(Brunkhorst et al., 2008). After the first safety analysis, involving 488 patients intensive insulin therapy was terminated early by the data and safety monitoring board, owing to an increased number of hypoglycemic events (12.1%), as compared with conventional insulin therapy (2.1%). No differences in mortality and morbidity between the intensive and conventional treatments groups were found, but patients in the intensive-therapy group tended to have longer stays in the ICU than did patients in the conventional therapy group. The NICE-SUGAR trial was a large, international, randomized trial comparing intensive glucose control, with a target blood glucose range of 81 to 108 mg/dl to conventional glucose control, with a target < 180 mg/dl(Finfer et al. 2009). Intensive glucose control increased the absolute risk of death at 90 days by 2.6 percentage points; this represents a number needed to harm of 38. The difference in mortality remained significant after adjustment for potential confounders. Severe hypoglycemia was significantly more common with intensive glucose control. Given this lack of reproducible results in a heterogeneous group of ICU patients, and concerns over excessive hypoglycemia, extremely tight glucose control cannot be considered standard of care in ICU patients.

The multicentre trials were unable to replicate the findings of the 2 Leuven trials and raised the possibility that intensive insulin therapy may even increase the risk of mortality and morbidity in ICU patients. The explanation for the disparate findings seems multifactorial. In the Leuven studies the rate of use of total parenteral nutrition was higher compared to the other studies. Intensive insulin therapy may increase mortality in patients receiving enteral nutrition, possibly related to the adverse effects associated with hypoglycemia(Marik & Preiser, 2010). In turn, high dose parenteral glucose administration in the absence of intensive insulin therapy results in hyperglycemia with associated organ failure and death. In the Leuven studies, adjustments of insulin dosage were exclusively based on blood glucose measured on arterial blood via a point-of-care blood gas/glucose analyzer, whereas other studies used samples obtained from different sites and measured with different devices. The conventional treatment differed among the studies(Gunst & Van den Berghe, 2010). In the Leuven studies higher glucose values were tolerated compared to the other trials and the beneficial effects of intensive insulin therapy in these studies may be obtained by preventing excessive hyperglycemia.

#### **4.2 Intensive insulin therapy in brain injury**

Hyperglycemia at the time of brain injury is associated with increased morbidity and mortality. A planned subgroup analysis in patients with isolated brain injury of the first

Hypoglycemia in Critically Ill Patients 85

The reliability of the POCT devices itself in critically ill patients is also poor. In a prospective observational study the performance of 3 different POCT devices was tested and compared with the glucose oxidase method in arterial blood samples(Hoedemaekers et al., 2008). To minimize preanalytical bias the measurements were performed simultaneously by an experienced laboratory technician under controlled circumstances using a single arterial blood sample. Paired samples from all 3 tested devices were inaccurate in 4.9-13.4% of measurements. Inaccurate glucose readings were most frequently falsely elevated, and occurred over the entire range of blood glucose values. Patients with inaccurate POCT glucose results were significantly older, had a higher disease severity score, and a higher ICU mortality compared with patients with accurate glucose values. The mechanism underlying the differences in glucose values between the different POCT systems and the glucose oxidase method in critically ill patients is unknown. Accu-Chek uses the glucosedehydrogenasepyrroloquinolinequinone method for glucose determination, which is not specific for glucose. This method misinterprets maltose, icodextrine (which is converted to maltose), galactose, and xylose as glucose, leading to erroneously elevated glucose levels (Schleis, 2007). In addition, a large number of drugs commonly used in the treatment of critically ill patients, such as acetaminophen, dopamine, and mannitol, interfere with a number of POCT test systems. Changes in hematocrit concentration can influence the results of POCT measurements. Depending on the point-of-care testing device that is used both overestimation and underestimation of the glucose values can occur in patients with low hematocrit levels(Karon et al., 2008). Glucose measurement in the ICU setting using these bedside devices can be inaccurate and potentially dangerous: inaccurate glucose readings are most frequently falsely elevated, resulting in misinterpretation of high glucose values with subsequent inappropriate

In the past decade continuous glucose measurement devices have been developed in order to make glycemic management safer and more efficient. Due to concerns regarding altered perfusion in critical illness, many have questioned the accuracy of such devices in ICUs. So far, both excellent and poor performance has been reported in ICU patients (Bridges et al,. 2010;Corstjens et al., 2006;Holzinger et al., 2009;Price et al., 2008). Until these matters are solved, continuous monitoring of interstitial glucose values should be used with caution in

Glucose control in the ICU is markedly different from that in an out-patient clinic. Severe illness induces dysglycemia, with potential detrimental effects of low, high and variable glucose values. The injured brain is particularly susceptible for changes in glucose concentrations. strict glucose control is not proven beneficial in neurological ICU patients and has a unacceptable risk of hypoglycemia. Glucose control in the neurological ICU should be focused on maintenance of a steady level of glucose between 8-10 mmol/l, avoiding large fluctuations. Glucose monitoring in neurological ICU patients is difficult and

Arabi, Y. M., H. M. Tamim, and A. H. Rishu. (2009). Hypoglycemia with intensive insulin

therapy in critically ill patients: predisposing factors and association with mortality.

insulin administration or masking of true hypoglycemia.

the ICU.

**6. Conclusion** 

requires special attention.

*Crit Care Med.* 37:2536-2544.

**7. References** 

Leuven study revealed that intensive insulin therapy resulted in lower intracranial pressures, less seizures and a better long-term rehabilitation. Strict glucose control also protected general ICU patients against critical illness polyneuropathy.

A number of small studies on glucose control in critically ill neurological patients have been published, but most of these trials were too small to achieve sufficient statistical power to demonstrate possible effects on neurological outcome or mortality. Intensive insulin therapy did not change the incidence of vasospasm, neurological outcome or mortality rates in patients with acute subarachnoid hemorrhage(Bilotta et al., 2007). A decrease in infection rate from 42 to 27% was observed in the patients with strict glucose control compared to conventional glucose control. No differences in neurological outcome or mortality rates were found in patients after severe traumatic brain injury(Bilotta et al., 2008;Coester et al., 2010). A trial in 483 patients undergoing elective or emergency brain surgery revealed that intensive insulin therapy significantly reduced the length of stay in the ICU (6 vs. 8 days), and the infection rate (25.7% vs. 39.3%) without a significant effect on neurological outcome or survival at 6 months(Bilotta et al., 2009). In the UK Glucose Insulin in Stroke Trial (GIST-UK) patients presenting within 24 hours of stroke onset were randomly assigned to receive glucosepotassium-insulin infusion aiming at a capillary glucose between 72-126 mg/dL or no glucose intervention(Gray et al., 2007). The trial was stopped due to slow enrolment after 933 patients were recruited. There was no significant reduction in mortality or neurological disability at 90 days, although the study was underpowered and alternative results could not be excluded. The Treatment of Hyperglycemia in Ischemic Stroke (THIS) trial revealed similar results(Bruno et al., 2008). Strict versus moderate glucose control did not improve outcome in patients after resuscitation from ventricular fibrillation(Oksanen et al., 2007). Intensive or conventional control of blood glucose levels in mechanically ventilated adult neurologic ICU patients resulted in a non-significant increase in mortality in the patients in the intensive insulin group (36 vs 25%), with no differences in functional outcome(Green et al., 2010).

The results from the studies on strict glucose control in unselected critically ill patients may not be directly applicable to patients with critical neurological disease because of the high sensitivity of the brain to the effects of hypoglycemia. Tight glucose control was complicated by an increased number of hypoglycemic events in all trials in critically ill neurological patients. Since studies in patients with acute brain injury did not show a beneficial effect of strict glucose control on mortality or neurological outcome, the markedly increased risk of hypoglycemia limits the safe use of intensive insulin therapy these patients.

### **5. Glucose monitoring in the ICU setting**

Detection of hypoglycemia is difficult in ICU patients since these patients are often sedated and incapable of communicating, thereby masking clinical symptoms and signs. Frequent glucose measurement is therefore required to titrate the amount of insulin and detect episodes of hypoglycemia. For practical reasons bedside point of care (POCT) devices are frequently used. The accuracy of the POCT monitoring is influenced in several ways, including both preanalytic and analytic parameters. Glucose concentrations may differ according to the blood sampling site (venous, arterial or capillary blood). In critically ill patients, capillary blood glucose measured by fingerstick is inaccurate(Critchell et al., 2007;Kanji et al., 2005). Capillary sampling led to both overestimation(Kanji et al., 2005) and underestimation(Atkin et al., 1991) of blood glucose values. The presence of shock, use of vasopressors and upper extremity edema were associated with the occurrence of inaccurate readings.

Leuven study revealed that intensive insulin therapy resulted in lower intracranial pressures, less seizures and a better long-term rehabilitation. Strict glucose control also

A number of small studies on glucose control in critically ill neurological patients have been published, but most of these trials were too small to achieve sufficient statistical power to demonstrate possible effects on neurological outcome or mortality. Intensive insulin therapy did not change the incidence of vasospasm, neurological outcome or mortality rates in patients with acute subarachnoid hemorrhage(Bilotta et al., 2007). A decrease in infection rate from 42 to 27% was observed in the patients with strict glucose control compared to conventional glucose control. No differences in neurological outcome or mortality rates were found in patients after severe traumatic brain injury(Bilotta et al., 2008;Coester et al., 2010). A trial in 483 patients undergoing elective or emergency brain surgery revealed that intensive insulin therapy significantly reduced the length of stay in the ICU (6 vs. 8 days), and the infection rate (25.7% vs. 39.3%) without a significant effect on neurological outcome or survival at 6 months(Bilotta et al., 2009). In the UK Glucose Insulin in Stroke Trial (GIST-UK) patients presenting within 24 hours of stroke onset were randomly assigned to receive glucosepotassium-insulin infusion aiming at a capillary glucose between 72-126 mg/dL or no glucose intervention(Gray et al., 2007). The trial was stopped due to slow enrolment after 933 patients were recruited. There was no significant reduction in mortality or neurological disability at 90 days, although the study was underpowered and alternative results could not be excluded. The Treatment of Hyperglycemia in Ischemic Stroke (THIS) trial revealed similar results(Bruno et al., 2008). Strict versus moderate glucose control did not improve outcome in patients after resuscitation from ventricular fibrillation(Oksanen et al., 2007). Intensive or conventional control of blood glucose levels in mechanically ventilated adult neurologic ICU patients resulted in a non-significant increase in mortality in the patients in the intensive insulin group (36 vs 25%), with no differences in functional outcome(Green et al., 2010). The results from the studies on strict glucose control in unselected critically ill patients may not be directly applicable to patients with critical neurological disease because of the high sensitivity of the brain to the effects of hypoglycemia. Tight glucose control was complicated by an increased number of hypoglycemic events in all trials in critically ill neurological patients. Since studies in patients with acute brain injury did not show a beneficial effect of strict glucose control on mortality or neurological outcome, the markedly increased risk of

protected general ICU patients against critical illness polyneuropathy.

hypoglycemia limits the safe use of intensive insulin therapy these patients.

edema were associated with the occurrence of inaccurate readings.

Detection of hypoglycemia is difficult in ICU patients since these patients are often sedated and incapable of communicating, thereby masking clinical symptoms and signs. Frequent glucose measurement is therefore required to titrate the amount of insulin and detect episodes of hypoglycemia. For practical reasons bedside point of care (POCT) devices are frequently used. The accuracy of the POCT monitoring is influenced in several ways, including both preanalytic and analytic parameters. Glucose concentrations may differ according to the blood sampling site (venous, arterial or capillary blood). In critically ill patients, capillary blood glucose measured by fingerstick is inaccurate(Critchell et al., 2007;Kanji et al., 2005). Capillary sampling led to both overestimation(Kanji et al., 2005) and underestimation(Atkin et al., 1991) of blood glucose values. The presence of shock, use of vasopressors and upper extremity

**5. Glucose monitoring in the ICU setting** 

The reliability of the POCT devices itself in critically ill patients is also poor. In a prospective observational study the performance of 3 different POCT devices was tested and compared with the glucose oxidase method in arterial blood samples(Hoedemaekers et al., 2008). To minimize preanalytical bias the measurements were performed simultaneously by an experienced laboratory technician under controlled circumstances using a single arterial blood sample. Paired samples from all 3 tested devices were inaccurate in 4.9-13.4% of measurements. Inaccurate glucose readings were most frequently falsely elevated, and occurred over the entire range of blood glucose values. Patients with inaccurate POCT glucose results were significantly older, had a higher disease severity score, and a higher ICU mortality compared with patients with accurate glucose values. The mechanism underlying the differences in glucose values between the different POCT systems and the glucose oxidase method in critically ill patients is unknown. Accu-Chek uses the glucosedehydrogenasepyrroloquinolinequinone method for glucose determination, which is not specific for glucose. This method misinterprets maltose, icodextrine (which is converted to maltose), galactose, and xylose as glucose, leading to erroneously elevated glucose levels (Schleis, 2007). In addition, a large number of drugs commonly used in the treatment of critically ill patients, such as acetaminophen, dopamine, and mannitol, interfere with a number of POCT test systems. Changes in hematocrit concentration can influence the results of POCT measurements. Depending on the point-of-care testing device that is used both overestimation and underestimation of the glucose values can occur in patients with low hematocrit levels(Karon et al., 2008). Glucose measurement in the ICU setting using these bedside devices can be inaccurate and potentially dangerous: inaccurate glucose readings are most frequently falsely elevated, resulting in misinterpretation of high glucose values with subsequent inappropriate insulin administration or masking of true hypoglycemia.

In the past decade continuous glucose measurement devices have been developed in order to make glycemic management safer and more efficient. Due to concerns regarding altered perfusion in critical illness, many have questioned the accuracy of such devices in ICUs. So far, both excellent and poor performance has been reported in ICU patients (Bridges et al,. 2010;Corstjens et al., 2006;Holzinger et al., 2009;Price et al., 2008). Until these matters are solved, continuous monitoring of interstitial glucose values should be used with caution in the ICU.

### **6. Conclusion**

Glucose control in the ICU is markedly different from that in an out-patient clinic. Severe illness induces dysglycemia, with potential detrimental effects of low, high and variable glucose values. The injured brain is particularly susceptible for changes in glucose concentrations. strict glucose control is not proven beneficial in neurological ICU patients and has a unacceptable risk of hypoglycemia. Glucose control in the neurological ICU should be focused on maintenance of a steady level of glucose between 8-10 mmol/l, avoiding large fluctuations. Glucose monitoring in neurological ICU patients is difficult and requires special attention.

### **7. References**

Arabi, Y. M., H. M. Tamim, and A. H. Rishu. (2009). Hypoglycemia with intensive insulin therapy in critically ill patients: predisposing factors and association with mortality. *Crit Care Med.* 37:2536-2544.

Hypoglycemia in Critically Ill Patients 87

Bridges, B. C., C. M. Preissig, K. O. Maher, and M. R. Rigby. (2010). Continuous glucose monitors prove highly accurate in critically ill children. *Crit Care* 14:R176. Brunkhorst, F. M., C. Engel, F. Bloos, A. Meier-Hellmann, M. Ragaller, N. Weiler, O. Moerer,

Bruno, A., T. A. Kent, B. M. Coull, R. R. Shankar, C. Saha, K. J. Becker, B. M. Kissela, and L.

Capes, S. E., D. Hunt, K. Malmberg, and H. C. Gerstein. (2000). Stress hyperglycaemia and

Coester, A., C. R. Neumann, and M. I. Schmidt. (2010). Intensive insulin therapy in severe

Corstjens, A. M., J. J. Ligtenberg, der Horst van, I, R. Spanjersberg, J. S. Lind, J. E. Tulleken, J.

de Court, M. Kleinholz, K. R. Wagner, and R. E. Myers. (1989). Fatal strokes in

de Court, R. E. Myers, and L. Schoolfield. (1988). Hyperglycemia enlarges infarct size in

de Galan, B. E., M. G. Netea, P. Smits, and J. W. van der Meer. (2003). Hypoglycaemia

Dungan, K. M., S. S. Braithwaite, and J. C. Preiser. (2009). Stress hyperglycaemia. *Lancet*

Egi, M., R. Bellomo, E. Stachowski, C. J. French, and G. Hart. (2006). Variability of blood

Egi, M., R. Bellomo, E. Stachowski, C. J. French, G. K. Hart, C. Hegarty, and M. Bailey.

Ergul, A., W. Li, M. M. Elgebaly, A. Bruno, and S. C. Fagan. (2009). Hyperglycemia, diabetes and stroke: focus on the cerebrovasculature. *Vascul.Pharmacol.* 51:44-49. Fan, J., Y. H. Li, M. M. Wojnar, and C. H. Lang. (1996). Endotoxin-induced alterations in

Feuerstein, D., A. Manning, P. Hashemi, R. Bhatia, M. Fabricius, C. Tolias, C. Pahl, M.

microdialysis study. *J.Cereb.Blood Flow Metab* 30:1343-1355.

traumatic brain injury: a randomized trial. *J.Trauma* 68:904-911.

resuscitation in severe sepsis. *N.Engl.J.Med.* 358:125-139.

diabetes: a systematic overview. *Lancet* 355:773-778.

randomized pilot trial. *Stroke* 39:384-389.

*Intensive Care Med.* 33:2079-2084.

373:1798-1807.

*Anesthesiology* 105:244-252.

diabetes. *Crit Care Med.* 36:2249-2255.

skeletal muscle. *Shock* 6:164-170.

hyperglycemic cats. *Stroke* 20:1707-1715.

cerebrovascular occlusion in cats. *Stroke* 19:623-630.

does not affect IL-1beta, IL-6, or IL-10. *Cytokine* 22:71-76.

M. Gruendling, M. Oppert, S. Grond, D. Olthoff, U. Jaschinski, S. John, R. Rossaint, T. Welte, M. Schaefer, P. Kern, E. Kuhnt, M. Kiehntopf, C. Hartog, C. Natanson, M. Loeffler, and K. Reinhart. (2008). Intensive insulin therapy and pentastarch

S. Williams. (2008). Treatment of hyperglycemia in ischemic stroke (THIS): a

increased risk of death after myocardial infarction in patients with and without

H. Meertens, and J. G. Zijlstra. (2006). Accuracy and feasibility of point-of-care and continuous blood glucose analysis in critically ill ICU patients. *Crit Care* 10:R135. Critchell, C. D., V. Savarese, A. Callahan, C. Aboud, S. Jabbour, and P. Marik. (2007).

Accuracy of bedside capillary blood glucose measurements in critically ill patients.

downregulates endotoxin-induced production of tumour necrosis factor-alpha, but

glucose concentration and short-term mortality in critically ill patients.

(2008). Blood glucose concentration and outcome of critical illness: the impact of

insulin-stimulated phosphorylation of insulin receptor, IRS-1, and MAP kinase in

Ervine, A. J. Strong, and M. G. Boutelle. (2010). Dynamic metabolic response to multiple spreading depolarizations in patients with acute brain injury: an online


Atkin, S. H., A. Dasmahapatra, M. A. Jaker, M. I. Chorost, and S. Reddy. (1991). Fingerstick

Auer, R. N., J. Hugh, E. Cosgrove, and B. Curry. (1989). Neuropathologic findings in three

Badjatia, N., M. A. Topcuoglu, F. S. Buonanno, E. E. Smith, R. G. Nogueira, G. A. Rordorf, B.

Bagshaw, S. M., R. Bellomo, M. J. Jacka, M. Egi, G. K. Hart, and C. George. (2009). The

Bemeur, C., L. Ste-Marie, P. Desjardins, L. Vachon, R. F. Butterworth, A. S. Hazell, and J.

Bemeur, C., L. Ste-Marie, and J. Montgomery. (2007). Increased oxidative stress during

Benarroch, E. E. (2010). Glycogen metabolism: metabolic coupling between astrocytes and

Bergsneider, M., D. A. Hovda, E. Shalmon, D. F. Kelly, P. M. Vespa, N. A. Martin, M. E.

Bilotta, F., R. Caramia, I. Cernak, F. P. Paoloni, A. Doronzio, V. Cuzzone, A. Santoro, and G.

Bilotta, F., R. Caramia, F. P. Paoloni, R. Delfini, and G. Rosa. (2009). Safety and efficacy of

Bilotta, F., A. Spinelli, F. Giovannini, A. Doronzio, R. Delfini, and G. Rosa. (2007). The effect

Blumberg, D., S. Hochwald, M. F. Brennan, and M. Burt. (1995a). Interleukin-6 stimulates gluconeogenesis in primary cultures of rat hepatocytes. *Metabolism* 44:145-146. Blumberg, D., S. Hochwald, M. Burt, D. Donner, and M. F. Brennan. (1995b). Tumor necrosis

Boumezbeur, F., K. F. Petersen, G. W. Cline, G. F. Mason, K. L. Behar, G. I. Shulman, and D.

Brealey, D. and M. Singer. (2009). Hyperglycemia in critical illness: a review. *J.Diabetes* 

S. Carter, C. S. Ogilvy, and A. B. Singhal. (2005). Relationship between hyperglycemia and symptomatic vasospasm after subarachnoid hemorrhage. *Crit* 

impact of early hypoglycemia and blood glucose variability on outcome in critical

Montgomery. (2005). Dehydroascorbic acid normalizes several markers of oxidative stress and inflammation in acute hyperglycemic focal cerebral ischemia in the rat.

Phelps, D. L. McArthur, M. J. Caron, J. F. Kraus, and D. P. Becker. (1997). Cerebral hyperglycolysis following severe traumatic brain injury in humans: a positron

Rosa. (2008). Intensive insulin therapy after severe traumatic brain injury: a

intensive insulin therapy in critical neurosurgical patients. *Anesthesiology* 110:611-

of intensive insulin therapy on infection rate, vasospasm, neurologic outcome, and mortality in neurointensive care unit after intracranial aneurysm clipping in patients with acute subarachnoid hemorrhage: a randomized prospective pilot trial.

factor alpha stimulates gluconeogenesis from alanine in vivo. *J.Surg.Oncol.* 59:220-

L. Rothman. (2010). The contribution of blood lactate to brain energy metabolism in humans measured by dynamic 13C nuclear magnetic resonance spectroscopy.

glucose determination in shock *Ann.Intern.Med.* 114:1020-1024.

cases of profound hypoglycemia. *Clin.Neuropathol.* 8:63-68.

hyperglycemic cerebral ischemia *Neurochem.Int.* 50:890-904.

emission tomography study. *J.Neurosurg.* 86:241-251.

randomized clinical trial. *Neurocrit.Care* 9:159-166.

*Care Med.* 33:1603-1609.

illness. *Crit Care* 13:R91.

*Neurochem.Int.* 46:399-407.

619.

224.

neurons. *Neurology* 74:919-923.

*J.Neurosurg.Anesthesiol.* 19:156-160.

*J.Neurosci.* 30:13983-13991.

*Sci.Technol.* 3:1250-1260.


Hypoglycemia in Critically Ill Patients 89

Hutchinson, P. J., M. T. O'Connell, A. Seal, J. Nortje, I. Timofeev, P. G. Al-Rawi, J. P. Coles,

Jeremitsky, E., L. A. Omert, C. M. Dunham, J. Wilberger, and A. Rodriguez. (2005). The impact of hyperglycemia on patients with severe brain injury. *J.Trauma* 58:47-50. Kamada, H., F. Yu, C. Nito, and P. H. Chan. (2007). Influence of hyperglycemia on oxidative

Kanji, S., J. Buffie, B. Hutton, P. S. Bunting, A. Singh, K. McDonald, D. Fergusson, L. A.

Karon, B. S., L. Griesmann, R. Scott, S. C. Bryant, J. A. Dubois, T. L. Shirey, S. Presti, and P. J.

in acute spontaneous intracerebral hemorrhage. *J.Neurol.Sci.* 255:90-94. Krinsley, J. S. (2006). Glycemic control, diabetic status, and mortality in a heterogeneous

Krinsley, J. S. (2009). Glycemic variability and mortality in critically ill patients: the impact

Krinsley, J. S. and A. Grover. (2007). Severe hypoglycemia in critically ill patients: risk

Kruyt, N. D., G. J. Biessels, R. J. de Haan, M. Vermeulen, G. J. Rinkel, B. Coert, and Y. B.

Kruyt, N. D., Y. W. Roos, S. M. Dorhout Mees, W. M. van den Bergh, A. Algra, G. J. Rinkel,

Lam, A. M., H. R. Winn, B. F. Cullen, and N. Sundling. (1991). Hyperglycemia and neurological outcome in patients with head injury. *J.Neurosurg.* 75:545-551. Lanzino, G., N. F. Kassell, T. Germanson, L. Truskowski, and W. Alves. (1993). Plasma

Mackenzie, I. M., T. Whitehouse, and P. G. Nightingale. (2011). The metrics of glycaemic

Marik, P. E. and J. C. Preiser. (2010). Toward understanding tight glycemic control in the

McCowen, K. C., P. R. Ling, A. Ciccarone, Y. Mao, J. C. Chow, B. R. Bistrian, and R. J. Smith.

community hospital. *Semin.Thorac.Cardiovasc.Surg.* 18:317-325.

of diabetes. *J.Diabetes Sci.Technol.* 3:1292-1301.

factors and outcomes. *Crit Care Med.* 35:2262-2267.

hemorrhage: a meta-analysis. *Stroke* 40:e424-e430.

*J.Neurosurg.* 79:885-891.

haemorrhage. *J.Neurol.Neurosurg.Psychiatry* 79:1382-1385.

control in critical care. *Intensive Care Med.* 37:435-443.

the insulin-signaling cascade. *Crit Care Med.* 29:839-846.

ICU: a systematic review and metaanalysis. *Chest* 137:544-551.

measurement in critically ill adults. *Crit Care Med.* 33:2778-2785.

*Neurochir.(Wien.)* 151:51-61.

38:1044-1049.

T. D. Fryer, D. K. Menon, J. D. Pickard, and K. L. Carpenter. (2009). A combined microdialysis and FDG-PET study of glucose metabolism in head injury. *Acta* 

stress and matrix metalloproteinase-9 activation after focal cerebral ischemia/reperfusion in rats: relation to blood-brain barrier dysfunction. *Stroke*

McIntyre, and P. C. Hebert. (2005). Reliability of point-of-care testing for glucose

Santrach. (2008). Evaluation of the impact of hematocrit and other interference on the accuracy of hospital-based glucose meters. *Diabetes Technol.Ther.* 10:111-120. Kimura, K., Y. Iguchi, T. Inoue, K. Shibazaki, N. Matsumoto, K. Kobayashi, and S.

Yamashita. (2007). Hyperglycemia independently increases the risk of early death

population of critically ill patients before and during the era of intensive glycemic management: six and one-half years experience at a university-affiliated

Roos. (2009). Hyperglycemia and clinical outcome in aneurysmal subarachnoid

and G. J. Biessels. (2008). High mean fasting glucose levels independently predict poor outcome and delayed cerebral ischaemia after aneurysmal subarachnoid

glucose levels and outcome after aneurysmal subarachnoid hemorrhage.

(2001). Sustained endotoxemia leads to marked down-regulation of early steps in


Finfer, S., D. R. Chittock, S. Y. Su, D. Blair, D. Foster, V. Dhingra, R. Bellomo, D. Cook, P.

Fogelholm, R., K. Murros, A. Rissanen, and S. Avikainen. (2005). Admission blood glucose

Frontera, J. A., A. Fernandez, J. Claassen, M. Schmidt, H. C. Schumacher, K. Wartenberg, R.

Godoy, D. A. and Napoli M. Di. (2007). Hyperglycemia in acute phase of spontaneous

Godoy, D. A., G. R. Pinero, S. Svampa, F. Papa, and Napoli M. Di. (2008). Hyperglycemia

Godoy, D. A., G. R. Pinero, S. Svampa, F. Papa, and Napoli M. Di. (2009). Early

Gottesman, I., L. Mandarino, and J. Gerich. (1983). Estimation and kinetic analysis of insulinindependent glucose uptake in human subjects. *Am.J.Physiol* 244:E632-E635. Gray, C. S., A. J. Hildreth, P. A. Sandercock, J. E. O'Connell, D. E. Johnston, N. E. Cartlidge,

Green, D. M., K. H. O'Phelan, S. L. Bassin, C. W. Chang, T. S. Stern, and S. M. Asai. (2010).

Gruetter, R., K. Ugurbil, and E. R. Seaquist. (1998). Steady-state cerebral glucose concentrations and transport in the human brain. *J.Neurochem.* 70:397-408. Gunst, J. and Berghe G. Van den. (2010). Blood glucose control in the intensive care unit:

Hancu, N., M. G. Netea, and I. Baciu. (1998). High glucose concentrations increase the tumor

Hoedemaekers, C. W., J. M. Klein Gunnewiek, M. A. Prinsen, J. L. Willems, and J. G. Van

Holzinger, U., J. Warszawska, R. Kitzberger, H. Herkner, P. G. Metnitz, and C. Madl. (2009).

glucometers in critically ill patients. *Crit Care Med.* 36:3062-3066.

intracerebral hemorrhage (sICH). *J.Neurol.Sci.* 263:228-229.

Insulin in Stroke Trial (GIST-UK). *Lancet Neurol.* 6:397-406.

*N.Engl.J.Med.* 360:1283-1297.

*Neurocrit.Care* 9:217-229.

*Neurocrit.Care* 13:299-306.

cells. *Rom.J.Physiol* 35:325-330.

*Med.* 

benefits and risks. *Semin.Dial.* 23:157-162.

change? *Neurocrit.Care* 10:150-153.

203.

study. *J.Neurol.Neurosurg.Psychiatry* 76:349-353.

Dodek, W. R. Henderson, P. C. Hebert, S. Heritier, D. K. Heyland, C. McArthur, E. McDonald, I. Mitchell, J. A. Myburgh, R. Norton, J. Potter, B. G. Robinson, and J. J. Ronco. (2009). Intensive versus conventional glucose control in critically ill patients.

and short term survival in primary intracerebral haemorrhage: a population based

Temes, A. Parra, N. D. Ostapkovich, and S. A. Mayer. (2006). Hyperglycemia after SAH: predictors, associated complications, and impact on outcome. *Stroke* 37:199-

and short-term outcome in patients with spontaneous intracerebral hemorrhage.

hyperglycemia and intravenous insulin-the rationale and management of hyperglycemia for spontaneous intracerebral hemorrhage patients: is time for

J. M. Bamford, O. F. James, and K. G. Alberti. (2007). Glucose-potassium-insulin infusions in the management of post-stroke hyperglycaemia: the UK Glucose

Intensive versus conventional insulin therapy in critically ill neurologic patients.

necrosis factor-alpha production capacity by human peripheral blood mononuclear

der Hoeven. (2008). Accuracy of bedside glucose measurement from three

Impact of shock requiring norepinephrine on the accuracy and reliability of subcutaneous continuous glucose monitoring. *Intensive Care Med.* 35:1383-1389. Honiden, S. and S. E. Inzucchi. (2010). Glucose Controversies in the ICU. *J.Intensive Care* 


Hypoglycemia in Critically Ill Patients 91

Rady, M. Y., D. J. Johnson, B. M. Patel, J. S. Larson, and R. A. Helmers. (2005). Influence of

Rothman, D. L., N. R. Sibson, F. Hyder, J. Shen, K. L. Behar, and R. G. Shulman. (1999). In

Rovlias, A. and S. Kotsou. (2000). The influence of hyperglycemia on neurological outcome

Salim, A., P. Hadjizacharia, J. Dubose, C. Brown, K. Inaba, L. S. Chan, and D. Margulies.

Schleis, T. G. (2007). Interference of maltose, icodextrin, galactose, or xylose with some blood

Simpson, I. A., A. Carruthers, and S. J. Vannucci. (2007). Supply and demand in cerebral

Suh, S. W., A. M. Hamby, and R. A. Swanson. (2007). Hypoglycemia, brain energetics, and

Turina, M., D. E. Fry, and H. C. Polk, Jr. (2005). Acute hyperglycemia and the innate

Ueki, K., T. Kondo, and C. R. Kahn. (2004). Suppressor of cytokine signaling 1 (SOCS-1) and

Van den, Berghe G., A. Wilmer, G. Hermans, W. Meersseman, P. J. Wouters, I. Milants,

Van den, Berghe G., A. Wilmer, I. Milants, P. J. Wouters, B. Bouckaert, F. Bruyninckx, R.

Van den, Berghe G., P. J. Wouters, R. Bouillon, F. Weekers, C. Verwaest, M. Schetz, D.

van, den Oever, I, H. G. Raterman, M. T. Nurmohamed, and S. Simsek. (2010). Endothelial

*Philos.Trans.R.Soc.Lond B Biol.Sci.* 354:1165-1177.

hypoglycemic neuronal death. *Glia* 55:1280-1286.

therapy in the medical ICU. *N.Engl.J.Med.* 354:449-461.

therapy in the critically ill patients. *N.Engl.J.Med.* 345:1359-1367.

predictor of outcome. *Am.Surg.* 75:25-29.

in patients with severe head injury. *Neurosurgery* 46:335-342.

glucose monitoring systems. *Pharmacotherapy* 27:1313-1321.

patients with or without diabetes mellitus. *Mayo Clin.Proc.* 80:1558-1567. Raichle, M. E. and M. A. Mintun. (2006). Brain work and brain imaging. *Annu.Rev.Neurosci.*

29:449-476.

27:1766-1791.

1633.

5446.

*Med.* 31:359-366.

2010:792393.

individual characteristics on outcome of glycemic control in intensive care unit

vivo nuclear magnetic resonance spectroscopy studies of the relationship between the glutamate-glutamine neurotransmitter cycle and functional neuroenergetics.

(2009). Persistent hyperglycemia in severe traumatic brain injury: an independent

energy metabolism: the role of nutrient transporters. *J.Cereb.Blood Flow Metab*

immune system: clinical, cellular, and molecular aspects. *Crit Care Med.* 33:1624-

SOCS-3 cause insulin resistance through inhibition of tyrosine phosphorylation of insulin receptor substrate proteins by discrete mechanisms. *Mol.Cell Biol.* 24:5434-

Wijngaerden E. Van, H. Bobbaers, and R. Bouillon. (2006a). Intensive insulin

Bouillon, and M. Schetz. (2006b). Intensive insulin therapy in mixed medical/surgical intensive care units: benefit versus harm. *Diabetes* 55:3151-3159. Van den, Berghe G., P. Wouters, F. Weekers, C. Verwaest, F. Bruyninckx, M. Schetz, D.

Vlasselaers, P. Ferdinande, P. Lauwers, and R. Bouillon. (2001). Intensive insulin

Vlasselaers, P. Ferdinande, and P. Lauwers. (2003). Outcome benefit of intensive insulin therapy in the critically ill: Insulin dose versus glycemic control. *Crit Care* 

dysfunction, inflammation, and apoptosis in diabetes mellitus. *Mediators.Inflamm.*


Meierhans, R., M. Bechir, S. Ludwig, J. Sommerfeld, G. Brandi, C. Haberthur, R. Stocker, and

Meyer, C., M. Stumvoll, J. Dostou, S. Welle, M. Haymond, and J. Gerich. (2002). Renal

Meyer, C., M. Stumvoll, S. Welle, H. J. Woerle, M. Haymond, and J. Gerich. (2003). Relative

gluconeogenesis in humans. *Am.J.Physiol Endocrinol.Metab* 285:E819-E826. Meyfroidt, G., D. M. Keenan, X. Wang, P. J. Wouters, J. D. Veldhuis, and Berghe G. Van den.

Mizock, B. A. (1995). Alterations in carbohydrate metabolism during stress: a review of the

Morohoshi, M., K. Fujisawa, I. Uchimura, and F. Numano. (1996). Glucose-dependent

Offner, H., S. Subramanian, S. M. Parker, M. E. Afentoulis, A. A. Vandenbark, and P. D.

Oksanen, T., M. B. Skrifvars, T. Varpula, A. Kuitunen, V. Pettila, J. Nurmi, and M. Castren.

Parkin, M., S. Hopwood, D. A. Jones, P. Hashemi, H. Landolt, M. Fabricius, M. Lauritzen, M.

Passero, S., G. Ciacci, and M. Ulivelli. (2003). The influence of diabetes and hyperglycemia on clinical course after intracerebral hemorrhage. *Neurology* 61:1351-1356. Pellerin, L., A. K. Bouzier-Sore, A. Aubert, S. Serres, M. Merle, R. Costalat, and P. J.

Pellerin, L. and P. J. Magistretti. (2004). Neuroenergetics: calling upon astrocytes to satisfy

Preiser, J. C., P. Devos, S. Ruiz-Santana, C. Melot, D. Annane, J. Groeneveld, G. Iapichino, X.

Price, G. C., K. Stevenson, and T. S. Walsh. (2008). Evaluation of a continuous glucose

peripheral immune system. *J.Cereb.Blood Flow Metab* 26:654-665.

injury. *Crit Care* 14:R13.

*Am.J.Physiol Endocrinol.Metab* 282:E428-E434.

mortality. *Crit Care Med.* 38:1021-1029.

monocytes in vitro. *Diabetes* 45:954-959.

fibrillation. *Intensive Care Med.* 33:2093-2100.

*J.Cereb.Blood Flow Metab* 25:402-413.

astrocytes: an update. *Glia* 55:1251-1262.

hungry neurons. *Neuroscientist.* 10:53-62.

216.

Glucontrol study. *Intensive Care Med.* 35:1738-1748.

literature. *Am.J.Med.* 98:75-84.

J. F. Stover. (2010). Brain metabolism is significantly impaired at blood glucose below 6 mM and brain glucose below 1 mM in patients with severe traumatic brain

substrate exchange and gluconeogenesis in normal postabsorptive humans.

importance of liver, kidney, and substrates in epinephrine-induced increased

(2010). Dynamic characteristics of blood glucose time series during the course of critical illness: effects of intensive insulin therapy and relative association with

interleukin 6 and tumor necrosis factor production by human peripheral blood

Hurn. (2006). Experimental stroke induces massive, rapid activation of the

(2007). Strict versus moderate glucose control after resuscitation from ventricular

G. Boutelle, and A. J. Strong. (2005). Dynamic changes in brain glucose and lactate in pericontusional areas of the human cerebral cortex, monitored with rapid sampling on-line microdialysis: relationship with depolarisation-like events.

Magistretti. (2007). Activity-dependent regulation of energy metabolism by

Leverve, G. Nitenberg, P. Singer, J. Wernerman, M. Joannidis, A. Stecher, and R. Chiolero. (2009). A prospective randomised multi-centre controlled trial on tight glucose control by intensive insulin therapy in adult intensive care units: the

monitor in an unselected general intensive care population. *Crit Care Resusc.* 10:209-


**Part 3** 

**Section C** 


**Part 3** 

**Section C** 

92 Diabetes – Damages and Treatments

Vriesendorp, T. M., Santen S. van, J. H. DeVries, Jonge E. de, F. R. Rosendaal, M. J. Schultz,

Watts, A. G. and C. M. Donovan. (2010). Sweet talk in the brain: glucosensing, neural networks, and hypoglycemic counterregulation. *Front Neuroendocrinol.* 31:32-43. Whitcomb, B. W., E. K. Pradhan, A. G. Pittas, M. C. Roghmann, and E. N. Perencevich.

intensive care unit populations. *Crit Care Med.* 33:2772-2777.

care unit. *Crit Care Med.* 34:96-101.

and J. B. Hoekstra. (2006). Predisposing factors for hypoglycemia in the intensive

(2005). Impact of admission hyperglycemia on hospital mortality in various

**5** 

Oren Tirosh

 *Israel* 

**Inflammation and Hypoglycemia:** 

*The Robert H Smith Faculty of Agriculture, Food and Environment,* 

Patients can be exposed to a variety of potentially life threatening acute inflammations mainly sepsis, which accounts to 9% of all death in the US. The prevalence of non-alcoholic fatty liver disease (NAFLD) is about 30% in the general population. Fatty liver is known to be more sensitive to endotoxins. It has been reported that metabolic aspects of sepsis and endotoxemia are suppression of the fatty acid beta-oxidation pathway and severe hypoglycemia. This can be due to lipotoxic effects following accumulation of free fatty acids in the liver and suppression of gluconeogenesis. In this chapter we will review the published facts about the development of hypoglycemic effects during sepsis and the possible connection of such an effect to the dysregulation of lipid metabolism. Secondly a possible redox related antilipotoxic cellular mechanism will be suggested. Such mechanism can alleviate the endotoxic hypoglycemic effect and is related to nitric oxide signaling. Nitric oxide signaling has been demonstrated to regulate the metabolic status of cells including upregulation of mitochondrial biogenesis, promoting liver glucose production and depending on the biological setting to protect cells against accumulation of oxidative damage, all possibly protect against development of hypoglycemia following liver injury.

Non-alcoholic fatty liver disease NAFLD comprises a spectrum of hepatic pathology, ranging from simple steatosis (SS), in which there is an increase of fat accumulation in hepatocytes, through steatohepatitis to cirrhosis (Farrell, GC et al., 2008). Primary NAFLD is associated with obesity, insulin resistance and metabolic syndrome, diabetes and dyslipidemia, while secondary NAFLD is associated with all forms of liver damage including viral infections autoimmune and heradetory disease, drugs, toxins and nutrition (parenteral nutrition , B12/folic acid deficiency etc.) (Musso, G et al., 2010) (Figure 1). Nonalcoholic steatohepatitis (NASH) is a progressive lesion in which steatosis is accompanied by hepatocyte injury and death, as well as hepatic infiltration by inflammatory cells. NASH-related liver damage often triggers liver fibrosis. In severe cases, NASH may progress to cirrhosis and possibly hepatocellular carcinoma (Lim, JH et al., 2006). NAFLD is one of the most common liver diseases worldwide, affecting all racial, ethnic, and age

**1. Introduction** 

**2.1 Introduction** 

**2. Non-alcoholic fatty liver disease** 

*Institute of Biochemistry, Food Science and Nutrition,* 

 *The Hebrew University of Jerusalem, Rehovot,* 

**The Lipid Connection** 
