The Importance of Glucose Regulation in the ICU
ACVIM 2008
Amy E. DeClue, DVM, MS, DACVIM
Columbia, MO, USA


Glucose is an important energy source for cells. In health, blood glucose concentrations are tightly regulated. Numerous hormones like insulin, glucagon, corticosteroids and catecholamines work in concert to maintain normal glucose homeostasis. Glucose is stored as glycogen in liver (25%) and muscle tissue (75%). Glycogenolysis can be initiated in the liver to release glucose into the blood stream during times of fasting. However, lacking glucose-6-phosphatase, muscle tissue is unable to release glucose into the blood stream. The liver is also capable of producing glucose from pyruvate, lactate, glycerol and proteins, a process known as gluconeogenesis. Maintaining blood glucose concentrations allows a continual supply of glucose to tissues that are glucose dependent for energy metabolism like brain, retina and germinal epithelium of the gonads.

Hyperglycemia is defined as a blood glucose concentration greater than 130 mg/dl in dogs and cats. A number of factors can lead to hyperglycemia. Traditionally, when considering the cause of hyperglycemia, one assumes there is a relative or absolute lack of insulin. However, hyperglycemia can result from production of counterregulatory hormones like glucocorticoids, glucagon, growth hormone or catecholamines, peripheral insulin resistance or excessive gluconeogenesis. Many different disease processes can lead to the production of counterregulatory hormones or peripheral tissue insulin resistance including trauma, infection and inflammation. Administration total or partial parenteral nutrition, dextrose containing fluids, vasopressors, some anesthetic drugs or glucocorticoids may also contribute to hyperglycemia in small animal patients.

Glucose and Critical Illness

During critical illness, transient hyperglycemia in a patient that lacks a previous history of diabetes mellitus is termed stress hyperglycemia. Stress hyperglycemia during critical illnesses like sepsis results from low to normal insulin concentration, increased counterregulatory hormone secretion, and peripheral tissue insulin resistance. In some instances insulin concentrations may be increased, yet hyperglycemia persists. This may be due, in part, to inhibition of insulin receptor substrate signaling and up-regulation of glucose transporters. Additionally, cortisol, epinephrine, norepinephrine, growth hormone and glucagon release antagonize the effects of insulin and lead to development of a catabolic state. Further, pro-inflammatory mediators play a role in induction of a hyperglycemic state during critical illness through inhibition of insulin release and promotion of hepatic gluconeogenesis.

Previously, the development of stress hyperglycemia was considered a reflection of the primary disease process and was generally left untreated until the blood glucose exceeded 200 mg/dl at which time insulin therapy was initiated. However, recent information from human and veterinary studies has indicated that even transient, mild hyperglycemia may have deleterious effects in the critically ill. Hyperglycemia is common in critically ill patients and has been correlated with increased morbidity and mortality in humans.1 This correlation is most notable in septic, cardiac, and neurologic patients.1-7 Further, there is evidence in human medicine that maintaining tight glycemic control provides a survival advantage for critically ill patients.

Glucose acts as an effective osmole and is capable if causing fluid shifting between compartments. Hyperglycemia pulls water from the intracellular compartment into the vasculature resulting in cellular dehydration. Hyponatremia is often associated with severe hyperglycemia because this hypotonic fluid shift causes a dilution of the blood sodium. Typically, for every 100 mg/dl increase in blood glucose above the reference interval there should be a 1.6 mEq/l decrease in plasma sodium concentration. Failure to observe this decrease in sodium indicates a free water deficit and hyperosmolality, especially in patients with severe hyperglycemia. Hyperglycemia that exceeds the renal transport maximum (180-200 mg/dl in dogs, 280-310 mg/dl in cats), results in glucosuria. Glucose inhibits normal urine concentrating mechanisms causing an osmotic diuresis. Dehydration and hypovolemia are possible sequelae to osmotic diuresis, especially in animals that are unable to drink adequate volumes of water to compensate. Ultimately, dehydration and hypovolemia lead to hypoperfusion and lactic acidosis. The contribution of osmotic diuresis-induced dehydration, hypoperfusion and lactic acidosis can be particularly devastating to critically ill animals.

There is strong evidence that glucose acts as an immunomodulating compound which is likely the reason that even mild hyperglycemia has an influence on mortality. Glucose promotes production of pro-inflammatory cytokines like tumor necrosis factor (TNF)-α and interleukin (IL)-6 in a dose dependent manner.8,9 In humans with naturally acquired sepsis, induction of hyperglycemia resulted in significantly higher plasma TNF-α and IL-6 concentrations for a significantly longer duration of time.10,11 Glucose promotes the activation of NF-kappa B, a transcription factor that is responsible for the transcription of numerous pro-inflammatory genes and induction of apoptosis. Activation of NF-kappa B is associated with many critical illnesses like sepsis, SIRS, MODS and ARDS. Similar pro-inflammatory effects have been noted in experimental animal models as well. Hyperglycemia enhances the pro-inflammatory and oxidative response to low dose endotoxin in rats.12 In rabbits, glucose loading increases LPS-induced TNF-α production.13

Hyperglycemia activates the endothelium and the coagulation system leading to a hypercoagulable state. Additionally, hyperglycemia is known to induce the release of fatty acids and promotes vasoconstriction both of which can lead to myocardial injury and dysfunction. In humans, hyperglycemia has been associated with an increased risk for wound infections as well. Increased risk of infection is likely due to the metabolic abnormalities associated with hyperglycemia which alter immune function through impairment of leukocyte adherence, chemotaxis, opsonization, phagocytosis, intracellular killing, and superoxide activity.

Insulin, on the other hand, antagonizes the effects of glucose and can act as an anti-inflammatory compound. Insulin is known to suppress TNF- α and the formation and release of reactive oxygen species. Insulin decreases the activation and nuclear translocation of transcription factor NF-kappa B and thus blunts inflammatory mediator production and cell apoptosis. Insulin also decreases ICAM-1, a cell receptor involved in leukocyte recruitment and activation.

Evidence in Veterinary Medicine

There is a limited amount of research pertaining to the clinical impact of hyperglycemia in non-diabetic small animals. Hyperglycemia is recognized in the small animal critical care population due to administration of dextrose containing fluids or alterations in glucose homeostasis. Dextrose containing fluid, specifically parenteral nutrition, has been implicated in the induction of hyperglycemia in non-diabetic dogs and cats in several studies.14-16 Importantly, in non-diabetic, critically ill cats receiving parenteral nutrition, hyperglycemia was associated with a poorer prognosis.15

Clinical research evaluating stress hyperglycemia in non-diabetic patients is somewhat limited, however there is some evidence that stress hyperglycemia develops in critically ill small animals and may be associated with a poorer prognosis. Brady, et al. evaluated plasma sodium and glucose concentrations in 59 dogs presenting for newly diagnosed congestive heart failure. In this study, the mean plasma glucose concentration was significantly lower in the survivors (100 ±13 mg/dl) versus nonsurvivors (128±52 mg/dl).17 Carbohydrate metabolism in critically ill cats has been evaluated by Chan, et al. This study of 26 non-diabetic cats admitted to the intensive care unit and 21 healthy control cats assessed blood glucose, lactate, cortisol, insulin, glucagon, epinephrine, norepinephrine and non-esterified fatty acid concentrations.18 Critically ill cats had significantly higher median blood glucose concentrations (183, range 51-321 mg/dl) than healthy controls (110, range 91-165 mg/dl).18 Critically ill cats also had significantly lower plasma insulin and significantly higher lactate, cortisol, glucagon, and norepinephrine concentrations.18 Syring, et al. investigated if hyperglycemia corresponded to the severity of head trauma or outcome in dogs and cats. Blood glucose concentrations were significantly associated with the severity of head trauma but were not associated with outcome.19 Hardie, et al. evaluated postsurgical blood glucose concentrations in dogs with sepsis and in healthy dogs. Fifty percent of dogs (n=8) with postsurgical plasma glucose concentrations greater than 150 mg/dl died while only 14% of dogs (n=7) with plasma glucose concentrations less than 150 mg/dl died.20 The difference in these groups was not significant (p=0.08), but this may be a reflection of the small sample size.20 Perhaps some of the most compelling evidence pertaining to hyperglycemia in the critically ill was a recent, clinical study evaluated blood glucose concentrations in a population of dogs admitted to an ICU.21 In this study, the incidence of hyperglycemia was 16%. The study found that nonsurvivors had significantly higher glucose concentrations (median 176, range 122-310 mg/dl) than survivors (median 139, range 121-191 mg/dl) and that dogs that developed hyperglycemia during hospitalization had a longer length of hospitalization than dogs that had hyperglycemia at presentation.21 In the same study, dogs with hyperglycemia had a higher incidence of septic complications.21

Monitoring Glucose in the ICU

Blood glucose concentrations in critically ill patients can quickly fluctuate. Because of this, frequent bedside monitoring (q 1-4 h) is recommended based on the clinical assessment of the patient. Bedside glucose monitoring can be accomplished using a handheld glucometer, i-STAT® handheld blood analyzer or the Guardian® Real-Time continuous glucose monitoring system. Hemorrhage, anemia and hypovolemia are potential complications associated with repeated phlebotomy in critically ill dogs and cats. Additionally, the stress of repeated phlebotomy itself can induce stress hyperglycemia. To circumvent some of these issues, it is recommended to place a central venous catheter for frequent blood collection. Although central venous catheters help alleviate stress and bruising from repeated phlebotomy, iatrogenic anemia or hypovolemia are still possible complications depending upon the size of the patient. There are also concerns about the potential thrombogenic nature of central venous catheters, especially in critically ill patients with conditions that may predispose them to thromboembolic disease. Nevertheless, central venous catheter placement should be considered in any patient that requires frequent blood sampling.

Continuous glucose monitoring is a new option for assessment of glucose homeostasis in dogs and cats.22,23 The continuous glucose monitor system (CGMS) consists of a flexible electrode sensor, recording device, docking station, computer and appropriate software. The system detects glucose concentrations in the interstitial space via a reaction between interstitial glucose and glucose oxidase. This reaction results in an electrical signal that is converted to a glucose concentration via a mathematical model. The interstitial glucose concentration is recorded in milligrams per deciliter every 10 seconds and reported as the mean interstitial glucose concentration every 5 minutes (figure 1). Interstitial glucose concentrations have excellent correlation with blood glucose concentrations with a lag time of about 10 min. The range of glucose detection for the CGMS is 40-400 mg/dL. Interstitial glucose concentrations can be evaluated for up to 72 hours with one probe. The development of the Medtronic Guardian® Real-Time CGMS (figure 2), which provides a continuous display of the interstitial glucose concentration, has opened new clinical applications for continuous glucose monitoring. The system is wireless, minimizing the impact on patient mobility and necessity for patient interaction. The real time monitor allows the clinician to detect trends and rapidly identify glucose fluctuations without while avoiding repeated phlebotomies for the patient. The real time continuous glucose monitor is ideal for monitoring critically ill patients.

Figure 1.
Figure 1.

Graph of interstitial glucose concentrations in a dog over 72 hours.

Figure 2. CGMS
Figure 2. CGMS


Managing Hyperglycemia in the ICU

Numerous clinical trials in human medicine have confirmed the importance of maintaining blood glucose concentrations within a tight range. In a landmark study in human medicine, maintaining blood glucose concentrations between 80-110 mg/dl via IV regular insulin administration (intensive insulin therapy) decreased mortality.2 In this study, intensive insulin therapy decreased bloodstream infections by 46%, acute renalfailure requiring dialysis or hemofiltration by 41%,the median number of red-cell transfusions by 50 %, andcritical-illness polyneuropathy by 44 %.2 Most importantly, controlling hyperglycemia has been shown to decrease mortality in the ICU unit by 42% and overall in hospital mortality by 34%.2 Further clinical investigation has indicated that blood glucose concentrations, not the amount of insulin given, is correlated to mortality indicating that hyperglycemia has a negative impact on patients with critical illness.24 The use of hypocaloric nutrition to decrease hyperglycemia has not been a successful strategy for controlling hyperglycemia in humans.

Currently, there are no studies evaluating the use of intensive insulin therapy to maintain normoglycemia in nondiabetic critically ill dogs or cats. However, for critically ill small animals with hyperglycemia (BG >130 mg/dl), insulin administration could be considered but should be approached cautiously and only with appropriate monitoring. Regular insulin (0.25-1 U/ kg/ day, IV as a constant rate infusion) can be used and the rate of infusion adjusted to achieve blood glucose concentrations within the target range (85-130 mg/dl). Any patient receiving IV regular insulin therapy should have blood glucose concentrations determined at least every 2-3 hours and should have electrolytes monitored a minimum of once daily. Potential side effects of insulin therapy include hypoglycemia, hypokalemia, hypophosphatemia and hypomagnesemia. Adequate nutrition and administration of IV fluids supplemented with potassium, phosphorus and magnesium will help avoid these complications. Additionally, drugs that may cause hyperglycemia should be avoided in critically ill small animals whenever possible. Examples of medications that may alter glucose homeostasis include glucocorticoids, androgens, ketamine, α2-agonists, catecholamines and morphine. Care should be taken to avoid dextrose containing fluids with the exception of the use of parenteral nutrition for patients that are unable to receive enteral nutrition. When dextrose containing fluids are used to supplement patients with hypoglycemia, the dextrose dose should be carefully titrated to maintain normoglycemia.


Hyperglycemia has profound effects on the immune system during critical illness. Any critically ill patient should be carefully monitored for hyperglycemia. Dextrose containing fluids or drugs that promote hyperglycemia should be avoided whenever possible. Little is known in veterinary medicine about the management of hyperglycemia in nondiabetic, critically ill patients. Further research to evaluate the mechanisms by which hyperglycemia promotes mortality and if therapeutic interventions such as insulin therapy will ameliorate the negative impact of hyperglycemia in dogs and cats are needed.


1.  Asadollahi K, et al. Qjm 2007;100:501-507.

2.  Van den Berghe G, et al. N Engl J Med 2001;345:1359-1367.

3.  Van den Berghe G, et al. Diabetes 2006;55:3151-3159.

4.  Young B, et al. Ann Surg 1989;210:466-472.

5.  Krinsley JS. Semin Thorac Cardiovasc Surg 2006;18:317-325.

6.  Norhammar AM, et al. Diabetes Care 1999;22:1827-1831.

7.  Zindrou D, et al. Diabetes Care 2001;24:1634-1639.

8.  Hancu N, et al. Rom J Physiol 1998;35:325-330.

9.  Morohoshi M, et al. Diabetes 1996;45:954-959.

10. Esposito K, et al. Circulation 2002;106:2067-2072.

11. Yu WK, et al. World J Gastroenterol 2003;9:1824-1827.

12. Ling PR, et al. Crit Care Med 2005;33:1084-1089.

13. Losser MR, et al. J Appl Physiol 1997;83:1566-1574.

14. Chan D, et al. J Vet Intern Med 2002;16:440-445.

15. Pyle S, et al. J Am Vet Med Assoc 2004;225:242-250.

16. Crabb S, et al. Journal of Veterinary Emergency and Critical Care 2006;16:S21-26.

17. Brady C, et al. Journal of Veterinary Emergency and Critical Care 2004;14:177-182.

18. Chan D, et al. Journal of Veterinary Emergency and Critical Care 2006;16:S7-13.

19. Syring R, et al. J Am Vet Med Assoc 2001;218:1124-1129.

20. Hardie E, et al. Am J Vet Res 1985;46:1700-1704.

21. Torre D, et al. J Vet Intern Med 2007;21:971-975.

22. Wiedmeyer C, et al. JAVMA 2003;223:987-992.

23. DeClue AE, et al. JAAHA 2004;40:171-172.

24. Finney SJ, et al. JAMA 2003;290:2041-2047.

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Columbia, MO

MAIN : ACVIM SAIM Generalist : Glucose Regulation
Powered By VIN