Hyperglycaemia in the critically ill patient: ideal treatment and impact of the condition

Abstract

This thesis is composed of four distinct related chapters focusing on glycaemia in the intensive care unit (ICU), its treatment, and the impact of the condition. Specifically, this research concentrates on approaches to glycaemic management, enteral feeding, and implications of hyperglycaemia in the critically ill. The work submitted comprises three literature reviews and five original studies. In the critically ill, delayed gastric emptying occurs frequently and is associated with adverse outcomes. Numerous techniques to quantify gastric emptying exist (Chapter 1.2). Energy dense feeds are sometimes administered to patients with delayed gastric emptying based on the rationale that volume is a determinant of gastric emptying. However, in health, tight regulation of gastric emptying occurs via ‘enterogastric feedback’, such that it is calorie load (not volume) that is the major determinant of gastric emptying. An analysis of previously obtained data suggested energy dense feed is associated with a slower emptying rate than a standard feed, resulting in similar caloric delivery (Chapter 1.3). Hyperglycaemia occurs frequently during critical illness in patients with pre-existing diabetes. The current approach to treatment of glycaemia in this group is to treat them identical to patients without diabetes. This strategy may be flawed as observational data suggest that the impact of acute glycaemia on outcomes is dependent on premorbid glycaemia (Chapter 2.2). To provide further information, a prospective sequential period pilot study was completed (Chapter 2.3). This study suggests that a more liberal approach may reduce hypoglycaemia episodes and that further trials of more liberal glucose targets are warranted. The treatment of hyperglycaemia during critical illness (for both patients with and without pre-existing diabetes) requires administration of insulin; however, this is not without risk. Therefore a prospective, randomised, cross over study in critically ill patients to determine the effect of glucose-dependent insulinotropic polypeptide (GIP) was performed. The administration of GIP in pharmacological doses when compared to placebo did not effect glycaemia, glucose absorption or gastric emptying (Chapter 2.4). Stress hyperglycaemia occurs frequently in critically ill patients but is not generally considered a risk factor for subsequent glucose intolerance. A systematic review and meta-analysis was conducted suggesting stress hyperglycaemia was associated with an increased risk of both prediabetes and diabetes (Chapter 3.2). Subsequently, a prospective cohort study was performed confirming this signal and providing mechanistic information regarding the progression to prediabetes and diabetes (Chapter 3.3). Patients during and recovering from critical illness as well as ambulant patients with diabetes frequently experience episodes of hypoglycaemia. The counter-regulatory response to hypoglycaemia is to accelerate gastric emptying, increasing carbohydrate absorption from the small intestine. A study was performed to understand whether recurrent episodes of hypoglycaemia diminish the standard counter-regulatory response (Chapter 4.2). In healthy volunteers, the acceleration of gastric emptying during acute hypoglycaemia in health did not appear to be affected by antecedent hypoglycaemia. In summary, this program of work has contributed new and important information in the fields of diabetes management, feeding in the critical care setting, and the implications of stress hyperglycaemia in the critically ill.