| |

Click on image to enlarge

Dissertations eBook from River Publishers

Decision Support for Blood Glucose Control in Critically Ill Patients

Author: Ulrike Pielmeier, Center for Model-Based Medical Decision Support, Department of Health Science and Technology, Aalborg University, Aalborg, Denmark

e-ISBN: 9788792329421

Price: Notify Me

Available: December 2012
Downloads: [1729]

Downloads: [1729]  
Technical | Layman

Layman Description:

A high proportion of patients suffering an acute stress such as stroke or severe illness may develop high blood sugar while in hospital. Stress-induced high blood sugar is a transient condition and may occur as a complication to all types of patients, also if the patient has no previous history of diabetes. Research studies show that untreated high blood sugar leads to prolonged illness and increases the mortality risk. Clinicians can choose to reduce the high blood sugar by treating the patient with insulin or by making nutritional adjustments that reduce the amount of carbohydrates fed to the patient. However, to find the optimal decisionis tricky as too much insulin may cause dangerously low blood sugar and energy needs must be considered. This dissertation describes the building of a computer program for use by clinicians and nurses to improve blood sugar control of the critically ill. The program goes by the name of ?Glucosafe? and was developed and tested by the author. It is a decision support system that incorporates a physiological model of human insulin kinetics and glucose metabolism. Based on the model, recommendations regarding insulin and nutrition are given by the program that minimize the risk of low or high blood sugar, undernourishment, or overfeeding. The theses summarizes how the model works, the design of the program for ease of use under clinical conditions, and the results from a clinical pilot study of the system in a Danish hospital setting, where blood sugar control was significantly improved by the use of Glucosafe.


Hyperglycemia in critical care