Blood glucose disturbance in patients encountered by emergency medical service

Hanna Vihonen

Research output: ThesisDoctoral Thesis


Background: Blood glucose disturbance is commonly seen in critically ill patients and it is associated with a poorer outcome. Aim of the study: To (a) describe how common blood glucose disturbance is in the prehospital setting and its association with short-term mortality and morbidity, (b) describe the very early mechanisms of stress-induced hyperglycaemia, and (c) examine whether blood glucose could be added as an additional parameter to a “track and trigger” scoring system for better discrimination of risk of death. Material and Methods: This study was conducted in areas of district of Helsinki and Uusimaa between 2006-2015 including one prospective and three retrospective sub-studies: 28 out-of-hospital cardiac arrest patients regardless of initial arrythmia type (III), 152 ST-elevation myocardial infarction patients (I), 3568 hypoglycaemic patients without diabetes from (II), and 27141 patients with all parameters to include in National Early Warning Score and a blood glucose measurement (IV). Results: Hyperglycaemia was common (76-93%) and hypoglycaemia less (2-11%), depending on the study design (I-IV). Of those, diabetes was not previously diagnosed in 62-71% (I, III, and IV). A greater change in blood glucose level during prehospital phase in ST-elevation myocardial infarction patients was associated with 30-day mortality (non-survivors +1.2±5.1 vs survivors -0.3±2.4 mmol/l [mean±SD], p=0.03) (I). When diabetes was not considered alcohol [41%, (CI95% 40-43)] was the most possible cause of hypoglycaemia episode encountered by emergency medical service (II). In out-of-hospital cardiac arrest patients hyperglycaemia was common (11.2 mmol/l, IQR 8.8-15.7) showing a decrease of 2.2 mmol/l, IQR -3.6 to -0.2 in blood glucose level from prehospital to hospital admission blood sampling 96 min (IQR 85-119) (III). Insulin (10.1 mU/l, IQR 4.2-5.2) and glucagon 141 ng/l, IQR 105-240 levels were low, and glucagon-like-peptide 1 level increased 2-8- fold (6.3 ng/ml, IQR 5.2-9.0) from fasting level (III). All studied biomarkers had great interindividual variation, and this was not associated with a change in blood glucose level (III). Blood glucose improved National Early Warning Score discrimination capability for risk of death at 24-hours and at 30-days Hyperglycaemia: 24-hour risk of mortality: OR 1.54, (1.11-2.12) and 30-day risk of mortality: OR 1.41, (1.20-1.66) and hypoglycaemia: 24-hour risk of mortality: OR 5.46, (2.87-9.64) and 30-day risk of mortality: OR 2.33, (1.47-3.52)  (IV). The association was improved according to likelihood ratio tests (p
Original languageEnglish
  • Kuisma, Markku, Supervisor
  • Nurmi, Jouni, Supervisor
Place of PublicationHelsinki
Print ISBNs978-951-51-5332-6
Electronic ISBNs978-951-51-5332-3
Publication statusPublished - 2019
MoE publication typeG5 Doctoral dissertation (article)

Bibliographical note

M1 - 74 s. + liitteet

Fields of Science

  • Blood Glucose
  • Mortality
  • Morbidity
  • ST Elevation Myocardial Infarction
  • Hyperglycemia
  • Hypoglycemia
  • Biomarkers
  • Glucagon
  • Glucagon-Like Peptide 1
  • Hydrocortisone
  • Glycated Hemoglobin A
  • Insulin
  • Diabetes Mellitus
  • Homeostasis
  • Insulin Resistance
  • Out-of-Hospital Cardiac Arrest
  • Emergency Medical Services
  • Critical Illness
  • Risk Assessment
  • 3126 Surgery, anesthesiology, intensive care, radiology

Cite this