Wilgen, Urs2009-02-132009-02-132009-02-13http://hdl.handle.net/10539/6105Abstract Mortality of patients requiring intensive care treatment for greater than 5 days has been shown to be about 20% worldwide. Hyperglycaemia is common in critically ill patients. Strict glucose control with insulin in critically ill patients was shown to reduce mortality and morbidity significantly. Several interrelated mechanisms are involved in the development of “stress hyperglycaemia” in critically ill patients. These include dextrose containing intravenous infusions and total parenteral nutrition; the counter regulatory hormones (catecholamines, cortisol, glucagon and growth hormone) which oppose the effects of insulin; nervous system signaling; increased insulin clearance; and excess production of cytokines that interfere with intracellular insulin signaling pathways. Aim of study: To determine if the cytokines TNFα, IL-6 and adiponectin are significant determinants of insulin resistance in critically ill patients. Methods: The study was a prospective observational study conducted in the intensive care unit (ICU) at the Chris Hani Baragwanath hospital. Forty sequential adult ICU admissions that met with the inclusion criteria were enrolled. Blood specimens were drawn for adiponectin, TNF, and IL-6 at the time of ICU admission, on day 3, day 7 and on discharge from the ICU. Demographic data and clinical data were recorded, and body mass index (BMI) and APACHE II scores were calculated on admission. Blood glucose was measured every 2 to 4 hours, recorded and a mean value was calculated over the 24 hour period. Insulin infusions were started when the blood glucose values exceeded 6.0mmol/l. Administration of insulin was according to a fixed sliding scale. The total amount of insulin administered intravenously over that 24 hour period was recorded. Other factors known to be related to insulin sensitivity, such as inflammation (as indicated by C-reactive protein), vii triglycerides, insulin, C-peptide and cortisol levels were also drawn in addition to the blood drawn for routine investigations. Results: Duration of stay in ICU correlated with severity of illness as assessed by the APACHE II score (r = 0.44, p = 0.004). There was no significant difference in the mean 24 hour plasma glucose concentration throughout the duration of stay in ICU, there were however significant differences in the amount of insulin administered to maintain normoglycaemia. The amount of administered insulin required was found to peak on day 3 and decline thereafter. The main determinant of insulin administered was mean glucose (r = 0.79, p < 0.00001). The measured insulin concentrations on admission correlated with mean plasma glucose (r = 0.41, p = 0.009) and C-peptide (r = 0.45, p = 0.004) levels. The main determinants of mean plasma glucose levels on admission were BMI (r = 0.38, p = 0.013) and serum cortisol (r = 0.41, p = 0.008) levels. Serum triglycerides levels showed a significant difference from admission to discharge, with values increasing from admission levels. Adiponectin levels showed a significant increase from admission to discharge. IL-6 levels showed a significant decrease. TNFα levels did not show statistically significant changes. No statistically significant correlations were found between the levels of TNFα or IL-6 and administered insulin. Adiponectin concentrations showed a negative correlation with amount of administered insulin on discharge (r = -0.457, p = 0.0049). There were significant gender differences in BMI, administered insulin on admission, serum cortisol and C-peptide concentrations, with females having higher values than males. BMI was shown to account for the gender differences in administered insulin and C-peptide levels. viii There were significant differences in IL-6 and TNFα concentrations between the survivors and nonsurvivors, with higher levels being seen in non-survivors. Adiponectin levels were lower in nonsurvivors, but this did not reach statistical significance. Conclusion: Although there was a demonstrable change in insulin sensitivity during the stay in ICU, there was no statistically significant association between the cytokines TNFα or IL-6 and insulin administration. There was a negative correlation between adiponectin concentrations and administered insulin on discharge. This data also demonstrates that mortality is associated with increased levels of proinflammatory cytokines.eninsulincytokinesICUcritical illnessThesis