Glycaemic variability and its association with enteral and parenteral nutrition in critically ill ventilated patients

Abstract

Background & aims: Extremes of dysglycaemia as well as glycaemic variability are associated with excess mortality in critically ill patients. Glycaemic variability is an increasingly important measure of glucose control in the intensive care unit (ICU) due to this association; however, there is limited data pertaining to the relationship between exogenous glucose from nutrition and glycaemic variability and clinical outcomes. The primary aim of this study was to determine if glycaemic variability is associated with an increase in mortality. Secondary objectives were to investigate any factors affecting glycaemic variability, and to characterise the role nutrition, particularly carbohydrate, plays as a contributing factor to glycaemic variability and other clinical outcomes (duration of ventilation and ICU length of stay). Methods: Data on patients in a combined medical/surgical tertiary Australian Intensive Care Unit (ICU), ventilated for >24 h and exclusively fed by artificial nutrition support was extracted from a clinical database of prospectively collected information over an 18 month period. Glycaemic variability was defined as the coefficient of variation (GV; standard deviation/mean of blood glucose levels x 100). Statistical analysis was performed using logistic regression, zero-truncated negative binomial and linear regression as appropriate to the distribution of the outcome variable using R software. Results: Data on up to 759 subjects was available. The average age of the study cohort was 56.9 years with a mean (standard deviation) APACHE III score of 72 (28). 66% of the study subjects were male. Glycaemic variability was associated with an increase in mortality (odds ratio 1.02; 95% CI: 1.00–1.04, p = 0.03). Factors associated with glycaemic variability included Acute Physiology and Chronic Health Evaluation III score (0.09, 0.06–0.11, p < 0.001), being male (−1.67, −2.97 to −0.38), p = 0.01) and mean units of insulin per day (0.08, 0.06–0.09, p < 0.001). There was no effect of any nutritional factor on glycaemic variability. Further exploratory analyses though showed that for those patients who required insulin during ICU admission, increased insulin dose was associated with increasing carbohydrate (incidence rate ratio (IRR) 1.003, 1.001–1.005, p = 0.001). Mean daily carbohydrate provision (grams) was associated with an increase in ventilation hours (IRR, 95% CI: 1.009, 1.008–1.009, p < 0.001) and length of intensive care unit stay (IRR, 95% CI: 1.007, 1.006–1.008, p < 0.001). Conclusion: This study confirms that GV was associated with excess mortality. Furthermore, administration of increasing doses of insulin was associated with increased GV. Increased carbohydrate intake was associated with an increased insulin requirement, as well as increased duration of mechanical ventilation and ICU length of stay. These findings provide important context for further prospective trials investigating the effect of carbohydrate provision in mechanically ventilated critically ill patients requiring artificial nutritional support.