Insulin-Treated Diabetes Reduces Dysglycemia-Related Mortality in Sepsis
Findings provide a rationale for an insulin-treated diabetes-specific approach to dysglycemia in patients with sepsis admitted to the ICU.
Patients with sepsis and a pre-existing diagnosis of insulin-treated diabetes (ITD) may show a different relationship between hospital mortality and highest glucose levels and glycemic variability in the first 24 hours than patients without ITD, according to a study published in the Journal of Critical Care.
In Australia and New Zealand, 11% to 15% of intensive care unit (ICU) admissions are comprised of patients with sepsis. Most patients with sepsis experience stress-induced dysglycemia (high peak glucose levels, elevated mean glucose levels, increased hypoglycemia or greater glucose variability), which has been associated with increased mortality. In contrast, the hyperglycemic threshold associated with mortality in patients with diabetes is unknown as no association between dysglycemia and mortality has been reported. This may be due to a biological adaptation to chronic hyperglycemia and glycemic variability. However, sepsis induces powerful metabolic derangements, which may cause the relationship between glycemia, diabetes, and outcome to be altered.
Using the Australia and New Zealand Intensive Care Society Adult Patient Database, in this observational cohort study researchers investigated the relationship between dysglycemia and hospital mortality in patients with and without a preadmission diagnosis of ITD between 2006 and 2015. A total of 90,644 septic patients including 5127patients diagnosed with ITD were included. The study investigated 4 domains of dysglycemia (highest, mean, and lowest blood glucose levels and glycemic variability: the absolute difference between the highest and lowest levels). The association between a preadmission diagnosis of ITD and hospital mortality in each domain was analyzed.
Results showed that patients in the ICU with sepsis and ITD have lower adjusted hospital mortality, with higher peak blood glucose levels in the first 24 hours while patients who do not have ITD have increased mortality (interaction P =.012). In the lowest quintile (<6.7 mmol/l), patients with ITD had the highest mortality rates while patients who did not have ITD had the lowest. Conversely, in the highest quintile (>11.6 mmol/l), patients with ITD had the lowest mortality while patients who did not have ITD had the highest. In addition, there was a 5% absolute mortality difference between patients with and without ITD in the highest quintile (21.4% vs 16.4%; P <.001). As early mean blood glucose and early lowest blood glucose levels increased, patients with ITD demonstrated a lower hospital mortality trend, while the opposite was seen in patients without ITD (P =.07 vs P =.11, respectively). Hypoglycemia, with a nadir blood glucose level of <5.1 mmol/l in the first 24 hours of ICU admission, was associated with an increased hospital mortality in both the ITD (24.4%) and non-ITD (24.5%) groups.
Finally, results showed replication of this significant difference when assessing glycemic variability (P =.048). Mortality increased in patients without ITD as early glycemic variability increased. However, no influence between increasing glycemic variability and mortality in patients with ITD was reported, even though patients with ITD had significantly higher early glycemic variability than patients who did not have ITD (54.2% vs 21.9% in highest quintile; P =.046).
Overall, findings imply a significant interaction between the presence of ITD and the relationship of dysglycemia with outcome, as ITD seems to attenuate injurious effects. The investigators concluded that “findings provide a rationale for an ITD-specific approach to dysglycemia management.”
Magee F, Bailey M, Pilcher DV, Mårtensson J, Bellomo R. Early glycemia and mortality in critically ill septic patients: Interaction with insulin-treated diabetes. J Crit Care. 2018;45:170-177.