Metabolic interventions in critically ill patients: impact on outcome in relation to ketogenesis and hypoglycemia

Major advances in intensive care medicine have greatly improved the outcome of critically ill patients, with the majority of patients surviving the acute insult. Despite a decreasing mortality rate, a considerable number of patients progress into a prolonged phase of critical illness, further requiring supportive care in order to survive. This persistent phase of critical illness contains a substantial mortality risk, driven by persistent multi-organ failure or severe ICU-acquired weakness. Therefore, the focus of intensive care specialists and researchers has shifted towards interventions that can enhance recovery or prevent complications that prolong critical illness. In this regard, during the last 2 decades, our research group has identified 2 metabolic interventions with such potential. Tight blood glucose control with intensive insulin therapy and tolerating a macronutrient deficit by withholding early PN both improved the outcome of critically ill patients. Furthermore, there is recent evidence that also ketone bodies may play a protective role during critical illness. Although ketogenesis has traditionally been reported as blunted during critical illness, there is recent evidence suggesting that a 12-h nutrient interruption can result in a ketogenic fasting response in critically ill patients. Critical illness is characterized by hyperinsulinemia and hyperglycemia, which may both suppress ketogenesis. Therefore, interventions that can decrease insulin and glucose levels may theoretically activate ketogenesis. Tight glucose control decreases the level of hyperglycemia but a higher a dose of insulin is used. However, during critical illness, insulin resistance in the liver, which is main site of ketone body production, is profound and difficult to overcome with insulin therapy. Therefore, the net impact of strictly controlling blood glucose levels, with intensive insulin therapy, on ketogenesis during critical illness remains a question. Besides tight glucose control, withholding early PN could theoretically also stimulate ketogenesis by reducing nutritional intake. However, late PN corresponds with a macronutrient restriction and not with full fasting. Consequently, the impact of this intervention on endogenous ketone body production also remains to be investigated. Both metabolic interventions not only reduce the level of hyperglycemia but also increase the risk of hypoglycemia, which has been associated with impaired outcome of critically ill patients. Especially the developing brain of critically ill children may be vulnerable to hypoglycemia. The general aim of this PhD project was to gain more insight in the impact of metabolic interventions on ketogenesis during critical illness and to investigate if ketone bodies may play a protective role in the outcome benefit of these interventions. In addition, we investigated if hypoglycemia independently associates with impaired short- and long-term outcome of critically ill children, and if such potential association could be affected by the timing of PN and the glucose control strategy. In a first and second objective, we investigated the effect of withholding early PN on ketogenesis. We demonstrated that late PN, as compared to early PN, increased 3HB levels in critically ill children and adults. Furthermore, this enhanced ketogenesis statistically mediated, at least partly, the beneficial impact of late PN on outcome of critically ill children. Conversely, 3HB did not associate with improved outcome of critically adults. However, the rise in ketone body concentrations was also much less pronounced than in critically ill children. Also in healthy individuals, the ketogenic response is known to be more pronounced in children than in adults. There are several suggested mechanisms whereby elevated levels of ketone bodies may enhance recovery of critically ill children. First, ketone bodies can serve as a super fuel for the brain, heart and skeletal muscle during physical stress. Besides its role as energy substrate, ketone bodies can exert several signaling functions and may activate pathways that are beneficial for recovery. Given that enhanced ketogenesis may have mediated part of the outcome benefit of late PN in critically ill children, strategies enhancing ketone body concentrations may mimic the beneficial impact of late PN on outcome. In a third objective, we assessed the effect of tight blood glucose control on ketone body concentrations in critically ill patients, in a context of early PN. Circulating 3HB concentrations were not affected by this metabolic intervention and stayed low throughout the ICU stay as well for critically ill children as adults. This suggests that the protective effects of tight glucose control in this context were not mediated by increased ketone availability. Whereas tight blood glucose control had no impact on ketogenesis in critically ill patients receiving early PN, it remains a question if ketogenesis can be enhanced by tight glucose in patients not receiving PN during the first week in the ICU. In this context, a large multicenter RCT, the TGC-fast study is currently ongoing. In this study, the impact of tight glucose control with intensive insulin therapy as compared to liberal glucose control on outcome of critically ill patients, not receiving supplemental PN until after the first week in the ICU, is investigated. In addition to the different study outcomes, the effect on ketogenesis will be studied by quantification of ketone bodies on collected blood samples. The beneficial metabolic interventions that are integrated in the TGC-fast study, late PN and tight glucose control, both increase the risk of hypoglycemia. As hypoglycemia during ICU stay has been associated with impaired outcome, it remains a question if tight glucose control is still beneficial in patients not receiving early PN. In the fourth objective we found that the occurrence of hypoglycemia in critically ill children independently associated with impaired executive functioning 4 years after randomization, irrespective of the nutritional strategy and the glucose control protocol. However, adjusting multivariable models for classical baseline risk factors may not fully capture the baseline risk of impaired neurocognitive outcome. In this context, a previous study reported elevated makers of neurological damage, S100B and NSE, upon ICU admission in patients with subsequent hypoglycemia, whereby these markers did not further increase after the hypoglycemic event. Therefore, it may be possible that additional adjustment for these baseline markers of neurological damage, NSE and S100B, may change our outcome results. In a sensitivity analysis, only episodes of spontaneous or recurrent hypoglycemia independently associated with worse neurodevelopmental outcome after 4 years. In conclusion, we demonstrated that withholding early PN enhanced ketogenesis in critically ill patients not receiving early PN. Activation of ketogenesis was most pronounced in critically ill children, in whom it statistically mediated part of the outcome benefit of the intervention. These data open perspectives for further investigating the impact of exogeneous ketone supplementation or ketogenic diets on outcome of critically ill patients. Hypoglycemia in ICU independently associated with worse neurocognitive outcome after 4 years, although additional analyses correcting for baseline markers of neurological damage are pending. Although it remains unclear whether a brief episode of iatrogenic hypoglycemia is harmful or not, it seems prudent to maximally prevent and adequately treat hypoglycemia.

Jaar van publicatie:2022

Toegankelijkheid:Open