Title: Tight glucose control with insulin in critically ill children: from acute endocrine mechanisms to long-term neurocognitive development
Other Titles: Strikte glycemiecontrole met insuline bij kritiek zieke kinderen: van acute endocriene mechanismen naar latere impact op neurocognitieve ontwikkeling
Authors: Gielen, Marijke
Issue Date: 25-Sep-2012
Abstract: <w:latentstyles deflockedstate="false" defunhidewhenused="true"  <w:lsdexception="" locked="false" priority="0" semihidden="false" critical="" illness="" is="" hallmarked="" by="" a="" wide="" array="" of="" metabolic="" and="" endocrine="" disturbances,="" among="" which="" severe="" dysregulation="" glucose="" homeostasis="" with="" subsequent="" hyperglycemia.="" although="" this="" “stress="" hyperglycemia”="" or="" “diabetes="" injury”="" has="" long="" been="" interpreted="" as="" beneficial="" response="" to="" the="" stress="" imposed="" any="" type="" illness,="" it="" in="" fact="" associated="" adverse="" outcome.="" landmark="" studies="" our="" research="" group="" demonstrated="" that="" targeting="" normal="" fasting="" blood="" levels="" intensive="" insulin="" therapy="" reduced="" morbidity="" mortality="" adult="" critically="" ill="" patients.="" however,="" intervention="" concomitantly="" increased="" incidence="" hypoglycemia,="" complication="" feared="" for="" treatment="" diabetes.="" subsequently,="" showed="" that,="" like="" adults,="" tight="" control="" pediatric="" since="" age="" are="" lower="" infants="" children="" than="" adults="" thus="" closer="" threshold="" not="" surprising="" increase="" hypoglycemia="" was="" even="" more="" pronounced,="" 25%="" patients="" experiencing="" at="" least="" one="" episode="" hypoglycemia. Critical illness is hallmarked by a wide array of metabolic and endocrine disturbances, among which a severe dysregulation of glucose homeostasis with subsequent hyperglycemia. Although this “stress hyperglycemia” or “diabetes of injury” has long been interpreted as a beneficial response to the stress imposed by any type of critical illness, it is in fact associated with adverse outcome. Landmark studies by our research group demonstrated that targeting normal fasting blood glucose levels with intensive insulin therapy reduced morbidity and mortality in adult critically ill patients. However, the intervention concomitantly increased the incidence of hypoglycemia, a complication feared for as in the treatment of diabetes. Subsequently, our group showed that, like in adults, tight glucose control with intensive insulin therapy reduced morbidity and mortality in pediatric critically ill patients. Since normal for age fasting blood glucose levels are lower in infants and children than in adults and thus closer to the threshold for hypoglycemia, it is not surprising that the increase in hypoglycemia with this intervention was even more pronounced, with 25% of the patients experiencing at least one episode of hypoglycemia. The central hypothesis of this PhD project puts forward that the short-term benefit of targeting age-adjusted normoglycemia with intensive insulin therapy in critically ill children during their stay in the intensive care unit (ICU) was obtained in part by an improvement of the endocrine profile of these children, without adverse effects on their neurocognitive development at long-term.In a first part, we focused on a potential stimulation of anabolism by intensive insulin therapy as a mediator of its beneficial effect on outcome. In this regard, both the growth hormone/insulin-like growth factor-I (GH/IGF-I) and thyroid axes are known to be suppressed during critical illness and have been linked to the hypercatabolic state of this condition. Although insulin administration has been shown to restore the disturbances in these axes in diabetes, we previously observed that tight glucose control with intensive insulin therapy failed to reactivate these axes in adult critically ill patients. We now hypothesized that relative underfeeding of adult critically ill patients as compared with infants and children may have prevented a reactivation of the GH/IGF-I axis and thyroid axis in adults, and that tight glucose control would still be able to stimulate these axes in pediatric critically ill patients. Unlike expected, however, this intervention further accentuated the suppression of both the GH/IGF-I and thyroid axes. Indeed, in a first study, where we thoroughly investigated the GH/IGF-I axis, we demonstrated that this intervention increased circulating growth hormone, but lowered bioavailable IGF-I, IGFBP-3 and the acid-labile subunit on day 3 of critical illness. The urea/creatinine ratio as a marker of catabolism was increased. We next considered that these observations may be explained by the increased incidence of hypoglycemia in the tight glucose control group, by induction of a counterregulatory response to reduce the glucose-lowering effect of IGF-I. However, in a carefully matched nested case-control study, we excluded this possibility. In a second study, we focused on the thyroid axis. Tight glucose control further accentuated the peripheral inactivation of thyroid hormone, expressed by a lowering of the T3/reverse T3 ratio or an exacerbation of the “low T3 syndrome”. Strikingly, this counterintuitive effect statistically explained at least part of the clinical outcome benefit of the intervention. The fact that “early catabolism”, with low bioavailable IGF-I and a low T3/reverse T3 ratio as seen in fasting, may provide benefit in the ICU is further supported by recent evidence on outcome improvement by early nutrient restriction in the adult ICU. In the pediatric ICU, the age-adjusted target for normoglycemia may have lowered blood glucose to the extent that it signaled through protective catabolic fasting pathways. In a second part, we assessed the impact of tight glucose control on the three domains of blood glucose dynamics (hyperglycemia, hypoglycemia and blood glucose variability) during ICU stay and investigated their independent relation to the survival benefit observed with tight glucose control. This third study showed that only avoidance of hyperglycemia, defined as higher than the age-adjusted normal fasting range, appeared to be the key factor in bringing about the outcome benefit of tight glucose control. Neither hypoglycemia, always quickly and effectively managed with prevention of rebound hyperglycemia, nor blood glucose variability were associated with increased mortality.In a third part, we investigated whether tight glucose control with intensive insulin therapy, which clearly improved short-term ICU outcome, does not have adverse effects on long-term. This question is definitely crucial for the brain, especially in children, as they are undergoing critical brain development. Children may therefore be particularly vulnerable to potential adverse effects on the brain evoked by hypoglycemia, often blamed for the neurocognitive deficits observed in patients with diabetes. However, also hyperglycemia could be harmful to the developing brain. Our fourth study thus focused on potential brain damage as assessed by serum biomarkers for damage to astrocytes (S100B) or neurons (neuron-specific enolase, NSE) during ICU stay. These markers were not altered by tight glucose control as compared with tolerating hyperglycemia. A nested case-control study further showed that these markers did not increase after a hypoglycemic event. In contrast, from ICU admission onward, patients experiencing hypoglycemia at any time during ICU stay revealed higher S100B and NSE levels than those who would not experience hypoglycemia. This suggests an increased incidence of hypoglycemia in more severely ill patients, rather than that hypoglycemia would be a risk for worse outcome on its own. Finally, we performed a fifth, long-term neurocognitive follow-up study, in which all 700 children originally included in the randomized controlled trial on tight glucose control were assessed four years after randomization. This study clearly revealed that tight glucose control during pediatric intensive care had not increased the incidence of very poor outcomes (death or severe disability) and had no adverse impact on neurocognitive development. Indeed, the intervention did not affect full scale IQ, other scores for intelligence, visual-motor integration, memory nor behavior as compared with usual care. Tight glucose control even improved motor coordination and cognitive flexibility, the latter up to the level of healthy children. Importantly, a separate analysis in a propensity score-matched subset of patients showed that brief hypoglycemia evoked by tight glucose control was not independently associated with worse neurocognitive outcome. Taken together, tight glucose control to normal for age fasting blood glucose levels in critically ill infants and children improved short-term outcome without compromising long-term survival and neurocognitive development assessed four years after PICU admission, despite the higher incidence of hypoglycemia. The short-term outcome benefits could not be explained by a reactivation of anabolism. On the contrary, the “early catabolism” accentuated by tight glucose control seemed to explain at least part of the outcome benefit observed with the intervention. The benefits of tight glucose control can probably only be achieved by targeting normal for age fasting blood glucose levels, which inherently increases the risk of hypoglycemia. However, when blood glucose levels are accurately monitored, hypoglycemia quickly and effectively managed, and insulin consequently titrated, potentially life-saving efforts to avoid hyperglycemia in critically ill children should not be abandoned solely because of the risk of brief hypoglycemia, especially if these episodes of hypoglycemia do not harm long-term neurocognitive development. Nevertheless, large multicentre studies targeting normal for age fasting blood glucose levels in critically ill infants and children with accurate glucose monitoring systems are needed to confirm the acute and long-term results of the initial proof-of-concept study in the Leuven pediatric ICU.
ISBN: 978 94 6165 063 4
Publication status: published
KU Leuven publication type: TH
Appears in Collections:Laboratory for Intensive Care Medicine (-)
Laboratory of Intensive Care Medicine

Files in This Item:

There are no files associated with this item.

Request a copy


All items in Lirias are protected by copyright, with all rights reserved.