| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
Metabolism, Endocrinology and Diabetes Division Department of Internal Medicine University of Michigan Ann Arbor, Michigan 48109
Address all correspondence and requests for reprints to: Elif Arioglu Oral, M.D., Assistant Professor of Medicine, Division of Endocrinology, Metabolism and Diabetes, University of Michigan, Ann Arbor, Michigan 48109. E-mail: eliforal{at}umich.edu.
Hyperglycemia and insulin resistance are hallmarks for the adaptive response in adult human organisms when survival is at stake due to critical illness (1, 2). These metabolic alterations are believed to be occurring acutely to ensure delivery of critical nutrients to the brain and other key organs to ensure the best chance of survival. An important question that remains unanswered is whether or not the beneficial effects of insulin treatment can be applicable to pediatric or geriatric age groups. The article written by van Waardenburg et al. (3) in this issue of the Journal begins to tackle this question. As such, this article opens a new door for detailed metabolic investigations in critical pediatric illness. The authors have chosen to study meningococcal septic shock and meningococcal sepsis without shock in a university-based intensive care unit (ICU) setting. They have included 10 patients with shock and six patients without shock, and evaluated the change in glycemia every 8 h within the first 72 h of admission to the ICU with the respective diagnoses. In addition, they studied an array of circulating hormones and factors known to be related to the acute inflammatory response at the end of 72 h (3). Given the scarcity of previous homogeneous data (4), this report, while quite preliminary in nature, sets the stage for more detailed studies. It also prepares a springboard for discussion and calls to design important new studies. The strongest points of the data presented include establishment of hyperglycemia in this distinct relatively homogenous and relatively hemodynamically stable population beginning within the first hours of admission and lasting through 72 h. Interestingly and in contrast to many adult populations with critical disease, the authors documented low insulin as well as low leptin levels in the setting of hyperglycemia in their unique study group with meningococcal septic shock. Other key findings that also take a first stab at explaining the potential mechanisms of adaptive responses in critically ill children include the following:
1) Positive correlation between plasma insulin and glucose levels which suggest that the pancreatic islets of the studied children are making an effort to compensate for the rising glucose levels. Taking this piece of information together with the presence of hyperglycemia, insulin deficiency, or the inability of the pancreatic islets to secrete insulin appears to be a more dominant result of meningococcal sepsis, especially in patients with circulatory failure.
2) Elevated plasma triglyceride levels with normal nonesterified fatty acids levels, suggesting a disproportionate increase in hepatic lipogenesis more than an increased level of lipolysis.
3) Elevated plasma levels of sTNF-R55 and 75 as well as sIL-1R2, C-reactive protein, and IL-6, which correlate well with the clinical severity of illness scores.
It remains unclear whether the inflammatory mediators studied in this article play a direct role in the etiology of hyperglycemia in critically ill children. The key point is to identify earlier phenomena, in an effort to determine the overall effects of modulating the activity of inflammatory mediators and to use this knowledge to develop targeted interventions. In addition to the putative effects of elevated proinflammatory cytokine levels, the authors provided alternative explanations for what seems to be a selective impairment of insulin secretion associated with the severity of acute meningococcal septic shock in children. The role of catecholamines on either systemic or splanchnic circulation could influence the degree of insulin secretion in these critically ill patients. Insulin secretion may also be affected by changes in gastrointestinal hormone levels such as glucagon-like peptide-1 and glucose dependent insulinotropic polypeptide (5).
Reviewing the literature on adults, the occurrence of hyperglycemia means that there is relative insulin deficiency that is unable to meet the peripheral demand. Although the initial phase of insulin resistance, hyperglycemia, and relative insulin deficiency may be beneficial, prolongation of these metabolic changes beyond a crucial time period appears to lead to deleterious effects. The duration of that threshold period may in fact be quite short. The landmark Leuven trial showed that in a cardiothoracic surgical ICU, reversal of hyperglycemia with the use of an intensive insulin infusion protocol proved to reduce mortality and morbidity and improve overall survival after cardiac surgery (2). This trial and the hope of improving outcome from critical illness, which remains on average around 50% at best estimates in the hospitalized ICU patients, fueled old scientific interest that dates back to the Claude Bernard era toward increasing our understanding of the impact of metabolic adaptive processes in critical illness.
Although the positive impact of the insulin therapy was shown over and over again in surgical settings in adult patients after the Leuven trial, the effects in more complicated medical critical illness appear to be more complex (1, 6, 7). The exact mechanisms of any observed benefit also remain elusive. Whether the critical component of therapy is reversal of hyperglycemia, correction of relative insulin deficiency, or overcoming aberrant signaling driven by impaired insulin signaling in insulin-dependent tissues still remains to be identified (8). More than likely, the currently adopted approach of early intensive insulin infusion protocol in the cardiothoracic and surgical ICUs tackles all three problems that are intricately connected.
While reviewing the findings of the paper by van Waardenburg et al., some important questions need to be considered. Why would insulin resistance not develop immediately in the face of critical illness in children? Do the pancreatic ß-cells in children with meningococcal sepsis acutely reduce insulin production, whereas in adults this capability is enhanced? If insulin production is acutely inhibited, why would such a mechanism occur?
Finding a teleological answer for the first question is easier. If the observations reported in this article are replicated by future studies, an important metabolic lesson may in fact be underscored for us: adults and children may be adapting quite differently to critical illness. After all, this lesson may not necessarily be so surprising because the energy investments of a not-developing human being are in fact quite different than the fully developed adult. The majority of the total energy expenditure of a healthy child between ages 2 and 4 is made toward growth and development, in contrast to less than 10% expanded for maintenance needs (9). It is also important to remember that growth and development can be halted if conditions are not optimal as observed in short stature of chronic medical illness. Thus, putting a break on the GH axis as demonstrated in many settings as well as in patients studied in the current article would lead to significant and perhaps sufficient energy conservation that would meet the basal requirements of critical organs without the need to redirect peripheral glucose utilization further. In other words, development of GH resistance in the face of severe systemic disease negates the need for peripheral insulin resistance, which appears to be the dominant adaptive metabolic change in the fully developed adult organism.
To answer whether children with meningococcal sepsis or other critical illness acutely reduce insulin production as an adaptive response or whether this occurs as a direct effect of a specific critical illness such as the meningococcal sepsis, future studies will have to be undertaken to study critically ill children for longer periods into their critical illness and also after recovery. It is possible that the systemic cytokines such as the ones studied in this article directly affect insulin secretion from the ß-cells without leading to ß-cell death (10, 11, 12). Also, the ß-cell reserve of a 3 yr old may not reach its maximal potential and is not able to mount the same response that a 30 yr old is able to mount. Also, other circulatory factors that regulate insulin secretion such as incretins can be studied fairly easily. It would be important to have a definite nutritional protocol as well as stratification of baseline weight for these latter studies because both body weight and nutritional factors alter the regulation of incretins. There are other potential mediators of insulin secretion such as neural factors and other hormonal regulators. The authors have evaluated a number of factors such as glucagon, catecholamine levels, and cortisol. However, the current study is a single snapshot; more detailed studies on secretory pattern and rhythm as well as regulation of secretion need to be undertaken. Also, investigation of many other hormonal factors not included in this study may be warranted. For example, circulating aldosterone levels were found to be unexpectedly low in children with acute meningococcal disease in 2004 by Lichtarowicz-Krynska et al. (13). Although researching neuroendocrine regulation is appealing for endocrinologists, other possibilities such as a direct infectious process in the ß-cells need to be ruled out.
Another point from the article by van Waardenburg et al. (3) that needs to be placed in perspective is the interpretation that circulating leptin levels measured from the patients studied in this study were low. Although there are well-established percentile levels for each gender, various body mass index (BMI) levels, and different ethnicities in adults (14), such detailed information is lacking in young children. Furthermore, for the adults, we do know that circulating leptin levels below 2 ng/dl are often associated with a constellation of metabolic as well as neuroendocrine and immunological aberrations that permit us to define true states of leptin deficiency (15). We simply do not have strong data for this definition in very young children in whom body adiposity is typically much lower than adult levels. In the current article, no information on the prior nutritional status of subjects including BMI (or BMI percentiles) was provided. Likewise, no simultaneous comparisons to healthy controls or other disease states were made while under similar enteral feeding regimen.
There is an accumulating line of evidence that focuses on the role of tissue macrophages on adipose tissue function and plasticity (16). These tissue macrophages are hypothesized to play an important function in orchestrating many functions of the adipose tissue including endocrine function. Thus, it may very well be conceivable that a generalized multisystemic inflammatory reaction would result in the direct inhibition of one or more adipocyte-derived hormones.
The authors rightfully shy away from a causality relationship between the low leptin levels and insulin levels. The literature on the direct effect of leptin on insulin secretion is quite controversial. There are in vitro islet perifusion studies and culture system reports indicating that leptin may directly inhibit insulin secretion from the islets in these systems (17). Rodent data from ob/ob mice and from Zucker rats suggest that in vivo administration of leptin may permit reversal of lipotoxicity on pancreatic islets and cause improved insulin secretion (18). Based on the information from children with extremely low levels of leptin, one would predict predominant hepatic insulin resistance coupled with peripheral insulin resistance in the setting of leptin deficiency. In addition, hepatic lipogenesis would also be dysregulated. The current data show mildly elevated plasma triglyceride concentrations and possibly dysregulated hepatic lipogenesis in the studied patients.
In conclusion, the work from the University Hospital Maastricht in The Netherlands is important in directing attention to the metabolism in pediatric critical illness. Yet, metabolic adaptation in very young children in the face of life-threatening illness cannot be fully understood without studying many related factors in much greater detail. The impact of GH resistance as well as regulation of dynamic GH secretion has to receive due attention. Other metabolic regulators such as insulin and leptin may prove to be important as speculated in the current paper; however, significantly more data are required before we can accept these conclusions. On the therapeutic front, current information only calls for caution while considering intensive insulin therapy in the pediatric ICUs. If such an approach will be undertaken, it should be done after careful consideration of the peripheral insulin responsiveness. Although metabolic modulation offers an exciting and currently unforeseen degree of potential impact in the ICU world, it is only logical that these therapies would need to be individualized for various age groups and perhaps genders. After all, in medicine one therapy rarely fits all.
Footnotes
Abbreviations: BMI, Body mass index; ICU, intensive care unit.
Received July 17, 2006.
Accepted July 19, 2006.
References
and interferon-
as growth factors to the human fetal ß-cell. J Clin Endocrinol Metab 73:1044–1050[Abstract]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
