Severely burned patients typically experience a systemic response expressed as increased metabolism, inflammation, alteration of cardiac and immune function, and associated hyperglycemia. Hyperglycemia has been associated with an increased risk of morbidity and mortality in critically ill patients. Until recently and for many years, hyperglycemia has been expectantly managed and considered a normal and desired response of an organism to stress. However, findings reported from recent studies now suggest beneficial effects of intensive insulin treatment for critically-ill patients. The literature on the management of hyperglycemia in severely burned patients is sparse, with most of the available studies involving only small numbers of burned patients. The purpose of this article is to describe the pathophysiology of hyperglycemia following severe burns and review the available literature on the outcome of intensive insulin treatment and other anti-hyperglycemic modalities in burned patients in an evidence-based-medicine approach.

IntroductionIntensive insulin treatment (IIT) has been shown to improve outcomes post-burn in severely burnt patients. However, it increases the incidence of hypoglycemia and is associated with risks and complications. We hypothesized that exenatide would decrease plasma glucose levels post-burn to levels similar to those achieved with IIT, and reduce the amount of exogenous insulin administered.MethodsThis open-label study included 24 severely burned pediatric patients. Six were randomized to receive exenatide, and 18 received IIT during acute hospitalization (block randomization). Exenatide and insulin were administered to maintain glucose levels between 80 and 140 mg/dl. We determined 6 AM, daily average, maximum and minimum glucose levels. Variability was determined using mean amplitude of glucose excursions (MAGE) and percentage of coefficient of variability. The amount of administered insulin was compared in both groups.ResultsGlucose values and variability were similar in both groups: Daily average was 130 ± 28 mg/dl in the intervention group and 138 ± 25 mg/dl in the control group (P = 0.31), MAGE 41 ± 6 vs. 45 ± 12 (respectively). However, administered insulin was significantly lower in the exenatide group than in the IIT group: 22 ± 14 IU patients/day in the intervention group and 76 ± 11 IU patients/day in the control group (P = 0.01). The incidence rate of hypoglycemia was similar in both groups (0.38 events/patient-month).ConclusionsPatients receiving exenatide received significantly lower amounts of exogenous insulin to control plasma glucose levels. Exenatide was well tolerated and potentially represents a novel agent to attenuate hyperglycemia in the critical care setting.Trial registrationNCT00673309.

Background Anesthetics used in burn and trauma animal models may be influencing results by modulating inflammatory and acute-phase responses. Accordingly, we determined the effects of various anesthetics, analgesia, and euthanasia techniques in a rodent burn model. Methods Isoflurane, ketamine-xylazine (KX), or pentobarbital, with or without buprenorphine, were administered prior to scald-burn in 72 rats that were euthanized without anesthesia by decapitation after 24 hours, and compared to unburned shams. In a second experiment, 120 rats underwent the same scald-burn injury using KX, and 24 hours later were euthanized under anesthesia or carbon dioxide (CO2). Additionally, we compared euthanasia by exsanguination versus decapitation. Serum cytokine levels were determined by ELISA. Results In the first experiment, isoflurane was associated with elevation of cytokine-induced neutrophil chemoattractant (CINC)-2, and monocyte chemotactic protein-1 (MCP-1), and KX and pentobarbital with elevation of CINC-1 and 2, interleukin (IL)-6 and MCP-1. Pentobarbital also decreased IL-1β. Interleukin-6 increased significantly when isoflurane or pentobarbital were combined with buprenorphine. In the second experiment, euthanasia performed by exsanguination under isoflurane was associated with reduced levels of IL-1β, CINC-1 and 2, and MCP-1, while KX reduced CINC-2 and increased IL-6 levels. Meanwhile, pentobarbital reduced levels of IL-1β and MCP-1, and CO2 reduced CINC-2 and MCP-1. Additionally, decapitation following KX, pentobarbital or CO2 decreased IL-1β and MCP-1, whilst we found no significant difference between isoflurane and controls. Euthanasia by exsanguination compared to decapitation using the same agent also led to modulation of several cytokines. Conclusions Differential expression of inflammatory markers with the use of anesthetics and analgesics should be considered when designing animal studies and interpreting results, as these appear to have significant modulating impact. Our findings indicate a brief anesthesia with isoflurane immediately prior to euthanasia by decapitation exerted the least dampening effect on the cytokines measured. Conversely, KX with buprenorphine may offer a better balance during longer procedures to avoid significant modulation. Standardization across all experiments that are compared and awareness of these findings is essential for those investigating the pathophysiology of inflammation in animal models.