Study of High Blood Sugars and Insulin in Hospitalized, Critically Ill Children (CIH)

Insulin Resistance Versus Absolute Insulin Deficiency: Evaluating the Mechanism of Hyperglycemia in Pediatric Critical Illness

BACKGROUND AND PURPOSE Critical illness hyperglycemia (CIH) - elevated blood glucose in the critically ill patient population - has gained much interest among health care providers over the past several years. Clinical studies in adults have documented a high rate of hyperglycemia in some post-surgical and medical intensive care units (ICUs). However, the primary reasons for interest in this topic are not due just to its high rate, but also to the fact that by returning the high glucose levels found in this population to normal with insulin therapy can dramatically improve clinical outcomes by decreasing both morbidity and long-term mortality. Because of this, aggressive glucose control has become common practice in adult ICU critical care management. Although there is substantial data describing the high incidence of CIH in adult patients, there is little information regarding this condition in children. A single retrospective study recently published also suggested a high incidence of CIH in children with critical illness secondary to both medical and surgical causes. It is yet to be determined if, like in adults, normalizing blood glucose levels with insulin improves outcomes in this pediatric population. Because evidence appears compelling that hyperglycemia is both common and detrimental in adults, many pediatric ICUs have likewise begun to focus on aggressively treating hyperglycemia in critically ill children. The proposed study is a prospective observational pilot study to occur in the Pediatric Intensive Care Unit (PICU) at Children's Healthcare of Atlanta at Egleston. This prospective pilot study is being done to evaluate the endocrine factors associated with, if not responsible for, CIH, and the changes that take place with the restoration of normal blood glucose levels by insulin therapy. To address these profound issues this study will pursue two interrelated Aims: Aim #1: To determine if critical illness hyperglycemia is associated with absolute insulin deficiency, peripheral insulin resistance, or both. Aim #2: To characterize the requirement of insulin required to initially restore and maintain normal blood glucose levels, and compare the changes in insulin that take place with this normalization in patients with CIH. We hypothesize that the hyperglycemic response to critical illness will be associated with abnormally low levels of insulin as compared to patients without critical illness hyperglycemia.

SUMMARY OF PROCEDURES This will be an approximately 12 month long, single center, prospective study of patients between ages 2 and 12 years old who meet criteria for CIH in our pediatric intensive care unit during the period of enrollment. Patients will be screened by physicians and staff and evaluated via inclusion and exclusion criteria, the evaluation will discussed with the family/legal guardian by appointed critical care fellow, attending or physician appointee, and the patient will be enrolled if informed consent is granted. Following informed consent approval from family/legal guardian, blood samples will be collected and sent for storage and processing in the clinical laboratory at Children's. Three patient population groups will be assessed in this study: Group A - Study Patients - At risk for CIH/with CIH Labs will be drawn in our target population at diagnosis of CIH and sent for insulin and C-peptide levels. These levels will be repeated 24 hours after initiation of insulin, and then followed every three days until insulin is no longer needed, then repeated once more 24 hours after insulin is off (see protocol). Group B - Control Patients - At risk for CIH/without CIH Insulin and C-peptide levels will be drawn in a control group of patients with the same risk factors but without CIH at the time of admission, 24 hours after admission, then in 3 days. Group C - Control Patients - Not at risk for CIH/without CIH The same protocol of lab schedules will be used in a second control group consisting of patients that are not deemed at risk for critical illness hyperglycemia. III. POTENTIAL RISKS Most patients who meet criteria usually have an arterial line, central venous catheter, or "blood drawing" IV. It will be rare that a separate venous puncture will be needed to collect samples. If the blood to be collected, approximately 2cc per stick, is believed to negatively impact the patient by either the patient or clinical caretakers, blood will not be drawn. Information collected in this study will contain Protected Health Information (PHI) on subjects. There is the potential risk of unintentional disclosure of this information. However, certain steps will be followed in this study minimalize PHI disclosure to help protect the subjects' privacy and confidentiality. Once collected, this information will be kept in a database that is password protected and with limited access on a password-protected network. Other than institutions required by law to have access to research records, only authorized research staff will have access to the research database associated with the study. Furthermore, no data will be identifiable to study subjects, as each subject will be assigned a unique study number when entered into a database. If published or presented, no identifying features will be provided. IV. POTENTIAL BENEFITS There may be no direct potential benefit to the subjects participate in this study. However, the information learned from this study may benefit other patients in the future. V. INCLUSION AND EXCLUSION CRITERIA Inclusion: - All patients between ages 2 and 12 years of age admitted to the pediatric ICU who meet criteria for critical illness hyperglycemia (ie: all patients on mechanical ventilation, all patients on vasoactive medications, patients deemed otherwise "at risk" for CIH by attending service) and are started on our standard CIH screening and treatment protocol will comprise the pool of potential study candidates. Exclusion: - Certain patients who meet criteria for critical illness hyperglycemia and thus are started on insulin will be excluded from this study: Age Patients less than 2 or greater than 12 years of age will be excluded from this study to decrease age variability. Pre-existing known endocrine disorder Patients with known or suspected Type 1 diabetes will be excluded because of their pre-existing absolute lack of insulin production secondary to autoimmune β cell destruction. Oncology patients Oncology patients will be excluded because of the effects of immunosuppressants on study markers. Renal replacement therapy Patients on continuous veno-venous hemofiltration (CVVH) or any type of dialysis (intermittent dialysis, peritoneal dialysis) will be excluded as it is unclear how different forms of renal replacement therapy affect insulin levels, and certain types of dialysis affect glucose and insulin levels (i.e. dextrose containing PD fluid). Non-adherence Inability or unwillingness of the legal guardian to provide consent, or unwillingness of the child to provide assent. VI. INFORMED CONSENT PROCESS Patients will be considered eligible for this study at the discretion of the pediatric intensivists or their appointees (i.e. pediatric critical care fellows or nurse practitioners). There will be discussion with the family of the background information, reasons for conducting the study, and the risks involved. At any time during the study there will be opportunity to be removed from the study without impact on care. After all questions have been answered, written informed consent will be obtained from the parents (or legal guardian), and signed by the intensivist or appointee. If the parent/subject decides not to participate, the child will continue with their standard of care procedures as originally planned without any further testing. As part of the ongoing informed consent process, whenever appropriate, the subjects participating in this study will be provided with additional pertinent information after initial time of consent.

In hyperglycemic patients who will be starting insulin infusions to control hyperglycemia, blood will be drawn just prior to initiation of insulin infusion for the following levels: insulin, glucose, and C-peptide. These levels will be re-drawn upon achieving euglycemia, at 24 hours following that, then every three days. Levels will be again drawn once the insulin infusion is stopped/when CIH has resolved, and 24 hours following discontinuation of insulin infusion. At each timepoint the patient's clinical status will be documented and significant interval changes (intubation/extubation, change in pressor need), amount of dextrose (mg/kg/hour) supplied, and other concurrent medicines and doses will be recorded.

For comparative controls, insulin, C-peptide, and glucose levels will be drawn from ICU patients aged 2-12 years at similar "risk" (mechanical ventilation or vasoactive medications) but without CIH. The above labs will be drawn and data gathered near the time of risk, 24 hours later, then in 3 days following, for a total of three timepoints.

In addition, other ICU patients aged 2-12 years that are deemed NOT at risk for critical illness hyperglycemia will also be evaluated to serve as a group not "at risk" but admitted to the PICU as a further control population. Like Group B, the above labs will be drawn and data gathered at the time consent is obtained, 24 hours later, then in 3 days following, for a total of three timepoints

We hypothesize that the hyperglycemic response to critical illness will be associated with abnormally low levels of endogenous insulin as compared to patients without critical illness hyperglycemia.

At time of consent, 24 hours, 72 hours, and if applicable every additional 72 hours until insulin discontinued. A final lab 24 hours after D/C of insulin will be performed.

Inclusion Criteria: - All patients between ages 2 and 12 years of age admitted to the pediatric ICU who meet criteria for critical illness hyperglycemia (ie: all patients on mechanical ventilation, all patients on vasoactive medications, patients deemed otherwise "at risk" for CIH by attending service) and are started on our standard CIH screening and treatment protocol will comprise the pool of potential study candidates. Exclusion Criteria: - Certain patients who meet criteria for critical illness hyperglycemia and thus are started on insulin will be excluded from this study: Age Patients less than 2 or greater than 12 years of age will be excluded from this study to decrease age variability. Pre-existing known endocrine disorder Patients with known or suspected Type 1 diabetes will be excluded because of their pre-existing absolute lack of insulin production secondary to autoimmune β cell destruction. Oncology patients Oncology patients will be excluded because of the effects of immunosuppressants on study markers. Renal replacement therapy Patients on continuous veno-venous hemofiltration (CVVH) or any type of dialysis (intermittent dialysis, peritoneal dialysis) will be excluded as it is unclear how different forms of renal replacement therapy affect insulin levels, and certain types of dialysis affect glucose and insulin levels (i.e. dextrose containing PD fluid). Non-adherence Inability or unwillingness of the legal guardian to provide consent, or unwillingness of the child to provide assent.