Clinical Equipoise Regarding Glycemic Control
Clinical Equipoise Regarding Glycemic Control
Objectives: To assess the willingness of pediatric intensivists to conduct a pediatric trial of blood glucose control, and to determine if self-reported practices were influenced by adult-specific data over the past 4 yrs. This was a follow-up to our previous 2005 survey.
Design: Electronic survey comprising a 30-item questionnaire.
Setting: North American PICUs that were members of, or connected to, the Pediatric Acute Lung Injury and Sepsis Network (n = 96 targeted institutions).
Participants: North American pediatric intensivists (n = 209).
Interventions: None.
Methods: We conducted a survey of North American PICUs using a Web-based questionnaire. Invitations were sent to 96 institutions in 37 states/provinces.
Results: Response rate was 68% (141/209). The median definitions of hyperglycemia (150 mg/dL) and hypoglycemia (≤60 mg/dL) were similar to our 2005 survey results. Self-reported practice patterns remain variable. Although 75% of clinician respondents denied a change in clinical practice based on the published literature, the preferred blood glucose target range increased from 80–110 mg/dL in 2005 to 90–140 mg/dL in 2009. Intensivists who preferred a blood glucose target of 80–110 mg/dL decreased from 43% to 6% (p < 0.001). Many respondents (45%) indicated that the acceptable severe hypoglycemia rate (% patients) for a protocol was ≤2.5%. The majority (93%) indicated they would be willing to enroll patients in a pediatric trial of blood glucose control.
Conclusions: Pediatric intensivists report that they control blood glucose with insulin in critically ill children and do not necessarily adopt adult-specific data or a single uniform blood glucose target. The published evidence does not adequately address PICU clinicians concerns. Unanswered questions and persistent variation in practice suggest a need for a multicenter clinical trial of blood glucose control in critically ill children.
The Institute of Medicine and Federal Coordinating Council for Comparative Effectiveness Research recently established Comparative Effectiveness Research trials in pediatrics as an institute priority. Although the number of critically ill pediatric patients is growing, the number of critical care pediatric randomized controlled trials is small. Without pediatric-specific studies, pediatricians may be forced to extrapolate from adult studies. However, the clinical contexts facing children are different than for adults and may alter patient outcomes.
Uncertainties about the risks and benefits of blood glucose (BG) control in critically ill adult patients increase the challenges pediatricians face in evaluating the evidence to guide the care of their patients. Studies on the benefits of BG control in critically ill adults are conflicting. The Normoglycemia in Intensive Care Evaluation–Survival Using Glucose Algorithm Regulation (NICE–SUGAR) and meta-analyses suggest an increase in hypoglycemia with a slight increase in harm with a BG target = 80–110 mg/dL. Although some pediatric clinicians believe a narrow BG target that is normal for age ("tight" glycemic control) is beneficial, others believe its risk of hypoglycemia outweighs any potential benefit. A single-center prospective study of BG control in pediatric patients reported reduced mortality and morbidity with tight glycemic control (goal BG target was normal for age) despite a high rate of severe hypoglycemia (SH, BG ≤40 mg/dL). This differs from the results of the NICE–SUGAR trial, the largest multicenter trial in adults.
It is likely that published adult data influence pediatric clinical practice. In 2005, we surveyed adult and pediatric intensive care clinicians and found wide variations in BG targets, hyperglycemia and hypoglycemia definitions, and clinical care algorithms. Since that time, a number of primarily adult ICU studies on BG control with insulin therapy were published. We undertook the current survey to assess the climate for conducting a study of BG control in critically ill children. We intended our results to inform the design of a pediatric multicenter comparative effectiveness trial of BG control. We, therefore, investigated pediatricians' adoption of published adult results regarding BG management. We explored whether pediatric intensivists reported a change in their practice from 2005 to 2009. We considered the study results reported here in the design of the current National Institutes of Health Heart and Lung Failure Pediatric Insulin Titration Trial (HALF-PINT; U01-HL107681).
Abstract and Introduction
Abstract
Objectives: To assess the willingness of pediatric intensivists to conduct a pediatric trial of blood glucose control, and to determine if self-reported practices were influenced by adult-specific data over the past 4 yrs. This was a follow-up to our previous 2005 survey.
Design: Electronic survey comprising a 30-item questionnaire.
Setting: North American PICUs that were members of, or connected to, the Pediatric Acute Lung Injury and Sepsis Network (n = 96 targeted institutions).
Participants: North American pediatric intensivists (n = 209).
Interventions: None.
Methods: We conducted a survey of North American PICUs using a Web-based questionnaire. Invitations were sent to 96 institutions in 37 states/provinces.
Results: Response rate was 68% (141/209). The median definitions of hyperglycemia (150 mg/dL) and hypoglycemia (≤60 mg/dL) were similar to our 2005 survey results. Self-reported practice patterns remain variable. Although 75% of clinician respondents denied a change in clinical practice based on the published literature, the preferred blood glucose target range increased from 80–110 mg/dL in 2005 to 90–140 mg/dL in 2009. Intensivists who preferred a blood glucose target of 80–110 mg/dL decreased from 43% to 6% (p < 0.001). Many respondents (45%) indicated that the acceptable severe hypoglycemia rate (% patients) for a protocol was ≤2.5%. The majority (93%) indicated they would be willing to enroll patients in a pediatric trial of blood glucose control.
Conclusions: Pediatric intensivists report that they control blood glucose with insulin in critically ill children and do not necessarily adopt adult-specific data or a single uniform blood glucose target. The published evidence does not adequately address PICU clinicians concerns. Unanswered questions and persistent variation in practice suggest a need for a multicenter clinical trial of blood glucose control in critically ill children.
Introduction
The Institute of Medicine and Federal Coordinating Council for Comparative Effectiveness Research recently established Comparative Effectiveness Research trials in pediatrics as an institute priority. Although the number of critically ill pediatric patients is growing, the number of critical care pediatric randomized controlled trials is small. Without pediatric-specific studies, pediatricians may be forced to extrapolate from adult studies. However, the clinical contexts facing children are different than for adults and may alter patient outcomes.
Uncertainties about the risks and benefits of blood glucose (BG) control in critically ill adult patients increase the challenges pediatricians face in evaluating the evidence to guide the care of their patients. Studies on the benefits of BG control in critically ill adults are conflicting. The Normoglycemia in Intensive Care Evaluation–Survival Using Glucose Algorithm Regulation (NICE–SUGAR) and meta-analyses suggest an increase in hypoglycemia with a slight increase in harm with a BG target = 80–110 mg/dL. Although some pediatric clinicians believe a narrow BG target that is normal for age ("tight" glycemic control) is beneficial, others believe its risk of hypoglycemia outweighs any potential benefit. A single-center prospective study of BG control in pediatric patients reported reduced mortality and morbidity with tight glycemic control (goal BG target was normal for age) despite a high rate of severe hypoglycemia (SH, BG ≤40 mg/dL). This differs from the results of the NICE–SUGAR trial, the largest multicenter trial in adults.
It is likely that published adult data influence pediatric clinical practice. In 2005, we surveyed adult and pediatric intensive care clinicians and found wide variations in BG targets, hyperglycemia and hypoglycemia definitions, and clinical care algorithms. Since that time, a number of primarily adult ICU studies on BG control with insulin therapy were published. We undertook the current survey to assess the climate for conducting a study of BG control in critically ill children. We intended our results to inform the design of a pediatric multicenter comparative effectiveness trial of BG control. We, therefore, investigated pediatricians' adoption of published adult results regarding BG management. We explored whether pediatric intensivists reported a change in their practice from 2005 to 2009. We considered the study results reported here in the design of the current National Institutes of Health Heart and Lung Failure Pediatric Insulin Titration Trial (HALF-PINT; U01-HL107681).
Source...