Prevalence of dysglycemia and association with outcomes in pediatric extracorporeal membrane oxygenation

Dysglycemia in Critically Ill Children Admitted to Jimma Medical Centre, Southwest Ethiopia

Background

Abnormal blood glucose level is one of the most frequently encountered problems in children with severe illnesses. However, its magnitude and outcome have rarely been determined in Ethiopia. We aimed to determine the magnitude, associated factors and outcome of dysglycemia in critically ill children admitted to Jimma Medical Center.

Methods

Prospective longitudinal study was conducted on children aged 28 days to 14 years admitted with critical illnesses at the different units of the Department of Pediatrics and Child Health of Jimma Medical Center, Southwest Ethiopia, from June to August 2019. Data were collected by trained medical personnel using structured questionnaire and analyzed using Statistical Package for Social Sciences (SPSS) windows version 20.0. Dysglycemia was considered whenever the child had a random blood sugar >150mg/dl or <45mg/dl.

Result

Dysglycemia was seen at admission in 139/481, 28.9% children; 24 (5.0%) had hypoglycemia whereas 115 (23.9%) had hyperglycemia. The factors associated with dysglycemia at admission were severe acute malnutrition (p=002, AOR=3.09, CI=1.18,7.77), impaired mental status (p=0.003, AOR=4.63, CI=1.68, 12.71), place of residence (p=0.01, AOR=1.85, CI=1.15-2.96) and presence of diarrhea on date of admission. Among the children who had dysglycemia at admission, 16/139, 11.5% died.

Conclusion

Dysglycemia is a common problem in critically ill children in the setting. Blood glucose level should be determined for all critically ill children, and routine empirical administration of dextrose should be minimized since most of the children with dysglycemia had hyperglycemia than hypoglycemia.

Admission blood glucose level and outcome in patients requiring venoarterial extracorporeal membrane oxygenation

Background

Patients with cardiogenic shock or cardiac arrest undergoing venoarterial extracorporeal membrane oxygenation (V-A ECMO) frequently present with blood glucose levels out of normal range. The clinical relevance of such findings in the context of V-A ECMO is unknown. We therefore investigated the prognostic relevance of blood glucose at time of cannulation for V-A ECMO.

Methods

We conducted a single-center retrospective registry study. All patients receiving V-A ECMO from October 2010 to January 2020 were included if blood glucose level at time of cannulation were documented. Patients were divided in five groups according to the initial blood glucose level ranging from hypoglycemic (< 80 mg/dl), normoglycemic (80–140 mg/dl), to mild (141-240 mg/dl), moderate (241–400 mg/dl), and severe (> 400 mg/dl) hyperglycemia, respectively. Clinical presentation, arterial blood gas analysis, and survival were compared between the groups.

Results

392 patients met inclusion criteria. Median age was 62 years (51.5–70.0), SAPS II at admission was 54 (43.5–63.0), and 108/392 (27.6%) were female. 131/392 were discharged alive (hospital survival 33.4%). At time of cannulation, survivors had higher pH, hemoglobin, calcium, bicarbonate but lower potassium and lactate levels compared to non-survivors (all p < 0.01). Outcome of patients diagnosed with particularly high (> 400 mg/dl) and low (< 80 mg/dl) blood glucose at time of V-A ECMO cannulation, respectively, was worse compared to patients with normoglycemic, mildly or moderately elevated values (p = 0.02). Glucose was independently associated with poor outcome after adjustment for other predictors of survival and persisted in all investigated subgroups.

Conclusion

Arterial blood glucose at time of V-A ECMO cannulation predicts in-hospital survival of patients with cardiac shock or after ECPR. Whether dysglycemia represents a potential therapeutic target requires further evaluation in prospective studies.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1007/s00392-021-01862-7.

Intraoperative dextrose rate during exploratory laparotomies in neonates and the incidence of postoperative hyperglycemia: A retrospective observational study

INTRODUCTION

Compared with the older pediatric population, neonates have greater perioperative morbidity and mortality. Difficulty with glucose regulation may be a contributing modifiable risk factor during perioperative anesthetic management. To mitigate the risk of hyperglycemia in neonates, some providers empirically halve the preoperative rate of dextrose-containing infusions during surgery.

AIM

To assess the association between halving the preoperative maintenance dextrose rate and postoperative euglycemia in neonatal intensive care unit patients undergoing exploratory laparotomies.

METHODS

Neonatal intensive care unit patients who underwent exploratory laparotomy under general anesthesia from 1/1/2014 to 11/21/2019 were included in this analysis. Hyperglycemia and hypoglycemia were defined as >150 mg/dL and <46 mg/dL. A calculated dextrose ratio was utilized to categorize patients into full and half intraoperative dextrose rate cohorts. Univariate analyses were performed with Fisher's exact test, the Wilcoxon rank sum test, or Spearman's correlation. Multivariable analyses with regression models were conducted after graphical evaluation of a predetermined set of independent variables.

RESULTS

107 patients were included in the full dextrose rate cohort and 96 patients in the half dextrose rate cohort with postoperative hyperglycemia occurring in 47 and 28 patients, respectively. On univariate analysis, halving the preoperative dextrose rate was associated with decreased postoperative hyperglycemia (odds ratio: 0.53; 95% CI: 0.28-0.98, P = 0.041). This association continued in the regression model (adjusted odds ratio: 0.49; 95% CI: 0.25-0.80, P = 0.008) after controlling for preoperative dextrose rate, preoperative serum glucose, preoperative pH, surgical duration, postmenstrual age at surgery, and the presence of necrotizing enterocolitis. Only one patient was hypoglycemic postoperatively, and they were in the full dextrose cohort.

CONCLUSION

Halving of preoperative dextrose rates intraoperatively during exploratory laparotomy in neonatal intensive care unit patients was associated with a decreased risk of postoperative hyperglycemia without substantially increasing the occurrence of postoperative hypoglycemia. The practice of halving preoperative dextrose rates may be an effective empirical approach for intraoperative glucose management in the high-risk neonatal population when blood glucose monitoring is challenging.

Short-Term Adverse Outcomes Associated With Hypoglycemia in Critically Ill Children

OBJECTIVES

Previous studies report worse short-term outcomes with hypoglycemia in critically ill children. These studies relied on intermittent blood glucose measurements, which may have introduced detection bias. We analyzed data from the Heart And Lung Failure-Pediatric INsulin Titration trial to determine the association of hypoglycemia with adverse short-term outcomes in critically ill children.

DESIGN

Nested case-control study.

SETTING

Thirty-five PICUs. A computerized algorithm that guided the timing of blood glucose measurements and titration of insulin infusion, continuous glucose monitors, and standardized glucose infusion rates were used to minimize hypoglycemia.

PATIENTS

Nondiabetic children with cardiovascular and/or respiratory failure and hyperglycemia. Cases were children with any hypoglycemia (blood glucose < 60 mg/dL), whereas controls were children without hypoglycemia. Each case was matched with up to four unique controls according to age group, study day, and severity of illness.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

A total of 112 (16.0%) of 698 children who received the Heart And Lung Failure-Pediatric INsulin Titration protocol developed hypoglycemia, including 25 (3.6%) who developed severe hypoglycemia (blood glucose < 40 mg/dL). Of these, 110 cases were matched to 427 controls. Hypoglycemia was associated with fewer ICU-free days (median, 15.3 vs 20.2 d; p = 0.04) and fewer hospital-free days (0 vs 7 d; p = 0.01) through day 28. Ventilator-free days through day 28 and mortality at 28 and 90 days did not differ between groups. More children with insulin-induced versus noninsulin-induced hypoglycemia had zero ICU-free days (35.8% vs 20.9%; p = 0.008). Outcomes did not differ between children with severe versus nonsevere hypoglycemia or those with recurrent versus isolated hypoglycemia.

CONCLUSIONS

When a computerized algorithm, continuous glucose monitors and standardized glucose infusion rates were used to manage hyperglycemia in critically ill children with cardiovascular and/or respiratory failure, severe hypoglycemia (blood glucose < 40 mg/dL) was uncommon, but any hypoglycemia (blood glucose < 60 mg/dL) remained common and was associated with worse short-term outcomes.