Variability in Emergency Management of Pediatric Diabetic Ketoacidosis.

OBJECTIVES: Our aim in this study was to identify pediatric patients presenting in diabetic ketoacidosis (DKA) who received initial treatment inconsistent with current guidelines and then to track associated complications. METHODS: This retrospective chart review examined 47 pediatric patients admitted to our institution with DKA, including those transferred from community hospitals. Primary outcome measures were exposure to treatment-related risk factors for cerebral edema (CE), including administration of an intravenous (IV) insulin bolus, IV bicarbonate, insulin within 1 hour of IV fluid start and non-NPO (nil per os) status. RESULTS: Seventy-five percent of patients were exposed to at least one management practice that deviated from guidelines. Thirty-four percent of patients were exposed to a treatment-related risk factor for CE, with a significantly higher prevalence in those presenting to community centres (52% vs 19%; p=0.02). There were no radiologically confirmed cases of CE. CONCLUSIONS: Despite the presence of multiple evidence-based guidelines, a significant proportion of children---and especially those who present to community centres---are exposed to practices that increase CE risk.

Hypoglycemia in unmonitored full-term newborns-a surveillance study.

BACKGROUND AND OBJECTIVES: Hypoglycemia monitoring is not recommended for most full-term newborns. We wished to determine the incidence, presentation and case characteristics of hypoglycemia in low-risk newborns. METHODS: With the assistance of the Canadian Paediatric Surveillance Program, we conducted a national study of severe hypoglycemia in apparently low-risk full-term newborns. Paediatricians who reported a case were sent a detailed questionnaire and the data were analyzed. RESULTS: All 93 confirmed cases were singletons, 56% were first-borns and 65% were male. An 8% rate of First Nations cases was twofold the population rate. Maternal hypertension rate was 23%, fourfold the general pregnancy rate. Maternal obesity was double the general pregnancy rate at 23%. Concerning signs or feeding issues were noted in 98% at the time of diagnosis. Median time to diagnosis was 4.1 hours. Mean blood glucose at intravenous (IV) start was 1.4 ± 0.5 hours (SD). Seventy-eight per cent had at least one of four potential stress indicators and were more likely to have early diagnosis (P=0.03). Major signs were present in 20%. Those cases presented later and had lower glucose levels (median=0.8 mmol/L versus 1.6 mmol/L, [P<0.001). Twenty-five per cent of cases had birth weight less than the 10th centile. Neurodevelopmental concern was reported in 20%. Of the 13 cases which had brain magnetic resonance imaging, 11 were abnormal. CONCLUSION: Hypoglycemia in unmonitored newborns is uncommon but is associated with significant morbidity. We provide a range of clues to help identify these newborns soon after birth. Widespread adoption of norm-based standards to identify small-for-gestational age infants is supported.