Fluid management in children with diabetic ketoacidosis

Abstract

Question

Previous research has indicated that rapid rehydration in children with type 1 diabetes who present with diabetic ketoacidosis could result in cerebral edema. I have been treating patients with diabetic ketoacidosis with gradual fluid replacement. With the risk of cerebral injury in these patients, should I continue management with slow fluid rehydration?

Answer

Recent research has shown that neither fluid infusion rate nor sodium chloride concentration increases risk of cerebral injury. However, it is possible for subtle brain injury to occur during treatment, regardless of the fluid administration strategy. The 2018 International Society for Pediatric and Adolescent Diabetes guidelines have been updated in light of this research.

Diabetic ketoacidosis (DKA) is the most common complication of type 1 diabetes (T1D) in children, with an incidence of 1% to 10% of diabetic patients per year in resource-rich countries.¹ Insufficient insulin levels and subsequent gluconeogenesis, lipolysis, and ketogenesis cause substantial hyperglycemia, which draws sugar and extracellular fluids into the urine, resulting in dehydration, ketones in the blood and urine, and acidosis.² The 2018 International Society for Pediatric and Adolescent Diabetes (ISPAD) definition for DKA is a blood glucose level of greater than 11 mmol/L, venous pH of less than 7.3 or serum bicarbonate level of less than 15 mmol/L, and either presence of ketonemia (blood β-hydroxybutyrate level ≥ 3 mmol/L) or moderate to high ketonuria.¹ A rare but severe complication of pediatric DKA is cerebral injury, occurring in about 1% of cases, with an estimated mortality rate of 20% to 25%.³

One-third of children present with DKA at diagnosis of T1D.⁴ A 2019 US survey reported that children 18 years of age or younger with a missed diagnosis of T1D at an initial health care visit were 18% more likely to develop DKA (relative risk [RR] = 1.18; 95% CI 1.05 to 1.32; P < .05).⁵ Thus, a presumptive diagnosis of T1D should be confirmed immediately if possible, although confirmatory testing should not delay treatment.⁶

In patients with established T1D, factors indicating DKA include febrile illness or gastroenteritis⁷; missed insulin injections owing to social and economic factors, as well as insulin compliance⁸; or technical malfunction of an insulin pump.⁷ In adolescents, poor compliance with an insulin regimen due to mental health or eating disorders,^9,10 or behaviour like the use of cannabis products or alcohol that results in poor metabolic control, might precipitate DKA.^11,12

Presentation of DKA might include dehydration, tachycardia, tachypnea, Kussmaul breathing (deep, laboured respirations), nausea, vomiting, abdominal pain, and altered level of consciousness.¹ Common imitators of DKA that can result in a missed diagnosis of diabetes include urinary tract infection, acute abdominal pain, febrile illness, and respiratory illness.¹³ Among 118 accounts of DKA in 52 children from a 2014 study, a missed diagnosis of DKA at the initial health care visit was associated with a significantly higher rate of cerebral injury (odds ratio of 4.47; 95% CI 1.79 to 11.89) and mortality (21% with delayed diagnosis compared with 3% with initial diagnosis; odds ratio of 8.40; 95% CI 1.90 to 58.56).¹³

Cerebral injury risk

Despite efforts to reduce risk of cerebral injury in children with DKA, the rate has remained relatively stable for the past 40 years.^14,15 Subclinical brain swelling, seen as a reduction in ventricle size on magnetic resonance imaging, has been reported in up to 54% of cases of DKA at onset of diagnosis, yet clinically overt cerebral injury often does not develop until 8 to 12 hours after beginning treatment.¹⁶ A previous study suggested that swelling was due to brain parenchyma preserving electrolyte and water balance during DKA, and that rapid rehydration could result in osmotic edema.¹⁵ However, DKA causes a deficit of extracellular fluid that limits peripheral perfusion.²

Fluid treatment

The balance between the need to replenish fluids and the fear that rapid rehydration could result in brain swelling has led for many years to a fluid management approach of slow, measured fluid replacement in children with DKA.^15,17 A case review in 1988 with 42 patients younger than 28 (mean age of 11 years) found an inverse correlation between overall rate of fluid administration and time of onset of brain herniation (r = -0.32, P = .04).¹⁸ However, patients with more severe DKA, who are more dehydrated, are at a higher risk of developing cerebral edema (CE).⁴ It is possible that those patients received overaggressive hydration compared with children who presented with milder symptoms.¹⁸

Case-control studies from 2001 and from 2005 failed to correlate rate of fluid administration with risk of intracerebral complications during treatment of DKA.^3,19 Moreover, evidence from studies of children with DKA and studies of rodent models indicate that DKA-induced hypoperfusion and neuroinflammation might result in vasogenic edema when fluids are administered during treatment, regardless of fluid rate.^16,20 Despite these findings, the 2014 ISPAD clinical practice guidelines recommended that initial fluid replacement take place over 4 to 6 hours in children with DKA to reduce risk of CE.²¹

Shift in fluid treatment

In 2018, Kuppermann et al and the Pediatric Emergency Care Applied Research Network conducted a 13-centre randomized controlled trial of 1389 cases of DKA,²² with the aim of determining whether the rate of fluid administration was associated with neurologic deterioration. Participants between 0 and 18 years of age were assigned to 1 of 4 fluid replacement regimens: fast rate (20 mL/kg over 12 hours followed by 10 mL/kg over 24 hours) with 0.45% normal saline (NS), slow rate (10 mL/kg over 48 hours) with 0.45% NS, fast rate with 0.9% NS, or slow rate with 0.9% NS. Neither fluid rate nor sodium chloride concentration increased the risk of neurologic deterioration during treatment (RR = 0.76; 95% CI 0.44 to 1.33; P = .34). Follow-up neurocognitive outcomes 2 to 6 months after DKA also did not differ among intervention groups (P = .06). Furthermore, of the 22 patients who were treated for CE, there was no significant difference in incidence of CE between groups (RR = 1.43; 95% CI 0.46 to 4.40; P = .53). However, they did report that subtle brain injury often occurred during treatment, regardless of fluid administration strategy, supporting the idea that DKA-induced injury and subsequent vasogenic edema occurs independent of fluid-administration rate.²⁰

The 2018 ISPAD DKA guidelines have been updated and recommend administration of 0.9% NS of 10 mL/kg over 30 to 60 minutes to restore peripheral circulation.¹ A second fluid bolus of 10 mL/kg might be given over the next 30 minutes if there are still signs of hypoperfusion. The ISPAD protocol recommends replacing the remaining fluid deficit over 24 to 48 hours with an intravenous (IV) drip of 0.9% NS at a rate of 4 to 6.5 mL/kg/hour (depending on the child’s weight), with an added 40 mmol/L of potassium chloride. Intravenous insulin should be initiated at a rate of 0.05 to 0.1 units/kg/hour, but not until 1 to 2 hours after IV fluids have begun, since it has also been shown that earlier insulin administration is associated with an increased risk of cerebral injury.¹

Conclusion

Diabetic ketoacidosis is a serious complication of T1D, caused by a lack of insulin. A severe risk factor of DKA in children is cerebral injury, although its exact pathophysiology remains unknown. It is important for primary care providers and emergency physicians to recognize early signs of insulin deficiency and T1D in order to prevent DKA and reduce the risk of cerebral injury. For many years, fluid replacement in DKA was administered gradually to avoid rapid osmotic changes in the brain. New evidence suggests that neither the infusion rate nor the sodium chloride content of IV fluids affects neurologic outcomes during treatment of DKA, and current guidelines have been updated to reflect this new information.

Child Health Update is produced by the Pediatric Research in Emergency Therapeutics (PRETx) program (www.pretx.org) at the BC Children’s Hospital in Vancouver, BC. Ms McGregor and Drs Metzger and Amed are members, and Dr Goldman is Director of the PRETx program. The mission of the PRETx program is to promote child health through evidence-based research in therapeutics in pediatric emergency medicine.

Do you have questions about the effects of drugs, chemicals, radiation, or infections in children? We invite you to submit them to the PRETx program by fax at 604 875-2414; they will be addressed in future Child Health Updates. Published Child Health Updates are available on the Canadian Family Physician website (www.cfp.ca).