How often do patients with DKA present with cerebral edema?

• 25-40% of pts with DMI experience DKA
• <1% of DKA presents with cerebral edema
• Interestingly, you may not see the Cushing's Triad as these pts have impaired cerebral auto-regulation
• Cerebral edema typically occurs 4-12 hrs after DKA treatment initiated, but some pts have edema upon arrival to ED

What is the mortality risk of cerebral edema?

• 30%-60% mortality rate and up to 33% of survivors with neurologic deficits

What is the pathophys of cerebral edema?

• Ischemia
• Dehydration + compensatory hyperventilation --> reduced blood flow + low PaCO2 --> vasoconstriction --> cerebral ischemia
• Inflammatory
• DKA induces non-infectious systemic inflammatory response with elevated levels of proinflammatory cytokines (IL-1, IL-6, IL-10, TNF), raised CRP, complement activation, generation of ROS, and reduced levels of antioxidants --> leaky blood-brain barrier --> vasogenic edema
• Hyperosmolar state
• Prolonged hyperglycemia --> prolonged serum hyperosmolarity --> brain producing idiogenic osmoles --> fluid shift into brain with improvement in hyperglycemia from DKA treatment (brain osmoles dissipate over 12-24 hrs after DKA therapy initiated)
• Ketones and acidosis
• Acetoacetate and B-hydroxybutyrate not only have osmotic effects that cross blood-brain barrier, but are also proinflammatory agents --> leaky blood-brain barrier

Which is a better rescue hyperosmolar therapy, mannitol or hypertonic saline (HTS)?

• A meta-analysis of hyperosmolar therapy in traumatic and non-traumatic brain injury with increased ICP showed that HTS was better than mannitol...but is this true for cerebral edema from DKA? ...not clear
• A case of mannitol therapy failure in DKA cerebral edema thought to be due to iatrogenic lowering of Na and subsequent worsening mental status
• IV mannitol given --> rise in extracellular osmolality --> influx of water from intracellular compartment --> dilutes and lowers serum Na and lowers serum osmolality transiently --> eventual renal clearance of mannitol (osmotic diuresis and elimination of free water) --> increase in serum Na and osmolality --> net effect of a single does of mannitol is no change or small net rise in osmolality
• Given the phenomenon of mannitol causing transiently low Na with worsening mental status...
• Some experts recommend initially using HTS
• If pt does not respond to HTS, blood-brain barrier must be significantly deteriorated --> only option is mannitol
• From 1999 to 2009, use of mannitol as a sole agent decreased from 98% to 49%, use of HTS as a sole agent increased from 2% to 39%, and use of combination therapy increased from 0% to 10%
• Pts who received combination therapy had higher mortality rate than pts who received single agents
• Interestingly, use of HTS as a sole agent was associated with higher mortality than mannitol alone, especially in pts on mechanical ventilation (retrospective study results so unclear why)
• However, mortality rate from 1999 to 2009 decreased by 83% with a transition from mannitol use to HTS use
• So which agent to use? Who knows!

Should we hyperventilate pts with DKA cerebral edema for neuroprotection?

• One article recommends to "avoid intubation unless it is for exhaustion...if intubation undertaken, consider a target level of PaCO2 appropriate for estimated CSF HCO3 concentration (...what?)...implication that "aggressive hyperventilation" should be avoided may be counter to the logic of adaptive physiology"
• Another article recommends "ventilating to PaCO2 levels existing at the time intubation is performed"

Should we limit fluid administration in pts with DKA cerebral edema?

• Unclear!
• Some authors recommend:
• Do not give fluid bolus unless hemodynamically unstable
• Do not continue to use boluses once circulatory stability is demonstrated
• Do not overestimate degree of dehydration (5-7% down usually, 10-15% down only if in shock)
• Replace fluid deficits evenly over 48hrs
• On-going FLUID (FLuid therapies Under Investigation in DKA) study is the first prospective randomized trial to evaluate fluid regimens for pediatric DKA

Should we administer NaHCO3 in pts with DKA cerebral edema and/or severe acidosis?

• No!
• Retrospective evidence of increased risk for cerebral edema and prolonged hospitalization in children
• In adults, evidence of transient worsening of ketosis, increased need for K supp, and no evidence of any clinical efficacy
• 2 RCTs did show improved acidosis resolution in the initial 2 hrs of therapy but no benefit beyond the 2 hrs
• No evidence of improved hemodynamic stability with the use of NaHCO3 in DKA

Should we increase insulin infusion rate if acidosis or hyperglycemia not correcting quickly?

• Our endocrinologist said not to increase rate over 0.1U/kg/hr as there is no evidence for benefit
• RCT in 1980 showed that 0.1U/kg/hr was as effective as 1.0U/kg/hr
• Recent RCT published in JAMA suggests 0.05U/kg/hr is non-inferior to 0.1U/kg/hr with respect to rate of glucose correction and resolution of acidosis
• However, some authors recommend that if pt fails to start to correct hyperglycemia in 2-4 hrs, increase to 0.15U/kg/hr (not evidence based)
• Do not let glucose decrease >50-100 per hr

Should we have a DKA evidence-based pathway in place?

• Seattle Children's Hospital implemented a DKA pathway (including management for cerebral edema) that led to:
• Reduction in the rate of hypokalemia
• Reduction in ICU stay
• Improved DKA education
• Faster initiation of DKA management
• Decreased incidence of cerebral edema
• Decreased use of bicarbonate

Sources

• 2014 - Improving Care for Pediatric DKA (Pediatrics)
• 2014 - HTS for Cerebral Edema in DKA - No Change Yet, Please (Pediatric Critical Care Med)
• 2014 - Low-dose vs. Standard-dose Insulin in Pediatric DKA (JAMA)
• 2014 - Cerebral Edema in Children with DKA - Vasogenic Rather Than Cellular? (Pediatric Diabetes)
• 2013 - Seattle Children's Hospital DKA v.2: Cerebral Edema 
• 2013 - Treating Cerebral Edema in DKA (Peds Critical Care Med)
• 2013 - Increasing Use of HTS Over Mannitol in the Treatment of Symptomatic Cerebral Edema in Pediatric DKA (Peds Critical Care Med)
• 2011 - Bicarbonate in DKA (Annals of Intensive Care)
• 2008 - Cerebral Edema in DKA (Peds Critical Care Med) 
• 2005 - Hyperventilation in Severe DKA (Peds Critical Care Med)
• 1980 - Comparison of High-dose and Low-dose Insulin in DKA in Children (Diabetes Care)

1. Glaser N, Barnett P, McCaslin I, et al. Risk factors for cerebral edema in children with diabetic ketoacidosis. The Pediatric Emergency Medicine Collaborative Research Committee of the American Academy of Pediatrics. N Engl J Med 2001; 344:264.