Pediatrics

Diabetic ketoacidosis/cerebral edema

OVERVIEW: What every practitioner needs to know

Are you sure your patient has diabetic ketoacidosis–related cerebral edema? What are the typical findings for this disease?

Cerebral edema is a potentially life-threatening complication of diabetic ketoacidosis (DKA) and is responsible for the majority of diabetes-related deaths in children.

Cerebral edema typically occurs after several hours of treatment with insulin and intravenous fluids but can also occur at the time of presentation of DKA before treatment is started.

The risk of cerebral edema is related to the severity of acidosis, hypocapnia, and dehydration at the time of presentation of DKA. Although severe, clinically apparent cerebral edema occurs in just 1% of DKA episodes in children, numerous studies have demonstrated that mild cerebral edema, associated with only minimal or no alterations in mental status, is present in the majority of children during DKA treatment.

The relationship between intravenous fluid treatment and the risk of DKA-related cerebral edema is frequently debated; however, there are no clear associations between the use of particular fluid treatment protocols and increased risk of DKA-related cerebral edema. At present, whether and how cerebral edema can be prevented is unknown. Treatment for clinically apparent cerebral edema typically involves use of hyperosmolar agents (mannitol or hypertonic saline).

DKA-related cerebral edema is a clinical diagnosis. Imaging studies may be helpful but are not always definitive. The most common symptoms of DKA-related cerebral edema include mental status changes (confusion, irritability, obtundation) associated with severe headache, recurrence of vomiting, seizures, hypertension, inappropriate slowing of heart rate, and/or signs of increased intracranial pressure.

Although severe clinically apparent cerebral edema is uncommon (approximately 1% of DKA episodes in children), recent data suggest that the majority of children with DKA have mild subclinical cerebral edema. In most cases, this edema is asymptomatic or associated with only minor changes in mental status. This mild edema is not clearly evident on imaging studies performed during episodes of DKA but can be detected when comparing imaging studies during an episode of DKA to imaging studies of the same patient after recovery from DKA.

Because DKA-related cerebral edema presents in a continuum of severities, with varying degrees of edema and varying degrees of mental status changes, deciding at which point a patient should be diagnosed with clinically relevant cerebral edema can be difficult.

When does diabetic ketoacidosis–related cerebral edema typically occur?

DKA-related cerebral edema occurs most commonly after several hours of DKA treatment with insulin and intravenous fluids, but can also occur at the time of presentation to the emergency department, before treatment is administered.

Which children are at greatest risk for diabetic ketoacidosis–related cerebral edema?

Epidemiologic studies demonstrate that DKA-related cerebral edema occurs most frequently in children with severe acidosis and severe hypocapnia as well as marked dehydration (high blood urea nitrogen concentrations In addition, a lesser rise in measured serum sodium concentration during DKA treatment as the serum glucose concentration falls has been identified as associated with greater risk for DKA-related cerebral edema. Because younger children tend to present with more severe acidosis and dehydration, DKA-related cerebral edema is more common in younger patients.

What other complications of diabetic ketoacidosis share some of these symptoms?

Other intracranial complications caused by DKA may present with symptoms similar to DKA-related cerebral edema. These include intracranial thromboses, cerebral infarctions, and cerebral hemorrhage. Seizures during DKA can also (rarely) be related to electrolyte imbalances or hypoglycemia. In addition, severe acidosis and/or severe hyperosmolality may cause depressed sensorium at the time of presentation of DKA.

Differentiating alterations in mental status resulting from metabolic alterations associated with DKA from altered mental status resulting from DKA-related cerebral edema can be difficult. Frequent and ongoing assessment of mental status changes in response to DKA treatment is helpful in this regard. Brain imaging studies may also provide helpful information, although these studies are not always abnormal in DKA-related cerebral edema (see below).

What caused this disease to develop at this time?

The causes of DKA-related cerebral edema are not well understood. Initially, many investigators hypothesized that DKA-related cerebral edema is caused by rapid declines in serum osmolality due to rapid infusion of hypotonic intravenous fluids. Data from clinical and laboratory studies, however, do not suggest that osmotic changes are central to the pathogenesis of DKA-related cerebral edema. Instead, more recent data suggest that DKA-related cerebral edema may possibly be caused by cerebral hypoperfusion before DKA treatment, with additional injury related to reperfusion occurring during treatment with insulin and intravenous fluids. Very recent studies also suggest that elevated levels of inflammatory mediators and other substances that can affect blood-brain barrier function may play a role.

What laboratory studies should you request to help confirm the diagnosis? How should you interpret the results?

DKA-related cerebral edema is a clinical diagnosis and there are no confirmatory laboratory tests. Imaging studies (see below) may be helpful but are not always definitive.

Would imaging studies be helpful? If so, which ones?

If clinically relevant DKA-related cerebral edema is suspected, brain imaging studies (computed tomography [CT] or magnetic resonance imaging [MRI]) should be strongly considered. Imaging studies may detect signs of edema (reduced ventricular size, inapparent basilar cisterns, or intensity abnormalities). More importantly, CT or MRI is important to rule out other causes of mental status changes during DKA, including cerebral infarction, cerebral hemorrhage, and/or cerebral vascular thrombosis.

Note that initial imaging studies in children with DKA-related cerebral edema may be normal and imaging findings consistent with edema may not be present until many hours after the onset of symptoms.In part for this reason, some investigators have hypothesized that edema and increased intracranial pressure may actually be consequences of cerebral injury (e.g., injury related to cerebral ischemia/reperfusion) rather than the main causes of cerebral injury during DKA.

Confirming the diagnosis

A clinical algorithm for diagnosis of DKA-related cerebral edema has been proposed in one study (Muir et al, see references below). This algorithm has not been validated in prospective clinical studies but does provide an extensive list of symptoms along with a classification of the relative importance of each. According to the algorithm, patients should be diagnosed with clinically relevant cerebral edema if they have any one of the "diagnostic" criteria, any two "major" criteria or one "major" and two "minor" criteria.

Each criterion must be observed after the initial fluid resuscitation.

Minor criteria: recurrence of vomiting, headache, lethargy/difficult to arouse from sleep, diastolic blood pressure reading greater than 90 mm Hg, age less than 5 years

Major criteria: altered mentation/fluctuating level of consciousness, sustained heart rate deceleration (> 20 beats/min) not related to improved intravascular volume or sleep state, age-inappropriate incontinence

Diagnostic criteria: abnormal motor or verbal response to pain, decorticate or decerebrate posturing, cranial nerve palsy, abnormal (neurogenic) respiratory pattern

If you are able to confirm that the patient has diabetic ketoacidosis/cerebral edema, what treatment should be initiated?

Mannitol (0.25-1g/kg) is the most frequently used treatment for DKA-related cerebral edema. Mannitol should be given as soon as a clinical diagnosis of DKA-related cerebral edema is made. Anecdotal evidence from case series of children with DKA-related cerebral edema suggests that prompt administration of mannitol may be of benefit, but data from randomized trials or other clinical studies are lacking.

The mechanism of action of mannitol is not well understood. Rapid clinical improvements in patients treated with mannitol suggest that effects other than direct osmotic removal of water from brain parenchyma may be important. Mannitol treatment is thought to improve blood viscosity and cerebral perfusion pressure, thereby improving cerebral blood flow.

Case reports of children with DKA-related cerebral edema also suggest that hypertonic (3%) saline may be beneficial as treatment for increased intracranial pressure. The relative effectiveness of hypertonic saline treatment in comparison to mannitol is unclear.

Intubation with hyperventilation to reduce intracranial pressure is sometimes used in patients with DKA-related cerebral edema. Because mental status is difficult to assess in sedated patients, however, intubation should be used only when absolutely necessary. In addition, one study found an association between intubation with hyperventilation and poor outcomes of DKA-related cerebral edema. For this reason, when possible, PCO2 levels in intubated patients with DKA-related cerebral edema should be maintained at the level expected for the patient's degree of acidosis and reduced further only when other options for treatment of elevated intracranial pressure have failed.

What are the adverse effects associated with each treatment option?

Treatment with mannitol can increase free water and electrolyte loss and result in electrolyte fluctuations. Careful attention to fluid and electrolyte balance after mannitol administration is essential. Mannitol treatment may have variable effects on blood pressure and may initially cause transient mild increases in intracranial pressure.

What are the possible outcomes of diabetic ketoacidosis–related cerebral edema?

DKA-related cerebral edema has high rates of mortality and permanent neurologic morbidity. Approximately 20%-25% of children in whom clinically relevant cerebral edema develops die and another 20%-25% suffer permanent brain injury. Adverse outcomes of DKA-related cerebral edema are more common among children with more profound abnormalities in mental status at the time of diagnosis of DKA-related cerebral edema and in those with higher blood urea nitrogen concentrations at the time of initial presentation with DKA.

What causes this disease and how frequent is it?

DKA-related cerebral edema occurs in 0.3%-0.9% of DKA episodes in children. The causes of DKA-related cerebral edema and cerebral injury continue to be debated. Initial hypotheses focused on the possible role of osmotic changes and infusion of hypotonic fluids, but data from clinical studies generally have not supported a major role for osmotic fluctuations in causing DKA-related cerebral edema. More recent data demonstrate that patients at greatest risk of DKA-related cerebral edema are those with greater dehydration and greater hypocapnia at presentation of DKA, raising the possibility that cerebral hypoperfusion might play a role.

Animal data suggest that cerebral metabolic alterations occurring during DKA are similar to those characteristic of hypoxic/ischemic brain injury. Furthermore, animal data demonstrate reduced cerebral blood flow and cytotoxic edema during untreated DKA. During DKA treatment with insulin and intravenous fluids, cerebral blood flow increases and vasogenic edema develops. Similarities between these findings and those typical of ischemia/reperfusion injury raise the possibility that cerebral injury caused by DKA might be due to cerebral hypoperfusion before treatment and the effects of reperfusion during treatment with insulin and intravenous fluids.

How do these pathogens/genes/exposures cause the disease?

The pathogenesis of DKA-related cerebral edema and cerebral injury is not well understood. It is known that children with greater hypocapnia, greater acidosis, and more severe dehydration are at highest risk for DKA-related cerebral edema. According to one hypothesis, hypocapnia and severe dehydration during DKA lead to reductions in cerebral blood flow. Hyperglycemia may also directly contribute to cerebral hypoperfusion, and animal studies demonstrate reduced cerebral blood flow during hyperglycemia. During DKA treatment with insulin and intravenous fluids, reperfusion of previously hypoperfused cerebral tissues occurs, resulting in reperfusion injury. Hyperglycemia has been shown to worsen many aspects of cerebral reperfusion injury and may similarly play a role in DKA-related cerebral injury.

What complications might you expect from the disease or treatment of the disease?

DKA-related cerebral edema has high rates of mortality and permanent neurologic morbidity. Although some patients recover fully, others may manifest increased intracranial pressure (some with signs consistent with "Cushing's triad": hypertension, bradycardia, and irregular respiration or apnea) and may progress to cerebral herniation. Even in the absence of increased intracranial pressure, however, focal or diffuse cerebral injury may occur.

Approximately 20%-25% of children with clinically relevant DKA-related cerebral edema die and another 20%-25% suffer permanent brain injury. Neurologic damage caused by DKA-related cerebral edema may be focal or more generalized, including intellectual impairments, memory dysfunction, speech/language impairments, hemiparesis/quadriparesis, hypopituitarism, and persistent vegetative state.

How can diabetic ketoacidosis–related cerebral edema be prevented?

At present, it is unclear whether or how DKA-related cerebral edema can be prevented. Much debate has centered on whether DKA-related cerebral edema can be prevented through the use of a particular intravenous fluid (and insulin) protocol for DKA treatment. An ongoing, large, multicenter clinical trial comparing intravenous fluid treatment protocols for pediatric DKA (the Pediatric Emergency Care Applied Research Network FLUID study, scheduled to be completed in 2016) may answer this question, but at present, the answer is unknown.

Because the risk of DKA-related cerebral edema is related to DKA severity (risk is greater in children presenting with greater acidosis, hypocapnia, and dehydration), prompt recognition of DKA and initiation of treatment at an early stage is important. For children with known diabetes, every effort should be made to educate parents and patients about symptoms of ketosis and the importance of prompt treatment when ketosis is detected at home. (See Diabetic Ketoacidosis chapter.)

What is the evidence?

Krane, E, Rockoff, M, Wallman, J. "Subclinical brain swelling in children during treatment of diabetic ketoacidosis". N Engl J Med . vol. 312. 1985. pp. 1147-51.

Glaser, N, Barnett, P, McCaslin, I. "Risk factors for cerebral edema in children with diabetic ketoacidosis". N Engl J Med . vol. 344. 2001. pp. 264-9.

Marcin, J, Glaser, N, Barnet, P. "Clinical and therapeutic factors associated with adverse outcomes in children with diabetic ketoacidosis-related cerebral edema". J Pediatr . vol. 141. 2002. pp. 793-7.

Muir, A, Quisling, R, Yang, M. "Cerebral edema in childhood diabetic ketoacidosis: natural history, radiographic findings and early identification". Diabetes Care . vol. 27. 2004. pp. 1541-6.

Glaser, N, Gorges, S, Marcin, J. "Mechanism of cerebral edema in children with diabetic ketoacidosis". J Pediatr . vol. 145. 2004. pp. 164-71.

Edge, J, Jakes, R, Roy, Y. "The UK case-control study of cerebral oedema complicating diabetic ketoacidosis in children". Diabetologia . vol. 49. 2006. pp. 2002-9.

Yuen, N, Anderson, S, Glaser, N. "Cerebral blood flow and cerebral edema in rats with diabetic ketoacidosis". Diabetes . vol. 57. 2008. pp. 2588-94.

Glaser, N, Yuen, N, Anderson, S. "Cerebral metabolic alterations in rats with diabetic ketoacidosis: effects of treatment with insulin and intravenous fluids and effects of bumetanide". Diabetes . vol. 59. 2010. pp. 702-9.

Lo, W, O'Donnell, M, Tancredi, D. "Diabetic ketoacidosis in juvenile rats is associated with reactive gliosis and activation of microglia in the hippocampus". Pediatr Diabetes. 2015 Jan 16.

Ma, L, Roberts, JS, Pihoker, C. "Transcranial Doppler-based assessment of cerebral autoregulation in critically ill children during diabetic ketoacidosis treatment". Pediatr Crit Care Med. vol. 15. 2014 Oct. pp. 742-9.

You must be a registered member of Clinical Pain Advisor to post a comment.

Sign Up for Free e-Newsletters