DKA Optic Ultrasound

Diabetic ketoacidosis (DKA) is a complication of type 1, or "insulin-dependent," diabetes
(T1DM) and is defined by a high blood glucose level (over 200 mg/dL) coupled with severe
acidosis. Children with T1DM lack adequate insulin production. Without insulin, glucose
cannot be used by cells to produce energy and blood glucose levels increase. As a result,
the body breaks down fat tissue for adequate energy production. This results in ketone and
acid production, and ultimately DKA. Often times DKA is present at the time of diabetes
diagnosis. Recent estimates suggest that nearly 30% of children that present with DKA are
newly diagnosed with diabetes. DKA can also occur as a result of poorly controlled T1DM or
poor adherence to insulin therapy. It is common for such patients to have recurrent episodes
of DKA, which may lead to permanent brain injury.

Cerebral edema (CE), or "brain swelling," can also occur with severe DKA. It is unclear what
causes CE in children with DKA. Proposed mechanisms include changes in blood flow to the
brain, inflammatory mediation, and alteration in the intracellular and extracellular fluid
composition in the brain. CE can be "clinically apparent" or "subclinical" in terms of
symptomatology. Children with clinically apparent DKA-related CE present with neurologic
deficits and altered mental status. These children are at risk for herniation of the brain
stem due to elevated pressures in and around the brain, and ultimately death. The neurologic
manifestations of clinically apparent DKA-related CE typically occur 4-12 hours after the
initiation of medical therapy. However, cases have been reported to occur prior to
initiation of therapy. Although clinically apparent DKA-related CE is rare, occurring in
0.5-1% of children with DKA, subclinical DKA-related CE is much more common with estimates
suggesting that up to 54% of children with DKA have underlying CE that goes clinically
undetected. Early identification of both clinically apparent and subclinical DKA-related CE
is critical as the time between initial neurologic symptoms and clinical deterioration can
be as little as 3 hours. Moreover, early identification of subclinical DKA-related CE may
lead to differences in medical therapy.

Current evaluation for DKA-related CE necessitates a high index of clinical suspicion and
often times such patients receive advanced brain imaging such as computed tomography (CT)
scans. While CT scans can identify CE, this imaging modality exposes patients to ionizing
radiation, is expensive, and may take considerable time in some instances. Moreover, CE can
occur in the absence of acute change on CT and some patients with subclinical CE may go
unidentified. Magnetic resonance imaging (MRI) has been used to identify subclinical
DKA-related CE, however this is often time-consuming, expensive, requires sedation, or may
be unavailable at some centers. Ocular sonography (ultrasound) is an alternative imaging
technique that can provide immediate diagnostic capability at the bedside and minimize
radiation exposure. This technique has been used to rapidly and accurately detect increased
brain swelling through measurement of the optic nerve sheath diameter (ONSD) in a number of
clinical situations including pediatric head trauma, hydrocephalus and ventriculoperitoneal
shunt malfunctions, and altitude sickness in adults.

It is believed, no study has used ocular ultrasound to examine the ONSD in the setting of
diabetes or DKA for identification of clinically apparent or subclinical CE. It is unclear
if patients with T1DM have a fundamentally different baseline ONSD measurement compared to
described normal values. It is also unclear how the ONSD varies during times of
diabetes-related illness, such as high blood glucose levels without criteria for DKA
compared to unequivocal DKA.

Investigators plan to conduct a prospective study including children aged 7-18 years that
present to the Children's Endocrinology Clinic and Children's Emergency Department. The
objective of this study is to assess the utility of sonographic measurements of the ONSD as
a tool for identification of DKA-related CE. Specifically, the primary aim of this study is
to measure and compare the mean ONSD in children with well-controlled T1DM to those
presenting to the Emergency Department (ED) with DKA and to those presenting to the ED with
T1DM and hyperglycemia without criteria for DKA. Investigators hypothesize that the mean
ONSD measurement is smaller in children with well-controlled T1DM compared to children
presenting to the ED with DKA and compared to children with T1DM and hyperglycemia without
criteria for DKA. Secondarily, investigators aim to compare the mean ONSD in children
presenting to the ED with DKA to those presenting to the ED with T1DM and hyperglycemia
without criteria for DKA. The study team hypothesizes that the mean ONSD measurement is
larger in children presenting to the ED with DKA compared children presenting with T1DM and
hyperglycemia without criteria for DKA.

Inclusion Criteria:

Ages 7-18 years English speaking

DKA group recruited from Children's ED. Criteria for DKA:

1. Hyperglycemia >200 mg/dL AND

2. Venous pH <7.30 AND/OR

3. Bicarbonate level <15 AND

4. Either positive urine or serum ketones

T1DM and hyperglycemia without criteria for DKA (recruited from Children's ED). Criteria
for inclusion:

1. Hyperglycemia (>200 mg/dL) not meeting above criteria for DKA

Well-controlled T1DM control group (recruited from Children's Endocrinology Clinic).
Criteria for inclusion:

1. Hemoglobin A1c <8% at time of clinic visit

2. No previous episodes of DKA other than at the time of diagnosis

3. Has never had a documented hemoglobin A1c >10%

Exclusion Criteria:

DKA and patients with T1DM and hyperglycemia without criteria for DKA

1. Treatment with more than 10 mL/kg IV fluids prior to transfer from outside hospital

2. Treatment with insulin prior to transfer from outside hospital

3. Patients with type 2 DM

4. Patients with Hyperosmolar Hyperglycemic Nonketotic State

5. Underlying neurologic condition predisposing to changes in ICP (hydrocephalus,
ventriculoperitoneal shunt, Chiari I malformation, Chiari II malformation,
pseudotumor cerebri, brain tumor)

6. Underlying condition predisposing to changes in intraocular pressure (glaucoma, eye
trauma)

Well-controlled T1DM group with poorly controlled disease, defined as:

1. Hemoglobin A1c >8% OR

2. >1 previous episode of DKA, including DKA at time of diagnosis OR

3. Hemoglobin A1c documented >10% any time in the past