Show Summary Details
Page of

# (p. 449) Diabetic Ketoacidosis

Diabetic Ketoacidosis
Chapter:
Diabetic Ketoacidosis
DOI:
10.1093/med/9780190862800.003.0078
Page of

PRINTED FROM OXFORD MEDICINE ONLINE (www.oxfordmedicine.com). © Oxford University Press, 2021. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

date: 23 October 2021

1. A. Introduction

1. a. Definition. Diabetic ketoacidosis (DKA) is a life-threatening complication of diabetes mellitus caused by a relative or absolute deficiency of insulin along with an excess of “stress hormones” (i.e., glucagon, epinephrine, or cortisol). This imbalance results in impaired cellular uptake of glucose, along with increased gluconeogenesis, lipolysis, and ketogenesis. Patients present with hyperglycemia, hyperosmolarity, significant dehydration from a glucose-induced osmotic diuresis, and acidosis from ketone production. DKA can occur in both type 1 and type 2 diabetes. The initial management is the same for both disorders.

2. b. Causes. Noncompliance with insulin therapy and inadequate insulin therapy are common causes of DKA. Systemic stressors such as myocardial infarction, stroke, infection, pregnancy, pulmonary embolism, and surgery may also lead to DKA in patients with diabetes.

2. B. Clinical Manifestations of Diabetic Ketoacidosis

1. a. Signs and symptoms. DKA may present with any of the following signs and symptoms.

1. i. Hypotension and tachycardia often reflect volume depletion, but sepsis should always be considered, especially if fever is present.

2. ii. Tachypnea is common and often reflects the hyperventilation needed to compensate for metabolic acidosis.

3. iii. Mild hypothermia is usually present. Thus, a mild temperature elevation serves as a strong indicator of an underlying infection.

4. iv. Neurologic abnormalities, including seizures and altered mental status, may be present.

5. v. Abdominal pain associated with nausea or vomiting can occur as a result of DKA. However, intraabdominal processes (e.g., cholecystitis) must be considered and excluded.

6. vi. Polyuria and polydipsia often precede other symptoms by 1–2 days and reflect the osmotic diuresis generated by glycosuria. The differential diagnosis for polyuria can be remembered as the “6 Ds”:

• MNEMONIC: Differential Diagnoses for Polyuria (“6 Ds”)

• Diabetes mellitus

• Diabetes insipidus

• Diuretics

• Diuretic phase of acute tubular necrosis (ATN)

• Drinking too much

• Darn! Too much calcium

1. b. Laboratory findings

1. i. Blood glucose levels greater than 300 mg/dL are usually found in conjunction with 4+ glucosuria on urinalysis.

2. ii. Ketonemia and ketonuria are present.

3. iii. Anion gap acidosis with a pH less than 7.3 is typical. The absence of an anion gap acidosis usually rules out DKA.

4. iv. Hyponatremia usually results from the hyperglycemia, which exerts an osmotic pull of water into the intravascular space, decreasing sodium concentration. Vomiting with fluid losses accompanied by free water replacement may also contribute to hyponatremia. Significant hyperlipidemia may accompany DKA, resulting in pseudohyponatremia (see Chapter 43).

5. v. Urine ketones. Ketones are concentrated in the urine, so the absence of ketonuria usually rules out DKA. Urine ketones are, however, nonspecific; therefore, a diagnosis of DKA requires other clinical criteria (i.e., acidosis, hyperglycemia).

6. vi. Profound potassium, magnesium, and phosphate depletion are typically present. The acidosis of DKA results in a compensatory movement of hydrogen ions intracellularly pushing potassium ions into the extracellular (i.e., intravascular) space. This movement may “normalize” or even elevate serum potassium despite significant total body depletion.

7. vii. An elevated blood urea nitrogen (BUN) and creatinine level usually result from prerenal azotemia.

8. viii. An elevated white blood cell (WBC) count with a left shift can occur as a stress response.

9. ix. An increased amylase level is common but reflects both salivary and pancreatic sources; therefore, amylase is not a good marker for pancreatitis in patients with DKA.

3. C. Approach to the Patient

1. a. Diagnose DKA. The diagnosis of DKA is made by the presence of acidosis, hyperglycemia, and serum ketonemia. All three components are required for the diagnosis.

2. b. Identify treatable underlying causes

1. i. The possibility of an underlying infection should be investigated, even if the patient does not have a fever. A chest radiograph, urinalysis, and urine and blood cultures should always be considered based on the clinical presentation.

2. ii. Whether to image the abdomen in patients with DKA and abdominal pain is a decision that must be based on clinical judgment and associated findings (e.g., fever, leukocytosis, liver test abnormalities).

3. iii. Patients with diabetes are at increased risk for myocardial infarction (with or without chest pain). This may be the cause, or the result, of DKA and should be considered early in the evaluation of a patient with DKA of uncertain etiology.

4. D. Treatment

1. a. Fluid replacement. Aggressive intravenous (IV) fluid replacement is the first line of action to help correct both volume depletion (5–6 L on average) and hyperglycemia.

1. i. How much? The most common mistake is not giving enough fluids. At least 1 L should be given in the first hour, often followed by 0.5–1 L/hr IV thereafter. Patients with cardiac dysfunction should have frequent lung examinations and oxygen saturation measurements to assess for pulmonary edema.

2. ii. What kind? A useful way of deciding between normal saline and half normal saline is to determine the corrected sodium concentration, which corrects for the dilution caused by hyperglycemia. For every extra 100 mg/dL of glucose over normal (100 mg/dL), the serum sodium concentration needs to be increased by approximately 1.6 mg/dL:

$Display mathematics$

1. 1. Use normal saline if the corrected sodium is less than or equal to 142 mg/dL or if hypotension is present.

2. 2. Use half normal saline in patients with a corrected serum sodium that is greater than 142 mg/dL. In such patients, normal saline (with a sodium concentration of 150 mg/dL) would keep the sodium high and contribute to persistent hyperosmolarity, which is correlated with poor outcomes.

2. b. Insulin. Because patients with DKA are significantly volume depleted, insulin is best given after or with fluid replacement. Giving insulin first may precipitate significant hypotension by moving glucose and water from the already depleted intravascular space to the intracellular space.

1. i. Insulin bolus. Regular insulin in a dose of 0.1–0.3 units/kg should be administered by IV route. An initial dose of 10 units IV avoids overdosage and is usually adequate until an insulin drip can be started.

2. ii. Insulin drip

1. 1. Initiation. A starting dose of 0.1 units/kg/hr is effective and can be titrated to keep the glucose level at 200–300 mg/dL.

2. 2. Continuation. The insulin drip must be continued until the anion gap is back to normal, even if the glucose level has fallen to near normal levels. Once the glucose level is below 250 mg/dL, hypoglycemia may result from insulin administration. Thus, change fluids from normal or half normal saline to 5% dextrose in normal saline or 5% dextrose in half normal saline and continue the insulin drip (adjusting the drip rate to keep the glucose level at 200–300 mg/dL) until the gap closes.

3. 3. Termination. The half-life of IV insulin is only 8 minutes, so subcutaneous insulin should be administered at least 2 hours before the insulin drip is stopped to ensure adequate overlap and bioavailability of insulin. Changing to subcutaneous insulin without adequate overlap may allow DKA to recur.

3. iii. Subcutaneous insulin. Both long-acting and short- or rapid-acting insulins should be used during the transition from IV to subcutaneous insulin therapy. A rapidly acting insulin analog, such as lispro, aspart, or glulisine, is generally preferred over regular insulin. While the patient is nothing per mouth (NPO), glucose level should be checked every 4 hours. However, once they are able to tolerate a diet, blood glucose testing should be switched to before meals and at bedtime. Subcutaneous insulin dosing should be based on oral (PO) intake and blood glucose levels. Optimal glucose control requires a combination of basal long-acting insulin and meal-specific dosing of rapidly acting insulin.

3. c. Electrolyte replacement

1. i. Potassium repletion is essential to prevent hypokalemia and potentially life-threatening arrhythmias. The potassium level will always fall when treating DKA as a result of insulin and fluid replacement and correction of acidosis.

1. 1. A potassium level below 4.5 mEq/L is usually the trigger for initiating replacement therapy. A potassium level below 4 mEq/L should alert you to aggressively replete potassium because the level is likely to continue to fall with insulin and fluid therapy.

2. 2. A potassium level of 4–5 mEq/L is the goal. Often, 20–40 mEq of potassium can be added to each liter of IV fluid, depending on the rate of IV fluid administration. Extra oral or IV potassium can also be given to keep the potassium level in the desired range. Generally, each 10 mEq of potassium given will raise the potassium level approximately 0.1 mEq/L (assuming normal renal function and no major change in volume or acid-base status). Always confirm that the patient has adequate urine output before administering potassium. (Watch your Pee’s and K!)

3. 3. Patients with renal insufficiency, hypotension, or significant volume depletion require careful potassium repletion to avoid hyperkalemia. Hypotensive or significantly volume-depleted patients should not receive standing potassium in their IV fluids because they may inadvertently receive an excessive potassium load if aggressive volume replacement.

2. ii. Magnesium replacement is frequently necessary and will help correct hypokalemia associated with hypomagnesemia. If the magnesium level is less than 1.6 mEq/L, IV magnesium sulfate may be given and repeated as needed. Like potassium, replete magnesium cautiously in the presence of renal failure.

3. iii. Phosphate replacement is often not required, but a level less than 1 mg/dL may be an indication for replacement with potassium phosphate (usually 9–15 mmol given at a maximal rate of 3 mmol/hr, repeated as needed). If the patient is able to take fluids orally, milk is very high in phosphate and an effective treatment of hypophosphatemia.

4. d. pH regulation. Routine therapy with sodium bicarbonate is not indicated because it increases hypokalemia, shifts the hemoglobin-oxygen dissociation curve to the left (increasing tissue hypoxia), increases intracellular acidosis, and does not confer any survival advantage. However, patients with symptomatic acidosis (e.g., cardiac pump dysfunction) or severe asymptomatic acidosis (pH <7) can sometimes be given sodium bicarbonate to keep the pH greater than 7.

5. e. Frequent monitoring is the key to treating patients with DKA.

1. i. An arterial line may be useful if frequent arterial blood gas and electrolyte measurements are needed.

1. 1. If the pH is less than 7.2, it should be followed closely until a steady upward trend is noted. When the pH reaches 7.2, less frequent evaluation can be instituted.

2. 2. Electrolytes (including glucose) are usually checked each hour initially and may be followed by checks every 2–4 hours after improvement is noted.

2. ii. Following electrolytes and anion gap is more useful than following the serum ketones.

## References

Kamel KS, Halperin ML. Acid–base problems in diabetic ketoacidosis. N Engl J Med 2015;372:546–54.Find this resource:

Umpierrez G, Korytkowski M. Diabetic emergencies—ketoacidosis, hyperglycaemic hyperosmolar state and hypoglycaemia. Nat Rev Endocrinol 2016;12:222.Find this resource: