Diabetic Ketoacidosis Case File
Eugene C. Toy, MD, Barry C. Simon, MD, Terrence H. Liu, MD, MHP, Katrin Y. Takenaka, MD, Adam J. Rosh, MD, MS
Case 5
A 19-year-old man is brought to the emergency department (ED) with diffuse abdominal pain, vomiting, and altered level of consciousness. The patient’s symptoms began several days ago, when he complained of “the flu.” His symptoms at that time included profound fatigue, nausea, mild abdominal discomfort and
some urinary frequency. Today he was found in bed moaning but otherwise unresponsive. His past medical history is unremarkable, and he is currently taking no medications. On physical examination, the patient appears pale and ill. His temperature is 36.0°C (96.8°F), pulse rate is 140 beats per minute, blood pressure is 82/40 mm Hg, and the respiratory rate is 40 breaths per minute. His head and neck examination shows dry mucous membranes and sunken eyes; there is an unusual odor to his breath. The lungs are clear bilaterally with increased rate and depth of respiration. The cardiac examination reveals tachycardia, no murmurs, rubs, or gallops. The abdomen is diffusely tender to palpation, with hypoactive bowel sounds and involuntary guarding. The rectal examination is normal. Skin is cool and dry with decreased turgor. On neurologic examination, the patient moans and localizes pain but does not speak coherently. Laboratory studies: the leukocyte count is 16,000 cells/uL, and the hemoglobin and hematocrit levels are normal. Electrolytes reveal a sodium of 124 mEq/L, potassium 3.4 mEq/L, chloride 98 mEq/L, and bicarbonate 6 mEq/L. BUN and creatinine are mildly elevated. The serum glucose is 740 mg/dL (41.1 mmol/L). The serum amylase, bilirubin, AST, ALT, and alkaline phosphatase are within normal limits. A 12-lead ECG shows sinus tachycardia. His CXR is normal.
⯈ What is the most likely diagnosis?
⯈ What is the next step?
ANSWER TO CASE 5:
Diabetic Ketoacidosis
Summary: This is a 19-year-old man with acute diabetic ketoacidosis. The patient has new-onset diabetes mellitus. The constellation of severe hyperglycemia, aniongap acidosis, and ketosis (manifested by the breath odor) is diagnostic. The other findings of dehydration, hyponatremia, hypotension, altered level of consciousness, and diffuse abdominal pain, are typical of a particularly severe episode of diabetic ketoacidosis.
- Most likely diagnosis: Diabetic ketoacidosis
- Next step: Management of the ABCs, including fluid resuscitation, the initiation of insulin therapy, and a careful search for any precipitating or concomitant illness.
- Recognize the clinical settings, the signs and symptoms, and complications of diabetic ketoacidosis.
- Understand the diagnostic and therapeutic approach to suspected diabetic ketoacidosis.
Considerations
This patient’s clinical presentation is typical for diabetic ketoacidosis. Morbidity may result from either underlying precipitating conditions, or from delayed or inadequate treatment. Prompt recognition, effective resuscitation, and diligent attention to fluid, electrolyte, and insulin replacement are essential. (Table 5–1 lists typical laboratory values in DKA). A comprehensive and thoughtful search for associated illnesses, along with frequent reassessment of the patient, will lead to the best outcome.
Approach To:
Suspected Diabetic Ketoacidosis
Diabetic ketoacidosis is a metabolic emergency. A delay in treatment leads to increased morbidity and mortality. In up to one-quarter of patients, DKA is the initial presentation of type I diabetes, so lack of a diabetic history cannot exclude the diagnosis. Most cases occur in patients with type I diabetes, though some patients with type II diabetes may develop DKA during severe physiologic stress. Some patients will present with the classical symptoms of diabetes, such as polyuria, polydipsia, and fatigue. Others will complain more of dyspnea related to the metabolic acidosis, or of the idiopathic but often severe abdominal pain that frequently accompanies DKA. Patients with an underlying infection or other precipitating illness may have symptoms predominantly from that process. As in our case presentation, some patients have such an altered sensorium that a history is entirely unobtainable.
DKA results from an absolute or severe relative lack of insulin, leading to a starvation state at the cellular level. Gluconeogenesis is stimulated even as glucose utilization falls. Hyperglycemia and ketoacidosis cause a profound osmotic diuresis and massive fluid shifts. The diuresis and acidosis cause severe electrolyte disturbances, with wasting of sodium, potassium, magnesium and phosphate. Acidosis, dehydration, hyperosmolality, and insulin deficiency can lead to potassium shifts into the extracellular space, so patients may have significant serum hyperkalemia at presentation, even with massive total body deficits of potassium. Nausea and vomiting can be severe and further cloud the clinical picture with variable superimposed acid-base and electrolyte disturbances.
Diagnosis is based on the triad of hyperglycemia, ketosis, and metabolic acidosis. The major differential diagnosis is hyperosmolar hyperglycemic state (HHS), which can present with very high glucose but slight or no acidosis. Starvation, pregnancy, alcoholic ketoacidosis, and various toxic ingestions can present with elevated serum ketones, but the glucose is normal or low. Patients can be rapidly screened for DKA with a bedside blood glucose measurement and a dipstick urinalysis. Except for the rare anuric patient, the absence of ketones in the urine reliably excludes the diagnosis of DKA. Serum ketones are commonly measured to confirm the diagnosis, but the absolute value is not as helpful because most labs measure only one of the several ketone bodies that may be present. Table 5–2 is a guide to the differential diagnosis of DKA.
Most patients with recurrent DKA have an underlying cause of the episode, and all symptoms require a thorough investigation. In some studies, less than 10% of episodes had no underlying illness. Infection is the most frequent trigger of DKA, but it has resulted from pancreatitis, myocardial infarction, stroke, and many drugs including corticosteroids, thiazides, sympathomimetics including cocaine, and some antipsychotic drugs. In children and adolescents, voluntary cessation of insulin for psychosocial reasons is a frequent and serious cause of DKA.
Patients in DKA can have massive fluid deficits, sometimes as much as 5 to 10 L. Shock is fairly common, and must be promptly treated with crystalloid infusion to
prevent further organ damage. Adults with clinical shock should receive an initial 2-L bolus of normal saline with frequent reassessment. In children, shock is treated with boluses of 20 mL/kg of normal saline. Although overaggressive hydration can present substantial complications later in the course of treatment, this concern takes a back seat to the reversal of shock. Untreated shock promotes multiple-organ dysfunction and contributes further to the severe acidosis seen in DKA.
Insulin is absolutely required to reverse ketoacidosis. Regular insulin is usually given by continuous IV infusion, though frequent IV boluses may be nearly as effective. Intramuscular injections are painful and less reliably absorbed when the patient is in shock. There is no role for long-acting insulin until the ketoacidotic state has resolved. (There are proposed protocols for treating mild and recurrent episodes of DKA with subcutaneous regular insulin, but those protocols have not yet been widely adopted.) The combination of rehydration and insulin will commonly lower the serum glucose much faster than ketones are cleared. Regardless, insulin infusion should continue until the anion gap has returned to normal. Dextrose should be added to the IV infusion when the serum glucose falls to 200 to 300 mg/dL (11.1-16.7 mmol/L) to prevent hypoglycemia, a common complication of treatment.
An insulin dose of 0.1 U/kg/h (5-10 U/h in the adult) is adequate for almost all clinical situations. This is sufficient to achieve maximum physiologic effect. Higher doses are no more effective, but do cause a higher rate of hypoglycemia. An initial bolus of insulin, equal to 1-hour infusion, is commonly given but has not been shown to hasten recovery or provide any other benefit. Insulin boluses are not recommended in pediatric patients. Insulin does bind readily to common medical plastics, so IV tubing should be thoroughly flushed with the drip solution at the start of therapy.
Patients in DKA generally have massive total body deficits of water, sodium, potassium, magnesium, phosphate, and other electrolytes. Specific laboratory values may vary widely depending on the patient’s intake, gastrointestinal and other losses, medications, and comorbid illnesses. It is not usually necessary to calculate exact sodium and water deficits and replacements, except in cases of severe renal failure. Simply reverse shock with normal saline, and then continue an infusion of halfnormal saline at two to three times maintenance. Remember that glucose should be added before the serum glucose falls to the normal range, usually when the glucose level reaches 250 mg/dL.
Potassium deficits are usually quite large, yet the serum potassium at presentation may be low, normal, or even high. If the potassium is elevated initially, look for and treat any hyperkalemic changes on the ECG. Give fluids without potassium until the serum K reaches the normal range, and then add potassium to the IV infusion. If the initial serum K is normal or low, potassium replacement can be started immediately. Magnesium supplementation may be necessary to help the patient retain potassium. Phosphate replacement has not been shown to improve clinical outcomes, but extremely low phosphate levels (<1.0 mg/dL) are known to cause muscle weakness and possibly rhabdomyolysis. For such patients, some of the potassium can be given in the form of potassium phosphate.
Because metabolic acidosis is such a prominent feature in DKA, some clinicians have administered substantial doses of sodium bicarbonate. Many studies have failed to demonstrate any improvement from this treatment even at surprisingly low serum pH values, but even skeptical physicians sometimes encounter a patient who is so acidotic that they feel compelled to give bicarbonate. There are multiple theoretical and observed complications from bicarbonate, including hypernatremia, hypokalemia, paradoxical CSF acidosis, and residual systemic alkalosis. Of course, in the case of hyperkalemia, bicarbonate might be lifesaving and should not be withheld.
Cerebral edema is a rare but devastating complication of DKA, seen most often and most severely in children. It almost always occurs during treatment, and is a leading cause of morbidity and mortality in pediatric DKA. It has been ascribed to the development of cryptogenic osmoles in the CNS to counter dehydration, which then draw water intracellularly during treatment, but this remains unproven. It has been variously associated with overhydration and vigorous insulin therapy, leading many pediatric centers to use extremely conservative, slow and low-dose treatment protocols for DKA. However, at least one large well-controlled study showed that the only reliable predictor of cerebral edema was the severity of metabolic derangements at presentation. Concern for cerebral edema should never be used as an excuse for the undertreatment of clinical shock. Once shock is reversed, pediatric DKA patients should be managed in consultation with an experienced pediatric endocrinologist or intensivist.
COMPREHENSION QUESTIONS
5.1 A 17-year-old adolescent boy who is a type I diabetic is brought in by his parents with concern about diabetic ketoacidosis. He has had several prior episodes of DKA. Which of the following is most diagnostic of DKA?
A. Polyuria, polydipsia, fatigue
B. Hypotension, dehydration, fruity breath odor
C. Hyperglycemia, ketosis, metabolic acidosis
D. Serum blood sugar of 600 mg/dL in the face of high concentrations of insulin
E. Elevated HCO3 and elevated glucose
5.2 A 28-year-old insulin-requiring woman is found in her apartment by her husband. She is stuporous and cannot provide any history. EMS is called and takes the patient to the emergency center, and a diagnosis of severe DKA is made. Her blood pressure is 80/40 mm Hg and heart rate 140 beats per minute. The glucose level is 950 mg/dL, potassium level 6 mEq/L, HCO3 4 mEq/L. Which of the following is the most appropriate initial treatment?
A. Administer 20 units regular insulin intramuscularly, and normal saline at 250 mL/h.
B. Begin an intravenous dopamine drip to raise BP above 90, then insulin at 10 U/h.
C. Initiate normal saline 2 L with KCl 20 mEq/L, insulin 10 U/h.
D. Provide an intravenous normal saline 2 L bolus, and start an insulin drip at 10 U/h.
5.3 The patient in Question 5.2 is undergoing therapy. Which of the following principles is most accurate in the treatment of DKA?
A. Isotonic saline with no dextrose should be used during the hospitalization because the patient is diabetic.
B. Typically, intravenous insulin and dextrose solution will need to be continued until the acidosis has resolved.
C. Potassium replacement is rarely necessary.
D. Sodium bicarbonate is helpful to resolve the anion gap more quickly.
5.4 The physician explains to a 25-year-old man who has recently been hospitalized with DKA that patients in DKA often have other illnesses or precipitating factors that initiated the ketoacidosis. Which of the following is the most common underlying etiology in DKA?
A. Asthmatic exacerbation
B. Cocaine use
C. Cholecystititis
D. Missed insulin doses
E. Urinary tract infection
ANSWERS
5.1 C. The triad of hyperglycemia, ketosis, and acidosis is diagnostic of DKA. Many other conditions cause one or two of the triad, but not all three. Although a fruity breath odor may suggest acetone, it is not reliably present and not all clinicians can distinguish it.
5.2 D. Fluid resuscitation via isotonic crystalloid solution to reverse shock, and IV insulin to reverse ketoacidosis, are the mainstays of therapy. Though most patients will require potassium, it should not be given while the serum K is elevated, and typically not until urine output is seen. Pressors have a limited role until the intravascular volume is restored.
5.3 B. The serum glucose often drops much more rapidly than the ketoacidosis resolves; insulin is necessary to metabolize the ketone bodies but dextrose prevents hypoglycemia. Potassium replacement is usually necessary, but should wait until hyperkalemia is excluded. Bicarbonate does not hasten resolution of DKA.
5.4 E. Many serious illnesses can precipitate an episode of DKA in the susceptible patient, including infection, stroke, myocardial infarction, pancreatitis, trauma, and surgery. Associated or precipitating illness should always be sought diligently. Urinary tract infection is the single most common underlying cause. Missed insulin doses are also common, but less common than infection.
CLINICAL PEARLS
⯈ Hyperglycemia, ketosis, and acidosis confirm the diagnosis of DKA and are enough to start fluids and insulin.
⯈ Patients in DKA are almost always dehydrated and have significant sodium and potassium deficits, regardless of their specific laboratory values.
⯈ Abdominal pain is a common feature in DKA and is usually idiopathic, especially in younger patients.
⯈ Most morbidity in DKA is iatrogenic.
References
Charfen MA, Fernández-Frackelton M. Diabetic ketoacidosis. Emerg Med Clin North Am. August
2005;23(3):609-628, vii.
Kitabchi AE, Umpierrez GE, Murphy MB, Kreisberg RA. Hyperglycemic crises in adult patients
with diabetes: a consensus statement from the American Diabetes Association. Diabetes Care.
2006;29(12):2739-2748.
Levin DL. Cerebral edema in diabetic ketoacidosis. Pediatr Crit Care Med. 2008;9(3):320-329.
Marcin JP, Glaser N, Barnett P, et al; American Academy of Pediatrics. Factors associated with adverse
outcomes in children with diabetic ketoacidosis-related cerebral edema. J Pediatr. 2002;141(6):793-797.
Mazer M, Chen E. Is subcutaneous administration of rapid-acting insulin as effective as intravenous
insulin for treating diabetic ketoacidosis? Ann Emerg Med. 2009;53(2):259-263.
Wolfsdorf J, Craig ME, Daneman D, et al; International Society for Pediatric and Adolescent Diabetes.
Diabetic ketoacidosis. Pediatr Diabetes. 2007;8(1): 28-43.
0 comments:
Post a Comment
Note: Only a member of this blog may post a comment.