Special Feature
Treatment Issues in Diabetic Ketoacidosis
By Stephanie Abbuhl, MD, FACEP, and Michael Chansky, MD, FACEP
Diabetic ketoacidosis (dka) is a common, life-threatening complication of diabetes mellitus familiar to all emergency physicians. Despite advances in our understanding of the pathophysiology and treatment of DKA, there remains a 2-5% mortality rate associated with this acute illness.1 Death may be due to severe metabolic derangements, coexisting illness, thromboembolic disease, or complications of treatment.
DKA develops as a result of insulin deficiency and the resultant counterregulatory hormone excess. The primary derangements in DKA are hyperglycemia, elevated serum ketones, and an anion gap metabolic acidosis. Hyperglycemia initially occurs in DKA because there is decreased peripheral use of glucose due to insulin deficiency. Cellular starvation then triggers increased levels of glucagon, catecholamines, cortisol, and growth hormone, which cause an increase in gluconeogenesis and glycogenolysis and worsening hyperglycemia. Since the cells are unable to use glucose, the body responds by breaking down protein and adipose stores to produce a usable fuel. Free fatty acids are released from the breakdown of fats and are transported to the liver where they are converted to ketone bodies. Beta-hydroxybutyrate and aceto-acetate are the primary ketone bodies providing substrate, and their accumulation leads to the anion gap metabolic acidosis seen in DKA.
Hyperglycemia causes an osmotic diuresis, volume depletion, and electrolyte losses. Volume depletion activates the renin-angiotensin-aldosterone system, worsening the renal potassium losses. In the kidney, chloride is retained in exchange for the ketoanions being excreted. The excretion of ketoanions represents a loss of potential bicarbonate and results in a non-anion gap hyperchloremic acidosis, especially evident in the recovery phase of DKA after intravenous NaCl therapy.
In severe insulin deficiency, the production of prostaglandins PGI2 and PGE2 by adipose tissue is increased.2 This prostaglandin release may explain, in part, the nausea, vomiting, and abdominal pain that are frequently seen in DKA patients, especially in children. Vomiting causes further potassium losses and worsens dehydration.
The diagnosis of DKA is often suspected at triage, and therapy with fluids should begin as soon as possible. These patients should be monitored and have at least one large bore intravenous (IV) line with normal saline (NS) started. It is ideal to have a second IV line of 0.45% NS at KVO. It is important to remember that the degree of hyperglycemia need not be profound. A significant number of patients with DKA have glucose concentrations less than 350 mg/dL.3 It has also been shown that DKA can occur in non-insulin dependent diabetic patients, especially in African American populations with new-onset diabetes.4
Fluid administration
Fluid resuscitation is the most important initial step in the treatment of DKA. Rehydration decreases blood glucose and ketone body concentration, reduces the level of counterregulatory hormones, and restores tissue perfusion, which improves the effectiveness of insulin. The average adult patient has a water deficit of approximately 100 mg/kg (5-10 liters).5 NS is recommended initially to prevent an excessively rapid fall in plasma osmolality, potentially causing free water entry into cells that are still hyperosmolar and precipitating cerebral edema. In general, the first liter of NS should be administered over approximately 30-60 minutes. Following this, a second liter is administered over 1-2 hours, the next two liters over 2-6 hours, and two more liters over 6-12 hours. These guidelines must be adjusted according to the individual case, the patient's hemodynamic status, renal function, body size, and response to ongoing therapy. One study has shown that patients without extreme volume depletion may be managed better with moderate fluid replacement (500 cc/hr for 4 hours) than with rapid rates (1 L/hr).6 Moderate replacement rates may avoid the excessive wash-out of ketone bodies that are metabolized to bicarbonate in the presence of insulin. After the initial resuscitation with NS, most authors favor alternating NS with 0.45% NS or using 0.45% NS alone. Overly aggressive fluid replacement should be avoided because of concerns that it may contribute to the development of ARDS and cerebral edema.
Insulin
The continuous infusion of the low-dose regular insulin with an infusion pump is simple and effective, and has become the standard-of-care in most institutions. After the initial fluid bolus, insulin is administered at 0.1 units/kg/hr. Some experts also suggest a bolus of insulin at 0.1 units/kg to initiate insulin therapy, while others feel this is unnecessary.5,7
Ideally, the plasma glucose should fall at a rate of approximately 100 mg/dL/hr. When the plasma glucose level falls to 250-300 mg/dL, glucose should be added to the IV fluid to allow continued insulin infusion until ketonemia has cleared and the anion gap has normalized.8 The initial four hours of management are the most critical; repeat glucose and electrolytes should be measured every 60-90 minutes. It is important to note that the nitroprusside method of measuring ketones in the serum and urine measures accto-acetate and acetone but not the predominant betahydroxybutyrate.5 During the recovery phase, there is conversion of betahydroxybutyrate to aceto-acetate and increasing levels of "ketones" are often measured in the serum and urine despite the appropriate resolution of ketonemia. Therefore, the anion gap should be used to monitor the improvement of DKA.
A small percentage of patients will not respond to low-dose continuous IV insulin. Failure to respond is most often due to infection and, therefore, a careful search for a source of infection should be undertaken.5 The infusion rate should be doubled or an IV bolus given at 0.2-0.4 units/kg.
Potassium
Patients with DKA are usually profoundly total-body potassium deficient due to insulin deficiency, metabolic acidosis, osmotic diuresis, and frequent vomiting. The serum potassium level on presentation can be high, normal, or low and does not necessarily accurately reflect total body stores. The initial serum potassium concentration is usually normal or high due primarily to the exchange of intracellular potassium for extracellular hydrogen ions in acidosis. If potassium concentrations are initially low, this indicates severe total body potassium depletion and large amounts of potassium replacement will be necessary, usually requiring several days for full replacement.
After the initiation of fluids and insulin in the early treatment of DKA, the serum potassium will rapidly fall. This is primarily due to the shift of extracellular potassium into cells due to insulin, but is also exacerbated by dilution with extracellular fluid replacement, correction of acidosis, and increased urinary losses. If careful attention is not paid to frequent monitoring of potassium levels during treatment, life-threatening hypokalemia can occur, causing cardiac arrhythmias, respiratory arrest, paralytic ileus, and rhabdomyolysis.
Virtually all DKA patients require potassium replacement. The important decision is when to initiate the replacement and at what rate. As a general guideline, if the initial serum potassium is greater than 3.5 mEq/L but less than 5.5 mEq/L and urine output has been established, potassium replacement at 10-15 mEq/hr for at least four hours is suggested. If the initial potassium is greater than 5.5 mEq/L, intravenous replacement should be withheld until the potassium concentration is less than 5.5 mEq/L and urine flow is confirmed. If the initial potassium is below 3.5 mEq/L, intravenous potassium replacement should begin with fluid resuscitation prior to insulin administration to avoid life-threatening hypokalemia.
Phosphate
Similar to potassium, most DKA patients have elevated phosphate concentrations, but are total-body phosphate depleted. With insulin therapy, phosphate re-enters the intracellular space and serum phosphate concentrations fall, usually reaching a nadir 24-48 hours after initiation of treatment. Intravenous phosphate administration has been associated with serious complications such as hypocalcemia, hypomagnesemia, metastatic soft tissue calcifications, renal failure, hypotension, and death. Fortunately, complications due to profound hypophosphatemia are rare in the treatment of DKA.8 In addition, IV phosphate has not been shown to improve outcome in these patients.9 In general, there is no role for the routine intravenous replacement of phosphate in the ED. Most moderate hypophosphatemia (1.0-2.5 mg/dL) can be treated with oral phosphate supplements such as Neutra-phos. If IV phosphate replacement is necessary in a patient with severe hypophosphatemia (< 1 mg/dL), this should be done by an experienced physician in a carefully monitored setting.
Magnesium
While it is recommended to monitor serum magnesium on presentation and for the first day of therapy, symptomatic hypomagnesemia appears to be rare. Replacement is only necessary if the serum magnesium is less than 1.2 mg/dL or if the patient is symptomatic. In patients with even mild renal failure, magnesium replacement should be done with extreme caution due to the risks of hypermagnesemia.
Bicarbonate
There is continued controversy surrounding the role of bicarbonate therapy in the treatment of DKA. Despite the lack of evidence, most texts continue to recommend bicarbonate for an arbitrary initial pH level, usually less than 7.1 or 7.0. However, recent studies suggest that the improvement in the metabolic parameters in DKA are either unchanged10 or delayed11 in patients receiving bicarbonate. Potential adverse effects of bicarbonate therapy include worsening hypokalemia, paradoxical cerebrospinal fluid acidosis, impaired oxyhemoglobin dissociation, delayed recovery from ketosis,11 an elevation of lactate levels,12 and possible precipitation of cerebral edema.13 It is useful to remember that acid production stops when insulin and fluids reverse ketogenesis. The metabolism of ketone bodies results in the endogenous production of alkali and the physiologic correction of acidosis. In a recent retrospective review, children with initial pH values as low as 6.73 promptly recovered from DKA without bicarbonate.12 Unless new studies show a clear benefit from bicarbonate in DKA, it seems reasonable to discontinue the routine use of bicarbonate therapy for most, if not all, DKA patients. Some experts feel that bicarbonate should still be considered in cases of profound acidemia and "impending cardiovascular collapse,"7 or in patients with an acidemia due to another cause, in addition to ketoacidosis.
Summary
The successful treatment of DKA depends largely on attention to detail, repeated labs, and patient reassessment in the first 4-6 hours of management. With this kind of careful monitoring, the most appropriate decisions can be made about the replacement of fluids, insulin, and electrolytes for each individual patient. (Dr. Chansky is Acting Chair of the Department of Emergency Medicine, Cooper Hospital, Camden, NJ)
References
1. Basu A, et al. Persisting mortality in DKA. Diabetes Med 1993;10:282.
2. Axelrod L. Diabetic ketoacidosis. Endocrinologist 1992; 2:375-383.
3. Burge MD, et al. Short-term fasting is a mechanism for the development of euglycemic ketoacidosis during periods of insulin deficiency. J Clin Endocrinol Metab 1993;76:1192-1198.
4. Pinhas-Hamiel D, et al. Diabetic ketoacidosis among obese African-American adolescents with NIDDM. Diabetes Care 1997;20:484-486.
5. Kitabchi AE, Wall BM. Diabetic ketoacidosis. Med Clin North Am 1995;79:9-37.
6. Adrogue HJ, et al. Salutary effects of modest fluid replacement in the treatment of adults with diabetic ketoacidosis. JAMA 1989;262:2108-2113.
7. Lebovitz HE. Diabetic ketoacidosis. Lancet 1995; 345:767-772.
8. Foster DW, McGarry JD. The metabolic derangements and treatment of diabetic ketoacidosis. N Engl J Med 1989;309:159.
9. Fisher JN, Kitabchi AE. A randomized study of phosphate therapy in the treatment of diabetic ketoacidosis. J Clin Endocrinol Metab 1983;57:177-180.
10. Morris LR, et al. Bicarbonate therapy in severe DKA. Ann Intern Med 1986;195:836.
11. Okuda Y, et al. Counterproductive effects of sodium bicarbonate in diabetic ketoacidosis. J Clin Endocrinol Metab 1996;81:314-320.
12. Green SM, et al. Failure of adjunctive bicarbonate to improve outcome in severe pediatric diabetic ketoacidosis. Ann Emerg Med 1998;31:41-48.
13. Krane EJ, et al. Subclinical brain swelling in children during treatment of diabetic ketoacidosis. N Engl J Med 1985;315:1147-1151.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content