VOL. 18 NO. 2 Spring 2000


Case Study: Diabetic Ketoacidosis Complications in Type 2 Diabetes

Craig D. Wittlesey, MD

A 48-year-old Hispanic woman with a long history of obesity, diabetes, dyslipidemia, and reactive airway disease presented to the hospital emergency department with a 5-day history of weakness, tactile fever, productive cough, nausea, and vomiting. Patient report and chart review confirmed that 2 years before this presentation, her diabetes had been managed with diet alone. In the past year, glipizide (Glucotrol), metformin (Glucophage), and ultralente insulin were added because of poor glycemic control.

On examination, her temperature was 99.1 F, blood pressure was 98/64 mmHg, pulse was 136, and respirations were 36. There was a strong smell of ketones in the exam room. The patient was drowsy but cogent. Her head and neck exam revealed poor dentition and periodontal disease. Her lung sounds were clear without wheezes or rhonchi. Her heart sounds were normal. The abdominal exam revealed mild epigastric tenderness to deep palpation but no rebound tenderness or guarding. Extremities were well perfused with symmetric pulses.

Laboratory results were remarkable for a room air arterial blood gas with pH of 7.12, pCO2 of 17 mmHg, and bicarbonate of 5.6 mEq/l. Urinalysis revealed 4+ glucose and 3+ ketones. Chemistry panel revealed a glucose of 420 mg/dl, BUN of 16 mg/dl, creatinine of 1.3 mg/dl, sodium of 139 mEq/l, chloride of 112 mEq/l, CO2 of 11.2 mmol/l, and potassium of 5.0 mEq/l. Chest X-ray revealed no infiltrate.


  1. Is this patient experiencing diabetic ketoacidosis (DKA)?
  2. What type of diabetes does this patient have?
  3. What is the etiology of DKA in this patient?
  4. What is the rationale for inpatient treatment?
  5. What are the expectations for out patient treatment?

DKA is defined as a plasma glucose level >250 mg/dl, plasma bicarbonate <15 mEq/l, pH <7.35, ketonemia, and an elevation in the anion gap. This patient clearly meets the criteria for DKA based on her blood glucose of 420 mg/dl, CO2 of 11.2, pH of 7.12, anion gap of 15.8, and obvious ketonemia.

The more intriguing question is what type of diabetes does this patient have? Typically, patients with diabetes have been classified as having either type 1 diabetes and type 2 diabetes based on their presumed etiologies. Type 1 diabetes is often an autoimmune disease associated with islet cell antibodies. It is generally associated with an absolute insulin deficiency and onset before the age of 30. It often occurs in people at or below their expected ideal body weight. In contrast, type 2 diabetes is associated with absence of circulating islet cell antibodies and relative insulin deficiency and occurs in people over age 40 who are likely to be obese.

People with type 1 diabetes may develop DKA due to an absolute deficiency of plasma insulin. In contrast, the expected outcome of decompensated type 2 diabetes is hyperosmolar nonketotic diabetic coma. Several recent studies have identified patients who have pre- and post-hospital courses more consistent with type 2 diabetes but who experience DKA. Relatively young, obese, African Americans seem to constitute the majority of such patients; however, DKA has been reported in American Indian, Hispanic, Japanese, and Caucasian patients with type 2 diabetes. In the majority of cases, no obvious precipitating cause is identified.

Insulin deficiency appears to be a requirement for the development of DKA in type 2 diabetes. Impaired endogenous secretion of insulin and resistance of insulin at the cellular level, presumed to be mediated through downregulation of the glucose-transport system, contribute to this insulin deficiency. This metabolic effect, thought to be a result of persistent hyperglycemia, has been termed "glucose toxicity." Correction of hyperglycemia can lead to dramatic improvements in insulin secretion and cellular activity.

The goal of initial treatment for patients with DKA is the same regardless of the type of diabetes. Correction of volume depletion, improved tissue perfusion, normalization of serum glucose, reduction in serum ketones, and attention to electrolyte balance are the cornerstones of therapy.

Fluid and insulin replacement are requirements. Profound volume depletion is the rule, not the exception; thus 12 liters of normal saline may be given rapidly. An insulin bolus of 0.1 U/kg followed by a constant infusion at 0.1 U/kg/h adjusted to prevent precipitous decline in serum glucose is advised. Insulin infusion should be continued as serum glucose normalizes to potentiate dissipation of ketones and normalization of pH. This can be readily accomplished by adding IV dextrose as the serum glucose approaches 250 mg/dl. Total body depletion of potassium almost always exists. Repletion of potassium should begin before insulin infusion if the initial potassium level is low. A normal or elevated potassium level does not imply an adequate total body potassium content. As in this case, potassium replacement should begin concurrent with insulin and fluid administration. Careful attention to the serum potassium level is mandatory, because shifts in pH and glucose can cause a significant flux in extracellular potassium.

Patients with decompensated type 2 diabetes and DKA will generally require temporary treatment with subcutaneous insulin. Short-acting insulin may be added as a supplement before meals with an intermediate- or long-acting insulin (0.30.4 U/kg) given one to two times daily. Careful follow-up is necessary to prevent hypoglycemia, which may occur as the effects of glucose toxicity wane. The total daily insulin dose may be reduced by 1015% at weekly intervals, provided that glycemic goals continue to be achieved. A significant proportion of patients will be able to discontinue insulin therapy in the weeks to months following their decompensation.

Clinical Pearls

  1. Patients with type 2 diabetes may present with DKA. This condition seems to require some degree of insulin deficiency and is made more likely by concurrent illness.
  2. Prolonged hyperglycemia may lead to impaired endogenous secretion of insulin and reduced efficiency of  insulin at the cellular level, a condition defined as glucose toxicity. This may result in insulin deficiency sufficient to cause DKA.
  3. Hospital therapy for DKA in type 2 diabetes is the same as for type 1 diabetes and hinges on fluid, electrolyte, and isulin repletion.
  4. Outpatient therapy for decompensated type 2 diabetes following DKA will likesly involve the combined use of short- and long-acting insulin.
  5. As the effects of glucose toxicity resove, a substantial proportion of patients with decompensated type 2 diabetes will be able to discontinue insulin and resume the use of oral hypoglycemic agents to achieve glycemic control.


Goldberg RB, Machado R: Atypical ketoacidosis in type 2 diabetes. Hosp Pract 33:105-108, 111-12, 117-18 passim, 1998.

Skyler JS: Insulin therapy in type 2 diabetes: who needs it, how much of it, and for how long? Postgrad Med 101:85-90, 92-94, 96, 1997.

Wilson C, Krakoff J, Gohdes D: Ketoacidosis in Apache Indians with non-insulin-dependent diabetes mellitus. Arch Intern Med 157:2098-2100, 1997.

Westphal SA: The occurrence of diabetic ketoacidosis in non-insulin-dependent diabetes and newly diagnosed diabetic adults. Am J Med 101:19-24, 1996.

Rossetti L, Giaccari A, Defronzo R: Glucose toxicity. Diabetes Care 13:610-30, 1990.

Yabe-Nishimura C: Aldose reductase in glucose toxicity: a potential target for the prevention of diabetic complications. Pharmacol Rev 50:21-33, 1998.

Craig D. Wittlesey, MD, is co-director of the Central Washington, Providence, Diabetes Care Center in Wapato, Wash.

Copyright 2000American Diabetes Association
Updated 4/00
For ADA Related Issues contact

For Technical Issues contact