How diabetic ketoacidosis develops

Diabetic ketoacidosis (DKA) is a medical emergency. All metabolic disturbances seen in DKA are the indirect or direct consequences of the lack of insulin (Fig 33.1). Decreased glucose transport into tissues leads to hyperglycaemia, which gives rise to glycosuria. Increased lipolysis causes overproduction of fatty acids, some of which are converted into ketones, giving ketonaemia, metabolic acidosis and ketonuria. Glycosuria causes an osmotic diuresis, which leads to the loss of water and electrolytes – sodium, potassium, calcium, magnesium, phosphate and chloride. Dehydration, if severe, produces pre-renal uraemia and may lead to hypovolaemic shock. The severe metabolic acidosis is partially compensated by an increased ventilation rate (Kussmaul breathing). Frequent vomiting is also usually present and accentuates the loss of water and electrolytes. Thus the development of DKA is a series of interlocking vicious circles all of which must be broken to aid the restoration of normal carbohydrate and lipid metabolism.

The most common precipitating factors in the development of DKA are infection, myocardial infarction, trauma or omission of insulin.

Treatment

The management of DKA requires the administration of three agents:

In most cases, rehydration and insulin therapy will correct the metabolic acidosis, and no further therapy is indicated. However, in the most severe cases when the hydrogen ion concentration is greater than 100 nmol/L, IV sodium bicarbonate may be indicated.

The detailed management of diabetic ketoacidosis is shown in Figure 33.2. The importance of good fluid balance charts, as in any serious fluid and electrolyte disorder, cannot be over-emphasized. The initial high input of physiological (0.9%) saline is cut back as the patient’s fluid and electrolyte deficit improves. Intravenous insulin is given by continuous infusion using an automated pump, and potassium supplements are added to the fluid regimen. The hallmark of good management of a patient with DKA is close clinical and biochemical monitoring.

Laboratory investigations

Initially, urine (if available) should be tested for glucose and ketones, and blood checked for glucose using a test strip. Venous blood should be sent to the laboratory for plasma glucose and serum sodium, potassium, chloride, bicarbonate, urea and creatinine. An arterial blood sample should also be sent for measurement of blood gases.

It is important to highlight a clinically important consequence of laboratory methodology here. The presence of ketone bodies in serum interferes with creatinine measurement; therefore serum creatinine can be falsely elevated in the acute stage. Reliable creatinine values are obtained only after ketonaemia subsides.

For reasons that are not entirely clear, amylase activity in serum is also increased in diabetic ketoacidosis. Pancreatitis should be considered as a precipitating factor only if there is persistent abdominal pain.

Blood glucose should be monitored hourly at the bedside until less than 15 mmol/L. Thereafter checks may continue 2-hourly. The plasma glucose should be confirmed in the laboratory every 2–4 hours. The frequency of monitoring of blood gases depends on the severity of DKA. In severe cases it should be performed 2-hourly at least for the first 4 hours. The serum potassium level should be checked every 2 hours for the first 6 hours, while urea and electrolytes should be measured at 4-hourly intervals (Fig 33.3).

Two other forms of severe metabolic decompensation may occur in diabetic patients. These are hyperosmolar non-ketotic (HONK) coma and lactic acidosis. Table 33.1 shows the principal features of these conditions in comparison with DKA.

Hyperosmolar non-ketotic (HONK) coma

Lactic acidosis