Hyperglycemic Diabetic Coma
Samir Malkani
Aldo A. Rossini
David M. Harlan
Michael J. Thompson
John P. Mordes
Introduction
The Acute Metabolic Complications of Diabetes: the Overlap Concept
The most urgent metabolic complications of diabetes are the four diabetic comas: hypoglycemia, diabetic ketoacidosis (DKA), hyperglycemic hyperosmolar syndrome (HHS), and alcoholic ketoacidosis (ethanol-induced hypoglycemia). These diagnostic possibilities must be considered in any lethargic or comatose patient. In addition to being life-threatening conditions, they account for thousands of hospitalizations and substantial costs [1]. Recognition of these diabetic comas is particularly important because these conditions are reversible with appropriate treatment. We use diabetic coma as a generic term that encompasses both frank coma and the milder metabolic abnormalities that precede loss of consciousness. This chapter considers the hyperglycemic crises; hypoglycemia and alcoholic ketoacidosis are discussed in Chapter 106.
 Figure 101.1. Neutralization of ketoacids. Hydrogen ion from ketoacids is neutralized by bicarbonate, producing carbonic acid that then decomposes to H2O and CO2. The latter is expelled by the lungs. The neutralized salts of ketone bodies are excreted in the urine. NAD, nicotinamide adenine dinucleotide; NADH, reduced form of NAD; AcAc, acetoacetate; Beta-OH, β-hydroxybutyrate. |
Although DKA and HHS are discussed separately, it is important to recognize that metabolic decompensation related to hyperglycemia can take many forms depending on the severity of insulin deficiency, underlying genetic predispositions, and intercurrent illnesses. There is frequent overlap in clinical phenotypes, and clinicians should be aware of this concept [1,2]. DKA can occur in a patient with type 2 diabetes; up to a third of patients with HHS have no prior history of diabetes [1]; both DKA and HHS can be complicated by lactic, uremic, or other form of metabolic acidosis, and ketoacidosis itself can occur in the setting of profound hypoglycemia [3]. These metabolic disturbances can overlap to yield both classical DKA and nonclassical presentations of HHS and other ketotic and acidotic states.
In a comatose patient if the blood glucose concentration is less than 50 mg per dL or if for any reason the blood glucose cannot be measured rapidly, the first diagnostic and therapeutic step should be the infusion of 50 mL of a 50% dextrose solution. The hypoglycemic patient who awakens is resuscitated; coma of any other origin is not adversely affected.
Diabetic Ketoacidosis
DKA comprises the triad of hyperglycemia, metabolic acidosis, and ketonemia. Any person with diabetes can develop DKA [4], but it most often occurs in those with type 1 diabetes. Before the discovery of insulin, most patients with type 1 diabetes died of DKA. With the advent of insulin and intensive care, mortality from DKA has fallen to less than 5% [5]. Deaths are associated with intercurrent heart disease or infection in older
patients, cerebral edema in younger patients, and, occasionally, therapeutic errors.
patients, cerebral edema in younger patients, and, occasionally, therapeutic errors.
Pathophysiology and Etiology
Normal Glucose Homeostasis
After a meal, pancreatic islet β cells release insulin into the circulation, enabling fuels to enter cells and activating enzymes for their storage or metabolism. Glucose enters most tissues only in the presence of insulin; erythrocytes, heart, and brain are exceptions. Glucose is stored in liver and muscle as glycogen. Some glucose is metabolized; some is converted into triglyceride.
In adipose tissue, insulin activates lipoprotein lipase, clears lipoproteins from the circulation, and stores them intracellularly. Insulin also inhibits the breakdown and release of previously stored fat. Insulin has similar effects on skeletal muscle, permitting both amino acids and glucose to enter cells for oxidation or storage [6,7].
During starvation, insulin concentrations decrease, catabolic pathways are activated, and stored fuels (glucose, amino acids, and fats) are mobilized to meet energy needs. Liver glycogen provides glucose for only several hours. Muscle glycogen is not directly available due to lack of glucose-6-phosphatase in muscle. To support plasma glucose, muscle glycogen undergoes anaerobic glycolysis, generating lactate that is converted into glucose in the liver. After glycogen stores are exhausted, the liver synthesizes glucose from muscle-derived amino acids through the process of gluconeogenesis [8]. To conserve muscle mass during starvation, glucose consumption is reduced and fatty acids released from adipose tissue become the principal fuel source. Some fatty acids are transformed by the liver into ketoacids [9].
The rate of catabolism is regulated by insulin. As circulating glucose concentration decreases, insulin concentration also decreases—but never to zero. Low insulin levels permit lipolysis and proteolysis while stimulating gluconeogenesis, and maintaining normal glucose concentration. Increased glucose concentration stimulates insulin secretion, which in turn reduces or halts catabolism. Precise regulation of insulin secretion, even in the absence of food intake, achieves continuous control of carbohydrate metabolism.
Abnormal Glucose Homeostasis
DKA can be viewed as a “super-fasted” state that occurs when there is insufficient insulin available to regulate carbohydrate metabolism [7]. Without insulin, glucose no longer enters most cells and is neither stored nor metabolized. Glucagon secretion is increased and hepatic glucose production increases without restraint. When the renal threshold for glucose is exceeded (180 to 200 mg per dL), an osmotic diuresis ensues and water and electrolytes are lost. If insulin deficiency persists, the stress-response hormones cortisol, epinephrine, norepinephrine, glucagon, and growth hormone are released and accelerate catabolism. Glucagon excess is responsible for oxidation of fatty acids to ketone bodies in the liver. Once this happens, DKA ensues with the life-threatening combination of hyperglycemia, acidemia, ketonemia, loss of free water, and depletion of electrolytes.
The cause of ketoacidosis is insulin deficiency. New onset type 1 diabetes commonly presents as ketoacidosis, but most cases occur in individuals known to have diabetes. Dietary indiscretion in a person with known treated diabetes may produce classic hyperglycemia, polydipsia, and polyuria but never ketosis. Ketonuria in any hyperglycemic diabetic patient should suggest the presence of DKA. Such patients must be carefully evaluated for the presence of acidemia. Ketoacidosis occurs most often in patients who have omitted their insulin or who have an intercurrent infection.
Infection and other stressors produce a state of insulin resistance, in part because of the presence of high levels of tumor necrosis factor α; infection may be the most common trigger of DKA in the ICU setting [10]. Severe stress occasionally causes ketosis in patients with type 2 diabetes [4]. African Americans with type 2 diabetes may be particularly susceptible to the development of ketosis [11,12]. Other factors that can precipitate ketosis include acute myocardial infarction, emotional stress, cancer, drugs that interfere with insulin release or action, pregnancy, menstruation, and various endocrinopathies. Occasionally, no precipitating factor can be identified.
Clinical Manifestations
Most patients with DKA are lethargic; about 10% are comatose [13]. They have lost large quantities of fluid; their skin, lips, and tongue are dry; and their eyes are soft to palpation. Postural hypotension is common, but shock is rare [14].
Patients with DKA have rapid deep (Kussmaul) respiration, and their breath has a sweet fruity odor. Some patients with new-onset DKA have been misdiagnosed as having psychological hyperventilation [15]. If a patient with DKA is not tachypneic, the physician should suspect that severe acidosis (pH < 7.1) is depressing the respiratory drive [16].
It is important to measure the temperature accurately. Because the patient is hyperventilating, rectal or tympanic temperature should be measured. Patients with DKA do not have fever unless an intercurrent process, usually infection, is present. Similarly, the rare cases of hypothermia in DKA are associated with sepsis [17].
Abdominal pain is common and may be accompanied by a tender guarded abdomen with diminished or absent bowel sounds. DKA should always be excluded when evaluating abdominal pain [18]. What may appear to be a surgical condition will resolve with correction of the acidosis.
Patients with DKA may be nauseous and vomit guaiac-positive coffee grounds–like material. This is probably due to gastric atony, distention, and rupture of mucosal blood vessels. Pleuritic chest pain may also be present. The cause is unknown, but it resolves with treatment of the DKA.
The nose and sinuses of all patients with DKA should be examined. Acute sinusitis and a black intranasal eschar should suggest mucormycosis, an opportunistic fungal infection that disseminates rapidly in acidotic patients. Mucormycosis is often fatal; survival requires prompt diagnosis [19].
DKA can complicate pregnancy. When DKA in pregnancy is due to new onset of diabetes, due to noncompliance in a woman known to have diabetes, or is complicated by infection, rates of fetal loss are high [20].
Laboratory Diagnosis
Hyperglycemia, acidemia, and ketosis in the appropriate clinical setting are the criteria for the diagnosis of DKA.
Blood Glucose
Normal plasma glucose concentration is 60 to 120 mg per dL (3.3 to 6.7 mmol). Whole blood glucose concentrations are 15% to 20% lower. Fingerstick blood glucose determinations are performed on whole capillary blood, and most meters correct for this offset. Calibrated glucose meters suitable for use in the ICU are accurate over a wide range of concentrations, but very high and low concentrations are less consistently accurate and should be confirmed by a clinical laboratory. Meters intended for home use may give less reproducibly accurate results [21].
In DKA, blood glucose concentration of 400 to 800 mg per dL is typical, but as many as 15% of cases of DKA may present
with blood glucose concentrations less than 300 mg per dL—so called euglycemic DKA [22]. Typically, these are younger patients with a high glomerular filtration rate (GFR). In one series, approximately 1% of patients with DKA presented with a blood glucose concentration less than 180 mg per dL and a bicarbonate concentration less than 10 mEq per L [23]. More often, the solute diuresis causes dehydration, decreases the GFR, and further increases circulating blood glucose concentration.
with blood glucose concentrations less than 300 mg per dL—so called euglycemic DKA [22]. Typically, these are younger patients with a high glomerular filtration rate (GFR). In one series, approximately 1% of patients with DKA presented with a blood glucose concentration less than 180 mg per dL and a bicarbonate concentration less than 10 mEq per L [23]. More often, the solute diuresis causes dehydration, decreases the GFR, and further increases circulating blood glucose concentration.
Electrolyte
Sodium
Serum sodium concentration is quite variable in DKA and must be interpreted in the context of serum glucose and lipid concentrations. If extremely abnormal, it may need special attention during management. Large amounts of sodium are lost during the osmotic diuresis of DKA, and the serum concentration does not necessarily reflect this loss. Because sodium resides principally in the extracellular fluid space, elevated sodium concentration may simply reflect the degree of dehydration and free water loss.
Abnormally low sodium concentrations may be due to the osmotic effect of large amounts of extracellular glucose. The osmotic activity of glucose, drawing free water from the intracellular to the extracellular space, produces a fall of 1.6 mEq per L of sodium for every increase of 100 mg per dL in blood glucose concentration more than 100 mg per dL [24]. The “corrected” serum sodium in a patient with a measured concentration of 135 mEq per L and a glucose concentration of 600 mg per dL is [1.6 × (6 – 1) + 135], or 143 mEq per L. The patient presenting with an elevated serum sodium concentration despite hyperglycemia has a severe total body free water deficit.
It is also important to be certain that abnormally low serum sodium concentrations in DKA are not factitious. Sodium resides only in the aqueous phase of plasma and when the nonaqueous constituents such as triglycerides increase substantially, the reported concentration of sodium will be spuriously low unless “ion-specific” technology is used for the measurement [4].
Chloride
Chloride concentrations are usually not helpful in the diagnosis of DKA, although they may provide useful information. Hyperchloremia may sometimes represent a more chronic ketoacidotic state [25] and may be associated with slower recovery [26]. Extremely low levels of chloride may result from vomiting [27]. Hyperchloremic acidosis can also occur during recovery from DKA as a consequence of the loss of neutralized ketone body salts [28].
Potassium
Potassium is the electrolyte that must be watched most carefully and often during therapy. All patients with DKA are at risk for life-threatening hypokalemia during treatment, despite the fact that the serum potassium concentration is usually elevated at presentation [26,29]. This elevation is due to catabolism of tissue, dehydration, and shifts of potassium from the intracellular to the extracellular space as hydrogen ions are buffered. An initially elevated serum potassium concentration should never obscure the fact that total body potassium loss (in the range of 200 to 700 mEq) occurs in ketoacidosis. The greatest potassium loss accompanies the osmotic diuresis of glucose. Additional losses are due to the excretion of ketone bodies as potassium salts, dehydration-induced secondary hyperaldosteronism, and vomiting. Potassium replacement early in the course of therapy for DKA is always necessary. It should be started as soon as the potassium concentration is at the upper end of the normal range because continued insulin therapy will invariably cause the potassium concentration to fall further. Normal or low concentrations of potassium early in ketoacidosis reflect a very severe potassium deficit.
Table 101.1 Calculations | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||
Magnesium
Like potassium, serum magnesium concentrations in patients with untreated DKA tend to be elevated initially, but they fall with subsequent hydration.
Bicarbonate
Serum bicarbonate concentration is low in ketoacidosis [16] because of neutralization of ketone bodies, which are acids. Bicarbonate buffer in the extracellular compartment represents the first line of defense in acid–base homeostasis. The process is summarized in Figure 101.1. Hydrogen ion (H+) from ketoacids is neutralized by bicarbonate, producing carbonic acid, water, and CO2. As CO2 is expelled through the lungs, the neutralized salts of the ketone bodies are excreted in the urine. In patients with established DKA, the serum bicarbonate concentration is less than 15 mEq per L.
Phosphorous
Elevated serum phosphate concentrations are common in untreated DKA; the mechanism is not clear. After therapy, there is a precipitous decline to subnormal levels [30]. It has been estimated that as much as 1 mM per kg of phosphate is lost during DKA. Hypophosphatemia of less than 0.5 mM per L has been described in both DKA and HHS [30].
Acidosis
Arterial blood gas and pH measurements are essential in the management of all but the mildest cases of DKA. The arterial pH in DKA is almost always less than 7.3. If arterial samples cannot be obtained, venous or capillary samples may be used, although they provide less information [31]. DKA classically presents as an anion gap acidosis. The anion gap should be calculated for all acidemic patients (Table 101.1). In addition to confirming the diagnosis of DKA, the anion gap can be used together with plasma ketone measurements to obtain important additional insight into the nature and severity of a given case [32]. More chronic ketoacidotic states may be associated with hyperchloremic rather than anion gap acidosis [25], probably as a consequence of the loss of neutralized ketone body salts [28]. Rare cases of DKA are complicated by intercurrent metabolic alkalosis, most often from severe vomiting [27,33].
Plasma Ketones and β-Hydroxybutyrate
Plasma ketones should be measured in all comatose patients with diabetes at the time of presentation. When the nitroprusside test is used, the results are usually expressed as the highest dilution of serum that gives a positive reaction. This test is always positive (> 1:2 dilution) in DKA, but its result may not reflect the full extent of ketogenesis. This is because the test measures only acetoacetate (AcAc) and acetone. It does not measure beta-hydroxybutyrate (BOHB), which, although a “ketone body,” is a hydroxyacid and not a ketone (Fig. 101.2). Normally, the BOHB-to-AcAc ratio is 3:1, but acidosis increases the ratio to 6:1 or even 12:1 as pH decreases. The BOHB-to-AcAc ratio at pH 7.1 is at least 6:1.
BOHB can be measured directly, and the test is available in many hospital laboratories. Measurement of BOHB concentration, if the result is available rapidly, can also be used to establish the diagnosis of DKA. The advantage of BOHB measurement derives from the fact that it is the major ketone
body and its concentration is a better indicator of the severity of ketoacidosis.
body and its concentration is a better indicator of the severity of ketoacidosis.
 Figure 101.2. Biochemical interrelationships of the ketone bodies. Acetoacetate and acetone are ketones, whereas β-hydroxybutyrate, although a “ketone body,” is β-hydroxy carboxylic acid and not a ketone. NAD+, nicotine adenine dinucleotide; NADH, reduced form of NAD. |
The results of plasma ketone, BOHB, anion gap, and arterial pH measurements can be used to determine whether a pure or mixed anion gap acidosis is present. The highest positive ketone dilution is multiplied by 0.1 mM per L to obtain an estimate of AcAc concentration; the BOHB can be measured directly. If a patient’s anion gap as calculated in Table 101.1 is greater than the estimated contribution of ketone bodies (AcAc plus BOHB), the presence of an additional unmeasured anion should be considered (e.g., lactate, salicylate, uremic compounds, methanol, or ethylene glycol; see Chapter 119).
Ketone body measurements are also useful for monitoring the resolution of DKA. In cases of severe acidosis, ketones initially rise rather than fall as the acidosis improves. This is due to conversion of BOHB back to AcAc. Clearance of ketone bodies occurs slowly; measurement of ketones and BOHB more often than every 12 hours is generally unnecessary.
It is worth noting that certain newer home blood monitors have the capacity to measure not only glucose but also “ketones.” These meters measure BOHB rather than AcAc using strips distinct from those used to measure glucose. They can warn patients of impending or established ketoacidosis prior to hospital presentation.
Blood Urea Nitrogen and Creatinine
The blood urea nitrogen (BUN) of patients with DKA is typically elevated to values between 25 and 50 mg per dL due not only to prerenal azotemia from volume depletion but also to increased ureagenesis. Patients with DKA are in a state of uncontrolled gluconeogenesis; the large quantities of amino acids released from muscle for conversion to glucose produce hyperaminoacidemia. These amino acids increase substrate availability for ureagenesis. Although the serum creatinine concentration usually reflects the degree of dehydration and prerenal azotemia in DKA accurately [34], spurious elevations occasionally occur because AcAc interferes with some older creatinine assays [35].
Complete Blood Count
Hematocrit and hemoglobin in DKA are usually high and in proportion to the degree of dehydration. Low values suggest preexisting anemia or acute blood loss. A characteristic hematologic finding in DKA is leukocytosis. White blood cell counts in the range of 15,000 to 90,000 per μL with a significant left shift often occur in the absence of intercurrent illness [13,36]. Leukocytosis and a left shift in DKA do not necessarily imply concurrent infection. The absence of leukocytosis suggests possible folic acid or vitamin B12 deficiency.
Triglycerides
Insulin deficiency impairs clearance of lipid from the circulation and accelerates hepatic production of very low-density lipoprotein (VLDL) [6]. In DKA, there is marked elevation of serum triglyceride concentrations that may be clinically obvious in the form of lactescent serum. With insulin therapy, this biochemical derangement reverses. If a patient can eat during the onset of DKA, hyperchylomicronemia may also be present.
Urine
Urinary glucose and acetone should be measured. If pyuria is present, a urine specimen should be sent for culture and sensitivity. To avoid iatrogenic infection, catheterization should be avoided unless the patient is comatose or anuric. A pregnancy test should be performed in women of childbearing age, as pregnancy can precipitate DKA.
Serum Amylase and Lipase
Other Laboratory Findings
Uric acid concentrations may be elevated during acute DKA [39] as a result of impaired renal function or competition from ketone bodies at sites of tubular secretion. Hepatic enlargement with fatty infiltration of parenchymal cells may occur during acute DKA. Increased levels of C-reactive protein and interleukin-6 may be indicative of underlying infection in DKA [40].
Treatment
Patients with severe DKA should be hospitalized in an intensive care unit (ICU). Delaying intensive care greatly increases morbidity, and detaining patients in the emergency room long after the diagnosis is established should be avoided. Treatment should be directed at three main problems—fluid, electrolytes, and insulin—in that order [41].
Recording of Data
The comprehensive flow sheet of vital signs, laboratory data, and treatment that is part of the modern electronic ICU greatly enhances management. For ICUs that do not have advanced capabilities, a comprehensive paper flow sheet is essential to follow the response to therapy.
Fluid Replacement
Fluid and electrolyte therapy always takes precedence over insulin administration in the treatment of DKA. As described later in “Complications” section of this chapter, insulin administration before volume and potassium repletion can cause shock and arrhythmias [42].
The free water deficit in adults with DKA generally ranges between 5 and 11 L, typically about 100 mL per kg, and is due primarily to the osmotic diuresis of glucose [41,43]. Vomiting and hyperventilation may also contribute to water loss. Initial fluid resuscitation should be an infusion of 0.9% saline. Approximately 2 L should be given during the first hour to restore blood volume, stabilize blood pressure, and establish urine flow. Another liter of 0.9% saline can typically be given during the next 2 hours. The subsequent rate of fluid replacement depends on individual clinical circumstances. During the first 24 hours, 75% of the estimated total water deficit should be replaced. Urine flow should be maintained at approximately
30 to 60 mL per hour. Fluid replacement after the first 2 L may be changed to hypotonic 0.45% saline if hypernatremia is present [44].
30 to 60 mL per hour. Fluid replacement after the first 2 L may be changed to hypotonic 0.45% saline if hypernatremia is present [44].
Electrolytes
Sodium, Chloride, and Potassium
Sodium and chloride are replaced together with free water as just described. Potassium must be added to the saline. Because serum potassium concentration does not accurately reflect total body potassium, replacement should be initiated early in treatment. Until the serum potassium concentration is known, replacement should be carried out cautiously. The recommended initial repletion rate is 20 mEq per hour as KCl or K3PO4. When the serum value is known, the rate of potassium administration can be adjusted. If a nasogastric tube is in place, electrolyte losses due to gastric suctioning must also be considered. Typical potassium deficits in DKA are 3 to 5 mEq per kg, but if hypokalemia or normokalemia is present at the time of admission, the deficit may be much higher, up to 10 mEq per kg.
Potassium concentration often falls precipitously after starting therapy. K+ shifts from the extracellular to the intracellular space in the presence of glucose and insulin. As acidemia resolves, buffered intracellular H+ is exchanged for extracellular K+, further lowering the serum potassium concentration. The electrocardiogram can be helpful in monitoring potassium treatment but cannot substitute for serum potassium determinations. A sudden reduction in serum potassium concentration can cause flaccid paralysis, respiratory failure, and life-threatening cardiac arrhythmias. If a patient in mild DKA is alert and able to tolerate liquids, potassium should be given orally.
Phosphate
Depletion of phosphate occurs in DKA. Initially, the concentration of phosphate is elevated, but levels may decrease to less than 1 mM per L within 4 to 6 hours of starting insulin treatment. Persistent severe hypophosphatemia can cause neurological disturbances, arthralgias, muscle weakness with respiratory impairment, rhabdomyolysis, and liver dysfunction [45].
Except when hypophosphatemia is severe (≤ 1.0 mg per dL), however, the need for phosphate replacement in DKA may be more theoretical than real. No studies have demonstrated that replacement of phosphate affects the course or outcome of ketoacidosis [46,47,48].
For treating severe hypophosphatemia, potassium phosphate (20 mEq K+; 16 mM PO43 –) can be added to replacement fluids in place of KCl. Because phosphate deficits in DKA average only 1.0 mM per kg, it is rarely necessary to administer more than one 5-mL ampule of potassium phosphate. Thereafter, potassium should be replaced as KCl. The hazards of parenteral phosphate administration include hypomagnesemia, hypocalcemia, and metastatic calcification [49]. If a patient with DKA can tolerate oral medication, phosphate-containing antacids (e.g., Neutra-Phos®) can be given.
Bicarbonate
Related posts:
Burn Management
Neuro-Oncological Problems in the Intensive Care Unit
Critical Care Problems in Kidney Transplant Recipients
Nutritional Therapy in the Critically Ill Patient
Diagnosis and Treatment of Anxiety in the Intensive Care Unit Patient
Healthy Work Environments: Necessary for Providers and Patients
Full access? Get Clinical Tree
