4: Acute metabolic complications

Published on 26/02/2015 by admin

Filed under Endocrinology, Diabetes and Metabolism

Last modified 22/04/2025

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 0 (0 votes)

This article have been viewed 1365 times

Section 4 Acute metabolic complications

Hypoglycaemia

Type 1 diabetes

Hypoglycaemia is the most-feared complication of therapy in insulin-treated patients. Unfortunately, iatrogenic hypoglycaemia is a common and serious hazard of treatment. It has a substantial clinical impact in terms of mortality, morbidity and quality of life. The risk of hypoglycaemia increases as glycaemic control is improved and in intensive insulin therapy it is the principal factor limiting the attainment of lower glycaemic targets.

Iatrogenic hypoglycaemia usually results from a mismatch between:

In the Diabetes Control and Complication Trial, the overall risk of severe hypoglycaemia was increased approximately 3-fold in the intensively treated group. This was observed despite strenuous efforts to exclude patients who were thought to be at high risk of hypoglycaemia. The relation between rate of severe hypoglycaemia and mean glycated haemoglobin (HbA1c) level was inverse and curvilinear. Factors predisposing to severe hypoglycaemia are presented in Table 4.1. However, in a significant proportion of hypoglycaemic episodes, a clear predisposing factor cannot be identified even with very careful scrutiny of the circumstances.

Table 4.1 Risk factors for severe hypoglycaemia

Other causes of hypoglycaemia in insulin-treated patients are presented in Table 4.2. Their effects in individual patients are variable.

Table 4.2 Causes of recurrent hypoglycaemia in insulin-treated patients

Changes in insulin pharmacokinetics

Altered insulin sensitivity Others

The symptoms and signs of acute hypoglycaemia may be divided into two main categories:

Autonomic (adrenergic) – arising from activation of the sympathoadrenal system (Table 4.3). During hypoglycaemia, the body normally releases adrenaline (epinephrine) and related substances. This serves two purposes:

Neuroglycopenic – resulting from inadequate cerebral glucose delivery (Table 4.3). Specific symptoms vary by age, duration of diabetes, severity of the hypoglycaemia and the speed of the decline. The symptoms in a particular individual may be similar from episode to episode, but are not necessarily so and may be influenced by the speed at which glucose levels are dropping, as well as the previous incidence of hypoglycaemia.

Table 4.3 Autonomic, neuroglycopenic and non-specific symptoms and signs in acute hypoglycaemia

Autonomic (adrenergic)
Neuroglycopenia
Non-specific

Under experimental conditions, adrenergic activation occurs at a higher plasma glucose concentration than the level at which cerebral function becomes impaired (2.7 mmol/L). Thus, the patient is often alerted to the falling plasma glucose concentration by adrenergic activation and is usually able to take corrective action.

Hypoglycaemia and the brain

Cerebral function is critically dependent on an adequate supply of glucose from the circulation. Glucose is transported into the brain across the blood–brain barrier by a facilitative glucose transporter protein, GLUT-1. Studies in humans indicate that the rate of glucose transport into the brain can be modified by changes in plasma glucose levels. In particular, antecedent hypoglycaemia causes upregulation of glucose transport, so that more glucose is transported across the blood–brain barrier during subsequent episodes of hypoglycaemia. This adaptive response has important clinical implications.

Cognitive impairment progresses ultimately to loss of consciousness as plasma glucose falls. Seizures and transient focal neurological deficits may occur.

The most serious consequence of acute hypoglycaemia is cerebral dysfunction with the risk of:

Clinically, hypoglycaemia may be usefully graded as follows:

Hypoglycaemia – counter-regulation

The physiological response to acute hypoglycaemia comprises:

Hypoglycaemia unawareness

Hypoglycaemic unawareness is particularly common in patients:

In addition, certain drugs and alcohol may impair a patient’s perception of these symptoms. β-Blockers are designed to blunt the β-effect of adrenaline and related substances. Hence, if hypoglycaemia occurs in someone who is using this type of drug, he or she may have reduced adrenergic warning symptoms such as tremor and palpitations. β-Blockers may also reduce adrenaline’s effect of stimulating the liver to make glucose, and therefore may lead to the hypoglycaemia being slightly more severe and/or more protracted. β-Blockers are not contraindicated in diabetes, but patients should be aware of these possible problems.

During hypoglycaemia, the body normally releases adrenaline and related substances. The reduction in the β-effect of adrenaline reduces the typical symptoms of hypoglycaemia and blunts liver glycogenolysis and gluconeogenesis. As a result the patient may not be aware that their glucose level is low and the liver produces less glucose.

Attenuation of the adrenaline response is usually due to the glycaemic threshold for the response being shifted to a lower plasma glucose concentration. This can be aggravated by previous episodes of hypoglycaemia.

‘Hypoglycaemia begets hypoglycaemia’ – antecedent hypoglycaemia alters the glycaemic threshold for counter-regulatory hormone secretion – the brain becomes ‘used to’ the low glucose concentration. Clinical studies have shown that intensive insulin therapy leads to symptoms that develop at lower plasma glucose levels. Neuronal glucose transporters increase in number in response to repeated hypoglycaemia. As a result, the hypoglycaemic threshold for the brain to signal adrenaline release falls. Consequently, the patient has less time between the onset of symptoms and the development of severe neuroglycopenia.

Recurrent hypoglycaemia, even if asymptomatic, is therefore a contraindication to intensive insulin therapy in patients with type 1 diabetes. Hypoglycaemic unawareness will sometimes disappear when the frequency of hypoglycaemic episodes has declined, but this is not always the case. With care and expert supervision, this may not necessarily compromise overall glycaemic control as judged by glycated haemoglobin concentrations.

As adrenaline release is a function of the autonomic nervous system, the presence of autonomic neuropathy causes the adrenaline release in response to hypoglycaemia to be lost or blunted. However, it is accepted that classical autonomic neuropathy is not usually responsible for loss of hypoglycaemic awareness.

Nocturnal hypoglycaemia

Hypoglycaemia in type 1 diabetes is particularly common during the night and often goes undetected. Prevention of severe nocturnal hypoglycaemic events remains one of the most challenging goals in the treatment of diabetes. With the prevention of severe hypoglycaemia, it is likely that more people would be able to move toward optimal glycaemic control. Insulin-induced hypoglycaemia is also implicated in occasional sudden death in young patients. Prolonged severe hypoglycaemia, often exacerbated by excessive alcohol consumption, may produce cerebral oedema and permanent brain damage. The risk of hypoglycaemia bars insulin-treated diabetics from certain occupations.

The frequent occurrence of nocturnal hypoglycaemia – which may affect more than 50% of patients and often goes unrecognized – may be an important cause of hypoglycaemia unawareness. In addition, the physiological responses to insulin-induced hypoglycaemia are impaired during stages 3–4 (slow wave) of sleep.

Conventional strategies to minimize nocturnal hypoglycaemia include:

Nocturnal hypoglycaemia often due to poor dietary compliance, and excess alcohol intake has been implicated in the sudden death of some young patients (so-called ‘dead in bed’ syndrome). The cause remains unproven, but catecholamine-mediated falls in plasma potassium concentration leading to cardiac arrhythmias have been implicated.

Treatment of insulin-induced hypoglycaemia

Grade 3–4 hypoglycaemia

Friends, colleagues or relatives may recognize the development of hypoglycaemia before patients themselves. A subtle change in appearance or behaviour may prompt a third party to encourage oral carbohydrate consumption. Unfortunately, cognitive dysfunction may lead to a negative or even hostile response. If the level of consciousness falls, it often becomes hazardous to try forcibly to administer carbohydrate by mouth. Alternatives include:

Recovery from hypoglycaemia may be delayed if:

If the development of cerebral oedema is suspected, computed tomography of the brain should be undertaken and the adjunctive treatment considered: intravenous dexamethasone (4–6 mg 6-hourly) or mannitol. However, evidence for the efficacy of these drugs, or for other measures such as controlled hyperventilation, is poor.

Behavioural problems specific to type 1 diabetes

Type 2 diabetes

Most of the research into hypoglycaemia has looked at hypoglycaemia in the insulin-deficient type 1 diabetic population. The occurrence of hypoglycaemia in the treatment of type 2 diabetes is well recognized, but is more protean in nature, having different risk factors and clinical features according to the nature of the antihyperglycaemic therapy, the extent of the insulin secretory deficit and the duration of diabetes.

The most common cause of hypoglycaemia in type 2 diabetes, resulting in significant physical and psychological morbidity, is iatrogenic, occurring with the use of insulin secretagogues (particularly sulphonylureas) and insulin therapy. These may overwhelm the normal defences that should protect against a significant fall in plasma glucose concentration, primarily by preventing a fall in circulating insulin. Risk of severe hypoglycaemia is further increased by any defects in the other systems for maintaining glucose concentrations. For example, defects in glucagon responses to hypoglycaemia develop in type 2 diabetes along with defects in the other stress responses. Specific therapies may worsen these defects; for example, sulphonylurea therapy sustains intrapancreatic insulin levels during hypoglycaemia, which may further impair glucagon responses.

Risk factors for individual episodes of hypoglycaemia in patients with type 2 diabetes include behavioural, physiological and therapeutic factors. The most common behavioural factor that precipitates individual episodes of severe hypoglycaemia is missed or irregular meals. Other lifestyle factors include alcohol, exercise and incorrect use of glucose-lowering medication (dose/timing).

Therapeutic/physiological factors associated with increased risk include older age, duration of diabetes, presence of co-morbidities, renal impairment, loss of residual insulin secretion, defective counter-regulation and loss of awareness of hypoglycaemia. The use of other medications may also increase risk.

Patient age also affects subjective awareness of hypoglycaemia. In the elderly, neuroglycopenic symptoms specifically related to articulation and coordination, which include unsteadiness, blurred or double vision, lack of coordination and slurred speech, are more common. There are experimental data to show that the adrenergic symptoms of hypoglycaemia decline with increasing age, whereas the tendency for cognitive dysfunction in hypoglycaemia increases.

Time of day is also important – even in the absence of pharmacological therapy, the lowest plasma glucose level of the day is just before the evening meal and unsuspected hypoglycaemia can occur at this time once drug therapy is started. Intensification of treatment targets will increase the risk of severe hypoglycaemia, although this effect will depend on the nature of the treatment used and the degree of insulin deficiency in the patient.

In addition, in type 2 diabetes hypoglycaemia has also been suggested as a private experience that is often not discussed with health-care providers. Hypoglycaemia may be a much greater burden for patients than health professionals are aware. Type 2 patients reported in a recent UK survey that they often received very little information and guidance from health professionals regarding adverse effects of their glucose-lowering regimens.

In patients on lifestyle adjustment and/or insulin-sensitizing treatments, the risk of hypoglycaemia is negligible. Based on patient reporting, the UK Prospective Diabetes Study (UKPDS) 73 showed rates of 0.1% and 0.3% for lifestyle and metformin, respectively, in patients receiving diet alone or monotherapy for 6 years from diagnosis. The recent Diabetes Outcome Progression Trial (ADOPT) reported rates of around 10% in patients on insulin sensitizers (metformin or a thiazolidinedione) over the 5 years of treatment (again all self-reported). Severe episodes were reported in very few patients (0.1%) on either treatment. Patients are more at risk of hypoglycaemia when an insulin sensitizer is combined with insulin or insulin secretagogues.

With the increasing drive for more strict glucose control in type 2 diabetes and new therapies that may carry different risks for hypoglycaemia from established therapies, a review of the risks of hypoglycaemia in type 2 diabetes is appropriate.

Risk factors for severe sulphonylurea-induced hypoglycaemia

Hypoglycaemia rates with the third-generation sulphonylureas (e.g. glimepiride), second-generation sulphonylureas (glipizide and gliclazide) and the metiglinides (e.g. repaglinide and nateglinide) appear to be lower than those with glibenclamide and chlorpropamide. This is thought to be related partly to duration of action, but there may be other contributory factors. For example, agents predominantly excreted via the kidney (e.g. glibenclamide) should be avoided in renal impairment and in the elderly. There is also some evidence for differential effects on insulin sensitivity.

The risk of hypoglycaemia in patients taking an α-glucosidase inhibitor or thiazolidinedione appears to be low, but this risk is increased significantly if these drugs are taken along with a sulphonylurea.

Mild symptomatic hypoglycaemia is not reported to have any serious clinical effects, apart from the potential for inducing defects in counter-regulatory responses and impaired awareness to subsequent hypoglycaemia. Nevertheless, people with diabetes are fearful of hypoglycaemia and even clinically trivial events may be sufficient to inhibit concordance with therapy.

Rates of hypoglycaemia with insulin vary according to the regimen and the stage of evolution of the person’s diabetes. In the UKPDS, patients who were newly diagnosed at the start of the study and randomized to insulin therapy reported ‘any’ hypoglycaemia rates of around 33% at year 1 and around 43% at year 10. Corresponding rates for severe episodes (defined as episodes requiring third-party help or medical intervention) were approximately 1.2% and 2.2%, respectively. In contrast, there is also some evidence to suggest that the frequency and severity of hypoglycaemia reduces with time as β-cell function deteriorates.

Newly emerging therapies based around enhancement of incretin action result in improved glycaemic control through a variety of mechanisms. The incretin, glucagon-like peptide (GLP)-1, released from the small intestine after eating, enhances insulin responses to glucose, as well as suppressing glucagon postprandially. The glucose-lowering effect of GLP-1 analogues and gliptins are glucose-dependent and therefore the hypoglycaemic risk of these agents appears much less.

Management

Recurrent symptoms suggestive of hypoglycaemia should prompt a reduction in dose or withdrawal of the medication responsible. Patients with severe sulphonylurea-induced hypoglycaemia should be admitted promptly to hospital; relapse following initial resuscitation with oral or intravenous glucose may necessitate prolonged infusions of dextrose. Delivery of an intravenous bolus of dextrose – a potent insulin secretagogue – will stimulate further insulin release, especially in patients with relatively well-preserved β-cell function. This predictable consequence of treatment, combined with the long duration of action of drugs such as chlorpropamide and glibenclamide, explains the tendency for hypoglycaemia to recur.

Treatment of hypoglycaemia in type 2 diabetes

The antihypertensive agent, diazoxide, and the somatostatin analogue, octreotide, offer a more direct approach by inhibiting stimulated endogenous insulin secretion; these drugs have been used successfully as adjuncts to intravenous dextrose. However, neither is licensed for this indication in the UK.

As high plasma insulin levels increase the transport of potassium into cells, serum potassium levels should always be monitored and intravenous supplements administered if hypokalaemia develops.

Diabetic ketoacidosis

Diabetic ketoacidosis (DKA) is a potentially life-threatening complication in type 1 diabetic patients. DKA results from an absolute shortage of insulin; in response, the body switches to burning fatty acids and producing acidic ketone bodies, which cause most of the symptoms and complications.

DKA may be the first symptom of previously undiagnosed diabetes, but it may also occur in known diabetics for a variety of reasons, such as poor compliance with insulin therapy or intercurrent illness. Vomiting, dehydration, deep gasping breathing, confusion and occasionally coma are typical symptoms. DKA is diagnosed with blood and urine tests; it is distinguished from other, rarer forms of ketoacidosis by the presence of high blood sugar levels.

DKA can affect patients in any age group. In some published series, female patients predominate. A small subgroup, predominantly females under the age of 20 years, may present with multiple episodes. Many episodes of DKA are avoidable with appropriate early action.

Mechanism

DKA arises because of absolute lack of insulin in the body. The lack of insulin and corresponding excess of glucagon leads to increased release of glucose by the liver from glycogen through gluconeogenesis. High glucose levels spill over into the urine, taking water and solutes (such as sodium and potassium) along with it in a process known as osmotic diuresis. Renal gluconeogenesis is also enhanced in the presence of acidosis. Glucose disposal by peripheral tissues such as muscle and adipose tissue is reduced by deficiency of insulin while raised plasma levels of catabolic hormones and fatty acids induce tissue insulin resistance. This whole process can lead to significant dehydration.

The absence of insulin leads to the release of free fatty acids from adipose tissue. Fatty acids are the principal substrate for hepatic ketogenesis and are converted to coenzyme A (CoA) derivatives prior to transportation into the mitochondria by an active transport system (the carnitine shuttle). In DKA, the hormonal imbalance strongly favours entry of fatty acids into the mitochondria and the preferential formation of ketone bodies:

Within the mitochondria, fatty acyl-CoA undergoes β-oxidation resulting in the formation of acetyl-CoA, which is then oxidized completely in the tricarboxylic acid cycle, utilized in lipid synthesis, or partially oxidized to ketone bodies. Acetone is formed by the spontaneous decarboxylation of acetoacetate. Although acetone concentration is increased in ketoacidosis, it does not contribute to the metabolic acidosis. Acetone is highly fat-soluble and is excreted slowly via the lungs.

In the absence of insulin-mediated glucose delivery, β-hydroxybutyrate can serve as an energy source for the brain, and is possibly a protective mechanism in case of starvation. The ketone bodies have a low pH and therefore cause a metabolic acidosis. The body initially buffers this with the bicarbonate buffering system, but this is quickly overwhelmed and other mechanisms such as hyperventilation compensate for the acidosis, to lower the blood carbon dioxide levels. This hyperventilation, in its extreme form, may be observed as ‘Kussmaul respiration’. In addition, ketones participate in osmotic diuresis and lead to further electrolyte losses.

Signs and symptoms

The symptoms of an episode of diabetic ketoacidosis usually evolve over the period of about 24 hours. Predominant symptoms are nausea and vomiting, pronounced thirst, excessive urine production and, particularly in the young, abdominal pain that may be severe. Those who measure their glucose levels themselves may notice hyperglycaemia. In severe DKA, breathing becomes laboured and of a deep, gasping character – Kussmaul respiration. Coffee ground vomiting (vomiting of altered blood) occurs in a minority of patients; this tends to originate from erosion of the oesophagus. In severe DKA, there may be confusion, lethargy, stupor or even coma.

On physical examination there is usually clinical evidence of dehydration, such as a dry mouth and decreased skin turgor. If the dehydration is profound enough to cause a decrease in the circulating blood volume, tachycardia and low blood pressure may be observed. Often, a ‘ketotic’ odour is present, which can be described as ‘fruity’ and is often compared to the smell of pear drops. If Kussmaul respiration is present, this is reflected in an increased respiratory rate. The abdomen may be tender to the point that an acute abdomen may be suspected, such as acute pancreatitis, appendicitis or gastrointestinal perforation.

Eruptive xanthomata and lipaemia retinalis are recognized complications that respond to treatment of the ketoacidosis. Serum transaminases and creatine phosphokinase may be non-specifically raised in DKA.

Diagnosis

Biochemical assessment Confirm diagnosis by bedside measurement of:

Venous blood is withdrawn for laboratory measurement of:

An arterial blood sample (corrected for hypothermia) is taken for:

Repeat laboratory measurement of blood glucose, electrolytes, urea, gases at 2 and 6 h

Other investigations As indicated by the circumstances:

BUN, blood urea nitrogen; DKA, diabetic ketoacidosis; ECG, electrocardiography; PCO2, partial pressure of carbon dioxide; PO2, partial pressure of oxygen.

In addition to the above, blood samples are usually taken to measure urea and creatinine (markers of renal function, which may be impaired in DKA as a result of dehydration) plus electrolytes. Furthermore, markers of infection (complete blood count, C-reactive protein (CRP)) and acute pancreatitis (amylase and lipase) may be measured. Given the need to exclude infection, chest radiography and urinalysis are usually performed. However, DKA is associated with the release of various cytokines that lead to increased markers of inflammation (increased white cell count and CRP) even in the absence of infection.

If, as a result of confusion, recurrent vomiting or other symptoms, cerebral oedema is suspected, computed tomography of brain should be performed to assess its severity and to exclude other problems such as stroke.

Criteria

DKA is distinguished from other hyperglycaemic diabetic emergencies by the presence of large amounts of ketones in blood and urine, and marked metabolic acidosis. Hyperosmolar non-ketotic state (HONK) or hyperglycaemic hyperosmolar syndrome (HHS) is much more common in type 2 diabetes and features increased plasma osmolarity (above 350 mosmol/kg) due to profound dehydration and concentration of the blood. Mild acidosis and ketonaemia may occur in HONK, but not to the extent observed in DKA. There is a degree of overlap between DKA and HONK, as in DKA the osmolarity may also be increased. In most situations it is relatively easy to classify a case as either DKA or HONK.

Ketoacidosis is not always the result of diabetes. It may also result from alcohol excess and from starvation; in both states the glucose level is normal or low. Metabolic acidosis may occur in diabetic patients for other reasons, such as poisoning with ethylene glycol or paraldehyde.

A 2006 American Diabetes Association statement (for adults) categorized DKA into one of three stages of severity:

Metabolic acidosis has a number of potential serious pathological effects:

DKA usually presents with an anion gap acidosis (anion gap typically 25–35 mmol/L). A wide variety of acid–base disturbances have been reported. Some causes are listed shown in Table 4.5. The anion gap is increased when plasma:

Table 4.5 Causes of anion gap acidosis

image

Potassium is not included because the plasma level of this ion may be altered significantly by acid–base disturbances. Proteins, phosphate, sulphate and lactate ions account for the normal anion gap of about 10–15 mmol/L. When the anion gap is increased, measurement of the plasma concentration of specific anions (e.g. ketone bodies, lactate) may confirm the aetiology of the acidosis.

Management

Hyperglycaemia

Delays in initiating therapy may be disastrous and the diagnosis should be considered in any unconscious or hyperventilating patient.

The main aims in the treatment of diabetic ketoacidosis are replacing the lost fluids and electrolytes while suppressing the high blood sugars and ketone production with insulin (Table 4.6). Admission to an intensive care unit or similar high-dependency area for close observation may be necessary.

Table 4.6 Guidelines for the management of severe diabetic ketoacidosis in adults

Fluids and electrolytes
Volumes:

Fluids:

Potassium:

Insulin By continuous intravenous infusion:

Other points

CT, computed tomography; CVP, central venous pressure; DKA, diabetic ketoacidosis; ECG, electrocardiography; IPPV, intermittent positive-pressure ventilation.

* 1.26% sodium bicarbonate (NaHCO3) = 12.6 g NaHCO3, 150 mmol each of Na+ and HCO3/L.

Fluid replacement

Despite a proportionally greater loss of body water, plasma sodium concentrations are usually normal or low (Table 4.7). However, in DKA, plasma electrolyte concentrations may be falsely depressed by grossly raised plasma lipid concentrations. Plasma should therefore be inspected for turbidity.

Table 4.7 Average electrolyte deficits in adults with diabetic ketoacidosis

Electrolyte Deficit (mmol)
Sodium 500
Chloride 350
Potassium 300–1000
Calcium 50–100
Phosphate 50–100
Magnesium 25–50

The amount of fluid depends on the estimated degree of dehydration (Table 4.8). If dehydration is so severe as to cause hypovolaemic shock or a depressed level of consciousness, rapid infusion of saline is recommended to restore circulating volume. When the circulating volume has been restored, this should be followed by gradual correction of interstitial and intracellular fluid deficits. Overenthusiastic fluid replacement may lead to respiratory distress syndrome and/or cerebral oedema.

Table 4.8 Assessment of hydration (over-estimation of dehydration is dangerous)

Degree of dehydration Clinical signs
Mild, 3% Only just clinically detectable
Moderate, 5% Dry mucous membranes, reduced skin turgor
Severe, 8% As above, plus sunken eyes, poor capillary return
+ Shock May be severely ill with poor perfusion, thready rapid pulse (reduced blood pressure is not likely and is a very late sign)

Table 4.9 Initial fluid replacement

Fluid Rate (mL/h) Time (h)
0.9% sodium chloride 1 L 1000 1
0.9% sodium chloride 1 L with potassium chloride 500 2
0.9% sodium chloride 1 L with potassium chloride 500 2
0.9% sodium chloride 1 L with potassium chloride 250 4
0.9% sodium chloride 1 L with potassium chloride 250 4
0.9% sodium chloride 1 L with potassium chloride 250 4
0.9% sodium chloride 1 L with potassium chloride 125 8
Total 7 L 25

Insulin

The aims of insulin treatment in DKA are:

Some guidelines recommend a bolus of insulin of 0.1 unit insulin per kilogram of body weight. This may be administered immediately after the potassium level is known. Insulin administration can lead to dangerously low potassium levels (see below). In order to reduce the theoretical risk of cerebral oedema, some guidelines recommend delaying the initiation of insulin until 1 h after fluids have been started.

In adults, insulin can be given at rate of around 6 units per hour to reduce the blood sugar level and suppress ketone production. Guidelines differ as to what dose to use when blood sugar levels start falling; some recommend reducing the dose of insulin (to 3 units/h) once the glucose concentration falls below 14.0 mmol/L. Below this blood glucose level it is often useful to start a glucose/K/insulin infusion.

Blood glucose concentrations tend to fall more slowly during treatment in patients with higher levels of catabolic hormones caused by infection or other serious illnesses.

Cerebral oedema

Cerebral oedema, which is the most dangerous complication of DKA, is probably the result of a number of factors (Table 4.11). Some authorities maintain that it is the result of over-vigorous fluid replacement, but the complication may develop before treatment has been commenced. It is more likely to occur in those with more severe DKA, in children, and in the first episode of DKA. Likely factors in the development of cerebral oedema are dehydration, acidosis, increased level of inflammation and coagulation. Together these factors lead to decreased blood flow to parts of the brain, which may then swell once fluid replacement is commenced. The swelling of brain tissue leads to raised intracranial pressure, which is reflected in a rising blood pressure and a falling heart rate, and ultimately herniation, where the swollen brain compresses vital structures in the brain stem, leading to death.

Table 4.11 Complications of diabetic ketoacidosis

Complication Clinical findings
Cerebral oedema This is unpredictable, occurs more frequently in younger children and newly diagnosed diabetics, and has a mortality rate of around 25%. Producing a slow correction of the metabolic abnormalities reduces the risk
Hypokalaemia This is preventable with careful monitoring and management
Aspiration pneumonia Use a nasogastric tube in semiconscious or unconscious patients
Gastric stasis A succussion splash may be evident on abdominal examination
Adult respiratory distress syndrome ARDS has been reported in patients with DKA, usually in patients under 50 years of age. Clinical features include dyspnoea, tachypnoea, central cyanosis and non-specific chest signs. Arterial hypoxia is characteristic and chest X-ray reveals bilateral pulmonary infiltrates. Management involves respiratory support with IPPV and avoidance of fluid overload
Thromboembolism Thromboembolic complications can cause mortality in DKA as a consequence of:

DIC has also been reported as a rare complication. The role of routine anticoagulation has not been clearly established in DKA and in the absence of other risk factors is not generally recommended

Rhinocerebral mucormycosis Rarely, an aggressive opportunistic fungal infection develops in patients with DKA or other metabolic acidoses. The lesion arises in the paranasal sinuses and rapidly invades adjacent tissues (nose, sinuses, orbit and brain). Treatment comprises correction of acidosis, surgical excision of affected tissue and parenteral antifungal agents. The course is often fulminant and the condition carries a high mortality rate Rhabdomyolysis Increased plasma myoglobin and creatine kinase concentrations may occur in DKA. However, clinically important renal complications of rhabdomyolysis are uncommon. Acute renal failure caused by rhabdomyolysis may be somewhat more common in the diabetic hyperosmolar non-ketotic syndrome but is nonetheless a rare complication

DIC, disseminated intravascular coagulation; DKA, diabetic ketoacidosis; IPPV, intermittent positive pressure ventilation.

Cerebral oedema, if associated with coma, necessitates admission to intensive care, artificial ventilation and close observation. The administration of fluids should be slowed. The ideal treatment of cerebral oedema in DKA is not established, but intravenous mannitol and hypertonic saline (3%) are often used (as in some other forms of cerebral oedema) in an attempt to reduce the swelling.

Diabetic hyperosmolar non-ketotic syndrome – hyperosmolar hyperglycaemic syndrome

Diabetic HONK syndrome is a characterized by:

It occurs in type 2 diabetes, is much less common than DKA, and is associated with a much higher mortality rate. The high mortality rate in part reflects the high incidence of serious associated disorders and complications. The preferred term used by the American Diabetes Association is hyperosmolar hyperglycaemic state (HHS).

Diagnosis

The insidious nature of the condition often leads to delays in diagnosis. The hyperosmolar non-ketotic syndrome must be considered in the differential diagnosis of any patient presenting with otherwise unexplained impairment of consciousness or focal neurological signs, dehydration or shock.

Table 4.12 Plasma osmolality

Plasma osmolality (the osmotic pressure exerted by a fluid across a membrane) can be measured in the laboratory (e.g. by freezing point depression) and estimated using the formula:
Plasma osmolality (mosmol/L) = 2 × (plasma Na+ + plasma K+) + plasma glucose + plasma urea
(where Na+, K+, glucose and urea are in mmol/L)
The values for Na+ and K+ are doubled to allow for their associated Cl anions.

Although total body sodium is reduced, plasma sodium concentration at presentation may be low, normal or high, depending on the degree of concomitant water deficit. The degree of dehydration is generally greater than in DKA. As in DKA, hypertriglyceridaemia and hyperglycaemia may falsely depress the sodium concentration.

Although renal impairment may lead to some retention of H+ ions, and hypotension may produce a degree of lactic acidaemia, plasma bicarbonate is usually above 15 mmol/L. Non-traumatic rhabdomyolysis may occasionally be severe enough to precipitate acute renal failure.

Treatment

Successful management of HONK depends on good general care of the unconscious patient and prompt recognition and treatment of underlying causes.

Lactic acidosis

Lactic acidosis is a condition characterized by low pH in body tissues and the blood accompanied by the build-up of lactate, and is considered a distinct form of metabolic acidosis. The condition typically occurs when cells receive too little oxygen (hypoxia), for example during vigorous exercise. In this situation, impaired cellular respiration leads to lower pH levels. Simultaneously, cells are forced to metabolize glucose anaerobically, which leads to lactate formation. Therefore, raised lactate levels are indicative of tissue hypoxia, hypoperfusion and possible damage. Normal fasting blood lactate concentrations range from approximately 0.5 to 1.5 mmol/L. Lactic acidosis is characterized by lactate levels > 5 mmol/L and serum pH < 7.35.

Pathophysiology

Most cells in the body normally metabolize glucose to form water and carbon dioxide. First, glucose is broken down to pyruvate through glycolysis. Then, mitochondria oxidize the pyruvate into water and carbon dioxide by means of the Krebs cycle and oxidative phosphorylation, which requires oxygen. The net result is ATP, the energy carrier used by the cell to drive useful work such as muscle contraction. When the energy in ATP is utilized during cell work, protons are produced. The mitochondria normally incorporate these protons back into ATP, thus preventing build-up of protons and maintaining neutral pH.

If the oxygen supply is inadequate, the mitochondria are unable to continue ATP synthesis at a rate sufficient to supply the cell with the required ATP. In this situation, glycolysis is increased to provide additional ATP, and the excess pyruvate produced is converted into lactate and released from the cell into the bloodstream, where it accumulates over time. Although increased glycolysis helps compensate for less ATP from oxidative phosphorylation, it cannot bind the protons resulting from ATP hydrolysis. Therefore, the proton concentration rises and causes an acidosis.

Lactate production is initially buffered intracellularly; for example, the lactate-producing enzyme lactate dehydrogenase binds one proton per pyruvate molecule converted. When such buffer systems become saturated, cells will transport lactate into the blood stream. Hypoxia causes both build-up of lactate and acidification, and lactate is therefore a good ‘marker’ of hypoxia, but lactate itself is not the cause of the low pH.

The signs of lactic acidosis are deep and rapid breathing, vomiting, and abdominal pain – symptoms that may easily be mistaken for other problems.

Causes

There are several different causes of lactic acidosis including:

Treatment and prognosis

The generally poor prognosis associated with severe lactic acidosis is determined largely by the severity of the underlying condition. An exception is lactic acidosis secondary to generalized epileptic convulsions; this is self-limiting and requires no specific treatment.

Despite controversy surrounding the theoretical and clinical benefits of alkali therapy, intravenous bicarbonate remains the mainstay of supportive treatment for cases of severe lactic acidosis. In most cases the best advice is to obtain expert help.

Intravenous bicarbonate. The role of bicarbonate is contentious. In one double-blind, placebo-controlled trial of intravenous administration of sodium bicarbonate, no improvement in cardiac haemodynamics occurred, although significant improvement in the arterial partial pressure of carbon dioxide (PaCO2) occurred. Animal models of lactic acidosis have shown that intravenous administration of bicarbonate may increase lactate production (particularly by the splanchnic bed), decrease portal vein flow, lower intracellular pH in muscle and liver, lower arterial pH, and worsen the cardiac output.

Sodium dichloroacetate. Dichloroacetate is the most potent stimulus of pyruvate dehydrogenase, the rate-limiting enzyme for the aerobic oxidation of glucose, pyruvate and lactate. Dichloroacetate may inhibit glycolysis and, thereby, lactate production. Dichloroacetate also exerts a positive inotropic effect that has been attributed to improvement in myocardial glucose use and high-energy phosphate production. Data from animal studies and one placebo-controlled double-blind clinical trial showed that dichloroacetate was superior to placebo in improving the acid–base status of the patients; however, the magnitude of change was small and did not alter haemodynamics or survival.