19 Gastrointestinal, Liver and Nutritional Alterations
After reading this chapter, you should be able to:
• describe the changes in normal gastrointestinal physiology and metabolism associated with critical illness
• integrate theoretical knowledge of the nutritional requirements, assessment of and potential for malnutrition in the critically ill with clinical practice and rationalise selected nutritional support strategies for specific patients
• identify patients at risk for the development of stress ulcers and rationalise therapeutic interventions for their prevention
• discuss the effects of critical illness on hepatic function and evaluate the consequences of liver dysfunction
• describe the treatment of liver failure, including liver support therapies and transplantation
• critically analyse the role of glycaemic control in the context of critical illness
• describe the physiological changes that occur during diabetic ketoacidosis and rationalise assessment and treatment strategies.
Gastrointenstinal Physiology
Digestion and absorption of nutrients such as carbohydrates, amino acids, minerals and water are key functions of the gastrointestinal system. Digestive enzymes are responsible for breaking down food into smaller substances that can be absorbed by the gastrointestinal tract. While some digestion begins in the oral cavity (for example, the breakdown of starch into sugar by salivary amylase), the stomach, pancreas, and small intestine secrete the most enzymes responsible for digestion (Table 19.1). The small bowel plays an important part in the digestion and absorption of these nutrients, where the processes of diffusion, facilitated diffusion, osmosis and active transport are responsible for absorption of 90% of all nutrients.1 The remaining 10% of nutrients are absorbed in the large intestine.
| Location | Enzymes | Target substance |
|---|---|---|
| Oral cavity | Salivary amylase (ptyalin) | Starch and glycogen |
| Bromelain | Protein | |
| Stomach | Pepsin | Proteins |
| Gelatinase | Proteoglycans in meat (gelatine and collagen) | |
| Gastric amylase | Starch | |
| Gastric lipase | Triglyceride | |
| Chymosin | Milk | |
| Pancreas | Trypsin, chymotrypsin, carboxypeptidase, elasatases | Proteins |
| Pancreatic lipase | Triglycerides | |
| Pancreatic amylase | Carbohydrates | |
| Small intestine | Sucrase | Sucrose |
| Lactase | Lactose | |
| Maltase | Maltose (into 2 molecules of glucose) | |
| Isomaltase | Maltose into isomaltose | |
| Intestinal lipase | Fatty acid |
The gastrointestinal tract also plays a role in immunity. It has a variety of mechanisms in place that prevent the movement of substances (other than nutrients, water and electrolytes) into the systemic circulation (see Table 19.2). In the setting of critical illness, where gastrointestinal hypoperfusion may be present, these protective functions may be diminished, so it is essential to understand the alterations in normal gastrointestinal physiology that occur during critical illness.
TABLE 19.2 Protective mechanisms of the gastrointestinal system and impact of critical illness1,3–12
| Mechanism | Action |
|---|---|
| Motility | Propels bacteria through the GI tract. In critical illness, motility may be altered because of enteric nerve impairment and altered smooth muscle function, inflammation (mediated by cytokines and nitric oxide), gut injury, hypoperfusion, medications (opioids, dopamine), electrolyte disturbances, hyperglycaemia, sepsis and increased intracranial pressure.3 |
| Hydrochloric acid secretion | Reduces gastric acidity and destroys bacteria. Parietal cells in the stomach produce hydrochloric acid and keep the intragastric environment relatively acidic (pH approx 4.0). An acidic pH has bactericidal and bacteriostatic properties,4 thus limiting overgrowth in the stomach. |
| Bicarbonate | Bicarbonate ions bind with hydrogen ions to form water and carbon dioxide, preventing the hydrogen ions (acid) from damaging the duodenal wall.5 |
| Bile salts | Bile salts provide protection against bacteria by breaking down the liposaccharide portion of endotoxins,6 thereby detoxifying gram-negative bacteria in the gastrointestinal tract. The deconjugation of bile salts into secondary bile acids inhibits the proliferation of pathogens and may destroy their cell walls.7 |
| Mucin production | Prevents the adhesion of bacteria to the wall of the GI tract. Mucous cells secrete large quantities of very thick, alkaline mucus (approximately 1 mm thick in the stomach). Glycoproteins present in the mucus prevent bacteria from adhering to and colonising the mucosal wall.8 |
| Epithelial cell shedding | Limits bacterial adhesion. The mucosal lining of the entire gastrointestinal tract is composed of epithelial cells that create a physical barrier to bacterial invasion. These cells are replaced approximately every 3–5 days9 limiting bacterial colonisation. |
| Zonea occludulns (tight junctions surrounding each cell in the epithelial sheet) | The junctions between epithelial cells provide a barrier to microorganisms. Intermediate junctions (zonula adherens) function primarily in cell–cell adhesion, while the tight junctions (zonula occludens) limit the movement of bacteria and toxins across the gut wall.10 |
| Gut-associated lymphoid tissue | Protection against bacterial invasion is provided by gut-associated lymphoid tissue,11 capable of cell-mediated and humoral-mediated immune responses.12 |
| Kupffer cells | Kupffer cells in the liver and spleen provide a back-up defence against pathogens that cross the barrier of the gastrointestinal wall and enter the systemic circulation.1 |
Alterations to Normal Gastrointestinal Physiology in Critical Illness
During critical illness, the digestion and absorption of nutrients may be altered. Gastric acid production is commonly thought to increase in critical illness, although evidence suggests that many critically ill patients do not hypersecrete gastric acid13 with increased gastric pH being observed in some critically ill patients, even in the absence of pharmacological inhibition of gastric acid secretion.14,15 The ability of the small intestine to absorb nutrients can be impaired during critical illness,16 although most critically ill patients appear to be able to tolerate enteral nutrition, making the clinical significance of impaired absorption unclear.
Some alterations to normal gastrointestinal physiology in critical illness relate to hypoperfusion and decreased oxygenation in this area and have high metabolic demands. Historically, gastrointestinal dysfunction in critical illness was described in relation to symptoms, such as gastrointestinal bleeding, mechanical obstruction, and pancreatitis17 resulting from ischaemia.18 However, the presence of covert ischaemia has resulted in a heightened interest in the prevention and early detection of gastrointestinal ischaemia in the critically ill, in an attempt to minimise ischaemia-related dysfunction.
Gastrointestinal Mucosal Hypoperfusion
The gastrointestinal system is particularly susceptible to alterations in regional blood flow and oxygen delivery because it has a higher critical oxygen delivery (DO2) than the rest of the body. Splanchnic vasoconstriction is also proportionally greater than other vascular beds and the countercurrent O2 exchange between vessels within the villi further contribute to decreased regional oxygen delivery.5
During shock states, decreased blood flow from vasoconstriction occurs in this region first. It is the last place to be restored following successful resuscitation.19 In shock states, the gastrointestinal system attempts to maintain adequate cellular oxygenation by increasing the amount of oxygen extracted from the blood. This increase in oxygen extraction may prevent serious compromise of tissue oxygenation even in the presence of reduced oxygen delivery.20
During periods of ischaemia and hypoxia, oxygen free-radicals are generated as byproducts of anaerobic metabolism. With successful resuscitation of the gastrointestinal tract, blood flow and oxygen delivery are restored but the oxygen free-radicals are liberated, contributing to the microvascular and mucosal changes characteristic of ischaemia and reperfusion of the gut mucosa.21
Consequences of Gastrointestinal Hypoperfusion
The consequences of gastrointestinal hypoperfusion are significant, and include disruption of the physical barrier to pathogens; disruption of chemical control of bacterial overgrowth; decreased peristalsis; and reduced immunological activities of gastrointestinal-associated lymphoid tissue. In health, all of these mechanisms work efficiently to contain bacteria within the gastrointestinal tract. During critical illness, however, reduced oxygenation contributes to decreased cellular function and failure of the protective mechanisms described in Table 19.2. Consequently, bacterial proliferation and translocation from the gastrointestinal tract to the systemic circulation may occur.22
Changes in gastrointestinal perfusion also has the capacity to affect hepatic perfusion, oxygenation and function. In approximately 50% of critically ill patients, ischaemic hepatitis or ‘shock liver’ occurs, which is evidenced by jaundice, elevation of liver function tests or overt hepatic dysfunction.23 Ischaemic hepatitis can vary from a mild elevation of serum aminotransferase and bilirubin levels in septic patients, to an acute elevation following haemodynamic shock. Ischaemic hepatic injury influences morbidity and mortality but remains underdiagnosed, probably because the clinical signs become apparent long after hypoperfusion has occurred. Physiological changes contributing to ischaemic hepatitis include changes to the portal and arterial blood supply as well as hepatic microcirculation. The degree to which the liver is damaged is directly related to the severity and duration of hypoperfusion, and both anoxic and reperfusion injury can damage hepatocytes and the vascular endothelium.23
Alterations to Normal Metabolism in Critical Illness
There is little information describing the changes to the exocrine function in the gastrointestinal system during critical illness, and it is uncertain how critical illness influences the metabolism of nutrients. While there is data to demonstrate that the secretion of hydrochloric acid by the parietal cells in the stomach is decreased, it is not certain whether the exocrine failure also extends to a decreased pepsin secretion. It is also possible that secretion of digestive enzymes might also be influenced by critical illness-induced pancreatitis, although clear data demonstrating this level of dysfunction are unavailable.16
Nutrition
Optimal nutritional support in the critically ill aims to prevent, detect and correct malnutrition, optimise the patient’s metabolic state, reduce morbidity and improve recovery.24 The metabolic response of stress or injury is hypermetabolism. There is an increased release of cytokines (e.g. interleukin-1, interleukin-6, tumor necrosis factor-α) and production of counter-regulatory hormones (e.g. catecholamines, cortisol, glucagon and growth hormone) that induce catabolism and oppose the anabolic effects of insulin.25 Hypercatabolism occurs with the imbalance between anabolism (i.e. the chemical process by which complex molecules, such as peptides, proteins, polysaccharides, lipids and nucleic acids, are synthesised from simpler molecules) and catabolism (i.e. the convergent process, in which many different types of molecules are broken down into relatively few types of end products). To compensate for the altered metabolic regulation, neuroendocrine stimulation increases the mobilisation and consumption of nutrients, such as glycogen and protein, from existing body stores. As the metabolic rate rises, nutritional requirements in critical illness are increased, characterised by a rise in resting energy expenditure and oxygen consumption, which in some critically ill patients can be increased by over 50%.26 Depletion of body energy stores result from alterations in protein, carbohydrate and fat metabolism.27 In addition to the rise in metabolic demands, patients who are critically ill often experience a concomitant fall in nutritional intake. The metabolic and nutrition alterations vary with the stress level, severity of illness, type of injury, organ dysfunction and nutrition status.25
To maintain normal cellular function, body cells require adequate amounts of the six basic nutrients: carbohydrates, fats and proteins to provide energy, vitamins, minerals and water to catalyse metabolic processes. Unlike normal metabolism, which preferentially uses carbohydrates and fats for energy, the hypermetabolic state associated with critical illness consumes proportionally more fats and proteins than carbohydrates to generate energy.28 As a consequence of the gluconeogenesis and the synthesis of acute-phase proteins, there is a decrease in lean body mass and negative nitrogen balance.
Consequences of Malnutrition
When adequate and timely nutrition support is not provided, body energy and protein depletion can occur with negative consequences on patient outcome.29 Critically ill patients require adequate nutrition to limit muscle wasting, respiratory and gastrointestinal dysfunction and alterations in immunity, all of which are associated with malnutrition.30 Respiratory support is often necessary during critical illness, and changes in respiratory muscle function and ventilatory drive may contribute to an increase in the number of ventilator days. Furthermore, infection rates may be increased in malnourished critically ill patients. The decrease in lean body mass and negative nitrogen balance is associated with delayed wound healing and a higher risk of infection.28
These complications contribute to increased length of stay, cost, morbidity and mortality.31 The degree of critical illness and hypercatabolism varies between patients and is often difficult to accurately determine. For this reason it is necessary to assess, as accurately as possible, the nutritional requirements of each individual patient.
Nutritional Assessment
The majority of studies report cumulative energy deficit or caloric debt is associated with worse clinical outcomes.32–35 Krishnan and colleagues,36 however, describe better clinical outcomes for patients fed fewer than the target nutrition goals when compared to those who received near target goals. Nutritional assessment includes patient history, physical examination and assessment of nutritional indices (see Table 19.3), but is often unreliable in the critically ill patient.37,38 Clinical judgement remains the most common way of assessing a patient’s nutritional status, and is shown to be as reliable as biochemical tests.39–41 Clinical judgement takes into consideration recent weight loss, reduced dietary intake, anorexia, vomiting, diarrhoea, muscle wasting and signs of nutritional deficiency.42 Appreciation of the importance of nutritional assessment and the impact of malnutrition in the critically ill informs management and is likely to improve outcomes.30
| Assessment | Limitations in critical illness |
|---|---|
| Subjective global assessment | Not validated in the critically ill40,43 |
| Biochemical markers: | |
Decreased sensitivity because of 20-day half-life; influenced by fluid balance/shifts42
Half-life of 8 days but lacks the sensitivity and specificity for determining nitrogen balance;43 influenced by fluid balance/shifts
Most sensitive with a half-life of 2 days,44 but changes may result from the metabolic response to illness rather than change in nutritional status; influenced by fluid balance/shifts
Determining Nutritional Requirements
Determining caloric requirements is largely dependent on energy expenditure, influenced by patient activity, stage of illness, type of injury and previous nutritional status.42 Indirect calorimetry is the ‘gold standard’ and most precise way of determining the nutritional requirements in critical illness.45 Energy expenditure is measured using the oxygen consumption obtained from carbon dioxide levels (PaCO2), or using a metabolic monitor. It is infrequently used in critical care settings, possibly because of the high equipment costs and unreliability in the critically ill.46
Calculating basal energy expenditure using the Harris-Benedict equation is a common, but less precise, method of determining nutritional requirements.42,47,48 The Harris-Benedict equation, and others, takes into account the age, height, weight and gender of the patient, with adjustments made for treatment, disease process and metabolic state. Importantly, these equations fail to find any significant benefits in outcomes, most likely because they do not measure energy requirement.49
The Prognostic Inflammatory Nutrition Index (PINI) uses the elevations in acute phase proteins (alpha-1-acid glycoprotein and C-reactive protein [CRP]) that occur with simultaneous reductions in transport proteins (albumin and pre-albumin) in a simple formula to stratify critically ill patients by risk of complications or death.50
Nutrition Support
For patients in ICU who are unable to take oral nutrition, enteral nutrition (EN), parenteral nutrition (PN) or combined EN and PN is available. The best method of providing nutrition to the critically ill who cannot have oral feeding is controversial. Infectious complications have been associated with PN when used alone,51 but no differences in infectious complications were seen with concurrent use of EN and PN.52 In a meta-analysis, PN was associated with reduced mortality when comparing PN with delayed EN despite the increased risk for infectious complications associated with PN.51 Meta-analyses are limited by the quality of the studies included in the analyses.53,54 Recent guidelines advocate early enteral nutrition53,55–59 but better evidence is needed.60
Enteral Nutrition
EN has benefits beyond the supply of nutrients to the body,61 including:
• gut-derived mucosal immunity62 and decrease in septic complications63–65
• preservation of gastrointestinal mucosal integrity66,67
• improved gastrointestinal mucosal cell growth and replacement68
Absence of enteral nutrients (despite the provision of PN) has been linked to atrophy of the intestinal villi, a reduction in the number of epithelial cells produced, reduced gastrointestinal mucosal thickness, and ineffective functioning of the intestinal brush border enzymes of the gastrointestinal mucosa.59,70–73 Stimulating and improving gastrointestinal immune function is an important goal of early EN.59 Early enteral feeding (within 48 hours) is recommended.55,56
Hypocaloric Intake in the Critically Ill
A significant number of hospitalised patients receiving EN do not have their nutritional needs met.70,71 Hypocaloric feeding in the first few days of critical illness may be beneficial,36,74–77 but results are conflicting.34,78–80 The belief that early enteral feeding prevents gut dysfunction independently of calorie intake81 perpetuates the acceptance of administration of EN below the nutrition target.33,70,82,83 In most cases, hypocaloric feeding is unnecessary and avoidable.84,85 Severe underfeeding over a short time particularly during the initial week of ICU stay is associated with the formation of an energy debt that leads to increased infections, complications and longer ICU stays.34 Factors that contribute to unintentional hypocaloric feeding include staffing shortages, unavailability of feeds/equipment, low priorities for feeding, fasting for clinical investigations, blockages in feeding tubes and variations in feed prescriptions.86 Delivery issues, such as elective interruption for investigative procedures or operations, contributed to hypocaloric feeding with only 76% of prescribed feeds delivered to critically ill patients.87 Similar results were observed in mechanically-ventilated patients,88 where more than 36% of patients received less than 90% of their caloric requirements.
Enteral Feeding Protocols
Enteral feeding protocols improve the delivery of enteral feeds87,89,90 and have been shown to improve clinical outcomes.83,91,92 But protocols vary widely between units and institutions,24,58,93–95 primarily as a consequence of the shortage of reliable and valid research into the effective delivery of enteral nutrition. In the absence of strong research evidence, rituals are embraced and rarely challenged.86 Furthermore, the implementation and sustainability of guidelines is influenced by multiple factors, e.g. clinicians, patients, context and contents of guidelines.96
Management of Enteral Feeding
Routes of enteral feeding
The insertion of enteral feeding tubes into the correct place in the critically ill can be difficult because of reduced cough reflex, altered sensorium and use of sedative and narcotic medications.97 Wide-bore nasogastric tubes (sump tubes) are most commonly used in the critically ill in the early stages of enteral feeding. Because long-term use of wide-bore tubes can contribute to sinusitis, a fine-bore feeding tube is often introduced if enteral feeding is expected to continue beyond a few days. Should prolonged enteral feeding be anticipated (longer than 1 month), gastrostomy, duodenostomy or jejunostomy tubes may also be used.98 Postpyloric feeding has not been shown to be beneficial over gastric feeding,99,100 but is useful for later enteral feeding in patients if gastric atony is present and the patient has persistent high gastric residual volumes.101
For some critically ill patients, gastric secretions may increase when small bowel feeding is initiated.102 A double-lumen tube is available, one lumen for gastric aspiration and decompression and the second for simultaneous jejunal feeding, but these tubes are not widely used in the clinical setting.103
Assessment of enteral feeding tube placement
Correct placement of enteral feeding tubes in the critically ill can be difficult.104,105 Misplacement of the feeding tube into the tracheobronchial tree are important complications of tube insertion.106 Additional complications such as infusion of tube feedings, pneumothorax, pneumonitis, hydropneumothorax, bronchopleural fistula, empyema and pulmonary hemorrhage have been reported.107–112 While confirmation of tube placement is routinely done with radiography, this approach does not prevent incorrect placement occurring during insertion; less reliable methods of confirming tube placement include the use of auscultation and aspiration, and other novel methods such as capnography.105,113
• Aspirate from critically ill patients who receive continuous feedings may have the appearance of unchanged formula, regardless of the site of the feeding tube, therefore this method should not be used.114
• Analysis of the pH of gastric secretions is not reliable. A pH of 0–5 may be used to indicate gastric placement of enteral feeding tubes, although this technique may be problematic for patients receiving histamine-2-receptor antagonists or proton pump inhibitors. If the aspirated fluid has a low pH, it may be assumed that the fluid originated in the stomach but the pH of fluid from an infected pleural space can also be acidic,115 therefore pH testing as a sole method to determine tube placement is not recommended.116
• End-tidal CO2 (ETCO2) detectors: capnometry and capnography. Capnometry and capnography use ETCO2 CO2 detectors to evaluate enteral tube placement but they are not used in routine clinical practice.107,117–122 Differentiating between oesophageal, stomach or intestinal placement is not possible.116
• Pepsin/trypsin. Measuring the concentrations of pepsin and trypsin in feeding tube aspirates can be used as a method of predicting tube placement however methods to measure pepsin and trypsin at the bedside are currently unavailable.123
In the absence of X-ray, several approaches should be used in combination to verify tube position. Metheny and colleagues114 found measuring: (a) length of tubing extending from the insertion site, (b) volume of aspirate from the feeding tube, (c) appearance of the aspirate, and (d) pH of the aspirate were able to correctly differentiate between gastric and bowel tube placement during continuous feedings in 81% of the predictions. Ongoing assessment of feeding tube placement is also essential, as feeding tubes may migrate after initial placement.
Feeding regimens
Once the enteral feeding tube is successfully placed, administration of the feeding solution can begin using a variety of methods, including bolus, intermittent and continuous enteral feeding (see Table 19.4). Bolus enteral feeding is rarely used in the critically ill, but it is less clear whether intermittent or continuous feeding is more beneficial.124–126 Because of inconclusive evidence regarding feeding regimens, decisions are best based on individual patient assessment and the clinician’s clinical judgement.
Commencing enteral feeding
The starting rate for enteral feeding is controversial, with suggestions in the range of 10–100 mL/h,86 and the commonest starting rate being 30 mL/h, despite there being no empirical data on which to base this recommendation. Increasing the rate of enteral feeding is equally variable, but strategies to progress patients towards meeting their daily caloric requirements should be employed. When a patient has experienced a prolonged period of starvation or total parenteral nutrition, the approach to enteral feeding is somewhat more reserved, as the risk of refeeding syndrome is increased.130–132 Although not common, this syndrome is associated with severe derangement in fluid and electrolyte levels (particularly hypophosphataemia, hypomagnesaemia and hypokalaemia), and may result in significant morbidity and mortality.
Managing complications of enteral feeding
Once enteral feeding is established, it is important to assess for such complications as:
This intolerance to enteral feeding can result in gastric distension, diarrhoea and increased GRV.87,133,134
Critically ill patients exhibit elevated gastric residual volume for a variety of reasons including feeding intolerance135–139 and reduced gastric motility.135,136,140 Monitoring tolerance to enteral feeding through the measurement of gastric residual volume has always been viewed as an important aspect of nursing management, although consensus on what constitutes a high gastric residual and any recommendations for interventions remain controversial. Ceasing feeds in response to gastric residual volume is questionable,141 particularly as a balanced enteral diet in itself has a prokinetic effect.142
Practice tip
In determining feeding intolerance, a single high gastric residual volume in the absence of physical examination or radiographic findings should not result in the cessation of enteral feeding. Persisting with enteral feeding has demonstrated benefits. It is thought that a balanced enteral diet, in itself, has a prokinetic effect.143
Development of diarrhoea is another complication for enterally fed patients, and is a common reason why enteral feeding is often reduced or ceased. Diarrhoea may contribute to fluid and electrolyte disorders, patient (and nursing) distress, and a higher cost of patient care.144 Unfortunately, defining diarrhoea is problematic, as it is a subjective assessment that relies on nursing interpretation rather than on quantifiable assessment of stool weight.145 There are various aetiologies for diarrhoea in the enterally fed, critically ill patient, including:
Probiotic administration may limit the development of diarrhoea,149 although its efficacy is yet to be established.150,151
Enteral feeding solutions present an excellent medium for the growth of microorganisms,152 and bacterial contamination of enteral feeds is common.153–155 Strategies to limit bacterial contamination of enteral feeding solutions include:
• meticulous preparation of feeding solutions and equipment156
• commercially prepared formula used in preference to decanted feeds157–159
• use of closed feeding systems93,160,161
• limiting the time feeding solution is kept at room temperature once opened and hang times93,148,157,162–168
• meticulous attention to hand washing and limiting manipulation of the enteral nutrition bags and delivery system at the bedside153,155
Prevention of pulmonary aspiration
An important complication of enteral feeding is the development of pulmonary aspiration and nosocomial pneumonia. Determining whether aspiration has occurred is difficult, even for experienced clinicians. High gastric residual volumes have been linked to the potential for pulmonary aspiration, although this has not been shown in research.141 Oropharyngeal secretions can contribute to nosocomial pneumonia and subglottic aspiration has improved outcomes.169 Nursing strategies to improve gastric emptying includes elevation of the head of the bed 30–45 degrees (unless otherwise contraindicated),170 reducing the likelihood of gastro-oesophageal reflux, which is present in up to 30% of patients in the supine position.
Prokinetic agents can improve gastric emptying and feeding tolerance, and avoid gastro-oesophageal reflux and pulmonary aspiration. Cisapride, erythromycin and metoclopramide have all been used clinically to improve gastrointestinal motility. A systematic review noted that, as a class of drugs, promotility agents have a beneficial effect on gastrointestinal motility in the critically ill patient.171 These prokinetic agents do, however, have undesirable effects. Use of erythromycin is associated with the development of bacterial resistance, and metoclopramide is associated with numerous systemic side effects. Erythromycin is more effective than metoclopramide in treating gastric intolerance among patients receiving enteral nutrition.172 However, combination therapy with erythromycin and metoclopramide is more effective than erythromycin alone in improving the delivery of enteral nutrition.173
Assessment of pulmonary aspiration
• The dye method involves the addition of blue food colouring to the enteral feeding formula, theoretically making it possible to visualise gastric contents if they have been inhaled into the tracheobronchial tree. However, the use of blue dye is poorly standardised and has a low sensitivity in detecting microaspiration.174 The use of methylene blue is not recommended because of associated side effects and high costs.175 There have been case reports of blue dye absorption describing discolouration of the skin, urine, serum and organs,176 and refractory hypotension and severe acidosis, suggesting poisoning by a mitochondrial toxin.177,178 These safety concerns, coupled with minimal benefits, have resulted in the recommendation that the practice of using blue food colouring in enteral feeding solutions be abandoned.179
• Measurement of glucose in tracheobronchial secretions is another method to detect pulmonary aspiration.180 As these secretions normally contain <5 mg/dL glucose, higher amounts of glucose may indicate the aspiration of glucose-rich enteral feeding formula.68However, differences in enteral feeding solutions affect the sensitivity of this method, with low glucose solutions being more difficult to detect. Also, patients not receiving enteral feeding can have detectable glucose in aspirates.181 This is further confounded by the presence of blood, which is closely associated with glucose values >20 mg/dL; consequently, any blood in the respiratory tract could contribute to a false-positive result.181 These findings led to the consensus that glucose monitoring in respiratory secretions should also be abandoned.179
• Measurement of pepsin in tracheobronchial secretions has been used in an animal study suggested that the detection of pepsin, a component of gastric secretions, may be useful in determining pulmonary aspiration.182 however, further investigation in acutely ill patients receiving enteral feeding is necessary.
Parenteral Nutrition
The appropriate use of PN in the context of critical illness continues to be debated.183–185 EN is the preferred method of nutritional support because it is less expensive and is associated with fewer infectious and metabolic complications than PN. However, it is not uncommon for critically ill patients to have difficulty in meeting daily caloric intake34,71 and this may necessitate supplementation of enteral nutrition with PN or the sole provision of nutritional support through parenteral means (as TPN). For patients who are unable to be fed by the enteral route and who were healthy prior to ICU admission, with no evidence of protein-calorie malnutrition, then it is recommended that PN be initiated after 3–7 days186 of hospitalisation.187 The lack of agreement on the efficacy of PN means that the use of this therapy varies both within and between countries.58,186,187
PN solutions contain carbohydrates, lipids, proteins, electrolytes, vitamins and trace elements. PN, whether supplementary or complete, provides daily allowances of nutrients and minerals. The components of PN are listed in Table 19.5. The addition of vitamins and trace elements to PN solutions is necessary, particularly as water-soluble vitamins and trace elements are rapidly depleted (see Table 19.6). Glucose is the primary energy source in PN solutions. Concentrations of 10–70% glucose may be used in PN solutions although the final concentration of the solution should be no more than 35%. The high concentration of PN solutions can cause thrombosis so PN is normally infused via a central venous catheter (CVC). Peripheral administration can be considered when the final solution concentration is 10–12%,188 but is not usually used in the context of critical illness because high volumes of PN would be required to meet caloric requirements.189
• Content varies in amino acid solutions, so content in relation to patient requirements needs to be considered.
• Monitoring of electrolyte status is essential, particularly serum phosphate levels if the amino acid solution used is phosphate-free.
• The balance between chloride and acetate is monitored, as administration of additional sodium or potassium may result in acid–base imbalances.
TABLE 19.6 Trace elements in TPN192
| Trace element | Action |
|---|---|
| Zinc | Wound healing |
| Iron | Haemoglobin synthesis |
| Copper | Erythrocyte maturation and lipid metabolism |
| Manganese | Calcium and phosphorus metabolism |
| Cobalt | Essential constituent of vitamin B12 |
| Iodine | Thyroxine synthesis |
| Chromium | Glucose utilisation |
Catheter insertion, ongoing care and replacement are similar to that with any other CVC. A dedicated CVC, or lumen of a multilumen CVC, should be used for PN.191,193 Manipulation of the CVC and tubing should be avoided to minimise infection of the catheter.
Routine monitoring of the patient’s fluid balance, glucose, biochemical profile, full blood count, triglycerides, trace elements and vitamins is necessary. The patient is also assessed for signs of complications associated with the administration of PN (see Table 19.7).
TABLE 19.7 Short-term metabolic complications associated with total parenteral nutrition
| Complication | Cause | Detection and treatment |
|---|---|---|
| Hyperosmolar coma | Occurs acutely if a rapid infusion of hypertonic fluid is administered. Infusion can cause severe osmotic diuresis, resulting in electrolyte abnormalities, dehydration and malfunction of the central nervous system. | Daily blood samples, accurate measurements of fluid balance, routine blood samples. Reduce infusion rate, correct electrolyte imbalances. |
| Electrolyte imbalance | Disturbances in serum electrolytes, particularly sodium potassium, urea and creatinine, may occur early in the treatment of TPN. Electrolyte imbalances can be caused by the patient’s underlying medical condition; requirements vary with individual patients’ needs. Can be caused by inadequate or excessive administration of intravenous fluids. | Daily blood samples taken early in treatment to detect abnormalities. Replacement fluid as required, extra intravenous fluids may be required during the stabilisation period. |
| Hyperglycaemia | Critically ill patients may be resistant to insulin because of the secretion of ACTH and adrenaline. This promotes the secretion of glycogen, which inhibits the insulin response to hyperglycaemia. | Monitor the patient’s blood sugar 4-hourly after commencement of treatment or as required. Monitor daily urinalysis for glucose and ketones. An insulin infusion may be required to keep blood sugar levels within prescribed limits. |
| Rebound hypoglycaemia | May occur on discontinuation of TPN because hyperinsulinism may occur after prolonged intravenous nutrition. A rise in serum insulin occurs with infusion, and thus sudden cessation of infusion can result in hypoglycaemia. | Glucose infusion rate should be gradually reduced over the final hour of infusion before discontinuing. Some patients may receive a 10% glucose solution after cessation of TPN. |
| Hypophosphataemia | Glucose infusion results in the continuous release of insulin, stimulating anabolism and resulting in rapid influx of phosphorus into muscle cells. The greatest risk is to malnourished patients with overzealous administration of feeding. Patients who are hyperglycaemic, who require insulin therapy during TPN or who have a history of alcoholism or chronic weight loss may require extra phosphate in the early stages of treatment. | Monitor phosphate levels daily. Hypophosphataemia will usually appear after 24–48 hours of feeding. Reduce the carbohydrate load and give phosphate supplementation. |
| Lipid clearance | Lipids are broken down in the bloodstream with the aid of lipoprotein lipase found in the epithelium of capillaries in many tissues. A syndrome known as fat overload syndrome can occur when infusion of lipid is administered that is beyond the patient’s clearing capacity, resulting in lipid deposits in the capillaries. | Blood samples should be taken after the first infusion commences (within 6 hours) to observe for lipid in the blood. |
| Side effects of lipid infusion | Some patients suffer symptoms either during or after an infusion of lipid mix parenteral nutrition. The exact cause is unknown. The patient may complain of headache, nausea or vomiting, and generally feels unwell. | Treat mild symptoms. If tolerated, the TPN solution of non-protein calories can be given in the form of glucose. However, it is essential that the regimen includes some fat to prevent the development of fatty acid deficiency. |
| Anaphylactic shock | This is a rare complication but may occur as a reaction to the administration of a lipid. | It may be necessary to administer adrenaline and/or steroids, and to provide supportive therapy as required. |
| Glucose intolerance | TPN using glucose as the main source of calories is associated with a rise in oxygen consumption and CO2 production. The workload imposed by the high CO2 production may precipitate respiratory distress in susceptible patients, particularly those requiring mechanical ventilation. | Observe patients for signs of respiratory distress. Provide non-protein calories in the form of glucose lipid mix. Slow initial rate of infusion. |
| Liver function | Abnormalities with liver function can be associated with TPN. May be attributable to hepatic stenosis with moderate hepatomegaly; patient may also develop jaundice. Liver function tests often return to normal after cessation of therapy; however, TPN can lead to severe hepatic dysfunction in neonates. | Monitor liver function tests twice weekly. There are several factors that may contribute to development of abnormal liver function tests. These most often occur after a period of time and appear to be more of a problem when there is an excess calorie intake or in glucose-based regimens. |
ACTH = adrenocorticotrophic hormone.
Stress-Related Mucosal Disease
The reported incidence of stress-related mucosal damage is variable194 and complicated by definitions of end points, difficulty in measuring the end points, and the heterogeneity of the patient populations.195 With occult bleeding (drop in haemoglobin level or positive stool occult blood test) as an endpoint, it is estimated that 15–50% of critically ill patients would be reported to have stress-related mucosal damage.196,197 Reported incidence is reduced to 25% or less when haematemesis or nasogastric lavage positive for bright red blood is used as an endpoint to describe clinically overt bleeding.198,199 The incidence of clinically significant bleeding, that is bleeding associated with hypotension, tachycardia, and a drop in haemoglobin level necessitating transfusion, is estimated to be 3–4%.194
Factors influencing the development of stress-related mucosal disease include splanchnic hypoperfusion200 which may influence mucosal ischaemia and reperfusion injury,201 maintenance of the gastric mucosa by sufficient microcirculation and the mucus-bicarbonate gel layer,202 decreased prostaglandin levels which impairs mucus replenishment and increased nitric oxide synthase which contributes to reperfusion injury and cell death.203 The protective mechanisms and factors which promote injury are detailed in Table 19.8.
TABLE 19.8 Factors contributing to stress-related mucosal disease204
| Factors | Mechanism | Action |
|---|---|---|
| Protective mechanisms | Mucosal prostaglandins | Protect the mucosa by stimulating blood flow, mucus and bicarbonate production205 Stimulate epithelial cell growth and repair |
| Mucosal bicarbonate barrier | Forms a physical barrier to acid and pepsin, preventing injury to the epithelium206 | |
| Epithelial restitution and regeneration | Epithelial cells rapidly regenerate but the process is highly metabolic and may be impaired by physiological stress206 | |
| Mucosal blood flow | Mucosal blood flow helps remove acid from the mucosa, supplies bicarbonate and oxygen to the mucosal epithelial cells207 | |
| Cell membrane and tight junctions | Tight junctions between mucosal epithelial cells prevents the back diffusion of hydrogen ions208 | |
| Factors promoting injury | Acid | Acid is a key issue in the pathogenesis of stress-related mucosal injury however not all critically ill patients hypersecrete acid.14,208 However small amounts of acid may still cause injury and the prevention of acid secretion has led to a reduction in injury209 |
| Pepsin | May cause direct injury to the mucosa210 Facilitates the lysis of clots211 |
|
| Mucosal hypoperfusion | Reduced mucosal blood flow results in reduced oxygen and nutrient delivery, making epithelial cells susceptible to injury.208 Contributes to mucosal acid-base imbalances Results in the formation of free radicals |
|
| Reperfusion injury | Nitric oxide, which causes vasodilation and hyperaemia, is released during hypoperfusion and results in an increase in cell-damaging cytokines | |
| Intramucosal acid–base balance | The mucus layer protects the epithelium and traps bicarbonate ions that neutralise acid thus a decrease in bicarbonate secretion results in intramucosal acidosis and local injury206 | |
| Systemic acidosis | Results in increased intramucosal acidity207 | |
| Free oxygen radicals | Generated as a result of tissue hypoxia, free oxygen radicals cause oxidative injury to the mucosa212 | |
| Bile salts | Bile salts reflux from the duodenum into the stomach and may have a role in stress-related damage although the exact mechanism is uncertain213 | |
| Heliobacter pylori | Conflicting results about the role of H. pylori as a cause of stress-induced mucosal disease in the critically ill214 |
Risk Factors for Stress-Related Mucosal Disease
A number of risk factors are associated with the development of stress-related mucosal disease, including respiratory failure requiring at least 48 hours of mechanical ventilation and coagulopathy,215 acute hepatic failure, hypotension, chronic renal failure, prolonged nasogastric tube placement, alcohol abuse, sepsis and an increased serum concentration of anti-Helicobacter pylori (H. pylori) immunoglobulin A.216
Mortality rates for critically ill patients who develop stress-related mucosal disease approximate 50–77% and higher than for those who do not develop this complication.200 Consequently, there is a strong imperative to implement stress-ulcer prophylaxis, particularly in those patients who are considered at risk.
Preventing Stress-Related Mucosal Disease
Prophylaxis for stress-related mucosal disease is often part of the care of the critically ill although evidence demonstrating an added benefit when this therapy is applied to those patients who are not identified as at risk for developing stress-related mucosal disease, is limited.201 Nevertheless, it is common for the majority of critically ill patients to receive some form of stress-ulcer prophylaxis during their episode of critical illness. There are a variety of pharmacological strategies that can be used to prevent stress ulcers from developing. These include antacids, sucralfate, histamine-2-receptor antagonists and proton pump inhibitors (PPIs).201
Antacids
Antacids directly neutralise gastric acid and have been shown to be effective in reducing significant stress-related bleeding.217 One of the disadvantages of this therapy is the time-intensive nature of administering antacids every 1–2 hours. Furthermore, antacids can contribute to further complications (e.g. aluminium toxicity, hypophosphataemia, diarrhoea or hypermagnesaemia). These factors have led to their infrequent use within the critical care setting.200,218,219
Histamine-2-Receptor Antagonists
Histamine-2-receptor antagonists (H2RAs) are commonly used in the critically ill to inhibit the production of gastric acid, which is achieved by the drug binding to the histamine-2 receptor on the basement membrane of the parietal cell.196 However, gastric acid secretion may also occur through stimulation of the acetylcholine or gastrin receptors present in parietal cells;220 therefore complete blocking of gastric acid production does not occur when H2RAs are used. A further limitation of H2RA is the development of tolerance that may occur within 72 hours of administration.221 Nevertheless, this pharmacological strategy to prevent stress-related mucosal disease remains commonplace in critical care.222
The decrease in gastric acidity as a result of H2RA use may be beneficial from the perspective of preventing stress-related mucosal disease, but changes in gastric pH could lead to bacterial overgrowth in the stomach, microaspiration, and consequently an increase in the incidence of nosocomial pneumonia,223 although there is some research that does not support this notion.209
Proton Pump Inhibitors
Proton pump inhibitors (PPIs) have a greater ability to maintain an increased intragastric pH than H2RAs.224 These drugs work by irreversibly binding to the proton pump, effectively blocking all three receptors responsible for gastric acid secretion by the parietal cell.196,201 PPIs are also able to limit vagally-mediated gastric acid secretion.200
Clinical evaluation of the efficacy of PPIs is somewhat limited; few studies have specifically studied the prophylactic use of PPIs for stress-related mucosal diseases12,225–227 and many are limited by small sample sizes. Although PPIs are similar to H2RA in the ability to raise the gastric pH above 4, a level considered adequate to prevent stress ulceration, PPIs are more likely to maintain the pH at greater than 6, which may be necessary to maintain clotting in those patients at risk of rebleeding from peptic ulcer.201
PPIs that may be administered intravenously include omeprazole, esomeprazole and pantoprazole. Omeprazole has the highest potential for drug interation and interferes with the metabolism of some drugs commonly used in intensive care, including cyclosporine, diazepam, phenytoin and warfarin.203 Pantoprazole has the lowest potential for drug interactions.200
Sucralfate
Sucralfate provides protection against stress-related mucosal disease through a number of mechanisms. Sucralfate provides a protective barrier on the surface gastric epithelium, stimulates mucus and bicarbonate secretion, stimulates epithelial renewal, improves mucosal blood flow and enhances prostaglandin release.196 Given orally or via a nasogastric tube, sucralfate is well tolerated but appears to be less effective than H2RAs in decreasing clinically significant bleeding.228 Earlier reports comparing sucralfate with ranitidine showed a decrease in the development of pneumonia in those patients receiving sucralfate; however, these findings were not supported in a subsequent Level I randomised controlled trial.228
Enteral Nutrition
It is thought that the presence of enteral feeding solution results in an increase in intragastric pH, thereby minimising acid injury. Several studies have demonstrated a lower incidence of stress-related bleeding in mechanically-ventilated229 and burn patients,230 while others were unable to show a significant effect on increasing gastric pH.231 A lack of well-designed prospective studies examining the role of enteral nutrition in stress-ulcer prophylaxsis prevents the use of this therapy as a sole therapeutic agent for this purpose.201
Liver Dysfunction
Related Anatomy and Physiology
The liver is the largest internal organ, weighing approximately 1200–1600 g in the adult. It receives approximately 25% of total cardiac output through a dual vascular supply consisting of the hepatic artery and portal vein.232 About 75% of the hepatic blood flow arises from the portal vein with the remaining 25% from the hepatic artery. Anatomically, the liver consists of 4 lobes: the major left and right lobes, and the minor caudate and quadrate lobes. The right lobe is considerably larger than the left lobe. Functionally, the liver is divided into eight segments each with their own inflow and outflow blood supply and biliary drainage. Hepatic lobules, or liver acini, are small units consisting of a single or double layer of hepatocytes arranged in plates interspersed with capillaries (sinusoids) that receive inflowing blood from the portal vein and hepatic artery. To safeguard the body from the entrance of toxins absorbed from the intestines, the sinusoids are lined by macrophages known as Kupffer cells. The hepatic vein then drains effluent blood from the liver into the general circulation.1
The arrangement of the circulation to the liver with its rich vascular architecture enables it to perform the vital functions of carbohydrate, fat and protein metabolism; production of bile to aid in digestion; the production, conjugation and elimination of bilirubin; immunological and inflammatory responses; glycogen storage; and detoxification of toxins and drugs.1
Mechanisms of Liver Cell Injury
Liver cell injury and death can occur either as a direct result of injury to the cell, resulting in cell necrosis, or as a result of ‘cellular stress’ and the triggering of apoptotic pathways, leading to ‘programmed cell death’.233 Major factors for the triggering of the apoptotic pathway are hypoxia with resulting ischaemia and reperfusion; reactive oxygen metabolites resulting from alcohol or drug ingestion; accumulation of bile acids resulting from cholestasis; and inflammatory cytokines such as tumour necrosis factor alpha (TNF-α).233 The apoptotic pathway results in the deconstruction of the cellular structure from the inside out, while necrosis results in cell rupture and release of cellular contents. Although these processes may overlap, it is thought that the apoptotic pathway is a way of preventing the inflammatory response that is triggered with cell necrosis. The activation of the inflammatory response results in secondary liver cell injury and contributes to the multiple organ dysfunction seen in liver failure.233,234
The degree and time course of liver cell damage from viral hepatitis depends on the immune response. Immune recognition and destruction of infected cells may result in either clearance of the virus or ongoing inflammation, cell death and fibrosis if the virus is not cleared. This process may progress over 20–40 years to cirrhosis and hepatocellular carcinoma.235 Chronic excessive alcohol intake may also result in a slower chronic course of liver injury that eventually results in cirrhosis, liver failure or hepatocellular carcinoma.236
Liver cells may also be injured by the toxic effects of drugs or their metabolites, as in paracetamol overdose, or by drugs at therapeutic doses (e.g. non-steroidal anti-inflammatory drugs, phenytoin, antimalarial agents). Other poisoning from the ingestion of mushrooms (e.g. Amanita phalloides), and from recreational drug use (e.g. ecstasy and amphetamines), may result in liver cell death and liver failure.237 Diseases of the biliary system such as primary biliary cirrhosis and primary sclerosing cholangitis also result in liver dysfunction and failure.238
The liver has a remarkable regenerative capacity. After injury and necrosis, liver cells rapidly regenerate around areas of surviving cells to restore the lost tissue whilst maintaining homeostasis during hepatic regeneration.234,239,240 However, with chronic injury, fibrosis or scarring occurs, resulting in the loss of the functional architecture and cell mass and ultimately in cirrhosis. Cirrhosis results in destruction of the normal liver vasculature, increased resistance to blood flow, and back pressure into the portal circulation. Dilation of the venous system leading into the liver results in the formation of varices.241
Liver cell injury may occur to such a degree that a critical amount of hepatic necrosis results in the failure of the liver to maintain metabolic, synthetic and clearance functions leading to death. Liver cell injury may also occur more slowly, giving rise to chronic liver injury.236
