First described in the 1970s, stress-related mucosal damage (SMRD) was once considered a common complication of critical illness. However, it is now questionable whether the early descriptive work on the epidemiology and pathophysiology of stress ulceration remains applicable to the modern-day intensive care patient. Early endoscopic studies reported gastroduodenal abnormalities in most patients admitted to an intensive care unit (ICU), with more than 20% experiencing a gastrointestinal bleed significant enough to warrant blood transfusion. With the evolution of critical care, a decade later, the number experiencing a clinically significant bleed had dramatically fallen. Although infrequent now, clinically significant bleeding (CSB) remains associated with significant increased mortality and excess length of ICU stay estimated at 4 to 5 days. However, preventative measures to reduce stress ulceration have no effect on mortality or length of stay.
The management of SMRD in critically ill patients remains a contentious issue for intensivists. The potential benefit of prophylactic treatment must be weighed against the adverse risks and cost associated with treatment. If the decision to treat is made, then there is further uncertainty about the choice and formulation of agent. Undoubtedly, certain subsets of critically ill patients are at greater risk, and the identification and treatment of this group are likely to result in safe, cost-effective therapy.
Definitions
SRMD represents upper gastrointestinal damage ranging from asymptomatic superficial subepithelial lesions through to deeper lesions and ulceration extending into the submucosa and muscularis propria, causing occult bleeding, overt bleeding, or CSB.
Overt bleeding manifests as hematemesis, gross blood, or coffee-ground-like material in the nasogastric aspirate, hematochezia, or melena. CSB is overt bleeding complicated by hemodynamic changes (hypotension, tachycardia, or orthostasis) or by the need for blood transfusion or surgery.
Pathophysiology
The pathophysiology of SRMD is complex, and uncertainty remains regarding the exact mechanisms. However, it is likely that it arises from an altered balance of adequate mucosal protection and gastric acid production. The synthesis of the protective barrier is weakened when gut perfusion is compromised. Its protective effect is impaired by the presence of bile salts and uremic toxins, all common findings in critically ill patients. Interestingly, intraluminal gastric hyperacidity has not been identified as a major contributor to the development of SMRD outside of head trauma and burn patients.
Under normal physiologic circumstances, defense mechanisms prevent the erosion of the upper gastrointestinal mucosal lining by the acidic intraluminal contents. A glycoprotein mucous layer lines the stomach and forms a physical barrier to hydrogen ion back-diffusion ( Fig. 18-1, A ). Bicarbonate is trapped in this protective layer and neutralizes hydrogen ions before they reach the gastric epithelial layer. Adequate perfusion and oxygen delivery maintain intramural pH and prostaglandin synthesis, which is necessary for maintenance of the protective barrier layer.
Shock is common in critically ill patients, and septic shock is the most frequent cause of death in intensive care. Early in the systemic inflammatory response, splanchnic blood flow is reduced, resulting in gastric intestinal mucosal hypoperfusion. This is exacerbated by absolute or relative hypovolemia and arterial hypotension. The combination of hypovolemia, redistribution of cardiac output, and intense splanchnic microvascular vasoconstriction results in hypoperfusion and tissue hypoxia. Hypoxia leads to uncoupling of oxidative phosphorylation. Energy is derived from anaerobic glycolysis resulting in regional lactic acidosis and a decrease in tissue pH.
Hypoperfusion initially causes an ischemic mucosal injury. Accumulation of oxygen-free radicals contributes to tissue inflammation and cell death. A reduction in prostaglandin synthesis results in breakdown in the protective mucosal barrier; the epithelial layer is exposed to hydrochloric acid and pepsin, and erosions ensue ( Fig. 18-1, B ).
In the severely physiologically stressed critically ill patient, the combination of hypovolemia, activation of the sympathetic nervous system, global and regional hypoperfusion, endogenous and exogenous vasoactive agents, the release of proinflammatory cytokines, and activation of coagulation create a milieu that favors gastrointestinal ulceration and impairs protective and healing mechanisms.
Gastric intraluminal acidity (pH < 4) is necessary for the generation of stress ulceration. Fasting and prolonged gastric transit times may contribute to a more acidic upper gastrointestinal tract. This increased duration and intensity of acid exposure may increase the likelihood of erosions and ulceration.
Epidemiology
SMRD was once considered an unavoidable complication of critical illness. Early endoscopic evidence based on 40 patients reported gastroduodenal abnormalities in almost 75% of patients admitted to an ICU, with more than 20% (9 of 40) of patients experiencing a gastrointestinal bleed significant enough to warrant blood transfusion. With the evolution of critical care, a decade later, the number experiencing a clinically significant bleed had fallen to 2 to 4%. Nonetheless, on the basis of the early endoscopic studies and the fact that CSB remains associated with significant morbidity and mortality, stress ulcer prophylaxis (SUP) has become a standard of care in patients admitted to the ICU, with the intervention endorsed by many professional bodies. It has been reported that 90% of ICU patients will receive some form of SUP. Consequently, it is difficult to ascertain the true incidence of SMRD in the modern-day intensive care patient not receiving SUP.
A recently published large cohort study involving more than 35,000 patients in more than 70 hospitals between 2003 and 2008 reported the risk of gastrointestinal hemorrhage at 4%. A meta-analysis from 2010 based on 17 studies performed between 1980 and 2004 estimates the risk of CSB from stress ulceration at 1% in ICU patients. The variation in prevalence likely relates to the clinical endpoints studied and how the conditions of overt bleeding and CSB are defined.
The downward trajectory in the prevalence of CSB as a result of SMRD is due, at least in part, to improvements in the overall management of the intensive care patient. The decline in prevalence is widely attributed to the development of early goal-directed therapy with rapid restoration of intravascular volume and organ perfusion pressure, the use of lung-protective ventilatory strategies with a shorter duration of mechanical ventilation, the institution of Surviving Sepsis Campaign guidelines, and early enteral nutrition.
Risk Factors
Not all critically ill patients are at equal risk for having gastrointestinal hemorrhage. In the prospective multicenter cohort study of 2252 intensive care patients by Cook and associates, two independent risk factors for CSB were identified: respiratory failure (requiring mechanical ventilation for >48 hours; odds ratio [OR] 15.6) and coagulopathy (platelets <50,000; international normalized ratio >1.5 or activated partial thromboplastin time >2 times the control; OR, 4.3).
There was a trend toward increased bleeding in patients with hypotension (OR, 3.7), sepsis (OR, 2.0), renal failure (OR, 1.6), and glucocorticoid use (OR, 1.5), but these did not reach statistical significance. In a later study of 1077 critically ill mechanically ventilated patients, using a multivariable analysis, the same group demonstrated that renal failure (OR, 1.16) was independently associated with CSB, whereas enteral nutrition (OR, 0.3) and prophylaxis with ranitidine (OR, 0.39) conferred significantly lower bleeding rates. Two factors that appear to be independently predictive of stress ulcer bleeding in trauma patients are severe injury, as defined by an Injury Severity Score greater than 16, and injuries to the central nervous system (brain and spinal cord). In an observational study by Maury and coworkers, Helicobacter pylori infection was found to be associated with a 20% absolute increase in risk in critically ill patients who developed upper gastrointestinal hemorrhage (36% vs. 16%; P = .04; Table 18-1 ).
|
† Independent risk factor in mechanically ventilated patient.
Management
The prevention or limitation of SRMD and stress ulceration begins with restoration of splanchnic perfusion and prompt effective treatment of the underlying condition. Early goal-directed therapy with fluid and catecholamine resuscitation has been shown to reduce mortality and multiorgan dysfunction in patients with severe sepsis and septic shock. In shocked patients with splanchnic hypoperfusion, adequate volume loading is likely to be the most important initial intervention. The types of fluids, resuscitation endpoints, and monitoring techniques remain controversial. These issues are covered elsewhere in this book.
Specific Stress Ulcer Prophylaxis
Although CSB from SRMD occurs infrequently, the associated morbidity and mortality warrant a preventative approach in at-risk patients. Specific pharmacologic anti-stress ulcer therapies can be broadly divided into four groups: antacids, cytoprotectants, H 2 -receptor antagonists, and proton pump inhibitors (PPIs) ( Fig. 18-2 ).
Antacids act locally by directly neutralizing gastric acid and transiently increasing intraluminal pH. Frequent oral administration is required. Adverse effects include vomiting, constipation, metabolic alkalosis, and a range of electrolyte disturbances. Antacids are less efficacious than H 2 -receptor antagonists and PPIs in reducing gastric acidity, and they are currently not recommended as prophylaxis.
Sucralfate is the most extensively studied of the cytoprotectant agents. It is a sulfated polysaccharide complexed with aluminum hydroxide, which adheres to epithelial cells to form a physical protective gel layer on the gastric mucosa, reducing direct acid contact. Sucralfate is administered orally or by nasogastric tube. It does not significantly alter intraluminal pH; this may confer benefit in terms of gastric bacterial colonization. Other proposed benefits of sucralfate include (1) stimulation of mucous and bicarbonate secretion, (2) stimulation of epidermal growth factor, and (3) improved mucosal blood flow and enhanced prostaglandin release. Adverse effects are reduced absorption of enteral feed and some medications (quinolones, theophylline, phenytoin, ranitidine, ketoconazole, digoxin). There is also the potential for bezoar formation, clogging of nasogastric tubes, need for feeding breaks, and increased serum aluminum levels in patients receiving renal replacement therapy. Because sucralfate acts directly on the stomach, administration distal to the pylorus is ineffective. Sucralfate is more effective than placebo but inferior to H 2 -receptor antagonists.
Gastric acid secretion is an active, energy-demanding process. Agents such as H 2 -receptor antagonists and PPIs, by diminishing energy-demanding gastric acid secretion, may protect against the development of stress ulcers related to hypoperfusion.
H 2 -receptor antagonists act by reversible competitive inhibition of histamine-stimulated acid production and decrease overall gastric secretions. Enteral and parenteral formulations are available. These agents require frequent dosing, and there is some evidence to suggest that continuous infusions of the intravenous formulations may achieve better pH control than bolus administration. The phenomenon of tachyphylaxis, displayed by H 2 -receptor antagonists, raises concerns about their suitability for longer term use in the critically ill. Seventy percent of patients will achieve a gastric intraluminal pH of more than 4 within 24 hours, but this falls to just 26% by day 3. Adverse effects include central nervous system disturbances, especially in elderly patients with intravenous administration. In rare instances, hematologic disorders such as thrombocytopenia have been associated with H 2 -receptor antagonists. Cimetidine and ranitidine cause inhibition of cytochrome P-450 metabolism that reduces the clearance of many drugs (e.g., warfarin and phenytoin). H 2 -receptor antagonists reduce the risk of CSB when compared with placebo.
PPIs are substituted benzimidazoles that inhibit the H + /K + ATPase (gastric hydrogen potassium adenosine triphosphate) enzyme on the parietal cell. This inhibits histamine-induced and vagally mediated acid secretion, making PPIs the most potent agents in raising the pH of gastric contents. PPIs irreversibly bind to the proton pump, and subsequent secretion of acid can only occur with the synthesis of new enzyme. There are enteral and intravenous formulations available. Patients do not have tolerance to the antacid effects of PPIs, with 100% of patients maintaining a gastric intraluminal pH of more than 4 at day 3. However, rebound acid hypersecretion is common after discontinuation. Adverse effects are generally mild (e.g., gastrointestinal upset or headache), but an association with Clostridium difficile diarrhea has been reported. PPIs are metabolized by the cytochrome P-450 enzyme system; therefore there is potential for drug interaction. Omeprazole interferes with metabolism of cyclosporine, diazepam, phenytoin, and warfarin and increases the metabolism of several antipsychotic drugs and theophylline. Pantoprazole undergoes dual-pathway metabolism in the liver to inactive metabolites through the cytochrome P-450 system and sulfate conjugation. This results in fewer drug interactions that make pantoprazole particularly useful in critically ill patients, who typically are on numerous medications. A meta-analysis performed by Alhazzani and colleagues in 2013 concluded that PPIs provide the most reliable and sustained control of gastric acidity, making them more effective than H 2 -receptor antagonists at reducing clinically significant and overt upper gastrointestinal bleeding. However, the investigators observed no corresponding reduction in mortality rates or length of hospital stay. An earlier large-scale cohort study performed between 2003 and 2008 involving more than 35,000 patients found a greater risk of gastrointestinal bleeding when PPIs were used compared with H 2 -receptor antagonists, but the validity of this trial was questioned because the authors did not differentiate between overt and clinically significant gastrointestinal bleeding. The Surviving Sepsis Campaign recommends the use of PPIs over H 2 -receptor antagonists for SUP.
In patients with bleeding peptic ulcers, who require a higher gastric pH to maintain clot stability, the results of two trials suggest that omeprazole infusion can maintain the intragastric pH higher than 6 for several days, whereas the initial effectiveness of the H 2 -receptor antagonists in keeping the pH above 6 is quickly lost, most likely as a result of tolerance. A meta-analysis of 11 trials compared the efficacy of PPIs and H 2 -receptor antagonists in reducing the rate of rebleeding in patients with bleeding peptic ulcer disease. PPIs were found to be more effective in preventing persistent or recurrent bleeding, but there was no significant difference in the need for surgery or mortality rate. Blood clot integrity is dependent on a pH higher than 6, which is only reliably achieved with PPIs. Questions remain outstanding regarding the appropriate PPI dosing regimen for patients with bleeding peptic ulcers. An intragastric pH higher than 6 during most of the 24-hour periods is a prerequisite for the control of bleeding in patients with active bleeding ulcers because platelet aggregation will not occur below a pH of 5.9 and is optimal at a pH of 7 to 8. In a crossover trial in 10 patients taking omeprazole, a 40-mg intravenous bolus was compared with an 80-mg bolus plus 8-mg-per-hour infusion with the outcome measure of mean intragastric pH. The two regimens were equivalent for the first 12 hours. When the time with an intragastric pH above 6 during the first 24 hours was considered, the 80-mg bolus and 8-mg-per-hour infusion were superior. A more recent trial involving more than 230 patients with bleeding ulcers comparing continuous infusion of PPI with twice-daily dosing of intravenous PPI failed to show a difference in rebleeding rates, need for surgical intervention, blood transfusion requirements, or mortality.
When CSB occurs, the patient should be hemodynamically assessed and appropriate volume resuscitation instituted. Early multidisciplinary involvement is recommended with angiographic embolization, endoscopic intervention, and surgical control of bleeding vessels all options for definitive treatment in patients with gastrointestinal bleeding.
Related posts:
What Lessons Have Intensivists Learned During the Evidence-Based Medicine Era?
What Is the Role of Invasive Hemodynamic Monitoring in Critical Care?
Should Fever Be Treated?
Should Exacerbations of COPD Be Managed in the Intensive Care Unit?
Are Anti-inflammatory Therapies in ARDS Effective?
Has Outcome in Sepsis Improved? What Has Worked? What Has Not Worked?
Full access? Get Clinical Tree
