Stress ulcers frequently are asymptomatic and come to clinical attention only when they manifest bleeding. Progression to stress ulcer syndrome (SUS) (i.e., clinical bleeding or intestinal perforation) is uncommon. When it occurs, GI bleeding typically presents within 14 days of the onset of physiologic stress or ICU admission [11]. Hematemesis, gross blood from the nasogastric tube, and melena are the usual presentations of SUS [1,2,3,4,12,13]. The true incidence of overt GI bleeding resulting from stress ulceration is difficult to ascertain and varies depending on the definition of bleeding and the category of ICU patients. The finding at preemptive endoscopy of a small amount of blood adjacent to an ulcer without clinically evident overt bleeding has been noted in 22% to 36% of ICU admissions [2,3]. Significant stress ulcer bleeding occurs in 2% to 7% of critically ill patients [14,15]. Patients with thermal injury from burns or with acute intracranial disease including head trauma and coma appear to be at increased risk of developing stress ulcer–related bleeding [3,12]. Evidence is surfacing that the incidence of stress ulcer bleeding is decreasing, from as high as 22% in the 1970s to 1.6% to 6.0% of patients admitted to ICUs in more recent studies [3,16,17,18,19]. Our recent experience suggests that SUS now occurs in less than 1% of ICU admissions [20]. This may be related to the common use of prophylactic agents and improved care of critically ill patients in the present-day ICU setting.

Stress ulceration with clinically significant bleeding is associated with a mortality rate as high as 50% to 80% [16,17,18,19,21], although death often is a result of the underlying disease and not directly linked to GI bleeding [16]. Thus, stress ulcer bleeding may serve more as a marker for the severely ill patient having a cascade of complications rather than represent a unique clinical entity.

Clinical risk factors commonly associated with SUS include mechanical ventilation, coagulopathy, major surgery, hemorrhagic shock, hypotension, trauma, and sepsis [11]. A statistically significant predisposition to stress ulceration has been demonstrated in ICU patients with coagulopathy or a requirement for prolonged mechanical ventilation (over 48 hours) [16,17]. Hypotension and shock also are more frequent among patients with bleeding attributed to stress ulceration but do not reach statistical significance as independent risk factors [16]. Patients with thermal injury from burns (> 35% body surface area) and patients with acute intracranial disease including head trauma and coma also are at increased risk of having stress ulcers (Curling’s and Cushing’s ulcers, respectively) [22,23]. Conditions that present a low risk for SUS include myocardial infarction, congestive heart failure, arrhythmias, chronic renal failure on dialysis, chronic obstructive pulmonary disease, and malignancy [11]. Recent studies emphasize that although the incidence of GI bleeding increases with up to two risk factors, additional risk factors do not further increase bleeding potential [24].

Stress ulcer is a misnomer in that many of the lesions appear as shallow mucosal erosions without the depth of an ulcer (Table 92.1). The earliest mucosal changes have been described in the most proximal parts of the stomach [25], but the process can eventually involve the distal stomach and duodenum [17,20]. Early mucosal changes include pallor, mottling, and submucosal petechiae which coalesce to form superficial linear erosions and ulcers. This can progress to diffuse mucosal damage with bleeding and in rare cases, perforation. On endoscopy, mucosal erosions have been found as early as 5 hours after ICU admission, and most are evident within 72 hours. Within 24 hours of ICU admission, they have the appearance of small (1 to 2 mm), round, shallow erosions, and by 48 hours, the lesions can be larger (2 to 25 mm), deeper, and associated with a clot. When patients bleed, 10 or more gastric mucosal erosions usually are observed at endoscopy [25,26]. The result is a diffuse area of involvement that can result in a spectrum of manifestations ranging from oozing of blood to massive hemorrhage or perforation [11,16].

Although the precise etiology of stress ulceration is not known, a number of factors have been implicated and studied in animal models. Animal studies using endoscopic and pathologic analyses have demonstrated progressive gastric mucosal injury from the physiologic stress of induced shock [27], a process that has been observed in humans under similar stress [1,27,28]. Gastric acid, but not hypersecretion, appears to be an essential prerequisite for the development of stress ulceration [29,30]. The classic lesion has been associated with a
normal or even decreased intraluminal gastric acid secretion [31,32,33], suggesting that a breakdown in the mucosal defense mechanism must also be present for stress ulceration to occur. Mucosal ischemia is thought to be the inciting event in the pathogenesis [31,32]. In experimental models, stress causes a decrease in gastric mucosal blood flow, resulting in decreased delivery of oxygen and nutrients and leading to a deficit in aerobic metabolism and high-energy phosphate compounds [30,34,35]. Diminished intramucosal blood flow decreases the availability of systemic bicarbonate to buffer back-diffused hydrogen ions, thereby allowing a fall in the intramural gastric pH [30,31,36]. Mucosal ischemia and subsequent reperfusion result in the formation of toxic oxygen-derived free radicals and superoxides [33,37,38] while decreasing the synthesis of cytoprotective prostaglandins [38], thereby creating a favorable scenario for mucosal damage.

Factors such as elevated gastrin and pepsin levels, increased intraluminal bile and urea concentrations, and decreased gastric motility from enhanced vagal stimulation appear to be contributory to the development of a stress ulcer but are inadequate by themselves to induce the lesion [31,39]. The impact, if any, of Helicobacter pylori on the risk and development of stress ulcer is unknown. H. pylori seropositivity has been found to be lower in patients with stress ulcer bleeding when compared with controls [40]. In fact, it is possible that some of the H. pylori–positive cases diagnosed as stress ulcer actually represent H. pylori–associated peptic ulcer disease.