Although the incidence of nonvariceal upper gastrointestinal bleeding (UGIB) has been decreasing worldwide, nonvariceal UGIB continues to be a significant problem. Even with the advent of advanced endoscopic procedures and potent medications to suppress acid production, UGIB carries significant morbidity and mortality. Some of the most common risk factors for nonvariceal UGIB include Helicobacter pylori infection, nonsteroidal antiinflammatory drugs (NSAIDs), aspirin, selective serotonin reuptake inhibitors, and other antiplatelet and anticoagulant medications. In patients with cardiovascular disease and kidney disease, UGIB tends to be more severe and has greater morbidity. Many of the newer NSAIDs have been removed from the market.
Key points
- •
Incidence of nonvariceal upper gastrointestinal bleeding (UGIB) has been decreasing worldwide. However, nonvariceal UGIB continues to be a significant problem.
- •
The most common risk factors for nonvariceal UGIB include Helicobacter pylori infection, nonsteroidal antiinflammatory medications (NSAIDs), aspirin, selective serotonin reuptake inhibitors, and other antiplatelet and anticoagulant medications.
- •
More recently observed important risks for UGIB are cardiovascular disease, heart failure, left ventricular assist devices, and renal failure.
- •
Despite the introduction of cyclooxygenase-2 inhibitors, which were introduced to decrease UGIB, bleeding from NSAIDs continues to be a problem.
- •
Introduction of newer antiplatelet agents and newer oral anticoagulants has contributed to the persisting incidence of UGIB.
Introduction
Upper gastrointestinal bleeding (UGIB) is defined as hemorrhage that involves the mouth to the duodenum proximal to the ligament of Treitz. Common causes of UGIB include peptic ulcer disease (PUD) and ulcers of the esophagus, erosions of the upper gastrointestinal (GI) tract, variceal bleeding, gastroesophageal reflux disease, Mallory-Weiss tears, vascular lesions, malignancy, and other less common causes. Between 40% and 50% of patients who have UGIB present with hematemesis and 90% to 98% with either melena or hematochezia. The disease is commonly divided into 2 types: nonvariceal and variceal, and the primary focuses of this article are the epidemiology and risk factors of nonvariceal bleeding. Even with the use of advanced endoscopic procedures and potent medications to suppress acid production, UGIB still carries significant morbidity and mortality. Although studies have reported a decreased incidence in UGIB in the United States (96–82 per 100,000) it still accounts for nearly 300,000 hospitalizations per year, with a mortality of approximately 5%. PUD still accounts for most nonvariceal bleeding (37%) and is higher among elderly men, those who use aspirin and nonsteroidal antiinflammatory drugs (NSAIDs), and in areas with high Helicobacter pylori prevalence. In the United States in 2004, there were about 700,000 ambulatory care visits with peptic ulcer as the first-listed diagnosis, with a total cost of approximately $1.4 billion.
Introduction
Upper gastrointestinal bleeding (UGIB) is defined as hemorrhage that involves the mouth to the duodenum proximal to the ligament of Treitz. Common causes of UGIB include peptic ulcer disease (PUD) and ulcers of the esophagus, erosions of the upper gastrointestinal (GI) tract, variceal bleeding, gastroesophageal reflux disease, Mallory-Weiss tears, vascular lesions, malignancy, and other less common causes. Between 40% and 50% of patients who have UGIB present with hematemesis and 90% to 98% with either melena or hematochezia. The disease is commonly divided into 2 types: nonvariceal and variceal, and the primary focuses of this article are the epidemiology and risk factors of nonvariceal bleeding. Even with the use of advanced endoscopic procedures and potent medications to suppress acid production, UGIB still carries significant morbidity and mortality. Although studies have reported a decreased incidence in UGIB in the United States (96–82 per 100,000) it still accounts for nearly 300,000 hospitalizations per year, with a mortality of approximately 5%. PUD still accounts for most nonvariceal bleeding (37%) and is higher among elderly men, those who use aspirin and nonsteroidal antiinflammatory drugs (NSAIDs), and in areas with high Helicobacter pylori prevalence. In the United States in 2004, there were about 700,000 ambulatory care visits with peptic ulcer as the first-listed diagnosis, with a total cost of approximately $1.4 billion.
Risk factors
The most common risk factors for nonvariceal UGIB include H pylori infection, NSAIDs/aspirin, and other antiplatelet and anticoagulant medication use. H pylori infection and NSAID use are both independent and synergistic risk factors for peptic ulcer–related bleeding. Studies from the 1980s and 1990s show that H pylori was present in more than 90% of patients with duodenal ulcers and approximately 70% of patients with gastric ulcers. However, more recent Western studies have suggested that there is a changing cause of PUD, with the overall PUD incidence decreasing and the proportion of H pylori –negative PUD increasing. A Dutch study of the incidence of duodenal and gastric ulcers in a district hospital found that the incidence of duodenal ulcer disease has been decreasing with a decline in the prevalence of H pylori . A review of 73 worldwide published studies from 1999 to 2008 evaluating patients with duodenal ulcers showed that 88% of patients were infected with H pylori , with a decrease to 77% infected when including the studies from 2003 to 2008. There has been an overall decline in the prevalence of H pylori –positive PUD and an increase in non-NSAID, non– H pylori PUD. According to a prospective multicenter study involving 32 hospitals in France, 40% of PUD was related to H pylori infection alone, 18.7% to gastrotoxic drugs alone (NSAIDs, aspirin, and/or cyclooxygenase [COX]-2 inhibitors), 19.8% had H pylori infection in the setting of gastrotoxic drugs, and 21.6% had neither H pylori infection nor gastrotoxic drug use. Therefore, although H pylori and gastrotoxic drugs make up approximately 80% of PUD and PUD-related bleeding, there is a significant subsection of idiopathic ulcers as well.
There has been a well-established correlation between NSAID use and the increased risk of UGIB. The prevalence of PUD in patients with regular NSAID use is approximately 15% to 30%. In addition to NSAID use independently increasing the risk of GI adverse effects, there are multiple other risk factors that have been associated with nonvariceal UGIB that are potentiated in the presence of NSAID use ( Fig. 1 ).
COX-2 inhibitors were first introduced in the hope that they would provide the analgesic and antiinflammatory benefits of nonselective NSAIDs without the adverse GI side effects. There is evidence that there is a reduced risk of upper GI complications and ulcers compared with those associated with nonselective NSAIDs. However, overall, there is still an increased risk of upper GI complications and symptomatic gastroduodenal ulcers with COX-2 selective inhibitors compared with no use at all.
Recently, a 2014 case series analysis of 7 population-based health care databases for 114,835 patients with UGIB analyzed various drug combinations with NSAIDs, low-dose aspirin, and COX-2 inhibitors. Monotherapy with nonselective NSAIDs was shown to increase the relative risk of UGIB by a factor of 4.3, compared with monotherapy with COX-2 inhibitors (relative risk of 2.9) and monotherapy with low-dose aspirin (relative risk of 3.1). Combination therapy also increases risk of UGIB, with the greatest increase seen with the combination of NSAIDs with corticosteroids ( Table 1 ). Other studies have also shown an increased risk for the combination of NSAIDs and aspirin, whereas other suggested risk factors from studies include alcohol consumption, history of prior peptic ulcers, and smoking.
| Drug Groups | Monotherapy | Combination with: | ||||||
|---|---|---|---|---|---|---|---|---|
| nsNSAIDs | COX-2 Inhibitors | Low-dose Aspirin | ||||||
| N | IRR (95% CI) | N | IRR (95% CI) | N | IRR (95% CI) | N | IRR (95% CI) | |
| No drug a | 69,664 | 1.00 (reference) | NA | NA | NA | |||
| nsNSAIDs | 3327 | 4.27 (4.11–4.44) | NA | NA | 416 | 6.77 (6.09–7.53) | ||
| COX-2 inhibitors | 635 | 2.90 (2.67–3.15) | NA | NA | 131 | 7.49 (6.22–9.02) | ||
| Low-dose aspirin | 4733 | 3.05 (2.94–3.17) | 416 | 6.77 (6.09–7.53) | 131 | 7.49 (6.22–9.02) | NA | |
| Corticosteroids | 1378 | 4.07 (3.83–4.32) | 244 | 12.82 (11.17–14.72) | 40 | 5.95 (4.25–8.33) | 190 | 8.37 (7.14–9.81) |
| SSRIs | 1793 | 2.06 (1.94–2.18) | 210 | 6.95 (5.97–8.08) | 65 | 5.82 (4.45–7.62) | 401 | 4.60 (4.09–5.17) |
| GPAs | 5279 | 1.61 (1.56–1.66) | 678 | 3.90 (3.59–4.24) | 95 | 2.37 (1.92–2.93) | 607 | 2.54 (2.32–2.78) |
| Aldosterone antagonists | 1211 | 3.27 (3.06–3.50) | 76 | 11.00 (8.63–14.03) | 10 | 4.02 (2.07–7.81) | 131 | 5.01 (4.13–6.08) |
| Ca channel blockers | 3546 | 1.57 (1.51–1.63) | 363 | 4.45 (3.98–4.98) | 77 | 3.11 (2.46–3.93) | 1123 | 3.07 (2.86–3.29) |
| Anticoagulants | 1760 | 3.01 (2.85–3.19) | 143 | 8.69 (7.30–10.35) | 21 | 5.01 (3.21–7.82) | 168 | 6.94 (5.86–8.22) |
| Antiplatelets (excluding aspirin) | 994 | 1.74 (1.61–1.87) | 87 | 6.50 (5.19–8.15) | 9 | 1.73 (0.87–3.44) | 246 | 5.49 (4.71–6.41) |
| Nitrates | 2572 | 2.55 (2.43–2.68) | 172 | 5.82 (4.97–6.82) | 49 | 5.09 (3.79–6.82) | 859 | 3.79 (3.51–4.10) |
There is growing evidence that use of selective serotonin reuptake inhibitors (SSRIs) is associated with an increased risk of UGIB, and that this risk is further increased by the concurrent use of NSAIDs and SSRIs. Previous reviews of a small number of studies have reported a substantial risk of UGIB with SSRIs. However, more recent studies have produced variable results. A meta-analysis of 4 observational studies showed an odds ratio (OR) of 2.36 (95% confidence interval [CI], 1.44, 3.85) for use of SSRI-associated UGIB. A more recent 2014 systematic review of 19 case-control studies showed a significant risk with an OR of 1.66 (95% CI, 1.44, 1.92). Moreover, the combination of SSRIs and NSAIDs further increases UGIB risk to an OR of 4.25 (95% CI, 2.82, 6.42) (see Fig. 1 ). Because serotonin induces platelet aggregation, it has been postulated that SSRIs increase UGIB risk by decreasing platelet-generated serotonin, resulting in impaired platelet aggregation and hemostasis. It has also been hypothesized that, because SSRIs increase gastric acid secretion, they may alternatively increase risks of peptic ulcers and UGIB from ulcers of the upper GI tract.
In patients with ulcers in whom tests for H pylori are negative and the history is also negative for NSAID or SSRI usage, important considerations are whether the tests for H pylori may be falsely negative because of concurrent use of proton pump inhibitors (PPIs); whether the patient might be taking an NSAID product that is unfamiliar to the patient (such as an over-the-counter NSAID); whether the patient might be taking an NSAID product that is part of a combination medication (such as combination analgesics or combination cough and cold medications); or whether the patient might have a gastric acid hypersecretory condition, such as Zollinger-Ellison syndrome.
Prevalence/Incidence
The overall reported incidence rates for acute UGIB vary from 48 to 160 cases per 100,000 population per year, with higher incidences reported among men and the elderly. The most common cause of acute UGIB continues to be PUD, with reported incidences as high as 67% of all acute UGIB ( Table 2 ).
| Cause | Incidence (%) |
|---|---|
| PUD | 20–67 |
| Erosive disease | 4–31 |
| Bleeding varices | 4–20 |
| Esophagitis | 3–12 |
| Mallory-Weiss tears | 4–12 |
| Malignancy | 2–8 |
| Vascular lesions | 2–8 |
| Esophageal ulcers | 2–6 |
| Idiopathic/unknown cause | 3–19 |
The overall incidence of acute UGIB, particularly PUD, decreased toward the end of the twentieth century and has now stabilized in the twenty-first century. In the United States, the incidence rates for hospital discharges that carried the diagnosis of PUD decreased from 1400 per 100,000 in 1979 to approximately 700 per 100,000 in 2004 ( Fig. 2 ). In addition, a 2012 article in the American Journal of Gastroenterology by Laine and colleagues indicated that there were 36 million inpatient visits for PUD across approximately 620 US hospitals from 2001 to 2009. Although alarming, these data revealed decreases in overall UGIB from 78.4 to 60.6 per 100,000 over that 8-year time frame. Furthermore, there was a decrease in peptic ulcer bleeding from 48.7 to 32.1 per 100,000 as well ( Fig. 3 ). Corroborating these earlier data are UGIB rates from a more recent 2014 longitudinal study of UGIB hospitalizations in the United States that used the Nationwide Inpatient Sample and indicated that nonvariceal UGIB incidence declined from 108 to 78 cases per 100,000 persons from 1994 to 2009, respectively ( Table 3 ).
Related posts:
Advances and Improvements in the Management of Upper Gastrointestinal Bleeding
Initial Assessment and Resuscitation in Nonvariceal Upper Gastrointestinal Bleeding
Endoscopic Diagnosis and Therapy in Gastroesophageal Variceal Bleeding
Injection and Cautery Methods for Nonvariceal Bleeding Control
Unusual Causes of Upper Gastrointestinal Bleeding
Tips and Tricks on How to Optimally Manage Patients with Upper Gastrointestinal Bleeding
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
