Skip to Main Content

Make the Diagnosis: Upper Gastrointestinal Bleed

Prior Probability of an Upper Gastrointestinal Bleed

Among patients who present with a gastrointestinal hemorrhage, an upper gastrointestinal bleed (UGIB) is more likely (incidence 63%, 95% CI, 51-73%) than a lower gastrointestinal bleed (LGIB). Among patients with UGIB, 36% (95% CI, 29-44%) require urgent intervention for severe bleeding.

Population in Whom UGIB Should Be Considered

Patients with gastrointestinal bleeding may present with visual evidence of blood loss such as hematemesis, hematochezia, melena, or coffee-ground emesis. Patients may not always reliably recognize melena as visual evidence of bleeding. A UGIB source should be considered in all patients with evidence of gastrointestinal blood loss.

Assessing the Likelihood of UGIB in Patients With Gastrointestinal Bleeding

Patients with a prior UGIB, younger patients (age < 50), or those with cirrhosis are more likely to have a UGIB than an LGIB (see Table 87-1). Although not all black stools represent melena, patients who report black stools are much more likely to have a UGIB source. The physician should inspect a stool sample when the patient has a GIB. Confirming melena is the most useful finding for identifying patients mostly likely to have a UGIB. When clots are seen in the stool, the patient is much less likely to have a UGIB. Because a negative nasogastric lavage cannot rule out a UGIB or an UGIB requiring intervention, many experts do not advocate for its use as a routine diagnostic test, and they instead favor endoscopy.

Table 87-1.Useful Findings for Assessing the Likelihood of a UGIB

Assessing the Likelihood That a Patient With Gastrointestinal Bleeding Is Having Severe Bleeding

The severity of the bleeding guides the need and urgency of subsequent procedures. A rapid pulse (> 100/minute) or significant anemia (hemoglobin < 8 g/dL) are the most useful findings to identify the patient suffering from severe bleeding (Table 87-2). Up to 22% of patients will have a Blatchford score of 0 (Table 87-3), and these patients are highly unlikely to have a severe hemorrhage.1-2 The Blatchford score can be determined without nasogastric lavage.

Table 87-2.Useful Findings for Assessing the Likelihood of a Severe UGIB

Pop-up div Successfully Displayed

This div only appears when the trigger link is hovered over. Otherwise it is hidden from view.