The established quality indicators for early management of upper gastrointestinal (GI) hemorrhage are based on rapid diagnosis, risk stratification, and early management. Effective preendoscopic treatment may improve survivability of critically ill patients and improve resource allocation for all patients. Accurate risk stratification helps determine the need for hospital admission, hemodynamic monitoring, blood transfusion, and endoscopic hemostasis before esophagogastroduodenoscopy (EGD) via indirect measures such as laboratory studies, physiologic data, and comorbidities. Early management before the definitive EGD is essential to improving outcomes for patients with upper GI bleeding.
Key points
- •
Early resuscitation including support for airway, breathing, and circulation is critical in the initial evaluation of a patient with a suspected upper gastrointestinal (GI) hemorrhage.
- •
Early risk stratification for mortality and rebleeding should be performed using validated decision tools such as the Glasgow-Blatchford Score or the Clinical Rockall Score.
- •
Early administration of proton pump inhibitors decreases the need for endoscopic intervention.
- •
Vasoactive medications such as somatostatin, vasopressin, and their analogues should be given in cases of acute variceal upper GI hemorrhage and considered in cases of acute nonvariceal upper GI hemorrhage.
- •
Improved survival has been shown to be associated with a restrictive strategy of blood transfusion.
Background
Acute upper gastrointestinal (GI) hemorrhage is a common presentation in hospital-based emergency departments (EDs) in the United States and around the world. According to data from the Healthcare Cost and Utilization Project, there were 863,000 US hospital admissions for GI hemorrhage in 2008, which included both upper and lower GI bleeding. The mean length of stay for patients discharged from the hospital with a diagnosis of GI hemorrhage is 4.5 days and the mean hospital charges are $26,210 per admission. Acute upper GI hemorrhage is a particularly severe manifestation of GI hemorrhage and is associated with mortalities ranging from 15% to 20%. Acute upper GI hemorrhage is blood loss that has a source proximal to the ligament of Treitz, and is divided into variceal and nonvariceal sources. On initial evaluation of a patient with a suspected acute upper GI hemorrhage, the physician must perform a rapid clinical assessment in order to estimate the source of bleeding and the severity. However, early management is a challenge because the patient presentation is variable, the risk stratification is imperfect, and the tools to stop the bleeding before an esophagogastroduodenoscopy (EGD) are limited.
Presentation
The most common presentation for an upper GI hemorrhage is hematemesis, coffee-ground emesis, or melena. For a lower GI hemorrhage, the most common presentation is bright red blood per rectum. However, the distinction between an upper and lower source of hemorrhage is not always obvious. Bright red blood per rectum may be the presenting symptom in an especially brisk upper GI hemorrhage. In addition, several conditions and medications may mimic a GI hemorrhage. Nosebleeds, dental bleeding, tonsilar bleeding, and red drinks or food can be mistaken for hematemesis; bismuth-containing medications such as Pepto-Bismol can mimic melena; and vaginal bleeding and gross hematuria can mimic bright red blood per rectum. Although the occult blood test can be used to confirm the presence of blood in the stool, false-positive occult blood tests can occur with red meat, turnips, horseradish, and vitamin C ( Table 1 ). Oral iron supplements do not cause a false-positive occult blood result. In addition, patients may have bleeding lesions and only present with systemic signs of blood loss such as weakness, light-headedness, chest pain, or syncope.
| Mimics for hematemesis | Nosebleeds, tonsilar bleeding, red drinks and food |
| Mimics for melena | Bismuth medications such as Pepto-Bismol |
| Mimics for bright red blood | Red food, vaginal bleeding, gross hematuria |
| False-positive occult blood test | Red meat, turnips, horseradish, vitamin C |
The most common type of acute upper GI hemorrhage is the nonvariceal hemorrhage, which includes peptic ulcer disease, Mallory-Weiss tears, and Dieulafoy lesions. Variceal hemorrhage usually occurs in patients with portal hypertension and is always considered high risk for rebleeding and mortality. Nonvariceal upper GI hemorrhage can present with a range of severity from self-limiting to life threatening. As part of the natural course of the disease, 80% of the nonvariceal hemorrhages resolve spontaneously but 10% lead to death. Upper endoscopy, also called EGD, is the diagnostic and therapeutic test of choice for patients who present with an upper GI hemorrhage; however, the EGD is not always feasible at initial presentation.
Initial management
As with all critically ill patients, early assessment begins with the A, B, Cs: airway, breathing and circulation. In the setting of GI hemorrhage, airway control is especially important because aspiration of vomited blood is associated with significant morbidity and mortality. The placement of an advanced airway may be indicated in cases of profuse vomiting or altered mental status. Rapid sequence intubation, a method of intubation that quickly induces unconsciousness and paralysis, is the ideal method to perform endotracheal intubation because it is associated with a reduced risk of aspiration. Breathing is evaluated by clinical criteria such as chest rise, absence of cyanosis, and normal oxygen saturation levels by pulse oximetry. Circulatory or hemodynamic status is assessed with blood pressure, heart rate, and signs of end organ hypoperfusion such as altered mental status, increased capillary refill, decreased urine output, and increased lactate ( Table 2 ). Elderly patients and patients on hypertensive medications may not become tachycardic in response to volume loss. Abnormal changes in vital signs measured while patients are lying, sitting, and standing (orthostatics) support the presence of volume loss and augments the information gathered from standard vital signs. Potential volume loss should be addressed with 2 large-bore (16 gauge or 18 gauge) intravenous angiocatheters to enable fluid resuscitation and targeted blood transfusion. The initial fluid replacement should be crystalloid and a complete blood count, metabolic panel, coagulation panel, and type and screen should be ordered with initial intravenous access.
| Class 1 | Class 2 | Class 3 | Class 4 | |
|---|---|---|---|---|
| Blood loss (mL) | <750 | 750–1500 | 1500–2000 | >2000 |
| Heart rate (BPM) | <100 | 100–120 | 120–140 | >140 |
| BP | Normal | Normal | Decreased | Decreased |
| Pulse pressure | Normal/Increased | Decreased | Decreased | Decreased |
| Respiratory rate (breaths/min) | 14–20 | 20–30 | 30–40 | >40 |
| Urine output (mL/h) | >30 | 20–30 | 5–15 | Negligible |
| Mental status | Slightly anxious | Mildly Anxious | Anxious, confused | Lethargic |
Once hemodynamic stability has been established, a thorough history and secondary survey should be conducted with specific questions regarding prior episodes of bleeding, including the amount of blood and color of the blood. However, a patient’s assessment of blood loss is frequently inaccurate. Comorbidities such as congestive heart failure, renal disease, liver disease, or vascular disease increase overall mortality risk and help determine the source of bleeding. Recent vascular surgery increases the risk for an aortoenteric fistula, a particularly severe cause of upper GI hemorrhage. In addition, medications such as steroids, nonsteroidal antiinflammatory drugs, and anticoagulants also increase the risk of bleeding. Patients should be examined for signs of liver failure that might indicate that the bleed has a variceal source. Stigmata of liver disease include vascular collaterals (caput medusa), ascites, asterixis, gynecomastia, hepatomegaly, jaundice, scleral icterus, vascular spiders (spider telangiectasias, spider angiomata), and splenomegaly.
Risk stratification
An early challenge in the initial evaluation is to distinguish the severe from the benign causes of upper GI hemorrhage. Lacking the ability to directly visualize the bleeding source with an EGD, surrogate diagnostic techniques such as clinical decision rules and nasogastric aspiration (NGA) are often used at initial presentation. Clinical decision rules such as the Clinical Rockall Score (CRS) and the Glasgow-Blatchford Score (GBS) that incorporate vital signs, laboratory values, and comorbidities to risk stratify patients have been derived and validated to identify low-risk patients before endoscopy ( Tables 3 and 4 ). Both have been validated to help predict high risk and low risk for rebleeding and mortality. The GBS has been shown to be useful in discharging some patients with a very low risk who can be followed as outpatients. In a prospective study of 676 patients, 16% had a GBS of zero and none of these patients were required to have an endoscopic interventions or died at outpatient follow-up. In addition to identifying some patients who may be managed as outpatients, the GBS can predict the risk of rebleeding, the need for transfusion, and the need for surgery. GBS seems to be superior to the CRS at predicting rebleeding but both GBS and CRS have been shown to be effective at predicting the risk of death. Limitations to the GBS are the low adoption by US physicians and the low specificity of the test. An additional limitation is that an endoscopic view (using traditional EGD) still seems to be desired by physicians and patients regardless of the risk score.
| Risk Marker | Score Value |
|---|---|
| BUN (mmol/L) | |
| 6.5–8.0 | 2 |
| 8.0–10.0 | 3 |
| 10.0–25.0 | 4 |
| ≥25.0 | 6 |
| Hemoglobin (g/L) for Men | |
| 120–130 | 1 |
| 100–120 | 3 |
| <100 | 6 |
| Hemoglobin (g/L) for Women | |
| 100–120 | 1 |
| ≤100 | 6 |
| SBP (mm Hg) | |
| 100–109 | 1 |
| 90–99 | 2 |
| <90 | 3 |
| Other Markers | |
| Pulse ≥100 BPM | 1 |
| Melena | 1 |
| Syncope | 2 |
| Hepatic disease | 2 |
| Cardiac failure | 2 |
| Score | ||||
|---|---|---|---|---|
| Variable | 0 | 1 | 2 | 3 |
| Age (y) | <60 | 60–79 | ≥80 | — |
| Shock | No shock | SBP ≥100 Pulse ≥100 BPM | Hypotension | — |
| Comorbidity | No major | — | Cardiac failure, ischemic heart disease, major comorbidities | Renal failure, liver failure, disseminated malignancy |
| Diagnosis | Mallory-Weiss tear, no lesion identified, and no stigmata | All other diagnoses | Malignancy upper GI tract | — |
| Major stigmata of recent hemorrhage | None or dark Spot only | — | Blood in GI tract, adherent clot, visible or bleeding vessel | — |
NGA is a diagnostic tool that has been used to risk stratify patients for many years. Despite the claim that insertion of the tube is the single most disliked procedure by patients in EDs, NGA is often used to predict the presence of a high-risk lesion that requires endoscopic hemostasis. The sensitivity of a bloody aspirate is estimated at 45% and specificity 72% for a high-risk lesion. False-positives can occur from epistaxis caused by the tube insertion and false-negatives occur when the lesion is in the duodenum or a nonbleeding visible vessel. In general, a positive NGA is a good indicator of an upper GI hemorrhage, whereas a negative NGA is insufficient to rule out an acute bleed. NGA has not been shown to have an effect on clinical outcomes such as mortality, surgery, and length of hospital stay, or transfusion requirements. The only clinical outcome that is affected by the NGA is a decreased time to endoscopy, which may reflect a belief that a positive aspiration indicates the need for expedited care. There is no evidence that placement of a nasogastric tube leads to bleeding by esophageal varices.
A novel approach to risk stratify an upper GI hemorrhage in the ED is to use video capsule endoscopy to directly visualize the bleeding. In 3 small ED-based studies, video capsule endoscopy showed good patient tolerance and good sensitivity for detecting acute upper GI hemorrhage. Potential advantages of video capsule endoscopy include improved patient tolerance compared with NGA, point-of-care results obtained without a specialist at the bedside, and decreased hospitalization rates. Potential barriers to adoption include the cost of the equipment, the need to train ED physicians to interpret the videos, the need to create a secure means to transmit images to on-call GI specialists, and the increased ED length of stay that may be required to use video capsule endoscopy in an ED. Further studies are needed to confirm the sensitivity of the test in multiple settings for detecting high-risk bleeding lesions and as a means to guide clinical decisions compared with standard of care.
Pharmacologic therapy
Although definitive management of a bleeding lesion requires endoscopic or surgical intervention, preendoscopic pharmacologic therapy may assist in achieving hemostasis. Six randomized controlled trials have been conducted to determine the effect of using preendoscopic proton pump inhibitors (PPIs) versus controls for patients with upper GI hemorrhage. Reducing acid secretion may stabilize an upper GI bleeding lesion and possibly improve clinical outcomes. PPIs reduce the stigmata of high-risk lesions 9-fold and reduce the need for endoscopic intervention 3-fold, but do not reduce mortality, need for blood transfusion, need for surgery, or rebleeding rates ( Table 5 ). The use of PPIs should not replace emergent endoscopy but may improve outcomes when started early. Future large-scale randomized control trials are still needed to determine whether acid suppression by intravenous PPIs promotes a benefit in mortality.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
