Stratifying Patients with Upper Gastrointestinal Hemorrhage
Stratifying Patients with Upper Gastrointestinal Hemorrhage
ABSTRACT & COMMENTARY
Source: Corley DA, et al. Early indicators of prognosis in upper gastrointestinal hemorrhage. Am J Gastroenterol 1998; 93:336-340.
This retrospective, observational study identified 17 variables that were possibly associated with an adverse outcome in patients who were admitted for upper GI bleeding. All patients in the study were endoscoped. The purpose of the study was to identify independent predictors of adverse outcome (before endoscopy is performed), and to develop a decision rule that would allow identification of patients at low risk for adverse outcome.
The investigators found five independent predictors of adverse outcome (defined as death, the need for operation, recurrent hematemesis, recurrent melena, or a falling hematocrit despite transfusion). The independent correlates were: 1) evidence of portal hypertension as defined by a history of cirrhosis and/or evidence of ascites on physical examination; 2) a history of hematemesis (red blood); 3) red blood on nasogastric tube aspiration; 4) initial systolic blood pressure lower than 100 mmHg; and 5) initial hematocrit lower than 30%. Patients with no risk factors had an adverse outcome rate of 9%. The adverse outcome rate in the subgroup of patients with zero or only one risk factor was 11.6%. Patients with two risk factors had a 34% risk of adverse outcome. Corley and colleagues conclude that their decision rule may permit triage of "low-risk" patients to less costly levels of care without compromising patient outcomes.
COMMENT BY GLENN C. FREAS, MD, JD, FACEP
This study presents some useful information, but it should be interpreted with caution. It may be quoted by our colleagues in medicine and GI, so it is important to know what it does and does not say.
Corley et al point out that this study should be replicated in other patient populations. This would enable their decision rule to be broadly applied. Another important point is that almost 14% of the patients with zero or only one risk factor not only had endoscopy but also had endoscopic therapy. This undermines the data for this group; that is, perhaps they had a good outcome because of therapeutic endoscopy.
Despite these observations, I think the study is useful. It clearly defines subgroups of populations who are at significant risk for rebleeding. It also helps us determine which patients may remain stable. Patients with two or more of the five correlates for adverse outcome should be considered critically ill. The risk factor of portal hypertension outperformed any of the other factors in predicting poor outcome. Conversely, absence of vomiting red blood and absence of red blood in the nasogastric aspirate were highly significant factors in predicting good outcome. Patients with no active comorbid conditions, who have none of the correlates for adverse outcome, can probably be admitted to a non-critical care setting and may not require emergent endoscopy. Nonetheless, remember that at least 9% of these patients may rebleed. Even the above observations and assumptions must be tempered by your patient population and your departmental/hospital resources. Remember what this study does not say. It does not define which patients need emergent endoscopy, and it does not say that patients with none of the five correlates can be safely discharged from the ED.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.