Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Several Factors Contribute to GI Bleeding Mortality Risk

Date

Patients who developed upper gastrointestinal (GI) bleeding during a hospital stay experienced worse adverse outcomes than those admitted for upper GI bleeding alone, French researchers found in a prospective study.

Currently hospitalized patients (dubbed “inpatients”) with upper GI bleeding showed a significantly higher mortality rate at 6 weeks than patients hospitalized for GI bleeding alone (dubbed “outpatients”), at 21.7% versus 8.8%, respectively (P<0.0001), as well as increased frequency of rebleeding, at 18.6% and 14.4% (P=0.015) wrote Weam El Hajj, MD, of the Groupe Hospitalier Intercommunal Le Raincy-Montfermeil in France, and colleagues.

Adjusted analyses found rebleeding, comorbidities, hemodynamic instability, severity of bleeding and simply being an “inpatient” were associated with mortality at 6 weeks, according to the findings in the United European Gastroenterology Journal.

Despite declining incidence of upper GI bleeding in patients over the past decades, the rates of rebleeding and mortality remained stable or increased slightly. Modifiable risk factors need to be identified to help prevent death, the authors said.

Researchers investigated the outcomes among “inpatients” and “outpatients” with upper GI bleeding (variceal and non-variceal). From November 2017 to October 2018, researchers collected data on 2,498 patients with upper GI bleeding from 46 hospitals. “Inpatients” were defined as patients who developed variceal or non-variceal bleeding at least 24 hours after hospitalization, and “outpatients” were defined as those who presented with bleeding upon admission.

Patients were included if they were age 18 or older, had upper GI bleeding, with hematemesis, and had melana or acute decreases in hemoglobin levels with blood in the stomach.

The primary outcomes included mortality and rebleeding rates, assessed at 6 weeks from onset. Secondary outcomes were hospital stay duration, and the requirement for radiological or surgical intervention.

The majority of participants were outpatients (75%). The average age of inpatients was about 73 and for outpatients, about 67. There were no differences in body mass index or sex between groups. Outpatients were more likely to be smokers and consumed more alcohol than inpatients.

Inpatients also had a significantly higher rate of comorbidities (39% vs 27%, respectively), and more inpatients had a Charlson score above 3 than outpatients (38.9% vs 26.6%; P<0.0001 for both).

Outpatients had an average hospital stay of 9 days and inpatients, 16 days. The requirement for radiological or surgical intervention did not differ among groups, the authors noted.

More inpatients were taking aspirin, steroids, and heparin, while more outpatients were taking oral anticoagulants and NSAIDs. At bleeding onset, significantly more inpatients were on proton pump inhibitors (PPIs) than outpatients (41.6% vs 27.5%). However, more outpatients were given intravenous PPIs than inpatients (87% vs 79%; P<0.0001 for both).

“Despite the more prevalent use of PPI among inpatients, their [upper gastrointestinal bleeding] was mainly related to peptic ulcer disease (PUD) and [esophagitis],” the authors explained. “This may be explained by the higher intake of aspirin and steroids, known to increase PUD-related hemorrhage risks especially in the elderly and hospitalized patients.”

Adjusted analyses found risk factors associated with 6-week mortality for all patients were rebleeding, a Charlson score of more than 3, hemodynamic instability, a pre-Rockall score of more than 5 and being an inpatient.

Independent risk factors for mortality among inpatients were prothrombin less than 50% and rebleeding, though bleeding-related mortality was lower among inpatients compared to outpatients (10.8% vs 20.6%, P=0.02).

“We found that mortality in outpatients was more likely to be directly related to [upper gastrointestinal bleeding] as opposed to inpatients where death resulted more commonly from other causes,” the authors stated.

When looking at patient groups separately, cirrhosis and antiplatelets were independent outcome predictors among outpatients, in addition to rebleeding, comorbidities, hemodynamic instability and severity of bleeding.

Limitations to this study included the difficulty in comparability of results to previous studies, since this study used a 6-week timeline for outcomes and previous studies commonly used 28-day timelines. Further, the initial cause of the hospitalization of inpatients was not documented, which could significantly affect the prognosis reported.

  • Zaina Hamza is a staff writer for MedPage Today, covering Gastroenterology and Infectious disease. She is based in Chicago.

Disclosures

This study received funding from the French Society of Gastroenterology.

The authors disclosed no conflicts of interest.

Facebook
Twitter
Reddit
LinkedIn
Email

More
articles

Join DBN Today!

Let DBN help guide you to success!

Doctors Business Network offers everything new and existing health care providers need to establish and build a successful career! Sign up with DBN today and let us help you succeed!

DBN Health News