Predicting Rebleeding in Patients with Idiopathic Peptic Ulcers

Using machine-based learning, a predictive score was derived and validated in patients with Helicobacter pylori–negative ulcers.

The major risk factors for peptic ulcers that bleed are H. pylori and nonsteroidal anti-inflammatory drugs (NSAIDs). However, idiopathic peptic ulcers, or IPUs, an increasingly prevalent category, have no identified risk factors and are associated with increased risk for further bleeding.

In an analysis of a large, prospectively collected database from Hong Kong, machine learning was used to build a model (called the IPU-ML) to predict IPU rebleeding. The training model, derived from 22,900 patients with IPU, was built on six clinical parameters: age; hemoglobin; and the presence of gastric ulcer, gastrointestinal diseases, malignancies, and infections. It was prospectively validated in 1300 patients with a bleeding IPU, among whom bleeding …

Reviewing Author

John R. Saltzman, MD

Disclosures

Consultant/Advisory BoardIterative Scopes

Editorial BoardsGastrointestinal Endoscopy; UpToDate

Leadership Positions in Professional SocietiesAmerican College of Gastroenterology (Trustee)

OtherVice President of Medical Affairs, 1Globe Health Institute