Initial Diagnostic Strategies for Helicobacter Pylori in Patients With Bleeding Peptic Ulcers Undergoing Endoscopy: A Cost-Effectiveness Analysis

Abstract

Background and aims: Helicobacter pylori (H. pylori) is a major cause of peptic ulcer disease (PUD) and upper gastrointestinal bleeding. Testing for and eradication of H. pylori reduces the risk of future PUD-related complications including readmission for gastrointestinal bleeding. Our aim was to determine the most cost-effective testing strategy for H. pylori in patients hospitalized with bleeding peptic ulcers.

Methods: We developed a Markov cohort model to compare the following 6 H. pylori testing strategies: no testing, histology, rapid urease test, stool antigen test, urea breath test (UBT), and serology. Histology and rapid urease test require biopsies, while stool antigen test, UBT, and serology do not. We assumed a 17% H. pylori prevalence in patients admitted with bleeding ulcers. Model outcomes included hospitalizations for rebleeds, number needed to treat to avoid another hospitalization, life expectancy, total cost, quality-adjusted life years, and incremental cost-effectiveness ratios.

Results: Compared to no testing, UBT resulted in a gain of 0.02 quality-adjusted life years, total cost savings of $2140 per patient, and 1675 hospitalizations avoided per 10,000 patients per year. Additionally, the number needed to treat to avoid an additional hospitalization over 35 years was 167. UBT was the preferred strategy as it was both less costly and more effective than no testing.

Conclusion: Our findings suggest that UBT is the cost-effective strategy to identify H. pylori in patients admitted with PUD. Noninvasive H. pylori testing at the point of care or during inpatient admission should be considered, as it presents limited risk to patients and offers potential clinical benefits.

Keywords: Cost-Effectiveness Analysis; GI Bleeding; H. pylori; Hospitalization; Peptic Ulcer Disease.

Figure

Markov Model schematic. All patients enter the Markov Model at age 65 hospitalized with an assumed first-time peptic ulcer bleed undergoing endoscopy and are tracked annually until age 100 or death. Patients receive 1 of 6 strategies, which may result in a true/false positive/negative: no testing, histology, RUT, SAT, UBT, or serology. Over the course of the model, patients can potentially become reinfected with H. pylori or experience another rebleed.

Figure A1

One-way Sensitivity Analysis of UBT vs SAT at 17% H. pylori Prevalence in Bleeding Peptic Ulcers.

Figure A2

Probabilistic Sensitivity Analyses of Strategies at 17% H. pylori Positive Prevalence in Bleeding Peptic Ulcers.