HELicobacter Pylori Screening to Prevent Gastrointestinal Bleeding in Patients With Acute Myocardial Infarction

Overview

Background: Potent antithrombotic therapy has improved prognosis for patients with acute myocardial infarction (MI) significantly, however, at a price of increased bleeding risk. Helicobacter pylori (H. pylori) infection commonly causes upper gastrointestinal bleeding (UGIB). If systematic screening for H. pylori significantly reduces the risk of UGIB and improves outcomes is unknown. Study design: A cluster randomized, cross-over, registry-based clinical trial using nationwide Swedish registries for study population definition and data collection. Population: Patients discharged alive after hospitalization for acute type 1 MI at up to 40 hospitals across Sweden. Regional PCI networks comprise 18 clusters. Clusters will be randomized to routine H. pylori screening or no screening for 1 year after which cross-over to the opposite strategy occurs for 1 year. The study ends after one additional year of registry-based follow-up, one year after the end of the second period. Intervention: All MI patients will routinely be offered screening for H. pylori by urea breath test. Controls: Standard clinical practice. Data will be collected from SWEDEHEART and national registries. For patients testing H. pylori positive, eradication therapy will be prescribed at the caring physician's discretion. The individual implementation of H. pylori screening, test result and eradication therapy prescription will be recorded in SWEDEHEART. All follow-up data is collected from SWEDEHEART and national registries. Outcome: Primary outcome is UGIB defined as hospitalization or an outpatient visit in specialized care with ICD codes corresponding to UGIB. The secondary endpoints (in hierarchical) order: 1. Net Adverse Clinical Events (NACE): All-cause death, UGIB, hospitalization for MI, or hospitalization for ischemic stroke. 2. Major Adverse Cardiovascular and Cerebrovascular Events (MACCE): CV death, hospitalization for MI, or hospitalization for ischemic stroke. 3. All-cause death. 4. CV death. 5. Hospitalization for MI. 6. Hospitalization for stroke. 7. Hospitalization for HF. 8. UGIB requiring blood transfusion.

Background: Despite progressively reduced mortality over the last decades, cardiovascular disease remains the most common cause of death in both men and women in Sweden and the world. Ischemic heart disease with its acute presentation myocardial infarction (MI) accounts for the majority of cases, approximately 25000 hospitalized patients in Sweden annually. In addition to early revascularization therapy, potent antithrombotic therapy is the basis for the reduction in cardiovascular events, however, at a price of increased risk of bleeding, typically upper gastrointestinal bleeding (UGIB) that result in substantial morbidity, mortality, and medical care cost. Consequently, antithrombotic therapy may be interrupted in these cases leading to an excessive risk of cardiovascular events; in particular in patients with high age or comorbidities who - by fear of bleeding - rarely receive full recommended treatment from the start. It is recognized that major bleeding events affect prognosis comparably to spontaneous ischemic complications. To optimize the sensitive trade-off between ischemia and bleeding, risk factor management is crucial. On top of established risk factors - high age, male sex, smoking, dyslipidemia, hypertension, hyperglycemia, physical inactivity - active infection with Helicobacter pylori (H. pylori) may be important for two reasons: First, as it commonly causes acute and chronic gastroduodenal lesions, concomitant anticoagulation or antithrombotic therapy aggravates the risk for bleeding, 2-fold with low dose aspirin, and 7-fold with dual antiplatelet therapy, which today is standard treatment for 12 months post MI. Non-invasive screening for H. pylori can be performed easily by urea breath test (UBT). If found positive, eradication by triple therapy is well established, recommended in risk individuals and believed to reverse the bleeding risk almost completely. Second, H. pylori has been proposed as a causal factor between atherosclerosis progression and plaque instability associated with a two-fold increased risk. H. pylori may hence be an overlooked risk factor for bleeding complications in MI patients, which potentially could be controlled by routine H. pylori screening. This would be anticipated not only to reduce the UGIB complications after MI but also to improve the adherence to dual antiplatelet therapy and consequently potentially also improve the cardiovascular prognosis in this group. In summary, the utility of routine H. pylori screening in a contemporary MI population remains undetermined as it is unknown if systematic H. pylori screening reduces the risk of bleeding and improves prognosis. OBJECTIVE - paradigm and main hypothesis The aim is to determine whether systematic screening for H. pylori after acute MI significantly reduces the risk of UGIB, cardiovascular events and mortality. STUDY POPULATIONS: 1. PRIMARY INTENTION-TO-TREAT (ITT) POPULATION All adult patients (\>18 years of age) registered in SWEDEHEART as discharged alive at a study site with a diagnosis of type 1 MI (including ICD-10 code I21 or I22) analyzed according to randomization. 2. SECONDARY PER-POPULATION (PP) analyses All adult patients (\>18 years of age) registered in SWEDEHEART as discharged alive at a study site with a diagnosis of type 1 MI (including ICD-10 code I21 or I22) who were screened for H. pylori (including subcategories screened y/n; screened positive y/n; eradicated y/n if screened positive) are compared to a latent population in the control group from study sites randomized to no-screening. OUTCOMES All clusters will be analyzed as randomized, irrespective of whether individual patients underwent screening, in accordance with ITT principle. All endpoints in HELP-MI SWEDEHEART are listed below. The primary endpoint is UGIB defined by appearance of ICD codes obtained from the mandatory National Patient Registry. No central adjudication of events will be performed. Secondary endpoints are all-cause and cardiovascular mortality, rehospitalization with MI, hospitalization with stroke or heart failure. MI during the initial hospital stay and readmission due to a nonfatal MI during the first month, will be collected from SWEDEHEART. After day 31, rehospitalization for MI will be captured in the National Patient Registry. All-cause death will be obtained from the Swedish population registry. Cardiovascular death will be obtained from the Cause-of-death registry. Hospitalization for heart failure and stroke will be obtained from the National Patient Registry. Tertiary endpoints in patients \<80 years of age participating in the routine SWEDEHEART post MI follow-up include symptoms (CCS, NYHA), health related quality of life (EQ5-D index, EQ-VAS, individuals components of EQ-5D), and change in blood glucose (plasma glucose, HbA1c) and blood lipids (total cholesterol, HDL, LDL, triglycerides) from baseline levels at index hospitalization to 6-10 weeks and 11-13 months follow-up. Health economics based on cost-effectiveness analysis will be performed after one year and including long-term modelling. The linkage with the National Patient Registry, the National Cause of Death Registry, National Prescribed Drug Registry (prescribed and dispensed treatments), and Statistics Sweden (socioeconomic data) will be done at the end of follow-up. Pre-specified subgroups are: * Age * Sex * Smoking status * Hypertension * Diabetes * Previous atherosclerotic cardiovascular disease (MI, stroke, peripheral arterial disease) * Anemia (previous diagnosis and at arrival during index hospitalization) * Previous gastroduodenal disease (UGIB, peptic ulcer disease, atrophic gastritis, mucosa-associated lymphoid tissue lymphoma, gastric cancer) * Previous cancer (excluding non-melanoma skin cancer) * Previous alcohol use disorder * Chronic or current kidney disease * MI patients by subcategory (STEMI; NSTEMI) * MI patients according to revascularization status * MI patients according to revascularization technique * Degree of coronary artery disease * Inflammation (CRP) * Concomitant medication (antithrombotic, anticoagulation, NSAID, serotonin reuptake inhibitors, cortisone, proton pump inhibitors) * Planned time on DAPT * Bleeding risk (using PRECISE DAPT) * Trial cluster * Socioeconomic and sociodemographic parameters (region of birth, education level, income) STATISTICS: All enrolled patients discharged alive will be included in the primary ITT analyses. Secondary per-protocol analyses will be performed. The primary analysis is a cluster-summary analysis. In a secondary sensitivity analysis, the outcome will be analyzed using Cox proportional hazards models. The primary analysis (cluster-summary using all follow-up time) of the primary and secondary endpoints will be adjusted for multiplicity using a hierarchical strategy, ordering the endpoints as indicated below. Prespecified subgroup analysis to determine heterogeneity of treatment effect will be performed but not all included in the primary report. The primary analysis will use all available follow-up time, censoring patients 2025-01-17. This corresponds to a maximum follow-up time of 3 years + 2 months. In a sensitivity analysis, follow-up will begin on the admission date rather than the discharge date. Patients who die during hospitalization will not be excluded. In a set of supplementary analyses, patients will be censored as follows: * After 30 days. * After 1 year. * Censor patients 1 year after the end of the inclusion time for their respective period (Period 1 or 2). * A landmark analysis, starting the follow-up 30 days after discharge, excluding patients with an event during these 30 days. All described endpoints will also be analyzed in all subgroups and study populations and reported as above at a later stage. Moreover, long-term (5 years, 10 years) follow-up is planned subsequently. A detailed statistical analysis plan with prespecified analyses will be published alongside the primary report.