In patients 60 years or above who have been prescribed aspirin in primary care, H. pylori eradication shields against aspirin-related peptic ulcer bleeding. But, this may not sustain in the long term.

A double-blind, randomized, placebo-controlled experiment (Helicobacter Eradication Aspirin Trial [HEAT]) was conducted to see if eliminating Helicobacter pylori (H. pylori) would offer protection against aspirin-related ulcer bleeding.

Individuals who had a positive C13 urea breath test for H. pylori at screening and were taking 325 mg or a less daily dose of aspirin (with 4 or more 28-day prescriptions in the previous year) were eligible. People getting gastroprotective or ulcerogenic medication were not included.  Routinely collected clinical data was used.

Oral 500 mg Clarithromycin, 400 mg Metronidazole, and 30 mg Lansoprazole (active eradication) or oral placebo (control) were given to patients at random (1:1) two times per day for a week. In primary and secondary care, follow-up included an examination of electronic data. In the intention-to-treat population, Cox proportional hazards techniques were used to evaluate the key endpoint, which was the amount of time before death or hospitalization owing to probable or definite peptic ulcer bleeding.

Overall, 30166 individuals underwent breath testing for H. pylori, 5367 had positive results, and 5352 were randomly allocated to either active eradication (n = 2677) or a placebo (n = 2675), with follow-up lasting a median of 5 years (Interquartile range [IQR] 3.9-6.4). Analysis across several time periods was necessary since the major endpoint indicated a considerable departure from the proportional hazards assumptions.

In the initial 2.5 years of follow-up, the active elimination group experienced a significantly lower occurrence of the primary outcome than the control group (6 episodes classified as definite or probable peptic ulcer bleeds, rate 0.92/1000 person-years versus 17 episodes, rate 2.61/1000 person-years; hazard ratio [HR] 0.35).

Following the adjustment of the competing risk of mortality, this benefit remained considerable. However, it was lost with longer follow-up (HR 1.31 in the time following the first 2.5 years). Noxious event reports were actively solicited. The most frequent event was taste disturbance (787 subjects).

Elimination of H. pylori was truly attained in a massive population of unselected older people getting 325 mg or less aspirin in primary care. H. pylori management was related to a considerable decline in the risk of hospitalization for ulcer bleeding. However, this advantage was lost over time.