Objective To evaluate the use, effectiveness and safety of Helicobacter pylori empirical rescue therapy in third and subsequent treatment lines in Europe.

Design International, prospective, non-interventional registry of the clinical practice of European gastroenterologists. Data were collected and quality reviewed until October 2021 at Asociación Española de Gastroenterología-Research Electronic Data Capture. All cases with three or more empirical eradication attempts were assessed for effectiveness by modified intention-to-treat and per-protocol analysis.

Results Overall, 2144 treatments were included: 1519, 439, 145 and 41 cases from third, fourth, fifth and sixth treatment lines, respectively. Sixty different therapies were used; the 15 most frequently prescribed encompassed >90% of cases. Overall effectiveness remained <90% in all therapies. Optimised treatments achieved a higher eradication rate than non-optimised (78% vs 67%, p<0.0001). From 2017 to 2021, only 44% of treatments other than 10-day single-capsule therapy used high proton-pump inhibitor doses and lasted ≥14 days. Quadruple therapy containing metronidazole, tetracycline and bismuth achieved optimal eradication rates only when prescribed as third-line treatment, either as 10-day single-capsule therapy (87%) or as 14-day traditional therapy with tetracycline hydrochloride (95%). Triple amoxicillin-levofloxacin therapy achieved 90% effectiveness in Eastern Europe only or when optimised. The overall incidence of adverse events was 31%.

Conclusion Empirical rescue treatment in third and subsequent lines achieved suboptimal effectiveness in most European regions. Only quadruple bismuth-metronidazole-tetracycline (10-day single-capsule or 14-day traditional scheme) and triple amoxicillin-levofloxacin therapies reached acceptable outcomes in some settings. Compliance with empirical therapy optimisation principles is still poor 5 years after clinical practice guidelines update.

Trial registration number NCT02328131.