Abstract
BACKGROUND: Updated 2016 Helicobacter pylori consensus guidelines recommend treatment for 14 days with concomitant therapy (proton-pump inhibitor (PPI)-amoxicillin-metronidazole-clarithromycin (PAMC) or bismuth-based quadruple therapy (PPI-bismuth-metronidazole-tetracycline, PBMT)) as first line, PBMT or PPI-amoxicillin-levofloxacin (PAL) as second or third line, and PPI-amoxicillin-rifabutin (PAR) as fourth line for 10 days. OBJECTIVES: This was a retrospective cohort study to describe and compare the efficacy of anti-Helicobacter treatment regimens over the periods 2007-2015 and 2016-2021 as well as antibiotic resistance. METHODS: A modified intention-to-treat (mITT) analysis was used to analyze the success rate of therapies. mITT includes all patients who were prescribed H. pylori treatment and had at least one follow-up test-of-cure. This included patients who could not complete treatment or were non-adherent with treatment. Risk factors for treatment failures were analyzed by univariate and multivariate logistic regression. Resistance testing was done in a small subset of patients. RESULTS: H. pylori-positive patients who received treatment in Edmonton, Alberta were included in a mITT analysis: 334/387(86%) from 2007 to 2015 and 193/199 (97%) from 2016 to 2021. During 2016-2021, 78% (150/193) of patients underwent cumulative guideline-based treatment with a successful cure in 80% (120/150) of patients. In those who were newly diagnosed, the cure rate was 88% (52/59) versus those with previous treatment failure 75% (68/91) (P < 0.05, risk difference [RD] 14%, 95% confidence interval [CI] 1.7-26.3%). The most effective first-line regimens were PAMC for 14 days (87% [45/52]) in 2016-2021 and sequential therapy in 2007-2015 (83% [66/80]) (P = 0.535, RD 4%, 95% CI -8.5-16.5%). When other treatments failed, success with PAR was 50% (2/4) from 2007 to 2015 and 57% (21/37) from 2016 to 2021. Recent (2016-2021) resistance rates to clarithromycin and metronidazole are high at 78% (50/64) and 56% (29/52), respectively. From 2007 to 2015, clarithromycin and metronidazole resistance rates were 80% (36/45) and 83% (38/46), respectively. Levofloxacin resistance increased significantly from 2007-2015 to 2016-2021 (28% [13/46] to 61% [35/57], P < 0.05, RD 33%, 95% CI 11.6-54.4%). CONCLUSIONS: Algorithmic treatment with PAMC first line followed by PBMT, PAL, and PAR cures H. pylori in 88% of newly diagnosed patients. PAR therapy shows suboptimal cure rates (50-57% success) but can be considered as third instead of fourth line given increasing levofloxacin resistance rates. Antibiotic resistance in H. pylori is common to clarithromycin, metronidazole, and levofloxacin and frequently accounts for treatment failures.