SIRS, In an apparently comprehensive review of gastrointestinal side effects of long-term acid suppression with proton pump inhibitors, Dr Laine et al. make no reference to our case control study published in the BMJ which reported that current omeprazole use was a strong risk factor for Campylobacter enteritis.1 Yet they include reference to another similar but essentially negative study, done in response to ours and published in a less well-known journal, which reported small and non-significant increases in risk.2 Given the relative profiles of the two journals, it would be difficult to accept that this omission reflected a failure of their literature search. Our study of 211 patients with notified Campylobacter infections, and 422 age and sex matched control from general practitioner records, found that current omeprazole users, but not former users, were at increased risk (odds ratio 11.7 (95% CI 2.5–54)).3 The study by Garcia Rodriguez and Ruigomez found a small increase in risk of Campylobacter gastroenteritis (odds ratio 1.7, 95% CI 1.0–3.0). In addition they also reported a small increase in risk for all types of bacterial gastroenteritis in current omeprazole users, which was not seen with current users of less potent anti-secretory drugs (odds ratio 1.6, 95% CI 1.0–2.4).2 While the differences in risk estimates between the two studies seem large, the confidence intervals overlap. Furthermore, because the Garcia Rodrigues and Ruigomez case-control study was nested within a cohort of ever users of anti-secretory drugs, it is possible that their study design selected-out subjects with previous gastroenteritis while on these drugs — as we found infection was most likely within the first month of treatment. We agree with Armstrong that this area merits further research, because use of powerful anti-secretory drugs is now widespread.4 If they are proved to be important risk factors for gastrointestinal infection, then this could represent a significant public health problem.
No takes yet. Share an insight, caveat, or question.
Neal et al. (2001) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: