Long-term PPI tied to higher pneumonia risk in older adults
Long-term proton-pump inhibitor (PPI) therapy was associated with an increased pneumonia risk among older adults, putting to rest the controversies about the validity of previously reported short-term harms of PPIs, according to a large, longitudinal analysis of electronic medical records.
A higher risk of pneumonia has been reported with short-term PPI use. However, one critique on the studies was that pneumonia risk was already higher in cases than control before initiation of PPI, which indicates confounding by pre-existing differences. [BMJ 2016;355:i5813; Gut 2014;63:552-558]
“Our own results suggest that although [associations between] short-term PPI and pneumonia may be confounded, longer-term associations in older adults in primary care were present and consistent across a range of analysis approaches ... independent of excess pneumonia rates immediately before first PPI receipt,” said principal investigator Professor David Melzer of the University of Exeter Medical School, Exeter, UK.
To adjust for pre-existing differences in pneumonia incidence between cases and controls before PPI initiation, the researchers used prior event rate ratio (PERR) to correct for unmeasured confounding. A total of 75,050 adults (aged ≥60 years) who received PPIs for 1 year or longer were matched by age and sex to similar number of controls who did not receive PPI. [J Am Geriatr Soc 2018;doi:10.1111/jgs.15385]
During the second year after PPI initiation, the therapy was associated with a significantly greater risk of incident pneumonia, even after adjusting for pretreatment pneumonia rates (PERR-adjusted hazard ratio [HR], 1.82, 95 percent confidence interval [CI], 1.27–2.54). According to the researchers, the HR of 1.82 suggests “a substantial excess long-term pneumonia risk”.
The risk remained consistently higher with PPI in prespecified subgroup analyses defined by age (HRs, 2.1 and 1.5 for age 60–74 and ≥75 years, respectively) and comorbidities (HRs, 2.8 and 1.6 for Charlson Comorbidity Index=0 and ≥1, respectively).
Similar results were seen in inverse probability-weighted models which were adjusted for 78 potential confounders including demographic features, medication, and healthcare usage.
“Our study adds to growing evidence that PPIs are not quite as safe as previously thought, although they are still a very useful class of medication for certain groups of patients,” said Melzer.
However, patients should not stop their PPIs without consulting healthcare professionals, the researchers advised.
“Caution is needed in interpreting the findings as our study is based on analysing data from medical records, so other factors may be involved,” said Melzer. Nonetheless, as incident pneumonia is relatively uncommon, a randomized controlled trial would require large number of participants to be powered, which poses challenge in recruitment of older adults.
“Thus, observational estimates in large representative populations based on robust statistical methods are likely to provide valuable estimates of effect (and are often the only available estimates) for informing safe prescribing practice, especially for older adults,” wrote the researchers. “More work is needed to identify individuals at highest risk.”