SYSTEMATIC REVIEW OF OBSERVATIONAL STUDIES FINDS INCREASED RISK OF FRACTURE AMONG OLDER ADULTS TAKING A PROTON PUMP INHIBITOR

Context

Proton pump inhibitors (PPIs) were initially considered to be potentially beneficial for bone as they were perceived to be able to inhibit acidic degradation by osteoclasts and thus have antiresorptive potential. However, this has been shown not to be the case, as the original studies were based on compounds differing from those used clinically.1 Studies of bone density have failed to show any major changes in bone density with PPIs. Short-term studies have failed to show an effect on calcium. The present study examines whether a relationship exists between the use of PPIs and histamine-2-receptor antagonists (H2RAs) and adverse effects on the skeleton.

Methods

Yu and colleagues applied standard meta-analytic techniques for their systematic review of published data from observational studies identified in PubMed, EMBASE, Web of Science and BIOSIS from 1970 to 2010, combined with contact with authors for supplementary data.

Findings

The study reported an excess risk of hip fractures (RR 1.30, 95% CI 1.19 to 1.43) with PPIs but not with H2RAs. An increased risk of any fracture and spine fractures was observed with PPIs. The estimates were statistical heterogeneous. An increased risk of hip fractures was seen both with more and less than 1 year of PPI exposure

Commentary/Recommendation

The use of PPIs in patients at risk of osteoporosis should be considered carefully. At present, the mechanism behind the association with fracture risk is not known, and, from the heterogeneity observed, a risk of confounding as an explanation for the observed associations still exists. Studies on pathophysiology such as changes in bone architecture through peripheral quantitative CT (pQCT) or bone biopsies and studies on bone turnover are needed.

References

Peter Vestergaard, Commentary on: Yu EW, Bauer SR, Bain PA, et al. Proton pump inhibitors and risk of fractures:a meta-analysis of 11 international studies. Am J Med 2011;124:519–26. In Evidence-Based Medicine April 2012 | volume 17 | number 2 |