Bone Abstracts

We retrospectively analyzed the incidence of hip re-fractures in Austrian hip fracture patients devoid of anti-osteoporotic drug treatment, taking vs not taking proton pump inhibitors (PPIs).

For 31,668 patients ≥50 years sustaining a hip fracture in Austria between July 2008 and December 2010, information on hip re-fractures with follow-up until June 2011 and on filed prescriptions of PPIs and anti-osteoporotic drugs between July 2007 and June 2011 was available. A total of 17,228 patients on PPIs not receiving anti-osteoporotic drugs like bisphosphonates were identified and categorized into five sub-groups: (1) PPIs began before or (2) after first fracture, (3) PPIs discontinued before first fracture or (4) began no sooner than 30 days after discharge from first fracture, and (5) PPIs taken only short-term in hospital and/or within 30 days after discharge from first fracture. Each subgroup was sex- and age-matched with 4568 control hip fracture patients receiving neither PPIs nor anti-osteoporotic drugs. Re-fractures were related to followed-up survival years and compared statistically.

Elevated numbers of hip re-fractures/1000 patient years (py) were associated with PPI medication vs control when PPIs were begun before (28.6 vs 16.7, P<0.0001), after (26.0 vs 16.8, P<0.0001), and not until 30 days after first fracture (33.6 vs 16.7, P<0.0001). By contrast, PPI use before first fracture only and under exclusively peri-operative short-term PPI use entailed fewer hip re-fractures per 1000 py among female patients relative to controls (7.7 vs 20.2, P<0.05, and 8.4 vs 22.5, P<0.05, respectively).

It is elusive whether elevated numbers of hip re-fractures among PPI users are causally linked to PPI effects or reflect greater comorbidity. Reduced re-fractures among patients on PPIs exclusively before hip fracture are likely due to few survival years, i.e. high mortality. This is not the case for short-term peri-operative PPI use, therefore not affecting or even lowering hip re-fracture incidence.