Proton pump inhibitors increase risk of bone fractures in men with cirrhosis: a population‐based study

• Published in: Alimentary pharmacology & therapeutics Vol. 52; no. 6; pp. 1042 - 1050
• Main Authors: Labenz, Christian, Wörns, Marcus‐Alexander, Adarkwah, Charles C., Galle, Peter R., Schattenberg, Jörn M., Kostev, Karel
• Format: Journal Article
• Language: English
• Published: England Wiley Subscription Services, Inc 01-09-2020
• Subjects: Adolescent; Adult; Aged; Aged, 80 and over; Case-Control Studies; Cirrhosis; Female; Fractures; Fractures, Bone - epidemiology; Fractures, Bone - etiology; Humans; Liver cirrhosis; Liver Cirrhosis - complications; Male; Middle Aged; Population studies; Population-based studies; Proton pump inhibitors; Proton Pump Inhibitors - adverse effects; Regression analysis; Risk Assessment; Sensitivity analysis; Young Adult
• ISSN: 0269-2813; 1365-2036
• DOI: 10.1111/apt.16008

Summary
Background
Bone fractures are a frequent complication in patients with cirrhosis. Proton pump inhibitors (PPIs) are among the most frequently prescribed medications and may impair bone quality and quantity.
Aims
To investigate whether PPI use predisposes patients with cirrhosis to bone fractures.
Methods
We performed a population‐based case‐control study exploring a sample of patients with cirrhosis derived from the Disease Analyzer database. In total, 1795 cirrhotic patients with fractures were compared to 10 235 cirrhotic patients without fractures. PPI use overall and the cumulative PPI dose 5 years prior to the index date were analysed. To estimate the association between PPI use and fractures, logistic regression analyses were performed taking cofounding factors into consideration.
Results
PPI use was more frequently seen in cirrhotic patients with fractures compared to controls (67.0% vs 53.4%, P < 0.001). In regression analyses, PPI use was associated with bone fractures after adjusting for important confounders (OR 1.34, 95% CI 1.20‐1.51, P < 0.001). Importantly, the strongest effect of PPIs on bone fractures was seen in men and patients below 70 years of age. On further sensitivity analyses, we observed a dose‐dependent effect for all PPIs with the strongest effect in cirrhotic patients receiving a dose of >50 000 mg during the 5 years prior to index date (OR 1.63, 95% CI 1.32‐2.03).
Conclusions
PPI use was associated with bone fractures in a dose‐dependent fashion in patients with cirrhosis. PPI use in these patients should be based on a careful risk‐benefit assessment.