Associations between acute kidney injury and bone fractures: a retrospective cohort study

Résumé fourni par la source

ABSTRACT Background Acute kidney injury (AKI) is common. An AKI episode may disrupt the normal mineral bone balance maintained by normal kidney function, thereby modifying the risk of developing bone fractures. However, it remains unclear whether an AKI episode is associated with the risk of bone fractures. Methods Using retrospective cohort study from an Australian Local Health District, we examined the association between an AKI episode and bone fractures using patient data between 2008 and 2017. Time-varying Cox proportional hazards and propensity-matched analysis were used to examine the association. Sensitivity analyses were undertaken to capture the impact of confirmed AKI status and AKI severity. Results Of 123 426 included patients, 14 549 (12%) had an AKI episode and 12 505 (10%) had a bone fracture. In the unadjusted analysis, AKI was associated with bone fractures [hazard ratio (HR) 1.99, 95% confidence interval (CI) 1.88–2.11]. This association persisted in the adjusted analysis (HR 1.50, 95% CI 1.41–1.59) and propensity-matched dataset (HR 1.71, 95% CI 1.59–1.83). The sensitivity analysis yielded similar results, with the AKI patients having a higher risk of fractures compared with no AKI patients in the adjusted analysis (HR 1.34, 95% CI 1.25–1.43) and in the propensity-matched dataset (HR 1.44, 95% CI 1.33–1.55). Similar results were seen in the subsidiary sensitivity analysis excluding patients without baseline creatinine. We did not find an increased risk of bone fractures with increasing AKI severity (P = .7). Interaction tests demonstrated a significant association between sex and age category with AKI status and fractures, but not CKD stage or osteoporosis. Conclusions AKI is associated with a greater risk of bone fractures. This could have implications for managing and screening for bone disease in patients post-AKI episode. This association should be examined in other cohorts and populations for verification.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.