Comorbidities, acute kidney injury and long-term mortality in elderly patients hospitalized because of hip fracture: a moderation analysis

Abstract

Introduction: Femoral fractures in elderly individuals present significant health challenges, often leading to increased morbidity and mortality. Acute kidney injury (AKI) during hospitalization further complicates outcomes, yet the interaction between AKI severity and comorbidities, as quantified by the Charlson Comorbidity Index (CCI), remains poorly understood in this population. This study aimed to assess the associations between AKI severity and the CCI and between AKI severity and one-year mortality postfemoral fracture in elderly patients.

Methodology: This study utilized data from the Multiparameter Intelligent Monitoring in Intensive Care (MIMIC-IV) database and focused on elderly patients (> 65 years) admitted with hip fractures. Patients were categorized based on AKI stage according to the KDIGO criteria and CCI scores. The primary outcome assessed was all-cause mortality one year after hospital discharge. The statistical analyses included logistic regression, Cox proportional hazards regression and moderation analysis with the Johnson-Neyman technique to evaluate associations between AKI and long-term mortality and between the CCI and long-term mortality.

Results: The analysis included 1,955 patients and revealed that severe AKI (stages 2 and 3) was independently associated with increased one-year mortality. Notably, the CCI moderated these associations significantly. A lower CCI score was significantly correlated with greater mortality in patients with severe AKI. The impact of severe AKI was greater for those with a CCI as low as 3, more than doubling the observed one-year mortality rate. In contrast, higher CCI scores (≥8) did not significantly impact mortality. Sensitivity analyses supported these findings, underscoring the robustness of the observed associations.

Conclusion: This study elucidates the complex interplay between AKI severity and comorbidities and long-term mortality in elderly hip fracture patients. These findings underscore the importance of considering both AKI severity and comorbidity burden in prognostic assessments and intervention strategies for this vulnerable population. Targeted interventions tailored to individual risk profiles may help mitigate the impact of AKI on mortality outcomes, ultimately improving patient care and outcomes. Further research is warranted to explore the underlying mechanisms involved and refine risk stratification approaches in this population.

Keywords: Acute kidney injury; Hip fracture; Mortality.

Fig. 1

Flowchart of the included patients from the MIMIC-IV database

Fig. 2

Cox survival curves for mortality according to AKI stage. Significant difference was observed between patients with no-AKI/stage 1 AKI and patients with AKI stage 2 or 3

Fig. 3

Moderation effects of the Charlson Comorbidity Index on the conditional association between severe AKI and one-year mortality. Note that the lower 95% confidence interval crosses the zero value (no association) at values near 8, and up to this value, there is no significant association

Fig. 4

Cox survival curves for one-year mortality in patients according to the Charlson comorbidity index (CCI). A significant difference was observed in the subgroup with a CCI < 8 according to AKI stage (Fig. 4a). In the subgroup with a CCI ≥ 8, there was no significant difference between patients according to AKI stage. Note that there is no AKI stage order in Fig. 4b