Association of Age, Frailty, and Strategy for Initiation of Renal-Replacement Therapy: A Post Hoc analysis of the STARRT-Acute Kidney Injury Trial

Introduction: This study was designed to assess the association of age and frailty with clinical outcomes in patients with severe acute kidney injury (AKI), according to accelerated and standard renal-replacement therapy (RRT) initiation strategies in the STARRT-AKI trial.

Methods: This was a secondary analysis of an international randomized trial. Older age was defined as ≥65 years. Frailty was assessed using the clinical frailty scale (CFS) score and defined as a score ≥5. The primary outcome was all-cause mortality at 90 days. Secondary outcomes included RRT dependence and RRT-free days at 90 days. We used logistic and linear regression and interaction testing to explore the impact of age and frailty on clinical outcomes.

Results: Of 2,927 patients randomized in the STARRT-AKI trial, 1,616 (55.2%) were aged ≥65 years (median [interquartile range] 73.9 [69.4-78.9]). Older patients had greater comorbid cardiovascular and chronic kidney disease, were more likely to be surgical admissions and to receive vasopressors at baseline. Older patients had higher 90-day mortality (50.4% vs. 35.6%, adjusted-odds ratio (OR), 1.81 [1.53-2.13], p < 0.001). There was no significant difference in RRT dependence at 90 days between older and younger patients (8.7% vs. 7.8%, adjusted-OR, 1.21 [0.82-1.79], p = 0.325). Patients with frailty had higher mortality; but no difference in RRT dependence at 90 days. There was no significant interaction between age and CFS score in relation to mortality, RRT dependence at 90 days, and other secondary outcomes. There was no significant difference in the proportion of patients who received RRT in the standard-strategy stratified by age groups (adjusted-OR, 0.85 [0.67-1.08], p = 0.180).

Conclusion: In this secondary analysis of the STARRT-AKI trial, older and frail patients had higher mortality at 90 days; however, there was no difference in RRT dependence. Mortality and RRT dependence were not modified by RRT initiation strategy in older or frail patients.