Sepsis‐associated acute kidney injury in patients with chronic kidney disease: Patient characteristics, prevalence, timing, trajectory, treatment and associated outcomes

Abstract Aim The features and outcomes of sepsis‐associated acute kidney injury (SA‐AKI) may be affected by chronic kidney disease (CKD). Accordingly, we aimed to compare SA‐AKI in patients with or without CKD. Methods Retrospective cohort study in 12 intensive care units (ICU). We studied the prevalence, patient characteristics, timing, trajectory, treatment and outcomes of SA‐AKI with and without CKD. Results Of 84 240 admissions, 7255 (8.6%) involved patients with CKD. SA‐AKI was more common in patients with CKD (21% vs 14%; p < .001). CKD patients were older (70 vs. 60 years; p < .001), had a higher median Charlson co‐morbidity index (5 vs. 3; p < .001) and acute physiology and chronic health evaluation (APACHE) III score (78 vs. 60; p < .001) and were more likely to receive renal replacement therapy (RRT) (25% vs. 17%; p < .001). They had less complete return to baseline function at ICU discharge (48% vs. 60%; p < .001), higher major adverse kidney events at day 30 (MAKE‐30) (38% vs. 27%; p < .001), and higher hospital and 90‐day mortality (21% vs. 13%; p < .001, and 27% vs. 16%; p < .001, respectively). After adjustment for patient characteristics and severity of illness, however, CKD was not an independent risk factor for increased 90‐day mortality (OR 0.88; 95% CI 0.76–1.02; p = .08) or MAKE‐30 (OR 0.98; 95% CI 0.80–1.09; p = .4). Conclusion SA‐AKI is more common in patients with CKD. Such patients are older, more co‐morbid, have higher disease severity, receive different ICU therapies and have different trajectories of renal recovery and greater unadjusted mortality. However, after adjustment day‐90 mortality and MAKE‐30 risk were not increased by CKD.