Association between baseline-stratified trajectory subgroups of serum albumin and outcomes in critically ill patients with early sepsis-associated acute kidney injury: a retrospective study

Purpose We aimed to explore the association between the serum albumin trajectory following early sepsis-associated acute kidney injury (SA-AKI) and the outcomes of sepsis patients in the intensive care unit (ICU). Methods We extracted data of patients with early SA-AKI from the Medical Information Mart for Intensive Care (MIMIC)-IV database. The dynamic levels of albumin within 7 days after AKI onset were used as the exposure factor in this study. The primary outcome was 365-day mortality. Group-based trajectory modeling (GBTM) was used to identify the dynamic characteristics of albumin levels. Kaplan–Meier (KM) survival curves, linear regression and Cox analysis were used to explore the potential associations between different trajectory groups and outcomes. Results A total of 1,914 patients with early SA-AKI were included in GBTM. We identified 5 baseline-stratified trajectory subgroups of albumin levels, including four zero-slope trajectories (at levels of 2.0 g/dL, 2.5 g/dL, 3.0 g/dL and 3.5 g/dL) and a slightly ascending trajectory (at a level of 4.1 g/dL). Compared with the steady-state trajectory at a level of 3.0 g/dL, patients with the steady-state trajectory at a level of 2.0 g/dL had higher 30-, 90-, 180-, and 365-day mortality risk and longer hospital length of stay (LOS). And patients with the trajectory at a level of 4.1 g/dL had higher 365-day mortality risk. Moreover, subgroup analysis indicated that the association between serum albumin trajectories and mortality risk interacted with age and sex. Further sensitivity analysis confirmed the robustness of these results through similar trends. Conclusion The baseline-stratified trajectory subgroups of serum albumin following early SA-AKI were associated with both short- and long-term mortality and hospital LOS in critically ill patients with sepsis, which can be used as a valuable indicator for guiding risk stratification and predicting prognosis in such patients.