Group-Based Trajectory Modeling of Acute Kidney Injury in the Intensive Care Unit: A Scoping Review

Acute kidney injury (AKI) is a common clinical syndrome with high heterogeneity in the intensive care unit (ICU). Traditional AKI diagnostic criteria, such as the KDIGO criteria, rely on static thresholds of renal function, which fail to capture the dynamic nature of kidney injury and recovery. Over the past ten years, Group-based trajectory model (GBTM) have has been increasingly adopted to overcome these limitations. GBTM is a finite mixture model for longitudinal data, which can use to determine subgroups and discuss differences between subgroups. By assuming the overall population consists of distinct subgroups with different underlying developmental trajectories, GBTM can accurately handle non-monotonic and irregularly spaced longitudinal data to classify clinical subphenotypes. This scoping review aims to map the current evidence landscape of AKI trajectory modeling in the ICU, evaluate the quality of these studies, and identify critical knowledge gaps to direct future medicine research. Furthermore, a crucial objective of this review is to evaluate the methodological quality and reporting transparency of existing studies. Due to the variations reporting quality in articles using GBTM, we will evaluate each study using guidelines for reporting on latent trajectory studies (The GRoLTS-Checklist).