Comparison of the predictive value of different non-insulin-based insulin resistance indices for acute kidney injury in patients with sepsis: a retrospective study.

OBJECTIVE

Insulin resistance (IR) is closely related to the development of acute kidney injury (AKI), but the preferred surrogate markers of IR have not been validated in patients with sepsis. This study aimed to evaluate the predictive value of triglyceride glucose index (TyG), metabolic score for IR (METS-IR) and triglyceride/high-density lipoprotein cholesterol (TG/HDL-C) for the development of AKI in patients with sepsis.

METHODS

Patients diagnosed with sepsis were retrospectively collected from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. The cohort was divided into based on the tertiles of the surrogate indices of IR and Kaplan-Meier curve was depicted the outcomes of each group. Correlations of these three surrogate markers of IR with sepsis-associated AKI and stage III AKI were evaluated through Cox regression models and restricted cubic spline (RCS).

RESULTS

Of the 997 septic patients enrolled in this study, 748 patients (75.03%) developed AKI and 286 (28.69%) eventually progressed to stage III AKI. Cox regression models and RCS showed that only METS-IR was significantly correlated with the development of AKI and stage III AKI in patients with sepsis in a non-linear positive manner. Subgroup analysis showed that the correlation between METS-IR and AKI was not remarkable in patients with heart failure (P = 0.925), chronic kidney disease (P = 0.284), and diabetes (P = 0.139).

CONCLUSIONS

The current study compared the three accepted surrogate indices of IR and the results suggested that METS-IR may prove to be an ideal risk stratification tool for sepsis-associated AKI, showing a non-linear positive correlation.