Objective

We sought to explore the linear or nonlinear relationship between preoperative iron metabolism and acute kidney injury following cardiac surgery (CSA-AKI) with cardiopulmonary bypass (CPB).

Methods

Patients who underwent cardiac surgery with CPB between December 2018 and April 2021 were retrospectively collected from Fuwai Hospital and Medical Information Mart for Intensive Cared dataset (MIMIC-IV). The measurements of iron metabolism included serum iron (SI), serum ferritin (SF), transferrin (TRF), transferrin saturation (TS), and total iron binding capacity (TIBC). Logistic regression and restricted cubic spline (RCS) were used for linear and nonlinear analysis. The primary outcome was postoperative AKI with 48 h after cardiac surgery.

Results

Of 10,639 patients screened (2420 in Fuwai Hospital and 8219 in MIMIC-IV dataset),1488 eligible patients were enrolled for the final analysis (Fuwai Hospital: n = 744, MIMIC-IV: n = 744).The incidence of AKI was 25.7% and 56.5%, respectively. Logistic regression showed that the levels of TRF (odds ratio (OR) = 1.53,95%CI:1.01–2.14, p = 0.012) and TIBC (OR = 1.05,95%CI:1.02–1.07, p < 0.001) were independent risk factor for AKI. Moreover, in the spline models adjusted with age (median:56), female, and history of diabetes, a significant statistical difference was observed between SI, SF, TRF, TS, TIBC (p for nonlinear < 0.05) and AKI in the Fuwai Hospital dataset. Additionally, the levels of SI (p for nonlinear 0.0364),SF (p for nonlinear 0.0461) were also in non-linear relationship with AKI in the MIMIC-IV dataset.

Conclusion

Iron metabolism markers (SI, SF, TS, TRF, and TIBC) displayed a nonlinear relationship with AKI by the RCS model (adjusted by age, gender, and history of diabetes). Notably, the MIMIC-IV dataset, which includes elderly patients, also demonstrated a nonlinear relationship between SI, SF and AKI. These findings highlight the potential therapeutic value of targeting proteins related to iron metabolism in patients with AKI.