Development of a multivariable clinical prediction model for acute kidney injury after valve replacement surgery: incremental value of the triglyceride-glucose body mass index

Background: Acute kidney injury (AKI) complicates cardiac valve replacement (CVR) in up to one-third of patients. The triglyceride-glucose-body mass index (TyG-BMI) is a composite marker derived from routine laboratory tests and anthropometric measurements. It reflects the combined burden of insulin resistance and obesity—two interrelated conditions that contribute to chronic low-grade inflammation, oxidative stress, endothelial dysfunction, and increased renal microvascular vulnerability. However, its association with AKI after CVR remains unclear. This study aimed to develop a multivariable clinical prediction model for AKI and evaluate the incremental value of incorporating TyG-BMI into that model. Methods: This single-center retrospective study enrolled consecutive patients aged 18–80 years undergoing first-time elective CVR with cardiopulmonary bypass (CPB) (November 2021–December 2022). Exclusion criteria were pre-existing chronic kidney disease (CKD) [estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2], preoperative AKI, dialysis/kidney transplantation, chronic liver disease, active malignancy, severe infection, infective endocarditis, death within 48 hours, concomitant CABG, or incomplete data. Preoperative blood samples were collected within 24 hours. TyG-BMI was calculated as Ln(TG (mg/dL)×FPG (mg/dL)/2)×BMI (kg/m2). Postoperative AKI was defined per KDIGO criteria (creatinine rise ≥26.5 µmol/L within 48 hours or ≥1.5× baseline within 7 days) and monitored daily for seven days. Univariate and multivariate logistic regression analyses were performed to identify independent risk factors. The incremental predictive value of TyG-BMI was assessed by comparing models with and without it using receiver operating characteristic (ROC) curve analysis. Restricted cubic splines (RCS) were used to explore nonlinear relationships. Results: Among 187 included patients, 64 (34.2%) developed postoperative AKI. Patients who developed AKI had higher preoperative TyG-BMI levels compared to those without [AKI (207.99±27.05) vs. non‑AKI (184.24±23.71); P<0.001]. After adjusting for potential confounders including age, sex, CPB time, aortic cross-clamp time, and preoperative creatinine, TyG-BMI remained independently associated with an increased risk of AKI [odds ratio (OR): 1.050 per unit increase, 95% confidence interval (CI): 1.026–1.076; P<0.001]. The predictive model incorporating TyG-BMI demonstrated a higher area under the curve (AUC) for AKI compared to the model without it (AUC: 0.786 vs. 0.735). RCS analysis showed a significant overall association (P-overall <0.001) but no evidence of nonlinearity (P-nonlinear =0.18). Conclusions: In this limited single-center retrospective cohort, the model with TyG-BMI, age, sex, and CPB time had a modest apparent AUC of 0.786 (optimism-corrected: 0.723). The gain over baseline was small (ΔAUC =0.051). These modest metrics and the retrospective design limit our findings to being preliminary. External validation in larger prospective cohorts is mandatory before clinical application. The weaker association observed in the highest TyG index group also needs independent confirmation.