#767 Utility of cystatin C as a predictor for postoperative acute kidney injury in non-cardiac surgeries [S1

Abstract

Abstract Background and Aims Serum creatinine has been widely utilized to estimate glomerular filtration rate (eGFR), although various biomarkers have been developed to address its limitations. Among these biomarkers, the clinical use of Cystatin C is increasingly gaining recognition. However, there is a lack of evaluation regarding the utility of cystatin C in predicting the risk of postoperative acute kidney injury (PO-AKI) in non-cardiac surgeries. Method A retrospective cohort study was conducted on patients who underwent non-cardiac surgeries more than 1 hour from five departments (general surgery, neurosurgery, obstetrics and gynecology [OBGY], orthopedics, and urosurgery) at Chung-Ang University Hospital between 2018 and 2020. Patients with missing baseline information, including creatinine and cystatin C, and those without follow-up serum creatinine levels were excluded. PO-AKI was defined as KDIGO-AKI criteria occurring within 7 days after surgery. The estimated glomerular filtration rate (eGFR) was calculated from 2012 and 2021 chronic kidney disease-epidemiology collaboration (CKD-EPI) equations. Logistic regression was used to develop a prediction model, and C-statistics and the Delong test were used to compare the performance. Baseline comorbidities (diabetes mellitus [DM] and hypertension), laboratory findings (hypoalbuminemia [serum albumin <3.5 g/dL], anemia [hemoglobin <12 g/dL for female or <13 g/dL for male], and hyponatremia [sodium <135 mEq/L]), medication usage within 90 days of surgery (nonsteroidal anti-inflammatory drug and renin-angiotensin-system [RAS] blocker), and operation information (Anesthesia type, emergency operation, and operation duration) were adjusted. Results Overall, 339 patients were enrolled, with 49 patients (14.5%) developed PO-AKI. The median age was 73.0 [64.0; 80.0]; eGFR-Cr 86.4 [65.5; 99.1] mL/min/1.73 m2, eGFR-Cystatin C 70.0 [45.2; 92.7] mL/min/1.73 m2; eGFR-Cr/Cystatin C 77.2 [54.5; 100.5] mL/min/1.73 m2. Patients who developed AKI had a higher prevalence of diabetes (AKI 49.0%, no AKI 28.3%; p-value 0.006) and prescription of RAS blockers (AKI 63.3%, no AKI 34.5%; p-value <0.001), longer operation duration (AKI 63.3%, no AKI 34.5%; p-value <0.001), and lower eGFR of all types (AKI: eGFR-Cr 69.8, eGFR-Cystatin C 48.0, eGFR-Cr/Cystatin C 57.2; no-AKI: eGFR-Cr 88.0, eGFR-Cystatin C 74.0, eGFR-Cr/Cystatin C 80.5 mL/min/1.73 m2; p-values <0.001). No statistical differences were observed in other laboratory findings between the two groups (Table 1). The model with eGFR-Cystatin C demonstrated the highest area under the curve (AUC) among the three models (eGFR-Cr: AUC 0.78, 95% confidential interval [CI] 0.702‒0.857; eGFR-Cystatin C: AUC 0.81, 95% CI 0.735‒0.877; eGFR-Cr/Cystatin C: AUC 0.80, 95% CI 0.724‒0.871). Only the eGFR-Cystatin C-based model showed significant improvement in prediction compared to the eGFR-Cr-based model in the Delong test (eGFR-Cystatin C vs eGFR-Cr: p-value 0.033; eGFR-Cr/Cystatin C vs eGFR-Cr: p-value 0.055, Fig. 1). Conclusion Our findings suggest the potential usefulness of cystatin C in predicting the risk of PO-AKI non-cardiac surgery in addition to conventional PO-AKI risk factors. Further studies including larger and external cohorts, are needed for validation.