Early serum cystatin C-enhanced risk prediction for acute kidney injury post cardiac surgery: a prospective, observational, cohort study.

Acute kidney injury (AKI) is a common post-cardiac surgery complication. It leads to increased morbidity and mortality. The aim of our study is to identify the prevalence and risk factors of AKI and to demonstrate if early postoperative serum cystatin C (sCyC) could accurately predict the development of AKI. We prospectively studied 628 patients undergoing elective cardiac surgery. Pre-morbid and operative variables known to be or potentially associated with AKI or other adverse outcomes were examined. AKI was defined according to Kidney Disease Improving Global Outcomes (KDIGO) creatinine criteria. Blood samples for biomarker measurement were collected at baseline, within 10 h of surgical completion and daily for three days. Logistic regression was used to assess predictive factors for AKI including 10 h sCyC. Model discrimination was assessed using receiver operator characteristic (ROC) curves. AKI occurred in 178 (28.3%) patients, Stage 1 in 17.5%, Stage 2 in 8.6% and Stage 3 in 2.2%. Mortality rose progressively with increased AKI stage (non-AKI 0.2%, Stage 1 1.8%, Stage 2 11.1% and Stage 3 35.7%). Age > 75 years, baseline estimated glomerular filtration rate (eGFR), proteinuria, diabetes mellitus, hypertension, hyperuricaemia, NYHA classification >2, recent myocardial infarction were associated with AKI in univariate analysis. A multivariate logistic model with clinical factors (age, eGFR, hypertension, NYHA classification >2, combined surgery and operation time) demonstrated moderate discrimination for AKI (area under ROC curve [AUC] 0.75). The 10 h postoperative sCyC levels strongly associated with AKI. After multivariable adjustment, the highest quartile of sCyC was associated with 13.1 - higher odds of AKI, compared with the lowest quartile. Elevated 10 h sCyC levels associated with longer hospital stay, longer intensive care unit stay and duration of mechanical ventilation. The addition of 10 h sCyC improved model discrimination for AKI (AUC 0.81). AKI following cardiac surgery was identified using KDIGO criteria in around one fourth of the patients. These patients had significantly increased morbidity and mortality. When added to prediction model, 10 h sCyC may enhance the identification of patients at higher risk of AKI, providing a readily available prognostic marker.