SP3.9 The role of urinary biomarkers in predicting post operative kidney function

Abstract

Abstract Aims Post-operative Acute Kidney Injury (AKI) has implications on recovery, conferring a significant risk of morbidity and mortality. A previous study of 488 major abdominal surgery patients demonstrated that urinary biomarkers could predict peri-operative AKI. The aim of this follow-up study was to assess the biomarkers (NGAL, KIM-1, DKK-3 and IGFBP-7*TIMP-2) for their ability to predict long-term changes in renal function. Methods Predictive ability for long term renal function derangement was assessed by classifying the participants change in their ‘CKD’ stage from the pre-operative status, determined by the measured eGFR at follow up as per the KDIGO criteria (at least 6 months from their operation). Biomarker performance was assessed utilising AUROC analysis. Biomarkers were also interrogated for their ability to predict mortality at one year. Results Follow up measurement of renal function was obtained in 446 of 488 patients. 40 patients died within 1 year of their operation. Median time of follow up renal function measurement was 398 days post operatively (IQR: 366 – 490 days). 169 patients developed an increase in their CKD stage. No biomarker showed good predictive ability for such a rise – all displayed AUC values of <0.6. Similarly, AUC values for the prediction of mortality within one year were low, with all AUC values <0.65. Discussion This study has determined that urinary AKI biomarkers are poor predictors for long term derangement in renal function and mortality at one year in the surgical patient. Such evidence suggests that biomarkers are best suited to detect changes in the immediate post operative period.