Diagnosis of cardiac surgery-associated acute kidney injury : differential roles of creatinine, chitinase 3-like protein 1 and neutrophil gelatinase-associated lipocalin : a prospective cohort study

Background: A common and serious complication of cardiac surgery prompting early detection and intervention is cardiac surgery-associated acute kidney injury (CSA-AKI). Urinary chitinase 3-like protein 1 (UCHI3L1) was found to predict AKI associated with critical illness in adults. Our aims were therefore to evaluate whether UCHI3L1 can also be used to predict AKI associated with elective cardiac surgery in adults, and to compare this predictive ability with that of urinary neutrophil gelatinase-associated lipocalin (UNGAL), more frequently assessed early serum creatinine (SCr) measurements, and various two-biomarker panels. Methods: This was a single-centre prospective cohort study at the eight-bed cardiac surgery ICU of Ghent University Hospital. AKI was diagnosed and classified according to the Kidney Disease|Improving Global Outcomes definitions for the diagnosis and staging of AKI, which are based on SCr and urine output (UO). Of the 211 enrolled elective cardiac surgery patients, we included 203 patients who had no AKI pre-operatively and at time of post-operative ICU admission (t1) in the primary endpoint analysis (i. e. AKI stage >= 1 within 48 h after t1), while 210 patients without AKI stage >= 2 pre-operatively and at t1 were included in the secondary endpoint analysis ( i. e. AKI stage >= 2 within 12 h after t1). Systemic and/or urine concentrations of Cr, CHI3L1 and NGAL were measured more frequently than SCr in routine early post-operative ICU practice. UO was monitored hourly in the ICU. Results: Within 48 h after t1, 46.8% of the patients had developed AKI (70.5% stage 1, 20.0% stage 2 and 9.5% stage 3). In the early post-operative period, only SCr was a good predictor of AKI within 48 h after t1 (primary endpoint). SCHI3L1 combined with either UCHI3L1 or UNGAL was a good predictor of AKI stage = 2 within 12 h after t1 (secondary endpoint). However, SCr and its absolute difference from pre- operative to early measures after surgery outperformed these combinations. Conclusions: We found that more frequent assessment of the functional biomarker SCr in the early post-operative ICU period (first 4 h) after elective cardiac surgery in adult patients had good to excellent predictive value for CSA-AKI, indicating that routine SCr assessment must become more frequent in order to detect AKI more early. This performance was in contrast with the inadequate predictive value of the urinary renal stress or damage biomarkers UCHI3L1 and UNGAL.Background: A common and serious complication of cardiac surgery prompting early detection and intervention is cardiac surgery-associated acute kidney injury (CSA-AKI). Urinary chitinase 3-like protein 1 (UCHI3L1) was found to predict AKI associated with critical illness in adults. Our aims were therefore to evaluate whether UCHI3L1 can also be used to predict AKI associated with elective cardiac surgery in adults, and to compare this predictive ability with that of urinary neutrophil gelatinase-associated lipocalin (UNGAL), more frequently assessed early serum creatinine (SCr) measurements, and various two-biomarker panels. Methods: This was a single-centre prospective cohort study at the eight-bed cardiac surgery ICU of Ghent University Hospital. AKI was diagnosed and classified according to the Kidney Disease|Improving Global Outcomes definitions for the diagnosis and staging of AKI, which are based on SCr and urine output (UO). Of the 211 enrolled elective cardiac surgery patients, we included 203 patients who had no AKI pre-operatively and at time of post-operative ICU admission (t1) in the primary endpoint analysis (i. e. AKI stage >= 1 within 48 h after t1), while 210 patients without AKI stage >= 2 pre-operatively and at t1 were included in the secondary endpoint analysis ( i. e. AKI stage >= 2 within 12 h after t1). Systemic and/or urine concentrations of Cr, CHI3L1 and NGAL were measured more frequently than SCr in routine early post-operative ICU practice. UO was monitored hourly in the ICU. Results: Within 48 h after t1, 46.8% of the patients had developed AKI (70.5% stage 1, 20.0% stage 2 and 9.5% stage 3). In the early post-operative period, only SCr was a good predictor of AKI within 48 h after t1 (primary endpoint). SCHI3L1 combined with either UCHI3L1 or UNGAL was a good predictor of AKI stage = 2 within 12 h after t1 (secondary endpoint). However, SCr and its absolute difference from pre- operative to early measures after surgery outperformed these combinations. Conclusions: We found that more frequent assessment of the functional biomarker SCr in the early post-operative ICU period (first 4 h) after elective cardiac surgery in adult patients had good to excellent predictive value for CSA-AKI, indicating that routine SCr assessment must become more frequent in order to detect AKI more early. This performance was in contrast with the inadequate predictive value of the urinary renal stress or damage biomarkers UCHI3L1 and UNGAL.A