Fig. 2
Combining functional and damage biomarkers simultaneously to delineate the spectrum of AKI. Patients of the primary analysis cohort (n = 203) who had no UNGALt1 or UNGALt3 concentration available were excluded. Missing UNGALt1 occurred in 5 patients, and missing UNGALt3 in 1 patient, resulting in a total of 197 patients. Following the recommendations of de Geus et al., acute tubular damage was defined as a CSA-NGAL score of 2 or greater; either as UNGALt1 or UNGALt3 ≥ 150 ng/ml or as ∆UNGALt3−t1 > 100 ng/ml with UNGALt3 ≥ 125 ng/ml [26]. Subclinical AKI was defined when there was acute tubular damage (according to the ‘de Geus criteria’) and absence of AKI according to the KDIGO definition. In this way 84.6% of AKI in our specific cohort (i.e. 77/[77 + 14]) was classified as AKI without acute tubular damage. Subclinical AKI, which was missed by KDIGO, occurred in 5.1% of the patients. AKI acute kidney injury, CSA cardiac surgery-associated, d day, KDIGO Kidney Disease|Improving Global Outcomes, t1 time of intensive care unit admission, t3 4 h after intensive care unit admission, UNGAL urinary neutrophil gelatinase-associated lipocalin