Abstract
AKI is a common and serious post-operative complication in patients undergoing cardiac surgery, and its incidence is particularly high among elderly patients. Cardiac surgery-associated AKI (CSA-AKI) represents the second most common cause of AKI in the intensive care unit, but its true incidence could be underestimated, especially in the elderly population. The current biomarkers of AKI are unreliable and delayed during acute changes in kidney function. In the setting of subclinical AKI (S-AKI), biomarkers of tubular damage, such as NGAL, seem to be an early indicator of kidney damage.
The aim of this study was to investigate NGAL utility in the SAKI diagnosis on the first 48 h after cardiac surgery in the elderly population and its helpfulness in predicting adverse clinical outcomes in comparison to current criteria for AKI.
This is an observational study of 72 patients aged 65 and above, admitted to San Bortolo's cardiac surgery department for elective cardiosurgical procedure enrolled over a 5-months period. All patients underwent peripheral venous sample 48 h after cardiac surgery to assess plasmatic creatinine (48Cr) and NGAL (48pNGAL) in addition to exams already foreseen by clinical practice. For each patient, we studied renal (ArenO), respiratory (ArespO) and cardiovascular (ACvO) outcomes during hospitalization as well as 30 days and 6 months mortality. Creatinine increase AKI (CrIAKI) was defined by 48CrI ≥ 0.3 mg/dL and subclinical acute kidney injury (SAKI) was defined by 48pNGAL ≥ 100 pg/dL.
The mean age was 74.2 ± 6.1 years, and 61.1% were male. Out of 72 patients, 22 were submitted to CABG, 34 to valvular procedures, 8 to mixed (valvular + CABG) procedures and 8 to aortic procedures with different procedures not included in the three aforementioned categories. At 30 days, mortality was 8.3% (6 patients) and at 6 months, mortality was 12.5% (9 patients). A total of 27 patients (37.5%) presented AKI according to KDIGO (4), and 4 (5.5%) needed RRT. SAKI was significantly associated with 30-day mortality (P = .0238), 6-months mortality (P = .002), adverse renal outcome ARenO (P = .004) and need for RRT (P = .005). CrIAKI was significantly associated with 30-day mortality (P = .009) and ARenO (P = .0001), but not with 6-month mortality nor need for RRT. ROC curves for renal, respiratory and cardiovascular outcomes and for 30-day and 6-month mortality, for both 48CrI and 48pNGAL values, are shown in Fig. 1.
In our study, we found NGAL being much more sensitive than creatinine increase for predicting adverse renal outcome. AKI incidence is high on the post-operatory period among the studied population: 37.5% presented AKI. A total of 34.7% of patients presented with CrIAKI and 37.5% presented with isolated SAKI. SAKI was significantly associated with ARenO, need for RRT, 30-day and 6-month mortality. Therefore, plasmatic NGAL can help identify those who will develop AKI and those who will need RRT and shows the potential to be an effective early biomarker of kidney injury.
Comments