Neutrophil gelatinase-associated lipocalin predicts acute kidney injury in critically ill patients

Introduction

Plasma and urinary concentrations of neutrophil gelatinase-associated lipocalin (NGAL) are early biomarkers of acute kidney injury (AKI) [1]. Most studies about the NGAL value to predict AKI were done in patients after cardiac surgery and not enough information about its usefulness in critically ill patients exists. To analyze the predictive value of plasma NGAL to predict AKI, evaluated by RIFLE score, in critically ill patients, particularly in those with severe sepsis and septic shock.

Methods

NGAL was measured in whole-blood (Triage^® NGAL Test; Inverness Medical, UK) at admission and after 48 hours in patients admitted to a general ICU. Patients were classified according to RIFLE score at admission and 24 and 48 hours thereafter and according to NGAL concentrations at admission using values suggested by Constatin and colleagues [2] (≤155 or >155 ng/ml). NGAL values at admission were compared in patients under systemic inflammatory response syndrome (SIRS), severe sepsis and septic shock.

Results

Sixty patients were included (28 female) with mean age 58 ± 18.6 years, SAPS II 40.6 ± 15.7, SOFA 6.9 ± 3.9, length of ICU stay of 11.4 ± 11.1 days and mortality of 23%. Forty patients developed SIRS, six severe sepsis and 14 septic shock. Eleven patients developed AKI (RIFLE F), 10 of them at ICU admission; extracorporeal renal therapies were required in six cases. Thirty patients have NGAL ≤155 ng/ml. When patients were classified according to RIFLE score at 24 hours of admission, NGAL values at admission were 180 ± 170 ng/ml in 36 patients with RIFLE 0, 243 ± 317 ng/ml in seven patients with RIFLE R, 303 ± 134 ng/ml in four patients with RIFLE I and 536 ± 421 ng/ml in nine patients with RIFLE F (P = 0.003). One patient was discharged before 24 hours (NGAL<60 ng/ml); whereas three patients who died before 24 hours had NGAL at admission of 578 ± 350 ng/ml. The area under the ROC curve of NGAL at admission for the prediction of AKI was 0.83 (95% confidence interval 0.70 to 0.97, P = 0.001), with an optimal cutoff value of 277 ng/ml with 73% sensitivity and 77% specificity. NGAL concentrations at admission were 160 ± 159 ng/ml in 40 patients with SIRS, 335 ± 206 ng/ml in six patients with severe sepsis and 540 ± 347 ng/ml in 14 patients with septic shock (P < 0.001).

Conclusions

Plasma NGAL concentrations measured at ICU admission appeared a useful predictor of AKI in critically ill patients; in addition, NGAL concentrations showed an increasing pattern from SIRS to severe sepsis and septic shock.