ABSTRACT
BACKGROUND: Interleukin-18 (IL-18) is a pro-inflammatory protein, which mediates ischaemic tubular injury, and has been suggested to be a sensitive and specific biomarker for acute kidney injury (AKI). The predictive value of IL-18 in the diagnosis, evolution, and outcome of AKI in critically ill patients is still unclear. METHODS: We measured urine IL-18 from critically ill patients at intensive care unit (ICU) admission and 24 h. We evaluated the association of IL-18 with developing new AKI, renal replacement therapy (RRT), and 90-day mortality. We calculated areas under receiver operating characteristics curves (AUCs), best cut-off values, and positive likelihood ratios (LR+) for IL-18 concerning these endpoints. Additionally, we compared the predictive value of IL-18 at ICU admission to that of urine neutrophil gelatinase-associated lipocalin (NGAL). RESULTS: In this study population of 1439 patients the highest urine IL-18 during the first 24 h in the ICU associated with the development of AKI with an AUC [95% confidence interval (CI)] of 0.586 (0.546-0.627) and with the development of Stage 3 AKI with an AUC (95% CI) of 0.667 (0.591-0.774). IL-18 predicted the initiation of RRT with an AUC (95% CI) of 0.655 (0.572-0.739), and 90-day mortality with an AUC (95% CI) of 0.536 (0.497-0.574). CONCLUSIONS: IL-18 had poor-to-moderate ability to predict AKI, RRT, or 90-day mortality in this large cohort of critically ill patients. Thus, it should be used with caution for diagnostic or predictive purposes in the critically ill.
Subject(s)
Acute Kidney Injury/urine , Interleukin-18/urine , Patient Outcome Assessment , Acute Kidney Injury/mortality , Acute Kidney Injury/therapy , Aged , Area Under Curve , Biomarkers/urine , Critical Illness , Female , Follow-Up Studies , Humans , Intensive Care Units , Male , Middle Aged , Predictive Value of Tests , Prospective Studies , ROC Curve , Renal Replacement Therapy/statistics & numerical dataABSTRACT
OBJECTIVE: Plasma neutrophil gelatinase-associated lipocalin (pNGAL) has been introduced as an early and sensitive biomarker of acute kidney injury (AKI), with an increased risk for renal replacement therapy (RRT) and adverse outcome in selected critically ill patient groups. Acute respiratory failure is the most common organ dysfunction in critically ill patients with an increased risk for AKI. Accordingly, we hypothesized that pNGAL would independently predict adverse outcome in a heterogeneous group of critically ill adult patients with acute respiratory failure. DESIGN AND SETTING: Prospective, multi-centre study in 25 Finnish intensive care units. PATIENTS AND METHODS: pNGAL was measured from critically ill patients with acute respiratory failure. We evaluated the predictive value of pNGAL for RRT, and hospital and 90-day mortality first separately, second in addition to the Simplified Acute Physiology Score (SAPS II), and third to RIFLE (Risk, Injury, Failure, Loss, End-Stage Renal Disease) AKI classification. Additionally, we assessed the factors associated with pNGAL by linear regression analysis. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: We included 369 patients. Median (interquartile range) baseline pNGAL was 169 (92-370) ng/ml. The areas under receiver operating characteristic curves of baseline pNGAL were as follows: 0.733 [95% confidence interval (CI) 0.656-0.810] for RRT, 0.627 (95% CI 0.561-0.693) for hospital, and 0.582 (95% CI 0.520-0.645) for 90-day mortality. Present infection, baseline creatinine, operative status, and pancreatitis were independently associated with baseline pNGAL. CONCLUSIONS: Baseline pNGAL gives no additional value into prediction of hospital and 90-day mortality compared with RIFLE or SAPS II, and has only moderate predictive power regarding RRT in critically ill adult patients with acute respiratory failure.
