A Combined Cardiorenal Assessment for the Prediction of Acute Kidney Injury in Lower Respiratory Tract Infections

BACKGROUND: The accurate prediction of acute kidney injury (AKI) is an unmet clinical need. A combined assessment of cardiac stress and renal tubular damage might improve early AKI detection. METHODS: A total of 372 consecutive patients presenting to the Emergency Department with lower respiratory tract infections were enrolled. Plasma B-type natriuretic peptide (BNP) and neutrophil gelatinase-associated lipocalin (NGAL) levels were measured in a blinded fashion at presentation. The potential of these biomarkers to predict AKI was assessed as the primary endpoint. AKI was defined according to the AKI Network classification. RESULTS: Overall, 16 patients (4%) experienced early AKI. These patients were more likely to suffer from preexisting chronic cardiac disease or diabetes mellitus. At presentation, BNP (334 pg/mL [130-1119] vs 113 pg/mL [52-328], P <.01) and NGAL (269 ng/mL [119-398] vs 96 ng/mL [60-199], P <.01) levels were significantly higher in AKI patients. The predictive accuracy of presentation BNP and NGAL levels was comparable (BNP 0.74; 95% confidence interval [CI], 0.64-0.84 vs NGAL 0.74; 95% CI, 0.61-0.87). In a combined logistic model, a joint BNP/NGAL approach improved the predictive accuracy for early AKI over either biomarker alone (area under the receiver operating characteristic curve: 0.82; 95% CI, 0.74-0.89). The combined categorical cut point defined by BNP >267 pg/mL or NGAL >231 ng/mL correctly identified 15 of 16 early AKI patients (sensitivity 94%, specificity 61%). During multivariable regression analysis, the combined BNP/NGAL cutoff remained the independent predictor of early AKI (hazard ratio 10.82; 95% CI, 1.22-96.23; P = .03). CONCLUSION: A model combining the markers BNP and NGAL is a powerful predictor of early AKI in patients with lower respiratory tract infection. (C) 2012 Elsevier Inc. All rights reserved. (C) The American Journal of Medicine (2012) 125, 168-175