Comparative accuracy of biomarkers for the prediction of hospital-acquired acute kidney injury: a systematic review and meta-analysis

Abstract

Background: Several biomarkers have been proposed to predict the occurrence of acute kidney injury (AKI); however, their efficacy varies between different trials. The aim of this study was to compare the predictive performance of different candidate biomarkers for AKI.

Methods: In this systematic review, we searched PubMed, Medline, Embase, and the Cochrane Library for papers published up to August 15, 2022. We selected all studies of adults (> 18 years) that reported the predictive performance of damage biomarkers (neutrophil gelatinase-associated lipocalin (NGAL), kidney injury molecule-1 (KIM-1), liver-type fatty acid-binding protein (L-FABP)), inflammatory biomarker (interleukin-18 (IL-18)), and stress biomarker (tissue inhibitor of metalloproteinases-2 × insulin-like growth factor-binding protein-7 (TIMP-2 × IGFBP-7)) for the occurrence of AKI. We performed pairwise meta-analyses to calculate odds ratios (ORs) and 95% confidence intervals (CIs) individually. Hierarchical summary receiver operating characteristic curves (HSROCs) were used to summarize the pooled test performance, and the Grading of Recommendations, Assessment, Development and Evaluations criteria were used to appraise the quality of evidence.

Results: We identified 242 published relevant studies from 1,803 screened abstracts, of which 110 studies with 38,725 patients were included in this meta-analysis. Urinary NGAL/creatinine (diagnostic odds ratio [DOR] 16.2, 95% CI 10.1-25.9), urinary NGAL (DOR 13.8, 95% CI 10.2-18.8), and serum NGAL (DOR 12.6, 95% CI 9.3-17.3) had the best diagnostic accuracy for the risk of AKI. In subgroup analyses, urinary NGAL, urinary NGAL/creatinine, and serum NGAL had better diagnostic accuracy for AKI than urinary IL-18 in non-critically ill patients. However, all of the biomarkers had similar diagnostic accuracy in critically ill patients. In the setting of medical and non-sepsis patients, urinary NGAL had better predictive performance than urinary IL-18, urinary L-FABP, and urinary TIMP-2 × IGFBP-7: 0.3. In the surgical patients, urinary NGAL/creatinine and urinary KIM-1 had the best diagnostic accuracy. The HSROC values of urinary NGAL/creatinine, urinary NGAL, and serum NGAL were 91.4%, 85.2%, and 84.7%, respectively.

Conclusions: Biomarkers containing NGAL had the best predictive accuracy for the occurrence of AKI, regardless of whether or not the values were adjusted by urinary creatinine, and especially in medically treated patients. However, the predictive performance of urinary NGAL was limited in surgical patients, and urinary NGAL/creatinine seemed to be the most accurate biomarkers in these patients. All of the biomarkers had similar predictive performance in critically ill patients. Trial registration CRD42020207883 , October 06, 2020.

Keywords: Acute kidney injury; Biomarker; Critically ill patient; Neutrophil gelatinase-associated lipocalin.

Fig. 1

The discriminative accuracy of the biomarkers to diagnose AKI (A) HSROCs for the AKI biomarkers. The global HSROCs depicting the discriminative accuracy of the biomarkers to diagnose AKI. The red point represents the observation and the circle represents the sample size. The asterisk “*” represents the estimate of HSROC, and the blue dotted circle around it indicates the 95% confidence interval. Among the biomarkers, NGAL, NGAL/Cr, L-FABP/Cr, TIMP-2 × IGFBP-7: custom, and TIMP-2 × IGFBP-7: 2 had good HSROCs (> 85–90%). (B) Heatmap plot depicting pairwise comparisons (row vs. column) of relative DOR between the biomarkers in the whole population. The contents of the diagonal are the values of the relative DOR. Red depicts a positive DOR, while yellow depicts no correlation. NGAL and NGAL/Cr had the best relative DOR of the biomarkers. (C) Heatmap plot depicting pairwise comparisons (row vs. column) of relative DOR between the biomarkers in the surgical subgroup. The contents of the diagonal are the values of the relative DOR. Red depicts a positive DOR, while yellow depicts no correlation. NGAL/Cr had the best relative DOR of the biomarkers. (D) Heatmap plot depicting pairwise comparisons (row vs. column) of relative DOR between the markers in the studies that did not use UO criteria. The contents of the diagonal are the values of the relative DOR. Red depicts a positive DOR, while yellow depicts no correlation. NGAL had the best relative DOR of the biomarkers. Abbreviations: AKI, acute kidney injury; Cr, creatinine; DOR, diagnostic odds ratio; HSROC, hierarchical summary receiver operating characteristic curve; IL-18, interleukin-18; KIM-1, kidney injury molecule-1; L-FABP, liver-type fatty acid-binding protein; NGAL, neutrophil gelatinase-associated lipocalin; TIMP-2 × IGFBP-7: tissue inhibitor of metalloproteinases-2 × insulin-like growth factor-binding protein-7; and UO, urine output