Diagnostic accuracy of TIMP-2 and IGFBP7 for detecting acute kidney injury in adult intensive care unit patients: a systematic review and meta-analysis

Acute kidney injury (AKI) is a frequent complication in hospitalized patients, particularly those admitted to the intensive care unit (ICU), which leads to high rates of morbidity and mortality. Under these conditions, tissue inhibitor metalloproteinases-2 (TIMP-2) and insulin-like growth factor-binding protein 7 (IGFBP7) have been identified as potential biomarkers. We conducted a systematic review and meta-analysis to evaluate the diagnostic accuracy of the TIMP2/IGFBP7 biomarkers for AKI in ICU patients, using data retrieved from PubMed, Embase, LILACS, Scopus, and Google Scholar. We use the Quality Assessment Tool for Diagnostic Accuracy Studies (QUADAS-2) to assess quality. We identified 2,530 publications and seven articles were included in this review, comprising a total of 2,676 participants. The overall summary receiver operating characteristic (SROC) analysis demonstrated good diagnostic performance (AUC=0.824), with a pooled sensitivity of 0.572 (95 % CI, 0.54-0.77) and specificity of 0.27 (95 % CI, 0.18-0.37). When analyzed by cut-off values, the 0.3 (ng/mL)²/1,000 threshold yielded higher sensitivity [0.815 (0.711-0.888)] but lower specificity [0.543 (0.500-0.585)], while the 2.0 (ng/mL)²/1,000 threshold showed the opposite pattern, with lower sensitivity [0.548 (0.384-0.702)] and higher specificity [0.838 (0.751-0.898)]. Heterogeneity was moderate to high (I²=69.8-79 %), partially explained by cut-off variation. The 0.3 cut-off demonstrated high sensitivity, whereas the 2.0 cut-off showed greater specificity. These results suggest that, clinically, a lower cut-off supports earlier AKI detection and reduces missed cases in critically ill patients, whereas a higher cut-off may be more appropriate when minimizing unnecessary interventions and resource utilization.