The urine biomarkers TIMP2 and IGFBP7 can identify patients who will experience severe acute kidney injury following a cardiac arrest: A prospective multicentre study.

Abstract

To determine whether the urine biomarkers tissue inhibitor of metalloproteinases-2 (TIMP-2) and insulin-like growth factor-binding protein 7 (IGFBP7) can identify patients who will develop severe acute kidney injury (AKI) soon after cardiac arrest. We performed a prospective, multicentre study in three French ICUs. The performance of [TIMP-2]*[IGFBP7] was assessed for urine samples collected a median [IQR] of 240 [169-315] minutes post-collapse. The primary end-point was severe AKI (KDIGO stage 3), within 48 h of admission. Of the 115 patients analyzed, 32 (28%) developed severe AKI. Eleven of these required renal replacement therapy. The median [IQR] baseline [TIMP-2]*[IGFBP7] level was higher in patients who developed severe AKI (1.57 [0.80-6.62] (ng/ml)²/1000) than in those who did not (0.17 [0.05-0.59] (ng/ml)²/1000; p < 0.001). The baseline [TIMP2]*[IGFBP7] predicted -severe AKI with an area under the curve [95% confidence interval (CI)] of 0.91 [0.84-0.95], an optimal cut-off value of 0.39 (ng/ml)²/1000, a sensitivity [95%CI] of 97% [84-100], and a specificity of 72% [61-82]. A cut-off of 2.0 (ng/ml)²/1000 yielded a specificity of 98% [92-100]. For predicting severe AKI, baseline [TIMP-2]*[IGFBP7] was significantly more discriminant than baseline SCr (AUC [95%CI]: 0.73 [0.63-0.84]; p = 0.005), and slightly but not significantly more discriminant than baseline UO (AUC [95%CI]: 0.86 [0.78‒0.94], p = 0.08) Combining the baseline [TIMP2]*[IGFBP7] with baseline SCr and UO significantly improved the latter markers' predictive performance. Urine [TIMP-2]*[IGFBP7] effectively identify patients with a risk of severe AKI. Below a cut-off of 0.39 (ng/ml)²/1000, the risk of severe AKI is low.