Table 3.

Studies of novel biomarkers

Study and PublicationBiomarkers StudiedSample SizeStudy DesignBiomarker Inclusion CriteriaPatient Type at Time of Biomarker MeasurementPrimary End PointSummary of Findings
Clinical trials
 Endre et al. 2010 (117)uGGT, uAP162Randomized double-blind placebo-controlled trial Evaluation of erythropoietin versus placebo to prevent AKI among critically ill adultsGGT×AP index ≥46.3Relative average plasma creatinine increase from baseline over 4–7 dNo statistically significant difference in primary outcome or in AKI incidence between placebo and treatment groups (AKI incidence 48.7% in placebo group versus 48.8% in treatment group; P>0.99)
 Zarbock et al. 2016 (45)pNGAL231Randomized clinical trial, single center Evaluation of early versus delayed initiation of dialysis for AKI in critically ill adults with KDIGO stage 2 AKIpNGAL >150 ng/mlMortality at 90 dReduced 90-d mortality in early initiation arm (39.3% versus 54.7%; HR 0.66; 95% CI, 0.45 to 0.97)
 Meersch et al. 2017 (51)TIMP-2*IGFBP7276Randomized controlled trial, single center Evaluation of KDIGO care bundle in patients with high risk of AKI after cardiac surgeryTIMP-2*IGFBP7 >0.3AKI within 72 h of cardiac surgeryReduced incidence of AKI in intervention arm (55% versus 71.7%; ARR 16.6%; 95% CI, 5.5 to 27.9)
 Göcze et al. 2018 (50)TIMP-2*IGFBP7125Randomized clinical trial, single center Evaluation of KDIGO care bundle in patients with high risk of AKI after major abdominal surgeryTIMP-2*IGFBP7 >0.3AKI within 7 d of abdominal surgeryNonsignificant reduction in AKI incidence in intervention arm (31.7% versus 47.5%; OR 1.96; 95% CI, 0.93 to 4.10) Significant reduction in AKI incidence in intervention arm in subgroup of patients with TIMP-2*IGFBP7 levels 0.3–2.0 (27.1% versus 48.0%; P=0.03)
 Zarbock et al. 2021 (118)TIMP-2*IGFBP7278Multicenter, multinational, randomized controlled trial Evaluation of adherence to KDIGO care bundle in patients with high risk of AKI after cardiac surgeryTIMP-2*IGFBP7 >0.3Adherence to KDIGO bundle protocolHigher rate of adherence to KDIGO bundle in intervention arm compared with control arm (65.4% versus 4.2%) Secondary end point: lower incidence of moderate and severe AKI in intervention group (14.0% versus 23.9%; ARR 10.0%; 95% CI, 0.9 to 19.1; P=0.03)
 Coca et al. 2014 (119)uNGAL, uIL-18, uKIM-1, uL-FABP, urinary albumin1199Prospective observation, multicenterHospitalized adults after cardiac surgery (1–3 d post-op)All-cause mortality (median follow-up 3 yr)In patients with clinical AKI, highest tertiles of peak uNGAL, uIL-18, uKIM-1, uL-FABP, urinary albumin associated with 2- to 3.2-fold higher risk of mortality compared with lowest tertiles
 Pike et al. 2015 (120)IL-6, IL-8, IL-10, IL-18, MMIF, TNFR-I, TNFR-II, DR-5817Prospective, nested observational cohort, multicenterCritically ill adults with AKI on KRTKidney recovery (alive and not on KRT by day 60 after hospital discharge), 60-d mortalityAUCs for kidney recovery: IL-6 0.61, IL-8 0.63, IL-18 0.58, MMIF 0.57 Clinical model AUCs for kidney recovery (0.73) and mortality (0.74). Addition of IL-8 to clinical model improved prediction of kidney recovery and mortality (0.76 and 0.78, respectively) IL-8 improved IDI and NRI for kidney recovery and mortality
 Koyner et al. 2015 (121)TIMP-2*IGFBP7692Secondary analysis of prospective observational multicenterCritically ill adults at time of ICU admissionComposite outcome of all-cause mortality or need for KRT at 9 moUnadjusted analysis: TIMP-2*IGFBP7 >2.0 associated with higher risk of end point (HR 2.11; 95% CI, 1.37 to 3.23) Multivariable analysis: TIMP-2*IGFBP7 >0.3 associated with end point only in patients who developed AKI (HR 1.44; 95% CI, 1.00 to 2.06, for levels 0.3 to ≤2.0; HR 2.16; 95% CI, 1.32 to 3.53, for levels >2.0)
 Parr et al. 2015 (122)uL-FABP, uIL-18, uKIM-1, uNGAL152Prospective observationalCritically ill adults with stage 1 AKIComposite outcome of persistent doubling of SCr (≥2 d), KRT, and mortalityAUCs for predicting composite outcome: uL-FABP 0.79, uIL-18 0.64, uKIM-1 0.62, uNGAL 0.65, combination of biomarkers 0.81 Clinical model AUC for composite outcome was 0.74; adding uL-FABP to clinical model improved AUC (0.82)
 Hollinger et al. 2018 (123)penkid583Prospective observationalCritically ill adults with sepsis or septic shockMAKE at 7 dpenkid concentration on admission associated with MAKE (aOR 3.3; 95% CI, 1.8 to 6.0)
 Xie et al. 2019 (124)TIMP-2*IGFBP7719Prospective observationalCritically ill adults with and without AKIIn-ICU mortality and initiation of CRRTAmong patients with AKI, those with elevated TIMP-2*IGFBP7 levels had higher in-ICU mortality (OR 2.087; 95% CI, 1.241 to 3.510) and more frequently reached composite end point of in-ICU mortality or CRRT initiation (OR 2.290; 95% CI, 1.401 to 3.744)
 Schunk et al. 2019 (67)DKK3733Prospective observationalAdults undergoing elective cardiac surgery (DKK3 measured preoperatively)AKI Secondary outcomes: persistent kidney dysfunction, KRT at 60 dAUC for postoperative AKI 0.783 (95% CI, 0.747 to 0.20). Adding DKK3 to clinical prediction model improved IDI and NRI Elevated DKK3 (>471 pg/mg) was associated with higher risk of persistent reduction in eGFR during follow-up compared with those with DKK3 <471 pg/mg (OR 2.01; 95% CI, 1.26 to 3.21)
 Legrand et al. 2019 (125)Cys C, pNGAL, uNGAL, penkid1207Prospective observational, multicenterAdult survivors of ICU who had received at least 24 h of mechanical ventilation or hemodynamic support Biomarkers measured at time of ICU dischargeAll-cause mortality at 1 yrBiomarker levels (Cys C, pNGAL, uNGAL, proenkephalin 119–159) all associated with increased all-cause mortality at 1 yr aOR for 1-yr mortality: uNGAL 2.08 (95% CI, 1.35 to 3.21); pNGAL 2.61 (95% CI, 1.71 to 3.97); Cys C 3.11 (95% CI, 1.88 to 5.16); proenkephalin 119–159 2.20 (95% CI, 1.44 to 3.38)
 Hoste et al. 2020 (126)pNGAL, uNGAL, TIMP-2*IGFBP7, uCCL14, penkid, uCHI3LI, Cys C, uL-FABP, uKIM-1, GST-π, IL-18331Prospective observational, multicenterCritically ill adults with stage 2–3 AKI (within 36 h of meeting KDIGO criteria)Development of persistent severe AKI (KDIGO stage 3) for ≥72 huCCL14 was most predictive of persistent stage 3 AKI (AUC 0.83; 95% CI, 0.78 to 0.87)
 Bagshaw et al. 2021 (68)uCCL14195Secondary analysis of prospective observational multicenterCritically ill adults within 36 h of onset of stage 2–3 AKIDevelopment of persistent severe AKI (KDIGO stage 3 for ≥72 h, or death or KRT occurring before 72 h)AUC for uCCL14 0.81 (95% CI, 0.72 to 0.89) Risk of persistent severe AKI was higher with higher values of uCCL14 KRT and/or death at 90 d was higher within tertiles of uCCL14 concentration
  • uGGT, urinary γ-glutamyl transpeptidase; uAP, urinary alkaline phosphatase; GGT, gamma-glutamyl transpeptidasealkaline phosphatase; pNGAL, plasma neutrophil gelatinase–associated lipocalin; KDIGO, Kidney Disease Improving Global Outcomes; HR, hazard ratio; 95% CI, 95% confidence interval; TIMP-2, tissue inhibitor of metalloproteinases–2; IGFBP7, IGF-binding protein 7; ARR, absolute risk reduction; OR, odds ratio; uNGAL, urinary neutrophil gelatinase–associated lipocalin; uIL-18, urinary IL-8; uKIM-1, urinary kidney injury molecule–1; uL-FABP, urinary liver-type fatty acid–binding protein; IL-6, plasma IL-6; IL-8, plasma IL-8; IL-10, plasma IL-10; IL-18, plasma IL-18; MMIF, macrophage migration inhibitory factor; TNFR-I, TNF receptor 1; TNFR-II, TNF receptor II; DR-5, death receptor–5; AUC, area under the curve; IDI, integrated discrimination improvement; NRI, net reclassification index; ICU, intensive care unit; SCr, serum creatinine; penkid, proenkephalin A 119–159; MAKE, major adverse kidney events (>50% increase in Cr from baseline, KRT, in-hospital death); aOR, adjusted odds ratio; CRRT, continuous renal replacement therapy; DKK3, dickkopf-3; Cys C, plasma cystatin C; uCCL14, urinary C-C motif chemokine ligand 14; uCHI3LI, urinary chitinase-3–like protein 1; GST-π, glutathione S-transferase–π.