Background 30%–40% of post-cardiac surgery patients develop acute kidney injury (AKI). AKI diagnosis uses raised serum creatinine levels which can take several days to rise.

Aim To systematically review the diagnostic value of novel blood biomarkers of AKI in children post-cardiac surgery, and identify those worthy of future research and inclusion into clinical practise.

Method Medline, EMBASE, CENTRAL and Web of Science were searched in October 2016 using the search strategy: AKI, cardiac surgery and biomarker. Two independent reviewers assessed the eligibility of the reports in Endnote based on predefined inclusion criteria (novel biomarker, clear definition of AKI, cardiac surgery in children [<18 year]) and exclusion criteria (no plasma/serum biomarker studied, no patients without AKI, non-English reports) at title, abstract and full paper screening stages. Eligible papers had data extracted into a custom-designed Excel spreadsheet and risk of bias was assessed using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS) tool. When possible meta-analysis was performed using MedCalc v16 where reports (n 3) on the same biomarker had an AUROC and 95% CI or standard error.

Results Of 3898 independent reports found, 21 met the inclusion criteria. 2 biomarkers had enough reports (n 3) to warrant independent comparison; Cystatin C [n=7] and plasma neutrophil gelatinase-associated lipocalin (pNGAL) [n=7]. Cystatin C had enough data to perform a meta-analysis ([n=5]; pooled AUROC=0.79 [95% CI: 0.73–0.85); I²=55%). pNGAL had conflicting evidence for its efficacy (AUROC range: 0.45–0.95). All other reported biomarkers showed varied performance, notably a combination of urine NGAL+Cystatin C showed good discrimination (AUROC=0.90). QUADAS analysis showed low risk of bias in all studies (score >8) as well as uncertainty bias due to lack of reporting on examiner blinding and participant withdrawals.

Conclusion Cystatin C may be a useful early diagnostic marker of AKI yet there is currently insufficient research for using any novel biomarker of AKI diagnosis after paediatric cardiac surgery. Large, multicentre studies with homogenous methodology should be undertaken to address this and to identify how they evaluate patient outcomes.