Prediction of urinary dickkopf-3 for AKI, sepsis-associated AKI, and PICU mortality in children

An overview of circulating and urinary biomarkers capable of predicting the transition of acute kidney injury to chronic kidney disease

INTRODUCTION

Acute kidney injury (AKI) defined by a substantial decrease in kidney function within hours to days and is often irreversible with higher risk to chronic kidney disease (CKD) transition.

AREAS COVERED

The authors discuss the diagnostic and predictive utilities of serum and urinary biomarkers on AKI and on the risk of AKI-to-CKD progression. The authors focus on the relevant literature covering evidence of circulating and urinary biomarkers' capability to predict the transition of AKI to CKD.

EXPERT OPINION

Based on the different modalities of serum and urinary biomarkers, multiple biomarker panel seems to be potentially useful to distinguish between various types of AKI, to detect the severity and the risk of AKI progression, to predict the clinical outcome and evaluate response to the therapy. Serum/urinary neutrophil gelatinase-associated lipocalin (NGAL), serum/urinary uromodulin, serum extracellular high mobility group box-1 (HMGB-1), serum cystatin C and urinary liver-type fatty acid-binding protein (L-FABP) were the most effective in the prediction of AKI-to-CKD transition regardless of etiology and the presence of critical state in patients. The current clinical evidence on the risk assessments of AKI progression is mainly based on the utility of combination of functional, injury and stress biomarkers, mainly NGAL, L-FABP, HMGB-1 and cystatin C.

Identification and validation of an explainable prediction model of acute kidney injury with prognostic implications in critically ill children: a prospective multicenter cohort study

Background

Acute kidney injury (AKI) is a common and serious organ dysfunction in critically ill children. Early identification and prediction of AKI are of great significance. However, current AKI criteria are insufficiently sensitive and specific, and AKI heterogeneity limits the clinical value of AKI biomarkers. This study aimed to establish and validate an explainable prediction model based on the machine learning (ML) approach for AKI, and assess its prognostic implications in children admitted to the pediatric intensive care unit (PICU).

Methods

This multicenter prospective study in China was conducted on critically ill children for the derivation and validation of the prediction model. The derivation cohort, consisting of 957 children admitted to four independent PICUs from September 2020 to January 2021, was separated for training and internal validation, and an external data set of 866 children admitted from February 2021 to February 2022 was employed for external validation. AKI was defined based on serum creatinine and urine output using the Kidney Disease: Improving Global Outcome (KDIGO) criteria. With 33 medical characteristics easily obtained or evaluated during the first 24 h after PICU admission, 11 ML algorithms were used to construct prediction models. Several evaluation indexes, including the area under the receiver-operating-characteristic curve (AUC), were used to compare the predictive performance. The SHapley Additive exPlanation method was used to rank the feature importance and explain the final model. A probability threshold for the final model was identified for AKI prediction and subgrouping. Clinical outcomes were evaluated in various subgroups determined by a combination of the final model and KDIGO criteria.

Findings

The random forest (RF) model performed best in discriminative ability among the 11 ML models. After reducing features according to feature importance rank, an explainable final RF model was established with 8 features. The final model could accurately predict AKI in both internal (AUC = 0.929) and external (AUC = 0.910) validations, and has been translated into a convenient tool to facilitate its utility in clinical settings. Critically ill children with a probability exceeding or equal to the threshold in the final model had a higher risk of death and multiple organ dysfunctions, regardless of whether they met the KDIGO criteria for AKI.

Interpretation

Our explainable ML model was not only successfully developed to accurately predict AKI but was also highly relevant to adverse outcomes in individual children at an early stage of PICU admission, and it mitigated the concern of the "black-box" issue with an undirect interpretation of the ML technique.

Funding

The National Natural Science Foundation of China, Jiangsu Province Science and Technology Support Program, Key talent of women's and children's health of Jiangsu Province, and Postgraduate Research & Practice Innovation Program of Jiangsu Province.

The role of urinary Dickkopf-3 in the prediction of acute kidney injury: a systematic review meta-analysis

BACKGROUND

To systematically evaluate the diagnostic efficacy of urinary Dickkopf-Related Protein 3 (DKK-3) in acute kidney injury and to explore the clinical application value of urinary DKK-3.

METHOD

English databases (PubMed, Embase, Cochrane, and WOS) and Chinese databases (VIP, WanFang data, and China National Knowledge Internet) were screened for relevant papers published before March 12, 2023. After literature screening and data extraction, quality assessment was performed according to the QUADAS-2 scoring system. Then, the combined diagnostic and predictive parameters were calculated using a bivariate mixed effect meta-analysis model. Deek's funnel plot asymmetry test assessed publication bias, and Fagan's nomogram plot was used to verify its clinical utility.

RESULT

A total of 5 studies involving 2787 patients were included in this meta-analysis, of which 4 focused on contrast-induced acute kidney injury (CI-AKI) and 1 focused on AKI associated with cardiac surgery. The analysis showed that urine Dickkopf-3 has high diagnostic accuracy for AKI, with a sensitivity of 0.55 (95% CI [0.41, 0.68]), specificity of 0.80 (95% CI [0.70, 0.87]), positive likelihood ratio (PLR) of 2.7 [1.8, 4.1], negative likelihood ratio (NLR) of 0.56 [0.42, 0.75], diagnostic odds ratio (DOR) of 5 [3, 9], and AUC of 0.74 [0.70-0.77]. We did not perform subgroup analyses for predictive value due to the small number of included studies.

CONCLUSION

Urinary DKK3 may have limited predictive ability for acute kidney injury, especially for AKI associated with cardiac surgery. Therefore, urinary DKK3 may serve as a potential predictor for AKI. However, clinical studies with larger samples are still needed for validation.

Kidney urinary biomarkers in patients with branched-chain amino acid and cobalamin metabolism defects

There is a clinical need for early detection of chronic kidney disease (CKD) in patients with organic acidurias. We measured kidney markers in a longitudinal study over 5 years in 40 patients with methylmalonic aciduria (Mut0 ), propionic aciduria (PA), cobalamin A (CblA), and cobalamin C (CblC) deficiencies. Neutrophil gelatinase-associated lipocalin (NGAL), calprotectin (CLP), kidney injury molecule-1 (KIM-1), dickkopf-3 (DKK-3), albumin and beta-2-microglobulin (B2MG) in urine, as well as cystatin C (CysC) in serum were quantified. In Mut0 patients, mean concentrations of B2MG, KIM-1, and DKK-3 were elevated compared with healthy controls, all markers indicative of proximal tubule damage. In PA patients, mean B2MG, albumin, and CLP were elevated, indicating signs of proximal tubule and glomerulus damage and inflammation. In CblC patients, mean B2MG, NGAL, and CLP were increased, and considered as markers for proximal and distal tubule damage and inflammation. B2MG, was elevated in all three diseases, and correlated with DKK-3 in Mut0 /CblA and with eGFR(CysC) and KIM-1 in PA patients, respectively. None of the markers were elevated in CblA patients. Significant deterioration of kidney function, as determined by steady increase in CysC concentrations was noted in seven patients within the observation period. None of the investigated biomarker profiles showed a clear increase or added value for early detection. In conclusion, we identified disease-specific biomarker profiles for inflammation, tubular, and proximal damage in the urine of Mut0 , PA, and CblC patients. Whether these biomarkers can be used for early detection of CKD requires further investigation, as significant kidney function deterioration was observed in only a few patients.