A clinical, proteomics, and artificial intelligence‐driven model to predict acute kidney injury in patients undergoing coronary angiography

Abstract

Background Standard measures of kidney function are only modestly useful for accurate prediction of risk for acute kidney injury (AKI). Hypothesis Clinical and biomarker data can predict AKI more accurately. Methods Using Luminex xMAP technology, we measured 109 biomarkers in blood from 889 patients prior to undergoing coronary angiography. Procedural AKI was defined as an absolute increase in serum creatinine of ≥0.3 mg/dL, a percentage increase in serum creatinine of ≥50%, or a reduction in urine output (documented oliguria of <0.5 mL/kg per hour for >6 hours) within 7 days after contrast exposure. Clinical and biomarker predictors of AKI were identified using machine learning and a final prognostic model was developed with least absolute shrinkage and selection operator (LASSO). Results Forty‐three (4.8%) patients developed procedural AKI. Six predictors were present in the final model: four (history of diabetes, blood urea nitrogen to creatinine ratio, C‐reactive protein, and osteopontin) had a positive association with AKI risk, while two (CD5 antigen‐like and Factor VII) had a negative association with AKI risk. The final model had a cross‐validated area under the receiver operating characteristic curve (AUC) of 0.79 for predicting procedural AKI, and an in‐sample AUC of 0.82 ( P < 0.001). The optimal score cutoff had 77% sensitivity, 75% specificity, and a negative predictive value of 98% for procedural AKI. An elevated score was predictive of procedural AKI in all subjects (odds ratio = 9.87; P < 0.001). Conclusions We describe a clinical and proteomics‐supported biomarker model with high accuracy for predicting procedural AKI in patients undergoing coronary angiography.

A long journey for acute kidney injury biomarkers

“…As AKI tends to occur in patients with common risk factors and in certain medical setting, risk prediction [53,[76][77][78]. Especially, Kashani et al tested the predictive value of novel urine biomarkers, TIMP-2 and IGFBP7, with ICU patients, and suggested a clinical model to predict the risk of AKI [37].…”

“…Although the study had limitations regarding AKI risk prediction, as their main goal was to incorporate new urinary markers as risk factors, the study showed that the inclusion of novel biomarkers could improve the robustness of a prediction model in early periods of critical care. Although there is no consensus regarding using such biomarkers to predict AKI, several have shown promising results [37,53,[78][79][80]. Therefore, a novel biomarker that directly reflects kidney injury may further improve prediction in the future.…”

Contributions from the 2019 Literature on Bioinformatics and Translational Informatics

“…They used an original confounder-controlled cross validation procedure for robust generalization estimation. Ibrahim et al, [11] used a statistical approach leveraging LASSO attribute selection and Monte-Carlo cross validation simulation to identify variables predictive of acute kidney injuries using proteomics data. The work by Adam et al, [3] indirectly tackles the question of machine learning.…”

Artificial Intelligence in Acute Kidney Injury Risk Prediction

“…Ibrahim et al developed a clinical and proteomics AKI risk predictor with an ML approach (least absolute shrinkage and selection operator (LASSO) with logistic regression) in a prospective study of 889 patients undergoing coronary angiography [106]. The risk predictor included a history of diabetes, blood urea nitrogen/creatinine ratio, c-reactive protein, osteopontin, CD5 antigen-like, and Factor VII and had an AUROC of 0.790 for predicting procedural AKI [106].…”