Serum Potassium Disorders Predict Subsequent Kidney Injury: A Retrospective Observational Cohort Study of Hospitalized Patients

Automated machine learning for early prediction of acute kidney injury in acute pancreatitis

BACKGROUND

Acute kidney injury (AKI) represents a frequent and grave complication associated with acute pancreatitis (AP), substantially elevating both mortality rates and the financial burden of hospitalization. The aim of our study is to construct a predictive model utilizing automated machine learning (AutoML) algorithms for the early prediction of AKI in patients with AP.

METHODS

We retrospectively analyzed patients who were diagnosed with AP in our hospital from January 2017 to December 2021. These patients were randomly allocated into a training set and a validation set at a ratio of 7:3. To develop predictive models for each set, we employed the least absolute shrinkage and selection operator (LASSO) algorithm along with AutoML. A nomogram was developed based on multivariate logistic regression analysis outcomes. The model's efficacy was assessed using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA). Additionally, the performance of the model constructed via AutoML was evaluated using decision curve analysis (DCA), feature importance, SHapley Additive exPlanations (SHAP) plots, and locally interpretable model-agnostic explanations (LIME).

RESULTS

This study incorporated a total of 437 patients who met the inclusion criteria. Out of these, 313 were assigned to the training cohort and 124 to the validation cohort. In the training and validation cohorts, AKI occurred in 68 (21.7%) and 29(23.4%) patients, respectively. Comparative analysis revealed that the AutoML models exhibited enhanced performance over traditional logistic regression (LR). Furthermore, the deep learning (DL) model demonstrated superior predictive accuracy, evidenced by an area under the ROC curve of 0.963 in the training set and 0.830 in the validation set, surpassing other comparative models. The key variables identified as significant in the DL model within the training dataset included creatinine (Cr), urea (Urea), international normalized ratio (INR), etiology, smoking, alanine aminotransferase (ALT), hypertension, prothrombin time (PT), lactate dehydrogenase (LDH), and diabetes.

CONCLUSION

The AutoML model, utilizing DL algorithm, offers considerable clinical significance in the early detection of AKI among patients with AP.

Stratification of Acute Kidney Injury Risk, Disease Severity, and Outcomes by Electrolyte Disturbances

Acute kidney injury (AKI) affects up to 30% of all hospitalized patients in Central Europe and the USA. New biomarker molecules have been identified in recent years; most studies performed so far however aimed to identify markers for diagnostic purposes. Serum electrolytes such as sodium and potassium are quantified in more or less all hospitalized patients. Aim of the article is to review the literature on the AKI predictive role of four distinct serum electrolytes in evolving/progressing AKI. The following databases were searched for references: PubMed, Web of Science, Cochrane Library, and Scopus. The period lasted from 2010 until 2022. The following terms were utilized: "AKI" AND "sodium" OR "potassium" OR "calcium" OR "phosphate" AND "risk" OR "dialysis" OR "recovery of kidney function" OR "renal recovery" OR "kidney recovery" OR "outcome". Finally, 17 references were selected. The included studies were mostly retrospective in nature. Particularly, hyponatremia has been shown to be associated with an overall poor clinical outcome. The association between dysnatremia and AKI is anything but consistent. Hyperkalemia and potassium variability are most likely AKI predictive. Serum calcium and AKI risk are associated in a U-shaped manner. Higher phosphate levels potentially predict AKI in non-coronavirus disease 2019 (COVID-19) patients. The literature suggests that admission electrolytes can offer valuable information about AKI onset during follow-up. Limited data are however available on follow-up characteristics such as the need for dialysis or the chance of renal recovery. These aspects are of particular interest from the nephrologist's perspective.