Abstract
Background
Acute kidney injury (AKI) has high morbidity and mortality in intensive care units (ICU). It can also lead to chronic kidney disease (CKD), more costs and longer hospital stay. Early identification of AKI is important.
Methods
We conducted this monocenter prospective observational study at West China Hospital, Sichuan University, China. We recorded information of each patient in the ICU within 24 h after admission and updated every two days. Patients who reached the primary outcome were accepted into the AKI group. Of all patients, we randomly drew 70% as the development cohort and the remaining 30% as the validation cohort. Using binary logistic regression we got a risk prediction model of the development cohort. In the validation cohort, we validated its discrimination by the area under the receiver operator curve (AUROC) and calibration by a calibration curve.
Results
There were 656 patients in the development cohorts and 280 in the validation cohort. Independent predictors of AKI in the risk prediction model including hypertension, chronic kidney disease, acute pancreatitis, cardiac failure, shock, pH ≤ 7.30, CK > 1000 U/L, hypoproteinemia, nephrotoxin exposure, and male. In the validation cohort, the AUROC is 0.783 (95% CI 0.730–0.836) and the calibration curve shows good calibration of this prediction model. The optimal cut-off value to distinguish high-risk and low-risk patients is 4.5 points (sensitivity is 78.4%, specificity is 73.2% and Youden’s index is 0.516).
Conclusions
This risk prediction model can help to identify high-risk patients of AKI in ICU to prevent the development of AKI and treat it at the early stages.
Trial registration TCTR, TCTR20170531001. Registered 30 May 2017, http://www.clinicaltrials.in.th/index.php?tp=regtrials&menu=trialsearch&smenu=fulltext&task=search&task2=view1&id=2573
A New Model for Caries Risk Prediction in Teenagers Using a Machine Learning Algorithm Based on Environmental and Genetic Factors
