Predicting AKI in patients with AMI: Development and assessment of a new predictive nomogram.

Abstract

Acute kidney injury (AKI) is a common complication of acute myocardial infarction (AMI) and is associated with both long- and short-term consequences. This study aimed to investigate relevant risk variables and create a nomogram that predicts the probability of AKI in patients with AMI, so that prophylaxis could be initiated as early as possible. Data were gathered from the medical information mart for the intensive care IV database. We included 1520 patients with AMI who were admitted to the coronary care unit or the cardiac vascular intensive care unit. The primary outcome was AKI during hospitalization. Independent risk factors for AKI were identified by applying least absolute shrinkage and selection operator regression models and multivariate logistic regression analyses. A multivariate logistic regression analysis was used to build a predictive model. The discrimination, calibration, and clinical usefulness of the prediction model were assessed using C-index, calibration plot, and decision curve analysis. Internal validation was assessed using bootstrapping validation. Of 1520 patients, 731 (48.09%) developed AKI during hospitalization. Hemoglobin, estimated glomerular filtration rate, sodium, bicarbonate, total bilirubin, age, heart failure, and diabetes were identified as predictive factors for the nomogram construction (P < .01). The model displayed good discrimination, with a C-index of 0.857 (95% CI:0.807-0.907), and good calibration. A high C-index value of 0.847 could still be reached during interval validation. Decision curve analysis showed that the AKI nomogram was clinically useful when the intervention was determined at an AKI possibility threshold of 10%. The nomogram constructed herein can successfully predict the risk of AKI in patients with AMI early and provide critical information that can facilitate prompt and efficient interventions.