A nomogram to predict congestive heart failure in patients with acute kidney injury: a retrospective study based on the MIMIC-III database.

Abstract

Object: Objective: Acute Kidney Injury (AKI) is a renal disease marked by diminished urine output and elevated serum creatinine levels. AKI has a global incidence rate of about 20%, with an average mortality rate of 23%. Cardiovascular disease emerges as one of the primary causes of death associated with AKI. We developed a nomogram to estimate the probability of patients with AKI developing congestive heart failure.

Method: We conducted a retrospective study of patients with AKI, using the MIMIC-III database. The patients were randomly divided into training and validation cohorts. Variables were selected via logistic regression, followed by the construction of the nomogram. The accuracy and sensitivity of the predictive model were verified using the Hosmer-Lemeshow test (HL) and the Area Under the Curve (AUC). The nomogram and SOFA scores were compared to APSIII using the Net Reclassification Index (NRI), Integrated Discrimination Improvement (IDI), Calibration curves, and Decision Curve Analysis (DCA).

Results: The final study included 9,174 individuals. The multivariate logistic regression revealed a correlation between age, Systolic Blood Pressure (SBP), Partial Pressure of Oxygen (PO2), hemoglobin, Blood Urea Nitrogen (BUN), Chloride (Cl^-), cardiac arrhythmias, valvular heart disease, pulmonary circulation disease, chronic pulmonary disease, and diabetes. These factors are strongly associated with the development of congestive heart failure. Based on these findings, we created a nomogram. This nomogram has a higher predictive effect than the SOFA score and the APSIII score (AUC = 0.751, SOFA: 0.659, APSIII: 0.62). Its verification through NRI, IDI, and DCA demonstrated that this nomogram offers superior specificity and clinical prognosis compared to the SOFA score and APSIII score.