Raw Data for Nonlinear Relationship Between Early Postoperative Pulse Pressure Variation and Acute Kidney Injury Risk in Elderly Cardiac Surgical Patients
Description
This dataset supports the study titled "Nonlinear Relationship Between Early Postoperative Pulse Pressure Variation and Acute Kidney Injury Risk in Elderly Cardiac Surgical Patients". It includes two cohorts derived from the MIMIC-IV and MIMIC-III databases, which are publicly available, de-identified critical care datasets maintained by the Massachusetts Institute of Technology. The primary dataset was extracted from MIMIC-IV, containing ICU admissions between 2008 and 2019 at Beth Israel Deaconess Medical Center. Patients included were aged ≥ 60 years and underwent cardiac thoracotomy surgery. Data include demographic variables, comorbidities, vital signs, laboratory test results, hemodynamic variables such as pulse pressure variation (PPV), vasopressor use, and AKI status based on KDIGO criteria. The validation dataset was extracted from MIMIC-III, which includes ICU admissions from 2001 to 2012 at the same institution. Similar variables were collected and harmonized to match those in the MIMIC-IV cohort. The purpose of the dataset is to explore the association between early PPV (within 12 and 24 hours after ICU admission) and the risk of postoperative acute kidney injury (AKI) in elderly patients undergoing cardiac surgery. The data were extracted using Structured Query Language (SQL), cleaned and processed in R (version 4.4.3). Variables with more than 20% missing data were excluded. Only complete cases were used in the final analysis. Access to the MIMIC databases was obtained after completion of human subjects research training and PhysioNet credentialing (Certification Number: 13567919). The dataset shared here has been further de-identified to ensure privacy compliance.
Files
Institutions
- Capital Medical University Affiliated Anzhen Hospital