Development and Validation of a Risk Prediction Model for Cardiac Surgery-associated Acute Kidney Injury Based on the Gut-Kidney Axis, Incorporating Gut Microbiota and Their Metabolites

Cardiac surgery-associated acute kidney injury (CSA-AKI) is one of the most common severe postoperative complications after cardiac surgery, which significantly increases the risks of prolonged hospitalization, chronic kidney disease and all-cause mortality. Current single-index risk scoring systems cannot fully predict individual CSA-AKI risk, and the underlying pathological mechanism remains incompletely clarified. Emerging evidence reveals that gut dysbiosis and altered intestinal metabolites participate in kidney injury via the gut-kidney axis, while few prospective cohort studies integrate multi-omics data to develop preoperative predictive models for CSA-AKI.

This prospective cohort study will consecutively recruit adult subjects receiving elective cardiac surgery at Peking Union Medical College Hospital. Basic demographic data, laboratory examinations, intraoperative and postoperative clinical parameters will be recorded at multiple perioperative time points. Fresh stool specimens and artery blood will be collected from all participants before surgery, 24 hours and 7 days after cardiac surgery for 16S rRNA gene sequencing and untargeted metabolomics analysis to characterize gut microbial composition and metabolite profiles.

The primary objective is to develop and internally validate a multi-dimensional predictive model combining clinical variables, gut flora biomarkers and metabolic signatures to stratify preoperative CSA-AKI risk of cardiac surgical patients. Secondary objectives include comparing the differences of intestinal microbiota and metabolites between patients with and without subsequent CSA-AKI, screening key microbial strains and differential metabolites related to postoperative renal impairment, and elucidate the molecular pathways of the gut-kidney axis mediating CSA-AKI occurrence.

All participants will receive standardized perioperative management following clinical guidelines. Subjects will be followed up until 7 days after surgery to identify CSA-AKI events defined by standardized serum creatinine criteria. The performance of the established prediction model will be evaluated by area under the receiver operating characteristic curve (AUC), calibration curve and decision curve analysis. All specimen testing and statistical analyses will be conducted following standardized omics and clinical research workflows. This study is approved by the institutional ethics committee, and written informed consent will be obtained from all enrolled patients prior to sample and data collection.

Inclusion criteria

• Aged 18 years or older;
• Scheduled for elective cardiac surgery;
• Surgery performed under cardiopulmonary bypass (CPB);
• Able to complete preoperative stool and blood sample collection independently;
• The patient and their family members voluntarily participate in this study, agree to the collection of preoperative stool samples, blood samples and clinical data, cooperate with follow-up visits, and sign the informed consent form.-

Exclusion criteria

（1) Patients with end-stage renal disease or dependence on renal replacement therapy before surgery; (2) Patients with a preoperative estimated glomerular filtration rate (eGFR) < 30 ml/min/1.73m² or severe chronic kidney disease (CKD stage 4-5); (3) Patients with a history of renal transplantation or unilateral nephrectomy; (4) Patients who received antibiotic treatment within 4 weeks before surgery; (5) Patients who took probiotic or prebiotic preparations within 4 weeks before surgery; (6) Patients with a previous history of intestinal surgery, such as colectomy and small bowel resection; (7) Patients with active inflammatory bowel disease, chronic diarrhea or constipation requiring pharmacological intervention; (8) Patients with preoperative intestinal obstruction, gastrointestinal bleeding, or those requiring fasting for more than 24 hours; (9) Patients undergoing emergency surgery (time from admission to surgery < 24 hours), or those unable to complete preoperative informed consent, dietary investigation and baseline sample collection; (10) Patients who received mechanical bowel preparation (e.g., enema) within 24 hours before surgery; (11) Patients with preoperative active infection or sepsis; (12) Patients treated with preoperative glucocorticoids or immunosuppressants (excluding routine doses); (13) Patients who died intraoperatively or within 24 hours postoperatively, or those who could not complete the evaluation of the primary outcome (postoperative 7-day acute kidney injury [AKI]); (14) Female patients who are pregnant, lactating or in menstruation; (15) Patients complicated with gastrointestinal malignant tumors or receiving systemic chemotherapy, targeted therapy or abdominal radiotherapy; (16) Patients with severe liver disease (Child-Pugh grade B or above).