PO-AKID-teller: An interpretable machine learning tool for predicting acute kidney injury requiring dialysis after acute type A aortic dissection surgery

Introduction

Estimation of an individual patient’s risk for postoperative acute kidney injury requiring dialysis can improve clinical decision making, support patient counseling, and plan future clinical trials. Based on routinely available preoperative and intraoperative features, we developed and validated a web-based risk calculator (named PO-AKID-teller) for calculating risk of acute kidney injury requiring dialysis after type A aortic dissection surgery. This tool can be easily applied at the bedside and provides a simple and interpretable estimation of risk.

Variables for model input

Preoperative variables:

Prior cardiovascular surgery: No or Yes

Serum creatinine: umol/L

Platelet: x109/L

The terminal site of dissection involvement: ascending aorta; aortic arch; thoracic descending aorta; abdominal aorta; iliac artery

Right coronary artery involvement: No or Yes

Intraoperative variables:

Urine output: mL

Estimated blood loss: mL