Exploring Acute Kidney Injury Following Aortic Dissection: A Comprehensive Review of Machine Learning Models for Predicting Risk, Management Strategies, Complications, and Racial and Gender Disparities.

Abstract

Both types of aortic dissection (AD), Stanford type A and type B, can result in complications such as acute kidney injury (AKI) and aortic rupture. Renal complications in AD arise from compromised renal perfusion affecting the renal arteries. Understanding the intricate connection between AD and AKI is crucial for navigating the complexities of tailored treatment and formulating specific management plans. Concerning machine learning models, in patients with type A aortic dissection, factors such as decreased platelet count on admission, increased D-dimer level, longer cardiopulmonary bypass duration, elevated white blood cell levels, the need for blood transfusion, longer aortic clamp time, extended surgery duration, advanced age, and an elevated body mass index were positively associated with the development of AKI. For the risk of AKI after type B aortic dissection, elevated Nt-pro brain natriuretic peptide, prolonged activated partial thromboplastin time, elevated admission systolic blood pressure, and a higher contrast agent requirement during operative repair were found to predict the risk. Male gender was associated with a higher risk of AKI, and nonwhite race was linked to a higher risk of AKI, a greater likelihood of requiring more urgent procedures, and lower levels of insurance coverage. The treatment of AKI following AD requires a multifaceted approach. Identifying and addressing the underlying cause, such as low blood pressure, renal artery involvement, or medication-induced injury, is crucial for effective management and preventing further kidney damage. Maintaining proper fluid balance is essential for improving renal perfusion, but careful monitoring is necessary to avoid complications. The evolving landscape of research, particularly in biomarkers and AI programs, reveals a promising role in predicting the risk for and managing AKI post-AD.