Intraoperative Hyperspectral Imaging Predicts Early Allograft Dysfunction and Overall Survival in Liver Transplantation.

Abstract

Objective: This study explored the novel application of hyperspectral imaging (HSI) for in vivo allograft perfusion assessment during liver transplantation (LT) and its potential value for predicting early allograft dysfunction (EAD), graft, and overall survival (OS).

Background: LT is a well-established therapy for acute and chronic liver diseases, with excellent outcomes. However, a significant proportion of recipients experience EAD, which affects graft and OS. EAD is associated with ischemia-reperfusion injury. HSI is a noninvasive imaging modality that provides information on tissue characteristics, such as tissue hemoglobin, water index, oxygenation, and perfusion.

Methods: We included all patients who underwent orthotopic LT with full-size allografts between 2019 and 2021. HSI was performed 15 minutes after reperfusion of the donor liver and subsequently analyzed. Furthermore, we collected data on postoperative graft function and clinical outcomes.

Results: A total of 73 LT recipients were included in this study. Around 56.9% had expanded criteria donors (N = 41). The mean model for end-stage liver disease score was 22 (±10). Eighteen patients (25%) had EAD. The statistical analysis demonstrated that recipients with EAD had significantly lower near-infrared (NIR) perfusion values after reperfusion. Recipients with low NIR had more pronounced reperfusion injury in postoperative laboratory studies. OS was significantly lower in recipients with low NIR than in those with high NIR (P = 0.049).

Conclusions: HSI is a promising, noninvasive tool, offering real-time, detailed graft perfusion assessment during LT. The fusion of spatial and spectral information is unique to HSI, making it an essential imaging technology for the further development of AI applications in surgery.