Related Experiment Video
Updated: Oct 7, 2026
Using a Chemical Biopsy for Graft Quality Assessment
Published on: June 17, 2020
Raman spectral signature uncovers predictive biomarkers for primary graft dysfunction in liver transplant: A
Edouard Wasielewski1, Antoine Castel2, Maëlle Guillout3
1Department of Hepatobiliary and Digestive Surgery, University Hospital of Rennes, University of Rennes, Rennes, France; University Hospital Center, INSERM, CIC1414, Hepatobiliary Surgery and Liver Transplantation Research Unit, Rennes, France; Inserm, Univ Rennes, OSS (Oncogenesis Stress Signaling) laboratory, UMR_S 1242, Eugène Marquis, F-35042, Rennes, France.
Background:
Graft selection remains a major challenge in liver transplantation, particularly in the context of ongoing organ shortages. Currently, few tools are available to predict graft quality and functional recovery before surgery. This study aimed to evaluate the ability of Raman spectroscopy to predict early graft function following liver transplantation.
Methods:
All patients aged over 18 who underwent liver transplantation at University Hospital of Rennes between November 2021 and October 2023 were eligible. The patients were divided into 2 groups, with primary graft dysfunction (PGD) and without PGD. Graft assessment was based on liver biopsies obtained at the time of revascularization.
Results:
A total of 133 patients were included. Baseline characteristics of donors and recipients did not differ significantly between the two groups. Cold ischemia time was significantly longer in the PGD group (420 vs. 383 min, P< 0.05). PGD was associated with longer intensive care unit stay (5.0 vs. 3.0 days, P = 0.05), more severe complications (51.6% vs. 25.5%, P < 0.01), and higher rates of retransplantation (6.5% vs. 0%, P = 0.05). Spectral analysis revealed a significantly lower protein-to-lipid ratio in the PGD group (1.92 ± 0.37 vs. 2.14 ± 0.42, P < 0.01). In multivariate analysis, this ratio was an independent predictor of PGD [odds ratio (OR) = 0.06; 95% confidence interval (CI): 0.01-0.29; P < 0.01], along with arterial hypertension (OR = 3.23, 95% CI: 1.07-10.40; P = 0.04). The multivariate model demonstrated good ROC performance for predicting PGD, with the area under the receiver operating characteristic curve of 0.78.
Conclusions:
This is the first study to demonstrate that Raman spectroscopy can predict PGD. A reduced protein-to-lipid ratio appears to be a significant independent risk factor, supporting the potential of this technique in graft evaluation.
