*Purpose: Viability assessment during normothermic machine perfusion (NMP) continues to lack objective metrics that facilitate accurate clinical decision making, in order to minimize discard rates while optimizing transplant outcomes. Real-time perfusate analysis with fluorescent spectroscopy was recently shown to accurately predict post-transplant outcomes during hypothermic oxygenated perfusion. However, presence of highly fluorescent hemoglobin (both free and cellular) intrinsically limits such promising optical approaches for viability assessment during NMP.

*Methods: To overcome this limitation, we used a custom-built, high-resolution resonance Raman spectrometer and analyzed the blood-based perfusate of 6 discarded human donor livers that were subsequently subjected to NMP. The raw perfusate Raman spectrum was measured after 1, 3, and 6 hours of perfusion using a 441nm excitation wavelength. Separately measured spectra of all perfusate components – including hemoglobin – were subtracted from the raw spectrum using a spectral processing algorithm. The residual Raman spectrum was compared to other viability parameters during NMP such as lactate, resistance, transaminases, bile production, and bile pH.

*Results: A clear residual spectrum was obtained from the perfusate with distinct peaks at ~1267, ~1505, and ~1620 cm^-1. The intensity of this spectrum significantly increased during perfusion for all livers. 4 out of 6 livers met the transplantable viability criteria that are used in European clinical trials. These livers had significantly lower peak heights at all time points compared to the 2 livers that performed poorly during NMP (P=0.0072, P=0.0012 and P=0.0004 at T=1, T=3 and T=6 hours).

*Conclusions: Raman spectroscopy enables optical analysis of blood-based perfusates. This may provide fast and objective viability assessment during NMP to enable the use of discarded and marginal livers.

« Back to 2020 American Transplant Congress