René Adam, Michele Tedeschi, Nicolas Golse, Daniel Cherqui, Andrea Laurenzi, Denis Castaing, Oriana Ciacio, Boris Tréchot, Antonio Sa Cunha, Daniel Pietrasz, Didier Samuel, Mylène Sebagh, Gabriella Pittau, Elena Fernandez-Sevilla, Eric Vibert, Marc-Antoine Allard, Maud Robert, Nicola Guglielmo, Cyril Cosse, Hôpital Paul Brousse, Centre Hépato-Biliaire [Hôpital Paul Brousse] (CHB), Hôpital Paul Brousse-Assistance Publique - Hôpitaux de Paris, DHU Hepatinov [Villejuif], Physiopathologie et traitement des maladies du foie, Hôpital Paul Brousse-Institut National de la Santé et de la Recherche Médicale (INSERM)-Université Paris-Saclay, Université Paris Descartes - Faculté de Médecine (UPD5 Médecine), Université Paris Descartes - Paris 5 (UPD5), Hôpital Edouard Herriot [CHU - HCL], Hospices Civils de Lyon (HCL), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon, Rythmes Biologiques et Cancers, Université Paris-Sud - Paris 11 (UP11)-Institut National de la Santé et de la Recherche Médicale (INSERM), Cardiovasculaire, métabolisme, diabétologie et nutrition (CarMeN), Institut National de la Recherche Agronomique (INRA)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Hospices Civils de Lyon (HCL), Laboratoire de Physique Nucléaire et de Hautes Énergies (LPNHE (UMR_7585)), Institut National de Physique Nucléaire et de Physique des Particules du CNRS (IN2P3)-Université Paris Diderot - Paris 7 (UPD7)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), and CCSD, Accord Elsevier
International audience; BACKGROUND & AIMS: Liver macrosteatosis (MS) is a major predictor of graft dysfunction after transplantation. However, frozen section techniques to quantify steatosis are often unavailable in the context of procurements, and the findings of preoperative imaging techniques correlate poorly with those of permanent sections, so that the surgeon is ultimately responsible for the decision. Our aim was to assess the accuracy of a non-invasive pocket-sized micro-spectrometer (PSM) for the real-time estimation of MS. METHODS: We prospectively evaluated a commercial PSM by scanning the liver capsule. A double pathological quantification of MS was performed on permanent sections. Initial calibration (training cohort) was performed on 35 livers (MS\textless/=60%) and an algorithm was created to correlate the estimated (PSM) and known (pathological) MS values. A second assessment (validation cohort) was then performed on 154 grafts. RESULTS: Our algorithm achieved a coefficient of determination R2=0.81. Its validation on the second cohort demonstrated a Lin's concordance coefficient of 0.78. Accuracy reached 0.91%, with reproducibility of 86.3%. The sensitivity, specificity, positive and negative predictive values for MS \textgreater/=30% were 66.7%, 100%, 100% and 98%, respectively. The PSM could predict the absence (\textless30%)/presence (\textgreater/=30%) of MS with a kappa coefficient of 0.79. Neither graft weight nor height, donor body mass index nor the CT-scan liver-to-spleen attenuation ratio could accurately predict MS. CONCLUSION: We demonstrated that a PSM can reliably and reproducibly assess mild-to-moderate MS. Its low cost and the immediacy of results may offer considerable added-value decision support for surgeons. This tool could avoid the detrimental and prolonged ischaemia caused by the pathological examination of (potentially) marginal grafts. This device now needs to be assessed in the context of a large-scale multicentre study. LAY SUMMARY: Macro-vacuolar liver steatosis is a major prognostic factor for outcomes after liver transplantation. However, it is often difficult for logistical reasons to get this estimation during procurement. Therefore, we developed an algorithm for a commercial, portable and affordable spectrometer to accurately estimate this content in a real-time fashion. This device could be of great interest for clinical decision-making to accept or discard a potential human liver graft.