Photoacoustic Tomography Assessments During Ex Vivo Normothermic Perfusion. A Novel and Noninvasive Modality to Evaluate Endothelial Integrity

The pancreases used were derived from a preliminary feasibility study and were employed following the principle of reuse. Animal euthanasia was performed in accordance with the law governing the protection of animals for scientific purposes (conditions of killing: Articles R214-98 to R214-98-1). The research protocol was approved by the French Ministry of Research (APAFiS nos 31507). All experiments adhered to the ARRIVE 2.0 recommendations and the European Directive 2010/63/EU on animal experimentation. In organ transplantation, ex vivo normothermic perfusion (EVNP) is increasingly being developed and used before transplantation, both to evaluate grafts and to recondition them. The assessment of grafts during EVNP is based on macroscopic evaluation of the organ, perfusion parameters, biochemical markers, and sometimes histology or organ function recovery (eg, urine output in kidney transplantation). To date, no strategy includes a morphological evaluation of vascularization or the integrity of the endothelial barrier. Photoacoustic tomography (PT) is an innovative modality for assessing organ vascularization, particularly microvascularization. This noninvasive imaging technique visualizes hemoglobin distribution, indirect assessing graft vascularization, both on the surface and deep within the parenchyma. An optical probe directly applied to the parenchyma emits a laser into the tissue with a wavelength corresponding to the excitation wavelength of hemoglobin. The absorption of this light by hemoglobin induces rapid thermal expansion, leading to the localized generation of ultrasound waves triggered by this excitation.1,2 The resulting acoustic signal is then recorded and, after reconstruction, allows visualization of hemoglobin and, indirectly, blood vessels, as well as hemorrhagic suffusions that may occur during ischemia/reperfusion injury (IRI).3 Here, we present the first images of the evaluation of pancreatic grafts by PT during EVNP (porcine model, use of pancreatic grafts due to their high sensitivity to IRI4). After 2 or 24 h of cold ischemia (static cold storage or hypothermic machine perfusion), pancreatic grafts were reperfused. The imaging platform for PT was provided by Deepcolor Imaging SAS (Nantes, France; http://www.deepcolorimaging.com). During reperfusion, PT was performed every 5 min during the first 30 min of reperfusion. A volumetric image of 1 × 1 × 1 cm was obtained during each analysis. The resolution is on the order of 100 µm. The perfusion fluid during EVNP consisted of autologous whole blood without additives. The reperfusion of 5 pancreatic grafts was evaluated (Figure 1). After 2 h of cold preservation, PT revealed the presence of hemoglobin along vessel-like structures during reperfusion according to EVNP. The signal appears at 5 min in structures approximately 3–4 mm in size as well as in those <1 mm. This supports the idea of revascularization in both medium-sized and small vessels at 5 min. The signal remains constant over the course of the following 30 min. The mesh-like structures of small vessels corresponded to pancreatic gland tissue, whereas larger vessels corresponded to duodenal parenchyma. No hemoglobin extravasation outside the vessels was observed during the first 30 min of reperfusion. After 24 h of preservation, the signal seems to appear progressively, which may correspond to a slower revascularization of the pancreas. We observed the appearance of additional imaging near the vessels, which may correspond to hemoglobin outside the vessels. In our opinion, this may reflect hemoglobin extravasation, secondary to endothelial barrier rupture and IRI.FIGURE 1.: Photoacoustic tomography evaluation of pancreatic transplants preserved for 2 or 24 h (SCS or HMP) before reperfusion according to ex vivo normothermic perfusion. The amount of hemoglobin is represented by the intensity of the photoacoustic signal. The colors represent the depth of the measurement and help to understand the 3D structure: (1) pancreatic parenchyma and (2) duodenal parenchyma. 3D, 3-dimensional; HMP, hypothermic machine perfusion; SCS, statical cold storage.PT may provide evidence of endothelial barrier function loss. Thus, PT could become a major marker of interest for graft evaluation during EVNP. It allows for in-depth vascularization evaluation while remaining noninvasive, nonionizing, and free of contrast agents, which is particularly valuable in EVNP. In contrast, PT has several technical limitations: although the acquisition time is <10 s, it does not allow for real-time assessment. Furthermore, light penetration into the parenchyma depends on the organ and is less effective in pigmented organs (eg, liver), impacting both image depth and the quality of the assessment obtained. Additionally, PT does not allow for determining whether the blood delivered to the graft is oxygenated or not. Finally, the future integration of PT technology into EVNP will first require a detailed description of the normal vascularization of organs and IRI as visualized by PT. ACKNOWLEDGMENTS The authors express their gratitude to Deepcolor Imaging SAS (Nantes, France, http://www.deepcolorimaging.com) for providing the platform, the time dedicated by their technical teams, and their scientific expertise. The authors thank the Center for Research in Transplantation and Translational Immunology for their scientific expertise (S. Le Bas-Bernardet, D. Minault, J. Herouvet, and G. Blancho), the French National Agency (Agence de la biomedicine) for grant funding, the French Urological Association (Association Française d'Urologie) for grant funding, the Nantes University Hospital for grant funding, the biochemical analysis laboratory of the Nantes University Hospital for logistical support, and the Institut Georges Lopez for grant funding and technical support.