Dual acellular subnormothermic machine perfusion for enhanced uterus preservation in a swine model Authors: Lucile Cabanel, Haïzam Oubari, Bradley W Ellis, Ludivine Dion, Vincent Lavoué, Carla Sousa, Loïc Van Dieren, Olivier Morel, Mark A Randolph, Curtis L Cetrulo, Korkut Uygun, Alexandre G Lellouch, Yanis Berkane, Basak E Uygun Published: 2026 ABSTRACT AbstractIntroductionMachine perfusion has rapidly developed in the last decade and demonstrated improved preservation of conventional and marginal organs by reducing ischemia and hypoxia injuries. The aim of this study was to assess whether subnormothermic machine perfusion improves graft preservation compared with SCS in a porcine uterus model.Material and MethodsSwine uteri were procured from circulatory‐dead donors and allocated into two groups. In the experimental group, the organs were preserved with 4 h‐dual subnormothermic machine perfusion (n = 4) while the control group (n = 4) underwent 4 h‐static cold storage. After preservation, all uteri received ex vivo whole blood normothermic reperfusion (NMP) to simulate transplantation. Perfusate samples were analyzed from uterine arteries and veins every 30 min for gas and metabolic analysis, and weight variations were recorded continuously during perfusion. Following simulated transplantation, uterine contraction was assessed after oxytocin injection, and macrovascular perfusion was evaluated using contrast angiography. Several biopsies were procured from different parts of the uterus for histological and immunohistochemistry analysis.ResultsIn the experimental group, the weight of the uteri and lactate difference between vein and artery decreased during subnormothermic machine perfusion. Weight gain was significantly higher in the control group (p = 0.03), which was consistent with interstitial edema observed on histological sections in this group. Uterine contraction evaluated through semi‐quantitative scoring was higher in the experimental group (p = 0.02).ConclusionsIn this porcine model, subnormothermic machine perfusion reduces edema and preserves contractility more effectively than static cold storage. We demonstrate its particular interest in recovering organs procured from deceased donors. In addition, subnormothermic machine perfusion may enable other applications, including graft assessment, treatment and tolerance induction.Keywords: machine perfusion, organ preservation, static cold storage, uterus transplantation, vascularized composite allotransplantation, VCA Abbreviations ischemia–reperfusion injuries machine perfusion static cold storage subnormothermic machine perfusion enzyme terminal deoxynucleotidyl transferase dUTP nick end uterus transplantation vascularized composite allotransplantation warm ischemia time Key Points Pre‐transplant uterus preservation remains limited by ischemia‐related injury and edema under static cold storage. In a porcine model, subnormothermic dual machine perfusion reduced edema, improved metabolic parameters, and preserved contractility, supporting its potential for deceased‐donor uterus transplantation. INTRODUCTION Uterus transplantation (UTx) is currently the only option for women with absolute uterine infertility to achieve pregnancy, including patients with Mayer‐Rokitansky‐Küster‐Hauser syndrome or prior hysterectomy. 1 , 2 The inaugural living‐donor uterine transplant was conducted in Saudi Arabia in 2000, 3 thereby establishing the foundation for this novel surgical technique. Unfortunately, complications related to the vascular system and graft necrosis led to hysterectomy 3 months later. Over a decade later, the first uterine transplant from a deceased donor occurred in Turkey. This procedure was notable for its successful embryo transfer, which unfortunately failed to develop into pregnancy over the course of several weeks. 4 A further 10 years elapsed before the first live birth following uterine transplantation. 5 Today, UTx has evolved into a clinical reality, with more than 140 transplants performed worldwide and live‐birth rates approaching 50%. 2 , 6 , 7 Unlike other vascularized composite allotransplantations (VCA), UTx is a temporary graft, removed after completion of childbearing, placing particular importance on graft procurement and preservation to ensure optimal function and reproductive outcomes. Despite growing clinical experience, UTx remains associated with important challenges, including graft rejection, early graft loss, and high incidence of thrombotic complications or hypoperfusion‐related complications. 2 , 8 , 9 In the second ISUTx registry report, 12‐month graft survival was approximately 75%, with early graft losses mainly attributed to hypoperfusion or thrombosis, and rejection episodes were frequent but reversible with antirejection therapy during the first 10 months. The mechanisms underlying thrombotic and hypoperfusion‐related complications remain incompletely characterized in clinical UTx. Further investigation of ischemia–reperfusion injury (IRI), endothelial dysfunction, and microcirculatory impairment may help clarify their respective contribution to graft dysfunction and guide future preservation strategies. 10 , 11 , 12 , 13 , 14 Living donors currently account for approximately 75% of UTx procedures, exposing healthy individuals to surgical morbidity and supporting further development of deceased‐donor strategies and improved preservation protocols. 2 , 15 Static cold storage (SCS) remains the standard preservation technique in UTx. However, machine perfusion (MP) has emerged as an effective dynamic strategy in solid organ transplantation, reducing ischemia and hypoxia injury and improving outcomes, particularly in marginal grafts. MP has demonstrated improved preservation of both conventional and marginal organs by reducing ischemia‐ and hypoxia‐related injury. 16 , 17 , 18 , 19 This dynamic preservation method is now widely employed in clinical practice for lung, heart, liver, and kidney transplantation, 19 , 20 , 21 as it allows to precondition the organ and minimizes ischemia, improving transplantation outcomes, 22 , 23 yet its advantages in UTx remain to be determined. Subnormothermic machine perfusion (SNMP) presents many advantages, including compatibility with acellular perfusate solutions and the convenience of room temperature systems, avoiding the need for bulky cooling or heating equipment, while still decreasing cell metabolism. It has demonstrated promising results in liver and other VCA models 24 , 25 , 26 , 27 but remains unexplored in the uterus. 28 Overall, the application of MP in UTx may yield several potential benefits, including a reduction in ischemia and hypoxia time or the modulation of IRI. To guarantee the significant contribution of such research prospects, pertinent preclinical surgical models are essential. In this study, we aimed to assess whether SNMP improves graft preservation compared with SCS in a porcine uterus model. To simulate early transplantation response and improve the relevance of IRI assessment, all organs underwent ex vivo normothermic reperfusion with autologous whole blood. MATERIAL AND METHODS Animals and experimental design All animals received humane care following the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and protocols 2020N000015 (04/28/23) and 2024N000128 (07/29/24) were approved by the Massachusetts General Hospital Institutional Animal Care and Use Committee (IACUC). Overall, 11 female Yucatan minipigs weighing between 30 and 45 kg were used for uterus procurement. To minimize the number of animals needed and according to the 3R principles, 29 other organs were collected by other research teams. Uterus procurement was performed either immediately after heart procurement or following euthanasia by exsanguination. Total warm ischemic time (WIT) was kept under 60 min (time between euthanasia and uterine ex vivo preservation). All animals received a full dose (100 IU/kg) of intravenous heparin and an additional half dose before euthanasia. Prior to complete circulatory arrest, a 500 cc CPDA blood bag (Kruuse, Langeskov, Denmark) was filled with arterial blood procured from the femoral artery or the descending aorta. At this point, the warm ischemia temperature gradually decreased until the preservation phase. The study design is represented in Figure 1. Following procurement, the uteri were allocated into the experimental or control groups (n = 4 per group). In the experimental group, the grafts underwent 4 h of SNMP with Modified Steen Solution (Steen+) followed by 2 h of normothermic machine perfusion (NMP) with whole blood (SNMP group). The control group underwent 4 h of SCS preservation in a Histidine‐tryptophan‐ketoglutarate (HTK) solution and 2 h of NMP with whole blood (SCS group). As additional controls, multiple biopsies were procured from the three remaining uteri immediately after procurement (fresh controls), after 4 h of SNMP, and after 4 h of SCS. Experimental study design. HTK, Histidine‐tryptophan‐ketoglutarate; NMP, Normothermic machine perfusion; SCS, Static cold storage; SNMP, Subnormothermic machine perfusion; WIT, Warm ischemia time. Uterus procurement Animals were placed in the supine position, and a 10‐cm median infraumbilical incision was performed. On each side, after identifying uterine vessels, an opening was created into the broad ligament laterally to the uterine vein. Ovarian and utero‐ovarian vessels were ligated, adnexa removed, and uterus freed from broad ligament and vagina with careful cervical coagulation to prevent leakage during perfusion. On the back table, uterine vessels were skeletonized, both uterine arteries were dilated, and 18‐G angiocatheters were inserted, followed by immediate heparinized saline flushing until clear outflow (Figure S1). The surgical technique has been described previously. 30 Dual‐SNMP and SCS preservation In the SNMP group, organs were connected to the perfusion system at room temperature (20°C) and perfused with Steen+, an optimized solution for VCAs detailed in Table S1. 31 , 32 Briefly, a deionized water base was supplemented with ions, albumin, glucose, steroids, insulin, and 15% bovine serum albumin (Sigma Aldrich, Saint Louis, MI, USA). Vancomycin (Hospira, Lake Forest, IL, USA) and Piperacillin‐Tazobactam (Auromedics, East Windsor, NJ, USA) were added to prevent bacterial proliferation, and heparin (Pfizer, New York, NY, USA) to prevent clotting. A pH range of 7.5–7.6 was targeted by adding sodium hydroxide (Sigma Aldrich, Saint Louis, MI, USA). This value was expressly high as the pH tended to fall as the machine was circulated and oxygenated with a 5% carbogen mixture. Two independent perfusion circuits were used for each uterine artery, following our preliminary study comparing single and dual‐SNMP. 30 The perfusion system was composed of a roller pump (MasterFlex® L/S®, Cole‐Parmer®, 77200‐32), an oxygenator (Affinity Pixie™ Oxygenation System, Medtronic, BBP241), a bubble trap (Radnoti Bubble Trap Compliance Chamber, Radnoti, 130149), and a pressure sensor (PM‐P‐1, Living Systems Instrumentation, PM‐P‐1), connected with 16G silicon tubing in which the Steen+ solution was circulated in a closed loop. Perfusate was oxygenated with carbogen gas (95% oxygen, 5% carbon dioxide). Uteri were placed on a homemade 3D‐printed holder mounted on a scale for continuous weight monitoring during perfusion. The set‐up is illustrated in Figure 2. Diagram of the machine perfusion system used for uterus preservation and reperfusion. (A) Diagram of the machine perfusion system for uterus machine perfusion. (B) Representative photograph of a porcine uterus undergoing machine perfusion. Perfusion was started using a low flow to maintain a constant pressure between 20 and 35 mmHg. Before connecting the organ to the perfusion setup, pressure values of the system without the uterus were noted at different flow rates (P system). During perfusion, values of the system including the graft were noted (P system + ut), and the resulting vascular pressure (P vasc, mmHg) was calculated as: The vascular resistance was calculated using the formula: where R is the vascular resistance (mmHg/mL/min), P vasc is the calculated vascular pressure (mmHg) according to Equation 1 and Q is the flow rate (mL/min). In the SCS group, uteri were manually flushed with 20 mL of 4°C HTK solution at approximately 2 mL/min flow rate. The organs were weighed before and after the HTK flush and were immersed in HTK in a sterile bag and cold stored at 4°C for 4 h. Whole blood reperfusion To simulate early transplantation and reveal IRI, the preservation phase was followed by ex vivo whole blood reperfusion using NMP. In the SNMP group, the Steen+ solution was removed from the reservoir and replaced with 500 mL of autologous whole blood. The organ‐holding basin was placed on a heating plate with an agitator, and the temperature was increased to 39°C to approximate the physiological temperature as closely as possible (37°C). In the SCS group, the uterus was connected to MP as described above. In both groups, the whole blood reperfusion lasted 2 h. Biochemical analyses during perfusion Perfusion and viability parameters were assessed every 30 min during both SNMP and whole blood reperfusion phases. Perfusate samples were collected from the two veins (outflows) and upstream of the arteries (inflows) for gas and biochemical analysis using an automatic analyzer (RAPIDPoint®500, Siemens, Germany). Lactate arteriovenous difference (Lactvein‐art, mmol/L) was calculated as follows: Similarly, potassium release (Potassiumvein‐art, mmol/L) was calculated using the formula: Oxygen consumption (O2,consumption, mmHg/mL/min/g) was calculated using the formula: where 0.0314 is Henry's constant in water at 20°C at 1 atm, pO 2 is the partial oxygen pressure (mmHg), Q is the arterial inflow rate (mL/min), and W is the initial weight of the organ (g). In the SNMP group, two additional venous outflow perfusate samples were collected in tubes (Nalgene™ General Long‐Term Storage Cryogenic Tubes) and were snap‐frozen in liquid nitrogen at the end of the preservation phase. The same process was performed at the end of the whole blood reperfusion phase using whole blood samples from both groups, which were snap‐frozen with and without prior centrifugation (1000 g for 10 min) for plasma storage. These frozen samples were analyzed to measure the lactate dehydrogenase activity in the plasma following reperfusion (Abcam 102 526 Lactate Dehydrogenase Assay Kit). Uterus contractility At the end of the whole blood reperfusion, oxytocin was injected into each uterine artery via the bubble trap [1 IU of lyophilized oxytocin powder (Sigma‐Aldrich®, USA) diluted in 1.5 mL sterile water], and the arterial pressure variation was recorded over 20 min. Uterine contraction was video recorded and independently scored by two blinded observers using a semi‐quantitative 0–3 scale: 0, no visible contraction and no relevant arterial pressure change (<10 mmHg variation); 1, arterial pressure increase >10 mmHg without visible contraction; 2, moderate visible contraction; and 3, major contraction with complete contraction of both uterine horns folding toward the uterine body. Both observers assigned identical scores for all specimens. This pragmatic scale was used to capture oxytocin‐induced contractile responsiveness, which has been used in previous ex vivo uterus perfusion models as a functional marker of uterine viability. Microvascular assessment Microvascular integrity was examined through X‐ray angiography in all replicates following reperfusion. Ten milliliters of contrast agent (Omnipaque, GE Healthcare, Marlborough, USA) were carefully injected through the uterine arteries, and images were acquired using a Powermobil C‐Arm (Siemens, Munich, Germany). Histology and immunohistochemistry After carefully flushing 10% formalin in the intravascular compartment for vessel preservation, several full‐thickness biopsies from horns, corpus, cervix, and uterine vessels were procured at the end of the reperfusion phase. The biopsies were fixed in formalin, dehydrated, paraffin‐embedded, sectioned, and stained with eosin and hematoxylin, and Masson's Trichrome. The sections were analyzed by light microscopy for edema, necrosis, and apoptosis. Apoptosis was analyzed using the enzyme terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay kit (CF540, Biotium, Fremont, CA, USA). Immunohistochemical CD31 staining was performed in large‐vessel sections, beyond 5 cm downstream of uterine artery cannulation, to assess endothelial integrity. Paraffin‐embedded sections were deparaffinized, subjected to heat‐induced antigen retrieval, and incubated overnight at 4–8°C with an anti‐CD31 antibody. Signal detection was performed using a biotin–streptavidin system with 3,3′‐diaminobenzidine as chromogen, followed by hematoxylin counterstaining. All stained slides were subsequently scanned using the Hamamatsu Nanozoomer Digital slide scanner (Hamamatsu Photonics K.K., Hamamatsu City, Japan) and digitally documented. Biopsies were procured from control uteri at H0 (1h‐WIT), after 4h‐SNMP and 4h‐SCS and processed similarly. Transmission electron microscopy (TEM) Small pieces (2 × 2 mm2) of porcine uterus horn tissue (originally fixed in formalin) were excised from ‘full‐thickness biopsies’ and transferred into 2.0% paraformaldehyde/2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.4) and allowed to fix for several hours at room temperature, then switched into fresh fixative and allowed to further infiltrate overnight at 4°C. Specimens were rinsed several times in 0.1 M cacodylate buffer, infiltrated for 1 h in 1% osmium tetroxide, rinsed multiple times again in cacodylate buffer, then dehydrated through a graded series of ethanols to 100%, followed by a brief (10 min) dehydration in 100% propylene oxide. Material was then transferred into a 2:1 mix of propylene oxide:eponate resin (Ted Pella, Redding, CA), allowed to infiltrate for 2 h at room temperature, then placed into a 1:1 mix of propylene oxide: Eponate resin for overnight infiltration at room temperature on a gentle rotator. The following day, specimens were transferred into fresh 100% Eponate and allowed to infiltrate for several hours on a gentle rocker table, then embedded in flat molds with fresh 100% Eponate resin and allowed to polymerize in a 60°C oven (24–48 h). Ultra‐thin (70 nm) sections were cut from embedded blocks using an Ultra 45° diamond knife (DIATOME/EMS, Fort Washington, PA) and a Leica EM UC7 ultramicrotome, collected onto formvar‐coated grids, stained with 2% uranyl acetate and Reynold's lead citrate, and examined in a JEOL JEM 1011 transmission electron microscope at 80 kV. Images were collected using an AMT digital imaging system with proprietary image capture software (Advanced Microscopy Techniques, Danvers, MA). Statistical analysis All data were recorded in Excel (Microsoft, Redmond, WA, USA), and all statistical analyses were performed using Prism (v. 10.3.0, GraphPad Software, La Jolla, CA, USA). Given the limited sample size (n = 4 per group), data are presented as individual values with median and interquartile range. Between‐group comparisons were performed using the Mann–Whitney U test, and within‐group comparisons across timepoints using the Wilcoxon signed‐rank test. A two‐tailed p‐value < 0.05 was considered statistically significant. RESULTS SNMP was evaluated as a dynamic preservation method versus SCS. In total, five uteri underwent SNMP (one for histologic control, and four were evaluated after 2h‐NMP). Similarly, five underwent SCS with the same allocation. Mean WIT was 57.5 ± 5 min in both groups. Perfusion parameters During preservation, perfusion parameters were available only for the SNMP group (Figure 3). Flow was adapted to maintain a pressure of 20–35 mmHg and increased slightly during the first hour, likely due to capillary recruitment, 33 , 34 , 35 from 4.8 ± 3.2 mL/min at H0 to 6.1 ± 3.9 mL/min at H4. Beyond the first hour of perfusion, flow, pressure, and vascular resistance remained stable. Profile of the perfusion parameters during 4h‐SNMP. (A) Arterial Flow rate (mL/min); (B) Arterial Pressure (mmHg); (C) Calculated Arterial Resistance (mmg/mL/min); (D) Weight change throughout the perfusion (% of the initial weight). The curves indicate the average for the n = 8 hemi‐uteri (except for weight, which is the weight of the whole uterus, i.e., n = 4), and error bars indicate the standard deviation of the mean. Mann–Whitney U‐test was performed to compare weight evolution, revealing a statistically significant reduction in initial weight between HO and H4 in the SNMP group (p = 0.02). Graft weight declined during SNMP between H0 and H4 (−20 ± 10%), which may be due to the high oncotic pressure of the Steen+, and stabilized after 2 h, whereas SCS grafts showed significant weight gain after 4 h (+14 ± 4%). During whole blood reperfusion, perfusion parameters were compared in both groups (Figure 4). Flow was adapted to maintain a pressure of 20–35 mmHg. For both groups, parameters remained stable up to 90 min of reperfusion. In the SCS group, pressure increased at H2. Similarly, arterial resistance increased at the end of blood reperfusion without reaching statistical significance. Throughout reperfusion, weight was significantly higher in the SCS group at each time point (p = 0.03). Profile of the perfusion parameters during 2h‐reperfusion in both groups. (A) Flow rate (mL/min); (B) Pressure (mmHg); (C) Calculated Arterial Resistance (mmHg/mL/min); (D) Weight change throughout the perfusion (% of the initial weight). In each group, the curves indicate the average for the n = 8 hemi‐uteri (except for weight, which is the weight of the whole uterus, i.e., n = 4), and error bars indicate the standard deviation of the mean. Mann–Whitney U‐test was performed to compare weight evolution. Throughout reperfusion, weight was significantly higher in the SCS group compared to the SNMP group, at each timepoint (p = 0.02). Biochemical parameters Biochemical parameters are presented in Figure 5. During SNMP preservation, lactate release showed a non‐significant decrease, progressively eliminating accumulated lactate during the WIT (Figure 5A). After 90 min, lactate release remained low and stable. Oxygen consumption remained relatively stable between 0.08 and 0.38 mmHg mL/min/g (min–max values) throughout SNMP (Figure 5C). Monitoring of main hypoxia markers during machine perfusion. The upper part corresponds to the preservation phase for the SNMP group (A, B, C), and the lower part to the reperfusion for both groups (D, E, F). (A, D) Lactate difference between artery and vein; (B, E) Potassium difference between artery and vein; (C, F) Calculated O2 consumption. The data represent the average for both sides of each uterus, and error bars show the standard deviation of the mean. During whole blood reperfusion, lactate release and oxygen consumption were comparable in both groups (Figure 5D, F). Potassium release decreased during the first hour in the SCS group but remained stable with SNMP, with no group difference thereafter (Figure 5E). Lactate dehydrogenase activity after reperfusion was higher in SCS than SNMP grafts, without statistical significance (n = 3 per group, p = 0.18; Figure S2). Functional evaluation In the SNMP group, all uteri responded with a contraction score ranging from 2/3 to 3/3. In the SCS group, the hormonal stimulation allowed increasing pressure, but without visual contraction, except for one graft. Based on the semi‐quantitative score, contraction was significantly higher in the SNMP group (p = 0.02, Figure 6). Video S1 shows a representative contraction scored 3/3 in an SNMP graft. Uterine contractility score (mean with standard error) at the end of reperfusion. Vascular integrity After blood reperfusion, contrast angiography was performed to evaluate the microvasculature. A representative replicate of both groups is shown in Figure S3. No differences were observed in the vascular tree distribution. However, vessel lumens appeared wider in the SCS group. Histology and immunohistochemistry analysis Histological findings are presented in Figure 7 and Figure S4. Compared with controls, horn samples showed no major changes after preservation and reperfusion. However, SCS grafts exhibited increased endometrial interstitial edema (black arrows) and more apoptotic cells (stained in orange‐brown nuclei in the TUNEL slides with orange arrows) before and after reperfusion, while the myometrium remained largely preserved. Vessel architecture appears well preserved in both groups, with intact intima, media, and adventitia, although lumen opening was observed after MP and cold preservation. Histological analysis of representative biopsies showed good preservation in the endometrium and the myometrium after 4h‐SNMP and 2h‐reperfusion with whole blood. Black arrows show interstitial edema in the endometrium in the SCS group. Orange arrows highlight apoptotic cells with the TUNEL staining. The bar on each image corresponds to the same scale, i.e., 250 pm. TUNEL labelling revealed minimal nuclear staining, limited to the inner vessel layer in the SNMP group. Homogenous CD31 labeling confirmed preserved endothelial lining along the vessel lumen (Figure S5). Electron microscopy analysis after reperfusion revealed well‐preserved smooth muscle ultrastructure in both groups, with organized myofibrils, no significant interstitial edema, and intact capillary endothelium (Figure 8). Examination of smooth muscle cell ultrastructure revealed gap junctions in both groups, typically multiple, discontinuous, yet clustered within the same intercellular interface, as illustrated in the micrographs. Electron microscopy analysis of uterine biopsies from the SNMP and SCS groups after 2h‐reperfusion with whole blood revealed better preservation of smooth muscle cell structure (black arrows) in the SNMP group. The smooth muscle cells' orientation is distinguishable as a tightly packed myofibril. No significant interstitial edema is observed in either group. No notable endothelial damage is detected in capillaries (red stars). Smooth muscle gap junctions are characterized by symmetrical cytoplasmic densification on both sides of the contact zone, a typical feature of uterine smooth muscle (red circles). No significant differences were found between groups. DISCUSSION This study provides the first description of a dynamic preservation strategy for the swine uterus using SNMP, with SCS as a control. Grafts were subsequently evaluated by ex vivo NMP with whole blood to simulate early transplantation conditions. SNMP was associated with reduced tissue injury, as shown by minimal edema reflected by stable uterine weight and confirmed by histological findings compared with SCS. This effect is likely related, at least in part, to the high oncotic pressure of the Steen+ solution, which allowed for preventing edema formation during SNMP. Dual perfusion of the uterine arteries was a key technical feature of this model, allowing independent adjustment of flow to each vascular territory and ensuring homogeneous organ perfusion, as confirmed by contrast angiography. This approach, developed following preliminary single‐versus‐dual SNMP experiments, 30 likely contributed to reduced shear stress and prevention of edema, 33 particularly in the bicornuate porcine uterus. A perfusion pressure of 20–35 mmHg was selected as a conservative low‐pressure setting for SNMP. Although this pressure is lower than physiological mean arterial pressure, the aim of the model was not to reproduce systemic arterial conditions, but to provide controlled graft preservation while limiting vascular overdistension, endothelial stress, tissue edema, and perfusate leakage in an isolated ex vivo uterus. The uterus is a high‐resistance, low‐flow organ during back‐table flushing, and physiological arterial pressures may not be appropriate in the absence of normal autoregulation and systemic venous return. Bilateral uterine artery perfusion at low pressure was therefore used to achieve stable perfusion while minimizing perfusion‐related injury. Although the human uterus, as a single uterine body, is not strictly symmetrical, anatomical variations may similarly benefit from dual perfusion. In contrast to uncontrolled MP, which may induce endothelial injury, 34 TEM revealed preserved endothelial integrity after SNMP, whereas abrupt arterial pressure increases during blood reperfusion in SCS grafts suggested endothelial damage induced during cold storage. These findings are consistent with observations in limb preservation models after SCS. 36 , 37 Metabolic monitoring further supported the benefits of SNMP. During SNMP, lactate and potassium levels progressively declined while oxygen consumption remained stable, reflecting lactate washout accumulated during warm ischemia and controlled metabolic activity at subnormothermic temperatures. No lactate rebound was observed during the 4‐h SNMP period, consistent with reports in non‐human primate and rodent limb models. 31 , 36 , 37 , 38 In contrast, elevated potassium levels in the SCS group reflected the composition of the HTK solution and were partially washed out during reperfusion, with a possible contribution from cold‐induced tissue injury. Higher lactate dehydrogenase activity after reperfusion in SCS grafts indicated increased cytolysis compared with SNMP, in line with previous VCA models showing metabolic recovery after SNMP following prolonged warm ischemia. 25 Histological analysis showed preserved myometrial architecture in both groups, suggesting that uterine smooth muscle tolerates 4 h of preservation regardless of method. However, increased apoptosis was observed in the endometrium after SCS, indicating greater sensitivity of this tissue to cold ischemia. Ultrastructural analysis confirmed intact smooth muscle organization, endothelial structures, and preserved gap junctions in both groups, supporting the maintenance of tissue architecture. Functionally, SNMP‐preserved uteri demonstrated superior contractility compared with SCS controls. After standardized oxytocin stimulation during normothermic reperfusion, all SNMP grafts showed clear contractile responses, whereas most SCS grafts exhibited only pressure increases without visible contraction. These findings indicate that SNMP better preserves the ability of the myometrium to translate hormonal stimulation into coordinated contraction. These results are consistent with previous studies in other muscle‐containing grafts, including non‐human primate forelimbs and swine whole‐eye transplantation models, where SNMP preserved muscle integrity and restored contractile function after prolonged warm ischemia. 38 , 39 , 40 In addition to its preservation benefits, SNMP may facilitate objective graft evaluation prior to transplantation. Continuous assessment of inflow and outflow metabolic parameters during perfusion could help identify suboptimal grafts and guide intraoperative decision‐making. 41 This feature is particularly relevant in the context of deceased donation and donation after circulatory death UTx, where graft quality and ischemic exposure are more variable. Several limitations of this study should be acknowledged. The absence of in vivo transplantation precludes evaluation of long‐term graft survival, immune‐mediated injury, and reproductive outcomes. Although ex vivo normothermic reperfusion with whole blood was used to simulate early transplantation, the relatively short reperfusion period may not fully reflect IRI or inflammatory responses. In addition, maintaining stable normothermic conditions was technically challenging. Nevertheless, ex vivo NMP is widely accepted as a robust preliminary platform for evaluating preservation strategies and has been validated in multiple organs and VCA models. 35 , 42 , 43 The limited sample size also restricts statistical power. However, the use of a large‐animal model, standardized protocols, and relevant control groups strengthens the translational value of these findings. The anatomical differences between the porcine and human uterus should also be acknowledged. As a bicornuate with long uterine horns, the swine model may influence perfusion dynamics and limit direct application to clinical UTx. Although the sheep uterus may provide a more human‐like anatomical model for in vivo transplantation studies, the swine uterus remains a relevant and established model for the technical development and preliminary evaluation of SNMP in this study. Moreover, swine uteri have been widely used in previous ex vivo perfusion studies assessing uterine viability, metabolism, and contractile responsiveness. While the porcine model is well suited for MP studies, it may be less optimal for retransplantation because of vascular pedicle length, despite allowing microsurgical anastomoses. Alternative models, such as procurement with the aorta as described by Dion et al., 44 have demonstrated prolonged hypothermic perfusion with preserved vascularization and may complement future investigations. CONCLUSION In conclusion, this study demonstrated that a simplified SNMP protocol permitted uterus preservation better than SCS for a period of 4 h in a swine model, with confirmed tissue viability and functional parameters. Further studies should concentrate on implementing dynamic preservation in an allotransplantation model. AUTHOR CONTRIBUTIONS Lucile Cabanel, Haïzam Oubari, Yanis Berkane, Alexandre G. Lellouch, Curtis L. Cetrulo Jr., and Basak E. Uygun contributed to the study design. Lucile Cabanel, Haïzam Oubari, Yanis Berkane, Loïc Van Dieren, and Mark A. Randolph performed the research. Data collection was carried out by Lucile Cabanel, Haïzam Oubari, Yanis Berkane, and Bradley W. Ellis. Data analysis was conducted by Lucile Cabanel, Haïzam Oubari, Yanis Berkane, and Basak E. Uygun. Manuscript writing and proofreading involved Lucile Cabanel, Haïzam Oubari, Ludivine Dion, Olivier Morel, Carla Sousa, Yanis Berkane, Alexandre G. Lellouch, Basak E. Uygun, Curtis L. Cetrulo Jr., Korkut Uygun, and Mark A. Randolph. The study was supervised by Korkut Uygun, Curtis L. Cetrulo Jr., Alexandre G. Lellouch, and Basak E. Uygun. All authors have read and approved the manuscript. FUNDING INFORMATION This work was partially funded by the National Institute of Health under award No R01AR082825 (BEU), National Science Foundation No ECC‐1941543 and Shriners Children's 84308 (YB). HO and YB received funding from the Fondation des Gueules Cassées. Support from Société Française de Chirurgie Plastique, Reconstructrice et Esthétique (SOFCPRE, France), and CHU de Rennes (France) to YB is greatly acknowledged. BWE is supported by the National Institutes of Health under award No F32HL176334. Imaging was performed in the Microscopy Core of the Program in Membrane Biology, which is partially supported by a Center for the Study of Inflammatory Bowel Disease Grant DK043351 and a Boston Area Diabetes and Endocrinology Research Center (BADERC) Award DK135043. CONFLICT OF INTEREST STATEMENT All the authors have no conflicts of interest to declare. ETHICS STATEMENT All animals received humane care following the National Institute of Health Guide for the Care and Use of Laboratory Animals, and the protocols (2020 N000015 and 2024 N000128) were approved by the Massachusetts General Hospital Institutional Animal Care and Use Committee. Supporting information Figure S1: Postoperative picture of the uterus after procurement and heparin‐saline flush. A: corpus; B: uterine horns; C: cervix; black arrows: uterine arteries; white arrows: uterine veins. Figure S2: Lactate dehydrogenase levels in plasma (mean with standard error) at the end of reperfusion. Figure S3: Representative angiography of the uterine vascular tree after 4 h preservation and 2 h reperfusion. A. Before injection of contrast agent for the SCS group; B. After injection of contrast agent for the SCS group; C. Before injection of contrast agent for the SNMP group; D. After injection of contrast agent for the SNMP group. Figure S4: Representative histologic analysis of vessel sections from fresh uteri, before and after reperfusion (H&E and TUNEL staining) showed a good preservation of vessel architecture with little damage after reperfusion in both groups. Figure S5: Representative histologic analysis of vessel sections using CD31 immunohistochemistry staining showed a good preservation of vessel architecture after reperfusion with endothelial cells all along the vessel lumen in both groups. Table S1: Steen+ solution composition. Video S1: Representative oxytocin‐induced contraction (score 3/3) in an SNMP‐preserved graft. ACKNOWLEDGMENTS The first author would like to thank Thomas Grinda for his unfailing help. DATA AVAILABILITY STATEMENT The data that support the findings of this study are available from the corresponding author upon reasonable request. REFERENCES Associated Data Supplementary Materials Figure S1: Postoperative picture of the uterus after procurement and heparin‐saline flush. A: corpus; B: uterine horns; C: cervix; black arrows: uterine arteries; white arrows: uterine veins. Figure S2: Lactate dehydrogenase levels in plasma (mean with standard error) at the end of reperfusion. Figure S3: Representative angiography of the uterine vascular tree after 4 h preservation and 2 h reperfusion. A. Before injection of contrast agent for the SCS group; B. After injection of contrast agent for the SCS group; C. Before injection of contrast agent for the SNMP group; D. After injection of contrast agent for the SNMP group. Figure S4: Representative histologic analysis of vessel sections from fresh uteri, before and after reperfusion (H&E and TUNEL staining) showed a good preservation of vessel architecture with little damage after reperfusion in both groups. Figure S5: Representative histologic analysis of vessel sections using CD31 immunohistochemistry staining showed a good preservation of vessel architecture after reperfusion with endothelial cells all along the vessel lumen in both groups. Table S1: Steen+ solution composition. Video S1: Representative oxytocin‐induced contraction (score 3/3) in an SNMP‐preserved graft. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.