Pig heart transplanted into baboon without any ischemic time

Surgeons at Columbia University have transplanted a genetically engineered pig heart into a baboon without ever letting the organ stop beating or go without blood. The donor heart was kept warm and continuously perfused through every stage of the operation—explant, transport, and implantation—eliminating the ischemic periods that have been unavoidable in every previous cardiac xenotransplantation experiment.

The technique, described by Daniel H. Wolbrom and colleagues, addresses one of the most stubborn obstacles in xenotransplantation: perioperative xenograft dysfunction, or PCXD. This rapidly developing loss of cardiac function within the first 48 hours after implantation occurs independently of hyperacute rejection and has been observed in 40% to 60% of orthotopic cardiac xenotransplantation experiments in nonhuman primates. Researchers attribute PCXD largely to ischemia-reperfusion injury, the damage inflicted when an organ is deprived of blood and then reperfused, a process the authors argue is amplified in cross-species transplants because species-dependent molecular incompatibilities provoke a more profound immune response.

Commercial ex vivo perfusion devices can limit cold ischemia time by keeping donor hearts circulating in chilled solution, but they do not eliminate ischemia entirely. The organs still undergo an obligatory ischemic period during initial arrest with cardioplegia and explantation, and again during implantation. These devices also cannot support smaller pediatric-sized hearts, which shrinks the pool of potential recipients. The Columbia team’s approach departs from this paradigm by never arresting the heart at all.

In the procedure, the donor pig was placed on a normothermic coronary perfusion pump primed with the donor’s own heparinized blood, running on a Spectrum Quantum Charlotte frame with a D100 oxygenator. Cannulas were placed in the right atrium, left ventricular apex, and proximal ascending aorta, with the right atrial and left ventricular lines connected via a Y-connector. The total circuit prime was 135 milliliters, including albumin, sodium bicarbonate, heparin, and Plasmalyte A, with flow rates maintained between 60 and 100 milliliters per minute. Surgeons ligated the superior and inferior vena cavae, clamped the aorta, and explanted the heart while antegrade flow into the proximal aorta continued to provide coronary blood flow. The beating organ was then carried in a sterile surgical bucket to an adjoining operating room, where it was implanted orthotopically into the recipient.

The donor heart spent 34 minutes on the perfusion circuit and remained in sinus rhythm throughout. After implantation, the baboon was weaned off cardiopulmonary bypass in sinus rhythm without inotropes or vasopressors. Echocardiography 60 minutes after separation showed normal right ventricular systolic function, mildly decreased left ventricular systolic function with a shortening fraction of 24% to 25%, and mild valvular regurgitation. Invasive monitoring recorded a right atrial pressure of 6 mmHg and pulmonary artery pressure of 32/12 mmHg. The xenograft sustained life-sustaining cardiac output, with a mixed venous oxygen saturation of 77% and an arterial blood gas showing pH 7.48. Histology revealed no morphologic evidence of hyperacute antibody-mediated rejection, acute T cell-mediated rejection, ischemic injury, or edema.

The survival experiment used a juvenile GGTA1 KO piglet weighing 5.0 kilograms, negative for porcine cytomegalovirus, as donor for a 5.3-kilogram male baboon. The recipient was screened beforehand and selected for low cytotoxicity of preformed natural antibody against GGTA1 null pig cells, with anti-non-Gal IgG and IgM indices below prespecified thresholds. Induction immunosuppression combined anti-CD20 antibody, rabbit anti-thymocyte globulin, and methylprednisolone, followed by weekly anti-C5 antibody, daily mycophenolate mofetil, and a tapering schedule of anti-CD40 antibody. Rapamycin was planned from day 30 to restrain graft growth but was delayed to day 43 because of infectious concerns. The donor heart remained beating at normothermia for 41 minutes on the ex vivo circuit before implantation, and the recipient was weaned from cardiopulmonary bypass after a total run time of 2 hours and 25 minutes.

The early postoperative course was rocky but recoverable. The animal required dobutamine to maintain adequate blood pressure, and serum lactate peaked around 12 mmol/L, indicating hypoperfusion. Initial echocardiography showed a left ventricular ejection fraction of 15.8% with moderate mitral regurgitation and moderate-to-severe tricuspid regurgitation. Within 12 to 14 hours, inotropic support had been substantially weaned and ejection fraction had risen to 50%, with improvement in mitral regurgitation. The baboon was successfully extubated and returned to monitored housing. By postoperative day 8, ejection fraction had climbed to 55.9%, and studies on days 14 and 21 showed stable graft function. Right heart catheterizations on days 8, 14, 28, and 44 demonstrated stable filling pressures throughout.

Signs of trouble emerged later. An echocardiogram on postoperative day 29 showed the first dysfunction, with ejection fraction declining to 43.9%, and by day 42 it had fallen further to 38%. A mass appeared in the left ventricular apex on day 21, near the site where a cannula had been inserted during ex vivo perfusion; the authors hypothesize the cannula caused an injury there, though histological examination was inconclusive. Around four weeks after transplant, the white blood cell count rose and the animal appeared lethargic, prompting treatment for a suspected central line infection with antibiotics and an antifungal. Maintenance mycophenolate mofetil was held for 48 hours and then reduced from 110 mg/kg/day to 40 mg/kg/day for a week. The intraventricular septum, which had shown essentially no change in thickness for several weeks, increased from roughly 0.4 to 0.8 centimeters by day 42, which the authors suspect reflected antibody-mediated rejection in the setting of the lowered immunosuppression. The recipient died on postoperative day 55, likely from antibody-mediated rejection secondary to the unplanned weaning of immunosuppression for the suspected infection. Biopsies taken on days 8 and 14 had shown only mild edema and minimal interstitial hemorrhage without evidence of active rejection or thrombotic microangiopathy; by day 28, C4d staining was focally positive with one focus of mononuclear infiltrate, and by day 44 histological features of antibody-mediated rejection were noted.

They also flag an underappreciated variable: an undetermined volume of donor porcine whole blood was transferred with the heart into the primate recipient, potentially eliciting innate and adaptive immune responses. Although GGTA1 knockout eliminates the Gal antigen, other xenoantigens capable of triggering antibody-dependent cytotoxicity remain, and interspecies incompatibilities in complement and coagulation regulation may amplify immune activation. The immunologic consequences of transient exposure to GGTA1 null porcine blood in this setting remain undefined.

The team is currently conducting experiments in several consecutive nonhuman primate recipients to determine whether the nonischemic approach improves long-term survival and reduces primary xenograft dysfunction, with long-term results to be published separately. The technique also has implications beyond xenotransplantation: early reports in human-to-human heart transplant using a similar beating heart technique have shown good results, though long-term outcomes have yet to be reported. For now, the study stands as a proof of feasibility—demonstrating that a pig heart can be removed, transported, and implanted while beating at normal body temperature, with no ischemic interval whatsoever, in a large-animal cross-species model.

Subject of Research: Nonischemic, normothermic beating-heart transplantation of genetically engineered pig hearts into nonhuman primates

Article Title: Nonischemic and normothermic beating cardiac xenotransplantation

Article References: Wolbrom, D. H., Matsumura, N., Moroi, M. K., Chauhan, I., Fujiwara, S., Qudus, S., Brewer, M., Ekanayake-Alper, D., Hajosi, D., Schweitzer, L., Takeda, K., Levy, R. J., Juergensen, S. M., Barry, O., Sekulic, M., Sachs, D. H., Sykes, M., Nowak, G., & Goldstone, A. B. (2026). Nonischemic and normothermic beating cardiac xenotransplantation. Frontiers in Transplantation, 5, 1984456. https://doi.org/10.3389/frtra.2026.1984456

Image Credits: AI Generated

DOI: 10.3389/frtra.2026.1984456

Keywords: cardiac xenotransplantation, GGTA1 knockout pigs, ischemia-reperfusion injury, normothermic perfusion, nonhuman primate model, perioperative xenograft dysfunction, ex vivo heart perfusion, orthotopic heart transplant, antibody-mediated rejection, immunosuppression, organ shortage, beating heart transplant