Perfusion systems are devices that pump fluid through living tissue or an organ, either inside the body or outside it, to deliver oxygen, nutrients, and other essentials while removing waste. They range from the room-sized heart-lung machines used during open-heart surgery to miniaturized microfluidic chips that mimic blood flow for drug testing. Although the underlying principle is always the same — keep cells alive by maintaining circulation — the engineering, the stakes, and the clinical trade-offs vary dramatically depending on what is being perfused and why.
How the Heart-Lung Machine Started It All
The idea of replacing the heart and lungs with a machine during surgery traces back to 1931, when a young surgeon named John Gibbon watched a patient die after a pulmonary embolectomy and began imagining a device that could oxygenate blood and pump it back into the body. Working with his wife Mary from the mid-1930s onward, Gibbon spent years refining prototypes, first keeping cats alive on bypass, then securing support from IBM to build a more advanced version. On May 6, 1953, he used the machine to repair an atrial septal defect in an 18-year-old woman, marking the first successful open-heart surgery supported by extracorporeal circulation.1PubMed. John H. Gibbon, Jr. Part I. The development of the first successful heart-lung machine That single operation launched the field of cardiac surgery as we know it. Other pioneers, including Lillehei and Kirklin, soon used cardiopulmonary bypass to repair large series of congenital heart defects, and the technology spread rapidly.2PubMed. Origins and Evolution of Extracorporeal Circulation: JACC Historical Breakthroughs in Perspective
Cardiopulmonary Bypass and the Inflammatory Trade-Off
Modern cardiopulmonary bypass (CPB) works by draining venous blood from the patient, passing it through an oxygenator that adds oxygen and removes carbon dioxide, and then pumping it back into the arterial system. This lets surgeons stop the heart entirely to work on it. Two main pump designs handle the job: roller pumps, which squeeze blood through tubing mechanically, and centrifugal pumps, which spin blood outward using a rotating impeller. Centrifugal pumps tend to be gentler on blood cells. In pediatric cardiac surgery, a randomized trial found that centrifugal pumps produced lower levels of free hemoglobin (a marker of red-cell damage), preserved more platelets, and triggered less inflammation than roller pumps.3PubMed. Superiority of centrifugal pump over roller pump in paediatric cardiac surgery: prospective randomised trial Still, centrifugal pumps require a larger volume of fluid to fill (the “prime volume”), which can be a problem in small children.4PubMed Central. The Impact of Roller Pump vs. Centrifugal Pump on Homologous Blood Transfusion in Pediatric Cardiac Surgery
The biggest inherent drawback of CPB is that it provokes a whole-body inflammatory response. When blood contacts the artificial surfaces of the tubing and oxygenator, the immune system reacts as though it has encountered a foreign invader, activating complement proteins, releasing cytokines, and drawing white blood cells into an aggressive state.5PubMed Central. Lung inflammatory response syndrome after cardiac-operations and treatment of lornoxicam Other factors pile on: surgical wounds, temperature changes, and the restart of blood flow to tissues that were temporarily deprived of oxygen. In most patients this inflammation stays mild and resolves quickly after surgery. In its worst form, however, it can escalate into systemic inflammatory response syndrome and major organ dysfunction.6International Journal of Surgery. The systemic inflammatory response syndrome and cardiopulmonary bypass Decades of engineering improvements — biocompatible coatings on tubing, smaller circuits that reduce blood-surface contact, and refined temperature management — have gradually reduced but never fully eliminated this problem.
ECMO and Longer-Term Life Support
Extracorporeal membrane oxygenation (ECMO) extends the concept of cardiopulmonary bypass from hours to days or even weeks. Instead of supporting a patient through a single operation, ECMO sustains people whose lungs or heart have failed so severely that conventional treatment cannot keep them alive. There are two configurations. Venoarterial (VA) ECMO drains blood from a vein, oxygenates it, and returns it to an artery, thereby supporting both the heart and the lungs. Venovenous (VV) ECMO drains and returns blood entirely through the venous system, providing lung support alone. VV ECMO was developed partly to avoid carotid artery ligation (tying off a major neck artery), which VA ECMO traditionally required in newborns, and to reduce the risk of pumping air or debris into the arterial system.7PubMed Central. Venovenous perfusion in ECMO for newborn respiratory insufficiency. A clinical comparison with venoarterial perfusion.
Because ECMO runs for extended periods, managing blood clotting becomes a constant balancing act. Patients are usually given anticoagulants to prevent clots from forming inside the circuit, but anticoagulation also raises the risk of dangerous bleeding. A systematic review of patients who ran on ECMO without any anticoagulant found that circuit-related clotting occurred in about 13% of cases, while roughly a third experienced some form of bleeding — a rate that was actually comparable to patients receiving standard anticoagulation.8PubMed Central. Thrombosis and bleeding in extracorporeal membrane oxygenation (ECMO) without anticoagulation: a systematic review That finding has fueled interest in whether certain ECMO patients might safely forgo anticoagulants altogether, though more research is needed before it changes standard practice.
Machine Perfusion for Transplant Organs
For decades, the standard way to preserve a donated organ between removal and transplant was static cold storage: pack it in ice, cool it to near freezing, and rush it to the recipient. This works reasonably well, but cold storage has limits. Cells still sustain damage from oxygen deprivation, and surgeons have no way to test whether an organ is actually functioning before implanting it. Machine perfusion changes the equation by actively pumping preservation fluid through the organ’s blood vessels, either cold or at body temperature, to maintain cellular activity and reduce injury.
The evidence is strongest for kidneys. A Cochrane review of 16 trials found high-certainty evidence that hypothermic machine perfusion (HMP) — cold, pulsatile pumping — reduces the risk of delayed graft function (the need for dialysis in the first week after transplant) compared to static cold storage. The benefit held for kidneys from both standard and higher-risk donors.9PubMed Central. Machine perfusion preservation versus static cold storage for deceased donor kidney transplantation A separate meta-analysis confirmed the reduction in delayed graft function, though it found no significant difference in rates of primary non-function, acute rejection, or patient survival.10BJS. Systematic review and meta-analysis of hypothermic machine perfusion versus static cold storage of kidney allografts on transplant outcomes Not every trial has shown the same benefit: one randomized trial of kidneys from donation-after-circulatory-death donors found no difference in delayed graft function between machine perfusion and cold storage.11American Journal of Transplantation. Controlled Nonheartbeating Donor Kidney Transplantation: A Randomized Controlled Trial Comparing Cold Pulsatile Machine Perfusion with Static Cold Storage But on balance, the pooled evidence favors machine perfusion, and it has become routine at many transplant centers.
Normothermic Perfusion and Rescuing Discarded Organs
Where hypothermic perfusion slows metabolism to reduce damage, normothermic machine perfusion (NMP) takes the opposite approach: it keeps the organ warm, fed with oxygenated fluid, and metabolically active. This has a major advantage — it lets clinicians watch the organ work in real time. If a liver is producing bile and clearing lactate, it is probably going to function after transplant. If it is not, the team can decide against using it before putting a patient through surgery.
This viability-testing capability is particularly valuable for marginal organs that would otherwise be discarded. In one study of previously rejected donor livers, NMP enabled successful transplantation of 71% of them, with every patient and graft surviving at 90 days.12Nature Communications. Transplantation of discarded livers following viability testing with normothermic machine perfusion Another study focused specifically on fatty livers — organs with moderate-to-severe fat deposits that are frequently turned down — and found that about 60% could be successfully transplanted after end-ischemic NMP with good outcomes.13PubMed Central. Viability assessment and transplantation of fatty liver grafts using end‐ischemic normothermic machine perfusion Given the chronic shortage of transplant organs worldwide, the ability to rescue even a fraction of discarded organs could meaningfully expand the donor pool.
NMP has also transformed heart transplantation from donors who die after circulatory arrest (as opposed to brain death). These hearts experience a period without blood flow that damages the muscle, and they were previously considered unusable. With ex vivo normothermic organ perfusion, the heart can be resuscitated and assessed outside the body. The trade-off is that the organ endures multiple rounds of blood-flow deprivation and restoration, which may contribute to higher rates of post-transplant ECMO support in these recipients.14PubMed. Beating Heart Transplant Procedures Using Organs From Donors With Circulatory Death
Early work on normothermic liver perfusion demonstrated its feasibility using four discarded human livers perfused for six hours. Biochemical markers showed minimal injury: lactate levels fell from high values to normal within hours, bile production remained steady, and tissue examined under the microscope showed well-preserved structure with no additional damage.15American Journal of Transplantation. Normothermic Machine Perfusion of Successively Discarded Human Donor Livers The field has moved rapidly since then, and normothermic perfusion is now being used clinically, not just experimentally.
What Goes Inside the Circuit
The fluid pumped through a perfusion system — the perfusate — matters just as much as the hardware. For normothermic perfusion, where the organ is metabolically active and consuming oxygen, the perfusate needs an oxygen carrier. Historically, that meant adding packed red blood cells, which work well but create logistical headaches: they require blood typing, cold-chain management, and donor screening. Researchers have been testing hemoglobin-based oxygen carriers (HBOCs) — synthetic solutions containing hemoglobin molecules stripped from red blood cells — as an off-the-shelf alternative. In a feasibility study of human kidney perfusion, kidneys perfused with HBOCs performed comparably to those perfused with red blood cells in terms of vascular flow, oxygen consumption, and energy restoration. Interestingly, the kidneys receiving red blood cells actually showed higher lactic acid levels, and there was no sign of tissue damage from the synthetic hemoglobin.16PubMed Central. Synthetic hemoglobin-based oxygen carriers are an acceptable alternative for packed red blood cells in normothermic kidney perfusion If HBOCs prove reliable across larger studies, they could simplify the logistics of machine perfusion programs considerably, particularly at smaller transplant centers that may not have easy access to blood products at all hours.
The Cost Question
Machine perfusion devices are expensive, and running them requires trained perfusionists and consumable supplies. Whether that investment pays for itself depends heavily on the organ and the setting. For kidneys, the economic picture is favorable. A cost-effectiveness analysis based on a large preservation trial found that machine perfusion was cheaper than cold storage at one year post-transplant — roughly $92,500 per standard-criteria kidney versus $104,000 for cold-stored ones — because fewer recipients needed costly dialysis after surgery.17PubMed. The cost-effectiveness of organ preservation methods in renal transplantation: US projections based on the machine preservation trial A separate European analysis found that machine perfusion was both less expensive and associated with better graft survival in about 85% of simulated scenarios.18American Journal of Transplantation. Cost-Effectiveness of Hypothermic Machine Preservation Versus Static Cold Storage in Renal Transplantation
Lungs tell a different story. Portable ex vivo lung perfusion (EVLP) systems are resource-intensive devices that recondition marginal donor lungs outside the body. At a low-volume US transplant center, median one-year costs were substantially higher with EVLP than with cold storage — roughly $918,000 versus $516,000 — and each quality-adjusted life year gained was over 1.5 times more expensive.19PubMed. Cost effectiveness of commercial portable ex vivo lung perfusion at a low-volume US lung transplant center Volume matters here: high-volume centers can spread the fixed costs over more cases and recoup savings from rescued organs that would otherwise have been discarded. For smaller programs, the math does not yet work out.
Beyond the Operating Room
Perfusion technology has spread well beyond cardiac surgery and organ transplantation. In cancer treatment, isolated limb perfusion delivers high concentrations of chemotherapy drugs directly to a limb affected by soft tissue sarcoma or melanoma, bypassing the rest of the body. The limb’s blood supply is temporarily separated from the general circulation, allowing drug doses that would be toxic if given systemically. One center’s 24-year experience with hyperthermic isolated limb perfusion found the approach effective for limb-sparing treatment and durable local control of soft tissue sarcoma, whether the tumor was primary, recurrent, or metastatic.20International Journal of Hyperthermia. Hyperthermic isolated limb perfusion in locally advanced limb soft tissue sarcoma: A 24-year single-centre experience
In aortic arch surgery, where the brain’s blood supply must be interrupted, retrograde cerebral perfusion pumps cold oxygenated blood backward through the veins of the head. This provides some metabolic support to the brain during the period of deep hypothermic circulatory arrest, when the body is cooled to the point where the heart is stopped and the normal circulation is halted entirely.21European Journal of Cardio-Thoracic Surgery. Deep hypothermic systemic circulatory arrest and continuous retrograde cerebral perfusion for surgery of aortic arch aneurysm While the technique’s neuroprotective benefit has been debated, it remains one of several strategies surgeons use to protect the brain during complex aortic procedures.
Perfusion in Biomanufacturing
Perfusion is not exclusively a medical technology. In biopharmaceutical manufacturing, perfusion cell culture means continuously feeding fresh nutrients to cells growing in a bioreactor while removing spent medium and harvested product. This stands in contrast to fed-batch processes, where everything stays in the tank until the run ends. One key enabling technology is alternating tangential flow (ATF) filtration, which uses cyclic reverse-flow pulses to prevent filters from clogging.22PubMed. Understanding and modeling alternating tangential flow filtration for perfusion cell culture Using ATF or similar approaches, researchers have achieved cell densities above 100 million cells per milliliter — far higher than conventional methods — while maintaining antibody production at comparable levels.23PubMed Central. Very high density of Chinese hamster ovary cells in perfusion by alternating tangential flow or tangential flow filtration in WAVE Bioreactor™-part II: Applications for antibody production and cryopreservation Higher cell densities in smaller bioreactors translate to a smaller manufacturing footprint, which could eventually make biologics production cheaper and more accessible.
Microfluidic Perfusion and Organ-on-a-Chip
At the smallest scale, perfusion systems have been miniaturized into microfluidic chips that replicate blood-flow conditions inside tiny channels. These “organ-on-a-chip” devices culture living human cells under controlled fluid shear stress — the mechanical drag of flowing liquid against cell surfaces — which profoundly affects how cells behave. One research group built a microchip that applied physiologically relevant shear stress to endothelial cells (the type that line blood vessels) while testing a light-activated cancer therapy. Compared to static cell cultures with no flow, the shear-stress environment altered the release of signaling molecules and significantly changed how cells responded to the treatment.24PubMed. Replicating endothelial shear stress in organ-on-a-chip for predictive hypericin photodynamic efficiency The implication is that drug-testing systems without perfusion may miss effects that would appear in the body, where cells experience continuous flow. As these platforms mature, they could reduce the need for animal testing and improve predictions of how drugs will perform in humans.
Ethical Tensions Around Normothermic Regional Perfusion
One of the most contentious frontiers in perfusion technology is normothermic regional perfusion (NRP) in organ donation after circulatory death. After a donor’s heart stops and death is declared, NRP restores warm, oxygenated blood flow to the abdominal or thoracic organs to improve their condition before removal. This improves organ quality, but it raises uncomfortable questions. The most debated is whether restarting circulation effectively negates the prior declaration of death by circulatory criteria. A second concern is that NRP may itself cause neurologic death in the donor. A third worry centers on the possibility that oxygenated blood could reach the brain through collateral vessels, raising uncertainty about whether neurologic function has truly and permanently ceased.25PubMed Central. The ethics surrounding normothermic regional perfusion in donors following circulatory death
These are not fringe objections. A scoping review of the ethical literature found that the implications of NRP for death determination appeared in over three-quarters of the sources examined, while its compatibility with the “dead donor rule” — the principle that organ procurement must not cause the donor’s death — was discussed in 60%.26PubMed Central. Ethical Issues in Normothermic Regional Perfusion in Controlled Organ Donation After Determination of Death by Circulatory Criteria: A Scoping Review In practice, NRP protocols typically include measures to prevent brain reperfusion, such as clamping the vessels that supply the head. But the ethical debate continues to shape how and whether NRP is adopted across different countries and transplant networks.
Perfusing the Brain After Death
Perhaps the most provocative perfusion experiment to date involved not preserving organs for transplant but restoring function in dead brains. In a landmark 2019 study, researchers at Yale developed a pulsatile perfusion system and a specially designed hemoglobin-based, acellular perfusate to pump through intact pig brains four hours after the animals were slaughtered. The results were striking: the perfusion system restored microcirculation, reduced cell death, and revived molecular and cellular activity, including spontaneous synaptic activity and active metabolism. Crucially, there was no global electrical brain activity of the kind associated with consciousness — the brains were not “waking up” — but the degree of cellular recovery far exceeded what anyone had thought possible hours after death.27PubMed Central. Restoration of brain circulation and cellular functions hours postmortem
The experiment did not aim to bring brains back to life, and the perfusate was designed to include chemicals that would block large-scale neural network activity. Its significance lies in demonstrating that the boundary between recoverable and permanently damaged tissue is far less fixed than previously assumed. For neuroscience, the technology offers a way to study intact brain circuits outside the body. For medicine, it raises the distant possibility of interventions to rescue brain tissue after prolonged oxygen deprivation — stroke, cardiac arrest, drowning. And for ethics, it adds another layer of complexity to questions about how we define death and when biological processes become truly irreversible.