Effective preservation is a requirement of techniques which involve transfer of viable organs and tissues. Graft damage must be minimized while the organ is still within the donor in preparation for graft procurement, during graft removal, during storage and transport of the graft, during the transplantation operation, and after reimplantation or revascularization. Some tissues, such as blood and skin, can be stored for several days or weeks. Vascularized autografts are usually reimplanted immediately, or within a few hours.

Allografts pose another problem. Not only must viability be maintained, but rejection of the grafts must be prevented by optimizing the match between donor and recipient tissue and by immunotherapy. These two objectives—preserving viability and preventing rejection—are closely linked. Optimal matching of antigens of graft and recipient requires a storage period of 24 h or more. Until recently storage for this length of time was only practicable for kidneys, but as prolonged storage becomes available for liver, pancreas, and heart, these grafts may also be allocated on the basis of improved tissue matching. Retrospective analysis has indicated that matching would improve the efficacy of cardiac transplants but this would require prolongation of the storage period beyond the current safe period of 6 h.

The process of clinical preservation usually begins with identification of the brain-dead, heart-beating cadaver donor. Effective preservation allows time for confirmatory tests of brain death, organization of operating teams for organ removal, typing and cross-matching of donor tissues against a pool of waiting recipients, excluding associated transplantable diseases such as infection or neoplasm in the donor, selecting and locating recipients and arranging their admission and preparation for surgery, arranging recipient operating teams, and time to transport organs over long distances, even between continents.

Good preservation also provides a window of diagnostic and therapeutic opportunity during the safe period of extracorporal storage. The time gained should be available for functional assessment of the graft to confirm its viability, and to predict early function after reimplantation. Good early function is obligatory in the case of heart and liver grafts. Immediate function of kidney and pancreas grafts, although not obligatory, is highly desirable and facilitates recipient management.

Effective immune manipulation of the recipient or of the graft itself during the period of storage is another (as yet largely unrealized) objective. Treatment of the stored organ to deplete it of antigen presenting cells, or other treatments to modify parenchymal and endothelial cell immunogenicity, could dramatically improve transplantation results.

Attempts to preserve human tissues and organs from putrefaction and decay after death began in antiquity with the embalming and mummifying techniques which were developed to a fine art by the Egyptian dynasties.

In the 1930s the classical experiments of Carrel and Lindbergh (Fig. 1) 679 applied perfusion techniques to preservation of organs for transplantation. They established ground rules for organ preservation by continuous ex-vivo perfusion—expert technology, perfect asepsis, and controlled biological conditions.

The Spanish Civil War marked the advent of blood banks and clinical application of tissue storage techniques. Cold storage was used to diminish metabolic demand. Fortunately blood, the first widely preserved and transplanted biological substance, lent itself well to extended storage. Refrigerated storage was possible for 3 weeks.

The discovery of cryoprotectants by Polge and coworkers in 1949 ushered in a major extension of preservation times for a variety of simple cells and tissues. Blood cells and gametes (even embryos) could be stored after freezing. Weeks or months later, they could be thawed and successfully reimplanted as autografts or allografts. Such freezing was only successful for isolated cells or undifferentiated multicellular embryos: complex and heterogeneous organs were highly sensitive to freezing damage. The biophysical problems of freezing large organs were subsequently defined by Pegg and other workers. To date no consistent success has been obtained with the use of freezing to preserve organs such as kidneys, liver, or hearts. Concepts of frozen humans waiting revival at an appropriate future time on another planet or in an after-life remain in the realms of science fiction.

Hypothermic organ preservation, stopping short of freezing, has given more promising results for whole organs. Most early work centred on the kidney. Simple cooling and ice storage gave reliable protection for several hours only. In the 1960s Pegg and Calne showed that preservation could be improved by a cold intravascular flush. Initially the blood within the organ was replaced with fresh blood, plasma, or extracellular-like flushing solutions containing colloid. In 1969 Collins demonstrated the clear superiority of an ‘intracellular’ electrolyte flushing solution, high in potassium, magnesium, and phosphate, low in sodium and chloride, and without colloid. Kidney storage for 24 h became clinically practical, provided that the organ did not suffer too much damage due to periods of warm ischaemia prior to cooling. In the early 1970s Belzer et al. extended Carrel's work on recirculating machine perfusion. Machine storage extended the time during which kidneys could be reliably preserved from one day to 3 days. However, simple hypothermic storage (after flushing) and continuous machine perfusion gave equivalent results for 24-h preservation—particularly if warm ischaemia was avoided. It also became clear that flushing solutions did not need to mimic intracellular composition to be effective: other solutions, even simpler in composition, based on citrate and on sucrose were shown to be as effective as Collins' solution (Table 1) 227. Clinical kidney preservation and transport could rely mainly on simple hypothermic storage after preliminary flushing with these solutions (Fig. 2) 680; compact cold perfusion machines using modifications of plasma were also available for more extended storage times. Belzer et al. gave a further major stimulus to preservation research in 1987, demonstrating that a complex multicomponent solution (University of Wisconsin (UW) solution) gave significantly enhanced preservation of several organs (particularly pancreas and liver). With a new colloid—hydroxyethyl starch, the UW solution was suitable for both simple flushing and recirculatory machine perfusion. Current research in many centres is evaluating the efficacy of the numerous components of UW solution, and attempting to improve it further. Modifications of UW solution have been successful in extending the time of preservation of pancreas, liver, kidney, and heart.

Cells of normally functioning organs derive their energy from the oxidation of substrates obtained from the circulating blood (glucose, fatty acids, amino acids, or ketones); in the case of liver and muscle energy is obtained from breakdown of endogenous glycogen. Energy is stored within cells as phosphate bonds of adenosine triphosphate (ATP), creatine phosphate, and other nucleotides. Cellular composition is maintained in homeostasis by numerous enzymic reactions acting in concert under the directions of hormones or key compounds such as cyclic adenosine monophosphate.

Ischaemia cuts off nutrients and oxygen supply to the organ. Continuing metabolic activity of the organ's parenchymal and other cells causes a cascade of events leading to irreversible cell damage and death. Metabolism becomes anaerobic; glycolysis causes depletion of high energy phosphate compounds. Degradation of ATP increases cellular levels of adenosine, inosine, and hypoxanthine. Depletion of the cell's energy stores inactivates Na⫀- K⫀-ATPase, the enzyme system controlling the sodium pump of the cell membrane. As fuel reserves disappear, sodium and chloride, freely permeable electrolytes which are normally actively excluded from the cell, diffuse into the cell down concentration gradients. As the osmotic force of non-permeable cellular proteins and anions is no longer balanced by the extrusion of sodium, water floods the increasingly swollen cell. Mitochondrial respiration is inhibited, and calcium enters the cytosol and mitochondria. Anaerobic metabolism temporarily uses glucose stores to generate ATP, but lactic acid is also produced, leading to progressive intracellular acidosis which activates lysosomal lytic enzymes leading to autolysis.

Most organs and cells can tolerate ischaemic hypoxia for 30 to 60 min without permanent damage. Parenchymal cells of most transplantable organs are generally similar in their tolerance to ischaemia. Rapidly metabolizing tissues are less tolerant: the heart is particularly vulnerable because it continues to beat until all energy reserves are depleted. Most organs are irreversibly damaged by 90 to 120 min of ischaemia at body temperatures. Vascular damage occurs along with the parenchymal effects: the cells of the vascular endothelial lining bear the brunt of this injury. These effects are slowed, but not reversed, by cooling (Fig. 3) 681.

Reperfusion injury is an added hazard which contributes to irreversible damage. Accumulation of metabolic end products such as hypoxanthine, under anaerobic conditions, can set the stage for reperfusion injury. When blood flow is restored, oxygen influx leads to the formation of toxic compounds such as hydroxen peroxide, and superoxide and hydroxyl radicals. These active free radicals of oxygen produce further cellular, membrane, and microvascular injury. Under normal circumstances these harmful radicals are short lived, as they are rapidly cleared by endogenous scavenging mechanisms. Ischaemia depletes these endogenous mediators.

Simple cooling markedly enhances ischaemic tolerance. All enzymic activity is temperature dependent—cooling diminishes metabolic activity, curtails oxygen demand, and slows degradation of energy stores. Hypothermia does not stop metabolism, but merely slows the metabolic clock and lessens the speed at which deterioration occurs. Cooling from 37°C (body temperature) to 0°C (storage temperature) extends the tolerance of most organs to ischaemia from between 1 and 2 h to about 12 h. Unfortunately cold does not slow all biological functions uniformly but causes discordance in a variety of metabolic processes which occur in concert at 37°C. Transmembrane passive diffusion of ions is not appreciably affected by hypothermia, while active transport mechanisms, such as those governed by Na⫀, K⫀-ATPase and Mg²⫀ Ca²⫀-ATPase are inhibited below 10°C. Hypothermia alone cannot prevent cell swelling during storage (Fig. 3) 681.

A major requirement for a cold flushing solution is therefore that it includes an impermeant to provide osmotic force to oppose cellular oedema (Fig. 5) 683. Large anions such as lactobionate (molecular weight 358 Da), or non-electrolytes such as the saccharides raffinose (molecular weight 505 Da) or sucrose (molecular weight 342 Da), or chelates of citrate and magnesium (molecular weight approximately 1000 Da), can achieve this. Glucose (molecular weight 180 Da) can permeate the cell slowly and can stimulate an undesirable production of lactic acid and hydrogen ions by anaerobic glycolysis. Glucose in flushing solutions can be replaced by impermeant sucrose with benefit—especially for liver and pancreatic grafting, where cell membranes are even more permeable to glucose. An effective buffer (phosphate, citrate, or histidine) to counter intracellular acidosis is a second major requirement. The importance of these two mechanisms is indicated by the fact that a solution consisting solely of sucrose with an added phosphate buffer (PBS) is remarkably effective in kidney preservation—almost as effective as the very much more complex UW solution.

The electrolyte composition of flushing solutions can vary widely. The freely diffusible anion chloride is preferably replaced by an impermeant anion (lactobionate, gluconate, or chelated citrate). Flushing solutions often have high potassium (100–130 mmol/l) and low sodium (10–30 mmol/l) concentrations. Since the high potassium concentration is cardioplegic, these solutions cannot be used for early systemic intravenous use, and a systemic leak before the time of organ flushing and retrieval is highly dangerous. A high potassium concentration is also vasoconstrictive, slowing flushing rates and the rate of organ cooling. Solutions with the sodium/potassium ratio reversed (Na 130 mmol/l, K 10–30 mmol/l), have been shown to be virtually as effective, provided suitable impermeants and buffers are present in the solution, and chloride concentration is kept low.

Magnesium has proved to be a useful additive in many solutions since it forms chelates with lactobionate and citrate which cannot pass through membranes. By contrast, calcium levels are kept low, or even excluded since cell damage is associated with calcium influx. Calcium (and magnesium) can precipitate in unstable solutions. Calcium is strongly chelated by lactobionate (and citrate), which may account partly for the usefulness of the latter substances in flushing solutions. This could be deleterious in heart preservation, contributing to harmful influx of calcium on reflow.

Preservation by simple hypothermic storage is limited by ultimate exhaustion of nutrients and accumulation of waste products. Preservation of kidneys can be extended to 3 days by simple storage using Collins, Citrate, PBS, and UW solutions. Preservation of liver and pancreas is more difficult, and was not consistently extended to 24 h until the advent of UW solution.

The multiple factors influencing the effectiveness of preservation were analysed by Southard et al., who produced a solution (UW) containing impermeants (raffinose, lactobionate), buffers (phosphate), free radical inhibitors and scavengers (glutathione, allopurinol), energy precursors (adenosine), vasoactive agents and hormones (steroids, insulin), and a colloid (hydroxyethyl starch). This complex solution was markedly more effective than other solutions (Collins', citrate, phosphate buffered sucrose) used for liver and pancreas preservation, and also for kidney and heart grafts. It was suitable both for cold-flush storage, and for continuous perfusion. Not all its components are equally important: hydroxyethyl starch can safely be omitted for cold flushing and simple hypothermic storage. High potassium levels are undesirable on several grounds, and reversing Na:K ratios does not significantly lower efficacy. Other additives, such as the buffer histidine, can improve preservation; the polysaccharide raffinose can be replaced by sucrose. Lactobionate is probably the most effective and most important component of UW solution. Gluconate (molecular weight 195 Da) appears slightly less effective as an alternative impermeant. Other adjuvants shown to be helpful include raffinose, glutathione, allopurinol, and adenosine. Steroids, insulin, and hydroxyethyl starch (HES) are of minor significance. Modified HES-free UW-derived solutions based on high concentrations of sodium lactobionate can give results similar or equal to the original UW solution.

Extension of experimental preservation by simple cold storage after flushing with modified UW solutions has been achieved for kidneys to 5 days, pancreas to 48 h, liver to 36 to 48 h, and for heart and lung towards 12 h. Recent clinical trials have confirmed the effectiveness of lactobionate-based UW solutions in preservation of kidney, liver, and pancreas; improved heart preservation for transplantation with UW flushing and storage has been demonstrated over standard cardioplegic solutions. Solutions with high concentrations of histidine, tryptophane, and &agr;-ketoglutarate (Bretschneider-HTK) have also given improved organ preservation.

Other agents shown to be helpful additives to flushing solutions (and which hold promise for further extension of preservation times) include other energy sources and buffers, calcium channel blockers (verapamil, diltiazem, trifluoperazine), and stable prostacyclin analogues.

Continuous cold perfusion by machine combines the benefits of hypothermia with continuous buffering, continuous washout of accumulating toxic metabolites, and continuous provision of oxygen and nutrients. Oxygen is more soluble at lower temperatures, and the reduced energy requirement of hypothermic organs can be provided by acellular perfusates. For all organs studied, viability has been maintained longest by continuous machine perfusion. A heat exchanger is needed to maintain temperature at 4 to 9°C, an atraumatic pump circulates the solution, and oxygenation is achieved either by simple surface diffusion or using a membrane oxygenator. An oxygen/carbon dioxide gas mixture, and an effective buffer are required to maintain a constant pH. A bubble trap and organ chamber complete the circuit (Fig. 6) 684. Materials coming in contact with the fluid must be sterile, non-toxic plastics or metals. The perfusate must contain colloid to maintain intravascular volume and prevent development of an ‘exploded’ extracellular fluid space. Cryoprecipitated plasma or albumin was initially used, but albumin gradually leaks into the interstitium and leads to weight gain, so this has been replaced by other oncotic colloids (gelatin, dextrans, hydroxyethyl starch). Metabolic fuels improve long-term preservation—glucose has been the substrate most commonly used, while adenosine, other precursors, or ATP itself have been used as additional sources of energy. Organ reperfusion requires rapid renewal of membrane pump activity, and quick regeneration of cellular fuels. Addition of other substrates more specific to individual organs can also improve function after storage (arachidonic acid, essential amino acids, fatty acids). Fuel additives become increasingly important as preservation times are prolonged beyond 24 h.

Hypothermic cell swelling due to membrane pump inhibition still occurs with cold perfusion below 15°C unless chloride is replaced by impermeants.

Continuous perfusion becomes limited after about 5 days. By this stage considerable damage has usually occurred to the vulnerable vascular endothelium, leading to irreversible cellular damage on reperfusion. Research continues into ways of further prolonging safe preservation, and of protecting the microcirculation during perfusion.

Organ perfusion at normal temperature can mimic normal function, but normothermic perfusion as a means of storage has not been possible for prolonged periods without seriously damaging the organ. Kootstra and his colleagues demonstrated that intermittent brief normothermic blood perfusion during cold storage improved preservation of both ice-stored and machine-perfused kidneys. The mechanisms of action are uncertain, but may relate to replenishing cellular energy stores and protecting the vasculature. The technique has not yet found clinical application.

Freezing is normally lethal to cells. Red cells, lymphocytes, spermatozoa, fertilized ova, embryos, and pancreatic islet cells can all be frozen and thawed satisfactorily in the presence of cryoprotectants. Simple tissues such as skin and cornea can also be preserved by freezing. Unfortunately freezing of organs, so attractive in concept in approaching true suspended animation and indefinite storage, has not proved feasible to date. The low surface area to volume ratio of organs compared to cells makes effective heat exchange a major problem. Cryoprotectants such as glycerol and dimethyl sulphoxide need to be present in high concentrations for adequate cryoprotection and this causes severe organ toxicity and vascular damage. Extracellular ice formation is innocuous when freezing cell suspensions, but can disrupt and severely damage the morphology of whole organs. The vascular system is especially vulnerable, and the attachment of vascular endothelial basement membrane is disrupted by freezing. Total vitrification of tissues is an alternative approach, but is also unsuccessful for whole organs. The tissue is first perfused with high concentrations of several cryoprotectants, so that on cooling neither intracellular nor extracellular freezing occurs. The solution vitrifies (i.e. solidifies into a glass state) at temperatures of approximately −120°C.

The function of the organ must be protected before and during procurement from the donor. Living donor grafts, such as kidney, are protected by maintaining blood and interstitial fluid volumes, by inducing a diuresis with mannitol during the operation, and by avoiding vascular trauma and vasospasm during mobilization and removal of the organ. Warm ischaemia is kept to a minimum.

Management of the brain-dead heart-beating organ donor follows similar protocols. Blood volume needs to be carefully maintained by transfusion and, when necessary, by central venous pressure monitoring. Diabetes insipidus is common in brain-dead patients; large volumes of fluid and careful monitoring are required. Over-transfusion and pulmonary oedema must be avoided in patients serving as heart and heart–lung donors. Renal support is given using dopamine (2–5 mg/kg.min). Systemic acidosis must be corrected. Mannitol 25 g, steroids 1 g, chlorpromazine 500 mg, and heparin 10 000 U are given just prior to in-situ aortic flush cooling. Nifedipine should be added if adrenaline or noradrenaline has been given to the donor. Expeditious and skilful organ removal from a well prepared, well hydrated cadaver donor with minimal warm ischaemia is the best guarantee of immediate organ function after grafting. The state of the organ prior to its removal in such instances becomes the best guide to subsequent early function.

In countries where kidney, heart, lung, liver, and pancreas transplantation is available up to 80 per cent of all cadaveric donors serve as donors of several organs. Criteria for the acceptance of donors for heart, lung, liver, and pancreas are more restricted than those for kidney transplantation. Transplanted hearts and livers must function immediately to support life, whereas kidney recipients can be supported by dialysis while temporary tubular necrosis recovers. An example of criteria required for kidney, liver, pancreas, and heart grafts is set out in Table 3 229.

General criteria require a donor to be free from cancer (except for treated skin cancers and central nervous system malignancies). Systemic embolization of primary brain tumours can occur after ventriculoatrial shunting; such patients should not be used as donors. Donors should also be free from systemic sepsis which is particularly likely to occur in those given intensive care for more than 7 days, and negative for hepatitis B surface antigen (HBsAg), hepatitis C serology, and human immunodeficiency virus (HIV) antibodies. Any patient with a known or suspected history of homosexual activity or intravenous drug usage should be excluded—the patient may be in the latent period before HIV antibodies develop.

A full postmortem examination should be performed following cadaveric organ donation. Occasionally unsuspected disease in the donor such as tuberculosis indicates the need for a period of drug prophylaxis in the recipient. Occult cancers may also be identified.

Techniques of multiple organ retrieval are outlined elsewhere. The donor organs are skeletonized. All organs require induction of rapid hypothermia to restrict initial warm ischaemic damage. The heart is stopped by infusion of a cold cardioplegic solution and removed first; rapid cooling of other organs is initiated by in-situ cold flushing via the aorta and portal vein. Liver, pancreas, then kidneys, are removed in sequence during in-situ flushing. Once the organs have been cooled and removed, the dissection is completed in a bath of ice slush when the vascular pedicles of each organ are prepared for transplantation. Final washout with 1 to 2 l of the organ preserving solution (UW, Citrate, EuroCollins, PBS, HTK) is performed at this stage until the effluent is macroscopically clear and the organ uniformly pallid.

Organ transplantation for end-stage disease is now well established worldwide in both developed and developing countries. Shortage of cadaver organs for transplantation remains a perennial problem, and is influenced by legal, ethical, cultural, and religious backgrounds of individual countries. The supply of organs from heart-beating brain-dead cadaver donors is inadequate to meet total demand. In developed countries the death rate from motor accidents has been lowered by public educational programmes, compulsory seat belt legislation, and other safety measures. Control of mortality from road crash epidemics has been an immensely impressive public health achievement; diminished numbers of brain-dead cadaver donors following road injury is an inevitable corollary.

Interest has thus rekindled in retrieval of organs from non-heart-beating cadavers. Organs from this source are unsuitable for liver, heart, or lung transplantation, but can be appropriate for kidney transplantation. Organ retrieval is facilitated by legislation governing early removal of organs after death, by public awareness of the problems, and by public support for organ transplantation. Kootstra and others in Europe and in the United States have shown that efficient organization in hospital emergency departments and in operating theatres can gain a significant number of adequately preserved kidneys from cadaver donors presenting with irreversible cardiac arrest. Special techniques include the insertion of aortic and caval balloon cannulae after groin cut-down, followed by rapid aortic delivery of large volumes of cold flushing solutions to provide in-situ cold perfusion of kidneys while preparations are being made for operative retrieval. Kidneys from these sources have provided very acceptable results; the frequency of delayed early function and of ultimately non-viable kidneys is higher than from heart-beating cadavers, but this additional source of cadaver organs has proved very valuable.

Most organs are subsequently immersed in cold preserving solution and stored at 0 to 4°C in a refrigerated container until reimplantation.

Kidneys are frequently transplanted within 24 h. This period can be extended safely for periods up to 36 to 48 h or longer, if such time is required to find and prepare an optimal recipient and to transport the refrigerated organ long distances. Liver and pancreas preservation with UW solution storage is effective for 12 to 24 h. Heart and heart–lung transplantation with current preservation solutions is restricted to a storage period of 5 to 6 h—increasing use of UW-derived solutions is likely to extend this time.

Organs are perfused continuously with a recirculating perfusate at 4 to 6°C. This technique is used less commonly now because of the increased effectiveness of preserving solutions for simple storage.

While management of the donor is of great importance to subsequent graft function, factors operating in the recipient during operation and after reperfusion may be equally significant. Failure of blood flow to return uniformly to all portions of previously ischaemic tissues (no reflow phenomenon) is a recognized sequel of extended ischaemic damage to organs in situ, to autografts, and to allografts. This phenomenon has been described following ischaemia of kidney, heart, muscle, brain, and other tissues. In recent years much attention has therefore focused on potential damage suffered by the graft during the implantation operation, and in the early period after revascularization.

Delayed restoration of blood flow to the renal cortex after reperfusion can contribute significantly to delayed graft function. Reperfusion of the stored kidney, rather than marking the welcome end of its ischaemic insult, may exacerbate the effects of ischaemia and lead to further cellular damage and early acute tubular necrosis.

The severity of reperfusion injury is dependent on the events of retrieval and storage. Injury can follow prolonged warm ischaemia, or extended cold preservation by either static or perfusion storage. The pathophysiology relates to continuing effects or oligaemic (hypoxic) hypothermic events during organ retrieval and storage under the fresh stimulus of return of oxygenated blood flow (Fig. 4) 682: pretreating the recipient, as well as the donor, can obviate some of these effects. Reperfusion injury can be exacerbated by gradual rewarming of the organ during reimplantation. Grafts should be kept cold with moistened cold packs during the recipient operation, and vascular anastomoses performed expeditiously. Addition of interposition arterial or venous jump grafts or other vascular repairs should be done as bench surgery on the preserved chilled organ. Technical failures of vascular anastomoses, requiring reclamping and revision, are potent contributors to reperfusion injury.

Simple hypothermic storage employs the following flush solutions: firstly, Collins' solution, with high concentrations of potassium, magnesium, phosphate, sulphate, and glucose. This extended preservation to 48 h, but precipitation of magnesium phosphate was a major problem. Subsequently, magnesium sulphate was omitted (EuroCollins solution) with no deleterious effect. Replacement of glucose by mannitol, sucrose, or raffinose improves function experimentally and in humans.

Citrate-based solutions, containing high concentrations of potassium, magnesium, citrate, sulphate, and mannitol were originally devised to overcome the limitations of Collins' solution. Both hypertonic and isotonic solutions are stable and provide successful clinical preservation for 48 h or more. High magnesium concentration is essential; magnesium citrate chelates provide the critical semipermeant component.

Recently, two other solutions have been shown to provide better renal preservation than the above which are nonetheless still widely used. These are phosphate buffered sucrose (PBS) and UW solution. The former is an isotonic solution containing only a phosphate buffer and sucrose, an impermeant solute. Clinically and experimentally this solution is highly effective. Preservation at 72 h was better with phosphate buffered sucrose than with Collins' or Citrate solutions.

UW solution was developed initially for pancreatic preservation. Subsequently, it was found to extend and improve preservation of the liver profoundly and became the preferred solution for multiorgan harvest, despite its higher cost. UW solution is highly effective in preserving dog kidneys for 72 h and rat kidneys for 48 h. Modifications of the solution, omitting the colloid (hydroxyethyl starch), are equally effective. High potassium concentration is not essential, and lactobionate can be replaced by gluconate. Clinically, a recent European Multicentre Trial indicated that kidneys preserved in UW solution produced a more rapid reduction of serum creatinine, higher creatinine clearance rate, and less dialysis when compared to EuroCollins (median preservation time 24 h, maximum 48 h). An additional advantage of use of UW solution was that sharing kidneys between centres to improve matching improved graft survival without any functional detriment for up to 48 h of storage. Other trials showed UW without HES to be equally beneficial.

Kidneys which have undergone more prolonged storage often exhibit poor early function for several days. Further extension of preservation requires additional resuscitative technology. Perfusion storage may resuscitate kidneys which have been subjected to adverse conditions before harvest or during storage. This could involve intermittent warm reperfusion with oxygen, nutrient amino acids, fatty acids, cofactors, and hormones. Normothermic reperfusion would be easier to apply clinically if a reliable synthetic perfusate could be developed.

Liver transplantation has been an established and effective treatment modality for end-stage liver disease since the early 1980s, but until recently livers were only able to be stored reliably for 12 h in Collins' or Citrate solutions. Preservation of the liver with EuroCollins or Citrate is less effective than preservation of the kidney for the hepatocyte is more permeable to glucose and mannitol, leading to reduced osmotic control and increased acidosis due to anaerobic glycolysis (EuroCollins). Pretreatment of the donor with chlorpromazine, or addition of chlorpromazine, diltiazem, or a stable prostacyclin analogue to the flush solution improves preservation.

A major advance in liver preservation occurred with introduction of UW solution. This solution provided effective preservation in the dog (48 h) and rat (30 h). Clinical preservation times have been safely extended to 24 h allowing procurement of liver grafts from distant cities, ample time for histopathological examination of the graft, more perfect recipient hepatectomy, and bench surgery of the graft to tailor adult grafts to fit one or more child recipients. The length of the preservation period within the range 4 to 24 h has not affected graft outcome in UW-preserved liver grafts. In contrast, the ischaemic period significantly affects EuroCollins livers preserved for this length of time.

Experimental studies have indicated that the essential ingredients of UW solution are the impermeant anion lactobionate, with additional osmolality provided by raffinose. Adenosine, allopurinol, and glutathione seem also beneficial. Omission of hydroxyethyl starch from UW solution is not detrimental to dog, rat, or human liver grafts. High potassium content has been shown to be unnecessary in 48-h preservation of the dog liver. Other buffers such as histidine can replace phosphate.

The hepatocyte is relatively insensitive to cold ischaemia; the primary damage is to the microvasculature and to the endothelial cells lining the sinusoid. Experimental studies indicate that although parenchymal cells remain viable during preservation for 48 h, a high proportion of endothelial cells are non-viable and this contributes to poor early function and mortality. UW solution helps to prevent cold-induced microcirculatory injury, but the mechanism of this protection is still speculative.

Prolonged preservation requires further refinement of solutions for static preservation, or development of perfusion preservation. Dog livers have been successfully stored for 72 h using continuous perfusion with modified UW solution: gluconate replaced lactobionate in this solution and calcium, adenine, glucose, and ribose were included.

Assessment of graft quality prior to implantation by measurement of ATP content, pH, energy charge, or morphology is not necessarily reliable; many of the detrimental changes occur after reperfusion. The volume and quality of bile flow, restoration of clotting, absence of lactate acidosis, blood amino acid clearance, and plasma concentration of bilirubin and aspartate aminotransferase can distinguish which grafts will recover.

During the recipient operation, it is advisable to replace the preservation fluid within the stored liver before releasing the suprahepatic caval clamp, by reflushing with balanced electrolyte solution, blood, or plasma, otherwise cardiac arrhythmias or arrest can occur due to the high concentrations of potassium, hydrogen ions, or pharmacological additives such as adenosine released into the circulation from the revascularized liver. Some centres use a warm rinse before release of the upper clamp; this avoids potential inhibition of cardiac activity from a large bolus of cold fluid.

Initial methods for pancreas preservation were based on those developed for kidney. The pancreas is more vulnerable to mechanical damage during retrieval, but is no more sensitive to either warm ischaemia or cold storage. The problems encountered in pancreatic grafting relate to its low circulatory flow rate and to vascular damage on storage and reperfusion. Thrombosis of vascular anastomses is more likely in organs subjected to poor preservation.

In early clinical studies, EuroCollins solution was successful, albeit for short periods of ischaemia: only about 10 per cent of grafts were preserved for longer than 12 h. Experimental preservation of the canine pancreas has been successful for 72 h (UW), and rat pancreas has been preserved for 48 h (UW, hydroxyethyl starch-free UW, and Citrate). Preservation times have been extended clinically using UW solution; this has markedly facilitated organization of transplant programmes. Clinically, graft survival has not been affected by extending preservation times to 24 h, although grafts stored for over 30 h show somewhat decreased graft survival. Retrospective analysis of transplant registry results showed that patients with better HLA-DR matching had significantly better graft survival. Prolonged effective preservation would allow allocation of grafts after matching.

Perfusion storage is inherently more difficult in the pancreas because of its low circulatory flow rate. Early studies using machinery designed for kidney perfusion were unsuccessful. Perfusion storage at low flow rates with UW-based solutions may in the future extend and improve pancreatic storage.

The interdependence of exocrine and endocrine cells has not been assessed. Ablation of the exocrine cells by polymer injection of the pancreatic duct does not appear to have any detrimental effect on 1-year graft survival, but long-term endocrine function may be compromised. Other transplant techniques involve reconstitution of pancreatic drainage via the intestine or the urinary bladder. Effective preservation of acinar cells helps minimize pancreatitis: elevated serum amylase is a common occurrence, and the level relates to preservation time. Acinar cell preservation provides a reliable and sensitive test of rejection. Monitoring of urinary amylase in bladder-drained pancreas transplants has become an accepted early indicator of rejection: falling urinary amylase indicates a need for more vigorous immunosuppression.

Heart preservation has employed simple static cold storage. As 85 per cent of the energy consumed by the heart fuels contraction of the myofibrils, cardioplegic arrest has been an essential feature of heart preservation. The development of heart transplantation from open heart surgery had a considerable influence on the strategies applied in heart preservation. Cardioplegic solutions developed for cardiac surgery provided safe preservation of myocardial function for only 5 to 6 h.

For transplantation, the heart is excised after induction of hypothermic cardioplegic arrest by in-situ flush with one of several standard cardioplegic solutions (Table 4) 230. The graft is then stored cold in the flush-out solution or in EuroCollins solution. Sinus node dysfunction (usually transient) is common after transplantation and is related to the duration of ischaemia. Microvascular injury has been reported in the presence of well-preserved myocytes.

Flushing solutions commonly used for the preservation of other organs are also cardioplegic, but generally contain much higher concentrations of potassium (over 100 mmol/l instead of 20 to 30 mmol/l) and do not contain any calcium. Initial reluctance to apply such solutions to heart preservation stems from earlier studies indicating that potassium-induced contraction band necrosis occurred with potassium concentrations above 40 mmol/l, and that a ‘calcium paradox’ effect in cardiac preservation was important (cardiac muscle incubated in calcium-free medium undergoes severe irreversible damage when reperfused with calcium-containing medium due to a massive influx of calcium). Warm ischaemic damage enhances calcium influx. The occurrence of the calcium paradox during hypothermia is controversial; but removal of calcium from St Thomas's solution has been shown to be detrimental, and the calcium paradox phenomenon has been seen in experimental cardiopulmonary bypass.

Recently, modifications of UW solution have been used clinically in open heart surgery and for transplantation, and in experimental studies. The range of available experimental models include heterotopic transplants to the abdomen or neck in small animals, isolated perfused working heart models, metabolic tissue analysis and histology, nuclear magnetic resonance spectroscopy, and allograft function in larger animals. Specimens of human atrial myocardium can be obtained for study during open heart surgery. Successful orthotopic transplants have been reported after 12 h of preservation with UW solution in primates, and clinical trials with UW solution have given superior donor heart preservation over that obtained by Stanford and other solutions.

Maintenance of high energy phosphate levels (ATP and creatine phosphate) has been another goal, but no linear correlation exists between the concentration of adenine nucleotides in preserved heart and the functional outcome. When the level of these high energy compounds falls below a certain level no functional recovery occurs, but inadequate preservation may occur at normal levels. Addition of ATP or its precursors to preserving solutions can improve cardiac function, but this does not correlate with maintenance of tissue ATP levels and may have been due to the vasodilatory properties of the additives.

Generation of oxygen free radicals has been implicated in ischaemic heart disease and in reperfusion injury. Metabolic inhibitors and free radical scavengers have been used to pretreat the donor and as additions to the preserving solution. Improvements in left ventricular function, lipid peroxidation, and platelet aggregation have been demonstrated. Prostacyclins also dilate coronary vessels, inhibit platelet aggregation, and stabilize lysosomal membranes. A stable prostacyclin analogue added to the cardioplegic solution has been shown to improve ventricular function after clinical cardiac transplantation.

More prolonged myocardial preservation may depend on techniques. Perfusion at low flow rates with oxygenated perfusate improved preservation of rabbit hearts using a modified UW solution incorporating polyethylene glycol (PEG). A colloid is essential in such perfusates; PEG may prove a suitable alternative to hydroxyethyl starch.

Simple cooling is sufficient for short-term preservation for heart–lung transplants. Lungs are flushed with cold EuroCollins solution via the pulmonary artery immediately after induction of cardioplegia. Flushing may be a source of injury and care must be taken to maintain pressure below that normally found in the pulmonary artery; lung distension is maintained by rhythmic ventilation at 50 to 70 per cent of normal inflation throughout the period of storage. Prostacyclin, superoxide dismutase, and catalase are beneficial when added to EuroCollins solution.

Clinically, treatment of the donor with prostaglandin E&sub1; followed by simple hypothermic flush with EuroCollins solution provides adequate 6-h preservation. In experimental heart–lung transplants in the dog, treatment of the recipient with superoxide dismutase also improved preservation. Use of a solution with lower potassium concentration may improve lung function: high potassium levels produce pulmonary vasospasm and may reduce the efficacy of the flush. Solutions containing colloid could be additionally beneficial in the lung.

Hypothermia and cardiopulmonary bypass with diluted blood is more complex, and has only been effective experimentally for 6 h. The addition of methylprednisolone, prostaglandin, and isoproterenol improved lung function.

Autoperfusion of heart–lung preparations at 30°C has afforded the longest lung preservation (12 h), but this procedure is cumbersome and limited in use. Adding metabolic substrates such as glucose or ribose enhances preservation, as does leucocyte depletion. Lung oedema, activation of complement, and subsequent pulmonary sequestration of leucocytes may play a role in lung dysfunction.

Transplantation of the lungs alone has met with increasing clinical success. Lung preservation using a EuroCollins flush is only successful for 5 to 6 h: after 12 h of storage pulmonary vascular resistance increases threefold and compliance is reduced; and after 24 h the lung is haemorrhagic and oedematous. Addition of methylprednisolone to the preserving solution and its administration to both the donor and recipient can allow storage to be extended to 12 h but not to 24 h.

The lung may require different preservation techniques from those used for other organs. Existing techniques have evolved primarily to maintain the viability of metabolically active cells (hepatocytes, myocytes, renal tubules). The maintenance of a well preserved microcirculation, essential for adequate lung function, may require a re-examination of available techniques. Blood perfusion at 25°C to 30°C with the addition of metabolites may afford optimum vascular preservation, without an irreversible energy deficit affecting parenchymal cells.

Clinical transplantation of the small intestine has not been widely applied, but several recipients of organ cluster grafts have survived for 12 months or more. The hazards of intestinal transplantation preclude it from being an alternative to home parenteral nutrition: the major problems are rejection and graft-versus-host disease.

Experimental preservation and transplantation of the small intestine was first attempted in the late 1950s. Storage for 24 h, after flushing both the vasculature and the lumen of the graft with EuroCollins' or histidine-based solutions, can maintain the physiological and pharmacological properties of the intestinal smooth muscle. Solutions containing dextrans have been equally effective. More prolonged preservation has not been successful. Treatment with free radical scavengers or allopurinol may improve preservation. The small intestine has the highest concentration of the enzyme xanthine oxidase: free radical damage has been implicated after warm ischaemic injury.

Reconstructive surgery increasingly uses vascularized autografts of composite tissues (skin, subcutaneous fat, muscle, and bone) to fill large defects. Severed limbs and digits are often replaced as autografts. These complex operative procedures can take many hours. Storage of the grafts has predominantly relied upon simple cooling by refrigeration, and wrapping of the tissues in cold saline-soaked packs during implantation.

Supplementary vascular flushing has not been widely used for fear of damage to the small vessels requiring microvascular suture. Experimental studies have not indicated that flushing gives any significant improvement over simple hypothermic storage—which adequately covers the periods (usually less than 12 h) required in clinical practice. Tolerance of different tissues to ischaemia varies: muscle is more sensitive than skin and bone. Morphology and viability of skin, muscle, and bone can be preserved after several days of storage at 4°C using simple hypothermia in a moist environment. In whole-limb replacement, flushing with an organ preservation solution can aid cooling and potentially improve preservation. Before the circulation is restored to replanted limbs or other tissues of large bulk, any solution with high potassium levels should be flushed out with balanced electrolyte solution or plasma, otherwise fatal hyperkalaemic cardiac arrest can occur after release of clamps.

Early in the 20th century it was realized that the cornea was not an inert transparent membrane, but a biologically active tissue needing careful handling if it was to survive the transplant operation. Pioneering efforts in the 1930s established the suitability of cadaver corneas for grafting and the feasibility of direct suturing of the graft to host tissues. The immunological privilege of the cornea was appreciated at an early stage. The cornea's function as a transparent reflecting and focusing complex requires the metabolic activity of a monolayer endothelium of mesodermally derived cells on the inner surface bathed by the aqueous humour.

For procurement from the cadaver donor, eyes are enucleated: this simple procedure can be done without elaborate preparation and need not be disfiguring. Originally corneas collected for transplantation were used immediately after the eye was removed, but storage in a moist environment at 4°C under sterile conditions allows penetrating (whole thickness) grafts to remain viable for up to 48 h after the death of the donor.

Intermediate storage times of up to 4 days can be achieved by the use of refrigerated tissue culture media. MK medium (developed by McCarey and Kaufman) containing dextran-40 and HEPES (N-2-hydroxyethylpiperazine- N-2-ethanesulphonic acid)-buffered tissue-culture fluid provides reliable preservation for 3 to 4 days. Storage in MK medium has achieved wide acceptance, but this still does not allow sufficient time for the ever more complex examinations required for the graft: examination of endothelial sheet integrity, disease screening (Creutzfeldt-Jakob syndrome, HIV antibodies, septicaemia), and tissue-typing, which may prove to be a valuable adjunct in corneal transplantation. Retrospective and prospective surveys have shown that results correlate with antigen matching. Further refinement of the tissue culture medium has resulted in the replacement of dextran-40 with chondroitin sulphate, which allows up to 14 days hypothermic preservation, with loss of only 3 per cent of the endothelial cells. Best preservation has been attained using 2.5 per cent chondroitin sulphate free of lower molecular weight moieties. Long-term storage can be achieved by cryopreservation of the isolated cornea. This technique was developed for prolonged eye banking, but has not gained universal acceptance.

Short-term hypothermic whole eye storage remains the most simple and convenient technique, particularly when there is no shortage of eyes, and in the United States this is still a common method of short-term banking of whole eyes. Tissue culture (MK) storage has gained acceptance by the majority of laboratories. A minority of departments use cryopreservation, and then only occasionally.

Preservation and transplantation of pancreatic islet cells have been extensively studied experimentally. Intraportal transplantation of islet extracts is now achieving some clinical success. Cellular preparations, after collagenase digestion and purification, have been successfully preserved by tissue culture for 24 h at 37°C, by simple ice cooling for 24 h, or for more prolonged periods by freezing with addition of cryoprotectives.

Harvested split skin grafts used as autografts or allografts can be preserved for approximately 2 weeks by simple wrapping and rolling in packs moistened with saline or tissue culture fluid. These are then stored in the refrigerator at 4°C.

Xenografts of pig skin can be similarly prepared and have been widely used for temporary cover of denuded sites. More prolonged preservation can be provided by freezing, or by culturing epidermal cells which can subsequently be reimplanted as a monolayer.

Autografts of fresh cancellous bone taken from iliac crest or other areas remain the most effective source of bone grafts, and provide viable osteoblasts and stimulate osteogenesis. Bone is a complex and active metabolic tissue, containing osteocytes bearing histocompatibility antigens nourished within a hydroxyapatite matrix. Stored allografts of bone can be cryopreserved for 12 months or more to provide a sterile source of bone matrix without viable cells. The biomechanical properties of bone do not seem affected by cryostorage, but preserved allografts are less active in stimulating osteogenesis than are fresh autografts.

Transplantation of large segments of cortical bone can provide a rigid bony matrix which is gradually replaced by creeping substitution of new bone. Transfer of viable cortical bone requires a vascularized graft which restores the bone's blood supply at its new site. Vascularized autografts of fibula and iliac crest can replace bony defects resulting from congenital anomalies, disease, or injury.

The metabolic needs of cartilage are chiefly met by diffusion of nutrients from synovial fluids; re-establishment of a blood supply is less critical than for bone. Chondrocytes also possess histocompatibility antigens and evoke an allograft rejection response. Intact cartilage survives better than isolated chondrocytes, showing the importance of the cartilage matrix. Experimentally, cartilage can be stored for 28 days in tissue culture medium at 4°C. The morphology of the cells and concentration of glycosaminoglycans and of collagen show no significant change, and the ability of cells to incorporate ³&sup5;S-sulphate into glycosaminoglycans does not diminish.

Cryopreservation at subzero temperatures with glycerol or dimethylsulphoxide produces a less favourable outcome, with loss of chondrocytes and conversion of hyaline cartilage to fibrocartilage. Isolated chondrocytes can be well preserved by freezing, suggesting poor penetration of cryopreservative.

Onlays or plugs of cadaveric articular cartilage fragments have been used as allografts in the management of degenerative arthritis. The methods remain experimental.

Bone marrow transplantation is now the preferred treatment for aplastic anaemias and for many leukaemias. Marrow allografts usually require minimal preservation, other than that provided by simple hypothermia as for blood collection. Invariably, living donors have been used. Extension of bone marrow transplantation in the future using cadaver donors could involve cryoprotective techniques similar to those used for freezing of blood. There is also an increasing use of autologous marrow grafts in cancer, the patient's marrow being removed and preserved before intensive chemotherapy and/or irradiation is given. The preserved marrow is then returned to the patient.

Autogenous veins are used to replace and bypass arteries in aortocoronary and peripheral vascular surgery. Occasionally autogenous arteries (such as the internal iliac) are used for short bypass procedures such as aortorenal grafts: in these situations the graft is a ‘vital’ one with the cellular tissues retaining their viability. The graft is usually prepared in heparinized saline at room temperature. Should the expected ischaemic period be longer than 1 h it is advisable to cool the vessel in iced saline. Blood vessels from cadaver donors are a very important source of extended arterial and venous conduits for renal, liver, and pancreatic allografts. These vessels are adequately preserved by simply hypothermic storage in standard flushing solutions. Vessels can be removed from cadaver donors, treated with glutaraldehyde, and stored indefinitely; they then become essentially collagenous tubes without apparent antigenic activity. This allografted non-living tissue does not cause a rejection reaction but will slowly degenerate, and a thrombosis or aneurysm can develop. It is an inferior option to the use of fresh autogenous vessels.

Preservation of organs and tissues for transplantation has entered a new and exciting phase with the development of multicomponent UW–lactobionate preserving solutions. The basic requirements of preservation remain rapid cooling aided by impermeants and buffers, but many other metabolic and pharmacological influences have been shown to enhance organ protection. Incorporation of such agents into preservation solutions, as well as their judicious use in preparation of the donor and in subsequent treatment of the recipient, has extended significantly preservation times of all organs and tissues. Although heart and heart–lung transplants still require rapid surgery with minimal storage times, prospects for extending safe preservation times for all vascularized organ grafts to 24 h or longer are now much closer. Research continues to look for more effective additives to preserving solutions, and to improve perfusion techniques, to extend safe preservation times even more. Preparation of the graft or cadaver donor by reduction of antigen presenting cells, as well as pregraft treatment of the recipient with selected monoclonal antibodies and other agents may also prepare the way for more successful transplantation. Advances in preservation will contribute significantly to such future developments.

Belzer FO, Southard JH. Principles of solid-organ preservation by cold storage. Transplantation 1988; 45: 673–6.

Calne RY, et al. Renal preservation by ice-cooling: An experimental study relating to kidney transplantation from cadavers. Br Med J 1963; 2: 651–5.

Carrel A, Lindbergh CA. The Culture of Organs. London: Hamish Hamilton Medical Books, 1938.

Collins GH, Bravo-Shugarman MB, Terasaki PI. Kidney preservation for transplantation: Initial perfusion and 30-hour ice storage. Lancet 1969; 2: 1219–22.

Flye MW. Principles of Organ Transplantation. Philadelphia: WB Saunders, 1989.

Marshall VC, Jablonski P, Scott DF. Renal preservation. In Morris PJ, ed. Kidney Transplantation. Principles and Practice. 3rd edn., Philadelphia: WB Saunders Company, 1988: 151–82.

Polge C, Smith AU, Parkes AS. Revival of spermatozoa after vitrification and dehydration at low temperature. Nature 1949; 164: 666.

Ross H, Marshall VC, Escott MO. 72-hour canine kidney preservation without continuous perfusion. Transplantation 1976; 21: 498–501.

Simmons RL, Finch ME, Ascher NL, Najarian JS, eds. Manual of Vascular Access, Organ Donation, and Transplantation. New York: Springer-Verlag 1984.

Todo S, Tsakis A, Starzl TE. Preservation of livers with UW or Euro-Collins solution. Transplantation 1988; 46: 925–6.

Wahlberg JA, et al. 72-hour preservation of the canine pancreas. Transplantation 1987; 43: 5–8.