The technical success of organ transplantation during the past decade has evolved in parallel with the surgical expertise which allows multiple organs to be obtained from a single cadaver donor. The procedure for multiple organ procurement has become widely adopted by transplant surgeons throughout the world, since this method was introduced by Starzl in 1983. Aside from the surgical advances however, the field of organ procurement has become the career interest of many medical personnel, who have functioned as co-ordinators of the organ donation process. This donation process now routinely entails a sequence of patient assessment and operative procedures, which includes determination of donor suitability, haemodynamic management prior to procurement, orchestration of operative events, and the preservation of the various organs.

Most organs for transplantation are procured from cadavers following the diagnosis of brain death. This diagnosis has facilitated the procurement of viable organs, free of the warm ischaemic injury seen when their removal follows the arrest of circulation. Nevertheless, the opportunity to salvage organs following cardiac asystole from otherwise suitable donors, remains inherently possible. Anaise has recently described an experimental procedure in dogs in which femoral vessel cannulation allows rapid exsanguination and installation of cold preservation fluid following cardiac arrest. Core visceral temperatures can be further decreased by continuous hypothermic perfusion of the abdominal cavity.

Following the cardiac death of many individuals, either in the emergency ward or on arrival at hospital, the Anaise method might allow sufficient in-vivo preservation, at least of the kidneys, to allow time for family permission to be obtained for organ removal. Thus far, this approach has received limited consideration because of the obvious public misunderstanding which might arise from its widespread implementation. However, these minimally invasive measures could be more comfortably initiated if individuals carried a donor card, indicating their prior consent to donation. They may also be applied in countries where consent for donation is presumed.

Although death may be defined in its simplest form as a permanent absence of brain function, this concept of death as it relates to brain activity is relatively new. Death was first characterized in terms of irreversible coma in 1968, following a report by an ad hoc committee of the Harvard Medical School, which was assembled to define brain death. Four essential criteria were necessary to establish coma as irreversible: unreceptivity and unresponsivity, no spontaneous movements or breathing, an absence of reflexes, and a flat electroencephalogram. These criteria have subsequently been modified to be less restrictive, but their accuracy has stood the test of extensive scrutiny. Prior to this report, death was only considered in terms of circulatory and respiratory function. With the development of sophisticated cardiorespiratory support systems, however, the notion of death based solely upon an absence of heartbeat became obsolete: arrest of heart activity could be considered a manifestation of death only by its terminal effect upon brain function. The routine resuscitation of patients following transient cardiac arrest forced a conceptual revision in the diagnosis of death.

Although Japan is a notable exception, the concept of brain death has become well accepted internationally during the past two decades. Homicide convictions have not been hampered by the brain death of a victim, and the medical progress report of brain death by Black greatly helped the medical profession to establish brain death as a diagnostic entity. Later, a commission of medical consultants to the President of the United States published guidelines for the determination of death, which acknowledged the concept of brain death and specified the criteria of brain death in terms of cerebral and brain-stem function. Many institutions subsequently adopted policies regarding brain death, with minor modifications appropriate for local medical practice, but incorporating the essential criteria of the presidential commission. For example, at the Massachusetts General Hospital, both clinical and laboratory criteria have been used to establish the essential elements of the diagnosis: cerebral unresponsiveness and brain-stem inactivity (Table 1) 221. An apnoea test has become a standard method of confirming irreversible brain-stem damage, although the level of Pco&sub2; which must be attained may vary. However, absent brain-stem function is also recognized clinically when pupillary light, corneal, oculocephalic, and oropharyngeal reflexes are absent.

The absence of electrocerebral activity as revealed by an EEG can be supportive of the diagnosis of brain death; absence of cerebral blood flow as demonstrated by a radionuclide or angiographic procedure is an alternative. However neither of these investigations are required, as stressed by Pallis, provided a careful evaluation of brain-stem death has been performed (Table 2) 222. In the United Kingdom the diagnosis of death is a diagnosis of brain-stem death.

A period of at least 24 h without neurological change is recommended before the diagnosis is made, especially for children less than 5 years of age. Screening for drugs which depress the central nervous system is also recommended. If the cause of coma is known with certainty and drug-related and metabolic causes have been excluded, a period of 6 h without clinical change is sufficient to allow a diagnosis of brain death in adults.

The use of barbiturates to reduce intracranial pressure and/or the seizure threshold, does not preclude the diagnosis of brain death, if the serum barbiturate level at the time of neurological examination is less than 1 mg per cent. The determination of brain blood flow may be useful in circumstances where barbiturate levels are higher.

The need to assess brain-stem activity for the diagnosis of brain death has precluded its application to anencephalic newborn infants, and to some irreversibly comatose individuals: these patient groups are both devoid of cerebral cortex activity, but show spontaneous breathing, which requires intact brain-stem function. An active debate has been waged as to whether brain-stem function alone, irrespective of cerebral function, should imply the presence of ‘life’, and this has gained attention following the use of an encephalic donors of hearts and kidneys. At this time however, society has not accepted a diagnosis of death which disregards brain-stem function, that is a capacity to breath spontaneously, and the procurement of organs from anencephalics prior to death has not been sanctioned.

A detailed social and medical history should be taken from the donor's family by the organ procurement agency representative (either physician or co-ordinator) who obtains consent for donation. In addition to being aware of the donor history, procuring surgeons should perform an appropriate examination which might bear upon donor suitability. This includes looking for extremity lacerations which have become secondarily infected, or needle marks indicative of drug abuse, and a rectal examination in donors over 60 years of age.

The cause of brain death should be ascertained prior to its pronouncement and certainly prior to organ procurement. Irreversible brain injury is usually due to head trauma, subarachnoid or intracranial haemorrhage, brain hypoxia of known aetiology, drug overdose, or primary brain tumour (glioblastoma, astrocytoma, etc.). A history of any other malignancy should be ruled out; these include metastatic melanoma, choriocarcinoma, and renal cell carcinoma, which can cause spontaneous intracranial haemorrhage and may not be apparent as a cause of brain death. Primary malignancies such as lymphoma, renal cell carcinoma, choriocarcinoma have been transmitted unknowingly by liver and renal allografts resulting in the death of organ recipients.

All potential donors should be evaluated for the possibility of multiple organ donation. There are certain age restrictions which apply to the donation of individual organs (Table 3) 223, although the persistent shortage of organ donors has led to these age limits being extended. Donor age has an impact upon successful transplantation, with recipients of organs from donors aged 16 to 65 years achieving the best clinical results. The use of kidneys from paediatric donors is controversial: children less than 3 years of age may not be suitable renal allograft donors unless en bloc paired kidneys are transplanted.

A past history of diabetes mellitus requiring treatment may not preclude organ donation; these patients should be evaluated on an individual basis and should not be excluded without consultation from local organ bank personnel (see Table 4 224). Similarly, a history of hypertension is no longer an absolute contraindication to donation, particularly of kidneys. If the hypertension is mild (diastolic pressure between 90 and 100 mmHg) and controlled with single agent therapy, a biopsy specimen during the period of renal preservation may be obtained and evaluated for the degree of intraparenchymal arteriosclerosis.

Potential donors with a history of autoimmune disorders must be carefully evaluated: the transmission of idiopathic thrombocytopenia purpura by liver transplantation has been reported. The 47-year-old recipient was subsequently cured by a second transplant.

Carbon monoxide poisoning from smoke inhalation can produce not only brain hypoxia but hypoxic myocardial damage that is not evident on echocardiography. Hearts obtained from victims of smoke inhalation have been observed to fail immediately following transplantation.

Each potential donor must be assessed for carriage of infectious agents to prevent the transfer of micro-organisms to allograft recipients. Systemic infection transmitted by a cadaveric donor organ can result not only in a loss of the allograft, but also in death of the immunosuppressed recipient.

The history and physical examination of a potential cadaveric donor may reveal a systemic infection present prior to or associated with the patient's death: the latter may have been hospital acquired. The cause of death may promote an infectious complication: victims of drowning often eventually die of pneumonia, and burn victims may develop cutaneous sepsis.

The presence of an active viral infection in the form of hepatitis, perineal herpes, encephalitis or meningitis, pneumonia, varicella zoster, or human immunodeficiency virus (HIV) infection, is an absolute contraindication to organ donation. A history of a specific viral infection does not preclude donation, however, but requires the exercise of clinical judgement, after a thorough gathering of historical information. Kidneys have been successfully transplanted from donors dying of Reye's syndrome, in which the influenza virus has been implicated. No adverse effects, in particular the development of Reye's syndrome, were evident in the allograft recipients.

A judicious approach is also applicable to potential donors with a history of hepatitis. Active viral hepatitis may be manifested by liver dysfunction at the time of donor death, and diagnosed either by the detection of hepatitis B antigenaemia, or serological reactivity to the hepatitis A or C viruses. The acquisition of the hepatitis B virus from renal allografts is well documented and patients positive for hepatitis B surface antigen should not donate their organs. However, a history of hepatitis B antigenaemia may not prohibit donation, if at the time of death the patient is hepatitis B antigen-negative and hepatitis B antibody-positive.

For many years, an unidentified viral pathogen (non-A, non-B hepatitis virus) had been implicated in the aetiology of hepatitis following transplantation, possibly transmitted from previously infected organ donors. Recently this elusive agent has been identified as hepatitis C virus. A flurry of clinical reports has documented the infectious potential of blood transfusions from hepatitis C antibody-positive donors. As a result, some blood banks have prohibited the transfusion of blood products from anti-HCV-positive donors. The question of transmission of hepatitis C virus to allograft recipients has also been raised, and organ procurement agencies have attempted to determine the prevalence of anti-HCV antibodies in their donor population. In a retrospective study of 716 consecutive donors to the New England Organ Bank between 1986 and 1990, 13 (1.8 per cent) of the donors tested were positive for antibodies to hepatitis C virus. Organs procured from these donors were transplanted into 29 recipients: hepatitis and/or liver failure was seen in 14 of the 29 (48 per cent) recipients (median onset 4.5 months after transplantation). Seroconversion for the virus occurred in four (29 per cent) of the 14 patients whose serum samples were available for testing before and after transplantation. All four patients developed liver disease. Thus, organ procurement agencies may restrict the acceptance of organs from donors positive for antibody to hepatitis C virus, unless the need of the recipient is so critical that transplantation is justified.

Since hepatitis A virus has not generally been considered a virulent pathogen for the allograft recipient, a donor history of hepatitis A virus infection has not been regarded as a contraindication to organ donation. A 3-month interval between active hepatitis A and death is accepted as sufficient to allow for organ donation, particularly if the serum titre of IgM antibodies to hepatitis A virus has returned to normal.

All potential organ donors should be assessed for HIV infection, as this virus can be readily transmitted by allografts. Screening for HIV by enzyme-linked immunoassay prior to organ donation is now routine; however, this assay for HIV antibodies may not detect recent onset HIV antigenaemia. Social circumstance must also be taken into account in the evaluation of potential cadaveric donors. A history of homosexuality or intravenous drug abuse may exclude a person from donation. Exclusion on the basis of social risk, may also be exercised for individuals who are homeless and alcoholic, prisoners, or immigrants from East Africa or Haiti, even if tests for hepatitis and HIV are negative.

Genital herpes virus infection at the time of death may raise a suspicion that the potential donor is a concomitant HIV carrier, even though the enzyme-linked immunosorbent assay does not reveal the presence of anti-HIV antibody. Aside from the implication of HIV infection, active herpes simplex may be transmitted through renal allografts to the immunocompromised recipient, and organ procurement from these individuals is best avoided.

Caution should also be exercised when potential donors have received multiple blood transfusions during resuscitation following trauma. HIV testing should be performed on a blood specimen obtained from the potential donor prior to the administration of transfusions since an infected individual may have a false-negative HIV test following repletion with multiple transfusions of blood.

Acute varicella zoster infection can be lethal in an immunosuppressed allograft recipient, and all potential allograft recipients should be screened for antibody to varicella-zoster virus prior to transplantation. Allograft recipients who are antibody negative are warned to avoid contact with individuals experiencing acute varicella, and individuals with primary or reactive varicella at the time of death should probably be excluded from donating organs.

Kidneys from donors shown to be serologically positive for cytomegalovirus may be the source of cytomegalovirus infection in allograft recipients, especially in seronegative recipients. Renal allografts may be preferentially disposed to transmission of this virus: a variety of cultured renal cell types will support growth of cytomegalovirus in culture, and donor-specific strains of the virus have been identified in renal allograft recipients by analysis of viral DNA. Cytomegalovirus has also been implicated in the rejection and atherosclerosis of cardiac allografts. Active cytomegalovirus infection, in the form of a pneumonia or hepatitis, therefore precludes organ donation. This type of cytomegalovirus infection may also raise the possibility of concomitant HIV infection, irrespective of the results of antibody testing.

Cystitis, as manifested by a positive culture of urine obtained through a Foley catheter, is not a contraindication to donation. If pyuria is observed, ureteral stump cultures may be obtained during the procurement procedure. While bladder contamination following Foley catheterization has not been associated with recipient infection following transplantation, a positive ureteral stump culture raises the possibility of active pyelonephritis, which is a contraindication to transplantation. Good surgical technique, with ligation of the distal ureters following ureteral transection, prevents potentially contaminated urine refluxing from the bladder into the open peritoneal cavity.

A history of pyelonephritis within 3 months of organ donation may increase the risk of transmission of bacteria to the recipient. Clinical judgement regarding the type of organism and verification of treatment should be exercised before recently infected kidneys are accepted for transplantation. Pseudomonas aeruginosa and Staphylococcus aureus have been reported to subsequently disrupt the vascular anastomosis of a renal allograft.

A history of chronic respiratory infection or acute pharyngitis does not exclude a patient from donating organs (with the exception of lung donation) unless a virulent organism such as Staphylococcus aureus is identified. If bacteraemia is not demonstrated by blood culture, donation may be considered to be safe.

Brain death due to mycotic aneurysm necessitates an evaluation of the cause: Candida albicans and Staphylococcus aureus are extremely dangerous to allograft recipients, and blood cultures from potential donors may be negative.

The death of a renal allograft recipient from disseminated tuberculosis transmitted through a cadaver kidney, obtained from a donor with unsuspected tuberculosis meningitis has been reported. The aetiology of the meningitis only became apparent several weeks after the donor's death, when mycobacteria grew in cultures of his cerebrospinal fluid. The donor's chest radiograph was normal. As well as documenting the transmission of tuberculosis via transplanted organs, this case underscores the danger of accepting organs from individuals with a diagnosis of meningitis of unknown aetiology.

A Venereal Disease Research Laboratory Test (VDRL) can detect non-specific antibodies directed against lipoidal antigens of Treponema pallidum. A VDRL or rapid plasma reagin test is routinely obtained on all individuals considered for organ and tissue donation. Although a positive test suggests exposure to T. pallidum, the lipoidal antigens used in the assay are found in a number of normal tissues, and a more specific test such as the fluorescent Treponema antibody absorption assay may be necessary to exclude a false-positive VDRL result. A positive reaction with the fluorescence assay is highly suggestive of infection.

Toxoplasma gondii is an indolent parasite, and infectivity of a potential donor may only become apparent after transplantation. Although toxoplasmosis may now be detected as an associated infection in patients with immunodeficiency syndromes such as AIDS, toxoplasmosis may be transferred from an unsuspected carrier who has died abruptly from trauma. In this latter circumstance detection of donor toxoplasmosis may only be accomplished retrospectively, following the detection of an elevated level of antibodies against Toxoplasma in stored donor blood. In general, retention of donor serum for such unforeseen developments may be prudent. Lethal toxoplasmosis has been transmitted to recipients of cardiac allografts from apparently healthy seropositive donors with unsuspected central nervous system disease. Thus, donor seropositivity to Toxoplasma gondii, identified in advance of transplantation, may be a contraindication to donation, especially for a seronegative recipient.

Candida albicans frequently colonizes the vagina and perineum of patients who are maintained on broad spectrum antibiotics for a long period of time. Thereafter, the monilia may gain entrance to the bladder, through an indwelling Foley conduit. Wound infection and vascular disruption may follow transplantation of organs contaminated with Candida, especially to diabetic recipients. Fatal candidal mediastinitis has been reported in recipients of lungs from donors with a heavy growth of Candida in cultures from the trachea.

Histoplasma and Cryptococcus have been transmitted to renal allograft recipients by organs from donors who died of intracerebral pathology, without evidence of a pulmonary infection. These organisms are difficult to eradicate from the central nervous system unless a protracted course of antifungal therapy (amphotericin B) is administered. Even with careful documentation of proper therapy however, a history of fungal infection (also including cocciodiomycosis and blastomycosis) should exclude an individual from donation, despite the apparent absence of infection at the time of death.

Chest tubes are not a contraindication to organ donation unless culture of the pleural fluid is positive for an organism which poses a high risk to the recipient, such as Pseudomonas aeruginosa or Staphylococcus aureus.

Peritoneal drains contaminated with enteric bacteria may assume more significance, especially if they reflect current peritonitis due to injured or devitalized bowel. A retroperitoneal dissection of each kidney can be accomplished through separate flank incisions if necessary. This approach may also be used when gastrostomy tube leakage into the abdominal cavity is suspected in patients who have been nutritionally depleted and who display poor wound healing.

Although perforations of the bowel repaired within 1 week of brain death probably exclude a patient from consideration as a donor, prior repair of the intestine may not, provided that a Gram stain of peritoneal fluid reveals no organisms, blood cultures are negative, and appropriate antibiotic coverage has been administered during the interval.

Extremity fractures in a potential donor may pose a hazard for allograft recipients, if metal appliances have been placed to stabilize the fracture, and/or there is evidence of cellulitis adjacent to the cutaneous exit site of the appliance. Once again, antibiotics should have been administered for at least 48 h and blood cultures must be free of bacteria before organ procurement is undertaken.

Burn victims who sustain tracheobronchial injury from smoke inhalation are at risk of pneumonitis. Burn injuries also compromise the procurement procedure since incisions cannot be made through damaged skin. Nevertheless, if incisions can be fashioned through areas of skin which have not been burned, organs may be procured from burn victims within 48 h of injury. A longer interval between injury and death increases the risk of septic contamination of visceral organs, through either a compromised respiratory tract or through burned skin.

Cellulitis adjacent to antecubital or subclavian vein intravenous lines signal the possibility of line infestation, which can seed visceral organs with bacteria such as Staphylococcus aureus. Management prior to procurement surgery entails the removal of the intravenous line, administration of appropriate antibiotics for at least 24 h following line removal, and the determination of negative blood cultures.

Any untreated bacteraemia noted around the time of death creates a significant risk of bacterial transmission to allograft recipients. Organisms such as Staphylococcus aureus and Pseudomonas aeruginosa are especially virulent, and have been noted to disrupt anastomotic suture lines to the allograft. Systemic infection of the donor with these organisms (e.g. pneumonia) is generally a contraindication to donation.

Traumatic injury to any visceral organ may initially be detected through laboratory testing of donor blood samples (Table 4) 224. Unremitting functional impairment because of shock or trauma generally precludes procurement of the specifically injured organ.

Candidates for heart donation are usually free of cardiac murmurs and display a normal 12-lead ECG. Pathological Q waves suggestive of infarction are a contraindication to donation; other electrocardiographic abnormalities, such as ST segment changes, do not exclude donation. Echocardiography may be performed to assess myocardial function and potential valvular pathology in patients requiring pressor support, or in whom a murmur is audible. Mitral valve prolapse without mitral regurgitation is not uncommon and alone does not preclude heart procurement.

Candidates for lung or heart and lung donation should usually be free of a smoking history and have clear lung fields on chest radiographs. A Gram stain and culture of the tracheobronchial secretion should be performed. Blood gas assessment or respiratory function is determined by ventilating the donor with 100 per cent oxygen and no more than 5 cm of positive end-expiratory pressure: under these conditions, the arterial oxygen pressure of a blood sample should be greater than 300 mmHg.

Many transplant centres are willing to accept livers from donors whose liver function tests are not within normal range. A 2- or 3-s elevation in the prothrombin time, and/or mild elevations in transaminase levels are not necessarily a contraindication to procurement. A careful history and serological testing for hepatitis B surface antigen and antihepatitis C antibody are important elements in the decision.

Small liver lacerations may be observed at the time of organ procurement from victims of trauma with normal blood transaminase levels. A small laceration with no bleeding or bile leakage from the site may not represent a contraindication to liver transplantation.

Liver function is especially susceptible to hypoxia: haemodynamically unstable patients requiring pressor support, with marginal arterial oxygen pressure (<100 mmHg), may not be satisfactory candidates for liver donation, particularly if the oxygenation cannot be improved because of pulmonary oedema. Consequential impairment of liver function may be discerned by an increase of at least 5 s in the prothrombin time and a transaminase level that is twice normal.

A coagulopathy detected by thrombocytopenia, low fibrinogen level, and a marked elevation of the prothrombin and partial thromboplastin times, suggests a disseminated intravascular coagulopathy, which is known to be associated with head injury, but may also be the result of sepsis. Once again, clinical judgement must be exercised in determining any possible cause of sepsis, impairment of organ function, and the potential for transmission of bacterial organisms. A biopsy of the renal allograft prior to transplantation may be necessary to exclude the presence of microthrombi within the renal parenchyma. Pulmonary dysfunction is a frequent cause of irresolvable hypoxia in patients with disseminated intravascular coagulation because of the deposition of microthrombi in the lungs.

The standard determinations of serum creatinine, blood urea nitrogen, urinalysis, and urine output (>1 ml/kg.min) are used to assess satisfactory donor renal function. Creatinine levels above 2.0 mg per cent are not necessarily a contraindication to kidney donation if the cause of renal dysfunction is transient hypotension, and if the creatinine level falls with intravenous fluid repletion. Chronic renal insufficiency, as noted by a persistent creatinine level of greater than 2.0 mg per cent, combined with a urinalysis revealing proteinuria and/or sediment casts, contraindicate renal donation.

A history of diabetes mellitus precludes pancreas donation, but does not necessarily prohibit renal donation, especially if the serum creatinine level is less than 2.0 mg per cent. Mild elevations in serum amylase may not be a reflection of pancreatic injury or pancreatitis, since the enzyme may be derived from a salivary gland source following trauma. Neither mild hyperamylasaemia nor mild hyperglycaemia are contraindications to pancreas donation.

Many organ procurement agencies have adopted a practice of extracting lymph nodes from the inguinal bed as soon as death has been declared and permission for organ donation has been granted. This practice has facilitated the identification of renal allograft recipients prior to kidney procurement, as accurate tissue typing from peripheral blood is usually difficult, and the subsequent preservation period of cold ischaemia.

A 6- to 8-cm incision is made below the inguinal ligament, along the course of the femoral vessels, using sterile technique. Three or four lymph nodes are more than sufficient for HLA typing, and these may be readily identified medial to the femoral vein.

In the same year in which the criteria were developed to determine brain death (1968), a Uniform Anatomical Gift Act (UAGA) was independently promulgated in the United States to facilitate the process of organ donation. The concept of the donor card was derived from the Uniform Anatomical Gift Act, as it permitted the decedent to certify an intent to donate, irrespective of the wishes of the surviving family. Nevertheless, a hierarchy of responsible family members was established for circumstances in which family approval would be requested. The order proceeded from spouse, to offspring, to parent, to sibling, and finally to guardian.

Unfortunately, few people have completed donor cards, even though renewal of a driver's licence affords a simple opportunity to do so. Moreover, the general policy of organ procurement agencies has been to obtain permission from the surviving family member, determined by the hierarchy noted above, even in instances where a donor card has been identified. Family wishes have prevailed despite properly indicated donor intent. It has also become a standard practice to obtain permission for organ procurement from the local medical examiner or coroner when potential donors have died from an unnatural cause. The medical examiner should be contacted in advance of organ procurement. The operative report may serve as evidence of the postmortem examination of the thoracic and abdominal cavities.

Measures such as the Uniform Anatomical Gift Act have not been successful in promoting a much needed voluntary organ donation. As a result, alternative approaches have been formulated to enhance the public awareness of the shortage of organs for transplantation. Legislation has been adopted, which makes a request for organ donation by hospital personnel mandatory in appropriate cases of brain death. However, this well-intentioned approach of ‘required request’ has not been successful in enlarging the donor pool in countries such as the United States, from the fixed rate of 16 donors/million population, to a rate estimated in some areas such as Austria to be as high as 38 to 50 donors/million population. This discouraging observation may mean that health care professionals have not accepted mandatory regulation and remain ambivalent about the concept of brain death and organ donation. The failure of ‘required request’ laws may also lead to the regulated process of consent being conducted by inconsistent and inexperienced personnel.

Presumed consent and financial incentives, although objectionable to many, remain options to boost the rate of organ donation in the future. Presumed consent for organ donation (organ removal takes place unless a surviving family member objects or the potential donor before death has objected) has been introduced in several European countries, resulting in an organ donation rate of approximately 25/million population.

Primary allograft function following transplantation is dependent upon careful maintenance of the donor after the declaration of brain death, the surgical technique of retrieval, and the duration and method of preservation.

Brain death elicits physiological responses which may lead to haemodynamic instability, including arrhythmias, a significant decrease in thyroid hormones (both T&sub3; and T&sub4;), and a reduction in circulating adrenal cortical and insulin levels. Hypothermia is common, as central neurological control of body temperature is lost; the resulting shift from aerobic to anaerobic metabolism leads to acidosis, which may be exacerbated by the peripheral vasoconstriction arising from pressor support.

Hypotension may develop as a consequence of cardiac arrhythmia, brain-stem herniation, or hypovolaemia. Brain death may also induce diabetes insipidus and polyuria, which exacerbates hypotension through a reduced blood volume. Diabetes insipidus becomes apparent as the serum osmolality (>300 mosmol) increases. Hypovolaemia is also the expected result of resuscitative measures intended to minimize cerebral oedema, prior to the development of brain death.

The goals of donor management are the prevention of hypotension, hypothermia, and hypoxia. If not already instituted, central venous pressure and arterial lines should be placed, as well as a Foley catheter, to monitor blood pressure, blood volume repletion and hourly urine output.

Crystalloid replacement with Ringer's lactate solution should be given until a systolic blood pressure of at least 90 mmHg is achieved. Administration of synthetic arginine vasopressin (1–2 units/h) will reduce urine output, conserving intravascular water, and may also contribute to haemodynamic stability by an independent mechanism.

Following extensive research in brain dead experimental animals, Novitsky has proposed the routine administration of a hormonal cocktail consisting of triiodothyronine, insulin, and cortisol, to assure haemodynamic stability and minimize systemic acidosis, but this approach remains controversial.

The hypovolaemia associated with diabetes insipidus may be worsened by haemorrhage associated with traumatic injury: scalp lacerations are a frequent source of unsuspected blood loss. Blood transfusions may be necessary, particularly if the haematocrit has fallen to less than 25 per cent. Volume repletion should raise the central venous pressure, thereby optimizing intracardiac filling pressures and cardiac function. If these measures fail to restore a satisfactory blood pressure, dopamine may be given at an initial constant intravenous rate of 2 &mgr;g/kg.min. This may be increased to 10 &mgr;g/kg.min if necessary; more than this may reduce renal and mesenteric blood flow, as vasoconstrictive effects are predominant at higher doses.

Because of their vasoconstrictive properties, alternative pressors such as phenylephrine or levophed are not recommended. If a urine output of at least 1 ml/min is not achieved following volume and pressor resuscitation of the systolic blood pressure to greater than 100 mmHg, then mannitol (12.5–25 g) and/or frusemide (furosemide) (20–40 mg) may be given intravenously. Metabolic acidosis should be corrected by bicarbonate injection.

Monitoring of the donor following the determination of brain death, usually includes the sampling of blood and urine for bacterial culture, particularly in those who have been in hospital for longer than 48 h. The routine administration of broad spectrum antibiotics such as cephalosporin is advisable, even in the absence of bacteraemia.

Following the transfer of the potential donor to the operating room, manually ventilated with 100 per cent oxygen, ventilator settings should be re-established to maintain the partial pressure of arterial oxygen at greater than 100 Torr. A warming blanket may be required to prevent the hypothermia (<32°C) associated with brain death, and its destabilizing effect upon cardiac function. The patient's arms should be maximally abducted, to permit a simultaneous dissection by thoracic and abdominal teams. Anaesthetic management of blood gases, central venous pressure, haematocrit, and urine output are essential for successful organ procurement. A haematocrit of less than 30 per cent should be restored by transfusion of warmed blood, especially if the patient is hypoxic or oedematous due to previously infused crystalloid solution. Bicarbonate administration may be necessary to correct metabolic acidosis.

A midline incision is made from the suprasternal notch to the pubis. As the sternum is split, bone wax is applied to minimize bleeding. If only intra-abdominal organs are being procured the sternal incision is omitted, and the midline incision is made from the xyphoid to the pubis. Some surgeons use a cruciate abdominal incision just above the umbilicus for extended exposure. Towel clips may be applied from the apex of each leaf of the abdominal incision, as it is folded to the opposing skin of the abdomen.

The organs of interest, whether thoracic or abdominal, must be inspected by the surgeon for unsuspected pathology or injury.

The thoracic teams generally begin the dissection by opening the pericardium and/or pleural cavities. The great vessels, including the vena cavae, the aortic arch, the innominate vessels, and the pulmonary artery are isolated. Both lungs may be mobilized and inspected by the division of the pulmonary ligaments.

The abdominal dissection may vary, depending upon the viscera to be procured. The liver is usually removed first, with the hepatic arterial circulation being established by dissection of the coeliac axis and superior mesenteric arteries. Replaced right or left hepatic arteries originating from the superior mesenteric or left gastric arteries must be preserved. When the donor is haemodynamically stable, the coeliac axis and superior mesenteric artery can be isolated to their branches. The portal vein is dissected and the inferior mesenteric vein may be cannulated to permit in-situ portal cooling, usually with Ringer's lactate solution. Alternatively, the splenic vein or superior mesenteric vein may be used for portal flushing if the pancreas is not required. The common bile duct is divided at the duodenal brim.

Pancreatic dissection may take place next. The spleen is dissected from its bed, and used as a handle to mobilize the tail and body of the pancreas. On the right side, a Kocher manoeuvre is used to dissect the C-portion of the duodenum and the head of the pancreas, again to the mesenteric vessels.

Each kidney and ureter is then mobilized by dissecting the poles beneath Gerota's fascia, the number and course of the left and right renal arteries being identified.

The donor is then anticoagulated systemically with 10000 to 20 000 units of heparin, and the distal aorta above the iliac bifurcation is cannulated. Rarely, an iliac artery may provide a lower pole renal artery branch: this anatomical aberration must be carefully assessed before aortic cannulation, and if present the contralateral iliac artery can be cannulated for prograde aortic cooling. A cannula may also be placed into the inferior vena cava below the renal veins, to provide a controlled exit for visceral perfusate. Some surgeons prefer to transect the inferior vena cava and allow the perfusate to empty into the abdominal cavity.

In the thoracic cavity, cannulas are finally placed into the main pulmonary artery for cardiopulmonary-plegia and prostacyclin infusion, and into the proximal ascending aorta for cardioplegia. The superior and inferior vena cava are ligated and transected as the plegia solutions are given.

Thoracic organs are excised first. Either cardiectomy is performed prior to lung removal (possibly as a double lung block); or the heart and lungs are removed en bloc and divided on a back table in the operating room. Alternatively the heart and lungs may be removed as a unit for transplantation to a single recipient. As the aortic arch is cross-clamped (usually with a staple device), perfusion of the intra-abdominal organs is begun through portal and aortic cannulas and this is continued during the thoracic excision.

The liver is the first intra-abdominal organ removed (Table 5) 225. Although the simultaneous procurement of the liver and pancreas from a single cadaver donor was once considered to be technically impossible because of the same blood supply, it is now routinely accomplished. Transection of the portal vein, 2 cm cephalad to its bifurcation into the splenic and superior mesenteric veins, provides sufficient length of portal vein for transplantation with both the liver and pancreas. The arterial supply can be managed by providing a Carrel patch of aorta either to the coeliac axis of the liver, or to the coeliac axis and superior mesenteric artery of the pancreas. Usually however, the coeliac axis and aortic patch are retained with the liver, and the iliac bifurcation graft from the donor is anastomosed to the splenic and superior mesenteric arteries of the donor pancreas.

As the liver is removed, aortic perfusion of the kidneys and pancreas with preservation fluid is continued: approximately 2 litres of preservation fluid are usually given through the aortic cannula.

The method of whole organ pancreas transplantation which is most commonly employed uses a segment of duodenum to drain exocrine secretions into the bladder. Since the donor duodenum must be transected (usually with the staple device), the pancreas is removed following en bloc nephroureteroectomy, to avoid contamination of either kidney with enteric organisms. The pancreas preparation therefore consists of a segment of duodenum, the entire pancreas, and the attached spleen.

Finally, both kidneys are removed en bloc with a cylinder of aorta and inferior vena cava, and with attached ureters of at least 10 cm in length. Once removed the kidneys may be divided for separate packaging and cold storage preservation, or they may be placed separately (or en bloc) on to a pulsatile perfusion machine.

The development of a consistently effective preservation fluid has dramatically changed the timing of renal, hepatic, and pancreatic transplant procedures. The preservation solution developed at the University of Wisconsin has extended the period of preservation to 48 h, permitting wider geographical sharing of renal allografts, which are allocated principally on the basis of HLA matching. The preservation time for liver and pancreas allografts has also been extended (Table 6) 226.

The objective of organ preservation is to cool the core temperature of the organ parenchyma so that the demand for oxygen and the requirement for energy (in the form of ATP) can be markedly reduced. During this period of ischaemia, cellular integrity of the parenchyma is maintained. These goals can be achieved by flushing of the organs with one of several preservative solutions at a temperature of 47°C.

The constituents of the University of Wisconsin solution include hydroxyethyl starch and lactobionate which suppress cell swelling, glutathione and MgSO&sub4; to stabilize cell membranes during the cold ischaemic period, and allopurinol to scavenge oxygen free radicals associated with reperfusion injury, and to stimulate ATP synthesis after preservation.

The debate as to whether renal preservation is best achieved by pulsatile perfusion or cold storage has largely subsided. The simplicity of cold storage is a major cost-saving asset. In a randomized study conducted by the New England Organ Bank, renal allografts cold stored for less than 36 h functioned as well as the paired kidney preserved by pulsatile perfusion.

Confronted with a persistently inadequate supply of organs from individuals who have died transplant clinicians have resorted to using living donors, not only for renal allografts but also for liver and pancreatic organs. Unrelated living donors of kidneys have been used when blood type compatibility and an emotional bond with the recipient exists. At one time, kidneys obtained from living related family members accounted for more than 30 per cent of renal transplants performed annually in the United States. The procurement of organs from living unrelated individuals in economically poor countries has been subject to unethical abuse. Nevertheless, living donation remains ethically accepted as a rewarding experience for the donor and a true benefit for the allograft recipient.

Improvements in immunosuppression and in the outcome of cadaver donor transplantation have reduced the proportion of organs obtained from living related donors in recent years. Between 1986 and 1990, 1753 (91.3 per cent) of the 1921 kidneys transplanted in Australia were from cadaver donors, 159 (8.3 per cent) were from living related donors, and 9 (0.5 per cent) were from living unrelated donors. In England, between 1989 and 1990, only 5.7 per cent of renal allografts were obtained from living donors. Thus, the number of living donors continues to fall far short of the demand of an increasing number of patients awaiting renal transplantation.

The operative risks of kidney donation during life are small; however, approximately 20 donors have died worldwide since the first successful living renal donation was performed in 1954.

Reports have appeared in the literature to suggest an increased incidence of hypertension in kidney donors many years after donation. However, although an increase in protein excretion and glomerular filtration has been noted immediately following removal of a kidney, renal function has remained well preserved, without evidence of long-term deterioration. In the past decade, complications have been observed in only five of more than 200 living donors at the Massachusetts General Hospital (two pulmonary emboli, one wound infection, one urinary tract infection, and one halothane hepatitis). All of these patients recovered without incident and there have been no deaths.

The success of organ transplantation is driving a continued research effort into the future use of xenografts. As immunosuppression improves, the possibility of successful xenografts will compel a social consideration of this approach. The persistent need for organs is unlikely to be resolved if only human donors are considered.

A Collaborative Study. An appraisal of the criteria of cerebral death. JAMA, 1977; 237: 982–6.

Alexander JW, Vaughan WK. The use of ‘marginal donors’ for organ transplantation. Transplantation, 1991; 51: 135–40.

Anaise D, et al. An approach to organ salvage from non-heartbeating cadaver donors under existing legal and ethical requirements for transplantation. Transplantation, 1990; 49: 290–4.

Arras JD, Shinnar S. Anencephalic newborns as organ donors: a critique. JAMA, 1988; 259: 2284–7.

Ascher NL, Bolman R, Sutherland DER. Multiple organ donation from a cadaver. In: Simmons RL, ed. Manual of vascular access, organ donation and transplantation. Berlin: Springer-Verlag, 1984: 105–43.

Bay WH, Hebert LA. The living donor in kidney transplantation. Ann Intern Med, 1987; 106: 719–27.Belzer FO, Glass NR, Sollinger HW, Hoffmann RM, Southard JH. A new perfusate for kidney preservation. Transplantation, 1982; 33: 322–3.

Black P McL. Brain death (first of two parts). N Engl J Med, 1978; 299: 338–44.

Black P McL. Brain death (second of two parts). N Engl J Med, 1978; 299: 393–401.

Black P McL. Brain tumors (second of two parts). New Engl J Med, 1991; 324: 1555–64.

Caplan AL. Ethical and policy issues in the procurement of cadaver organs for transplantation. N Engl J Med, 1984; 311: 981–3.

Curran WJ, Hyg SM. The brain-death concept: judicial acceptance in Massachusetts. N Engl J Med, 1978; 298: 1008–9.

Dakshinamurty KV, Date A, Jacob CK, Kirubakaran MG, Shastry JCM. Hepatitis in living-donor renal allograft recipients. Transplantation, 1988; 46: 926–7.

Darby JM, Stein K, Grenvik A, Stuart S. Approach to management of the heartbeating ‘brain dead’ organ donor. JAMA, 1989; 261: 2222–8.

Diethelm AG. Ethical decision in the history of organ transplantation. Ann Surg, 1990; 211: 505–20.

Evans RW. Organ donation: facts and figures. Dialysis Transplant, 1990; 19: 234–40.

Friedland GH, Klein RS. Transmission of the human immunodeficiency virus. N Engl J Med, 1987; 317: 1125–34.

Goodman JL. Possible transmission of herpes simplex virus by organ transplantation. Transplantation, 1989; 47: 609–13.

Gottesdiener KM. Transplanted infections: donor-to-host transmission with the allograft. Ann Intern Med, 1989; 110: 1001–16.

Ho M, Suwansirikul S, Dowling JN, Youngblood LA, Armstrong JA. The transplanted kidney as a source of cytomegalovirus infection. N Engl J Med, 1975; 293: 1109–12.

Iwai A, Sakano T, Uenishi M, Sugimoto H, Yoshioka T, Sugimoto T. Effects of vasopressin and catecholamines on the maintenance of circulatory stability of brain-dead patients. Transplantation, 1989; 48: 613–17.

Kuo G, et al. An assay for circulating antibodies to a major etiologic virus of human non-A, non-B hepatitis. Science, 1989; 244: 362–4.

Laquaglia MP, et al. Impact of hepatitis on renal transplantation. Transplantation, 1981; 32: 504–7.

Levey AS, Hou S, Bush, Jr. HL. Kidney transplantation from unrelated living donors. N Engl J Med, 1986; 314: 914–16.

Levin SD, Whyte RK. Brain death sans frontieres. N Engl J Med, 1988; 318: 852.

Mani MK. Kidney transplantation from unrelated living donors. N Engl J Med, 1986; 315: 714.

Marsh CL, Perkins JD, Sutherland DER, Corry RJ, Sterioff S. Combined hepatic and pancreaticoduodenal procurement for transplantation. Surg Gynecol Obstet, 1989; 168: 254–8.

Medearis, Jr. DN, Holmes LB. On the use of anencephalic infants as organ donors. N Engl J Med, 1989; 321: 391–3.

Merrill JP, Murray JE, Harrison JH, Guild WR. Successful homotransplantation of the human kidney between identical twins. JAMA, 1956; 160: 277–82.

Novitzky D, Cooper DKC, Chaffin JS, Greer AE, DeBault LE, Zuhdi N. Improved cardiac allograft function following triiodothyronnine therapy to both donor and recipient. Transplantation, 1990; 49: 311–6.

Olthoff KM, et al. Comparison of UW solution and Euro-collins solutions for cold preservation of human liver grafts. Transplantation, 1990; 49: 284–90.

Pallis C. Brainstem death: the evolution of a concept. In: Morris PJ, ed. Kidney Transplantation: Principles and Practice 3rd edn. Philadelphia: W. B. Saunders, 1988; 123–50.

Pereira BJG, Milford EL, Kirkman RL, Levey AS. Transmission of hepatitis C virus by organ transplantation. N Engl J Med, 1991, 325: 454–60.

Pirsch JD, et al. Living-unrelated renal transplantation: results in 40 patients. Am J Kidney Dis, 1988; 12: 499–503.

Ploeg RJ. Kidney preservation with the UW and Euro-collins solutions. Transplantation, 1990; 49: 281–4.

Prottas J. Shifting responsibilities in organ procurement: a plan for routine referral. JAMA, 1988; 260: 832–3.

Rao KV, et al. Influence of cadaver donor age on posttransplant renal function and graft outcome. Transplantation, 1990; 49: 91–4.

Ratner LE, Flye MW. Successful transplantation of cadaveric en-bloc paired pediatric kidneys into adult recipients. Transplantation, 1991; 51: 273–5.

Report of the Ad Hoc Committee of the Harvard Medical School to examine the definition of brain death. A definition of irreversible coma. JAMA, 1968; 205: 85–8.

Report of the Medical Consultants on the Diagnosis of Death to the President's Commission for the Study of Ethical Problems in Medicine and Biomendical and Behavioral Research. Guidelines for the determination of death. JAMA, 1981; 246: 2184–6.

Ruder H, Schaefer F, Gretz, N, Mohring S, Scharer K. Donor kidneys of infants and very young children are acceptable for transplantation. Lancet, 1989; ii: 168.

Sadler AM, Sadler BL, Statson EB. ‘The uniform anatomical gift act’. JAMA, 1968; 206: 2501–6.

Shewmon DA, et al. The use of anencephalic infants as organ sources. JAMA, 1989; 261: 1773–82.

Singer PA, et al. Ethics of liver transplantation with living donors. N Engl J Med, 1989; 321: 620–2.

Southard JH, et al. Important components of the UW solution. Transplantation, 1990; 49: 251–7.

Starzl TE, et al. A flexible procedure for multiple cadaveric organ procurement. Surg Gynecol Obstet, 1984; 158: 223–30.

Stuart FP, Veith FJ, Cranford RE. Brain death laws and patterns of consent to remove organs for transplantation from cadavers in the United States and 28 other countries. Transplantation, 1981; 31: 238–44.

Sutherland DER, Goetz FC, Najarian JS. Pancreas transplants from related donors. Transplantation, 1984; 38: 625–33.

Task Force for the Determination of Brain Death in Children. Guidelines for the determination of brain death in children. Ann Neurol, 1987; 21: 616–17.

Teperman L, et al. The successful use of older donors for liver transplantation. JAMA, 1989; 262: 2837.

The Medical Task Force on Anencephaly. The infant with anencephaly. N Engl J Med, 1990; 322: 669–74.

Truog RD, Fletcher JC. Anencephalic newborns—can organs be transplanted before brain death? N Engl J Med, 1989; 321: 388–91.

Van Der Poel CL, et al. Infectivity of blood seropositive for hepatitis C antibodies. Lancet, 1990; 335: 558.