Show Summary Details
Page of

Death and dying 

Death and dying
Death and dying

Carl Waldmann

, Andrew Rhodes

, Neil Soni

, and Jonathan Handy

Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2021. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

date: 04 December 2021

Chapter contents

Confirmation of death using neurological criteria (brainstem death)

Death is defined as the irreversible loss of the capacity to breathe combined with the irreversible loss of the capacity for consciousness. Irreversible cessation of brainstem function produces this clinical state and equates with the death of the individual. The brainstem contains the cranial nerve nuclei, cardiorespiratory control centres, the reticular activating system, and the neural pathways that relay sensory and motor impulses between the periphery and the higher cortical centres. Destruction of these centres by any pathological process underlies the process of confirming death using neurological criteria (brainstem death).


Described in 1968 by the Harvard Medical Committee and adopted in the UK in 1976 following the Conference of the Royal Colleges and their Faculties, the permanent and irreversible loss of the function of the brainstem equates to the death of an individual, with the permanent functional death of the brainstem constituting brain death, a position since adopted by the courts of England and Wales. A UK Code of Practice for the confirmation and diagnosis of death was published by the Department of Health and Academy of Royal Medical Colleges in 2008.


Brainstem death may follow a severe global insult such as hypoxia, a direct insult such as trauma, or focal ischaemia resulting from vertebral/basilar artery pathology. The most common causes of brainstem death are traumatic brain injury, cerebrovascular accidents (including subarachnoid haemorrhage), cerebral tumours, meningitis, and hypoxic brain injury.

The process of confirming death by neurological testing


All of the following condition must be met before neurological testing can be undertaken.

  • The cause of the coma must be established as an irreversible structural cause of brain injury.

  • The patient must be deeply comatose, unresponsive, apneic, and dependent on mechanical ventilation.

  • The possibility of neuromuscular blockade should be excluded by means of a peripheral nerve stimulator or the eliciting of deep tendon reflexes. Ventilatory depression due to drugs or spinal cord injury must be excluded.

  • Potentially reversible causes of brainstem depression must be excluded: hypothermia (<34°C), sedative drugs, alcohol, circulatory, endocrine, and metabolic disturbances. Deviations outside normal values that are clearly a consequence of brainstem death rather than its cause do not preclude the diagnosis.

Establishing the irreversible loss of brainstem reflexes

This may only take place in a setting when all preconditions described in the section on ‘Preconditions’ have been met.

  • The pupils are fixed with no direct or consensual response to light. The pupils need not be dilated or equal in size and shape for the diagnosis to be confirmed.

  • Bilateral corneal reflexes are absent.

  • There is no motor response within the cranial nerve distribution to a painful stimulus applied centrally or peripherally. Spinal reflexes may persist in patients confirmed dead according to neurological criteria.

  • The oculovestibular (caloric) reflex is absent. This is performed by injecting 50 ml of ice-cold water over 1 min into each external auditory meatus whilst the eyelids are held open. There should be no eye movement, principally nystagmus, during or after the injection, and the eyes should be observed for 1 min after each injection. Direct access to the tympanic membrane must be confirmed on each side using an auriscope, and the head should be flexed at 30° to the horizontal plane. The diagnosis is not invalidated if it cannot be performed on one side because of localized trauma.

  • There should be no gag reflex or cough reflex in response to a suction catheter being passed into the pharynx or down the endotracheal tube (ETT) to stimulate the carina.

Apnoea testing

The responsiveness of the respiratory system to hypercarbia is the final test performed to confirm brainstem death, and should only be performed if all the preceding tests confirm the absence of brainstem function. It is recommended that end-tidal carbon dioxide (ETCO2) monitoring is utilized to minimize the risk of excessive hypercarbia and/or rapid changes in CO2 tension, so limiting the risk of further injury to potentially recoverable brain tissue. The patient should be preoxygenated with 100% oxygen for 5 min. With a peripheral capillary oxygen saturation (SpO2) >95%, minute volume ventilation is reduced and the ETCO2 allowed to rise slowly. Once the ETCO2 is >6.0 kPa, a level high enough to exceed the normal threshold for stimulation of respiration, an arterial blood gas (ABG) is taken to confirm that the PaCO2 is >6.0 kPa. If cardiovascular stability is maintained, the patient should then be disconnected from the ventilator for 5 min. Hypoxia is prevented by insufflating oxygen at 5 l/min via a suction catheter placed down the ETT with continual SpO2 monitoring. While disconnected, the patient should remain apnoeic and is continually observed for any respiratory effort (abdomen and chest exposed). If, after 5 min, there has been no spontaneous respiratory effort, a further confirmatory ABG is taken to ensure that the arterial partial pressure of carbon dioxide (PaCO2) has increased from the starting level by >0.5 kPa. The ventilator should then be reconnected and the minute volume returned to the setting that allows blood gases to return to the target value for the patient’s further management. In patients with pre-existing chronic respiratory disease who rely on a hypoxic drive to respiration, a mild but significant acidosis should be achieved by allowing the PaCO2 to rise to a point where the pH is <7.35.

Timing of testing

Testing should not be considered until at least 6 h after the onset of apnoeic coma or loss of the last cranial nerve reflex. When the cause is hypoxia following a cardiac arrest or severe circulatory insufficiency, brainstem testing should only be undertaken when there is no possibility of a reversible or treatable underlying cause being present. More time before testing may also be required in patients with neuromuscular disorders, in patients receiving steroids for space-occupying lesions, or those receiving prolonged fentanyl infusions, and those with lesions primarily located in the brainstem.

Conduct of testing

Neurological testing to confirm death should be performed by two medical practitioners, both of whom should have held full registration with the General Medical Council (GMC) for at least 5 years. One should be a consultant, and both should be competent in the procedure. Neither should have any connection to the transplant team. The doctors may perform the tests independently or together. There is no statutory time interval that must pass between the two sets of tests; it is a matter of clinical judgement to decide when sufficient time has elapsed between tests to satisfy all those involved. Two sets must always be undertaken to remove the risk of observer error and to reassure the family.

Other considerations

  • Death is confirmed after completion of the second set of tests.

  • The time of death is then recorded as the time at which the first set of tests was completed.

  • Documentation of the results should be on the appropriate form, approved by the Faculty of Intensive Care Medicine.

  • The coroner should be informed of most of these patients owing to the underlying diagnosis, mechanism of injury, or if organ donation is to take place.

  • Effective communication with relatives enables understanding of the indications for and components of brainstem death testing and its implications. Some relatives find it helpful to observe the tests and this may help alleviate emotional distress.

Ancillary tests

No additional tests are required in the UK to confirm brainstem death, although ancillary tests such as computed tomographic angiography, transcranial Doppler, or radioisotope scanning may be used to establish the absence of cerebral bloodflow in cases of diagnostic uncertainties. Electroencephalography or brainstem evoked potential testing are not usually helpful in establishing the diagnosis.


  • Depending on the underlying pathology, there may be uncertainty as to when the brain injury can be considered irreversible and brainstem testing can take place.

  • With altered pharmacokinetics in critically ill patients, uncertainty may surround the length of time that has to elapse before the effects of sedative agents, particularly thiopental.

  • No specific ranges can be set within which biochemical variables must fall to be considered ‘normal’ and compatible with brainstem death testing, although recommendations as such have been made by the Academy of Royal Medical Colleges.

Subsequent actions

After confirming death using neurological criteria, subsequent actions depend on whether patients had expressed a wish to donate their organs after death. If so, the maintenance of normal physiological parameters is vital to ensure optimal organ viability. This includes correcting the physiological consequences of brainstem death, including cardiovascular instability (hypotension and arrhythmias), hypothermia, diabetes insipidus, disseminated intravascular coagulation, hyperglycaemia, acidosis, and other metabolic and endocrine abnormalities. If organ donation is not an option, then, following consultation with the relatives, withdrawal of respiratory support may take place at a time acceptable to all parties.

Withdrawing and withholding treatment

Across Europe, 13.5% of patients admitted to an intensive care unit (ICU) die. Of these patients who are going to die, 72.6% had limitations applied to their treatment. This equates to 10% of all ICU admissions in this study population. Therefore, decisions about withdrawing and withholding treatment occur constantly. In the European Society of Intensive Care Medicine (ESICM) ‘Ethicus’ paper, it was argued that there were three methods that could be adopted by intensivists that could lead to a patient’s death:

  1. 1. Withholding treatment was defined as a decision that was made not to start or increase a life-sustaining intervention.

  2. 2. Withdrawing treatment was defined as a decision that was made actively to stop a life-sustaining intervention presently being given.

  3. 3. Active shortening of the dying process was defined as a circumstance in which someone performed an act with the specific intent of shortening the dying process. This is unlawful in the UK, but not in certain jurisdictions in Europe and elsewhere in the world. Certain authors hold that the ‘double effect’ may fall into this category.

Generally (with some exceptions, e.g. the Jewish faith), it is argued that withdrawing and withholding treatment are morally and ethically the same.

In the USA, the President’s Commission for the Study of Ethical problems in Medicine and Biomedical Research has said, ‘[The] healthcare professional has an obligation to allow a patient to choose from among the medically acceptable treatment options … or to reject all options. No one, however, has an obligation to provide interventions that would, in his or her judgement, be counter-therapeutic.’ It is establishing whether treatment options are countertherapeutic, or futile, that is difficult.


One way of establishing whether a treatment is countertherapeutic is by using the concept of futility. Unfortunately, this is difficult to define, with the courts saying, ‘Futility is a subjective and nebulous concept which, except in the strictest physiological sense, incorporates value judgements.’ Futility can be defined in a number of ways, including (and modifying) Ardagh’s definitions.

  • Physiological futility exists when a procedure cannot bring about its physiological objective. For example, it is physiologically futile if the administration of adrenaline to a hypotensive patient fails to result in an increase in blood pressure.

  • The benefit-centred definition of futility has been defined as consisting of quantitative and qualitative considerations. First, the quantitative estimate of futility is one in which an intervention is considered futile if it has failed in the last defined number of times attempted. For example, should a condition produce a predicted mortality of 99.9%, then 1 patient in 1000 will live if treated. It is not possible to determine which patient will survive and which one will die in advance of treatment. In this example, treatment is futile for 999 patients, but not futile for one individual who could survive. The qualitative component is defined as occurring where the patient’s resulting quality of life falls well below the threshold considered minimal by general professional judgement. Note that possible considerations of the patient may be ignored, which is precisely the situation that worried Mr Burke so that he brought a case against the GMC. Mr Burke requested a declaration that it would be unlawful to withdraw treatment (in the form of artificial nutrition and hydration) against his wishes. The Court of Appeal ruled against him, saying, ‘Ultimately, however, a patient cannot demand that a doctor administer a treatment which the doctor considers is adverse to the patient’s clinical needs.’

  • Operationalizing or cost-based futility. This definition of futility is when treatment is so unlikely to succeed that many people, both professional and lay persons, would consider it not worth the cost. Cost is relative and depends on society’s needs and wishes, e.g. in the aftermath of the Ladbroke Grove rail crash, it has been estimated that the Train Passenger Warning System installed would save ~38 lives over 25 years, at a cost of £15.4 million per life saved. In the UK, case law suggests that, where a treatment is felt to be clinically appropriate, then it may not be withheld solely on the basis of that cost.

Doctors rarely use the physiological definition of futility in common practice. The latter two definitions of futility mentioned above follow utilitarian principles such as those espoused by Savulescu and Shaw and advocated by the US President’s Commission (mentioned in the section on ‘Withdrawing and withholding treatment’), but incorporate value judgements. These value judgements may be subject to challenge by relatives, those with power of attorney, and the courts. It has been argued that, ‘The principle that each individual is entitled to an equal opportunity to benefit from any public health care system, and that this entitlement is proportionate neither to the size of their chance of benefiting, nor to the quality of the benefit, nor to the length of lifetime remaining in which that benefit may be enjoyed.’

Best interests

If futility is not as useful as it first appears, the best interests of the patient can be considered.

Competent patients

  • A competent patient (see section on ‘Organ donation after brain death’) should be involved with the decision and a discussion should be held with the patient. As Harris has said, ‘Each person’s desire to stay alive should be regarded as of the same importance and as deserving the same respect as that of anyone else, irrespective of the quality of their life or its expected duration.’

  • There are some individuals who wish to die rather than continue treatment. Recent examples in medicine include the cases of Diane Pretty and Ms B.

  • In justification of this wish to die, consider Kuhse’s thought experiment: ‘A truck driver and his co-driver had an accident on a lonely stretch of road. The truck caught fire and the driver was trapped in the wreckage of the cabin. The co-driver struggled to free him, but could not do so. The driver, now burning, pleaded with his colleague—an experienced shooter—to take a rifle, which was stowed in a box on the back of the truck, and shoot him. The co-driver took the rifle and shot his colleague. Was what the co-driver did morally reprehensible? Did he act wrongly? Students who are presented with this case will generally answer both questions in the negative. The reasons for their intuitions are not hard to find. In this case, the agent was not motivated by personal gain, but by compassion.’

  • If we accept this, then we can accept the underlying premise that it is sometimes in the best interests of the patient to die.

Patients who lack capacity

If the patient is not competent, then it is important to consult widely to discover the person’s past and present wishes and feelings, the beliefs and values that would be likely to influence his or her decision, and any other factors that heor she would be likely to consider. Once this has been done, the doctor can then consider whether further treatment of this patient is justified or whether it is in the patient’s best interests to cease treatment and be permitted to die. The key is to consider the individual concerned and not merely the pathology.


Although every case of withholding or withdrawing treatment is different, it is sensible to have departmental guidelines drawn up to guide inexperienced practitioners. The emphasis of care changes from one of cure to one of symptom control. It is likely that analgesics and anxiolytics are continued. It is further likely that other agents such as antisialogogues are added. The precise treatment will vary according to the individual patient.


If we get it wrong, then we risk the censure of the courts and the GMC.

  • An intensive care doctor withdrew ventilatory support from a patient and administered 20 mg diazepam. The patient subsequently died. Consequently, the doctor was struck off the medical register for gross professional misconduct. The Fitness to Practise Panel found proven that her actions were, ‘clinically unjustified, inappropriate, premature and not in the patient’s best interests’.

  • The doctor involved came to the view that further treatment of the patient was futile because the patient was not responding to her treatment and his ventilatory requirements were increasing. Despite disagreement from the family, the doctor withdrew treatment.

  • The doctor discussed her opinion with the patient’s family, but appears not to have discussed the matter with the patient. Yet, he ‘was not in imminent danger of dying and was recorded as being conscious and orientated’. The doctor knew, however, that the family did not wish treatment to be withdrawn. A family’s requests are not binding on clinicians because doctors have to promote their patient’s best interests. Nevertheless, the GMC specifically censured the doctor for failure to treat the patient in his best interests.

Consider the busy hospital, with high pressure on beds; the consultant, unable to take time to establish a rapport with the family; working with limited access to an appropriate peer group. There are a number of factors that could help minimize the occurrence of unfortunate episodes.

  • Doctors should establish local, regional, and national networks of support to which to turn for advice.

  • The counsel of a trusted colleague is invaluable in clinical practice.

  • It has been suggested that the presence of a hospital-based clinical ethics committee would help clinicians and families in this terrible position. Such a forum would facilitate dialogue between the parties, but not provide a didactic answer.

Finally, there may be a way to resolve these conflicts through legislation. In the state of Texas, the Texas Advance Directives Act 1999 provides that there is a mandated period of 10 days in which care may not be withdrawn. During this time, doctors and relatives attempt to find an alternative hospital that would be prepared to continue treating the patient. Only if no medical facility is found can there be any consideration of withdrawal of medical care against the wishes of the family. The experience of this Act suggests that most disagreements between families and medical staff are resolved in the 10-day ‘cooling off’ period.


The pattern of death in modern society is changing from occurring at home to occurring in hospital. The general practitioner will refer the patient to hospital when the patient’s condition is not able to be managed in a home environment. Once in hospital, should the patient’s condition worsen, he or she becomes more likely to be referred to ICU. It is likely that more people will die on ICU as an ever-increasing number are referred. Death is traditionally perceived by the medical profession as a failure, as opposed to being seen as in the best interests of the patient. If death is accepted as a best interest, then the management of that death is paramount. Quality of death is as important as quality of life. Intensivists have been given a new goal: to provide excellence in end-of-life care. We must be ready to meet this challenge.

Organ donation after brain death (DBD)

Organ transplantation is one of the most significant medical advances of recent times, with >90% of organ recipients surviving 1 year. This success has led to a situation worldwide whereby demand for organs outstrips supply. The transplantation of organs in suboptimal conditions is a significant cause of graft failure. It is therefore vital that comprehensive physiological support is given to a potential organ donor to ensure that donated organs reach the recipients in optimal condition. Clinicians caring for an organ donor have a duty of care and obligation to both the donor and the recipient to ensure that this is the case.

Identification of the potential organ donor

Organ donation is widely supported, with approximately 21 million (33%) of the UK’s population actively expressing a wish to donate their organs after death by registering on the UK Transplant’s Organ Donor Register. Few of these people will die in an ICU and, of those who do, only 6% will have death confirmed using neurological criteria. To ensure that the wishes of dying patients are met, any patient under the age of 85 years who has been confirmed dead using neurological criteria should be considered as a potential organ donor and their relatives should be offered the option. There are very few absolute contraindications to organ donation:

  • Presence of malignancy (except certain primary tumours of the central nervous system).

  • Systemic sepsis.

  • Confirmed diagnosis of human immunodeficiency virus.

  • Known or suspected classical or variant Creutzfeldt–Jakob disease.

To avoid missing potential organ donors, all such patients should be discussed with the specialist nurse in organ donation (SNOD) regarding suitability, as ‘marginal’ organs may still be transplanted if the clinical situation of the recipient demands it. Until recently hepatitis B- or C-positive donors were not considered suitable for donation, but organs from these patients are now transplanted into hepatitis B- or C-positive recipients.

Approaching a potential donor’s family

Once a potential donor has been identified, the SNOD should be notified as early as possible so that the suitability of the patient can be determined, avoiding the situation whereby a patient’s relatives are approached with a request for organ donation only to find subsequently that they are not suitable. They will also check the organ donor register to see if the patient had registered their wish to donate their organs after death. Permission for donation may be required from the coroner or procurator fiscal, depending on the circumstances of the death, and should be sought prior to approaching the donor’s relatives. There are three key stages to approaching the family of a potential organ donor:

  1. 1. Planning.

  2. 2. Confirming that a family have understood and accepted their loss (breaking bad news) before donation is raised with the family.

  3. 3. Discussing donation.

Professional guidance advocates that a multidisciplinary team is involved in planning the approach and discussing organ donation with the patient’s relatives. The medical and nursing staff involved in the care of the patient and the SNOD should routinely be involved and a faith representative, when appropriate. No member of the transplant team should be involved with the request for organ donation. Relatives should be fully informed of progress at every stage, and time taken to explain the need for ongoing therapeutic interventions aimed at optimizing the function of potentially transplantable organs following the confirmation of death.

Pathophysiological changes after death according to neurological criteria

Following brainstem death, potentially transplantable organs are subject to a combination of cardiovascular instability and hormonal changes, which can compromise organ viability and post-transplant function (Table 33.1). It is therefore imperative that potential organ donors are actively managed physiologically to maximize the chance of successful organ retrieval and subsequent transplantation.

Table 33.1 Incidence of pathophysiological changes after death according to neurological criteria



Diabetes insipidus


Disseminated intravascular coagulation


Cardiac arrhythmias


Pulmonary oedema


Metabolic acidosis


Cardiovascular changes

Impending brainstem death is associated with intense autonomic activity and massive catecholamine release. The resulting increases in heart rate, blood pressure, cardiac output, and peripheral vascular resistance lead to widespread myocardial ischaemic damage. Impaired adenosine triphosphate (ATP) production, increased free radical production, and defective oxidative metabolism exacerbate the cellular damage and worsen myocardial function. ST segment and T-wave changes are common, along with atrial or ventricular arrhythmias and conduction abnormalities. These changes reflect loss of vagal tone, sympathetic overactivity, myocardial ischaemia, electrolyte abnormalities, and/or the administration of exogenous catecholamines. As vasomotor regulation is lost, profound vasodilatation and myocardial depression result, with consequent hypotension. This hypotension may be potentiated by other factors (Box 33.1). Asystolic cardiac arrest is likely to occur within days if no other organ support except mechanical ventilation is provided.

Pulmonary changes

Pulmonary injury may result from trauma, fluid overload, ventilator-induced lung injury, infection, atelectasis, oedema, or as part of any other coincidental disease process. Neurogenic pulmonary oedema resulting from elevated pulmonary capillary hydrostatic pressure as a result of acute left ventricular dysfunction and increased capillary permeability as a result of a generalized proinflammatory state is common.

Endocrine changes

Failure of the anterior and posterior pituitary leads to reduced circulating levels of antidiuretic hormone, cortisol, T3, and T4. Diabetes insipidus occurs in up to 65% of potential organ donors and is characterized by diuresis, hypovolaemia, and metabolic derangement (plasma hyperosmolality, hypernatraemia, hypomagnesaemia, and hypocalcaemia). Lack of T3 may compound cardiovascular collapse through depletion of high-energy phosphates, whilst a reduced blood cortisol level impairs the donor stress response. Hyperglycaemia resulting from reduced insulin secretion, the administration of exogenous catecholamines to counter hypotension, and the administration of glucose-containing fluids to manage hypernatraemia is common. Left untreated, hyperglycaemia leads to increased plasma osmolality, metabolic acidosis, and osmotic diuresis with consequent cardiovascular instability.

Temperature regulation

Hypothermia is a common feature and results from loss of hypothalamic temperature-regulating mechanisms, a reduction in heat production owing to a decreased metabolic rate and increased heat loss because of peripheral vasodilatation.

Haematological changes

The release of tissue thromboplastin and other mediators from severely damaged brain tissue may cause a coagulopathy, disseminated intravascular coagulation being present in 30% of organ donors.

Immunological changes

The process of brainstem death alters the immunogenicity of organs by a variety of poorly understood mechanisms. This may explain why acute rejection episodes are more frequent and more severe in recipients of organs from heart-beating donors compared with those receiving organs from living donors.

Clinical management of the potential DBD

Once death has been confirmed using neurological criteria, there is an alteration in the emphasis of patient care. Whereas therapies have previously been aimed at preserving brain function and life, they are now directed to optimize the function of transplantable organs. High-quality routine general intensive care measures should continue, including strict asepsis, nutrition, regular tracheal suctioning, physiotherapy, turning, and the active treatment of infection or arrhythmias. Additionally, specific therapies to control or reverse the pathophysiological abnormalities that may damage or impair the function of transplantable organs should be instigated. Invasive monitoring should be continued or commenced. If new lines are inserted, the arterial cannula should ideally be sited in the left radial or brachial artery and central venous in the right internal jugular vein as a result of the order in which the great vessels are ligated during organ retrieval. Donor optimization led by an intensivist can increase the number of transplantable organs from DBD donors.

Cardiovascular support

Blood pressure and fluid management are key components of a donor optimization strategy that will lead to successful organ retrieval and transplantation. Suggested haemodynamic targets are:

  • Cardiac index 2.2–2.5 l/min/m–2.

  • Mean arterial pressure (MAP) >70 mmHg.

  • Pulse rate 60–100/min, ideally sinus rhythm.

Haemodynamic management combines careful fluid management with the administration of vasoactive agents to target a MAP of >70 mmHg. Optimal monitoring techniques remain undefined and choice of modality will be guided by local protocols and personal preference. Serial echocardiography may be utilized to observe left ventricular performance and assessment of filling status. Care must be taken to avoid excessive fluid administration, which in the setting of raised pulmonary capillary hydrostatic pressure and increased capillary permeability can exacerbate pulmonary oedema, with a consequent negative effect on the potential for lung transplantation. The use of protocolized fluid regimens has not been shown to have benefit in terms of number of organs donated or recipient mortality. Evidence for a primary vasoactive agent is lacking but vasopressin is increasingly considered to be the vasoconstrictor of choice. Unlike other inotropes and vasoconstrictors, it does not deplete myocardial ATP levels whilst increasing blood pressure, maintaining cardiac output, and countering diabetes insipidus. If, prior to death, a patient was receiving noradrenaline, then a switch to vasopressin should be considered after confirmation of death. If, despite fluid and vasopressin administration, a target MAP cannot be achieved, then second-line vasoactive agents such as noradrenaline can be added, although high-dose noradrenaline infusions (>0.05 µg/kg) are associated with cardiac graft dysfunction. Prolonged administration of exogenous catecholamines should be avoided, as rapid depletion of myocardial ATP stores and detrimental vasoconstriction in donor organs may result. If high doses of vasoactive drugs are required, ‘hormone replacement’ (see section on ‘Endocrine support’) may be considered. Arrhythmias are common and should be aggressively treated, along with any associated hypokalaemia or hypomagnesaemia.

Respiratory support

If the lungs are being considered for transplantation then, to minimize the risk of oxygen toxicity, the fractional inspired oxygen concentration should be kept as low as possible to achieve a PaO2 >10 kPa. Early bronchoscopy, frequent suctioning, and the use of protective lung ventilation strategies should be employed to limit the incidence of ventilator-induced lung injury and maximize the chance of successful lung transplantation. A PEEP of 5–10 cmH2O will prevent alveolar collapse, whilst higher levels may induce hypotension in an inadequately fluid-resuscitated patient. The judicious use of fluid resuscitation to ensure end-organ perfusion while minimizing the accumulation of extravascular lung water will prevent any exacerbation of pre-existing pulmonary oedema and worsening of the arterial alveolar oxygen gradient.

Endocrine support

Diabetes insipidus occurs in up to 65% of patients following brainstem death, leading to hypovolaemia and hypernatraemia. Correction requires administration of appropriate intravenous (IV) fluid therapy and a vasopressin infusion or desmopressin. Desmopressin, which acts preferentially on vasopressin-2 receptors in the collecting tubules, is associated with a lower incidence of acute tubular necrosis and graft failure than vasopressin and is preferred unless vasopressin is required for blood pressure support.

Hyperglycaemia results in increases in counter-regulatory hormones, changes in carbohydrate metabolism, the infusion of glucose-containing solutions, reduced insulin secretion, and peripheral resistance to insulin. This should be treated with an insulin infusion as per standard intensive care protocols.

Other hormone replacement should be considered in organ donors with cardiovascular instability, escalating vasoactive drug requirements, and a worsening acidosis.

  • Some cardiothoracic transplant centres recommend using T3 to reduce inotrope requirements and help restore cardiovascular stability. T3 administration may be associated with an increase in the number of retrieved organs and improved function post-transplantation; however, routine administration is not currently recommended.

  • The administration of high-dose methylprednisolone has been advocated to reduce vasoactive drug requirements, to attenuate the effects of proinflammatory cytokines, and to reduce the accumulation of extravascular lung water. It improves oxygenation in the donor and is associated with increased rates of successful lung retrieval and transplantation, although a recent systematic review found a lack of robust evidence to justify routine administration.

On the basis of the failure of the hypothalamopituitary axis, the concept of ‘hormone resuscitation’ involving the routine administration of vasopressin, methylprednisolone, and T3 has been advocated as its use may be associated with an increase in the possibility of the kidney, heart, liver, lung, and pancreas being transplanted. There is also an improvement in the short-term function of transplanted hearts. Exactly which constituents carry the most benefit has not been established, so whether all or some of vasopressin, T3, or methylprednisolone are administered is determined by the preferences of the transplant centre involved.

Renal support

Renal perfusion pressure should be optimized by maintaining a MAP >70 mmHg, with fluid resuscitation and/or vasoactive drugs. Episodes of hypotension should be treated aggressively as they are associated with acute tubular necrosis and failure of the transplanted kidney.

Haematological support

Disorders of blood coagulation are common following the release of thromboplastin and other mediators from necrotic brain tissue. Combined with hypothermia, acidosis, and the dilution of coagulant factors following IV fluid therapy, a profound coagulopathy can result. This should be rapidly corrected by administering blood products in line with local policies. Heparin administration may be requested by some transplant centres to limit the formation of microthrombi within organs but its benefit is unproven. Four units of blood should be available prior to multiorgan procurement since the surgery may involve significant blood loss.

Temperature support

Traditional guidance has always been to reverse the hypothermia commonly seen in patients following brainstem death, targeting normothermia by increasing the ambient temperature, heating and humidifying inspired gases, administering warmed IV fluids, and using forced air warming blankets. However, a recent study found that a temperature of 34–35°C was associated with a significant reduction in the requirement for dialysis in the first 7 days after kidney transplantation when compared to normothermia. Whether this is generalizable to other organs is unknown, and further research is required before a recommendation to target moderate hypothermia prior to organ retrieval following the confirmation of death by neurological criteria can be made.

Other considerations

Following confirmation of death, documentation should be continued as previously. The time of death is recorded as the time that the first set of tests confirming death according to neurological criteria is completed. If a post mortem is to be performed, local policy may dictate that intravascular cannulae, the ETT, and the urethral catheter are left in place; otherwise, they may be removed. Last offices are performed by the nursing staff with help from the SNOD, and the relatives should be given the opportunity to view the body after retrieval surgery if they wish to do so.

Meeting the wishes of individuals and their relatives to donate organs after death is an important aspect in bereavement care. The attitudes of medical staff in explaining the need for ongoing therapies despite the donor being declared dead will lead to greater understanding on the part of the relatives, many of whom gain comfort from respecting their relative’s wishes after death and consolation in knowing that some good has come from their loss. Psychological and pastoral support offered to relatives by the ICU team and the SNOD may help to alleviate feelings of guilt, anger, or remorse. They should be invited to return and discuss any issues with ICU staff or the SNOD in the future. It is common practice for relatives to be informed of outcomes of any organs retrieved from the donor.

Organ donation after circulatory determination of death


Organ transplantation offers patients with end-stage organ failure an improved quality of life and increased life expectancy. The practice of organ donation following death confirmed by circulatory criteria (DCD) involves organ retrieval from donors following cardiorespiratory arrest rather than retrieval after confirmation of death by neurological criteria. This is not a new concept. Prior to 1968, when the Harvard Medical Committee recognized that death resulted from irreversible damage to the brainstem and the introduction in 1976 of clinical tests to confirm death using neurological criteria, organs for transplantation were routinely retrieved from donors following cardiorespiratory arrest. The result of this declaration was that transplant centres rapidly switched to transplanting organs from patients confirmed dead using neurological criteria as organs were not subject to a warm ischaemic time, and the practice of DCD rapidly declined. However, with the increasing demand for transplantable organs and the demonstration that kidneys retrieved from DCDs have the same long-term outcome as those retrieved from donors following brainstem death, DCD donation has been reintroduced and contributes a substantial proportion of transplanted organs.


In the UK, the demand for transplantable organs continues to outstrip supply, with over 6044 people waiting for an organ transplant at the end of March 2018. Contributory factors to the growing demand for organs include:

  • An ageing population.

  • An increase in the prevalence of renal failure.

  • Advances in transplant technology.

While the demand for transplantable organs increases, the actual number of donors proceeding following neurological death is declining. This pattern is likely to continue for two reasons:

  • Fewer young people are dying of a traumatic brain injury or catastrophic cerebrovascular event owing to the combined effect of public safety campaigns and advances in safety-related technology.

  • Improvements in the ICU management and outcome of traumatic brain injuries means that fewer people fulfil neurological criteria for confirming death.

In addition, 41% of relatives approached with a request for organ donation decline the offer, meaning that not all potential donors are converted into actual donors. The reintroduction of DCD schemes forms part of a wider strategy to address the imbalance between organ supply and demand. Not only do they provide an opportunity for patients to become organ donors even if they do not fulfil the criteria for neurological death and provide obvious benefit to the recipient, they also allow relatives to witness the observable ending of life as represented by cessation of the heart beat.

Controlled organ DCD


Potential DCDs can be divided into five categories according to the modified Maastricht classification (Table 33.2). Categories I, II, and V are described as ‘uncontrolled’ and categories III and IV as ‘controlled’. The ‘controlled’ aspect refers to the fact that the patient’s death is anticipated and therefore the donation process can be better planned and coordinated. In the ICU setting, DCDs are usually controlled.

Table 33.2 The modified Maastricht classification of DCDs

Category I

Dead on arrival

Category II

Unsuccessful resuscitation

Category III

Anticipated cardiac arrest

Category IV

Cardiac arrest in a brainstem dead donor

Category V

Unexpected cardiac arrest in an ICU patient

Reproduced from Continuing Education in Anaesthesia Critical Care & Pain, 11, 3, Dunne K., Doherty P. Donation after circulatory death, pp. 82–86, Copyright © Elsevier 2011.

Potential DCD

Patients suitable for DCD are typically, but not exclusively, those who have experienced a catastrophic brain injury but who do not fulfil the neurological criteria for death. They are dependent on life-sustaining support, yet continued medical intervention is not considered to be in the patient’s overall best interests. As a result, withdrawal of active treatment has been planned and it is anticipated that cardiorespiratory arrest will follow swiftly and predictably. Patients with other diagnoses in whom treatment withdrawal is planned may also be suitable and should also be discussed with the SNOD to assess suitability. Absolute contraindications to DCD are the same as those described for donation following death according to neurological criteria.

Organs suitable for donation

Kidneys are relatively tolerant of warm ischaemia and, other than a longer requirement for post-transplant dialysis, achieve similar long-term outcomes to those retrieved from donors verified as dead according to neurological criteria. Other organs retrieved from DCDs that are successfully transplanted include the liver, pancreas, lungs, and, more recently, the heart.

Withdrawal of active treatment

The decision to withdraw life-sustaining therapy should be made in consultation with the patient’s relatives and the referring medical team in accordance with guidelines from the Intensive Care Society, the British Medical Association, and the GMC. It is a decision that must be made entirely separate from any consideration of organ donation. To avoid a conflict of interest, no member of the transplant team should be involved in any aspect of the decision to withdraw treatment or how the withdrawal is managed. Agreement on the exact timing of treatment withdrawal is then reached between the critical care team, the patient’s relatives, and the retrieval team. The dignity, wellbeing, and comfort of the dying patient remain paramount and the process itself should take place in line with local policy. A significant number of DCDs undergoing withdrawal of life-supporting treatments have the potential of progressing to brain death, and consideration should be given to delaying the treatment withdrawal to allow this to happen, if acceptable to the relatives, since more organs are transplanted from DBD than DCD.

There should be no difference in the process of withdrawal because organ donation is being considered, although there will necessarily need to be manipulation of the timing of withdrawal to facilitate donation. This may involve stopping mechanical ventilation, vasoactive drug support, supplementary oxygen, and/or removing the ETT. Sedative or opioid infusions should be started as appropriate. Where possible, the process of withdrawal should take place within the ICU. In circumstances where the ICU is too distant from theatres for DCD to be possible, treatment withdrawal may have to take place in the theatre complex. If treatment withdrawal does occur outside an ICU environment, the same level of critical care nursing expertise in the management of the dying patient must be provided.

Once a decision to withdraw life-sustaining therapy has been made, antemortem interventions that are of no benefit to the dying patient but are designed to improve organ viability can be considered. Examples include the administration of heparin or the placement of vascular cannulae to facilitate rapid cold perfusion. Such interventions, however, are not currently recommended routine practice in the UK. Blood sampling from indwelling intravascular lines may be undertaken to allow tissue typing and serology purposes.

Once life-sustaining therapy is withdrawn, the transplant team need to be kept informed of any prolonged periods of hypotension (systolic blood pressure <50 mmHg), hypoxia (arterial oxygen saturation <80%), or anuria, so that they can assess whether organ donation remains a viable option. The period of time following withdrawal of life-sustaining therapy within which cardiorespiratory arrest must occur to allow organ donation to proceed is organ-specific. For the liver, this time period must not exceed 1 h yet for the kidney in some cases it can extend up to 4 h. If death has not occurred within this time frame, organ donation is abandoned while palliative care is continued.


Discussion as to the suitability of the patient as a potential DCD should take place with the SNOD before approaching the relatives. This avoids the relatives agreeing to the patient becoming an organ donor only to find subsequently that they are not suitable. Even if concerns exist about contraindications, potential donors should be referred, leaving the decision about suitability to the transplant team. Agreement from the coroner/procurator fiscal should also be obtained (if necessary) before any approach to the family. A planned, collaborative approach to relatives should be undertaken and DCD should not be brought up until the family have understood and accepted the futility of continuing life-sustaining therapy. The practicalities of DCD should be explained, with emphasis on the following points:

  • Death may occur quickly after treatment withdrawal and that, if organ donation is to be possible, the time relatives can have with their relative at this stage is limited.

  • If the dying process is prolonged, organ donation may not proceed owing to an unacceptably long warm ischaemic time.

  • Retrieved organs may not always be suitable for transplantation if perfusion has failed or they are damaged during the retrieval process.

  • There will be a further opportunity for relatives to spend time with their loved one after organ retrieval has taken place.

  • The relatives can stop the donation process at any stage.

The SNOD has a central role in facilitating effective communication between the critical care team, retrieval team, operating theatre staff, and relatives. The SNOD is responsible for ensuring a smooth transition from withdrawal of life-sustaining therapy to confirmation of death and organ retrieval.

Confirming death using circulatory criteria

Once cardiorespiratory arrest has occurred after withdrawal of life-sustaining therapy, death is confirmed after a 5-min period of continuous asystole, apnoea, and unconsciousness as recommended by the Academy of Medical Royal Colleges. In the ICU, this observation is facilitated by the continued use of existing monitoring modalities, including electrocardiograph and intra-abdominal pressure. Any return of cardiac or respiratory activity during this period prompts a further 5 min of observation from that point. After 5 min, a lack of response to supraorbital pressure and the absence of pupillary and corneal reflexes is confirmed.

Management of the DCD after death

Once death has been confirmed, relatives may spend up to 5 min with the patient before transfer to the operating theatre for organ retrieval. This period is necessarily brief to minimize the warm ischaemic time. Should relatives require a longer period of time with the patient then the donation process should be reviewed and, if necessary, abandoned. Post-mortem interventions such as chest compressions and full cardiopulmonary bypass designed to reduce the warm ischaemic time risk inadvertently restoring the cerebral bloodflow and should not be initiated. Once the retrieval process has been completed, relatives are given the opportunity to spend a further period of time with the deceased if required.

Care of relatives of the DCD

It is central to the principles of organ donation that donation be carried out to meet the wishes of the deceased and also to bring comfort to the relatives. All medical staff have a role in offering support to the relatives before, during, and after the donation process. Some relatives may wish to be present at the time of death following withdrawal of life-sustaining therapy, and this option should be incorporated into DCD protocols. Pastoral care and bereavement counselling should be made available to relatives if required.


Despite the endorsement of DCD programmes by professional and regulatory bodies, some critical care practitioners remain concerned about the ethics and legality of aspects of DCD. The areas of concern revolve around reaching a decision that continued treatment is futile, the associated potential for conflict of interest in the context of DCD, uncertainties over the time at which death can be confirmed using cardiorespiratory criteria, and the possibility of the spontaneous return of cardiac function after asystole (the Lazarus phenomenon). These issues have all been addressed in a recent Department of Health publication.

Multiple choice questions and further reading

Death and dying Interactive multiple choice questions to test your knowledge on this chapter and additional further reading can be found in Appendix Chapter 33 Multiple choice questions and further reading