LLST is defined as the omission or withdrawal of part or all life-support procedures like mechanical ventilation, hemofiltration, hemodialysis, and vasoactive medication. Over three decades ago, the Hasting Center defined life-support treatment as medical, technical, procedural or pharmacological interventions administered to a patient to delay the moment of death regardless of whether this treatment targets the underlying disease or actual biological process.

For healthcare providers LLSTs are clinical decisions and futility judgments consisting of weighed prognostic assessments on the ineffectiveness of a healthcare intervention. A procedure is considered futile when its application in a given clinical situation is ineffective and useless regarding the effect wanted. The main problem is the term futility—well-defined theoretically—but not so easy to identify in the clinical practice. Limitation of life-support treatment includes the decisions to omit and withdraw life-support treatment by healthcare providers on grounds of respect to the patient and his dignity. A different thing that does not come from the healthcare providers but from the patients themselves is refusal to receive treatment. With the actual ethical standards, promoting and protecting the patient autonomy makes him free to reject all sorts of life-support treatment even if prescribed by a doctor and even when the refusal can be life-treatening. Fig. 2 shows conceptual differences between refusing treatment and LLST.

Fig. 2.

Differences between the decisions made to refuse treatment and LLST.

(0.08MB).

Traditionally and arbitrarily the types of limitation of life-support treatment have been divided into 4 great categories: disregarding the patient admission to an ICU; omitting treatments; keeping therapies; withdrawing therapies.

The most common type of LLST is to omit treatment, being the order to refuse cardiopulmonary resuscitation the most common of all followed by the omission of preferably invasive treatments, and less frequently treatment withdrawal. Due to disease severity the mortality rate of treatment withdrawal is between 96% and 99%; treatment omission stays at 81%, and no further treatment, 44%.25 But there are new contributing factors in the decision-making process; the EPIPUSE study that analyzed LLST in Spanish ICUs in patients with long stays confirmed that in our country LLST happens 35% of the time with an associated hospital mortality rate of 93%. Also, this study confirmed that LLSTs are implemented in one third of the patients with complications or adverse events with organ repercussions that start during the second week of ICU admission.26 Although they are really well-categorized there is a huge variability in their application. A systematic review shows great differences in the ICUs of different countries, indicative of the existing needs for ethical training and research.27

Terminal extubation (TE) is a type of withdrawing treatments previously started agreed with the patient or his representatives. It is performed in patients with failed treatments, in the absence of new and effective treatments, and when respiratory support is a disproportionate and senseless life-support treatment within the new therapeutic plan. In a high percentage of patients, TE should be performed in a relatively short period of time before the patient’s death due to disease progression. Death cannot be the direct cause of the treatments administered so TE should not be performed in patients with neuromuscular relaxation. These treatments should not be used so that the patient looks more placid after weaning from respiratory support. Muscle relaxants do not have sedative-analgesic effects and can make the examination of the patient difficult regarding pain or comfort. If the patient is already on these drugs, we should wait until the effect is gone. In some patients the lack of respiratory support may cause signs of respiratory failure44 due to an obstructed airway and dyspnea. These situations should be anticipated and avoided with the proper treatment like advanced sedation. That is why TE is often associated with some sort of analgesic-sedative treatment at adequate doses to avoid such symptoms. Depending on the cause of respiratory failure, oxygen, corticoids, diuretics, bronchodilators and/or opiods can be prescribed to treat the symptoms.45 Also, relatives46 and healthcare providers47 should be informed that the patient has developed these symptoms.

Associating LLST with death is a common mistake. The fact that most patients on LLSTs die due to the severe course of their disease does not mean that death will occur inexorably like former landmark studies have reported.48 One of the main goals here is to facilitate comfort, symptom control, and accompaniment of the loved ones. ICU humanization projects have tried to improve ICUs by relaxing the visiting hours and making the relatives part of the healthcare process.49 Therefore, it is essential to anticipate disease progression after the implementation of LLST decisions. Also, to contemplate action plans coordinated with other specialties like palliative care or hospital units to continue non-invasive treatments.

When LLST has been decided, the healthcare plan should focus on avoiding suffering, respecting the patient’s dignity, and preventing or resolving the conflicts derived from the decision-making process. The comprehensive management of critically ill patients includes prevention, diagnosis, monitoring, and treatment of these patients plus proper palliative care (PC) after the LLST decision has been made.50

PC involves the administration of appropriate drugs to alleviate pain or dyspnea, family accompaniment, taking care of the patient’s needs, and facilitating a comfortable environment avoiding noises or unnecessary alarms. It consists of giving more importance gradually to care instead of treatment or diagnostic tests and making the relatives participate in that goal not only during the process of dying accompaniment but providing care during the process of mourning after the patient’s death. Table 3 shows the fundamental aspects of palliative care at the ICU setting. It is essential to plan LLSTs in advance, what treatments will be limited or withdrawn and how the events that may unfold and the presence of the family will be handled. Also, plan the duration of the visits; the number of people that will be accompanying the patient; inform the family of the different events that may occur like prolonged agony, survival of the patient, etc.; and teach them how to cope with them. Also, it is advisable to offer psychological support to the family. If the patient survives and is discharged from the ICU, it is advisable to promote the continuity of PC at the hospital floor unit with specific PC services only if appropriate and if prescribed by the healthcare team. By implementing comprehensive PC with the aspects mentioned above, the ICU can become a better place to die with personnel properly trained to treat the symptoms, accompany and comfort family and patient. It is a significant change that broadens the services provided in response to specific needs to ultimately improves the quality of end-of-life care. To achieve this we still have to break through SOME barriers; Table 4 shows the main difficulties and possible solutions for the integration of PC at the ICU setting.

Intensive care to facilitate organ donation (ICOD) is the result of the changes made in the doctor-patient relation and affects 3 different dimensions: medical, ethical, and socicultural.50 ICOD should be included routinely in end-of-life care and are applicable to 2 different domains: the process of donation after brain death and the process of controlled/expected asystole donation. In both cases, the process of donation is built on a prior decision of limitation on criteria of futility often proposed by the treating physician and agreed with the patient and/or his relatives. Only when this decision has been made, the possibility of donation can be discussed. The identification and careful selection of patients eligible for ICOD is essential: patients with devastating irreversible neurological injuries with poor prognosis not eligible for ICU admission with curative purposes and who may evolve to brain death. Also, patients whose only reason for admission is being donors to avoid unnecessary or inappropriate admissions51 within a framework of excellent communication with the family and the medical teams.52 In patients who don’t meet the criteria for brain death, all procedures should be terminated. When appropriate, the possibility of asystole donation should be offered.

Expected asystole donation should be included as part of end-of-life care, once the LLST decision has been made, and in patients who may die after withdrawing life-support.

The healthcare provider-patient relation is based on confidence and mutual sincerity. The patient is the actual holder of information and the only one entitled to decide whether to inform other people of his situation and status.53 A patient’s decision not to inform any third parties should be respected. Still, if appropriate, he should be persuaded that other people can participate in certain decisions not forseen in the instructions given by him when he cannot make any decisions by himself. However, in most cases, the status of the patient won’t let him participate in any end-of-life decisions at the ICU and relatives will become the main interlocutors unless the patient picked someone else.

There is an ongoing flow of information at the ICU not limited to a single day. In any case, if the course of the disease suggests that we are in an end-of-life situation, an interview will be necessary including a summary of what happened previously and what motivated the actual situation plus a treatment plan following the patient’s preferences. This information exchange should take place in an adequate setting.54 There is no unanimous criterion on what the best way to expose a patient’s diagnosis is. As a matter of fact, several studies warn that patients and relatives not always understand what doctors are trying to tell them. Several methods for decision-making or deliberating at the ICU setting have been suggested. Added to the healthcare team necessary consensus, we think LLST decisions should be shared with the patient and his relatives.55–60

Table 5 shows the recommendations established to make shared end-of-life healthcare decisions.61 The conflicts or discrepancies that may arise should be solved. Sometimes the family needs time to assimilate the information, postpone decisions, solve doubts or contact another physician who may have assisted the patient in the past and with whom patient and relatives have a relationship of trust. Also, the relatives need to be informed of the possible consequences of the actions that will unfold like breathing patterns prior to death that don’t necessarily mean suffering for the patient. Although the actual legislation and our own tradition consider the decisions made by relatives or legal representatives valid, several studies suggest that they do not reflect the actual wishes of the patient.62–64 We should explain and guide the relatives that they need to make a decision on behalf of the patient based on the decisions he would have made by himself. The information and the decisions made in these interviews should be included in the patient’s clinical history. No written informed consent is required for these proceedings that are nothing but an adjustment of treatment. Telling bad news and making end-of-life decisions is something intensivists and other healthcare providers involved in the management of critically ill patients should learn and train.65 The possibility of providing spiritual support to patients and relatives should be explored too; most families are thankful when offered even if they did not ask or want it.66

Measuring quality is essential to guarantee and improve end-of-life care in the critically ill patient. The key domains of high-quality palliative care have been defined by patients, relatives,67 and expert consensus.60,68 These domains include the effective management of physical, psychological, and spiritual symptoms, and the timely and sensitive communication on the adequacy of care in relation to the situation of the patient, prognosis, and values. Also, match treatment to the patient’s preferences, care for the family needs, plan care transitions, and support to healthcare providers. Quality of end-of-life care has also been assessed through surveys including providers, both doctors and nurses, and relatives and validated in the United States69 and, recently, Europe.70 Several studies have reported the margin for improvement of many ICUs on a global scale.71

Quality assessment can be performed through indicators of structure, process and result.10,72 Most proposals describe process indicators, but there is less consensus on result indicators.73 Also, we need studies on the process-results correlation guaranteeing the effectiveness, efficiency, and viability of the different interventions proposed over the last few years to improve the quality of end-of-life care. Also, we need to know what interventions can lead to unwanted consequences.74,75 Ever since 2005, SEMICYUC has been elaborating quality indicators—recently updated—including a significant number of end-of-life care indicators 76 as shown on Table 6. Measuring these indicators periodically allows us to identify situations of improvement, compare results, and establish better practices to guarantee excellent end-of-life care.

Healthcare providers can suffer from stress after being involved in end-of-life situations; ethical conflicts; constantly seeing the patients’ suffering; perceiving potentially inappropriate or futile care; inadequately communicating with the team; and having to care for the family needs. On many occasions, healthcare providers have to face these situations without the proper training in clinical ethics. These work-related stress factors can trigger specific syndromes like moral suffering, perception of inappropriate care, compassion-induced fatigue, secondary traumatic stress, and occupational burnout, among others.77 On many occasions, these syndromes are closely related to one another and overlap in some professionals. They can have a negative personal and professional impact and repercussions in the quality of care provided to patients and relatives, favor professional absenteeism, and increase costs.78

To have a healthy working space we need multimodal interventions aimed at improving communication skills, collaboration, effective decision-making processes, adequate ratios of professionals, personal recognition, and real leadership. Training in bioethics and interdisciplinary debates during and after conflicting situations may reduce the appearance of these symptoms.79 Healthcare providers need to be aware of risky situations and ask for help when needed. Similarly, individual strategies should focus on promoting the healthcare providers’ physical and emotional self-esteem, and resilience.

The valuable contribution of bioethics to intensive medicine is the introduction of values in the complex decision-making process to improve these decisions. LLST means omitting or withdrawing futile treatments at a certain point after confirming the mismatch between the therapeutic means and their goals. Today, LLST is a common clinical practice at the ICU setting. Fighting therapeutic obstinacy is a primary goal of end-of-life care. The recommendations established by SEMICYUC Bioethics Working Group emphasize the importance of shared healthcare planning including donation-oriented intensive care, and new types of LLST. At the same time, they promote integrating palliative care in the ICU setting and improving communication and family accompaniment plans. These recommendations have been elaborated by SEMICYUC Bioethics Working Group to reduce the variability of clinical practice observed in LLST stratgies and contribute to improve the end-of-life care of critically ill patients.