Pain Control in the Care of the Dying

Introduction

If death is a defeat, then it won`t be a high priority. But if a good death is the culmination of a good life then it must be a priority. ( )

One of the main forces behind the development of the palliative care movement and hospices was the needless suffering of dying patients, not the least being poor pain control in the last days or hours of life. ( and ). The expertise of palliative care lies in control of pain and symptoms together with the wider consideration of the patient and family. The World Health Organization’s definition of palliative care is shown in Box 76-1 ( ). The traditional medical model of illness/treatment/cure is displaced in palliative care to a model and attitude focused on “care.” This palliative care model embraces the physical, psychological, social, and spiritual domains of care.

Fear of dying in pain is still common and was the reason for the majority of requests for euthanasia in one study from a hospice in the Netherlands, one that not only cares for inpatients but also consults with a large number of family doctors. However, once reassured by effective consultation and/or treatment, few requests were repeated and the patients died peacefully of their illness ( ).

Other studies document the fact that pain at the end of life can be controlled, usually without sedation, by teams experienced in the multiprofessional practice of palliative care; these are presented in detail in the . It is sadly true that not all patients benefit from this experience. In some developing countries this is because of lack of resources, including opioid availability, but it is not only in such deprived conditions that patients suffer relievable terminal distress. In the United Kingdom in response to the influence of the new medical specialty of palliative medicine, care of the dying is now a core objective of the medical undergraduate curriculum ( ). This is reflected in research suggesting that education in palliative care is increasing in medical schools ( ). However, there is still a general lack of knowledge and skills within the health care community, including myths concerning the use of narcotics ( ; see Chapter 31 ), such as addiction, respiratory failure, and the association of morphine with imminent death.

An Overview of the Evidence

The first clinical study of the way that patients die was reported by Sir William Osler in his lecture “Science and Immortality” (1906). Forms were completed in his wards in the Johns Hopkins Hospital by nurses, and they are preserved in the Osler Library at McGill University, Montreal. Osler states:

In our modern life the educated man dies … generally unconscious and unconcerned. I have careful records of about 500 death bed studies, particularly with reference to the modes of death and the sensations of the dying. The latter alone concerns us here. Ninety suffered bodily pain or distress of one sort or another, 11 showed mental apprehension, 2 positive terror, 1 expressed spiritual exaltation, 1 bitter remorse. The great majority gave no signs one way or the other; like their birth, their death was a sleep and a forgetting. ( )

Death at that time was likely to have come at an earlier age than most deaths today, often following infections. Osler’s work points to not just physical pain but also psychological and spiritual distress at the end of life.

The have been translated into many languages. The efficacy of the guidelines was examined in 401 dying patients by . They reported that at the time of death, only 3% of patients experienced severe or very severe pain whereas 52% had no pain at all, 24% had only mild or moderate pain, and 20% were unable to rate their pain intensity. Forty-four percent of patients required parenteral drugs. Additional adjuvant drugs were used in 90% of patients. This study was carried out by a department of anesthesiology and a pain clinic on 45% of the patients treated in a general ward. The World Health Organization guidelines form the basis of the European Association for Palliative Care guidelines for cancer pain ( ).

Some earlier studies of the last days of life ( , , , ) showed that although the majority of patients dies peacefully, sufficient attention was not always given earlier to what had by then become a multisystem disease or deterioration. Hinton’s report was a uniquely detailed study in which the physical discomfort and mental state of 102 dying patients were compared with 102 patients in the same wards with diseases in the same systems that were serious but not fatal. He found that although pain was adequately controlled in 82% of 82 patients with malignant disease, vomiting and nausea were relieved in 63% and dyspnea in only 18%. These symptoms were more common in patients dying of renal or cardiac failure—a total of 14 patients in all. Eleven percent of his patients were unable to be roused in the last week of life, 34% were unconscious for at least 24 hours and 60% for 6 or 9 hours before death, and only 6% were conscious just before they died. The act of dying itself was rarely distressing. Hinton noticed a significant degree of depression in 46%, with a rise in the last week or two of life. This contrasts with a group of 77 patients in his later home care study, for which he found serious depression in only 5% of the patients and a decrease in the proportion who had serious depression as death approached ( ). Many of his patients were eventually admitted to St. Christopher’s Hospice, but the families afterward felt that this was an extension of home care and were satisfied with what they themselves had achieved and the way that the patients finally died. A further study of their retrospective observations showed that they tended to overestimate the pain when compared with their reports to him during the patient’s illness ( ).

Not all pain in cancer patients is directly related to the disease. Some pain is experienced because of the effects of the disease, for example, bowel obstruction. In such circumstances a mixed pain picture, partly attributable to the tumour and partly to the colic of obstruction, is experienced by patients. A clinical and pathological study of 40 patients with intestinal obstruction caused by far-advanced abdominal and/or pelvic malignant disease was conducted. Surgical intervention was feasible in only two cases. The remaining 38 patients were managed medically without intravenous fluids and nasogastric suction. Obstructive symptoms, including intestinal colic, vomiting, and diarrhea, were effectively controlled by drugs ( ).

In contrast, carried out a prospective study of 120 patients dying at home and found that more than 50% had suffered what they considered unendurable pain during their last days, which was controlled only by means of sedation. Pain was the problem in 31 patients, dyspnea in 33, delirium in 11, and vomiting in 5. In a guest editorial, asked whether this might be a different population and pointed out that many hospices are able to admit patients when control at home breaks down, often through carer exhaustion, as reported by . By contrast, followed 200 consecutive patients in an integrated hospice inpatient and home care program. Around 40% died in their own homes, but in identifying and treating the same symptoms as in the Ventafridda study, they found that with ready awareness of the problems that may arise in the last 48 hours of life it was possible to keep their patients comfortable to the end. Thirty percent of their patients were conscious until death, 38% became unconscious from 12–24 hours before death, 7% became unconscious 24–48 hours prior to death, and 10% were unconscious for more than 48 hours. He wrote in some detail of the organic brain disorders that may be manifested as delirium but noted that this did not follow rapid escalation of the use of analgesics or other drugs. As he stated regarding its management, all unnecessary drugs should be discontinued, but sedation and antispasmodic treatment may be needed, and the combination of a neuroleptic and a benzodiazepine is indicated ( ).

A further comparable study would be home care patients receiving hospice care during admission to St. Christopher’s Hospice for the last 48 hours of life ( ). This included 9.8% of all admissions over a 6-month period, with 34 patients being admitted from home and 13 transferred from local hospitals. Most of the patients from home were being visited by the hospice’s own team of specialist nurses and were being cared for by family and friends; admission was arranged for symptom control and nursing or because carers (often elderly) had become exhausted.

All but three of the patients required an opioid, most receiving it parenterally in the last few hours. In general, low doses gave rapid control of symptoms, and only one patient needed escalating doses. Dyspnea presented more of a problem than pain did. That most of these very sick patients were at home so late in their illness makes them a different group from those in acute wards.

Of Boyd’s patients in the hospice, all were reported to have died peacefully, but 16 (34%) needed intervention for one or more problems in the 3 hours before death ( ). In none of these patients, however, was pain the problem. It would seem that the main problem is changing gear, ceasing to try to prolong life, and accepting that death is now inevitable and that the duty of the physician and the entire multiprofessional team is to relieve the patient’s suffering and support the family. This has become more difficult as technology has developed and as people have sometimes seemed to believe that death must be due to failure of medicine to preserve life.

The Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatment ( ) has been the most extensive report published, with a total of 9105 adults hospitalized with one or more of nine life-threatening diagnoses and an overall 6-month mortality rate of 47%. The second phase of this study documented the results of the intervention of specially trained nurses in an attempt to improve communication and care and found that there was no overall improvement, including the level of pain reported. Half the patients who died had moderate or severe pain during most of their final 3 days of life in the intervention group of 2652 patients, and there was no difference in their experience in comparison to the control group. An editorial ( ) notes that “apparently no component of the SUPPORT intervention directly addressed the problem of inadequate pain control.” Communication between physicians and patients remained poor; for example, the former failed to implement patients’ refusal of cardiopulmonary resuscitation (CPR), and they misunderstood patients’ preferences regarding CPR in 80% of cases. These findings have led to a number of calls to address education, which has resulted in better practice in decision making and the control of pain and distress ( ).

A study of an integrated palliative care service by looked at the last 3 days of life of 50 consecutive patients, with particular concern for what they defined as “dignity in dying.” Personal function, including the ability to speak lucidly, cognitive function, and continence of urine and feces, was maintained at least to a moderate degree in the majority of patients. Of the 29 with pain, it was estimated that 15 had good control, 12 had moderate control, and 2 had poor control. Dose escalation did not generally occur. The same team repeated this study with 132 consecutive patients. Here, pain was the most common major symptom, with good relief being achieved in 56.5%, moderate control in 42%, and poor relief in 1.5%. Escalating doses of morphine or sedative were not required, and nearly 80% were able to recognize family or friends (personal communication). This study counters the arguments that patients in the dying phase are given high doses of opioids and that this dose often requires escalation for control of symptoms and consequently foreshortens life.

The issue that opioids given for control of symptoms may shorten life was addressed in a study by . They retrospectively analyzed the pattern of opioid use in the last week of life in 238 consecutive patients who died in a palliative care unit. The median dose of opioid was low (26.4 mg) in the last 24 hours of life, and patients who received increases in opioid doses at the end of life did not have shorter survival than those who received no increases.

A study by reviewed the care of 160 patients who died in a hospice over a 1-year period. These patients were evaluated for pain control every 4 hours in the last hours to days of life. The results showed that 54% had no pain, 32% had one episode of pain, 9% had two episodes of pain, and 5% had three or more episodes of pain in the last 48 hours of life. Overall, there was a trend for pain to diminish toward death. This study—in line with the findings of and —reinforces the importance of patients in the dying phase having appropriate pain medication prescribed, including regular analgesia if indicated and additional “as-required” analgesic prescribed for episodic pain.

The role of hospital palliative care teams and their impact on pain control have been described by . A non-equivalent control group design using a quota sample investigated 100 cancer patients who had been admitted to the hospital for symptom control. Fifty patients received intervention by a hospital palliative care team and 50 received traditional care. Data were collected by using the palliative care assessment (PACA) tool on three occasions. The PACA tool uses a 4-point scale of 0–3:

• 0: Absence of symptom

• 1: Symptom present, not affecting daily life

• 2: Symptom present, moderate effect on daily life

• 3: Symptom present, daily life dominated by symptom

The mean pain score in the group without hospital palliative care team intervention was 2.08 at initial assessment and 1.74 at 1 week. The group with hospital palliative care team intervention had a mean pain score of 2.32 at initial assessment and a mean score of 1.00 at 1 week. Both groups showed a statistically significant improvement in their pain control, and there was also a statistically significant difference in favor of the patients who had received intervention by a hospital palliative care team. This study suggests that with intervention, the pain of cancer patients can be improved in a hospital setting, but it can be achieved most significantly by involvement of a hospital palliative care team.

From this selection of evidence it would seem that in experienced hands, adequate pain relief can be achieved in the great majority of patients. However, the SUPPORT study and reports of patients with diseases other than cancer show that much needs to be done before it will be possible to say that all dying patients are given the relief that has been shown to be possible ( ). Better community care and readiness in all settings to appreciate that death is both inevitable and imminent are called for if we are to be satisfied that “competent care for the dying” is responding to need and can counter demands for euthanasia and physician-assisted suicide ( ). Dependence and loss of dignity are also cited as reasons for such requests, but relief of pain can make a difference here too and needs to be taught widely.

In published a book titled Changing Gear—Guidelines for Managing the Last Days of Life , which gives clear guidance for the recognition, assessment, and treatment of dying patients. It recognizes relatives’ needs and calls for a collaborative multiprofessional approach, which is especially needed for the rare intractable situation. Specific guidance for doctors related to end-of-life care with incorporation of care of the dying was published by the .

Diagnosing Dying

Recognizing that death is approaching may not be easy, but experienced nurses usually make more reliable predictions than other professionals do because of their closer contact with the patient. They note a deepening feeling of profound weakness, increasing irritability of mood, and sensitivity to “minor” discomfort. Patients may become increasingly apprehensive and need continual contact and reassurance. Terminal restlessness may be caused by various factors such as pain, dyspnea, metabolic disturbances leading to confusion, urinary retention, or an inability to move any longer without assistance. Some of this can be corrected and causes should be sought and treated. Families are understandably concerned if these symptoms are not controlled. Nurses (and others) are anxious that their patients not be sedated to the point where they cannot recognize or speak to their families, and a careful balance must be maintained between an individual’s need and the drugs and dosages used. Although it can sometimes be difficult to achieve symptom control without causing drowsiness (which in many cases is a consequence of increasing weakness), it is important to keep families informed at all times.

When it is recognized that patients are in the dying phase, the team can then refocus care appropriately for the needs of patients and their families ( Box 76-2 ; ).

In the dying phase, only medications that control or prevent distressing symptoms should be used, including opioids, anxiolytics, antiemetics, and the treatment of respiratory tract secretions. Appropriate conversions from the oral preparation to the subcutaneous route should be made as described later in this chapter. Inappropriate interventions, such as blood tests and measurement of vital signs, should be discontinued. It should be ensured that arrangements are not made for CPR because this constitutes a futile inappropriate medical treatment. Recent discussion on “necessary sustenance” when a patient is dying tends to ignore the fact that patients gradually lose their urge to eat and that the feeling of thirst is generally relieved by the slow administration of oral fluids and giving ice to suck and by scrupulous mouth care rather than by the administration of intravenous fluids. Such intravenous infusions are often continued until death in seriously ill patients because it is thought that electrolyte imbalance and dehydration may cause distress. However, the infusion may cause discomfort and distress to the patient, act as a barrier to relatives, and divert attention of the attending medical and nursing personnel from care of the patient to electrolyte and fluid balance ( ).

In the dying phase, regular observations should be made regarding pain and symptom control. In an inpatient setting this should be done every 4 hours. In the community there are specific challenges in caring for dying patients, including ensuring the availability of drugs on a 24-hour basis. Ideally, anticipatory prescribing should ensure that drugs are on hand in the patient’s home and that nursing services are available to provide care for the patient, support for the family, and administration of medication as appropriate. If a patient dies within a short time after an injection, many families (and indeed many nurses) need reassurance that this was not the cause of death and that “the proper medical treatment that is administered and that has an incidental effect on determining the exact moment of death is not the cause of death in any sensible use of the term” ( ).

In the dying phase, skilled communication by health care professionals is essential. Both the patient and family must be encouraged to express fears and anxieties, sometimes elicited by direct questioning. This is especially important when the family comes from a different culture. It is essential that the family understand that the patient is now dying and that specific religious or spiritual needs are assessed and met. Frameworks that underpin and empower health care professionals in the care of dying patients, such as the Liverpool Care Pathway for the Dying Patient (LCP), have been developed ( ); they promote anticipatory prescribing of analgesics and regular assessment of pain control in dying patients.

Fear and Impending Death

Fear is an occasionally overwhelming symptom of impending death and calls for physical contact and adequate medication. All the common fears of dying, of separation from loved ones, of uncompleted responsibilities, of dependence, which may be stronger than the fear of pain or mutilation or even an unfocused fear of the mystery of death, may be exacerbated ( ). As Parkes suggested, although one cannot deal effectively with everyone`s fear, “there are no cases for whom nothing can be done.” Fears of separation and loss are proper causes of grief, but if expressions of appropriate sorrow are encouraged, the patient can often move on to a deeper enjoyment of the life left. Fear of failure may lead to what has been termed “the life review,” in which long-standing problems are sometimes worked through at surprising speed ( ). Crises of all kinds can lead to different forms of acceleration. Fear for dependents is frequently realistic, and instituting or planning practical arrangements can bring comfort to all who are involved. Fear of losing physical function “seems to derive mainly from fantasies of the effects of this loss of control on those around.” This is an obvious component of the widespread fear of incontinence. Parkes also found that patients who had been in pain were often more afraid of the perceived disgrace associated with crying or screaming aloud than they were of the pain itself. This fear of physical pain is often unrealistic—or should be made so by effective treatment of both the patient and others around the patient. Such fear should be listened to and the possibilities of relief explained and given.

Fear of mutilation or physical deterioration may be helped by the attitude of the staff toward the weakening body. This attitude is reflected in verbal reassurances that the essence of this person remains and is recognized and respected. Fear of the unknown is helped when the known, as seen by the patient, is rendered attentive and reassuring. Trust and faith in life and death are interwoven, and both are enhanced by the attitude of those around. It is most important that the staff be at ease, with some confidence that both life and death are meaningful and that death is a necessary and fitting end to the accomplishment of living. A supportive atmosphere is best created with few if any words.

Pain and Sudden Death

Death from injury in an accident often appears to witnesses to be painless. pointed out that those who have been rescued from death by drowning, even after apparently hopeless hours of artificial respiration, always say that before losing consciousness they experienced no suffering whatever. reported that of patients admitted to an emergency department, 37% had no pain in the initial phase of injury, but they pointed out that 40% reported very severe pain.

Sudden death from coronary occlusion may also be painless, although some of those who survive long enough to speak may refer to severe pain. An elderly nurse died after a series of myocardial infarctions. Talking with a friend, she suddenly interrupted her to say, quite calmly, “I need one of my pills,” and died without another breath or sign of distress. The same evidently happens after some cerebrovascular accidents. Many would choose this way of dying, although research on bereavement suggests that it is more difficult for survivors to come to terms with sudden death than with the slower, expected death when there has been an opportunity to bid farewell and resolve outstanding difficulties.

Lewis discussed the apparent painlessness of some traumatic deaths in his essay “On Natural Death.” He wrote:

Pain is useful for avoidance, for getting away when there’s time to get away, but when it is endgame, and no way back, pain is likely to be turned off, and the mechanisms for this are wonderfully precise and quick. If I had to design an ecosystem in which creatures had to live off each other and in which dying was an indispensable part of living, I could not think of a better way to manage.

Total Pain

The concept of “total pain” was first described in the literature by Saunders ( ). It was observed in clinical practice that some patients suffering pain of clear physical origin reported more severe pain or an exacerbation of the pain that appeared to be related to

• Psychological “pain” (e.g., difficulty coping with the diagnosis of cancer or difficulty coping with a short prognosis)

• Social “pain” (e.g., sadness at leaving family and friends)

• Spiritual “pain” (e.g., angry with God or “Why me, why now?” type of questioning)

A further clinical example of total pain is when a patient describes the pain as being “all over” and is unable to localize the pain to any one part of the body.

In patients who are in pain, an in-depth clinical history that includes physical, psychological, social, and spiritual aspects is essential, together with a complete physical assessment. In such cases, optimizing oral analgesics, the use of anesthetic techniques, and oncological as well as surgical interventions may well be required to address the physical element of the total pain. The approach to managing total pain includes detailed attention to pain control together with skilled psychosocial and spiritual interventions. Multidisciplinary health care professionals can work together to alleviate the patient’s distress. This is a specific area of expertise in specialist palliative care.

Patients who have total pain can also “displace” this total pain onto the team looking after them. Caring for patients with pain can be distressing for health care professionals, and teams not used to dealing with this situation can become dysfunctional because of the apparent failure to address the patient’s pain. It is therefore important that such patients and teams have access to specialist palliative care services.