34 - Healthcare providers responses to the death of a child | Oxford Textbook of Palliative Care for Children (Liben, Oxford Textbook of Palliative Care for Children)

Editors: Goldman, Ann; Hain, Richard; Liben, Stephen

Title: Oxford Textbook of Palliative Care for Children, 1st Edition

> Table of Contents > Section 3 - Symptom care > 32 - Symptom control at the end-of-life

Symptom control at the end-of-life

Dawn Davies

de Vlaming Debbie

Palliative care for children is best thought of as a continuum. Sometimes the management of pain and other symptoms might be relatively unaltered leading up to the child's death. Often in these situations, symptom management at the end of life may involve intensification of management strategies, using higher doses or more frequent administration of previously used medications (please see specific preceding chapters for more details). Not infrequently, though, new symptoms may arise close to the end of life, or symptom progression may be such that symptoms are no longer controlled by previously effective interventions. The following is intended to address symptom control at the very end of life: the hours to days prior to death.

There is a need for caregivers to develop a proactive approach to the what ifs that families face close to the time of their children's deaths. This anticipatory planning is arguably even more important for families caring for their child at home, and especially for those living in rural areas, or at a distance from their usual health care resources. Contingency plans should be in place for the management of new symptoms. Unexpected or unanticipated symptoms may, in turn, lead caregivers and families alike to rethink the setting in which the child is being cared for. It is often during this period that families that were previously coping very well at home, with in home supports, become frightened, and may wish for admission to hospital, where staff that they know and trust is present.

Because the average parent has very little, if any, direct experience with death, it is our duty as professionals to guide parents honestly through the dying process of their child, anticipating their needs and questions. Well-developed written care plans need to be in place, and provision of ongoing and flexible plans for the changing situation of the child and family is key. Many parents have stated that feeling like there's a plan, and a next step, are crucial to a feeling of coping and control. These plans, in the context of this chapter, need to include a next step for symptom management, and a location for care provision.

As with all human needs, physical needs have to be met as a priority. If the child is in pain, or is distressed by some other symptom, very little can be achieved in enhancing the life of the child and family. Unfortunately, a few landmark papers suggest that uncontrolled symptoms and suffering at the end of children's lives are much too frequent [1, 2].

Paradoxically, there is a risk that in pursuing control of symptoms, we can become too interventionist in our approach. For a majority of families, it becomes a priority to avoid hospital admission, where possible. Even turning up for an X-ray may be unwelcome, and most will prefer to avoid blood tests, even if these can be done without attending hospital. In formulating a therapeutic strategy, we need to rely heavily on clinical judgment, rather than investigations, and often, to proceed on the basis of probability, rather than certainty.

Optimum pain and symptom control is the foundation of excellent palliative care, and is the cornerstone that facilitates attainment of all the other goals in care at the end of life. Intensive symptom management, particularly in the last few hours or days of life, is often necessary in order to allow for the peaceful death of the child. It is also crucial for the family and friends that symptoms do not spiral out of control in those last precious moments. The family's feeling that its child did not suffer is paramount, and is the one small comfort that parents often discuss as the only solace to be found, in their monumental loss and devastation. Alternatively, parents have sometimes described being haunted by memories of their child's suffering, when symptom management has been less than optimal. These are the memories that become the family's lasting legacy, and therefore, care of the child and family must be as seamless as possible.

Setting for care in the final phase of life

As the child's condition changes, parents or other caregivers may experience sudden changes of mind as to where and how they want to care for their child, and every attempt should be made to honour these wishes to the extent possible.

P.498

Anticipating and planning for sudden hospital re-admission or discharge pre-emptively can be very helpful, thereby allowing maximal flexibility for the child and family (cross ref Place of care chapter, chapter 37, Brook).

Throughout the child's illness, flexibility in approach to care should be optimized such that the child and family lead as normal a life as possible. Having a resource person (also known as a key-worker ) to coordinate all transitions within the medical system, to home, and back to hospital, has been identified as a crucial part of well-coordinated palliative care. Very often, a passport on which all health care providers record their assessments and treatments, and which is carried by the patient's family through various settings, is helpful. In the absence of such a document, a letter carried by the parents, describing their child's diagnosis and treatment plan (including Do Not Attempt Resuscitation orders, where applicable) is useful. This can facilitate optimal treatment, reflecting the family's wishes, and prevents the family members from having to re-tell their story to yet another practitioner.

Decision-making about the setting in which the child will be cared for, and eventually die, is emotionally charged. Some parents will embrace the opportunity to keep their child at home with the family, in spite of challenges and fears. Others find the dual role of parent and proxy health care provider to be unduly burdensome, or undesirable; they may want to be just Mom/Dad. Some parents, after a period of intensive care-giving at home, will seek help either in a hospice or a hospital setting, at the end of life. Other families fear the impact that a death at home would have on surviving siblings, and many parents have voiced concerns about being able to remain in the family home if their child died there. There is often fear of the unknown, in terms of parents ability to gauge what their own reaction will be at the time of the child's death, and parents often will state that they cannot bring themselves to even contemplate this in advance. The help of Child Life specialists, social workers, and psychologists can be instrumental in helping families with these issues.

A cautionary note: we, as health care professionals, may sometimes underestimate the ability of families to cope at home in providing palliative care, and we need to refrain from making paternalistic judgments in how we present options to families. Sometimes complex care offered in a hospital, or even in intensive care settings, can be simplified or modified, to allow for safe and comfortable care in the home. The more complicated the situation is, the more advance planning will be needed for a discharge to home. As long as families are aware of the risks, benefits and resources available in various settings, these very personal decisions should be supported, as long as there are no clear medical contra-indications. A good bridge to transitions between hospital and home is a trial run.

The patient may technically remain an in-patient, but has passes to home or an alternate site of care, to see how care in this other setting would work for the child and the family. This also allows time for set-up of the out-patient setting for equipment and pharmacy needs. Some hospitals and intensive care units have care-by-parent units, or rooming-in facilities, in which parents provide all of their child's care, with the benefit of having health care professionals who are well known to them easily available. This often bolsters the confidence of parents upon leaving the hospital.

On the other hand, some families feel that they will want to remain in hospital indefinitely, and may be overwhelmed by the care required by their child. Parents often state that they are afraid to be at home with their child. Often, the further the family lives from immediate medical or nursing assistance, the more this fear is intensified. Over time this may change, as parents first assist in, and then gradually learn to perform independently, the different procedures required. Some parents will eventually master and embrace tasks they initially thought would be overwhelming. In these cases, some families will opt stepwise for passes out of the hospital, evolving to a full discharge home. Appropriate palliative home-care support must be set up in the home as part of a well-constructed discharge plan.

Even in very small towns, palliative care teams can evolve around a particular child; these teams are often composed of local on-call doctors, nurses, pharmacists, and others, who will rally around the family when given the extreme rarity of being called on to perform such a service. Specialized pediatric palliative care teams can be consulted directly by local care-givers as needs change. However, there are rare situations in which the setting decided upon is not ideal, despite everyone's best efforts:

Case

Y.K. is a 13-year-old girl who lives in a very remote northern Canadian community of 200 people. There is no road access during winter; access is by flight only. There is a nursing station, with two registered nurses on duty to care for the community at all times, in alternating shifts. The nearest hospital is 700 miles away.

Y.K. has a diffuse brainstem glioma. She is cognitively intact, but has diffuse cranial nerve dysfunction, which has lead to recurrent episodes of aspiration pneumonia. She is cachectic, very weak and bed-bound, and requires total care. Mom is a single parent, and is busy with a job that allows her to be sole provider for her children. Y.K. was residing at home with her Mom and 16-year-old brother until 3 weeks ago. Since then, she has developed Gram-negative bacteremia and urinary tract infection, intolerance of her gastrostomy feeds, vomiting, weight loss, and back pain. She was initially treated in a secondary level center, attended by a pediatrician whom the patient and mother trust. She has now been transferred to a tertiary care center, hundreds of miles from

P.499

home, for further supportive treatment, and is recovering well. She is much more comfortable, and her deterioration has been slow, over months. But the dilemma for this family is emerging. She is well enough to return to the secondary hospital that referred her, which is much closer to home, and is a place where the family has many psychosocial supports. Mom would like to stay in the secondary hospital with Y.K., but cannot manage this financially, as she has already applied for, and been granted, a year off work to attend to her ill child. There is a 16-year-old sibling who is very close to his dying sister, but the family's relocation for her treatment had a negative impact on him, and the mother is not willing to risk further trauma to him by moving again. It is likely that they will be able to visit Y.K. only for one week of every month from the home community.

Conversely, although aware of her mother's concerns and her own medical situation, Y.K. herself expresses a strong desire to return to her home. This child requires frequent suctioning and attention through the night. There are no professionals to provide in-home assistance, either in her care, or for general respite or homemaking for Mom. Mom is resourceful, dependable, and very aware of their situation. It is very upsetting for her to contemplate being separated from her young daughter at this crucial time, but currently, she feels there is no alternative. She has friends and family in her village, but they all have young children, and she fears that despite people's best intentions, she will be on her own in caring for her child. She states that she is scared by the prospect of taking Y.K. home under such circumstances, and fears for the welfare of her child if she became suddenly ill again, and for her own ability to cope with what may still be a lengthy illness. She is also very concerned about her 16-year-old son, who has been left in the care of relatives much of the time, and Mom states that he is having an increasingly difficult time with separations from her.

However, even in situations like this one, contingency plans to make the best of the situation are still possible. In the above scenario, the patient was able to identify a wish list of things to do in the secondary hospital, and Mom was able to devise a plan, involving friends and relations who could serially spend time with Y.K.A comprehensive plan communicating the patient's desires, dislikes, wants and needs was composed, and there was good nursing and physician transfer of her care. The palliative and oncology teams remained available for consultation, to family and local health care providers alike. Lastly, there was an understanding that if opportunities arose when Y.K. could return home for even short periods of time, they would be availed of, and appropriate equipment would be set up at home in anticipation of such possibilities. Air travel was sponsored by a local airline, in response to requests by the treatment team, to ensure that the child's brother as well as her mother would be able to visit her monthly.

Lines of communication

When a family takes their child home for end-of-life care, it is essential that some sort of continuous on-call coverage be arranged. If a family is not well supported at home, hospitalization is the default position, and this may not be in keeping with its hopes and goals. Different communities will have differing availability of home-health nurses, but some do provide 24-h-a-day on-call service; this service is often the first-line'for families questions or concerns. Regardless of the nursing support available, it is also imperative that a physician or physician group provide continuous on-call coverage as the child's condition changes. In smaller communities, where a family has very close ties to a local general practitioner, that person may elect to be available to the family at all times, given the extreme rarity of the situation. In larger centers, in which caring for a dying child is a less rare occurrence, an on-call group of physicians often shares this responsibility. In many situations, pharmacists, psychologists, clergy, and social workers may be available for home visits. Neighbors, especially those with health care backgrounds, can provide enormous support to families. Regardless of the particular demographics of the health care community, in most cases a system can be devised such that parents know whom to call, day and night, for a variety of concerns. This list should be clearly written and kept handy, as should written care plans, to foster optimal communication.

Changes in the aim of treatment

Periods of transition are difficult for people in general. The transitions involved in redefining the goals of therapy for a child who is dying, then, is often an extremely emotional time for parents and caregivers alike. Often, the end of a child's life is preceded by a period of sometimes quite aggressive efforts to save this child's life. This transition, whether it be over hours, days or months, can be hard to understand. Therapies that were used at an earlier period in the child's illness, or thought to be essential, may no longer have a place in providing a benefit at the end of life, and the new essential medications are often analgesics, anticonvulsants, anxiolytics, antiemetics, and so on.

At the end of life, there are particular issues that recurrently cause distress. Whether, or how, to feed/hydrate their children can cause much distress for families, especially if they are unprepared for the eventuality that the children may either refuse, or become unable, to eat and drink (see also Chpater 4, Ethics; and Chapter 24, Feeding in palliative care). There is a spectrum of responses to this situation, seen in both caregivers and parents. The parents often see the provision of food and

P.500

drink to their child as the most basic of human comforts, and to be unable to offer this simple type of care is often traumatic. Parents will respond uniquely, with some expressing their view that they can't just let him starve , and others seeing this event as a part of the overall picture, in which all of the child's bodily functions are shutting down. It is helpful to explore feelings about this in advance, and to avoid a crisis-oriented approach of solving the problem only when it arises. The evidence suggests that dying persons do not suffer adverse consequences, when they no longer feel like eating/drinking [3]. Patients sometimes complain of adverse effects, such as abdominal pain and distention, when artificially provided hydration or nutrition is given. This knowledge is empowering to parents in their decision making. Similarly, many parents feel that their children's survival might be lengthened if nutrition is artificially provided. For children who have had a long and complicated medical course prior to death, requests for total parenteral nutrition at the end of life are not unheard of. Gently informing the parent that providing nutrition at this stage of the child's illness will not contribute to longer survival is reassuring [4].

Families may seek other supportive treatments very late in the child's illness. For oncology patients, the need for blood transfusions may have become rather routine to the family. Families will sometimes have a hard time letting go of this therapy, especially when they have seen marked improvement in the child's wellbeing at an earlier stage of her illness [5, 6]. Parents need to feel that they have done everything that they can to save their child, and in turn, to ensure that they are as comfortable as possible in their dying. As physicians, we recognize that further treatment with packed red cells will have no beneficial effect in the last few days of life. However, we have our medical knowledge, and our practical experience, on which to base this rationale. Families are at a distinct disadvantage in terms of the ability to understand this. When physicians deny their child a therapy that has been routine and previously effective, a negative dynamic can develop in which the parent may fear the team is giving up. Sometimes a natural consequences approach can be helpful in resolving this issue. If it appears that an impasse is being built around issues such as this, it may be most helpful to support the parents in their wishes, with certain provisos built into the administration of the therapy in question. For example, in the case of transfusions, it can be useful to help families understand that the hemoglobin level is no longer relevant, but that the patient's symptomatic relief provided by the transfusion will be the outcome to be evaluated. Most often, it is easier for parents to let go of such therapies, if they can see for themselves that these have become ineffective. The need to feel that they have done everything possible to ease their child's suffering is immense, and a human, rather than a medical approach to these discussions can have a positive impact.

Alternatively, as the child is seen to be dying by the parents, the situation may suddenly seem intolerable, and parents may ask, indirectly or directly, that the physician end their child's suffering/life. Such requests are not well-reported in the literature. Again, these requests are often borne of a fear or belief that the child is suffering. If the child is suffering from a symptom that has not been controlled, this plea should be taken as an indication that more needs to be done. If the physician feels at his or her limit in offering any further ideas, it behooves that physician to consult with others who may have more expertise, to get this child's symptoms under control.

However, sometimes such a request is made at a stage when the patient appears to be comfortable, but is no longer able to interact. Each physician will have to find his or her own way to explain, as sensitively as possible, that this action cannot be undertaken. In this situation, the distilling of the conversation to a very human level (i.e., I as a human being vs. I as her doctor ) cannot end her life has been helpful. There is certainly no panacea, and every situation will have individual determinants, which should guide the explanation that makes the most sense to the particular family.

Anticipating likely symptoms

Parents need very specific guidance as to possible and probable symptoms that may emerge, and need to be educated in stepwise plans for control of these. The medical literature suggests that physicians are unable to predict how much time is left for patients in their care, and it is recommended that no guesswork be made regarding specific timelines [7]. If a family expects death to occur at a certain time, anxiety often ensues, whether life expectancy is over-or under-estimated. In any case, families need anticipatory guidance as to how their child may change, with regard to both his or her physical appearance and physical signs of deterioration, as well as possible changes in his or her behavior and interactions. It is common that parents are unaware of what dying may actually look like, and the experience can be devastating, if they are unprepared for the changes that they may see in their child. Parents should be made aware that the child's activity level will decrease, and the time spent sleeping will increase, as will the degree of difficulty in awakening; sometimes, the child will be deeply unconscious for hours to days before his or her death. Similarly, families should be prepared for the physical signs that they may witness, including pallor/cyanosis or mottling of the skin,

P.501

and progressive cooling of the extremities, usually from distal to proximal. They should know that the child might have noisy respirations as he or she loses control of his or her oropharyngeal muscles and secretions. This should be treated, if distressing to the family, while reassuring them that the child is unaware of and unbothered by the death rattle, as some laypersons know it. Management strategies include positioning the child on his side, with the head slightly lower, by placing the bed in a slight Trendelenberg position. Medications such as glycopyrrolate, hyoscine given parenterally or transdermally, or the use of 1% atropine eye drops given sublingually, can all be helpful. Playing soft music in the background can be a useful distraction.

Management of some particularly problematic symptoms will follow later in the chapter. However, it is essential not just that families are prepared for what is likely to happen today, but also that health care providers be more long-sighted in guiding the families as to what symptoms are probable or possible, as the disease progresses. To that end, advance planning must occur, if the child is to remain adequately managed at home for the duration of his or her illness. A fairly common problem is that of sudden onset or exacerbation of a symptom in out-of-hours periods, such as evenings, nights, or weekends, at the end of life. For children cared for in a hospital or hospice setting, this can be very frightening for a family, if it had not been prepared for such a turn of events. However, in this setting, despite the worry and fear that may have arisen, there will at least be the presence of health care professionals, with immediate access to needed medications.

The same situation occurring in the family home at an inopportune time is much more problematic. Even if the family is well supported by a home health agency or program, accessing needed medication, as circumstances change, is potentially problematic. There are a number of ways to circumvent this problem, but it requires planning. There are situations where the family resides in an urban setting, with physicians, home health nurses, and pharmacies available on a 24 h-a-day, 7 days-a-week basis. However, although access to medications may be 24-h/day, there are the practical problems and inconvenience of dispensing the medication, and having someone pick it up from the pharmacy.

Another approach is that of the Emergency Drug Box . This approach to optimizing palliative care in non-business hours seeks to keep families at home when that is their desire. Certainly, within adult populations, the great majority of patients have indicated that they would prefer to die at home, surrounded by loved ones [8]. It is the wish of many pediatric patients and their parents as well. However, if symptoms suddenly emerge or intensify, hospitalization will likely follow, if adequate resources are unavailable at home. Lack of ready access to new medications is an enormous obstacle to the child experiencing pain, dyspnea, or some other distressing symptom.

Emergency drug boxes can be instrumental in preventing undesired hospital admissions, in the last few hours or days of life (Personal communication: Winnipeg Regional Health Authority, Palliative Care Program; Capital Health Regional Palliative Care Program, Edmonton). These locked boxes are most often released only after consultation of the home health nurse with the most responsible physician. Specific orders should be obtained from the physician before any medication is administered. In turn, medications administered from the emergency drug box are recorded, and education of the family/patient is achieved as new medications are administered.

Emergency drug boxes most often contain sublingual formulations, suppositories, transdermal patches or injectable medications, on the assumption that the patient may no longer be in a condition where oral medications would be practical or possible. The medications contained often include a variety of steroids, benzodiazepines, anti-emetics, anti-convulsants, anti-psychotics and opioids. These are often delivered to the family home by a courier or a taxi company, with whom arrangements have been made in advance, thus freeing family members from having to leave their ill child, at a very distressing time, to go to the pharmacy.

However, many communities are without such programs. In these situations, it is imperative that the treatment team anticipates possible symptoms that may emerge, based on the patient's disease. Medications should be available readily for all the symptoms that are considered possible. It is prudent to ensure that families are given prescriptions for, and receive hands-on education about, the administration of these drugs. The home health nurse is instrumental in reviewing the storage, preparation, and administration of these drugs with the family. If this review doesn't occur, there is a potential for parents to forget, between the time of discharge from hospital and the time the symptom arises, the appropriate medication for each symptom. Safety in a home setting must be reinforced, particularly if there are other small children in the home. All medications should be kept in an easy-to-organize box, such as a tackle or tool box, and be kept locked, whenever not in use.

What follows is specific information in regards to management of a variety of symptoms that often emerge or intensify in the final hours or days of life. There is an intentional emphasis on the practical aspects of how to institute each specific therapy, as the calculations required in mixing parenteral medications in the middle of the night are often vexing, particularly to practitioners who may not be handling these medications routinely. There is an attempt to present a broad range of

P.502

approaches, recognizing that treatment will be provided in settings ranging from homes in small rural settings to homes in large tertiary centers.

Lastly, parents should be made aware of changes that they will see in their child's body at the time of death, and should be supported through these. They should know that secretions may appear at the mouth/nose, and that the child may be incontinent of stool or urine. They should be reassured that there is no right way to behave, and they should be encouraged to do what they feel is best for them. Families privacies should be respected, and health care staff should be close at hand, but should not impose. A family may choose to bathe, hold, or dress its child. Other cultural or religious rites may occur around the child's body. The family members should be reassured that there is no rush, and that family members can spend as long as they need to with their child, family, and friends. Families are often very bewildered at this time, and may seek assistance for practical tasks, such as making phone calls to other extended family, selecting a funeral home, etc.

Preparedness through the dying process, and in the immediate period after death, contributes to the parents sense of control. Their centrality in decision-making empowers them. As devastating as the situation is, knowing what potential problems they may face, having contingency plans, and knowing which people to contact for support, makes it less frightening.

Management of pain exacerbations

This chapter presupposes that the child may previously have been on oral opioids, with good pain control. However, in the final hours and days of life, a sudden crescendo of pain is not uncommon, and must be anticipated; at such times, patients frequently become too ill to manage oral medications. The following describes a step-by-step approach to transition to other forms of analgesia if this occurs, such that physicians less familiar with provision of these medications can administer them with confidence, if their patients require such a change.

Morphine

Of the short half-life opioids (duration of action 3 4 h), morphine is the most commonly used, and is often considered to be first line in treating severe pain, because of availability and familiarity. Starting intravenous and subcutaneous (IV/SC) doses are in the range of 0.05 0.1 mg/kg/dose, typically given every four hours; as an intermittent bolus dose for the opioid na ve child. If the child has previously been on an oral preparation of morphine with well-controlled symptoms, the parenteral dose should be calculated as a half to a third of the oral dose, due to significant first-pass effect of enterally administered morphine through the hepatic circulation.

It is important to note that the child may still need rescue or breakthrough doses of immediate-release morphine in this situation. Generally, the breakthrough dose given is approximately 10% of the total daily morphine requirement, and may be given as frequently as every hour. If the child requires 3 or more doses of rescue medication in a day, the regular dose of medication should be increased by an increment that reflects the total amount of morphine used in the preceding 24-h period. However, if there is rapid escalation of the pain, the regular morphine dose should be increased by 20 50%, and frequent breakthrough doses given to achieve pain control. If such dose-escalation does not achieve good effect, or dose-limiting side effects are encountered, rotation to another opioid may be indicated.

The subcutaneous route means that many children, even without IV access, can be treated with a morphine infusion, avoiding the peaks and valleys of intermitttent boluses. One standard approach to mixing this infusion is to put 1.25 child's weight (kg)= dose in mg, to put in 250 ml of IV fluid. For example, for a 10 kg child:

This is a 5 g/kg/ml solution. Conventional starting rates for a morphine infusion are 10 40 g/kg/h. Therefore, using the above formula, the infusion would run at 2 8 ml per hour. It is important that breakthrough doses of analgesia are available. Because the child has a significant systemic level of morphine while on an infusion, boluses should be reduced in magnitude, compared to giving intermittent IV boluses in the absence of an infusion. Therefore, breakthrough doses of 0.02 mg/kg (20 mCg/kg) are considered safe for children on an infusion running at the starting rates above. If the child is requiring more morphine than the range listed previously, then a breakthrough dose equal to the hourly dose, given as a bolus over 10 15 min, is usually adequate.

It is noted that, like all high potency opioids, morphine does not have a ceiling, or upper-limit dose. The amount of drug should be titrated against the pain, providing the minimally effective dose to offer comfort, and the caretakers should against be vigilant for detecting dose-limiting side effects. This applies regardless of the route administered, and regardless of whether it is given as a patient-controlled analgesia (PCA), infusion, or intermittent bolus. It should also be noted that IV access is not necessary, and morphine may also be given subcutaneously, if other routes are not possible or desired, either as intermittent boluses or as a constant infusion. The easiest way to use the same site for multiple doses is to insert an Insuflon, a small flexible catheter, into the subcutaneous

P.503

tissue, and fix it in place with a transparent dressing. These may be left in situ for 3 7 days, while vigilance is maintained to catch signs of infection, induration, or medication leakage from the site. Otherwise, a 25-gauge butterfly needle can be secured in place, and repeated doses of medication can be given.

Diamorphine

Diamorphine is not yet available in some parts of the world, including most of North America. Elsewhere, it is one of the main opioids used in palliative medicine. Pharmacologically, it is little different from morphine, into which it is rapidly metabolised. It is no more addictive than morphine, but is 1.5 2 times as potent.

The major advantage of diamorphine is that it is much more soluble in water than morphine. This enables large doses of diamorphine to be prepared in very small volumes of water. Given parenterally, therefore, it is more practical than morphine at very high doses. Current practice in countries where diamorphine is available is often to use morphine for oral medication and diamorphine for subcutaneous or intravenous use. The 24-hourly dose of parenteral diamorphine is calculated from that of oral morphine as follows:

Hydromorphone (Dilaudid)

An alternative high potency opioid is hydromorphone. Hydromorphone is five times more potent than morphine [9]. Given the preceding, a reasonable starting intermittent parenteral bolus dose is 0.01 0.02 mg/kg/dose. It is also a short half-life opioid, like morphine, and therefore is often given every four hours. If the patient has previously been on oral hydromorphone, the parenteral dose should be calculated as a half to a third of the oral dose, because of first pass metabolism.

If an infusion is desired, it can be prepared using a similar standard formula as used in the preparation of a morphine infusion. Because hydromorphone is five times as potent as morphine, the amount of hydromorphone used is one fifth the amount of morphine, for the same volume of IV fluid. Therefore, 0.25 weight of the patient (kg)= amount of drug in mg, to be placed in 250 ml of IV fluid. Again, in the case of a 10 kg infant:

This yields a concentration of 10 mCg/ml, or 1 mCg/Kg/ml. Therefore, running the infusion at 2 8 ml will give a dose of 2 8 mCg/kg/h.

Fentanyl

Fentanyl is an opioid with a half-life that is ultra-short, and for this reason, intermittent dosing is not possible. Therefore, one must use either an IV infusion, or a transdermal form. Regardless of the medication used regularly, the use of a rapid-titration protocol for IV fentanyl has been described, for sudden uncontrolled cancer pain [10]. The more lipophilic nature of this drug, compared to morphine and hydromorphone, yields faster access through the blood-brain barrier, and therefore, a faster onset to peak effect. However, caution for this approach is warranted, given the possibility of respiratory depression in uncontrolled settings.

Fentanyl infusions typically start in the range of 0.5 1 g/kg/h. As such, perhaps the easiest and safest standardized formula for preparation is:

In the example of a 14 kg patient:

Therefore, for any weight of patient, running the infusion by this method at 0.5 ml/0.5 mL/h equals 0.5 mCg/kg/h, 1 ml/h equals 1 mCg/kg/h, and so on.

The equi-analgesic dose of fentanyl, compared to morphine, has not been accurately determined. In acute use, fentanyl is thought to have 100 times the potency of morphine. However, with prior regular use of opioid medications, fentanyl has been found on average to be 68 85 times more potent than morphine, although the limited studies available, searching for relative potency of these drugs, vary widely. Studies suggest that constipation and laxative use may be less common with fentanyl analgesia [11, 12, 13].

Transdermal fentanyl

If the patient is of a sufficient size and is not opioid na ve, transdermal fentanyl patches can be a very desirable method of anal-gesia. Transdermal fentanyl is commercially available as Duragesic patches in many countries. This patch is applied to the skin, and works by creation of a reservoir of drug that diffuses at a constant rate into the skin, and then, into the systemic circulation. It can take approximately 12 h to reach a systemic level sufficient to provide analgesia. During the first 12 h of application, regular doses of the opioid being rotated from

P.504

should be continued. Similarly, when the patch is removed, it takes up to 12 h for plasma levels to decline sufficiently to institute a new regular opioid medication. Rescue doses of breakthrough analgesic should be used through this period of time.

Transdermal fentanyl patches are indicated for stable pain, as fentanyl is not a medication that one can quickly titrate. However, these patches are convenient, and easy to use. Pediatric patients often prefer this modality, because it dispenses with the need to take oral medication, or to be attached to an IV or SC system, normalizing life to some extent [14, 15].

Caution is needed if the patient becomes febrile, as the increased skin temperature can increase absorption by up to 30%. The fever should be controlled with anti-pyretics as able. The child should be watched for side-effects of somnolence, and for respiratory depression through this period. The dose should be reduced if these effects are observed, or a different drug, or a different route, should be tried if the child is already on the lowest-dose patch, until the fever subsides.

Currently, fentanyl patches are available in 25, 50, 75, and 100 g/h formulations. Because the lowest dose patch requires a morphine equivalent of a daily dose of at least 45 mg of oral morphine, this route is often impossible to use in small children. The patches may be used in an additive fashion, to achieve doses higher than 100 /h. For example, a patient may wear a 100 mCg/h patch with a 50 mCg/h patch, to achieve a total dose delivery of 150 mCg/h.

Care must be taken in choosing the application site, ensuring the child will not play with, pick at, bite, or eat the patch. Therefore, a position over the middle of the back, between the scapulae, is often a good choice in a small or cognitively-impaired child. The patches are designed such that cutting them disrupts the reservoir and is contra-indicated. There is no good evidence to support the practice of giving half-doses of the smallest patch by placing an occlusive dressing onto the skin, and then placing the patch over this, so that only half of the patch contacts the skin directly.

The patch should be applied to clean skin washed with water only, without soap or detergent. If hair is present, it should be clipped close to the level of the skin, and not shaven, as abrasion can increase the absorption of the medication. To apply, the patch should be placed against the skin, with even pressure from the applicant's hand for 30 s after the protective backing is removed. When the patch is taken off, it should be folded onto itself, skin-side inwards, and placed in a Sharps or other safety container. The hands of the caregiver should be washed with soap and water, to prevent accidental absorption.

Methadone

Methadone is a synthetic long half-life opioid, that differs somewhat from the previously described opioids. It has a long but unpredictable half-life, varying from 5 100 h, depending on the individual on whom it is used, with an average half-life, of around 24 h [16]. In Canada, prescribing of methadone is highly regulated, and one must apply for a methadone license through Health Canada. In other countries, methadone is more easily accessible. In the United Kingdom, for example, most general practitioners can prescribe it.

Intractable pain

Patients sometimes experience opioid dose-limiting side effects and severe pain simultaneously, a situation distressing to patient, family, and clinicians alike. In these situations, it is necessary to look quickly for other modalities, that may augment or replace systemic opioid/adjuvant therapy. Examples would be referral to anaesthesiology for consideration of a variety of regional blocks, epidural or intrathecal catheter placement, as deemed necessary.

Another strategy is the provision of low-dose ketamine. Ketamine is a dissociative anaesthetic (certain areas of the brain are affected, while others are unaffected, preserving normal muscle tone and respiration). When ketamine is used as anaesthetic doses, psychomimetic side effects, including hallucinations and nightmares, are described. There are a number of protocols for the use of low-dose, continuous ketamine infusions in refractory cancer pain. However, a recent Cochrane Database Systematic Review of ketamine as an adjuvant to opioids for cancer pain, has concluded that despite broad positive results, there are only two high quality randomized, controlled trials (RCT). The authors conclude that available evidence is insufficient to conclude that ketamine improves the effectiveness of opioid treatment in cancer pain [28, 29]. Further studies are therefore required. However, the following clinical experience, and those of others [30, 31, 32], suggest consideration of this modality as an alternative, if sequential opioid rotations have been ineffective.

Case

A 3-year-old female, with metastatic neuroblastoma, had a known relapse of localized disease in her left femur, which grew visibly over a 10-day period. She was initially given oral morphine, with good effect. Over time, this pain proved recalcitrant to increasing doses, and she was rotated to methadone. This briefly controlled her pain, but she began experiencing dose-limiting side effects. She became very sedated, picked frequently at the air, suggesting visual hallucinations, and her respiratory rate, usually 20 28, dropped to 14. Simultaneously, her pain escalated, interfering with her sleep, and causing her to cry and yell owie frequently through one day. Palliative local radiotherapy was being organized, but the full effect from that would take some days to achieve, and would not help with the immediate pain crisis. Given

P.505

the lack of efficacy from two types of opioids, and the experience of significant side effects, it was felt that rotation to a third opioid would be unlikely to optimize her pain control. At this point, the help of the anaesthesia service was sought. A continuous keta-mine infusion was suggested. The methadone was stopped thereafter, and she was started on a ketamine infusion, at a dose of 0.1 mg/kg/h. She still received boluses of intermittent morphine on the first day of ketamine therapy, but these were needed very infrequently thereafter. Her pain control was greatly improved by this therapy. She was fully conscious, and happy to play in her bed, with improved freedom from incident pain associated with movement of her affected leg. Her sleep was much better, and her interaction with her family normal. As her disease progressed, the ketamine infusion was increased to 0.2 mg/kg/h, and eventually, a morphine infusion was restarted, to a maximum of 60 mCg/kg/h, with good effect. She died peacefully in her sleep, with no signs of toxicity from either the ketamine or the morphine.

Dyspnea

Acute or progressive dyspnea is a relatively common symptom in dying children, particularly those with intrinsic lung disease, such as malignant metastases or cystic fibrosis [33], or those with extrinsic processes, such as neuro-muscular disease, or heart or kidney disease. This may be an increasingly prominent symptom as death approaches, and can signify the beginning of a fairly rapid deterioration of the body's ability to compensate against the disease process.

During the initial assessment of the patient, it may be helpful to start oxygen therapy, to assess for symptomatic relief. If the patient has a disease process in which problematic dyspnea/hypoxemia is anticipated, it is helpful to have oxygen organized for home use before it is actually needed, as getting it into the home requires significant time in many instances. Increased air movement over the child's face, by way of a fan or open window, is sometimes helpful. Another simple measure is to change the patient's position. Trying to sit the child upright, and providing distraction or massage, may relieve some of the associated anxiety. If wheeze is detected, a trial of inhaled bronchodilator is warranted, and easy to provide [34].

Dyspnea is a profoundly anxiety-provoking symptom to patient, family and health care professionals alike. Careful assessment of the cause of the dyspnea must be undertaken, to ensure that treatment of reversible causes such as pulmonary edema, pneumonia, pleural effusions, and ascites, is considered. If treatment of the above conditions with diuretics, antibiotics, chest tubes, or paracentesis respectively is ineffective, medical management should include use of systemic opioids for dyspnea [35]. These should be dosed as per initial doses for analgesia, and titrated to effect, although some patients may require smaller doses than for analgesia.

The use of nebulized opioids continues to be a matter of debate, using current evidence. Since the discovery of opioid receptors in the respiratory tract, opioids have been administered by nebulization. Some case reports reflect a mild beneficial effect on dyspnea [36, 37]. The mechanism by which this modality may contribute to the subjective sensation of breathlessness remains unclear [38]. Others have argued that benefit is not a peripheral effect in the respiratory tract, but is a result of systemic absorption. This argument seems to have gained some validity in a study done in which plasma levels of morphine were measured in healthy subjects, after nebulized or oral doses of morphine were administered, demonstrating that nebulization is a rapid, but inefficient, method of administering morphine [39]. A small double-blind randomized placebo controlled study done by Noseda et al also concluded that subjects benefited from saline or morphine via a placebo effect, and that morphine had no specific effect on dyspnea [40]. Questions about the possible role of nebulized opioids would likely be better answered by larger, well-constructed clinical studies.

If dyspnea persists despite treatment with opioids, concomitant benzodiazepines or other sedatives should be considered. Co-administration of theses drugs is not extensively reported in the literature, but there are some relevant studies [41, 42]. However, in the face of persistent dyspnea, the theoretical risks and benefits of combining these therapies should be evaluated. Ultimately, treating the panicked patient and family will usually override fears of respiratory depression, or misgivings about sedative side effects, in most situations.

If the dyspnea has developed slowly, over days to weeks, anemia may be a contributing factor, and relative risks and benefits of packed red blood cell transfusion should be contemplated, if this is detected [43]. If there is known intrathoracic tumour, initiation of systemic dexamethasone should be considered. In cases of muscle weakness causing dyspnea, referral to a pulmonary physician for evaluation of positive pressure ventilation, using nasal or facial masks, may be considered.

Bleeding/hemorrhage

Bleeding is an extremely distressing symptom to patients and families. Where sudden hemorrhage is thought to be a possibility, families should be counseled about it. In an exsanguinating hemorrhage, be the bleeding internal or external, it is paramount to discuss with the family that the patient will become rapidly unconscious, and will not suffer. This should be reinforced in an anticipatory fashion, and health care providers must ensure they restate this, if they are present during the actual event, to minimize the panic and distress

P.506

that will arise on the part of those witnessing such bleeding. A simple measure for nursing the child at risk of hemorrhage, is the use of dark fabrics (pajamas, clothing, bedding, towels, etc.), which provides much better camouflage than does light hospital linen.

In the case of brisk, but not exsanguinating, bleeding, the child will no doubt be very panicked and distressed, and a dose of intravenous or intramuscular midazolam will rapidly lead to the patient becoming unconscious, and prevent them from bearing witness to such a distressing event. The possibility of using this modality should be discussed with parents in advance, when time permits, to ensure that they understand that the goal of treatment is not to hasten the death of their child, but to allow them freedom from awareness of what is happening to them. Very often nursing and other staff need this explained in advance, too, to allow time for discussion about concerns and comfort level in providing this treatment.

For patients in whom oozing of blood is a problem, the use of the anti-fibrinolytic drug transexamic acid may be beneficial. If the bleeding is localized to oral mucosa, this can be delivered as a 5% solution mouthwash in which 30 ml is used TID. Dissolving a 500 mg tablet in 5 ml of water yields this concentration. If the bleeding is more widespread (i.e., recurrent GI bleeding) systemic use, giving 20 mg/kg enterally or intravenously every eight hours, may be beneficial in preventing breakdown of clot. Bleeding cutaneous ulcers may respond to topical application of 1 : 1000 epinephrine.

Seizures

Seizures are extremely frightening to parents, and many will state, after witnessing a first seizure, that they thought their child was dying. Seizures are, unfortunately, not rare as children die. This may be directly related to underlying disease in some cases (e.g., brain metastases or metabolic diseases), or may be related to end-stage hypoxemia, or electrolyte disturbance. It is helpful for the parents to know, first and foremost, that people are not conscious when they have generalized seizures, distressing as they may be to watch. The knowledge that their child is not suffering during this time is helpful to parents. For any patient who has a disease with potential for seizure activity, the parents must be guided through what a seizure would look like, and to have a specific plan in place, should this occur. This would include immediate positioning of the child in a safe place, dispelling of the myth that the child will swallow his tongue, and learning that no foreign objects should be placed in the mouth, in an attempt to open it. The parents should be instructed in the use of either rectal benzodiazepine administration, or in buccal application of a sublingual lorazepam tablet, holding it firmly against the buccal mucosa until it is dissolved. Rectal diazepam is commercially available in a variety of doses, but unfortunately, the cost can be prohibitive. An alternative is the rectal use of diluted injectable diazepam or lorazepam, given via a 10 ml syringe with the first 5 cm of a small feeding tube (remainder cut away), to deliver the medication rectally. Parents should also know in advance who to call for assistance.

If the seizure does not respond the first dose of medication, parents should be instructed on further doses and dosing intervals when they call for help. In rare instances, the seizure may be refractory to repeated doses of benzodiazepine. For a patient being cared for at home, this situation may require transfer by ambulance to the closest hospital, for further management with IV medications. For a family that lives remotely, and wishes that its child be kept at home no matter what occurs, it may be helpful to include pentobarbital (Nembutal) in an emergency drug kit. This drug has fallen out of favour, given the genesis of better and safer anti-epileptic/sedative medications, but still has a useful role, given per rectum, or parenterally, in this dire situation [44]. Use of this medication in such a setting would require significant advance planning, discussion, and consultation.

Superior vena cava obstruction syndrome (SVCO)

This complication arises from compression of the superior vena cava (SVC), preventing venous drainage from the head, arms, and upper trunk. Clinical features of SVCO include dyspnea from tracheal edema, headache secondary to cerebral edema, swelling of face, neck, arms and hands, visual changes, and dizziness [45]. It can arise from compression or invasion by tumor, or by the formation of a thrombus around a central subclavian catheter. In the former case, it may occur over weeks or months, allowing collateral drainage to develop. Occasionally, it occurs rapidly over days, and requires urgent treatment. Radiotherapy can be very helpful within the first two weeks of treatment, if the SVCO is caused by extrinsic tumour. In cases where radiotherapy is not possible, or is not efficacious, placement of a stent, as an interventional radiology technique, can also be considered. In the case of a thrombus of the vessel, relief of the obstruction may be obtained by removal of the catheter, although embolization is a risk if the line is left in situ, or as a complication of its removal. For patients receiving palliative care, the risks and benefits of invasive procedures need to be weighed. Factors in the decision would include the burden of suffering arising from the SVCO, the risk of proposed therapy, other comorbidities, predicted length and quality of remaining life, and so on. As a temporizing measure, a course of dexamethasone may

P.507

be helpful in short-term tumour shrinkage, and this may be particularly pertinent for children in whom this occurs very late in the courses of their lives.

Spinal cord compression

Compression of the spinal cord is a palliative care emergency, as a significant number of patients will survive for a significant length of time after this occurs, and have the added devastation of paraplegia to contend with, if it is not adequately recognized and immediately treated. In adults, the great majority of patients develop this complication secondary to extension of vertebral body metastases into the epidural space [45]. Back pain, often worsened by the Valsalva maneuver, should alert the physician or nurse of this possibility, when caring for a child with cancer. Other warning signs are unpleasant sensations in limbs. Sensory and motor loss occur later, and sphincter disturbance is a late sign. The final stage is cord ischemia, that occurs rapidly over hours. High dose dexamethasone, given immediately, will reduce spinal cord edema, and may temporarily prevent the onset of cord ischemia. Radiotherapy is the definitive treatment, and surgery is indicated for spinal instability in patients who are early in their course. Again, when this symptom occurs very close to the end of life, corticosteroids alone may be employed, provided they do not cause undesired side effects such as emotional lability. The symptoms associated with spinal cord compression can be severe, and intervention with surgery and/or radiotherapy should be discussed with the family, and (where appropriate) the patient, even when he or she is unlikely to survive more than a few weeks. For many, the benefits of such intervention will be outweighed by the possibility of prolonged hospitalisation. Interventions that are commenced more than 48 h after the onset of symptoms of spinal cord compression are highly unlikely to relieve symptoms, and aggressive management should not usually be recommended.

Sedation for intractable distress

Despite the progress made in pain and symptom management over the last few decades, distressing situations still occur, in which symptoms remains uncontrolled in the face of all reasonable efforts to alleviate them. The suffering incurred by patients and families in this circumstance is immense, and the option of providing sedation to the patient, for relief of refractory symptoms, is often entertained at this stage.

Unfortunately, despite descriptions of terminal and palliative sedation in recent medical literature, there remains much confusion about the intended meaning. Descriptions range from mild intermittent sedation at one end of the spectrum, in which patients are able to communicate at least episodically, to deep continuous sedation, in which the patient is likely to remain unconscious until death, at the other end [46, 47]. There has been a recent move to rename this therapy, sedation for intractable distress in the dying patient. Use of the term terminal sedation has been discouraged, because of the misinterpretation of intent (connotation of euthanasia), and because it does not convey the gravity of the severe suffering which this type of sedation serves to lessen. Clearly, when titrated against a particular symptom, this form of sedation can be seen as morally justifiable, as would be the array of potent opioids we bring to palliative care under the auspices of the principle of double effect [48]. In short, provision of relief from a distressing symptom is the morally defensible intent, even if there are foreseeable, but unintended, harmful effects, including respiratory depression, or even the hastening of death. This is in stark contrast to the intent of euthanasia, which is to end suffering by means of ending life. Unfortunately, a class of these sedative agents (barbiturates) is used in Holland for performing euthanasia, and has been recommended for assisting suicide, further adding to the misunderstanding. A recent survey of palliative care experts from eight countries showed 89% agreement that sedation for relief of intractable symptoms is sometimes necessary, and that it was felt to be effective in 90% of cases recalled. Of these experts, 90% were not in support of legalized euthanasia [49]. Importantly, a study in a specialist palliative care unit showed that sedative use was not associated with shorter survival, suggesting that the doctrine of double effect rarely has to be invoked to justify sedative prescribing at the end of life [50].

Before sedation is considered for symptom control, the advice of colleagues with expertise in palliative symptom management and/or anaesthesia should be sought, to ensure that the symptom is truly refractory. In the relatively rare circumstance where this proves to be the case, careful discussions must be held with the parents, involved health care professionals, and the patient, where appropriate. The goal of therapy relief of suffering from the symptom must be clearly demarcated from euthanasia. It must be made clear that aside from this alternative, there are no further medical options. It is imperative that the staff attending the family be fully informed about the implementation of this therapy, and related myths dispelled. Otherwise, unhealthy situations can evolve, in which parents are challenged by staff, or equally inappropriately, parents feel the burden of having to help health care professionals cope. Under no circumstance should parents be made to feel that they are choosing between their child's comfort and their child's life another possible innuendo the parents may feel, if professional caregivers are not well-prepared.

Pharmacologic options for provision of sedation for intractable symptoms include benzodiazepines, neuroleptics,

P.508

barbiturates and general anaesthetics. Kenny and Frager [51] provide specific guidelines for use of these medications in children at the end of life. Starting doses should be conservative, given the concomitant administration of other centrally acting medications, and can be titrated upwards to effect. A frequently given benzodiazepine is lorazepam, in sublingual (SL), intravenous (IV), or subcutaneous (SC) form. Starting doses of 0.02 0.05 mg/kg are reasonable, and may be given every 6 8 h. Sublingual tablets may be held against the buccal mucosa until absorbed. Diazepam is irritating SC, but may be given IV (0.1 mg/kg), or per rectum (PR) 0.3 mg/kg, in the same time interval.

Neuroleptics, such as haloperidol (0.01 mg/kg), or methotrimeprazine (leromepromazine) (0.1 mg/kg), may be given IV/SC every 8 h.

If symptoms are refractory, despite rapid escalation in 20 50% aliquots from suggested starting doses, advice from anaesthesiologists, regarding use of barbiturates such as pentobarbital, and general anaesthetics such as propofol or ketamine, is suggested.

References

1. Wolfe, J., Grier, H., Klar, N. et al. (2000). Symptoms and suffering at the end of life in children with cancer. N England J Med 342: 326 33.

2. The SUPPORT Principle Investigators (1995). A controlled trial to improve care for seriously ill hospitalized patients: the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments. JAMA 274:1591 8.

3. Anonymous (1998). Guidelines for Analgesic Drug Therapy. In: Cancer pain relief and palliative care in children, pp. 24 8. World Health Organization, Geneva.

4. Kyriakides, K., Hussain, S.K., Hobbs, G.J. (1999). Management of opioid-induced pruritus: a role for 5-HT3 antagonists? Br J Anaesth. 82:439 41.

5. McGrath, P.J., Finley, A. (1996). Attitudes and beliefs about medication and pain management in children. J Palliat Care 12:46 50.

6. Schecter, N.L. (1989). The undertreatment of pain in children: an overview. Pediat Clin North Am 36:781 794.

7. Liben, S. (1996). Pediatric palliative medicine: obstacles to overcome. J Palliat Care 12:24 8.

8. Cantwell, P., MacKay, S., Macmillan, K., Turco, S., McKinnon, S., Read-Paul, L. (1998). Pain. In: Brenneis C., Perry B., Read-Paul L., Bruera E., ed. 99 Common questions (and answers) about Palliative care: a nurses handbook. pp. 20. Regional Palliative Care Program, Edmonton.

9. Pereira, J., Bruera, E. (2000). Table of equianalgesic dose's of opioids. Palliative care handbook, pp. appendix B. Alberta Cancer Board, Edmonton.

10. Collins, J.J., Dunkel, I.J., Gupta. S.K. et al. (1999). Transdermal fentanyl in children with cancer pain: Feasibility, tolerability and pharmakokinetic correlates. J Pediatr 134:319 23.

11. Hunt, A., Goldman, A., Devine, T., Phillips, M. (2001). Transdermal fentanyl for pain relief in a pediatric palliative care population. Palliat Med 15:405 12.

12. No authors cited (2000). Evidence-Based recommendations for medical management of chronic non-maligant pain. Reference Guide for clinicians for the treatment of chronic non-malignant pain. pp. 39 40. College of Physicians and Surgeons of Ontario.

13. Gagnon, B., Bruera, E. (1999). Differences in the ratios of morphine to methadone in patients with neuropathic pain versus non-neuropathic pain. J Pain Symptom Manage 18:120 25.

14. Ripamonti, C., Zecca, E., Bruera, E. (1997). An update on the clinical use of methadone for cancer pain. Pain 70:109 15.

15. Ripamonti, C., Groff, L., Brunelli, C., Polastri, D., Stavrakis, A., De Conno F. (1998). Switching from morphine to oral methadone in treating cancer pain: what is the equianalgesic dose ratio? J Clin Oncol 16:3216 21.

16. Bruera, E., Pereira, J., Watanabe, S., et al. (1996). Opioid rotation in patients with cancer pain; a retrospective comparison of dose ratios between methadone, hydromorphone, and morphine. Cancer 78:852 7.

17. Wiser, A.W., Miser, J.S. (1985). The use of oral methadone to control moderate and severe pain in children and young adults with malignancy. Clin J Pain 1:243 8.

18. Shir, Y., Shenkman, Z., Shavelson, V., Davidson, E., Rosen, G. (1998). Oral methadone for the treatment of severe pain in hospitalized children: a report of five cases. Clin J Pain 14:350 53.

19. Zeltzer, L., Bush, J., Chen, E., Riveral, A. (1997). A Psychobiologic approach to pediatric pain: Part II: prevention and treatment. Curr Probl Pediatr 27:264 84.

20. Berde, C. (1991). The treatment of pain in children., In: Bond MR, Charlton J.E. and Woolf C.J. ed. Proceedings of the VIth World Congress on Pain, pp. 435 40. Elsevier Science Publishers.

21. Collins, J.J. (1996). Intractable pain in children with terminal cancer. J Palliat Care 12:29 34.

22. Cooper, M.G., Keneally, J.P., Kinchington, D. (1994). Continuous brachial plexus neural blockade in a child with intractable cancer pain. J Symptom Manage 9:277 81.

23. Gurnani, A., Sharma, P.K., Rautela, R.S., Bhattacharya, A. (1996). Analgesia for acute musculoskeletal trauma: low-dose subcutaneous infusion of ketamine. Anaesth Intensive Care 24:32 6.

24. Klepstad, P., Borchgrevink, P., Hval, B., Flaat, S., Kaasa, S. (2001). Long-term treatment with ketamine in a 12-year-old girl with severe neuropathic pain caused by a cervical spinal tumor. J Pediatr Hematol Oncol 23:616 19.

25. Clark, J.L., Kalan, G.E. (1995). Effective treatment of severe cancer pain of the head using low-dose Ketamine in an opioid-tolerant patient. J Pain Symptom Manage 10:310 14.

26. Robinson, W.M., Ravilly, S., Berde, C., Wohl M.E. (1997). End-of-life care in cystic fibrosis. Pediatrics 100:205 9.

27. Regnard, C.F., Tempest, S. (1998). Respiratory Problems. In: A guide to symptom relief in advanced Disease, Fourth Edition, pp. 32 34. Hochland and Hochland Ltd, Cheshire, England.

28. Kvale, P.A., Simoff, M., Prakash, U.B. (2003). Palliative Care. Chest 123:284S-311S.

29. Cohen, S.P., Dawson, T.C. (2002). Nebulized morphine as a treatment for dyspnea in a child with cystic fibrosis. Pediatrics 110, e38.

P.509

30. Coyne, P.J., Viswanathan, R., Smith, T.J. (2002). Nebulized fentanyl citrate improves patients' perception of breathing, respiratory rate, and oxygen saturation in dyspnea. J Pain Symptom Manage 23:157 60.

31. Zebraski, S.E., Kochenash, S.M., Raffa, R.B. (2000). Lung opioid receptors: pharmacology and possible target for nebulized morphine in dyspnea. Life Sci 66:2221 31.

32. Masood, A.R., Thomas, S.H. (1996). Systemic absorption of nebulized morphine compared with oral morphine in healthy subjects. Br J Clin Pharmacol 41:250 2.

33. Noseda, A., Carpiaux, J.P., Markstein, C., Meyvaert, A., de Maertelaer, V. (1997). Disabling dyspnoea in patients with advanced disease: lack of effect of nebulized morphine. Eur Respir J 10:1079 83.

34. Zeppetella, G (1998). The palliation of dyspnea in terminal disease. Am J Hosp Palliat Care 15:322 30.

35. Fainsinger, R.L., Waller, A., Bercovici, M. et al. (2000). A multicentre international study of sedation for uncontrolled symptoms in terminally ill patients. Palliat Med 14:257 65.

36. Gleeson, C., Spencer, D. (1995). Blood transfusion and its benefits in palliative care. Palliat Med 9:307 13.

37. Collins, J.J., Devine, T.D., Dick, G.S. (2002). The measurement of symptoms in young children with cancer: the validation of the Memorial Symptom Assessment Scale in children aged 7 12. J Pain Symptom Manage 23:10 16.

38. Wilwerding, M.B., Loprinzi, C.L., Mailliard, J.A. et al. (1995). A randomized, crossover evaluation of methylphenidate in cancer patients receiving strong narcotics. Support Care Cancer 3:135 8.