Chapter 67: Hospice & Palliative Medicine

Homes for the dying or, as they were soon to be called, hospices, were established in Ireland and France in the nineteenth century. However, it was not until 1967 that the first modern hospice, Saint Christopher’s Hospice, was founded in London. There, Dr. Cicely Saunders, a former nurse and social worker who had earned a medical degree, helped establish the underlying philosophy of hospice and palliative medicine. She emphasized clinical excellence in pain and symptom management; care of the whole person, including physical, emotional, social, and spiritual needs; and the need for research in this newly developing field of medicine. Interdisciplinary team care became the norm, as it became clear that no one physician, nurse, social worker, or chaplain could address all the needs of the terminally ill person. Further, although the focus of care was clearly on the dying individual, the needs of the family were also addressed.

In 1982, the Congress created the Medicare Hospice Benefit (MHB), and in 1986, the benefit was made permanent. By 2007, 4700 hospice programs were providing healthcare services to the terminally ill and their families throughout the United States.

Eligibility criteria for hospice enrollment through the MHB require that patients waive traditional Medicare coverage for curative and life-prolonging care related to the terminal diagnosis and be certified by their physician and the hospice medical director as having a life expectancy of 6 months or less if the disease runs its usual course. Recertification periods within the MHB allow for reexamination of hospice eligibility. If the hospice medical director believes that the patient has a life expectancy of ≤6 months if the disease runs its usual course, the patient may be recertified as eligible for the MHB even if the patient has already been receiving the benefit for 6 months or longer.

The goal of hospice care is to relieve suffering and improve the patient’s and family’s quality of daily life. To achieve those goals, hospice care has come to be defined as holistic, patient-, and family-centered rather than disease-centered. Hospice provides a team composed of those trained to care for problems in a holistic manner: physician, nurse, social worker, chaplain, bereavement counselor, nursing assistant, and volunteer. The hospice team meets weekly, under the direction of the hospice medical director, to review the care plans of all patients. The hospice program is charged with providing medications for the relief of physical distress, durable medical equipment, supplies, a multidisciplinary team to provide care, and bereavement support before and after the patient’s death.

Palliative medicine has developed as a medical subspecialty in the United States since the mid-1990s, bringing a “hospicelike” approach to patients with serious illnesses regardless of prognosis or their interest in pursuing life-prolonging treatments. The goals of palliative care programs are similar to those of the hospice: pain and symptom control; emotional, social, and spiritual support of patients and families; and facilitation of clear and compassionate communication regarding goals of care. Early involvement of palliative care has been shown to significantly improve symptom management, quality of life, mood, and, in one study, survival. Furthermore, palliative care has been shown to increase patient and family satisfaction and reduce costs by limiting the use of high-technology care.

Many models of palliative care are under development. There are palliative care consultation teams in hospitals, nursing homes, and outpatient clinics. The growth of palliative care consultation programs since 2004 or so has been significant and mirrors the growth of the hospice since 1974. The number of hospital-based palliative care consultation programs has increased linearly from 632 (15%) in 2000 to 1027 (25%) in 2003 and continues to grow annually. Many larger hospitals and all VA institutions have consultative palliative care services. Many hospitals have also established inpatient palliative care units.

Palliative medicine fellowships are undergoing rapid development as well. In the early 1990s, the first US palliative medicine fellowship was initiated. By June 2006, 49 programs offered 119 fellowships for advanced training. In September 2006, the American Board of Medical Specialties officially recognized the field of palliative medicine as a subspecialty.

Optimal symptom control is an important cornerstone of palliative medicine, because uncontrolled symptoms increase patients’ and caregivers’ distress. Poorly controlled symptoms often detract from patients’ quality of life, impair their interactions with loved ones, and limit their ability to attend to important issues at the end of life. Many studies have documented the high frequency of symptoms in patients with serious illnesses and the tendency for symptoms to increase in intensity as a disease progresses. The following discussion reviews management of some common symptoms. As with most medical problems, successful management of symptoms starts with a careful history and physical examination, with therapy directed at identifiable underlying causes.

Pain

Pain can be classified physiologically as nociceptive (somatic or visceral) or neuropathic. (Table 67-1). Pain can occur directly from the underlying illnesses (such as tumor involvement, cytokine release, and vasculopathy), as a consequence of therapy (particularly chemotherapy, radiation therapy, and invasive procedures), or from pathologies that are not directly related to the primary disease processes. It is important to remember that pain is a subjective experience, and it is essential to respect and accept the complaint of pain as characterized by the patient. Pain is influenced by psychosocial and spiritual issues, and effective pain management requires a multidisciplinary approach.

Unlike opioids, nonsteroidal anti-inflammatory drugs (NSAIDs) and acetaminophen have an analgesic ceiling effect. The use of opioid-nonopioid combinations therefore is limited by the dose of the NSAIDs or acetaminophen. Despite this fact, NSAIDs are effective pain medication, especially for inflammatory conditions. Their use can decrease the amount of opioids required and hence decrease the incidence of opioid side effects. Unless contraindicated, all pain protocols should include a NSAID or acetaminophen.

The general principles of pain management with opioids are as follows:

1. Assess pain using a standardized pain scale: Most commonly used is a 0–10 scale, where 0 indicates no pain and 10 represents the worst pain imaginable. Numerous validated scales can be used for patients who cannot communicate–either because they are intubated, are preverbal children, or are cognitively impaired adults.

2. In opioid-naive patients, start with a short-acting opioid (such as morphine, oxycodone, or hydromorphone) to control acute, moderate to severe pain. Unless there are contraindications, morphine is the agent of choice given its low costs and variable routes of administration [oral (PO), intravenous (IV), subcutaneous (SQ), intramuscular (IM), or rectal (PR)]. Short-acting opioids can be given as frequently as the time to peak onset of action: every 60–90 minutes for oral immediate-release formulations and every 10–15 minutes for IV formulations. For severe pain, IV administration is recommended given the faster onset of action and greater ease of titration. Conversion to oral opioids can occur once the pain is controlled.

3. Determine whether the dose is adequate, and adjust accordingly: The dose should be titrated at least every 24 hours if the pain is moderate, and as often as every 4 hours if the pain is severe, particularly while using intravenous opioids. Dose increases should be made by 25–50% for moderate pain, and by 50–100% for severe pain. There is no specific limit to opioid doses. These agents should be titrated until pain is controlled or side effects develop.

4. Further titration or rotation of opioids is achieved by first calculating the oral morphine equivalent (OME) of the previous 24 hours’ worth of opioid use. Equianalgesic tablets for opioids are readily available (Table 67-2); however, these tablets often differ slightly. It is important to remember that oral and parenteral doses are not equal because of oral medicines’ first-pass hepatic metabolism– for example, 1 mg of IV morphine sulfate equals 3 mg of oral morphine sulfate, and 1 mg of parenteral hydromorphone equals 4 mg of oral hydromorphone.

5. Determine the dosing schedule: Chronic pain deserves scheduled pain medication, not just as-needed dosing: 66–75% of the patient’s stable 24-hour OME needs should be given as a long-acting formulation.

6. Determine the breakthrough dose. Breakthrough pain is defined as a transitory exacerbation of pain that occurs on top of an otherwise stable persistent pain. An adequate breakthrough dose is calculated as 10–15% of the total daily long-acting opioid, and it should be given every 3 hours as needed. Whenever possible, the same opioid agent should be used for both short- and long-acting administration [eg, sustained-release morphine 150 mg PO Q12h (by mouth every 12 hours) and immediate-release morphine 30–45 mg PO Q3h, as needed).

7. During administration of opioid therapy, the patient’s renal function should be closely monitored, as toxic metabolites can accumulate with decreased kidney function, leading to alterations in mental status, ­myoclonic jerks, and seizures. Opioids that are safe to use in renal failure include hydromorphone, fentanyl, and methadone.

8. Reasons for rotating to a different opioid include renal failure, side effects, and a need to change the route of administration (such as changing from oral morphine to transdermal fentanyl). When rotating to a different opioid, the dose should be reduced by 25–50% to account for incomplete cross-tolerance.

Fentanyl patches should not be used alone for acute severe pain. Because of the delayed onset of effect (12 hours) and long half-life of this formulation, which allows for titrations only every 48–72 hours, it cannot be titrated quickly for rapid pain control. If the pain is severe or unpredictable, or if opioid requirements are unknown or increasing rapidly, patient-controlled analgesia (PCA) may be indicated. With PCA infusion pumps, opioids can be infused intravenously or subcutaneously at a continuous basal rate programmed by the care provider, while the administrations of bolus doses are controlled by the patient. There is theoretically no risk of overdose since the patient will fall asleep before serious signs of overdose occur, and for this reason, the patient’s family or visitors should be carefully educated to avoid pushing the PCA button on the patient’s behalf.

Guidelines on management of opioid-induced side effects are as follows:

1. Nausea/vomiting: Opioids can cause nausea and vomiting by decreasing gastrointestinal motility. Patients usually develop a tolerance to these side effects. It may be helpful to schedule an antiemetic for a few days, then change to as-needed administration. Dopamine receptor antagonists are most effective (haloperidol 0.5–2 m Q6–12h or metoclopramide 10–40 mg Q6h).

2. Sedation: This is a commonly encountered side effect, although tolerance typically develops over time to this as well. Downtitration of opioids, rotation of opiates, and use of adjuvant, nonopioid pain therapies should be considered. Other sedating agents should be eliminated whenever possible. If sedation persists despite these interventions, the addition of a central nervous system (CNS) stimulant can be helpful (methylphenidate 2.5–5 mg in the morning and at noon, or modafinil 100–200 mg daily).

3. Constipation: This is one of the most common side effects. Since patients rarely develop tolerance, if at all, a bowel regimen must be started for most patients with initiation of opioid therapy. A bowel stimulant (senna) is the most commonly used agent. Methylnaltrexone is a peripherally acting opioid receptor antagonist that has been approved for refractory constipation in patients receiving opioid therapy. It does not reverse central analgesia.

4. Delirium/confusion/hallucinations: In this setting, opioid dose reduction, rotation to a different opioid, and initiation of neuroleptic therapy (haloperidol 0.5–1 mg BID-QID or olanzapine 2.5–5 mg daily BID should be considered).

5. Allergic reaction: True allergic reactions are rare. Allergylike symptoms are usually secondary to mast cell activation and subsequent histamine release.

6. Respiratory depression: Tolerance to the respiratory depressant effects occurs rapidly; thus, opioids can be used safely when titrated to pain control, even in patients with underlying emphysema. Naloxone administration is indicated if (a) the patient is somnolent and difficult to arouse or (b) respiratory rate is <8/min or oxygen saturation is <92% and the respiratory rate is <12/min. A dilution of 0.4 mg naloxone (one ampoule, 1 mL) in 9 mL of normal saline to yield 0.04 mg naloxone per milliliter should be prepared, with administration of this diluted nalxone in 1–2-mL increments (0.04–0.08 mg) over 1–2-minute intervals until a change in alertness is observed. Giving naloxone in this way should not cause pain to return or opioid withdrawal. Remember that the half-life of naloxone is shorter than the half-life of most opioids, so respiratory depression may recur and a naloxone drip may be needed.

A challenge in prescribing opioid pain medications is to correctly differentiate a patient in pain from a patient with a substance abuse disorder. The key lies in understanding the specific characteristics of tolerance, physical and psychological dependence, and pseudoaddiction. Tolerance is a state of adaption in which exposure to a drug induces changes that result in a decreased effect of the drug dose over time. Physical dependence is a state of adaption that is manifested by a specific withdrawal syndrome when the drug is discontinued suddenly. Most patients on chronic opioids will develop physical dependence. If the need arises for a rapid decrease in opioid dose, administering 25–50% of the stable dose can prevent withdrawal symptoms. Both tolerance and physical dependence cannot be used to differentiate between a patient in pain and a patient with a substance abuse disorder. Psychological dependence or addiction is a primary, chronic, neurobiological disease with genetic and psychosocial factors influencing its development. It is characterized by certain behaviors, including loss of control, compulsive drug use, craving, and continued use of the drug despite harm. Research suggests that opioids used to treat pain rarely lead to psychological dependence. Pseudoaddiction describes a patient’s behavior when the pain is undertreated. Some of these behaviors, such as “clock watching” or having “excessive pain,” may mimic addiction. However, a distinguishing feature of pseudoaddiction not seen in true addiction is the resolution of these behaviors when pain is effectively treated.

A. Adjuvant Pain Therapies

Patients with neuropathic pain occasionally respond to opioids alone; however, many require the addition of adjuvant pain medications. Commonly used adjuvants for neuropathic pain include tricyclic antidepressants (TCAs), serotonin and norepinephrine reuptake inhibitors (SNRIs), anticonvulsants, and antiarrhythmics. The choice of an adjuvant is usually dictated by the individual drug side effect profile, the potential for drug interactions, and the previous drug therapy. The secondary amines, nortriptyline and desipramine, are generally better tolerated than amitriptyline. The analgesic effects of TCAs occur at lower doses and usually within several days, as compared with the antidepressant effects. Data on use of selective serotonin reuptake inhibitors (SSRIs) for neuropathic pain are not convincing; however, recent studies suggest that SNRIs may be as beneficial at tricyclic agents for pain control. Of the anticonvulsants, gabapentin, pregabalin, carbamazepine, and valproic acid are commonly used for neuropathic pain. Carbamazepine and valproic acid are cost-effective but have a higher risk of drug interactions and toxicity compared with gabapentin and pregabalin. Gabapentin requires more frequent dosing, slower titration secondary to sedation, and dose adjustments for renal insufficiency. Antiarrhythmics, topical lidocaine, and oral mexiletine have also been used successfully for neuropathic pain. For adjuvant pain medications, standard initial dosing and titration guidelines should be followed, although lower-than-usual doses have been effective for pain control. In elderly patients, it is generally safer to start at low doses and titrate at a slower rate.

Corticosteroids, benzodiazepines, and anticholinergics are also used as adjuvant pain medication. Corticosteroids, by decreasing tumor-associated edema and by their anti-inflammatory effects, are useful for pain because of multiple pathologies, including bone metastasis, liver capsule distention from metastasis, and conditions in which the tumor is compressing sensitive structures. Benzodiazepines and baclofen are indicated for pain from spasticity. Anticholinergics can relieve colic due to intestinal obstruction.

In addition to drug therapy for pain control, interventions such as palliative radiation therapy for bone metastasis, nerve blockage (eg, celiac plexus block for pancreatic cancer), palliative surgical resection, or immobilization of fractures should be considered. Before undertaking such interventions, the patient’s overall prognosis and the effectiveness of less invasive measures should be considered. Complementary therapies are often used in hospice and palliative care for treatment of pain and other symptoms. Some of these therapies are described in Table 67-3.

Nausea & Vomiting

Nausea and vomiting entail complex physiologic processes and are triggered by activation of one of four main pathways. An understanding of these pathways may aid in choosing an effective antiemetic regimen.

1. The chemoreceptor trigger zone (CTZ) is located at the area postrema of the medulla. It lies outside the blood-brain barrier and is therefore able to sample emetogenic toxins, drugs, or metabolic abnormalities such as uremia or hypercalcemia. It further receives input from the gastrointestinal tract. The main receptors involved include dopamine type 2 receptors (D2), 5-hydroxytryptamine type 3 receptor (5-HT-3), and neurokinin type 1 receptor (NK1).

2. Pathways from the vestibular apparatus respond to vertigo and visuospatial disorientation. The main receptors involved include muscarinic acetylcholine receptors (m-Ach) and histamine type 1 receptors (H1).

3. Peripheral pathways (vagus nerve and splanchnic nerves) mediate nausea triggered by activation of visceral chemoreceptors (local toxins) and serosal mechanoreceptors (stretch of organs and capsules). The main receptors involved include H1 and m-Ach receptors. Enterochromaffine cells release 5-HT-3 when damaged by interventions such as chemotherapy or radiation therapy.

4. Cortical pathways respond to increased intracranial pressure, sensory stimuli (smell, pain), and psychogenic stimuli (anxiety, memory, conditioning).

Each of these pathways sends signals to the vomiting center (VC), which triggers nausea and vomiting when thresholds are reached. The main receptors involved at the VC are H1, m-Ach, and 5-HT-2.

Treatment should focus on correcting underlying causes as well as choosing antiemetics to target specific receptors and pathways involved. Commonly used antiemetics are described in Table 67-4. If nausea or vomiting is persistent, severe, or refractory, it is recommended to schedule regular administrations of an antiemetic agent. A second or third antiemetic targeting a different receptor may be added (scheduled dosing, or dosed as needed).

Nausea and vomiting also may be presenting symptoms of a malignant gastrointestinal obstruction. The patent should be evaluated for invasive procedures such as venting gastrostomy, intraluminal stent, or surgical diversion. With partial obstruction, the use of metoclopramide and dexamethasone along with a low-fiber diet can provide significant symptom relief for several weeks or longer. When an obstruction becomes complete, therapy is directed at decreasing intestinal motility and decreasing secretions using anticholinergics and somatostatin analogs. Metoclopramide is contraindicated in complete bowel obstruction as it increases gastrointestinal motility and leads to exacerbation of painful abdominal cramping.

Dyspnea

Dyspnea, like pain, is a subjective experience, and can be present with or without hypoxia. With a broad differential existing for dyspnea, reversible causes should always be considered first. The optimal therapy is aimed at the presumed etiology. Palliative therapy can involve chemotherapy, radiotherapy, thoracentesis, pericardiocentesis, and bronchial stent placement. General measures such as providing a fan, keeping the room temperature cool, use of relaxation techniques, or using a careful trial of supplemental oxygen (for hypoxic patients) can help dyspneic patients. Available palliative drug therapies include steroids, opioids, bronchodilators, diuretics, anxiolytics, antibiotics, and anticoagulants. All these drugs can be used in combination, depending on the etiology of dyspnea.

Opioids can relieve breathlessness, although the mechanism is unclear. Opioid administration, dose, frequency, and titration are the same as for pain control. The use of nebulized morphine sulfate is not more effective than placebo. Opioids can increase exercise tolerance and reduce dyspnea in patients with chronic obstructive airways. Fear of addiction or fear of respiratory depression should not preclude a trial of opioids in this population. Starting at low doses, carefully titrating the dose to achieve symptom control, and close monitoring allow for safe and effective use.

Steroids are useful for dyspnea from bronchospasm and tumor-associated edema. Specific indications include malignant bronchial obstruction, carcinomatous lymphangitis, and superior vena cava syndrome. Dexamethasone can be started at 4 mg twice daily and subsequently reduced to the lowest effective dose. Dexamethasone is more potent and has lower mineralocorticoid activity than other steroids, resulting in less fluid retention.

Some patients with dyspnea express disturbing fears of suffocation and choking. Understandably, anxiety often coexists with chronic dyspnea. Anxiety can heighten breathlessness, making symptom control more difficult. The use of anxiolytics such as benzodiazepines and phenothiazines can help treat dyspnea associated with a high component of anxiety. Lorazepam, 0.5–1 mg, can be tried initially. If patients show benefit, long-acting diazepam or clonazepam can then be prescribed. Low-dose chlorpromazine has also shown benefit in relieving both dyspnea and anxiety.

Anorexia & Cachexia

Anorexia (poor appetite) and cachexia (involuntary weight loss regardless of caloric intake or appetite) are prevalent distressing symptoms in patients with advanced illnesses. Mechanisms of the anorexia-cachexia syndrome include an aberrant inflammatory response, generated by disease-host reactions, as well as neurohormonal dysfunction. Exacerbating factors include delayed gastric emptying, constipation, nausea, depression, mucositis, thrush, and even ill-fitting dentures.

Little effective drug therapy is available. Megestrol acetate, a progestin, has been shown to increase appetite and result in weight gain; however, weight gain is due largely to accumulation of adipose tissue and not lean muscle mass. Doses start at 160 mg/day and can be titrated to 800 mg/day if required. Side effects include thromboembolic events and adrenal insufficiency on abrupt cessation. Corticosteroids, such as dexamethasone, can be prescribed as an appetite stimulant for patients in whom side effects of long-term steroid use are of less concern. Beneficial effects tend to be limited to several weeks. Significant weight gain is not seen with corticosteroids in this population. Dexamethasone can be started at 2–4 mg daily, with titration to 16 mg daily if required. The lowest effective steroid dose should always be used. Androgens dronabinol and growth hormones have been effective for patients with AIDS-associated anorexia and cachexia.

Nutritional support, parenteral and enteral, has not been shown to prolong survival in patients with advanced cancer who are not candidates for disease-specific therapy. Exceptions include patients with head and neck cancer undergoing radiation therapy or patients with gastrointestinal dysfunction and otherwise good performance status.

Asthenia

Asthenia is well described in patients with cancer, but it also affects patients with end-stage organ dysfunction. It is one of the most disabling symptoms yet at the same time remains underrecognized and poorly treated. It is characterized by excessive physical, emotional, and cognitive fatigue that is not relieved by rest. Etiologies are often multifactorial, and treatment should be aimed at treating underlying and reversible causes, including hypoxia, anemia, electrolyte disturbances, dehydration, insomnia, or uncontrolled symptoms (such as pain, nausea, vomiting, and constipation). Unfortunately, when disease-specific therapy is not effective, asthenia is difficult to palliate. Education of the patient and family, normalization of the experience, counseling about energy conservation, mild exercise programs, light exposure therapy, and cognitive behavioral therapy can be effective. Symptomatic drug therapy includes corticosteroids (dexamethasone 2–4 mg PO once or twice daily) and psychostimulants (methylphenidate 2.5–5 mg PO once or twice daily). To lessen potential insomnia at night, these drugs should be administered early in the day.

Depression

There is a common assumption that symptoms of depression are normal or expected in patients facing life-threatening illness. This thought promotes the underdiagnosis of depression and, in turn, its undertreatment. Depressive states exist on a continuum from normal sadness that accompanies life-limiting disease to major affective disorders. It is important that clinicians differentiate among these levels of distress.

Besides a personal or family history of psychiatric illness, risk factors for depression in palliative care patients include young age, poor social support, worsening illness, high symptom burden, poor functional status, and the use of certain medications such as corticosteroids, interferon, interleukin-2, and some chemotherapeutic agents.

Diagnosing depression in physically healthy patients depends heavily on the presence of neurovegetative symptoms such as decreased appetite, loss of energy, insomnia, loss of sexual drive, and psychomotor retardation. Given that these symptoms are frequently present in patients with advanced illness, they are less reliable for diagnosing depression in this patient population. Elements of the history that can be helpful in diagnosing depression include anhedonia, feelings of hopelessness, worthlessness, helplessness, excessive guilt, and suicidal ideation. A quick and effective way to screen for depression is to ask the following two questions: “Are you feeling down, depressed or hopeless most of the time over the last 2 weeks?” and, “Have you found that little brings you pleasure or joy over the last 2 weeks?”

Treatment of depression in the palliative care population is similar to the general population. Since antidepressants take as much as 4–8 weeks to show effect, it is reasonable to start treatment with a psychostimulant in patients with a prognosis of <6 months. Effects are usually felt within 1–2 days and include improved mood, energy, appetite, and cognition. Other effective treatment options include psychotherapeutic interventions.

It is important to differentiate depression from preparatory or anticipatory grief, which is defined as the “grief that a terminally ill person has to undergo in order to prepare himself for his final separation from this world” (Elisabeth Kübler-Ross). Features include withdrawal from loved ones, rumination about the past, sadness, crying, and anxiety. In contrast to depression, mood changes are typically transient, and patients maintain the capacity for experiencing pleasure. Further, patients’ self-image is typically not disturbed, and they are able to maintain hope, although the focus of their hope may shift.

Anxiety

Anxiety is a common symptom in patients facing serious illness. It is characterized by worries or fears stemming from one’s perception of a threat in either the present or the future. Anxiety may be present as part of a primary psychiatric disorder, including generalized anxiety disorder, panic disorder, adjustment disorder, phobias, and acute or posttraumatic stress disorders. Anxiety may also be a secondary component of other symptoms, including pain, shortness of breath, or nausea. Furthermore, it can be a sign of drug withdrawal, including alcohol, opioids, benzodiazepines, antidepressants, and nicotine; and in addition, it can be an adverse drug reaction from medications such as corticosteroids, psychostimulants, and some antidepressants. Anxiety may have metabolic causes such as hyperthyroidism or syndromes of adrenergic or serotonergic excess. Often anxiety is also a sign of existential or psychosocial concerns about disease progression, disability, loss, dying, legacy, family, finances, and religion or spirituality.

Several formal screening tools for anxiety exist. The Edmonton Assessment Scale and the Hospital Anxiety and Depression Scale are most frequently used. Besides a thorough history and physical exam, the evaluation of a patient with anxiety should include an assessment of prior episodes of anxiety, depression, posttraumatic stress disorder (PTSD), and alcohol and drug use. It is helpful to ascertain if there are specific thoughts or situations triggering anxiety. Symptoms that can be misdiagnosed as anxiety include agitated delirium, cardiac arrhythmias, or akathisia, which is a symptom induced by dopamine receptor blocking agents such as metoclopramide or antipsychotics and manifests as an unpleasant sensation of restlessness.

The first step of treatment is active listening, with efforts to normalize the patient’s experience and provide support. Pharmacologic treatment options include medications such as SSRIs (selective serotonin reuptake inhibitors) and benzodiazepines. Longer-acting benzodiazepines such as clonezepam are usually preferred, as they are less likely to cause rebound anxiety. The patient’s subjective level of distress is the primary indication for the initiation of pharmacologic treatment. Nonpharmacologic treatment options include psychotherapeutic interventions such as cognitive behavioral therapy or supportive expressive therapy. The use of relaxation techniques or guided imagery is also helpful.

Delirium

Delirium is a frequently overlooked diagnosis, and it is associated with increased morbidity and mortality. It is characterized by an alteration of consciousness with a reduced ability to focus, sustain, or shift attention. Additional features are changes in cognition (such as memory deficits, language disturbances, and disorientation) and the development of perceptual disturbances. These changes are acute in onset and typically fluctuate throughout the day. Further symptoms include changes in the sleep/wake cycle, altered psychomotor activity, or delusions. A formal diagnosis of delirium also requires evidence by history, physical exam, or laboratory data that the changes in consciousness and cognition are caused by the direct physiological consequences of a medical condition.

The main differential diagnosis of delirium is dementia, which, unlike delirium, is characterized by little or no clouding in consciousness, an insidious onset, and a chronic, progressive course. Furthermore, hypoactive delirium may be mistaken for depression; conversely, early stages of hyperactive delirium may be mistaken for anxiety or extrapyramidal symptoms.

Treatment is focused on reversing or treating underlying causes, such as dehydration, infection, hypoxia, electrolyte disturbances (particularly hypercalcemia), metabolic disturbances, urinary retention, constipation, and brain metastases. The medication list should be reviewed for offending medications and drug-drug interactions. Benzodiazepines, anticholinergics, and antihistamines are often implicated in precipitating delirium and should be discontinued or decreased as much as possible. If opioids are suspected as the culprit agent, and the treatment doses cannot be reduced because of pain needs, then rotation to a different opioid at a dose reduced by 25–50% is recommended

Simple environmental measures are often underutilized and include frequent orientation (including use of clocks, calendars, caregivers, and pictures of loved ones), ample day/night indicators (well-illuminated rooms during the day and reduced noise and light at nighttime), and treatment of hearing and vision problems.

Pharmacologic interventions include the use of antipsychotics (Table 67-5). Haloperidol is most commonly used.

Palliative care physicians observe that their patients have needs that transcend physical pain, social disruptions, and psychiatric disorders. Chochinov (2006) notes, “More ubiquitous aspects of suffering—including psychological, existential, or spiritual distress—are not necessarily well understood or researched, nor do they necessarily engender a well-considered response.”

The spiritual dimensions in palliative care include consideration of the patient’s religious practices (eg, prayer, sacraments, and rituals) but also attend to what may be called the existential concerns of the patient. Chochinov lists these as “an overwhelming sense of hopelessness, existential or spiritual angst; loss of sense of dignity; sensing oneself a burden to others; or a waning of one’s will to live and a growing desire for death.” To this list can be added concerns such as the loss of a sense of meaning, a paralyzing sense of guilt and regret, broken relationships with loved ones, difficulties with one’s concept of divinity or a sense of difficulty in a relationship with a personalized deity, feelings of anger, feelings of grief, and feelings of despair. Unsatisfied religious needs and unresolved existential concerns cause spiritual distress and can also result in psychiatric disorders such as depression or anxiety, as well as an increased sense of physical pain for the patient.

Recent studies have examined the importance of spirituality to physicians and to patients who are seriously or terminally ill, and whether and how physicians should address the spiritual concerns of these patients. [For a brief summary see Astrow and Sulmasy (2004).] Holmes and colleagues (2006) considered the results of some of these studies, conducted a study of their own, and came to the following conclusions: “It seems that patients tend to desire a sophisticated and somewhat controlled relationship with PCPs [primary care providers] around spirituality: They want their concerns cared about but not discussed or talked about, and they want to be prayed for but not with. These results indicate that, instead of discussing spiritual issues, PCPs may more appropriately “care” for the spiritual concerns of their patients by simply asking and listening … and leave the more active roles to others who have specific training in this area.”

A study conducted by MacLean and colleagues (2003) concludes that patients’ “desire for spiritual interaction increased with increasing severity of illness setting and decreased with referring to more-intense spiritual interactions.”

Puchalski and Sandoval (2003) offer a simple assessment tool, known by its acronym, FICA, for assessing a patient’s spirituality needs:

F: Faith, belief, meaning. Ask: “Do you consider yourself spiritual or religious?” “Do you have spiritual beliefs that help you cope with stress?” If the patient answers no, the physician might ask, “What gives your life meaning?”

I: Importance and influence. Ask: “What importance does your faith or belief have in your life?” “Have your beliefs influenced you in how you handle stress?” “Do you have specific beliefs that might influence your health care decisions?”

C: Community. Ask: “Are you a part of a spiritual or religious community? Is this of support to you, and how?” “Is there a group of people you really love or who are important to you?”

A: Address/action in care. Ask: “How should the healthcare provider address these issues in your healthcare?” Appropriate action might involve referral to chaplains, clergy, and other spiritual care providers.

Often the presence of spiritual suffering will emerge as patients tell their stories. To attend to this suffering requires that the healthcare professional be aware of and attentive to her/his own spirituality. The decision to share personal insights, experiences, and resources must be done with sensitivity and compassion, and with respect for the patient’s faith tradition and practice.

Consultation with and referral to a chaplain or other faith-based community leader is often appropriate. Professional chaplains are trained to understand and respect religious and cultural diversity and to assist patients and families in dealing with a wide variety of spiritual issues. Meador (2004) notes that “the clergy member of the team brings an interpretive, liturgical, and communal sense of spiritual care from her or his pastoral formation unique to that vocational formation.”

Victor Frankl, a psychiatrist whose writings were based on his experience in a Nazi concentration camp, observed: “Man is not destroyed by suffering; he is destroyed by suffering without meaning.” Palliative care therefore extends beyond the physical dimensions of suffering to attend to the spiritual suffering that may be present. A physician can attend to this dimension of healing by offering a listening ear, a word of kindness, and a referral to a professional spiritual caregiver.

Physicians treating patients with serious illness encounter several difficult conversations with the patients and their families. These conversations span the entire disease course, from the diagnosis through the disease progression, and finally, end-of-life care. Challenging aspects of these conversations include the breaking of serious news and discussions of prognosis, goals of care, and advance directives. These conversations often culminate in the emotionally challenging decision to transition from curative or life-prolonging therapies to hospice care. While easing difficult situations, effective communication also results in improved patient adjustment to illness, increased adherence to treatments, lessened pain and other physical symptoms, reduced anxiety, decreased conflict, and avoidance of ineffective, often invasive treatments.

Responding to Emotions

Difficult conversations are very complex and elicit a wide range of emotions and reactions in patients and their families. These responses may include shock, withdrawal, sadness, crying, denial, fear, anger, and acceptance. During difficult conversations, clinicians tend to focus on objective, medical data; however, it is crucial to notice and respond to emotions. Understandably, difficult conversations represent a form of danger to patients. When facing danger, we usually respond with a fight-or-flight reaction, while more cognitive, analytical responses get shifted into the background. This is the reason why patients commonly report that they “did not hear anything” after their clinician conveyed serious news. The key at this point in the conversation is to slow down and respond to emotions so the “system can cool down” and patients are able to cognitively process important medical data.

Empathy is the process of recognizing and responding to emotions in others. It starts with noticing and identifying the patient’s emotion; clinicians should importantly not try to fix or quiet a patient’s emotion. The next step includes acknowledging the emotion. This can be done nonverbally (such as through eye contact, changes in body position, or touch) or through explicit statements. The acronym NURSE summarizes ways to respond verbally to emotions (Table 67-6).

Another powerful way to respond to emotions is to normalize the experience. Examples include “Anyone receiving such news would feel anxious and sad,” or, “It is completely expected to feel devastated when one’s life changes so dramatically.”

Breaking Serious News

Conversations in which serious news is delivered by the clinician to the patient and family occur not only at the time of diagnosis but also when the illness progressed to the point at which it no longer responds to therapies. The SPIKES (set up, perception, invitation, knowledge, emotion, summarize/strategy) protocol is a six-step protocol that can be used in these stressful situations:

1. Set up—prepare for the conversation. Prior to the conversation, all medical data and information about possible treatment options need to be reviewed. The conversations should take place in a quiet place where everybody who is attending can sit down. Pagers and cell phones need to be silenced, and tissues should be in the patient’s reach.

2. Assess the patient’s perception. This step can provide an important window into the patient’s perspective, especially if a clinician is meeting the patient for the first time. It may guide how much more information the patient may need. A simple way to assess a patient’s understanding is to ask, “What have other doctors told you so far about your illness?” This step may not be necessary if the clinician knows the patient very well.

3. Ask for an invitation to talk about the news. This is a very valuable step, as patients in these situations can often feel out of control. Simply asking the question, “Are you ready to talk about this?” gives patients a little bit of control and signals an intent to work cooperatively. It is also important to assess how much the patient wants to know, taking into account cultural, religious, social, or personal issues.

4. Knowledge–disclose the news. Information needs to be communicated in a way that helps the patient to process and understand. It is helpful to preface serious news with a warning statement, such as “The test result came back and there is some serious news that we need to talk about” or “I am sorry to tell you that ….”. Clear language without medical jargon should be used. The information needs to be provided in small, understandable amounts; details can be filled in later.

5. Respond to the patient’s emotions. A wave of emotions will follow hearing the news. Ways on how to respond have been described above.

6. Summarize the conversation and discuss the strategy (treatment plan)

   It is important to ask for questions, summarize what has been discussed, and to outline next steps. Providing information also in writing can reduce confusion and anxiety.

Discussing Prognosis

Discussing prognosis is difficult; telling patients that they will die from their illness may feel, at its worst, like handing down a death sentence. It may often seem easier to defer these conversations and instead focus on symptoms and other care-related issues. For this reason, clinicians may have conflicting feelings about these conversations, because they understand that, although difficult and challenging, conversations that provide information about prognosis will allow the patient and family to prepare appropriately. Patients may also have contradictory wishes for their care. Most patients report that they want to be included in decision making and receive as much information as possible. However, they also want their care providers to support their wishes.

Having conversations about prognosis has been shown to impact advance care planning, particularly with regard to do-not-resuscitate orders and timely referrals to hospice. However, physicians are poor prognosticators and tend to be overly optimistic. The accuracy of prognostication improves with experience, but it worsens as the duration of the patient-physician relationship increases.

Prognostication is not commonly taught in medical school or during postgraduate training. In addition to observational data culled from patients with specific diseases, several models and scales are available to aid in forming prognoses. Some of these models are disease-specific, such as the Seattle heart failure model, a six-month- mortality calculator for patients on maintenance hemodialysis, and the Mortality Risk Index for patients with dementia. An example of a non-disease specific tool to assess prognosis is the Palliative Performance Scale (PPS, Table 67-7). Several studies showed correlations of PPS scores with survival in palliative care patients (including patients with different diagnoses and in different care settings).

A conversation about prognosis entails four steps: (1) negotiating the content, (2) providing information, (3) acknowledging the patient’s and family’s reaction to the news, and (4) checking for understanding.

1. Negotiating the content: Even though most patients are interested in hearing about prognosis, not all want the same level of detail. Therefore, it is helpful to start the conversation by asking, “How much do you want to know about your prognosis?” A clinician can further explain the range of possible information provided by saying, “Some people want to hear every detail, some want to focus on the big picture, and others may rather not discuss prognosis at all. What would be best for you?” If a patient is interested in talking about prognosis, it is recommended to ask if the patient would like to hear specific statistics or rather be informed about the best, worst, and most likely case scenarios.

2. Providing information: The information needs to be given straightforwardly and slowly, in digestible pieces and with sufficient pauses to allow the patient to absorb it and then react and ask relevant questions. Information about time should be given in ranges, such as hours to days, days to weeks, weeks to months, or months to years.

3. Acknowledging the patient’s and family’s reaction to the news: The patient and family will most likely have an emotional reaction of some considerable intensity. Ways to respond verbally to emotions have been outlined in Table 67-6. Acknowledging the emotion can lead to a deepening of the conversation.

4. Checking for comprehension: Given the complexity of these conversations, patients and families may hear only the good or bad aspects. Ways to check in and ascertain what information was understood include, “Tell me what you are taking away from this conversation,” or, “Tell me what you will tell your spouse/friend about this conversation.”

Family members may turn to care providers to help them understand what to expect when their loved one is dying. Patients who are very close to the end of life display a series of signs that predict the closeness of their deaths. One of the first signs is a decrease in engagement with one’s surroundings and in communication. As the body is preparing to die, the patient decreases his/her oral intake. This is often the most difficult change for family members, as they may become concerned that their loved one is suffering and starving. Often it is enough to explain to them that this is part of the normal dying process. Most of the time, patients do not report being hungry or thirsty, although they may complain of a dry mouth. In this case, effective mouth care with moistened sponges and a lip balm is sufficient.

As the dying process continues, the patient will become progressively somnolent, with fewer and shorter awake periods. Sometimes the patient may become confused or restless. The clarity of hearing and vision is also often seen to decrease. Family members sometimes ask if their loved one is still able to hear them. Hearing is the last one of the five senses to be lost, and family members should be invited to continue talking to their loved one.

As oral intake decreases and metabolic changes continue, urine output will decrease and the urine will be more concentrated. When very close to death, patients may develop urinary and bowel incontinence. Other physical changes include alterations in body temperature and blood pressure, increased perspiration, and skin changes, such as mottling or a pale, yellowish pallor. Breathing changes also occur, including increased, decreased, or irregular respirations, as well as periods of apnea. As the patient is becoming weaker and more somnolent, s/he is no longer able to clear the throat or cough. Secretions begin to pool in the throat directly above the vocal cords. As air is passing by these secretions, the sound produced may be loud and rattling, often referred to as “death rattle.” Family members who never have witnessed somebody dying may become concerned and wonder if the patient is “drowning.” Very often, comparing the death rattle to snoring can comfort them; like snoring, this sound may be disturbing to the person hearing it, but is not uncomfortable to the person producing it. Measures to decrease the death rattle include simply repositioning the head or the use of anticholinergic medications (see Table 67-8). Deep suction is not recommended, as it is uncomfortable and can lead to bleeding.

Medical care should be simplified as much as possible. Laboratory tests, radiologic procedures, and other interventions should be done only if they will result in improvement of the patient’s comfort. Nonessential medications should be discontinued. Artificially provided hydration and nutrition are seldom necessary or helpful for the dying patient. Administration of parenteral fluids may result in progressive edema, lung congestion, increased oral secretions, and frequent urination with attendant discomfort and distress. Experienced hospice professionals note no increase in discomfort or suffering with the naturally occurring dehydration that accompanies the dying process. Frequently used medications to address symptoms of the dying patient are listed in Table 67-8.

Additionally, of great importance are nursing interventions, such as daily bathing, good mouth care, and application of artificial tears and lubricating ointment to the eyes, as well as comfortable positioning in the bed with pillows placed under the calves or other areas of support. Family members may be instructed in these nursing interventions and participate in the care of their loved one. This often is very meaningful and comforting to both the patient and family members.

Grief, which is a normal reaction to loss, is the process of adjusting to a difficult reality. Both patients and families experience grief prior to and in anticipation of death; families and friends, of course, grieve after the death of a loved one. Attention to grief and bereavement is an often neglected part of excellent end-of-life care. Clinicians can play an important role in facilitating healthy grief and assessing for complicated grief.

Grieving individuals experience a variety of difficult emotional reactions, which usually occur in waves. Grief is triggered or exacerbated predictably by new losses, such as decline in functional status, as well as significant life events, including anniversaries and holidays. But grief can also worsen unpredictably by seemingly trivial events. Elisabeth Kübler-Ross first described five stages of grief—denial, anger, bargaining, depression, and acceptance—which may occur in any order and usually peak within 6 months following a loss. Somatic symptoms of grief include insomnia, ­dizziness, anorexia, nausea, restlessness, generalized weakness, and shortness of breath.

The death of a loved one is likely one of the most stressful human experiences, yet most people cope without needing professional interventions. A few individuals have more pronounced symptoms and may experience a persistent, debilitating phenomenon referred to as complicated grief (CG). The symptoms of CG include longing for the loved one; trouble accepting the death; feeling uneasy about moving on with one’s life; inability to trust others after the death; excessive bitterness or anger about the death; persistent feeling of being shocked, stunned, or numb; intense feeling of loneliness; feeling that life is empty or meaningless without the deceased; and frequent preoccupying thoughts about the deceased. These symptoms persist after 6 months, and they cause impairments in daily function. Predisposition to CG seems related to insecure attachment styles, an unstable sense of self, weak parental bonding in childhood, female gender, low perceived social support, death in an intensive care unit, and insufficient preparation for the loss (such as after an unexpected, traumatic death). Bereavement after the death of a child should always be considered CG. No interventions have been shown to prevent CG, but there is evidence from a recent meta-analysis that it is responsive to cognitive behavioral or group therapy.

Clinicians can play an important role in facilitating a healthy grieving process. They can assess patient and family risk factors for difficulties in grieving, and they can provide psychosocial resources as needed. Factors that have been associated with better bereavement outcomes include effective symptom management and open and honest communication about the course of the illness, prognosis, and advance care planning. Timely hospice enrollment has also been shown to positively affect bereavement outcomes.

Even after a patient’s death, clinicians can facilitate healthy grieving. Throughout the course of a serious illness, clinicians become an integral part in patients’ and families lives. Therefore, it is a very appreciated and meaningful act of kindness to not end this relationship suddenly when a patient dies, and rather to make a condolence call, write a condolence letter, or even attend the funeral or memorial service. Screening for insomnia, hypertension, and substance abuse, as well as complicated grief and other psychiatric illnesses, is indicated.

Books

Journals