Pain Management in Palliative Care Oncology Patients Jonathan R. Gavrin, MD, and Erin M. McMenamin, ANP-BC

Corresponding author Jonathan R. Gavrin, MD Symptom Management and Palliative Care (SYMPAC), University of Pennsylvania Health System, 1108 Penn Tower, Philadelphia, PA 19104, USA. E-mail: [email protected] Current Pain and Headache Reports 2008, 12:257–261 Current Medicine Group LLC ISSN 1531-3433 Copyright © 2008 by Current Medicine Group LLC

Palliative care is an interdisciplinary approach to relieving aversive symptoms in people with life-threatening illnesses; it aims to improve the lives of patients and their loved ones, the “patient–family unit.” Palliative care should occur in parallel with all other medical interventions. Indeed, good symptom management is important in helping patients cope with the unpleasantness associated with potentially curative or life-prolonging interventions; it is absolutely essential near the end of life. Unrelieved pain is the symptom that people fear the most. In most cases, adequate pain relief can be achieved with systemic medications alone. When systemic medications fail, due to inadequate analgesia or burdensome side effects, invasive techniques may complement, or replace, systemic therapy. Using a case-based format, we illustrate some complex issues that clinicians face and offer strategies to improve the lives of oncology patients with pain.

Introduction “The goal of palliative care is to prevent and relieve suffering, and to support the best possible quality of life for patients and their families, regardless of their stage of disease or the need for other therapies, in accordance with their values and preferences. Palliative care is both a philosophy of care and an organized, highly structured system for delivering care. Palliative care expands traditional disease-model medical treatments to include the goals of enhancing quality of life for patient and family, optimizing function, helping with decision-making, and providing opportunities for personal growth. As such, it can be delivered

concurrently with life-prolonging care or as the main focus of care.” —American Academy of Hospice and Palliative Medicine [1••].

This statement is a strong endorsement that palliative interventions should commence when a person is diagnosed with a potentially lethal disease, and it lays out a philosophy that weaves palliative care intimately with other medical interventions intended to cure disease or prolong life. This has become increasingly more important as advances in antineoplastic therapies have had a positive effect on survival in oncology patients. We treat people for longer periods of time, who often have diseases that have advanced far beyond what was seen in years past. Palliative care contributes to higher quality of life (QOL) and better allows patients to tolerate the frequently unpleasant symptoms associated with curative or life-prolonging therapies [2••]. Longer life and more aggressive therapies have created a significant subset of patients who have aversive symptoms, including pain that can be exceedingly difficult to control. Although the World Health Organization analgesic ladder is effective in 80% to 90% of cancer patients [3], systemic therapy alone frequently is insufficient to control painful symptoms. This article uses a case-based approach to describe some of the clinical and logistic problems we face on a regular basis in our attempts to control pain in cancer, with special emphasis on advanced disease. We make no pretense to present a “cookbook” or algorithmic approach to pain management in palliative care, but rather, illustrate the complexities of symptom management and the value of interdisciplinary, multimodal therapy for analgesia in cancer patients.

Case A 59-year-old man with a history of gastroesophageal reflux disease, social alcohol intake, and moderate cigarette smoking developed adenocarcinoma of the esophagus. Original treatment had been resection of the lesion with gastric pull-through, followed by outpatient chemotherapy and radiation therapy. Shortly after surgery, the patient developed a post-thoracotomy pain syndrome for which his oncologist referred him to

258

Cancer Pain

a pain management specialist associated with the palliative care program. A combination of radiofrequency neuroablative techniques, small amounts of systemic opioids, and a well-placed transdermal lidocaine patch provided adequate relief for the patient to participate in most social activities. Simultaneously, as is the practice of the oncology department where he was treated, he was offered consultation with a psychiatrist, also associated with the palliative care program. He saw the psychiatrist on several occasions, but neither he nor the physician felt any need for medicines or psychotherapeutic counseling. Indeed, the patient had quite a positive outlook on life, remaining cheerful and polite even through the most difficult of times. When hospitalized, or when seen in clinic, a close female friend always accompanied him, although the two were not romantically involved. The patient did well for about a year until he developed recurrent disease. Evaluation at that time revealed widespread metastatic disease to the abdominal viscera, including the liver, multiple bones in the axial and thoracic skeleton, and soft tissues in the neck. His pain slowly increased, with the worst symptoms in both subcostal regions, the periumbilical area, and bilaterally in the anterior pelvic rims. Over a period of about 4 weeks, the pain management specialist titrated the systemic analgesics upward to achieve comfort; the patient’s pain stabilized on a regimen of fentanyl transdermal patches, 300 μ g/hour, and oral immediate-release oxycodone, 150 mg at 3- to 4-hour intervals for breakthrough, or incident, pain. The patient again did well but only for about 2 weeks. Despite the transdermal fentanyl and, typically, 900 mg of additional short-acting oxycodone per day, the patient arrived at clinic one day, doubled over with uncontrolled pain. The palliative care team admitted him to the hospital to manage his visceral and bony pain; the treating oncologist began to explore further therapies that might slow the disease in an attempt to prolong life or reduce pain. Using standard conversion tables, the palliative care team came up with an equianalgesic dose of approximately 100 mg of hydromorphone intravenously per day. Because the patient’s pain was uncontrolled, doses were not adjusted downward to account for incomplete cross tolerance; the team started a patient-controlled analgesia (PCA) pump with the following settings: 2 mg/hour basal, 4-mg bolus, and 6-minute lockout. The patient did not feel additional relief despite successive doubling of the PCA doses until the pump settings were at 16 mg/hour basal rate, 32-mg bolus, and 6-minute lockout. Three clinician-delivered 96-mg doses of intravenous (IV) hydromorphone in a 10-minute period gave the patient transient relief (about 20 minutes). Careful questioning revealed no evidence of opioid-induced hyperalgesia. After obtaining a baseline electrocardiogram (ECG), the palliative care team administered methadone, 10 mg intravenously, with

little noticeable effect. Because the patient was still in a pain crisis and little headway was being made with systemic analgesics, one of the anesthesiologists on the team placed an epidural catheter at the T9–T10 level, running 0.1% bupivacaine (without epinephrine) via patient-controlled epidural analgesia (PCEA) with settings of 6 mL/hour basal rate, 4-mL bolus, and 10minute lockout (later increased to 8 mL/hour basal rate, 6-mL bolus, and 10-minute lockout). The patient fi nally achieved acceptable (although incomplete) relief while also continuing the hydromorphone PCA; his use was 450 to 550 mg/day. Suspicious of opioid tolerance, the team initiated standing oral methadone on the third hospital day, starting with 20 mg every 6 hours and eventually escalating to 120 mg every 6 hours over the next week. The patient’s hydromorphone use decreased to about 350 mg per day. Diagnostic imaging revealed further extension of the intra-abdominal lesions but nothing specifi cally amenable to surgical, chemo-, or radiotherapeutic interventions. The patient remained comfortable, and because of his personal resistance to the placement of an indwelling neuraxial catheter, the team decided to discontinue the epidural temporarily to see if systemic analgesics alone would be sufficient to relieve the pain. In less than 2 hours, the patient had uncontrolled pain again. The anesthesiologist restarted the PCEA, after an appropriate loading bolus of 12 mL of 1% plain lidocaine; the patient’s comfort returned to an acceptable level in about 10 minutes. At the urging of the psychiatrist whom he had seen at the time of diagnosis, and with the support of his ever-constant companion, the patient fi nally agreed to the placement of an indwelling epidural catheter in the hopes that he could go home. His methadone dosage was 120 mg orally every 6 hours; serial ECGs showed no significant increase in the QT interval, a potentially dangerous complication from methadone [4, 5•]. He continued to use IV hydromorphone by PCA. Over several days, his total consumption dropped into a range of 30- to 50-mg/24 hours. Although the patient could have gone home at that time with PCEA and PCA, he did not want to be encumbered by two separate external pumps (an understandable concern). Therefore, the palliative care team increased the oral methadone to 150 mg every 6 hours and added morphine to the medication mix in the epidural catheter. Occasional rescue boluses from the PCEA gave the patient acceptable relief and allowed him to leave the hospital. Once home, with every-other-day visits by nurses from the home infusion service with which the palliative care team frequently worked, the patient did well for about 3 weeks, returning to the palliative care clinic on two occasions for routine follow-up; no analgesic changes were necessary. The patient was cheerful, interactive, ambulatory, and able to perform his own daily care activities, as he had through all of his hospital stays.

Pain Management in Palliative Care Oncology Patients

Then, one night the patient fell. After transportation to the emergency department (ED) at our hospital, he was found to be febrile and mildly impaired cognitively with a mini-mental state examination score of 25/30. His pain had escalated. Evaluation in the ED showed small bilateral pleural effusions but no significant extension of intra-abdominal disease since the previous imaging studies 3 weeks earlier. The white blood cell count was not elevated, and there was no obvious source of infection. Over the following 4 days, the patient’s mental status deteriorated. All cultures for bacterial disease were negative. The patient became comatose. Although he could not verbalize, it appeared that the patient again suffered from uncontrolled pain, manifested by soft moaning from time to time at rest and loud vocalizations, as well as combative behavior, when turned. Despite initiation of haloperidol, as well as rapid escalations in the epidural and systemic analgesics, the palliative care team could not control what appeared to be ever-increasing delirium and discomfort. On many occasions the patient clearly had stated that he wanted a peaceful and dignified death. His advance directive designated his companion as the surrogate decision maker. Representing his wishes, that nothing extraordinary be done and that if his symptoms could not be controlled he would choose comfort and hygiene over prolongation of the dying process, his surrogate asked the team to institute whatever measures would be necessary to control the delirium and the apparent pain. The palliative care team started a midazolam drip, but this too failed to relieve the escalating delirium even at doses of 25 mg/hour. After further consultation with the surrogate decision maker, the team initiated palliative sedation with phenobarbital, 200 mg intravenously, administered over 20 minutes on three different occasions in a 2-hour period, accompanied by a 25-mg/hour continuous IV infusion. The patient remained calm, never regained consciousness, and died 36 hours later. The only other medical intervention was to start scheduled doses of glycopyrrolate to dry out the secretions that the patient no longer could handle once he began actively to die; this eliminated the “death rattle” and brought peace to the patient’s companion and to all the care providers.

Discussion This case is notable for the failure of large quantities of systemic analgesics to fully control tumor-related pain and raises many of the issues inherent to managing pain in the palliative care oncology patient. We will use the case to illustrate some of the many ways that coordinated palliative care contributes to the well-being of cancer patients. The literature abounds with references to improved QOL when aversive symptoms are managed appropriately [6•]. Often, patients with advanced disease have

Gavrin and McMenamin 259

a complex constellation of symptoms, requiring referral to a palliative care team that can use an interdisciplinary approach to provide relief. In the aforementioned case, for example, the psychiatrist associated with the team was able to assist the patient with the decision to place an indwelling epidural catheter. The symptom management team also can facilitate consults to surgical, medical, and radiation oncologists in search of other palliative interventions; none was available for the patient described earlier. The patient’s companion also was an important link in the total care of the patient. This is consistent with the philosophy of treating the “patient–family unit” in palliative care. The companion was the proxy for the patient if the capacity to make decisions was lost. It is essential to include potential surrogate decision makers in the care of the patient throughout the course of treatment in order to ensure that they understand the wishes of the patient, as well as the anticipated disease course. Systemic analgesics alone usually are effective in providing patients with adequate analgesia, even at the end of life. Opioids are the mainstay of pain relief for patients with advanced cancer. Analgesia is achieved when opioids bind with the receptors in the brain and spinal cord. Possible side effects common to all opioids include sedation, respiratory depression (rare except in overdose), nausea (often transient), pruritus, acute confusional states (delirium), and constipation, to which the body usually does not accommodate, thus requiring a scheduled bowel regimen. The opioids have similar analgesic and side effect profi les across populations, but these responses can be quite variable and idiosyncratic. A priori there is no way to know which opioid will be best for each patient, and thus therapy must be tailored to the individual patient and circumstance; a simple illustration of this is the fact that all patients do not require the same dosage of opioid in order to achieve analgesia. Generally, oral opioids are preferred due to the ease of administration. When anatomic limitations, such as gastrointestinal obstruction and inability to swallow, or high dosage requirements prohibit the use of transdermal delivery systems, then patients require subcutaneous (SC), IV, or neuraxial analgesia. Systemic (SC/IV) opioids are used at home when necessary and appropriate. Insurance payers must be amenable to payment for this route before initiation. Every palliative care team should have a social worker who can help with the logistics of preauthorization, home care, and payments. Support by an infusion therapy company also is important in order to ensure the safe and effective administration of the opioid. A common misconception regarding treatment of pain with systemic opioids is that there is no ceiling dose, a myth perpetuated in many standard texts. Empirically, although infrequent and often toward the end of life, pain specialists all have had patients who experience no additional analgesia from escalating doses of opioids.

260

Cancer Pain

Sometimes, these patients require rotation to another opioid, often methadone. Although not well studied in humans, investigators believe that methadone acts as an antagonist at the N-methyl-D-aspartate receptor and consequently may prevent or reverse development of tolerance and central sensitization [7–9]. This property often makes methadone an effective analgesic in patients who seem to have reached a ceiling effect with other opioids. Torsades de pointes is a side effect sometimes seen in patients using high doses of methadone; baseline and follow-up ECGs are essential because this is life-threatening if untreated. Neuraxial therapy is initiated when the patient is experiencing untoward side effects from systemic therapy, the dosage requirements are prohibitive, or the patient has a pain syndrome (almost anywhere below the clavicles) that would be amenable to treatment with a local anesthetic agent, with or without added opioid. Local anesthetic agents (generally bupivacaine) in low concentrations block sensory input, while usually sparing motor nerves, preserving the ability to ambulate. Often a combination of opioid and a local anesthetic agent is used in neuraxial catheters. The amount of opioid required to provide analgesia through this route is lower due to higher concentrations of medication at the dorsal horn. This route often avoids systemic side effects of opioids; opioids do not interfere with motor function. A recent meta-analysis compared epidural analgesia with systemic opioids for patients undergoing surgeries; there is some evidence that epidural analgesia is superior to systemic opioids for at least 3 days postoperatively, but the superiority of epidural analgesia after that is not clear. The authors pointed out that there are insufficient data to determine whether epidural analgesia provides superior quality of recovery, QOL, or satisfaction in the long term [10•]. Patients are the most reliable raters of their own pain. Often, typically toward the end of life, patients cannot provide reliable pain ratings. In those cases, one must rely on behaviors as indicators of pain, or the ratings of a caregiver and/or family member. Studies have found low levels of agreement between patients and caregivers [11]. Some investigators recommend the use of a behavioral scale [12••]. They report that there is no one scale against which all others are compared, but recommend several scales, depending on the patient population. Good general advice is to treat any impaired patient “as if” they were cognitively intact; that is, if a diagnostic or therapeutic intervention would have hurt a communicative patient, it will be just as painful to a noncommunicative one. Often patients fear uncontrolled pain at the end of life, and they request a comfortable, dignified death, even if interventions might hasten the dying process. A knowledgeable and experienced team typically can manage symptoms to provide patients with the desired comfort and dignity at death. However, there are times

when, despite involvement of the most skilled clinicians, aversive symptoms are intractable. In those unusual cases, especially when death is fairly imminent, patients or their surrogates may opt for deep sedation to control symptoms. If the surrogate requests palliative sedation, that request must be consistent with the patient’s values, namely “what the patient would have wanted” in such a situation. Therefore, it is wise to establish the patient’s values well before any cognitive impairment ensues. Although palliative sedation can be a somewhat controversial topic [13•,14 ,15 ], instituting it to control intractable symptoms is legally and ethically appropriate, even if death may be hastened in the attempt to treat the symptoms [16 ,17 ]. This is consistent with the Rule of Double Effect, which derives from Catholic moral theology and dates back, at least, to Thomas Aquinas [18]. Every effort should be made to treat the symptoms using standard care before the initiation of palliative sedation.

A Word About Corticosteroids We have not specifically addressed the use of corticosteroids, primarily dexamethasone, in palliative care. To do so would have been outside the scope of this article. Nevertheless, it is important to note that corticosteroid therapy often is the “fi nal common pathway” for treatment of many symptoms, including (most importantly) pain and refractory nausea. The mechanism of action remains unknown.

Conclusions Palliative care improves the quality of medicine for our sickest and most vulnerable patients and their loved ones. Optimal delivery of palliative care requires an interdisciplinary team with experts from various medical specialties, nurses, chaplains, social workers, and rehabilitation experts. Key to palliative care is the recognition that it is offered simultaneously with, and parallel to, all other appropriate medical treatment. The ultimate goal of medicine is to offer cure or meaningful prolongation of life. Palliative care experts are important contributors because they can make even the most unpleasant treatments bearable. When all options to cure disease or prolong life have been exhausted, “aggressive” palliative care becomes appropriate. In this article, we have used the case of a patient with difficult-to-control pain to illustrate many of the issues that arise during the course of severe, life-threatening illness, from a common postsurgical neuropathic pain to a complex pain syndrome due to extensive and accelerating cancer.

Disclosures No potential conflicts of interest relevant to this article were reported.

Pain Management in Palliative Care Oncology Patients

References and Recommended Reading Papers of particular interest, published recently, have been highlighted as: • Of importance •• Of major importance 1.••

American Academy of Hospice and Palliative Medicine: Statement on Clinical Practice Guidelines for Quality Palliative Care. Available at: http://www.aahpm.org/positions/quality. html. Accessed March 2008. This “must-see” website contains a wealth of clinically oriented material on palliative care. It provides access to educational materials, as well as details about meetings, funding opportunities, and board certification requirements. 2.•• Teno JM: Outcomes assessment in palliative care. In Principles and Practice of Palliative Care and Supportive Oncology. Edited by Berger AM, Shuster JL, Von Roenn JH. Philadelphia: Lippincott Williams and Wilkins; 2006:869– 876. Dr. Teno, a leading expert on outcomes research in palliative care, has been a driving force in the establishment and justification of palliative care programs around the country. This chapter is from a quality textbook that should be on the shelf of every clinician engaged in palliative care. 3. World Health Organization: WHO’s pain ladder. Available at: http://www.who.int/cancer/palliative/painladder/en/. Accessed March 2008. 4. Pearson EC, Woosley RL: QT prolongation and torsades de pointes among methadone users: reports to the FDA spontaneous reporting system. Pharmacoepidemiol Drug Saf 2005, 14:747–753. 5.• Sticherlinga C, Schaera BA, Ammannb P, et al.: Methadoneinduced torsade de pointes tachycardias. Swiss Med Wkly 2005, 135: 282–285. This important, oft-cited resource outlines one of the most perplexing and dangerous side effects of methadone. It provides a good entry point into other methadone-related literature. 6.• Jocham HR, Dassen T, Widdershoven G, Halfens R: Quality of life in palliative care cancer patients: a literature review. J Clin Nurs 2006, 15:1188–1195. This comprehensive literature review on QOL in palliative care is written from a nursing perspective and underscores the value of an interdisciplinary palliative team. 7. End of Life/Palliative Education Resource Center: Fast Fact and Concept #75: Methadone for Pain. Fast Facts 2002. Available at: http://www.mywhatever.com/cifwriter/library/ eperc/fastfact/ff75.html. Accessed March 2008.

8.

Gavrin and McMenamin 261

Bruera E, Sweeney C: Methadone use in cancer patients with pain: a review. J Palliat Med 2002, 5:127–138. 9. Callahan RJ, Au JD, Paul M, et al.: Functional inhibition by methadone of N-methyl-D-aspartate receptors expressed in xenopus oocytes: stereospecific and subunit effects. Anesth Analg 2004, 98: 653– 659. 10.• Liu SS, Wu CL: The effect of analgesic technique on postoperative patient-reported outcomes including analgesia: a systematic review. Anesth Analg 2007, 105:789– 808. It has been difficult to show improvement in outcomes using epidural, rather than systemic, analgesia. The article suffers from all the pitfalls inherent in meta-analyses, but the authors offer a window into the value of postoperative epidural analgesia and provide some grounding for the use of longer-term epidural analgesia in select patients. 11. Werner, P, Cohen-Mansfield J, Watson V, Pasis S: Pain in participants of adult day care centers: assessment by different raters. J Pain Symptom Manage 1998, 15:8–17. 12.•• van Herk R, van Dijk M, Baar FP: Observation scales for pain assessment in older adults with cognitive impairments or communication difficulties. Nurs Res 2007, 56:34– 43. Pain assessment in cognitively impaired patients is difficult at best. Investigators have used a variety of tools, none of which has proven to be completely satisfactory. This article is essential for clinicians who care for such patients. 13.• Demme RA, Singer, EA, Greenlaw J, Quill TE: Ethical issues in palliative care. Anesthesiol Clin 2006, 24:129–144. Dr. Demme has established himself as a thoughtful and probing clinical bioethicist. This pragmatically oriented article outlines some of the most difficult issues in palliative and end-of-life care. 14. Quill TE, Byock IR: Responding to intractable terminal suffering: the role of terminal sedation and voluntary refusal of food and fluids. ACP-ASIM End-of-Life Care Consensus. Panel American College of Physicians-American Society of Internal Medicine. Ann Intern Med 2000, 132:408– 414. 15. Quill TE, Lo B, Brock DW: Palliative options of last resort: a comparison of voluntarily stopping eating and drinking, terminal sedation, physician-assisted suicide, and voluntary active euthanasia. JAMA 1997, 278: 2099–2104. 16. Washington v Glucksberg, 117 S. Ct. 2258, 138 L. Ed. 2d 772 (1997). 17. Vacco v Quill, 521 US 793 (1997). 18. Aquinas T: Summa Theologica (II-II, Qu. 64, Art. 7). 13th century.