Purpose of This PDQ Summary
Overview

Highlights of Patient Management

Pain Assessment

Initial Assessment Patient Self-Report Physical Examination Assessment of the Outcomes of Pain Management

Pharmacologic Management

Basic Principles of Cancer Pain Management Acetaminophen and Nonsteroidal Anti-inflammatory Drugs Opioids         Opioid types         Principles of opioid administration         Route of administration         Drugs and routes to be avoided         Side effects of opioids Adjuvant Drugs

Physical and Psychosocial Interventions

Physical Modalities Cognitive-Behavioral Interventions

Antineoplastic Interventions

Radiation Therapy Radiofrequency Ablation Surgery

Invasive Interventions

Nerve Blocks Neurologic Interventions Management of Procedural Pain

Discharge Planning
Treating Elderly Patients
Get More Information From NCI
Changes to This Summary (07/10/2008)
Questions or Comments About This Summary
More Information

Purpose of This PDQ Summary

This PDQ cancer information summary provides comprehensive, peer-reviewed information for health professionals about the pathophysiology and treatment of pain. This information is reviewed regularly and updated as necessary by the PDQ Supportive and Palliative Care Editorial Board ¹.

Information about the following is included in this summary:

• Assessment.
• Pharmacological management.
• Intervention.

This summary is intended as a resource to inform and assist clinicians and other health professionals who care for cancer patients during and after cancer treatment. It does not provide formal guidelines or recommendations for making health care decisions. Information in this summary should not be used as a basis for reimbursement determinations.

This summary is also available in a patient version ², which is written in less technical language, and in Spanish ³.

Overview

The International Association for the Study of Pain defines pain ⁴ as an unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of such damage. Cancer pain can be managed effectively through relatively simple means in up to 90% of the eight million Americans who have cancer or a history of cancer. Unfortunately, pain associated with cancer is frequently undertreated.[1]

Although cancer pain or associated symptoms cannot always be entirely eliminated, appropriate use of available therapies can effectively relieve pain in most patients. Pain management improves the patient’s quality of life throughout all stages of the disease. Patients with advanced cancer experience multiple concurrent symptoms with pain; therefore, optimal pain management necessitates a systematic symptom assessment and appropriate management for optimal quality of life.[2] Despite the wide range of available pain management therapies, data are insufficient to guide their use in children, adolescents, older adults, and special populations.[3]

State and local laws often restrict the medical use of opioids to relieve cancer pain, and third-party payers may not reimburse for noninvasive pain control treatments. Thus, clinicians should work with regulators, state cancer pain initiatives, or other groups to eliminate these health care system barriers to effective pain management. (These and other barriers to effective pain management are listed below.) Changes in health care delivery may create additional disincentives for clinicians to practice effective pain management.

Barriers to Effective Pain Management

• Problems related to health care professionals:
  • Inadequate knowledge of pain management.
  • Poor assessment of pain.[4]
  • Concern about regulation of controlled substances.
  • Fear of patient addiction.
  • Concern about side effects of analgesics.
  • Concern about patients becoming tolerant to analgesics.



• Problems related to patients:
  • Reluctance to report pain.
  • Concern about distracting physicians from treatment of underlying disease.
  • Fear that pain means disease is worse.
  • Concern about not being a “good” patient.
  • Reluctance to take pain medications.
  • Fear of addiction or of being thought of as an addict (this fear may be more pronounced in minority patients).[5]
  • Worries about unmanageable side effects.
  • Concern about becoming tolerant to pain medications.
  • Poor adherence with the prescribed analgesic regimen.[6]



• Problems related to the health care system:
  • Low priority given to cancer pain treatment.
  • Inadequate reimbursement.
  • The most appropriate treatment may not be reimbursed or may be too costly for patients and families.
  • Restrictive regulation of controlled substances.
  • Problems of availability of treatment or access to it.
  • Opioids unavailable in the patient’s pharmacy.

Flexibility is the key to managing cancer pain. As patients vary in diagnosis, stage of disease, responses to pain and interventions, and personal preferences, so must pain management. The recommended clinical approach outlined below emphasizes a focus on patient involvement.

1. Ask about pain regularly. Assess pain and associated symptoms systematically using brief assessment tools. Assessment should include discussion about common symptoms experienced by cancer patients and how each symptom will be treated.[2,3] Asking a patient to identify his or her most troublesome symptom is also of clinical value because the most troublesome symptom is not always the most severe, as demonstrated in a survey of 146 patients in the palliative phase of treatment for lung, gastrointestinal, or breast cancer.[7]



2. Believe patient and family reports of pain and what relieves the pain. (Caveats include patients with significant psychological/existential distress and patients with cognitive impairment.)[8,9]



3. Choose pain-control options appropriate for the patient, family, and setting.



4. Deliver interventions in a timely, logical, coordinated fashion.



5. Empower patients and their families. Enable patients to control their course as much as possible.

Effective pain management is best achieved by a team approach involving patients, their families, and health care providers. The clinician should:

• Initiate prophylactic anticonstipation measures in all patients before or during opiate administration. (Refer to the Constipation ⁵ section in the Side Effects of Opioids ⁶ section of this summary for more information.)



• Discuss pain and its management with patients and their families.



• Encourage patients to be active participants in their care.



• Reassure patients who are reluctant to report pain that there are many safe and effective ways to relieve pain.



• Consider the cost of proposed drugs and technologies.



• Share documented pain assessment and management with other clinicians treating the patient.



• Know state/local regulations for controlled substances.

References

1. Weiss SC, Emanuel LL, Fairclough DL, et al.: Understanding the experience of pain in terminally ill patients. Lancet 357 (9265): 1311-5, 2001.  [PUBMED Abstract]
   
   
2. Meuser T, Pietruck C, Radbruch L, et al.: Symptoms during cancer pain treatment following WHO-guidelines: a longitudinal follow-up study of symptom prevalence, severity and etiology. Pain 93 (3): 247-57, 2001.  [PUBMED Abstract]
   
   
3. Patrick DL, Ferketich SL, Frame PS, et al.: National Institutes of Health State-of-the-Science Conference Statement: Symptom Management in Cancer: Pain, Depression, and Fatigue, July 15-17, 2002. J Natl Cancer Inst 95 (15): 1110-7, 2003.  [PUBMED Abstract]
   
   
4. Bruera E, Willey JS, Ewert-Flannagan PA, et al.: Pain intensity assessment by bedside nurses and palliative care consultants: a retrospective study. Support Care Cancer 13 (4): 228-31, 2005.  [PUBMED Abstract]
   
   
5. Anderson KO, Richman SP, Hurley J, et al.: Cancer pain management among underserved minority outpatients: perceived needs and barriers to optimal control. Cancer 94 (8): 2295-304, 2002.  [PUBMED Abstract]
   
   
6. Miaskowski C, Dodd MJ, West C, et al.: Lack of adherence with the analgesic regimen: a significant barrier to effective cancer pain management. J Clin Oncol 19 (23): 4275-9, 2001.  [PUBMED Abstract]
   
   
7. Hoekstra J, Vernooij-Dassen MJ, de Vos R, et al.: The added value of assessing the 'most troublesome' symptom among patients with cancer in the palliative phase. Patient Educ Couns 65 (2): 223-9, 2007.  [PUBMED Abstract]
   
   
8. Allen RS, Haley WE, Small BJ, et al.: Pain reports by older hospice cancer patients and family caregivers: the role of cognitive functioning. Gerontologist 42 (4): 507-14, 2002.  [PUBMED Abstract]
   
   
9. Bruera E, Sweeney C, Willey J, et al.: Perception of discomfort by relatives and nurses in unresponsive terminally ill patients with cancer: a prospective study. J Pain Symptom Manage 26 (3): 818-26, 2003.  [PUBMED Abstract]
   
   

Pain Assessment

Failure to assess pain is a critical factor leading to undertreatment. Assessment involves both the clinician and the patient. Assessment should occur:

• At regular intervals after initiation of treatment.
• At each new report of pain.
• At a suitable interval after pharmacologic or nonpharmacologic intervention, e.g., 15 to 30 minutes after parenteral drug therapy and 1 hour after oral administration.

Identifying the etiology of pain is important to its management. Clinicians treating patients with cancer should recognize the common cancer pain syndromes (see table below). Prompt diagnosis and treatment of these syndromes can reduce morbidity associated with unrelieved pain. Distinct cultural components may need to be incorporated into a multidimensional assessment of pain.[1-4] A comprehensive review of cancer pain with a focus on neuropathic pain provides an overview of pain pathophysiologies and an extensive review of available and investigational pharmacotherapies.[5]

The goal of the initial assessment of pain is to characterize the pathophysiology of the pain and to determine the intensity of the pain and its impact on the patient’s ability to function. For example, one study evaluated the association between psychological distress and pain in 120 patients with advanced cancer. Pain intensity and pain that interfered with walking ability, normal work, and relations with other people as measured by the Brief Pain Inventory (Greek version) were found to be significant predictors of anxiety as measured by the Hospital Anxiety and Depression Scale on multivariate analysis. Using the same tools, the authors also found pain that interfered with enjoyment of life was a predictor of depression.[6] Factors that may influence analgesic response and result in persistent pain include changing nociception due to disease progression, intractable side effects, tolerance, neuropathic pain, and opioid metabolites.[7] The following are essential to the initial assessment:

• Detailed history.



• Physical examination.



• Psychosocial assessment.[8]



• Diagnostic evaluation.

The experience of cancer pain is complex and includes physical, psychosocial, and spiritual dimensions. There is no universally accepted pain classification measure that would assist with predicting the complexity of pain management, particularly for cancer pain patients, who may be more difficult to treat. Clinicians and researchers lack a common language to discuss and compare outcomes of cancer pain assessment and management. Oncologists use the tumor, nodes, metastases (TNM) system to provide a universal language to describe a variety of cancers. The need for a similar classification system for cancer pain resulted in the development of the Edmonton Staging System.[9,10] This system has been further refined in two reports that have gathered construct validity evidence using an international panel of content experts [11] and a multicenter study to determine interrater reliability and predictive value.[12] The development of an internationally recognized classification system for cancer pain could play a significant role in improving the assessment of cancer pain, allow a more meaningful assessment of clinical prognosis and treatment, and better enable researchers to compare results with regard to cancer pain management.

The mainstay of pain assessment is the patient self-report; however, family caregivers are often used as proxies for patient reports, especially in situations in which communication barriers exist, such as cognitive impairment or language difficulties. Family members who act as proxies typically, as a group, report higher levels of pain than patient self-reports, but there is individual variation.[13] Differences in clinician assessment of pain intensity are also significant. A retrospective review of 41 patient charts using pain ratings of palliative care consultants as the gold standard found high agreement with assessments performed by bedside nurses (registered nurses [RNs] and clinical nurse assistants [CNAs]) when pain was not present or was mild but poor agreement for moderate or severe pain (sensitivity: RNs, 45%; CNAs, 30%).[14]

Pain assessment tools may be unidimensional or multidimensional. Multiple assessment tools exist. Among the more commonly used bedside tools are numeric rating scales, verbal rating scales, visual analog scales, and picture scales.[15,16] To enhance pain management across all settings, clinicians should teach families to use pain assessment tools in their homes. The clinician should help the patient to describe:

Pain

• Listen to the patient’s descriptive words about the quality of the pain; these provide valuable clues to its etiology. Elicit the temporal features including onset, duration, and diurnal variation. Ask about breakthrough or episodic pain (a transitory increase in pain that occurs in addition to persistent pain). Some patients may have episodic pain without persistent pain.[17]

Location

• Ask the patient to indicate the exact location of the pain on his or her body, or on a body diagram, and whether the pain radiates.

Intensity or Severity

• Encourage the patient to keep a log of pain-intensity scores to report during follow-up visits or by telephone. Examples of simple self-report pain-intensity scales include the simple, descriptive, numeric, and visual analog scales.

Aggravating and Relieving Factors

• Ask the patient to identify factors that cause the most pain and also what relieves the pain.

Cognitive Response to Pain

• Cognitive appraisals of pain can be based on a range of psychological variables such as perceived control, meaning attributed to pain experience, fear of death, and hopelessness.[18] All these variables appear to contribute to the experience of cancer pain and suffering. A study of women with metastatic breast cancer found that although the site of metastasis did not predict the intensity of pain report, greater depression and the belief that pain represented the spread of disease significantly predicted the degree of pain experienced.[19] It was also reported that patients who thought that their pain represented disease progression reported more pain-related interference with function.[20]

Cognitive Impairment

• Note behavior that suggests pain in patients who are cognitively impaired or who have communication problems relating to education, language, ethnicity, or culture. Cognitive impairment itself and the degree of cognitive impairment may impact patient self-report of pain. Preliminary data suggest that mild degrees of cognitive impairment are associated with increased intensity of pain-report in older patients with cancer who are receiving hospice care.[13] In contrast, cognitively impaired nursing home residents are less likely to report pain. Use appropriate (e.g., simpler or translated) pain assessment tools.

Goals for Pain Control

• Document the patient’s preferred pain assessment tool and the goals for pain control (such as scores on a pain scale).



• Encourage use of the pain diary: The daily pain diary is a well-established tool in symptom management research and in clinical practice. Benefits of using a pain diary include heightened awareness of pain, guidance for pain management behaviors, enhanced sense of control, and a tool for communication.[21] It is difficult to get good pain-diary compliance with adolescents who are experiencing intense chronic pain.

A thorough physical examination is required to determine the pathophysiology of pain. Specific features of the neurologic examination such as altered sensation (hypoesthesia, hyperesthesia, hyperpathia, allodynia) in a painful area are suggestive of neuropathic pain. Physical findings of tumor growth and metastasis are also important to identify.

Changes in pain pattern or the development of new pain should trigger diagnostic evaluation and modification of the treatment plan. Persistent pain indicates the need to consider other etiologies (e.g., related to disease progression or treatment) and alternative (perhaps more invasive) treatments.

Pain-related outcomes: Clinicians should document and be aware of outcomes of pain therapy. It is helpful to think of pain-related outcomes as primarily measured in two ways: decreased pain intensity and improvement in psychosocial functioning. Using rating scales of pain intensity at its worst and on average and using pain interference scales can help clinicians monitor outcomes. Measurement of the percentage of pain relief is also useful, though measuring patient satisfaction is less useful because of the low expectations patients sometimes hold for pain control.[22,23]

Drug-taking outcomes: Clinicians prescribing chronic opioids should also monitor and document patients’ drug-taking behaviors. Outcomes related to addiction in cancer patients are rare but nonetheless should be periodically assessed; these assessments can be reassuring to patients. Tolerance and dependence are not addiction related. Documentation of patients’ compliance with regard to changes in dosing and duration of prescriptions is essential in all pain practice.

The clinical assessment of drug-taking behaviors in medically ill patients with pain is complex. Aberrant drug-taking behavior from cancer pain management is related to premorbid history of drug addiction and the likelihood of other pain treatment. A pilot questionnaire was used to characterize drug-related behaviors and attitudes in cancer and AIDS patients. Despite limitations, this study highlights wide potential variation among different palliative care populations in patterns of past and present aberrant drug-taking behaviors and the need for a clinically useful screening approach. The implications for psychosocial and pharmacological management of symptoms such as pain, as well as any aberrant behavior, remains unclear.[24-26]

Previous drug abuse is likely to lead to specific needs for appropriate dosing during cancer pain therapy. A prospective open-label study compared morphine dosage and effectiveness in AIDS patients with and without previous substance abuse. Results demonstrated that both groups benefited, but patients with a history of drug use require and will tolerate substantially higher morphine doses to achieve stable pain control.[27] This study should increase confidence in providing appropriate pain management to patients who have a history of drug use.[28]

References

1. Chung JW, Wong TK, Yang JC: The lens model: assessment of cancer pain in a Chinese context. Cancer Nurs 23 (6): 454-61, 2000.  [PUBMED Abstract]
   
   
2. Cleeland CS, Nakamura Y, Mendoza TR, et al.: Dimensions of the impact of cancer pain in a four country sample: new information from multidimensional scaling. Pain 67 (2-3): 267-73, 1996.  [PUBMED Abstract]
   
   
3. Greenwald HP: Interethnic differences in pain perception. Pain 44 (2): 157-63, 1991.  [PUBMED Abstract]
   
   
4. Bates MS, Edwards WT, Anderson KO: Ethnocultural influences on variation in chronic pain perception. Pain 52 (1): 101-12, 1993.  [PUBMED Abstract]
   
   
5. Fine PG, Miaskowski C, Paice JA: Meeting the challenges in cancer pain management. J Support Oncol 2 (6 Suppl 4): 5-22; quiz 23-4, 2004 Nov-Dec.  [PUBMED Abstract]
   
   
6. Mystakidou K, Tsilika E, Parpa E, et al.: Psychological distress of patients with advanced cancer: influence and contribution of pain severity and pain interference. Cancer Nurs 29 (5): 400-5, 2006 Sep-Oct.  [PUBMED Abstract]
   
   
7. Mercadante S, Portenoy RK: Opioid poorly-responsive cancer pain. Part 1: clinical considerations. J Pain Symptom Manage 21 (2): 144-50, 2001.  [PUBMED Abstract]
   
   
8. Otis-Green S, Sherman R, Perez M, et al.: An integrated psychosocial-spiritual model for cancer pain management. Cancer Pract 10 (Suppl 1): S58-65, 2002 May-Jun.  [PUBMED Abstract]
   
   
9. Bruera E, MacMillan K, Hanson J, et al.: The Edmonton staging system for cancer pain: preliminary report. Pain 37 (2): 203-9, 1989.  [PUBMED Abstract]
   
   
10. Bruera E, Schoeller T, Wenk R, et al.: A prospective multicenter assessment of the Edmonton staging system for cancer pain. J Pain Symptom Manage 10 (5): 348-55, 1995.  [PUBMED Abstract]
    
    
11. Nekolaichuk CL, Fainsinger RL, Lawlor PG: A validation study of a pain classification system for advanced cancer patients using content experts: the Edmonton Classification System for Cancer Pain. Palliat Med 19 (6): 466-76, 2005.  [PUBMED Abstract]
    
    
12. Fainsinger RL, Nekolaichuk CL, Lawlor PG, et al.: A multicenter study of the revised Edmonton Staging System for classifying cancer pain in advanced cancer patients. J Pain Symptom Manage 29 (3): 224-37, 2005.  [PUBMED Abstract]
    
    
13. Allen RS, Haley WE, Small BJ, et al.: Pain reports by older hospice cancer patients and family caregivers: the role of cognitive functioning. Gerontologist 42 (4): 507-14, 2002.  [PUBMED Abstract]
    
    
14. Bruera E, Willey JS, Ewert-Flannagan PA, et al.: Pain intensity assessment by bedside nurses and palliative care consultants: a retrospective study. Support Care Cancer 13 (4): 228-31, 2005.  [PUBMED Abstract]
    
    
15. Jensen MP, Karoly P: Measurement of cancer pain via patient self-report. In: Chapman CR, Foley KM, eds.: Current and Emerging Issues in Cancer Pain: Research and Practice. New York, NY: Raven Press, 1993, pp 193-218. 
    
    
16. Hølen JC, Hjermstad MJ, Loge JH, et al.: Pain assessment tools: is the content appropriate for use in palliative care? J Pain Symptom Manage 32 (6): 567-80, 2006.  [PUBMED Abstract]
    
    
17. Mercadante S, Radbruch L, Caraceni A, et al.: Episodic (breakthrough) pain: consensus conference of an expert working group of the European Association for Palliative Care. Cancer 94 (3): 832-9, 2002.  [PUBMED Abstract]
    
    
18. Mystakidou K, Tsilika E, Parpa E, et al.: Exploring the relationships between depression, hopelessness, cognitive status, pain, and spirituality in patients with advanced cancer. Arch Psychiatr Nurs 21 (3): 150-61, 2007.  [PUBMED Abstract]
    
    
19. Spiegel D, Bloom JR: Pain in metastatic breast cancer. Cancer 52 (2): 341-5, 1983.  [PUBMED Abstract]
    
    
20. Daut RL, Cleeland CS: The prevalence and severity of pain in cancer. Cancer 50 (9): 1913-8, 1982.  [PUBMED Abstract]
    
    
21. Schumacher KL, Koresawa S, West C, et al.: The usefulness of a daily pain management diary for outpatients with cancer-related pain. Oncol Nurs Forum 29 (9): 1304-13, 2002.  [PUBMED Abstract]
    
    
22. Rhodes DJ, Koshy RC, Waterfield WC, et al.: Feasibility of quantitative pain assessment in outpatient oncology practice. J Clin Oncol 19 (2): 501-8, 2001.  [PUBMED Abstract]
    
    
23. Hwang SS, Chang VT, Kasimis B: Dynamic cancer pain management outcomes: the relationship between pain severity, pain relief, functional interference, satisfaction and global quality of life over time. J Pain Symptom Manage 23 (3): 190-200, 2002.  [PUBMED Abstract]
    
    
24. Passik SD, Kirsh KL, McDonald MV, et al.: A pilot survey of aberrant drug-taking attitudes and behaviors in samples of cancer and AIDS patients. J Pain Symptom Manage 19 (4): 274-86, 2000.  [PUBMED Abstract]
    
    
25. Kirsh KL, Whitcomb LA, Donaghy K, et al.: Abuse and addiction issues in medically ill patients with pain: attempts at clarification of terms and empirical study. Clin J Pain 18 (4 Suppl): S52-60, 2002 Jul-Aug.  [PUBMED Abstract]
    
    
26. Passik SD, Kirsh KL, Whitcomb L, et al.: A new tool to assess and document pain outcomes in chronic pain patients receiving opioid therapy. Clin Ther 26 (4): 552-61, 2004.  [PUBMED Abstract]
    
    
27. Kaplan R, Slywka J, Slagle S, et al.: A titrated morphine analgesic regimen comparing substance users and non-users with AIDS-related pain. J Pain Symptom Manage 19 (4): 265-73, 2000.  [PUBMED Abstract]
    
    
28. Whitcomb LA, Kirsh KL, Passik SD: Substance abuse issues in cancer pain. Curr Pain Headache Rep 6 (3): 183-90, 2002.  [PUBMED Abstract]
    
    

Pharmacologic Management

Basic Principles of Cancer Pain Management

The World Health Organization (WHO) has described a three-step analgesic ladder as a framework for pain management.[1] It involves a stepped approach based on the severity of the pain. If the pain is mild, one may begin by prescribing a Step 1 analgesic such as acetaminophen or a nonsteroidal anti-inflammatory drug (NSAID). Potential adverse effects should be noted, particularly the renal and gastrointestinal adverse effects of the NSAIDs. If pain persists or worsens despite appropriate dose increases, a change to a Step 2 or Step 3 analgesic is indicated. Most patients with cancer pain will require a Step 2 or Step 3 analgesic. Step 1 can be skipped in those patients presenting at the onset with moderate-to-severe pain in favor of Step 2 or Step 3. At each step, an adjuvant drug or modality such as radiation therapy may be considered in selected patients. WHO recommendations are based on worldwide availability of drugs and not strictly on pharmacology.

Analgesics should be given “by mouth, by the clock, by the ladder, and for the individual.”[1] This requires regular scheduling of the analgesic, not just as needed. In addition, rescue-doses for breakthrough pain need to be added. The oral route is preferred as long as a patient is able to swallow. Each analgesic regimen should be adjusted for each patient’s individual circumstances and physical condition.

NSAIDs are effective for relief of mild pain and may have an opioid dose–sparing effect that helps reduce side effects when given with opioids for moderate-to-severe pain. Acetaminophen is included with aspirin and other NSAIDs because it has similar analgesic potency, though it lacks peripheral anti-inflammatory activity.[2] Side effects can occur at any time, and patients who take acetaminophen or NSAIDs, especially elderly patients, should be followed carefully.[3-5] There is growing debate about whether NSAIDs are useful and have significant opioid-sparing effects. One meta-analysis [6] suggests that the usefulness of NSAIDs is limited and that they do not significantly spare opioid doses. Another study suggests that NSAIDs are useful and reduce the need for opioid dose increases; however, only patients with pain progression after 1 week of opioid stabilization were selected for the study.[7]

The coxibs are a subclass of NSAIDs designed to selectively inhibit cyclooxygenase-2 (COX-2).[8] Development of these drugs was based on the hypothesis that COX-2 was the source of prostaglandins E₂ and I₂, which mediate inflammation, and COX-1 was the source of the same prostaglandins in gastric epithelium, with the potential advantage over traditional NSAIDs of less gastrointestinal ulceration and bleeding and the absence of platelet inhibition. Direct comparisons between COX-2 inhibitors are few. A systematic meta-analysis of COX-2 inhibitors compared with traditional NSAIDs or different COX-2 inhibitors for postoperative pain suggests that rofecoxib, 50 mg, and parecoxib, 40 mg, are equipotent to traditional NSAIDs for postoperative pain after minor and major surgical procedures and have a longer duration of action after dental surgery. Rofecoxib was found to provide superior analgesic effect compared with celecoxib, 200 mg. There were insufficient data to comment on toxicity.[9]

There are three coxibs that were approved by the U.S. Food and Drug Administration (FDA): celecoxib, rofecoxib, and valdecoxib. On September 30, 2004, rofecoxib was withdrawn from the market after a study demonstrated that subjects in a colon cancer prevention trial taking the drug at higher-than-typical doses on a long-term basis had a significant increase in the incidence of serious thromboembolic complications. The question that remains unanswered is whether the increased risk applies to all COX-2 inhibitors, with the caution that the burden of proof rests with those who might claim that this is a problem for rofecoxib alone and does not extend to other coxibs.[8,10] On April 7, 2005, valdecoxib was withdrawn from the market. FDA is also asking manufacturers of all marketed prescription NSAIDs, including celecoxib (Celebrex) to revise the labeling (package insert) for their products to include a boxed warning, highlighting the potential for increased risk of cardiovascular events and/or the serious, potentially life-threatening gastrointestinal bleeding associated with use of these drugs.

Dosage

• Use patient response to determine the effective dosing interval for aspirin, acetaminophen, and other NSAIDs listed in the Dosing Recommendations for Acetaminophen and NSAIDs ⁸ table. When pain relief is not attained with the maximum dosage of one NSAID, try other drugs within this category before abandoning NSAID therapy.

Route of administration

• Use readily available oral tablets, capsules, or liquid. During intervals of nausea and vomiting, use suppositories. Ketorolac tromethamine is the only NSAID available for parenteral use.

Contraindications

• Patients taking NSAIDs are at risk for platelet dysfunction that may impair blood clotting. The table below lists NSAIDs with minimal antiplatelet activity.

Other side effects

• Follow patients carefully for adverse effects, which range from mild gastrointestinal discomfort to more serious problems including the following:
  • Gastric ulceration.
  • Hepatic dysfunction.
  • Myocardial infarction.
  • Renal failure.

  Because both NSAIDs and other drugs (e.g., warfarin, methotrexate, digoxin, cyclosporine, oral antidiabetic agents, and sulfonamide-containing drugs) are highly protein-bound, there is potential for altered efficacy or toxicity when they are given simultaneously.

Opioids, the major class of analgesics used in management of moderate-to-severe pain, are effective, are easily titrated, and have a favorable benefit-to-risk ratio.

The predictable consequences of long-term opioid administration—tolerance and physical dependence—are often confused with psychological dependence (addiction) that manifests as drug abuse. This misunderstanding can lead to ineffective prescribing, administering, or dispensing of opioids for cancer pain. The result is undertreatment of pain.[11]

Clinicians may be reluctant to give high doses of opioids to patients with advanced disease because of a fear of respiratory depression. Many patients with cancer pain become opioid tolerant during long-term opioid therapy. Therefore, the clinician’s fear of shortening life by increasing opioid doses is usually unfounded.

Opioids are classified as full morphine-like agonists, partial agonists, or mixed agonist-antagonists, depending on the specific receptors to which they bind and their activity at these receptors. The benefits of using opioids and the risks associated with their use vary among individuals.

Morphine is the most commonly used opioid in cancer pain management, largely for reasons of availability and familiarity;[12] however, it is useful to be familiar with more than one type of opioid. Wide interindividual variability in response to both the analgesic and adverse effects of opioids is recognized.[13] Some patients may not experience adequate pain control despite appropriate dose adjustments, while others may develop intolerable adverse effects to one particular opioid (see below). Alternative opioids include hydromorphone, oxycodone, methadone, and fentanyl. Knowledge of several medications and formulations give the caregiver much more flexibility in tailoring a regime to a particular patient’s needs.

Short-acting opioids are generally recommended when opioid therapy is being initiated for the first time or when patients are medically unstable or the pain intensity is highly variable. Once stable, patients can be switched to a controlled-release or slow-release formulation. This is more convenient and promotes compliance. (Refer to the Approximate Dose Equivalents for Opioid Analgesics ¹⁰ table in the Principles of Opioid Administration ¹¹ section of this summary.)

Full agonists

• Morphine, hydromorphone, codeine, oxycodone, hydrocodone, methadone, levorphanol, and fentanyl are classified as full agonists because their effectiveness with increasing doses is not limited by a ceiling. Full agonists will not reverse or antagonize the effects of other full agonists given simultaneously.

Morphine

• The most commonly used opioid, morphine, is readily available in several forms, including sustained-release (8–24 hours duration of effectiveness) formulations for oral administration.

Other agonists

• For the patient who experiences dose-limiting side effects with one oral opioid (e.g., hallucinations, nightmares, dysphoria, nausea, or mental clouding), other oral opioids should be tried before abandoning one route in favor of another.

Methadone

• Methadone has had a revival in interest for the management of cancer pain. Published reports have been in the form of case reports,[14-20] outcome surveys,[21-25] and reviews.[26-28]Success has been reported with oral, intravenous (IV), and suppository methadone use. Subcutaneous methadone has been reported to cause tissue irritation at the injection site but has been used effectively in some patients without clinically significant local toxicity.[29]

  Methadone is a synthetic opioid agonist that has been reported to have a number of unique characteristics. These include excellent oral and rectal absorption, no known active metabolites, prolonged duration of action resulting in longer administration intervals, and lower cost than other opioids. Methadone is available as a pill, an elixir, and for parenteral use. Methadone has an average oral bioavailability of approximately 80% (range, 41%–99%).[30]

  Morphine is the international gold standard for first-line treatment of cancer pain. Methadone, however, can be considerably less expensive than existing rapid-release or sustained-release morphine or other opioid options. A randomized trial of 103 patients compared the effectiveness and side effects of morphine and methadone as first-line treatments for cancer pain. The outcome of successful pain management was similar for both groups; however, there were significantly more opioid-related dropouts in the methadone group. This study did not demonstrate superior analgesic effectiveness or overall tolerability of methadone over morphine as a first-line treatment for cancer pain. Despite this finding, the authors of this report suggested that study limitations did not allow definitive conclusions that methadone could not be a useful first-line opioid. Further research exploring other doses and schedules of methadone should still be explored.[31]

  Because of its long and unpredictable half-life and relatively unknown equianalgesic dose as compared with other opioids, methadone has been generally used by pain specialists with experience in its use. The utility of methadone in cancer pain and difficult cancer pain syndromes such as neuropathic pain has become more widely appreciated and has gained increasing acceptance for use in hospital and hospice settings and by clinicians who are not pain specialists.[32] The methadone preparation widely used in the United States is a racemic mix of the d-isomer and l-isomer of methadone. The d-isomer has antagonist activity at the N-methyl-D-aspartate (NMDA) receptor and may be beneficial in controlling neuropathic pain.

  Another controversy related to methadone concerns possible prolongation of QTc interval, leading to torsades de pointes and ventricular arrhythmia. A number of studies have raised the concern that methadone may be associated with prolonged QT interval and may lead to torsades de pointes. Several retrospective case reports suggest that parenteral methadone or oral methadone in high doses could be associated with this adverse effect.[33-36] Chlorobutanol has been implicated; although this substance is present in parenteral solutions, it is not found in oral formulations.[37] Another series of 132 patients taking methadone revealed statistically significant mean increases in QTc of 10.2 to 13.2 milliseconds, yet no episodes of torsades de pointes were reported.[36] This result raises the issue of the clinical significance of this effect. In another retrospective review of 520 patients treated with methadone for cancer pain, no change in QTc was seen in the 56 patients who had electrocardiograms 3 months before and after starting methadone.[38,39] Avoidance of concomitant medications that prolong QT interval [37] or that share common metabolism pathways with methadone [38] is recommended. In high-risk situations, clinicians could consider electrocardiogram monitoring and other clinical precautions such as correcting electrolyte abnormalities.

  When converting from another opioid to methadone, the calculated equianalgesic dose ratio of methadone varies depending on the oral morphine-equivalent daily dose (MEDD) of the previous opioid.[40,22] One guideline for choosing an appropriate initial dose of methadone based on the oral MEDD of the previous opioid is shown in the table below. For example, a patient who has been using sustained-release morphine at 80 mg every 8 hours (240 mg/day) would be appropriately switched to methadone at a dose of 10 mg every 8 hours (30 mg/day, an 8:1 conversion ratio). In contrast, a patient who is taking sustained-release morphine at a total daily dose of 60 mg/day might be switched to an oral methadone dose of 5 mg every 8 hours (15 mg/day, a 4:1 conversion ratio).

  Method 1: Initial Methadone Dose Based on Oral MEDD*
  Enlarge ¹²

  Oral MEDD (mg/d)	Initial Dose Ratio (oral morphine:oral methadone)
  <30	2:1
  30–99	4:1
  100–299	8:1
  300–499	12:1
  500–999	15:1
  >1,000	20:1 or greater**

  
  

  *Reprinted with permission from Fisch and Cleeland [41]

  **Great caution must be used when converting to methadone when very high opioid doses have been used. Often, only a portion of the total opioid dose is converted initially, with further conversions taking place over several days to weeks.

  
  

  To be conservative, one might estimate that methadone is roughly twice as potent when administered via IV versus oral administration. Thus, a patient with well-controlled pain on a stable oral methadone dose of 10 mg every 8 hours might be given IV methadone at an initial dose of 5 mg every 8 hours if IV use is necessary. Subcutaneous use of methadone may cause skin irritation in some patients but has been used successfully.

  In addition to the method described in the table above, several methods of switching to methadone have been proposed.[42,22,43-46] Some rely on patient-controlled analgesia with fixed doses and flexible intervals, some require fixed intervals and fixed doses, while others stagger the conversion over a few days. Whatever method is chosen, this kind of switch can be safe and effective as long as regular assessments are provided over time, and there is an appreciation of the equianalgesic dose ratio of methadone to morphine in opioid-tolerant patients.

  • Method 2: Staggered or 3-day Switchover

    One approach calls for a gradual switch over 3 to 5 days to decrease the risk of relative overdosing. An equianalgesic dose of methadone is first calculated, using an equianalgesic dose ratio of morphine to methadone of 10:1 (i.e., methadone being approximately ten times more potent than morphine). The caveat in using a ratio of 10:1 is that variations in ratios have been noted, depending on the dose of the previous opioid. The ratio may be much higher (12:1 or even higher) in patients being switched from high doses of morphine to methadone. The following example is given to illustrate this method: A patient who is on the equivalent of 450 mg per day of oral morphine (quick-release morphine 75 mg orally every 4 hours) needs to be switched to methadone. Using a ratio of 10:1, the predicted equivalent daily oral dose of methadone, once the switch is completed, will be 45 mg. On day 1 of the switch, the daily morphine dose is reduced by one third to approximately 300 mg per day (morphine 50 mg orally every 4 hours) and one third of the predicted daily methadone dose is added, divided into three doses per 24 hours (i.e., methadone 5 mg orally every 8 hours). Morphine continues to be given for rescue doses. On day 2 of the switch the patient is reassessed, and if no problems have developed, the morphine dose is reduced by another third (i.e., morphine 25 mg orally every 4 hours) and the methadone dose is increased by another third (i.e., methadone 10 mg orally every 8 hours). On day 3, the patient is reassessed. If there are complications such as significant somnolence, but the pain is still not under good control, the methadone dose is increased to 15 mg every 8 hours and the morphine is discontinued. On day 3, methadone or a short half-life opioid is added as a rescue dose as needed. The rescue dose is calculated at 5% to 15% of the total daily dose. On day 3, if the patient has good pain control but shows signs of relative overdosing such as significant somnolence, the morphine is discontinued without any increase in the methadone dose (i.e., it remains at the day 2 level or may even be decreased if needed).

  
  
  
  • Method 3: Ad Libitum [42]

    This approach calls for the previous dose to be discontinued and a single fixed-dose of methadone to be given at the start, calculated using an equianalgesic dose ratio of morphine to methadone of 10:1 (i.e., morphine 10 mg being roughly equivalent to 1 mg of methadone), but to a maximum of 50 mg of methadone per dose. After the initial single priming dose, the same dose is administered every 3 hours as needed. The clinician observes the patient and when the demand for rescue doses reduces or stabilizes (indicating steady-state being reached), which is usually on day 4 to 7, the daily requirement is calculated and the dose is given every 8 to 12 hours.

  
  
  
  • Method 4: Initial Priming Followed By Variable Conversion [43]

    In this method, an opioid-naïve patient is started on 3 to 5 mg of methadone every 8 hours and a nonnaïve patient is started on a dose of methadone that is equivalent to 50% of the estimated daily morphine dose. These doses are initially given for 3 days. Once the patient has acceptable pain relief for 6 to 8 hours, the dose is changed to a single fixed-dose once a day and rescue doses are given as needed. This method is probably best suited for opioid-naïve patients (in relatively unlikely situations where more frequently used opioids such as morphine are not available) or patients who are, for one reason or another, being switched from relatively low doses of morphine or other opioids.

  
  
  
  • Method 5: German Model [45]

    This method is suggested when patients are being switched from high equivalent daily doses of morphine (>600 mg orally per day). The morphine or other opioid the patient is receiving is stopped. Methadone at a dose of 5 to 10 mg orally is started every 4 hours and rescue-doses of 5 to 10 mg every hour are allowed as needed. On the second to third days of the switch, the methadone dose is increased by up to 30% every 4 hours until sufficient pain relief is achieved and no significant adverse effects are noted. After exactly 72 hours following the switch to methadone, the dose is changed from every 4 hours to every 8 hours, and the interval of rescue doses is increased to every 3 hours as needed at the same single dose as established on days 2 to 3. The dose can then be increased by up to 30% if further upward titration is required.

  
  
  

  In some countries there are restrictions that do not apply to other opioids on the ability of physicians to prescribe methadone. In the United States, this pertains to methadone for maintenance of addiction. Methadone is not restricted when used for pain management; however, physicians should carefully document the use of methadone.[47] It should be noted that ratios are different for switching from methadone to a morphine-like opioid.[22]

Meperidine (Demerol)

• Useful for brief courses (a few days) to treat acute pain, meperidine is not recommended in treating persistent cancer pain because of its short duration of action (2.5–3.5 hours) and its neurotoxic metabolite, normeperidine. Accumulation of this metabolite, particularly when renal function is impaired, causes central nervous system (CNS) stimulation that may lead to delirium or seizures. Seizures are typically preceded by development of multifocal myoclonus, which can be used as a warning sign.

Tramadol

• Tramadol can be considered an atypical opioid analgesic that has a dual action. It is a weak mu opioid agonist that also inhibits the reuptake of norepinephrine and serotonin.[48,49] It is believed that both mechanisms work synergistically to provide analgesic benefit with a potency that is approximately one-tenth that of morphine [50] and approximately equivalent to codeine. The most common side effects reported with tramadol are drowsiness, constipation, dizziness, nausea, and orthostatic hypotension.[48] There is also a risk of precipitating seizures in patients with a previous history or in patients who are receiving medications that could reduce the seizure threshold. The use of other serotonergic medications (e.g., selective serotonin reuptake inhibitors [SSRIs] and serotonin-norepinephrine reuptake inhibitors [SNRIs]) together with tramadol has the potential to increase the risk of the serotonin syndrome. Tramadol is available in short- and long-acting formulations and in fixed combination with acetaminophen. The recommended starting dose of oral tramadol is 50 mg 1 or 2 times a day, with gradual titration up to a maximum of 400 mg per day.[48] There is also the option of using tramadol via the rectal or subcutaneous route in patients who are unable to tolerate oral medication.[51,52]

Partial agonists

• Partial agonists such as buprenorphine have less effect than full agonists at opioid receptors. They are subject to a ceiling effect and thus are less effective analgesics.

Mixed agonist-antagonists

• Mixed agonist-antagonists block or are neutral at one type of opioid receptor while activating a different opioid receptor. Mixed agonist-antagonists are contraindicated for use in the patient receiving an opioid agonist because they may precipitate a withdrawal syndrome and increase pain. Mixed agonist-antagonists include pentazocine (Talwin), butorphanol tartrate (Stadol), dezocine (Dalgan), and nalbuphine hydrochloride (Nubain). Their analgesic effectiveness is limited by a dose-related ceiling effect.

Most patients with cancer pain require fixed-schedule dosing to manage the constant pain and prevent the pain from worsening.[53] An Italian study of patients whose baseline pain was well controlled on morphine when admitted to a palliative care unit found that most episodes of breakthrough pain were rapidly controlled with IV morphine equivalent to 20% of the calculated equianalgesic total daily dose. Adverse effects were uncommon.[54] An as-needed rescue dose (breakthrough dose) should be combined with the regular fixed-schedule opioid to control the episodic exacerbation of pain, often referred to as breakthrough pain. When this pain is elicited by an action such as weight-bearing, breathing, or defecation, it is termed incident pain. Rescue or breakthrough doses can be given hourly or more frequently as needed, depending on route of administration, pharmacokinetic properties of the drug, and presence or absence of side effects. The breakthrough dose is generally calculated to be 10% to 20% of the total dose of the fixed schedule.[55] Adherence rates are improved when patients are prescribed around-the-clock opioids compared with as-needed prescribing.[56] Preliminary data suggest that the intensity of incident pain related to bone metastases may be diminished by increasing the dose of the scheduled opioid above that needed for control of baseline pain, while maintaining it below that associated with the development of limiting side effects.[57]

Dosage

• The appropriate dosing interval is determined by the opioid and formulation used. The analgesic effects of short-acting oral opioids such as morphine, hydromorphone, codeine, and oxycodone begin within a half hour after administration and last for approximately 4 hours. The dosing interval of these drugs is usually 4 hours. In patients given controlled-release formulations of morphine, hydromorphone, codeine, or oxycodone, relief should begin in 1 hour, peak in 2 to 3 hours, and last for 12 hours (controlled-release hydromorphone and codeine are not available in the United States); these formulations are usually prescribed in 12-hour intervals. A small group of patients, however (10%–20% of those on 12-hour controlled-release formulations), may require administration every 8 hours. The analgesic effect of transdermal fentanyl begins approximately 12 hours after the application of the patch, peaks in 24 to 48 hours, and lasts for approximately 72 hours. Patches are therefore changed every 72 hours. In a select group of patients who consistently experience end-of-dose failure despite increases in the patch doses, the dosing interval can be increased to every 48 hours (<10% of patients on fentanyl patches). Transdermal fentanyl is not recommended for control of acute pain or poorly controlled pain because there is a delayed onset of action until reaching steady-state either with new use or with a change in the dose. Patients receiving transdermal fentanyl may be switched to a continuous IV or subcutaneous infusion of fentanyl using a conversion ratio of 1:1 to facilitate more rapid titration.[58]

Dose titration

• To date, dose titration is largely patient-driven, as determined by the balance of analgesia with side effects. For example, while morphine dose correlates with peak-and-trough plasma concentrations of a parent drug and its metabolites morphine-3-glucuronide and morphine-6-glucuronide, studies are conflicting with regard to the association between plasma levels of morphine and its metabolites versus analgesia as measured by pain scores.[59] The strong opioid agonists have no maximum dose or ceiling dose. The appropriate dose is the amount of opioid that controls pain with the fewest side effects. Dose titration should continue until good pain relief is achieved or intolerable side effects develop that cannot otherwise be controlled. The goal is to achieve a favorable balance between analgesia and side effects through gradual adjustment of the dose. If analgesic tolerance appears to be occurring, the dose can be increased or consideration given to switching the opioid, especially if higher doses are required.

  The severity of the pain and the opioid formulation chosen determine the rate of titration. The dose of immediate-release formulations can be increased on a daily basis if necessary until pain relief is adequate. Among patients receiving relatively low doses of opioids, those with uncontrolled moderate-intensity pain require daily increases of between 25% and 50% to their previous dose, while patients with severe uncontrolled pain may require a higher increase. At higher opioid doses, increases of 20% to 30% would be more prudent. Rapid dose escalation requires close monitoring for both efficacy and side effects. Preliminary data suggest that titration with sustained-release daily morphine is equivalent to titration with immediate-release morphine administered every 4 hours by an expert group of clinicians, but standard practice is to use a short-acting opioid for initial titration.[60]

  Occasionally, doses may need to be reduced or, rarely, stopped. This may occur when patients become pain free as a result of cancer treatment, including treatments such as nerve blocks and radiation therapy. Another time to consider reducing the dose is when a patient experiences significant opioid-related sedation that is accompanied by good pain control. In situations where interventions achieve complete pain relief, rapid opioid tapering rather than abrupt discontinuation is recommended and usually adequate.

Different types of opioids

• The debate regarding whether any individual opioid causes fewer side effects or is more effective is characterized by much speculation but little clinical evidence. These inconclusive findings have prompted expert working groups of the European Association of Palliative Care to recommend that there is currently little evidence of the clinical superiority of one opioid over another regarding the side-effect profile and/or analgesia.[12,13] Even constipation and other side effects may be positively affected by a switch. Compared with morphine, fentanyl may cause less constipation.[61,62] Studies suggesting that oxycodone and hydromorphone may cause less nausea and hallucinations than morphine [63] are juxtaposed with other studies that found no significant differences between them.[64-66] One study found that transdermal fentanyl was better tolerated than sustained-release oral morphine and equally effective.[67]

Tolerance

• Assume that patients actively abusing heroin or prescription opioids (including methadone) have some pharmacologic tolerance that will require higher starting doses and shorter dosing intervals.

Opioid therapy in special populations

• Health professionals should check current recommendations for opioid use in older people, children, people who are cognitively impaired, and known or suspected drug abusers.

Opioid switching

A series of case reports have demonstrated the clinical problem of inadequate pain control with escalating opioid doses in the presence of dose-limiting toxic effects, including hallucinations, confusion, hyperalgesia, myoclonus, sedation, and nausea.[17,23,68-70] It was suggested that these problems could be managed by switching to an alternative opioid, with the result being improved pain management and decreased toxic effects. The improvement with opioid switching, although predominantly demonstrated initially with morphine, has also been reported with other opioids.[71-74] A retrospective review over a 1-year period in a pediatric oncology center supports efficacy of this technique in children, with resolution of adverse opioid effects, largely pruritus, achieved in 90% of patients, while maintaining pain control.[75]

• Guidelines for switching from one opioid to another

  Guidelines for opioid switching are intended to reduce the risk of relative overdosing or underdosing as one opioid is replaced by another. These guidelines require a working knowledge of an equianalgesic-dose table.[13,76] The equianalgesic-dose table provides only a broad guide for dose selection when switching from one opioid to another. Wide ranges in interindividual responses to the various opioids have been noted.[76] Therefore, because of incomplete cross-tolerance in most cases, the calculated dose-equivalent of a new drug must be reduced by 25% to 50% to ensure safety. These figures are based on clinical experience rather than empiric data. The selection of an alternative opioid is largely empirical. There is little clinical evidence to indicate that one opioid has therapeutic superiority over another opioid. A patient, for example, who requires a switch from morphine to another opioid can be switched to hydromorphone, oxycodone, fentanyl, or methadone.[77-79] In one prospective study of 186 cancer patients being treated with morphine, 25% did not respond and required switching to another opioid (oxycodone). The primary reasons for switching included pain, confusion, drowsiness, nightmares, and nausea. Of the 47 patients who required switching to an alternative opioid, 37 (79%) obtained good relief. This result provides beginning evidence for the prevalence of the need to switch, as well as determining the success rate once switching occurs.[80] Patients should be followed closely after a switch and should be reassessed, and the new opioid dose should be adjusted according to the intensity of pain and lack or presence of adverse effects.

Note: The values that appear in the table below are NOT recommended starting doses. Opioid doses are highly variable and should be based on the individual’s previous responses and overall condition. Important cautions are contained in the footnotes.