Abeloff's Clinical Oncology, 4th Edition

Part II – Problems Common to Cancer and its Therapy

Section A – Symptom Management and Palliative Care

Chapter 36 – Cancer Pain

Stuart A. Grossman,
Suzanne Nesbit



Major Presenting Symptom of Malignancies



Affects more than 30% of patients undergoing antineoplastic therapy



Moderate to severe pain occurs in over 70% of patients during the later phases of their illness



Significantly affects quality of life



Frequently managed poorly

Etiology of Complication



Can be of nociceptive, neuropathic, or sympathetically maintained origin



Due to direct tumor involvement (70%), evaluation or therapy (20%), or illness unrelated to the malignancy (<10%)

Evaluation of the Patient



Determining the etiology of pain is key to appropriate therapy



Treat pain aggressively during evaluation



Fully evaluate the pain using a careful history and physical examination, validated pain assessment scales, and selected laboratory tests

Grading of the Complication



Measurements of pain intensity using validated pain assessment scales



Results should be recorded serially as an integral part of the medical record




85% of patients can be well palliated using simple, inexpensive, “low-technology,” oral analgesics



Addition of appropriate adjuvant pain medications, alternate routes of opioid administration, antineoplastic therapy, nonpharmacologic approaches, neurostimulatory techniques, regional analgesia, and neuroablative procedures provides excellent palliation for nearly all patients with cancer pain



Pain is one of the most common and dreaded symptoms associated with cancer. It occurs in one-quarter to one-half of patients with newly diagnosed malignancies, in one-third of those undergoing treatment, and in more than three-quarters with advanced disease. Overall, 75% of patients with cancer experience pain severe enough to require treatment with opioids during their illness.[1] Unrelieved pain directly affects patients’ daily activities, quality of life, and psychologic status. The importance of this symptom and the availability of excellent analgesic therapies make it imperative that health care providers be adept at the evaluation and treatment of cancer pain ( Table 36-1 ).

Table 36-1   -- Landmark Literature on Cancer Pain



Foley KM: N Engl J Med 1985;313:84–95. A consideration of changing attitudes toward the use of narcotic analgesics and novel methods of managing cancer pain.



WHO: Cancer Pain Relief, 1986. A guide to pain relief that also served to set cancer pain in context.



Twycross RG: Oncology 1988;2:35–43. An early review of the literature in cancer pain management offered a guide to drugs and dosages.



Cleeland CS, et al: N Engl J Med 1994;330:595–596. A study of 1308 cancer outpatients concluded that despite published guidelines for pain relief, many patients receive inadequate analgesia.



Grossman SA, Staats PS: Oncology 1994;8:93–107. A detailed look at state-of-the-art management of cancer pain.



Levy MH: N Engl J Med 1996;335:1124–1132. An in-depth review of the pharmacologic treatment of cancer pain.



Koshy RC, et al: Support Care Cancer 1998;6:430–437. A discussion of the complex issues involved in cancer pain management in developing countries.



NCCN: Guidelines for Adult Cancer Pain. Available at <http://www.nccn.org> First published in 1998, an expert consensus panel offers guidelines on currently accepted approaches to treatment.



Miaskowski C, et al: APS Clinical Practice Guidelines Series, No. 3, 2005.



An updated guideline for the management of cancer pain in adults and children.



Smith TJ, et al: J Clin Oncol 2002;20:4040–4049. In this study, an implantable intrathecal drug delivery system improved pain control and survival in patients with refractory cancer pain.

From Grossman SA, Dunbar EM, Nesbit SA: Cancer pain management in the 21st century. Oncology 2006;20:1333–1340.





Pain in patients with malignancies is a complex and often recurring process that occurs from many causes. Ninety percent of pain in patients with cancer results from the tumor or its evaluation or therapy, whereas less than 10% is due to unrelated illnesses. In 70% of patients, pain develops from tumor invading or compressing soft tissue, bone, or neural structures. The common pain syndromes that result are listed in Table 36-2 . The remaining 20% of cancer pain occurs from diagnostic and therapeutic procedures that these patients undergo in the process of evaluation and treatment.[2] Examples of these procedures include venipuncture, bone marrow aspiration and biopsy, endoscopy, lumbar puncture, invasive radiologic procedures, surgery, chemotherapy, and radiation therapy.

Table 36-2   -- Etiology of Pain in Cancer Patients



Direct tumor involvement (70%)



Invasion of bone



Invasion or compression of neural structures



Obstruction of hollow viscus or ductal system of solid viscus



Vascular obstruction or invasion



Mucous membrane ulceration or involvement



Cancer-induced syndromes (<10%)



Paraneoplastic syndromes



Pain associated with debility (i.e., bedsores, constipation, rectal or bladder spasm)



Other (i.e., postherpetic neuralgia)



Diagnostic or therapeutic procedures (20%)



Procedure-related pain (i.e., bone marrow aspiration or biopsy, lumbar puncture)



Acute postoperative pain or postsurgical syndromes (i.e., postmastectomy, postthoracotomy, postamputation syndromes)



After radiation (i.e., injury to plexus or spinal cord, mucositis, enteritis)



After chemotherapy (i.e., mucositis, peripheral neuropathy, aseptic necrosis)



Pain unrelated to the malignancy or its treatment (<10%)

From Grossman SA, Baumohl L: Evaluation and management of cancer pain. In Bone RC (ed): Current Practice of Medicine. Philadelphia, Current Practice, 1993, p 18.1.




Surgery is a frequent cause of pain in patients with cancer and can consist of biopsy, removal, or debulking of a tumor or management of a complication of the tumor or its treatment, such as a small-bowel obstruction. These procedures are associated with postoperative pain and injury to local nerves, which can produce neuromas and chronic pain syndromes that are severe and difficult to manage. Surgically induced nerve injuries are most commonly seen after breast cancer surgery, thoracotomy, radical neck dissection, and limb amputation.[3] Postmastectomy syndrome occurs in 4% to 10% of all women undergoing this type of breast cancer surgery. It is most frequent in patients with postoperative complications or keloid formation and is characterized by a constricting, burning sensation in the posterior arm, axilla, and anterior chest. It can develop immediately after the procedure or months later and can be complicated by the secondary development of a frozen shoulder. Postthoracotomy syndrome occurs after nerve injury secondary to rib retraction and typically presents as an aching, burning sensation in the incisional area with local tenderness, sensory loss, and occasional autonomic changes. Injury to local nerves after a radical neck dissection can produce tightness and burning dysthesias in the area of sensory loss and acute, lancinating pain. The loss of neck musculature from this surgery can also result in a “droopy shoulder,” thoracic outlet syndrome, and suprascapular nerve entrapment.

Chemotherapy and radiation also produce significant pain in patients with cancer. Phlebitis, mucositis, hemorrhagic cystitis, and peripheral neuropathy are common complications of antineoplastic agents. Glucocorticoids, administered as a component of therapy, can cause aseptic necrosis of the hip, and severe perineal pain when given rapidly in high doses. Examples of radiation-induced pain include mucositis, local skin reactions, enteritis, proctitis, fibrosis with nerve entrapment syndromes, and radiation myelopathy. Electric shocklike sensations that accompany flexion of the neck (Lhermittes syndrome) can last for months after radiation to the spinal cord. Painful peripheral nerve tumors can also follow radiation therapy, especially in patients with neurofibromatosis.

Patients with malignancies are also predisposed to painful infections. Common examples include pneumonia, urinary tract infection, wound infections, candida esophagitis, oral or genital herpes, and herpes zoster.

Current Status of Cancer Pain Management

Studies from hospices and from World Health Organization demonstration sites suggest that 85% of patients with cancer pain can be well palliated using oral opioids. A wide array of effective options exist for the remaining 15% of patients. These include parenteral, transdermal, transmucosal, or intraspinal opioids, glucocorticoids, anti-inflammatory and adjuvant medications, antineoplastic therapies, and anesthetic and neurosurgical procedures.

Although proper use of available therapeutic approaches should result in excellent pain control in nearly 95% of patients with cancer pain, cancer pain remains grossly undertreated throughout the world.[4]In most countries, the unavailability of oral opioids is a major contributing factor. [5] [6]

Even in countries such as the United States and the United Kingdom, where a wide range of opioid analgesics and routes of administration are available, studies suggest that cancer pain is undertreated. A survey of oncologists highlights their reluctance to assess pain routinely and prescribe appropriate analgesics. These findings prompted the creation of cancer pain initiatives in most states and the development of cancer pain guidelines and algorithms by the American Society of Clinical Oncology, the American Pain Society, and the U.S. Public Health Service's Agency for Health Care Policy and Research. [7] [8] [9] [10] [11] In addition, to improve the overall management of pain in the United States, the Joint Commission on Accreditation of Healthcare Organizations has recently set new standards for pain management that are required for continued accreditation.[12]

Barriers to the Provision of Adequate Analgesia

Many reasons have been cited for the inadequate treatment of patients with cancer pain in developed nations. Some of them relate directly to health care providers, including



Failure to appreciate the intensity of the pain their patients are experiencing



Reluctance to evaluate the etiology of the pain



Lack of training in pain management



Excessive concern regarding the regulatory oversight of opioid prescribing

One major barrier to the provision of adequate analgesia in patients with cancer is the failure of health care providers to appreciate the intensity of patients’ pain. This occurs because pain is entirely subjective and can only be experienced and quantified by the patient. There are no pathognomonic findings on physical examination, and laboratory studies can be normal. Assessment of pain is further complicated by the complexities surrounding death and dying and the possibility that patients with chronic, severe pain might not appear or act uncomfortable. In one study examining health care provider perceptions of patient pain, pain intensity was quantitatively assessed using a visual analog scale in 103 consecutive patients admitted to the solid-tumor service of a large cancer center.[13] Each patient's primary care nurse, house officer, and oncology fellow rated his or her perceptions of their patient's pain intensity using the same pain rating instrument as the patient. The results ( Fig. 36-1 ) demonstrate a lack of correlation between the patient's and the health care provider's perception of patient pain. Furthermore, the concordance between patient and health care provider pain intensity scores was highest when patients had no pain and lowest when the patients were experiencing severe discomfort. Similar results have been obtained in studies of patients with cancer and their next of kin and in burn patients.


Figure 36-1  Correlation of health care provider and patient perceptions of patient pain.  (Data from Grossman SA, Sheidler VR, Swedeen K, et al: Correlation of patient and caregiver ratings of cancer pain. J Pain Symptom Manage 1991;6:53. Copyright U.S. Cancer Pain Relief Committee.)




There are many reasons why health care providers might be unaware of the pain their patients are experiencing. Because pain is entirely subjective, its presence and intensity must be communicated to health care providers by patients. Patients, however, might not discuss their pain if they expect cancer to be painful or if they are concerned about opioid addiction, tolerance, or side effects or about diverting their physician's attention from treating the tumor. In addition, they might be reluctant to admit to themselves or others that their pain has worsened, knowing that this could signify progression of the cancer. Health care providers also contribute to the lack of communication by neglecting to emphasize their interest or abilities in controlling pain and by failing to use validated pain assessment tools. Serial numeric pain ratings in the medical record will foster the necessary dialogue between patients and health care providers about pain management issues.

These issues are greatly magnified in children, the elderly, or individuals with a history of drug abuse. Children have special difficulty in communicating pain intensity, and their unique pain management needs have been relatively neglected. Special pain assessment tools are required, and the child's age and developmental level must be considered when planning assessment or interventions. Many elderly patients also find it difficult to communicate their discomfort to health care professionals, have multisystem disease, and are especially sensitive to the adverse effects of analgesics.[14] Those with cancer and a current or prior history of drug abuse often have difficulty finding health care providers who believe their reports of pain and who will provide the high doses of analgesics required in these opioid-tolerant individuals.[15]

Another barrier to the provision of adequate analgesia relates to the training of medical professionals. The principles of cancer pain management receive little attention in academic centers and relevant scientific societies. Medical school courses and textbooks typically focus on diseases rather than symptoms, and pain management issues are infrequently highlighted at rounds, educational conferences, or in the formal curriculum of those training to care for patients with cancer. These circumstances leave many health care professionals eager to concentrate on medical problems they feel competent to handle. In addition, the scarcity of research abstracts on cancer pain at the scientific meetings of physician oncologic societies reinforces the notion that pain control is a topic of limited importance.

The lack of training and emphasis on cancer pain management manifests in many ways, including physicians’ lack of opioid-prescribing skills, failure to evaluate the etiology of cancer pain, and excessive concerns regarding the regulatory oversight of opioid prescribing. In one study, 81 physician trainees, given a hypothetical patient case that included performing opioid conversions, 75% calculated a dose that was only one-third the correct dose, and only 5% calculated the dose correctly.[16] Another study of experienced oncology nurses indicated that they lack the capacity to recognize physician orders that could result in serious over- or underdosing of opioids. These difficulties in calculating equianalgesic doses have prompted the development of software to facilitate opioid conversions.[17]

Many physicians and nurses consider “cancer pain” a diagnostic entity that requires opioids, without a formal evaluation of the etiology of the pain. Although this approach can provide relief, it is often ineffective and can lead to indefensible medical practices. For example, progressive back pain in a patient with metastatic lung cancer can occur from a postobstructive pneumonia or tumor invasion of the esophagus, liver, spleen, pleura, pericardium, rib, vertebrae, intercostal nerves, brachial plexus, leptomeninges, or epidural space. Each of these diagnostic possibilities can be associated with a different therapeutic approach or sense of urgency. To provide opioids without evaluation would be an error in such a patient with an impending epidural cord compression. Furthermore, many common cancer pain syndromes might be better treated with therapies tailored to a patient's individual pain problem. These may include local radiation, nerve blocks, glucocorticoids, anticonvulsants, or surgery to maximize analgesia, minimize side effects, and improve quality of life.

Physicians, pharmacists, and nurses caring for patients with cancer must be willing to prescribe, dispense, and administer the opioids in doses required to alleviate their pain. Drug enforcement agencies often discourage opioid prescribing, however, in an attempt to reduce the illegal diversion of these drugs. Unfortunately, many health care professionals with limited knowledge and experience in the treatment of cancer pain react to the perceived threat of investigation by law enforcement agencies with dramatic decreases in opioid prescribing.[18] This perception further contributes to the undertreatment of patients with cancer pain.


A comprehensive assessment of cancer pain is the first important step toward optimal pain relief.[2] This evaluation should provide the clinician with sufficient information to carry out the following tasks:



Estimate the severity of pain.



Form a clinical impression regarding the etiology of the pain.



Determine the need for further diagnostic studies.



Formulate therapeutic recommendations that take into account the patient's overall medical and psychosocial status ( Table 36-3 ).

Table 36-3   -- Components of a Comprehensive Assessment of Cancer Pain



Detailed history of current pain problem



Catalogue of pain (number and locations)



Information for each pain



Intensity (0–10)



Locations and radiation



Onset and changes over time



Temporal pattern (constant, intermittent, etc.) and quality (burning, etc.)



Exacerbating and relieving factors



Associated neurologic or vasomotor abnormalities



Other associated factors



How the pain interferes with the patient's life



Current therapeutic modalities (schedule, efficacy, side effects)



Prior therapeutic modalities (schedule, efficacy, side effects)



Oncologic history






Presentation: date, stage, sites of involvement



Antineoplastic therapies: dates, types, doses, toxicities, and response to each therapy



Current sites of disease: stable, responding, or progressive



Patient expectations and goals



Medical history—may be affected by pain therapies



Coexisting diseases



Medications and allergies



Substance abuse history



Other constitutional symptoms (i.e., anorexia, fatigue, sedation and other changes in mental status, nausea, vomiting, dysphagia, dyspnea, constipation, urinary and sexual function, depression, dry mouth, ability to take medications by mouth, presence of a central venous catheter)



Personal and social history



Background: age, educational, employment, marital, residential, religious, cultural, ethnic



Current status: functional status, caregivers and their health and availability, support system



Physical examination



Review of additional information



Medical records, radiologic/laboratory studies



Family members and physicians and/or nurses who know the patient and his or her illness



Differential diagnosis



Recommendations regarding workup and therapy




From Grossman SA: Cancer pain assessment: a continual challenge. Support Care Cancer 1994;2:105.




As with any serious medical condition, the assessment of cancer pain requires a detailed history, physical examination, and review of available records, laboratory data, and imaging studies. The special challenges associated with the assessment of cancer pain include the entirely subjective nature of pain, the complex multisystem involvement in patients with advanced malignancies, and the ever-changing clinical situation in this patient population.

A detailed pain history is the cornerstone of the assessment. This can be complex, because 75% of patients with advanced cancer have several concurrent painful sites, and nearly one-third have four or more separate pain problems. [19] [20] Each distinct pain must be identified and characterized. Pertinent information should include its intensity, location, radiation, how and when it began, how it has changed over time, and what makes it better or worse. The quality of each pain, its temporal pattern, whether it is associated with neurologic or vasomotor abnormalities, how it interferes with the patient's life, and an account of the successes and failures of current and prior therapeutic modalities also provide valuable insight.

Many instruments have been developed to aid in pain assessment.[21] These attempt to characterize and quantify the quality and/or intensity of a patient's pain and represent the best available means to document the discomfort and to follow the results of therapy serially. Each instrument has its shortcomings, but several have been validated in patients with cancer pain and incorporated into clinical practice. Most contain a variant of the unidimensional visual analog scale and a schematic representation of the body for the patient to indicate where their pain is located. The McGill Pain Questionnaire is comprehensive, but too awkward and time-consuming for most oncology patients in a clinical setting. [22] [23] [24] The Wisconsin Brief Pain Inventory, which can be completed in 15 minutes, provides information on the characteristics, severity, and location of the pain, its interference with normal life functions, and the efficacy of prior therapy. The Memorial Pain Assessment Card can be completed in less than a minute and features scales for the measurement of pain intensity and pain relief.[25] It is also designed to provide insight into global suffering or psychologic distress. The Hopkins Pain Rating Instrument is a validated plastic version of the visual analog scale that obviates the need for the paper, pencil, ruler, and measurements associated with the standard visual analog scale.[26] This simplifies repeated pain intensity measurements, making it easier to reassess the efficacy of therapeutic endeavors on a continuing basis.[27]

A complete oncologic history is also essential, because 90% of cancer pain is related to the malignancy or cancer treatment. The histology, presentation, stage, sites of involvement, and natural history and the history of surgery, radiation, chemotherapy, and hormonal treatments will help shape a therapeutic approach. In addition, it is important to note whether the malignancy is responding to therapy, stable, or progressing. A general medical history is also helpful, because pain treatments can affect coexisting medical problems, exacerbate constitutional symptoms, interact with other medications, or be contraindicated because of allergies. For example, a patient with painful bone metastases and severe peptic ulcer disease would not be an ideal candidate for potent anti-inflammatory agents. Opioids can be problematic in patients with severe benign prostatic hypertrophy or severe obstructive pulmonary disease and carbon dioxide retention. Likewise, knowledge that a patient tolerates food or fluids poorly by mouth, has an indwelling venous access device, or admits to substance abuse might influence decisions about the best way to control that patient's pain. The patient's age, functional status, social support, education, residence, health insurance, finances, and religious and cultural background might also figure prominently in planning therapy. A careful neurologic and physical examination also provides important clues as to the etiology of the pain. Added insight can come from a review of available laboratory and imaging data, from medical records, and from discussions with family members and physicians who are familiar with the patient and his or her illness.

The history, physical examination, and review of other available data should provide the clinician with sufficient information to formulate a differential diagnosis for each of the patient's distinct pains and to make recommendations regarding the workup and therapy for each. Based on this initial impression, analgesic therapy should be initiated. The nature of the treatment prescribed might depend on the clinician's judgment regarding the origin of the pain. Somatic, visceral, neuropathic, and sympathetically maintained pain are each approached somewhat differently ( Table 36-4 ). Prompt institution of therapy reassures patients that their pain will receive immediate attention, ensures patient comfort for diagnostic studies, and can provide information on the accuracy of the clinician's assessment. Excellent pain relief suggests an accurate initial diagnosis and appropriate therapy, whereas suboptimal control might prompt a new treatment approach or a search for a different etiology of the pain.

Table 36-4   -- Classification of Cancer Pain




Primary Therapies


Constant, aching, gnawing, often well localized

Bone metastases

Treatment of tumor, anti-inflammatory agents, analgesics


Constant, aching, often associated with nausea

Pancreatic cancer

Treatment of tumor, analgesics, nerve blocks


Paroxysmal shocklike pain on top of a burning, constricting sensation

Plexopathy or postherpetic neuralgia

Treatment of tumor, analgesics, TENS, nerve blocks

Sympathetically maintained

Severe burning, squeezing, or constricting with local edema

Reflex sympathetic dystrophy

Sympathetic blockade, physiotherapy, adjuvant analgesics

From Grossman SA, Staats PS: The current management of pain in patients with cancer. Oncology 1994;8:93.

TENS, transcutaneous electrical nerve stimulation.





One of the most difficult aspects of cancer pain management is that the patient's clinical situation is rarely static. The patient's underlying malignancy, antineoplastic therapy, and psychosocial status change continually during the course of the illness. As a result, the etiology and intensity of each new or worsening pain must be reassessed. The toxicities of the analgesics should also be evaluated periodically, because they can affect quality of life substantially. If significant toxicities are recognized, alternate approaches with a lower toxicity profile can be attempted.


Nearly 85% of patients with cancer pain can achieve good control of their pain with conventional oral medications. More aggressive or invasive therapies should provide pain relief to an additional 10% of patients, leaving only a small fraction of patients with cancer with inadequate relief. [28]

Pharmacologic Therapy

Pharmacologic approaches are the most commonly used treatments for cancer pain, because they are effective, safe, and usually inexpensive. These are classified as nonopioids, opioids, and adjuvant analgesics. Twenty years after the World Health Organization's analgesic ladder was introduced, it still provides a framework for analgesic prescribing. [29] [30] [31] Aspirin, acetaminophen, or nonsteroidal anti-inflammatory drugs (NSAIDs) are preferred for mild to moderate pain.[32] If these do not provide adequate analgesia, codeine, oxycodone, or hydrocodone will frequently provide excellent relief (Table 36-5 ). For persistent or severe pain, codeine (or its congener) is replaced by a potent opioid, such as morphine ( Table 36-6 ). Drug substitution should be considered before an entire class of agents is abandoned, because patients frequently tolerate one NSAID or opioid better than another. In addition, patients with severe pain might need a strong opioid as initial therapy to ensure rapid pain relief.[33]

Table 36-5   -- Opioids for Mild to Moderate Pain



Equi-Analgesic Dose (mg)[*]

Peak Effect (hr)

Duration of Effect (hr)







Ceiling for analgesia reached at doses >240 mg/day orally.












No ceiling dose if given without fixed combinations; parenteral formulation not available.






Only available as fixed combination with acetaminophen or aspirin.






100 mg napsylate = 65 mg hydrochloride salt. Not recommended for treatment of cancer pain.

Modified from Grossman SA, Gregory E: Cancer pain. In Kirkwood MT, Lotze MT, Yasko JM (eds): Current Cancer Therapeutics. Philadelphia, Current Medicine, 1994, p 290.


Approximate potency relative to 10 mg of parenteral morphine.



Table 36-6   -- Strong Opiates for Moderate to Severe Cancer Pain



Equi-Analgesic Dose (mg)[*]

Duration of Effect (hr)






No ceiling dose if given without fixed combinations; parenteral formulations not available.










Many oral formulations for individual patient needs.















Good choice for SQ due to potency.















Not preferred due to CNS toxic metabolite that accumulates in renal failure.










Long T1/2 (11 hr) necessitates slow dose titration. Drug accumulation may occur.










Short T1/2 (<1 hr). TD dose titration difficult with depot in SQ adipose tissue. TD fentanyl 25 μg/hr ≈ 45 mg/day oral morphine.










Despite long T1/2 (15 to >150 hr), duration of analgesia is not prolonged; however, drug accumulation can result in toxicities. Caution is warranted when converting to methadone in patients with high opioid tolerance.





Now available as immediate-release formulations.











Modified from Grossman SA, Gregory E: Cancer pain. In Kirkwood MT, Lotze MT, Yasko JM (eds): Current Cancer Therapeutics. Philadelphia, Current Medicine, 1994, p 290, with permission.

IV, intravenous; PO, oral; SQ, subcutaneous; TD, transdermal; (?), unknown.



Approximate potency relative to 10 mg of parenteral morphine. Caution: Cross-tolerance between opioids is incomplete. Use caution when performing equianalgesic conversions. Titrate to clinical response.

Ripamonti C, Groff L, Brunelli C, et al: Switching from morphine to oral methadone in treating cancer pain: what is the equianalgesic dose ratio? J Clin Oncol 1998;16:3216–3221; Moryl N, Santiago-Palma J, Kornick C, et al: Pitfalls of opioid rotation: substituting another opioid for methadone in patients with cancer pain. Pain 2002;96:325–328; Bruera E, Neumann CM: Role of methadone in the management of pain in cancer patients. Oncology 1999;13:1275–1282; Pereira J, Lawlor P, Vigano E, et al: Equianalgesic dose ratios for opioids: a critical review of proposals for long term dosing. J Pain Symptom Manage 2001;22:672–687; Bruera E, Sweeny C: Methadone use in cancer patients with pain: a review. J Pall Med 2002;5:127–138.



The site of action of the nonopioids is primarily the peripheral nervous system. These agents are not associated with physical dependence, tolerance, or addiction, and they have a maximum dose associated with analgesia. Many are available in combination with a weak opioid. The anti-inflammatory component of aspirin and of the NSAIDs and COX-2 inhibitors is often useful for patients with somatic pain from bone metastasis, inflammation, or mechanical compression of tendons, muscles, pleura, and peritoneum, and for nonobstructive visceral pain. Because some of these agents can affect platelet and renal function or act as antipyretics, they should be administered thoughtfully to patients receiving chemotherapy. COX-2 inhibitors allow NSAIDs to be used with less risk of gastrointestinal bleeding and platelet dysfunction. However, the mounting evidence of increased cardiovascular risk calls into question the overall safety of all NSAIDs. Caution must be exercised in using these agents in patients with a history risk of cardiovascular disease. [34] [35] [36] In addition, it is important to recognize that sustained high doses of acetaminophen can cause renal and hepatic damage, especially when combined with more than 2 ounces of alcohol per day or with other agents that cause liver damage or induce hepatic microsomes.

Most patients with moderate-to-severe pain rely primarily on opioid analgesics for the management of their cancer pain. The vast majority of patients can be managed with oral opioids. These are best given “around the clock” to keep pain under control. Although tolerance to these agents occurs, tumor progression is the most common reason for increasing opioid requirements. Tolerance can easily be overcome by raising opioid doses. Addiction is extremely rare in patients with cancer who are taking opioids for pain relief ( Table 36-7 ). Most opioid side effects can be managed without excessivedifficulty ( Table 36-8 ). Constipation should be anticipated and treated prophylactically.[37]

Table 36-7   -- Important Definitions in the Treatment of Cancer Pain






A normal physiologic response to chronic opioid administration characterized by development of the abstinence syndrome on abrupt withdrawal of opioids.



A potential problem in virtually all patients receiving moderate to high doses of opiates.






A normal pharmacologic response to chronic opioid therapy characterized by the development of a relative resistance to analgesic and other effects of the drug.



Overcome by increasing the dose administered.






Abnormal behavior pattern characterized by an all-consuming desire to obtain opioids for reasons other than pain relief. This often occurs at the expense of the patient's physical, social, and environmental well-being.



Extraordinarily rare in patients with cancer pain.



Not to be confused with “pseudo-addiction,” which is behavior commonly seen in patients who are undertreated and in pain attempting to obtain appropriate analgesia.

Modified from Grossman SA, Gregory E: Cancer pain. In Kirkwood MT, Lotze MT, Yasko JM (eds): Current Cancer Therapeutics. Philadelphia, Current Medicine, 1994, p 290, with permission.

Definitions Related to the Use of Opioids for the Treatment of Pain 2001. (Available at http://www.ampainsoc.org; www.asam.org).





Table 36-8   -- Management of Common Opioid Side Effects

Side Effect


Specific Agents


Begin bowel program when initiating therapy.

Stool softeners


Combinations of agents may be useful.




Bulk laxatives







Nausea and vomiting

Treat with antiemetics, especially phenothiazine and anticholinergic agents.






Switch to another opiate.




5HT3 antagonists








Use of stimulants












Treat with antihistamines.



Consider another opiate, avoiding morphine.






Switch to another opiate or lower opiate dose.






Use of anxiolytics



Avoid meperidine, especially in patients with impaired renal function.


Withdrawal symptoms

Taper dose by one-half every other day when discontinuing.


Modified from Grossman SA, Gregory E: Cancer pain. In Kirkwood MT, Lotze MT, Yasko JM (eds): Current Cancer Therapeutics. Philadelphia, Current Medicine, 1994, p 290, with permission; O'Mahony S, Coyle N, Payne R: Current management of opioid-related side effects. Oncology 2001;15:61–77; Cherny N, Ripamonti C, Pereira J, et al: Strategies to manage the adverse effects of oral morphine: an evidence-based report. J Clin Oncol 2001;19:2542–2554.




The opioids have their primary effect centrally, where they interfere with pain perception. They can be classified into three groups:



Morphine-like opioid agonists that bind competitively with μ and κ receptors (e.g., codeine, fentanyl, hydromorphone, morphine, oxycodone, and methadone)



Opioid antagonists that have no agonist receptor activity (e.g., naloxone)



Mixed agonists-antagonists (e.g., pentazocine and butorphanol) or partial agonists (e.g., buprenorphine).[38]

The mixed agonist-antagonist drugs have limited utility in cancer pain because of their side effect profiles and their propensity to induce opioid withdrawal in patients who have received opioid agonists. Proper opioid prescribing is critical to patients with cancer ( Box 36-1 ), who often require high doses of opioids for long periods of time.[39]

Box 36-1 


Tenets of Opioid Prescribing



Order opioids on a scheduled “around-the-clock” basis to optimize relief.



Order an as-needed opioid to treat breakthrough or incident pain. For example, if a patient is taking morphine elixir, 100 mg orally every 4 hours, order an additional 25 to 50 mg of oral morphine elixir every 2 hours as needed for pain.



Initiate a prophylactic bowel regimen at the same time opioids are prescribed. Patients usually require a combination of detergent and stimulant cathartics to treat opioid-induced constipation.



Treat opioid-induced nausea and vomiting with aggressive antiemetic management. This includes giving patients antiemetics on an around-the-clock basis. Patients often become tolerant to this side effect several days after beginning opioids.



Once baseline opioid requirements are determined, sustained-release opioid preparations can be used to reduce the number of pills taken each day.



Teach the patient and family about the purpose and benefits of opioids to allay their fears about side effects and addiction. This instruction will improve patient compliance.



Frequent assessment of pain relief is paramount during the opioid titration period. Titrate doses based on the patient's report of pain relief and/or the amount of as-needed opioid that has been required for patient comfort.



Maximize the doses of one opioid before changing to another agent or route. Changes should be made primarily because of toxicities. For example, a patient taking 200 mg of controlled-release morphine every 12 hours and 200 mg of immediate-release morphine daily for breakthrough pain should have the dose of controlled-release morphine increased to 300 mg every 12 hours if he or she is not experiencing significant opioid side effects. This approach is more likely to be beneficial than beginning titration with subcutaneous or intravenous morphine or oral hydromorphone.



Refer to equianalgesic tables when initiating or changing a patient's analgesic regimen (see Tables 36-5 and 36-6 [5] [6]).



Avoid chronic administration of intramuscular or rectal opioids.



Do not use chronic administration of meperidine, which can be associated with the accumulation of normeperidine, a neurotoxic metabolite.

Several classes of drugs that are used primarily for conditions other than pain have been found to be useful adjuvant analgesics in specific circumstances ( Table 36-9 ). Antidepressants and anticonvulsants can be effective in neuropathic pain. Psychostimulants can decrease opioid-induced sedation. Glucocorticoids are effective anti-inflammatory agents and are also used to reduce pain associated with brain edema and epidural metastases. Muscle relaxants, anxiolytic, antispasmodic, and neuroleptic agents also are administered for specific indications.[40] Bisphosphonates reduce the incidence of skeletal complications, particularly in patients with myeloma and breast cancer. [41] [42] Caution must be exercised in the use of adjuvant drugs that have sedative properties, in that the dose of opioids should not be compromised by the toxicities of these secondary agents.

Table 36-9   -- Commonly Used Adjuvant Analgesics for Cancer Pain

Drug Category



Common Toxicities



Neuropathic pain










Sedation, dry mouth, constipation, postural hypotension, urinary retention

Begin with low doses (10–25 mg); increase dose every few days; expect to see pain relief within several days, mood elevation within several weeks.


Neuropathic pain, myoclonic jerks









Valproic acid







Drowsiness, dizziness, nausea, rash, bone marrow depression

Use loading dose with phenytoin; monitor platelets with carbamazepine.


Opioid-induced sedation










Nervousness, irritability, insomnia, dizziness, dry mouth

Give early in the day to avoid insomnia; do not use if patient is already delirious or confused.


Spinal cord compression, increased intracranial pressure, visceral distention










Gastritis, insomnia, fluid retention, hyperglyceroximal myopathy, increased appetite


Muscle relaxants

Muscle spasm













Sedation, dizziness, nausea, weakness, confusion



Muscle spasm, myoclonus, anxiety, insomnia
















Sedation, delirium, hypotension, headache, respiratory depression

Not analgesics; synergistic effect with opioids can cause respiratory depression.


GI or bladder spasm



Diphenoxylate and atropine






Scopolamine patch




Sedation, dry mouth, constipation



Delirium, agitation, nausea and vomiting, hiccoughs













Sedation, orthostatic hypotension, confusion, extrapyramidal reactions

Useful for symptoms other than pain; methotrimeprazine has analgesic properties


Bone pain






Zoledronic acid

Hypocalcemia, fever, GI disturbances, anemia

Delays time to painful skeletal events; also used with analgesics for bone pain.

GI, gastrointestinal.




Although the vast majority of outpatients can be managed with oral opioids, alternative routes of analgesic administration are sometimes needed. Subcutaneous, intravenous, transdermal, transmucosal, or intraspinal opioids can be delivered by intermittent bolus, continuous infusion, or a combination of both, as is frequently the case with patient-controlled analgesia. These alternative routes of administration should be considered when a patient has one of the following conditions:



Intractable vomiting or bowel obstruction, making oral therapy impractical



Ineffective pain relief despite titration to toxicity with several oral opioids



Unacceptable toxicities to several oral opioids



Such high opioid requirements that oral administration is impractical

The costs associated with these routes of opioid administration must be considered carefully. In addition, care must be taken not to transform the home unnecessarily into a complex health care setting.

Subcutaneous opioid injections administered through a butterfly or subcutaneous needle on a fixed schedule are used commonly by hospices as an effective, less expensive alternative to continuous intravenous or subcutaneous infusions. A transdermal system for opioid administration can be beneficial for some patients. Although transdermal fentanyl provides patients with continuous drug delivery, it does not eliminate the need for additional analgesics for breakthrough pain. The slow onset of action of fentanyl and the uncertainties associated in conversion from other opioids have led many to reserve transdermal fentanyl for patients with stable opioid requirements who do not have significant incidental pain. [43] [44]

Oral transmucosal fentanyl citrate (OTFC) can be effective for patients with incident or breakthrough pain, for whom rapid onset and short duration of action are desired.[45] This opioid is available as a lozenge or a buccal tablet. When placed into the mouth, a portion of the fentanyl is absorbed rapidly through the oral mucosa, while the remainder is swallowed and absorbed through the gastrointestinal tract. The onset of analgesia can be as soon as five minutes. The optimal dose for this delivery system is found through titration and is not predicted by the around-the-clock dose of opioids. The effervescent transmucosal fentanyl tablet has a higher bioavailability than oral transmucosal fentanyl citrate, and the manufacturer recommends converting doses of the two products.[46]

Intraspinal opioids produce analgesia without blocking other sensory, motor, or sympathetic functions. These can be delivered into the epidural space through a tunneled external catheter or to the subarachnoid space or lateral ventricles using a totally implanted pump.[47] Because the total daily dose of intraspinal opioid is one-tenth to one-hundredth of parenteral opioid, it is associated with fewer systemic toxicities. Chronic epidural or intrathecal opioids are invasive, expensive, and frequently ineffective in patients requiring high doses of systemic opioids. Tolerance, pruritus, urinary retention, and nausea and vomiting occur in as many as 20% of patients receiving spinal opioids. Respiratory depression is unusual. The addition of low doses of anesthetic agents or agents such as clonidine to intrathecal and epidural opioids could add considerably to pain relief. Ziconotide, a selective N-type calcium channel blocker, is approved for intrathecal analgesia in patients with pain refractory to other treatments.[48] Intraspinal opioids are generally used after documentation of the failure of maximal doses of opioids through other routes.[49] Consensus guidelines are available to guide the practitioner with proper drug selections for intraspinal analgesia.[50]

Antineoplastic Therapy

Antineoplastic therapy can provide analgesia if it reduces the size of lesions invading or compressing normal tissues. Radiation therapy is the treatment of choice for most patients with local pain from tumor progression. It is frequently administered to patients with symptomatic bone, brain, epidural, and plexus metastases. Systemic radiopharmaceuticals such as strontium-89 and samarium are also used for the treatment of pain from bone metastases.[51] Chemotherapy can provide substantial pain relief in malignancies that respond to this therapeutic modality. Surgery can be effective in relieving pain from intestinal obstruction, pathologic fractures, and obstructive hydrocephalus.

Nonpharmacologic Therapy

Neurostimulatory techniques, such as transcutaneous electrical nerve stimulation, are safe, noninvasive, relatively inexpensive, and easily added to other analgesic approaches.[52] Transcutaneous electrical nerve stimulation could provide short-term benefits in patients with cancer, and a 2- to 4-week trial will often determine its clinical utility. Nonpharmacologic approaches such as progressive muscle relaxation, massage, use of heat or cold, guided imagery, biofeedback, hypnosis, and acupuncture are useful adjuncts to pain management.[53] Although psychotherapy is indicated for an associated depression, unrelieved pain can result in depression that is best treated with analgesic therapies.

Invasive Therapy

Although most cancer pain can be well controlled using the techniques listed in the foregoing discussions, some pain remains refractory, and some patients have persistent adverse effects from opioids despite aggressive therapy with psychostimulants, antiemetics, and laxatives. The side effects can be severe enough that patients might refuse to take sufficient medication to relieve their pain. Adding adjuvant medications, changing to another opioid, or using continuous intravenous or subcutaneous infusions to reduce “peak” levels might be helpful. In selected patients, regional analgesia or neuroablative procedures might permit the doses of pharmacologic agents to be reduced substantially. In particular, spinal metastases may be treated with minimally invasive therapies such as vertebroplasty, kyphoplasty, or radiofrequency ablation.[54] Neurosurgery has been shown to be the first-line therapy over radiation therapy in the management of a solitary spinal metastasis.[55] These invasive approaches should be considered under the following conditions:



If significant pain persists at doses of analgesics that are associated with intolerable side effects



If excessive toxicities result from opioid analgesics



If a careful assessment suggests that a low-risk procedure is likely to result in excellent analgesia

Regional Analgesia

Regional analgesia can be achieved with long-acting local anesthetics that provide pain relief for 3 to 12 hours, neurolytic agents (alcohol or phenol) that produce analgesia for weeks to months, or opioids injected into the epidural or subarachnoid space ( Table 36-10 ). Diagnostic blocks with local anesthetics are usually performed before neurolysis. This permits the anesthesiologist to determine the response to local therapy and allows the patient to decide whether the “numbness” that replaces the pain is tolerable. If the pain can be relieved temporarily with local anesthetics, alcohol or phenol can be injected into the subarachnoid or epidural space to destroy nociceptive fibers in the dorsal rootlets, thus simulating a surgical rhizotomy. Although injections of these neurolytic agents are commonly called “permanent blocks,” pain relief usually lasts several months. Neurolytic blocks can be particularly useful in the thoracic region, where they are associated with few motor complications. In the cervical and lumbar regions, nearly 20% of patients develop motor and/or sphincter dysfunction, which can be permanent. In patients with pre-existing lower extremity paralysis, colostomy, or nephrostomy tubes—cases in which loss of motor or sphincter function might be less critical—lumbar neurolysis might be worthwhile. Other potential side effects of these procedures include hypotension, toxic reactions from accidental intravenous or subarachnoid administration, or pneumothorax after needle placement. Neurolysis is usually restricted to patients with a limited life expectancy, because it can produce a painful neuritis that becomes clinically apparent only months after the procedure.

Table 36-10   -- Regional Anesthetic Techniques


Types of Blocks




Local anesthetic blocks


Intercostal nerve block

Determine etiology of pain and the response and side effects following local therapies.

Analgesic effect will last only hours.


Treatment of sympathetically maintained pain

Stellate ganglion block

Sympathetically maintained pain

Repeated blocks might be needed.


Trigger point injections

Trigger point injection

Myofascial pain syndrome

Repeated blocks might be needed.

Neurolytic (alcohol or phenol) blocks


Intercostal nerve blocks

Chest wall tumor

Pain relief usually lasts several months.



Celiac plexus block

Pancreatic cancer

Pain relief usually lasts several months.




Pain localized to two or three dermatomes

Pain relief usually lasts several months



Intrathecal neurolysis



From Grossman SA, Staats PS: The current management of pain in patients with cancer. Oncology 1994;8:93, with permission.




Neurolytic blocks are used in selected patients who have localized or regional pain. Percutaneous celiac plexus neurolysis is an outpatient procedure associated with few risks; it alleviates pain originating in the pancreas, stomach, gallbladder, or other upper abdominal viscera in most patients. One randomized, prospective, double-blind trial comparing celiac plexus injections of alcohol or saline and systemic analgesic therapy in patients with unresectable pancreatic cancer showed greater pain relief with the block. Opioid consumption and quality of life were not statistically significant, however.[56] In other settings, it has been shown to decrease opioid requirements. [57] [58] Although pain can recur months after a celiac block, subsequent blocks are often associated with excellent pain relief. Less commonly used neurolytic procedures include intercostal blocks (chest wall or rib pain), neuroaxial blocks (pain in two to three dermatomes), Gasserian ganglion neurolysis (pain in the anterior two-thirds of the head), brachial plexus blocks (for patients with pre-existing limb paralysis), and ganglion impar and superior hypogastric blocks (for lower abdominal and pelvic pain).

Neuroablative Procedures

Neuroablative procedures are performed infrequently on patients with cancer because of the success of more conservative approaches. The most commonly performed procedures are radiofrequency ablation and the open unilateral anterolateral cordotomy, percutaneous cordotomy, and commissural myelotomy. An open cordotomy is usually performed through a T2 or T3 laminectomy and produces excellent pain relief in the lower part of the body in 80% of patients. A 5% to 10% mortality rate and significant morbidity in an additional 15% of patients are reported for this procedure. Hemiparesis, urinary retention, sexual impotence, unmasking pain on the opposite side of the body, and late sensory abnormalities are not infrequent. Bilateral cordotomies are associated with higher complication rates. Percutaneous cordotomy is safer and provides excellent pain relief; however, pain recurs within 3 months in 50% of patients. A commissural myelotomy can be considered in selected patients who experience bilateral pelvic and perineal pain. This involves a laminectomy and surgical division of the crossing fibers of the spinal cord. Although it can result in pain relief with sphincter sparing, few neurosurgeons have extensive expertise with this procedure.


Difficult-to-manage pain problems are most common in patients with any of the following conditions:



Pain of neuropathic origin



Episodic or incidental pain



Impaired cognitive or communicative skills



A history of substance abuse

Referral to an experienced multidisciplinary cancer pain team can be helpful if initial attempts to control pain in patients with these underlying problems are unsuccessful.

Patients with Pain of Neuropathic Origin

Any injury to the peripheral or central nervous system can cause neuropathic pain. This is often characterized by paroxysms of shocklike pain on top of a burning or constricting sensation. Neuropathic pain in patients with cancer commonly arises from tumor invading or compressing peripheral nerve, nerve plexus, or spinal cord. It can occur as a result of surgery, radiation, or chemotherapy as exemplified by postmastectomy and postthoracotomy syndromes, radiation-induced plexopathies, and chemotherapy-induced neuropathies. [59] [60] Neuropathic pain also can accompany disorders that are unrelated to the tumor or its treatment, such as diabetes mellitus, nerve entrapment syndromes, and herpes zoster.

Providing adequate relief from neuropathic pain can be difficult even for the most experienced physicians. Although this pain might improve on opioids, it seems to respond less well to these agents than does nociceptive pain. Optimal therapy for neuropathic pain often depends on opioids used in combination with a variety of nonopioid “adjuvant” analgesics (see Table 36-9 ).[61] Tricyclic antidepressants have been studied most extensively in this situation. Although the most convincing efficacy data is with amitriptyline, this agent is associated with significant anticholinergic effects and sedation. Other drugs in this class with a more favorable toxicity profile include desipramine and nortriptyline. Duloxetine, a serotonin and dual serotonin-norepinephrine reuptake inhibitor, has substantial effect in neuropathic pain.[62]

Anticonvulsants are also helpful in the management of neuropathic pain, particularly if the pain has lancinating qualities. The doses of these agents are similar to those used for the control of seizures. Care must be taken to avoid abrupt withdrawal, because this could induce seizures. Randomized controlled trials have demonstrated the efficacy and tolerability of gabapentin for the treatment of postherpetic neuralgia and painful diabetic neuropathy.[63] Other studies suggest that it might help in the management of neuropathic pain secondary to cancer or its treatment.[64] This agent is well tolerated, no drug-drug interactions have been identified, and the average effective dose is 1800 to 3600 mg/day. A newer anticonvulsant, pregabalin, has a mechanism of action and side effects similar to gabapentin. Maximum doses are achieved within 1 to 2 weeks, offering an advantage over gabapentin.[65] Reports support the use of carbamazepine, valproic acid, and diphenylhydantoin, although the myelosuppression associated with carbamazepine might compromise the ability to administer concurrent chemotherapy.

Systemically administered local anesthetics have been used for the treatment of neuropathic pain. Anesthetic creams that produce few systemic side effects are also available. Topical lidocaine patches are an effective means to contol neuropathic pain that is confined to a small area. Capsaicin, a neurotoxin that selectively destroys nociceptors, is also manufactured as a topical preparation and provides relief in some patients. If oral agents and topical creams are ineffective, afferent input can be reduced with transcutaneous electrical nerve stimulation or regional anesthetic techniques such as long-term epidural catheters or intrathecal pumps for the delivery of local anesthetics. Neurolytic blocks, more invasive neurostimulatory techniques, or even neurosurgical procedures might be indicated in extreme situations.

Patients with Episodic or Incidental Pain

It is widely recognized that some patients experience transient, but severe, exacerbations of their pain. These can be difficult to treat and are often very troubling to the patient. These exacerbations can occur as a result of an inadequate analgesic regimen. For example, a patient who receives opioids every 6 hours and has good relief for only 4 hours needs a change in regimen to ensure that opioid levels will not fall below the analgesic threshold after 4 hours. This change is best accomplished by providing the agent more frequently or by administering sufficient doses of a sustained-release preparation. Episodic pain associated with voluntary or involuntary movements poses a more difficult therapeutic problem. Examples of these “incidental pains” are seen in patients with pelvic metastases or pathologic fractures who have severe pain with walking or sitting, patients with rib metastases who experience stabbing chest pain with movement or coughing, or patients with esophageal, rectal, or bladder lesions with pain on swallowing, defecation, or urination, respectively. Involuntary precipitants can include bowel or ureteral distention. In a recent study of incident pain, nearly three-quarters was directly related to neoplastic lesions, 20% resulted from of antineoplastic therapy, and the remainder was unrelated to the tumor or its treatment. [66] [67] [68]

Proper management of these patients requires a comprehensive assessment to determine the origin of the pain. Therapy directed at the underlying etiologic factors is most likely to provide pain relief. Relieving a bowel obstruction, repairing or splinting a fracture, treating a local metastatic lesion with radiation therapy, or performing a neurolytic block for a painful rib lesion provides better analgesia than opioids. The frequency and severity of incidental pain might also be reduced significantly by anti-inflammatory agents or corticosteroids in bone or nerve compression pain and by anticonvulsants or tricyclic antidepressants in neuropathic pain. In addition, agents that reduce the frequency of precipitating events should be used. These include antitussives, laxatives, antiperistaltic drugs, or agents that reduce muscle spasms. Physiotherapy can be useful in musculoskeletal complications, and the cognitive and psychologic approaches can be helpful to patients with these pains. Rarely, patients require invasive anesthetic or neurosurgical approaches for relief of these transient but severe pains. Local anesthetic injections might predict whether a patient is likely to respond to chemical neurolysis or a destructive neurosurgical procedure. Continuous epidural anesthetics and opioids might also be helpful in carefully selected patients.

Many of the approaches listed in the foregoing discussion might not be effective, or even possible or advisable, in the context of a patient's illness. In such situations, opioids remain the mainstay of therapy. The baseline dose of opioid can be escalated until pain relief or intolerable side effects occur. Although this approach might produce relief, patients are often excessively sedated during the intervals between the severe pains. Alternatively, patients might elect to take supplemental analgesics (usually short-acting opioids) 30 to 60 minutes before they know a precipitating event is likely to occur. If the pain is unpredictable, the additional medications are taken as soon as the pain begins. The optimal timing of opioids for episodic pain is difficult to achieve. Episodic pain is characterized by a quick onset and short duration. Parenteral opioids, while having a faster onset of action than the oral route, are not practical in the home setting. Alternatively, the transmucosal fentanyl products may provide the best option. The doses of these supplemental opioids must be determined from the patient's baseline opioid requirements. It is common to begin with 5% to 10% of the total daily opioid dose ordered every 2 to 3 hours as needed. The dosing of transmucosal formulations, as previously mentioned, is determined through titration and not based on the baseline opioid requirement. [45] [69]

Patients with Impaired Cognitive or Communicative Function

The difficulties that physicians and nurses have determining the intensity of pain in patients with cancer have been described previously in this chapter. Problems in conveying pain intensity are greatly magnified in patients who cannot communicate with their health care providers or who are cognitively impaired. These deficits complicate the assessment of both pain intensity and pain relief. Some patients are unable to speak the language of the health care provider, whereas others have severe neurologic deficits, such as an expressive aphasia. As previously noted, children and the elderly have special difficulty communicating pain intensity.[70] Patients with severe cognitive deficits present obvious problems in assessing an entirely subjective symptom. Delirious patients with cancer are often restless, moaning, and unable to convey the intensity, nature, or even location of their pain. These patients require a review of correctable factors contributing to the delirium. Neurologic events, infections, trauma, bladder distention, fecal impaction, hypoxia, or metabolic abnormalities are common. The patient's drug regimen should be simplified, and all agents with anticholinergic properties should be discontinued. If the patient is on an opioid, reducing the dose, switching agents, or using a continuous infusion or sustained-release preparation to avoid wide fluctuations in drug levels might result in improvement.[71]

Patients with a History of Substance Abuse

The principles of cancer pain assessment and management in patients with a current or prior history of drug abuse are similar to those for any patient with cancer pain.[15] These individuals should not remain in pain as a result of this complicating medical problem. Patients with a history of drug abuse, however, often have difficulty finding physicians who believe their reports of pain and who will provide the high doses of analgesics required in these opioid-tolerant individuals. As a result, they can become angry, frustrated, and more persistent in their demands for opioids. This constellation of symptoms is also seen in patients who do not have a history of drug abuse but who have severe, untreated pain. Their preoccupation with obtaining analgesics is referred to as pseudo-addiction and tends to disappear rapidly with appropriate pain therapy.

A frank discussion of major issues relating to the proper use of opioid analgesics for pain management with the patient, the patient's family, and the drug counselor is important. Pain management agreements often help to ensure that all parties understand and agree to the same principles and plans for therapy. Oral agents are preferred, or regional pain management techniques that could limit the need for opioids can be considered. Opioids such as methadone might be prescribed in lieu of morphine or hydromorphone.


Pain is common in patients with cancer and remains one of the most feared aspects of this illness, despite the excellent therapies that are available to provide pain relief. Cancer pain commonly results from tumor compressing or invading soft tissue, bone, or nerves or from diagnostic or therapeutic endeavors. The key to optimal pain management rests with a thorough assessment of the patient's pain. This involves a determination of pain intensity, an evaluation of the etiology of the pain, a carefully considered therapeutic plan, and repeated assessments of pain relief after therapeutic interventions. The vast majority of cancer pain can be well controlled with therapies readily available to most physicians. These include nonopioid analgesics, opioid analgesics, adjuvant medications, antineoplastic therapies, nonpharmacologic approaches, and noninvasive neurostimulatory techniques. Regional anesthetic or neurosurgical approaches should be considered for selected patients who continue to experience pain after an adequate trial of the foregoing therapies or who have unrelenting toxicities from these agents. They should also be considered for patients whose pain suggests that an inexpensive, low-risk procedure is likely to result in excellent analgesia. Examples of such cases include pancreatic cancer pain and thoracic pain in a dermatomal distribution. Referral to an experienced multidisciplinary pain team might be required in situations that are known to pose special challenges in pain management. These include patients with neuropathic pain, episodic or incident pain, impaired cognitive or communicative capabilities, or a history of substance abuse.

Cancer pain remains undertreated despite evidence that a careful assessment of cancer pain and the appropriate use of available therapies should result in excellent relief in nearly 95% of patients. Providing optimal cancer pain relief tests the skills and commitment of physicians, nurses, and pharmacists, because the diagnostic, therapeutic, and social issues in these patients are complex and constantly changing. Meeting these challenges can be satisfying, because patients and family are grateful to find that pain can usually be alleviated. Furthermore, as noted in the American Society of Clinical Oncology's policy statement on cancer pain, “patients with cancer have a right to effective treatment of pain” and the “evaluation and treatment of cancer pain are an integral part” of each caregiver's responsibilities.[7]


  1. van den Beuken-van Everdingen MH, de Rijke JM, Kessels AG, et al: Prevalence of pain in patients with cancer: a systematic review of the past 40 years.  Ann Oncol2007; 18:1437-1449.
  2. Portnow J, Lim C, Grossman SA: Assessment of pain caused by invasive procedures in cancer patients.  J National Comprehensive Cancer Network2003; 3:435-439.
  3. Fitzgibbon DR, Chapman CR: Cancer pain: assessment and diagnosis. In Loeser JD, et al (eds): Bonica's Management of Pain, 3rd ed. 2001.
  4. Gilson AM, Ryan KM, Joranson DE, Dahl JL: A reassessment of trends in the medical use and abuse of opioid analgesics and implications for diversion control: 1997–2002.  J Pain Symptom Manage2004; 28:176-188.
  5. Pain & Policy Studies Group : Improving Cancer Pain Relief in the World: Report for 2003,  Madison, WI, University of Wisconsin Pain & Policy Studies Group/WHO Collaborating Center for Policy and Communications in Cancer Care, 2004.
  6. Pain & Policy Studies Group : Achieving Balance in Federal and State Pain Policy: A Guide to Evaluation,  3rd ed.. Madison, WI, University of Wisconsin Paul P Carbone Comprehensive Cancer Center, 2006.
  7. Ad Hoc Committee on Cancer Pain : Cancer pain assessment and treatment curriculum guidelines.  J Clin Oncol1992; 10:1976-1982.
  8. Gordon DB, Dahl JL, Miaskowski C, et al: American Pain Society recommendations for improving the quality of acute and cancer pain management.  Arch Intern Med2005; 165:1574-1580.
  9. Miaskowski C, Cleary J, Burney R, et al: Guideline for the management of cancer pain in adults and children. APS Clinical Practice Guidelines Series 3, 2005.
  10. Jacox A, Carr DB, Payne R, et al: Management of cancer pain: clinical practice guideline no. 9. AHCPR Publication No. 94-0592,  Rockville, MD, Agency for Health Care Policy and Research, U.S. Dept of Health and Human Services, Public Health Service, 1994.
  11. Panchal SJ: Adult Cancer Pain V.1.2006. National Comprehensive Cancer Network guidelines.  Available at <http://www.nccn.org>
  12. Curtiss CP: JCAHO: meeting the standards for pain management.  Orthop Nurs2001; 20:27-30.
  13. Grossman SA, Sheidler VR, Swedeen K, et al: Correlation of patient and caregiver ratings of cancer pain.  J Pain Symptom Manage1991; 6:53-57.
  14. Cleary JF, Carbone PP: Palliative medicine in the elderly.  Cancer1997; 80:1335-1347.
  15. Passik SD, Kirsh KL: Opioid therapy in patients with a history of substance abuse.  CNS Drugs2004; 18:13-25.
  16. Mortimer JE, Bartlett NJ: Assessment of knowledge about cancer pain management by physicians in training.  J Pain Symptom Manage1997; 14:21-28.
  17. Grossman SA, Nesbit S: Hopkins Opioid Program. Available at <http://www.hopkinsopioidprogram.org> and http://www.hopweb.org.
  18. Hoffman DE, Tarzian AJ: Achieving the right balance in oversight of physician opioid prescribing for pain: the role of state medical boards.  J Law Med Ethics2003; 31:21-40.
  19. Cohen MZ, Easley MK, Ellis C, et al: Cancer pain management and the JCAHO's pain standards: an institutional challenge.  J Pain Symptom Manage2003; 25:519-527.
  20. 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. 15–17 July 2002.  J Natl Cancer Inst2003; 95:1110-1117.
  21. Chibnal JT: Pain assessment in cognitively impaired and unimpaired older adults: a comparison of four scales.  Pain2001; 92:173-186.
  22. Melzack R: The McGill pain questionnaire: major properties and scoring methods.  Pain1975; 1:277-299.
  23. Graham C, Bond SS, Gerkovich MM, Cook MR: Use of the McGill pain questionnaire in the assessment of cancer pain: replicability and consistency.  Pain1980; 8:377-387.
  24. Kremer EF, Atkinson JH, Ignelzi RJ: Pain measurement: affective dimensional measure of the McGill pain questionnaire with a cancer pain population.  Pain1982; 12:153-163.
  25. Fishman B, Pasternak S, Wallenstein S, et al: The Memorial pain assessment card: a valid instrument for the evaluation of cancer pain.  Cancer1987; 60:1151-1158.
  26. Grossman SA, Sheidler VR, McGuire DB, et al: A comparison of the Hopkins Pain Rating Instrument with standard visual analogue and verbal descriptor scales in patients with cancer pain.  J Pain Symptom Manage1992; 7:196-203.
  27. Rhodes DJ, Koshy R, Sheidler VR, et al: Feasibility of quantitative pain assessment in outpatient oncology practice.  J Clin Oncol2001; 19:501-508.
  28. 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.  Pain2001; 93:247-257.
  29. Ferreira KASL, Kimura M, Teixeira MJ: The WHO analgesic ladder for cancer pain control, twenty years of use. How much pain relief does one get from using it?.  Support Care Cancer2006; 14:1086-1093.Published online 8 June 2006.
  30. Meldrum M: The ladder and the clock: cancer pain and public policy at the end of the twentieth century.  J Pain Symptom Manage2005; 29:41-54.
  31. Ripamonti C, Dickerson ED: Strategies for the treatment of cancer pain in the new millennium.  Drugs2001; 61:955-977.
  32. McNicol E, Strassels S, Goudas L, et al: Nonsteroidal anti-inflammatory drugs, alone or combined with opioids, for cancer pain: a systematic review.  J Clin Oncol2004; 22:1975-1992.
  33. MacPherson RD: The pharmacological basis of contemporary pain management.  Pharmacol Ther2000; 88:163-185.
  34. Helin-Salmivaraara A, Virtanen A, Vesalainen R, et al: NSAID use and the risk of hospitalization for first myocardial infarction in the general population: a nationwide case-control study from Finland.  Eur Heart J2006; 27:1657-1663.
  35. Mukherjee D, Nissen SE, Topol EJ: Risk of cardiovascular events associated with selective COX-2 inhibitors.  JAMA2001; 286:954-959.
  36. Rouff G, Lema M: Strategies in pain management: new and potential indications for COX-2 specific inhibitors!.  J Pain Symptom Manage2003; 25S:S21-S31.
  37. McNicol E, Horowicz-Mehler N, Fisk RA, et al: Management of opioid side effects in cancer-related and chronic noncancer pain: a systematic review.  J Pain2003; 4:231-256.
  38. Jaffe JH, Martin WR: Opioid analgesics and antagonists.   In: Gilman AG, Rall TW, Nies AL, et al ed. Goodman and Gilman's The Pharmacological Basis of Therapeutics,  8th ed.. New York: Pergamon Press; 1990:485-521.
  39. Ashburn MA, Lipman AG, Carr D, Rubingh C: Principles of Analgesic Use in the Treatment of Acute Pain and Cancer Pain,  5th ed.. Glenview, IL, American Pain Society, 2003.
  40. Lussier D, Huskey AGB, Portenoy RK: Adjuvant analgesics in cancer pain management.  Oncologist2004; 9:571-591.
  41. Major PP, Lipton A, Berenson J, Hortobagyi G: Oral bisphosphonates: a review of clinical use in patients with bone metastases.  Cancer2000; 88:6-14.
  42. Wong R, Wiffen PJ: Bisphosphonates for the relief of pain secondary to bone metastases.  Cochrane Database Syst Rev2002;CD002068
  43. Ripamonti C, Fagnoni E, Campa T, et al: Is the use of transdermal fentanyl inappropriate according to the WHO guidelines and the EAPC recommendations? A study of cancer patients in Italy.  Supportive Care Cancer2006; 14:400-407.
  44. Ahmedzai S, Brooks D: Transdermal fentanyl versus sustained-release oral morphine in cancer pain: preference, efficacy, and quality of life. The TTS-Fentanyl Comparative Trial Group.  J Pain Symptom Manage2006; 13:254-261.
  45. Zeppetella G, Ribeiro MDC: Opioids for the management of breakthrough (episodic) pain in cancer patients.  Cochrane Database Syst Rev2006Jan; 25(1):CD004311.
  46. Portenoy RK, Taylor D, Messina J, Tremmel L: A randomized, placebo-controlled study of fentanyl buccal tablet for breakthrough pain in opioid-treated patients with cancer.  Clin J Pain2006; 22:805-811.
  47. Ballantyne JC, Carwood CM: Comparative efficacy of epidural, subarachnoid, and intracerebroventricular opioids in patients with pain due to cancer.  Cochrane Database Syst Rev2005;CD005178.doi: 10.1002/14651858.CD005178.
  48. Staats PS, Yearwood T, Charapata SG, et al: Intrathecal ziconotide in the treatment of refractory pain in patients with cancer or aids.  JAMA2004; 291:63-70.
  49. Smith TJ, Staats PS, Deer T, et al: Randomized clinical trial of an implantable drug delivery system compared with comprehensive medical management for refractory cancer pain: impact on pain, drug-related toxicity, and survival.  J Clin Oncol2002; 20:4040-4049.
  50. Hassenbusch SJ, Portenoy RK, Cousins M, et al: Polyanalgesic Consensus Conference 2003: an update on the management of pain by intraspinal drug delivery—report of an expert panel.  J Pain Symptom Manage2004; 27:540-563.
  51. Silberstein EB, Eugene L, Saenger SR: Painful osteoblastic metastases: the role of nuclear medicine.  Oncology2001; 15:157-163.
  52. Grond S, Radbruch L, Meuser T, et al: Assessment and treatment of neuropathic cancer pain following WHO guidelines.  Pain1999; 79:15-20.
  53. Cohen AJ, Menter A, Hale L: Acupuncture: role in comprehensive cancer care—a primer for the oncologist and review of the literature.  Integr Cancer Ther2005; 4:131-143.
  54. Halpin RJ, Bendok BR, Liu JC: Minimally invasive treatments for spinal metastases: vertebroplasty, kyphoplasty, and radiofrequency ablation.  J Support Oncol2004; 2:339-351.
  55. Patchell RA, Tibbs PA, Regine RF, et al: Direct decompressive surgical resection in the treatment of spinal cord compression caused by metastatic cancer: a randomised trial.  Lancet2005; 366:643-648.
  56. Wong GY, Schroeder DR, Carns PE, et al: Effect of neurolytic celiac plexus block on pain relief, quality of life, and survival in patients with unresectable pancreatic cancer.  JAMA2004; 291:1092-1099.
  57. Lillemoe KD, Cameron JL, Kaufman HS, et al: Chemical splanchnicectomy in patients with unresectable pancreatic cancer.  Ann Surg1993; 217:447-455.
  58. Mercadante S: Celiac plexus block versus analgesics in pancreatic cancer pain.  Pain1993; 52:187-192.
  59. Martin LA, Hagen NA: Neuropathic pain in cancer patients: mechanisms, syndromes, and clinical controversies.  J Pain Symptom Manage1997; 14:99-117.
  60. Mantyh PW: Cancer pain and its impact on diagnosis, survival and quality of life.  Nature Rev Neurosci2006; 7:797-809.
  61. Dworkin RH, Backonja M, Rowbotham MC, et al: Advances in neuropathic pain. Diagnosis, mechanisms, and treatment recommendations.  Arch Neurol2003; 60:1524-1534.
  62. Goldstein DJ, Lu Y, Detke MJ, et al: Duloxetine vs. placebo in patients with painful diabetic neuropathy.  Pain2005; 116:109-118.
  63. Wiffen PJ, McQuay HJ, Edwards JE, Moore RA: Gabapentin for acute and chronic pain.  Cochrane Database Syst Rev2005;CD005452. doi: 10.1002/14651858.CD005452.
  64. Bosnjak S, Jelic S, Susnjar S, Luki V: Gabapentin for relief of neuropathic pain related to anticancer treatment: a preliminary study.  J Chemother2002; 14:214-219.
  65. Gilron I, Flatters SJ: Gabapentin and pregabalin for the treatment of neuropathic pain: a review of laboratory and clinical evidence.  Pain Res Manag2006; 11(Suppl A):16A-29A.
  66. Zeppetella G, Ribeiro MD: Pharmacotherapy of cancer-related episodic pain.  Expert Opin Pharmacother2003; 4:493-502.
  67. Payne R: Recognition and diagnosis of breakthrough pain.  Pain Med2007; 8(Suppl 1):S3-S7.
  68. Svendsen KB, Andersen S, Arnason S, et al: Breakthrough pain in malignant and non-malignant diseases: a review of prevalence, characteristics and mechanisms.  Eur J Pain2005; 9:195-206.
  69. Mercadante S, Radbruch L, Caraceni A, et al: Episodic (breakthrough) pain.  Cancer2002; 94:832-839.
  70. Balducci L: Management of cancer pain in geriatric patients.  J Support Oncol2003; 1:175-191.
  71. Herr KA, Garand L: Assessment and measurement of pain in older adults.  Clin Geriatr Med2001; 17:457-478.