30 Opioid Therapy in Chronic Nonmalignant Pain
The Massachusetts General Hospital Handbook of Pain Management
Opioid Therapy in Chronic Nonmalignant Pain
Scott M. Fishman and Jianren Mao
Thou only givest these gifts to man, and thou hast the keys of Paradise, O just, subtle and mighty opium!
—Thomas De Quincey (1785–1859)
II. Treatment endpoints
III. Side effects
2. Nausea and vomiting
5. Respiratory depression
IV. Opioid tolerance, physical dependence, and addiction
2. Physical dependence and withdrawal
3. Opioid addiction and pseudoaddiction
Chronic nonmalignant pain (CNMP) refers to a broad spectrum of chronic pain states associated with a variety of disease processes. CNMP can be categorized as follows:
Chronic nociceptive pain (e.g., arthritic pain)
Neuropathic pain [e.g., complex regional pain syndrome (CRPS), postherpetic neuralgia]
Pain with mixed nociceptive and neuropathic components (e.g., low back pain with radiculopathy)
Pain with an unknown etiology, termed idiopathic pain (e.g., fibromyalgia)
Opioids are still the gold standard of the currently available analgesics, yet their use in CNMP is controversial. The decision to initiate opioid therapy in patients with CNMP should be patient specific and based on the rationale for treatment, observable treatment endpoints, and the potential for side effects and addiction. Each of these considerations is discussed in this chapter. The general pharmacology and dosing schedules for opioid therapy as well as issues of opioid addiction, abuse, and adherence are discussed in Chapter 9 and Chapter 35. In general, long-acting opioids are preferred. Short-acting opioids are avoided so that patients can learn to utilize other means of controlling their pain when they experience breakthrough pain.
Whether or not long-term opioid treatment is beneficial for patients with CNMP is debated, with a scarcity of convincing scientific data to persuasively argue for either side. Since the endpoint of analgesic therapy is often difficult to quantitatively monitor, managing chronic opioid therapy can be a formidable undertaking, fraught with the challenges of measuring the positive and negative impact of treatment on quality of life. Patients who use opioids for CNMP range from those using consistent amounts of opioid with little variation to those whose needs are seemingly never satisfied and whose dosages frequently escalate. Resistance to chronic opioid therapy is often founded on social, medical, and legal stigmata. Concerns in this patient population include questions of efficacy, adverse side effects, toxicity, addiction and abuse potential, tolerance and physical dependence, fear of regulatory scrutiny, possible neuropsychiatric deterioration, reinforcement of pain behavior, and even possible immunosuppression.
Opioid therapy can be an integrated part of a multidisciplinary approach for CNMP management, which includes any possible combination of interventions with many other analgesic drug groups, interventional procedures, and other modalities of psychological and physical rehabilitation. Much of the debate concerning the role of opioid therapy in CNMP management has been centered on the issue of whether opioid therapy should be used as a first-line treatment. Should pain physicians withhold opioid therapy until other treatment options are exhausted? Although there is a lack of consensus on this important issue, opioid therapy tends to be used as a second-line treatment for CNMP for the following reasons:
CNMP may respond to nonopioid pharmacologic interventions such as nonsteroidal anti-inflammatory drugs (NSAIDs) for arthritic pain and anticonvulsants or tricyclic antidepressants for postherpetic neuralgia.
At times, interventional analgesia can be more effective than chronic drug management; for example, sympathetic nerve blockade may provide better therapeutic outcomes than opioids in certain types of CNMP (e.g., CRPS types I and II, postherpetic neuralgia).
Considering the significant side effects and liability profiles of opioid treatment (see later), the assessment of risks versus benefits often suggests that some patients with CNMP are not good candidates for opioid therapy.
Opioid therapy should not, however, be withheld in patients with CNMP whose pain is not relieved by other interventions. Although the effectiveness of opioid therapy in certain types of CNMP remains controversial (in Chapter 35, see the section “Opioid Responsiveness”), there is no evidence suggesting that opioid therapy is contraindicated in these circumstances. Preclinical studies have shown a right shift of the opioid dose–response curve in neuropathic pain, suggesting that higher opioid doses may be required for patients primarily suffering from neuropathic pain. Thus, a limiting factor for opioid therapy in neuropathic pain treatment may be related to the development of significant side effects associated with the high doses of opioids needed. It is possible that methadone may be an optimal choice as an opioid for neuropathic pain because of its N-methyl-D-aspartate (NMDA) blocking effects.
In summary, opioid therapy may be initiated in patients with CNMP when a thorough evaluation of the risk-to-benefit ratio supports this intervention. This requires a clear sense of purpose; clear, observable endpoints of treatment; and an exhaustive trial of other potentially effective therapies that have less risk. In addition, combination analgesia (e.g., an opioid in combination with an NSAID) can have opioid-sparing effects by taking advantage of nonopioid analgesic actions that have synergistic effects with the opioids, increasing analgesia at lower doses of each, and helping reduce the side effects of each.
II. TREATMENT ENDPOINTS
There are two critical issues related to treatment endpoints in opioid therapy for CNMP:
What should be considered a positive outcome after a trial of opioids?
When should opioid therapy be discontinued (or tapered) if the treatment is either effective or ineffective?
Clinical studies in this area are limited.
Markers of benefit in patients treated with opioids for CNMP include subjective pain reduction and objective evidence of improvement in functional status and quality of life. However, psychological and social factors, as well as the status of coexistent disease, may influence pain perception, suffering, and entitlement, crucially altering the overall assessment. Unfortunately, not all these issues will improve concomitantly and proportionately following the initiation of opioid therapy.
For example, it is well recognized that psychological factors may influence pain perception. Conceivably, the reduction of pain from an opioid trial may not be as robust in a patient who has not adequately resolved other psychological amplifiers of pain perception. Also, pain reduction and improvement of functional status resulting from effective opioid analgesia may not be simultaneous. Thus, other than requiring objective evidence of efficacy, determining treatment endpoints during an opioid trial usually requires some flexibility in considering the many possible variations in functional gain and improvement in quality of life. Pain reduction is always subjective and, as such, can only serve as a single aspect of adequate chronic opioid therapy.
Consider the patient who has daily pain rated 6 on a pain severity scale of 1 to 10, with significant disability associated with the pain. Although opioid therapy may reduce subjective pain scores by only one point (from 6 to 5), the treatment is clearly successful if there is evidence of increased function such as return to work and improved ability to participate in physical rehabilitation. On the other hand, an opioid trial characterized by subjective reports of marked pain relief but no observable functional gains, and possibly even signs of functional loss such as sedation, loss of job, dysfunctional interpersonal relationships, or diminished physical activity, suggests that the improvement in subjective analgesia has not improved the quality of life and may have worsened it overall.
When function worsens, it is imperative to assess the possible contribution of opioid side effects, including addiction. Signs of dysfunction always warrant consideration of, but do not confirm, possible addiction (see later). It is only with careful assessment of objective treatment endpoints that side effects can be recognized and managed, allowing chronic opioid therapy to be safe, with minimal adverse effects.
Another critical issue is when to discontinue opioid therapy if the treatment is deemed unsatisfactory. Determination of treatment failure requires consideration of many possible contributing factors, including the following:
Inappropriate dosing schedule
Improper drug delivery route
Opioid-insensitive pain relating to the nature of the pain generator (e.g., neuropathic pain)
Involvement of unresolved contributors to pain, such as physical, psychological and social disability
The development of significant side effects limiting the dose escalation
Some patients appear resistant to one opioid and sensitive to another; a possible solution, therefore, is to try a different one.
How long effective opioid therapy should be continued remains a question with little science to guide decision making and no clear consensus among practitioners. Pharmacologic tolerance to opioids can develop during treatment (see later) and may necessitate an increase in dose to maintain the same therapeutic effect. Although some clinical studies have suggested a plateau of opioid dose requirement following an initial escalation, it is possible that progressive dose escalation may be required during prolonged opioid treatment. This implies that periodic opioid dose escalation would be expected even in patients experiencing effective pain relief. Clearly, decisions regarding the duration of effective opioid therapy should be made on the basis of each patient’s need, with full consideration of treatment efficacy relative to adverse effects as well as to the progression or regression of underlying pathology. Once opioid therapy is started, it may not be possible to know how much pain would be present without it, unless opioids are discontinued (by slow taper.)
III. SIDE EFFECTS
The most common side effects of opioids are constipation, nausea, vomiting, sedation, pruritus, and respiratory depression. Any adverse effects from opioids may significantly limit therapy and some can present with life-threatening consequences. Unfortunately, there are few ways to predict which patients will experience which side effects, and which particular opioids will produce them. It is sensible to expect side effects and to take preventive action. As not all opioid-related side effects can be prevented, patients should be followed closely with a high level of suspicion. Effective management includes anticipation of adverse effects and preventive measures (such as laxatives for constipation), choosing the best medication with careful administration, and clear communication with the patient, family, or nurse to ensure prompt recognition of and response to adverse effects.
Constipation is the most common side effect of opioids. Whereas tolerance develops to most other side effects, it does not develop to constipation, which can be expected throughout the duration of opioid administration. Preventive therapy with cathartics and adequate fluid intake is the mainstay of therapy and should be offered at the time opioids are started and continued throughout opioid treatment. Stool softeners and bulking agents such as bran or psyllium derivatives alone will be inadequate because opioid-related constipation results from decreased gut motility. Thus, active stimulating laxatives are effective and passive ones are not.
Severe constipation may respond to oral administration of an opioid antagonist such as naloxone. This maneuver exploits the extensive metabolism of naloxone after oral administration that limits its systemic bioavailability. Unfortunately, there is uncertainty about the dosing regimen. It is suggested that an initial oral naloxone dose should not exceed 5 mg. Start with 1.2 mg to 2.4 mg orally (four to six small ampules) every 4 hours until the first bowel movement, or for five doses. If ineffective, another series with a higher dose (3 to 5 mg per dose) may be tried.
Oral naloxone usually works only when the constipation is related solely to opioids. When concurrently used nonopioid drugs contribute to this side effect, the constipation is not reversed by naloxone (e.g., the anticholinergic side effect of tricyclic antidepressants).
Because constipation can be mitigated by direct effects of opioid antagonists on the bowel, opioids that are delivered without direct bowel contact may induce less constipation. Some evidence shows that certain opioid products that are absorbed without contact to the gastrointestinal (GI) tract, such as transdermal fentanyl, may induce less constipation than equivalent oral morphine. In one study of cancer patients, the incidence of constipation was reduced by up to two thirds by switching from oral morphine to transdermal fentanyl.
Drugs that cause diarrhea are usually well balanced with constipating opioids. Misoprostol (Cytotec), a drug that is marketed to protect the gastric mucosa from NSAID toxicity, is commonly associated with diarrhea but is a compelling addition to combination therapy with opioids and NSAIDs.
2. Nausea and vomiting
Although nausea and vomiting is a common early side effect of opioids, severe protracted nausea and vomiting caused solely by opioids is rare. Addition of antiemetics, or reduction of the opioid dose to the minimum that produces acceptable analgesia, is usually effective. A change in the route of administration may also alleviate symptoms. Fortunately, nausea and vomiting often becomes less significant within several days of administering opioids, at which time antiemetic therapy can be discontinued.
Although it remains unclear why one opioid should produce nausea and vomiting in an individual patient while another does not, the change to a different opioid often reduces or eliminates emetic side effects. A history of severe nausea with previous opioid treatment may prompt pretreatment with an antiemetic or avoidance of the previous offending opioid. There is rarely a need to discontinue or avoid opioid treatment, since the nausea and vomiting is usually responsive to dose reduction and/or antiemetic treatment.
Opioid-induced nausea and vomiting is thought to result from activation within the medullary chemoreceptor trigger zone (CTZ), a brainstem area responsible for afferent input to the emetic center. This anatomic area is rich in specific receptors responsive to various neurotransmitters. Effective antiemetic agents include the antihistamines (e.g., hydroxyzine), serotonin antagonists (e.g., ondansetron), dopamine antagonists (e.g., droperidol, haloperidol, and metoclopramide), and anticholinergics (e.g., scopolamine). It is not clear which of these drug classes is most effective for opioidinduced nausea and vomiting, so the choice is usually made on a try-it-and-see basis.
Opioid-related nausea may be related to orthostasis or ambulation, perhaps suggesting vestibular involvement. There are many other causes of nausea and vomiting that can occur concomitantly but which are unrelated to opioids. These include chemotherapy (particularly cisplatin), radiation therapy, metastases (particularly to brain and GI tract), increased intracranial pressure, peptic ulcer disease, esophagitis, gastritis, electrolyte and acid–base imbalance, uremia, liver disease, infection, pregnancy, and fear or anxiety. Since predicting the most effective antiemetic is not always possible, an agent may be chosen for its secondary benefits such as its promotility, sedative, antipruritic, anxiolytic, or antipsychotic properties.
Opioid-related sedation is very common, and it can indicate either excess drug or even delirium. Opioid-induced sedation is usually temporary, resolving over time as the patient accommodates to a new opioid drug or a new dose. In those with significant sedation, the opioid dosage should be reduced to the minimal level required for adequate analgesia.
In some cases, where sedation occurs late in treatment, medication may be accumulating and, if so, it will be necessary to either increase the dose interval or change to a different agent. Also, in such cases, consider other causes of sedation such as other sedating drugs and encephalopathy. For unremitting sedation that limits therapeutic options, stimulants such as dextroamphetamine or caffeine can reduce sedation.
Pruritus is uncommon in association with oral opioid use, somewhat more common in patients treated with intravenous (IV) or intramuscular opioids, and frequent in patients treated with intrathecal and epidural opioids. Such effects often vary with the specific agent and dosage. Parenteral opioid-induced pruritus is usually mild, although it may rarely be moderate to severe. Fortunately, tolerance usually occurs quickly. Opioid-related pruritus is often localized to the face and less often to the perineum, but it can become generalized. The mechanism is not well understood. Suggested hypotheses include mu receptor stimulation, histamine release, local excitation of dorsal horn neurons, and central migration of spinal opioids to the brainstem.
Opioid-induced pruritus may respond to a change in opioid agent. Mini-doses of naloxone are effective for opioid-related pruritus and should not interfere with analgesia (5 µg/kg IV every 10 minutes; repeat as needed, and hold if analgesia decreases). Antihistamines may also be effective. Since nonsedating antihistamines are less effective than sedating antihistamines, the antipruritic efficacy of antihistamine therapy may be related in part to sedation.
5. Respiratory depression
Depressed respiration is one of the most feared opioid side effects. Tolerance to opioid-induced respiratory depression usually develops early in the course of chronic therapy, so with long-term therapy, respiratory depression is rarely a problem. However, depression of respiratory drive may occur more rapidly when oral or intravenous opioids are combined with epidural or intrathecal opioids. Likewise, combining opioids with other sedating drugs can worsen the respiratory depressant effect of the opioid.
Significant acute respiratory depression can be managed with the opiate receptor antagonist naloxone. The dosage of naloxone for treating respiratory depression is 0.04 mg IV (one 0.4-mg ampule diluted in 10 mL) repeated every few minutes until a response is seen. Although naloxone may provide a brisk response, its duration of action is short, often requiring frequent dosing or continuous IV administration. Particular care must be given to the rapid administration of naloxone in a patient who has had prolonged opioid exposure, as this can precipitate an aggressive and possibly dangerous withdrawal. In special cases, particularly in a patient predisposed to pulmonary edema (congestive heart failure, adult respiratory distress syndrome), reversal of opioid actions can promote pulmonary edema.
IV. OPIOID TOLERANCE, PHYSICAL DEPENDENCE, AND ADDICTION
Pharmacologic tolerance and physical dependence are pharmacologic properties of a drug and are not synonymous with addiction. Both can develop following opioid treatment. Sustained analgesia at stable doses is also seen, especially in patients treated with opioids for CNMP. It is not clear why tolerance and physical dependence develop in some individuals and not others, although these phenomena tend to be associated more with the rapidly accelerating and large doses used to treat cancer pain than with those used to treat CNMP. (See Chapter 35, the section “Opioids and Addiction,” for a full account of distinguishing between physical dependence, tolerance, and addiction.)
Tolerance occurs when a fixed dose of opioid produces decreasing analgesia so that a dose increase is required to maintain a stable effect. Just as tolerance develops to analgesic effects, so it may also occur to opioid side effects. Either changing the opioid or increasing the dose can usually compensate decreased analgesic efficacy in a tolerant patient. A patient who has become tolerant to one opioid drug may respond with adequate analgesia to another. Mechanisms underlying this clinical observation are not completely understood; it may relate to the differential opioid receptor profiles of different opioid agonists. Equianalgesic dosages are not applicable in the opioid-tolerant patient. When starting a new opioid agent in a tolerant patient, begin with half the equianalgesic dose and titrate to effective analgesia.
Recent studies have demonstrated the involvement of the NMDA receptor in mechanisms of opioid tolerance. It may be beneficial to combine an NMDA receptor antagonist with an opioid to attenuate tolerance. Such agents include dextromethorphan and ketamine. Appropriate ratios for such combinations for tolerance reduction are yet to be determined. In addition, methadone may be useful in opioid tolerance presumably because of its agonist effects on opioid receptors and its antagonism at NMDA receptors.
2. Physical dependence and withdrawal
Physical dependence applies to many drugs that are and are not addictive (e.g., morphine, clonidine.) Physical dependence relates to the expression of a withdrawal syndrome upon sudden drug cessation. It may reflect a biochemical adaptation from chronic exposure to a drug. Thus, duration of opioid treatment is probably a significant factor contributing to the development of opioid-related physical dependence. Cellular and intracellular mechanisms of opioid-related physical dependence have been proposed, including a rebound increase in cyclic adenosine monophosphate (cAMP), increased release of endogenous NMDA receptor agonists (namely, glutamate and aspartate), and changes in the NMDA receptor property following repeated opioid exposure. Fear of opioid dependence may lead to undertreatment.
Opioid withdrawal is rarely life-threatening and usually occurs with a systematic progression of symptoms. The least severe withdrawal symptoms typically appear earlier than the most severe. Withdrawal begins with increased irritability, restlessness, anxiety, insomnia, yawning, sweating, rhinorrhea, and lacrimation, and it progresses to dilated pupils, gooseflesh, tremor, chills, anorexia, muscle cramps, nausea, vomiting, abdominal pain, agitation, fever, tachycardia, and other features of heightened sympathetic activity. Laboratory data may reveal leukocytosis, ketosis, metabolic acidosis and respiratory alkalosis, and electrolyte imbalance.
The withdrawal syndrome may be seen with discontinuation or antagonism of any opioid. However, sudden discontinuation of shorter-acting opioids such as morphine or hydromorphone is more likely to produce withdrawal symptoms than discontinuation of agents with longer plasma half-lives, such as methadone or transdermally administered fentanyl. Slow, systematic tapering of opioids at a rate of 10% to 15% every 48 to 72 hours can usually prevent withdrawal. Usually a 2- to 3-week period is necessary for completion of the taper. Once the tapering is accomplished, withdrawal may be reversed by reintroducing the opioid at dosages of 25% to 40% of the previous daily dose.
Slow weaning of bedtime doses may help avoid the sleep disturbances that are often associated with opioid cessation. Signs of sympathetic hyperactivity may be treated with sympatholytics. Effective sympatholytics include clonidine (0.2 to 0.4 mg/day) and beta-blockers although these agents can produce hypotension. Clonidine may produce sedation and its anti-withdrawal effects may be antagonized by tricyclic antidepressants.
When abrupt discontinuation of a chronic opioid is necessary, clonidine detoxification can effectively blunt objective findings of sympathetic hyperactivity. It remains controversial whether clonidine or other sympatholytics increase or mask subjective symptoms of withdrawal such as anxiety, insomnia, and restlessness. Such treatment is often begun at dosages of 10 to 20 µg/kg/day in three divided doses, with subsequent adjustments to reduce signs of withdrawal while limiting hypotension. Clonidine may be maintained for between 4 days for short-acting opioids and 14 days for long-acting opioids. Tapering of clonidine can then occur over 4 to 6 days.
3. Opioid addiction and pseudoaddiction
This topic is discussed in greater detail in Chapter 35. Opioid addiction is a disorder characterized by compulsive use of opioids resulting in physical, psychological, and/or social dysfunction to the user and continued use despite that dysfunction. Opioid addiction should be distinguished from pseudoaddiction, a phenomenon that results from undertreatment. Whereas addiction is marked by dysfunction with the use of the drug that stimulates the disease, pseudoaddiction resolves when increasing the drug that the patient seeks, resulting in improved function.
To date, there is no consensus on whether, when, and how opioid therapy should be administered in CNMP patients. Rational and individualized opioid treatment regimens should be formulated according to the general guidelines presented here. Safe opioid therapy requires a program for ongoing close monitoring of functional gains and possible side effects.
American Pain Society. Principles of analgesic use in the treatment of acute pain and chronic cancer pain, 2nd ed. Clin Pharm 1990;9:601–611.
Carr DB. Pain. In: Firestone LL, Lebowitz P, Cook C, eds. Clinical anesthesia procedures of the Massachusetts General Hospital, 3rd ed. Boston: Little Brown, 1988:571–585.
Foley KM. Controversies in cancer pain. Medical perspectives. Cancer 1989;63: 2257–2265.
Fishman SM, Yang J, Reisfield G, et al. Adherence monitoring and drug surveillance with chronic opioid therapy. J Pain Symptom Manage 2000;20:293–307.
Fishman SM, Bandman T, Edwards A, Borsook D. The opioid contract in the management of patients on chronic opioid therapies. J Pain Symptom Manage 1999;18:27–37.
Mao J. NMDA and opioid receptors: Their interactions in antinociception, tolerance and neuroplasticity. Brain Res Rev 1999; 30:289–304.
Mao J, Price DD, Mayer DJ. Mechanisms of hyperalgesia and opioid tolerance: A current view of their possible interactions. Pain 1995;62:259–274.
National Institutes of Health Consensus Development Conference: The integrated approach to the management of pain. J Pain Symptom Manage 1987;2:35–41.
Portenoy RK. Chronic opioid therapy in nonmalignant pain. J Pain Symptom Manage 1990;5:S46–S62.
Wall PD, Melzack R, eds. Textbook of pain, 3rd ed. New York: Churchill Livingstone, 1998.