Addiction, Pain, and Public Health website
Chronic Pain: 2. The Case for Opiates
Opioid medications allow us to treat chronic pain as aggressively as we would any pathogen, but we must first overcome ingrained misconceptions about patients' motivations for seeking treatment and about the addictive properties of the drugs. With controlled use, the newer sustained-release formulations give real hope for safe and sustained pain relief.
Pain must be regarded as a disease...
and the physician's first duty is action--heroic action--to fight disease.
Opioid medications were once withheld from suffering cancer patients because of fear of addiction, exaggerated concern about side effects, or, in some cases, doubt about the morality of treatment. Less than 50 years ago, some medical textbooks discussed the need for patients to experience pain and suffering at the end of life so that they would relate to the agony of Christ and prepare for redemption. Although few physicians still hold these views, many continue to imply that pain should be accepted without complaint, telling their patients that "after all, pain is not going to kill you."
There is growing evidence,
however, that too much pain can cause damage and even death. When pain is
controlled, medications for the underlying disease or disorder tend to work
better. Opioid analgesia is one of the most prolife therapies that we have to
offer patients with cancer pain, and there is no reason to think that patients
with other diseases are any less deserving of relief or that their pain is any
less amenable to treatment. Although there is currently no ideal analgesic for
chronic pain, medications that act on µ-opioid receptors are the closest thing
that we have.
Assessment of Pain
One of the main problems in assessing patients with chronic pain is that the physical examination and laboratory tests often do not provide the information necessary to gauge severity and assess outcomes. Various survey instruments and visual analogue scales that allow precise measurements of pain are available but used only rarely. Pain is generally assessed indirectly, which why it is so important to listen to--and believe--patients when they say that they are in pain.
Some physicians apparently have difficulty with that. Many of my patients with chronic pain have been refused treatment by previous caregivers who apparently believed that their pain was not real. Even after undergoing painful procedures and surgeries that failed to bring relief, some of these patients were labeled as drug-seekers when they continued to ask for help. They had to contend not only with the pain but also with feelings of frustration, isolation, and abandonment by those on whom they had most relied.
In some cases, physicians may be well informed about pain mechanisms but lack an organized approach to the individual assessment of pain. A comprehensive evaluation of patients with chronic pain syndromes can be time-consuming and often requires more data than can be obtained in a few brief clinic visits. I have found the following operational format to be particularly useful, both in gauging the severity of pain and in determining the degree of disability:
1) The patient's perception. Asking the patient to keep a pain diary that includes numerical scales can help to objectify the pain. If it is understood that the physician will review the diary carefully, the patient will not have to act out a month's worth of pain at every appointment. The diary can also be an important aid in identifying exacerbating or ameliorating factors and developing more effective strategies to cope with the pain such as behavioral changes or the preemptive use of analgesics in certain situations.
2) The patient's emotional state and somatic preoccupation. This relates to the degree to which the patient remains focused on bodily symptoms to the exclusion of other issues and often can be best assessed by interviewing a close family member.
3) Functional status at home. The first things that many patients in pain stop doing are usually non-work-related activities such as going out with family and friends, attending church, or engaging in hobbies. Some patients continue to report pain or discomfort even though their condition has improved. By keeping track of daily activities, both patient and physician have some measure of how disabling the pain actually is.
4) Functional status at work. The number of work days missed and the specific work activities curtailed because of pain are also useful indices of pain severity. Since these variables can be expected to change with analgesic treatment, they provide a way to gauge the patient's response to different therapies.
5) Use of analgesic
medications. If the patient is given an adequate supply of effective
short-acting rescue medications and told to take them as needed, the number
consumed can be a measure of pain. It can also be a way to assess whether the
patient is benefiting from other medications or nonpharmacologic treatments. The
physician should make it plain that the other treatments are not designed to get
the patient to stop using the pain medication but to stop needing it.
Setting Goals of Treatment
It is important that the
physician and patient collaborate in developing the goals to guide treatment and
the means to assess progress. In many cases, there is no realistic hope of cure,
and patients must come to terms with the fact that treatment will probably
continue for a long time. At first, the goals may be as simple as sleeping
through the night, but as the patient's condition improves, more ambitious
goals, such as returning to work or participating in recreational activities,
may be attainable. In addition to reviewing the patient's diary and keeping
track of the various functional indicators, the physician must take the time to
discuss the patient's personal goals--what he or she has been missing because of
pain and most wants to be restored.
Treating Suffering as Well as Pain
The ultimate goal in treating chronic pain is for patients to reclaim control of their lives, and, to do that, they must be relieved of suffering as well as pain. Issues such as sadness over lost opportunities, guilt for being a burden to others, and feelings of inadequacy or abandonment contribute to the suffering of many patients with chronic pain and deserve attention. Ensuring that the patient obtains good psychological care is just as important as providing analgesic medications.
Unfortunately, many patients
with chronic pain see referral to a psychologist as an invalidation of the
physical nature of their pain. After years of hearing their disease or disorder
referred to as functional or somatoform, they may need to be convinced that it
is common for chronic pain to have an impact on many aspects of their lives,
including their relationships with family and friends. In referring my patients
for psychological assessment, I encourage them to recognize that psychological
health is a vital aspect of well-being.
Initiation of Opioid Therapy
In the United States, up to 90% of the prescriptions written for opioids are for noncancer pain. The efficacy and safety of these drugs in treating chronic pain syndromes has been demonstrated many times over. Most patients with chronic pain of moderate or greater severity who have not gotten good relief with disease-specific treatments or nonopioid analgesics should at least have a trial of an opioid medication, no matter what the cause of the pain. One of the most important ground rules for such a trial, as well as for subsequent treatment, is that a single physician must take full responsibility for establishing the protocol and writing all prescriptions.
Opiate-naive patients are usually started with a short half-life drug (e.g., hydrocodone, hydromorphone, oxycodone, codeine, or morphine). Because of their rapid clearance, these drugs must be taken every three to four hours. For severe pain, the usual starting dose is 10 to 15 mg of hydrocodone or oxycodone, 2 to 4 mg of hydromorphone, 30 to 60 mg of codeine, or 15 to 30 mg of morphine.
The common strategy in
treating chronic pain with opioid analgesics is to rely on "as-needed"
intermittent dosing, but that does not usually provide sufficient coverage. As a
result, the patient is subjected to periods of anxiety and pain that are not
only unnecessary but also contribute to the patient's distrust of the
Posttrial Therapy: Sustained-Release Opioids
For patients with continuous pain who respond well to opioids, long-acting sustained-release agents are indicated. They can be started once a good response to a short half-life opioid has been obtained and any side effects have been managed. The initial daily dose should be roughly equivalent to the average daily amount of short-acting opioid that provided relief.
Oxycodone is available in an eight- to 12-hr formulation, morphine in both eight- to 12- and 24-hr formulations, and fentanyl in a 72-hr controlled-release skin patch. Hydromorphone formulations allowing once-a-day dosing are expected to be on the market by next year, and several other sustained-release opioids are in various stages of testing.
These drugs have many important advantages over their short half-life counterparts. Because serum levels remain steady, miniwithdrawals or rebound reactions do not occur, as they sometimes do with long-term use of short-acting opioids. Sleep patterns are also more normal; with careful titration, the intermittent sedation that occurs with high-peaking, short-acting drugs is largely avoided.
Furthermore, because dosing is more convenient, compliance increases. By reducing the number of daily pills and eliminating fluctuations in serum levels, sustained-release opioids may help patients to shift their focus away from somatic concerns. Several studies have shown that patients using sustained-release opioids for pain are more satisfied with outcomes and have a measurably higher quality of life.
These drugs also provide
less reinforcement of inappropriate use. Because they are more difficult to
convert into injectable forms, they have a lower potential for diversion and a
lower street price than short-acting opioids. The vast majority of prescription
opioid abuse or drug diversion cases investigated by the Drug Enforcement
Administration, state medical boards, and other regulatory or law enforcement
agencies involve short-acting opioids--sustained-release opioids are rarely a
problem. A recent study by D.E. Joranson and colleagues showed that with
increased use of sustained-release opioids, medical emergencies related to abuse
of prescription opioids have decreased.
A careful review of the patient's medical history is the first essential in choosing a medication. Patients with chronic pain can often tell what has worked in the past and what has not. In one study of cancer patients using opioid analgesics, 44% required trials of two or more opioids, and 20% required three or more, before achieving satisfactory pain relief without intolerable side effects. As in selecting any drug, both adverse and beneficial effects must be evaluated in the context of the individual patient's current physical condition.
Morphine. Of the sustained-release opioids, morphine is considered by many to be the gold standard, probably because it was the first sustained-release drug that allowed adequate dosing. The designation is more historic than clinically significant, however. In comparison to other opioids, morphine generally causes more nausea and pruritus. It also has the disadvantage of being relatively hydrophilic, which delays its transport across the blood-brain barrier. In addition, morphine has several active metabolites whose levels can vary with renal or hepatic function. The drug apparently induces its own metabolism, which makes it difficult to maintain a steady serum level.
Fentanyl. Because of these problems, fentanyl, a derivative of meperidine, was specifically designed to substitute for morphine in operative anesthesia. It is lipophilic and therefore faster acting. Of greater import, however, is that its potency is hundreds of times greater than morphine's. Microgram amounts, administered by a transdermal patch, can provide pain relief for up to 72 hours. Because fentanyl does not traverse the gastrointestinal tract, it causes less constipation than other sustained-release opioids and does not induce liver metabolism through first-pass effects. It also does not appear to have active metabolites.
Oxycodone is sometimes thought of as a weak opioid, probably because it has been formulated in low-dose pills. Like fentanyl, however, it is more potent than morphine and is associated with fewer adverse effects. A sustained-release formulation, introduced about five years ago, has quickly become the most popular opioid for treatment of chronic noncancer pain in the United States.
Oxycodone is actually a prodrug. The active metabolite, oxymorphone, is produced in the liver by the enzymatic activity of cytochrome P450 2D6. (This is the same enzyme responsible for activating two other prodrugs, hydrocodone and codeine). Approximately 10% of the population have genetically low levels of P450 2D6 and thus require higher than usual doses of the prodrug to obtain pain relief. Less than optimal effects may also be expected in patients taking neuroleptics, quinine, or selective serotonin reuptake inhibitors such as fluoxetine that inhibit P450 2D6 activity.
There have been sporadic reports of agitation, sleeplessness, and dysphoria in patients taking high doses of oxycodone, which suggests that, in addition to µ-opioid effects, the drug also has kappa-opioid effects. It may thus have some added advantage for treatment of visceral pain, which appears to respond to kappa-receptor agonists.
Hydromorphone. The clinical trials of sustained-release hydromorphone indicate that it has analgesic effects similar to those of morphine but that it has fewer side effects. None of hydromorphone's metabolites are pharmacologically active, which will make this drug particularly useful for patients with renal failure.
Methadone and Levorphanol. By virtue of their intrinsically long half-lives, methadone and levorphanol are also useful for treatment of chronic pain. Both have significant nonopioid effects that may contribute to their utility as pain relievers, especially for neuropathic pain. Methadone appears to be a potent N-methyl-D-aspartate (NMDA)-receptor blocker. Levorphanol also has some NMDA inhibitory effect and may, in addition, inhibit reuptake of serotonin and noradrenaline.
These drugs are generally reserved for second-line treatment, however, because they are difficult to titrate and have delayed-onset side effects. To control pain effectively, methadone must be administered at intervals (6-8 hr) shorter than its metabolic half-life (15-90 hr). As serum levels of the drug increase, so does the risk of toxicity, primarily sedation.
Many U.S. physicians are under the impression that it is illegal to prescribe methadone without a license from the federal government, but that is the case only if methadone is used in a maintenance program for treatment of drug addiction. When prescribed for pain, methadone is subject to the same regulations as other high-potency opioid medications. One practical advantage of methadone and levorphanol is that they are both relatively inexpensive.
Rescue Medication. Most patients taking long-acting opioids should be supplied with a fast-acting rescue opioid to treat pain that breaks through the regular medication schedule. Oral rescue agents may be needed as often as every two to four hours. The usual dose is the analgesic equivalent of 5% to 15% of the 24-hour baseline dose of the long-acting opioid (Table 1).
Choosing the right opioid for a particular patient is usually a matter of guesswork. Even though opioids have been in use as long as any other class of medications, the base of knowledge on how to use them most effectively is surprisingly small, particularly with regard to combining them with nonopioid medications and with each other.
Much of the human research has involved patients who were not in pain or who experienced limited episodes of acute pain. As a result, we often see disparities between the guidance of the medical literature and empiric clinical practice. That tables of drug equivalence are still published attests that many clinicians think opioids are interchangeable. Imagine the reaction if someone were to publish a conversion table for antibiotics showing how to convert milligrams of ampicillin into milligrams of gentamicin!
The opioid medications currently in use act largely through the µ-opioid receptor, but nearly all of them stimulate kappa- and delta-opioid receptors and some nonopioid receptors as well. In addition to their inhibitory functions, some opioid receptors appear to have stimulatory functions that may be responsible for some adverse effects of opioids. According to recent studies, these stimulatory effects can be inhibited by miniscule doses of opioid antagonists such as naltrexone and nalmefene. As we get more sophisticated in using these medications, we may be able to combine opioid agonists and antagonists both to enhance pain relief and to attenuate dependence and other side effects. This exciting field of analgesic research has recently been reviewed by R. F. Crain and S. M. Shen.
While there is general agreement that µ-opioid receptors predominate in mediating opioid-induced analgesia, we still do not know the consequences of stimulating other opioid receptors, including how they affect different groups of patients. Kappa-opioid agonists, for example, produce greater analgesia in women than in men. Until recently, most clinical pain research was conducted in men to reduce the variability caused by hormonal fluctuations.
There are also wide
metabolic variabilities within groups of pain patients that may be determined by
genetics or influenced by interacting drugs. Yet we have little information to
guide the selection of dosages. For patients on chronic opioid therapy, it may
soon be reasonable to periodically assess serum levels of the drugs and their
metabolites to guide therapy, much as we do with antiarrhythmics and other
Titration Against Pain
The daily consumption of the
rescue drug can be an indicator of how much the sustained-release drug is
falling short of providing adequate relief. By titrating the sustained-release
drug accordingly, the minimum dose needed to ameliorate the pain can usually be
quickly established. Patients sometimes do well at the beginning of opioid
therapy and then seem to lose ground within a few weeks. In those who have been
severely limited in their activities, the recurrence of pain is not necessarily
a sign of growing tolerance to the medication--the patient may be experiencing
more pain because of increased activity and should be reassured that more
medication is required to alleviate the pain of someone with a busy schedule
than of someone lying in bed all day. Maintaining close communication with
patients and families and explicitly laying out the criteria for evaluating the
effects of analgesic medications can go a long way toward defusing the anxiety
that often accompanies visits to the physician.
The most feared side effect of opioid medications is respiratory depression. This does not occur, however, when the drugs are titrated against the patient's pain, probably because the pain signals activate respiratory centers in the brain that counterbalance depressive effects (Figure 1). In the event of an overdose, the activity of various brain centers is affected in an orderly, progressive fashion, with the cortex affected before the brainstem (see Figure 1, "Chronic Pain: 1. A New Disease?"). The patient will thus become unresponsive and obtunded before the opioid level is high enough to suppress the respiratory centers. One can be assured that a patient who is awake and complaining of pain is not in any imminent danger of respiratory depression.
If the source of pain is abruptly removed, an opioid dose that is well-tolerated can suddenly become sedating. That is clinically relevant when sending a patient taking opioids for an anesthetic or neurolytic procedure. Otherwise, respiratory depression rarely occurs. The main challenges of opioid therapy have to do with managing such common side effects as nausea, vomiting, itching, or somnolence. Although typically resolved within a few days to a week, they can be temporarily incapacitating.
Nausea occurs in 10% to 40% of patients treated with opioids. If they are getting good pain relief, there is no need to withdraw or reduce the medication. Antiemetics such as promethazine or prochlorperazine, both of which are available as pills and suppositories, are usually effective when used three to four times a day for a few days. If the nausea is primarily postprandial or related to bloating and early satiety, metoclopramide (10-20 mg tid) is the drug of choice. For some patients, antihistamines such as hydroxyzine are also useful as antiemetics. In refractory cases, low doses of haloperidol or olanzapine at bedtime can be very helpful.
Hives or itching may occur at the beginning of therapy with certain opioids as a result of their direct effect on mast cells. These problems are more commonly seen with the naturally occurring opiates (e.g., codeine or morphine) than with synthetics. Although hives and itching are not generally considered allergic reactions, relief can often be obtained with antihistamines. Patients with a positive history of such symptoms can be preemptively given a nonsedating antihistamine for the first week of opioid therapy.
Initiation or escalation of opioid medications can produce sedation or somnolence in some patients. However, those with severe pain may feel more alert or normal. If the patient is getting good pain relief but experiences sedation, the patient and family should be counseled to find ways to limit activities such as driving, working, meal preparation, and child care while the opioid is being titrated. If the situation cannot be tolerated, the dose can be reduced by 25% until the sedation lifts. Should it persist, other sedating medications, including antihistamines and antiemetics used to counteract other side effects, may have to be reduced in dose or eliminated.
A few patients experience myoclonus with opioid use. If discomforting, the condition can be treated with a benzodiazepine. Clonazepam is a good choice, because it is less sedating than other drugs in the same class. Opioids can sometimes cause urinary retention and increased bladder capacity due to increased sphincter tone. With high doses, peripheral edema may also occur as a result of antidiuretic hormone release, in which case a short course of diuretics is usually effective. There is no evidence of any organ toxicity attributable to the prolonged use of sustained-release opioids.
The one common persistent
side effect of opioid use is constipation, which is mediated by opioid receptors
in the bowel. More than half of patients on sustained-release opioids experience
constipation requiring specific therapy. It is important to get patients on a
good bowel regimen as soon as possible and to teach them to adjust their bowel
medications as needed. Many patients start out with a stool softener (e.g.,
docusate) and a mild fiber-based cathartic (e.g., senna). Osmotic laxatives such
as milk of magnesia, polyethylene glycol, magnesium citrate, or lactulose will
probably also be needed and may as well be started early. In contrast to
stimulant laxatives, osmotic diuretics are safe and non-habit-forming. Patients
have to find the dose that they can safely take nightly to produce a firm stool
in the morning. Persistent constipation can be a serious problem requiring
emergency treatment. I tell my patients to let me know if they have not had a
bowel movement for at least three days. At that point, oral magnesium citrate or
a sodium phosphate enema may be indicated.
Tolerance to Opioids
That most adverse side effects of opioids resolve on their own is an indication of growing tolerance with continued use. Tolerance can also be conferred by other factors. Severe pain, for example, allows patients to tolerate the sedative effects of opioids. Whether tolerance develops to the pain-relieving effects of opioids is a matter of controversy. Most of the data on opioid tolerance and physical dependence in humans involves subjects who were not in pain. Studies of patients with chronic pain who have taken opioids for a long time indicate that once the dose required for pain relief is established, it generally remains stable unless the underlying disease progresses.
Often, when a specific
treatment or adjunctive therapy begins to work, patients who have been taking
opioids for some time will begin to feel somnolent or sedated, or will not
require rescue medications for long periods. That is the time to consider slowly
tapering the long-acting pain medication.
Physical Dependence on Opioids
With long-term use of
opioids, patients will experience physical symptoms (abdominal cramping,
sweating, nausea, diarrhea, irritability) if the medication is abruptly
withdrawn or the dose is markedly reduced. This type of physical dependence is
not limited to opioids but can occur with other drugs such as antihypertensives
and steroids. It is a medical condition and should not be taken as a sign of
psychological or spiritual weakness. Withdrawal symptoms are easily avoided by
using a tapering regimen when lowering the dose. This can nearly always be done,
without discomfort, in an outpatient setting. When necessary, however,
withdrawal symptoms can usually be relieved by slowing the taper or using small
doses of clonidine or a benzodiazepine.
Appropriate Use Versus Abuse
Nothing is intrinsically good or evil, but its manner of usage may make it so -- St. Thomas Aquinas
The fear of drug abuse and drug addiction is the major reason that physicians are reluctant to prescribe opioid medications for patients in severe pain. The inappropriate use of a medication for a nonmedicinal problem is drug abuse. Using a pain medication to get high or euphoric is clearly inappropriate, as is using drugs to escape family or other problems that should be dealt with by other means. If a patient's physical pain has prevented him or her from living life fully, using a medication that allows a return to normal activities cannot be called drug abuse.
The appropriate role of
medicine is to prolong and maintain life, promote function, and provide comfort
from symptoms of disease. It is up to the physician to determine whether the
prescribed medications are being used to participate in life or to escape from
it. The patient's mood and activities, and the reports of family members, can be
helpful indicators. Health care facilities are beginning to use validated
quality-of-life instruments that should make the assessment of appropriate and
inappropriate use easier. Appropriate use of pain medications can significantly
increase the quality of life; inappropriate use invariably decreases it.
Addiction and Pseudoaddiction
Taken to the extreme, drug abuse can become drug addiction, a driving force that leads to compulsive, socially inappropriate, or even dangerous behaviors. The overwhelming majority of drug addicts report that their addiction began with recreational drug use. Medical use of opioids is generally not associated with addiction.
The most important predictor of continued abuse or addiction is previous substance abuse. If a physician prescribes a pain medication in good faith, anyone who leaves that physician's care with an addiction probably already had a problem when the treatment began. That is not to say that addiction disorders are not an important concern--they are serious health problems that must be evaluated and treated. If substance abuse or addiction is strongly suspected, the patient should be referred for evaluation by a psychologist or psychiatrist who has had experience working with chronic pain patients.
patients as addicts can alienate them from their caregivers and family, deepen
their isolation, and prolong their suffering. Denied the pain treatment to which
they are entitled, patients often say that they feel isolated, anxious, and even
desperate. The obsessive and manipulative behaviors that these feelings
engender, which can sometimes be confused with addiction, are called
pseudoaddiction. Russell Portenoy's tabulation of drug-seeking behaviors is a
useful reference for discriminating between addiction and pseudoaddiction (Table
2). I go over this table with my patients so that they are specifically aware of
what I consider to be inappropriate behaviors.
Opioids are usually reserved as a last resort for the treatment of pain, but it may be time to consider using them to rescue patients in severe pain who have not responded to disease-specific treatments or mild analgesics. Once some relief is achieved, adjuvant medications and nonpharmacologic or more aggressive approaches can be tried. In my experience, most treatments work better when there has been some initial pain relief.
Medical training places such a strong emphasis on curing disease that we sometimes do not pay as much attention as we should to patients who cannot be cured. This is particularly true of patients whose condition has not improved despite years of traditional therapy. Even if they cannot be cured, however, they can be healed. There may be no effective treatment for the underlying illness, but there is almost always an effective treatment for the pain.
One thing that I learned in medical school that is still true today is that it is very easy to become judgmental when faced with a patient whose suffering is difficult to understand. Unfounded assumptions are harmful and can rob a suffering patient of hope. With our current knowledge of how pain is generated and alleviated, it is both disrespectful to the patient and a breach of medical ethics not to provide what is clearly needed. When a patient in chronic pain seeks our help, the first question we should ask ourselves is not whether we should provide an analgesic but whether we can in good conscience leave that person in pain. To quote Marcia Angell, "Few things a doctor does are more important than relieving pain. . . pain is soul destroying. No patient should have to endure intense pain unnecessarily. The quality of mercy is essential to the practice of medicine; here, of all places, it should not be strained."
Angell M. The quality of mercy (editorial). N Engl J Med 306:98, 1982
Ashburn MA, Rice I.J. The Management of Pain. Churchill Livingstone, New York, 1998
Brookoff D. Abuse potential of various opioid medications. J Gen Intern Med. 8:688, 1993
Crain SM, Shen KF. Antagonists of excitatory opioid receptor functions enhance morphine's analgesic potency and attenuate opioid tolerance/dependence liability. Pain 84:121, 2000
Graven S et al. Opioids in chronic nonmalignant pain: A criteria-based review of the literature. In Proceedings of the 9th World Congress on Pain, Devor M, Rowbotham MC, Weisenfeld-Hamlin Z (Eds). IASP Press, Seattle, 2000, pp 965-972
Joranson DE et al. Trends in medical use of opioid analgesics. JAMA 283, 1710, 2000.
Kalso E, McQuay HJ, Wiesenfeld-Hallin Z (Eds). Opioid Sensitivity of Chronic Noncancer Pain. IASP Press, Seattle, 1999
Larusso K et al. Use of sustained-release opioids after short-acting opioids. Paper presented at the 17th annual meeting of the American Pain Society, November 1998, San Diego, Calif.
Portenoy RK. Opioid therapy for chronic nonmalignant pain. In Progress in Pain Research and Management. IASP Press, Seattle, 1994, pp 595-620
[This Article in
PDF print format]