This site is intended for health professionals only

The use of opioids in chronic pain management


Gerald Aronoff
Adjunct Associate Professor
Department of Psychiatry
Duke University Medical Center
Medical Director
Carolina Pain Associates
Charlotte, NC
E:[email protected]

The use of opioids for the management of severe acute pain and postoperative pain is well established, is generally not controversial and has been demonstrated to provide good analgesia as well as concurrent improvement in morbidity, treatment course and shortened hospital stays. Furthermore, pre-emptive analgesia has been shown to decrease postoperative pain and, possibly, decrease the risk of chronic pain.(1–3) Yet chronic pain is woefully undertreated because physicians fear:

  • Addiction to pain medications.
  • Societal and regulatory censure.
  • Diversion of drugs to the street.

Clinical principles
Long-term opioid treatment for cancer pain is helpful in 70–90% of patients.(4,5) It is associated with improvement in morbidity and psychosocial distress and overall improvement in quality of life with adequate analgesia. It is unusual for treatment to be compromised by analgesic tolerance, physical dependence or dysfunctional drug use in individuals without a prior history of substance abuse.(6) Despite these observations, it has been widely established that, even with cancer patients, pain is grossly undertreated.(7) Cleeland et al found that 42% of patients with pain in cancer centres in the eastern USA received ­inadequate analgesic therapy.(8)

Opioid treatment for nonmalignant pain
While there is no doubt that opiates are often the drugs of choice for severe acute pain and chronic cancer pain, their use in chronic nonmalignant pain is appropriate but must be selective.(9)

Treatment principles
Chronic pain and chronic pain syndrome are not diagnoses but rather refer to a heterogeneous population of individuals with persistent pain. Although the World Health Organization (WHO) analgesic ladder paradigm(10) was developed for use with chronic cancer pain, these principles can be followed for noncancer pain, appropriately combining
acetaminophen/nonsteroidal anti-inflammatory drugs (NSAIDs)/cyclo-oxygenase (cox)-2 and adjuvants. Opioids can be well managed in the vast majority of patients with chronic pain.

Twycross has emphasised that, in a cancer pain population, increasing requirements for opioid analgesia generally represents not tolerance but rather progression of the disease.(11) It should also be noted that the dosage of opioid frequently can be decreased through balanced analgesia utilising adjuvants and peripherally acting non-narcotic analgesics. Judicious use of these drug combinations not only can enhance analgesia but can also diminish adverse side-effects and treat symptoms other than pain. Nonpharmacological pain management techniques of proven efficacy should also be employed to reduce suffering and improve analgesic efficacy.

Much has been learned about opioid use in chronic pain, especially from clinical studies and treatment of cancer pain. Insights from the cancer population include the following:

  • Unrelieved pain is associated with increased morbidity and psychosocial distress.
  • Effective analgesia can reverse these and improve quality of life.
  • Management problems related to tolerance or physical dependency are rare.
  • Addiction is rare without a prior history of substance abuse.

It is clear that inadequately treated pain is associated with increased morbidity and psychosocial distress, and that providing effective analgesia can reverse these and improve the quality of life. We now know that, in appropriately selected patients, opioids have a low morbidity (often less than with NSAIDs) and a low addiction potential. Although tolerance may occur in some cases, patients generally become tolerant to bothersome side-effects (other than constipation), more so than to analgesic effects.

Evidence from cancer studies suggests that, when patients who are clinically stable on a certain opioid dose request a dose escalation, it may be more related to progression of their disease than to tolerance. When principles from the WHO analgesic ladder paradigm(12) are followed appropriately, combining NSAIDs, adjuvants and opioids, we can manage the vast majority of patients with severe chronic pain. This paradigm was developed to afford patients with cancer pain the maximum benefit with the lowest drug dosage to minimise adverse effects. The model is also useful with noncancer pain. Progression up the ladder is based on pain severity. Although one of the therapeutic advantages of nonopioid treatment is the absence of tolerance and physical dependence, disadvantages relate to its ceiling effect at high doses as well as the risk of target organ pathology. These caveats are also true for the opioid-combination analgesics, in which case the toxicity generally occurs in relation to the nonopioid component. As one continues to ascend the ladder, the opioids can generally be titrated to the desired effect without concern for ceiling effects.

It is generally accepted that:

  • Nociceptive pain is the most opioid-responsive. (13–15)
  • Neuropathic and central pain are less but variably responsive.(16–18)
  • Psychogenic pain (pain disorder associated with psychological factors), although less common, is generally unresponsive and inappropriate for chronic opioid treatment.

Optimal management of chronic pain depends on comprehensive assessment. We still have much to learn about identifying which patients are most appropriate for which treatments (eg, improving treatment specificity). In a joint consensus statement, the American Academy of Pain Medicine (AAPM) and the American Pain Society (APS) published Guidelines for the Use of Opioids for the Treatment of Chronic Pain.(19) These, as well as various guidelines, are attempts to assist practicing physicians faced with complex chronic pain patients in need of analgesic treatment.

Clinical use of opioid analgesics
Opioids can be classified as agonists, agonist–antagonists and antagonists. In this section, only the agonist and agonist–antagonist groups are discussed since the antagonist drugs are not used as analgesics. Established principles of opioid use are:

  • Moderate-to-severe pain is treated with opioids, with or without adjuvants.
  • Opioids produce analgesia by binding to opioid receptors both within and outside the CNS.
  • All mu-agonists can cause constipation, sedation, nausea, confusion, urinary retention and respiratory depression.
  • Do not combine mu-opioid agonists with mixed agonist–antagonists. This can precipitate withdrawal symptoms and increase pain.
  • For moderate-to-severe pain of short duration, use immediate-release opioids (or combinations) with a short duration of action. These have a quicker onset of action.
  • For moderate-to-severe persistent pain, use sustained-action opioids. They are more convenient for patients, provide stable analgesic levels and more consistent analgesia.
  • When using a sustained-release opioid, or while waiting to reach steady-state levels, use short-acting opioids for breakthrough pain.
  • Consider using adjunctive medications for side-effects such as pruritus, nausea, sedation and constipation.

Drug regimens should be simplified by prescribing based on the anticipated duration of the pain:

  • Use long-acting or sustained-acting opioids for continuous pain.
  • Dose to effect: with long-acting or sustained-acting opioids, it is generally better to increase the dose rather than the frequency. There are exceptions, however. For example, if the patient has adequate analgesia (without adverse effects) for eight hours using a sustained-action opioid recommended for every-12-hour dosing but has end-of-dose failure, the clinician might try to increase the dose to the next-higher dosage and continue the every-12-hour dosing. However, the problem is then often that the patient will get excessive medication with an increased risk of adverse effects and still have end-of-dose failure. Therefore, there are two alternative scenarios likely to have a better clinical outcome:

–   Use the sustained-acting medication on an every- eight-hour regimen, carefully monitoring efficacy and side-effects.
–   Use a short-acting medication for end-of-dose failure.

Opioid agonists
Opioid agonists generally have similar clinical effects when given at equianalgesic doses. The differences among the individual drugs are important because a patient may have an adverse reaction or poor response to one of the opioids but not to others. For example, Eckhardt et al(20) found that about 7–10% of Caucasians are poor responders to codeine because they lack the cytochrome P450 enzyme CYP2D6, responsible for converting codeine to its active metabolites, including morphine.

Usually, the reason for choosing one agonist over another is that one confers some advantage – for instance, in route of administration or in the time–action of the agent. It should be emphasised, however, that at equianalgesic dosages the reasons for choosing one agonist over another are the specific adverse effects of a particular opioid for an individual patient and the differences in oral absorption or time action of analgesic effect.

Studies on mu-opioid pharmacology(21) have indicated that:

  • Mu-receptors mediating systemic morphine analgesia differ from those responsible for respiratory depression and constipation.
  • Mu-receptors mediating methadone, M6G, heroin and fentanyl analgesia differ from those responsible for morphine analgesia.
  • Mu-receptor subtypes may help explain genetic polymorphism and:

–    The variability in response of individual patients to various opioid analgesics.
–    Differences in side-effects among patients.
–    Incomplete cross-tolerance among mu-opioid analgesics.
–    The clinical utility of “opioid rotation”.

The use of opioids in pregnancy can result in neonatal withdrawal symptoms. This is a special problem with longer-acting opioids such as methadone, because the abstinence syndrome can be delayed until after the infant leaves the hospital. Also, the physician must acknowledge the possibility that long-term exposure to systemic opioids can adversely affect central nervous system (CNS) development, and this possibility must be weighed against the adverse effects in the infant of poorly controlled maternal pain.

The agonists are useful in relieving the anxiety that is normally associated with severe pain. The emotional reaction to pain is changed at usual clinical dosages; this may largely be due to the action of the agonists on the limbic system. Other CNS effects include analgesia, drowsiness, mood changes, mental clouding and, rarely, respiratory depression. Non-CNS adverse effects of the agonist opioids have been discussed above.

In the supine patient, therapeutic doses of agonists do not generally have significant effects on blood pressure or cardiac rate and rhythm. However, these substances cause vasodilatation, so watch for the development of orthostatic hypotension and syncope in susceptible individuals.

Morphine is the standard to which other opioid agonists are often compared. The reader is referred elsewhere for discussion of some of the comparative characteristics of opioid agonists.(22)

Article continues below this sponsored advert
Cogora InRead Image
Explore the latest advances in respiratory care at events delivered by renowned experts from CofE

Studies have found that, with postoperative pain, 10mg of D-amphetamine combined with morphine is twice as potent as morphine alone, and a combination with 5mg is 1.5 times as potent as morphine alone.(23) It also has been found that, in proportion to its dose, D-amphetamine generally improves performance that has been diminished by morphine.

Agonist–antagonist opioids
The agonist–antagonist group consists of ­pentazocine, butorphanol, nalbuphine and buprenorphine. The raison d’être for this class of drugs is not that they provide better analgesia than the agonist opioids, but rather that they were thought to have a much lower incidence of physiological dependence and addiction, and that they are less likely to cause respiratory depression.(24) However, this has not proven to be true. Although somewhat diminished compared with the agonist drugs, there is a definite abuse potential and the possibility of a withdrawal syndrome after chronic use.(25)

Clinical experience also suggests that, for severe pain, analgesic efficacy is better with the mu- agonists than with the agonist–antagonists or partial agonists, as there is a ceiling effect with these agents. Pentazocine and nalorphine (generally to a lesser extent) have been associated with psychotomimetic effects not reversible by naloxone.(25) The incidence of psychotomimetic effects seems to be reduced (although still present) with the newer members of this class, and in clinical trials this is described as approximately 1%. Exercise caution with these medications. Patients should not have taken opioids for a period of time depending on the half-life of the narcotic, as the agonist– antagonist drug may precipitate an abstinence syndrome. Pain medicine practitioners generally do not recommend this class of opioids for use with chronic nonmalignant pain.

Selecting the appropriate opioid analgesic
The selection process is actually based on several empirical factors, including an attempt to maximise therapeutic efficacy while minimising adverse effects. Patient compliance is likely to be greatest when drug administration is easy, the clinical desired effects (analgesia, decreased suffering and improved function) are reliable and side-effects are absent or minimal. Traditionally, the oral route has been preferred, and options have included pills, capsules, liquids and sprinkle administration.

In recent years, there has been a significant increase in transdermal and oral transmucosal administration of opioids. Some patients, unable to tolerate oral administration, may be appropriate for either rectal administration or a G-tube. It is beyond the scope of this article to discuss interventional approaches for opioid usage that include intramuscular, intravenous, subcutaneous, epidural and intrathecal routes.

Sustained-acting and long-acting opioids
Traditional opioid treatment begins with short-acting opioids. This provides more immediate pain relief and allows the physician to determine the minimally effective daily analgesic daily dosage via a process of titration. Most pain medicine physicians then switch to longer-acting opioids. Some physicians begin opioid treatment for chronic pain with sustained-acting or long-acting opioids. Controlled-release opioid preparations are among the most important recent innovations in analgesic treatments. They help patients to:

  • Achieve a steady level of satisfactory analgesia throughout the day.
  • Sleep through the night. Their advantages include:
  • Extended release over eight to 12 hours or longer.
  • Minimal fluctuations in opioid blood levels.

In opioid-naive patients, I strongly recommend beginning with a low dosage and titrating carefully based on analgesic effect and side-effects.

When the physician considers the use of opioids in a chronic pain patient with moderate-to-severe pain, the advent of sustained-acting opioid preparations improves the likelihood of getting good analgesia with a lower incidence of adverse effects. Randomised studies support their efficacy for chronic pain management.(26,27) Sustained-acting oxycodone (OxyContin), morphine (Avinza, Kadian, MS Contin, Oramorph SR) and fentanyl (Duragesic Transdermal System) enable the patient to reach a steady state while avoiding the fluctuations in blood levels, often either contributing to inadequate analgesia (with low blood levels) or adverse effects (with high blood levels). For years, MS Contin had been the standard for sustained-action morphine preparations, and it continues to be widely used because of lower cost when compared with Kadian or Avinza. However, pharmacokinetic profiles are noted to be superior with the newer drugs, with less fluctuation in plasma morphine levels.(28)

Another reason to administer these medications “around the clock” rather than prn is to minimise drug-seeking behaviour by making medication usage contingent upon complaining about severe pain. The side-effect profile of these medications appears to be significantly lower than repeated dosing with short-acting opioids. Clinical studies suggest that the opioid dose must be individualised based on indications, contraindications, efficacy and side-effects, and that there is no absolute ceiling effect or maximal dose for severe pain.(28)

McCaffrey and Portenoy have noted that opioid responsiveness cannot be determined unless the dose reaches treatment-limited toxicity.(29) This is consistent with my clinical experience as well as my findings from performing many chart reviews that suggested clinicians had determined the patient’s pain to be opioid-nonresponsive when, in fact, a subtherapeutic dose was likely administered in the absence of adverse effects.

In the event that the drug trial is limited by treatment side-effects, it is essential to try alternative opioids before determining that the pain is not opioid-responsive. Principles of opioid dosing in establishing equivalency dosing when switching from one opioid to another will be not be discussed in detail. Studies indicate that there is incomplete cross-tolerance when patients are switched from one mu-opioid to another.(21) Its occurrence is impossible to predict reliably, and it may involve interactions between genetic factors, concomitant medications and other medical factors. Therefore, as a general rule(28,30) physicians are advised to determine equianalgesic opioid doses and to then:

  • Reduce the equianalgesic dose by less than 50% if the pain is severe.
  • Reduce the equianalgesic dose by more than 50% in a medically frail patient.
  • Reduce the equianalgesic dose by less than 50% if the same drug is being used by a different route.
  • Reduce fentanyl less.
  • Reduce the equianalgesic dose of methadone by 75–90%.

Around-the-clock control of chronic pain has also been noted to be useful with severe rheumatological chronic pain problems. Based on the research, Roth advocates avoiding the “wind up” that leads to intractable pain progression by using potent opioids in intractable arthritic chronic pain.(31)

Methadone, used for decades as a heroin substitute for drug addicts being maintained on narcotics, has also been useful for chronic pain. However, it carries the burden of its pharmacokinetics. It has good oral bioavailability, and single-dose studies have estimated an oral:parenteral potency ratio of 1:2. Its analgesic effects last from four to eight hours (six to 12 hours after steady state is reached). Its pharmacological half-life is far longer, however, at times being 72 hours or more (some sources go as high as
190 hours), with potential toxicity from drug accumulation as patient and physician attempt to reach an analgesic “steady state”. Because of its undesirable pharmacokinetics, methadone must be used even more cautiously and monitored more carefully than other opioids to avoid toxicity.

There have been studies suggesting that, because of its NMDA receptor antagonism, methadone may be “the most effective opioid against neuropathic pain or pain related to tolerance”.(32)

Generally, in chronic pain patients, physicians should avoid using meperidine (Demerol) because of its short analgesic half-life and the toxicity of its principal metabolite, normeperidine, which accumulates with repeated dosing. Also, do not combine methadone with monoamine oxidase inhibitors.

Avoid pentazocine because of its psychotomimetic effects and its lower efficacy compared with pure agonists.

Breakthrough pain
Chronic pain is generally subdivided into persistent (or around-the-clock) pain and breakthrough pain. Portenoy and Hagen described these pains in cancer patients in 1990.(33) Persistent pain is the baseline pain that chronic pain patients experience. As noted above, when this pain is severe, it is generally managed with sustained-action, time-released (or long-acting) opioids, sometimes in conjunction with peripherally acting nonopioid analgesics and adjuvant analgesics.  Breakthrough pain is a flare-up or acute exacerbation of moderate to severe pain in otherwise-controlled baseline pain and is distinguished from poorly controlled persistent pain. Breakthrough pain should be distinguished from continuous uncontrolled pain as well as from acute episodic pain. Most patients who have persistent pain and breakthrough pain are able to acknowledge that there are multiple factors influencing their pain. These range from the disease itself to the patient’s activity level, as well as other physical and psychological factors. Clinicians are advised to take a detailed history to define the factors that influence the baseline persistent pain as well as the episodes of breakthrough pain. Breakthrough pain is further subdivided into incident pain associated with movement or activities, idiopathic or spontaneous pain not having a definable pattern and end-of-dose failure that occurs because the analgesic action of the around-the-clock medication is inadequate to contain the pain until the next scheduled dose. Generally, this is not considered actual breakthrough pain and it is often best managed by an adjustment in the around-the-clock medication.

Multiple surveys and studies have documented that more than half of chronic cancer pain patients experience significant breakthrough pain,(33) with some studies indicating a prevalence of breakthrough pain as high as 89%.(34) It is generally of rapid onset, brief and in >40 % of cancer patients begins in less than four minutes.(33) Portenoy(35) and others(36–38) have found that breakthrough pain often predicts a poor patient outcome associated with patient dissatisfaction with treatment, decreased levels of function and increased levels of anxiety and depression. Unrelieved breakthrough pain increases the economic burden placed on patients and the healthcare system because of increased hospitalisations and more medical visits.(39)

Although oral agents are most commonly used to treat breakthrough pain, gastrointestinal (GI) absorption prevents rapid absorption for most agents and is therefore often inadequate for rapid-onset breakthrough pain. Other possibilities include rectal and transdermal routes, as well as multiple invasive techniques such as intramuscular, subcutaneous, intravenous, epidural and spinal. Invasive techniques such as intravenous administration and patient-controlled analgesia (PCA) provide rapid onset of action but are expensive and require technical expertise.

Oral transmucosal fentanyl (OTFC) was developed to provide rapid onset and relatively short duration of action via a convenient and noninvasive delivery system.(40–43) It is FDA-approved for the treatment of breakthrough pain cancer patients, but because it is so well tolerated it is increasingly being used for severe rapid-onset breakthrough pain in noncancer patients. It is significantly different from other oral agents (pills, capsules and liquids) that:

  • Are affected by delayed GI absorption and first-pass metabolism in the liver.
  • Have delayed entry into the CNS associated with hydrophilic opioids and delayed onset of analgesia.

OTFC is rapidly absorbed from the buccal mucosa because it is highly lipophilic, coupled with the characteristics of the oral mucosa physiology, which include a large surface area, uniform temperature, high permeability and good vascularisation.(44) Persistent moderate-to-severe pain is best treated with medications administered around the clock to provide relatively constant blood levels. As discussed above, this generally involves an opioid agent with a prolonged absorption, such as a controlled-release oral medication or a transdermal medication. The dosing intervals are adjusted to ensure that blood concentrations do not fall greatly at the end of the dosing interval. The goal is to maintain blood levels above the analgesic threshold to relieve the persistent pain without causing overmedication.

When the clinician attempts to eliminate breakthrough pain by raising the around-the-clock analgesic dosage, there is often an increased incidence of side-effects, including excessive sedation. For this reason, it is generally recommended that breakthrough pain be treated separately from the around-the-clock pain.


  1. Anesthesiology 1992;77:439-46.
  2. Lancet 1993;342:73-5.
  3. Br J Anesth 1995;74:396-9.
  4. Pain 1996; 63:65-76.
  5. Patt RB, Burton AW. In: Aronoff GM, editor. Evaluation & Treatment of Chronic Pain. 3rd ed. Baltimore: Lippincott Williams & Wilkins; 1999. p. 337-76.
  6. J Pain Symptom Manage 1996;11:203-17.
  7. Ann Intern Med 1993;119:121-6.
  8. N Engl J Med 1994;330:592-6.
  9. Curr Rev Pain 2000;4:112-21.
  10. JAMA 1995;274:1870-3.
  11. Int J Clin Pharmacol Ther Toxicol 1974;9:184-98.
  12. JAMA 1995;274:1870-3.
  13. BMJ 2001; 322:1-7.
  14. Lancet 1996; 347:143-7.
  15. Arch Int Med 2000;160:853-60.
  16. Neurology 2002;58:554-63.
  17. Anesth Analg 2001;92:488-95.
  18. Pain 2001; 90:46-55.
  19. American Academy of Pain medicine and American Pain Society consensus statement 1997. The use of opioids for the treatment of chronic pain.
  20. Pain 1998; 76:27-33.
  21. Neurosci 2001;7:220-31.
  22. Aronoff, GM. In: Handbook on Pharmacological Management of Chronic Pain. Wavecrest Publishing; 2005. p. 275-8.
  23. N Engl J Med 1977;296:712-5.
  24. Psychopharmacol 1978;60:25-8.
  25. Jaffe J, Martin WRP. In: Gillman AG, Goodman LS,  Gilman A, editors.  The pharmacological basis of therapeutics.  10th ed. McGraw-Hill Publishing Co. New York; 2001. p. 569-620.
  26. J Pain Symptom Manage 1998;16:220-9.
  27. J Pain 2001;2:197-204.
  28. Hewitt DJ. Principles of opioid therapy (dosing). Presented at: The International Conference on Pain and Chemical Dependency. 6-8 June 2002; New York, NY, USA.
  29. McCaffery M, Portenoy RK. Overview of three groups of analgesics. In: McCaffery M, Posero C, editors. Pain:  clinical manual. 2nd ed. St Louis, MO:Mosby; 1999. p. 108-28.
  30. Portenoy RK. Opioid ­prescribing to patients with and without ­chemical ­dependency.  Presented at:  The International Conference on Pain and Chemical Dependency.  6-8 June 2002; New York, NY, USA.
  31. Roth SH. A new role for opioids in the treatment for arthritis. Drugs 2002;62:255-63.
  32. Mercadante S, Casuccio A, Fulfaro F, et al. Switching from morphine to methadone to improve analgesia and tolerability in cancer patients: a prospective study. J Clin Oncol 2001;19:2898-904.
  33. Portenoy RK. Symposium on Breakthrough Pain at the 15th Annual Scientific Meeting of the American Pain Society, 14–15 November 1996, Washington DC, USA.
  34. J Pain Sympt Manage 1998;16:179-83.
  35. Pain 1990;41:273-81.
  36. Pain 1989;27:203-9.
  37. Pain 1992;50:151-5.
  38. Pain 1994;59:141-5.
  39. J Pain 2002;3:38-44.
  40. Anesth Analg 1999;89:732-8.
  41. Proc Am Soc Clin Oncol 1998;16:3238-45 (Abstract 180).
  42. J Clin Oncol 1998;16:3238-45.
  43. Pain 1999;79:303-12.
  44. Am Acad Pain Med 2005;6:4.

Be in the know
Subscribe to Hospital Pharmacy Europe newsletter and magazine