McMaster University

Michael G. DeGroote
National Pain Centre

Scope of Search

Canadian Guideline for Safe and Effective Use of Opioids for Chronic Non-Cancer Pain

Cluster 1: Deciding to Initiate Opioid Therapy

R05. Recommendation Statement

No. Recommendation Keyword
R05 Before initiating opioid therapy, ensure informed consent by explaining potential benefits, adverse effects, complications and risks (Grade B). A treatment agreement may be helpful, particularly for patients not well known to the physician or at higher risk for opioid misuse. (Grade C). Risks, adverse effects, complications
  • R05. Discussion
  • R05. Summary of Peer-Reviewed Evidence

R05. Discussion

1. Informed Consent

A discussion about potential benefits, adverse effects, complications, and risks helps the physician and patient make a joint decision on whether to proceed with opioid therapy. (See Appendix B-4 for opioid information for patients).

1.1 Goal Setting: Potential Benefits and Patient Expectations

Before starting opioids, the physician should ensure the patient’s expectations are realistic. The goal of opioid therapy for chronic non-cancer pain is rarely the elimination of pain, but rather an improvement in function or a reduction of pain intensity by at least 30%. Before starting opioids, a discussion with the patient about specific goals related to pain reduction and functional improvement should address any unrealistic expectations. These agreed-on goals should be documented in the patient’s record; they are critical in determining that opioids are effective and should be monitored over time.

1.2 Adverse Effects

The most common adverse effects are listed in Table B-5.1.

Table B-5.1 Adverse Effects of Opioids

Note: From randomized trials, excluding enrichment design trials, results show a clinically important difference (Diff>10%) and are statistically significant (P<0.05).

Adverse effect Number of Studies Incidence in Opioid Group Incidence in Placebo Group Difference (95% CI)
Nausea 38 28% 9% 17% (13% to 21%) P<0.00001
Constipation 37 26% 7% 20% (15% to 25%) P<0.00001
Somnolence / drowsiness 30 24% 7% 14% (10% to 18%) P<0.00001
Dizziness/vertigo 33 18% 5% 12% (9% to 16%) P<0.00001
Dry-skin / itching / pruritus 25 15% 2% 10% (5% to 15%) P<0.0001
Vomiting 23 15% 3% 11% (7% to 16%) P<0.00001

Adverse effects where the difference was not clinically important (Diff <10%) and/or not statistically significant (P>=0.05) include: dry-mouth, headache, sexual dysfunction, hot flushes, loss of appetite, abdominal pain, fatigue, sleeplessness/insomnia, sweating, blurred vision/confusion, muscle contractions, diarrhea, ataxia, edema, difficulty urinating, restless legs, application site reaction, heart burn, anxiety, weakness.

1.3 Medical Complications

Information about medical complications associated with LTOT is reported in non-randomized trials (RCTs are short-term: 3 months). There is no evidence regarding the frequency of medical complications, the relationship between length of time on opioids and occurrence of medical complications, or whether the complications are permanent or transient. Patients should be informed about potential long-term use medical complications such as neuroendocrine (hypogonadism and amenorrhea), sleep apnea (central sleep apnea or worsening of obstructive sleep apnea), and opioid-induced hyperalgesia.

1.3.1 Neuroendocrine Abnormalities

Neuroendocrine abnormalities and erectile dysfunction can be experienced with LTOT (Ballantyne 2003, Daniell 2006). One recently published randomized trial found that the incidence of sexual dysfunction after morphine happened in 11% (Khoromi 2007). However, two other randomized trials suggested that patients taking opioid medications reported better sexual function, which was likely an improvement of well-being (Arkinstall 1995, Watson 2003). In summary, in the short term, the patient may notice improvement in sexual function (as a consequence of improved analgesia), but in the long term, opioids may cause neuroendocrine dysfunction.

1.3.2 Sleep Apnea

Opioids can aggravate not just central sleep apnea, but frequently also significantly aggravate obstructive sleep apnea. High opioid doses may contribute to sleep movement disorders including myoclonus and sometimes choreiform movement, and in combination with benzodiazepines and other drugs may significantly contribute to oxygen desaturation (Zgierska 2007, Mogri 2008, Farney 2003). Consider a sleep study for patients using high-dose opioids, opioid in combination with other sedating drugs, elderly patients, obese patients, and patients with somnolence.

1.3.3 Opioid-induced Hyperalgesia (OIH)

OIH is a paradoxical hyperalgesia resulting from LTOT. It is characterized by pain sensitivity (hyperalgesia and allodynia) in the absence of overt opioid withdrawal. It is distinct from tolerance in that pain extends beyond the area of initial complaint. It is also known as opioid neurotoxicity or opioid-induced pain sensitivity (OIPS) (Chu 2006, Ballantyne 2003).

1.4 Risks

Explain the potential risks of opioid therapy and provide reassurance on how the risks can be managed. See Table B-5.2.

Table B-5.2 Opioid Risks

Risk Actions for the Physician  Information for the Patient  Directions for the Patient and Family 
  • Start with a low dose, titrate gradually, and monitor frequently. See Table B-9.1: Opioid Suggested Initial Dose and Titration.
  • Be cautious when prescribing benzodiazepines (see Recommendation 06).
  • For patients at higher risk of overdose*
    • Initial dose should not exceed 50% of the suggested initial dose, and dose increments should be more gradual (See Table B-9.1)
    • Consider a 3-day "tolerance check": contact the patient 3 days after starting the opioid to check for signs of oversedation.  
  • Opioids are safe over the long term, BUT can be dangerous when starting or increasing a dose.
  • Overdose means thinking and breathing slows down. This could result in brain damage, trauma, and death.
  • Mixing opioids with alcohol or sedating drugs greatly increases the risk of overdose.
  • Contact a physician on early signs of overdose: slurred or drawling speech, emotional lability, ataxia, "nodding off" during conversation or activity.
  • Avoid mixing prescribed opioids with alcohol or sedating drugs.
  • Avoid driving a vehicle or operating equipment/heavy machinery until a stable dose is reached.
  • If you interrupt your medication schedule for three days or more for any reason, do not resume taking it without consulting a physician. 

Ask questions about the following to determine risk of opioid diversion: 

  • History of alcohol or substance abuse (patient and/or household member)
  • Transient or unstable housing
  • Vulnerability and dependence on caregivers  
  • Sharing prescribed medication with others is illegal, and could harm the other person.
  • While the patient's opioid dose is safe, it may be dangerous for other people.
  • Adolescents may abuse prescription opioids and sometimes pilfer drugs from the family medicine cabinet  
  • Do not give your prescribed medication to any other person: This is illegal, and the drug could harm the other person.
  • Store your medication in a secure place with limited access to guard against others' (e.g., adolescents) illicit use.
  • Inform your physician if you feel your medication is insecure, or if you feel any pressure about sharing.  
Addiction Use appropriate screening tools to determine risk of addiction. 
  • Addiction means that a person uses the drug to "get high" and cannot control the urge to take the drug.
  • However, most patients do not get high from taking opioids, and addiction is unlikely if addiction risk factors are low: those at greatest risk have a history of addiction.
  • Withdrawal symptoms can occur in any patient taking opioids regularly: they do not indicate addiction. 
Do not let unfounded fears of addiction stop you from taking your medication. Take your medication strictly as prescribed and do not stop the medication without informing a doctor.  
Withdrawal  If a decision is made to discontinue opioid therapy, the opioids should be tapered under medical supervision (see Appendix B-12).  
  • Opioid withdrawal symptoms are flu-like, e.g., nausea, diarrhea, and chills.
  • Withdrawal is not dangerous but it can be very uncomfortable.
  • Withdrawal can occur in any patient who takes opioids regularly, and it does not mean that the patient is addicted.   
Do not abruptly discontinue your medicationas this can cause uncomfortable withdrawal symptoms.

* Patients at higher risk of opioid overdose are those with:

  • Renal or hepatic impairment: Caution is advised, because opioids are metabolized in the liver and excreted through the renal system (Tegeder 1999, Foral 2007). Morphine is contraindicated in renal insufficiency.
  • Chronic obstructive pulmonary disease (COPD) and sleep apnea: Opioid use may be a risk factor for central sleep apnea (Mogri 2008). Tolerance to the respiratory depressant effects of opioids develops slowly and incompletely, putting COPD patients at risk for respiratory depression with a higher dose increase.
  • Sleep disorders: Sleep disorders, including insomnia and daytime sleepiness, are common among opioid users (Zgierska 2007). They may reflect the effects of pain, or the sedating effects of opioids, or concurrent depression.
  • Cognitive impairment: Opioids should be avoided in cognitively impaired patients who live alone, unless ongoing medication supervision can be arranged.

2. Treatment Agreement / Contract

Contracts are widely used in the long-term administration of potentially abusable substances. These agreements are intended to improve adherence and to enhance the therapeutic relationship by initiating an alliance between the patient and the physician. A contract is defined as an "explicit bilateral commitment to a well-defined course of action." Responsible parties in the contract usually have a clearly stated understanding of their individual obligations.

Contracts attempt to improve treatment through disseminating information, facilitating an agreed-on course, and enhancing adherence. The treatment agreement often includes clear descriptions of medication use and abuse, as well as the consequences for violating the contract.

2.1 Treatments Agreements: Oral or Written
  • Written treatment agreements are chosen particularly for patients the physician does not know well, or who are at higher risk for misuse. A written agreement is usually signed by both patient and physician, with a copy provided to the patient.
  • Oral treatment agreements should be documented in the patient’s chart.
2.2 Treatments Agreement Contents

The agreement usually outlines responsibilities and boundaries for both the patient and physician. (See Appendix B-5 for an example of a treatment agreement.) For example, a treatment agreement typically includes the following:

  • states that the patient:
    • will not give opioids to others
    • will not receive opioids from other sources
    • will store the medication in a safe place
    • will comply with scheduled visits and consultations
    • will provide urine samples for drug screens when requested
  • states that the physician:
    • will not normally refill the prescription ahead of schedule if the patient runs out
    • may cease opioid prescribing if the patient does not abide by the agreement.
  • identifies one single prescribing physician: All physicians involved in the patient’s care should agree on a designated prescribing physician, and whenever possible, identify an alternate physician to continue prescribing a patient's medication in the event that the primary prescribing physician is unavailable.
  • identifies one dispensing pharmacy.

R05. Summary of Peer-Reviewed Evidence

1. Non-randomized trials describe medical complications.

1.1 Hypogonadism

Opioids influence the hypothalamic-pituitary-adrenal axis and the hypothalamic-pituitary-gonadal axis. Morphine has been reported to cause a strong, progressive decline in the plasma cortisol level in adults. Opioids interfere with the modulation of hormonal release, including an increase in prolactin and a decrease in luteinizing hormone, follicle-stimulating hormone, testosterone, and estrogen. Testosterone depletion has been demonstrated in heroin addicts and in patients receiving methadone maintenance therapy. The collective effects of the hormonal changes may lead to decreased libido, aggression, and drive; amenorrhea or irregular menses; and galactorrhea (Ballantyne 2003).

Most randomized trials reviewed did not inquire about sexual dysfunction. The few studies that did so were of too short duration to allow for the development of any endocrinological abnormalities. In these studies, the authors inquired about sexual activity by using the Pain Disability Index (PDI). This index consists of 7 self-reported disability subscales, one of which refers to sexual activity; each scale is graded from 0 to 10, where 0 = no disability and 10 = total disability. This scale is not adequate to validly identify sexual dysfunction. Only two studies give a specific score on the dimension of sexual activity. In the first study using this measure (Arkinstall 1995), with 46 patients randomly assigned to receive CR codeine or placebo, the PDI score for the “sexual activity” subscale was 4.1 and 6.3, respectively. In the other (Watson 2003), which involved 45 patients, the score was 3.4 for controlled-release oxycodone and 4.5 for placebo. Both studies, therefore, suggested that patients taking opioid medications reported better sexual function than those taking placebo.

However, the PDI is a patient-rated global rating of function, does not measure variables such as libido, sexual dysfunction or gonadal function, or opportunity for sexual activity, and by itself cannot be used to estimate risk of hypogonadism. It is more likely that improvement of well-being secondary to better pain control by the use of opioids, accounted for this reported positive result in those studies.

One recently published trial (Khoromi 2007) found that the incidence of sexual dysfunction after morphine happened in 11% (of 28 completers of the study, out of 55 randomized), 0% in the nortriptyline group, 4% in the combination (morphine plus nortryptiline) and 0% in the placebo group. It is not possible to draw conclusions about the differences among these four groups because 1) this information is drawn from the completers of the study, and 2) these subgroup analyses do not have statistical power to detect any meaningful difference. Nevertheless, it was interesting to note that most recent studies are starting to ask participants about sexual dysfunction as a possible adverse event from opioids.

1.2 Sleep apnea

Patients on long-term sustained-release opioids show a distinctive pattern of sleep-disordered breathing that is different from the disturbances usually observed in subjects with obstructive sleep apnea (OSA). The oxygen desaturation is more severe and respiratory disturbances are long during NREM sleep (Farney 2003). In another study, even a short-term ingestion of opioid analgesic precipitated central sleep apnea in patients with chronic pain receiving long-term opioid therapy (Mogri 2008). There is also evidence that opioids may complicate underlying sleep apnea and make continuous positive airway pressure (CPAP) therapy less effective (Mogri 2008).

1.3 Opioid-induced hyperalgesia

Many studies were conducted in healthy volunteers with experimental pain, opioid addicts on methadone program and on perioperative exposures to opioids. There is one prospective study conducted on chronic pain patients (low-back pain) after one month of oral morphine therapy (Chu 2006). These authors showed evidence for the development of analgesic tolerance and OIH using a cold pressor test and experimental heat pain to measure pain sensitivity.

2. Evidence for Treatment Agreements

Overall, there is evidence to support the use of treatment agreements, although from non-randomized studies (Arnold 2006). One small study found that treatment agreements improve compliance (Fishman 2000), while another found that primary-care physicians were more willing to prescribe opioids to patients if the pain-medicine physician also signed an agreement (“trilateral contract”) (Fishman 2002).