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 2: Conducting an Opioid Trial

R08. Recommendation Statement

No. Recommendation Keyword
R08 During an opioid trial, select the most appropriate opioid for trial therapy using a stepped approach, and consider safety (Grade C). Stepped opioid selection
  • R08. Discussion
  • R08. Supporting Evidence

R08. Discussion

The most appropriate drug for an opioid trial depends on the patient’s clinical profile and individual circumstances. The following tables have been prepared to assist prescribers in selecting the most appropriate opioid.

Table B-8.1 Stepped Approach to Opioid Selection

  Mild-to-Moderate Pain Severe Pain
First Line codeine or tramadol morphine, oxycodone or hydromorphone
Second Line morphine, oxycodone or hydromorphone fentanyl
Thrid Line   methadone

Table B-8.2 Safety Issues to Consider When Selecting Opioids

Note: This table highlights safety issues for specific agents; for comprehensive information, prescribers should consult the individual drug monographs.

Agent Safety Issues
  • Use with caution for breast-feeding women: some rapidly convert codeine to morphine, placing the infant at risk of morphine toxicity. (See Recommendation 19.)
  • Lower risk of overdose and addiction than stronger opioids. (See Supporting Evidence item 1.)
  • Associated with seizures in patients at high seizure risk, or when combined with medications that increase serotonin levels, e.g., SSRIs.
  • Lower risk of overdose and addiction than stronger opioids. (See Supporting Evidence item 1.)
Morphine Avoid for patients with renal dysfunction: an active metabolite of morphine (M-6 glucoronide) can accumulate to toxic levels in patients with renal impairment. (See Supporting Evidence item 2.)
Oxycodone, Hydromorphone, Hydrocodone Use with caution for patients at higher risk for opioid misuse and addiction: experimental studies and surveys of drug users suggest that oxycodone, hydromorphone and hydrocodone may have a higher abuse liability than morphine. (See Supporting Evidence item 3.)


1) Before starting fentanyl, obtain a complete history of opioid use within the last 2 weeks to ensure the patient is fully opioid tolerant. Tolerance can be assumed if the patient is on a moderate, stable dose of a strong opioid, i.e., a total daily dose of at least 60–90 mg/day morphine equivalence daily for at least 2 weeks. This dose should be scheduled rather than p.r.n. (at least b.i.d. for CR or q.i.d. for IR). See Supporting Evidence item 4.)

2) Do not switch from codeine to fentanyl regardless of the codeine dose, as some codeine users may have little or no opioid tolerance.

3) Maintain the initial dose for at least 6 days: use extra caution with patients at higher risk for overdose, e.g., elderly, patients on benzodiazepines.

4) Advise the patient as follows:

  • Be alert for signs of overdose: (e.g. slurred or drawling speech, emotionally labile, ataxia, nodding off during conversation or activity) if detected, remove the patch and seek medical attention.
  • Apply as prescribed: do not apply more than one patch at a time or change more often than directed.
  • Avoid heat sources such as heating pads, electric blankets, saunas, heated waterbeds, hot baths, sunbathing.
  • Dispose of patches securely: a used patch contains large amount of fentanyl and could be dangerous to others. e.g., children or abusers could "recycle" by cutting into small pieces and sucking the pieces.
Methadone Use methadone to treat pain only if holding a written Health Canada exemption. Titration is hazardous due to its very long half life leading to bio-accumulation. (See Supporting Evidence item 5.)
Meperidine (Demerol®)

Not recommended for use in CNCP:

a) oral meperidine has poor bioavailability and is less effective than codeine, and

b) normeperidine can accumulate with frequent use of parenteral doses of meperidine, causing seizures and delirium. (See Supporting Evidence item 6.)

Acetaminophen-opioid combinations Use with caution to avoid acetaminophen toxicity. FDA (U.S.) recommends a maximum daily dose of 3.2 grams acetaminophen for adults = 10 tablets/day for opioid/ acetaminophen combinations. The manufacturer recommends a lower dose for tramadol/acetaminophen (8 tablets/day). (See Supporting Evidence item 7.) Heavy drinkers should be advised to use acetaminophen with extra caution.

Table B-8.3 Other Formulations and Preparations

Formulation/ Preparation Safety Issues
CR formulations Titrate with caution to avoid overdose and misuse: each CR tablet can contain a much higher opioid dose than IR formulations, and can easily be converted to IR by biting or crushing the tablet. (See Supporting Evidence item 8.)
Parenteral opioids Parenteral opioids are not recommended for use in CNCP: parenteral route has higher risk of overdose, abuse and addiction, and infection.

R08. Supporting Evidence

1. Codeine and Tramadol

1.1 Codeine and tramadol may have a lower abuse risk than more potent opioids.

Codeine has a lower risk of abuse and addiction than stronger opioids. For example, one national U.S. study found that codeine and other low potency opioids have low ratios of abuse to prescription use, relative to oxycodone, hydromorphone and hydrocodone. Abuse rates were measured from Drug Abuse Warning Network data (Dasgupta 2006). Tramadol also has a low risk of addiction, and experimental studies suggest that it has fewer psychoactive effects than other opioids (Preston 1991, Cicero 2005).

2. Morphine

2.1 Morphine can cause toxicity in patients with renal dysfunction.

For example, one cross-sectional study demonstrated that M-6 glucoronide, an active metabolite of morphine, accumulated in the serum of patients with renal dysfunction when morphine was administered orally or subcutaneously. The degree of accumulation was related to the morphine dose and the extent of renal impairment (Peterson 1990).

3. Oxycodone, Hydromorphone and Hydrocodone

3.1 There is evidence that oxycodone and hydromorphone have a higher abuse liability than other opioids. This is based on phase-2 studies, patient surveys, and studies of treatment programs.

One study found that prescription opioid misusers ranked controlled-release oxycodone, and immediate-release hydromorphone and oxycodone as the most desirable of 14 different opioid formulations. The study used a validated opioid attractiveness scale (Butler 2006). A national surveillance study of addiction experts, law enforcement agencies and poison control centers identified hydrocodone and both immediate-release and controlled-release oxycodone as by far the most commonly abused opioids in the United States (Cicero 2007).

Only a few controlled studies have been conducted comparing opioids on their abuse liability. Two placebo-controlled studies compared the psychoactive effects of oral morphine to oral oxycodone in non-drug abusing volunteers. The studies found that oxycodone had greater reinforcing effects at equi-analgesic doses to morphine (Zacny 2003, Zacny 2007). Another controlled trial found that oxycodone, hydromorphone and hydrocodone had equivalent abuse liability (Walsh 2008). The clinical significance of these studies for chronic pain patients is not certain because volunteers may experience different psychoactive effects than actual pain patients (Lamb 1991).

It is also possible that the prevalence of oxycodone abuse may simply reflect its popularity as an opioid analgesic. In an analysis of data from the Drug Abuse Warning Network, oxycodone, hydromorphone and morphine had similar rates of overdoses and other events after controlling for the potency of the opioid and the amounts prescribed in kg (Dasgupta 2006).

4. Fentanyl

4.1 Fentanyl can cause significant cognitive impairment in non-tolerant opioid patients.

Experimental studies in volunteers have found that cognitive impairment caused by acute intravenous fentanyl administration was greater than that caused by moderate doses of alcohol (Zacny 1992, Schneider 1999).

4.2 Fentanyl has contributed to numerous overdose deaths.

Fentanyl was a contributing cause in 100 overdose deaths in Ontario between 2002 and 2004. In 54 of the deaths, fentanyl intoxication was the sole cause of death. Deaths occurred from both therapeutic and illicit use (Martin 2006).

Fentanyl-laced heroin appeared simultaneously in various parts of the United States, beginning in 2005. In Chicago, in the first half of 2006, 55 drug overdose cases (resulting in 12 deaths) have been attributed to fentanyl-laced heroin (Fodale 2008). Fentanyl toxicity is related in 92% of fentanyl-related deaths and is attributed partially due to cytochrome P450 3A4*1B and 3A5*3 variant alleles, resulting in variable fentanyl metabolism: the homozygous CYP3A5*3 have impaired metabolism of fentanyl (Fodale 2008). In July 2005, the FDA issued a public health advisory calling attention to an increase in the number of fentanyl patch-related overdoses and deaths, particularly among patients ignoring the product’s boxed warnings and instruction for use (Federal Drug Administration 2007).

4.3 CNCP patients on codeine at risk for overdose when switched to fentanyl.

Up to 10% of Caucasians lack the enzyme CYP450 2D6 that converts codeine to morphine and therefore when switching from codeine to fentanyl, regardless of the codeine dose, caution is required as patients may have little or no opioid tolerance (Tyndale 1997, Romach 2000, Howard 2002).

5. Methadone

5.1 Methadone for pain is more effective than placebo, but has not been shown to be more effective than other opioids.

Sandoval (2005) conducted a systematic review of methadone for CNCP. The review included 21 studies (1 small randomized trial, 13 case reports, and 7 case series) and concluded that pain improvements were meaningful in 59% of the patients in the uncontrolled studies. The randomized trial demonstrated a statistically significant improvement in pain for methadone (20 mg/day) compared to placebo. Side effects were considered minor. One controlled trial found no difference in analgesic efficacy between morphine and methadone in cancer patients with respect to pain management (Bruera 2004). A similar trial found no difference between methadone, oral morphine and transdermal fentanyl 25 ucg/hour, although methadone titration was more difficult (Mercadante 2005).

5.2 Physicians must hold an exemption from Health Canada before prescribing methadone for pain.

Methadone has been associated with numerous overdose deaths in pain patients. Methadone analgesic use has increased sharply in the US, with a seven-fold rise from 1997 to 2004 (Sims 2007). This has been accompanied by a 17-fold increase in methadone overdose deaths (Shields 2007, Sims, 2007). Federal law requires that a physician hold a written exemption from Health Canada before prescribing methadone for analgesia. The specific process to apply for a methadone exemption varies by jurisdiction, and may include submission of a letter of support from the applicable medical regulatory authority before Health Canada will provide a methadone exemption. A physician may be able to receive an exemption to prescribe methadone under various circumstances, including if “mentored” by an experienced methadone prescriber. Physicians should confirm the methadone prescribing requirements of the jurisdiction where they practice.

6. Meperidine (Demerol®)

6.1 Repeated parenteral doses of meperidine are associated with adverse neurological events.

In one study of hospitalized patients receiving parenteral meperidine, 14% had neurological adverse events such as confusion or seizures. The risk of an adverse event was associated with the cumulative meperidine dose, renal insufficiency, and benzodiazepine use (Seifert 2004).

7. Acetaminophen-opioid Combinations

7.1 Acetaminophen is a common cause of hepatotoxicity; risk increases with alcohol use.

Acetaminophen toxicity causes the majority of cases of acute liver failure in the U.S., (Krenzelok 2009, Amar 2007). Sub-clinical liver toxicity has been shown to occur even with doses below 4 gm/day (Krenzelok 2009, Arundel and Lewis 244-54). To reduce toxicity, the FDA in the U.S. revised their maximum daily acetaminophen dose downward, from 4 gm/day to 3.2 gm/day. Alcohol competes for the same metabolic pathway as acetaminophen so heavy drinkers are at higher risk for toxicity. Chronic alcohol use is an independent risk factor for mortality in acetaminophen poisoning (Schmidt 2002).

8. CR Formulations

8.1 CR opioids are available in high-dose formulations which increase their risk of abuse and overdose.

CR opioids contain much higher opioid doses than acetaminophen-opioid combinations (e.g., one OxyContin® 80 mg tab = 16 Percocet® tablets). This increases the risk of both overdose and addiction. Controlled experimental studies indicate that the psychoactive effects of an opioid are dose related (Lamas 1994). Studies using non-drug-abusing volunteers have found dose-related reinforcing psychoactive effects with oral doses of 5, 10, and 20 mg of hydrocodone, and 10, 20, and 30 mg of oxycodone (Zacny 2003, 2005).

CR opioids can easily be converted to IR by crushing or biting the tablet. The outer layer of the OxyContin® tablet (but not other Contin tablets) is an IR formulation, containing 1/3 of the total dose.