Oral morphine is the standard strong opioid used in cancer pain, and it continues to be the benchmark 'step III' analgesic for severe pain in the World Health Organization (WHO) analgesic ladder, against which newer drugs are measured. [1-2] A number of alternative strong opioids are available in Australia, including oxycodone, hydromorphone, fentanyl, methadone, and buprenorphine.
What is known
There is evidence that alternative strong opioids such as methadone, [3,4] oxycodone, [5,6] hydromorphone [7,8] and fentanyl [9,10] are effective alternatives to morphine, although there are few trials which have been designed to allow direct comparison between different opioids. [1,2] There are no important differences between oral morphine, oral oxycodone, and oral hydromorphone, and according to recently released EAPC guidelines any of these can be used as the first line strong opioid.  Transdermal fentanyl has been preferred over slow release oral morphine by cancer patients in clinical trials, and may cause less constipation.  However, there is weak negative evidence regarding the use of transdermal fentanyl over the gold standard of oral morphine as a first-line drug, and there is negative evidence related to the use of transdermal buprenorphine first line in this population.  A systematic review of tramadol has identified significant problems in the evidence base for this drug, and concludes that it should not be considered as a strong (“step III”) opioid, nor is there evidence to support its use as an alternative to paracetamol / codeine for mild to moderate pain (“step II”). 
The practice of 'opioid switching' is done to improve analgesia in the minority of patients who do not respond to increasing doses or who have severe adverse effects. Inter-individual genetic differences are likely to be significant in patients’ individual responses to opioid medications. This may be a rationale for opioid rotation.  However the practice is not supported by strong evidence. [14-16,40] Studies which have been done have methodological problems, but suggest that 50% of patients who are switched may experience some clinical improvement.  The recent EAPC guidelines offer a weak recommendation for opioid switching. 
A recent systematic review has identified evidence to support various opioid conversion ratios, but points out that the clinical complexity of the situations in which these decisions are often made limits the potential for studying all of the relevant factors in detail.  One challenging area of practice relates to opioid rotation to and from methadone, as there are a number of different protocols available in the literature. A study of conversion ratios used to switch patients to methadone from other opioids revealed that, of the many methods in clinical use, none seems to be superior to any other.  Clinical expertise and careful monitoring are needed in the use of methadone. 
Managing 'breakthrough' pain
Consensus recommendations on breakthrough pain are available.  Providing an immediate-release opioid as needed to be used as a 'rescue dose' is recommended [1,19] and, in a change from previous advice that breakthrough opioids should be prescribed in a fixed ratio to background opioids, individual titration of breakthrough doses is now recommended in these guidelines.  The opioid preparation which has been best studied for use as a breakthrough medication is the transmucosal fentanyl lozenge, and it was shown to be effective.  The route of administration of an opioid breakthrough is significant, and where possible should match the time profile of the pain. Even though it is the most commonly recommended approach to managing breakthrough pain , there is little evidence to support the oral route as effective in providing breakthrough analgesia.  Clinicians should also address and treat reversible causes of breakthrough pain and background analgesia and adjuvant medications reviewed. [18,40] A weak recommendation for pre-emptive treatment of predictable episodes of pain with immediate release opioids that have short half-lives is also included in the recent evidence-based recommendations from the EAPC. 
Managing 'crescendo' pain
There is little good evidence to guide the practice of escalating opioid doses for severe or 'crescendo' cancer pain, and a range of strategies of combining fixed and variable doses, at different intervals, are used.  Whilst the dosing strategies used in studies varied, and so no evidence-based recommendations can be made, all the strategies studied were safe and effective in relieving pain in the majority of patients within 24 hours.
Combining strong opioids
Weak evidence is available to support the practice of using several strong opioids in combination, based on a small number of studies. However there are some theoretical advantages, and in clinical practice it is quite common for a different opioid to be used for background and for breakthrough dosing. 
Neuropathic pain is a significant problem for many palliative care patients, and has been thought to be only partially responsive to opioids. However, in a recent systematic review opioids were found to be effective in relieving neuropathic pain in patients followed over eight to ten weeks.  Adverse effects were common but not severe, however combining opioids and other adjuvants with CNS side-effects (antidepressants or anticonvulsants) CNS adverse events are more likely.  Tramadol has also been shown to be effective at treating neuropathic pain in a systematic review.  A systematic review of methadone in cancer pain provided no evidence that methadone has particular benefits for treating neuropathic pain beyond other opioids, but the evidence base is small. [3,4]
Routes of administration for opioids
The subcutaneous route has been shown to be as effective and safe as intravenous administration of opioids, while the rectal route is an alternative although less feasible and acceptable. [26,40] Intravenous infusion can be considered as a second line alternative to subcutaneous administration when that route is contraindicated, and techniques of patient- controlled analgesia may be used for either route of administration.  A review which compared transdermal opioids (fentanyl and buprenorphine) with slow release oral morphine suggested that the overall adverse effect profile was not significantly different, although constipation may possibly be less with transdermal opioids.  Transdermal opioids are recommended as a non-invasive option for patients who are unable to swallow.  Topical opioids have been studied for use in ulcers (both malignant and non-malignant) and in oropharyngeal mucositis, and appear likely to be effective, although the evidence is of varied quality, and further investigation is needed. 
Managing opioid adverse effects, whether by preventing or treating them, is an important aspect of clinical practice. A systematic review of nausea and vomiting associated with opioids identified evidence to support the practice of opioid switching or changing the route of administration to manage nausea thought to be due to opioids.  The subsequent EAPC guidelines recommend the use of antidopaminergic drugs (eg, haloperidol or metoclopramide) for treatment of this problem. 
Problems such as cognitive changes associated with opioids have also been studied,  and demonstrated objectively. Limited evidence supports the use of methylphenidate to treat central side effects of opioids (sedation and cognitive disturbance). [31,40] Pharmacokinetic evidence supports fentanyl, alfentanil, and methadone as safer in renal failure, with caveats, however there is little high level evidence to guide practice,  and EAPC guidelines are that fentanyl and buprenorphine are the appropriate first choice in patients with severe renal failure.  The prevalence of drug interactions with methadone was studied,  but little information was available that specifically relates to a palliative care population.
What this means in practice
Choosing the first line strong opioid
Morphine has been widely recommended due to its efficacy, easy availability, versatility, cost, and the extensive clinical experience of using it in palliative care patients; however there is no evidence that it is superior to other opioids. [16-17]
Morphine, oxycodone and hydromorphone are are clinically similar in regard to efficacy and adverse effects. [2,40] Choice of opioid is based on specific properties of the drug and preparation (eg, versatility of various routes of administration, availability of sustained and immediate release formulations, cost to the patient and the healthcare system).  Methadone and transdermal preparations are not recommended as first-line strong opioids. [2,10] Morphine, oxycodone and hydromorphone all have metabolites that are likely to accumulate in renal impairment. However, of these agents, there is extensive experience of successfully using hydromorphone in renal patients. Unwanted side effects due to accumulation of morphine metabolites may be satisfactorily dealt with by either increasing the dosing interval or reducing the 24 hour dose or by switching to an alternative opioid. [32,40] Methadone and fentanyl have less risk of accumulation. However methadone has complex pharmacokinetics and a long and unpredictable half-life  and fentanyl has less versatility in the preparations which are available, and significant potential for drug interactions. 
In the past, initiation of strong opioids with immediate release preparations has always been recommended, but the evidence also suggests that initiation with long-acting opioids can be safe and effective, and can improve compliance. [34,37] Suggested starting doses are oral morphine 30 mg/24 hr in opioid-naıve patients, or 60 mg/24 hr in patients already using a WHO step II opioid.  Doses should be divided into either once or twice daily according to the long-acting morphine preparation used, or divided and given fourth hourly if an immediate release preparation is used.
Breakthrough pain and opioid titration
Recommendations continue to evolve. Based on studies using the oral transdermal fentanyl lozenge for breakthrough pain, it appears that the ideal ratio of breakthrough dose to background dose is not fixed and should be individualised. [18,21] Ratios of one sixth to one twelfth of the background dose may be used as a starting point for prescribing breakthrough doses. There is little evidence to guide the titration of opioid analgesia.  Dose increments of at least 25% are usually required when opioids are titrated upwards for poorly controlled pain. 
Opioid rotation or switching
Opioid dose conversion ratios are not fixed, but are affected by the clinical context of the switch and the setting of care. [14,16,39-40] When switching opioids, frequent clinical review and reassessment of pain and adverse effects is the safest approach. Based on studies of opioid switching, it is important to individualise all opioid conversions.