The Pharmaceutical Journal Vol 262 No 7043 p620-624
May 01, 1999 Continuing education

Continuing education

issues in . . . Pain control in palliative care

By Mary Allen, BSc, MRPharmS, and Ros Taylor, MA, MBBChir

We continue our series looking at key issues in a number of disease and treatment areas with a discussion of the management of pain control in palliative care

Step three	Strong opioids,eg, morphine, +/– adjuvants
Step two	Weak opioids, eg, codeine, dihydrocodeine, +/–paracetamol +/– adjuvants
Step one	Simple analgesics, eg, paracetamol (1g four times a day) +/– adjuvants

Figure 1: Stepwise approach to analgesia in cancer pain (adapted from the WHO guidelines)

The World Health Organisation guidelines on cancer pain advocate the use of analgesia "by the mouth, by the clock and by the analgesic ladder". Patients should continue to take appropriate medication orally for as long as possible, they should receive doses at prescribed times rather than on a "when required" basis, and they should be prescribed medication which is appropriate to control the pain.¹ Often paracetamol is adequate.
Morphine is the recommended strong opioid for the management of severe pain in advanced cancer² (see Figure 1, step 3). The combined use of modified-release morphine preparations plus immediate-release morphine for breakthrough pain allows effective symptom control for most patients. As the disease progresses, subcutaneous diamorphine infusion provides an option if the oral route is no longer appropriate, for example, if the patient is unable to swallow or has protracted vomiting. Diamorphine is more soluble than morphine, providing a smaller volume for infusion or injection. Useful reference sources for the treatment of cancer pain are listed on p624.
Morphine's side effects include nausea and vomiting, sedation and hallucinations, which frequently reduce within a few days, and constipation, which usually does not. Appropriate stimulant laxatives such as co-danthrusate help and should be prescribed routinely for patients taking morphine. Respiratory depression is an adverse effect which is rarely a problem with careful dose titration.
Morphine dosage should be titrated against pain to achieve analgesia. There is not thought to be a ceiling to its analgesic effect, but for some patients there may come a stage at which further titration is not possible because of unacceptable side effects, resulting in a small minority whose pain is difficult to control.
Careful assessment of presenting pain is essential. Pain is a complex symptom, and one of the main principles of the palliative approach is to understand both the physical nature of pain and the psychological, emotional and spiritual needs of individual patients. Identifying the underlying cause of a given pain is crucial - and patients may present with more than one pain.
Pain which is neuropathic in origin may not be fully responsive to morphine, and the use of other agents, such as non-steroidal anti-inflammatory drugs and other co-analgesics, may bring about better pain relief (see later).
The phenomenon of "opiophobia" may result in an underuse of morphine. Doctors may under-prescribe, nurses may under-administer and patients may be reluctant to take the drug. Some doctors and nurses believe that it should be reserved for use only when death is imminent, and patients fear that the prescribing of morphine means that they will die soon. However, it may be used safely for months or years.
Patients (and professionals) need reassurance that the therapeutic range of morphine allows for safe dose increases if required. Fear of dependence or tolerance to the analgesic effect of morphine is largely unfounded and should not be a reason to withhold morphine.
Emotional, spiritual and cultural matters should not be underestimated in their relevance to the perception of pain: helping a patient to come to terms with underlying psychosocial concerns is as important as good dose titration. Where morphine has been thought to be ineffective, it is frequently because these issues have not been addressed or because the dose was too low or too infrequent, or because of failure adequately to address breakthrough pain.

Opioid rotation

In practice, some patients unable to tolerate morphine have benefited from a switch to another strong opioid. Switching from one opioid to another to produce improved analgesia with reduced side effects is termed opioid rotation. Drugs used in this way include fentanyl, hydromorphone, methadone and oxycodone. Possible reasons for the reported benefits include differing pharmacokinetics and pharmacodynamics, agonist effects at different receptors, and incomplete cross-tolerance.

Pharmacokinetic differences Oral opioids are generally rapidly absorbed, although some are subject to first pass metabolism in the liver. Since there may be considerable variation in relevant enzyme activity in different individuals, the effective dose for a given patient may be difficult to predict. Most opioids are well absorbed from subcutaneous and intramuscular sites, avoiding first pass loss. The transdermal route is useful for fentanyl. Most opioids are metabolised to glucuronides and excreted renally. Metabolites were originally thought to be inactive, but it is now thought that some, especially morphine-6-glucuronide, do possess analgesic potency. Differing ratios of active to non-active opioid metabolites may contribute to the benefits seen in patients switched from one opioid to another.

Receptor subtype differences and incomplete cross-toleranceOpioid agonists produce their analgesic effect by binding to specific pain receptors, located primarily in the brain and spinal cord. Opioid drugs may have full agonist, partial agonist or antagonist activity at these receptors. Activity may be further complicated by the existence of different subtypes of opioid receptors, so it is possible for a compound to have an agonist effect on one receptor subtype, and a partial agonist or antagonist effect at another.
At least three different subtypes have been identifed: mu (μ), delta (δ) and kappa (κ) receptors. Most strong opioids, like morphine, are strong μ agonists. However, differing potencies at different receptor subtypes may have a role to play in the apparent benefits of opioid rotation.
Tolerance to the side effects of opioids frequently occurs and patients with previous exposure to opioids often suffer fewer of the side effects associated with initial treatment, or do so to a lesser degree. Cross-tolerance occurs when tolerance to one drug results in tolerance to another. This is sometimes incomplete with respect to some actions of the drug, whether beneficial or adverse.

Panel 1:Some pitfalls of opioid rotation Failure to use morphine correctly or properly to address its side effects Failure to address psychosocial concerns Failure to consider reasons for the increase in pain Failure to consider other, possibly more appropriate, means of pain control Unpredictability in cross-tolerance and in dose conversions leading to an increase in adverse effects, such as increased sedation, with inadequate analgesia Poor understanding of the underlying mechanisms An inconsistent approach: "chopping and changing" which results in poor pain management

In opioid rotation, the trick is to find a second opioid for which cross-tolerance to the beneficial effects is less than cross-tolerance to its side effects. The mechanisms of cross-tolerance are not fully understood but are thought to be associated with the different receptor subtypes. Dose equivalents are sometimes unpredictable,³ and so careful management is required, particularly with higher opioid doses. Tables of potency ratios need to be carefully interpreted, particularly where high doses are involved (see Table 1).
There is a school of thought in the UK that adverse effects of morphine indicate that the dose should be lowered and non-opioid adjuvants added first, rather than switching to alternative opioids.⁴
Some opioids may also have activity at other sites: methadone is thought to have some NMDA (N-methyl-D-aspartate) antagonist activity, which may explain its good profile in neuropathic pain (see later).
Panel 1 lists some pitfalls of opioid rotation.

Alternative opioids in current use

Transdermal fentanyl is a useful alternative to morphine, particularly in those patients with severe unmanageable morphine-induced constipation, and those unable to swallow tablets. As fentanyl is often less constipating than morphine, any laxative dose should be halved and subsequently adjusted according to need, especially when switching from high doses of morphine. Some patients may experience debilitating diarrhoea as an effect of morphine withdrawal.
Transdermal fentanyl has a long half-life: steady-state plasma concentrations are achieved after 36-48 hours and the elimination half-life is 17 hours or longer. This needs careful consideration in dose titration and when changing patients from or to an oral or parenteral opioid. Dose conversions for fentanyl and morphine are such that 25mg/hour of fentanyl is equivalent to up to a 24-hour 135mg morphine dose (ie, 30-135mg/day). Table 2 shows appropriate oral morphine and subcutaneous diamorphine breakthrough doses when using fentanyl patches.

Table 1: Oral opioid potencies
Oral opioid	Potency ratio to morphine	Oral opioid	Potency ratio to morphine
Codeine	0.05	Diamorphine	1
Dihydrocodeine	0.1	Oxycodon	2
Dextropropoxyphene	0.1	Dextromoramide	2
Tramadol	0.1	Phenazocine	5
Pethidine (meperidine)	0.1	Levophanol	5
Dipipanone	0.5	Hydromorphone	7.5
Morphine	1	Buprenorphine (sublingual)	50
Fentanyl (transdermal)	70-150*

*manufacturer recommends 150 but in some patients the ratio may be only half this
Adapted from A Guide to Symptom Control. Regnard C, Tempest S. 4th ed. Hochland and Hochland, 1998.

Adapted from West Herts Community Trust palliative guidelines (with permission from Dr Kilian Dunphy)

Fentanyl patches are supplied in four strengths and should normally be changed every three days. They can be used in combination to achieve tailored doses. If more than one patch is used, they should be changed on the same day to avoid confusion. When treatment is first started the patient may need rescue doses of morphine for the first 24 hours.
If transferring from, or to, other opioid delivery systems, the 12-hour rule should be observed to ensure adequate analgesia without overdose. Since it will be at least 12 hours before the effects of transdermal fentanyl are felt, the first patch should be applied at the same time as the last modified release morphine tablet is given to ensure adequate analgesic cover. Breakthrough pain should be treated using morphine solution or fast-release tablets. The fact that there is no breakthrough preparation of fentanyl is a drawback, as often the reason for choosing fentanyl is to avoid morphine or the oral route, yet patients may require large doses of drugs for breakthrough pain. Ideally, a patch with a booster button would be a boon. Breakthrough preparations of fentanyl, eg, a nasal spray, are currently under trial.
Since the elimination half-life of fentanyl is around 17 hours, subcutaneous diamorphine should not be started until 12 hours after the withdrawal of the last patch and patients should be carefully monitored. Many areas have local guidelines for treatment switches, available through local specialist palliative care centres. Useful guidelines for fentanyl use appear in the current edition of the Palliative Care Formulary.⁵
A newer - or more accurately, re-introduced - alternative to oral morphine in the UK is hydromorphone. This drug has been widely used in the US for cancer pain. The oral analgesic potency of hydromorphone is about seven and a half times that of morphine.⁶ In the UK, hydromorphone is available as immediate-release capsules (1.3 and 2.6mg, equivalent to 10 and 20mg morphine respectively) and in a range of strengths as controlled-release capsules. These may be swallowed whole or their contents sprinkled on cold soft food. Patients with hepatic or renal impairment and the elderly may require lower doses so care is needed in titrating doses. Like morphine, hydromorphone may cause nausea and vomiting and constipation. These may be alleviated with appropriate antiemetics and laxatives. Several studies have demonstrated an improved side effect profile in patients unable to tolerate adverse effects of morphine, including a retrospective study by De Stoutz et al⁷ looking at toxicity reduction in 80 patients by opioid switch to hydromorphone and other opioids.
Methadone is both a μ and δ receptor agonist, and it is also thought to have NMDA-receptor antagonist activity⁸ (see ketamine). It has been successfully used at relatively low doses in some patients unable to tolerate morphine. It is useful in patients with renal failure who may accumulate morphine-6-glucuronide, developing excessive drowsiness and/or delirium. The analgesic and adverse effects of methadone are unaltered in renal failure. However, its pharmacokinetics are complex and both dose and interval need titrating for individual patients.
Oxycodone is a μ agonist similar to morphine. In suppository form it has provided a useful alternative to subcutaneous infusions. An oral formulation is expected to become available in the UK later this year. Despite the successes seen with opioid rotation in a small number of patients, its theoretical basis is still poorly understood. More research is needed to establish an evidence base in clinical practice. In the meantime, practitioners should consider carefully any underlying reasons for apparent morphine failure before switching medication.

Neuropathic pain

Although skill and expertise in the use of opioids is increasing, there remains a proportion of complex pain problems that are not so opioid sensitive. These often fall into the group known as neuropathic pain and refer to pain arising in situations where there is compression or damage to the nervous system, as opposed to nociceptive pain where the nervous system is intact. As many as 34 per cent of cancer patients have neuropathic pain.⁹ Pain arising in an area of the body where the overlying skin feels numb or oversensitive is often diagnostic of neuropathic pain and the patient often describes the pain as "burning", "stabbing" or "tingling".
In cancer patients, neuropathic pain may be caused by treatment, eg, brachial plexus arm pain secondary to radiotherapy, or it may be directly due to tumour invasion of nerves, eg, sacral recurrence of rectal cancer invading the lumbo-sacral plexus causing severe pelvic and sciatic pain. Before medication options are described, it should be remembered that other treatment modalities, such as TENS (transcutaneous nerve stimulation), acupuncture and nerve blocks may have a role in management.

Management of neuropathic pain

Ideally, the tumour causing the problem should be treated as the first-line approach to pain control (with anti-tumour modalities such as chemotherapy or hormonal manipulation). However, residual pain will need analgesia, and opioids should be tried first. It is now suggested that there is a continuum of opioid sensitivity¹⁰ and nerve pain is at the least sensitive end of the spectrum.
To achieve total pain control, adjuvants or co-analgesics (drugs whose primary function is not analgesic) are used together with the opioid. A wide range of agents has been tried for neuropathic pain (see Table 3).

Table 3: drugs used in neuropathic pain
Medication group	Drug	Starting dose
Tricyclic antidepressants	Amitriptyline	10-25mg at night
Anticonvulsants	Carbamazepine Sodium valproate Clonazepam	100mg bd 100-200 mg bd 0.5mg bd
Steroids	Dexamethasone	8mg daily
NSAIDs	Diclofenac	50mg tds
Antiarrhythmics	Flecainide Mexiletine	50-100mg bd 50mg bd
Others	Clonidine Baclofen Tramadol	50mg tds 5mg bd 50mg qds
NMDA receptor antagonists	Ketamine	50-100mg per day sc under specialist supervision
Other opioids	Methadone	5mg tds under specialist supervision

Evidence for tricyclic antidepressants and anticonvulsants as adjuvants has been reviewed by McQuay.^11,12 Amitriptyline, sodium valproate and carbamazepine are all in widespread use and remain the first choice of adjuvant therapy for nerve pain. Amitriptyline, with its mixed receptor activity, has an effect on pain that is independent of its antidepressant effect. The dose is titrated, depending on adverse effects, but often only 10-25mg at night is needed. The role of the newer antidepressants such as the SSRIs is more controversial in pain control, although there is a drive to prove their value as their side effect profile is an advantage over older drugs.
Most evidence supports carbamazepine as the first choice anticonvulsant for nerve pain, but sodium valproate may be effective and is better tolerated. Alternative anticonvulsants are being tried, such as gabapentin and clonazepam, but evidence is scant.
Class I anti-arrhythmic drugs, such as flecainide and mexiletine (also known as systemic local anaesthetics) are sometimes used in neuropathic pain. They are thought to work centrally by suppressing spinal cord neurones activated by pain fibre stimulation. Their efficacy has been proven in non-malignant nerve pain but is less convincing for cancer pain.¹³ High dose steroids, typically dexamethasone 8 to 16mg per day, may have an analgesic role by reducing peri-neural compression, but toxicity precludes their long-term use. Other miscellaneous drugs in use include clonidine and baclofen and of increasing interest is the dissociative anaesthetic agent, ketamine. It is likely that in the future, an analgesic ladder solely for neuropathic pain will be developed to provide a logical framework.

Role of ketamine in neuropathic Pain

Ketamine is an antagonist of the NMDA (N-methyl-D-aspartate) receptor which is widely spread throughout the central nervous system, particularly the spinal cord. Stimulation of pain fibres in the periphery cause release of excitatory amino acids such as glutamate and aspartate, which in turn activate the NMDA receptor complex. A phenomenon known as "wind-up" then occurs,¹⁴ producing a magnified pain response which, clinically, is associated with features of neuropathic pain such as allodynia (pain produced by a stimulus that is not normally painful, eg, light touch) and hyperalgesia (an exaggerated and prolonged pain response to a mildly painful stimulus). If these features are present, suggesting that NMDA receptors have been activated, then NMDA antagonists such as ketamine may have a role in inhibiting the response.
Ketamine has been used as an anaesthetic for over 30 years, but only for the past few years as an analgesic in palliative care. (Analgesic properties were recognised in the 1970s.) The analgesic dose is actually only 100th of the anaesthetic induction dose and, with doses currently in use in palliative care, there is little sedation.
Ketamine itself is poorly absorbed by mouth - oral bioavailability is 16 per cent compared with 90 per cent by the parenteral route. It is metabolised in the liver to norketamine, which is a potent active metabolite and, interestingly, norketamine levels are higher with oral than parenteral administration. This may explain the increasing reports of ketamine being used effectively as an oral analgesic.¹⁵ However, most experience is with a continuous parenteral infusion. Doses vary but most reports use 100-600mg/24 hours. Case studies have reported ketamine to be beneficial in many instances of benign neuropathic pain, AIDS neuropathy, spinal pain, skin pain from burns and complex cancer pains.¹⁶ Adverse effects are unusual at doses of less than 300mg per day, but they include vivid dreams, disorientation and dizziness, which may force discontinuation of treatment. These adverse effects may be helped by small doses of midazolam (5 to 15mg/24 hours) or haloperidol (2 to 4 mg/24 hours). If a patient is already taking opioids when ketamine is started, the opioid dose should be reduced by a third as ketamine can increase the "sensitivity" of the nervous system to opioid drugs¹⁷ and toxicity will ensue. If using a subcutaneous infusion via a syringe driver, the ketamine should be mixed with saline rather than water. Prolonged subcutaneous use can be very irritant to the skin, requiring frequent changes of needle site. For oral administration, the injection solution has to be used and the bitter taste disguised with juice. At present, ketamine is not licensed in the UK for analgesic use or for oral administration. It is only available on a named patient basis in the community from the manufacturer, Parke-Davis (tel 01703 620500). The company can be contacted for further supply details. There have been no placebo controlled trials using ketamine for cancer pain, just case reports involving a small number of patients.¹⁸ It remains a specialist drug, only to be commenced by palliative care physicians, but its use is increasing.

Bone pain and bisphosphonates

Bone metastases are extremely common in cancer and, at any one time, up to 20,000 women in the UK may be living with the pain of bone secondaries from breast cancer. In addition, about 85 per cent of men with advanced prostate cancer and 65 per cent of lung cancer patients will also develop bone metastases. So the scale of palliation needed to maintain the quality of life in these painful scenarios is vast, especially as the survival time in breast and prostate cancer may be several years.¹⁹
The first line treatment of bone pain is usually radiotherapy if the patient is well enough. With regard to medication, a combination of NSAIDs and opioids will usually be an effective combination. However, in recent years, an increasing amount of attention has been paid to bisphosphonate drugs.
This group of drugs was developed industrially to bind calcium in various chemical processes and they have found a wide range of use, for example, as descalers and constituents of toothpaste. In the past 20 years, it has been found that bisphosphonates are powerful inhibitors of bone resorption, reducing the activity of the bone cells known as osteoclasts. Excessive osteoclastic bone cell activity is the key pathological feature in all bone metastases, causing "lytic" holes in the bones as in breast cancer or "sclerotic" areas of thickened abnormal bone as in prostate cancer.
The role of bisphosphonates has developed from treating benign bone disease (such as osteoporosis and Pagets disease) to having multiple uses in malignant bone problems. They are now the treatment of choice for the hypercalcaemia of malignancy as well as being key agents in the analgesic management of bone secondaries.
There is also now evidence that bisphosphonates actually reduce the skeletal complications of metastases, such as fractures in breast cancer and myeloma, reducing the need for radiotherapy and surgery (see Table 4). Lipton's group looked at over 600 women with advanced breast cancer receiving hormone or cytotoxic therapy, who had bone secondaries.²⁰ Half were given a placebo infusion for two years and half received monthly pamidronate boluses. The survival of both groups was the same, but skeletal problems such as vertebral collapse and spinal cord compression were significantly reduced in the treated group. Pain and analgesic use were also decreased. Studies with prostate cancer also suggest an analgesic role.²¹

Table 4: Role of bisphosphonates in malignancy
Clinical situation		Clinical value
Treatment of hypercalcaemia of malignancy in cancer		Proven Treatment of choice
Prevention of skeletal morbidity		Proven in myeloma
		Positive data in breast cancer
		No data in other tumours
		Intravenous route preferred
Relief of bone pain		Positive data in breast cancer
		Little data in prostate cancer
		Intravenous route required
Treatment of osteoporosis		Increases bone mass
		Probably prevents fractures
		Oral therapy sufficient
Prevention of bone metasteses		Speculative
		Animal data only

Reproduced and adapted from reference 19 with the permission of Leeds medical information

At least eight bisphosphonates of varying potency are available but most of the trials and current experience involve clodronate and pamidronate. Both of these drugs are now licensed for supportive treatment in myeloma and breast cancer. They are usually given by intermittent intravenous infusions (eg, 90mg pamidronate four-weekly) on an outpatient basis. Newer, more potent variations such as ibandronate are under investigation. An oral preparation of clodronate exists but oral absorption is poor; even on an empty stomach less than 2 per cent of the dose is absorbed. The economic implications of using bisphosphonates more systematically for all ases of metastatic bone disease are huge,²² as drug costs alone would be over £2,000 per year, per patient. In the US, bisphosphonates are already widely used in breast cancer but in the UK it is fairly random as to whether a patient might be offered regular treatment.
Further cost benefit analyses are underway.

Conclusion

Significant progress has been made in treating pain and in other symptom control over the past two decades. Consistency of approach is essential to ensure that patients receive the benefits of these improvements, whatever the care setting, and to avoid risk of confusion among prescribers. The principles of assessment and review should continue to underpin the management of pain control in cancer.
In October, 1996, the NHS Executive issued guidance²³ on continuity of palliative care services and stressed the need for co-operation at all levels across primary and secondary care and between statutory and voluntary sectors.
The guidance recognises that continuity of care will be assisted by locally agreed core formularies and guidelines, together with locally developed schemes to ensure that medicines used in palliative care are available when they are needed.
In January, 1997, the Royal Pharmaceutical Society's working party report on pharmaceutical care of cancer patients in the community²⁴ published in the wake of the Calman-Hine report on commissioning care care services,²⁵ recognised the contribution that pharmacists can and should make to this continuity of care.

Mrs Allen is clinical pharmacy adviser at the Hospice of St Francis, Berkhamsted and chairman of the Hospice and Palliative Care Pharmacists Association and Dr Taylor is medical director at the Hospice of St Francis, Berkhamsted