Show Summary Details
Page of

Pharmacological approaches to pain. 1: ‘By the ladder’—the WHO approach to management of pain in palliative care 

Pharmacological approaches to pain. 1: ‘By the ladder’—the WHO approach to management of pain in palliative care
Pharmacological approaches to pain. 1: ‘By the ladder’—the WHO approach to management of pain in palliative care

Richard D. W. Hain

and Stefan J. Friedrichsdorf

Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2016. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 20 January 2020


This chapter sets out some of the principles of pharmacological pain management in children's palliative care. The fundamental precept that links them is that of simplicity. The default intervention should always be the simplest one that will achieve the desired therapeutic effect. This is arguably an important principle for good clinical care generally, but it becomes paramount during the palliative phase, when the focus is on carefully weighing the potential benefit of any intervention against its burden. Treatment will inevitably be made more complex by polypharmacy, by the use of intravenous, epidural, or intrathecal routes rather than enteral, transcutaneous, or subcutaneous ones, or by the prescription of esoteric medications that are unfamiliar to those providing bedside care.

That is not to say that such approaches are never necessary. For individual children the balance of benefit and burden may make complexity unavoidable. However, the default approach should always be the simplest effective one. What follows is an approach based on these principles. It also considers some of the situations in which that approach might not be sufficient, and some of the alternatives that should be considered.

The principles of the WHO ladder approach: what it is

In the 1980s, the World Health Organization (WHO) identified a global problem in managing pain in adults with cancer.1 A number of uncertainties and misconceptions were recognized regarding pain and how it should be treated. Prominent among these was a widespread concern about the use of major opioids. At that time, major opioids were often seen primarily as drugs of addiction that should usually be avoided unless there was no alternative. Children were, by common consent, considered particularly vulnerable to the adverse effects of opioids, and therefore these drugs were often withheld. This overcautious approach was often justified by early studies which seemed to suggest that children experienced pain less intensely than adults.2

In attempting to address some of the confusion and misunderstanding, the WHO drew up a simple and rational stepwise approach to the management of cancer pain in adults.1 The guidelines were subsequently republished, with little modification, for children.3 The WHO approach (see Figure 20.1) is based on the assumption that, for most children, pain will gradually increase as their illness progresses, and that this increase in pain intensity should be matched by the stepwise introduction of progressively stronger analgesics. This is particularly true for children with cancer; there will usually be a more gradual increase in pain intensity in children with non-malignant life-limiting conditions.

 Figure 20.1 Pain management in palliative care: the WHO pain ladder. Reproduced from with permission of the World Health Organization.

Figure 20.1
Pain management in palliative care: the WHO pain ladder. Reproduced from with permission of the World Health Organization.

On the first step are simple analgesics, on the second step are minor or weak opioids, and on the third step are major or strong opioids. One aim of this model is to avoid cycling through alternative medications of the same potency as an alternative to selecting a stronger class of drug. If step 2 is no longer effective, a major opioid on step 3 is required.

It is perhaps best to consider the guidelines to be the articulation of important basic principles in a framework, rather than a set of detailed recommendations. The four principles that underpin the WHO recommendations are often summarized as follows:

  • By the ladder—enabling a stepwise approach to treatment, commencing with non-opioids and increasing to strong opioids (see Figure 20.1). The level at which a child enters the ladder is determined by the child's needs, the intensity of pain, and the response to previous treatments.

  • By the clock—regular scheduling ensures a steady blood concentration, reducing the peaks and troughs of ‘as needed’ (PRN) dosing.

  • By the appropriate route—use the least invasive route of administration. The enteral (oral or gastrostomy) route is usually convenient, non-invasive, and cost-effective.

  • By the child—individualize treatment according to the child's pain and response to treatment.

The WHO pain ladder: what it isn't

The WHO ladder approach to pain was the result of expert consensus, rather than having a basis in evidence.4 Empirical evidence for its validity5 is sparse, but is available even in children.6 This has led to important debates about the premise of the guidelines,4 some of which are addressed later in this chapter.

However, other reservations stem from a lack of understanding of the guidelines and their purpose. It has been pointed out, for example, that they are focused on pharmacological approaches.7 It is self-evident in palliative care that pain, like all symptoms, occurs simultaneously in all domains of a child's experience. It will always have ramifications in emotional, psychosocial, and existential/spiritual domains. Furthermore, problems that occur in any of these other domains will also influence pain. Pain is always ‘total’8 and pharmacological management is, of course, not the sole or even always the most important approach to managing it.9 Comprehensive pain control for children requires tailoring to the needs of the individual child, and integrating pharmacological approaches (non-opioids, opioids, adjuvant analgesia) with non-pharmacological supportive and rehabilitative methods.

The WHO approach does not aim to address these broader issues. Its intention is to set out a simple, logical, and accessible approach to the pharmacological elements of pain relief in palliative care. This is not a drawback of the pain ladder, but simply a statement and recognition of its range.

Some authors have sought to modify the ladder—for example, by adding a fourth step that includes neurolytic procedures or patient-controlled analgesia (PCA).10 Arguably, this misses the point of the WHO ladder, which is that it is a framework of principles on which pain-relieving procedures that are potentially relevant to palliative care can, by definition, already be accommodated. Neurolytic procedures are, for example, effectively an adjuvant therapy that should be considered at each step, rather than interventions that should be reserved for when major opioids fail. Similarly, when properly used, PCA is not an alternative to the WHO ladder, but a further way to deliver it. It does not replace step 3, but is rather a way of ensuring both regular background analgesia and rapid access to breakthrough analgesia when it is required.

Any procedure that is accommodated by the WHO pain ladder, or which can be modified so that it is so accommodated, should be considered to have a potential place in the management of pain in children's palliative care.

‘By the clock’: selecting the right frequency

One of the fundamental principles of the WHO pain ladder for children is that opioid medication should always be given ‘by the clock’ rather than being available only when it is needed. It is an important basic principle of pain management in palliative care that strong opioids should always be given regularly rather than only ‘as necessary.’

One of the defining principles of management of pain in palliative care is that its primary intention is to prevent the child from experiencing pain in the first place. Any prescription that relies on the occurrence of pain to trigger an analgesic intervention is inherently unsatisfactory. To allow access to analgesia to be contingent upon reporting the need for it means that a child must inevitably experience pain and can never be free from it.

Furthermore, it is comparatively unusual for pain that, at its worst, is intense enough to need a major opioid, to disappear completely at other times. There are exceptions to this, particularly when managing episodic pain (see below), but in general, if pain is severe enough to warrant major opioids, it is likely to require them to be given on a regular basis. The aim is to secure and maintain a steady serum concentration of analgesic that provides adequate pain relief for most of the time. Prescriptions that allow opioids to be given only ‘as needed’ result in a cycle of under-medication and pain, with periods of over-medication and toxicity.11 This is particularly relevant with regard to the use of step 3—that is, major opioids.

There are sound theoretical reasons for always giving major opioids regularly. There is evidence12 , 13 that single doses of morphine are considerably less effective than repeated doses, with the effectiveness of the first dose of morphine being less than that of the fifth or sixth regular dose. The explanation for this is thought to be accumulation of the active metabolite, morphine-6-glucuronide (M6G), which has a longer half-life and therefore accumulates after repeated dosing. The analgesic mechanisms of morphine and M6G may not be the same.14 Regular repeated doses are therefore usually considered mandatory when prescribing major opioids.

The exact dosage interval is largely governed by the half-life of the medication and its duration of action. The half-life of morphine in children is probably somewhat shorter than that in adults,15–18 but the significance of this is unclear when it is compared with the wide variability observed between individual patients.19 , 20 Certainly in practice a 4-hourly dosing schedule for immediate-release oral morphine seems to work well in children, as it does in adults.

Exceptions to the ‘by-the-clock’ rule

There are apparent exceptions to the 4-hourly ‘by-the-clock’ rule. Where the half-life of a drug is increased, the dosage interval should also be lengthened to accommodate it. For children in the neonatal period, for example, renal clearance of morphine and its active metabolite M6G is less than that in older children or adults.21 For children with poor renal function from other causes, such as those dying from renal failure, clearance of morphine may also be reduced, and adjustments should be made to the dosing interval. One approach is to reduce the dosing interval of the immediate-release morphine to 8- or 12-hourly. However, if renal dysfunction is particularly severe, it may be preferable to give morphine as needed, with no regular dose at all. Although this appears to break one of the golden rules of the WHO pain ladder, reduced clearance of morphine and M6G means that the child will maintain an acceptable serum level, even with intermittent dosing.

In summary, when considering the appropriate frequency with which analgesia should be prescribed, the most important characteristics are those of the drug itself. However, it is also important to consider the characteristics of the patient, in particular as a result of developmental considerations or coexisting disease that may have an impact on clearance of the drug.

Breakthrough pain: fluctuating pain vs. insufficient background

An important distinction needs to be made between two situations in which additional analgesia may be required in the face of regular analgesia. The aim of pain management in palliative care is to match the dose of analgesia to the level of pain. The corollary is that pain can be said to occur because there is a mismatch between the two. This may, of course, be because the background analgesia is not quite adequate. However, it may also be because the underlying pain itself is fluctuating. The most obvious example would be incident pain (e.g. severe pain caused by movement as a result of pathological fracture or hip dislocation), but it can also be the result of pain that inherently fluctuates in intensity (e.g. colicky abdominal pain).

In palliative care, terminology has not always made the distinction between pain that is the result of inadequate background, and pain that is inherently fluctuating in intensity.22–24 Breakthrough pain has been usefully classified as incident, idiopathic, or end-of-dose failure,22 and these terms will be used for the purposes of this chapter.

Incident pain

Incident pain22–24 refers to pain caused by movements that would not normally be painful. It is particularly characteristic of pain caused by pathological fractures in children with cancer, or by dislocation, particularly of the hips, in children with non-malignant life-threatening conditions.

The ideal agent for relieving incident pain would have a rapid onset and a rapid offset of action. That is to say, it would be immediately accessible to the patient, able to deliver powerful analgesia quickly, and then be rapidly eliminated.

Unfortunately, no such ideal agent is available at present. Practical limitations imposed by formulation mean that, in practice, it is usually difficult to ensure that the peaks of pain and analgesia coincide.

Inhaled Entonox, a combination of oxygen and nitrous oxide, has been used to treat short-lived pain in children, with considerable success.25–27 Unfortunately, it can be difficult to arrange access to Entonox in the home. Addiction and megaloblastic anaemia are listed among its adverse effects. It is also unsuitable for a significant proportion of children with life-limiting conditions, as it requires both the physical and cognitive ability to coordinate inhalation.

Buccal diamorphine or oral morphine are simple to administer in the home, and are often adequate if given in anticipation of a procedure that is known to be painful. However, the time to peak effect is likely to be at least 20 minutes, even if the drug is made immediately available to the child.

Sublingual fentanyl28 , 29 and its congeners are more rapidly absorbed across the oral mucosa, and offer a combination of rapid absorption and ease of administration. They are probably the most effective opioid agents currently available for incident pain.

Where it is possible, orthopaedic intervention can be dramatically effective in avoiding incident pain. Neurolytic procedures may also have a place in management of the individual patient. These adjuvant procedures should always be considered alongside other breakthrough measures.

Idiopathic breakthrough pain

Idiopathic breakthrough pain refers to pain that breaks through with no obvious cause. It is in effect a diagnosis of exclusion, and in the absence of evidence linking it with incident pain, it should probably be treated as a form of end-of-dose failure (see below).

End-of-dose failure

The intention of regular background opioids is to maintain a serum concentration that is largely adequate to prevent pain occurring. It is rarely possible for this to be complete, and indeed for patients who never need breakthrough analgesia it might be inferred that the background dose is higher than they need. However, if the need for breakthrough doses is frequent, this indicates that serum concentrations of opioids are falling below the levels needed to maintain analgesia in the outpatient at that particular time. This describes the phenomenon of ‘end-of-dose failure.’ By convention, two doses or fewer in any two consecutive 24-hour periods are usually regarded as acceptable, while any more frequent requirements for breakthrough indicate that the background dose should be reviewed and increased.

The ideal agent for management of end-of-dose failure would also have a rapid onset, since the aim is to minimize as far as possible the length of time for which the patient has to experience pain. However, its offset does not need to be rapid, and indeed it could be argued that a longer half-life is desirable, as it reduces the risk of further pain occurring until the dosage of background analgesia is reviewed.

The first-line opioid for end-of-dose failure is enteral morphine. Its pharmacokinetics are relatively predictable and well understood, and it is simple to use. Its opioid-receptor profile is complementary to that of fentanyl, and it is usual practice in palliative medicine to prescribe morphine for breakthrough pain, even when fentanyl is used for background analgesia.

Methadone has been used for breakthrough pain in patients with cancer,30–35 with good effect. Methadone ‘as needed’ is probably safer than regular methadone, which is subject to unexpected accumulation,36 but it still needs to be used with some caution if the patient is requiring frequent doses. Common practice in some centres, in adults and children, is to use methadone as needed alongside an alternative opioid for background analgesia, in order to ensure that the methadone concentration does not exceed safe serum levels. As always, this should be done under careful supervision.36

PCA offers the patient immediate access to analgesia at the time of end-of-dose failure, and short time to peak concentration. It also facilitates review of the background dose by electronically recording the doses that the patient has required. It is important that some form of regular analgesia is maintained alongside these breakthrough doses, either as a background infusion of the PCA or by some alternative means, such as a transcutaneous patch or slow-release enteral formulation.

‘By the appropriate route’

Oral or other enteral routes

The enteral route is always preferable if it is available and practical. Enteral medications can be administered without advanced skill, and are relatively easy to titrate. They must be acceptable to children, so important considerations are palatability, tablet size, solution volume, and frequency of administration.

Palatability has been improved by the manufacture of better-tasting flavoured vehicles, and a variety of syrups, ice creams, jams, and sauces can be used. Manufacturers have moved away from only providing medications in tablet form, which made it difficult for children who were unwilling or unable to swallow tablets, and have developed concentrated solutions for children who require high doses of a medication. The frequency with which a medication needs to be administered can be prohibitive to children. In general, analgesics that need to be taken every 4 hours are impractical for anything but short-term administration. However, many analgesics are now available in sustained-release preparations. The preparation of some products means that even the contents of an opened capsule can be mixed with soft foods without loss of the slow-release formulation. However, if these granules are actually chewed, they will in effect become immediate-release preparations, so this often remains an unpredictable approach.

The enteral route is not appropriate if the child:

  • prefers a different route of administration

  • is actively vomiting

  • is unable to comply (e.g. because they are drowsy or unconscious)

  • has significant gastrointestinal or swallowing dysfunction, with a risk of aspiration, as might be seen in children with neurological impairment

  • is experiencing a severe pain crisis, when parenteral administration may become necessary for rapid titration.

Increasingly, children with life-limiting conditions will have a gastrostomy, particularly outside ACT/RCPCH category I. Although this simplifies enteral access, in that it avoids problems of palatability, it can give rise to new problems, as the physical characteristics of some formulations mean that they cannot be given by this route. An important area for study in children's palliative care would be the development of ‘gastrostomy-friendly’ formulations of important symptom control medications.

Subcutaneous route

Parenteral routes should be used when rapid titration is required or when a child is unable to tolerate the enteral route. Several types of device are available to provide parenteral delivery of analgesics by continuous infusion, intermittent boluses, or both. Opioids can be used alone or administered in combination with other medications, such as anti-emetics, benzodiazepines, and corticosteroids, among others. PCA is a refinement of this approach.

The pharmacokinetics and efficacy of subcutaneous delivery are equivalent to those of intravenous delivery (see below). The usual volume limit for a subcutaneous infusion is 3 ml/hour, but this does not normally pose a problem, because of the availability of many highly potent and/or soluble opioid preparations, particularly diamorphine, but also hydromorphone and oxymorphone.

A short-gauge indwelling non-metallic needle system, most often placed over the upper arm or abdominal wall, provides intermittent and/or continuous delivery of medication. The chest can also be used as a site for needle placement, but care must be taken, as there is a small risk of causing a pneumothorax. If irritation, erythema, or induration occurs at the insertion site, the needle site should be changed. There are no firm rules as to how long a needle should remain in situ. Anecdotally, non-metallic needles have been successfully used for periods of up to 21 days. However, the site should be checked on a regular basis and carefully examined in the event of poor pain control, as inflammation of the area may be uncomfortable, as well as impairing drug delivery.

Intravenous routes

The need for repeated and usually increasingly difficult re-siting of a peripheral cannula, combined with the challenge of safely maintaining it outside a hospital environment, mean that this route is usually impractical for a child who is being cared for at home. A long-term percutaneous catheter or other central venous device is often used in children who have difficult venous access or a need for long-term therapy, such as blood products or parenteral nutrition. It can also be conveniently used for palliative interventions, although this should be balanced against the risk of repeatedly accessing the line.

Fortunately, the much simpler subcutaneous route is equally suitable for most medications in the palliative phase. The intravenous route, whether central or peripheral, is rarely necessary.

Other routes

One of the keys to delivering effective medication to children is to find a route that is acceptable to them. The transdermal route has provided a valuable alternative to injections when enteral dosing is not possible.

A number of other routes have also been used in paediatric palliative medicine. The transmucosal route has been used, particularly for diamorphine and midazolam. It avoids first-pass metabolism, and can provide rapid intervention for breakthrough pain, anxiety, or dyspnoea. Buccal (transmucosal) midazolam can be of particular value in breaking the cycle of acute shortness of breath and anxiety.

A variation on the buccal transmucosal route is the nasal approach. This has been used for many years to deliver some hormones, and again has found a place in delivering midazolam, either through a spray37 or simply through a small-dose syringe with no needle.

Opioids delivered through a nebulizer have been the subject of considerable study in the management of dyspnoea.38 Although a consensus seems to be emerging that they are not generally effective, individual patients do seem to derive some benefit.39 The effects of nebulized local anaesthetics on cough have been examined in adults.40

Any face mask or nebulizer should be used with caution. Many children find it claustrophobic and intolerable, and bronchospasm may, albeit rarely, complicate administration via any nebulized route.

Other routes

In many cultures, children dislike rectal administration. However, the rectal route can be useful as a means of avoiding parenteral administration in the short term if the child is unable to tolerate enteral analgesics. The strength of analgesic suppositories is limited, but enteral sustained-release or immediate-release preparations can be given via the rectum at the usual intervals. The rectal route is usually contraindicated during episodes of neutropenia and thrombocytopenia, because of the risk of infection and bleeding, respectively.

Intramuscular administration can rarely be justified in children, given the number of alternatives that are now available. Although the absorption profile of analgesia given by this route means that a single bolus can deliver sustained serum levels for longer than other parenteral routes, it is inevitably painful to administer. This forces the child in pain to have to consider whether the pain that they are already experiencing is severe enough to outweigh the discomfort of the remedy. This is an invidious choice that they should not have to make, and it ensures that many children will remain in pain.

‘By the child’: finding the right dose

One of the central principles of pain management and palliative care is that the correct dose of the analgesic is that which achieves effective pain relief, and is well tolerated, in an individual patient. In this respect, analgesic prescribing in palliative care is rather different from the way in which medications are used in other contexts, where the expectation is typically that one standard dose per kilogram will be enough to achieve the desired effect.

The initial prescription for analgesics is not necessarily even a dose that is expected to be effective, but a starting point from which it is expected that titration will occur until the pain is under control. It is often helpful to make this clear to the child and their family, so that they do not feel discouraged if the initial prescription is not quite enough to control the pain. In effect, the prescription of major opioids involves three phases.

  1. 1 Initiation. A starting dose is selected, either on a dose-per-kilogram basis or as a conversion from existing opioid requirements.

  2. 2 Titration. During this period, which typically lasts for a few days, the aim is to match the degree of pain with enough drug to provide analgesia, but without exceeding this and incurring unnecessary adverse effects.

  3. 3 Maintenance. During this period, which often lasts for several weeks, months, or even years, a reasonably stable dose of medication has been reached.

The distinction between the three phases is not quite as clear as this might suggest. In practice, the maintenance phase often represents a period of slow titration. Disease progression (and perhaps opioid tolerance) means that opioid requirements will change, and a process of continual review is necessary even during the maintenance phase, to ensure that adequate analgesia is achieved.

It is occasionally necessary to telescope the titration phase into a few hours or minutes by slow and careful infusion of parenteral opioid in cases where there is severe pain that requires urgent intervention (see later). This is a relatively rare procedure in children, but has been described more often in adults.41–43

Each of the three phases—initiation, titration, and maintenance—is characterized by specific considerations of dosing and formulation.


Anxiety often surrounds the initial prescription of major opioids, particularly for those who have relatively little experience of using them in children. In practice there are two ways to achieve a safe and appropriate initial dose.

Dose per kilogram

If the child is not already receiving opioids, the dose should be calculated on the basis of the child's weight.

Conversion from existing opioid dose

For children who are already receiving opioids (minor or major), there is likely to be some tolerance, and a dose-per-kilogram approach will often underestimate the child's true requirements. Instead, the dose of opioids already required by the child should be used as a guide to the appropriate initial dose of the new drug. Where this occurs as a child moves from step 2 (minor opioids) to step 3 (major opioids), this is complicated by the fact that the need to move to step 3 implies that the child's pain is not yet adequately controlled. In that situation, it is often useful to calculate the dose using both methods. Where the two calculations yield different doses, the higher figure is usually the more appropriate one.

Calculating the initial dose using a ‘dose-per-kilogram’ approach

In countries where major opioids are available to children, most formularies will offer a suitable dose per kilogram of the child's weight. This is based on the assumption that the volume of distribution per kilogram is the same in children as it is in adults. For example, if 10 mg of morphine given to a 70 kg adult results in a suitable and effective serum concentration, then it is assumed that half that dose given to a 35 kg child will have the same result. This assumption appears to be correct.15–17 Certainly, in practice, dosing guidelines based on this approach seem to work reasonably well.

Local formulary recommendations vary, but most recommend as a starting dose an equivalent to enteral morphine 0.5–1.5 mg/kg/24 hours. There is relatively little direct evidence from paediatric studies, but one study seems to suggest that this will usually result in adequate analgesia with little toxicity.16

The importance of prescribing regular and breakthrough medication has already been emphasized. Immediate-release enteral morphine is the preferred first-line major opioid, and one-sixth of the total daily dose should be prescribed regularly 4-hourly. It is often undesirable and unnecessary to wake the child to give them the night-time dose, and some clinicians will double the dose before bedtime to make up for this missing dose.

The breakthrough dose should be the same as the regular 4-hourly dose (i.e. one-sixth of the total daily dose). It is important to explain to families that although the breakthrough dose and the regular dose are the same, they perform two different functions. The regular dose is ‘to try to keep the pain away’, and the breakthrough dose is ‘to treat pain if it happens despite that.’ This is important, because without this understanding the parents may withhold a breakthrough dose if it is needed just before or just after a regular one, rather than giving the extra dose in addition.

The breakthrough dose serves two purposes. It ensures that analgesia is available to the child should the regular analgesia be inadequate. It also provides some measure of the child's requirement for such additional analgesia, which allows a process of rational and safe titration.

The frequency with which breakthrough medication is made available varies from one centre to another. Traditional practice was to offer the breakthrough dose as needed 4-hourly, but increasingly it is being offered as often as is necessary, even up to hourly. Once a breakthrough dose has been given, there is perhaps little to be gained by giving a further dose within an hour of a previous one, since it can take 30–60 minutes for the effect of an enteral dose to become apparent.

Opioids should usually be started enterally, unless there is no alternative. Where it is thought necessary to commence them using a parenteral formulation, standard dosing protocols are usually available, or again a dose can be calculated by conversion from any existing opioid requirements. Rarely, it may be necessary to intervene more urgently, and the appropriate dose can be established on the basis of rapid titration to the child's requirements (see below).

Calculation of an initial dose by conversion from existing opioid requirements

A further important and fundamental concept in the pharmacology of analgesia in palliative care is that of analgesic equivalence among opioids. Most major opioids work in the same way on the same group of receptors, but with differing potency. This means that the analgesic effectiveness of any opioid can be expressed in terms of how it compares with other opioids. For example, fentanyl is 75 times as potent as morphine when both are given parenterally.

The route should also be taken into consideration. Morphine is twice as potent when given by the parenteral route as it is when given orally, so parenteral fentanyl is 150 times as potent as oral morphine.

The concept of analgesic equivalence can also be extended to minor opioids. For example, codeine is approximately one-tenth as potent as morphine, whereas pethidine is about one-sixth as potent. Equivalency can be made more complex if an opioid has more than one analgesic action. For example, enteral tramadol has approximately one-fifth the opioid potency of morphine, but has additional non-opioid analgesic properties that make its effects less predictable.

By convention, the potency of all opioids can be expressed in terms of their equivalence to enteral morphine (see Table 20.1). This enables appropriate conversions to be made not only between morphine and other opioids, but between different non-morphine opioids. For example, since hydromorphone is approximately 5 oral morphine equivalents (OME), and oxycodone is approximately 2 OME, it is clear that hydromorphone must be two and a half times as potent as oxycodone. A patient on oxycodone who wishes to convert to hydromorphone would therefore be expected to have the same pain relief if the dose were divided by two and a half (see Example 20.1).

Table 20.1 Some oral morphine equivalents (note that there is variability both in the published evidence and in individual patients, and the data are mainly from adults with cancer)


Potency relative to oral morphine


Morphine PO


Morphine SC or IV


Diamorphine PO


Diamorphine SC or IV


Fentanyl transdermal (OFTC) SC or IV


Absorption from mouth is transmucosal, not enteral

Hydromorphone PO


Hydromorphone SC


Codeine PO


Tramadol PO


Non-opioid analgesic effects may make it more potent in practice

Buprenorphine transdermally


Oxycodone PO


Oxycodone SC or IV




Complex, depends on dose (see text)

PO, orally; SC, subcutaneously; IV, intravenously; OTFC, oral transmucosal fentanyl citrate.

However, it should be borne in mind that demonstration of opioid equivalence is not an exact science. There is often a range of documented equivalence. As always, calculations based on published data do not obviate the need for careful review of effectiveness and tolerability in an individual patient.

Example 20.1 Conversion

A patient is receiving 15 mg of enteral oxycodone in 24 hours, and needs to change to hydromorphone. Published tables suggest that the relative potency of enteral oxycodone is 2 OME (i.e. twice the potency of enteral morphine) and the relative potency of hydromorphone is 5 OME (i.e. five times the potency of enteral morphine).

Oral morphine equivalent of oxycodone = 15 x 2 = 30 mg of enteral morphine.

Enteral hydromorphone equivalent = 30 ÷ 5 = 6 mg of enteral hydromorphone.

So 15 mg of enteral oxycodone are equivalent to 6 mg of enteral hydromorphone. Therefore the total daily dose of hydromorphone should in theory be 6 mg. In practice, the dose should be further reduced to take account of incomplete cross-tolerance.

Opioid substitution, rotation, and switching

It may sometimes become necessary to change from one major opioid to an alternative, ideally of a different class.18 , 44 , 45 Changes in opioid inevitably risk jeopardizing good pain control, and should be considered sparingly. In practice, substitution is usually for one of the following two reasons.

  • Pain control is not yet adequate on the current opioid with appropriate adjuvant therapy, but further increases in opioid dose are limited by toxicity, particularly neuroexcitability.

  • It becomes clear that an alternative opioid might offer specific advantages over the current one (e.g. converting from enteral morphine to the more convenient transcutaneous fentanyl patch, or introducing methadone to treat neuropathic pain).

The new dose is calculated on the basis of oral morphine equivalence. Changing to a new opioid allows a reduction in equivalent dose, because the child is not equally tolerant of the new opioid—a phenomenon termed ‘incomplete cross-tolerance.’ A child who has become partially tolerant of the analgesic effects of morphine, for example, may well be less tolerant of those of fentanyl. Understanding this phenomenon is central to an understanding of the effectiveness and management of opioid substitution.

The effectiveness of opioid rotation or substitution in this way depends in part on the different adverse effects profiles of different opioids. However, it is mainly due to the fact that incomplete cross-tolerance means that changing to a new opioid allows a reduction in the total opioid dose, without any loss of analgesia. The dose reduction is conventionally 25%. When converting from one major opioid to another, there are therefore two stages in the calculation (see Example 20.2).

  1. 1 calculation of an equi-analgesic dose of the new opioid, based on oral morphine equivalency (see Table 20.1)

  2. 2 a 25% reduction in that dosage in order to reduce toxicity.

If substitution is for a reason other than toxicity, incomplete cross-tolerance means that there is a theoretical risk of opioid overdosage occurring with the new opioid. In practice, this has not been observed in children.46

Rarely, it is necessary to titrate rapidly against pain in order to establish an appropriate starting dose. The indication is for severe pain for which a more measured approach would condemn the child to prolonged suffering. In this situation, the parenteral route is the most appropriate one. The opioid should be infused slowly over 30 minutes or so until analgesia is achieved. A number of methods can be used to calculate the total daily dose required once this has been achieved. Perhaps the simplest43 is to assume that it represents the equivalent of a single 4-hourly dose, and accordingly give six times this dose in 24 hours (see Example 20.3). This can be given by any appropriate route, and as any opioid, so long as appropriate conversions in dose are made.

Example 20.2 Opioid substitution

After titration, a child who is receiving a subcutaneous infusion of morphine receives 500 mg in 24 hours, but is becoming toxic, with neuroexcitability, sweating, and myoclonus. The decision in made to switch to an alternative opioid. Parenteral fentanyl (which has 150 times the analgesic potency of enteral morphine) is selected because it is a synthetic opioid with a structure different from morphine.

Step 1: Calculating the theoretical equi-analgesic dose of fentanyl

Oral morphine equivalent of subcutaneous morphine = 500 x 2 = 1000 mg of enteral morphine.

Parenteral fentanyl equivalent of 1000 mg of enteral morphine = 1000 mg ÷ 150

= 6.67 mg of parenteral fentanyl in 24 hours.

Step 2: Reducing the dose by 25% to account for incomplete cross-tolerance

25% of the parenteral fentanyl dose = 0.25 x 6.67 = 1.67 mg of parenteral fentanyl.

The final dose of parenteral fentanyl, taking into account both equi-analgesic potency in theory and incomplete cross-tolerance in practice, is:

6.67 – 1.67 = 5 mg in 24 hours.

(Cross-check: 5 x 150 = 750 mg oral morphine equivalent, i.e. 75% of the original 24-hourly opioid requirements.)


The purpose of the titration phase is to match the dose of analgesia prescribed with the degree of pain experienced by the patient. If pain is counteracted by adequate analgesia, it is also true that some of the effects of analgesia are countered by pain. The risk of clinically significant respiratory depression, for example, is small if the dose of opioid prescribed is not excessive in relation to the degree of pain experienced by the patient. Titration is the means by which the correct dose of analgesia is determined so as to keep the child comfortable without unnecessary toxicity.

The opioid of choice during the titration phase is enteral morphine. Other enteral medications with relatively short half-lives, such as hydromorphone, methadone, or oxycodone, could also be used. Formulations with long half-lives or slow-release delivery systems, such as fentanyl patches or slow-release preparations of morphine or oxycodone, would not be appropriate for titration.

The essence of titration is ongoing review of the regular opioid dosage, based on the amount of breakthrough medication that the child has required. Once appropriate initial doses of regular and breakthrough opioid have been selected, the prescription should if possible be left for 48 hours (see Figure 20.2) and then reviewed.

 Figure 20.2 Review cycle in titration and maintenance phase.

Figure 20.2
Review cycle in titration and maintenance phase.

If the child has required only one or two breakthroughs in each 24-hour period, the dose of regular analgesia is probably about right and no alteration needs to be made. If, on the other hand, the child has needed more than this, the total daily dose of the regular opioid prescription should be increased by the amount of breakthrough that has been required (see Example 20.4). It is imperative that each time the total daily dose of regular opioid is reviewed, the breakthrough dose should also be increased, so that it remains approximately one-sixth of the total daily dose. This is because as tolerance develops, the breakthrough dose will become progressively less effective unless it is increased in proportion to the regular dose.

After a further 48 hours, the process should be repeated and once again the total daily dose of regular opioid amended in line with the child's requirements for breakthrough opioid. In this way it is usually possible to find a 24-hour regular dose that allows the child to need only a small number of breakthrough doses.

This approach has a number of advantages. It allows the 24-hour regular opioid dose to be precisely titrated against the child's experience of pain. Furthermore, increases in regular opioid dose can be made with confidence, as they only reflect what the child has already received over the previous 48 hours with no ill effect. This can be reassuring for parent and professional alike.

The reason for delaying changes to the regular opioid dosage for 48 hours is to allow the opioid to reach steady state at the new dose. Too frequent changes in dosage can result in increases being made disproportionately to the degree of pain, so that the child in effect receives too high a dose of opioids and is therefore at risk of some of the adverse effects. However, there are clearly situations in which more rapid titration is necessary—for example, if the pain is severe, or if there is rapid disease progression such that the severity of pain is increasing faster than 48-hour titration would allow.

The role of PCA in palliative care is controversial, and is considered later in this chapter. When used to manage pain in palliative care, PCA must of course obey the fundamental principles of the WHO pain ladder. In an acute setting, PCA is often used with little or no background infusion. In palliative care, this would not be appropriate—it would not be ‘by the clock.’ However, PCA is potentially an ideal way to initiate and titrate opioids at step 3. It allows immediate access to analgesia when it is needed, and automatically provides a detailed record of analgesic requirements that can facilitate 48-hour review.

If titration of opioids fails to have an adequate impact on the pain, this may be because the pain is inherently wholly or partially insensitive to opioids. The first situation is extremely unusual, but the second is seen not infrequently. Neuropathic pain is one cause. Although opioids remain the most effective analgesic agent for neuropathic pain, additional approaches (e.g. adjuvants or neurolytic procedures) may improve the situation.

There are many other reasons why pain may be relatively resistant, which are beyond the scope of this chapter but are considered elsewhere in the book. Among these, perhaps the most important one to consider is that pharmacological management of pain tends to address mainly the physical aspects of pain. If physical elements are making only a small contribution to pain, the latter will typically respond only poorly to the approaches to prescription and titration described earlier.18 , 47 The WHO guidelines point out that ‘attempting to relieve pain without addressing the patient's non-physical concerns is likely to lead to frustration.’3

Example 20.3 Rapid titration

A child with severe pain related to a sarcoma requires 6 mg of intravenous morphine to achieve analgesia.

Oral morphine equivalent = 2 x 6 = 12 mg.

Assume that this represents the 4-hourly breakthrough requirement.

Total 24-hourly requirement = 6 x 12 = 72 mg oral morphine equivalent.

Example 20.4 Prescription, titration, and maintenance of opioids

Day 0: A child who weighs 30 kg commences 1 mg/kg/24 hours enteral morphine (i.e. 30 mg/24 hours). This is prescribed as an immediate-release enteral preparation of morphine. This regular dose is given as 5 mg every 4 hours. The additional breakthrough dose is one-sixth of the total daily dose (i.e. 5 mg), prescribed as needed enterally 1- to 4-hourly.

Day 1: The child requires four breakthrough doses, each of 5 mg.

Day 2: The child requires two breakthrough doses, each of 5 mg.

Review after 48 hours: The child has needed an average of three breakthrough doses each day, which implies that the regular dose is not yet enough and needs to be increased.

  • Regular dose: The increase in the 24-hourly regular dose should be 3 x 5 = 15 mg, so the new regular total daily dose is 30 + 15 = 45 mg/24 hours.

  • Breakthrough dose: The new breakthrough dose remains one-sixth of the total daily dose, so is 45 ÷ 6 = 7.5 mg prescribed as needed 1- to 4-hourly.

Day 3: The child requires two breakthrough doses, each of 7.5 mg.

Day 4: The child requires one breakthrough dose of 7.5 mg.

Review after 48 hours: The child has needed an average of 1.5 breakthrough doses each day, which implies that the regular dose is still not enough and needs to be increased further.

  • Regular dose: The increase in the 24-hourly regular dose should be 1.5 x 7.5 = 10 mg, so the new regular total daily dose is 45 + 10 = 55 mg/24 hours.

  • Breakthrough dose: The new breakthrough dose remains one-sixth of the total daily dose, so is 55 ÷ 6 = 9 mg prescribed as needed 1- to 4-hourly.

Days 5–8: The child requires only occasional additional breakthrough doses.

Review every 48 hours: Few additional breakthrough doses have been necessary, which implies that the opioid prescription is now well matched to the child's pain. This marks the end of the titration phase and the beginning of the maintenance phase.

  • Regular dose: Change to a more convenient formulation. The total daily morphine dose is 55 mg, which is approximately equal to 30 mg twice daily of 12-hourly slow-release morphine formulation.

  • Breakthrough dose: The breakthrough dose remains one-sixth of the total daily dose, so is 60 ÷ 6 = 10 mg prescribed as needed 1- to 4-hourly. This continues to be an immediate-release preparation.

Continue regular review process, including changing needs for breakthrough pain, tolerability of the drug and formulation, and the possibility of narcotization.

Maintenance and review

For most children, the titration phase will come to an end when it is no longer necessary to keep adjusting the total daily dose of regular opioid. At this point it is usually helpful to change to a long-acting formulation if this has not already been done. In paediatric palliative medicine, slow-release morphine preparations that can be given 12- or 24-hourly are first-line treatment, and in most countries fentanyl patches are second-line therapy. The relative potency of fentanyl means that the patches should really only be prescribed to a child who is receiving more than 30 or 40 mg of oral morphine equivalent for breakthrough pain in 24 hours. In practice, however, fentanyl patches seem to be well tolerated even when they represent an increase in the dose of opioid.

As has been seen, there is no theoretical basis for insisting that the opioid prescribed for breakthrough pain should be the same as that prescribed for background pain. There is indeed a theoretical advantage in combining opioids. For example, enteral morphine (primarily a µ1-opioid-receptor agonist) with fentanyl (primarily a µ-2-opioid-receptor agonist) may in theory give better overall blockade of receptors than either alone. Nevertheless, there is a countervailing argument in favour of simplicity and avoiding polypharmacy that should also be considered in individual patients.

The pharmacokinetics of some slow-release preparations of morphine are different in children to those in adults.16 It appears that the absorption of slow-release morphine formulations is erratic, and their slow-release nature may be less reliable. For this reason, slow-release preparations of morphine in children are sometimes given 8- rather than 12-hourly. This rather reduces their usefulness, and for children who require 8-hourly slow-release morphine it may be better to consider an alternative. Clinical experience is that in some children it is necessary to change fentanyl patches every 48 hours rather than every 72 hours, in order to avoid increasing requirements for breakthrough towards the time of the next patch change. Children and their parents sometimes report that the patch has ‘dried out’, and report a greater need for breakthrough doses in the last 24 hours before a patch change.

Of course, even during the maintenance phase the dose should be subject to review (see Example 20.4). For most children the maintenance phase is not a true plateau, but a much more gradual gradient as their pain increases. Further dose increases will be needed mainly as a result of disease progression, but perhaps also due to the development of tolerance.

Sometimes the dose may need to be revised downwards and reduced, rather than being increased. The need for such a decrease can be inferred from the sudden development of toxicity. A child who is receiving an appropriate dose of opioid for the degree of pain that they are experiencing will typically experience few of the side effects that are countered by pain, particularly drowsiness and respiratory depression.

Should adverse effects from opioids develop unexpectedly despite careful titration, the cause may be narcotization. There are three common reasons why narcotization occurs.

  • The pain has become suddenly less severe. This is usually because some other therapeutic intervention has been effective. For example, radiotherapy to a malignant metastasis may reduce or even abolish pain at that point completely. Similarly, a nerve block or epidural anaesthetic may provide effective analgesia such that a previously appropriate dose of opioid is now too high. Less commonly, progression of the disease itself can paradoxically provide symptom relief. For example, nerve damage may initially cause pain, but then as it progresses it may instead cause regional anaesthesia.

  • Clearance of the opioid may be acutely impaired. The commonest cause is a sudden deterioration in renal function, resulting in accumulation of morphine and its metabolites. The cause may not be immediately obvious. Renal function may be impaired both by the underlying disease and by some of the interventions, palliative or otherwise, that are introduced in its management.

  • Interactions with other drugs may be occurring. Polypharmacy is common during the palliative phase, even in the paediatric specialty. It is not always possible to predict the effect of a certain combination of drugs on a child's conscious level. The introduction of psychoactive drugs such as midazolam or phenothiazines may induce drowsiness that can masquerade as opioid toxicity.

Management of adverse effects

Opioids are the main pharmacological tool available for managing pain in palliative medicine, and the majority of children will have a favourable outcome. However, excessive adverse effects will be experienced by a significant minority,15 , 48 , 49 and when they occur these can jeopardize the optimal management of pain. Children are likely to refuse medication that causes distressing side effects, even if it is effective. Adverse effects should be anticipated, specifically addressed with the patient and their family, and relieved as rapidly as possible.

Most evidence-based practice is derived from research in adults.44 There is little evidence in children, but it seems reasonable to extrapolate these principles to paediatric patients, so long as this is done with caution.

The adverse effects can be dealt with in one of two ways:

  • a reduction in the total opioid dose (especially by substitution)

  • symptomatic management of the adverse effect.

Dose reduction of systemic opioids

Unless there has been a reduction in pain, it is unlikely that a simple reduction in the dose of opioid can be made without jeopardizing effective pain relief. However, the development of tolerance of the analgesic effects of one opioid does not necessarily mean that tolerance will have occurred equally to all of them. This provides the opportunity to reduce the total opioid dose by switching to an alternative opioid. This is more likely to be effective if the opioid is of a different class, as incomplete cross-tolerance is more probable. The technique of opioid ‘switching’ or ‘rotating’ was considered earlier in this chapter.

Symptomatic management of adverse effects

This involves the use of drugs to prevent or control adverse effects, and it adds to the child's medication burden. For example, all children who are prescribed a major opioid should also be prescribed stimulant and softening laxatives. Generally speaking, additional medications should be avoided if possible, as children may be reluctant to take yet another medication, and polypharmacy often increases the risk of further adverse effects or drug interactions.

The effects of underlying disease and of other drugs can mimic opioid toxicity. This is particularly true of neuropsychiatric problems such as drowsiness, cognitive dysfunction, and myclonus.44 As an initial step, it is important to ensure that opioids are indeed the cause. Substitution of a different opioid or replacement with an alternative, non-opioid approach may be required.

If these fail, it may be necessary to begin specific therapy. The agents recommended for symptomatic management of the adverse effects of opioids are often based on anecdotal experience and clinical observation. They lack support from prospective studies of efficacy or toxicity over the long term, or systematic evaluation of retrospective data.

Nausea and vomiting

This is probably relatively rare in children, although its incidence may be underestimated,15 and anti-emetics are not normally prescribed prophylactically. Alternative explanations should usually be sought if nausea or vomiting occurs on established treatment. Haloperidol is usually recommended for opioid-induced symptoms in adults.


Constipation is extremely common among children who are receiving opioids, and should be anticipated and prophylactic laxatives prescribed as soon as the opioid prescription has been decided. A combination of stimulant and softener should be used. Co-danthrusate (a combination of danthron and docusate), a combination of magnesium hydroxide and senna, or Movicol® are all effective. Lactulose, which is favoured by paediatricians in many countries for the treatment of constipation in children, is not appropriate for opioid-induced constipation.

There is good evidence that, if given enterally, the opioid antagonist naloxone may help to reverse opioid-induced constipation,50–56 although there is a risk that it will reverse analgesia as well.54 Combination formulations containing opioid and naloxone are available for adults,57–62 and have given mixed results. Other antagonists, particularly methylnaltrexone,63 have a place in the treatment of chronic opioid-induced constipation where laxative treatment has failed.64 Transdermal fentanyl65–67 and buprenorphine68 probably cause less constipation than morphine.


The management of pruritus caused by opioids is considered in more detail in Chapter 29. It is not uncommon in infants and young children,15 , 46 and seems to occur particularly around the nose and face. Substitution of an alternative opioid is usually effective.


The patient and/or their family should be informed that there might be an initial period of drowsiness when opioid therapy is commenced or increased, but this will generally subside within a few days. Unfortunately, occasionally the sedative effect persists, and contributing factors such as hepatic, renal, or central nervous system disease should be considered. Reduced renal function can result in the accumulation of morphine's centrally acting metabolite, M6G, and an opioid such as fentanyl, that is not as dependent on renal excretion, may be better tolerated. Opioid rotation has been found to have a positive effect on the prevalence and severity of drowsiness, as has a change from enteral to subcutaneous morphine.

Psychostimulants such as methylphenidate (familiar to most paediatricians in its role of managing hyperactivity disorders) have had a demonstrable effect in a number of adult studies. They have also been found to be effective in adolescents with cancer.69 However, in all of the studies the use of psychostimulants has been associated with adverse effects such as hallucinations, delirium, psychosis, decreased appetite, and tremor. Psychostimulants are contraindicated when there is a history of psychiatric disorders, and relatively contraindicated in patients with substance abuse problems or paroxysmal tachyarrhythmia. There is little information about their use for this indication in younger children.

Cognitive impairment

Some patients who are receiving opioids, especially young children, become agitated rather than sedated. It may be appropriate to check their renal and/or hepatic function and adjust the dose of opioid accordingly. The adequacy of pain control dictates the approach that should be taken. Good pain control allows a trial reduction in the dose of opioid or lengthening of the dosing interval, but poor control is an indication for opioid rotation. The latter approach has been supported by a prospective study in adults.69


Myoclonus is an involuntary muscle contraction that occurs while conscious. It can occur with high opioid doses or in patients on long-term opioid therapy, due to the accumulation of neurotoxic metabolites such as morphine-3-glucuronide. It is an idiosyncratic reaction. Empirical approaches that have been used70 , 71 are based on benzodiazepines, particularly midazolam or clonazepam, or muscle relaxants, such as baclofen or dantrolene. Valproic acid has also been used for myoclonus.72 Opioid rotation (accompanied, as always, by an appropriate reduction in total opioid dose) can also be helpful.

Urinary retention

Urinary retention can be caused by any opioid given by any route, but anecdotally is more frequent when the opioid is given epidurally or spinally, often after rapid dose escalation. In one small series15 it occurred in one in seven children. Interventions to counter the effect include the application of external bladder pressure, starting a low-dose infusion, or the use of intermittent catheterization. Other options include substitution with an alternative opioid, such as fentanyl, or trying a cholinomimetic agent, such as bethanechol, to stimulate effective bladder contractions.

Respiratory depression

Respiratory depression due to opioid administration in children is much feared but grossly overestimated. Pain is an effective stimulant of respiratory drive, and apnoea is highly unlikely if titration is carried out appropriately as described previously. An exception is when narcotization occurs, usually as a result of a sudden reduction in pain stimulus or in the capacity to clear opioids. It is rarely necessary to administer the opioid antagonist naloxone, and since doing so carries the certainty of simultaneously reversing analgesia, this should not be undertaken lightly.

Physiological dependence, tolerance, and addiction

Many of the misconceptions about opioid use centre on concerns about dependence, tolerance, and addiction. Such misconceptions are so common that they must be proactively addressed and reviewed with all health professionals, patients, and families. Dependence and tolerance are physiological events and are not indicative of addiction, which is a psychological phenomenon.

Physiological dependence may occur. Dependence is the occurrence of physical symptoms attributable to withdrawal when the dose of opioids is reduced too quickly. If the dose of opioid needs to be reduced (e.g. if there has been another analgesic intervention and narcotization is anticipated), this should be done slowly. One practical approach is to reduce the dose by 25% every 2 days, aiming to reduce the dose to zero over the course of 1 or 2 weeks.

Tolerance is defined as a requirement for increasing opioid doses to maintain the same degree of pain relief. There are many mechanisms underlying this phenomenon, including alterations in G-protein expression and opioid-receptor down-regulation. The most common cause of increased opioid requirements is advancing disease, rather than tolerance. The extent to which tolerance occurs in a therapeutic setting at all is unclear, but in any case the solution during the terminal phase is simply to increase the dose or, if this is not possible due to adverse effects, to substitute a different opioid.

Addiction is defined as an individual's craving for opioids for their psychological impact rather than for pain relief. It is probably no more likely to occur in patients who are receiving opioids for pain than in the general population.73 It is important that irrational fear of addiction does not interfere with proper prescribing as part of good palliative care, and similarly that useful therapeutic medications such as diamorphine do not remain unavailable to these vulnerable patients because of misunderstanding among law makers.


The key points with regard to pain management in palliative care are summarized in Box 20.1.

The WHO approach: debates and controversies

Do we really need step 2?

The WHO pain ladder describes an intermediate step between simple analgesia and major opioids, in which a group of drugs that are referred to as ‘minor opioids’ is recommended. On the face of it, this is an uncontentious description. It refers to a group of drugs that, like morphine, act at opioid receptors, but which are less potent than morphine. Minor opioids are also sometimes considered to be distinct because they are thought to have a ‘ceiling effect.’

In reality, neither of these descriptions reliably characterizes one group of opioids over another, and some regard the second step of the WHO ladder as redundant.74 The potency of opioids at the opioid receptor is a spectrum, rather than being quantally divided into two distinct groups. A large dose of codeine acts at the opioid receptor in essentially the same way as a small dose of morphine.

This lack of precision in characterizing step 2 is well illustrated by buprenorphine. This opioid is available in a range of transcutaneous patch sizes, and at one end of the spectrum the opioid equivalency is roughly that of a moderate dose of codeine. Buprenorphine can therefore legitimately be considered—as it has been in this chapter—to be a minor opioid. However, for the larger patch doses the equivalency of buprenorphine clearly corresponds to significant doses of morphine. Buprenorphine therefore becomes, in reality as well as in effect, a major opioid. Although buprenorphine illustrates the problems with characterizing step 2 well, the same could be said of almost any opioid—a small dose is minor, but a large dose would be major.

The distinction that is sometimes made between minor and major opioids on the basis of a ceiling effect is also unsatisfactory. A ceiling effect describes the phenomenon where there is a change in the relationship between dose and response, such that beyond a certain dose of the drug there is no further increase in its effectiveness. It is the full occupancy of opioid receptors by a partial agonist that causes a true pharmacological ceiling effect because, at any given point in time, its occupancy of opioid receptors prevents other opioids from having access to them to a degree that is clinically significant.

Once again, this does not adequately define a separate step on the pain ladder. The only commonly used opioid that has a true ceiling effect is buprenorphine, and it does so only at doses well above those used in normal therapeutic practice.75 , 76 Once again, defining a separate step for minor opioids is not readily supported by their pharmacology.

Solutions to the problem that involve changing the name of this group of opioids (e.g. to ‘weak opioid’ or ‘opioid for mild to moderate pain’) refocus the categorization from pharmacology to empiricism, but they do not address the fundamental contradiction of making the distinction in the first place.

There are arguably two reasons for retaining the current WHO structure. The first is that its primary purpose was to disseminate an easily understood and simple approach to the management of pain in palliative care that could be easily adopted by specialists in any culture. The three-step approach is intuitive and easily learned. Perhaps removing the second step would undermine that simplicity and accessibility.

However, perhaps more convincing is the argument that for many patients the distinction between major and minor opioids is a very real one. For many, morphine, diamorphine, or any of the other major opioids carry with them the well-known stigma of approaching death. Although it is of course the responsibility of clinicians to explore and address these issues, it is a matter of common experience that some patients will always find compliance with major opioids difficult, simply because they are major opioids. Codeine and tramadol, inferior though they may be to major opioids from a purely pharmacological point of view, are often more acceptable from a psychological perspective.

Since there is no point in prescribing the right drug if the patient will simply refuse to take it, it does seem that the paediatric palliative care establishment may need to reluctantly accept the need for a step 2 that contains opioids characterized by their greater acceptability to patients.

Morphine as first-line treatment

At its outset, morphine was made the mainstay of the WHO pain ladder approach to management of pain in palliative care. Since then, experience, evidence, and confidence in the use of major opioids have grown. This has led many to feel that morphine should be replaced by one of the more modern major opioids.

There have been a number of potential candidates. Fentanyl has been proposed because of its significantly greater potency,77 and its higher level of lipophilicity, which means that it can readily access the central nervous system. Methadone is favoured by some32 for its broad receptor profile and comparatively low cost.

It is hard to see why an alternative is being sought. Morphine is easy to take, is well tolerated by most patients, and its adverse effects are well understood. Where the adverse effects are such as to make morphine intolerable, there are established strategies, and an increasing array of alternatives. Furthermore, morphine is almost universally low in cost and, so long as its more soluble alter ego—diamorphine—is readily available, there are rarely problems with its administration.

There are few clinical trials comparing potential competitors with morphine. Studies of novel opioids have usually compared their effects with placebo or with other non-morphine opioids. In the absence of evidence that they are superior to morphine in effectiveness and/or tolerability, other opioids can surely never be a rational alternative first-line treatment.

Does patient-controlled analgesia (PCA) have a role in palliative care?

On the face of it, PCA seems to break two of the golden rules of pain management in palliative medicine. It requires that the patient experiences pain before receiving analgesia (i.e. it is not ‘by the clock’), and the need for intravenous access and sophisticated equipment means that it is rarely the simplest option (i.e. it is not ‘by the appropriate route’). However, on closer inspection neither of these is necessarily conclusive, and PCA potentially offers important benefits over other approaches in palliative care.

PCA describes a range of techniques in which the individual patient is able to deliver a bolus dose of analgesia to him- or herself as soon as pain occurs. The commonest technique is to use an electronic pump mechanism that delivers a background infusion, and has a button that can be pressed the moment that pain occurs. A bolus dose is then delivered intravenously. The background rate, the magnitude of the bolus dose, the interval between bolus doses, and the total number of doses that can be given during a single 24-hour period can all be preset by the prescriber.

It has been suggested10 that the WHO guidelines should be updated to include PCA in the management of pain in palliative care. As yet there is no clinical evidence to support the use of PCA over more conventional approaches in paediatric palliative care. However, there are good reasons for believing that it might be effective. Another golden rule of palliative care is that analgesia should be ‘by the child.’ One of the great strengths of PCA is its immediacy. One of the threats to effective analgesia in palliative care is simply the time that elapses between the patient expressing the fact that they are in pain and being able to access pain relief. This is particularly likely in the paediatric population, as children need first to convince their parents of their pain, and then to wait while the parents retrieve the medication themselves, or need to go through the nurse or doctor to be able to do so. In a well-run family or hospital units, this may take as little as 10 minutes, but 10 minutes is still an unacceptable time period for a child to remain in pain.

PCA abolishes this delay. Furthermore, the psychological benefit of being able to respond instantly to pain by accessing analgesia is significant. For many children, particularly teenagers and young adults, a sense of powerlessness to intervene when pain occurs can amplify the total experience of pain. By giving the patient access to immediate analgesia as soon as it is needed, PCA helps to avoid this spiral of pain and powerlessness.

From the perspective of the patient, PCA is quite simple. It has been used successfully by children,78 even by those with neurological impairment.79 From the family's perspective there is a single infusion and a single response to the occurrence of pain. Again, this simplicity represents an advantage of the use of PCA in pain management in palliative care.

Children who are using PCA receive less opioid in total than those who are using continuous infusion,78 but this argument is probably not relevant to the use of PCA in palliative care. One of the basic principles of the use of opioids in palliative care is that there is no maximum dose except that imposed by toxic effects. The total dose of opioids received by a patient is of no relevance unless toxic effects occur. So long as there is no evidence of toxicity, it does not matter if the patient receives more opioids using one method than using another.

Therefore the relevant question that should be addressed by research to evaluate the use of PCA in palliative care is not whether the total amount of opioid can be reduced by its use compared with more conventional approaches, but whether PCA can provide equivalent analgesia at less cost in terms of adverse effects. This has not been shown, and on the face of it seems unlikely.

There have been reports of patients and families tampering with PCA,80 and this could be considered a disadvantage. In reality, however, this is no different to tampering with any other technique that is used to manage pain and palliative care. Any approach can be abused, and there is nothing to suggest that PCA is any more vulnerable to abuse than other techniques.

A more substantial argument against the use of PCA in palliative care relates to its dependence on the patient's experiencing pain before accessing analgesia. As has been discussed, the aim of ‘by-the-clock’ pain management in palliative care is to avoid the patient experiencing pain wherever possible. It is one of the golden rules of palliative pain management that the patient should, as far as possible, be prevented from experiencing pain in the first place. PCA as it is often used does not achieve this. On the contrary, it relies on the fact that the patient will need to experience pain in order to obtain analgesia.

However, this is not an insurmountable problem. Although it is often not used, PCA technology can deliver adequate background doses to allow prevention of pain. There seems to be no reason why, so long as this is done, PCA should not be used effectively in palliative care in line with the principles of background analgesia and titration set out in the WHO guidelines. PCA would, in effect, become a refinement of the continuous syringe driver technique that is already used extensively in palliative care.

A further argument against the use of PCA in palliative care, which has some merit, is that compared with enteral morphine or subcutaneous infusions of other opioids, it is expensive and complex to set up. It is usually used intravenously, it can be difficult to obtain equipment in some centres, and it requires trained supervision which may not be available to patients outside hospital, or in densely populated areas.

Even where it is available, the arrangements that are needed to organize it safely can risk delaying the patient's discharge home. Any potential delay in discharge from hospital, or unnecessary medicalization of care once the patient is at home, needs to be weighed carefully against the undoubted advantages of this technique. However, these are largely system issues, and it seems likely that, in centres where the use of PCA would be practicable and become established in palliative care, the process would become simpler over time.

Use of the intravenous route is suboptimal in palliative care. Peripheral lines require re-siting by nursing or medical staff, which typically becomes increasingly difficult as the condition progresses. Central venous line access is rare, with the exception of children with cancer, and in any case the process of accessing it risks infection and admission to hospital, both of which are undesirable outcomes in palliative care.

For these reasons, the subcutaneous route remains the usual preferred route for parenteral infusions in palliative care in children, as it is in adults. If PCA is to become established practice in palliative care in children, it should be shown that it can be delivered subcutaneously. Anecdotal reports suggest that it can be used in this way, and indeed it seems unlikely that it would present significant problems, as the infusion volume is typically small even for boluses. This makes PCA, if anything, particularly well suited to subcutaneous administration.

The advantages and drawbacks of the use of PCA in paediatric palliative care, and the potential solutions, are summarized in Box 20.2.

Thus the role of PCA in the management of pain in a palliative care context has yet to be defined. There are plausible and powerful reasons for considering it to be a useful modality for the future. However, it will need modification if it is to be useful for paediatric palliative care. It is not acceptable to allow patients to experience pain before they can gain access to analgesic, and the background opioid infusion must be based on WHO recommendations. It should be piloted using the subcutaneous rather than the intravenous route, so that intravenous access is not a prerequisite. The benefits of its use should always be carefully weighed against the burden of its complexity, and potentially the delay to discharge home. However, PCA unquestionably offers exciting potential for the future.

Are non-steroidal anti-inflammatory drugs (NSAIDs) really adjuvants?

By definition, an adjuvant has always been considered to be a drug that, although it can provide pain relief in certain situations, is not itself inherently analgesic. This definition is appropriate for carbamazepine, for example, which is of proven analgesic efficacy in many forms of neuropathic pain, but has no place in managing pain outside this indication. However, it is not equally appropriate for NSAIDs, which have considerable inherent analgesic activity,74 and this is reflected in a later version of the WHO guidelines for adults.81

Furthermore, by their very definition, adjuvants cannot be ranked in order of analgesic potency. However, NSAIDs have a well-documented range of potency, from ibuprofen at one end through to ketorolac at the other.

This illustrates a weakness in the WHO pain ladder. Because the aim is for it to remain simple and accessible, the ladder does not allow for the possibility of non-opioid analgesics that are of equal potency to opioids. There is no way of accommodating powerful non-opioid analgesics in the ladder except by considering them to be adjuvants.

Perhaps this problem can be addressed simply by reconsidering the definition of adjuvant in this context. The word ‘adjuvant’ was in fact originally used simply to refer to a medication that is added alongside the main therapy. When the term is used in this way, it adequately describes the role of NSAIDs, and indeed many other interventions that are broadly analgesic and have a range of potency. It would then include neurolytic procedures and NSAIDs, as well as its more conventional description of interventions such as antidepressants and anti-epileptics for neuropathic pain, chemotherapy or radiotherapy for pain caused by tumour, and anxiolytics or muscle relaxants to relieve the pain of muscle spasm and cerebral irritation.

Some conceptual distinction might still need to be made between adjuvants that are inherently analgesic, and those that are analgesic only in specific pain situations. It should also be borne in mind that the same medication can have both adjuvant and analgesic properties. For example, ketamine is primarily an adjuvant, and its use should usually be restricted to management of neuropathic pain where more conventional measures have failed. However, ketamine also has analgesic properties of its own. This illustrates an important principle, namely that even if the traditional terms ‘adjuvant’ and ‘analgesic’ do not adequately classify drugs, they may usefully distinguish between different individual effects of the same drug.

Summary and conclusion

The WHO pain ladder was the first rational and systematic approach to the management of pain in palliative medicine in adults or children. Its aim was to provide logical and simple structure through a set of principles—not to be exhaustive or to provide detailed guidelines. Its intention was to teach, rather than to instruct.

Now, 25 years after its publication, the WHO approach remains the yardstick against which new interventions in the management of pain in palliative medicine should be measured. However, this does not mean that the guidelines should not be subjected to critical evaluation and review. Such scrutiny is particularly important when applying them to children, who were not the population for whom the original guidelines were designed.

Some of the criticisms that have been levelled at the WHO approach seem unfair. It does not claim to address non-pharmacological dimensions of management, but this does not mean that their importance was not recognized by its authors. New techniques such as PCA are not mentioned by the WHO guidelines, but the guidelines are flexible enough to accommodate new techniques that are suitably adapted.

Other criticisms have more substance. The definition of an adjuvant seems perverse in the light of what we now know about the analgesic properties of, for example, NSAIDs. The distinction between minor and major opioids, which is a central feature of the original pain ladder, now seems arbitrary.

Nevertheless, the WHO approach remains the basis for pain management in palliative care, in adults and in children. Perhaps this is for two reasons. The first is that it is simple and intuitive. The second is that—with all its flaws—it seems to work.


Some of the material for this chapter formed part of a chapter in the first edition of the Textbook. The authors are extremely grateful to Dr Ross Drake, both for his contributions to that earlier chapter and for his valuable and constructive comments on earlier drafts of the current one.


1. Ventafridda V et al. A validation study of the WHO method for cancer pain relief. Cancer 1987; 59: 850–6.Find this resource:

2. McGraw MB. Neural maturation as exemplified in the changing reactions of the infant to pin prick. Child Dev 1941; 12(1): 31–42.Find this resource:

3. World Health Organization. Cancer Pain Relief and Palliative Care in Children. Geneva: World Health Organization; 1998.Find this resource:

    4. Vargas-Schaffer G. Is the WHO analgesic ladder still valid? Twenty-four years of experience. Can Fam Physician 2010; 56(6): 514–17.Find this resource:

    5. Stjernsward J. WHO cancer pain relief programme. Cancer Surv 1988; 7(1): 195–208.Find this resource:

    6. Zernikow B et al. Paediatric cancer pain management using the WHO analgesic ladder – results of a prospective analysis from 2265 treatment days during a quality improvement study. Eur J Pain 2006; 10(7): 587–95.Find this resource:

    7. Glare P. Choice of opioids and the WHO ladder. In: Davis MP, Glare P, Quigley C, Hardy JR, eds. Opioids in Cancer Pain, 2nd edn. Oxford: Oxford University Press; 2009. p. 279.Find this resource:

      8. Ong CK, Forbes D. Embracing Cicely Saunders's concept of total pain. BMJ 2005; 331(7516): 576–7.Find this resource:

      9. Kuttner L. Integrative Methods to Relieve Pain and Suffering. Oxford: Oxford University Press; 2005.Find this resource:

        10. Armstrong EJ, Jenkins AJ, Sebrosky GF, Balraj EK. An unusual fatality in a child due to oxycodone. Am J Forensic Med Pathol 2004; 25(4): 338–41.Find this resource:

        11. Miaskowski C et al. Principles of Analgesic Use in the Treatment of Acute Pain and Cancer Pain. Glenview, IL: American Pain Society; 2008.

        12. Hanks GW et al. Explanation for potency of repeated oral doses of morphine? Lancet 1987; 2: 723–6.Find this resource:

        13. Hoskin PJ et al. M6G and its analgesic action in chronic use. Clin J Pain 1989; 5(2): 199–200.Find this resource:

        14. Okura T et al. Different effects of morphine and morphine-6beta-glucuronide on formalin-evoked spinal glutamate release in conscious and freely moving rats. Neurosci Lett 2007; 415(2): 169–73.Find this resource:

        15. Mashayekhi SO, Ghandforoush-Sattari M, Routledge PA, Hain RD. Pharmacokinetic and pharmacodynamic study of morphine and morphine 6-glucuronide after oral and intravenous administration of morphine in children with cancer. Biopharm Drug Dispos 2009; 30(3): 99–106.Find this resource:

        16. Hunt AM, Joel S, Dick G, Goldman A. Population pharmacokinetics of oral morphine and its glucuronides in children receiving morphine as immediate-release liquid or sustained-release tablets for cancer pain. J Pediatr 1999; 135(1): 47–55.Find this resource:

        17. Hain RD, Hardcastle A, Pinkerton CR, Aherne GW. Morphine and morphine-6-glucuronide in the plasma and cerebrospinal fluid of children. Br J Clin Pharmacol 1999; 48(1): 37–42.Find this resource:

        18. Hain RDW, Miser A, Devins M, Wallace WHB. Strong opioids in pediatric palliative medicine. Pediatr Drugs 2005; 5(1): 1–9.Find this resource:

        19. Neumann PB, Henriksen H, Grosman N, Christensen CB. Plasma morphine concentrations during chronic oral administration in patients with cancer pain. Pain 1982; 13(3): 247–52.Find this resource:

        20. Nahata MC, Miser AW, Miser JS, Reuning RH. Variation in morphine pharmacokinetics in children with cancer. Dev Pharmacol Ther 1985; 8(3): 182–8.Find this resource:

        21. Bouwmeester N, Anderson BJ, Tibboel D, Holford NH. Developmental pharmacokinetics of morphine and its metabolites in neonates, infants and young children. Br J Anaesth 2004; 92: 208–17.Find this resource:

        22. Payne R. Recognition and diagnosis of breakthrough pain. Pain Med 2007; 8 (Suppl. 1): S3–7.Find this resource:

        23. Portenoy RK, Payne D, Jacobsen P. Breakthrough pain: characteristics and impact in patients with cancer pain. Pain 1999; 81(1–2): 129–34.Find this resource:

        24. Farrar JT. Incident pain: definition, diagnosis, and therapy. In: Perry MC, ed. American Society of Clinical Oncology Educational Book. Alexandria, VA: American Society of Clinical Oncology; 1999. pp. 402–4.Find this resource:

          25. Harrop JE. Management of pain in childhood. Arch Dis Child Educ Pract Ed 2007; 92(4): ep101–8.Find this resource:

          26. Bruce E, Franck L, Howard RF. The efficacy of morphine and Entonox analgesia during chest drain removal in children. Paediatr Anaesth 2006; 16(3): 302–8.Find this resource:

          27. Michaud L et al. Nitrous oxide sedation in pediatric patients undergoing gastrointestinal endoscopy. J Pediatr Gastroenterol Nutr 1999; 28(3): 310–14.Find this resource:

          28. Weinstein SM, Messina J, Xie F. Fentanyl buccal tablet for the treatment of breakthrough pain in opioid-tolerant patients with chronic cancer pain: a long-term, open-label safety study. Cancer 2009; 115(11): 2571–9.Find this resource:

          29. Rhiner M, Palos G, Termini M. Managing breakthrough pain: a clinical review with three case studies using oral transmucosal fentanyl citrate. Clin J Oncol Nurs 2004; 8(5): 507–12.Find this resource:

          30. Davies D, DeVlaming D, Haines C. Methadone analgesia for children with advanced cancer. Pediatr Blood Cancer 2008; 51(3): 393–7.Find this resource:

          31. Nicholson AB. Methadone for cancer pain. Cochrane Database Syst Rev 2007; Issue 4: CD003971.Find this resource:

          32. Bruera E et al. Methadone versus morphine as a first-line strong opioid for cancer pain: a randomized, double-blind study. J Clin Oncol 2004; 22(1): 185–92.Find this resource:

          33. Ripamonti C, Bianchi M. The use of methadone for cancer pain. Hematol Oncol Clin North Am 2002; 16(3): 543–55.Find this resource:

          34. Layson-Wolf C, Goode JV, Small RE. Clinical use of methadone. J Pain Palliat Care Pharmacother 2002; 16(1): 29–59.Find this resource:

          35. Bruera E, Sweeney C. Methadone use in cancer patients with pain: a review. J Palliat Med 2002; 5(1): 127–38.Find this resource:

          36. Shaiova L et al. Consensus guideline on parenteral methadone use in pain and palliative care. Palliat Support Care 2008; 6(2): 165–76.Find this resource:

          37. Knoester PD et al. Pharmacokinetics and pharmacodynamics of midazolam administered as a concentrated intranasal spray. A study in healthy volunteers. Br J Clin Pharmacol 2002; 53(5): 501–7.Find this resource:

          38. Polosa R, Simidchiev A, Walters EH. Nebulised morphine for severe interstitial lung disease. Cochrane Database Syst Rev 2002; Issue 3: CD002872.Find this resource:

          39. Cohen SP, Dawson TC. Nebulized morphine as a treatment for dyspnea in a child with cystic fibrosis. Pediatrics 2002; 110(3): e38.Find this resource:

          40. Hansson L, Midgren B, Karlsson JA. Effects of inhaled lignocaine and adrenaline on capsaicin-induced cough in humans. Thorax 1994; 49(11): 1166–8.Find this resource:

          41. Soares LG, Martins M, Uchoa R. Intravenous fentanyl for cancer pain: a “fast titration” protocol for the emergency room. J Pain Symptom Manage 2003; 26(3): 876–81.Find this resource:

          42. Harris JT, Suresh Kumar K, Rajagopal MR. Intravenous morphine for rapid control of severe cancer pain. Palliat Med 2003; 17(3): 248–56.Find this resource:

          43. Mercadante S et al. Rapid titration with intravenous morphine for severe cancer pain and immediate oral conversion. Cancer 2002; 95(1): 203–8.Find this resource:

          44. Cherny N et al. Strategies to manage the adverse effects of oral morphine: an evidence-based report. J Clin Oncol 2001; 19(9): 2542–54.Find this resource:

          45. Noyes M, Irving H. The use of transdermal fentanyl in pediatric oncology palliative care. Am J Hosp Palliat Care 2001; 18(6): 411–16.Find this resource:

          46. Drake R, Longworth J, Collins JJ. Opioid rotation in children with cancer. J Palliat Med 2004; 7(3): 419–22.Find this resource:

          47. Al-Shahri MZ, Molina EH, Oneschuk D. Medication-focused approach to total pain: poor symptom control, polypharmacy, and adverse reactions. Am J Hosp Palliat Care 2003; 20(4): 307–10.Find this resource:

          48. Vella-Brincat J, Macleod AD. Adverse effects of opioids on the central nervous systems of palliative care patients. J Pain Palliat Care Pharmacother 2007; 21(1): 15–25.Find this resource:

          49. Glare P, Walsh D, Sheehan D. The adverse effects of morphine: a prospective survey of common symptoms during repeated dosing for chronic cancer pain. Am J Hosp Palliat Care 2006; 23(3): 229–35.Find this resource:

          50. Arpino PA, Thompson BT. Safety of enteral naloxone for the reversal of opiate-induced constipation in the intensive care unit. J Clin Pharm Ther 2009; 34(2): 171–5.Find this resource:

          51. Akkawi R et al. Effect of oral naloxone hydrochloride on gastrointestinal transit in premature infants treated with morphine. Acta Paediatr 2009; 98(3): 442–7.Find this resource:

          52. Netzer P et al. The effect of naloxone-3-glucuronide on colonic transit time in healthy men after acute morphine administration: a placebo-controlled double-blinded crossover preclinical volunteer study. Aliment Pharmacol Ther 2008; 28(11–12): 1334–41.Find this resource:

          53. Tofil NM, Benner KW, Faro SJ, Winkler MK. The use of enteral naloxone to treat opioid-induced constipation in a pediatric intensive care unit. Pediatr Crit Care Med 2006; 7(3): 252–4.Find this resource:

          54. Liu M, Wittbrodt E. Low-dose oral naloxone reverses opioid-induced constipation and analgesia. J Pain Symptom Manage 2002; 23(1): 48–53.Find this resource:

          55. Hawkes ND et al. Effect of an enteric-release formulation of naloxone on intestinal transit in volunteers taking codeine. Aliment Pharmacol Ther 2001; 15(5): 625–30.Find this resource:

          56. Meissner W et al. Oral naloxone reverses opioid-associated constipation. Pain 2000; 84(1): 105–9.Find this resource:

          57. Schutter U et al. Innovative pain therapy with a fixed combination of prolonged-release oxycodone/naloxone: a large observational study under conditions of daily practice. Curr Med Res Opin 2010; 26(6): 1377–87.Find this resource:

          58. Mueller-Lissner S. Fixed combination of oxycodone with naloxone: a new way to prevent and treat opioid-induced constipation. Adv Ther 2010; 27(9): 581–90.Find this resource:

          59. Montesano F et al. Therapeutic switch to buprenorphine/naloxone from buprenorphine alone: clinical experience in an Italian addiction centre. Clin Drug Investig 2010; 30 (Suppl. 1): 13–19.Find this resource:

          60. Magnelli F et al. Safety and efficacy of buprenorphine/naloxone in opioid-dependent patients: an Italian observational study. Clin Drug Investig 2010; 30 (Suppl. 1): 21–6.Find this resource:

          61. Lowenstein O et al. Efficacy and safety of combined prolonged-release oxycodone and naloxone in the management of moderate/severe chronic non-malignant pain: results of a prospectively designed pooled analysis of two randomised, double-blind clinical trials. BMC Clin Pharmacol 2010; 10: 12.Find this resource:

          62. Clemens KE, Mikus G. Combined oral prolonged-release oxycodone and naloxone in opioid-induced bowel dysfunction: review of efficacy and safety data in the treatment of patients experiencing chronic pain. Expert Opin Pharmacother 2009; 11(2): 297–310.Find this resource:

          63. Yuan CS. Methylnaltrexone mechanisms of action and effects on opioid bowel dysfunction and other opioid adverse effects. Ann Pharmacother 2007; 41(6): 984–93.Find this resource:

          64. Deibert P, Xander C, Blum HE, Becker G. Methylnaltrexone: the evidence for its use in the management of opioid-induced constipation. Core Evid 2010; 4: 247–58.Find this resource:

          65. Clark AJ et al. Efficacy and safety of transdermal fentanyl and sustained-release oral morphine in patients with cancer and chronic non-cancer pain. Curr Med Res Opin 2004; 20(9): 1419–28.Find this resource:

          66. Radbruch L et al. Constipation and the use of laxatives: a comparison between transdermal fentanyl and oral morphine. Palliat Med 2000; 14(2): 111–19.Find this resource:

          67. Staats PS, Markowitz J, Schein J. Incidence of constipation associated with long-acting opioid therapy: a comparative study. South Med J 2004; 97(2): 129–34.Find this resource:

          68. Mercadante S, Casuccio A, Tirelli W, Giarratano A. Equipotent doses to switch from high doses of opioids to transdermal buprenorphine. Support Care Cancer 2009; 17(6): 715–18.Find this resource:

          69. Yee JD, Berde CB. Dextroamphetamine or methylphenidate as adjuvants to opioid analgesia for adolescents with cancer. J Pain Symptom Manage 1994; 9(2): 122–5.Find this resource:

          70. Mercadante S. Pathophysiology and treatment of opioid-related myoclonus in cancer patients. Pain 1998; 74(1): 5–9.Find this resource:

          71. Holdsworth MT et al. Continuous midazolam infusion for the management of morphine-induced myoclonus. Ann Pharmacother 1995; 29(1): 25–9.Find this resource:

          72. Chapman AG. Valproate and myoclonus. Adv Neurol 1986; 43: 661–74.Find this resource:

          73. Foley KM. Current issues in the management of cancer pain: Memorial Sloan-Kettering Cancer Center. NIDA Res Monogr 1981; 36: 169–81.Find this resource:

          74. McQuay H, Moore A. Acute pain: conclusion. In: McQuay H, Moore A, eds. An Evidence-Based Resource for Pain Relief. Oxford: Oxford University Press; 1998. pp. 187–92.Find this resource:

            75. Mercadante S, Ferrera P, Villari P. Is there a ceiling effect of transdermal buprenorphine? Preliminary data in cancer patients. Support Care Cancer 2007; 15(4): 441–4.Find this resource:

            76. Christoph T et al. Broad analgesic profile of buprenorphine in rodent models of acute and chronic pain. Eur J Pharmacol 2005; 507: 87–98.Find this resource:

            77. Hoya Y, Okamoto T, Yanaga K. Evaluation of analgesic effect and safety of fentanyl transdermal patch for cancer pain as the first line. Support Care Cancer 2010; 18(6): 761–4.Find this resource:

            78. Birmingham PK et al. Patient-controlled epidural analgesia in children: can they do it? Anesth Analg 200; 96(3): 686–91.Find this resource:

              79. Grandinetti CA, Buck ML. Patient-controlled analgesia: guidelines for use in children. Pediatr Pharmacother 2000; 6(11): 1–4.Find this resource:

                80. Pasero C, McCaffery M. Authorized and unauthorized use of PCA pumps: clarifying the use of patient-controlled analgesia, in light of recent alerts. Am J Nurs 2005; 105(7): 30–31, 33.Find this resource:

                81. World Health Organization. Cancer Pain Relief with a Guide to Opioid Availability, 2nd edn. Geneva: World Health Organization; 1996.Find this resource: