1856
`
`Opioid Rotation for Cancer Pain
`RationaleandClinicalAspects
`
`Sebastiano Mercadante, M.D.1,2
`
`1 Department of Anesthesia and Intensive Care,
`Pain Relief and Palliative Care, La Maddalena
`Clinic, Palermo, Italy.
`
`2 Pain Relief and Palliative Care Unit, SAMOT, Pal-
`ermo, Italy.
`
`Presented in part at the International Meeting on
`Pain Control and Regional Anesthesia, Jaipur, In-
`dia, February 9 –13, 1998.
`
`Address for reprints: Dr. Sebastiano Mercadante,
`M.D., SAMOT, Via Liberta` 191, 90143 Palermo,
`Italy.
`
`Received February 25, 1999; revision received
`June 1, 1999; accepted June 1, 1999.
`
`© 1999 American Cancer Society
`
`BACKGROUND. Some patients with cancer pain may develop uncontrolled adverse
`effects, including generalized myoclonus, delirium, nausea and emesis, or severe
`sedation before achieving adequate analgesia during opioid dose titration. Sequen-
`tial therapeutic trials should be considered to determine the most favorable drug.
`METHODS. Recent literature was taken into account when reviewing the rationale
`and potential of opioid rotation.
`RESULTS. When aggressive attempts to prevent adverse effects fail, drug rotation
`should be considered, because sequential therapeutic trials can be useful in iden-
`tifying the most favorable drug. Different mechanisms, including receptor activity,
`the asymmetry in cross-tolerance among different opioids, different opioid effica-
`cies, and accumulation of toxic metabolites can explain the differences in analgesic
`or adverse effect responses among opioids in a clinical setting.
`CONCLUSIONS. When pain is relieved inadequately by opioid analgesics given in a
`dose that causes intolerable side effects despite routine measures to control them,
`treatment with the same opioid by an alternative route or with an alternative
`opioid administered by the same route should be considered. Opioid rotation may
`be useful in opening the therapeutic window and for establishing a more advan-
`tageous analgesia/toxicity relationship. By substituting opioids and using lower
`doses than expected according to the equivalency conversion tables, it is possible
`in the majority of cases to reduce or relieve the symptoms of opioid toxicity in
`those patients who were highly tolerant to previous opioids while improving
`analgesia and, as a consequence, the opioid responsiveness. Cancer 1999;86:
`1856 – 66. © 1999 American Cancer Society.
`
`KEYWORDS: cancer pain, opioids, tolerance, opioid toxicity, opioid rotation, mor-
`phine metabolites, methadone, transdermal route.
`
`Most cancer patients develop significant pain, requiring opioid
`
`therapy during the course of their illness.1 Cancer pain can be
`controlled adequately in most patients with oral analgesics, which
`usually are provided “around the clock” with rescue doses for break-
`through pain.2 Nevertheless, some patients develop uncontrolled ad-
`verse effects, including generalized myoclonus, delirium, nausea and
`emesis, or severe sedation before achieving adequate analgesia dur-
`ing dose titration. Aggressive attempts to prevent adverse effects
`should be made in all patients on opioids, and only when these
`interventions fail should drug rotation be considered. Sequential ther-
`apeutic trials with different opioids can be useful in identifying the
`most favorable drug.3–5 This review examines the rationale and the
`clinical aspects of opioid rotation on the basis of recent research. This
`term may be confusing, because its meaning should include the use
`of multiple opioids at intervals. However, the substitution of another
`opioid for a previous one to obtain a more favorable response largely
`
`1
`
`

`

`has been reported as opioid rotation, so that, in this
`review, the term “opioid rotation” is used with this
`meaning.
`
`BASIC SCIENTIFIC EVIDENCE
`Receptor Aspects
`Except for differences in pharmacokinetics, the m-re-
`ceptor opioid agonists have been considered similar to
`morphine with respect to their mechanism of action.
`However, several studies suggest that opioids may ex-
`ert different receptor activities.6 The knowledge of
`these receptor aspects can be useful in explaining the
`differences in analgesic or adverse effect responses
`among opioids in the clinical setting.
`
`Tolerance
`The repeated administration of opioids leads to the
`development of tolerance. Analgesic tolerance is de-
`fined pharmacologically as a reduced potency of the
`analgesic effects of opioids after repeated administra-
`tion or the need for higher doses to maintain the same
`effect. Consequently, tolerance may shift the dose re-
`sponse curve to the right. This may result in a greater
`risk of side effects due to an increase in opioid toxic
`metabolites.
`Pharmacodynamic, pharmacokinetic, or psycho-
`logical processes are involved in the development of
`tolerance.7 Pharmacodynamic tolerance often is as-
`sumed to refer to a diminution of analgesic effects
`related to a process of neural adaptation.8 Because
`opioid analgesia is mediated through a complex inter-
`action of different receptors, nociceptive function, to
`some extent, is a mixed-receptor function.9 Tolerance
`development for an analgesic and a toxic effect has
`been shown to be dissociated in time.6,10 The magni-
`tude of cross-tolerance between opioids is different
`for the analgesic and the toxic effects. The develop-
`ment of tolerance to adverse effects is the favorable
`side of the coin, because this condition represents a
`widening of the therapeutic window, inducing a right-
`ward shift of the toxicity curve.11
`The variability in analgesic or adverse effect re-
`sponse to different opioid analgesics is relatively
`common and probably is due to an incomplete
`cross-tolerance among opioids. This phenomenon
`frequently is attributed to differential opioid recep-
`tor affinities.12 Tolerance develops independently at
`each receptor subtype in response to the binding of
`a drug and its intrinsic activity. Changes in the
`receptor-effect relation also may occur over the
`course of the illness with prolonged morphine ex-
`posure.13 Indeed, the mechanism of pain may influ-
`ence the pattern of responses produced by different
`opioids due to a different affinity or efficacy of these
`
`Opioid Rotation/Mercadante
`
`1857
`
`drugs at different receptor levels (see below). It also
`has been suggested that the phenomenon depends
`on genetic factors. Consequently, the individual re-
`ceptor profile and the specific clinical conditions
`may influence the final effects in terms of analgesia
`or side effects.5,13–19 Moreover, the asymmetry ob-
`served in the development of tolerance also may be
`due to changes in pharmacokinetics, including ad-
`sorption, distribution, metabolism, and specific dis-
`ease.
`
`N-methyl-D-aspartatereceptoractivation
`The mechanism of the pain may influence the pattern
`of responses produced by different opioids. Several
`common factors in neuropathic pain and tolerance
`have been found. A given dose of morphine produced
`less antinociception in nerve-injured rats, and the an-
`tinociceptive dose response curve was shifted to the
`right after intervention. Animals made tolerant to
`morphine analgesia become hyperalgesic. Therefore,
`a neuropathic pain state may be equivalent to that of
`the development of morphine tolerance, with a similar
`activation of N-methyl-D-aspartate (NMDA) receptors
`and the consequent intracellular events. Development
`of hyperalgesia and the rightward shift of the mor-
`phine antinociceptive dose response curve were pre-
`vented by the administration of NMDA receptor an-
`tagonists.20 These findings may explain the rationale
`for using drugs that inhibit some steps of the neuronal
`hyperexicitability, such as the NMDA antagonist ket-
`amine and dextromethorphan. In animal and human
`studies, these substances have been reported to re-
`duce hyperalgesia and tolerance development subse-
`quent to neural injury.20,21 Some opioids have been
`reported to have relevant anti-NMDA activity. Metha-
`done has been demonstrated to exert a potent inhibi-
`tion of MK-801 binding nearly equipotent with dex-
`tromethorphan and to show an NMDA antagonist
`activity.22,23 Therefore, it potentially may be more ef-
`fective in circumstances in which a NMDA receptor
`activation develops, such as in case of tolerance or in
`the presence of a neuropathic pain mechanism.
`
`Opioidefficacy
`The phenomenon of asymmetric cross-tolerance also
`may be due to differences in agonist efficacies rather
`than receptor selectivities. To generate a given effect,
`it is necessary to occupy a number of receptors out of
`the total population, the so-called “fractional receptor
`occupancy”. The number of receptors to be occupied
`is inversely proportional to the intrinsic activity.24 The
`amount of the remaining unoccupied receptors (re-
`ceptor reserve) depends on this property: The larger
`the receptor reserve, the greater the intrinsic efficacy.
`
`2
`
`

`

`1858
`
`CANCER November 1, 1999 / Volume 86 / Number 9
`
`According to this theory, different drugs may produce
`equivalent pain relief while occupying different pro-
`portions of the available receptors. A low intrinsic
`efficacy agonist, such as morphine, occupies and pro-
`duces a greater tolerance effect due to its high occu-
`pancy requirements compared with high-efficacy ago-
`nists, such as sufentanil, which has a higher receptor
`reserve.24 –27
`The concept of agonist efficacy related to dose
`response changes with progressive increases in stim-
`ulus intensity has been debated recently.25,26,28 It has
`been demonstrated that the relative potency of sufen-
`tanil to morphine increases as tolerance develops.19
`With an increase in stimulus intensity, sufentanil
`showed a smaller shift in the dose response curve than
`morphine, which showed a greater reduction in the
`maximum effect and increased occupancy require-
`ments.25 The greater shift in morphine dose response
`relative to sufentanil when stimulus intensity rises
`supports the receptor occupancy theory. Several opi-
`oids, including methadone, fentanyl, and sufentanil,
`have been demonstrated to have a much higher effi-
`cacy than morphine due to a higher receptor reserve
`than morphine.13,19,26,29
`These observations may explain the loss of effi-
`cacy with increasing doses of morphine. Switching to
`an alternative opioid with a higher efficacy may re-
`verse the reduced response to morphine.
`
`The Role of the Metabolites
`Metaboliteproductionandanalgesicresponse
`Patients’ analgesic response may depend on their
`morphine to metabolite ratio.30 Morphine-6-glucuro-
`nide (M6G) is an active metabolite of morphine that
`has analgesic properties.31 Conversely, morphine-3-
`glucuronide (M3G), the major metabolite of mor-
`phine, has a negligible affinity for opioid receptors and
`is considered to be devoid of analgesic activity. On the
`contrary, experimental studies have shown that M3G
`produces neuroexcitatory
`and antianalgesic
`ef-
`fects.32,33 Controversies exist about its role in the de-
`velopment of tolerance to M6G antinociceptive ef-
`fects.30,32,33
`Concentration ratios of the metabolites to mor-
`phine are higher after prolonged oral morphine ther-
`apy than after parenteral administration of morphine
`because of a high first pass metabolic clearance to
`M3G and M6G.34 –36 Because elimination of M6G has
`been demonstrated to be related closely to renal func-
`tion,37–39 renal failure may affect morphine and its
`metabolite pharmacokinetics with different mecha-
`nisms, including the rate of oral absorption and he-
`patic biotransformation, the elimination of morphine
`and its metabolites, and the diffusion across the
`
`TABLE 1
`Prevalent Symptoms Due to Morphine Metabolite Toxicitya
`
`Opioid receptor
`(morphine-6-glucuronide)
`
`Drowsiness
`Nausea and emesis
`Coma
`Respiratory depression
`
`a Mixed syndromes are observed more often.
`
`Nonopioid receptor
`(morphine-3-glucuronide)
`
`Impaired cognitive function
`Myoclonus, seizures
`Hyperalgesia
`—
`
`blood-brain barrier.37 Hepatic disease or a variation in
`morphine metabolism may explain cases in which a
`higher concentration of morphine than M6G has been
`found.39 Conversely, M6G accumulation also may oc-
`cur in patients with normal renal function due to an
`increased capacity in M6G synthesis or enteroepathic
`recirculation. Increased M6G levels also were found 19
`days after morphine discontinuation in patients who
`were treated with antibiotics, probably facilitating the
`intestinal reabsorption of M6G.40
`On clinical grounds, the M6G concentration in
`plasma and cerebrospinal fluid (CSF) during mor-
`phine therapy suggests that it makes an important
`contribution to the analgesic efficacy of the parent
`drug, although the degree to which M6G contributes
`to the clinical effects of morphine and its exact po-
`tency are yet to be quantified. Because M3G may
`oppose the analgesic effect of morphine or M6G, a
`negative relation between the degree of pain relief and
`CSF M3G concentration may be expected.41 Nonethe-
`less, evidence showing that M3G may antagonize the
`analgesic actions of morphine or influence the sever-
`ity of adverse effects is unsubstantiated.42,43
`
`Toxicity
`The role of morphine metabolites in the pathogenesis
`of late opioid toxicity has been described.44 –53 The
`potential mechanisms of morphine metabolite toxicity
`are listed in Table 1. High doses of morphine or its
`metabolites may favor the appearance of neuroexcita-
`tory effects, such as hyperesthesia and myoclonus, via
`a spinal antiglycinergic effect at a nonopioid receptor
`site. Conversely, hyperalgesia merely may be the con-
`sequence of an exacerbation of preexisting pain.45 The
`increase in the morphine glucuronides to morphine
`ratio would likely augment the potential for clinical
`consequences from metabolites. These findings have
`clear clinical
`implications, particularly during pro-
`longed morphine use and in the presence of renal
`impairment or dehydration, when these metabolites
`may accumulate. Elevated M6G levels were associated
`with severe side effects, particularly in the setting of
`
`3
`
`

`

`metabolic dysfunction. Nevertheless, the increasing
`M6G:morphine ratio was not determinant in produc-
`ing myoclonus and cognitive impairment, whereas in-
`creasing age, elevated bilirubin level, and elevated
`LDH level were associated significantly with cognitive
`impairment.47 Normorphine, another morphine me-
`tabolite, has been reported to be responsible for some
`of the neurologic side effects of high doses of mor-
`phine, resembling the neurotoxic properties described
`for normeperidine.50,51
`When it is administered orally, hydromorphone is
`metabolized extensively by the liver to 3-glucuronide
`(H3G) and 6-glucuronide (H6G) with a wide interpa-
`tient variability. Based on their molecular structure
`and their enhanced water solubility, H3G and H6G
`can be expected to accumulate during chronic admin-
`istration or in the presence of renal insufficiency.52–54
`Taking these clinical observations into account, it
`was suggested that the accumulation of toxic metab-
`olites during chronic opioid therapy can lead to severe
`adverse effects, even in patients with apparently nor-
`mal renal function. If metabolites are considered to
`play a relevant role, then switching opioids may allow
`for the elimination of the responsible breakdown.
`
`INDICATIONS FOR OPIOID ROTATION
`Opioid responsiveness can be defined by the degree of
`analgesia achieved during dose escalation either to
`intolerable side effects or to the development of ade-
`quate analgesia.18 Several factors can interfere with an
`appropriate opioid analgesic response in the course of
`the illness. These include the progression of the dis-
`ease and the development of tolerance, the appear-
`ance of intractable side effects, the type and temporal
`pattern of pain, morphine metabolites, pharmacoki-
`netic and pharmacodynamic factors, and individual
`factors that are not well known.55 Consequently, the
`term opioid responsiveness includes the extremely
`large variability that characterizes the continuum of
`responses going from the easy and immediate
`achievement of analgesia to unresponsiveness. Opioid
`responsiveness should not be judged on the analgesic
`response to one opioid and should be assessed after a
`trial of one or more alternative opioids. When the pain
`is relieved inadequately by opioid analgesics given in a
`dose that causes intolerable side effects despite rou-
`tine measures to control them, patients should be
`considered for treatment with the same opioid by an
`alternative route or with an alternative opioid admin-
`istered by the same route.56 –59 Consequently, the ma-
`jor indication for switching opioids is represented by
`poorly controlled pain with unacceptable adverse ef-
`fects due to opioid toxicity, rapid development of tol-
`erance,
`refractory pain, or difficult pain syn-
`
`Opioid Rotation/Mercadante
`
`1859
`
`dromes.56,57 In a recent survey, the indications for
`opioid rotation included cognitive failure in 39% of
`patients, hallucinations in 24% of patients, uncon-
`trolled pain in 16% of patients, myoclonus in 11% of
`patients, nausea in 9% of patients, and local irritation
`from methadone suppositories in 1% of patients.10
`Before starting opioid rotation with an alternative
`drug, an appropriate assessment should include a re-
`view of the clinical situation and pain syndromes, the
`use of an adjuvant analgesic to deal with the altered
`sensorium secondary to opioid toxicity using specific
`drugs, and the correction of any contributing abnor-
`mal biochemistry, including hydration status.60
`
`CLINICAL EXPERIENCES
`Switching the Route of Administration
`Although the optimal route for the administration of
`morphine is by mouth, this route influences the mor-
`phine to metabolite ratio both in plasma and in CSF.
`Concentrations of M3G and M6G in relation to mor-
`phine were found to be greater after oral administra-
`tion than after intravenous administration of mor-
`phine.34,35 Patients who received oral morphine had a
`three-fold higher prevalence of myoclonus than pa-
`tients who received morphine by the parenteral
`route.49 Thus, switching from oral administration to
`parenteral administration may reduce metabolite for-
`mation and, thus, toxicity due the accumulation of
`metabolites.
`
`Parenteralroute
`Twenty-eight percent of cancer patients with ad-
`vanced disease who were referred to a pain clinic
`required parenteral opioids at some point during the
`course of their illness.60 To improve the convenience
`of the treatment regimen in the setting of adequate
`pain control, to diminish side effects in the setting of
`controlled pain, to reduce the invasiveness of therapy,
`and simultaneously to improve pain control and re-
`duce opioid toxicity, the route of opioid delivery was
`changed in 31% of patients. In selecting the route of
`administration, the severity of the pain and the rapid-
`ity with which relief was required were found to be
`critical considerations.59 The parenteral route facili-
`tates rapid titration of the opioid dose.61 When stable
`pain control is achieved, patients can be switched to
`noninvasive routes of administration.59
`About 70% of patients will benefit from the use of
`an alternate nonoral route for opioid administration
`before death, because cognitive failure and the inabil-
`ity to swallow make it impossible to continue oral
`administration. If patients are unable to take drugs
`orally, then the preferred alternative routes are sub-
`cutaneous, rectal, and transdermal.2 Subcutaneous,
`
`4
`
`

`

`1860
`
`CANCER November 1, 1999 / Volume 86 / Number 9
`
`continuous administration of opioids appears to be
`safe and effective and makes home management sim-
`ple with the use of simple devices for continuous
`administration, providing effective analgesia with a
`low incidence of adverse effects. Blood levels are com-
`parable to those obtained during intravenous admin-
`istration.62 The relative potency ratio of oral morphine
`to subcutaneous morphine is about 1:2.2
`Low dose, continuous, subcutaneous morphine
`improved both pain and quality of life compared with
`conventional, intermittent, oral or subcutaneous mor-
`phine application. Side effects were infrequent and
`mild.63 Epidural and subcutaneous morphine admin-
`istration provided better pain relief with fewer adverse
`effects compared with oral morphine treatment. Both
`treatments turned out to be comparable in terms of
`both effectiveness and acceptability. These results also
`indicate that no significant benefits are achieved by
`administering morphine epidurally compared with
`the subcutaneous route.64 Using selective criteria,
`only a minority of patients require the use of spinal
`treatment.65
`However, subcutaneous administration may not
`be advisable in patients with generalized edema who
`easily develop erythema or abscesses at the skin site,
`with coagulation disorders or with poor peripheral
`circulation.2 Other opioids should be preferred to
`morphine when high dosages are required because of
`their solubility. A dose of ,2 mL can dissolve about 1 g
`of hydromorphone as well as diamorphine. A prolon-
`gation in the duration of sites of infusion was obtained
`after initiating an equianalgesic dose of diamor-
`phine.66 Although methadone administration by the
`subcutaneous route has been shown to be an effective
`alternative, it results in topic adverse skin reactions,
`requiring either site change or discontinuation of sub-
`cutaneous methadone.67
`
`Rectalandsublingualroutes
`Apart from subcutaneous opioid administration, the
`rectal route is a commonly used alternative method of
`opioid delivery for a systemic effect. The rectal route
`has the potential to partially avoid the first-pass effect
`and is advantageous in some cases because the inser-
`tion of needles and the use of portable pumps are
`unnecessary. The differences found between the oral
`and rectal equianalgesic ratios suggest lower bioavail-
`ability by the rectal route, although similar analgesic
`levels were achieved.68 In opioid-naive patients, the
`pharmacokinetic effect of a single dose of rectal meth-
`adone was similar to that reported by the oral route.69
`The preparation, the use of surfactants, and the pres-
`ence of feces inside the rectum influence the bioavail-
`ability of drugs administered by this route. Interindi-
`
`vidual variability in rectal bioavailability also is due to
`the adsorption of the drug in the feces, spread of the
`drug to the upper part of the rectum and transporta-
`tion via the portal system to the liver, poor drug re-
`lease from the suppository, poor absorption from the
`rectal mucosa, and poor dissolution in small volumes
`of rectal fluid.70 Moreover, rectal administration can
`be uncomfortable, and progressive titration may be
`difficult because of the limited availability of commer-
`cial preparations.
`Highly lipophilic drugs, like methadone, fentanyl,
`and buprenorphine, are absorbed sufficiently sublin-
`gually and can be useful alternatives for patients who
`are not able to swallow. The sublingual administration
`of methadone results in a higher absorption than that
`of morphine (34%) due to its lipophylic characteris-
`tics. Alkalinization of the oral cavity to pH 8.5 results
`in an increase in its absorption.71 The buccal, sublin-
`gual, and nebulized routes of administration of mor-
`phine are not recommended, because the absorption
`seems to be unpredictable.2
`
`Transdermalroute
`A revitalized interest in transdermal route administra-
`tion has been developed recently. Substitution with
`transdermal (TTS) fentanyl has been reported as ben-
`eficial for patients with disabling adverse effects from
`morphine. Improved bowel function, relief of nausea
`and emesis, improved sleep quality, and morning vig-
`ilance associated with adequate pain relief have been
`reported. Most patients estimated that the fentanyl
`TTS therapy was superior to the previous therapy with
`oral morphine.72,73
`The product information recommends a mor-
`phine/fentanyl TTS conversion ratio of 150:1 in mg/
`day.74 However, recommendations for switching from
`morphine to transdermal fentanyl have been reported
`to be inaccurate.75 Direct conversion from oral mor-
`phine to transdermal fentanyl with a ratio of oral
`morphine to transdermal fentanyl of 100:1 (mg:mg/
`day) has been demonstrated to be safe and effective.
`Nonetheless, this ratio may result in subanalgesia in
`.60% of patients. The need for escape medication
`after the switch-over to fentanyl TTS was greater com-
`pared with morphine therapy. A regression analysis
`revealed an effective morphine to transdermal fenta-
`nyl ratio of 70:1.73 The reduction of the equianalgesic
`dose, using a ratio of 100:1, may prevent overdosage
`due to the interindividual differences of opioids while
`requiring a prolonged phase of dose finding. As a
`result of
`the slow pharmacokinetics,
`the patient
`should receive the last medication of the long-acting
`opioid with the first application. A reduced dose of the
`previous opioid may avoid withdrawal symptoms after
`
`5
`
`

`

`the substitution with TTS fentanyl, and escape medi-
`cation should be available during the first days of
`treatment. In this regard, oral transmucosal fentanyl
`may be a useful alternative for breakthrough pain
`events because of the rapid onset and short duration
`of action.76 In patients receiving continuous subcuta-
`neous administration of morphine, the dose should be
`reduced progressively in the first 12 hours after the
`fentanyl patch application.
`Patients with stable pain who already are receiv-
`ing regular doses of opioids and with low to medium
`opioid dose requirements are considered the ideal
`candidates for such a route of administration.77 Dys-
`phagia and nausea and/or emesis due to tumoral in-
`volvement of the upper gastrointestinal tract may rep-
`resent a useful indication.
`Problems in obtaining adequate pain relief in the
`first 72 hours and unstable pain syndromes requiring
`rapid dose escalation or reduction are common limits
`of the transdermal route of fentanyl. To account for
`possible incomplete cross-tolerance and interindi-
`vidual variability, a reduction of the calculated dose by
`a fraction of one-half or one-third and a rapid dose
`titration phase have been proposed when switching
`from morphine to TTS fentanyl.78 A day-to-day change
`of the fentanyl dosage has been proposed to accelerate
`the effective dose finding in uncontrolled cancer
`pain.79 However, the risk of overdosage and the costs
`also should be considered. The extended area of ap-
`plication required for the administration of high doses
`may be a limiting factor and is considered unsuitable
`in patients with unstable pain syndrome. Withdrawal
`symptoms and transient and mild cutaneous reactions
`to the patch can occur.73 A short-lived opioid with-
`drawal syndrome occurs in up to 10% of patients who
`are converted from different types of opioids to trans-
`dermal fentanyl. Gastrointestinal symptoms appear to
`be particularly common.80 – 82
`
`Switching the Opioid
`Toxicity may increase over time without any change in
`the opioid dose. Cancer patients can develop severe,
`persistent adverse effects due to high M6G levels even
`when they are receiving only small doses of morphine.
`Opioid metabolites have been claimed to generate
`these late effects. Moreover, dehydration can lead to a
`reduced intravascular volume and slower glomerular
`filtration, resulting in renal failure. Hydration can re-
`verse states of confusion due to the accumulation of
`opioid metabolites and can facilitate metabolite elim-
`ination, promoting dyuresis, when switching to an
`alternative opioid. Hypodermoclysis is a safe and ef-
`fective method of hydration in cancer patients with
`advanced disease.83 The reduction of the opioid dose
`
`Opioid Rotation/Mercadante
`
`1861
`
`may result in symptom improvement. However, pain
`becomes uncontrolled in most cases. Dissimilar side-
`effect profiles and a completely different intensity of
`side effects at any given level of analgesia have been
`reported in patients undergoing treatment with ap-
`parently comparable drugs.27,84 – 87
`Methadone often is used for opioid rotation.
`Methadone has the advantage of extremely low cost
`and lack of known active metabolites. Nevertheless, its
`long and unpredictable half-life and a relatively un-
`known equianalgesic dose compared with other opi-
`oids require much expertise in cancer pain manage-
`ment.86 Because methadone elimination is influenced
`minimally by renal clearance, renal insufficiency may
`represent a good indication for its use in cases of
`presumed metabolite accumulation and consequent
`toxicity with morphine administration.38,83,84,88 More-
`over, methadone is a cost-effective opioid when high
`doses of opioids are necessary.53,89 –91 Furthermore,
`methadone has been shown to have anti-NMDA prop-
`erties and may be more effective in circumstances in
`which NMDA receptor activation develops, such as in
`patients with a neuropathic mechanism associated
`with tolerance.28,29,82 For this extended activity, meth-
`adone can be considered as an opioid at a broad
`spectrum.28
`Methadone has been reported to be useful in re-
`storing opioid responsiveness in patients whose pain
`ceases to be controlled by morphine or diamor-
`phine.3,4,61,92–95 A significant improvement in pain
`control with minimal toxicity has been evidenced with
`very high doses of methadone.90 Patients with persis-
`tent pain and opioid adverse effects limiting the dose
`of
`intravenous morphine
`and hydromorphone
`achieved excellent pain relief without significant ad-
`verse effects at doses of intravenous methadone much
`lower than those suggested by the opioid conversion
`charts.61,66 A regaining of sensitivity to methadone by
`temporary rotation to hydromorphone has recently
`been observed in one cancer patient. An up-regulation
`of the receptors after the reduction in the level of
`stimulation from methadone or the synthesis of addi-
`tional receptors have been postulated to explain this
`finding.96 The clinical benefit will depend on the de-
`gree to which cross-tolerance exists to analgesia as
`well as side effects. In a retrospective comparison of
`dose ratios between opioids, the dose ratios between
`morphine and hydromorphone, and vice versa, were
`found to be 5.33 and 0.28, respectively, whereas the
`hydromorphone to methadone ratio was found to be
`1.14:1.0 (5–10 times higher than expected). Moreover,
`the hydromorphone to methadone ratio correlated
`with total opioid dose, the dose ratio being higher in
`patients receiving higher doses of hydromorphone.7 A
`
`6
`
`

`

`1862
`
`CANCER November 1, 1999 / Volume 86 / Number 9
`
`very limited cross-tolerance between methadone and
`hydromorphone was observed in an elderly cancer
`patient who was switched from 84 mg daily of subcu-
`taneous hydromorphone to 90 mg daily of oral meth-
`adone. This patient developed respiratory depression
`and noncardiogenic pulmonary edema that were re-
`sponsive to naloxone and methadone discontinua-
`tion.97 A highly significant difference in dose ratios
`between morphine to hydromorphone rotation and
`hydromorphone to morphine rotation has been
`found, indicating that the opioid to which the patient
`is rotated is relatively more potent. Unlike opioid ro-
`tation between hydromorphone and methadone,71 the
`dose ratios between morphine and hydromorphone
`were not dose dependent.98
`In a recent survey, morphine, hydromorphone,
`and methadone were used as alternative drugs. Im-
`provement was noted in 69% of patients with cogni-
`tive failure, in 33% of patients with hallucinations, in
`70% of patients with uncontrolled pain, and in 100% of
`patients with myoclonus. Renal failure was associated
`with opioid toxicity in 20% of patients only at the time
`of opioid rotation. Because renal failure may share
`common symptoms with opioid toxicity, patients may
`not benefit from opioid rotation in such a clinical
`situation.10
`The rotation of opioids has been shown to im-
`prove delirium after escalation of the opioid dosage
`has lead to the occurrence of a change in cognition or
`clouding of consciousness.93,99,100 Because the im-
`provement is progressive over the course of several
`days after a change from morphine, a long-acting me-
`tabolite of morphine has been considered responsible
`for causing delirium. A severe organic brain syndrome
`characterized by symptoms of central irritability, such
`as myoclonus, hallucinations, and agitation in the
`presence of mild renal failure, disappeared after the
`discontinuation of morphine and initiation of hydro-
`morphone.101 Significant
`improvement
`in mental
`state and nausea and emesis occurred after a change
`from morphine to oxycodone administered by the
`subcutaneous route in patients who had experienced
`acute delirium from morphine.102 Patients who devel-
`oped organic hallucinosis with significant doses of
`hydromorphone benefited from a change in the type
`of opioid used, including diamorphine, methadone,
`morphine, and haloperidol administration.103 Three
`patients who had experienced severe central nervous
`system adverse effects on a high dose of hydromor-
`phone were switched t