Establishing “Best Practices” for Opioid Rotation: Conclusions of an Expert Panel

418 Journal of Pain and Symptom Management Vol. 38 No. 3 September 2009

Special Article

Establishing ‘‘Best Practices’’ for OpioidRotation: Conclusions of an Expert PanelPerry G. Fine, MD, and Russell K. Portenoy, MD, for theAd Hoc Expert Panel on Evidence Review and Guidelines for Opioid RotationDepartment of Anesthesiology, Pain Research Center (P.G.F.), University of Utah School of Medicine,

Salt Lake City, Utah; and Department of Pain Medicine and Palliative Care (R.K.P.), Beth Israel

Medical Center, New York, New York, USA

Abstract
Opioid rotation is a strategy applied during opioid therapy for pain that refers to a switchfrom one opioid to another in an effort to improve clinical outcomes (benefits or harms). Itbegins with the selection of a new drug at a starting dose that minimizes potential risks whileideally maintaining analgesic efficacy. The selection of a starting dose must be informed byan estimate of the relative potency between the existing opioid and the new one. Clinicallyrelevant estimates of relative analgesic potency have been codified in the ‘‘equianalgesic dosetable,’’ which has been used with little modification for more than 40 years. New informationabout relative potency and the growing implementation of long-term opioid therapy forchronic pain provided a strong rationale for the convening of an expert panel to discuss thescientific foundation to opioid rotation and the elements that now should inform a clinicalguideline for this practice. The panel affirmed both the value and the limitations of thecurrent equianalgesic dose table and proposed a guideline intended to promote safety duringopioid rotation. J Pain Symptom Manage 2009;38:418e425. � 2009 U.S. CancerPain Relief Committee. Published by Elsevier Inc. All rights reserved.
Key Words
Opioid, opioid rotation, relative potency, equianalgesic dose, pain management
This project, including the convening of an expertpanel and the preparation of this article, was sup-ported by an unrestricted grant from Endo Pharma-ceuticals to the Department of Pain Medicine andPalliative Care, Beth Israel Medical Center, NewYork, New York. The authors declare no financialconflicts with respect to this work.

See Appendix for members of the Ad Hoc ExpertPanel.

Address correspondence to: Perry G. Fine, MD, Depart-ment of Anesthesiology, Pain Research Center, Uni-versity of Utah School of Medicine, 615 ArapeenDrive, Suite 200, Salt Lake City, UT 84108, USA.E-mail: [email protected]

Accepted for publication: June 22, 2009.

� 2009 U.S. Cancer Pain Relief CommitteePublished by Elsevier Inc. All rights reserved.

IntroductionOpioid therapy for acute or chronic pain

requires individualization of the dose, with theobjective of identifying a favorable balance be-tween analgesia and side effects. Opioid-relatedadverse effects are common and may be treat-ment-limiting. ‘‘Opioid rotation,’’ a plannedswitch from one opioid to another in an effortto improve outcomes, is one strategy in this set-ting. Widespread use of the approach hasevolved over many years despite the lack of evi-dence-based guidelines or a uniformly accepted‘‘best practice’’ derived from expert opinion.1

Opioid rotation begins with the selection ofa safe and reasonably effective starting dose

0885-3924/09/$esee front matterdoi:10.1016/j.jpainsymman.2009.06.002

mailto:[email protected]

Vol. 38 No. 3 September 2009 419Opioid Rotation Guidelines

for the new opioid. Once initiated, the new ther-apy must be individualized through the processof dose titration and treatment of adverseeffects. Given the large differences in potencyamong opioid drugs, the selection of a startingdose must be informed by an estimate of the rel-ative potency between the existing opioid andthe new one. Ideally, clinicians shouldimplement the opioid switch at an initial dosethat does not result in adverse effects or absti-nence, and maintains efficacy. In clinical prac-tice, determination of the optimal initial dosewhen rotating opioids is a challenge.

More than 40 years ago, methods for measur-ing relative opioid potency were developed.2

Data from numerous studies based on this meth-odology were subsequently adapted into ‘‘equia-nalgesic dose tables’’ to help guide conversionsbetween opioids. However, the original relativepotency assays had limitations and new evidenceon relative potency is now available. The purposeof this article is to summarize the findings of anexpert panel (see Appendix) that reviewed theevidence pertaining to relative opioid potencyto develop guidelines that comport with the cir-cumstances of everyday clinical practice.

MethodsAn interdisciplinary expert panel with clini-

cal and research expertise in opioid pharma-cology was convened to review the evidenceand formulate recommendations. The existingliterature on relative potency was collated andcritiqued as background information.3 Thepanel met for a full day of discussions, whichwere taped and transcribed. The transcriptwas then reviewed for key conclusions relatedto the definition of opioid rotation, thestrengths and limitations of the existing litera-ture on relative potency, the clinical consider-ations relevant to the clinical application ofrelative potency estimates, and the elementsof a new guideline for opioid rotation. Notingthat there have been no prospective clinical tri-als to evaluate the impact of opioid rotation onclinical outcomes in patients with acute orchronic pain, the panel endorsed the needfor a guideline based on best evidence andclinical experience, prioritizing safety in thepractice of opioid rotation. All panel membersprovided independent editorial input, andconsensus views are summarized in this article.

Conclusions of the Expert PanelThe interdisciplinary panel reached con-

sensus on an array of issues related to thepractice of opioid rotation. These areas ofagreement supported the development ofa guideline.

Definition of Opioid RotationThe panel agreed that the goal of opioid

rotation is to improve therapeutic effectivenessduring opioid therapy. The following defini-tion was proposed:

Opioid rotation (or switching) is a change inopioid drug or route of administration with thegoal of improving outcomes.

The panel agreed on several points that mayclarify or expand the definition.

� Opioid rotation is best viewed as one strat-egy among many to address unsatisfactoryoutcomes following opioid administrationor dose escalation. As such, it is importantto note that opioid rotation may or maynot be the best approach at any point intime, that more than one switch may be re-quired to obtain satisfactory therapeuticoutcomes, and that some patients willnot respond well even to trials of multipleopioid drugs.1 Given the lack of clear evi-dence regarding clinical benefits of opioidrotation, the decision to undergo a trial ofopioid rotation should be based ona shared decision-making approach thatemphasizes the substantial uncertaintiesin estimating benefits and harms.� Both the immediate and long-range goals

of opioid rotation are to establish an opi-oid regimen that is more effective thanthe prior therapy. Effectiveness encom-passes improved analgesic efficacy, re-duced adverse effects, and/or improvedtreatment-related outcomes associatedwith physical and/or psychosocial func-tioning or quality of life.� The definition of opioid rotation is pre-

mised on treatment principles that shouldgeneralize to a broad population ofpatients and provide sufficient clinicalflexibility to address the large degree ofindividual variation encountered indiverse clinical settings.

420 Vol. 38 No. 3 September 2009Fine and Portenoy

Indications for Opioid RotationThe panel identified potential indications

for a switch in the existing opioid therapy:

� Occurrence of intolerable adverse effectsduring dose titration� Poor analgesic efficacy despite aggressive

dose titration� Problematic drug-drug interactions� Preference or need for a different route of

administration� Change in clinical status (e.g., concern

about drug abuse or the development ofmalabsorption syndrome) or clinicalsetting that suggests benefit from anopioid with different pharmacokineticproperties� Financial or drug-availability considera-

tions

The group excluded pain crises from the listof potential indications, because the manage-ment of these complex clinical scenarios wasbeyond the scope of a guideline focused solelyon opioid rotation. It also agreed that the term‘‘poor opioid responsiveness’’ to describe a clin-ical situation that would justify a switch to analternative opioid should be used cautiously,because some clinicians perceive this term toimply that a patient is not responsive toopioids in general.

Clinical Considerations in the Practiceof Opioid Rotation

The expert panel next focused on a numberof clinical considerations relevant to the prac-tice of opioid rotation:

� To optimize outcomes, the approachshould begin with an assessment of anarray of factors that may influence deci-sion making relevant to the selection ofa new drug and initial dose, the processof dose individualization, and other fac-tors that may help ensure that the newtherapy is optimized. These includedemographic factors, such as age andrace, relevant disease- and treatment-related factors, comorbid medical condi-tions, and concomitant pharmacotherapy.� Implementation of opioid rotation also

must consider the clinical care environ-ment (e.g., outpatient, inpatient, long-

term care, hospice) and psychosocialcircumstances.� In considering which specific opioid

should be tried next, clinicians shouldweigh the patient’s history of any drug sen-sitivities or experience with specific drugs,drug characteristics that may increase ordecrease safety or efficacy given the pa-tient’s clinical status, drug characteristicsthat may offer previously unrealized bene-fits unrelated to pain relief (e.g., conve-nience, improved adherence, lessreliance on oral administration, or accessto a regular nonopioid drug in a combina-tion product), and problems related to fi-nancial issues or insurance.� If an opioid is selected that may require

enhanced knowledge for safe prescribing,such as methadone or buprenorphine, cli-nicians should ensure that skills are ade-quate, obtain appropriate consultation,or refer to persons with expertise in pre-scribing these drugs.

In discussing these considerations, panelmembers emphasized several specific observa-tions. For example, the distinction betweenacute and chronic pain has not been previ-ously emphasized in clinical discussions of opi-oid rotation, but may represent an importantissue. To reduce the risk of unintentional over-dose when pain intensity may be changingquickly or rapid titration may be needed aftera change in drugs, opioid rotation in the set-ting of acute pain management usually shoulduse a short-acting drug rather than an ex-tended-release formulation or methadone.

The panel also discussed the myriad of socialcircumstances that may influence drug selec-tion, starting dose, or the protocol applied todose titration. The decision to recommendone drug over another may be influenced, forexample, by recognition that a patient liveswith a substance abuser who may complicate ef-forts to protect the prescription, or an elderlycaregiver who may not be able to monitor thepatient. It was noted that the need for opioidswitching may be driven by formulary restric-tions, commonly encountered in Medicaid pro-grams, managed care plans, long-term carefacilities, or hospice programs. Clinicians whopractice in those settings should have opioid ro-tation guidelines that protect their patients.


The panel also observed that withdrawal im-mediately after the switch to a new opioid hasreceived little attention in the literature. Mostclinicians have little experience in managingacute withdrawal and many appear to have lit-tle recognition of the more subtle manifesta-tions of protracted withdrawal, such asdysphoria, fatigue, or sleep disturbance.4 Thepanel advised that clinicians who frequentlyoffer opioid rotation should be prepared torecognize and manage opioid withdrawal syn-drome, and they may need additional educa-tion about this issue.

The final observation highlighted by thepanel was that opioid rotation must be viewedin the larger context of opioid therapy forpain. Long-term therapy for chronic noncancerpain remains controversial, and recent evidence-based guidelines indicate the importance oflinking routine risk assessment to optimal phar-macotherapy.5 This type of guidance is founda-tional to any of the best practices that togethercomprise the therapy, including opioid rotation.

Need to Reevaluate the EquianalgesicDose Table

The panel recognized the importance of theequianalgesic dose table, and the value of hav-ing the table represent the results of well-con-trolled trials. An extensive review of relativepotency studies, however, highlighted boththe limitations of the existing data and thechallenges inherent in applying them to opi-oid rotation in the clinical setting.3 For exam-ple, almost all trials of relative potency wereshort-term trials conducted in patients withacute postoperative pain or patients with can-cer pain on low-dose opioids, and may notbe directly applicable to patients with chronicnoncancer pain on relatively high doses. Al-though the panel agreed that the currentequianalgesic dose table should be used untilan alternative is created, it also concludedthat the use of the conventionally acceptedconversion ratios without adjustments for theindividual patient would be dangerous, andthat a modern guideline for opioid rotationmust emphasize the goal of safety by specifyingthe potential for dose adjustments after calcu-lation of the equianalgesic dose. The conver-sion ratios included in the table are merelya broad indicator of relative analgesic potency,which must be considered in tandem with

other factors when switching from one opioidregimen to another.

The expert panel discussed the viability ofa new equianalgesic dose table that wouldinclude all the opioids now used in practiceand would have conversion ratios that incorpo-rated the type of dose adjustments that mightbe included in a modern guideline for opioidrotation. Although these ‘‘adjusted’’ ratioswould no longer be directly representative ofdata from randomized controlled trials, theycould be applied to opioid rotation withoutrequiring stepwise calculations, and for thisreason, should reduce the risk of error. Giventhe complexity of this pharmacology, a newequianalgesic dose table would likely replacethe single conversion ratio with a matrix of fre-quently applied ratios, and would presumablybe best suited for an electronic medium.

The panel identified numerous gaps in the lit-erature on relative potency, each of which com-plicates efforts to create a new equianalgesicdose table. Some drug pairs have been evalu-ated in several trials, which have yielded incon-sistent relative potency ratios, or ratios shownto change with direction of the switch or theduration of treatment.3 Many influences on po-tency, such as genetically determined differ-ences in drug metabolism,6 have not yet beenevaluated in relative potency studies, and theirimpact can only be inferred. Although the cur-rent opioid dose is likely to have an effect onthe ratio necessary to select an equianalgesicdose of any opioid, this has been confirmedonly for conversions to methadone,7,8 and inthe case of this drug, the effect of dose on rela-tive potency is assumed to be greater than wouldbe the case with other drugs. Owing to the lackof data from studies of these and other factors,a revised table would necessarily include ratioslargely informed by clinical judgment and expe-rience rather than evidence.

These challenges notwithstanding, the ex-pert panel concluded that it would be worthpursuing the development of a more sophisti-cated equianalgesic table that would incorpo-rate a guideline for dose adjustment basedon the existence of factors that could influ-ence relative potency. If created, studies couldbe designed to validate the model incorpo-rated into the table, thereby demonstratingits utility overall while potentially testing thevalidity of each element.


Guideline for Opioid RotationIn the absence of a simple approach for revis-

ing the equianalgesic dose table, the expertpanel emphasized the need for a guideline fo-cused on opioid rotation that would continueto rely on the existing equianalgesic dose table,but promote safety through dose adjustmentsbased on the best evidence available and expertopinion. In publications that include referenceto the use of equianalgesic doses to switch opi-oid drugs, reference to an appended guidelineshould be encouraged (Table 1).

The guideline for opioid rotation uses exist-ing equianalgesic dose tables as a reasonablestarting point, though it is difficult to predictwhether an individual patient will react to anopioid switch as anticipated.3 To reduce therisk of unintentional overdose, the conversionratio calculated for a patient undergoing opi-oid rotation should be adjusted based on clin-ical assessment of risk.9 To address risk,strategies for safe use of the equianalgesicdose table should involve a two-step process:

� Step 1: Calculate an automatic safety factor.� Step 2: Calculate an additional dose ad-

justment based on the assessed patientcharacteristics.

The safety factor (Step 1) may be conceptu-alized as an automatic reduction in the

Table 1Guideline for Opio

Step 1� Calculate the equianalgesic dose of the new opioid based on the� If switching to any opioid other than methadone or fentanyl, ide

lower than the calculated equianalgesic dose.� If switching to methadone, identify this window at 75%e90% low

very high opioid doses (e.g., 1000 mg morphine equivalents/dto methadone at doses of 100 mg or greater per day; considermonitoring.

� If switching to transdermal fentanyl, calculate dose conversiopackage insert for these formulations.

� Select a dose closer to the lower bound (25% reduction) or the ureduction window on the basis of a clinical judgment that the eqrespectively, to the specific characteristics of the opioid regimen� Select a dose closer to the upper bound (50% reduction) of th

of the current opioid regimen, is not Caucasian, or is elde� Select a dose closer to the lower bound (25% reduction) of the

or is undergoing a switch to a different route of systemic d

Step 2� Perform a second assessment of pain severity and other medical or

an additional increase or decrease of 15%e30% to enhance theconversely, unlikely to cause withdrawal or opioid-related side e

� Have a strategy to frequently assess initial response and titrate the� If a supplemental ‘‘rescue dose’’ is used for titration, calculate this

an appropriate interval; if an oral transmucosal fentanyl formulower doses irrespective of the baseline opioid dose.

equianalgesic dose within a narrow window.This automatic reduction is justified on the ba-sis of extensive experience demonstrating thatthe calculated equianalgesic dose commonlyunderstates the actual potency of the newdrug because of individual variation and theimpact of incomplete cross-tolerance in thechronic treatment setting.1,9 Based on panelconsensus, the window to apply to mostswitches is a reduction of 25%e50% of the cal-culated equianalgesic dose.

The expert panel endorsed three exceptionsto this automatic 25%e50% reduction in thecalculated equianalgesic dose. First, whenswitching to methadone, evidence of higher-than-anticipated potency in the clinical settingsuggests that the automatic reduction in thecalculated dose should be substantially greater,usually 75%e90%.10 Although this steepreduction probably is not needed when theswitch to methadone is occurring from a rela-tively low-dose opioid regimen, the decisionto use a smaller reduction requires particularlycareful monitoring after the change. Many cli-nicians use the 75%e90% reduction in allcases, recognizing that initial underdosing islikely and that dose titration will be necessary.Some clinicians opt to alter the automatic re-duction by applying a stepwise reduction basedon the dose of the regimen before the switch

id Rotation

equianalgesic table.ntify an ‘‘automatic dose reduction window’’ of 25%e50%

er than the calculated equianalgesic dose. For individuals onay or higher), great caution should be exercised in convertinginpatient monitoring, including serial electrocardiogram

ns based on the equianalgesic dose ratios included in the

pper bound (50% reduction) of this automatic dose-uianalgesic dose table is relatively more or less applicable,or patient.

e reduction if the patient is receiving a relatively high doserly or medically frail.

reduction if the patient does not have these characteristicsrug administration using the same drug.

psychosocial characteristics to determine whether to applylikelihood that the initial dose will be effective for pain, orffects.dose of the new opioid regimen to optimize outcomes.

at 5%e15% of the total daily opioid dose and administer atlation is used as a rescue dose, begin dosing at one of the


to methadone, using standard low, medium,and high conversion ratios depending on thecurrent opioid dose.3,7

Second, the original studies of transdermalfentanyl led to the development of a conver-sion table from oral or parenteral opioids totransdermal fentanyl. This one-way conversionchart incorporated a safety factor, and subse-quent experience supported the conclusionthat the equianalgesic ratios printed in thelabel were conservative enough that an addi-tional automatic reduction in the calculatedequianalgesic dose was not required.11

Third, studies have confirmed that a largeproportion of patients obtain satisfactoryresults when treatment with the newer oraltransmucosal fentanyl citrate formulationsare initiated at the lowest available doses, irre-spective of the baseline opioid regimen.12 Thisobservation suggests that these formulations,which are used as supplemental treatmentsfor breakthrough pain, should not be includedin an opioid rotation guideline and shouldalways be initiated at one of the lower dosesin practice.

How much to adjust the calculated analgesicdose (i.e., a reduction of 25%e50% of thecalculated equianalgesic dose) should bebased on a clinical judgment about the likeli-hood that the dose ratio in the equianalgesictable applies to the patient in question. Manycharacteristics of the patient or the analgesicregimen suggest that the conversion ratio in-cluded in the table may not be fully applica-ble.3 A larger reduction (e.g., 50% reductionin most cases) might be appropriate, for exam-ple, if the current opioid regimen uses a rela-tively high dose or if the patient hasadvanced age or renal disease. Patients ofnon-Caucasian race may be more sensitive toopioid effects for various reasons,3 and thischaracteristic may also suggest the use of thishigher dose reduction. In contrast, a smallerreduction (e.g., 25% reduction in most cases)might be appropriate when the patient is ona relatively low-dose regimen and is perceivedto have characteristics comparable to the clin-ical populations that were studied in the earlyrelative potency assays. Adjustment closer tothe lower bound also is reasonable when theswitch to a new regimen involves changingroutes of administration without changingthe drug.

The expert panel supported the use of a sec-ond evaluation (Step 2) for dose adjustment,which would be applied after the automatic re-duction in the calculated equianalgesic dose isselected. This second step requires an assess-ment focusing on the severity of the pain atthe time of the change and the existence ofother medical or psychosocial factors that po-tentially alter potency or shift the likelihoodthat the initial dose of the new drug will be an-algesic, relatively free of adverse effects, andunlikely to precipitate withdrawal. In manycases, the second assessment will concludethat the initial adjusted dose (Step 1) can beused as the starting dose. In some cases, how-ever, the second evaluation may suggest thatan additional change in this dose, usually inthe range of 15%e30%, would be prudent.

For example, a patient undergoing a switchfrom morphine to hydromorphone may firstbe considered for an initial (Step 1) automatic25% reduction in the calculated equianalgesicdose. If the second assessment (Step 2) indi-cates that pain is very severe, however, a reason-able judgment would be to eliminate thisreduction. In another case, a patient undergo-ing a switch to hydromorphone may first beconsidered for an initial (Step 1) automatic25% reduction in the calculated equianalgesicdose, and the second assessment reveals mod-erate pain, mild confusion, and the use ofmultiple other drugs. These patient-specificobservations from the second evaluation(Step 2) may lead to the decision to reducethe dose by an additional 15%.

The expert panel acknowledged that theseelements of the guideline were likely to be var-iably implemented, given the lack of high-quality evidence to determine relative opioidpotency in individual patients. The recommen-dations are intended to reduce risks associatedwith opioid rotation, but provide no guaranteethat the initial dose of the new drug is ade-quate. Accordingly, the panel also emphasizedthat a guideline for opioid rotation must pres-ent a strategy for titration of the dose after thechange to a new drug is initiated. Dependingon the approach selected by the clinician,this may or may not involve a coadministeredshort-acting supplemental dose, often termedthe ‘‘rescue’’ dose. If a rescue dose is used, itconventionally is initiated at 5%e15% of thetotal daily dose of the new medication and


titrated as the baseline dose is increased.As noted, however, the oral transmucosal fen-tanyl formulations represent an important ex-ception to this empirical approach, andusually are started at one of the lower doses ir-respective of the baseline opioid dose.

Need for ResearchWell-designed studies that compare out-

comes among persons who undergo opioid ro-tation and those who are managed with doseescalations of the current opioid, opioid with-drawal, or other strategies are needed to clarifyrisks and benefits of opioid rotation. Studiesshould be conducted to determine who ismore likely to benefit from opioid rotationand the effects of applying different dose-conversion strategies. Studies that assess rela-tive potency estimates in different populations,during treatment with newer formulations,during very short- (e.g., acute pain settings)vs. long-term therapy, and in patients on rela-tively high doses of opioids, would be valuable.The impact of prior opioid dose on thepotency of a new drug which appears to beparticularly important, when methadone10 isadministered, should be studied with otherdrugs as well. The potential for bidirectionalchange in relative potency should be investi-gated across varied pairs of drugs. Sources ofvariation that may systematically alter potency,including demography, pain-related factors,and disease-related factors, remain to beinvestigated.3

SummaryAlthough opioid rotation is a common prac-

tice, review of the existing literature and dis-cussion by an expert panel revealedsubstantial limitations in the pertinent evi-dence and a lack of clear consensus abouta ‘‘best practice’’ approach. This effort hasyielded a proposed two-step guideline for opi-oid rotation, which emphasizes a safe strategyfor switching drugs in diverse populationsthat is tailored to the assessments of potential

benefits and harms in individual patients. Fu-ture studies are needed to expand the evi-dence base and refine the guideline.

References1. Quigley C. Opioid switching to improve pain

relief and drug tolerability. Cochrane DatabaseSyst Rev 2004;(3):CD004847.

2. Houde R, Wallenstein S, Beaver W. Evaluationof analgesics in patients with cancer pain. ClinPharm 1966;1:59e97.

3. Knotkova H, Fine PG, Portenoy RK. Opioidrotation: the science and the limitations of equianal-gesic dose table. J Pain Symptom Manage 2009;38(3):426e439.

4. Shi J, Zhao LY, Epstein DH, Zhang XL, Lu L.Long-term methadone maintenance reduces pro-tracted symptoms of heroin abstinence andcue-induced craving in Chinese heroin abusers.Pharmacol Biochem Behav 2007;87:141e145.

5. Chou R, Fanciullo GJ, Fine PG, et al. Clinicalguidelines for the use of chronic opioid therapy inchronic noncancer pain. J Pain 2009;10(2):113e130.

6. Caraco Y, Sheller J, Wood AJ. Impact of ethnicorigin and quinidine coadministration on codeinedisposition and pharmacodynamic effects. J Phar-macol Exp Ther 1999;290:413e422.

7. Lawlor PG, Turner KS, Hanson J, Bruera ED.Dose ratio between morphine and methadone inpatients with cancer pain: a retrospective study. Can-cer 1998;82(6):1167e1173.

8. Ripamonti C, Groff L, Brunelli C, et al. Switch-ing from morphine to oral methadone in treatingcancer pain: what is the equianalgesic dose ratio?J Clin Oncol 1998;16(10):3216e3221.

9. Pereira J, Lawlor P, Vigano A, Dorgan M,Bruera E. Equianalgesic dose ratios of opioids: a crit-ical review and proposals for long-term dosing.J Pain Symptom Manage 2001;22:672e687.

10. Gazelle G, Fine PG. Methadone for pain. J Pall-iat Med 2004;7(2):303e304.

11. Duragesic� (Fentanyl Transdermal System)Full Prescribing Information: Ortho-McNeil-JanssenPharmaceuticals, Inc., Titusville, NJ, USA. 2008.

12. Portenoy RK, Taylor D, Messina J, Tremmel L. Arandomized, placebo-controlled study of fentanylbuccal tablet for breakthrough pain in opioid--treated patients with cancer. Clin J Pain 2006;22:805e811.


Appendix

Ad Hoc Expert Panel on Evidence Review and Guidelines for Opioid Rotation

Panel members and authors:

Russell K. Portenoy, MD, Co-Chair (Beth Israel Medical Center, New York, NY)

Perry G. Fine, MD, Co-Chair (University of Utah, Salt Lake City, UT)

Roger Chou, MD (Oregon Health Sciences University, Portland, OR)

Ricardo A. Cruciani, MD, PhD (Beth Israel Medical Center, New York, NY)

Deborah Gordon, RN, MS, FAAN (University of Wisconsin Hospitals and Clinics, Madison, WI)

Charles E. Inturrisi, PhD (Weill Cornell Medical College, New York, NY)

Kenneth Jackson, PharmD (Pacific University, Hillsboro, OR)

Michael Moskowitz, MD (Bay Area Pain Medical Associates, Mill Valley, CA)

Gavril Pasternak, MD, PhD (Memorial Sloan-Kettering Cancer Center, New York, NY)

Srinivasa Raja, MD (John Hopkins School of Medicine, Baltimore, MD)

Lynn Webster, MD (Lifetree Medical, Inc., Salt Lake City, UT)

Donna S. Zhukovsky, MD (M. D. Anderson Cancer Center, Houston, TX)

Documents

Establishing “Best Practices” for Opioid Rotation: Conclusions of an Expert Panel