See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/51744577

QT Interval Screening in Methadone Maintenance Treatment: Report of a

SAMHSA Expert Panel

Article  in  Journal of Addictive Diseases · October 2011

DOI: 10.1080/10550887.2011.610710 · Source: PubMed

CITATIONS

43READS

259

13 authors, including:

Some of the authors of this publication are also working on these related projects:

Opioid physiology View project

Holter detection of sleep disordered breathing View project

Mary Jeanne Kreek

The Rockefeller University

596 PUBLICATIONS   23,398 CITATIONS   

SEE PROFILE

Davendra Mehta

Icahn School of Medicine at Mount Sinai

55 PUBLICATIONS   1,669 CITATIONS   

SEE PROFILE

Mark Haigney

Uniformed Services University of the Health Sciences

117 PUBLICATIONS   2,879 CITATIONS   

SEE PROFILE

Bonnie Wilford

Coaliltion On Physician Education in Substance Use Disorders

25 PUBLICATIONS   470 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Mark Haigney on 29 September 2015.

The user has requested enhancement of the downloaded file.

This article was downloaded by: [Mark Haigney]On: 26 January 2012, At: 07:33Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Journal of Addictive DiseasesPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/wjad20

QT Interval Screening in Methadone MaintenanceTreatment: Report of a SAMHSA Expert PanelJudith A. Martin MD a , Anthony Campbell RPh, DO b , Thomas Killip MD c , Margaret Kotz DOd , Mori J. Krantz MD e , Mary Jeanne Kreek MD f , Brian A. McCarroll DO, MS g , DavendraMehta MD, PhD h , J. Thomas Payte MD i , Barry Stimmel MD, FASAM h , Trusandra Taylor MD,MPH j , Mark C.P. Haigney MD, FAHA k & Bonnie B. Wilford MS la BAART Turk Street Clinic, San Francisco, CAb Center for Substance Abuse Treatment of the Substance Abuse and Mental Health ServicesAdministration, Rockville, MDc Beth Israel Medical Center, New York, NYd University Hospitals of Cleveland, OHe Denver Health Medical Center, Denver, COf Rockefeller University, New York, NYg BioMed Behavioral Healthcare, Inc., Sterling Heights, MIh Mt. Sinai School of Medicine, New York, NYi Colonial Management Group, Inc., Orlando, FLj JEVS Human Services, Philadelphia, PAk Uniformed Services University of the Health Sciences, Bethesda, MDl JBS International, Inc., Easton, MD

Available online: 25 Oct 2011

To cite this article: Judith A. Martin MD, Anthony Campbell RPh, DO, Thomas Killip MD, Margaret Kotz DO, Mori J. Krantz MD,Mary Jeanne Kreek MD, Brian A. McCarroll DO, MS, Davendra Mehta MD, PhD, J. Thomas Payte MD, Barry Stimmel MD, FASAM,Trusandra Taylor MD, MPH, Mark C.P. Haigney MD, FAHA & Bonnie B. Wilford MS (2011): QT Interval Screening in MethadoneMaintenance Treatment: Report of a SAMHSA Expert Panel, Journal of Addictive Diseases, 30:4, 283-306

To link to this article: http://dx.doi.org/10.1080/10550887.2011.610710

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form toanyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses shouldbe independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,

proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly inconnection with or arising out of the use of this material.

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

Journal of Addictive Diseases, 30:283–306, 2011Copyright c© Taylor & Francis Group, LLCISSN: 1055-0887 print / 1545-0848 onlineDOI: 10.1080/10550887.2011.610710

SAMHSA REPORT

QT Interval Screening in MethadoneMaintenance Treatment:

Report of a SAMHSA Expert Panel

Judith A. Martin is affiliated with BAART Turk Street Clinic, San Francisco, CA. Anthony Campbellis affiliated with the Center for Substance Abuse Treatment of the Substance Abuse and Mental HealthServices Administration, Rockville, MD. Thomas Killip is affiliated with the Beth Israel Medical Center,New York, NY. Margaret Kotz is affiliated with the University Hospitals of Cleveland, OH. Mori J. Krantzis affiliated with the Denver Health Medical Center, Denver, CO. Mary Jeanne Kreek is affiliated with theRockefeller University, New York, NY. Brian A. McCarroll is affiliated with BioMed Behavioral Healthcare,Inc., Sterling Heights, MI. Davendra Mehta is affiliated with the Mt. Sinai School of Medicine, New York,NY. J. Thomas Payte is affiliated with Colonial Management Group, Inc., Orlando, FL. Barry Stimmel isaffiliated with Mt. Sinai School of Medicine, New York, NY. Trusandra Taylor is affiliated with JEVS HumanServices, Philadelphia, PA. Mark C.P. Haigney is affiliated with the Uniformed Services University of theHealth Sciences, Bethesda, MD. Bonnie B. Wilford is affiliated with JBS International, Inc., Easton, MD.

Address correspondence to: Bonnie B. Wilford, MS, JBS International, Inc., 210 Marlboro Avenue, Suite31, PMB 187, Easton, MD 21601 (E-mail: BBWilford@aol.com).

The authors thank Anthony Campbell, RPh, DO, and Bonnie B. Wilford, MS, for project management,Lilian Hockensmith and Mary A. Kelly, MSLS, for research support, Ellen Dreyer, RN, and Gwen Littman,MD, for editorial support, and Adriana Padget for administrative support. The Substance Abuse and MentalHealth Services Administration (SAMHSA) funded the work of the Panel. All members of the Expert Panelhad access to the information, participated in its interpretation, and had an opportunity to review and modifythe report.

Consistent with requirements of the Accreditation Council on Continuing Medical Education (ACCME),members of the Expert Panel and project staff were asked to report any relevant financial relationships. Thefollowing Panel members, ex officio members, guests, and staff reported no known or potential conflicts ofinterest: Mr. Bowman, Dr. Campbell, Dr. Fan, Dr. Haigney, Dr. Kahn, Dr. Killip, Dr. Kreek, Mr. Lubran, Dr.Martin, Dr. Payte, Dr. Rappaport, Dr. Stimmel, Dr. Taylor, and Mrs. Wilford.

Dr. Gourevitch reported research support from Alkermes, Inc. Dr. Kotz reported honoraria paid to herspouse by Abbott Laboratories and Forrest Laboratories. Dr. Krantz reported conference support from Reckitt-Benckiser, Inc. Dr. McCarroll reported honoraria from Reckitt-Benckiser, Inc. Dr. Mehta reported honorariafrom Medtronic and Boston Scientific. Dr. O’Keeffe reported consulting fees from Catalyst PharmaceuticalPartners, Inc., and Reckitt-Benckiser, Inc., and an ownership interest in Kingston Pharmaceuticals, Inc.

The views, opinions, and content of this document are those of the Expert Panel members and otherreferenced sources and do not necessarily reflect the views, opinions, or policies of the Substance Abuse andMental Health Services Administration (SAMHSA) or any other part of the U.S. Department of Health andHuman Services (DHHS).

The conclusions of the Panel have not been adopted by SAMHSA as either Treatment ImprovementProtocols (TIPs) or certification standards. Accordingly, the contents of this report do not constitute Federalguidelines.

This report is intended to enhance patient care, but it does not supplant clinical judgment in the care ofindividual patients. Moreover, the advice given here may not apply to all patients or clinical situations.

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

284 JOURNAL OF ADDICTIVE DISEASES

Judith A. Martin, MDAnthony Campbell, RPh, DO

Thomas Killip, MDMargaret Kotz, DOMori J. Krantz, MD

Mary Jeanne Kreek, MDBrian A. McCarroll, DO, MSDavendra Mehta, MD, PhD

J. Thomas Payte, MDBarry Stimmel, MD, FASAMTrusandra Taylor, MD, MPH

Mark C.P. Haigney, MD, FAHABonnie B. Wilford, MS

ABSTRACT. In an effort to enhance patient safety in opioid treatment programs, the SubstanceAbuse and Mental Health Saervices Administration convened a multi-disciplinary Expert Panel onthe Cardiac Effects of Methadone. Panel members (Appendix A) reviewed the literature, regulatoryactions, professional guidances, and opioid treatment program experiences regarding adverse cardiacevents associated with methadone. The Panel concluded that, to the extent possible, every opioidtreatment program should have a universal Cardiac Risk Management Plan (incorporating clinicalassessment, electrocardiogram assessment, risk stratification, and prevention of drug interactions) forall patients and should strongly consider patient-specific risk minimization strategies (such as carefulpatient monitoring, obtaining electrocardiograms as indicated by a particular patient’s risk profile,and adjusting the methadone dose as needed) for patients with identified risk factors for adversecardiac events. The Panel also suggested specific modifications to informed consent documents, patienteducation, staff education, and methadone protocols.

KEYWORDS. Methadone maintenance treatment, QTc, cardiac risk, patient screening

BACKGROUND AND PURPOSE

Drug-induced long QT syndrome occurswith a wide range of medications, includingmethadone.1,2 Prolongation of the QT intervalserves as a surrogate marker for the risk of de-veloping Torsades de Pointes (TdP), a relativelyrare and potentially lethal ventricular arrhyth-mia. The association between methadone, pro-longed QT interval, and TdP came into sharpfocus in 2006, when the U.S. Food and DrugAdministration (FDA) issued a physician safetyalert regarding fatalities and cardiac arrhythmiasassociated with methadone (Appendix B).3 Thiswas followed by a warning in the manufacturer’sproduct labeling.4 Warnings about the associ-ation between methadone and cardiac arrhyth-

mias also are catalogued in Thompson Reuters’MICROMEDEX5 and a web site that dynami-cally archives QT-prolonging drugs.6

Methadone has a long history as an effectivetreatment for opioid addiction and thus is in wideuse in opioid treatment programs in the UnitedStates. Therefore, it is of concern that severalrecent reports suggest that methadone prolongsthe QT interval when used at doses higher than60 to 120 mg/day, which often are required insuch treatment.6–9 Also of concern is evidenceof a significant knowledge gap among opioidtreatment program medical staff regarding therisk of QT prolongation and TdP associated withmethadone.10

The prevalence of QT interval prolonga-tion or TdP in patients who are treated with

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

SAMHSA Report 285

methadone has not been firmly established,11

although marked prolongation of the correctedQT (QTc) interval appears to occur in only 2%of opioid treatment program patients.12 To ad-dress the topic, the Substance Abuse and Men-tal Health Services Administration (SAMHSA)convened a multi-disciplinary Expert Panel onthe Cardiac Effects of Methadone and chargedit with evaluating the available evidence andformulating recommendations to enhance thecardiac safety of patients in opioid treatmentprograms. The Panel also reviewed the contextfor the situation. This included a discussion ofmethadone’s unequivocal effectiveness in treat-ing opioid addiction, current regulatory actionsand recommendations from private sector orga-nizations, guidance documents from other coun-tries, and opioid treatment program providers’awareness of QT and TdP.

Panel members are cardiologists with back-grounds in addiction medicine, electrophysiol-ogists, addiction treatment specialists, medicaleducators and researchers, and experts in drugdevelopment. Ex officio members include rep-resentatives of scientific and regulatory author-ities (including the National Institute on DrugAbuse and the Food and Drug Administration[FDA]) and stakeholder organizations that sharea commitment to assuring the safety of patientsin opioid addiction treatment.

The principles that guided the Panel’s ap-proach included a recognition that methadonehas been associated with a reduction in over-all mortality,1,2 has few therapeutic alternatives,and is cost-effective. Therefore, the Panel op-erated on the premise that methadone must re-main widely available in the United States forthe treatment of opioid addiction.

The purpose of the Panel’s work, as reflectedin this report, is solely to provide a workabletemplate for opioid treatment program admin-istrators and clinical staff who are attemptingto develop and implement cardiac safety stan-dards for methadone induction and maintenancetreatment in federally certified opioid treatmentprograms.

SAMHSA encourages providers to considerthe Panel’s report and to take action to the extentthat they are clinically, administratively, and fi-nancially able to do so. However, nothing in the

report is intended to create a legal standard ofcare for any opioid treatment program, or an ac-creditation requirement, or to interfere with thejudgment of individual clinicians who are re-sponsible for delivering patient care.

METHODS

Panel members reviewed the available infor-mation at an initial meeting in December 2007,after which a draft document was prepared bythe Panel’s writing group. A second meeting ofthe Panel was convened in July 2008 and a thirdin June 2009.

Literature Search

As part of the Panel’s review, a comprehen-sive literature search was performed via MED-LINE and EMBASE (covering articles publishedbetween 1966 and July 2008) for publicationsaddressing the cardiac effects of methadone.English-language manuscripts were reviewed,as were official opioid treatment guidelines pub-lished in Canada and the United Kingdom, back-ground articles on QT prolongation and TdP pro-vided by Panel members, and the findings ofrelevant meetings.

Members of the Panel also reviewed in-formation from regulatory authorities, consid-ered the challenges involved in applying cardiacsafety recommendations within opioid treatmentprograms, examined data regarding physicianawareness of the cardiac effects of methadone,and evaluated contextual materials regarding thecardiac effects of the methadone derivative levo-alpha-acetyl-methadyl (LAAM), which is nolonger marketed.13 The objectives of this processwere to broadly synthesize available evidenceregarding the cardiac effects of methadone andthen to distill that information into a practicalframework for improving cardiac safety in opi-oid treatment programs.

Field Review

In 2009, SAMHSA distributed more than 470copies of the draft document to stakeholder or-ganizations and to individual experts for field

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

286 JOURNAL OF ADDICTIVE DISEASES

review. Copies were also distributed to partic-ipants in workshops at the 2009 AATOD andASAM annual meetings. The field review draftwas marked “Not for Reproduction or Distri-bution.” Responses were requested by May 31,2009. A total of 51 responses were received (an11% response rate). All of the comments andsuggested changes were compiled and circulatedto the Panel for consideration at a meeting in July2009.

As part of the review process, the draft docu-ment was submitted to a law firm specializing inprofessional liability matters, with a request fora determination as to whether the language of thedocument would tend to increase the exposure ofopioid treatment programs and individual clini-cians to legal action and, if so, for suggested lan-guage modifications that would minimize suchexposure. The law firm complied with these re-quests, and their suggested language has beenincorporated herein.14

Finally, the draft and the supporting literaturewere sent to a health care statistician, who wasasked to determine whether the articles on whichthe Panel based its findings and conclusions metstandard tests for statistical significance. The re-sults of that examination also are reflected in thisdraft.15

This version of the report incorporates com-ments received through the field review, as wellas from literature published subsequent to prepa-ration of the earlier drafts, and revisions agreedon at the Panel’s June 2009 meeting. It super-sedes any previous drafts.

BACKGROUND

Methadone Pharmacology

Methadone is a potent full mu and delta opi-oid agonist with good oral bioavailability and along duration of action. Its actions may mimicthose of the endogenous opioids, enkephalins,and endorphins and affect the release of otherneurotransmitters, such as acetylcholine, nore-pinephrine, substance P, and dopamine.16 Thisaccounts for its analgesic and antitussive prop-erties, respiratory depression, sedation, decreasein bowel motility, increase in biliary tone, hor-

mone regulation, and increase of prolactin andgrowth hormone release, miotic pupils, nausea,and hypotension.17

Methadone is also a noncompetitive antago-nist at the N-methyl-D-aspartate (NMDA) recep-tor. There is a great deal of interest in this recep-tor because it may be linked to pain processingand spinal neural plasticity, although its exactrole and how it can be manipulated awaits furtherclarification.18 On the other hand, methadone’shigh inter-individual variability and potential ad-verse effects—particularly respiratory depres-sion and death—make a fundamental knowledgeof its pharmacokinetics and pharmacodynamicsimperative to the safe use of this important med-ication.19

Pharmacokinetic studies have shown that oralmethadone has a delayed onset of action. Fol-lowing oral administration, time to achievepeak plasma drug concentration is 2.5 hoursfor methadone in solution and 3 hours formethadone in tablet form. Oral bioavailabilityis approximately 85% (range: 67% to 95%), orthree times that of morphine. Although there areno data on tissue distribution for methadone inhumans, the distribution to various tissues hasbeen shown to be extensive in animal models.This is consistent with the high volume of dis-tribution in humans (4.2 to 9.2 L/kg in opioidsaddicts and 1.7 to 5.3 L/kg in chronic pain pa-tients.20–25

Methadone is biotransformed in the liver bythe cytochrome P450-related enzymes (primar-ily by the 3A4 and—to a lesser extent—the 2D6and 1A2 systems) to two N-demethylated bi-ologically inactive metabolites, which undergoadditional oxidative metabolism.26 At physio-logical pH, 86% of methadone is bound toplasma proteins, predominately to α1-acid gly-coprotein (AAG). AAG is an acute-phase re-active protein and the plasma level fluctuateswith various physiologic and pathologic con-ditions, such as stress, opioid addiction, can-cer, and concomitant administration of certainmedications.

The clinical implication of increased levelsof AAG is that such an individual may beprotected from the toxic effects of a dose ofmethadone, as compared with a healthy, casualuser of methadone who would not have elevated

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

SAMHSA Report 287

levels of AAG. Unlike morphine, methadoneis biotransformed rather than conjugated in theliver. At daily doses less than 55 mg, the ma-jority of the metabolites are cleared via thefecal route. Methadone is metabolized by thetype I cytochrome P450 group of enzymes. Themain enzyme responsible for N-demethylationof methadone is CYP3A4, with lesser involve-ment from CYP1A2, CYP2B6, and CYP2D6.18

Current evidence suggests that CYP2B6 mayplay a significant role in metabolism as well. Themain product of N-demethylation, 2-ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine (EDDP),is inactive. The activities of these cytochromeenzymes, especially CYP3A4, can be in-duced or inhibited by other drugs or by themethadone itself, accounting for the large in-dividual variations in methadone pharmacology(Appendix D). Although the primary metabo-lite of methadone is inactive, methadol andnormethadol are two minor metabolites pro-duced in small amounts that have similar phar-macologic activity to methadone.25 Renal excre-tion is variable and is pH dependent. At a urinepH above 6, renal clearance is only 4% of thetotal drug elimination. When urine pH drops be-low 6, the unchanged methadone excreted by therenal route is approximately 30% of the total ad-ministered dose. Despite this, methadone doesnot accumulate in patients with renal failure andis poorly removed by hemodialysis. The renalexcretion of the primary metabolite, EDDP, isnot pH-dependent.

Methadone undergoes a biphasic pattern ofelimination: slow distribution or α-eliminationphase (in 8 to 12 hours) and a ß-eliminationphase (in 30 to 60 hours). The α-eliminationcorrelates with the duration of analgesia that istypically 6 to 8 hours. The plasma level in theß-elimination phase is sub-analgesic but is suf-ficient to prevent withdrawal symptoms.24

QT Interval and Torsades de Pointes

The QT interval is a measure of the time be-tween the start of the Q wave and the end of theT wave in the heart’s electrical cycle. The QTinterval is dependent on the heart rate in an ob-vious way (the faster the heart rate, the shorterthe QT interval) and must be adjusted to aid in-

terpretation. Normal values for the QT intervalare between 0.30 and 0.44 (0.45 for women) sec-onds. Prolonged QT interval has been definedas >450 ms for men and >460 to 470 ms forwomen.27,28

QT interval can be measured by differentmethods, such as the threshold method, in whichthe end of the T wave is determined by thepoint at which the component of the T wavemerges with the isoelectric baseline or the tan-gent method in which the end of the T wave isdetermined by the intersection of a line extrap-olated from the isoelectric baseline and the tan-gent line which touches the terminal part of theT wave at the point of maximum downslope.29

Generally, the QT interval is corrected forits natural dependence on the heart rate usingBazett’s formula:30 QTc = QT interval (in mil-liseconds) divided by the square root of the pre-ceding RR interval (in seconds). Although thisformula is likely to overcorrect in the setting ofhigh heart rates,31 it is nevertheless a reasonablemethod for screening purposes, with the provisothat patients remain supine for approximately 5minutes prior to electrocardiogram (ECG) ac-quisition.

Prolongation of the QT interval occurs whenthere is an interruption in the normal balance andflow of ions in the myocardium, increasing thetime it takes for repolarization to occur.32 A QTinterval of 390 to 420 ms in men or 400 to 440ms in women is considered normal (Figure 1).33

Causation of the prolonged QT interval in-volves the human ether-a-go-go-related gene(hERG) and the subunit of the voltage-gatedpotassium channels (found predominantly inthe myocardium) for which it encodes.34 Thesechannels are the predominant facilitators of thedelayed-rectifier potassium ion currents, whichcause repolarization. Abnormalities in thesechannels have been shown to lead to prolongedaction potentials that are expressed as long QTintervals on the electrocardiogram.35

The health risks associated with a prolongedQT interval are not clear. Although individualswith a prolonged QT interval often are asymp-tomatic, some may develop palpitations, syn-cope, seizures, or cardiac arrest.36 QT intervalprolongation can also lead to the potentially fa-tal reentrant arrhythmia known as the TdP.33

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

288 JOURNAL OF ADDICTIVE DISEASES

FIGURE 1. Prolonged QT interval on electrocar-diogram.

Torsades de Pointes is a French term that lit-erally means “twisting of the points.” TdP isa potentially lethal form of ventricular arrhyth-mia named for its hallmark feature: polymorphicQRS complexes that appear to twist around theisoelectric line.

However, the mere presence of QT intervalprolongation may not lead to TdP.37 Some med-ications, such as amiodarone, are commonly as-sociated with a prolonged QT interval but rarelywith TdP.38 Conversely, TdP can occur in indi-viduals whose QT interval is within the normalrange.34

Torsades usually occurs in bursts that arenot sustained; thus, the rhythm strip usuallyshows the patient’s baseline QT prolongation.Although frequently self-terminating, TdP maydegenerate into ventricular fibrillation and canlead to sudden death.

QT prolongation is the mandatory substratefor development of TdP and is the most com-monly scrutinized pharmacologic adverse ef-fect evaluated during the safety phase of drugdevelopment and post-marketing surveillance(Figure 2).31,39

FIGURE 2. Torsades de Pointes.

Drug-induced QT prolongation or TdP of-ten result from a confluence of factors. Roden40

compiled a list of factors that may increase thelikelihood of a patient developing QT intervalprolongation and subsequent progression to TdP,such as female gender, hypokalemia, a historyof drug interactions, underlying cardiac con-ditions, unrecognized congenital long QT syn-drome, and predisposing DNA polymorphisms.Although these risk factors have not been stud-ied in patients receiving methadone, they shouldbe taken into consideration when treating thisspecific population.

Preexisting QT prolongation appears to bea particularly important risk factor for drug-induced arrhythmia. In a meta-analysis of 1,288participants from 181 clinical trials of the antiar-rhythmic drug sotalol, in which the incidence ofTdP was 2%, an increased QT at baseline (meanQT = 455 ± 11 ms in those who experiencedTdP versus 428 ± 1 ms in those who did not,p < .003) was the most consistent predictor ofTdP.41

The DIAMOND-HF study—a multivariateanalysis of those receiving the QT-prolongingdrug dofetilide—produced similar findings,42

with a baseline QT greater than the upper quar-tile value of 479 ms associated with a relativerisk for death of 1.9 (95% CI = 1.0–3.6, p =.05).

Although many drugs prolong the QT inter-val, drug-induced TdP is associated with a muchsmaller number of medications and is accom-panied often by an additional predisposing riskfactor,31 such as bradycardia.43,44 Women have aslightly longer QT interval than men and are atgreater risk for arrhythmia due to QT-prolongingcompounds.45 Threshold values of 430 ms formen and 450 ms for women have been pro-posed as the upper limit of normal.46 Despitethese variable definitions, the international reg-ulatory guidance for drug development suggests

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

SAMHSA Report 289

a gender-independent categorical threshold forQT prolongation of 450 ms.47

The chiral nature of the methadone moleculealso may be important in understanding QT in-terval prolongation associated with methadone.In addition to CYP3A4, methadone is metab-olized by CYP2B6 and CYP2D6.48–50 Stereos-electivity of CYP2B6 toward S-methadone hasbeen demonstrated in vitro51 and in vivo.52,53 Itmay be that slow metabolizers of CYP 2B6 whopossess a ∗6/∗6 genotype have difficulty me-tabolizing S-methadone but not R-methadone.Eap et al.53 demonstrated that S-methadoneblocks the hERG channel more potently thanR-methadone does and that slow metaboliz-ers of CYP 2B6 with a ∗6/∗6 genotype weremore likely to have a prolonged QT inter-val than did patients without the ∗6/∗6 geno-type (extensive metabolizers). This is clinicallysignificant because approximately 6% of Cau-casians and African Americans possess the ∗6/∗6genotype.52,53

Although there is disagreement over the exactdegree of risk posed by drug-related QT prolon-gation, it is generally accepted that significantrisk of TdP occurs at measurements greater than500 ms.54 The paucity of long-term studies ofQT-prolonging drugs in large populations makesit difficult to assign a relative risk to a QT > 500ms, but in patients with the long QT syndrome, aQT >500 ms was associated with an odds ratiofor syncope or sudden death of 4.22 (95% CI =1.12–15.80, p = .033), presumably due to TdP.55

Rationale for Screening

The reason for the variability in QT intervalsamong patients treated with methadone is notcompletely clear, which complicates the processof screening for risk. Ehret et al.56 posited thatsome other factors might have contributing roles,as they found that long QT intervals seemedto be associated with the use of CYP3A4 in-hibitors, low potassium concentrations, and ab-normal hepatic function. They concluded thatthe occurrence of ECG changes is more likely inpatients receiving methadone, and other phys-iological factors are likely to contribute to themanifestation of abnormal cardiac function.

Other limitations of QT interval screening in-clude selection of a methodology for rate correc-tion at extremes of heart rate; choosing betweenfully manual, semi-automatic or automated QTmeasurements; and the limited predictive valueof QT prolongation for arrhythmia risk at anindividual level. Despite these limitations, QTinterval screening is the current standard for as-sessing cardiac drug safety in the major domainsof medicine: clinical trials, cardiology practice,drug development, and regulatory assessmentfor drug withdrawal or manufacturers’ labelingchanges.57 This type of screening does not nec-essarily require a specialist and has been judgedappropriate for primary care settings.28

ADVICE FROM RELEVANTAGENCIES AND ORGANIZATIONS

There are no current, widely accepted rec-ommendations from U.S. authorities for the pre-vention and treatment of methadone-induced QTinterval prolongation. As noted earlier, the FDAissued a physician safety alert regarding fatal-ities and cardiac arrhythmias associated withmethadone,3 which was followed by a “blackbox warning” in the manufacturer’s product la-beling.4 Although the revised product label sug-gests careful monitoring of patients with pro-longed QT intervals, it does not specify whatform of monitoring is appropriate.

Thompson Reuters’ MICROMEDEX issomewhat more concrete in suggesting ECGmonitoring of patients with known cardiac con-duction abnormalities or those at increased riskfor such abnormalities.5 Relevant advice fromthe American Association for the Treatmentof Opioid Dependence and from Canadian andBritish health authorities is summarized in Ap-pendix B.

RESULTS OF THE LITERATUREREVIEW

The literature linking methadone to QT prol-ongation and TdP was organized into the follow-ing categories: experimental (in vitro) studies,clinical case series, forensic/toxicologic studies,

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

290 JOURNAL OF ADDICTIVE DISEASES

cross-sectional investigations, and prospectivecohort studies or randomized trials. In addition,the published studies were scrutinized for therelationship between methadone dose or serumconcentration and cardiac repolarization. An ab-breviated discussion of the prinicipal findingsfollows (also refer to Appendix D).

Experimental Studies

As noted earlier, the most common cause ofdrug-induced QT prolongation and TdP is block-ade of the hERG, which encodes the rapid com-ponent of the delayed-rectifier potassium ioncurrent (Ikr).34 Blockade of this cardiac ion chan-nel prolongs the terminal portion of the cardiacaction, potentially causing delayed repolariza-tion, which manifests as QT interval prolonga-tion on the surface ECG.

Methadone has been shown to be a potent in-hibitor of the hERG channel, capable of achiev-ing 50% in vitro inhibition (IC50%) of Ikr at con-centrations between 1 and 20 µM, dependingon experimental conditions—a level that can beattained in clinical practice (Katchman et al.,2002).58 The ratio of the IC50% to maximal serumconcentration (Cmax) is a better predictor of ar-rhythmia risk and is identical for methadone andLAAM, but an order of magnitude higher thanfor buprenorphine, another opioid approved foruse in addiction treatment.58

Major interindividual variability in serum lev-els for any given dose due to variable hep-atic clearance makes a priori prediction of QTeffect problematic.48 Specifically, methadoneis metabolized by the cytochrome P450 sys-tem, and inhibitors of this enzyme can sig-nificantly increase plasma area-under-curvemeasurements. In addition, unsuspected poly-morphisms in the gene for the hERG channeloccur in 2% of the healthy population and maybe associated with increased sensitivity to hERGchannel blockade by methadone or similarcompounds.59

Beyond its effect on cardiac repolarizationvia blockade of hERG channels, methadonepossesses additional properties that may pre-dispose patients to the development of TdP.For example, risk of TdP increases in thesetting of bradycardia—an effect that hasbeen confirmed clinically.60 Methadone ap-

pears to exhibit negative chronotropic ef-fects through two key mechanisms: calciumchannel antagonism61,62 and anti-cholinesteraseproperties.63–65

Clinical Case Series

As early as 1973, a group of patients addictedto heroin were evaluated for predisposing riskfactors for sudden cardiac death.66,67 The au-thors observed QT prolongation in several pa-tients taking methadone who were also usingillicit drugs. At the time, there was no clini-cal evidence that methadone possessed cardiacproperties.

Only three decades later was the associationbetween very high doses of methadone and TdPdocumented in a North American case series of17 patients who experienced TdP.66 The pati-ents’ mean (± SD) daily dose of methadonewas 397 ± 283 mg, and their mean (± SD)QT interval was 615 ± 77 ms. None of thepatients died, but an implantable pacemaker ordefibrillator was placed in 14 patients. The re-searchers concluded that doses of methadonegreater than 60 mg a day can lead to TdP. How-ever, because of the small sample size, lack ofa control group, and absence of testing for con-genital long QT syndrome, they cautioned thattheir research did not establish a causal relation-ship between methadone use and TdP. (In a laterpublication, Krantz et al.68 further analyzed thesame data and determined that methadone usewas the only variable predictive of QT inter-val increases and that the increases were dose-dependent.) Since that time, a growing body ofevidence has emerged to suggest a clinical as-sociation between methadone, QT prolongation,and TdP.69–87

For example, Justo et al.88 performed a meta-analysis of 14 case reports (totaling 40 pa-tients dependent on opioids) of methadone-associated TdP to identify factors associatedwith methadone-related QT interval prolonga-tion and TdP. Only patients being treated inmethadone programs for opioid addiction wereincluded. The mean (± SD) QT interval dur-ing or immediately after TdP was 598 ± 75 ms.The mean (± SD) daily methadone dose was231 ± 201 mg (range: 60 to 1000 mg daily).The most common risk factor identified was a

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

SAMHSA Report 291

methadone dose greater than 60 mg a day (n =39, 98%). Other common risk factors includedthe use of other medications known to increaseserum methadone levels or trigger TdP (n =22, 55%), HIV infection (n = 16, 40%), hy-pokalemia (n = 14, 35%), female sex (n = 13,33%), cirrhosis or renal failure (n = 11, 28%),and heart disease (n = 9, 33%). The mean num-ber of risk factors was 3.5 per patient, with allpatients having at least one risk factor and 85%having two or more.

Through a search of the FDA’s MedWatchsystem, Pearson and Woosley89 identified 5,503methadone-related adverse events reported tothe FDA between 1969 and 2002. Of these, 59were related to QT interval prolongation or TdP.Of the 59 cases, 28 resulted in hospitalizationand 5 resulted in death. Interestingly, 56 of thecases were reported in the past 2 years of the re-view. The authors proposed several explanationsfor this, including the fact that the arrhythmiascan be confirmed only through ECG monitor-ing, which was not routinely conducted in theearlier years of methadone treatment. They alsofound that the abnormal cardiac rhythm eventsreported occurred over a large range of doses.(It is important to note that only a fraction ofdrug-related serious adverse events are voluntar-ily reported to MedWatch, so the true number ofarrhythmia episodes attributable to methadoneis not known.)

Patel et al.90 described 8 methadone-maintenance patients who presented withaborted sudden death or TdP and requiredplacement of implantable cardioverter defibril-lators. The patients were receiving high doses ofmethadone (mean: 204 ± 173 mg/day) and werefollowed longitudinally for a mean of 27 months.Six of the 8 patients continued methadone ther-apy: 1 died and 3 others received electroshockfor recurrent TdP. This case series suggests thatmethadone-treated patients who experience TdPare at substantial risk for recurrent arrhythmiaif methadone is continued, particularly at higherdoses.

Toxicologic Studies

Chugh et al.91 conducted a study of patientswho had sudden cardiac death and were assessedby a medical examiner. Case participants in-

cluded those with therapeutic blood methadoneconcentrations (<1 mg/L) on postmortem tox-icological evaluation; controls included partic-ipants who did not take methadone. Cases inwhich higher methadone levels were presentwere presumed to be overdoses and were ex-cluded, as were cases in which recreational druguse was determined to be the cause of death.A total of 22 patients with sudden cardiac deathand therapeutic methadone levels were identifiedand compared with 106 patients in the controlgroup. Based on autopsy findings, it was deter-mined that a structural cardiac abnormality thatcould have caused sudden cardiac death (e.g.,coronary artery disease, severe left ventricularhypertrophy, hypertrophic cardiomyopathy) waspresent in only 23% of the methadone patients,compared with 60% of patients in the controlgroup (p = .002). Rates of use of other medica-tions (such as benzodiazepines, antidepressants,anticonvulsants, antihistamines, and muscle re-laxants) were similar in both groups. Althoughit was not possible to determine how many ofthe study participants had died from respiratorydepression unrelated to a cardiac event, the au-thors concluded that the results of the study sup-ported the presence of cardiac risk associatedwith methadone given at therapeutic levels.

It should be emphasized that this study is onlyinferential because central nervous system de-pression with resulting anoxia also can cause ar-rhythmias. However, the findings are consistentwith methadone’s potent proarrhythmic effectsin vitro and the statistical reality that a small butsignificant proportion of arrhythmic events willbe fatal.92,93

Cross-Sectional Studies

Cross-sectional studies were available forfour ambulatory and one inpatient cohort.Maremmani et al.94 conducted a restrospec-tive analysis of 83 patients being treated withmethadone maintenance for their heroin addic-tion for at least 6 months. At the time of ECGrecording, all patients had urine toxicology nega-tive for opioids, cocaine, and amphetamines andnone were taking other medications associatedwith QT interval prolongation. Eighty-three per-cent of the patients had QT interval prolongationgreater than the reference value for individuals

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

292 JOURNAL OF ADDICTIVE DISEASES

of the same sex and age, although only two pa-tients had QT intervals exceeding 500 ms. Nobaseline ECGs or plasma methadone levels wereobtained. The mean daily dose of methadone was87 mg (range: 10 to 600 mg daily), but no cor-relation was found between the QT interval andthe methadone dose.

Ehret et al.56 examined ECG changes in a ret-rospective study of 527 former heroin users. Ofthese, 280 were excluded for not having ade-quate records available, for abusing methadone,or for having severe heart disease. The remaining247 individuals were separated into two groups:methadone maintenance patients (n = 167)and patients not receiving any pharmacotherapy(n = 80). Twenty-seven patients in themethadone maintenance group had a QT intervalof 500 ms or longer compared with no patientsin the control group (p < .001). The relationshipbetween dosage and QT interval was weakly as-sociated, but a higher daily methadone dose wasassociated with significantly greater QT intervalprolongation (p < .01). In this study, the lowestmethadone dosage found to increase the QT in-terval above 500 mswas 30 mg per day. In themethadone maintenance group, 6 patients expe-rienced TdP. These patients were taking 40 to200 mg of methadone per day and had a QTinterval of 430 to 750 ms. Of note, two of thepatients who experienced TdP had QT intervalswithin the normal range.

The largest cross-sectional comparative studyperformed to date analyzed 393 methadone-treated patients and 43 buprenorphine-maintained patients in Copenhagen.95 Theauthors observed QT prolongation (> 440 ms)in 32% of methadone-treated patients, but nonein buprenorphine-maintained patients. Of themethadone-treated patients, all of whom werereceiving doses > 100 mg/day, 8 had a QTgreater than 500 ms. Similar findings weredocumented in a cohort of chronic pain patients(n = 104) receiving methadone, 33% of whomhad QT prolongation, defined as > 430 in menand > 450 in women.96

Prospective and Randomized Trials

In the largest prospective cohort study todate, Martell et al.97 evaluated 167 new en-

trants into methadone maintenance treatment.ECGs were blinded to dose and time inter-val. Oral methadone induction resulted in asignificant increase in the mean QT interval,averaging 12.4 ± 23 ms at 6 months, whichpersisted (10.7 ± 30 ms) at 12 months. A sim-ilar increase in QT interval dispersion (9.5 ±18.6 ms, p < .0001), which is a marker for het-erogeneous cardiac repolarization, was also ob-served from baseline to 6 months in the studycohort.98

Wedam et al.99 conducted a randomized con-trolled trial of 220 patients with opioid depen-dence to evaluate the changes in QT intervalsof patients previously enrolled in a random-ized, blinded study of methadone, levomethadyl,and buprenorphine. Patients’ ECG data wereanalyzed, and QT threshold values of 470 and490 ms were selected for men and women, re-spectively. Controlling for the use of other med-ications, hypokalemia, and renal insufficiency,the investigators found that 28% of patientsin the levomethadyl group, 23% of patients inthe methadone group, and no patients in thebuprenorphine group exceeded this thresholdduring the 16-week study (p < .001). Twenty-one percent of the levomethadyl-treated patientsand 12% of the methadone-treated patients hada QT interval increase from baseline of > 60 msat some point in the study. This difference wassignificant when compared with buprenorphine(p < .001). There was no significant differenceobserved in the QT interval of patients receiv-ing methadone versus levomethadyl when us-ing Bazett’s formula. However, the researchersfound a progressive prolongation in the QT inter-val of patients receiving methadone even whentheir dose remained stable (p = 0.01). This trendwas not significant among patients treated withlevomethadyl, nor was there a significant differ-ence in interval prolongation between men andwomen.

Evidence of Dose-Dependent Effects

Several studies have attempted to quan-tify the magnitude of the relationship betweenmethadone dose and delayed repolarization. Forexample, a 2007 article by Atkinson et al.71 re-ported QT normalization when the methadone

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

SAMHSA Report 293

was discontinued or the dose reduced. In a se-ries of patients with TdP, oral methadone dosewas modestly correlated with the absolute QT in-terval recorded at the time the patient presentedwith arrhythmia.68 With intravenous methadone,a significant linear relationship between QT andmethadone dose was noted.100 The correlationbecame more robust in the subset of patientswho also were using cocaine (r = +0.4, p =.03). This is consistent with a synergistic effectof combined methadone and cocaine on hERGchannel blockade.101

In the Copenhagen study, Fanoe et al.95 re-ported the effects of methadone and buprenor-phine on QT interval in 450 patients addictedto heroin. The authors obtained ECGs approxi-mately 24 hours after the last dose of methadoneand found a significant association between QTinterval and methadone dose in both sexes. Theassociation existed regardless of the correctionformula used (p < .001). There was no as-sociation between QT interval and buprenor-phine dose. Overall, 32% of patients treatedwith methadone had a QT interval greater than440 ms. There was no significant associationbetween QT interval and age or length oftime in treatment. The investigators also askedthe patients about histories of syncope overthe prior year and found that patients receiv-ing higher doses of methadone reported moreepisodes of syncope. Syncope also was found tobe related to longer QT intervals. Limitationsof the study included lack of baseline ECGsand the absence of information regarding othermedications being used by the study partici-pants.

Peles et al.102 published the results of a studyof 138 patients (71% male) on methadone main-tenance for heroin dependence for a mean of4.4 years (range: 0.3 to 10.7 years). ECGswere performed and serum methadone levelswere measured for all patients approximately24 hours after their last methadone dose. Themean ± SD methadone dose was 170.9 ±50.3 mg daily (range: 40 to 290 mg daily), with80.4% of patients receiving greater than 120 mgdaily. The mean ± serum methadone concen-tration was 708.2 ± 363.1 Ngami (range: 110to 2350 Ngami). The mean ± SD QT intervalwas 418.3 ± 32.8 ms (range: 330 to 520 ms).

During the month before the study, 29.7% ofthe patients had urine toxicology results posi-tive for opiates and 22.5% had toxicology pos-itive for cocaine. Neither the methadone dosenor the serum methadone concentration corre-lated with QT interval. Of the three patients whohad QT intervals greater than 500 ms, two haddied by the time of the 2-year follow-up, al-though neither death was attributed to cardiaccauses. None of the 19 patients with QT inter-vals of 450 to 500 ms had any cardiac prob-lems.

With regard to serum levels, Martell et al.97

prospectively demonstrated that the QT changefrom baseline to 12 months after initia-tion of methadone was significantly cor-related with both trough and peak serummethadone concentrations. Similar relationshipshave been observed with the methadone deriva-tive, LAAM.103 Taken in the aggregate, the avail-able literature supports a dose-dependent effectof methadone and its derivative, LAAM, on car-diac repolarization. This creates a safety/efficacyparadox, given that higher doses of methadonemay reduce illicit drug use, yet place patients atgreater risk for TdP.104

Summary

Based on evidence published in the peer-reviewed literature,67–85 the Panel concluded thatthe relationship between methadone and QTprolongation is causal. Moreover, available ev-idence suggests that both oral and intravenousmethadone hydrochloride have an independentassociation with QT prolongation.

Prolongation of the QT to >500 ms is thoughtto confer significant risk with respect to ar-rhythmias. In all but one study of patients en-rolled in methadone maintenance treatment, aQT >500 ms was seen in 2% of those en-rolled. Wedam et al.99 found a higher incidence,which they estimated at 10%. If one accepts themore conservative estimate of 2%, one wouldpredict that an estimated 5,000 of the 250,000participants currently enrolled in opioid treat-ment programs are in need of interventions forcardiac risk reduction. An additional 40,000 to60,000 opioid treatment program patients likelyhave a QT between 450 and 500 ms and may

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

294 JOURNAL OF ADDICTIVE DISEASES

have some lesser (but nevertheless elevated)risk. Undercurrent illnesses or use of additionalQT prolonging agents may substantially pro-long the QT in these participants and exposethem to arrhythmia, so increased vigilance iswarranted.

CONTEXT AND CONCLUSIONS

Through the steps described earlier, the Ex-pert Panel arrived at a series of conclusions,which are summarized below. The following fac-tors were considered by the Panel members inreaching their conclusions.

• Use of methadone for the treatment ofopioid addiction in the United States isconfined to Federally certified opioid treat-ment programs, most of which are free-standing outpatient clinics. The ExpertPanel discussed specifics of opioid treat-ment program care that could present bar-riers to cardiac risk assessment or leadto unintentional consequences when suchguidelines are applied. Panel membersagreed that any steps recommended mustpreserve patients’ access to care.

• Most opioid treatment programs are notlicensed, staffed, or funded to provide car-diac screening services. Therefore, mostopioid treatment programs would have tochange their intake and outreach proce-dures to include screening for cardiac risk,which may result in the need to arrange al-ternative treatments or delay initiation ofcare. The Expert Panel supports promptaccess to treatment for the disease of ad-diction and recommends that every effortbe made to treat patients appropriately andsafely.

• Compared with the range of medicationsavailable to treat chronic pain, opioid treat-ment program physicians’ choice of ther-apeutic agent is limited by Federal regu-lations to oral methadone, naltrexone, orsublingual formulations of buprenorphineand buprenorphine/naloxone.

• An optimal strategy for identifying andreducing QT-associated risk has not yet

been established. Screening for risk isthe current standard of care. In patientswho have dose-related QT prolongation,it remains unknown how much additionalrisk of relapse is associated with reduc-ing the methadone dose. However, highermethadone maintenance doses are associ-ated with better treatment retention andoutcomes.

• Protocols that call for patient ECGs and re-view of cardiac risk may require consulta-tion with a cardiologist or other physician,yet such services may not be reimbursableunder existing contracts.

• There is compelling evidence that the ma-jority of physicians who direct treatmentin opioid treatment programs are not fullyaware of methadone’s association with ad-verse cardiac events. In a survey of medicaldirectors of all accredited opioid treatmentprograms in the United States, only 41%(95% CI = 37–45) of 692 physicians whoresponded were aware of methadone’s QT-prolonging properties. Only 24% (95% CI= 21–27) recognized the potential risk forTdP.10

The Panel’s conclusions are intended to assistclinicians and program administrators in devel-oping cardiac safety standards for methadone in-duction and maintenance treatment in Federallycertified opioid treatment programs. However,the conclusions provide only general guidanceand are not intended to supplant clinical judg-ment in treating individual patients, nor do theyrepresent Federal requirements or accreditationstandards.

Conclusions Regarding ClinicalProcedures

Expert Panel members agreed that, to the ex-tent possible, every opioid treatment programshould have a cardiac risk management plan thatincorporates the following elements.

Clinical Assessment

The assessment conducted at intake shouldinclude a complete medication history; per-sonal and family history of structural heart dis-

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

SAMHSA Report 295

ease (including long QT syndrome, sudden car-diac death, myocardial infarction, heart failure);any personal history of arrhythmia or syncope;and use of QT-prolonging drugs, including pre-scribed medications and illicit drugs such as co-caine.

ECG Assessment

The Expert Panel considered whether rou-tine ECGs to measure the QT interval shouldbe performed on every patient within 30 daysof admission. After extensive discussion and re-view of the evidence, the Panel members andex officio members could not reach agreementon the merits of such a statement, largely be-cause of concerns over the level of resourcesinvolved in implementing routine ECGs and theabsence of clear evidence for the effectivenessof the practice in achieving meaningful reduc-tions in methadone-associated cardiac events.Of those who expressed an opinion, nine werein favor of routine screening within the first 30days, whereas four felt such routine screeningis not necessary. (Voting in favor: Drs. Kotz,Krantz, Martin, Mehta, and Stimmel [Panelmembers]; Mr. Bowman and Drs. Haigney, Khanand O’Keeffe [Ex Officio members]. Votingagainst: Drs. Kreek, McCarroll, Payte and Tay-lor [Panel members].) Therefore, routine screen-ing of every patient within 30 days of admissionto treatment is not part of the consensus-drivenconclusions presented here. The Panel did agreethat a baseline ECG at the time of admission andwithin 30 days should be performed on patientswith significant risk factors for QT prolonga-tion, including a history of cardiac arrhythmiaor prolonged QT interval; symptoms sugges-tive of arrhythmia, such as episodes of syncope,dizzy spells, palpitations, or seizures; medica-tion history; family history of premature deathor any other historical information suggestive ofa possible cardiac arrhythmia. Additional ECGsshould be performed annually or whenever themethadone dose exceeds 120 mg/day.

In addition to scheduled ECGs, any patientwho experiences unexplained syncope or gener-alized seizures should have an ECG. If markedQT prolongation is documented, TdP should be

suspected and the patient hospitalized for moni-toring (through telemetry).

For most opioid treatment program patients,it is likely that automated (computer-generated)measurements of QT interval provide reasonableestimates of arrhythmia risk. However, if thereis uncertainty about the presence of significantQT prolongation in a particular patient, it is pru-dent to repeat the ECG or to have the tracinginterpreted by a cardiologist.

Risk Stratification

If the QT interval is more than 450 ms butless than 500 ms, methadone may be initiatedor continued, accompanied by a risk-benefitdiscussion with the patient and more frequentmonitoring. For methadone-maintained patientswith marked QT prolongation (≥500 ms), strongconsideration should be given to adoption of arisk minimization strategy (such as reducing themethadone dose, eliminating other contributingfactors, transitioning the patient to an alternativetreatment such as buprenorphine, or discontinu-ing methadone treatment).

Routine echocardiography to assess for struc-tural heart disease is not indicated in opioid treat-ment programs, nor is genetic testing for congen-ital long QT syndrome. That having been said,unexplained symptoms of syncope or seizuresthat emerge during therapy require urgent eval-uation.

Drug Interactions

Physicians should be aware that interactionsbetween methadone and other medications alsohave QT-prolonging properties, as does concur-rent use of drugs that slow the elimination ofmethadone (Appendix C).

Conclusions Regarding AdministrativeProcedures

Cardiac risk that is related to methadoneshould be incorporated into the informed con-sent document presented to patients at intake.

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

296 JOURNAL OF ADDICTIVE DISEASES

Conclusions Regarding Patient Education

Patients should receive educational materialsthat explain, in lay language, cardiac risk and itsrelationship to QT interval. Part of this enhancededucation involves the use of a consent form thataddresses cardiac concerns.

Conclusions Regarding Staff Education

Opioid treatment program medical directorsand clinical staff should be educated about therisks posed by prolonged QT interval and trainedin assessing patients for risk of TdP and othercardiac problems.

Conclusions Regardingthe Use of Methadone

The Panel affirms that methadone can be usedwith reasonable assurance that it is effective andthat its benefits exceed its risks, providing thatthe potential for QT prolongation is recognized,that patients receive ECG screening at indicatedintervals, and that appropriate clinical action istaken in the presence of significant QT prolon-gation.

Implementation Issues

The Panel acknowledges that acting on theconclusions presented here will pose challengesto many opioid treatment programs. Identifyingclinically relevant QT prolongation remains dif-ficult, given the variability of ECG machine mea-surements and difficulty in defining the preciserisk a prolonged QT portends for any given in-dividual. The Panel recognizes that opioid treat-ment programs will be challenged to integratecardiac arrhythmia risk assessment into the careof opioid-addicted patients without reducing ac-cess to vital addiction treatment services. ThePanel is also aware that not all methadone main-tenance treatment providers will be capable ofadministering an ECG to every patient in all thecircumstances suggested above.

Opioid treatment programs and otherproviders are encouraged to consider these con-clusions to the extent that they are practically orfinancially capable of doing so. Nothing in thisreport is intended to create a legal standard of

care for any opioid treatment program or to in-terfere with clinical judgment in the practice ofmedicine.

REFERENCES

1. Center for Substance Abuse Treatment. Methadone-associated mortality: report of a national assessment, May2003. http://www.csdp.org/research/methadone.samhsa204.pdf. Accessed February 16, 2011.

2. Center for Substance Abuse Treatment. Methadonemortality: a reassessment. Report of the Meetingand Follow-up Activities, July 2007. Rockville, MD:CSAT, Substance Abuse and Mental Health ServicesAdministration, 2007; 32–6. http://www.dpt.samhsa.gov/pdf/Methadone Report 10%2018%/2007 Brief%20w%20attch.pdf. Accessed February 16, 2011.

3. Food and Drug Administration. Public health advi-sory: methadone use for pain control may result in deathand life-threatening changes in breathing and heart beat.http://www.fda.gov/Drugs/DrugSafety/PostmarketDrugSafety/SafetyInformationforPatientsandProviders/DrugSafetyInformationforHealthcareProfessionals/PublicHealthAdvisories/ucm/24346.htm. Accessed February 16, 2011.

4. Roxane Laboratories. Dolophine hydrochlorideCII (methadone hydrochloride tablets, USP). http://www.accessdata.fda.gov/drugsatfda docs/label/2006/006134s0281lbl.pdf. Accessed February 16, 2011.

5. Thompson Reuters MICROMEDEX R©. www.thomsonhc.com. Accessed February 22, 2011.

6. Center for Education and Research on Therapeu-tics (CERT). Drugs with risk of torsades de pointes.http://www.azcert.org/medical-pros/drug-lists/list-01.cfm.Accessed February 16, 2011.

7. Ancherson K, Clausen T, Gossop M, et al. Preva-lence and clinical relevance of corrected QT interval pro-longation during methadone and buprenorphine treatment:a mortality assessment study. Addiction 2009; 104:993–99.

8. Wallner C, Stollberger C, Lhavin A, et al. Electro-cardiographic abnormalities in opiate addicts. Addiction2008; 103:1994.

9. George S. Methadone-associated QT prolongationand torsades de pointes. Br J Hosp Med (London) 2007;68:24.

10. Krantz MJ, Rowan S, Schmittner J, et al. Physicianawareness of methadone’s cardiac effects: a national sur-vey. J Addict Dis 2007; 26:79–85.

11. Athanasos P, Farquharson AL, Compton P, PsaltisP, Hay J. Electrocardiogram characteristics of methadoneand buprenorphine maintained subjects. J Addict Dis 2008;27:31–5.

12. Krantz MJ, Martin J, Stimmel B, Mehta D, HaigneyMCP. QTc interval screening in methadone treatment. AnnIntern Med 2009; 150:387–95.

13. Food and Drug Administration (FDA). MedWatchsafety information and adverse event reporting program,

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

SAMHSA Report 297

safety information: ORLAAM (levomethadyl hydrochlo-ride acetate) withdrawn from market March 2001.http://www.fda.gov/Safety/MedWatch/SafetyInformation/SafetyAlertsforHumanMedicalProducts/UCM/53332.htm.Accessed February 16, 2011.

14. Hodgson Russ Associates. Personal communication,June 14, 2009.

15. Pullum T. Department of statistics, University ofTexas at Austin. Personal communication, June 15, 2009.

16. Inturrisi CE. Pharmacology of methadone and itsisomers. Anesthesiology 2005; 71:435–7.

17. Anderson IB, Kearney TE. Use of methadone. WestJ Med 2000; 172:43–6.

18. College of Physicians of Ontario (CPSO).Methadone maintenance guidelines. Toronto, Ontario,Canada: Author, 2004.

19. Fishman SM, Wilsey B, Mahajan G, et al.Methadone reincarnated: novel clinical applications withrelated concerns (review). Pain Med 2002; 3:339–48.

20. Anggaard E, Gunne L-M, Holmstrand J, et al. Dis-position of methadone in methadone maintenance. ClinPharmacol Ther 1974; 17:258–66.

21. Dole VP, Kreek MJ. Methadone plasma level: sus-tained by a reservoir of drug in tissue. Proc Natl Acad SciU S A 1973; 70:10.

22. Inturrisi CE, Verebely K. The levels of methadonein the plasma in methadone maintenance. Clin PharmacolTher 1972; 13:633–7.

23. Kling MA, Carson RE, Borg L, et al. Opioid re-ceptor imaging with positron emission tomography and[18F]cyclofoxyl in long-term, methadone-treated formerheroin addicts. J Pharmacol Exper Ther 2000; 295:1070–6.

24. Kreek MJ. Plasma and urine levels of methadone. NY State J Med 1973; 73:2773–7.

25. Peng PW, Tumber PS, Gourlay D. Review article:perioperative pain management of patients on methadonetherapy. Can J Anaesth 2005; 52:513–23.

26. Sullivan HR, Smits SE, Due SL, et al. Metabolism ofd-methadone: isolation and identification of analgesicallyactive metabolites. Life Sciences 1972; 11:1093–104.

27. Committee for Proprietary Medicinal Products(CPMP). Points to consider: the assessment of the potentialfor QT interval prolongation by non-cardiovascular medic-inal products. The European Agency for the Evaluation ofMedicinal Products. December 1997 [CPMP/986/96].

28. Al-Khatib SM, Allen LaPointe NM, Kramer JM,et al. What clinicians should know about the QT interval.JAMA 2003; 289:2120–27.

29. Panicker GK, Karnad DR, Joshi R, et al. Compari-son of QT measurement by tangent method and thresholdmethod. Ind Heart J 2006; 58:487–88.

30. Bazett HC. An analysis of the time-relations of elec-trocardiograms. Heart 1920; 7:353–70.

31. Food and Drug Administration. Guidance for indus-try: E14 clinical evaluation of QT/QTc interval prolon-

gation and proarrhythmic potential for non-antiarrhythmicdrugs. Rockville, MD: Author, 2005.

32. Nerbonne JM, Kass RS. Molecular physiol-ogy of cardiac repolarization. Physiol Rev 2005; 85:1205–53.

33. Merri M, Benhorin J, Alberti M, et al. Electrocar-diographic quantitation of ventricular repolarization. Cir-culation 1989; 80:1301–08.

34. Stringer J, Welsh C, Tommasello A. Methadone-associated QT interval prolongation and torsades depointes. Am J Health Syst Pharm 2009; 66:825–33.

35. Fenichel RR, Malik M, Antzelevitch C, et al. Drug-induced torsades de pointes and implications for drugdevelopment. J Cardiovasc Electrophysiol 2004; 15:475–95.

36. Sanguinetti MC, Tristani-Firouzi M. hERG potas-sium channels and cardiac arrhythmia. Nature 2006;440:463–69.

37. Montanez A, Ruskin JN, Hebert PR, et al. ProlongedQTc interval and risks of total and cardiovascular mortalityand sudden death in the general population: a review andqualitative overview of the prospective cohort studies. ArchIntern Med 2004; 164:943–48. [Erratum, Arch Intern Med.2004; 164:1796.]

38. Yoshida H, Sugiyama A, Satoh Y, et al. Comparisonof the in vivo electrophysiological and proarrhythmic ef-fects of amiodarone with those of a selective class III drug,sematilide, using a canine chronic atrioventricular blockmodel. Circ J 2002; 66:758–62.

39. Redfern WS, Carlsson L, Davis AS, et al. Relation-ships between preclinical cardiac electrophysiology, clini-cal QT interval prolongation and Torsades de Pointes for abroad range of drugs: Evidence for a provisional safetymargin in drug development. Cardiovas Res 2003; 58:32–45.

40. Roden DM. Long QT syndrome: reduced repolar-ization reserve and the genetic link. J Intern Med 2006;259:59–69.

41. Soyka LF, Wirtz C, Spangenberg RB. Clinical safetyprofile of sotalol in patients with arrhythmias. Am J Cardiol1990; 65(Suppl):74A–81A.

42. Brendorp B, Elming H, Jun L, et al.; for the DI-AMOND Study Group. QT interval as a guide to selectthose patients with congestive heart failure and reducedleft ventricular systolic function who will benefit from an-tiarrhythmic treatment with dofetilide. Circulation 2001;103:1422–27.

43. Kurita T, Ohe T, Shimizu W, et al. Early after depo-larization in a patient with complete atrioventricular blockand torsades de pointes. Pacing Clin Electrophysiol 1993;16:33–8.

44. Wesley RC Jr, Turnquest P. Torsades de Pointes afterintravenous adenosine in the presence of prolonged QT-syndrome. Am Heart J 1992; 123:794–96.

45. Ebert SN, Liu XK, Woosley RL. Female gender as arisk factor for drug-induced cardiac arrhythmias: evaluation

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

298 JOURNAL OF ADDICTIVE DISEASES

of clinical and experimental evidence. J Womens Health1998; 7:547–57.

46. Garson A Jr. How to measure the QT interval: whatis normal? Am J Cardiol 1993; 72:14B–16B.

47. Roden DM. Drug-induced prolongation of the QTinterval. N Engl J Med 2004; 350:1013–22.

48. Eap CB, Buclin T, Baumann P. Interindividual vari-ability of the clinical pharmacokinetics of methadone: im-plications for the treatment of opioid dependence. ClinPharmacokinet 2002; 41:1153–93.

49. Kharasch ED, Hoffer C, Whittington D, Sheffels P.Role of hepatic and intestinal cytochrome P450 3A and2B6 in the metabolism, disposition, and miotic effects ofmethadone. Clin Pharmacol Ther 2004; 76:250–69.

50. Gerber JG, Rhodes RJ, Gal J. Stereoselectivemetabolism of methadone N-demethylation by cytochromeP450 2B6 and 2C19. Chirality 2004; 16:36–44.

51. Crettol S, Deglon JJ, Besson J, et al. Methadoneenantiomer plasma levels, CYP2B6, CYP2C19, andCYP2C9 genotypes, and response to treatment. Clin Phar-macol Ther 2005; 78:593–604.

52. Crettol S, Deglon JJ, Besson J, et al. ABCB1 andcytochrome P450 genotypes and phenotypes: influence onmethadone plasma levels and response to treatment. ClinPharmacol Ther 2006; 80:668–81.

53. Eap CB, Crettol S, Rougier JS, et al. Stereoselectiveblock of hERG channel by (S)-methadone and QT intervalprolongation in CYP2B6 slow metabolizers. Clin Pharma-col Ther 2007; 81:719–28.

54. Bednar MM, Harrigan EP, Ruskin JN. Torsades dePointes associated with nonantiarrhythmic drugs and obser-vations on gender and QT. Am J Cardiol 2002; 89:1316–19.

55. Brink PA, Crotti L, Corfield V, et al. Phenotypicvariability and unusual clinical severity of congenital long-QT syndrome in a founder population. Circulation 2005;112:2602–10.

56. Ehret GB, Voide C, Gex-Fabry M, et al. Drug-induced long QT syndrome in injection drug users receivingmethadone: high frequency in hospitalized patients and riskfactors. Arch Intern Med 2006; 166: 1280–87.

57. DePonti F, Poluzzi E, Cavalli A, et al. Safety ofnon-antiarrhythmic drugs that prolong the QT interval orinduce Torsades de Pointes: an overview. Drug Saf 2002;25:263–86.

58. Katchman AN, McGroary KA, Kilborn MJ, et al.Influence op opioid agonists on cardiac human ether-a-go-go-related gene K(+) currents. J Pharmacol Exper Ther2002; 303:688–94.

59. Ackerman MJ, Tester DJ, Jones GS, et al. Ethnicdifferences in cardiac potassium channel variants: implica-tions for genetic susceptibility to sudden cardiac death andgenetic testing for congenital long QT syndrome. MayoClin Proceedings 2003; 78:1479–87.

60. De Bels D, Staroukine M, Devriendt J. Torsadesde Pointes due to methadone. Ann Intern Med 2003;139:E156.

61. Seyler DE, Borowitz JL, Maickel RP. Calcium chan-nel blockade by certain opioids. Fundam Appl Toxicol1983; 3:536–42.

62. Lee CH, Berkowitz BA. Calcium antagonist activityof methadone, l-acetylmethadol, and l-pentazocine in therat aortic strip. J Pharmacol Exp Ther 1977; 202:646–53.

63. Rendig SV, Amsterdam EA, Henderson GL, Ma-son DT. Comparative cardiac contractilities of six narcoticanalgesics: morphine, meperidine, pentazocine, fentanyl,methadone,and l-alpha-acetylmethadol (LAAM). J Phar-macol Exp Ther. 1980; 215:259–65.

64. Eikenburg DC, Stickney JL. Anti-cholinesterase ac-tivity of 1-alpha-acetylmethadol: relationship to bradycar-dia. Gen Pharmacol 1979; 10:195–200.

65. Stickney JL. Cardiac effects of 1-alpha-acetylmethadol. 1. Chronotropic effects in vitro. ToxicolAppl Pharmacol 1977; 40:23–32.

66. Krantz MJ, Lewkowiez L, Hays H, et al. Torsadesde Pointes associated with very high-dose methadone. AnnIntern Med 2002; 137:501–04.

67. Lipski J, Stimmel B, Donoso E. The effect of heroinand multiple drug abuse on the electrocardiogram. AmHeart J 1973; 86:663–68.

68. Krantz MJ, Kutinsky IB, Robertson AD, et al. Dose-related effects of methadone on QT prolongation in a se-ries of patients with Torsades de Pointes. Pharmacotherapy2003; 23:802–05.

69. Pimentel L, Mayo D. Chronic methadone therapycomplicated by Torsades de Pointes: a case report. J EmergMed 2008; 34:287–90.

70. Luthi B, Huttner A, Speck RF, et al. Methadone-induced Torsades de Pointes after stopping lopinavir-ritonavir. Eur J Clin Microbiol Infect Dis 2007; 26:367–69.

71. Atkinson D, Dunne A, Parker M. Torsades de pointesand self-terminating ventricular fibrillation in a prescriptionmethadone user. Anaesthesia 2007; 62:952–55.

72. Wong SC, Roberts JR. Case files of the DrexelUniversity Medical Toxicology Fellowship: methadone-induced QTc prolongation. J Med Toxicol 2007; 4:190–94.

73. Routhier DD, Katz KD, Brooks DE. QTc prolonga-tion and Torsades de Pointes associated with methadonetherapy. J Emerg Med 2007; 32:275–78.

74. Iskandar SB, Abi-Saleh BS, Mechleb BK, et al.Methadone and Torsades de Pointes: case report and re-view of the literature. Tenn Med 2007; 100:35–7.

75. Falconer M, Molloy D, Ingerhaug J, et al. Methadoneinduced Torsades de Pointes in a patient receiving antiretro-viral therapy. Ir Med J 2007; 100:631–32.

76. Lamont P, Hunt SC. A twist on Torsades: a pro-longed QT interval on methadone. J Gen Intern Med 2006;21:C9–C12.

77. Osticherling C, Schaer B, Ammann P, et al.Methadone-induced cases of Torsades de Pointes tachy-cardias. Swiss Med Wkly 2005; 135:282–85.

78. Krantz MJ, Garcia JA, Mehler PS. Effects ofbuprenorphine on cardiac repolarization in a patient with

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

SAMHSA Report 299

methadone-related Torsades de Pointes. Pharmacotherapy2005; 25:611–14.

79. Ower K, Morley-Forsler P, Moulin D. Fluctuat-ing QTc interval in an asymptomatic patient treated withmethadone for chronic pain. J Opioid Manag 2005; 1:73–6.

80. Porter BO, Coyne PJ, Smith WR. Methadone-relatedTorsades de Pointes in a sickle cell patient treated forchronic pain. Am J Hematol 2005; 78:316–17.

81. Rademacher S, Dietz R, Haverkamp W. QT pro-longation and syncope with methadone, doxepin, and abeta-blocker. Ann Pharmacother 2005; 39:1762–3.

82. Almehmi A, Malas AM, Yousufuddin M, et al.Methadone-induced Torsades de Pointes in a patient withnormal baseline QT interval. W V Med J 2004; 100:147–48.

83. Decerf JA, Gressens B, Brohet C, et al. Canmethadone prolong the QT interval? Intensive Care Med2004; 8:1690–1.

84. Piguet V, Desmeules J, Ehret G, et al. QT intervalprolongation in patients on methadone with concomitantdrugs. J Clin Psychopharmacol 2004; 24:446–48.

85. Lucchini L, Barbaro G, Barbarini G. Methadone andQTc prolongation in HIV-infected patients. Am J Cardiol2004; 94:147–48.

86. Walker P, Klein K, Kasza L. High dose methadoneand ventricular arrhythmias: a report of three cases. Pain2003; 103:321–24.

87. Gil M, Sala M, Anguera I, et al. QT prolongationand Torsades de Pointes in patients infected with humanimmunodeficiency virus and treated with methadone. AmJ Cardiol 2003; 92:995–7.

88. Justo D, Gal-Oz A, Paran Y, Goldin Y, Zeltser D.Methadone-associated Torsades de Pointes (polymorphicventricular tachycardia) in opioid-dependent patients. Ad-diction 2006; 101:1333–38.

89. Pearson EC, Woosley RL. QT prolongation and Tor-sades de Pointes among methadone users: Reports to theFDA spontaneous reporting system. PharmacoepidemiolDrug Saf 2005; 14:747–53.

90. Patel AM, Singh JP, Ruskin J. Role of implantablecardioverter-defibrillators in patients with methadone-induced long QT syndrome. Am J Cardiol 2008;101:209–11.

91. Chugh SS, Socoteanu C, Reinier K, et al. Acommunity-based evaluation of sudden death associatedwith therapeutic levels of methadone. Am J Med 2008;121:66–71.

92. Salle P, Rey JL, Bernasconi P, Quiret JC, LombaertM. Torsades de Pointes: Apropos of 60 cases. Ann CardilAngeiol (Paris) 1985; 34:381–88.

93. Milon D, Daubert JC, Saint-Marc C, Goufault J.Torsades de Pointes: Apropos of 54 cases. Ann Fr AnesthReanim 1982; 1:513–20.

94. Maremmani I, Pacini M, Cesaroni C, et al. QTc inter-val prolongation in patients on long-term methadone main-tenance therapy. Eur Addict Res 2005; 11:44–9.

95. Fanoe S, Hvidt C, Ege P, et al. Syncope and QTprolongation among patients treated with methadone forheroin dependence in the city of Copenhagen. Heart 2007;93:1051–55.

96. Cruciani RA, Sekiner R, Homel P, et al. Measure-ment of QTc in patients receiving chronic methadone ther-apy. J Pain Symptom Manage 2005; 29:385–91.

97. Martell BA, Arnsten JH, Krantz MJ, et al. Impact oforal methadone on cardiac repolarization and conductionin opioid users. Am J Cardiol. 2005; 95:915–18.

98. Krantz MJ, Lowery CM, Martell BA, et al. Effectsof methadone on QT interval dispersion. Pharmacotherapy2005; 25:1523–29.

99. Wedam EF, Bigelow G, Nuzzo P, et al. QT in-terval effects of methadone, levomethadyl acetate, andbuprenorphine in a randomized trial. Arch Intern Med 2007;167:2469–75.100. Kornick CA, Kilborn MJ, Santiago-Palma J, et al.

QTc interval prolongation associated with intravenousmethadone. Pain 2003; 105:499–506.101. Ferreira S, Crumb JR, Carlton CG, et al. Effects

of cocaine and its major metabolites on the HERG-encoded potassium channel. J Pharmacol Exp Ther 2001;299:220–6.102. Peles E, Bodner G, Kreek MJ, Rados V,

Adelson M. Corrected-QT intervals as related to methadonedose and serum level in methadone maintenance treatment(MMT) patients: a cross-sectional study. Addiction 2007;102:289–300.103. Huber A, Ling W, Fradis J, et al. Comparison of the

effects of methadone and LAAM on the electrocardiogram.Drug Alcohol Depend 2001; 63:S70.104. Krantz MJ, Mehler PS. QTc prolongation:

methadone’s efficacy-safety paradox. Lancet 2006;368:356–57.105. American Association for the Treatment of Opioid

Dependence. AATOD policy on QT interval screening.New York, NY: Author, 2009.106. Medicines and Healthcare Products Regulatory

Agency. Current problems in pharmacovigilance, Vol. 31.London: Author, 2006.107. Department of Health (England) and the devel-

oped administrations. Drug misuse and dependence: UKguidelines on clinical management. London: Departmentof Health (England), the Scottish Government, WelshAssembly Government and Northern Ireland Executive,2007.108. Botstein P. Is QT interval prolongation harmful?

A regulatory perspective. Am J Cardiol 1993; 72:50B–52B.

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

300 JOURNAL OF ADDICTIVE DISEASES

APPENDIX A

SAMHSA Expert Panel on CardiacEffects of Methadone

PANEL MEMBERSAnthony Campbell, RPh, DOMedical AdvisorDivision of Pharmacologic TherapiesCenter for Substance Abuse Treatment, SAMHSARockville, Maryland

Thomas Killip, MDBeth Israel Medical CenterMilton and Carroll Petrie DivisionNew York, New York

Margaret Kotz, DODirector, Addiction Recovery Servicesand Professor of of Psychiatry,Case Western Reserve University School of MedicineCleveland, Ohio

Mori J. Krantz, MDAssociate Professor of MedicineDivision of CardiologyUniversity of Colorado School of Medicine,and Director, Colorado Prevention CenterDenver Health Medical CenterDenver, Colorado

Mary Jeanne Kreek, MDProfessor and HeadThe Laboratory of the Biology of Addictive DiseasesThe Rockefeller UniversityNew York, New York

Judith Martin, MDMedical DirectorBAART Turk Street ClinicSan Francisco, California

Brian A. McCarroll, DO, MSMedical DirectorBioMed Behavioral Healthcare, Inc.Sterling Heights, Michigan

Davendra Mehta, MD, PhDProfessor of Medicine,and Director, Electrophysiology SectionMt. Sinai School of MedicineNew York, New York

J. Thomas Payte, MDMedical DirectorColonial Management Group, Inc.Orlando, Florida

Barry Stimmel, MD, FASAM (Panel Chair)Dean Emeritus of Graduate Medical EducationMt. Sinai School of MedicineNew York, New York

Trusandra Taylor, MD, MPHMedical DirectorJEVS Human ServicesPhiladelphia, PA

EX OFFICIO MEMBERSPaul Bowman, CMANew England DirectorNational Association of Methadone AdvocatesNorth Quincy, Massachusetts

Amina Chaudhry, MD, MPHMedical AdvisorDivision of Pharmacologic TherapiesCenter for Substance Abuse Treatment, SAMHSARockville, Maryland

Marc N. Gourevitch, MDDr. Adolph and Margaret Berger Professor of Medicineand Professor of Psychiatry,and Director, Division of General Internal MedicineNew York University Langone Medical CenterNew York, New York

Mark C. P. Haigney, MD, FAHADirector of Cardiologyand Professor of MedicineUniformed Services University of the Health SciencesBethesda, Maryland

Roberta Kahn, PhDDivision of Pharmacotherapiesand Medical Consequences of Drug AbuseNational Institute on Drug AbuseBethesda, Maryland

Robert Lubran, MS, MPADirector, Division of Pharmacologic TherapiesCenter for Substance Abuse Treatment, SAMHSARockville, Maryland

Charles O’Keeffe, PhDProfessor, Department of Epidemiologyand Community HealthVirginia Commonwealth UniversityRichmond, Virginia

Bob Rappaport, MDDirector, Division of Anesthesia, Analgesia,and Rheumatology ProductsFood and Drug AdministrationSilver Spring, Maryland

Alina Salvatore, RPh, MSPublic Health AdvisorDivision of Pharmacologic TherapiesCenter for Substance Abuse Treatment, SAMHSARockville, Maryland

APPENDIX B

Relevant Advice from Private-SectorOrganizations and Government Agencies

American Association for the Treatment of OpioidDependence (AATOD; United States)

AATOD’s guidelines on cardiac risk, issued in 2009,include the following general and specific steps:

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

SAMHSA Report 301

1. Opioid treatment programs should develop compre-hensive cardiac arrhythmia risk management plansthat specify the threshold and frequency for ECGscreening and monitoring.

2. A personal medical history of long QT syndrome,cardiac conduction defects, arrhythmias, syncopeepisodes, seizures, palpitations, dizziness and light-headedness, and family history of long QT syn-drome, cardiac conduction defects, arrhythmias,syncope episodes, seizures and sudden or unex-pected death should be part of a medical assessmentprior to admission to an opioid treatment program.

3. Electrolytes—in particular hypokalemia, hypo-magnesaemia and patients on medications thatcan induce these conditions (diuretics andlaxatives)—should be included in the medical as-sessment conducted at admission.

4. Patient records should note any history of clinicallysignificant bradycardia or cardiac disease (such asCHF and reduced ventricular function).

5. All prescribed medications should be reviewed priorto induction onto methadone treatment. Particularattention should be given to medications that aresubstrates of CYP3A4 or CYP3D26 and those thatblock HERG channel currents, as well as over-the-counter agents, herbal preparations, and dietary sup-plements.

6. Toxicology screens should be reviewed for the pres-ence of illicit drugs, particularly those that add tocardiac risk such as cocaine and amphetamines.

7. Medically fragile patients (including the elderly;patients with advanced heart, liver, or kidney dis-ease; patients with advanced HIV/AIDs; patientswho are taking opioid analgesics for chronic pain;and patients with a history of poor, extensive orrapid metabolism of methadone) should be closelymonitored.

8. Patient consent to methadone treatment should in-clude information about the cardiac risk associatedwith continued use of illicit drugs, particularly drugsdiluted with quinine.

9. Physicians who work in opioid treatment programs(as well as those in pain management) should beeducated about the risk of QTc/TdP in methadone-maintained patients.105

College of Physicians and Surgeons of Ontario(CPSO; Canada)

In its Methadone Maintenance Guideline,18 the CPSOsuggests that an ECG be performed when the methadonedose exceeds 150 mg/day. It recommends a repeat ECGwhen the dose approaches 180 to 200 mg.

The Canadian guideline further proposes tapering themethadone dose and referring the patient to a cardiologistif the QTc interval exceeds 470 ms.

Food and Drug Administration (FDA; United States)The issues described in this communication have been

addressed in product labeling.FDA ALERT [11/2006]:

Death, Narcotic Overdose, and Serious Cardiac Arrhyth-mias

“FDA has reviewed reports of death and life-threateningadverse events such as respiratory depression and cardiacarrhythmias in patients receiving methadone. These adverseevents are the possible result of unintentional methadoneoverdoses, drug interactions, and methadone’s cardiactoxicities (QT prolongation and Torsades de Pointes).Physicians prescribing methadone should be familiar withmethadone’s toxicities and unique pharmacologic prop-erties. Methadone’s elimination half-life (8–59 hours) islonger than its duration of analgesic action (4–8 hours).Methadone doses for pain should be carefully selected andslowly titrated to analgesic effect even in patients who areopioid-tolerant. Physicians should closely monitor patientswhen converting them from other opioids and changing themethadone dose, and thoroughly instruct patients how totake methadone. Healthcare professionals should tell pa-tients to take no more methadone than has been prescribedwithout first talking to their physician.”

Medicines and Healthcare Products RegulatoryAgency (MHRA; United Kingdom)

MHRA recommends monitoring patients on methadonedoses greater than 100 mg per day. However, the MHRArecommendation does not define the monitoring approachto be used.106 The UK guideline on clinical managementof drug abuse and addiction incorporates the MHRA ap-proach and suggests that patients be informed of the reasonscardiac monitoring is recommended,107 as follows:

“A2.1 Drug-induced prolongation of the QT interval.The QT interval is measured on an ECG∗ from the begin-ning of the QRS complex (caused by contraction of theventricular mass) until the end of the T wave (caused bythe return of the ventricular mass to the resting state). TheQT corrected (QTd) interval is the QT interval (in millisec-onds) corrected for heart rate using a standard formula,such as, Bazett’s formula: QTc (ms) = QT (ms) / RR1/2– QT divided by the square root of the R-R interval. QTccalculators are available on the Internet.

“The QTc interval is a useful indicator of risk of poly-morphic ventricular tachycardias, or torsade de pointeswhich can be fatal. QTc interval prolongation beyond nor-mal limits (440 ms for men and 470 ms for women) isassociated with increased risk of cardiac arrhythmias andsudden death, especially above 500 ms.108

“Various psychotropic medications have recently beenidentified as causing QT prolongation and sudden death. Inthe past decade this has become the most common reasonfor a drug to be withdrawn from the market. In the drugtreatment field, this was the reason for levacetylmethadol(LAAM or ORLAAM) being withdrawn.”12

APPENDIX C

Drugs that May Interact with Methadoneto Elevate Cardiac Risk

A limited number of clinical studies have investigatedinteractions between methadone and specific drugs; there-fore, some interactions are predicted based on lower levels

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

302 JOURNAL OF ADDICTIVE DISEASES

of evidence, such as case reports, laboratory experiments,or pharmacologic principles. The various levels of evidenceare denoted in the table as follows:

Level 1: Interaction demonstrated via published clinicalstudies and/or by the well-established and specificpharmacology of methadone metabolism.

Level 2: Based on published clinical case series reportsand/or laboratory investigations in animals or tissues(in vitro).

Level 3: Proposed in the literature, but predicted fromgeneral pharmacologic principles and/or sporadicanecdotal cases.

Generic name Trade Name (Examples) Mechanism of Action Level of Evidence

Amiodarone Cordarone, Nexterone,Amiadarone HCl injection

Proposed due to CYP450inhibition.

Level 3

Amitriptyline Elavil Possible increased TCA toxicity;uncertain effect on methadone.

Level 2

Clarithromycin Biaxin Strong CYP3A4 inhibition. Level 3Cocaine and crack — Methadone elimination

acceleratedLevel 2

Desipramine Norpramin Possible increased TCA toxicity;uncertain effect on methadone.

Level 2

Doxepin Sinequan, Prudoxin, Zonalon Possible increased TCA toxicity;uncertain effect on methadone.

Level 2

Erythromycin Benzamycin, EES, Emgel,Erythrocin, Ilosone, Ilotycin

Strong CYP3A4 inhibition. Level 3

Fluoxetine Prozac, Serafem Variable CYP450 enzymeinhibition.

Level 2

Imipramine Tofranil Possible increased TCA toxicity;uncertain effect on methadone.

Level 2

Ketoconazole Extina, Nizoral, Xolegel, Predicted due to CYP3A4inhibition.

Level 3

Nortriptyline Pamelor Possible increased TCA toxicity;uncertain effect on methadone

Level 2

Paroxetine Asimia, Paxil Variable CYP450 enzymeinhibition.

Level 2

Lansoprazole, amoxicillinand clarithromycin

Prevpac CYP3A4 inhibition (containsclarithromycin).

Level 3

Protriptyline Vivactil Possible increased TCA toxicity;uncertain effect on methadone

Level 2

Sertraline Zoloft Variable CYP450 enzymeinhibition.

Level 2

Tricyclic antidepressants Aventyl Possible increased TCA toxicity;uncertain effect on methadone.

Level 2

SOURCES: Adapted from Leavitt SR, Toombs JD & Kral L. Methadone-drug interactions. Pain Treatment Topics Jan. 20, 2006; Table 5; andLeavitt SR, Bruce RD, Eap CB, Kharasch E, Kral L, McCance-Katz E & Payte JT. Methadone-drug interactions. 3rd ed. Pain Treatment TopicsNov. 2005, pp. 18–24.

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

AP

PE

ND

IXD

Sum

mar

yof

Maj

orSt

udie

s*of

Met

hado

nean

dP

rolo

nged

QT

Synd

rom

e

Aut

hors

Jour

nal

Type

ofSt

udy

Num

ber

Stud

yPo

pula

tion

(MM

TP

orPa

in)

Oth

erM

edic

atio

nsE

valu

ated

orPr

esen

tO

ther

Dru

gsof

Abu

seE

valu

ated

Oth

erM

edic

alC

ondi

tions

Eva

luat

edM

ajor

Find

ings

Not

es,I

nclu

ding

Lim

itatio

ns

Chu

ghet

al.,

2008

Am

eric

anJo

urna

lof M

edic

ine

22ca

ses

com

pare

dw

ith10

6co

ntro

ls

All

deat

hsin

Port

land

met

ropo

litan

area

over

4ye

ars

with

‘the

rape

utic

’le

vels

ofm

etha

done

pres

ent,

com

pare

dw

ithno

n-m

etha

done

case

s.55

%w

ere

taki

ngm

etha

done

for

pain

cont

rol.

Yes

Pers

ons

with

evid

ence

ofm

etha

done

over

dose

orot

her

recr

eatio

nald

rug

use

excl

uded

Stru

ctur

alhe

artd

isea

seon

lyA

mon

gm

etha

done

case

s,ev

iden

ceof

stru

ctur

alhe

artd

isea

sein

23%

,co

mpa

red

with

60%

ofco

ntro

ls

The

reis

nosi

ngle

cut-

off

leve

lat

whi

chm

etha

done

beco

mes

toxi

cfo

ral

lper

sons

.

Cru

cian

iet

al.,

2005

Jour

nalo

fPa

inan

dSy

mpt

omM

anag

e-m

ent

Cro

ss-

sect

iona

l10

4to

tal

Bot

hM

MT

Pan

dpa

inY

es(n

otin

clud

edin

am

ultiv

aria

tean

alys

is)

No

Exc

lude

dw

ithco

ngen

ital

long

QT

synd

rom

e,im

plan

ted

pace

mak

er,

atri

alfib

rilla

tion,

orw

ide

QR

Sco

mpl

exon

prio

rE

CG

Seru

mel

ectr

olyt

esm

easu

red

33%

had

QT

cpr

olon

gatio

n(d

efine

das

QT

c>43

0m

sfo

rm

ales

and

450

ms

for

fem

ales

)N

opa

tient

had

QT

c>50

0ms

Sign

ifica

ntdo

sere

spon

sew

asse

enin

men

onm

etha

done

<12

mon

ths

No

cont

rolg

roup

Smal

l,ex

plor

ator

yst

udy

Ehr

etet

al.,

2006

Arc

hive

sof

Inte

rnal

Med

icin

e

Cas

e-co

ntro

l16

7m

etha

done

case

s80

cont

rol

Hos

pita

lized

IDU

onm

etha

done

vers

usco

ntro

lID

Uno

ton

met

hado

ne

Yes

,but

nots

peci

fied

Yes

HIV

Hep

atiti

sC

Hep

atiti

sB

Stru

ctur

alhe

atdi

seas

e

16.2

%of

met

hado

nepa

tient

sha

dQ

TC

≥50

0ms

com

pare

dto

0%of

cont

rols

.3.

6%of

thos

ein

the

met

hado

negr

oup

had

TdP

.Q

Tc

was

wea

kly

but

sign

ifica

ntly

corr

elat

edw

ithm

etha

done

dose

.In

am

ultiv

aria

tere

gres

sion

anal

ysis

,QT

prol

onga

tion

was

asso

ciat

edw

ithhi

gher

met

hado

nedo

se,

low

erpo

tass

ium

,low

erpr

othr

ombi

ntim

e,an

dco

-med

icat

ion

with

CY

P3A

4in

hibi

tion.

No

form

alas

sess

men

tof

stru

ctur

alhe

art

dise

ase

Oth

ersu

bsta

nce

use

was

self

-rep

orte

d

(Con

tinu

edon

next

page

)

303

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

(Con

tinu

ed)

Aut

hors

Jour

nal

Type

ofSt

udy

Num

ber

Stud

yPo

pula

tion

(MM

TP

orPa

in)

Oth

erM

edic

atio

nsE

valu

ated

orPr

esen

tO

ther

Dru

gsof

Abu

seE

valu

ated

Oth

erM

edic

alC

ondi

tions

Eva

luat

edM

ajor

Find

ings

Not

es,I

nclu

ding

Lim

itatio

ns

Fano

eet

al.,

2007

Hea

rtC

ross

-se

ctio

nal

450

(52%

)of

trea

tmen

tpo

pula

tion

inC

open

hage

n

OT

P(m

etha

done

vers

usbu

pren

orph

ine)

No

Yes

Coc

aine

,can

nabi

s,ill

icit

opio

ids,

illic

itbe

nzod

iaze

pine

s(b

utno

tinc

lude

din

the

mul

tivar

iate

anal

ysis

)

Pers

ons

with

atri

alfib

rilla

tion

orflu

tter,

bund

lebr

anch

bloc

k,bi

gem

iny,

orpa

cem

aker

wer

eex

clud

ed-s

erum

pota

ssiu

mw

asch

ecke

d

Met

hado

nedo

sew

assi

gnifi

cant

lyco

rrel

ated

with

QT

cfo

rbo

thm

enan

dw

omen

.N

oas

soci

atio

nw

asfo

und

betw

een

bupr

enor

phin

edo

sean

dQ

Tin

terv

al.

Ina

mul

tivar

iate

anal

ysis

,m

etha

done

dose

was

asso

ciat

edw

ithpr

olon

ged

QT

inte

rval

,and

seru

mpo

tass

ium

was

nega

tivel

yas

soci

ated

with

prol

onge

dQ

T.O

dds

ofse

lf-r

epor

ted

sync

ope

also

wer

e1.

2tim

eshi

gher

whe

nm

etha

done

dose

was

incr

ease

dby

50m

g.

Bup

reno

rphi

negr

oup

was

youn

ger

and

had

shor

ter

dura

tion

ofth

erap

y

Kor

nick

etal

.,20

03In

tern

atio

nal

Ass

ocia

-tio

nfo

rth

eSt

udy

ofPa

in

Pros

pect

ive

(cha

rtre

view

)ov

er20

mon

ths

N=

47(i

vm

etha

done

)N

=35

(mor

phin

e)

Can

cer

pain

Unk

now

nif

othe

rm

edic

atio

nsw

ere

pres

ent.

Acc

ordi

ngto

artic

le,

info

rmat

ion

was

colle

cted

.M

orph

ine.

Mor

phin

eN

one

Met

hado

nein

com

bina

tion

with

chlo

robu

tano

lis

asso

ciat

edw

ithQ

Tin

terv

alpr

olon

gatio

n.

IVm

etha

done

,w

hich

cont

ains

chlo

robu

tano

l,w

asev

alua

ted.

Chl

orob

utan

olal

one

can

affe

ctQ

Tin

terv

al.S

mal

lsa

mpl

esi

ze.

304

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

Kra

ntz

etal

.,20

03Ph

arm

aco-

ther

apy

Ret

rosp

ectiv

eca

sese

ries

anal

ysis

N=

179

patie

nts

wer

ere

ceiv

ing

met

hado

nefo

rop

ioid

depe

nden

cy(6

wer

efr

omC

olor

ado)

;8pa

tient

sw

ere

rece

ivin

gm

etha

done

for

chro

nic

pain

from

one

cent

er.A

llpa

tient

sha

dbe

enho

spita

lized

from

1996

–200

1w

ithto

rsad

esde

poin

tes.

Ola

nzap

ine,

fluox

etin

e,le

vace

tylm

etha

dol,

nelfi

navi

r,am

itrip

tylin

e

Coc

aine

,alc

ohol

Stru

ctur

alhe

artd

isea

seA

rela

tions

hip

was

foun

dbe

twee

nda

ilym

etha

done

dose

and

the

QT

inte

rval

.

Smal

lstu

dypo

pula

tion,

retr

ospe

ctiv

est

udy

desi

gn,

and

inhe

rent

sele

ctio

nbi

as.

Mar

emm

ani

etal

.,20

05

Eur

opea

nA

ddic

-tio

nR

esea

rch

Cro

ss-

sect

iona

l83

MM

TP

prog

ram

inIt

aly.

InM

MT

Pfo

rat

leas

t6m

onth

sSt

able

dose

for

atle

ast4

mon

ths

No

othe

rm

edic

atio

nskn

own

topr

olon

gQ

Ton

boar

dR

epor

ted

allh

adno

rmal

elec

trol

ytes

and

nutr

ition

alst

atus

(val

ues

not

prov

ided

)

Att

ime

ofE

CG

,all

nega

tive

for

mor

phin

e,co

cain

em

etab

olite

s,an

dam

phet

amin

esA

llse

lfre

port

ed“h

abitu

al”

alco

hol

use

(not

defin

ed)

Not

repo

rted

Rep

orte

dal

lhad

norm

alel

ectr

olyt

esan

dnu

triti

onal

stat

us(v

alue

sno

tpro

vide

d)

83.1

%ha

dQ

Tm

ore

prol

onge

dth

anag

ean

dse

xm

atch

edco

ntro

lsO

nly

2su

bjec

ts(2

.4%

)ha

dQ

T>

500

ms

No

rela

tions

hip

obse

rved

betw

een

QT

and

met

hado

nedo

se

Smal

lsam

ple

size

No

data

onpl

asm

ale

vels

Mar

tell

etal

.,20

05A

mer

ican

Jour

nal

ofC

ardi

-ol

ogy

Pros

pect

ive

160

MM

TP

Ant

idep

ress

ants

;Ca+

Cha

nnel

Blo

cker

s;A

ntir

etro

vira

ls;

Diu

retic

s;Ph

enyt

oin

Coc

aine

;ET

OH

;to

bacc

oH

epat

itis

C;H

IVin

fect

ion

Posi

tive

corr

elat

ion

betw

een

seru

mM

etha

done

conc

.an

dm

agni

tude

ofQ

Tpr

olon

gatio

n

Eff

ecto

fm

edic

alco

nditi

ons

orpr

escr

iptio

nm

edic

atio

nson

QT

prol

onga

tion

coul

dno

tbe

tota

llyex

clud

ed(C

onti

nued

onne

xtpa

ge)

305

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

(Con

tinu

ed)

Aut

hors

Jour

nal

Type

ofSt

udy

Num

ber

Stud

yPo

pula

tion

(MM

TP

orPa

in)

Oth

erM

edic

atio

nsE

valu

ated

orPr

esen

tO

ther

Dru

gsof

Abu

seE

valu

ated

Oth

erM

edic

alC

ondi

tions

Eva

luat

edM

ajor

Find

ings

Not

es,I

nclu

ding

Lim

itatio

ns

Pele

set

al.,

2006

Soci

ety

for

the

Stud

yof

Add

ic-

tion

Pros

pect

ive.

Patie

nts’

med

ical

char

tsw

ere

also

revi

ewed

retr

ospe

ctiv

e-ly

for

pres

crib

edm

edic

a-tio

nsin

the

peri

odbe

fore

the

stud

yw

aspe

rfor

med

.C

ross

sect

iona

l.

N=

138

MM

T(m

ustb

ein

trea

tmen

tfo

rat

leas

t100

days

inad

ditio

nto

bein

gon

stea

dym

etha

done

dose

sfo

rat

leas

t14

days

).Ph

ysic

ians

enco

urag

edpa

tient

sre

ceiv

ing

high

met

hado

nedo

ses

(ove

r12

0m

g/da

y)to

part

icip

ate.

)

Ben

zodi

azep

ines

,op

iate

s,am

phet

amin

es.

Oth

erm

edic

atio

nspr

esen

t:ur

sode

oxyc

holic

acid

,spi

rono

lact

one,

colc

hici

ne,

salb

utam

ol,

theo

trim

e,ip

ratr

opiu

mbr

omid

e,tr

azod

one,

met

form

in,t

hyro

xin

sodi

um,

clon

azep

am,

amox

icill

in,

fluox

etin

e,di

azep

am,

fluvo

xam

ine,

insu

lin,p

enflu

rido

l,en

oxa-

pari

n,es

cita

lopr

am,

amitr

ipty

line,

mel

aton

in,

halo

peri

dol,

bipe

ride

n,pr

opan

olol

,asp

irin

,m

irta

zapi

me

sodi

um,v

alpr

oate

,ac

etyl

salic

ylic

acid

,si

mva

stat

in,

ram

ipri

l,is

osor

bide

.

Ben

zodi

aze-

pine

s,op

iate

s,co

cain

e,ca

nnab

is,

amph

etam

ines

.

HC

V,H

IVN

oco

rrel

atio

nbe

twee

nQ

Tin

terv

alan

dm

etha

done

dose

san

dse

rum

leve

ls;

how

ever

,sig

nific

ant

corr

elat

ion

betw

een

met

hado

nedo

sean

dQ

Tin

terv

alw

ere

foun

din

patie

nts

who

wer

eur

ine

posi

tive

for

coca

ine.

Eve

nth

ough

the

stud

yw

ascr

oss-

sect

iona

l,a

grea

ter

prop

ortio

nof

“hig

h-do

se”

patie

nts

(120

mg/

day)

wer

ein

clud

ed,

limiti

ngth

ege

nera

lizab

il-ity

ofth

efin

ding

.Stu

dypo

pula

tion

limite

dto

one

prog

ram

inTe

lA

viv,

Isra

el;

not

mul

ti-ce

nter

.N

oba

selin

eE

CG

s.

Wed

amet

al.,

2007

Arc

hive

sof

Inte

rnal

Med

icin

e

Ran

dom

ized

cont

rolle

dtr

ial

220

MM

TP

N/A

Alc

ohol

;her

oin;

coca

ine

N/A

Com

pare

dto

levo

met

hady

lan

dm

etha

done

,bu

pren

orph

ine

isas

soci

ated

with

less

QT

prol

onga

tion

Abs

ence

ofpl

aceb

oar

mto

asse

ssra

ndom

inci

denc

eof

QT

prol

onga

tion

∗E

xclu

des

case

stud

ies

and

case

serie

s.

306

Dow

nloa

ded

by [

Mar

k H

aign

ey]

at 0

7:33

26

Janu

ary

2012

View publication statsView publication stats