Review Biological

Psychiatry

Medication Treatment of Opioid Use Disorder

James Bell and John Strang

ABSTRACT
Opioid use disorder (OUD) is a chronic, relapsing condition, often associated with legal, interpersonal, and
employment problems. Medications demonstrated to be effective for OUD are methadone (a full opioid agonist),
buprenorphine (a partial agonist), and naltrexone (an opioid antagonist). Methadone and buprenorphine act by
suppressing opioid withdrawal symptoms and attenuating the effects of other opioids. Naltrexone blocks the effects
of opioid agonists. Oral methadone has the strongest evidence for effectiveness. Longer duration of treatment allows
restoration of social connections and is associated with better outcomes. Treatments for OUD may be limited by poor
adherence to treatment recommendations and by high rates of relapse and increased risk of overdose after leaving
treatment. Treatment with methadone and buprenorphine has the additional risk of diversion and misuse of medi-
cation. New depot and implant formulations of buprenorphine and naltrexone have been developed to address issues
of safety and problems of poor treatment adherence. For people with OUD who do not respond to these treatments,
there is accumulating evidence for supervised injectable opioid treatment (prescribing pharmaceutical heroin).
Another medication mode of minimizing risk of overdose is take-home naloxone. Naloxone is an opioid antagonist
used to reverse opioid overdose, and take-home naloxone programs aim to prevent fatal overdose. All medication-
assisted treatment is limited by lack of access and by stigma. In seeking to stem the rising toll from OUD, expanding
access to approved treatment such as methadone, for which there remains the best evidence of efficacy, may be the
most useful approach.
Keywords: Buprenorphine, Heroin, Methadone, Naloxone, Naltrexone, Opioid use disorder
https://doi.org/10.1016/j.biopsych.2019.06.020

Three medications are currently registered in the United States and a lower likelihood of, return to treatment—a measure of
(and many other jurisdictions) for use in medication-assisted relapse (5). The duration of methadone treatment has been
treatment (MAT) of opioid use disorder (OUD). Most experi- demonstrated to be a linear, nonthreshold predictor of
ence has been with methadone, which remains the gold outcome (6); longer episodes of treatment are associated with
standard against which other medications have been better outcomes.
compared. Methadone is a full opioid agonist, producing dose- Slowly tapering doses of methadone or buprenorphine have
dependent analgesia, sedation, and risk of respiratory been used in medically managed withdrawal from opioids
depression in overdose. It has a long but variable half-life, (“detoxification”). However, short-term abstinence is not pre-
usually around 22 hours, but ranging from 13 to 50 hours (1). dictive of long-term remission (7). Given the consistent evi-
Buprenorphine is categorized as a partial agonist at the mu dence of the importance of treatment duration, the likelihood of
opioid receptor. It has a high affinity for the mu opioid receptor; relapse with treatment discontinuation, and increased risk of
low doses produce typical opioid effects, but higher doses overdose after detoxification (8), offering medically managed
prolong rather than increase opioid agonist effects (2). withdrawal alone increases risks with uncertain benefit. A
Naltrexone is a mu opioid receptor antagonist, blocking the flexible approach to patients who request detoxification is to
effects of opioids (and precipitating withdrawal if administered initiate MAT with the option of continuing or reducing slowly to
to a person currently dependent on opioids). It has been used abstinence, depending on patient’s progress (9).
clinically as an aid to sustaining abstinence, blocking the The objectives of long-term management include reduced
pleasurable effects of opioids and reducing the risk of relapse risk of death and disease, improvement in mental health and
after impulsive use of opioids. outlook, and restoration of social role impaired through such
Follow-up studies of people seeking treatment demonstrate issues as unemployment, disrupted family relations, and
that the course of OUD tends to be a chronic, relapsing one, involvement with the criminal justice system (4). These objec-
particularly for people with poor social supports or mental tives are most likely to be achieved if patients stop or markedly
health issues (3). Treatment of OUD does not fit an acute care reduce their use of illicit opioids (10,11). Most research on the
paradigm, with the objective of cure, but is better conceptu- effectiveness of MAT has been undertaken with methadone. In
alized as the management of a chronic condition (4). Longer suppressing the use of illicit opioids, methadone treatment is
episodes of continuous treatment are associated with later, more effective than short-term treatment or no treatment (12).

ª 2019 Society of Biological Psychiatry. 1

ISSN: 0006-3223 Biological Psychiatry - -, 2019; -:-–- www.sobp.org/journal
Biological
Psychiatry Medication Treatment of Opioid Use Disorder

Methadone treatment attracts and retains in treatment more drugs, and if the dose is raised too rapidly, respiratory depres-
people with OUD than drug-free approaches (12,13). The sion, which is potentially fatal, results. During induction, the blood
suppression of illicit opioid use that follows entry into treatment level achieved by a stable dose progressively increases over the
results in a reduced risk of fatal overdose for as long as people first week, owing to tissue binding and to methadone‘s long half-
remain in treatment (14,15). Reduction in crime (16,17) and life. A dose that was tolerated on day 1 may produce fatal res-
subjective improvement in quality of life (18) also occur through piratory depression in nontolerant patients on day 2 or 3. Clinical
reduced use of illicit opioids during methadone treatment. guidelines recommend that methadone doses need to start low
Premature dropping out of treatment is common, and and be slowly increased to avoid fatalities in the first month of
people often cycle in and out of short terms of MAT (19,20). In treatment (15). Increasing doses of buprenorphine produce little
part, poor response to MAT is due to variations in treatment or no increase in opioid effects. This “ceiling effect” reduces the
delivery, which often deviates from what has been demon- risk of respiratory depression in the event of overdose. Induction
strated to be effective (21,22). In particular, use of subthera- onto buprenorphine is associated with a significantly lower risk of
peutic doses of methadone and buprenorphine has overdose than induction onto methadone (15).
compromised treatment effectiveness (23). The importance of While adhering to MAT, patients are at reduced risk of
adequate doses reflects the pharmacological mechanism of overdose; however, there is a significantly increased risk of
action of methadone and buprenorphine—suppressing opioid fatal overdose in the month after leaving any form of treatment
withdrawal, attenuating the effects of injected opioids, and for OUD (5,15,31). Benzodiazepine use is common among
protecting against overdose (23). people with OUD, and concurrent use of benzodiazepines and
A single daily dose of 30 mg methadone is sufficient to methadone is associated with an increased risk of fatal over-
block the emergence of opioid withdrawal for at least 24 hours dose and emergency room presentations (32,33).
in most people with OUD (24). However, at this daily dose,
many patients persist in injecting street opioids. Higher doses QT Interval Prolongation
of methadone induce opioid tolerance, reducing the rewarding
Methadone prolongs the QT interval in dose-dependent
effect of injected opioids. A pooled review of randomized
fashion, and high-dose methadone has been associated with
controlled trials reported that a high dose (60–100 mg/day
the ventricular tachycardia torsade de pointes (34). This is rare
methadone) was more effective in producing abstinence from
complication, and there is little consensus on the implications
nonprescribed opioids, as detected by urine testing, than were
for treatment programs (35). Buprenorphine appears to have
middle (40–60 mg) or low (,40 mg) doses (25).
minimal impact on the corrected QT interval (36).
Buprenorphine at dose .8 mg/day has been reported to
suppress opioid withdrawal for 24 hours in opioid-dependent
subjects (26). However, to suppress use of illicit opioids dur- Diversion
ing treatment, higher doses of buprenorphine are more effec- As prescribing of methadone or buprenorphine in a jurisdiction
tive. Buprenorphine binds to opioid receptors, producing increases, there is an increase in overdose deaths from these
prompt blockade of other opioids. Positron emission tomog- drugs among people not in treatment (37). Some degree of
raphy scanning indicates that buprenorphine 2 mg produced diversion is inevitable when prescribing agonist medication
41% inhibition of carfentanil binding, 16 mg produced 80% (38), and much of the diverted mediation goes to people with
inhibition, and 32 mg produced 84% inhibition (27). Consistent OUD who are not in treatment. Measures to minimize diversion
with these laboratory findings, a meta-analysis of published include ensuring good access to treatment and administering
trials concluded that buprenorphine doses .16 mg/day were doses under direct observation, thus restricting unsupervised
more effective than doses ,16 mg at retaining subjects in doses to people who meet criteria of stability (38). There have
treatment (28). not been reports of problems associated with diversion and
An oral dose of 50 mg naltrexone produces mu opioid re- misuse of naltrexone.
ceptor blockade lasting 24 to 36 hours. It is difficult to induct In some countries (e.g., United Kingdom, Australia), meth-
people onto naltrexone because of the need for a period of adone and buprenorphine are administered under direct
abstinence before initiating antagonist treatment; once in observation. In these jurisdictions, both medications can be
treatment, there is a high rate of early discontinuation (29,30). prescribed in primary care and dispensed (with observation) by
Methadone and buprenorphine produce “drug-liking” re- community pharmacists. In the United States and France,
sponses in people with OUD, and people who miss doses methadone is only available via clinic-based programs and is
experience opioid withdrawal. These responses contribute to administered mainly under direct observation, but buprenor-
holding people in treatment. Naltrexone produces no positive phine can be prescribed by trained doctors, with medication
opioid effects, and this may contribute to erratic not directly observed. This model of treatment is made
compliance, early dropout, and a resulting increased risk of possible because buprenorphine diversion is associated with
fatal overdose (31). significantly less risk of fatal overdose than diverted metha-
done (39). Office-based treatment improves access, reduces
SAFETY OF MAT stigma, and requires less costly infrastructure.
The provision of buprenorphine without supervised admin-
Respiratory Depression and Overdose istration potentially risks diversion. An extreme example is that
Methadone produces respiratory depression if administered to buprenorphine diverted from France created a black market in
nontolerant individuals. Methadone treatment involves inducing a Georgia, where injected buprenorphine became the primary
higher level of tolerance than that occurring during use of illicit drug of abuse (40). A combination of buprenorphine and

2 Biological Psychiatry - -, 2019; -:-–- www.sobp.org/journal

 Biological
Medication Treatment of Opioid Use Disorder Psychiatry

naloxone has been marketed to minimize intravenous misuse. buprenorphine. It confirmed a significantly higher death rate
The rationale is that naloxone taken orally or sublingually has during induction onto methadone compared with buprenor-
only low bioavailability, whereas if the medication is crushed phine, and no difference during treatment, but in the month
and injected, naloxone attenuates the opioid agonist effect and after leaving treatment, drug-related mortality was significantly
precipitates withdrawal in people dependent on opioids. There lower for the methadone cohort (adjusted drug-related mor-
is conflicting evidence as to the effectiveness of the combi- tality rate ratio, 0.50; 95% confidence interval, 0.29–0.86).
nation in deterring intravenous misuse; the added
naloxone appears to reduce, but does not abolish, intravenous New Formulations: Long-Acting Buprenorphine and
misuse (41). Naltrexone Formulations
Enhancing retention in treatments for OUD, especially with
COMPARING EFFECTIVENESS OF MEDICATIONS naltrexone and buprenorphine, is key to improving long-term
USED IN MAT outcomes. One approach to improving adherence has been
development of sustained-release preparations. Two depot
Large-scale observational studies (15) and pooled analysis of injections of buprenorphine, and one buprenorphine implant,
randomized trials (12) concur that retention on buprenorphine have been tested in clinical trials. One trial of depot bupre-
is inferior to retention on methadone. The impact of differences norphine compared it with sublingual BNx and reported that
in retention has been demonstrated in a large trial that the depot was not inferior (47). The second trial, using a
randomized participants to methadone or sublingual different depot, reported that it was superior to placebo (48).
buprenorphine-naloxone (BNx) at flexible doses (42). At 24 One trial of buprenorphine implants has been reported,
weeks, 74% remained on methadone and 46% remained on demonstrating noninferiority to sublingual BNx (49). Based on
BNx, a clinically and statistically significant difference. Doses these studies, it is not possible to conclude that any of these
higher than 16 mg BNx and .60 mg of methadone were depot preparations is superior in efficacy to the others, nor that
associated with fewer opioid-positive urine tests. Non- depot preparations are superior to sublingual buprenorphine.
prescribed opioid use was significantly lower among BNx than In all trials of depot buprenorphine, more than half of urine
methadone participants during the first 9 weeks, but not samples were opioid positive in the active treatment groups, in
thereafter. This reflects the pharmacological differences part owing to high dropout rates.
between the full agonist and partial agonist—methadone A sustained-release injectable formulation of naltrexone,
suppresses the response to illicit opioid use in a dose- and designed to block opioid effects for 1 month, was tested in an
time-dependent fashion, through the progressive development open-label trial. Opioid-dependent individuals were random-
of tolerance (43), whereas buprenorphine promptly occupies ized to either sustained-release naltrexone or sublingual BNx
mu receptors and produces a degree of receptor blockade. (50). There were more dropouts during induction in the
However, the benefits of early suppression of illicit opioid use naltrexone group (72%, compared with 94% in BNx). On an
were offset by poor retention. Even with the highest BNx dose intention-to-treat analysis, relapse at 24 weeks was 65% for
level of 30 to 32 mg, the retention rate was less than the rate in naltrexone and 57% for buprenorphine; this difference was
the methadone group, and approximately 30% of the attributable to induction failures. Among participants suc-
participants continued illicit opioid use. Significantly more cessfully inducted, 24-week relapse events were similar across
BNx participants left the trial because they no longer wished study groups. Opioid-negative urine samples (p , .0001) and
to participate in their assigned treatment (25.6% vs. 12.4%; opioid-abstinent days (p , .0001) favored BNx compared with
p , .001). the depot among the intention-to-treat population but were
A qualitative study of a subset of participants in this trial (44) similar across study groups among the per-protocol popula-
reported that for some, withdrawal symptoms or negative re- tion. A smaller Norwegian trial (51) reported similar results,
actions continued beyond the induction period, despite although on the retained-in-treatment comparison, naltrexone
dosage adjustments. Some participants reached a dose of 32 was superior to BNx in suppressing heroin use.
mg but continued to feel sick. These results are consistent with Overall, depot preparations of buprenorphine tested to date
findings from a double-blind, double-dummy comparison of have not performed better than sublingual BNx. Retention in
methadone and buprenorphine (45), in which respondents on depot naltrexone was better than usually observed in studies
buprenorphine reported significantly more withdrawal symp- using oral naltrexone, but not superior to sublingual BNx. As
toms and less positive opioid effects from their medication with all forms of treatment for OUD, the naltrexone depot
than did subjects receiving methadone. probably does not avoid the risk of overdose after it ceases to
A follow-up study to the trial comparing BNx and metha- deliver a therapeutic blood level, as there have been case re-
done was conducted a mean of 4.5 years postrandomization; ports of fatalities following naltrexone depot (52).
73% of participants were interviewed and asked to provide a
urine sample (46). Opioid use at follow-up was significantly
higher among participants originally randomized to BNx, COMPARING MAT WITH NONMEDICAL
mainly due to less participation in treatment. At follow-up, APPROACHES TO OUD
mortality was not different between the two randomized con- Despite the limitations of MAT, it has distinct advantages over
ditions. However, clinical trials may be underpowered to detect other approaches to treatment of OUD. An observational study
differences in mortality. A large data-linkage study from from the United States (53) looked at participation in MAT
Australia (15) investigated mortality and overdose mortality in using Medicaid data on people 22 years of age or younger with
32,033 people who had commenced methadone or a diagnosis of OUD. The end point was retention in treatment

Biological Psychiatry - -, 2019; -:-–- www.sobp.org/journal 3

Biological
Psychiatry Medication Treatment of Opioid Use Disorder

as shown by no Medicaid claims within 60 days. Early intro- HAT is a treatment for individuals with injecting OUD who
duction of any medication was associated with better retention have failed to derive benefit from other treatments. It involves
than drug-free approaches; median retention was 123 days in supervised injection of pharmaceutical heroin up to 3 times
buprenorphine, 150 days in naltrexone, 324 days in metha- daily, with concurrent oral methadone (or other long-acting
done, and 67 days among youths who received only behavioral opioid) available to minimize withdrawal in the interdosing in-
health services. MATs were each independently associated tervals. The rationale for prescribing pharmaceutical heroin is
with lower attrition from treatment compared with receipt of to enable some individuals with the most entrenched addiction
behavioral health services alone. This study has limitations, to stop use of illicit opioids, breaking the link with associated
being nonrandomized, and with an end point that did not health risks and criminal activities. HAT patients often have
clearly identify how subjects were doing. However, it has the histories of poor therapeutic relationships in previous treat-
advantage of being a real-world study. Receipt of medication ments; trials demonstrate that with skilled staff to provide
helped hold young people with OUD in treatment. Naltrexone support and supervision, HAT can hold previously treatment-
and buprenorphine were similarly effective in retaining people, resistant subjects in structured treatment (62).
although not as effective as methadone. The major adverse events associated with HAT are respi-
ratory depression and seizures, both of which tend to occur
MAT IN PREGNANCY within minutes after injection. Pooled analysis from trials in-
dicates that there was a significantly higher risk of serious
OUD during pregnancy is associated with many adverse out-
adverse events linked to medication in the HAT groups
comes, including an increased risk of stillbirth, prematurity,
compared with methadone groups, but no deaths attributable
intrauterine growth retardation, and prolonged hospital stay
to prescribed medication, as overdoses were managed by
postdelivery (54). Management of the pregnant woman with
supervising staff (60).
OUD is optimized by use of a coordinated, specialist team
Randomized controlled trials are short-term studies and
including social services, addiction medicine, obstetrics, and
give little indication about long-term outcomes. Much re-
neonatology (55). When used in conjunction with comprehen-
mains to be learned about the place of HAT in a treatment
sive care, methadone and buprenorphine have been demon-
system. The cost is several times more than with buprenor-
strated to improve treatment outcomes in pregnant women
phine or methadone maintenance (although not as expensive
with OUD (56). The objective of medication management is to
as long-term residential rehabilitation or prison); however,
provide a stable intrauterine milieu, minimizing episodes of
cost-benefit analyses have reported that HAT is cost-
opioid intoxication and opioid withdrawal. Outcomes when
effective, as other treatments deliver such poor results for
methadone was continued throughout pregnancy were supe-
these patients (63).
rior to those observed in women managed with methadone
withdrawal during pregnancy (56).
Babies exposed to methadone in utero are at high risk of NALOXONE
neonatal abstinence syndrome, characterized by irritability and
Naloxone is an opioid antagonist; intravenous or intramuscular
autonomic dysfunction. Recent data demonstrate that bupre-
naloxone promptly reverses opioid effects and is routinely
norphine is a safe alternative to methadone in pregnancy and is
used to reverse opioid overdose. Preprovision of a take-home
associated with significantly lower severity of neonatal absti-
naloxone (THN) kit to injecting drug users, plus training the
nence syndrome (57).
recipients in emergency management of the overdose situa-
Whereas use of methadone and buprenorphine in preg-
tion, is an intervention being introduced in many countries. The
nancy is now accepted, there is less experience with
objective is to reduce opioid overdose fatalities. The rationale
naltrexone in managing OUD in pregnancy. Opioid withdrawal
is that most overdoses occur in the presence of others, with
during pregnancy is associated with an increased risk of fetal
observers often taking actions (that are not always well
distress and miscarriage, making it risky to initiate naltrexone
informed) to reverse overdose. Target populations have shown
during pregnancy (58).
high levels of willingness to engage actively with emergency
resuscitation.
HEROIN-ASSISTED TREATMENT THN projects report real-world emergency use of about
In the mid-1990s, confronted with a heroin epidemic, Swiss 10% of naloxone distributed. Two recent, independently con-
clinicians introduced a highly structured treatment, involving ducted Bradford-Hill public health analyses (64,65) both
the injecting of pharmaceutical heroin up to 3 times daily under concluded, on the basis of published evidence from a range of
direct observation (59). A subsequent series of randomized sources, that THN reduces the risk of fatal overdose.
trials provided the evidence for this treatment. Pooled analysis The development of new concentrated naloxone nasal
of trials showed that more than half of apparently “untreatable” sprays (three different products now in use globally) should
or poorly responding patients substantially disengage from use increase the acceptability of wider THN. The unit dose used in
of street heroin within 6 months, with reductions in criminal emergency room and ambulance setting is 0.4 mg, often
involvement and improvement in health and well-being (60). administered intramuscularly, with repeat dosing as necessary.
From these findings emerged a new treatment modality, now All THN kits include more than one dose (typically a twin-pack).
available in Canada and some European countries: heroin- The new naloxone nasal sprays deliver doses ranging from 1 to
assisted treatment (HAT). Injected hydromorphone has also 4 mg per spray, with approximately 50% bioavailability (66).
been trialed and appears to be of similar efficacy as injected There remains much to be learned about use of THN.
heroin (61). Differing local drug problems (such as fentanyl availability in

4 Biological Psychiatry - -, 2019; -:-–- www.sobp.org/journal

 Biological
Medication Treatment of Opioid Use Disorder Psychiatry

North America) may mean that higher doses of naloxone are (21). In the current North American opioid epidemic, there is a
required. Higher doses of naloxone can precipitate withdrawal lack of treatment spaces. For 2 decades, U.S. authorities have
in dependent users, so a dose-titration approach, starting with sought to expand treatment, particularly buprenorphine treat-
lower doses, is proposed in other countries (67). THN depends ment. However, in 2012, a decade after treatment expansion
on competence by lay responders, and there are also differ- began, it was estimated that the maximal potential availability
ences in the importance attached to training. Some advocates of methadone or buprenorphine treatment was about 60% of
concentrate on attention to respiratory depression and the the number of individuals with OUD (70), and actual availability
provision of assisted breathing, whereas others present much fell well short of potential. About half of Drug Addiction
fuller training including cardiopulmonary resuscitation. THN Treatment Act (DATA)-waived physicians actually prescribed
schemes generally stress the importance of a prompt call for buprenorphine; of these prescribers, the majority did not pre-
an ambulance call alongside assisted breathing. scribe to their maximum patient limit. A further recent response
Cost often obstructs THN provision. In some countries, THN has been the much wider provision of THN so that when
kits are provided at no cost to the individual recipient, whereas overdose occurs, fatal outcome can be avoided, and the Food
other countries require the individual to meet the subsidized or and Drug Administration is now considering whether emer-
full-price cost. Naloxone itself is inexpensive, with naloxone gency naloxone packs should be co-prescribed with some or
ampoules costing a dollar, whereas newer formulations can all opioid prescriptions (71).
cost hundreds of dollars.

WHICH MEDICATION FOR WHICH PATIENT? CONCLUSIONS

There is consistent evidence for the greater pharmacological Methadone has the best evidence for effectiveness; bupre-
effectiveness of methadone in retaining people in structured norphine and naltrexone also have evidence of effectiveness,
treatment. The distinct advantages of buprenorphine and provide distinct different benefits, and are important treatment
naltrexone is that they do not require daily attendance for su- options. Evidence of whether the new, sustained-release for-
pervised administration, making them valuable options for in- mulations of buprenorphine and naltrexone bring distinct
dividuals who can benefit from less structured treatment. advantage is not yet available. Some people with injecting
There is currently no evidence to predict which patient will do OUD and a poor response to conventional treatments can
best with which medication. The main guide is patient prefer- benefit from the more intensive treatment of HAT. Provision of
ence. A flexible approach to treatment is optimal, adjusting naloxone to people not currently in treatment may be one
medication and treatment conditions according to response. measure to reduce the overdose toll.
Stepped care, in which people commence treatment on Expanding access to methadone treatment through allow-
buprenorphine but transfer promptly to methadone if not ing methadone prescribing in primary care and dispensing in
responding, has been demonstrated to be as effective as pharmacies has been proposed as the most useful approach
initiating treatment with methadone (68). A French study to expanding access to effective treatment in the United
confirmed that individuals not responding to buprenorphine States, where currently methadone treatment is more highly
who were transferred to methadone had significantly improved restricted than in many other countries (72). Based on the
outcomes (16). In people not responding to methadone, greater pharmacological efficacy of methadone in holding
transfer to HAT can result in improved outcomes (60). people in treatment, this seems a timely recommendation.

RESPONDING TO THE NORTH AMERICAN OPIOID ACKNOWLEDGMENTS AND DISCLOSURES

EPIDEMIC JS's contribution to the study was funded by the National Institute for Health
North America is in the grip of an “opioid crisis.” Escalating Research (NIHR) Maudsley Biomedical Research Centre (BRC) (Grant No.
IS-BRC-1215-20018). The views expressed are those of the author(s) and
opioid overdose deaths are attributable to overlapping epi-
not necessarily those of the NHS, the NIHR, or the Department of Health and
demics—a longer-phase epidemic of prescription opioid use Social Care.
with associated deaths, a more recent epidemic of illicit heroin JB in the last 5 years has been a paid consultant for Martindale Pharma,
use and mortality, and most recently, a sharp-onset epidemic and for Indivior. He has received funding support to attend conferences from
of use of illicitly manufactured fentanyl. Age-standardized Indivior. JS is a researcher and clinician who, through his university, has
mortality rates for fentanyl, fentanyl analogues, and tramadol worked with various pharmaceutical companies to identify new or improved
treatments and his employer (King’s College London) has received grants,
were 0.3 per 100,000 in 1999, 1.0 per 100,000 in 2013, 1.8 per
travel costs, and/or consultancy payments, which have included consulta-
100,000 in 2014, 3.1 per 100,000 in 2015, and 6.2 per 100,000 tions with, in the past 3 years, Martindale, Indivior, Mundipharma, Camurus,
in 2016. In contrast, mortality rates for heroin increased from and Molteni Farma. The university and clinical services have received sup-
0.7 per 100,000 in 1999, to 1.0 per 100,000 in 2010, to 4.9 per plies of study medications for trials from companies including Camurus and
100,000 in 2016; and for methadone, mortality rates were 0.3 also iGen. King’s College London has also registered intellectual property on
per 100,000 in 1999, 1.8 per 100,000 in 2006, and 1.0 per a novel buccal naloxone formulation, naming JS and colleagues, and he was
earlier named in a patent registration by a pharmaceutical company
100,000 in 2016 (69).
regarding concentrated nasal naloxone spray.
OUD underlies the growing opioid overdose crisis in North
America. Opioid overdose deaths could be reduced by
ensuring that people with OUD have access to treatment that ARTICLE INFORMATION
is affordable and of adequate quality such that patients remain From the National Addiction Centre, Institute of Psychiatry, Psychology and
in treatment and have a greater likelihood of good outcomes Neuroscience, King’s College London, London, United Kingdom.

Biological Psychiatry - -, 2019; -:-–- www.sobp.org/journal 5

Biological
Psychiatry Medication Treatment of Opioid Use Disorder

Address correspondence to James Bell, M.D., Uniting MSIC, 66 Dar- 19. Bell J, Burrell T, Indig D, Gilmour S (2006): Cycling in and out of
linghurst Road, Kings Cross, Potts Point, NSW 2011, Australia; E-mail: treatment; participation in methadone treatment in NSW, 1990-2002.
[email protected]. Drug Alcohol Depend 81:55–61.
Received Mar 16, 2019; revised and accepted Jun 25, 2019. 20. Zhang L, Zou X, Zhang D, Li X, Zhao P, et al. (2015): Investigation of
repeat client drop-out and re-enrolment cycles in fourteen methadone
maintenance treatment clinics in Guangdong, China. PLoS One
10:e0139942.
REFERENCES 21. Barnett PG, Trafton JA, Humphreys K (2010): The cost of concordance
1. Inturrisi C (2005): Pharmacology of methadone and its isomers. with opiate substitution treatment guidelines. J Subst Abuse Treat
Minerva Anestesiol 71:435–437. 39:141–149.
2. Walsh SL, Eissenberg T (2003): The clinical pharmacology of bupre- 22. D’Aunno T, Vaughan TE (1992): Variation in methadone treatment
norphine: Extrapolating from the laboratory to the clinic. Drug Alcohol practices: Results from a national study. JAMA 267:253–258.
Depend 70:S13–S27. 23. Bell J (2014): Pharmacological maintenance treatments of opiate
3. Hser Y, Hoffman V, Grella C, Anglin MD (2001): A 33-year follow-up of addiction. Br J Clin Pharmacol 77:253–263.
narcotic addicts. Arch Gen Psychiatry 58:503–508. 24. Walsh SL, Strain EC (2006): Pharmacology of methadone. In:
4. McLellan AT, McKay JR, Forman R, Cacciola J, Kemp J (2005): Strain EC, Stitzer M, editors. The Treatment of Opioid Dependence.
Reconsidering the evaluation of addiction treatment: From retro- Baltimore, MD: John Hopkins University Press, 59–76.
spective follow-up to concurrent recovery monitoring. Addiction 25. Faggiano F, Vigna-Taglianti F, Versino E, Lemma P (2003): Methadone
100:447–458. maintenance at different dosages for opioid dependence. Cochrane
5. Bell J, Trinh L, Butler B, Randall D, Rubin G (2009): Comparing Database Syst Rev 3:CD002208.
retention in treatment and mortality in people after initial entry to 26. Kuhlman JJ, Levine B, Johnson RE, Fudala PJ, Cone EJ (1998):
methadone and buprenorphine treatment. Addiction 104:1193– Relationship of plasma buprenorphine and norbuprenorphine to
1200. withdrawal symptoms during dose induction, maintenance and with-
6. Zhang Z, Friedmann PD, Gerstein DR (2003): Does retention matter? drawal from sublingual buprenorphine. Addiction 93:549–559.
Treatment duration and improvement in drug use. Addiction 98:673– 27. Greenwald MK, Johanson C, Moody DE, Woods JH, Kilbourn MR,
684. Koeppe RA, et al. (2003): Effects of buprenorphine maintenance dose
7. Moos RH (2007): Theory-based processes that promote the remission on mu opioid receptor availability, plasma concentrations, and
of substance use disorders. Clin Psychol Rev 27:537–551. antagonist blockade in heroin-dependent volunteers. Neuro-
8. Wines JD, Saitz R, Hortone NJ, Lloyd-Travaglini C, Samet JH (2007): psychopharmacology 28:2000–2009.
Overdose after detoxification: A prospective study. Drug Alcohol 28. Fareed A, Vayalapalli S, Casarella J, Drexler K (2012): Effect of
Depend 89:161–169. buprenorphine dose on treatment outcome. J Addict Dis 31:8–18.
9. Lintzeris N, Bell J, Bammer G, Jolly D, Rushworth L (2002): 29. Morgan JR, Schackman BR, Leff JA, Linas BP (2018): Injectable
A randomised, controlled trial of buprenorphine in the management naltrexone, oral naltrexone, and buprenorphine utilization and
of short-term ambulatory heroin withdrawal. Addiction 97:1395– discontinuation among individuals treated for opioid use disorder in a
1404. United States commercially insured population. J Subst Abuse Treat
10. Dennis ML, Scott CK, Funk R, Foss MA (2005): The duration and 85:90–96.
correlates of addiction and treatment careers. J Subst Abuse Treat 30. Tucker T, Ritter AJ (2000): Naltrexone in the treatment of heroin
28(suppl 1):S51–S62. dependence: A literature review. Drug Alcohol Rev 19:73–82.
11. Gossop M, Marsden J, Stewart D, Treacy S (2001): Outcomes after 31. Degenhardt L, Larney S, Kimber J, Farrell M, Hall W (2015): Excess
methadone maintenance and methadone reduction treatments: Two- mortality among opioid-using patients treated with oral naltrexone in
year follow-up results from the National Treatment Outcome Australia. Drug Alcohol Rev 34:90–96.
Research Study. Drug Alcohol Depend 62:255–264. 32. Sun EC, Dixit A, Humphreys K, Darnall BD, Baker LC, Mackey S
12. Mattick RP, Breen C, Kimber J, Davoli M (2014): Buprenorphine (2017): Association between concurrent use of prescription opioids
maintenance versus placebo or methadone maintenance for and benzodiazepines and overdose: Retrospective analysis. BMJ
opioid dependence. Cochrane Database Syst Rev 2:CD002207. 356:j760.
13. Teesson M, Mills K, Ross J, Darke S, Williamson A, Havard A (2007): 33. McCowan C, Kidd B, Fahey T (2009): Factors associated with mortality
The impact of treatment on 3 years’ outcome for heroin dependence: in Scottish patients receiving methadone in primary care: Retrospec-
Findings from the Australian Treatment Outcome Study (ATOS). tive cohort study. BMJ 338:b2225.
Addiction 103:80–88. 34. Krantz MJ, Lewkowiez L, Hays H, Woodroffe MA, Robertson AD,
14. Clausen T, Anchersen K, Waal H (2008): Mortality prior to, during and Mehler PS (2002): Torsade de pointes associated with very-high-dose
after opioid maintenance treatment (OMT): A national, prospective methadone. Ann Intern Med 137:501–504.
cross-registry study. Drug Alcohol Depend 94:151–157. 35. Treece JM, Al Madani M, El Khoury G, Khraisha O, Martin JE,
15. Kimber J, Larney S, Hickman M, Randall D, Degenhardt L (2015): Baumrucker SJ, et al. (2018): Comprehensive review on methadone-
Mortality risk of opioid substitution therapy with methadone versus induced QT prolongation and torsades. J Pharmacol Pharmacother
buprenorphine: A retrospective cohort study. Lancet Psychiatry 9:66–75.
2:901–908. 36. Wedam EF, Bigelow GE, Johnson RE, Nuzzo PA, Haigney MCP (2007):
16. Carrieria P, Vilotitcha A, Nordmann S, Lions C, Michel L, Mora M, QT-interval effects of methadone, levomethadyl, and buprenorphine in
et al. (2017): Decrease in self-reported offences and incarceration a randomized trial. Arch Intern Med 167:2469–2475.
rates during methadone treatment: A comparison between patients 37. Strang J, Hall W, Hickman M, Bird SM (2010): Impact of supervision of
switching from buprenorphine to methadone and maintenance methadone consumption on deaths related to methadone overdose
treatment incident users (ANRS-Methaville trial). Int J Drug Policy (1993-2008): Analyses using OD4 index in England and Scotland. BMJ
39:86–91. 341:c4851.
17. Russolillo A, Moniruzzaman A, McCandless LC, Patterson M, 38. Bell J (2010): The global diversion of pharmaceutical drugs. Opiate
Somers JM (2018): Associations between methadone maintenance treatment and the diversion of pharmaceutical opiates: A clinician’s
treatment and crime: A 17-year longitudinal cohort study of Canadian perspective. Addiction 105:1531–1537.
provincial offenders. Addiction 113:656–667. 39. Bell J, Butler B, Lawrance A, Batey R, Salmelainen P (2009):
18. De Maeyer J, Vanderplasschen W, Camfield L, Vanheule S, Sabbe B, Comparing overdose mortality associated with methadone and
Broekaert E (2011): A good quality of life under the influence of buprenorphine treatment. Drug Alcohol Depend 104:73–77.
methadone: A qualitative study among opiate-dependent individuals. 40. Cleaver H (2007): Georgian drug misusers switch to Western heroin
Int J Nursing Stud 48:1244–1257. substitute. BMJ 334:821.

6 Biological Psychiatry - -, 2019; -:-–- www.sobp.org/journal

 Biological
Medication Treatment of Opioid Use Disorder Psychiatry

41. Degenhardt L, Larance B, Bell J, Winstock A, Lintzeris N, Ali R, et al. 56. Jones HE, O’Grady KE, Malfi D, Tuten M (2008): Methadone mainte-
(2009): Injection of OST in Australia following the introduction of a nance vs. methadone taper during pregnancy: Maternal and neonatal
mixed partial agonist-antagonist opioid medication. Med J Aust outcomes. Am J Addict 17:372–386.
191:161–165. 57. Jones HE, Kaltenbach K, Heil SH, Stine SM, Coyle MG, Arria AM, et al.
42. Hser Y-I, Saxon AJ, Huang D, Hasson A, Thomas C, Hillhouse M, et al. (2010): Neonatal abstinence syndrome after methadone or buprenor-
(2014): Treatment retention among patients randomized to buprenor- phine exposure. N Engl J Med 363:2320–2331.
phine/naloxone compared to methadone in a multi-site trial. Addiction 58. Jones HE, Chisolm MS, Jansson LM, Terplan M (2013): Naltrexone in the
109:79–87. treatment of opioid-dependent pregnant women: The case for a
43. Volavka J, Verebely K, Resnick R, Mule S (1978): Methadone dose, considered and measured approach to research. Addiction 108:233–247.
plasma level, and cross-tolerance to heroin in man. J Nerv Ment Dis 59. Rehm J, Gschwend P, Steffen T, Gutzwiller F, Dobler-Mikola A,
166:104–109. Uchtenhagen A (2001): Feasibility, safety, and efficacy of injectable
44. Teruya C, Schwartz RP, Mitchell SG, Hasson AL, Thomas C, heroin prescription for refractory opioid addicts: A follow-up study.
Buoncristiani SH, et al. (2014): Patient perspectives on buprenorphine/ Lancet 358:1417–1423.
naloxone: A qualitative study of retention during the starting treatment 60. Strang J, Groshkova T, Uchtenhagen A, van den brink W, Haasen C,
with agonist replacement therapies (START) study. J Psychoactive Schecter M, et al. (2015): Heroin on trial: Systematic review and meta-
Drugs 46:412–426. analysis of randomised trials of diamorphine prescribing as treatment
45. Mattick RP, Ali R, White J, O’Brien S, Wolk S, Danz C (2003): Bupre- for refractory heroin addiction. Br J Psychiatry 207:5–14.
norphine versus methadone maintenance therapy: A randomized 61. Oviedo-Joekes E, Guh D, Brissette S, Marsh DC, Nosyk B, Krausz M,
double-blind trial with 405 opioid-dependent patients. Addiction et al. (2010): Double-blind injectable hydromorphone versus diac-
98:441–452. etylmorphine for the treatment of opioid dependence: A pilot study.
46. Hser Y-I, Evans E, Huang D, Weiss R, Saxon A, Carroll KM, et al. J Subst Abuse Treat 38:408–411.
(2016): Long-term outcomes after randomization to buprenorphine/ 62. Bell J, Belackova V, Lintzeris N (2018): Supervised injectable opioid
naloxone versus methadone in a multi-site trial. Addiction 111:695– treatment for the management of opioid dependence. Drugs 78:1339–
705. 1352.
47. Lofwall MR, Walsh SL, Nunes EV, Bailey GL, Sigmon SC, 63. Strang J, Groshkova T, Metrebian N (2012): Recent Evidence and
Kampman KM, et al. (2018): Weekly and monthly subcutaneous Current Practices of Supervised Injectable Heroin Treatment in Europe
buprenorphine depot formulations vs daily sublingual buprenorphine and Beyond. EMCDDA Monograph. Luxembourg: Publications Office
with naloxone for treatment of opioid use disorder a randomized of the European Union.
clinical trial. JAMA Intern Med 178:764–773. 64. McDonald R, Strang J (2016): Are take-home naloxone programmes
48. Haight BR, Learned SM, Laffont CM, Fudala PJ, Zhao Y, Garofalo AS, effective? Systematic review utilizing application of the Bradford Hill
et al. (2019): Efficacy and safety of a monthly buprenorphine depot criteria. Addiction 111:1177–1187.
injection for opioid use disorder: A multicentre, randomised, double- 65. Olsen A, McDonald D, Lenton S, Dietze P (2014): Evidence for take-
blind, placebo-controlled, phase 3 trial. Lancet 393:778–790. home naloxone programs: A Bradford Hill analysis. Drug Alcohol Rev
49. Rosenthal R, Ling W, Casadonte P, Vocci F, Bailey GL, Kampman K, 33:49–50.
et al. (2013): Buprenorphine implants for treatment of opioid depen- 66. McDonald R, Lorch U, Woodward J, Bosse B, Johnson H, Mundin G,
dence: Randomized comparison to placebo and sublingual bupre- et al. (2017): Pharmacokinetics of concentrated naloxone nasal spray
norphine/naloxone. Addiction 108:2141–2149. for opioid overdose reversal: Phase-I healthy volunteer study. Addic-
50. Lee JD, Nunes EV, Novo P, Bachrach K, Bailey GL, Bhatt S, et al. tion 113:484–493.
(2018): Comparative effectiveness of extended-release naltrexone 67. Neale J, Strang J (2015): Naloxone–does over-antagonism matter?
versus buprenorphine-naloxone for opioid relapse prevention (X:BOT): Evidence of iatrogenic harm after emergency treatment of heroin/
A multicentre, open-label, randomised controlled trial. Lancet opioid overdose. Addiction 110:1644–1652.
391:309–318. 68. Kakko J, Gronbladh L, Svanborg KD, von Wachenfeldt J, Ruck C,
51. Tanum L, Solli KK, Latif ZE, Benth JS, Opheim A, Sharma-Haase K, Rawlings B, et al. (2007): A stepped care strategy using buprenorphine
et al. (2017): Effectiveness of injectable extended-release naltrexone and methadone versus conventional methadone maintenance in her-
vs daily buprenorphine-naloxone for opioid dependence: A random- oin dependence: A randomized controlled trial. Am J Psychiatry
ized clinical noninferiority trial. JAMA Psychiatry 74:1197–1205. 164:797–803.
52. Saucier R, Wolfe D, Dasgupta N (2018): Review of case narratives from 69. Hedegaard H, Warner M, Miniño AM (2017): Drug Overdose Deaths in
fatal overdoses associated with injectable naltrexone for opioid the United States, 1999–2016. NCHS Data Brief, no. 294. Hyattsville,
dependence. Drug Saf 41:981–988. MD: National Center for Health Statistics.
53. Hadland SE, Bagley SM, Rodean J, Silverstein M, Levy S, 70. Wakeman SE, Barnett ML (2018): Primary care and the opioid-overdose
Larochelle MR, et al. (2018): Receipt of timely addiction treatment and crisis - Buprenorphine myths and realities. N Engl J Med 379:1–4.
association of early medication treatment with retention in care among 71. Food and Drug Administration (2019): FDA approves first generic
youths with opioid use disorder. JAMA Pediatr 172:1029–1037. naloxone nasal spray to treat opioid overdose. FDA News
54. Kennare R, Heard A, Chan A (2005): Substance use during pregnancy: Release April 19. Available at: https://www.fda.gov/news-events/
risk factors and obstetric and perinatal outcomes in South Australia. press-announcements/fda-approves-first-generic-naloxone-nasal-
Aust N Z J Obstet Gynaecol 45:220–225. spray-treat-opioid-overdose. Accessed April 21, 2019.
55. Alto WA, O’Connor AB (2011): Management of women treated 72. Samet JH, Botticelli M, Bharel M (2018): Methadone in primary care -
with buprenorphine during pregnancy. Am J Obstet Gynecol One small step for Congress, one giant leap for addiction treatment.
205:302–308. N Engl J Med 379:7–8.

Biological Psychiatry - -, 2019; -:-–- www.sobp.org/journal 7