Download Frequency, Indications, Outcomes, and Predictive Factors of Opioid

632
Journal of Pain and Symptom Management
Vol. 37 No. 4 April 2009
Original Article
Frequency, Indications, Outcomes,
and Predictive Factors of Opioid Switching
in an Acute Palliative Care Unit
Sebastiano Mercadante, MD, Patrizia Ferrera, MD, Patrizia Villari, MD,
Alessandra Casuccio, BS, Giuseppe Intravaia, RN, and Salvatore Mangione, MD
Pain Relief and Palliative Care Unit (S.M., P.F., P.V., G.I.), La Maddalena Cancer Center;
and Palliative Medicine (S.Me., S.Ma.), Department of Clinical Neuroscience (A.C.), and Department
of Anesthesiology and Intensive Care (S.Ma.), University of Palermo, Palermo, Italy
Abstract
The aim of this study was to prospectively evaluate the frequency, indications, outcomes, and
predictive factors associated with opioid switching, using a protocol that had been clinically
applied and viewed as effective for many years. A prospective study was carried out on
a cohort of consecutive cancer patients who were receiving opioids but had an unacceptable
balance between analgesia and adverse effects, despite symptomatic treatment of side effects.
The initial conversion ratio between opioids and routes was as follows (mg/day): oral
morphine 100 ¼ intravenous morphine 33 ¼ transdermal fentanyl 1 ¼ intravenous
fentanyl 1 ¼ oral methadone 20 ¼ intravenous methadone 16 ¼ oral oxycodone
70 ¼ transdermal buprenorphine 1.3. The switch was assisted by opioids used as needed,
and doses were changed after the initial conversion according to clinical response in an acute
care setting. Intensity of pain and symptoms associated with opioid therapy were recorded. A
distress score (DS) was calculated as a sum of symptom intensity. A switch was considered
successful when the intensity of pain and/or DS, or the principal symptom necessitating the
switch, decreased to at least 33% of the value recorded before switching. One hundred
eighteen patients underwent opioid substitutions. The indications for opioid switching were
uncontrolled pain and adverse effects (50.8%), adverse effects (28.8%), uncontrolled pain
(15.2%), and convenience (4.2%). Overall, 103 substitutions were successful. Ninety-six
substitutions were successful after the first switching, and a further substitution was
successful in seven patients who did not respond to the first switch. The mean time to achieve
dose stabilization after switching was 3.2 days. The presence of both poor pain control and
adverse effects was related to unsuccessful switching (P < 0.004). No relationship was
identified between unsuccessful switching and the opioid dose, opioid sequence, pain
mechanism, or use of adjuvant medications. Opioid switching was an effective method to
improve the balance between analgesia and adverse effects in more than 80% of cancer
patients with a poor response to an opioid. The presence of both poor pain relief and adverse
effects is a negative factor for switching prognosis, whereas renal failure is not. J Pain
Symptom Manage 2009;37:632e641. Ó 2009 U.S. Cancer Pain Relief Committee.
Published by Elsevier Inc. All rights reserved.
Address correspondence to: Sebastiano Mercadante,
MD, Pain Relief and Palliative Care Unit, La
Maddalena Cancer Center, Via San Lorenzo 312,
Ó 2009 U.S. Cancer Pain Relief Committee
Published by Elsevier Inc. All rights reserved.
90146 Palermo, Italy. E-mail: terapiadeldolore@
la-maddalena.it
Accepted for publication: December 13, 2007.
0885-3924/09/$esee front matter
doi:10.1016/j.jpainsymman.2007.12.024
Vol. 37 No. 4 April 2009
Opioid Switching in an Acute Palliative Care Unit
633
Key Words
Opioid switching, cancer pain, palliative care
Introduction
Opioids are the mainstay therapy for moderate to severe cancer pain, and most patients respond favorably to opioid therapy. However, in
some patients, the response may be characterized by adverse effects severe enough to compromise benefit. Previous experience has
shown that failure to respond to one opioid
does not mean failure to respond to all opioids, and opioid switching may lead to better
pain control and/or a decrease in the intensity
of disabling adverse effects.1e4
Although the keystone to the rationale behind opioid substitution is incomplete crosstolerance, the exact reason why opioid substitution is successful remains unclear. In some
patients, poorly responsive pain may arise because analgesic tolerance develops more completely than tolerance to adverse effects. As
a consequence, escalating the dose may reach
a level at which the adverse effects become predominant. Thus, the benefit of a switch from
one opioid to another opioid could depend
on cross-tolerance to the analgesic effects
being less than cross-tolerance to the adverse
effects. The clinical challenge is that it is
impossible to know in advance if the balance
between analgesia and adverse effects will
improve after opioid substitution. In addition,
the optimal dose of the alternative opioid
chosen to initiate a switch is uncertain, as it
will depend on a series of factors, including
individual response, pain mechanism, pharmacogenetics, and degree of cross-tolerance.5
The need to change opioid may occur in the
following clinical conditions: 1) pain is controlled but the patient experiences intolerable
adverse effects; 2) pain is not adequately controlled, but it is impossible to increase the
dose because of the concomitant presence of
adverse effects; 3) pain is not adequately controlled by rapidly increasing the dose of opioids, although the drug does not produce
adverse effects. This last point remains controversial, as further increasing doses could potentially yield appropriate analgesia. However,
rapid opioid escalation has been recognized
as a negative factor for the clinical response;6
4) finally, some patients are switched for convenience or patient preference, despite reporting an acceptable balance between analgesia
and adverse effects.
Protocols for opioid substitution have been
empirically proposed for many years, and
each has a logical rationale.4 The aim of this
study was to prospectively evaluate the frequency, indications, and responses to opioid
switching in patients with a poor response to
an opioid, who were admitted to an acute palliative care unit. We sought to evaluate the outcomes associated with a protocol that has been
clinically used and viewed as effective for many
years, and to detect some possible factors influencing outcomes.
Patients and Methods
A prospective study was carried out in a sample of consecutive patients admitted to an
acute palliative care unit over a period of two
years. Informed consent was obtained from patients or from relatives in cases of cognitive failure, and institutional approval was obtained to
conduct the study. Patients with advanced cancer who were receiving opioids and required
a switch in opioid therapy were included.
There were four categories of patients:
1) Patients with unacceptable adverse effects
despite good pain control. To be eligible,
these patients had to present at least one
relevant symptom, such as drowsiness,
confusion, or myoclonus, with intensity
$ 2 on a scale from 0 to 3 (see below),
or other symptoms (constipation, dry
mouth) rated as severe.
2) Patients with a poor analgesic response
despite having their dose doubled in
one week.
3) Patients with both poor pain control and
adverse effects.
4) Patients who were switched because of
preference and/or convenience (e.g.,
dysphagia limiting the oral route), or
other reasons.
The switch took place in an acute palliative
care unit. The choice of opioid sequence
634
Mercadante et al.
was based on previous drugs already tried and
targeted individual needs. The intravenous
route or transdermal drugs were used when
the oral route was unavailable. Rescue doses
were used, intravenously or orally, using the
equivalent of 1/6 of the daily dose, to support
the switching. The same opioid was mainly
used as a rescue dose, unless the route was impracticable (e.g., vomiting), or if the opioid
was unavailable for the alternative route of
administration.
The initial conversion ratios used among opioids and routes of administration were based on
department policy; known drug availability for
oral, transdermal, and intravenous routes of
administration; and previous experiences.7e14
Ratios were expressed as oral morphine equivalents (mg/day): oral morphine 100 ¼ intravenous morphine 33 ¼ transdermal fentanyl
1 ¼ intravenous fentanyl 1 ¼ oral methadone
20 ¼ intravenous methadone 16 ¼ oral oxycodone 70 ¼ transdermal buprenorphine 1.3. After the initial dose, the subsequent doses were
flexible and were changed according to patients’ needs in an attempt to find the best
balance between pain and opioid-related
symptoms.
Adjuvant drugs, previously administered to
control symptoms due to illness or treatment,
were continued at the same doses during the
switching. Non-opioid analgesics were also
continued if previously administered, at the
same doses. No patient received anticancer
therapy during the course of the study. All patients were strictly monitored by a team of doctors and nurses experienced in palliative care.
In our unit, pain and symptoms are recorded
by nurses four times a day, as a routine, and
four rounds or more are performed by the
team to explore clinical changes. A physician
on duty and a team component on call were
available. Daily doses were changed, according
to the amount of drugs consumed as rescue
doses during the previous day and clinical
judgment, to achieve the best clinical balance
between analgesia and adverse effects.
The following data were recorded:
1) age, gender, primary cancer and known
metastases, pain causes and mechanisms,
performance status;
2) pain syndromes determined on the basis
of clinical history, anatomical site of
3)
4)
5)
6)
Vol. 37 No. 4 April 2009
primary tumor and known metastases,
physical examination, and available
investigations;
opioid doses before the switch;
daily opioid doses at the time of
stabilization;
symptoms associated with opioid therapy
or commonly present in advanced cancer
patients, such as nausea and vomiting,
drowsiness, confusion, constipation, dry
mouth, myoclonus, and sweating. Symptom intensity was assessed using a scale
from 0 to 3, corresponding to a verbal scale
(not at all, slight, a lot, awful). A distress
score (DS) also was calculated as a sum of
symptom intensity. Although never validated, this score has been previously used
in studies for determining the ‘‘weight’’
of adverse effects. The aim of using a sum
of intensities is justified by the high variability of symptom intensity in individual
patients. This score can determine the
general improvement of symptoms when
the evaluation of the changes in intensity
of a single symptom, which differ with
each patient, make a global evaluation,
and statistics for a group of patients, practically impossible. Moreover, it is not rare
to switch a patient for more than one symptom. An important decrease of the principal symptom that necessitated the
switching is another parameter to take
into consideration. Thus, the parameters
used to define a successful switch included
both DS and change in the principal symptom. Symptoms were assessed by the
patient, whenever possible; when patients
had severe cognitive failure, a proxy
evaluation was used.
pain intensity was measured using the patient’s self-report on a numerical 0e10
scale.
Opioid, dose, route, pain intensity, and
symptoms were recorded before switching
(T0) and at the end of switching (time of stabilization) (see below). In addition, the number
of changes in the planned daily doses, the time
to reach a stable daily dose (defined as the first
of two consecutive days requiring no more
than two rescue doses [time of stabilization]),
and time from admission to hospital discharge
(T-discharge), were recorded.
Vol. 37 No. 4 April 2009
Opioid Switching in an Acute Palliative Care Unit
Data Analysis
Clinical judgment, as reported in some articles,15 was not considered a meaningful
method to evaluate outcomes in this study.
For this reason, other parameters previously
used in similar studies were chosen.9,12,14 A
switch was considered successful when the intensity of pain and/or DS, or score given to
the principal symptom that necessitated the
switch, decreased at least 33% below the baseline value recorded before switching within
seven days of the switch. This time was allowed
if a positive trend was observed; otherwise, the
switch was considered unsuccessful after four
days to avoid further suffering for the patients.
A decrease of 33% has been found to be a clinically relevant effect,16 and has been increasingly used to assess rapid clinical changes
after specific interventions.12,14
Risk factors examined as potentially contributing to unsuccessful switching were pain type,
cause of switching, opioid doses, use of adjuvant
drugs, renal failure, age, gender, and primary
tumor. A serum creatinine level > 1.5 times
the upper limit of normal for our laboratory
was considered as abnormal renal function.
Frequency analysis was performed using Chisquared tests. The paired Wilcoxon signedrank test was used to compare pain intensity
scores and symptom intensity scores across
time periods. One-way analysis of variance
(ANOVA) and Mann-Whitney U statistic tests
were used for parametric and nonparametric
analysis, respectively. All P values were twosided and P values less than 0.05 were considered to indicate statistical significance.
Results
Three hundred forty-five patients were admitted for pain control during a period of
two years. One hundred eighteen patients underwent opioid switching (34.2%). No relationships between the need for opioid
switching and pain type, type of opioid, use
of adjuvant drugs, biochemical parameters,
or opioid doses were found. Patient participation in rating symptoms was high (94%); the
remaining evaluations were provided by
proxies.
Ninety-six substitutions (81%) were successful after the first switch, and a further substitution was successful in seven patients who did
635
not respond to the first switch. Overall, 103
substitutions (87%) were effective, as indicated
by a decrease of at least 33% in pain intensity
and/or of DS (or the leading symptom)
compared with the baseline value recorded before switching, within a reasonable period of
time (commonly 4e7 days). A second attempt
to substitute opioid was ineffective in the remaining 15 patients (see below). Data regarding the most frequent sequences of opioid
switching are listed in Table 1. Original tables
regarding all the sequences of opioid switching are available in the online version of this
journal.
Causes of Opioid Switching
The presence of both uncontrolled pain and
adverse effects was the most frequent indication to switch (60 substitutions, 50.8%). The
presence of uncontrollable adverse effects (at
least one symptom with an intensity scored as
‘‘a lot,’’ that is, 2 on a scale from 0 to 3) with
adequate pain control was the reason for
switching in 34 substitutions (28.8%). Uncontrolled pain despite rapid opioid escalation
was the cause of switching for 18 substitutions
(15.2%). Convenience was the main cause of
switching in five substitutions (4.2%).
Opioid Sequences
Most patients were switched from fentanyl
(53) or morphine (44), and the majority of patients were switched to methadone (60 patients). No differences among the different
opioid sequences in determining a successful
switch were observed (P ¼ 0.243).
The mean time to achieve dose stabilization
after switching was 3.2 days (95% confidence
interval [CI] 2.7e3.6). Patients switched to
buprenorphine, from morphine to fentanyl,
from methadone to other opioids, and from
and to oxycodone had a shorter time to
achieve dose stabilization in comparison with
other opioid sequences (P s < 0.001). Globally,
switching to transdermal drugs, including
fentanyl and buprenorphine, required a shorter time for achieving dose stabilization in comparison with the other sequences (P ¼ 0.0015).
Time for stabilization was significantly longer
in patients who switched for concomitant
adverse effects and poor pain control than in
patients who switched for adverse effects
(P ¼ 0.004). Changes in doses were more
636
Table 1
Principal Sequences of Opioid Switching
No. of
Patients
15
20
11
9
51.5 (44e59)
11 Men
8 Women
60 (54e65)
9 Men
6 Women
56 (51e60)
16 Men
4 Women
62 (57e67)
7 Men
4 Women
64 (55e73)
6 Men
3 Women
Reason
for
Switch
P/AE ¼ 14 4.3 TFE to
AE ¼ 3
ORME
P¼2
P/AE ¼ 8 2.4 TFE to
P¼7
ORMO
P/AE ¼ 10 389 ORMO
AE ¼ 6
to ORME
P¼4
P/AE ¼ 4 245 ORMO
AE ¼ 4
to TFE
C¼3
P/AE ¼ 6 53 ORME
AE ¼ 3
to other
opioids
T0 Distress
Scoreb
T Stabilization
Distress
Score
T0 Pain
Intensityb
T Stabilization
Pain Intensity
Number
of Dose
Changes
Days to
Achieve
Stabilization
Days to
Discharge
68 ORME
(17.1)
4.9 (4.3e5.6) 2.3 (1.7e2.8)
6.2 (5.2e7.2) 2.0 (1.4e2.6)
3.1 (2.3e3.9)
4.1 (3.3e4.9) 5.2 (4.5e5.9)
455 ORMO
(24.5)
6.9
(6.1e7.6)
3.9 (2.9e4.8)
3.5 (0.8e6.3) 1.9 (1.2e2.7)
2.6 (1.5e3.7)
4.1 (2.2e5.9) 5.9 (4.3e7.6)
56 ORME
(5.3)
6.9
(5e8.2)
3.5 (3.0e4.6)
2.7 (2.0e3.2) 2.5 (2.1e2.9)
3.0 (2e4)
4.3 (3.1e5.4) 6.0 (4.8e7.1)
2.5 TFE
(10.3)
5.7
(4.8e6.9)
2.7 (1.4e4.5)
2.5 (1.6e3.3) 2.2 (1.3e3.1)
0.4 (0.1 to 0.8) 1.8 (0.5e3.1) 4.8 (2.9e6.7)
(33)
6
3.7 (2.9e4.4)
5.1 (2.9e7.4) 2.3 (1.3e3.3)
0.3 (e0.1e0.7)
Mercadante et al.
19
Age
and
Gender
T0 Dose
T Stabilization
(mg/day)
Dose (mg/day)
and route
of second
of first opioid
opioid at
at the time
stabilization
of switch
(% of change
to 2nd
from initial
opioid
dosea)
2.6 (0.5e4.6) 5.1 (1.8e8.4)
(4.8e7.1)
Vol. 37 No. 4 April 2009
All data are expressed as mean (95% CI).
TFE ¼ transdermal fentanyl; ORME ¼ oral methadone; ORMO ¼ oral morphine; P ¼ uncontrolled pain; AE ¼ adverse effects; P/AE ¼ uncontrolled pain and adverse effects; C ¼ convenience.
Doses of drugs administered intravenously were converted to oral and transdermal routes (see conversion ratios in Patients and Methods section).
a
Change from the initial dose of the second opioid.
b
Distress score and pain intensity were measured on 0e10 scales.
Vol. 37 No. 4 April 2009
Opioid Switching in an Acute Palliative Care Unit
frequently observed in patients switched to
methadone, and were less frequently reported
in patients switched to transdermal fentanyl
and buprenorphine (P ¼ 0.005). Switching to
buprenorphine was more frequently performed for adverse effects (P < 0.0005), mainly
gastrointestinal.
No relationship between the starting opioid
dose and dose at stabilization after switching
(P ¼ 0.810), or time to achieve stabilization
(P ¼ 0.064) was found. Hospital discharge was
proportional to the time needed to achieve
dose stabilization (P < 0.0005).
The presence of both pain and adverse effects as indications for opioid switching was
more frequently observed in patients who
were switched from fentanyl and morphine
to methadone (P ¼ 0.005). No relationship between the reasons to switch and the opioid
dose was found.
Unsuccessful Switching
In 22 patients, opioid switching failed to
provide a clinical benefit at first instance. A
second-line substitution was performed in 11
patients. Seven of these patients benefited
from a second-line switch to other opioids.
One of these, who had been switched from
oxycodone, which became unavailable in pharmacies, to transdermal fentanyl (convenience), was switched back to oxycodone
when it again became available, because of
the development of sweating and worsening
of pain intensity after starting transdermal
fentanyl.
Four patients did not benefit from a secondline switch, although one seemed to improve,
but relatives were not satisfied. Three of them
were subsequently treated intrathecally. Eleven
(42%) of these substitutions were from fentanyl
to methadone (which was 37% of the total substitutions from fentanyl to methadone) and
were more frequent than the other sequences
(P < 0.0005). Twenty-three percent of substitutions were from morphine to methadone
(23% of the total substitutions from morphine
to methadone), and 12% were from fentanyl
to morphine (18% of the total substitutions
from fentanyl to morphine). The presence of
both poor pain control and adverse effects was
more often found as the reason for substitution
in patients who were unsuccessfully switched
(P ¼ 0.004). Other factors, including gender
637
(P ¼ 0.602) and age (P > 0.05), did not influence the switching outcome. Similarly, no relationship was found between failure of the
switch and the opioid dose (P > 0.05), type of
opioid (P ¼ 0.337), or pain type (P ¼ 0.504).
Renal failure was present in 16% of patients,
but this finding was not significantly related to
the failed switching, as creatinine values did
not influence outcome (P > 0.05). Finally, no
differences in the previous use of adjuvant
drugs were found in the different opioid sequences and no relation with the use of a particular adjuvant and failed switching was observed
(P ¼ 0.613).
Globally, eight patients subsequently underwent an intrathecal treatment, which was effective in most cases. One of these patients was not
able to tolerate even minimal doses of intrathecal
opioids (0.5 mg of intrathecal morphine), and
received just a local anesthetic infusion for his
pain. Of the 22 patients who were considered failures, eight patients (36%) were particularly
advanced and had a short survival, requiring, in
some cases, terminal sedation.
Discussion
The aim of this study was to evaluate the frequency and indication for opioid switching.
We also sought to evaluate the outcome of an
approach used for many years and some possible factors influencing the outcome in a typical
clinical scenario, similar to that of an intensive
pain relief and palliative care unit, where physicians typically face very complex situations.
In this context, a flexible approach, starting
with a priming dose, followed by change in
doses according to the clinical situation has
been used for many years. This method has
been shown to decrease the time for achieving
the desired stabilization but also requires strict
surveillance and intensive monitoring.
Frequency and Causes of Opioid Switching
The frequency of opioid switching reported
in this study was relatively high. In a prospective
cohort study of 412 palliative care patients, only
49 opioid substitutions (11.9%) were recorded.17 However, only about 25% of patients
were treated in hospices or palliative care units,
and switching occurred more often in inpatients than in outpatients. Thus, the observed
differences can be attributed to the different
638
Mercadante et al.
settings and more stringent admission criteria,
as demonstrated by the high opioid basal doses
recorded in this study. In another study of 186
subjects receiving morphine, 48 patients
(25%) failed to obtain a satisfactory analgesic
response and required opioid switching.15 Insufficient efficacy was the main reason to switch
in the former study,17 whereas in the latter, the
distinction of indications for switching was unclear.15 A high rate of opioid substitution was reported in other settings, mostly because of
adverse effects.18,19 In a tertiary unit for palliative care, of 200 successive admissions, 80 patients underwent opioid substitution, most of
them being switched for central adverse effects,20 and 80 of 100 patients underwent
changes in drug, route or both before discharge
or death, usually for poor pain control.21 In the
present study, the majority of opioid substitutions were due to both uncontrolled pain and
adverse effects (about 80%), reflecting the complexity of admitted patients.
The most frequent opioid sequence was
from fentanyl to other opioids, reflecting the
high use of this preparation in Italy, and which
is mainly prescribed by oncologists;22 methadone was more often used as the first option
for opioid switching, reflecting local experience with this drug. The presence of both
pain and adverse effects was more frequently
observed as an indication before switch to
methadone. Of interest, preference was given
to transdermal drugs or the intravenous route
to accomplish opioid switching in some particular conditions, where gastrointestinal effects
may preclude the oral route. Switching to
transdermal drugs was usually done because
of adverse effects and/or convenience, clinical
situations where an acceptable pain control exists, and conversion ratios are more reliable.
This was expected, given that these modalities
are not generally used for patients who possibly need rapid changes in doses, thus also explaining timing to achieve stabilization and
the relative dose stability.
In recent years, some studies have attempted
to find possible factors involved in opioid response. In a retrospective analysis of risk factors
for opioid switching, the need for opioid
switching was not influenced by age, gender,
or the routine use of adjuvants or co-analgesics,
except corticosteroids.19 In a retrospective
analysis, older patients, high platelet count,
Vol. 37 No. 4 April 2009
and high white cell count were found to predict
morphine intolerance requiring a switch to another opioid, and renal function was not significantly correlated to the need to switch,15
probably because of the variable time for blood
sampling. The concomitant use of antiemetics
(5HT3 antagonists), b-blockers and proton
pump inhibitors; tumor of the lower gastrointestinal tract; and recent chemotherapy were
considered risk factors for opioid switching,23
although no clear explanation was provided
for these findings. In this study, no biochemical
markers were identified in patients who required opioid substitution. Similarly, pain
type, adjuvants, opioid dose and type, age, gender, the use of corticosteroids, and other variables did not influence the need to switch.
Some differences could be attributed to the
heterogeneity of palliative care units, their admission policies, and therapeutic protocols.
Outcome
In this study, we tried to use objective and
reproducible data in clinical assessment. A
change of 33% in symptom intensity may validly assess significant clinical benefit.16 Moreover, given the diversity in opioid adverse
effect patterns, we used DS, which represents
the global ‘‘weight’’ of adverse effects. This
metric has been feasibly used in previous studies to evaluate clinical changes after opioid
switching. In a palliative care population, it
would be impractical to select patients with
a specific symptom prevalence to be switched
to another opioid.
Opioid switching was generally effective and
no differences in opioid sequence, starting
opioid dose, dose at stabilization after switching, or time to achieve stabilization were
found. In most patients, opioid switching, independently from the opioid sequence,
proved to be effective, given that 81% and
87% of opioid substitutions were successful after one or two switchings, respectively. The
mean time to achieve dose stabilization after
switching was relatively short (about three
days on average), probably because of careful
monitoring. As expected, hospital discharge
was proportional to the time needed to
achieve dose stabilization, although in some
cases other reasons may prolong hospital stay.
Of interest, the pre-switching opioid doses
did not influence the time to achieve
Vol. 37 No. 4 April 2009
Opioid Switching in an Acute Palliative Care Unit
stabilization or the final doses of the second
opioid. However, the numbers may have been
too small to detect a relationship.
Similarly, pain type did not influence outcome, or the final doses of methadone, confirming previous findings that failed to show
a difference in the ratios of patients with neuropathic or non-neuropathic pain syndromes.24
However, these findings should not be considered definitive, because of the low number of
patients who failed switching.
Switching to methadone may require a more
careful approach. Changes in doses were more
frequently reported in patients switched to
methadone (about three dose changes),
whereas they were less frequently reported in
patients switched to transdermal drugs, fentanyl and buprenorphine. This was probably because of the reasons for switching to
transdermal drugs, mainly the presence of gastrointestinal adverse effects or convenience,
conditions which do not require relative dose
increases of the second opioid for uncontrolled pain. This can also explain the shorter
time for achieving dose stabilization observed
when switching to transdermal drugs.
The initial conversion ratios, based on clinical studies and personal experience,7e14 were
changed based on patients’ titration and opioid dose consumption, to achieve the best balance between pain intensity and adverse
effects. This resulted in large variability of final
doses in all the opioid sequences, so that the
final conversion ratio for dose stabilization
was very likely to be at least marginally different. This finding was expected.
The low number of patients for each sequence does not allow comparison. The variables involved, including drugs used, level of
tolerance, and genetics, are so numerous that
any attempt to draw any scientific conclusion
would be speculative. This observation confirms the importance of the individual response, so that the initial dose should be
considered just as an approximate starting
point to be assessed and evaluated over time,
with the appropriate dose changes, rather
than the definitive result of a mere calculation.
Of interest, some patients switched from
methadone required larger increases in doses
of the second opioid (mean 33%), confirming
previous data from a small series, where the final doses of fentanyl were found to be 30%
639
higher than the initial dose.12 Worsening pain
with severe adverse effects after switching from
methadone to another opioid has been reported.25 This observation is not substantiated
by general experience during opioid switching,
even when switching from methadone to another opioid, which is less frequently used.1,2
When very high doses of opioids are administered, it is impossible to quantify any approximate dose conversion ratio for methadone
(one study reported the effective dose as 1%
of the calculated conversion ratio when switching from extraordinary doses of parenteral
morphine, specifically 21,600 mg/day)26). The
reasons why patients switched from methadone
require substantially higher doses of the second
opioids, other than wide individual responses,
may reside in the possible anti-N-methyl-D-aspartate properties of methadone.27 This does not
mean that patients will be unresponsive to
other opioids.
Predictive Factors
Renal failure may render the treatment
more at risk to develop adverse effects, due
in some cases to the occurrence of metabolite
accumulation. In this study, renal failure was
not found to be a negative prognostic factor.
Although some patients who failed opioid
switching had renal failure, some others were
successful. For example, 11 patients were successfully switched from morphine and other
opioids to methadone or fentanyl. Unfortunately, in a previous study on prognostic factors for opioid switching, patients with
abnormal renal function were excluded.15
Our experience suggests that patients with opioid adverse effects related to renal impairment
could benefit from opioid switching. This observation confirms the data from previous
studies reporting that patients with renal impairment had a high success rate from opioid
switching.18e20
Few other studies have assessed factors that
may influence outcome. In a previous survey,
myoclonus and cognitive failure were the indications with the highest success rate.19,20 The
presence of a neuropathic pain type did not influence the outcome, and the use of drugs presumably more effective for neuropathic pain,
such as methadone, did not change the chances
of success or the final doses, replaying previous
retrospective data that failed to demonstrate
640
Mercadante et al.
a different efficacy of methadone in patients
with neuropathic pain.24 Also, the use of drugs
commonly prescribed for neuropathic pain,
such as antidepressants and anticonvulsants,
or corticosteroids did not influence the need
to switch and the outcome. The only factor
that had a significant relationship with unsuccessful switching was the presence of both
poor pain control and adverse effects. However,
as mentioned earlier, this was also the principal
indication to switch, and a relevant percentage
of patients still had a benefit.
Conclusion
The frequency of opioid switching identified
in the survey may be relevant in a tertiary unit,
with a selected population. Concomitant presence of poor pain control and adverse effects
is the most frequent indication for opioid
switching. This indication, even though associated with fewer changes of success, may still
benefit from opioid switching in a substantial
percentage of patients. No specific parameters
were found to be associated with the need for
opioid switching. The initial conversion ratios
between opioids and routes used in the unit,
followed by opioid dose changes based on
a careful monitoring of the clinical events, allowed the achievement of a good balance between analgesia and adverse effects in 87% of
patients in a relatively short time. Failure
more commonly occurred when patients were
close to death, when other factors may probably play a negative role.
Although an initial conversion ratio was the
option, opioid conversion should not be
a mere mathematical calculation, but part of
a more comprehensive assessment of opioid
therapy, evaluating the underlying clinical situation, pain and adverse effect intensity, comorbidity, and concomitant drugs, and excluding
any possible pharmacokinetic factors that
could limit the effectiveness of a certain
drug.4 More prudent guidelines must be applied in less intensively monitored environments, particularly when patients are
switched from high doses of opioids. Although
the conversion ratios could be higher, time to
reach stabilization could be consequently slower. Thus, it would be important to select complex patients at high risk to be admitted to
specialized centers.
Vol. 37 No. 4 April 2009
There are limitations to this study, including
lack of a comparison group and blinding, and
the setting of acute palliative care, which may
not be easily reproducible, along with provider
strict surveillance and team expertise. Moreover, a rigid protocol was excluded to allow
the necessary flexibility to obtain the best clinical balance between analgesia and adverse effects in patients with difficult conditions.
Finally, a percentage of patients were much advanced and required terminal sedation, generally because of the development of delirium,
which possibly was not necessarily associated
to the opioid use. Future research should include controlled studies with a large number
of patients presenting a similar pattern of adverse effects, to be switched to a single alternative opioid or to an alternative route by using
the same opioid, or with the aim of assessing
different switching sequences.
References
1. Bruera E, Pereira J, Watanabe S, et al. Opioid rotation in patients with cancer pain. A retrospective
comparison of dose ratios between methadone, hydromorphone, and morphine. Cancer 1996;78:
852e857.
2. Lawlor P, Turner K, Hanson J, Bruera E. Dose
ratio between morphine and methadone in patients
with cancer pain: a retrospective study. Cancer 1998;
82:1167e1173.
3. Ripamonti C, Groff L, Brunelli C, et al. Switching from morphine to oral methadone in treating
cancer pain: what is the equianalgesic dose ratio?
J Clin Oncol 1998;16:3216e3221.
4. Mercadante S, Bruera E. Opioid switching: a systematic and critical review. Cancer Treat Rev 2006;
32:304e315.
5. Pasternak GW. Multiple opiate receptors: deja
vu all over again. Neuropharmacology 2004;
47(Suppl 1):312e323.
6. Mercadante S, Portenoy RK. Opioid poorly responsive cancer pain. Part 1. Clinical considerations. J Pain Symptom Manage 2001;21:144e150.
7. Donner B, Zenz M, Tryba M, Strumpf M. Direct
conversion from oral morphine to transdermal fentanyl: a multicenter study in patients with cancer
pain. Pain 1996;64:527e534.
8. Ahmedzai S, Brooks D, TTS-Fentanyl Comparative Trial Group. Transdermal fentanyl versus sustained release oral morphine in cancer pain:
preference, efficacy, and quality of life. J Pain Symptom Manage 1997;13:254e261.
Vol. 37 No. 4 April 2009
Opioid Switching in an Acute Palliative Care Unit
9. Mercadante S, Casuccio A, Calderone L. Rapid
switching from morphine to methadone in cancer
patients with poor response to morphine. J Clin
Oncol 1999;17:3307e3312.
10. Bruera E, Belzile M, Pituskin E, et al. Randomized, double-blind, cross-over trial comparing safety
and efficacy of oral controlled-release oxycodone
with controlled-release morphine in patients with
cancer pain. J Clin Oncol 1998;16:3222e3229.
11. Benitez-Rosario MA, Ferla M, Salinas-Martin A,
Martinez-Castillo LP, Martin-Ortega JJ. Opioid
switching from transdermal fentanyl to oral methadone in patients with cancer pain. Cancer 2004;
101:2866e2873.
12. Mercadante S, Ferrera P, Villari P, Casuccio A.
Rapid switching between transdermal fentanyl and
methadone in cancer patients. J Clin Oncol 2005;
23:5229e5234.
13. Santiago-Palma J, Khojainova N, Kornick C,
et al. Intravenous methadone in the management
of chronic cancer pain: safe and effective starting
doses when substituting methadone for fentanyl.
Cancer 2001;92:1919e1925.
14. Mercadante S, Porzio G, Fulfaro F, et al. Switching from transdermal drugs: an observational ‘‘n of
1’’ study of fentanyl and buprenorphine. J Pain
Symptom Manage 2007;34:532e538.
15. Riley J, Ross J, Rutter D, et al. No pain relief
from morphine? Individual variation in sensitivity
to morphine and the need to switch to an alternative opioid in cancer patients. Support Care Cancer
2006;14:56e64.
16. Farrar JT, Berlin JA, Strom BL. Clinically important changes in acute pain outcome measures: a validation study. J Pain Symptom Manage 2003;25:406e411.
17. Muller-Busch H, Lindena G, Tietze K,
Woskanjan S. Opioid switch in palliative care, opioid
choice by clinical need and opioid availability. Eur J
Pain 2005;9:571e579.
18. Ashby M, Martin P, Jackson K. Opioid substitution to reduce adverse effects in cancer pain management. Med J Austr 1999;170:68e71.
641
19. Kloke M, Rapp M, Bosse B, Kloke O. Toxicity
and/or insufficient analgesia by opioid therapy:
risk factors and the impact of changing the opioid.
A retrospective analysis of 273 patients observed at
a single center. Support Care Cancer 2000;8:
479e486.
20. de Stoutz N, Bruera E, Suarez-Almazor M. Opioid rotation for toxicity reduction in terminal cancer patients. J Pain Symptom Manage 1995;10:
378e384.
21. Cherny N, Chang V, Frager G, et al. Opioid
pharmacotherapy in the management of cancer
pain. Cancer 1995;76:1288e1293.
22. Ripamonti C, Fagnoni E, Campa T, Brunelli C,
De Conno F. Is the use of transdermal fentanyl inappropriate according to the WHO guidelines and
the EAPC recommendations? A study of cancer patients in Italy. Support Care Cancer 2006;14:
400e407.
23. Riley J, Ross J, Rutter D, et al. A retrospective
study of the association between haematological
and biochemical parameters and morphine intolerance in patients with cancer pain. Palliat Med 2004;
18:19e24.
24. Gagnon B, Bruera E. Differences in the ratios of
morphine to methadone in patients with neuropathic pain versus non-neuropathic pain. J Pain
Symptom Manage 1999;18:120e125.
25. Moryl N, Santiago-Palma J, Kornick C, et al. Pitfalls of opioid rotation: substituting another opioid
for methadone in patients with cancer pain. Pain
2002;96:325e328.
26. Hagen N, Swanson R. Strychnine-like multifocal
myoclonus and seizures in extremely high-dose opioid administration: treatment strategies. J Pain
Symptom Manage 1997;14:51e58.
27. Mercadante S, Portenoy RK. Opioid poorly responsive cancer pain. Part 2. Basic mechanisms
that could shift dose-response for analgesia. J Pain
Symptom Manage 2001;21:255e264.