Content uploaded by Rainer Sabatowski
Author content
All content in this area was uploaded by Rainer Sabatowski on Mar 05, 2018
Content may be subject to copyright.
Constipation and the use of laxatives: a comparison
between transdermal fentanyl and oral morphine
Lukas Radbruch, Rainer Sabatowski, Georg Loick Pain Clinic, Department of Anaesthesiology, University of
Cologne, Carsten Kulbe Department of Anaesthesiology, Gilead Hospital, Bielefeld, Mario Kasper,
Stefan Grond and Klaus A Lehmann Pain Clinic, Department of Anaesthesiology, University of Cologne,
Cologne
Abstract: Constipation and the use of laxatives were investigated in patients with
chronic cancer pain treated with oral morphine and transdermal fentanyl in an open
sequential trial. Forty-six patients were treated with slow-release morphine 30–1000
mg/day for 6 days and 39 of these patients were switched to transdermal fentanyl
0.6–9.6 mg/day with a conversion ratio of 100:1. Median fentanyl doses increased from
1.2 to 3.0 mg/day throughout the 30-day transdermal treatment period. Twenty-three
patients completed the study. Two patients died from the basic disease while treated
with transdermal fentanyl, 12 patients were excluded for various reasons, and not
enough data for evaluation were available for two patients. Mean pain intensity
decreased slightly after conversion although the number of patients with breakthrough
pain or requiring immediate-release morphine as a rescue medication was higher with
transdermal fentanyl.
The number of patients with bowel movements did not change after the opioid switch
but the number of patients taking laxatives was reduced significantly from 78–87% of
the patients per treatment day (morphine) to 22–48% (fentanyl). Lactulose was used
mainly and was reduced most drastically, but other laxatives were also used less
frequently.
In this study transdermal fentanyl was associated with a significantly lower use of
laxatives compared to oral morphine. The difference in the degree of constipation
between the two analgesic regimens should be confirmed in a randomized double-blind
study that takes into account both constipation and use of laxatives.
Key words: analgesics, opioid; administration, cutaneous; cathartics; constipation;
fentanyl; morphine
Resumé: Nous avons étudié la constipation et l’utilisation de laxatifs chez des patients
douloureux cancéreux chroniques traités par morphine orale et fentanyl transdermique
dans une étude séquentielle ouverte. Quarante-six patients ont pris de la morphine à
libération prolongée 30–1000 mg/jour pendant 6 jours et 39 d’entre eux ont ensuite été
mis sous fentanyl transdermique 0.6–9.6 mg/jour selon un facteur de conversion de
100:1. Les doses moyennes de fentanyl sont passées de 1.2 à 3.0 mg/jour durant les
30 jours de son utilisation. Vingt-trois patients ont terminé l’étude. Deux sont décédés
Palliative Medicine 2000; 14: 111–119
© Arnold 2000 0267–6591(00)PM290OA
Address for correspondence: Dr Lukas Radbruch, Klinik
für Anästhesiologie und Operative Intensivmedizin, Universität
zu Köln, 50924 Köln, Germany. E-mail: Lukas.Radbruch@
Uni-Koeln.de
04pm290.qxd 16/3/00 8:53 am Page 111
Introduction
Opioids are the mainstay of symptomatic cancer
pain management, and morphine has been
described as the gold standard.1–3 However, side-
effects are common and often have to be treated
with adjuvant medications. Whereas some side-
effects, such as nausea or sedation, are reported
mainly with initial dose titration of morphine,
constipation does not seem to diminish with
longer-term therapy. Even if the need for a regular
intake of laxatives does not stress the patient too
much, it remains a barrier for the use of morphine
with nonspecialized physicians treating cancer
patients.
Other opioids may show less constipation as a
side-effect. The anti-diarrhoeal effect was shown to
be significantly less for fentanyl than for morphine.4
This may be related to differences in the distribu-
tion of the opioids in blood and brain depending on
the lipid solubility.5Transdermal administration of
fentanyl also has the advantage of bypassing the
intestinal opioid receptors, and has been reported
to cause less constipation than oral morphine.6–8
However, the main topic in these studies has been
pain relief and not constipation. We therefore inves-
tigated constipation and the use of laxatives in
patients with chronic cancer pain treated with oral
morphine and transdermal fentanyl.
Methods
From June 1995 to January 1996 patients with
cancer pain and adequate pain relief from potent
opioids were included in an open multicentre study.
Patients were treated with oral slow-release
morphine for at least 6 days (morphine phase), until
they reported stable pain intensity scores of 40 or
less on a visual analogue scale (0 = no pain, 100 =
worst pain imaginable) for at least 2 days. Analgesic
therapy then was switched from oral morphine to
transdermal fentanyl (fentanyl phase). Fentanyl
doses were calculated with a conversion table based
on a 100:1 dose ratio.7If the calculated fentanyl dose
was higher than 2.4 mg/day = 100 µg/h (more than
270 mg/day slow-release morphine), more than one
patch was used. Fentanyl patches were changed
regularly after 3 days. Fentanyl doses were
increased when patients reported inadequate pain
relief or had to take more than six rescue
medications per day. The study was terminated after
30 days of transdermal therapy. Patients were free
to continue with transdermal fentanyl application. In
the morphine phase as well as in the fentanyl phase
oral immediate-release morphine was available as a
rescue medication for treatment of breakthrough
pain. Patients documented pain intensity on a visual
analogue scale three times daily (endpoints 0 = no
pain, 100 = worst pain imaginable), as well as
112 L Radbruch et al.
de leur maladie alors qu’ils étaient sous fentanyl, 12 ont été exclus pour diverses raisons,
et pour 2 autres les résultats étaient insuffisants pour être exploités. L’intensité moyenne
de la douleur a légèrement diminué après mise sous fentanyl bien que la
consommation de doses de secours sous forme de morphine à libération immédiate
pour douleurs inopinées ait été alors supérieur.
Le nombre de patients ayant un transit n’a pas varié lors du changement de traitement
mais le nombre de patients prenant des laxatifs a baissé de façon significative pour
passer de 78–87% (sous morphine) à 22–48% (sous fentanyl). Le lactulose était
essentiellement utilisé et a été réduit de la façon la plus drastique, mais les autres
laxatifs étaient, eux, peu utilisés.
Dans cette étude, la prescription du fentanyl transdermique s’est accompagnée d’un
usage de laxatifs significativement inférieur à celui lié à la presciption de morphine. La
différence du degré de constipation entre les 2 antalgiques devrait être confirmée par
une étude randomisée en double-aveugle, prenant à la fois en compte la constipation
et l’utilisation de laxatifs.
Mots-clés: antalgiques opioïdes; voie transdermique; laxatifs; constipation; fentanyl;
morphine
04pm290.qxd 16/3/00 8:53 am Page 112
frequency of pain attacks, intake of rescue
medications and use of laxatives in a pain diary.
Constipation was evaluated from the patients’ self-
assessment of frequency and consistency of
defaecation in the pain diary. When switching from
morphine to fentanyl and at the end of the study
patients completed the quality-of-life questionnaire
of the European Organization for Research and
Treatment of Cancer (EORTC-QLQ-C30).9This
questionnaire uses 30 items to assess nine
symptoms and six dimensions of quality of life.
Blood pressure, heart rate, respiratory rate and
skin reaction at the site of application were docu-
mented by the treating physician on day 0, 6, 12, 18,
24 and 30.
Patients who completed the study until day 17 or
longer were included in an intraindividual compar-
ison of laxative intake. For each patient the per-
centage of days with laxative medication in the
morphine phase was calculated and compared with
the percentage of days with laxatives for the period
of days 12–17 of transdermal therapy. Patients were
excluded from the analysis if evaluative data were
available for less than 3 days of the morphine ther-
apy. The differences in the percentages of days were
analysed with the Wilcoxon matched pair signed
rank test at a 5% significance level.
The study was accepted by the ethics committee
of the University of Cologne. Patients had to give
written consent before participation in the study.
Results
Forty-six patients were included in the study (Table 1).
Seven patients were excluded in the morphine phase
because stable analgesia could not be achieved
(three patients), no laxatives were needed, consent
was withdrawn, a chemotherapeutic trial was
started or an emergency operation had to be
performed (one patient each).
Transdermal therapy was started for 39 patients.
Insufficient data in the morphine phase were doc-
umented for two patients, so only 37 patients
were available for evaluation, of whom only 23
completed the 30-day study period. Two patients
died in the fentanyl phase (days 22 and 27) from
their underlying disease. Twelve patients were
excluded between days 3 and 30. Four of these
patients did not wish to continue transdermal
therapy. Insufficient analgesia led to withdrawal of
two patients, two patients were excluded when an
operation was indicated. Two patients showed signs
of withdrawal, such as restlessness, on days 3 and 8
and were excluded.
One patient with astrocytoma was suspected to
have stolen drugs from the ward: a urine analysis
showed intake of bromazepam and tramadol. Nei-
ther drug had been prescribed, and the study was
discontinued.
A patient with lung cancer was excluded on day
27. The patch was changed by the family doctor at
home for the first time. By mistake he had applied
Constipation and the use of laxatives 113
Table 1 Demographic data for 46 patients
Patients 46
Male 29
Female 17
Median Range
Age 57.5 years 31–83 years
Weight 62.5 kg 40–96 kg
Height 172 cm 146–185 cm
Patients entering Patients receiving transdermal Patients completing the
Site of primary cancer study (n= 46) fentanyl (n= 39) study (n= 23)
Gastrointestinal 11 10 5
Oral cavity/throat 10 9 7
Urogenitary tract 9 6 4
Respiratory tract 7 7 4
Breast 3 3 2
Haematological/lymphatic system 3 2 1
Other 3 2 0
04pm290.qxd 16/3/00 8:53 am Page 113
the new patch using adhesive tape without remov-
ing the protective cover. The patient described
slight symptoms of withdrawal and consequently did
not wish to continue the study.
Doses between 30 and 1000 mg/day morphine
(median 180 mg/day) were given in the morphine
phase, and were switched to between 0.6 and 9.6
mg/day fentanyl (median 1.8 mg/day) initially. One
patient was switched from 960 mg/day oral mor-
phine to 9.6 mg/day transdermal fentanyl (four
patches with 100 µg/h). Median fentanyl doses
increased throughout the study period to 3 mg/day
(range 0.6–16.8 mg/day; Figure 1). Median pain
intensity scores were reduced slightly after opioid
switch (Figure 2). Between 14% and 30% of the
patients needed rescue medications on treatment
days with slow-release morphine (Figure 3). With
transdermal fentanyl more patients reported
breakthrough pain and required rescue medication
(28–56% of the patients per day). Patients received
up to 180 mg of immediate-release morphine as res-
cue medication in the morphine phase and up to
210 mg in the fentanyl phase.
The frequency of bowel movements did not
change significantly in the study period, but the
intake of laxatives was reduced significantly (Figure
4). Patients used up to four different laxatives
and enemas. Lactulose was used most frequently
and was reduced most markedly, but other laxatives
were used less frequently also (Figure 5). The
consistency of the stools was described as pasty or
liquid by one-quarter of the patients in the
morphine phase. This group increased to 35% of
the patients in the first 6 days following conversion
to transdermal fentanyl, but decreased in the next
3 weeks to 20%.
The intraindividual comparison of the use of lax-
atives was possible for 28 patients. The Wilcoxon
rank test showed a significant reduction between
days −6 and −1 (morphine phase) and days 12 and
17 (fentanyl phase, P <0.001). The frequency of lax-
ative use decreased in 23 of these patients and
increased in only two patients with transdermal fen-
tanyl. Laxative medication did not change for three
patients.
There were no significant changes in blood pres-
sure, heart rate or respiratory rate in the study peri-
od except for a patient with a respiratory rate of 4
per min reported by the treating physician on day
30. This patient had been treated with a maximum
of 16.8 mg transdermal fentanyl per day. Pre-final
deterioration with sedation and somnolence
caused by the progression of the hypopharynx car-
cinoma had been reported 8 days before. In spite of
the low respiratory rate, transdermal treatment was
continued until the patient died 7 days later with-
out signs of opioid overdose. A connection of this
single low respiratory value with the opioid thera-
py cannot be excluded, but was thought unlikely by
the treating physician.
Mild to moderate skin reaction such as erythema
at the application site after removal of the patch was
observed in five patients.
In the EORTC quality-of-life questionnaire
symptom scores showed only minor differences
except for constipation, where the score decreased
114 L Radbruch et al.
Figure 1 Daily dose of transdermal fentanyl for 23 patients (median and range)
04pm290.qxd 16/3/00 8:53 am Page 114
from 57 on day 0 to 27 on day 30. Physical, role,
emotional and cognitive functioning decreased
during the study period, while social functioning
and quality of life remained unchanged.
Discussion
For a long time slow-release morphine seemed to be
the only therapeutic option for treatment of chronic
severe cancer pain. During the last years several other
potent opioids have become available in convenient
application forms. Transdermal administration of
fentanyl has its obvious advantages for patients who
cannot take their analgesics by the oral route, either
because of intractable nausea and vomiting or
because they are not able to swallow. Over and above
this it has been postulated that transdermal fentanyl
causes less constipation and therefore may be
advantageous for other patients as well.10
Constipation and the use of laxatives 115
Figure 2 Pain intensity on a visual analogue scale with the endpoints 0 = no pain and 100 = worst pain imaginable for 23
patients (mean and standard deviation, three assessments per patient per day). Days −6 to 0: treatment with oral slow-release
morphine; days 1 to 30: treatment with transdermal fentanyl.
Figure 3 Rescue medication and breakthrough pain. For each day the percentage of patients needing rescue medication
and of patients reporting breakthrough pain is calculated for the 23 patients completing the study. Days −6 to 0: treatment
with oral slow-release morphine; days 1 to 30: treatment with transdermal fentanyl.
04pm290.qxd 16/3/00 8:54 am Page 115
The lipophilic drug fentanyl passes the blood–
brain barrier very rapidly and reaches its site of action
in the central nervous system more easily than the
hydrophilic morphine. Analgesic effects are reached
with doses that have little intestinal action. In a study
with rats fentanyl and morphine doses were calcu-
lated that produced analgesia in 50% of the animals
as well as doses that reversed castor oil-induced diar-
rhoea in 50% of the animals.4After subcutaneous
administration the analgesic dose exceeded the
antidiarrhoeal dose only slightly in the case of fen-
tanyl (ED50analgesia = 1.1 ×ED50antidiarrhoea), but
markedly in the case of morphine (ED50analgesia =
36 ×ED50antidiarrhoea). After oral administration
the difference was less pronounced (fentanyl:
ED50analgesia = 1.7 ×ED50antidiarrhoea; mor-
phine: ED50analgesia = 6.2 ×ED50antidiarrhoea).
Other reasons for a lower incidence of constipation
may be related to the transdermal route that bypass-
es the enteral opioid receptors.
The influence of morphine and fentanyl on con-
stipation has been compared in several clinical tri-
als. Significantly less constipation with transdermal
fentanyl was found in an open cross-over study with
116 L Radbruch et al.
Figure 4 Bowel movements and use of laxatives. For each day the percentage of patients using laxatives and of patients
reporting bowel movements is calculated for the 23 patients who completed the study. Days −6 to 0: treatment with oral
slow-release morphine; days 1 to 30: treatment with transdermal fentanyl
Figure 5 Laxatives used by 23 patients (multiple entries). Days −6 to 0: treatment with oral slow-release morphine; days 1
to 30: treatment with transdermal fentanyl.
04pm290.qxd 16/3/00 8:54 am Page 116
202 patients from palliative care centres in the UK.6
EORTC symptom scores were used to assess con-
stipation. Mean score after fentanyl was 20.7 com-
pared to 36.6 after 15 days of morphine treatment.
The scores for diarrhoea were 15.4 after fentanyl
and 8.6 after morphine. However, the conversion
ratio chosen was 150:1, much higher than the
equianalgesic conversion rate of 70:1 found by Don-
ner et al.7Correspondingly with fentanyl treatment
47% of the patients needed at least one dose adjust-
ment compared to 27% with morphine. In the fen-
tanyl phase, patients used more immediate-release
morphine than in the morphine phase. More
importantly, no information on the use of laxatives
is given, so the reduction of constipation may easi-
ly have been influenced by other factors.
In another study from the same group the inci-
dence of constipation was reduced significantly with
transdermal fentanyl therapy compared to a previ-
ous stabilization phase with slow-release or imme-
diate-release morphine.11 The duration of the
stabilization phase ranged from 0 to 11 days, and
again no information on laxative medication is
given.
The study design used by Donner and colleagues
was similar to our own. Thirty-eight patients were
treated with slow-release morphine for 6 days and
then switched to transdermal fentanyl for 15 days.7
Patients reported constipation on 59% of days with
morphine. This was reduced to 35% of the days with
transdermal fentanyl. Concomitantly, the medica-
tion with laxatives was reduced from 62% to 38%
of days. However, assessment of constipation was
not standardized. Side-effects such as constipation
were elicited by questioning and control of the pain
diary. As in the study of Ahmedzai et al., the initial
dose of transdermal fentanyl may have been inad-
equate for many patients. Half of the patients need-
ed a dose increase after the conversion to
transdermal therapy, although a conversion ratio of
100:1 was choosen in this study.
More information is available from another mul-
ticentre study comparing an initial stabilization
phase with oral slow-release morphine over 1 week
and the following open-ended treatment phase with
transdermal fentanyl in 53 patients.12 Forty per cent
of the patients receiving morphine and 42% with
transdermal therapy received laxatives. Nearly all
laxatives were started during initial morphine titra-
tion and were continued throughout the study. With
this regimen, four patients (8%) reported consti-
pation with morphine therapy and none with fen-
tanyl.
In an open clinical trial 18 patients were followed-
up for 3 months after the switch from oral morphine
to transdermal fentanyl.13 Twelve patients reported
constipation at the beginning of the study and con-
stipation was the reason for the opioid switch in five
of these patients. Eight of the 12 patients reported
an improvement after a period of 14 days which last-
ed to the end of the study period. Again no data on
the use of laxatives are given. In a case report four
of five cancer patients reported less constipation
with transdermal fentanyl after conversion from
oral oxycodone or morphine.14 Two studies from
our pain clinic used intravenous patient-controlled
analgesia for dose titration of transdermal fen-
tanyl.8,15 Both report a reduction of constipation
compared with the pre-study situation but no
direct comparison with oral morphine can be
extracted from the data.
The incidence of constipation is much lower if
only spontaneous reports of the patients are evalu-
ated. In a study comparing transdermal fentanyl to
placebo, the incidence of constipation was 3%.16
Constipation is not even mentioned as a specific
side effect in a randomized open study comparing
slow-release morphine and transdermal fentanyl.17
Self-assessment of more or less constipation by the
patient may be influenced mainly by the need to
take laxatives and must not necessarily indicate a
change in frequency or consistency of defaecation.
This has not been considered in the trial designs of
the studies using self-assessment forms such as the
EORTC quality-of-life questionnaire.
In the studies discussed no information is given
about the laxatives used. Only in one study do the
authors state that they use mainly docusate or senna
preparations.12 As lactulose is used frequently in the
study centres participating in our trial, it is not unex-
pected that this laxative was used most frequently.
However, lactulose has been described as a rather
weak laxative.18 Its laxative effect could have been
negligible and this could have influenced the num-
ber of days with laxatives in the morphine phase.
Even if our study had been designed primarily to
investigate the use of laxatives and constipation with
transdermal fentanyl, and had avoided the mistakes
discussed above, some factors remain that could
have influenced our results.
Constipation and the use of laxatives 117
04pm290.qxd 16/3/00 8:54 am Page 117
The study design was open, and so prejudices
from staff or patients may have entered. The site of
the primary tumour was situated in the gastro-
intestinal tract for a quarter of the patients, and in
these patients constipation may easily have been
caused by tumour growth.
It may be questioned whether the conversion
from morphine to fentanyl really was equianalgesic.
Thus, less constipation may have been the conse-
quence of lower equianalgesic opioid dosage. We
used the conversion ratio 100:1 described by Don-
ner et al.7This ratio was kept even for a patient with
high morphine doses, where transdermal therapy
started with four 100 µg/h patches. Conversion
tables based on this ratio are included in the prod-
uct information for fentanyl patches in Germany.
With this ratio patients receive more fentanyl than
with the more cautious ratio of 150:1 used in the
study of Ahmedzai and Brooks.6The ratio used in
our study is closer to but still below the equi-
analgesic ratio of 70:17found in other studies after
longer treatment periods. This equianalgesic ratio
is confirmed by the increase in the median fentanyl
dose throughout the study. Comparing the median
dose of 2.4 mg/day on day 15 with the median mor-
phine dose before conversion, a ratio of 75:1 could
be calculated from our data. However, it must be
presumed that at least part of the dose escalation is
due to progressive tumour spread.
On the other hand, the increase in the number of
patients needing immediate-release morphine for
breakthrough pain in the fentanyl phase may have
reduced the positive effect of transdermal therapy
on constipation, and use of laxatives through daily
doses of immediate-release morphine remained low
for most patients.
Results may have been influenced by the high
number of patients who dropped out of the study.
However, this is a drawback of all studies with can-
cer patients and long study periods. Ahmedzai and
Brooks reported a 50% drop-out rate in a 30-day
study period.6In another study only 38 from 98
patients completed the study.7
Contrary to other studies frequency and consis-
tency of defaecation were not changed with trans-
dermal fentanyl therapy after switching from oral
slow-release morphine in our study. However, the
use of laxatives was reduced significantly with
transdermal fentanyl. The difference in the degree
of constipation between the two analgesic regimens
should be confirmed in a randomized double-blind
study that takes into account both constipation and
use of laxatives.
Acknowledgements
This study was supported by a research grant from
the Janssen-Cilag GmbH, Neuss, Germany. We
would like to thank our co-workers in the Pain
Clinic of the Department of Anaesthesiology of the
University of Cologne and all physicians in the other
study centres. Without their support this study
would not have been possible.
References
1 Hanks GW, DeConno F, Ripamonti C et al.
Morphine in cancer pain: modes of administration.
Br Med J 1996; 312: 823–26.
2 World Health Organization. Cancer pain relief.
Geneva: World Health Organization, 1986.
3 World Health Organization. Cancer pain relief: with
a guide to opioid availability, 2nd edn. Geneva:
World Health Organization, 1996.
4 Megens AA, Artois K, Vermeire J, Meert T,
Awouters FH. Comparison of the analgesic and
intestinal effects of fentanyl and morphine in rats.
J Pain Sympt Manage 1998; 15: 253–58.
5 Herz A, Teschemacher HJ. Activities and sites of
antinociceptive action of morphine-like analgesics
and kinetics of distribution following intravenous,
intracerebral and intraventricular application. Adv
Drug Res 1971; 6: 79–119.
6 Ahmedzai S, Brooks D. Transdermal fentanyl versus
sustained-release oral morphine in cancer pain:
preference, efficacy, and quality of life. The TTS-
Fentanyl Comparative Trial Group. J Pain Symptom
Manage 1997; 13: 254–61.
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: 527–34.
8 Grond S, Zech D, Lehmann KA, Radbruch L,
Breitenbach H, Hertel D. Transdermal fentanyl in
the long-term treatment of cancer pain: a
prospective study of 50 patients with advanced
cancer of the gastrointestinal tract or the head and
neck region. Pain 1997; 69: 1–8.
9 Aaronson NK, Ahmedzai S, Bergmann B et al. The
European Organization for Research and Treatment
of Cancer QLQ-C30: a quality-of-life instrument for
use in international clinical trials in oncology. J Natl
Cancer Inst 1993; 85: 365–76.
118 L Radbruch et al.
04pm290.qxd 16/3/00 8:54 am Page 118
10 Jeal W, Benfield P. Transdermal fentanyl. A review
of its pharmacological properties and
therapeutic efficacy in pain control. Drugs 1997; 53:
109–38.
11 Ahmedzai S, Allan E, Fallon M et al. Transdermal
fentanyl in cancer pain. J Drug Dev 1994; 6: 93–97.
12 Sloan PA, Moulin DE, Hays H. A clinical evaluation
of transdermal therapeutic system fentanyl for the
treatment of cancer pain. J Pain Symptom Manage
1998; 16: 102–11.
13 Slappendel R, Lako SJ, Crul BJ. Gastrointestinal
side effects diminish after switch over from
morphine to transdermal fentanyl. Ann Oncol 1994;
5: 200.
14 Slover R. Transdermal fentanyl: clinical trial at the
University of Colorado Health Sciences Center. J
Pain Symptom Manage 1992; 7: S45–47.
15 Zech DF, Grond SU, Lynch J, Dauer HG,
Stollenwerk B, Lehmann KA. Transdermal fentanyl
and initial dose-finding with patient-controlled
analgesia in cancer pain. A pilot study with 20
terminally ill cancer patients. Pain 1992; 50: 293–301.
16 Kongsgaard UE, Poulain P. Transdermal fentanyl for
pain control in adults with chronic cancer pain. Eur
J Pain 1998; 2: 53–62.
17 Wong JO, Chiu GL, Tsao CJ, Chang CL.
Comparison of oral controlled-release morphine
with transdermal fentanyl in terminal cancer pain.
Acta Anaesthesiolog Sin 1997; 35: 25–32.
18 Sykes NP. Constipation and diarrhoea. In: Doyle D,
Hanks G, MacDonald M eds. Oxford textbook of
palliative medicine, 2nd edn. Oxford: Oxford
University Press, 1998: 513–26.
Constipation and the use of laxatives 119
04pm290.qxd 16/3/00 8:54 am Page 119
