This Article Has Been Retracted

Eur J Anaesthesiol 2014; 31:3540

ORIGINAL ARTICLE

Intravenous nonopioid analgesic drugs in chronic low

back pain patients on chronic opioid treatment
A crossover, randomised, double-blinded, placebo-controlled
study
Leonore Wetzel, Markus Zadrazil, Tatjana Paternostro-Sluga, Georg Authried,
Sibylle Kozek-Langenecker and Gisela Scharbert

BACKGROUND Addition of nonopioid analgesic drugs MAIN OUTCOME MEASURES Primary outcome was as
reduces pain and opioid requirements in acute low back pain. follows: VAS pain intensity (0 to 100 mm) at inclusion,
In noncancer chronic low back pain (CLBP), the efficacy of a before and within 30 min after infusion. Secondary outcomes
combined regimen to reduce breakthrough pain has not were as follows: RolandMorris questionnaire, McGill pain
been proven so far. questionnaire and a test panel of physical functioning for
spinal mobility, muscular endurance, balance and coordina-
OBJECTIVE Evaluation of the effects of intravenous (i.v.)
tion. The differences in means of the above assessments
nonopioid analgesic drugs on pain intensity and lumbar
among the groups were analysed.
mobility in CLBP patients on chronic opioid therapy.
RESULTS We found an improvement in VAS from the day of
DESIGN Randomised, placebo-controlled, double blinded,
inclusion to the day of each appointment. We observed no
crossover study.
improvement in pain intensity (VAS) or in any of the physical
SETTING Vienna General Hospital, Austria, from December functioning tests immediately before versus after adminis-
2002 to May 2004. tration of the four i.v. drugs. Reductions in sensory, affective
and cognitive dimensions of the McGill pain questionnaire
PATIENTS Thirty-six adults with CLBP on chronic opioid
were statistically significant in the diclofenac group. A trend
therapy. Inclusion criteria are as follows: American Society
of McGill pain questionnaire improvement existed in the other
of Anesthesiologists physical status less than 3, visual
groups.
analogue scale (VAS) more than 4 and no known allergy to
any of the used drugs. CONCLUSION The present data show that the anti-
cipation of an i.v. infusion of nonopioid analgesic drug
INTERVENTION After written informed consent and VAS
improves VAS significantly, probably through expectation-
assessment, any oral nonopioid analgesic drug (NSAIDs,
related mechanisms. However, single dose i.v. infusions of
metamizol, paracetamol) was replaced by placebo 10 days
nonopioid analgesic drugs fail to improve pain intensity and
before the first test infusion as a washout period.
spinal mobility in CLBP patients on chronic opioid treatment,
Coanalgesics (anticonvulsants, antidepressants) were main-
even immediately after the infusion.
tained. Each patient received randomly four i.v. test infusions
of diclofenac 75 mg (and orphenadrine 30 mg), parecoxib Published online 17 October 2013
40 mg, paracetamol 1 g and isotonic saline. A washout time
of 72 h was allowed between each infusion.

From the Department of Anaesthesia, Intensive Care and Pain Management (LW, MZ, GA, GS), Department of Physical Medicine and Rehabilitation, Medical University
Vienna (TP-S), Department of Anaesthesia and Intensive Care, Evangelical Hospital Vienna (SK-L), Vienna, Austria
Correspondence to Leonore Wetzel, MD, Department of Anaesthesia, Intensive Care and Pain Management, Vienna, Austria
E-mail: leonore.wetzel@meduniwien.ac.at

0265-0215 2013 Copyright European Society of Anaesthesiology DOI:10.1097/EJA.0b013e328365ae28

Copyright European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.

 This Article Has Been Retracted
36 Wetzel et al.

Background assessed at inclusion. The pain intensity on VAS at

Chronic low back pain (CLBP) is a complex bio-psycho- inclusion was the baseline value.
social problem requiring multidisciplinary and multi-
General exclusion criteria were any additional
modal treatment that includes nonpharmacological
treatment (psychotherapy, orthopaedic aids, physio-
therapy, drugs and self-care.1 In the pharmacological
therapy, acupuncture, use of depot steroids in the last
part of the treatment, rational combinations of
three months), cancer-related pain, positive pregnancy
analgesic drugs with different mechanisms of action
test, American Society of Anesthesiologists (ASA)
are proposed to achieve improved efficacy and safety
physical status at least 3, allergy or contraindications to
compared with equianalgesic doses of single drugs.24
the tested substances, specific inflammatory or progress-
Addition of nonopioid analgesic drugs (NSAIDs, meta-
ive metabolic disorders, fibromyalgia, autoimmune dis-
mizol, paracetamol) reduces pain and opioid requirement
eases, inflammatory rheumatic disorders, radiculopathy,
in cancer pain patients and in acute back pain.57
spinal trauma or vertebral fracture caused by osteoporosis
This approach was adopted for noncancer chronic pain
in the past 6 months, relevant cardiopulmonary restric-
patients.
tions, severe kidney or liver function disorders, acute
Breakthrough pain is highly prevalent in opioid-treated duodenal or ventricular ulcer or psychiatric disorder.
patients with noncancer CLBP compared with cancer
Preexisting nonopioid analgesic drugs (NSAIDs, meta-
groups.7 In the primary care setting in Europe,
mizol, paracetamol) were replaced by oral placebo 10 days
this breakthrough pain in CLBP patients is commonly
before the first test infusion as a washout period and
managed with a series of intravenous infusions of
were maintained until 1 week after the last test infu-
nonopioid analgesics. Patients receive intravenous
sion. Coanalgesics (anticonvulsants, antidepressants) and
infusions over 30 min, for example. During the proce-
opioids were maintained. Medication for the treatment of
dure of intravenous access, patients probably feel
internal comorbidity remained unchanged.
increased attention from the medical staff in comparison
to a normal consultation. After inclusion, four appointments for the study were
fixed. At every appointment, each patient received
Nonopioid analgesic drugs are heterogeneous in pharma-
randomly the test infusions of 250 ml administered over
cology and include nonselective cyclooxygenase (COX)-I
30 min consisting of diclofenac 75 mg (and orphenadrine
and COX-II inhibitors, selective isoenzyme COX-II
30 mg; Neodolpasse; Fresenius Kabi Austria GmbH,
inhibitors, nonselective inhibitors of prostaglandin syn-
Graz, Austria), parecoxib 40 mg (Dynastat; Bristol-Myers
thesis and other mostly unknown mechanisms such as
Squibb, Wien, Austria), paracetamol 1 g (Paracetamol;
paracetamol (acetaminophen).8 The aim of the present
Bristol-Myers Squibb, Wien, Austria; acetaminophen)
study was to evaluate the effect of nonopioid analgesic
and normal saline (placebo). The sequence of test
drugs on breakthrough pain intensity and lumbar
infusions was assigned randomly using a computer-
mobility in CLBP patients on chronic oral sustained-
generated table. The time interval between test infusions
release opioid therapy.
was 72 h. The elimination half-life for the chosen drugs
are 1 to 2 h for diclofenac and paracetamol, 8 h for
parecoxib and 17 h for the muscle relaxant orphenadrine.
Materials and methods
After four times the half-life value, more than 90% of the
Ethics
drugs are eliminated and so we calculated elimination as
Ethical approval for this study (Ethical Committee N8
four times the longest half-life of 17 h; this equals 68 h.
411/2002) was provided by the Ethical Committee of the
To make the planning easier, we waited 3 days between
Medical University of Vienna, Austria on 9 December
each infusion. Patients and observers were blinded to the
2002.
test infusion. The doses of diclofenac, parecoxib and
paracetamol represent the manufacturers recommended
Data collection was at the pain clinic of the Vienna
dose for single rescue injection. The duration of the test
University Hospital, Austria. After written informed
infusion was standardised at 30 min in order to guarantee
consent, 40 adult male and female consecutive out-
study blinding. This infusion speed represents common
patients were included. The patients were referred from
clinical practice. A pilot study in 10 patients confirmed
their primary care doctor or specialist to our pain manage-
no difference in efficacy between parecoxib infusions
ment department. The patients were aged more than
administered over 30 and 10 min.
19 and less than 65 years with CLBP (at least 6 months) of
intensity at least 4 on a visual analogue scale (VAS) At every appointment, pain intensity on the VAS
according to the Quebec Task Force definition9 on stable and lumbar mobility (tests are described below) were
WHO class II or III medication (at least 4 months). The investigated immediately before and within 30 min after
study design was crossover, randomised, double-blinded each test infusion. Patients assessed their pain intensity
and placebo-controlled. The increase of pain intensity at on VAS (10-cm line, 0 no pain, 10 worst pain imagin-
least VAS 4 was regarded as breakthrough pain and was able) as a primary outcome. A validated German version

Eur J Anaesthesiol 2014; 31:3540

Copyright European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
 This Article Has Been Retracted
Intravenous nonopioid analgesic drugs in chronic low back pain 37

of the McGill pain questionnaire was used to assess values before each test infusion were assessed using
cognitive, affective and sensory dimensions of pain analysis of variance for repeated measures. Post-hoc
(PRI pain rating index, total number of words chosen comparisons were made by a paired t-test. Differences
to describe pain sensation).10 The cross-cultural between preinfusion and postinfusion values were
adaptation of the RolandMorris questionnaire for assessed using the paired t-test. Data are expressed as
German-speaking patients was used to assess long-term mean  standard deviation (SD); P <0.05 was considered
mean score changes in the functional status (values 0 to statistically significant. Power analysis was performed
24 items chosen).11 based on a two-sided significance level of 5% and a power
of 90%. Forty patients were required to demonstrate a
Patients underwent a clinical examination assessing tests
clinically relevant difference of pain intensity of at least
of physical functioning for spinal mobility, muscular
1.5 cm, which means 15% on VAS between oral WHO
endurance, balance and coordination by the use of the
I medication and pausing of this therapy (0.05 a,
fingertip-to-floor test, lumbar flexibility by the modified
two-tailed; 0.1 b, one-tailed). For the statistical analysis
Schobers test, endurance by the 6-min walk test, coordi-
program, SPSS version 16.0 was used.
nation by the timed up-and-go test and balance by the
functional reach test.
Results
Fingertip-to-floor test Forty patients with a mean (SD) age of 55  12 years, a
The distance from the middle finger to the floor is mean weight of 82  14 kg and a mean height of
determined with a tape during forward bending with 173  9 cm were included. Patients received either weak
the knees, arms and fingers fully extended (normal opioids (n 31) or potent opioids (n 9) on a stable basis
values: maximum 10 cm).12 (at least 4 months). Four patients dropped out without
giving reasons. Personal data are shown in Table 1.
Modified Schobers test
One line is marked in the standing position 5 cm below Table 1 Personal data and the distribution of treatment of chronic
the cranial margin of the sacrum and another line 15 cm low back pain patients by our pain management department
above the first line. The change in the distance between Number of patients
the two lines is measured before and after maximal Men 22
forward bending with the knees extended (normal range: Women 18

5 cm).13 Mean age in years (min to max)  SD 57.2 (25 to 83)  12.8
Antidepressants 17
Anticonvulsant 8
Six-minute walk test Tramadol 31
For the 6-min walk test, patients were asked to walk Morphine 9
Pain diagnosis
rapidly up and down a corridor in our hospital of known
Pseudoradicular lumboischialgia 24
length (30 m) for exactly 6 min. The maximum distance Radicular lumboischialgia 6
achieved was determined. The mean 6-min walk Failed back surgery 8
distance in women was 367 m (range: 249 to 479 m) Scoliosis 1
Vertebral endplate compression 1
and 400 m in men (range: 280 to 532 m).14 Treatment time in our department
1 year 12
Timed up-and-go test 2 years 11
3 years 12
This test measures the time taken to stand up from 5 years 3
a chair, walk around a marker on the floor at a distance 7 Years 2
of 3 m, return to the chair, and sit down (normal Duration of illness (low back pain)
values: 10 s).15 The test assesses spinal mobility and 2 years 4
3 years 8
coordination. 4 years 10
5 years 9
Functional reach test 7 years 5
15 years 4
This test measures the difference between arms length Daily tramadol dose at inclusion date
and maximal forward reach using a fixed base of 200 mg 4
support and is determined with a yardstick (normal range 300 mg 3
depending on age and sex between 30 and 45 cm).16 The 400 mg 14
600 mg 10
test assesses balance and coordination. Mean values of Daily morphine dose at inclusion date
three measurements for each test were determined. 30 mg 1
60 mg 3
120 mg 3
Statistics 180 mg 2
All data were tested for normal distribution using
the KolmogorovSmirnov test. Differences in baseline Twenty five percent of patients were treated with WHO III and 75% with WHO II.

Eur J Anaesthesiol 2014; 31:3540

Copyright European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
 This Article Has Been Retracted
38 Wetzel et al.

Fig. 1 an improvement in the VAS score between the day

of inclusion and before drug infusions of 17% for
VAS at inclusion day NaCl, 16% for paracetamol, 19% for parecoxib and
VAS before infusion 19% for diclofenac orphenadrine. As shown in Fig. 1,
VAS after infusion the improvement in pain intensity on the VAS after
the nonopioid administration is insignificant. Table 2
10 shows the data of the functional tests for spinal mobility,
** * ** * ** ** ** **
9
muscular endurance, balance and coordination. None of
8
the tests improved significantly after the administration
7
of the infusions. A significant improvement in the total
number of words chosen (PRI) to describe pain sensation
VAS (cm)

6
in sensory, affective and cognitive dimensions of the
5
McGill pain questionnaire was found in the diclofenac
4
group as shown in Fig. 2. As also shown in Fig. 2, there
3
was only an insignificant trend of sensory pain improve-
2
ment in the other groups. The RolandMorris disability
1
questionnaire showed no statistically significant change
0
Diclofenac+
in functional outcome over time (13.6  5.1 before and
NaCI Paracetamol Parecoxib orphenadrine 14.8  5.2 after four cross-over infusions).

Pain intensity on visual analogue scale (VAS) at inclusion day, and

before and after the test infusions. The figure shows statistically
significant improvements in pain intensity between values at the
inclusion day and administration days, but not between the values
immediately before and after administration of infusions. Statistically
significant compared with baseline (P <0.001). Statistically significant
compared with baseline (P <0.01).
Test infusions were administered intravenously for
30 min without any complications such as pain at the
intravenous injection site or along the vein. The mean
treatment time in our department and the time of illness
are shown in Table 1.
The mean baseline value of pain intensity of the
study population at inclusion was 6.5  1.3 cm on VAS.
Immediately before infusion of the first analgesic
drug, the mean VAS value was 4.8  2.0 cm for NaCl,
4.9  2.0 cm for paracetamol, 4.6  2.0 cm for parecoxib
and 4.6  2.0 cm for diclofenac orphenadrine. After the
test infusions, pain intensity on the VAS scale was
5.1  2.1 cm for NaCl, 5.1  2.1 cm for paracetamol,
4.5  2.0 cm for parecoxib and 4.5  2.0 cm for diclofe-
nac orphenadrine. This is shown in Fig. 1. The
significant improvement in pain intensity on the VAS
scale is clearly visible in the boxplot graph. There was
Discussion
We found a highly significant improvement in pain
intensity on VAS from baseline (at inclusion) to all days
of infusion. This indicates that the anticipation of getting
an intravenous infusion improved VAS significantly,
probably through expectation-related mechanisms as
described in a work of Benedetti et al.17 These expec-
tations of additional care activate endogenous opioids and
change the pain modulation network. Benedetti et al.17
showed that destruction of the prefrontal lobes in
Alzheimer patients leads to a loss of endogenous
expectation and placebo mechanisms. These losses make
an analgesic treatment less effective and more difficult.
Several other studies have demonstrated a positive effect
of placebo and expectation in irritable bowel syndrome,
chronic neuropathic pain and psychiatric patients.1820
If the appointment itself improves the patients pain
intensity, it should be used as part of the treatment rather
than giving additional NSAID infusions. Side-effects of
nonopioids are well known.
Our data also show that additional nonopioids do not
relieve breakthrough pain as expected. All infusions
failed to improve pain intensity on VAS, spinal mobility
and physical functioning in CLBP patients on stable

Table 2 Functional tests before and after the test infusions

Diclofenac R
NaCl Paracetamol Parecoxib orphenadrine
Before After Before After Before After Before After
Fingertip-to-floor 26.7  12.0 27.9  13.0 29.6  11.4 27.3  2.5 28.3  12.4 28.4  13.3 27.0  10.8 27.0  12.3
distance (cm)
Schobers test (cm) 19.9  1.7 19.7  1.9 20.1  1.8 20.2  2.4 19.9  1.7 19.9  1.9 20.3  1.7 20.3  1.8
Timed up-and-go test (s) 11.9  4.8 11.4  4.5 11.1  3.9 11.2  4.1 12.1  4.8 11.7  4.6 11.3  3.9 11.3  3.8
Functional reach test (cm) 85.0  12.0 84.5  11.8 84.9  11.0 85.4  11.4 84.9  11.4 85.4  10.8 83.4  8.3 84.4  9.6
Six-minute walk test (m) 383  117 398  120 379  116 376  128 374  132 390  127 398  120 381  109

Data are mean  SD. There were no statistically significant differences.

Eur J Anaesthesiol 2014; 31:3540

Copyright European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
 This Article Has Been Retracted
Intravenous nonopioid analgesic drugs in chronic low back pain 39
Fig. 2
PRI before infusion
PRI after infusion
25
* symptoms when attempting to document progress.
20
15
PRI
10
5 examinations and questionnaires was completed within
0 improvement in the total number of words chosen in the
Diclofenac+
NaCI Paracetamol Parecoxib
orphenadrine
Total number of items/words chosen in the McGill pain Questionnaire
(PRI pain rating index) before and after the test infusions.
Statistically significant (P <0.05).
opioid therapy (Table 2). Even repeated administration
of nonopioid analgesics after three cross-over sessions
showed no benefit (Table 2). Sometimes, the pain on
VAS increased after infusion as in the NaCl and
paracetamol groups (Fig. 1). There is no explanation
for an increase in VAS after infusion, but the pain relief
of the expectations lasted for the full study period.
Back pain is a common and very costly condition
associated with disability and absence from work.2,21
Chapman et al.2 even call it the Sisyphus of challenges
for the clinician. National guidelines for the manage-
ment of low back pain recommend the use of NSAIDs as
one option for symptomatic relief in the early stage to
prevent disability and to support early return to normal
activities.1,7,22 There is strong evidence that various
types of NSAIDs are equally effective in acute pain.22,23
However, there is low-quality evidence that NSAIDs are
more effective than placebo.24 In a recent review, Roelofs
et al.23 suggest that NSAIDs are effective for short-term
symptomatic relief in patients with acute and CLBP
without sciatica.
This Sisyphus of challenges often leads to the
practice of giving a patient infusions that are not always
evidence-based. Nonopioid analgesic drugs remain the
most frequently prescribed medications worldwide and
are widely used for patients with low back pain at various
stages of the disease.22
The goals of pain treatment in CLBP are to relieve pain
and to enable patients to start physiotherapy and
rehabilitation, work and normal activities.1,2 Pain syn-
drome-specific tests of physical functioning together with
Eur J Anaesthesiol 2014; 31:3540
Copyright European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
patient-oriented questionnaires may be the most
important tools clinicians have at their disposal to identify
the effectiveness of different low back pain therapies.25,26
Furthermore, tests of physical functioning may help
patients to focus on their functional ability rather than
At the same time, interaction between the patient and
the medical staff is more intense when functional tests
and questionnaires are used and could have a benefit
on the patients psychological well being. The set of
validated tests used in the present study proved
useful to measure spinal mobility, muscular endurance,
balance and coordination. Performance of these physical
20 min. The present data show a statistically significant
McGill Pain Questionnaire (PRI) only after diclofenac
administration. We observed an insignificant improve-
ment in psychological well being after the other
nonopioid infusions (Table 1). We do not have any
explanation for the improvement only in the diclofenac
group.
One limitation of our study is the sample size. A larger
sample would allow subgroup analyses for differences
between patients with either weak or strong opioids.
Without subgroup analysis, the inclusion of patients with
strong and weak opioids is a shortcoming of this study.
In addition, we did not perform tests of physical func-
tioning at the date of inclusion, which is also a limitation
of this study. However, our primary goal was not to
analyse any expectation-related mechanisms. We only
found this interesting outcome at the end of our evalua-
tion.
Conclusion
Our study enforces the theory that CLBP is a complex
bio-psycho-social problem requiring multidisciplinary
and multimodal treatment. The positive effect of the
expectation-related mechanisms could be a helpful
tool in the treatment of CLBP patients to reduce
pain intensity. Further studies need to confirm these
placebo and expectation-related mechanisms. Our find-
ings challenge the usefulness of repetitive intravenous
infusions with nonopioid analgesic drugs in CLBP
requiring stable long-term opioid therapy.
Acknowledgements relating to this article
Assistance with the study: Dr Paul Nilges, Chair of the Pain Center
in Mainz, helped the authors with some literature and inter-
pretation.
Financial support and sponsorship: this work was financially
supported by the Department of Anesthesiology, General Intensive
Care and Pain Control, Vienna Medical University, Vienna, Austria.
Conflicts of interest: none.
Presentation: none.
 This Article Has Been Retracted
40 Wetzel et al.
References
1 Chou R, Qaseem A, Snow V, et al. Diagnosis and treatment of low
back pain: a joint clinical practice guideline from the American College
of Physicians and the American Pain Society. Ann Intern Med 2007;
147:478491.
2 Chapman JR, Norvell DC, Hermsmeyer JT, et al. Evaluating common
outcomes for measuring treatment success for chronic low back pain.
Spine 2011; 36 (21 Suppl):S54S68.
3 Kalso E, Allan L, Dobrogowski J, et al. Do strong opioids have a role in the
early management of back pain? Recommendations from a European
expert panel. Curr Med Res Opin 2005; 21:18191828.
4 Reid MC, Engles-Horton LL, Weber MB, et al. Use of opioid medications for
chronic noncancer pain syndromes in primary care. J Gen Intern Med
2002; 17:173179.
5 Deyo RA. Conservative therapy for low back pain. Distinguishing useful
from useless therapy. JAMA 1983; 250:10571062.
6 Amlie E, Weber H, Holme I. Treatment of acute low-back pain with
piroxicam: results of a double-blind placebo-controlled trial. Spine 1987;
12:473476.
7 van Tulder MW, Koes BW, Bouter LM. Conservative treatment of acute and
chronic nonspecific low back pain. A systematic review of randomized
controlled trials of the most common interventions. Spine 1997; 22:2128
2156.
8 Portenoy RK, Bennett DS, Rauck R, et al. Prevalence and characteristics
of breakthrough pain in opioid-treated patients with chronic noncancer
pain. J Pain 2006; 7:583591.
9 Rossignol M, Abenhaim L, Bonvalot Y, et al. Should the gap be filled
between guidelines and actual practice for management of low back pain in
primary care? The Quebec experience. Spine 1996; 21:28932898.
10 Stein C, Mendl G. The German counterpart to McGill Pain Questionnaire.
Pain 1988; 32:251255.
11 Wiesinger GF, Nuhr M, Quittan M, et al. Cross-cultural adaptation of the
Roland-Morris questionnaire for German-speaking patients with low back
pain. Spine 1999; 24:10991103.
12 Frost M, Stuckey S, Smalley LA, Dorman G. Reliability of measuring trunk
motions in centimeters. Phys Ther 1982; 62:14311437.
13 Macrae IF, Wright V. Measurement of back movement. Ann Rheum Dis
1969; 28:584589.
Eur J Anaesthesiol 2014; 31:3540
Copyright European Society of Anaesthesiology. Unauthorized reproduction of this article is prohibited.
14 Enright PL, McBurnie MA, Bittner V, et al. The 6-min walk test: a quick
measure of functional status in elderly adults. Chest 2003; 123:387398.
15 Podsiadlo D, Richardson S. The timed Up & Go: a test of basic functional
mobility for frail elderly persons. J Am Geriatr Soc 1991; 39:142148.
16 Duncan PW, Weiner DK, Chandler J, Studenski S. Functional reach:
a new clinical measure of balance. J Gerontol 1990; 45:M192M197.
17 Benedetti F, Arduino C, Costa S, et al. Loss of expectation-related
mechanisms in Alzheimers disease makes analgesic therapies less
effective. Pain 2006; 121:133144.
18 Petersen GL, Finnerup NB, Nrskov KN, et al. Placebo manipulations
reduce hyperalgesia in neuropathic pain. Pain 2012; 153:12921300.
19 Alphs L, Benedetti F, Fleischhacker WW, Kane JM. Placebo-related effects in
clinical trials in schizophrenia: what is driving this phenomenon and what can
be done to minimize it? Int J Neuropsychopharmacol 2012; 15:10031014.
20 Lee H-F, Hsieh J-C, Lu C-L, et al. Enhanced affect/cognition-related brain
responses during visceral placebo analgesia in irritable bowel syndrome
patients. Pain 2012; 153:13011310.
21 Ott E, Nussmeier NA, Duke PC, et al. Efficacy and safety of the
cyclooxygenase 2 inhibitors parecoxib and valdecoxib in patients
undergoing coronary artery bypass surgery. J Thorac Cardiovasc Surg
2003; 125:14811492.
22 van Tulder MW, Scholten RJ, Koes BW, Deyo RA. Nonsteroidal anti-
inflammatory drugs for low back pain: a systematic review within the
framework of the Cochrane Collaboration Back Review Group. Spine
2000; 25:25012513.
23 Roelofs PD, Deyo RA, Koes BW, et al. Nonsteroidal anti-inflammatory
drugs for low back pain: an updated Cochrane review. Spine 2008;
33:17661774.
24 Kuijpers T, van Middelkoop M, Rubinstein SM, et al. A systematic review on
the effectiveness of pharmacological interventions for chronic nonspecific
low-back pain. Eur Spine J 2011; 20:4050.
25 Campbell H, Rivero-Arias O, Johnston K, et al. Responsiveness of
objective, disease-specific, and generic outcome measures in patients
with chronic low back pain: an assessment for improving, stable, and
deteriorating patients. Spine 2006; 31:815822.
26 Deyo RA, Andersson G, Bombardier C, et al. Outcome measures for
studying patients with low back pain. Spine 1994; 19 (18 Suppl):2032S
2036S.