Rheumatology 2008;47:1741–1746 doi:10.

1093/rheumatology/ken317
Advance Access publication 12 August 2008

Review

Amitriptyline in the treatment of fibromyalgia: a systematic review

of its efficacy
B. Nishishinya1,2,3, G. Urrútia1,4, B. Walitt5, A. Rodriguez6, X. Bonfill1,4, C. Alegre6 and G. Darko5

The objective of this study was to assess the efficacy and safety of amitriptyline as a treatment of FM. A comprehensive computerized
search in Medline (Pubmed), EMBASE and The Cochrane Library was performed. Randomized controlled trials (RCTs) comparing
amitriptyline vs placebo in adult patients suffering from FM were identified, the methodological quality was assessed and the results of the
main outcomes were evaluated. Ten RCTs were identified. Large clinical variability and statistical heterogeneity precluded quantitative meta-
analysis. Overall, the study quality was moderate to high. Amitriptyline 25 mg/day (six RCTs) demonstrated a therapeutic response compared
with placebo in the domains of pain, sleep, fatigue and overall patient and investigator impression. This benefit was generally seen at 6–8
weeks of treatment but no effect was noted at 12 weeks. Amitriptyline 50 mg/day (four RCTs) did not demonstrate a therapeutic effect
compared with placebo. Neither dose of amitriptyline had an effect on tender points count. No clear statements on adverse events with
amitriptyline can be made due to inconsistencies in data among the studies. A definitive clinical recommendation regarding the efficacy of
amitriptyline for FM symptoms cannot be made. There is some evidence to support the short-term efficacy of amitriptyline 25 mg/day in FM.
There is no evidence to support the efficacy of amitriptyline at higher doses or for periods >8 weeks. More stringent RCTs with longer follow-
up periods are required to determine the long-term efficacy and safety of the amitriptyline and define its role in the multidisciplinary
management of FM.

KEY WORDS: Fibromyalgia, Amitriptyline, Randomized controlled trials, Systematic review, Tricyclic anti-depressants.

Introduction effectiveness of TCAs is related to modulation of the monoamine

FM is a common chronic pain condition that affects >2% of the
adult population in the developed world [1–4]. FM is a chronic
condition of pain, stiffness and tenderness of the muscles, tendons
and joints in the absence of clinically apparent pathology. These
painful symptoms often co-exist with restless and unrefreshing
sleep, fatigue, headaches, anxiety, depression, exercise intolerance
and neurocognitive and neuroendocrine dysfunction [5]. FM has
substantial impact on quality of life (QoL) [6] and adversely
affects both social and work functioning [7].
While FM is one of the most common musculoskeletal pain
syndromes, its pathogenesis is not known. Recent evidence
implicates the central nervous system as key in maintaining pain
and other somatic symptoms of FM [8–11]. Current treatment
approaches to FM pain attempt to influence these central
mechanisms. This had led to a range of medical treatments
being used, including anti-depressants, opioids, NSAIDs, seda-
tives, muscle relaxants and anti-epileptics.
Several studies that reported tricyclic anti-depressants (TCAs)
are effective compared with placebo in the treatment of FM
[12–15] established amitriptyline and other TCAs as the corner-
stone of drug therapy in FM for 20 yrs. Amitriptyline is currently
recommended by experts [16] in the context of a multidisciplinary
strategy that includes both pharmacological and non-pharmaco-
logic therapies in the management of FM. It is believed that the
1
Iberoamerican Cochrane Center, Public Health and Clinical Epidemiology
Service, Hospital de la Santa Creu i Sant Pau, Universitat Autònoma de
Barcelona, 2Universitat Autònoma de Barcelona, Barcelona, 3Kovacs
Foundation, Palma de Mallorca, 4CIBER Epidemiologı́a y Salud Pública
(CIBERESP), Spain, 5Washington Hospital Center, WA, USA and 6Hospital de
la Vall d’Hebron, Barcelona, Spain.
Submitted 28 February 2008; revised version accepted 7 July 2008.
Correspondence to: G. Urrútia, Centro Cochrane Iberoamericano, Hospital de la
Santa Creu i Sant Pau, C/Sant Antoni Maria Claret 171, 08041 Barcelona, Spain.
E-mail: gurrutia@santpau.es
1741
ß The Author 2008. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org
neurotransmitters serotonin and norepinephrine [17–19].
However, the role of these neurotransmitters in FM pathogenesis
is still unknown.
This systematic review aims to synthesize the evidence from
randomized controlled trials (RCTs) into a comprehensive
statement about the known effectiveness of amitriptyline as a
symptomatic treatment for FM.
Methods
Selection criteria
We selected studies for inclusion if they met the following criteria:
RCTs that assessed the efficacy of amitriptyline in patients having
a clinical diagnosis of FM or fibrositis by any recognized criteria
[5, 20–23]. Trials needed to be double-blinded and compare any
dose of amitriptyline with placebo. Trials comparing active
treatments without a placebo control arm were excluded.
Literature search
We performed an exhaustive computerized literature searches in
the following databases EMBASE (via OVID; 1991–2007), The
Cochrane Library; 2007, Issue 2 and MEDLINE (via PubMed;
1966–2007). The search combined the terms ‘fibromyalgia’ and
‘fibrositis’ with a validated filter for RCTs [24]. This search
strategy was part of a broader previous global search of all
pharmacological and non-pharmacological treatments of FM,
which is why amitriptyline and TCA were not included as discrete
search terms.
Two independent reviewers screened the search results, select-
ing the studies that fulfilled the inclusion criteria. We examined
databases of ongoing trials and checked the reference lists of
relevant studies to identify any additional trial. All trials captured
by our search that utilized amitriptyline were selected for this
review. There was no language restriction. The only non-English
paper identified was disqualified by eligibility criteria.
1742 B. Nishishinya et al.
Data abstraction and quality assessment
Two independent reviewers abstracted the relevant data from the
included trials, and a third one solved possible disagreements.
Data was extracted regarding patients’ and intervention char-
acteristics, methodological quality and results for each group of
participants.
Methodological quality was assessed by two independent
reviewers by using the Schulz’s scale [25] and the Jadad’s score
[26]. These applications were used to assess the risk of bias in the
results of the study by considering the particulars of random
sequence generation, concealment of randomization assignments,
baseline homogeneity of patient groups and intention to treat
analysis. The quality of each study was classified as ‘high’ or ‘low’,
depending on the Jadad’s score (‘low’ ¼ 0–2 and ‘high’ ¼ 3–5).
Discrepancies were settled through discussions with a third
reviewer. A consensus was reached about the effectiveness and
safety of the intervention for each study.
Results
Description of the studies
Fifteen studies with amitriptyline were identified, 10 of which met
the inclusion criteria [27–36] (n ¼ 615 patients). Three were cross-
over trials [29, 31, 34] while seven had a parallel design [27, 28, 30,
32, 33, 35, 36]. Five trials were excluded [37–41]. There was
agreement between reviewers about the selection of included/
excluded studies. Figure 1 shows the trial flow.
Among the 10 included RCTs that compared amitriptyline vs
placebo, four studies administered the dose of 50 mg/day [27, 28,
30, 31] and six the dose of 25 mg/day [29, 32–36]. Seven of the
studies used the ACR [5] as the diagnostic criteria while the three
studies that predated these consensus criteria used either the
Smythe criteria [20] or the Yunus criteria [21]. Eight of the studies
were 8-week long, one study was 12-week long and one study was
24-week long. Sample sizes ranged from 22 to 126 patients, with a
mean of 61 patients per study. Sample size calculation method was
adequately reported in eight studies. The most common outcomes
measures in the studies included patient global assessment
measured by Visual Analogic Scale (VAS) (9/10), pain measured
by VAS (8/10), tender point counts (8/10) and physician global
assessment measured by VAS (7/10). Other outcomes, including
QoL, functional status and depression, were not routinely
measured. The few studies that did make such measurements
tended to use different scales, further complicating comparisons.
Most studies used acetaminophen or paracetamol for pain control
as a co-intervention. The effect of these treatments on symptoms
was not felt to interfere with comparisons between amitriptyline
and placebo. Six studies were sponsored by medical associations
or foundations, two by a pharmaceutical company and two
studies did not state any funding source.
Baseline demographic characteristics and risks factors were well
balanced in the studies. Most patients were females (82–100%) of
ages that ranged from 36.7 to 53.4 yrs. Between-study differences
were noted in FM duration (2.5–15.6 yrs) and baseline pain VAS
(4.2–7.3).
All studies reported dropout rates that ranged from 3% to
28.8%, with no differences between the amitriptyline and placebo
groups. Two studies [27, 34] showed a dropout rate >20% (22.2
and 28.8%, respectively). The adverse event frequencies were well
described in 7/10 studies.
Table 1 describes the main characteristics of the included
studies.
All included trials were judged to be of high methodological
quality according to the Jadad’s scale (Table 2). Despite this, there
were substantial methodological differences among the trials.
Three of the studies utilized a cross-over design while the other
seven were comparison trials with a parallel design. Only six
3020 references
screened for retrieval
3005 references were
excluded for not fulfilling
inclusion criteria or were
duplicated
15 potentially appropriate
studies to be included in the
Five articles excluded :
review
• Two open RCTs [37, 38]
• Two duplicated data from other studies [39, 40]
• One RCT compared amitriptyline vs laser [41]
10 studies included in the
systematic review
FIG. 1. Identification of eligible RCTs.
studies reported the details of the randomization method and five
studies reported the process of allocation concealment. All studies
were reported to be double-blind but one study did not provide
information about the blinding method used. Three studies
applied an intention-to-treat analysis while one study did not
provide information concerning the population analysis applied.
Due to clinical and statistical heterogeneity and data-reporting
issues of the included studies, our planned meta-analytic approach
had to be abandoned. One major factor was the fundamental
inability to combine the three cross-over trials with the other seven
parallel design trials. This occurred because none of the cross-over
trials provided separate baseline data previous to the cross-
over therapy. In addition, many of the parallel design trials
reported either incomplete results or provided results only in
graphical form (3/7). In most cases, there was not enough raw
data to justify a meta-analysis. Our attempts to contact the
authors to obtain this missing data were not fruitful. Finally, in
the few cases where meta-analysis was thought to be possible (up
to three RCT), a high statistical heterogeneity (I2 > 50–60%) was
identified. We surmise that this occurred due to clinical hetero-
geneity of basal pain levels and differences in FM duration.
Therefore, the forest plots of these results were not informative.
Thus, we have only provided individual study results in tabular
form (Tables 3 and 4).
Clinical results
Overall, pain significantly improved in the group treated with
amitriptyline compared with placebo in 3/8 studies. Improvements
in fatigue (3/6), sleep disturbance (5/7), patient global assessment
(5/9) and physician global assessment (4/7) were also noted.
Tender point counts improved in only 1/8 studies.
Amitriptyline 25 mg was consistently reported to be more
effective compared with placebo than amitriptyline 50 mg.
Amitriptyline 25 mg showed a significant improvement in pain
(3/5) [29, 32, 34], sleep disturbance (4/5) [29, 32–34], fatigue (3/5)
[29, 32, 33] and physician (4/5) [29, 32–34] and patient (5/6) [29,
32–34, 36] global assessment. Morning stiffness improved in one
of two studies [32]. Other less frequent outcomes are presented in
Tables 3 and 4. Furthermore, in the trials where no differences
TABLE 1. Description of the included studies

Dose Study Duration

References N (mg/day) design (weeks) Outcomes Co-interventions
Carette et al. [27] 70 50 RCT, 9 Morning stiffness (min), Pain (VAS 0–10), total myalgic score (TMS Likert 1–5), sleep quality Acetaminophen
parallel (Likert 1–3), patient assessment (Likert 1–5), physician assessment (Likert 1–5)
Carette et al. [28] 126 50 RCT, 24 Pain (VAS 0–10, McGill Pain Questionnaire 0–78), Fatigue (VAS 0–10), sleep (VAS 0–10), Feeling Acetaminophen
parallel on awakening (VAS 0–10), morning stiffness (VAS 0–10), global assessment of fibromyalgia
(VAS 0–10), sickness impact profile (SIP), HAQ (0–3), Arthritis Impact Measurement Scales (AIMS),
physician global assessment (VAS 0–10), total myalgic score (TMS Likert 1–5)
Fors et al. [30] 55 50 RCT, 4 Pain (VAS 0–10), State-Trait Anxiety Inventory-Trait (STAI-T 20–80), Beck Depression Inventory In addition, patients were also
parallel (BDI 0–21), Automatic Negative Thoughts Questionnaire (ATQ-30) Scores randomly assigned to three
psychological interventions:
(i) pleasant guided
imagery (PI) audiotape;
(ii) attention imagery (AI); and

Efficacy of amitriptyline in fibromyalgia

(iii) a control group
with a blank audiotape.
Scudds et al. [31] 36 50 RCT, 4 The total myalgic score (TMS Likert 1–5), pain threshold, pain tolerance, pain (McGill Pain Acetaminophen
cross-over Questionnaire, MGPQ 0–78), patient assessment of global treatment (Likert 1–5)
Carette et al. [29] 22 25 RCT, 8 EEG changes, pain (VAS 0–10), fatigue (VAS 0–10), sleep (VAS 0–10), global severity Acetaminophen
cross-over (VAS 0–10), point tenderness, total myalgic score (TMS Likert 1–5)
Ginsberg et al. [32] 46 25 RCT, 8 Proportion of responders in each treatment group. Paracetamol
parallel A responder must have at least three of the following four criteria: (i) at least 50% improvement
in patient global assessment, (ii) at least 50% improvement of physician global assessment,
(iii) at least 50% improvement in pain and (iv) at least 25% reduction in tender point score
Pain (VAS 0–10), patient assessment of global treatment (VAS 0–10), wake up tired (Likert 1–5),
asleep disturbance (VAS 0–10), fatigue (VAS 0–10), physician global assessment (VAS 0–10),
morning stiffness (min)
Goldenberg et al. 62 25 RCT, 6 Pain (VAS 0–10), morning stiffness (VAS 0–10), fatigue (VAS 0–10), sleep (VAS 0–10), physician Acetaminophen
[33] parallel global evaluation (VAS 0–10), tender point score (Likert 0–2)
Goldenberg et al. 31 25 RCT, 6 Pain (VAS 0–10), morning stiffness (VAS 0–10), fatigue (VAS 0–10), sleep (VAS 0–10), feeling Acetaminophen
[34] cross-over refreshed upon awakening (VAS 0–10), global well-being (VAS 0–10), fibromyalgia impact
questionnaire (FIQ 0–100), Beck Depression Inventory (BDI 0–21), physician global evaluation
(VAS 0–10), tender point score (Likert 0–2)
Hannonen et al. 87 25–37.5 RCT, 12 Proportion of responders as assessed by physicians, global health (VAS 0–10), pain (VAS 0–10), Paracetamol
[35] parallel sleep quality and quantity (VAS 0–10), fatigue (VAS 0–10), sheehan’s disability scales
(VAS 0–10), Nottingham Health Profile (NHP), tender points (0–18), physician’s clinical
impression of the severity (CIS Likert 3, 0,
3), physician’s clinical global impression of tolerability
(CGI Likert 1–4)
Heymann et al. [36] 80 25 RCT, 8 Verbal evaluation scale for global improvement (VSGI Likert 1–5) Acetaminophen
parallel Fibromyalgia impact questionnaire (FIQ 0–100)
Tender points (0–18)

1743
1744 B. Nishishinya et al.

TABLE 2. Assessment of methodological quality of studies

JADAD Losses to Method of Method of

References N score follow-up n/N (%) randomization Concealment blinding ITT analysis Funding source
Carette et al. [26] 70 4 11/70 (15.7) Unclear Unclear Adequate No Arthritis Society
Carette et al. [28] 126 5 30/124 (22.2) Adequate Adequate Adequate Yes Canadian Arthritis Society
and Merk Frosst Canada
Carette et al. [29]a 22 5 2/22 (9) Adequate Unclear Adequate No Canadian Arthritis Society
and Merk Frosst Canada
Fors et al. [30] 55 4 3/55 (5.5) Adequate Adequate Adequate Unclear Norwegian University of
Science and Technology
Ginsberg et al. [32] 46 4 5/46 (10.9) Unclear Unclear Adequate Yes Not specified
Goldenberg et al. [33] 62 3 2/62 (3.2) Unclear Unclear Unclear No Arthritis Foundation,
Multipurpose
Goldenberg et al. [34]a 31 5 2/31 (6.5) Adequate Adequate Adequate No Lot Page Found,
Newton-Wellesley Hospital.
Newton, Massachusetts
Hannonen et al. [35] 87 5 25/87 (28.8) Adequate Adequate Adequate Yes Roche Oy, Finland
Heymann et al. [36] 80 5 10/80 (12.5) Adequate Adequate Adequate No Not specified
Scudds et al. [31]a 36 4 3/36 (8.3) Unclear Unclear Adequate No The Arthritis Society Studentship

a
Cross-over study.

TABLE 3. Results: amitriptyline 50 mg

Patient’s global Pain Tender points Physician’s global Sleep disturbance Fatigue
References Study type assessment (VAS) (0–18 ACR) assessment (VAS) (VAS) (VAS)
Carette Parallel ND between ND between group ND between groups ND between Better AMI (P ¼ 0.02) Not measured
et al. [27] (9 weeks) groups Intragroup dif: AMI Intragroup dif: AMI groups (Improvement:
yes, Pcb no no, Pcb no 70% AMI versus
40% Pcb)
Carette Parallel ND between ND between groups ND between ND between ND between ND between
et al. [28] (24 weeks) groups Intragroup dif: AMI groups groups groups groups
yes, Pcb yes
Fors Parallel Not measured ND between Not measured Not measured Not measured Not measured
et al. [30] (4 weeks) groups
Scudds Cross-over ND between groups Not measured Not measured Not measured Not measured Not measured
et al. [31] (4 weeks) Intragroup dif: AMI
yes, Pcb no

ND: No Difference; AMI: Amitriptyline; Pcb: Placebo Intragroup differences: not specified when not reported. Information in bold indicates positive results where the difference observed between
amitriptyline and placebo is statistically significant.

between amitriptyline 25 mg and placebo were observed, a
statistically significant improvement was observed between base-
line and final measurements within the amitriptyline arm but not
in the placebo arm. There were no reports of either the lack of
significant improvement or significant worsening of symptoms
with amitriptyline 25 mg.
Interestingly, the clinical effects seen with amitriptyline 25 mg
were not seen with amitriptyline 50 mg. Amitriptyline 50 mg only
demonstrated significant improvement in sleep disturbance
compared with placebo in one of the two studies [27]. No
significant differences were observed in any other outcomes
assessed in the four trials using this dose. The data regarding
differences between baseline and final measurements within each
treatment arm is both heterogeneous and scant, making mean-
ingful comment on intragroup differences difficult.
Some of the standard measurements currently used in FM trials
were not well represented in the included studies. The
Fibromyalgia Impact Questionnaire (FIQ) was assessed in only
two studies with contradictory results [34, 36]. Other standard
outcome measures of functional ability, QoL and psychological
status were also poorly represented. Only four studies assessed
these alternative outcomes and they did so in a heterogeneous
manner.
Only six studies rigorously reported adverse events. In those
studies, the mean adverse event rate was 51.84% (2.8–95%) with
any dose of amitriptyline and 36.63% (2.8–80%) in the placebo
group. Three studies had a large percentage of adverse effects even
in the placebo group (62–80%), while two had very few events or
none in that group. Moreover, the adverse events rate was greater
in the placebo group compared with the amitriptyline group in
two studies, further casting doubts on the reliability of data
collection and reporting. However, all adverse effects reported
were either mild to moderate; no severe or life threatening events
such as cardiac arrhythmias were reported. The most frequently
reported adverse events were somnolence, dry mouth, gastro-
intestinal symptoms and weight gain. There were no differences in
the percentage of patients that withdrew from the studies due to
treatment side-effects between any dose of amitriptyline and
placebo groups.
Discussion
This systematic review was designed to determine the efficacy and
safety of amitriptyline as a treatment of FM symptoms.
Symptomatic improvement was seen in the amitriptyline 25 mg
studies for most outcome variables. No such benefits were seen
with amitriptyline 50 mg. However, problems with the quality of
included studies and limitations of the analysis make definitive
statements premature.
The interesting difference in efficacy by dose was not an
anticipated finding in this study. We had expected a dose-related
effect on FM symptoms with amitriptyline. A normal dose–
response curve has been demonstrated in the treatment of
depression and neuropathy with amitriptyline; the low doses of
amitriptyline used in FM therapy are typically subtherapeutic in
the treatment of the aforementioned disorders [42, 43]. This lack
of efficacy seen at higher doses of amitriptyline in our systematic
review does not appear to be caused by dose-related increases in
 Efficacy of amitriptyline in fibromyalgia 1745

Better AMI (P < 0.001)

Better AMI (P < 0.001)

ND: No Difference; AMI: Amitriptyline; Pcb: Placebo; Intragroup differences: not specified if not reported; Mean values (S.D.) for AMI and Pcb groups are provided at the end of the study;
AMI: Change from baseline in the AMI group (in order to assess the
Better AMI (P < 0.05)
side-effects or drop-outs. In fact, no important differences were

No raw data (graphs)

ND between groups
ND between groups
Fatigue (VAS)
noted in side-effects or drop-outs between the 25 mg and the 50 mg

AMI yes, Pcb no

AMI: 5.62  3.07

(
AMI:
2.22)
dosing regimen. Prior studies do bolster this observation. More
Pcb: 7.64  1.8

(
AMI:
3.5)

(
AMI:
3.2)

Not measured
AMI: 3.8  2.5
Pcb: 5.9  2.2

Intragroup dif:
potent monoamine inhibitors, such as monoamine oxidase
inhibitors (MAOIs) have failed to provide symptomatic relief in
FM [40, 44], while less potent monoamine inhibitors, such as
tramadol, have been shown to help [45]. If this dose-related
observation is demonstrated in future studies, it may provide a
Better AMI (P ¼ 0.003)

Better AMI (P < 0.001)

Better AMI (P < 0.001)

hint into the physiological underpinnings of FM.
Better AMI (P < 0.05)

No raw data (graphs)

Sleep disturbance

ND between groups
These results may, in part, explain one of the notorious clinical

Pcb: 74.6  2 3.9

AMI yes, Pcb no

AMI: 3.93  3.14

AMI: 57.0  34.8

problems that bedevil FM treatment. Tachyphylaxis with
Pcb: 6.51  2.69
(
AMI:
3.56)
(VAS)

(
AMI:
2.6)

(
AMI:
3.9)

Not measured
AMI: 2.6  3.1
Pcb: 5.1  3.0

Intragroup dif:
(
AMI:
11)
amitriptyline is a well-known clinical problem, with amitriptyline
often losing its clinical efficacy after 3 months [46]. This study
suggests that dose escalation practices may contribute to this
problem. However, the short duration of the RCTs prevent any
definitive comment about this issue or other clinical questions
about the long-term efficacy and safety in the treatment of the
Better AMI (P < 0.001)

Better AMI (P < 0.001)

Better AMI (P < 0.05)

Better AMI (P ¼ 0.04)

No raw data (graphs)

chronic symptoms of FM. Future trials should focus on

assessment (VAS)
Physician’s global

(
AMI: aprox –4.0)

ND between groups

determining both the proper dose and the long-term efficacy

AMI yes, Pcb no
AMI: 4.81  2.81

AMI: 64.2  25.2

Pcb: 6.36  1.59

Pcb: 74.7  19.9

(i AMI:
1.59)

and safety of amitriptyline in the treatment of FM.

(
AMI:
3.5)

(
AMI:
0.9)

Not measured
AMI: 3.5  2.1

Intragroup dif:
Pcb: 5.9  1.7

Adverse events were frequently reported with amitriptyline at

all doses. The events appear to be generally mild and an infrequent
cause of study withdrawal. No severe adverse events, such as
arrhythmia, were reported in any of the trials. However, the
disparity in adverse event rates among the studies makes
Better AMI (P < 0.001)

interpretation challenging. The wide variability suggests that

relevance of the effect). Information in bold indicates positive results where the difference observed between amitriptyline and placebo is statistically significant.
ND between groups
ND between groups

ND between groups

ND between groups

ND between groups

these estimates of adverse events are a function of study design

Tender points

AMI yes, Pcb yes

(0–18 ACR)

AMI yes, Pcb no

AMI yes, Pcb no

AMI no, Pcb no

and data collection and may not well reflect the adverse event rate
AMI: 10.2  4.8
Pcb: 13.9  3.1
(
AMI:
4.6)

Intragroup dif:
Intragroup dif:

Intragroup dif:

Intragroup dif:

in general clinical use. Also, the adverse events rate was greater in
the placebo group compared with the amitriptyline group in two
studies, further casting doubts on the reliability of data collection
and reporting. However, few studies using a dose of 25 mg in the
treatment of chronic pain have showed that amitriptyline has no
substantial adverse effects at this low dose [47].
Better AMI (P < 0.001)
Better AMI (P < 0.05)

Better AMI (P ¼ 0.02)

The limitations of this systematic review need to be considered

ND between groups
ND between groups
Pain (VAS)

in interpreting these results. First, most trials did not assess for
AMI yes, Pcb no

AMI yes, Pcb no

AMI: 5.07  3.22

AMI: 64.4  28.3

Pcb: 7.13  2.41

Pcb: 81.5  16.5

(i AMI:
2.05)

(
AMI:
3.5)

(
AMI:
4.0)

Not measured
AMI: 3.8  2.4

Intragroup dif:

Intragroup dif:
Pcb: 7.0  1.3

depression or anxiety disorders. Both depression and anxiety are

commonly associated with FM [48], making it a possible
confounder in these studies. There were also important issues
regarding methodological quality of the RCTs reviewed. While
the included studies were rated as high methodological quality
according to the Jadad’s scale, this measure does not capture
Better AMI (P < 0.001)

Better AMI (P < 0.001)

Better AMI (P ¼ 0.009)

Better AMI (P < 0.05)

Better AMI (P ¼ 0.02)

No raw data (graphs)

many quality issues important in performing a systematic review.

assessment (VAS)

ND between groups

(Improvement: 87%
Patient’s global

Issues related to length of follow-up, sample size, drop-out rates

AMI: 61.60  29.5
Pcb: 76.80  24.8

AMI vs 55% Pcb)

AMI yes, Pcb no
AMI: 5.47  3.03
Pcb: 7.11  2.14
(i AMI:
1.80)

AMI: 3.90  2.3

Pcb: 6.80  1.8

and data reporting are essential to meaningful systematic reviews

(
AMI:
3.8)

(
AMI:
3.0)

(
AMI:
5.0)

Intragroup dif:

are not captured by current quality scales. Overall, these RCTs

would be rated as fair to poor on the aforementioned issues [49],
which would undermine the strength of the apparently favourable
results for amitriptyline 25 mg. Finally, while a consistent
therapeutic effect with amitriptyline 25 mg was seen in nearly all
outcomes, the clinical relevance of this effect remains unclear. Due
to problems with heterogeneity and data reporting, we were only
able to perform a qualitative analysis based on reported P-values
AMIþnaproxen vs naproxen vs
Compares four arms: AMI vs

from each individual trial. We would have much preferred to been

able to combine the raw data together to perform a quantitative
Study type

Cross-over (6 weeks)
Cross-over (8 weeks)

meta-analysis. Such an analysis would have provided a more

Parallel (12 weeks)

precise estimate of the effect size of amitriptyline on the different

Parallel (8 weeks)
Parallel (8 weeks)

Parallel (6 weeks)
TABLE 4. Results: amitriptyline 25 mg.

FM symptom domains. These estimates would have provided a

better understanding of the clinical relevance of the statistical
improvements reported.
placebo

Our methodology only considers amitriptyline as a mono-

therapy, which gives little insight into its efficacy as part of a
multi-disciplinary treatment plan that couples combinations of
pharmacological therapies with non-pharmacological cognitive
and physical therapies that are currently recommended in clinical
Goldenberg

Goldenberg
References

Hannonen

Heymann
et al. [29]

et al. [32]

et al. [33]

et al. [34]

et al. [35]

et al. [36]
Ginsberg

treatment guidelines. To date, only observational studies have

Carette

examined the use of amitriptyline as part of a multi-disciplinary

programme. These studies tend to report favourable results,
1746 B. Nishishinya et al.
but suffer from major biases that preclude their inclusion in this
systematic review.
In conclusion, there is some evidence supporting the efficacy of
amitriptyline 25 mg for the short-term treatment of FM symp-
toms, although better studies are needed to specify the magnitude
of the effect and its clinical relevance. There is no evidence
supporting the use of higher doses of amitriptyline or using it for
periods longer than 8 weeks.
Rheumatology key messages
 Despite its consistent effect, the evidence supporting the use of
amitriptyline 25 mg for the short-term treatment of FM symptoms is
at best moderate or even poor.
 The clinical relevance of these findings remains uncertain.
 There is no evidence to support the use of amitriptyline at higher
doses or for periods >8 weeks.
Acknowledgements
Funding: The authors received a grant by the Agència d’Avaluació
de Tecnologia i Recerca Mèdiques (146/24/2004), Catalonia.
Disclosure statement: The authors have declared no conflicts of
interest.
References
1 Wolfe F, Cathey MA. Prevalence of primary and secondary fibrositis. J Rheumatol
1983;10:965–8.
2 Wolfe F, Cathey MA. The epidemiology of tender points: a prospective study of 1520
patients. J Rheumatol 1985;12:1164–8.
3 Wolfe F, Ross K, Anderson J, Russell IJ, Hebert L. The prevalence and
characteristics of fibromyalgia in the general population. Arthritis Rheum
1995;38:19–28.
4 Croft P, Schollum J, Silman A. Population study of tender point counts and pain as
evidence of fibromyalgia. Br Med J 1994;309:696–9.
5 Wolfe F, Smythe HA, Yunus MB et al. The American College of Rheumatology 1990
criteria for the classification of fibromyalgia. Report of the multicenter criteria
committee. Arthritis Rheum 1990;33:160–72.
6 Bernard AL, Prince A, Edsall P. Quality of life issues for fibromyalgia patients.
Arthritis Care Res 2000;13:42–50.
7 Paulson M, Norberg A, Soderberg S. Living in the shadow of fibromyalgic pain: the
meaning of female partners’ experiences. J Clin Nurs 2003;12:235–43.
8 Bennett RM. The rational management of fibromyalgia patients. Rheum Dis Clin
North Am 2002;28:181–99.
9 Clauw DJ, Crofford LJ. Chronic widespread pain and fibromyalgia: what
we know, and what we need to know. Best Pract Res Clin Rheumatol
2003;17:685–701.
10 Mease P. Fibromyalgia syndrome: review of clinical presentation, pathogenesis,
outcome measures, and treatment. J Rheumatol Suppl 2005;75:6–21.
11 Price DD, Staud R. Neurobiology of fibromyalgia syndrome. J Rheumatol Suppl
2005;75:22–8.
12 Arnold LM, Keck PE Jr, Welge JA. Antidepressant treatment of fibromyalgia. A meta-
analysis and review. Psychosomatics 2000;41:104–13.
13 Goldenberg DL. Pharmacological treatment of fibromyalgia and other chronic
musculoskeletal pain. Best Pract Res Clin Rheumatol 2007;21:499–511.
14 O’Malley PG, Balden E, Tomkins G, Santoro J, Kroenke K, Jackson JL. Treatment
of fibromyalgia with antidepressants: a meta-analysis. J Gen Intern Med
2000;15:659–66.
15 Rossy LA, Buckelew SP, Dorr N et al. A meta-analysis of fibromyalgia treatment
interventions. Ann Behav Med 1999;21:180–91.
16 Buckhardt CS, Goldenberg D, Crofford L et al. Guideline for the management of
fibromyalgia syndrome pain in adults and children. Glenview, IL: American Pain
Society (APS), 2005.
17 Goldenberg DL. A review of the role of tricyclic medications in the treatment of
fibromialgia syndrome. J Rheumatol Suppl 1989;19:137–9.
18 Godfrey RG. A guide to the understanding and use of tricyclic antidepressants in the
overall management of fibromyalgia and other chronic pain syndromes. Arch Intern
Med 1996;156:1047–52.
19 Moldofsky H, Scarisbrick P, England R, Smythe H. Musculosketal symptoms and
non-REM sleep disturbance in patients with ‘‘fibrositis syndrome’’ and healthy
subjects. Psychosom Med 1975;37:341–51.
20 Smythe HA. Fibrositis and other diffuse musculoskeletal syndromes. In: Kelley WN,
Harris ED Jr, Ruddy S, Sledge CB, eds. Textbook of rheumatology, 1st edition,
Philadelphia: WD Saunders, 1981:485–93.
21 Yunus M, Masi AT, Calabro JJ, Miller KA, Feigenbaum SL. Primary fibromyalgia
(fibrositis): clinical study of 50 patients with matched normal controls. Semin Arthritis
Rheum 1981;11:151–71.
22 Yunus M, Masi AT, Calabro JJ, Shah IK. Primary fibromyalgia. Am Fam Physician
1982;25:115–21.
23 Yunus M. Primary fibromyalgia syndrome: current concepts. Compr Ther
1984;10:21–8.
24 Glanville JM, Lefebvre C, Miles JN, Camosso-Stefinovic J. How to identify
randomized controlled trials in MEDLINE: ten years on. J Med Libr Assoc
2006;94:130–6.
25 Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias.
Dimensions of methodological quality associated with estimates of treatment effects
in controlled trials. J Am Med Assoc 1995;273:408–12.
26 Jadad AR. Assessing the quality of reports of randomized clinical trials: is blinding
necessary? Control Clin Trials 1996;17:1–12.
27 Carette S, Mc.Cain G, Bell D, Fam A. Evaluation of Amitriptyline in primary fibrositis.
A double-blind, placebo-controlled study. Arthritis Rheum 1986;29:655–9.
28 Carette S, Bell M, Reynolds J et al. Comparison of amitriptyline, cyclobenzaprine,
and placebo in the treatment of fibromyalgia. A randomized, double-blind clinical trial.
Arthritis Rheum 1994;31:32–40.
29 Carette S, Oakson G, Guimont C, Steriade M. Sleep electroencephalography and the
clinical response to amitriptyline in patients with fibromyalgia. Arthritis Rheum
1995;9:1211–7.
30 Fors EA, Sexton H, Gotestam KG. The effect of guided imagery and amitriptyline on
daily fibromyalgia pain: a prospective, randomized, controlled trial. J Psychiatr Res
2002;36:179–87.
31 Scudds RA, McCain GA, Rollman GB, Harth M. Improvements in pain respon-
siveness in patients with fibrositis after successful treatment with amitriptyline.
J Rheumatol Suppl 1989;19:98–103.
32 Ginsberg F, Mancaux A, Joos E, Vanhove P, Famaey JP. A randomized placebo-
controled trial of sustained-release amitriptyline in primary fibromyalgia. J Muscul
Pain 1996;4:37–47.
33 Goldenberg D, Felson D, Dinerman H. A randomized, controlled trial of amitriptyline
and naproxen in the treatment of patients with fibromyalgia. Arthritis Rheum
1986;29:1371–7.
34 Goldenberg D, Mayskiy M, Mossey C, Ruthazer R, Schmid C. A randomized, double-
blind crossover trial of fluoxetine and amitriptyline in the treatment of fibromyalgia.
Arthritis Rheum 1996;39:1852–9.
35 Hannonen P, Malminiemi K, Yli-Kerttula U, Isomeri R, Roponen P. A randomized,
double-blind, placebo-controlled study of moclobemide and amitriptyline in the
treatment of fibromyalgia in females without psychiatric disorder. Br J Rheumatol
1998;37:1279–86.
36 Heymann RE, Helfenstein M, Feldman D. A double-blind, randomized controlled
study of amitriptyline, nortriptyline and placebo in patients with fibromyalgia. An
analysis of outcome measures. Clin Exp Rheumatol 2001;19:697–702.
37 Isomeri R, Mikkelsson M, Latikka P, Kammonen K. Effects of amitriptyline and
cardiovascular Fitness Training on pain in Patients with Primary Fibromyalgia.
J Muscul Pain 1993;1:253–60.
38 Jaeschke R, Adachi J, Guyatt G, Keller J, Wong B. Clinical usefulness of amitriptyline
in fibromyalgia: the results of 23 N-of-1 randomized controlled trials. J Rheumatol
1991;18:447–51.
39 Heymann RE, Quaresma M, Helfenstein M, Feldman D. A double-blinded,
randomized, controlled study between amitriptyline, nortriptyline and placebo in
patients with fibromyalgia: analysis of the outcome measures. Rev Bras Reumatol
1998;38:119–27.
40 Hannonen P, Malminiemi K, Yli-Kerttula U, Isomeri R, Roponen P. A randomized
double-blind placebo controlled study of moclobemide and amitriptyline in fibro-
myalgia. Scand J Rheumatol 1998;27:246.
41 Gur A, Karakoc M, Nas K, Cevik R, Sarac J, Ataoglu S. Effects of low power
laser and low dose amitriptyline therapy on clinical symptoms and quality of
life in fibromyalgia: a single-blind, placebo-controlled trial. Rheumatol Int
2002;22:188–93.
42 Saarto, PJ Wiffen. Antidepressants for neuropathic pain. Cochrane Database of
Systematic Reviews 2005; CD005454. DOI: 10.1002/14651858.CD005454.
43 Mihajlovic GS, Milovanovic DR, Jankovic SM. Comparison of efficacy and safety
between individualized and empiric dose regimen of amitriptyline in the treatment of
major depressive episode. Psychiatry Clin Neurosci 2003;57:580–5.
44 Nicolodi M, Sicuteri F. Fibromyalgia and migraine, two faces of the same mechanism.
Serotonin as the common clue for pathogenesis and therapy. Adv Exp Med Biol
1996;398:373–9.
45 Bennett RM, Kamin M, Karim R, Rosenthal N. Tramadol and acetaminophen
combination tablets in the treatment of fibromyalgia pain: a double-blind, randomized,
placebo-controlled study. Am J Med 2003;114:537–45.
46 Clauw DJ, Crofford LJ. Chronic widespread pain and fibromyalgia: what we know,
and what we need to know. Best Pract Res Clin Rheumatol 2003;17:685–701.
47 McQuay HJ, Carroll D, Glynn CJ. Low dose amitriptyline in the treatment of chronic
pain. Anaesthesia 1992;47:646–52.
48 Burckhardt CS, Clark SR, Bennett RM. The Fibromyalgia Impact Questionnaire:
development and validation. J Rheumatol 1991;18:728–33.
49 Jüni P, Witschi A, Bloch R, Egger M. The hazards of scoring the quality of clinical
trials for meta-analysis. J Am Med Assoc 1999;282:1054–60.