CLINICAL REVIEW ARTICLE

Antidepressants in Inammatory Bowel Disease:

A Systematic Review
Benjamin J. D. Macer, MSc,* Stephanie L. Prady, PhD,* and Antonina Mikocka-Walus, PhD* ,

Background: Antidepressants are commonly used to treat symptoms of anxiety and depression in inammatory bowel disease (IBD). Recent studies
suggest a link between IBD activity and an individuals emotional state which raises the possibility that antidepressants may potentially modify the
disease course of IBD. This systematic review thus primarily aims to evaluate the efcacy of antidepressants on IBD activity, and secondarily, on anxiety
and depression.
Methods: MEDLINE, EMBASE, Cochrane (IBD Group), CINAHL, AMED, PsycINFO, and OpenGrey were searched from 1990 onward with no
restrictions on study design. A quality appraisal was conducted using several scales as appropriate for each study design. A narrative synthesis was also
conducted.
Results: Fifteen eligible studies included in the review (1 randomized controlled trial, 2 cohorts, 1 casecontrol, 1 cross-sectional survey, 1 qualitative, 2
audits, 1 case series, and 6 case reports) examined a range of antidepressants. Twelve studies suggested that antidepressants have a positive impact on
IBD course. Nine studies reported anxiety and depression as an outcome, of these 8 reported benecial effects of antidepressants. Most of the studies
were deemed to be at low risk of bias, apart from the case reports, which were at high risk of bias.
Conclusions: This research indicates that antidepressants may have a benecial effect on IBD course. However, it is currently not possible to determine
their efcacy for certain because of the lack of randomized trials. Further trials using objective measures of IBD activity, longer follow-up periods, and
larger sample sizes are needed.
(Inamm Bowel Dis 2017;23:534550)
Key Words: antidepressants, anxiety, depression, inammatory bowel disease, systematic review

D epression and anxiety have a negative effect on disease

course in inammatory bowel disease (IBD). A recent sys-
tematic review of 86 studies found that adults with IBD are more
likely to develop anxiety and depression before IBD onset, and
rates of anxiety and depression are higher in patients with IBD
than in the general population, and higher in those with active IBD
compared with inactive IBD (66.4% versus 28.2% respectively for
anxiety, and 34.7% versus 19.9% for depression1).
Antidepressants are often used to treat the anxiety and
depression that is commonly experienced by patients with IBD,
a case-note audit found that 28.9% of patients with IBD in
a public tertiary hospital had used antidepressants at some point
in their life.2 Antidepressants have also been shown to be effective
in treating gastrointestinal symptoms associated with some other
Supplemental digital content is available for this article. Direct URL citations
appear in the printed text and are provided in the HTML and PDF versions of this
article on the journals Web site (www.ibdjournal.org).
Received for publication November 14, 2016; Accepted January 17, 2017.
From the *Department of Health Sciences, University of York, York, United King-
dom; and School of Psychology, Deakin University, Burwood, Victoria, Australia.
The authors have no conict of interest to disclose.
Address correspondence to: Antonina Mikocka-Walus, PhD, School of
Psychology, Deakin University, 221 Burwood Highway, Burwood, 3125 Victoria,
Australia (e-mail: mikocka@deakin.edu.au).
Copyright 2017 Crohns & Colitis Foundation of America, Inc.
DOI 10.1097/MIB.0000000000001059
Published online 6 March 2017.
disorders. A systematic review and meta-analysis looking at the
effect of antidepressants and psychological therapies on irritable
bowel syndrome, a functional gastrointestinal disorder, found
antidepressants to have efcacy over placebo in the improvement
of somatic bowel symptoms (relative risk 0.67; 95% CI,
0.580.77) with similar effects observed for both selective sero-
tonin reuptake inhibitors and tricyclic antidepressants.3 A system-
atic review of animal models of colitis has found that desipramine
and uoxetine reduce the risk of colitis and improve inammatory
markers, with little evidence of adverse effects.4
A previous systematic review published 10 years ago
examined the effect of antidepressants in the treatment of IBD
and found 12 publications, none of which were randomized
controlled trials (RCTs).5 The review suggested that 16/20 pa-
tients experienced benecial effects on physical IBD symptoms
as a result of antidepressants, but conclusions were limited
because of the observational nature of the research and very small
samples of patients.
Given that psychological factors play an important role in
IBD activity and antidepressants have been reported to have anti-
inammatory properties6,7, antidepressants have the potential to
be an adjuvant treatment for IBD. Despite the lack of conclusive
evidence on efcacy or effectiveness, antidepressants are already
prescribed in the treatment of somatic IBD symptoms2 and thus it
is timely to review the role they may play in IBD management.

534 | www.ibdjournal.org Inamm Bowel Dis
Volume 23, Number 4, April 2017

Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.
Inamm Bowel Dis
Volume 23, Number 4, April 2017
The aim of this study is to (1) examine the evidence on the
impact of antidepressants on disease activity, and (2) their impact
on comorbid symptoms of anxiety and depression in IBD.
MATERIALS AND METHODS
Search Strategy
We searched MEDLINE, EMBASE, Cochrane (Cochrane
Inammatory Bowel Disease and Functional Bowel Disease
Group), CINAHL, AMED, and PsycINFO. Search strategies
were compiled with the assistance of an academic librarian.
Articles published before 1990 were not included. No restrictions
were placed on language during the searches although for
practical reasons, it was only possible to include English language
articles. An example search strategy is presented in the Appendix,
Supplemental Digital Content 1, http://links.lww.com/IBD/B467.
Searches were conducted on third June 2016 by one author (B.J.
D.M.).
The reference lists of included articles were scanned and 1
journal (Gastroenterology) hand searched. Titles and abstracts of
retrieved studies were screened for inclusion. The full text of
potentially relevant articles was obtained, and the inclusion and
exclusion criteria were applied.
Inclusion Criteria
Studies were included if they met the following criteria:
1. Contained human participants, clinically diagnosed with
any form of IBD (i.e., Crohns disease [CD], ulcerative
colitis [UC], or intermediate colitis based on clinical, his-
tological, radiological, or endoscopic criteria).
2. Participants could be any age and any sex.
3. Participants were prescribed or took any of the following
antidepressants: tricyclics, monoamine oxidase inhibitors,
serotonin reuptake inhibitors, serotonin and noradrenaline
reuptake inhibitors, or atypical antidepressants. Antidepres-
sants could be used both with and without other treatments,
apart from other pharmacological psychiatric treatments
(such as anxiolytics). Standard care was assumed.
4. Any comparator.
5. Any study design.
6. Contained an outcome measure of remission or anxiety/
depression outcome (see Outcome Measures below).
Exclusion Criteria
1. Participants were prescribed or took any other form of
medication used to treat depression or anxiety; such as
herbal medicines and anxiolytics alone.
Outcome Measures
For studies to be included in the review, they had to include
at least 1 of the following primary or secondary outcomes:
Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.
Antidepressants in Inammatory Bowel Disease
Primary Outcome
1. Remission measured through changes in disease activity
indices as per respective cut-off values, as dened by study
authors (e.g., Crohns Disease Activity Index [CDAI], Sim-
ple Clinical Colitis Activity Index [SCCAI]), using calpro-
tectin, colonoscopy, or other similar measures (e.g., blood).
Secondary Outcomes
1. Anxiety and depression symptoms, as measured through
using any relevant diagnostic interview technique or
screening scale.
Data Extraction
Data pertaining to the sample, methods, and results were
extracted from each of the included studies by 1 author (BJDM).
Quality Assessment
Other than human participants this review applied no
restrictions on study design; therefore, the variety of study
designs necessitated the use of several different quality assess-
ment tools. The Cochrane Risk of Bias tool was used for
randomized trials;8 this was based on 8 questions which can be
addressed with either Low risk or High risk. The
NewcastleOttawa Scale (NOS) was used for observational
studies (casecontrol and cohorts),9 for which a study can score
a possible of 8 points, a higher score signies a lower risk. The
National Institute of Health (NIH) quality assessment tool was
used for audits, case reports, and case series.10 Another NIH
quality assessment tool was used to assess the quality of cross-
sectional surveys.11 Both of the NIH tools used gave a nal
quality rating of Good, Fair, or Poor. Qualitative studies
were assessed using the Critical Appraisal Skills Programme
(CASP) tool.12 The CASP tool has 10 questions which can be
answered Yes, Cant tell, or No; a Yes would imply
a low risk of bias.
Data Analysis
A narrative synthesis was used to describe and compare the
studies. A meta-analysis was planned but not performed because
of heterogeneity of study design and outcomes.
RESULTS
A total of 2193 studies were retrieved with 1840 screened
after duplicates were removed (Fig. 1). Fifteen studies were
included in the review: 1 placebo-controlled RCT, 1 prospective
and 1 retrospective cohort study, 1 retrospective casecontrol
study, 1 cross-sectional survey, 1 qualitative study, 1 report on
a clinical case note audit, 1 audit, 1 case series, and 6 case reports.
The follow-up period of the studies varied from 6 weeks to 11
years. The majority of the studies were from the United States
www.ibdjournal.org | 535
Macer et al Inamm Bowel Dis
Volume 23, Number 4, April 2017

mid-to-high risk of bias, primarily because it contained an

irritable bowel syndrome comparative cohort which was irrele-
vant for this review. The casecontrol study14 was deemed to be
at low risk of bias. The main weakness of this study was the
representativeness of the participants because they were sampled
from a single tertiary care IBD center in London. In the cross-
sectional survey,15 6 of the 7 relevant categories received
a Yes on the NIH tool. The reasons for not receiving
a Yes on the other category were because it was not possible
to determine if 50% of eligible persons took part in the study.
The study was given an overall quality rating of Good indi-
cating a low risk of bias. The single qualitative study16 met all
the 9 criteria as set out by the CASP assessment tool, and the
study was deemed to be at low risk of bias.
The NIH tool was used to quality assess the 2 audits, the
case series, and the 6 case reports. The report on a case note audit2
met all the criteria apart from length of follow-up, as this was not
applicable; scoring Good overall deeming it at low risk of bias.
The other audit17 was deemed at high risk of bias, only receiving
a Yes in 3 of the 9 categories.
The case series18 was at low risk of bias only being marked
down because the length of follow-up was inadequate. Of the case
reports the study quality was generally poor, so a high risk of bias.
A weakness of all the case reports, of which 2 were abstracts, is
that the outcome measures were not clearly dened, with often
incompletely reported results.

FIGURE 1. PRISMA ow diagram.
(n 8) and Australia (n 3), with 1 study each from England,
Iran, New Zealand, and India.
Quality Assessment
Quality assessments of each individual study are shown in
Tables 14. The RCT13 was at low risk of bias with only high-risk
scores from the sections assessing attrition bias.
Using the NOS for nonrandomized studies, Yanartas et al.
(2016) was at low risk of bias, and Iskandar et al (2014) was at
Narrative Synthesis
Of the 15 included studies, 14 (93%) addressed the primary
outcome measure of remission and 10 (67%) addressed the
secondary outcomes of anxiety/depression. See Table 5 for
a description of each study and Table 6 for results.
RCT
The RCT was conducted between 2013 and 2014 in Iran.13
Forty-four participants were randomly allocated to either be pre-
scribed duloxetine (60 mg once a day) or a placebo for 12 weeks.
Anxiety and depression were measured using the Hospital Anx-
iety and Depression Scale (HADS) and symptom severity using
Lichtiger Colitis Activity Index (LCAI). Five patients were lost to

TABLE 1. Quality Assessment of RCTCochrane Risk of Bias Tool

Incomplete Incomplete
Blinding of Outcome Data Outcome Data
Random Blinding of Outcome Blinding of Addressed Addressed
Sequence Allocation Participants Assessment Outcome (Attrition Bias) (Attrition Bias) Selective
Generation Concealment and Personnel (Detection Bias) Assessment (Short-term (Longer-term Reporting
(Selection (Selection (Performance (Patient-reported (Detection Outcomes Outcomes (Reporting
Bias) Bias) Bias) outcomes) Bias) [26 wk]) [.6 wk]) Bias)

Daghaghzadeh et al13 Low risk Low risk Low risk Low risk Low risk High risk High risk Low risk

536 | www.ibdjournal.org

Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.
Inamm Bowel Dis
Volume 23, Number 4, April 2017 Antidepressants in Inammatory Bowel Disease

follow-up in the intervention group and 4 in the control group,

Total (/8)

Total
(/8)

6
leaving a total of 35 participants (UC: 22; CD: 13) in the analysis.

8
5
Symptom severity signicantly improved in the interven-

Nonresponse
tion group compared with the control group (P 0.02). Depres-

Rate (/1)
Follow-up of
Adequacy of

Cohorts (/1)
sion and anxiety also improved signicantly in the intervention

*
* group compared with the control group (depression P 0.041;
* anxiety P 0.049).

of Ascertainment for
Cohorts
Same Method

Controls (/1)
Enough for Outcomes
Was Follow-up Long

Cases and
The retrospective cohort study included 81 participants
to Occur? (/1)

taking tricyclic antidepressant (UC: 23; CD: 58)19 who were fol-

*
lowed over 11 years using outpatient records from a Gastroenter-
*
*

ology practice in St. Louis, Missouri. Baseline symptom severity

was assessed on a 4-point Likert scale (0 no symptoms to 3
severe, disabling symptoms) with no signicant difference
Assessment of
Exposure (/1)

between disease types. Antidepressant treatment responses were

TABLE 2. Quality Assessment of Cohort Studies and CaseControl StudyNewcastleOttawa Scale

graded using an established 4-point scale (0 no improvement to

Outcome

3 complete satisfaction). Patients with UC responded signi-

*
*

cantly better than patients with CD at rst follow-up (time frame

not stated), mean 1.86 (SEM 0.13) for UC and 1.26 (0.11) for CD
Comparability

(P 0.003). Eighty-three percent of patients with UC had at least

and Controls on

or Analysis (/2)
Comparability

of the Design

a moderate symptomatic improvement on tricyclic antidepressant,

the Basis
of Cases
*

compared with 50% of patients with CD (P 0.01). At the

**

second follow-up (time frame not stated), there was no signicant

difference between the disease types (CD: 1.31 [SEM 0.16]; UC:
Selection

1.47 [0.17], P 0.76) or on whether they had at least a further

moderate symptom response, (CD 56%; UC 40% P 0.16).

*

Denition of
Controls (/1)

The prospective study20 followed 67 patients (UC: 36; CD:

31) from an IBD-specic Gastroenterology outpatient clinic at

Cohort Study

a hospital in Istanbul, between June 2013 and June 2014. The

CDAI and Modied Mayo Score (MMS) were used to measure
*
*

disease activity for CD and UC, respectively, as well as C-reactive

blood count. Anxiety and depression were assessed using HADS
Controls (/1)
Selection of

and Structured Clinical Interview for DSM Disorders (SCID-I).

Nonexposed Cohort (/1)

Antidepressant treatment was not associated with a signif-

Selection of the

icant improvement in CDAI compared with the control group

(mean improvement 262.9 [SD: 99.5], P 0.57), nor was it

*

associated with improvement in MMS (21.6 [3.4], P 0.926).

Representativeness
of the Cases (/1)

However, a signicant improvement was seen in anxiety and

depression when compared with the control group, P 0.001

and P 0.017, respectively.

of Exposed Cohort (/1)

Casecontrol Study
Represent-ativeness

The casecontrol study retrospectively compared 58 partic-

Adequate? (/1)
Is the Case

ipants, 29 (UC: 14; CD: 15) who were sampled from an adult and
Denition
*
*

pediatric IBD center in London, United Kingdom.14 In the inter-

*

vention group (n 29), antidepressants were used to treat mood

disorders; the matched controls (n 29) received no antidepres-
sants therapy; patients were matched based on age, sex, disease
Goodhand et al14

type, medication at baseline, and relapse rate in year 1. Patients

Yanartas et al20
Iskandar et al19
Cohort Studies

CaseControl

were assessed the year before and the year after initiation of
antidepressant therapy.
Study

Outcomes included number of relapses, number of endo-

scopic procedures, number of outpatient attendances and hospital

www.ibdjournal.org | 537

Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.
 538

Macer et al
Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.

| www.ibdjournal.org

TABLE 3. Quality Assessment of Cross-sectional Survey, Audits, Case Series, and Case ReportsNational Institute of Health Tool
Cross-sectional Objective Population Specied and Participation Rate Uniform Selection and Sample Size/Power Exposure(s) Measured Before the Sufcient
Survey Stated Dened? $50% Recruitment Estimate Outcome(s) Timeframe

Mikocka-Walus15 Yes Yes CD Yes N/A N/A N/A

Exposure Exposure Measures Dened, Exposure(s) Outcome Measures Dened, Outcome Confounding Quality
Cross-sectional Appropriately Valid, Reliable and Consistently Assessed Valid, Reliable and Consistently Assessors Attritrion Variables Measured Rating (Good/
Survey Measured Implemented .Once Implemented Blinded #20% and Adjusted for Fair/Poor)

Mikocka-Walus15 Yes Yes N/A Yes N/A N/A N/A Good

Audits, Case- Question/ Population Intervention Outcome Measures Dened, Sufcient Statistical Results Quality
series & Case Objective Specied and Consecutive Subject Clearly Valid, Reliable, and Consistently Length of Methods Well Well Rating (Good/
Report Stated Dened Cases Comparability Described Implemented Follow-up Described Described Fair/Poor)

Mikocka- Yes Yes Yes Yes Yes Yes N/A Yes Yes Good
Walus et al2
Ramos Rivers No No N/A Yes No No Yes Yes No Poor
et al17
Walker et al18 Yes Yes Yes Yes Yes Yes No Yes Yes Good
Kane et al21 Yes No No No Yes No Yes N/A No Poor

Inamm Bowel Dis
Volume 23, Number 4, April 2017

Kast22 Yes Yes N/A N/A Yes No Yes NR Yes Poor
Kast and No Yes No No Yes No CD NR Yes Poor
Altschuler23

CD, cannot determine; N/A, not applicable; NR, not reported.

Inamm Bowel Dis
Volume 23, Number 4, April 2017 Antidepressants in Inammatory Bowel Disease

TABLE 4. Quality Assessment of Qualitative StudyCritical Appraisal Skills Programme Tool

Was the
Was there Recruitment Was the Data Has a Relationship
a Clear Is Was the Research Strategy Collected in Between Researcher Have Ethical
Statement of Qualitative Design Appropriate Appropriate to the a Way that and Participant Been Issues Been
the Aims of the Method to Address the Aims Aims of the Addressed the Adequately Taken Into
Research? Appropriate? of the Research? Research? Research Issue? Considered? Consideration?

Mikocka- Yes Yes Yes Yes Yes Yes Yes

Walus et al16

Is Qualitative Was the Recruitment Strategy Was the Data Is There a Clear
Method Appropriate to the Aims of the Analysis Sufciently Statement of How Valuable is
Appropriate? Research? Rigorous? Findings? the Research?

Mikocka-Walus Yes Yes Yes Yes Showed 1/3 patients to have

et al16 perceived improvements.
Informed future RCTs.

admissions, and number of courses of steroids. Fewer relapses and
courses of steroids in the year after starting an antidepressant were
experienced in the intervention group than in the year before {1
(04) (median [range]) versus 0 (04), P 0.002; 1 (03) versus
0 (04), P , 0.001, respectively}. The controls showed no changes
between years 1 and 2 in relapses (1 [04] versus 1 [03], respec-
tively) or courses of steroids (1 [02] versus 0 [03]). There was
a signicant difference between the 2 groups for number of re-
lapses (P 0.03), but not for course of steroids (P 0.07).
Cross-sectional Survey
The cross-sectional on-line survey, advertised between
March 2012 and April 2013, included 98 participants (UC: 32;
CD: 48; intermediate colitis: 3; missing type: 15) from a non-
clinical population recruited through the Australian IBD advocacy
and support group.15 Participants were required to be taking anti-
depressants or had previously been on antidepressants since their
IBD diagnosis. The aim of the study was to explore the use and
type of antidepressants currently prescribed to IBD sufferers, their
effects on symptoms, and experiences of them.
Participants had been taking antidepressants for an average
of 4 (SD: 3.9) years, with a range of 4 weeks to 15 years. Of those
individuals taking antidepressants, 79% reported perceived im-
provements; however, 67% had observed no change in perceived
disease activity. Disease activity was found to improve in 25% of
participants. The study also showed perceived psychological well-
being had improved in 87% of participants.
Qualitative Study
The qualitative study interviewed 15 participants taking
antidepressants, sampled from The Royal Adelaide Hospital,
a tertiary teaching hospital in South Australia.16 The interviews,
conducted between January and March 2011, were
semistructured, containing open-ended questions relating to IBD
history, reasons for antidepressant therapy and details of the ther-
apy, acceptance of the treatment, side effects, impact on IBD and
quality of life, and attitudes toward taking part in future trials.
The study showed that antidepressants helped disease course
(n 5), reduced pain and frequency of bowel movements (n 3),
and reduced the frequency of symptoms or are up (n 3). Con-
versely, n 10 reported that antidepressants did not inuence dis-
ease course, although the authors did concede that it was difcult to
distinguish between the effectiveness of different treatments. The
study also showed that the majority of the participants had a positive
attitude toward antidepressants (n 9). Twelve of the participants
stated they would take part in further trials; 2 did not want to change
their antidepressant treatment because of their success with it.
Audits
The case-note audit was conducted at the center where the
participants from the qualitative study, mentioned above, were
sampled. This retrospective analysis was from an IBD database at
an Australian tertiary hospital, and assessed participants for type,
frequency, and impact of antidepressant therapy on IBD course.2
The audit showed that from 287 participants (UC: 95; CD:
179; intermediate colitis: 13), 51 (18%) were currently taking
antidepressants. Within the 51 taking antidepressants, 15 (30%)
individuals had inactive disease but presented with symptoms
such as pain or diarrhea, consistent with functional bowel
disorders, 11 (22%) were in full remission with no disease
activity, 2 (0.01%) had active disease, and the data for 23 (45%)
participants were not recorded. Seventy-one patients had a history
of antidepressant use, 45 (63%) were prescribed for anxiety and
depression, or both; 10 (14%) were for somatic complaints, and
no data were available for the remaining 16 (22.5%) patients.
While on antidepressants, 19 (28%) had inactive disease but had

www.ibdjournal.org | 539

Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.
 540

Macer et al
TABLE 5. Summary Table of the Included Studies
Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.

| www.ibdjournal.org

Measurement and Assessment

Author (Year), Depression and
Country Study Design Participants Study Details Disease Type IBD Anxiety Follow-up

Daghaghzadeh Placebo- 35 participants between 1865 Two groups. intervention UC 22; Disease duration, mean (SD): Depression and 12 wk
et al,13 Iran controlled years old (mean [SE] age: group (n 17) took CD 13 intervention6.49 (3.27) anxiety: mean (SD)
RCT 38 [8.08]), with no are up duloxetine 3060 mg once yrs; control8.17 (4.29) score across both
over previous 6 months. per day for 12 weeks. yrs (P 0.538). groups: 9.22 (3.45)
Selected from the GI clinic Control group (n 18) was and 8.17 (4.29),
of Alzahra Hospital placebo controlled, the respectively.
(Isfahan) between 2013 and subjects received placebo in Measured using
2014. the same form and packages HADS.
as duloxetine for the same
length of time. All
participants also received
mesalazine, 24 mg daily.
Experimental group n 17 Randomization: A third-party Symptom severity measured
(47% females); control physician using tables of using LCAI.
group n 18 (44% females) random numbers conducted
the randomization.
Blinding: A psychologist who
was not informed about
grouping of the subjects
assessed questionnaire
scores.
Yanartas et al,20 Prospective 67 participants (43 [64%] Participants had psychiatric UC 36; CDAI and MMS for the Major depression 6 mo
New Zealand cohort women) older than 17 years interviews using SCID-I. CD 31 assessment of disease (43.2%) and
old, mean age was 40.71 6 Participants also had SF-36 activity in patients with CD generalized anxiety
12.71 yrs, followed up in and ASEX tests for or UC, respectively. Along disorder (15%)
the IBD-specic assessing QoL and sexual with CRP, complete blood using HADS.

Inamm Bowel Dis
Volume 23, Number 4, April 2017

gastroenterology outpatient
clinic at Marmara
University Hospital between
June 1, 2013 and June 1,
2014.
Assessments before and after 6
dysfunction. count, and routine blood
bio-chemistry were
collected on all visits
months; 47 completed
antidepressant therapy
(group A), 20 did not (group
B). Most common
antidepressants used were
sertraline (21.0%) and
escitalopram (15.8%).
 Inamm Bowel Dis
Volume 23, Number 4, April 2017
TABLE 5. (Continued)
Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.

Measurement and Assessment

Author (Year), Depression and
Country Study Design Participants Study Details Disease Type IBD Anxiety Follow-up

Iskandar et al,19 Retrospective 81 participants with IBD. mean Outpatient electronic medical UC 23; Baseline symptom severity Data were self-reported. 11 years
USA cohort (SD) age: 41.3 6 1.7; records were reviewed to CD 58 was assessed on a 4-point
69.1% females identify patients over an 11- Likert scale (0 no
year period between July symptoms, 3 severe,
2000 and June 2011. disabling symptoms). AD
treatment responses were
graded using an
established 4-point scale
(0 no improvement; 3
complete satisfaction).
TCA median dose Depression, n 23
(amitriptyline, nortriptyline, (28.4%). Anxiety,
desipramine) 25 mg, range n 5 (6.2%). Both,
10150 mg. TCA dose n 10 (12.3%).
increase by second follow-
up, 29/81 (35.8%).
Currently taking biologics
(22.4%), immuno-
modulators (31.0%) and 5-
ASA (12.1%).
Goodhand Retrospective 58 participants divided equally Patients with IBD using ADs to UC 14; Median age at diagnosis, NR 2 years
et al,14 USA casecontrol into 2 groups (n 29 IBD treat concomitant mood CD 15 (in years (range): AD group
and n 29 matched disorders. Citalopram 20 mg each group) 26 (1372); Controls
controls). Seventeen (2060 mg) and uoxetine 29 (1262). Median
females in each group. From 20 mg (2060 mg) were the disease duration, years
a tertiary adult and pediatric most commonly used ADs; (range): AD group5.2
IBD center located in other SSRIs were used and (140); Controls4.2
London. TCAs, NaSSa, and SNRIs. (131).

Antidepressants in Inammatory Bowel Disease

Patients were already using
corticosteroids, 5-ASA,
immunosuppressive agents,
and anti-TNF.
www.ibdjournal.org |
Controls did not receive ADs
and were matched based on
sex, age at diagnosis 65
years, and disease duration
63 years was sought and
then screened in detail to
match for disease

phenotype, baseline
medications, surgeries, and
relapse rate in year 1.
541
 542
TABLE 5. (Continued)
Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.
| www.ibdjournal.org
Author (Year),
Country Study Design Participants
Mikocka- Cross-sectional 98 participants (76 [78%]
Walus,15 survey female) from a national IBD
Australia advocacy and support group
accessed the survey. Mean
(SD) age: 37.7 (11.9).
Participants were currently
taking a mixture of or solely
conventional and alternative
treatments.
Mikocka-Walus Qualitative 15 participants taking ADs (9 Semistructured interviews were UC 2; CD Time since diagnosis ranged
et al,16 interview [60%] females) selected
Australia from a case-note audit.
Mean (SD) age: 45.8
(17.11) years.
Most common symptoms: pain Open-ended questions were
(86.7%), diarrhea (66.7%),
nausea (33.3%), fatigue
(26.7%), bloating (26.7%),
and difculties tolerating
medications (20.0%).
Study Details Disease Type
Questions in the survey were UC 32;
related to type and dosage of CD 48; IC
ADs; perceived outcome of
the treatment; perspectives
and experiences with the use
of ADs as well as views on
the interactions between AD
treatment and their disease
course; respondents
acceptability of trials with
ADs.
conducted.
asked about IBD history,
reasons for taking Ads, and
details of this therapy (type,
dose, length of treatment,
etc.), acceptance of this
treatment, patients
observations in relation to
side-effects and impact on
IBD (e.g., impact on pain,
frequency of bowel
movement), observed
impact on QoL, attitudes
toward ADs, and attitudes
toward future trials with the
use of ADs.
Macer et al
Measurement and Assessment
Depression and
IBD Anxiety Follow-up
Time with IBD symptoms, As diagnosed by N/A
mean (SD): 13.7 (9.5) a clinician.
3; unknown years. Time since IBD
type 15 diagnosis, mean (SD): 9.2
(8.8).
Depression (n 25);
anxiety (n 10);
both (12).
Self-reported data. N/A
12; colitis of from 3 to 30.5 yrs, mean
undetermined (SD) 16.8 (8.9). The
etiology 1 number of current
symptoms reported per
patient ranged from 1 to
7, mean (SD) 3.5 (2.0).
Depression or
depressed mood,
reported by 10
patients (66.7%), and
Inamm Bowel Dis
Volume 23, Number 4, April 2017
anxiety or anxious
mood, reported by 7
patients (46.7%).
 Inamm Bowel Dis
Volume 23, Number 4, April 2017
TABLE 5. (Continued)
Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.

Measurement and Assessment

Author (Year), Depression and
Country Study Design Participants Study Details Disease Type IBD Anxiety Follow-up

Mikocka-Walus Report on 287 (143 [50%] females). Patients details were collected UC 95; See Table 2 As diagnosed by N/A
et al,2 clinical case- Mean (SD) age: 47 (17). from an IBD database at an CD 179; clinicians.
Australia note audit Australian tertiary hospital. IC 13
Patients taking the following: Details on frequency, type, and Depression (45%);
5-ASA, azathioprine, outcome of AD treatment in combined depression
biologics, corticosteroids, terms of IBD course were and anxiety disorder
metronidazole, salazopyrin, collected. (23.5%)
pain killers, and
benzodiazepines.
Ramos Rivers Audit (Abstract) 855 IBD, mean age 47 6 15 Electronic medical records UC 352; History of GI surgery, N/A 4 yrs
et al,17 USA (422 [52%] females) (EMRs) were used to CD 503 46.7%
identify frequency and
classes of AD use. For the
most frequently used ADs,
differences in QoL (SIBDQ)
and IBD activity between
pts taking ADs and those
who did not during that
same 4-year period were
evaluated.
IBD activity measured using
HBI/UCDAI
Walker et al,18 Case series 8 IBD participants, 18 years Patients interviewed using Not specied GI symptom interview All participants 8 weeks
USA old or older. Selected from NIMH DIS, GI symptom diagnosed with
tertiary care medical faculty interview and the Briere major depression.
in Seattle. Child Maltreatment Conrmed by
Interview (history of Hamilton Depression
childhood abuse and Rating Scale

Antidepressants in Inammatory Bowel Disease

neglect), SF-36, Tri- (HAM-D)
dimensional Personality
Questionnaire.
Patients treated with
paroxetine, 20 mg for rst
month. Second month 2
patients moved 40 mg. At
www.ibdjournal.org |

the end of follow-up patients

re-interviewed, SF-36 and
HAM-D.
543
 544

Macer et al
TABLE 5. (Continued)
Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.

| www.ibdjournal.org

Measurement and Assessment

Author (Year), Depression and
Country Study Design Participants Study Details Disease Type IBD Anxiety Follow-up

Kane et al,21 Case report 4 participants (2 women, Bupropion 100 mg for CD NR As diagnosed by 6 weeks
USA 2 unspecied) depression or smoking a clinician.
cessation for at least 6 Depression (n 2)
weeks.
Kast,22 USA Case report 33-year-old woman. Phenelzine 15 mg 3 times daily CD 18-year history of CD. Has Major depressive NR
for 1 month, then 30 mg undergone 3 bowel episodes and anxiety,
3 times daily after for 2 resections and had 10 as diagnosed by
years. watery bowel movements a clinician.
with severe abdominal
cramping daily.
Currently taking 75 mg
azathioprine, 60 mg
prednisone, and 3
acetaminophen/oxycodone
tablets daily.
Kast and Case report 44-year-old woman. Taking Bupropion 150 mg 3 times CD Woman10-year history of Femaleepisode of Female
Altschuler,23 uoxetine 40 mg every day daily for depression (female) IBD (CDAI202). major depression, at
USA for depression, and and smoking cessation superimposed on least 19
mesalamine 500 mg twice (male). a chronic mild months
a day. depressed state
(dysthymia). As
diagnosed by
a clinician.
45-yr-old man. Man20-year history of Male
IBD with multiple NR
surgeries, including 4

Inamm Bowel Dis
Volume 23, Number 4, April 2017

small bowel resections
(CDAI275).
Scott et al,24 Case report 42-year-old black man. 80 mg/d amitriptyline CD Severe are up of CD, pain Sertraline previously 6 weeks
USA Prescribed 6- administered 8/10 on visual analog prescribed for major
metacaptopurine, intramuscularly scale despite morphine. depression,
prednisolone, and total discontinued after 19 days unsuccessful.
parenteral nutrition. because of pain at injection Amitriptyline was
site. successful.
Afterward 150 mg
amitriptyline gel was
applied to patients chest at
bedtime.
Inamm Bowel Dis
Volume 23, Number 4, April 2017 Antidepressants in Inammatory Bowel Disease

functional symptoms; 12 (17%) had active disease; and 9 (13%)

5-ASA, 5-aminosalicylic acid; ADs, antidepressants; ASEX, Arizona Sexual Experience Scale; CDAI, Crohns Disease Activity Index; CRP, C-reactive protein; DIS, Diagnostic Interview Schedule; GI, gastrointestinal; HADS,
Hospital Anxiety and Depression Scale; HAM-D, Hamilton Depression Rating Scale; HBI, Harvey-Bradshaw Index; IC, intermediate colitis; LCAI, Lichtiger Colitis Activity Index; NaSSa, noradrenergic and specic serotonergic
antidepressants; NR, not reported; QoL, quality of life; SCID-I, Structured Clinical Interview for DSM Disorders; SF-36, Short Form 36; SIBDQ, Short Inammatory Bowel Disease Questionnaire; SNRI, serotonin and noradrenaline
Follow-up

6 weeks
had inactive disease.

NR
The other study (reported as an abstract)17 we have classied
as an audit but may be better described as a series of annually
conducted cross-sectional surveys of patient IBD activity and anti-

6 months of chronic, watery, Severe depression, as

Depression and

Generalised anxiety

depressant use in an IBD clinic. The results showed that in 855 IBD
Anxiety

diagnosed by

diagnosed by
participants (UC: 352; CD: 503), the mean IBD activity decreased
disorder, as

a clinician.

a clinician.
Measurement and Assessment

over 4 years, independent of serotonin reuptake inhibitor use.

Case series
The one case series included18 studied 8 IBD participants from
watery bowel movements

a Gastroenterology tertiary care center in Seattle, attending from

nonbloody diarrhea. 46

March to October 1993. Participants were screened and selected if

they were diagnosed with major depression, using Hamilton Depres-
IBD

sion Rating Scale and Structured Clinical Interview for DSM

Disorders.
per day

Participants were interviewed at baseline using the National

Institute of Mental Health (NIMH) Diagnostic Interview Schedule
reuptake inhibitor; SSRI, selective serotonin reuptake inhibitor; TNF-a, tumour necrosis factoralpha; UCDAI, Ulcerative Colitis Disease Activity Index.
NR

for Children (DISC), a structured interview process used to

determine current and lifetime diagnoses of a number of
Disease Type

psychiatric disorders. Participants also had gastrointestinal symp-

tom interviews.
All participants were then treated with paroxetine and
received follow-up interviews at 8 weeks. Disease activity was not
UC

CD

reported in the study. Depression improved signicantly when

comparing participants pre- and post-data, P 0.0001 (pretreat-
mirtazepine and sertraline.
Patient received mirtazapine

ment 29.0 [SD 7.7]; posttreatment 8.1 [6.1]).

Patient received the ADs
Study Details

Case Reports
(15 mg) at night.

Of the 6 included case reports,2126 2 were reported as ab-

stracts only,25,26 1 of the case reports describes an individual with
UC,25 and the rest of the studies refer to patients with CD. The
patient with UC had generalized anxiety disorder and was treated
with mirtazapine (15 mg) at night and after 6 weeks had relief from
bloody diarrhea, rectal pain, and anxiety. The other abstract
received immunomodulators

mirtazapine, and sertraline.

adalimumab, aripiprazole,
26-year-old man. Previously

described a 64 year-old-man with 6 months of 4 to 6 watery bowel

and courses of steroids

movements per day.26 The patient was receiving mirtazapine and

Medications included
Participants

sertraline for severe depression; when the dosage was changed to

64-year-old man.

be taken at night, the patient had relief from IBD symptoms.

without relief.

One case report found phenelzine to reduce bowel movements

from 10 watery movements per day to one soft movement per day
and without any cramping.22 The participant was tapered off any
other medications, and the symptoms only returned when phenelzine
was stopped after 2 years. Another report found no improvements on
Study Design

IBD course after treatment with transdermal amitriptyline; however,

(Abstract)

(Abstract)
Case report

Case report

no adverse effects were observed.24 The 2 remaining studies reported

TABLE 5. (Continued)

CDAI; in both these studies, all the patients (n 6) achieved remis-

sion with antidepressant treatment.21,23
Dixit,25 India
Author (Year),

Kahn,26 USA

DISCUSSION
Joshi and

Most studies (80%) included in this systematic review

Country

reported antidepressants to have a benecial effect on IBD course,

and 60% reported a benecial effect on anxiety and depression

www.ibdjournal.org | 545

Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.
 546

Macer et al
TABLE 6. Summary of the Primary and Secondary Outcomes in the Included Studies
Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.

| www.ibdjournal.org

Author (Year), Secondary OutcomeAnxiety

Country Study Design Primary OutcomeIBD and Depression Conclusions

Daghaghzadeh Placebo-controlled Severity of symptoms signicantly improved Depression signicantly improved compared Duloxetine recommended for disease
et al,13 Iran RCT compared with control (P 0.02). Intervention: with control (P 0.041). Intervention: mean activity, anxiety, and depression.
mean (SE) 6.23 (1.00) to 4.52 (0.54); control: mean (SE) 8.64 (0.89) to 7.47 (0.80); control: mean
(SE) 7.50 (0.80) to 6.83 (0.69). (SE) 9.77 (0.75) to 10.50 (1.18).
Anxiety signicantly improved compared with
control (P 0.049). Intervention: mean (SE)
7.94 (1.03) to 6.11 (0.99); control: mean (SE)
8.38 (1.04) to 8.50 (1.14).
Yanartas et al,20 Prospective cohort AD treatment was found to be associated with Depression (HAD-D) improved. Intervention: ADs recommended for disease
New Zealand an improvement in CDAI in patients with IBD. 10.62 (3.61) to 3.35 (4.01); control: 11.55 activity, anxiety, and depression.
Intervention: 197.41 (130.60) to 101.08 (2.85) to 10.15 (3.51).
(65.88) (P 0.011); control: 58.50 (74.94) to
83.50 (62.68) (P 0.710). No signicant
difference was observed between groups (P
0.570).
MMSIntervention: 2.71 (3.05) to 0.94 (1.91) (P Anxiety (HAD-A) improved. Intervention: 12.38
0.054); control: 2.78 (3.42) to 1.77 (1.98) (P (4.38) to 5.97 (4.45); control: 11.40 (4.60) to
0.464). No signicant difference was observed 11.05 (4.40).
between groups (P 0.926).
CRP decreased insignicantly in both groups.
Intervention: 6.58 6 13.89 to 4.61 6 4.03
(P 0.324); control: 4.30 6 3.79 to 4.35 6 3.47
(P 0.949). No signicant difference was
observed between groups (P 0.656).
Iskandar et al,19 Retrospective Likert baseline severity scores (CD: 2.07 6 0.03, Not measured Low-dose TCAs recommended for
USA cohort UC: 2.03 6 0.04, P 0.67). Patients with UC management of residual symptoms
responded signicantly better to TCA therapy, in IBD patients with minimal
1.86 6 0.13 for UC and 1.26 6 0.11 for CD inammation.

Inamm Bowel Dis
Volume 23, Number 4, April 2017

(P 0.003). 83% of patients with UC had at
least a moderate symptomatic improvement on
TCA, compared with 50% of patients with CD
(P 0.01).
No signicant difference at the second follow-up
visit. Mean response score of 1.31 6 0.16 for CD
and 1.47 6 0.17 for UC, P 0.76. At the second
visit, 56% of CD group and 40% of UC group had
at least a further moderate symptom response,
P 0.16.
 Inamm Bowel Dis
Volume 23, Number 4, April 2017
TABLE 6. (Continued)
Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.

Author (Year), Secondary OutcomeAnxiety

Country Study Design Primary OutcomeIBD and Depression Conclusions

Goodhand et al,14 Retrospective Fewer relapses and courses of steroids in the year Not measured ADs recommended for disease
USA casecontrol after starting an AD than in the year before (1 [04] activity.
(median [range]) versus 0 [04], P 0.002; 1
[03] versus 0 [04], P ,0.001, respectively); the
controls showed no changes between years 1 and 2
in relapses (1 [04] versus 1 [03], respectively) or
courses of steroids (1 [02] versus 0 [03]).
Mikocka-Walus Cross-sectional Respondents reported taking an AD for an average of Psychological well-being had improved in 87% ADs recommended for anxiety and
(2014), Australia survey 4 (SD 3.9) years ranging from 4 wk to 15 yrs. (n 55) of participants. depression.
79% reported perceived improvements despite 67%
observing no change in disease activity. Disease
activity improved in 25% of participants.
Mikocka-Walus Qualitative ADs improved QoLprimarily psychological, as Three (20%) patients noted how they believed the ADs recommended for anxiety and
et al,16 Australia interview well as social and biological. reduction in feelings of stress mediated the depression.
positive inuence of the AD on IBD course.
5 (33%)helped disease course; 3 (20%)reduction
in pain and frequency of bowel movements; 10
(66%)did not inuence disease course, but
difcult to distinguish between treatments; 3 (20%)
reduction in frequency of symptoms or are ups
Mikocka-Walus Report on clinical 51 currently taking ADs. 71 received ADs in the past. Of the 51 patients currents taking ADs, 45% were ADs recommended for disease
et al,2 Australia case-note audit taking them for depression and 23% for activity.
a combined anxiety and depression disorder.
Disease activity on ADs (n 51): 15 (29%)
inactive disease but presented with symptoms
such as pain or diarrhea, consistent with functional
bowel disorders; 11 (22%)full remission with no
disease activity; 2 (0.04%)active disease; 23
(45%)no data were recorded

Antidepressants in Inammatory Bowel Disease

Ramos Rivers Audit (Abstract) There was a difference in proportion of poorer SIBDQ Not measured ADs are not recommended for disease
et al,17 USA (OR 22.88, 95% CI 8.8958.89, P , 0.0001) activity.
and higher IBD activity (OR 6.34, 95% CI
2.9113.80, P , 0.0001) in those taking SSRIs
versus those who did not but not in proportion with
CRP in those taking SSRIs (OR 1. 78, 95% CI
0.923.42, P 0.09). Mean IBD activity decreases
www.ibdjournal.org |

over time, independent of SSRI use.

Walker et al,18 USA Case series Not measured Mean (SD) HAM-D improvement (pretreatment ADs recommended for anxiety and
29.0 6 7.7; posttreatment 8.1 6 6.1, P depression.
0.0001).
547
 548

Macer et al
TABLE 6. (Continued)
Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.

| www.ibdjournal.org

Author (Year), Secondary OutcomeAnxiety

Country Study Design Primary OutcomeIBD and Depression Conclusions

Kane et al,21 USA Case report Decrease in CDAI to ,150 within 6 weeks (without Not measured Bupropion recommended for disease
other changes to IBD medication). activity.
Kast,22 USA Case report First 7-days bowel movements described as soft, 34 Depression responded well. Phenelzine recommended for disease
per day with cramping. After increase to 30 mg, 1 activity and depression.
bowel movement per day with no cramping. Other
medication tapered off. After 2 years phenelzine
stopped, 6 weeks later admitted to hospital with
CD relapse.
Kast and Case report Female: 19-month remission, any attempts to stop Femalemajor depression remitted. The Bupropion recommended for disease
Altschuler,23 bupropion were associated with relapse. CDAI 0. baseline dysthymia remained. activity and depression.
USA Mesalamine was tapered off.
Male: CDAI 45. 34 episodes of diarrhea daily due
to ilealcecal value.
Scott et al,24 USA Case report Patients abdominal pain remained unchanged, Psychiatrist determined patients depression had Amitriptyline no effect on IBD.
assessed by visual analog scale, but no adverse not responded adequately. Although man
events were associated with transdermal stated his mood had improved at the end of 6-
amitriptyline. week therapy.
Joshi and Dixit,25 Case study After 2 weeks, decreased urgency of defecation and Improvement in anxiety features in 2 weeks. Mirtazapine recommended for disease
India (Abstract) reduced tenesmus were reported. After 6 weeks, After 6 weeks, patient had relief from anxiety activity and anxiety.
there was complete resolution of bloody diarrhea features.
and rectal pain.
Kahn,26 USA Case study Treatment ineffective. Became effective when Not measured Night dosing of mirtazapine and
(Abstract) psychiatrist changed sertraline to bedtime dosing. sertraline recommended for disease
activity.

ADs, antidepressants; CDAI, Crohns Disease Activity Index; CI, condence interval; CRP, C-reactive protein; HAD-A, Hospital Anxiety and Depression Scale-A; HAD-B, Hospital Anxiety and Depression Scale-B; MMS,
Modied Mayo Score; OR, odds ratio; QoL, quality of life; SSRI, selective serotonin reuptake inhibitor; TCAs, tricyclic antidepressants.

Inamm Bowel Dis
Volume 23, Number 4, April 2017

Inamm Bowel Dis
Volume 23, Number 4, April 2017
levels. Despite this encouraging nding, due to the limitations of
the observational study designs included, no rm conclusions can
be drawn about the efcacy or effectiveness of antidepressants in
IBD. Nevertheless, judging by the success of antidepressant
treatment in functional gut disorders and particularly the improve-
ments in bowel functions and abdominal pain,3,27 but also by
a signicant proportion of patients with IBD actively using anti-
depressants (between 10% and 30%)2,28,29 antidepressants have
a role to play in IBD management. Whether this is because they
inuence the inammatory processes or simply because they
improve mood is hard to decipher at present, and their role in
IBD should be further investigated.
To the authors knowledge, only one other similar system-
atic review has been conducted.5 The systematic review included
12 relevant articles; however, the authors found a paucity of high-
quality data. None of the included articles were RCTs; 5 of them
were not primary research and the same group conducted 7 of the
studies. The previous review, while acknowledging the poor
methodological quality of the included studies, concluded that
the results suggest that antidepressants have the potential to be
used to help certain individuals cope with the emotional comor-
bidities of IBD; such as anxiety and depression, improve quality
of life, and possibly have a benecial effect on the IBD course. In
the 10 years since this review was conducted, the evidence seems
to have improved slightly. One RCT was included in this review
but it had some limitations which may have biased the results. It
should be noted that another small trial has been published in
recent weeks,30 reporting no impact of uoxetine on disease activ-
ity over 12 months in CD but observing some potentially positive
impact of this antidepressant on the cytokine proles.
There is much speculation around the potential mechanism
of action of antidepressants in altering the course of IBD. Three of
the included studies13,14,20 hypothesized that the improvements
seen in patients could be because of the anti-inammatory prop-
erties observed in antidepressants.31 There is evidence that anti-
depressants can lower circulating levels of tumor necrosis
factoralpha and this could potentially explain the positive ef-
fects of antidepressants on IBD course.32,33 Alternatively, and
most probably, improvements seen can be a direct result of the
reduction in the symptoms of anxiety and depression as a result of
antidepressants. The braingutmicrobiome research reviewed
elsewhere points toward this explanation.3436 However, further
research is required to conclusively determine the exact mecha-
nism or mechanisms of action.
Current Guidelines
A recent review of the international evidence-based guide-
lines on managing IBD and its comorbid psychosocial issues37
concluded that psychological distress should be screened for and
treated appropriately, with psychotherapy/psychopharmacother-
apy offered if required. The dominance of observational studies
in this review precludes a judgment on the efcacy of antidepres-
sants on IBD course, but results indicate the possibility of an
effect which needs experimental verication.
Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.
Antidepressants in Inammatory Bowel Disease
Limitations of Included Studies
The majority of the included studies were observational,
uncontrolled, and nonrandomized. Only 3 studies had follow-up
periods at 2 years or more, 5 studies had follow-up periods of 12
weeks or less. IBD often takes longer than 12 months to go through
cycles of relapse and remission. Population-based studies have
shown that after 5 years of being diagnosed as in remission, nearly
100% of patients have relapsed;38 therefore, follow-up periods that
are shorter than this are not likely to capture long-term effectiveness.
Many of the studies had small sample sizes and only
sampled participants from a single source; therefore, participants
are unlikely to be representative of the IBD population as a whole.
Furthermore, all studies did not account for differences by sex in
their analyses, which is important because women may be at
greater risk of anxiety and depression than men.39
The nal limitation of the included studies is the study
design. Only 1 RCT was included and 6 (40%) were case; 2
studies were incompletely reported conference abstracts.
Strengths and Limitations of This Review
There were a number of strengths to this review, the rst
being its comprehensive literature search which included an
extensive search string and a large number of databases, including
gray literature. The review was also adapted to account for the
differing study designs using a range of quality assessment tools.
Despite these strengths, there were a number of limitations
to the review. Because of the limited resources, it was not possible
to have a second reviewer at either the screening or data extraction
stages of the review. The review was also limited by only
including articles published in English. However, only 30 non-
English publications were excluded and based on the percentage
of relevant English articles once titles and abstracts were screened
(2.9%), it would be unlikely that the non-English language
publications would have yielded further studies.
Future Research
Further RCTs are required to improve understanding of the
impact of antidepressants on IBD course. Trials should aim to recruit
larger numbers of participants, and analyses should take account of
potential sex differences. Future trials should also prioritize objective
measures of disease activity (i.e., calprotectin, colonoscopy) over
subjective (i.e., disease activity indices) when assessing IBD activity.
The previous systematic review5 recommended that future
research should differentiate between CD and UC; this recommen-
dation has not changed in light of this review ndings. Finally, longer
follow-up periods (at least 5 years) are required to more accurately
determine the efcacy of antidepressants therapy on disease course.
CONCLUSIONS
Antidepressants are commonly used by patients with IBD;
however, based on the ndings from this systematic review, it is
not possible to determine for certain whether antidepressants have
a benecial effect on the course of IBD. The state of research has
www.ibdjournal.org | 549
Macer et al
improved over the last 10 years; however, nearly all the evidence
comes from observational studies where cause and effect are
difcult to attribute. Further properly conducted RCTs with
validated measures, larger samples, and adequate follow-up
periods are required to accurately determine the efcacy of
antidepressants on improving disease course.
ACKNOWLEDGMENTS
We are very grateful to David Brown, Academic Liaison
Librarian at the University of York for his help in compiling the
search strategy.
REFERENCES
1. Mikocka-Walus A, Knowles SR, Keefer L, et al. Controversies revisited:
a systematic review of the comorbidity of depression and anxiety with
inammatory bowel diseases. Inamm Bowel Dis. 2016;22:752762.
2. Mikocka-Walus AA, Gordon AL, Stewart BJ, et al. The role of antide-
pressants in the management of inammatory bowel disease (IBD): a short
report on a clinical case-note audit. J Psychosom Res. 2012;72:165167.
3. Ford AC, Quigley EM, Lacy BE, et al. Effect of antidepressants and
psychological therapies, including hypnotherapy, in irritable bowel syn-
drome: systematic review and meta-analysis. Am J Gastroenterol. 2014;
109:13501365; quiz 66.
4. Mikocka-Walus A, Clarke D, Gibson PR. Can antidepressants inuence
the course of inammatory bowel disease? The current state of research.
Eur Gastroeneterol and Hepatol Rev. 2009;5:16.
5. Mikocka-Walus AA, Turnbull DA, Moulding NT, et al. Antidepressants
and inammatory bowel disease: a systematic review. Clin Pract Epidemiol
Ment Health. 2006;2:24.
6. OBrien SM, Scott LV, Dinan TG. Antidepressant therapy and C-reactive
protein levels. Br J Psychiatry. 2006;188:449452.
7. Sluzewska A, Rybakowski JK, Laciak M, et al. Interleukin-6 serum levels
in depressed patients before and after treatment with uoxetine. Ann N Y
Acad Sci. 1995;762:474476.
8. Higgins JPT, Altman DG, Gtzsche PC, et al. The cochrane collabora-
tions tool for assessing risk of bias in randomised trials. BMJ. 2011;343:
d5928.
9. Wells G, Shea B, OConnell D, et al. The Newcastlescale for
assessing the quality of nonrandomised studies in meta-analyses.pdf.
2000. Available at: http://www.medicine.mcgill.ca/rtamblyn/Readings/
TheNewcastle-Scaleforassessingthequalityofnonrandomisedstudiesinmeta-
analyses.pdf. Accessed February, 2017.
10. NIH. Quality Assessment Tool for Case Series Studies. 2014. Available at:
https://www.nhlbi.nih.gov/health-pro/guidelines/in-develop/cardiovascular-
risk-reduction/tools/case_series. Accessed February, 2017.
11. NIH. Quality Assessment Tool for Observational Cohort and Cross-sectional
StudiesNHLBI, NIH. 2016. Available at: https://www.nhlbi.nih.gov/health-
pro/guidelines/in-develop/cardiovascular-risk-reduction/tools/cohort. Accessed
February, 2017.
12. CASP. Critical Appraisal Skills Programme (CASP). 2016. Available at:
http://www.casp-uk.net/. Accessed February, 2017.
13. Daghaghzadeh H, Naji F, Afshar H, et al. Efcacy of duloxetine add on in
treatment of inammatory bowel disease patients: a double-blind con-
trolled study. J Res Med Sci. 2015;20:595601.
14. Goodhand JR, Greig FIS, Koodun Y, et al. Do antidepressants inuence
the disease course in inammatory bowel disease? A retrospective case-
matched observational study. Inamm Bowel Dis. 2012;18:12321239.
15. Mikocka-Walus A, Andrews JM. Attitudes towards antidepressants
among people living with inammatory bowel disease: an online
Australia-wide survey. J Crohns Colitis. 2014;8:296303.
16. Mikocka-Walus AA, Gordon AL, Stewart BJ, et al. A magic pill? A
qualitative analysis of patients views on the role of antidepressant therapy
in inammatory bowel disease (IBD). BMC Gastroenterol. 2012;12:93.
17. Ramos Rivers CM, Binion DG, Youk A, et al. The longitudinal impact of
serotonin reuptake inhibitors on quality of life and disease activity in
550 | www.ibdjournal.org
Copyright 2017 Crohns & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.
Inamm Bowel Dis
Volume 23, Number 4, April 2017
adults with inammatory bowel disease (IBD). Gastroenterology. 2014;
1:S-456.
18. Walker EA, Gelfand MD, Gelfand AN, et al. The relationship of
current psychiatric disorder to functional disability and distress in
patients with inammatory bowel disease. Gen Hosp Psychiatry.
1996;18:220229.
19. Iskandar HN, Cassell B, Kanuri N, et al. Tricyclic antidepressants for
management of residual symptoms in inammatory bowel disease.
J Clin Gastroenterol. 2014;48:423429.
20. Yanartas O, Kani HT, Bicakci E, et al. The effects of psychiatric treatment
on depression, anxiety, quality of life, and sexual dysfunction in patients
with inammatory bowel disease. Neuropsychiatr Dis Treat. 2016;12:
673683.
21. Kane SV, Altschuler EL, Kast RE. Crohns disease remission on bupro-
pion. Gastroenterology. 2003;125:1290.
22. Kast RE. Crohns disease remission with phenelzine treatment.
Gastroenterology. 1998;115:10341035.
23. Kast RE, Altschuler EL. Remission of Crohns disease on bupropion.
Gastroenterology. 2001;121:12601261.
24. Scott MA, Letrent KJ, Hager KL, et al. Use of transdermal amitriptyline
gel in a patient with chronic pain and depression. Pharmacotherapy.
1999;19:236239.
25. Joshi PA, Dixit K. Case report: use of mirtazapine in treating ulcerative
colitis. Indian J Psychiatry. 2013;55:S118.
26. Kahn T, Laponis R. Down in the dumps: selective serotonin reuptake
inhibitors as a cause of diarrhea. J Gen Intern Med. 2014;29:S348.
27. Ford AC, Talley NJ, Schoenfeld PS, et al. Efcacy of antidepressants and
psychological therapies in irritable bowel syndrome: systematic review
and meta-analysis. Gut. 2009;58:367378.
28. Haapamaki J, Tanskanen A, Roine RP, et al. Medication use among
inammatory bowel disease patients: excessive consumption of antide-
pressants and analgesics. Scand J Gastroenterol. 2013;48:4250.
29. Fuller-Thomson E, Sulman J. Depression and inammatory bowel
disease: ndings from two nationally representative Canadian surveys.
Inamm Bowel Dis. 2006;12:697707.
30. Mikocka-Walus A, Hughes PA, Bampton P, et al. Fluoxetine for mainte-
nance of remission and to improve quality of life in patients with Crohns
disease: A pilot randomized placebo-controlled trial. J Crohns Colitis.
2016. pii: jjw165.
31. Tynan RJ, Weidenhofer J, Hinwood M, et al. A comparative examination
of the anti-inammatory effects of SSRI and SNRI antidepressants on LPS
stimulated microglia. Brain Behav Immun. 2012;26:469479.
32. Kast RE. Anti- and pro-inammatory considerations in antidepressant use
during medical illness: bupropion lowers and mirtazapine increases circu-
lating tumor necrosis factor-alpha levels. Gen Hosp Psychiatry. 2003;25:
495496.
33. Kast RE. Evidence of a mechanism by which etanercept increased TNF-
alpha in multiple myeloma: new insights into the biology of TNF-alpha
giving new treatment opportunities-the role of bupropion. Leuk Res. 2005;
29:14591463.
34. Mayer E, Bradesi S, Gupta A, et al. The brain-gut axis and psychological
processes in IBD. In: Knowles SR, Mikocka-Walus A, eds. Psychological
Aspects of Inammatory Bowel Disease: A Biopsychosocial Approach.
London, United Kingdom: Routledge; 2015:2029.
35. Sexton K, Bernstein C. Stress, distress and IBD. In: Knowles SR,
Mikocka-Walus A, eds. Psychological Aspects of Inammatory Bowel
Disease: A Biopsychosocial Approach. London, United Kingdom:
Routledge; 2015:1019.
36. De Cruz P. Microbiota and psychological processes in IBD. In: Knowles
SR, Mikocka-Walus A, eds. Psychological Aspects of Inammatory
Bowel Disease: A Biopsychosocial Approach. London, United Kingdom:
Routledge; 2015:1019.
37. Hauser W, Moser G, Klose P, et al. Psychosocial issues in evidence-based
guidelines on inammatory bowel diseases: a review. World J Gastroenterol.
2014;20:36633671.
38. Burisch J, Jess T, Martinato M, et al. The burden of inammatory bowel
disease in Europe. J Crohns Colitis. 2013;7:322337.
39. Fuller-Thomson E, Sulman J. Depression and inammatory bowel
disease: ndings from two nationally representative Canadian surveys.
Inamm Bowel Dis. 2006;12:697707.