The effect of TNF-alpha blockers on psychometric measures in ankylosing spondylitis patients a preliminary observation

Page 1

ORIGINAL ARTICLE

The effect of TNF-alpha blockers on psychometric measuresin ankylosing spondylitis patients: a preliminary observation

Ozden Arısoy • Cemal Bes • Cigdem Cifci •

Mustafa Sercan • Mehmet Soy

Received: 14 June 2012 / Accepted: 4 January 2013 / Published online: 20 January 2013

� Springer-Verlag Berlin Heidelberg 2013

Abstract There is a high co-morbidity between chronic

inflammatory disorders and depression. Proinflammatory

cytokines like TNF-a seem to play a central role in the

pathogenesis of these disorders, and its neutralization pro-

vides a potent treatment for inflammatory disorders. Few

studies showed that TNF-a blockers also caused an

improvement in depressive symptoms associated with these

chronic inflammatory disorders. To evaluate the effective-

ness of TNF-a blockers on symptoms of ankylosing spon-

dylitis (AS), depression, anxiety and quality of life, 9 AS

patients resistant to classical therapy were enrolled and

followed-up at 2nd and 6th weeks after a TNF-a blocker

was started. Hamilton Depression and Anxiety Scales

(HAM-D, HAM-A), Hospital Depression and Anxiety

Questionnaire (HAD), Quality of Life Scale (SF36) and AS

severity index (BASDAI) were applied to the patients at

weeks 0, 2 and 6. ESR and CRP were evaluated to monitor

biological disease activity. There was a significant reduc-

tion in HAM-D (p = 0.00), HAM-A (p = 0.00), HAD

anxiety scores (p = 0.02) and a significant improvement in

SF36 physical function (p = 0.00), physical role limitations

(p = 0.00), bodily pain (p = 0.05), general health

(p = 0.01), vitality (p = 0.03) and emotional role limita-

tions (p = 0.00) subscales, BASDAI scores (p = 0.00),

ESR (p = 0.00) and CRP (p = 0.00). Change in clinical

disease activity (BASDAI) was not correlated with change

in depression–anxiety scores, while change in biological

disease activity (CRP) was correlated with change in

depression–anxiety scores. TNFa blockers may have a

potential antidepressant effect besides its anti-inflammatory

effect that seems to be independent of its clinical effect.

Keywords Ankylosing spondylitis � TNF-alpha blockers �Depression � Anxiety � Quality of life � Cytokines

Introduction

Ankylosing spondylitis (AS) is a systemic progressive,

chronic inflammatory disease primarily affecting the axial

skeleton causing pain, physical disability, functional

impairment and reduced quality of life (QOL) [1]. It may

also cause psychiatric symptoms since it has a substantial

effect on the mood of patients [2]. Although reduced

physical functioning is well recognized in AS, its impact

on social and mental health has not received that much

attention [1].

Among a few studies performed in the literature, Barlow

et al. [3] showed that psychological disorders frequently

coexist with AS. They found that 1/3 of patients with AS

reported symptoms of depression [3]. A significant reci-

procal association between disease status and anxiety or

depression in AS was reported as well [4, 5].

Generally, medically ill patients are at increased risk for

depression [6]. Depression in those patients is often

O. Arısoy (&) � M. Sercan

Department of Psychiatry, Medical Faculty, Abant Izzet Baysal

University, Bolu, Turkey

e-mail: [email protected]

C. Bes

Department of Rheumatology, Bakırkoy State Hospital, Istanbul,

Turkey

C. Cifci

Department of Psychiatry, Beypazarı State Hospital, Beypazarı,Turkey

M. Soy

Department of Rheumatology, Hisar Intercontinental Hospital,

Istanbul, Turkey

123

Rheumatol Int (2013) 33:1855–1864

DOI 10.1007/s00296-013-2671-x

Page 2

attributed to the physical impairment that accompanies the

illness, but recent data suggest that other processes like

immune activation may play a role in the pathogenesis of

depressive symptoms [7, 8]. Especially, the actions of

proinflammatory cytokines (PIC), such as interleukin

(IL)-1, tumor necrosis factor-alpha (TNF-a) and interferon-

gamma (IFN-c), may account for various psychopatho-

logical changes associated with major depression [9]. This

view is supported by various findings. Preclinically,

administration of PIC’s in animals induced sickness

behavior characterized by sleepiness, fatigue, loss of

appetite and decreased libido, which are behavioral alter-

ations very similar to the symptoms of depression in

humans [10]. Clinically, administration of PIC’s to cancer

or hepatitis C patients has been found to induce depressive

symptomatology [11]. Also, several medical illnesses,

characterized by chronic inflammatory responses, like

rheumatoid arthritis (RA) and psoriasis, have been reported

to be accompanied by depression [12]. Finally, activated

inflammatory pathways have been shown in clinically

depressed population. Meta-analysis revealed significantly

higher levels of TNF-a, IL-1 and IL-6 in depressed patients

compared to healthy controls [13]. Interestingly, antide-

pressants have been shown to have an anti-inflammatory

activity by reducing TNF-a levels [14, 15]. All these

findings show that immune stimulation induces depression-

like signs and symptoms through the release of PIC’s, and

reducing the effect of PIC’s may reverse depressive

symptoms.

Proinflammatory cytokines are also thought to play a

part in the pathogenesis of several immune-mediated

chronic inflammatory disorders like RA, AS, Crohn’s dis-

ease (CD) and psoriasis [16, 17]. Studies have shown that

TNF-a and IL1b were increased [18–20] and IL-10 was

decreased in AS [20, 21]. This immune activation leads to

synovial inflammation and eventually to the destruction of

cartilage and bone in AS patients [22]. These changes in

cytokines are correlated with disease activity, suggesting

that PIC’s may play an important role during active

inflammation in AS [20]. Especially, TNF-a is the master

cytokine for other inflammatory cytokines, and its neu-

tralization leads to reduced secretion of other PIC’s [23].

This targeted bench-to-bedside research led to the devel-

opment of TNF-a blockers. These agents have been proven

to be a potent treatment for especially patients with AS

who continue to have symptoms of active inflammation

despite medication with nonsteroidal anti-inflammatory

drugs at the maximum possible dose. Insufficient control of

disease activity as indicated by pain, stiffness and decrease

of function is the most common clinical reason for starting

anti-TNF therapy [24].

It is well known that TNF-a blockers are also very

effective in the treatment of other chronic inflammatory

disorders like RA, CD, ulcerative colitis (UC) and psoriasis

[25]. They decrease disease activity and increase QOL in

these patients [1, 26–28]. But, recent evidence has shown

that TNF-a blockers were also effective to decrease

depressive symptoms associated with these disorders. In

one of these trials, Tyring et al. [29] showed that etanercept

caused a 50 % improvement in depressive scores in pso-

riasis patients and the improvement in depression was less

correlated with skin clearance or joint pain. In CD, inflix-

imab also caused a significant decrease in depression

scores, fatigue and a significant increase in QOL [30–32].

There is only one study about the effect of TNF-a blockers

in AS patients up to now, and in that study, Ertenli et al.

[33] showed that infliximab caused a significant decrease in

depression–anxiety scores and disease activity and a sig-

nificant increase in QOL, and this change in disease

activity was not correlated with changes in depression and

anxiety scores.

These observations together with the theoretical back-

ground reported in this article leads to the hypothesis that

TNF-a blockers may reduce the effect of PIC’s and reverse

depressive symptoms associated with chronic inflammatory

disorders both of which are thought to be associated with

increased levels of TNF-a [34]. So, in this study, we aimed

to evaluate the effectiveness of TNF-a blockers on symp-

toms of AS, depression, anxiety and QOL and to see

whether improvement in depression, anxiety and QOL is

correlated with improvement in disease activity or not.

Methods

Ethical approval

In this longitudinal study, 9 patients with AS were asses-

sed. The trial was conducted in accordance with ethical

principles of the Declaration of Helsinki. The study was

approved by the Local Ethics Committee. All patients gave

written informed consent.

Patient recruitment

Patients with AS who were clinically resistant to conven-

tional therapy with disease-modifying antirheumatic drugs

and who were eligible candidates for TNF-a therapy were

invited to participate in the study, and a TNF-a blocker

treatment was started to the eligible patients between

January and April 2009 in the Rheumatology Clinic of

Abant Izzet Baysal University. Diagnosis of AS was based

on American College of Rheumatology criteria. Patients

older than 18 years of age with an active disease of

BASDAI[4 were included in the study. Exclusion criteria

were pregnancy or breast feeding, previous treatment with

1856 Rheumatol Int (2013) 33:1855–1864

123

Page 3

TNF-a blockers, current treatment with psychotropics and

systemic disorders like chronic obstructive pulmonary

disease, congestive heart failure, stroke, cancer, active

tuberculosis or tuberculosis history in the previous 2 years

or other chronic inflammatory disease like psoriasis and

inflammatory bowel disease. Among 9 patients included in

the study, two of them received adalimumab 40 mg/kg

every other week and seven of them received infliximab

5 mg/kg at 0, 2nd and 6th week.

Rheumatological assessment

Turkish version of BASDAI (Bath Ankylosing Spondylitis

Disease Activity Index) was used to assess clinical disease

activity at 0, 2 and 6 weeks. BASDAI is a visual analog

scale (0–10) for 5 major symptoms of AS over last week

(fatigue, spinal pain, peripheral joint pain, enthesitis,

intensity/duration of morning stiffness). Lower scores

indicate less active disease. Turkish reliability and valida-

tion of BASDAI were made by Akkoc et al. [35]. Biological

disease activity was evaluated with erythrocyte sedimen-

tation rate (ESR) and C-reactive protein (CRP) at 0, 2 and

6 weeks. ESR and CRP were conventionally determined.

Psychiatric assessment

A nonstructured psychiatric clinical interview was per-

formed. According to this clinical interview, seven of the

patients had no history of psychiatric disorder, but one had

an ongoing panic disorder and another had depression.

Depression, anxiety levels and quality of life were assessed

with four questionnaires before treatment and after treat-

ment at 2nd and 6th weeks.

Hospital Anxiety and Depression Scale (HADS)

It is a Likert-type self-evaluation scale used to scan anxiety

and depression in physically ill patients. It consists of two

subscales and 14 items; 7 items investigate depression

symptoms (HAD-D) and 7 items investigate anxiety

symptoms (HAD-A). The items are scored from 0 (no

distress) to 3 (maximum distress). The cut-off point for the

anxiety subscale is 10 and 7 for the depression subscale.

Higher scores reflect higher levels of anxiety and depres-

sion. The validity and reliability of the Turkish version

were performed by Aydemir et al. [36].

Hamilton Depression Rating Scale (HAM-D)

It is a clinician-administered scale widely used for

assessing the severity of depressive symptoms. The present

study used the 17-item version (HAM-D-17). Items are

rated on a Likert scale. The cut-off score is 17 for

depression and higher scores reflect higher depression. Its

reliability and validity for the Turkish population were

assessed by Akdemir et al. [37].

Hamilton Anxiety Rating Scale (HAM-A)

It is a clinician-administered scale designed for assessing

the severity of anxiety. It is composed of 14 items rated on

a Likert scale. Higher scores reflect higher anxiety. Its

validity in the Turkish population was assessed by Yazıcıet al. [38].

Short Form-36 (SF-36)

It is the most preferred instrument in measuring quality of

life (QOL). It is a self-rated instrument consisting of 36

items, which provides assessment in 8 domains in the last

four weeks: physical functioning, social functioning, role

limitations due to emotional problems (role-emotional),

role limitations due to physical problems (role-physical),

bodily pain, vitality, mental health and general health

perception. The score ranges between 0 and 100, and

higher score represents better QOL. The reliability and

validity study of the Turkish version were performed by

Kocyigit et al. [39].

Statistical analysis

SPSS version 13.0 was used for statistical analysis. Data

were expressed as mean (SD) or frequency (%). Normal

distribution was tested by Kolmogorov–Smirnov test.

After Mauchly’s test of sphericity, repeated measures

analysis of variance (ANOVA) was performed on the

normally distributed variables among three consecutive

measurements of depression, anxiety and QOL to explore

the difference between baseline, 2nd week and 6th week.

If sphericity could not be assumed, the Greenhouse–

Geisser adjustment was used for the numerator and

denominator degrees of freedom in the F test. Nonpara-

metric Freidman test was applied for CRP values, since

2nd week CRP value did not distribute normally. When a

significant time effect was demonstrated in the Friedman

test, the Wilcoxon signed-rank test for paired samples was

used as a follow-up procedure to make post hoc pairwise

comparisons. Correlations of baseline scores and corre-

lation of change scores in disease activity, biological

activity and psychopathological status were explored by

using Pearson rank correlation test. Change scores were

determined by taking the difference of means between

week 6 and week 0.

Rheumatol Int (2013) 33:1855–1864 1857

123

Page 4

Results

Demographic data

There were 8 males and 1 female in the study group. Mean

age was 39.4 ± 10.1 (20–57) and mean disease duration

was 10.6 ± 7.6 years (Table 1).

Comparison of baseline, 2nd and 6th week scores

The mean values of psychiatric questionnaires (HADS,

HAM-D, HAM-A, SF-36) and clinical (BASDAI) and

biological disease activity (ESR, CRP) are given in

Table 1, and changes of these scores are shown in Figs. 1,

2 and 3.

For self-rating scales, only 3 patients out of 9 patients

(3/9) had high depression scores on HADS at baseline. On

follow-up, 2 patients were over cut-off levels at 2nd and

6th weeks. There was not a significant change over time in

HAD–D scores (p = 0.39). For anxiety, there were 2

patients with high anxiety levels at baseline, but on follow-

up, there were no patients with high anxiety scores at 2nd

week and there were only 1 patient with high anxiety levels

at 6th week. There was a significant change in anxiety

Table 1 Comparison of baseline, 2nd and 6th week scores

Baseline (1) 2nd week (2) 6th week (3) p Difference between

Age mean (SD) 39.4 (10.1)

Sex (male/female) (8/1)

Disease duration (years) mean (SD) 10.6 (7.6)

HAD-D mean (SD) 5.1 (3.2) 4.2 (3.0) 3.3 (3.2) 0.39 NS

HAD-A mean (SD) 7.4 (2.3) 4.2 (3.3) 4.8 (3.2) 0.02 1–2, 1–3

HAM-D mean (SD) 11.9 (3.5) 3.4 (2.8) 3.7 (2.7) 0.00 1–2, 1–3

HAM-A mean (SD) 10.2 (3.5) 2.8 (2.6) 3.7 (2.7) 0.00 1–2, 1–3

SF-36

Physical functioning mean (SD) 17.5 (4.1) 25.3 (4.1) 25.3 (3.9) 0.00 1–2, 1–3

Role limitations physical mean (SD) 4.8 (1.2) 6.8 (1.5) 6.6 (1.8) 0.00 1–2, 1–3

Bodily pain mean (SD) 6.2 (1.9) 8.1 (1.4) 8.5 (1.8) 0.05 1–2, 1–3

General health mean (SD) 12.4 (2.2) 17.8 (4.9) 18.4 (4.6) 0.01 1–2, 1–3

Vitality (energy) mean (SD) 14.00 (3.9) 18.6 (3.6) 18.6 (3.9) 0.03 1–2, 1–3

Social functioning mean (SD) 6.9 (2.4) 8.2 (1.8) 8.1 (1.7) 0.42 NS

Role limitations emotional mean (SD) 3.9 (1.1) 5.4 (0.9) 5.1 (1.3) 0.00 1–2, 1–3

Mental health mean (SD) 20.5 (4.8) 24.2 (4.4) 23.3 (4.7) 0.25 NS

ESR mean (SD) 28.2 (22.2) 4.9 (2.8) 5.8 (5.5) 0.00 1–2, 1–3

CRP mean (SD) 22.9 (15.0) 3.3 (0.6) 5.2 (4.9) 0.00 1–2, 1–3

BASDAI mean (SD) 7.0 (1.1) 2.2 (1.6) 2.1 (1.6) 0.00 1–2, 1–3

HAD-D Hospital Anxiety and Depression Scale—depression, HAD-A Hospital Anxiety and Depression Scale—anxiety, HAM-D Hamilton

Depression Scale, HAM-A Hamilton Anxiety Scale, SF-36 Short Form-36, ESR erythrocyte sedimentation rate, CRP C-reactive protein, BASDAI

Bath Ankylosing Spondylitis Disease Activity Index, SD standard deviation, NS nonspecific

Bold values indicate statistically significant results with a p value of *p \ 0.05

Fig. 1 Change of depression and anxiety scores by time. HAD-

D Hospital Anxiety and Depression Scale—depression, HAD-A Hos-

pital Anxiety and Depression Scale—anxiety, HAM-D Hamilton

Depression Scale, HAM-A Hamilton Anxiety Scale. *p \ 0.05

1858 Rheumatol Int (2013) 33:1855–1864

123

Page 5

scores by time (p = 0.02). The difference was prominent

between baseline and 2nd week and between baseline and

6th week. There was no significant difference in HAD-A

scores between week 2 and week 6.

For clinician-administered scales, there was only 1

patient who was above cut-off level in HAM-D scale. On

follow-up, there were no patients with high depression

scores. There was a significant change over time in HAM-D

scores (p = 0.00). The difference was significant between

baseline and week 2 and between baseline and week 6

assessments. Similarly, anxiety scores dropped significantly

with time (p = 0.00). The change in HAM-A scores was

significant between baseline and week 2 and between

baseline and week 6. There were no significant differences

in HAM-D and HAM-A scores between 2nd and 6th week.

For SF-36 scores, there was a significant increase in all

subscales except social functioning and mental health

(Table 1; Fig. 3). The mean scores that increased signifi-

cantly were physical functioning (p = 0.00), role limita-

tions due to physical problems (p = 0.00), bodily pain

(p = 0.05), general health (p = 0.01), vitality-energy

(p = 0.03) and role limitations due to emotional problems

(p = 0.00), and the changes in these scores were significant

between baseline and week 2 and between baseline and

week 6, but not between week 2 and week 6. The mean

scores for social functioning and mental health increased

also over time, but the differences were not significant

(p = 0.42 and p = 0.25, respectively).

Clinical and biological disease activity as measured by

BASDAI, ESR and CRP decreased significantly over time

as well (p = 0.00, p = 0.00 and p = 0.00, respectively)

(Table 1; Fig. 2). The change was significant between

baseline and week 2 and between baseline and week 6, but

again there was no significant difference between week 2

and week 6.

Correlation of baseline scores

Baseline HAD-D score was negatively correlated with

vitality (p \ 0.01) and mental health (p \ 0.05) subscales

of SF-36. Baseline HAD-A was positively correlated with

ESR (p \ 0.05). Basal HAM-D score was negatively cor-

related with vitality (p \ 0.05) and role limitations due to

emotional problems (p \ 0.05). Basal CRP was negatively

correlated with physical function subscale of SF-36

(p \ 0.05). Basal BASDAI score was not correlated with

depression, anxiety or any subscale of SF-36. (Table 2).

Correlation of change scores

Change in HAD-D score was positively correlated with the

change in HAD-A score (p \ 0.01) and change in CRP

(p \ 0.05), while negatively correlated with role limita-

tions due to physical problems (p \ 0.05), bodily pain

(p \ 0.05), vitality (p \ 0.01) and mental health

(p \ 0.01). Change in HAD-A score was again positively

correlated with change in CRP (p \ 0.05), while negatively

correlated with change in physical function (p \ 0.05),

bodily pain (p \ 0.05) and vitality (p \ 0.05) subscales of

SF-36 (Table 3).

Change in ESR was positively correlated with change in

BASDAI score (p \ 0.05) but negatively correlated with

changes in physical function (p \ 0.01) and bodily pain

(p \ 0.05). Change in CRP was also negatively correlated

with change in physical function (p \ 0.05), role

Fig. 2 Change of disease activity by time. ESR erythrocyte sedi-

mentation rate, CRP C-reactive protein, BASDAI Bath Ankylosing

Spondylitis Disease Activity Index. *p \ 0.05

Fig. 3 Change of quality of life scores by time. PF physical

functionality, RLP role limitations physical, BP bodily pain, GH

general health, VIT vitality, SF social functioning, RLE role

limitations emotional, MH mental health. *p \ 0.05

Rheumatol Int (2013) 33:1855–1864 1859

123

Page 6

Ta

ble

2C

orr

elat

ion

of

bas

elin

esc

ore

s

HA

D-D

HA

D-A

HA

M-D

HA

M-A

ES

RC

RP

BA

SD

AI

SF

-36

PF

SF

-36

RL

PS

F-3

6B

PS

F-3

6b

GH

SF

-36

VIT

SF

-36

SF

SF

-36

RL

ES

F-3

6M

H

HA

D-D

2.8

74

**

2.7

42*

HA

D-A

.73

6*

HA

M-D

2.7

29

*2

.76

4*

HA

M-A

ES

R.7

11*

CR

P2

.76

3*

BA

SD

AI

SF

-36

PF

SF

-36

RL

P.9

53*

*

SF

-36

BP

SF

-36

GH

SF

-36

VIT

.81

8*

.89

2*

*

SF

-36

SF

.92

6*

*

SF

-36

RL

E

SF

-36

MH

HA

D-D

Ho

spit

alA

nx

iety

and

Dep

ress

ion

Sca

le—

dep

ress

ion

,H

AD

-AH

osp

ital

An

xie

tyan

dD

epre

ssio

nS

cale

—an

xie

ty,

HA

M-D

Ham

ilto

nD

epre

ssio

nS

cale

,H

AM

-AH

amil

ton

An

xie

tyS

cale

,

SF

-36

Sh

ort

Fo

rm-3

6,

ES

Rer

yth

rocy

tese

dim

enta

tio

nra

te,

CR

PC

-rea

ctiv

ep

rote

in,

BA

SD

AI

Bat

hA

nk

ylo

sin

gS

po

nd

yli

tis

Dis

ease

Act

ivit

yIn

dex

,S

F-3

6P

Fp

hy

sica

lfu

nct

ion

alit

y,

SF

-36

RF

P

role

lim

itat

ion

sp

hy

sica

l,S

F-3

6B

Pb

od

ily

pai

nS

F-3

6G

Hg

ener

alh

ealt

h,

SF

-36

VIT

vit

alit

y,

SF

-36

SF

soci

alfu

nct

ion

ing

,S

F-3

6R

LE

role

lim

itat

ion

sem

oti

on

al,

SF

-36

MH

men

tal

hea

lth

Bo

ldv

alu

esin

dic

ate

stat

isti

call

ysi

gn

ifica

nt

resu

lts

wit

ha

pv

alu

eo

f*

p\

0.0

5

*p

\0

.05

;*

*p\

0.0

1

1860 Rheumatol Int (2013) 33:1855–1864

123

Page 7

limitations due to physical problems (p \ 0.01), bodily

pain (p \ 0.01), general health (p \ 0.01) and vitality

(p \ 0.05). Change in BASDAI score was negatively cor-

related with the change in bodily pain (p \ 0.05) and the

change in general health (p \ 0.05).

Change in physical function subscale of SF-36 was

positively correlated with change in bodily pain (p \ 0.01).

Change in role limitations due to physical problems

was also positively correlated with change in bodily

pain (p \ 0.05), vitality (p \ 0.05) and mental health

(p \ 0.05). Change in bodily pain was positively correlated

with the change in general health (p \ 0.01) and change in

vitality (p \ 0.05). Finally, change in vitality was posi-

tively correlated with the change in mental health

(p \ 0.01).

Discussion

This is the second study which examined the effect of

TNF-a blockers on depression, anxiety and QOL, in active

AS pts. Frequency of depressed patients in our study was

11 % according to HAM-D score and 33 % according to

HAD-D score. But official diagnostic criteria for MD were

not applied. Depression frequency in patients with chronic

medical diseases is about 6–14 % [6], while depression

frequency in AS is found to be about 33 % [3] In that

sense, our finding is in line with the literature.

Depression and anxiety scores measured by HAM-D,

HAM-A and HAD-A significantly decreased after first and

second injection of TNF-a blockers as early as 2nd week.

Minderhoud et al. [32] reported a significant effect of inflix-

imab infusion on depression after 4 weeks, but we observed

this response to depression much earlier as by 2nd week.

Quality of life scores also increased significantly in all

subscales except social functioning and mental health after

the first and second injections in our study. Social func-

tioning, mental health and role limitations due to emotional

problems are the three subscales which are most sensitive to

the change in depression and which have the most mental

factor component [39], while physical functioning, role

limitations due to physical problems and bodily pain are the

three subscales which are most sensitive to the physical

factor. Vitality and general health are sensitive to both

physical and mental factors [39]. In that sense, we can say

that especially physical function improved significantly in

our study, but, also factors like vitality, general health and

role limitations due to emotional problems increased as

well, indicating an improvement in mental factor parallel to

the decrease in depression and anxiety scores. Correlation

analysis also revealed that improvement in depression and

anxiety was associated more with improvement in both

physical and mental components of QOL, while improve-

ment in biological and serological disease activity was

associated rather with improvement in physical component

of QOL only. So, anxiety and depression seem to contribute

Table 3 Correlations of change scores

Change

from day

0 day 42

HAD-D HAD-A HAM-D HAM-A ESR CRP BASDI SF36 PF SF36

RLP

SF36 BP SF36

GH

SF36

VIT

SF36

SF

SF36

RLE

SF36

MH

HAD-D .864** .735* 2.764* 2.731* 2.919** 2.849**

HAD-A .695* 2.771* 2.798* 2.770

HAM-D

HAM-A

ESR .749* .724* 2.872** 2.825*

CRP 2.746* 2.909** 2.931** 2.883** 2.780*

BASDAI 2.759* 2.869*

SF36 PF .858**

SF36 RFP .784* .808* .760*

SF36 BP .924** .744*

SF36 GH

SF36 VIT .861**

SF36 SF

SF36 RFE

SF36 MH

HAD-D Hospital Anxiety and Depression Scale—depression, HAD-A Hospital Anxiety and Depression Scale—anxiety, HAM-D Hamilton Depression Scale, HAM-A Hamilton

Anxiety Scale, SF-36 Short Form-36, ESR erythrocyte sedimentation rate, CRP C-reactive protein, BASDAI Bath Ankylosing Spondylitis Disease Activity Index, SF-36 PF

physical functionality, SF-36 RLP role limitations physical, SF-36 BP bodily pain SF-36 GH general health, SF-36 VIT vitality, SF-36 SF social functioning, SF-36 RLE role

limitations emotional, SF-36 MH mental health

Bold values indicate statistically significant results with a p value of *p \ 0.05

* ‘ p \ 0.05; ** p \ 0.01

Rheumatol Int (2013) 33:1855–1864 1861

123

Page 8

to decreased QOL caused by the disease itself [40], and

improvement in psychological status seems to be associated

with a more global improvement in QOL.

Correlation analysis also revealed that baseline clinical

disease activity (BASDAI) was not associated with base-

line depression and anxiety, but baseline serological dis-

ease activity (ESR) was associated with baseline anxiety

scores (HAD-A). Previous studies showed that clinical or

serological disease activity was associated with the severity

of depression and anxiety in AS [4, 5, 41]. But Persoons

et al. [30] found no correlation between clinical disease

activity and depression scores neither at baseline nor after

infliximab infusion in CD similar to our study. Likewise,

Ertenli et al. [33] found no correlation between baseline

clinical disease activity and baseline depression–anxiety

scores as well, but they found a positive correlation of

baseline ESR with basal HAD-D scores. So, it seems that

increase in ESR is probably associated with an increase in

depression and anxiety scores, suggesting a role of

inflammation in depression and anxiety.

As for correlation of change scores, we could not find an

association between change in clinical disease activity

(BASDAI) and change in depression–anxiety scores, which

was similar to Ertenli et al.’s [33] finding. But apart from

their studies, we found that change in serological disease

activity (CRP) was correlated with change in depression

and anxiety scores, again suggesting a role of inflammation

in depression and anxiety.

In view of all these findings, it seems reasonable to

suggest that improvement in depression is not associated

with clinical improvement in our study. Similar to our

findings, it was shown that infliximab in CD caused a rapid

and significant decrease in depression scores, prior to any

improvement in the intestinal pathology [30, 32]. These

data were consistent with our findings in AS patients,

because change in the activity of clinical disease was not

associated with the change in depression scores. If

improvement of depression is not related to improvement

of clinical disease activity, then this may suggest

involvement of other mechanisms in improvement of

depression associated with AS.

One of these mechanisms might be the correction of

inflammation. Because, PIC release in inflammation might

be involved in the pathogenesis of depression by causing

disturbances in neurotransmitter metabolism, by increasing

hypothalomopituitary axis (HPA) activity and decreasing

production of relevant central nervous system growth fac-

tors. PIC’s may cause reduction in tryptophan availability

for serotonin synthesis by activating indoleamine-2,3-

dioxygenase (IDO) enzyme [42]. IDO switches the syn-

thesis of serotonin from tryptophan to kynurenine and

quinolinic acid instead of serotonin and by this way

decrease serotonin synthesis [43]. IL-1 and TNF-a are

potent inducers of IDO (43). IL-1 and TNF-a increase the

activity of the serotonin transporter as well. [44]. Espe-

cially, TNF-a contributes to a decrease in synaptic avail-

ability of serotonin. This decrease in serotonin levels

eventually leads to depressive symptoms. All the above

changes are frequently observed in depressive disorders.

Thus, TNF-a blockers, by causing a decrease in TNF-alevels, might increase serotonin levels and therefore correct

depression via this mechanism. In our study, although we

have not studied PIC levels like TNF-a, a decrease in CRP

levels—which is also considered as a PIC—was associated

with improvement in both depression and anxiety. Tuglu

et al. [15] also showed a decrease in CRP levels in depressed

patients with antidepressant treatment. So, as our results also

suggest, TNF-a blockers might be effective in treatment of

depressive symptoms accompanying AS. This may mean

that TNF-a blockers may be promising targets for the

treatment of MD associated with chronic inflammatory

disorders or in otherwise healthy MD patients who have

increased levels of PIC’s. Accordingly, a randomized, dou-

ble-blind, placebo-controlled phase IV trial of infliximab in

treatment-resistant MD patients has been recently completed

by Raison et al. [45] to test the potential antidepressive effect

of TNF-a blockers in a psychiatric population.

Limitations

The major limitations of our study was its small sample

size, absence of a control group, short follow-up interval,

nonstructured psychiatric interview, scarcity of depressed

patients and absence of biochemical measurements of

cytokines. If the number of depressed or anxious AS

patients would have been higher, the findings might have

been more precious. Also, though we showed a decrease in

depression and anxiety scores irrespective of BASDAI

scores, improvement in mental state can still be due to

improvement in pain and stiffness, so it would be much

better to test the potential antidepressant effects of TNF-

alpha blockers in resistant depressive patients without a

chronic inflammatory disorder like AS.

Conclusion

Despite these limitations, our results show that TNF-ablockers may be effective in the treatment of depressive

symptoms accompanying AS. But to clarify the underlying

mechanism of this potential antidepressive effect of TNF-ablockers, change in cytokine levels should also be analyzed

in these patients. This may help to better understand the

role of cytokines in the pathogenesis of depression in active

AS. Also, future studies should be designed to further

1862 Rheumatol Int (2013) 33:1855–1864

123

Page 9

clarify the role of antiinflammatory drugs as potential new

therapeutic agents for the treatment of MD.

Conflict of interest The authors declare that they have no conflict

of interest.

References

1. Braun J, Sieper J (2007) Ankylosing spondylitis. Lancet 369:

1379–1390

2. Ortancil O, Konuk N, May H, Sanli A, Ozturk D, Ankarali H

(2010) Psychological status and patient-assessed health instru-

ments in ankylosing spondylitis. J Clin Rheumatol 16:313–316

3. Barlow JH, Macey SJ, Struthers GR (1993) Gender, depression

and ankylosing spondylitis. Arthritis Care Res 6:45–51

4. Zink A, Braun J, Listing J, Wollenhaupt J (2000) Disability and

handicap in rheumatoid arthritis and ankylosing spondylitis-

results from the German rheumatological database. J Rheumatol

27:613–622

5. Martindale J, Smith J, Sutton CJ, Grennan D, Goodacre L,

Goodacre JA (2006) Disease and psychological status in anky-

losing spondylitis. Rheumatology 45:1288–1293

6. Evans DL, Charney DS, Lewis L et al (2005) Mood disorders in

the medically ill: scientific review and recommendations. Biol

Psychiatry 58:175–189

7. Raison CL, Capuron L, Miller AH (2006) Cytokines sing the

blues: inflammation and the pathogenesis of depression. Trends

Immunol 27:24–31

8. Leonard BE (2010) The concept of depression as a dysfunction of

the immune system. Curr Immunol Rev 6:205–212

9. Schiepers OJ, Wichers MC, Maes M (2005) Cytokines and major

depression. Prog Neuropsychopharmacol Biol Psychiatry

29:201–217

10. Dantzer R (2001) Cytokine-induced sickness behavior: mecha-

nisms and implications. Ann N Y Acad Sci 933:222–234

11. Patten SB (2006) Psychiatric side effects of interferon treatment.

Curr Drug Saf 1:143–150

12. Dantzer R, Capuron L, Irwin MR et al (2008) Identification and

treatment of symptoms associated with ınflammation in medically

ıll patients. Psychoneuroendocrinology 33:18–29

13. Dowlati Y, Herrmann N, Swardfager W et al (2010) A meta-

analysis of cytokines in major depression. Biol Psychiatry

67:446–457

14. Reynolds JL, Ignatowski TA, Sud R, Spengler RN (2005) An

antidepressant mechanism of desipramine is to decrease tumor

necrosis factor-alpha production culminating in increases in

noradrenergic neurotransmission. Neuroscience 133:519–531

15. Tuglu C, Kara SH, Caliyurt O, Vardar E, Abay E (2003)

Increased serum tumor necrosis factor-alpha levels and treatment

response in major depressive disorder. Psychopharmacology 170:

429–433

16. Bradley JR (2008) TNF-mediated inflammatory disease. J Pathol

214:149–160

17. Braun J, Sieper J (2003) Overview of the use of the anti-TNF

agent infliximab in chronic inflammatory diseases. Expert Opin

Biol Ther 3:141–168

18. Lange U, Teichmann J, Stracke H (2000) Correlation between

plasma TNF-alpha, IGF-1, biochemical markers of bone metab-

olism, markers of inflammation/disease activity, and clinical

manifestations in ankylosing spondylitis. Eur J Med Res

29:507–511

19. Sonel B, Tutkak H, Duzgun N (2002) Serum levels of IL-1 beta,

TNF-alpha, IL-8, and acute phase proteins in seronegative

spondyloarthropathies. Joint Bone Spine 69:463e467

20. Chou CT, Huo AP, Chang HN, Tsai CY, Chen WS, Wang HP

(2007) Cytokine production from peripheral blood mononuclear

cells in patients with ankylosing spondylitis and their first-degree

relatives. Arch Med Res 38:190–195

21. Kozaci LD, Sari I, Alacacioglu A, Akar S, Akkoc N (2010)

Evaluation of inflammation and oxidative stress in ankylosing

spondylitis: a role for macrophage migration inhibitory factor.

Mod Rheumatol 20:34–39

22. Frech T (2007) Treatment of ankylosing spondylitis: focus on

etanercept. Biologics 1:45–51

23. Lin J, Ziring D, Desai S et al (2008) TNFalpha blockade in

human diseases: an overview of efficacy and safety. Clin

Immunol 126:13–30

24. Braun J, Pham T, Sieper J, Davis J, van der Linden S, Dougados

M, van der Heijde D; ASAS Working Group (2003) International

ASAS consensus statement for the use of anti-tumour necrosis

factor agents in patients with ankylosing spondylitis. Ann Rheum

Dis 62:817–824

25. Silva LC, Ortigosa LC, Benard G (2010) Anti-TNF-a agents in

the treatment of immune-mediated inflammatory diseases:

mechanisms of action and pitfalls. Immunotherapy 2:817–833

26. Loftus EV, Feagan BG, Colombel JF et al (2008) Effects of

adalimumab maintenance therapy on health-related quality of life

of patients with Crohn’s disease: patient-reported outcomes of the

CHARM trial. Am J Gastroenterol 103:3132–3141

27. Fernandez Lison LC, Vazquez Domınguez B, Luis Fernandez J,

Moreno Alvarez P, Fruns Gimenez I, Liso Rubio J (2008) Quality

of life of patients with rheumatoid arthritis undergoing out-

patient treatment with TNF inhibitors. Farm Hosp 32:178–181

28. Abalos-Medina GM, Ruiz-Villaverde G, Sanchez-Cano D, Ruiz-

Villaverde R, Ocana-Peinado F, Villaverde-Gutierrez C (2010)

Functional level and quality of life in ankylosing spondylitis,

pilot study after 16 weeks TNF blocker treatment. Rev Esp Ge-

riatr Gerontol 45:331–334

29. Tyring S, Gottlieb A, Papp K et al (2006) Etanercept and clinical

outcomes, fatigue, and depression in psoriasis: double-blind

placebo-controlled randomized phase III trial. Lancet 367:29–35

30. Persoons P, Vermeire S, Demyttenaere K et al (2005) The impactof major depressive disorder on the short- and long-term outcome

of Crohn’s disease treatment with infliximab. Aliment Pharmacol

Ther 22:101–110

31. Lichtenstein GR, Bala M, Han C, DeWoody K, Schaible T (2002)

Infliximab improves quality of life in patients with Crohn’s dis-

ease. Inflamm Bowel Dis 8:237–243

32. Minderhoud IM, Samsom M, Oldenburg B (2007) Crohn’s dis-

ease, fatigue, and infliximab: is there a role for cytokines in the

pathogenesis of fatigue? World J Gastroenterol 13:2089–2093

33. Ertenli I, Ozer S, Kiraz S et al (2010) Infliximab, a TNF-alpha

antagonist treatment in patients with ankylosing spondylitis: the

impact on depression, anxiety and quality of life level. Rheumatol

Int. doi:10.1007/s00296-010-1616-x

34. Berthold-Losleben M, Heitmann S, Himmerich H (2009) Anti-

inflammatory drugs in psychiatry. Inflamm Allergy Drug Targets

8:266–276

35. Akkoc Y, Karatepe AG, Akar S, Kirazli Y, Akkoc N (2005) A

Turkish version of the Bath Ankylosing Spondylitis Disease

Activity Index: reliability and validity. Rheumatol Int 25:280–284

36. Aydemir O, Guvenir T, Kuey L et al (1997). Reliability and

validity of Turkish version of Hospital Depression and Anxiety

Questionnaire. Turk Psik Derg 8:280–287 (Turkish)

37. Akdemir A, Orsel S, Dag I et al (1996) Reliability and validity of

Turkish version of Hamilton Depression Scale. Psikiyatri Psiko-

loji Psikofarmakoloji Dergisi 4:251–259 [Turkish]

38. Yazıcı MK, Demir B, Tanrıverdi N et al (1998) Interrater reli-

ability and validity of Turkish version of Hamilton Anxiety Scale.

Turk Psikiyatri Dergisi 9:114–117 [Turkish]

Rheumatol Int (2013) 33:1855–1864 1863

123

Page 10

39. Kocyigit H, Aydemir O, Olmez N et al (1992) Reliability and

validity of Turkish version of Short Form 36 (SF-36). Ilac ve

Tedavi Dergisi 12:102–106 [Turkish]

40. Baysal O, Durmus B, Ersoy Y et al (2011) Relationship between

psychological status and disease activity and quality of life in

ankylosing spondylitis. Rheumatol Int 31:795–800

41. Eren I, Sahin M, Cure E et al (2007) Interactions between psy-

chiatric symptoms and disability and quality of life in ankylosing

spondylitis patients. Arch Neuropsychiatry 44:1–9

42. Wichers M, Maes M (2002) The psychoneuroimmuno-patho-

physiology of cytokine-induced depression in humans. Int J

Neuropsychopharmacol 5:375–388

43. Myint AM, Kim YK (2003) Cytokine-serotonin interaction

through IDO: a neurodegeneration hypothesis of depression. Med

Hypothesis 61:519–525

44. Zhu CB, Blakely RD, Hewlett WA (2006) The proinflammatory

cytokines interleukin-1beta and tumor necrosis factor-alpha acti-

vate serotonin transporters. Neuropsychopharmacol 31:2121–2131

45. Raison CL, Rutherford RE, Woolwine BJ, Shuo C, Schettler P,

Drake DF, Haroon E, Miller AH (2012) A randomized controlled

trial of the tumor necrosis factor antagonist infliximab for treat-

ment-resistant depression: the role of baseline inflammatory

biomarkers. Arch Gen Psychiatry 3:1–11. doi:10.1001/2013

1864 Rheumatol Int (2013) 33:1855–1864

123

Page 11

Copyright of Rheumatology International is the property of Springer Science & BusinessMedia B.V. and its content may not be copied or emailed to multiple sites or posted to alistserv without the copyright holder's express written permission. However, users may print,download, or email articles for individual use.