Psychiatry Research | Depression
Added on 2022-08-25
7 Pages8301 Words22 Views
|
|
|
Contents lists available at ScienceDirect
Psychiatry Research
journal homepage: www.elsevier.com/locate/psychres
Interleukin-6 and depressive symptom severity in response to physical
exercise
Catharina Lavebratta,b,1
, Matthew P. Herringc,d,⁎,1
, Jia Jia Liu a,b
, Ya Bin Weia,b , Davide Bossolie ,
Mats Hallgrenf , Yvonne Forsellf
a Department of Molecular Medicine and Surgery, Neurogenetics Unit, Karolinska Institutet, Stockholm, Sweden
b Center for Molecular Medicine, Karolinska Institutet, Stockholm, Sweden
c Department of Physical Education and Sport Sciences, University of Limerick, Ireland
d Health Research Institute, University of Limerick, Ireland
e Department of Statistics, University of Padua, Padua, Italy
f Department of Public Health Sciences, Karolinska Institutet, Stockholm, Sweden
A R T I C L E I N F O
Keywords:
Regassa
MADRS
Cytokine
Inflammation
Exercise intervention
Depression
A B S T R A C T
Elevated IL-6 has been implicated in depression. The anti-inflammatory effects of exercise may be associated
with its clinical efficacy for depression. We determined if serum IL-6 levels were altered by 12 weeks of physical
exercise, and if IL-6 levels were associated with baseline depression severity and change in depression severity
in response to exercise. Data from 116 adults (42.7 ± 11.5y) with mild-to-moderate depression (Patient Health
Questionnaire > 9) who participated in the physical exercise arm of the Regassa RCT (www.regassa.se) were
analyzed. Participants were requested to complete three 60-min exercise sessions weekly for 12 weeks. Blood
samples were provided at baseline and post-intervention following an overnight fast and were analyzed for
serum levels of IL-6 using ELISA. IL-6 values were logarithm-transformed. Higher baseline serum IL-6 levels
were significantly associated with reduced depression severity after exercise. Reduced IL-6 levels following
exercise were significantly associated with parallel reductions in depression severity. These findings are
consistent with a previously reported association between reduced serum IL-1β levels and reduced depression
severity following 12 weeks of physical exercise in 105 depressed adults. Findings support associations between
IL-6, depressive symptoms, and exercise response, and provide support for the plausible involvement of IL-6 in
the antidepressive effect of exercise.
1. Introduction
Depression is a major public health burden for which successful
treatment remains limited (Chisholm et al., 2016; Vigo et al., 2016).
Thus, there is continued interest in alternative and complementary
treatment strategies for depression. The antidepressive effects of
exercise, an inexpensive and accessible potential adjuvant therapy,
are well-established and comparable to other empirically-supported
treatments in mild-to-moderate depression (Cooney et al., 2013).
Physical exercise training significantly improves the severity of depres-
sive symptoms among healthy adults (Conn, 2010), patients with a
chronic illness (Herring et al., 2012), and clinically depressed patients
(Schuch et al., 2016b), and reduces other associated signs and
symptoms of depressive disorders (Herring et al., 2011). Exercise has
fewer negative side effects than antidepressant medication, and may
also reduce metabolic abnormalities overrepresented among persons
with depression (Vancampfort et al., 2015).
Though exercise has shown promise as an adjuvant therapy against
depression, the mechanisms underlying the positive effects of exercise
remain understudied. Alterations in levels of several neurotransmitters
and neurotrophic peptides have been implicated as plausible mechan-
isms of the antidepressive effects of exercise, including serotonin (5-
HT), noradrenaline, galanin, and brain-derived neurotrophic factor
(Erickson et al., 2011; Greenwood and Fleshner, 2011; Sciolino et al.,
2012; Strohle et al., 2010). Other related mechanisms include changes
in brain anatomy (e.g., alterations in hippocampal volume) and serum
oxidative stress markers (e.g., GPX) (Schuch et al., 2016a).
Inflammation is another factor that putatively links exercise to
improvement in depression. Increased inflammation resulting from
peripheral immune activation, partly due to increased total body and
http://dx.doi.org/10.1016/j.psychres.2017.03.012
Received 7 November 2016; Received in revised form 1 March 2017; Accepted 6 March 2017
⁎ Corresponding author at: Department of Physical Education and Sport Sciences, University of Limerick, PESS 1045, Limerick, Ireland.
1 Equal contributions.
E-mail address: matthew.herring@ul.ie (M.P. Herring).
Psychiatry Research 252 (2017) 270–276
Available online 07 March 2017
0165-1781/ © 2017 Elsevier B.V. All rights reserved.
MARK
Psychiatry Research
journal homepage: www.elsevier.com/locate/psychres
Interleukin-6 and depressive symptom severity in response to physical
exercise
Catharina Lavebratta,b,1
, Matthew P. Herringc,d,⁎,1
, Jia Jia Liu a,b
, Ya Bin Weia,b , Davide Bossolie ,
Mats Hallgrenf , Yvonne Forsellf
a Department of Molecular Medicine and Surgery, Neurogenetics Unit, Karolinska Institutet, Stockholm, Sweden
b Center for Molecular Medicine, Karolinska Institutet, Stockholm, Sweden
c Department of Physical Education and Sport Sciences, University of Limerick, Ireland
d Health Research Institute, University of Limerick, Ireland
e Department of Statistics, University of Padua, Padua, Italy
f Department of Public Health Sciences, Karolinska Institutet, Stockholm, Sweden
A R T I C L E I N F O
Keywords:
Regassa
MADRS
Cytokine
Inflammation
Exercise intervention
Depression
A B S T R A C T
Elevated IL-6 has been implicated in depression. The anti-inflammatory effects of exercise may be associated
with its clinical efficacy for depression. We determined if serum IL-6 levels were altered by 12 weeks of physical
exercise, and if IL-6 levels were associated with baseline depression severity and change in depression severity
in response to exercise. Data from 116 adults (42.7 ± 11.5y) with mild-to-moderate depression (Patient Health
Questionnaire > 9) who participated in the physical exercise arm of the Regassa RCT (www.regassa.se) were
analyzed. Participants were requested to complete three 60-min exercise sessions weekly for 12 weeks. Blood
samples were provided at baseline and post-intervention following an overnight fast and were analyzed for
serum levels of IL-6 using ELISA. IL-6 values were logarithm-transformed. Higher baseline serum IL-6 levels
were significantly associated with reduced depression severity after exercise. Reduced IL-6 levels following
exercise were significantly associated with parallel reductions in depression severity. These findings are
consistent with a previously reported association between reduced serum IL-1β levels and reduced depression
severity following 12 weeks of physical exercise in 105 depressed adults. Findings support associations between
IL-6, depressive symptoms, and exercise response, and provide support for the plausible involvement of IL-6 in
the antidepressive effect of exercise.
1. Introduction
Depression is a major public health burden for which successful
treatment remains limited (Chisholm et al., 2016; Vigo et al., 2016).
Thus, there is continued interest in alternative and complementary
treatment strategies for depression. The antidepressive effects of
exercise, an inexpensive and accessible potential adjuvant therapy,
are well-established and comparable to other empirically-supported
treatments in mild-to-moderate depression (Cooney et al., 2013).
Physical exercise training significantly improves the severity of depres-
sive symptoms among healthy adults (Conn, 2010), patients with a
chronic illness (Herring et al., 2012), and clinically depressed patients
(Schuch et al., 2016b), and reduces other associated signs and
symptoms of depressive disorders (Herring et al., 2011). Exercise has
fewer negative side effects than antidepressant medication, and may
also reduce metabolic abnormalities overrepresented among persons
with depression (Vancampfort et al., 2015).
Though exercise has shown promise as an adjuvant therapy against
depression, the mechanisms underlying the positive effects of exercise
remain understudied. Alterations in levels of several neurotransmitters
and neurotrophic peptides have been implicated as plausible mechan-
isms of the antidepressive effects of exercise, including serotonin (5-
HT), noradrenaline, galanin, and brain-derived neurotrophic factor
(Erickson et al., 2011; Greenwood and Fleshner, 2011; Sciolino et al.,
2012; Strohle et al., 2010). Other related mechanisms include changes
in brain anatomy (e.g., alterations in hippocampal volume) and serum
oxidative stress markers (e.g., GPX) (Schuch et al., 2016a).
Inflammation is another factor that putatively links exercise to
improvement in depression. Increased inflammation resulting from
peripheral immune activation, partly due to increased total body and
http://dx.doi.org/10.1016/j.psychres.2017.03.012
Received 7 November 2016; Received in revised form 1 March 2017; Accepted 6 March 2017
⁎ Corresponding author at: Department of Physical Education and Sport Sciences, University of Limerick, PESS 1045, Limerick, Ireland.
1 Equal contributions.
E-mail address: matthew.herring@ul.ie (M.P. Herring).
Psychiatry Research 252 (2017) 270–276
Available online 07 March 2017
0165-1781/ © 2017 Elsevier B.V. All rights reserved.
MARK
visceral adiposity, can produce sickness behavior with symptoms
consistent with depression (Dantzer and Kelley, 2007). Elevated serum
or plasma levels of interleukin-6 (IL-6), IL1β, tumor necrosis factor α
(TNFα) are the most consistently replicated inflammation-related
findings in depression (Dowlati et al., 2010; Miller et al., 2009).
Indeed, Shelton and colleagues recently reported significant differences
in IL-6 between patients with Major Depressive Disorder (MDD) and
controls after adjusting for BMI (Shelton et al., 2015). Elevated IL-6
levels were also found in cerebrospinal fluid of patients with depres-
sion, and in the frontal cortex of a rat model with depression-like
behavior (Lindqvist et al., 2009; Wei et al., 2016). More recently, a
meta-analysis of 82 studies showed elevated peripheral levels of
cytokines, including IL-6, among patients with Major Depressive
Disorder compared to healthy controls (Kohler et al., 2017).
Inflammatory processes may also indirectly affect the pathophysiology
of depression by influencing neurotransmitter systems (Schwarcz et al.,
2012). For example, pro-inflammatory cytokines upregulate trypto-
phan metabolism in the kynurenine pathway generating functional
ligands (e.g, kynurenic acid and quinolinic acid) which affect NMDA
and α7 nicotinic receptors to also influence dopamine signaling
(Schwarcz et al., 2012). Activation of the kynurenine pathway has
been implicated in depressive symptoms (Schwarcz et al., 2012).
Exercise influences pro-inflammatory factors (Eyre and Baune,
2012; Eyre et al., 2013; Petersen and Pedersen, 2005), having the
potential to alter the number and function of immune cells (Walsh
et al., 2011a). Several inflammatory markers are reduced following
increased physical activity and reduced energy intake (Petersen and
Pedersen, 2005). Moreover, the protective effect of exercise on chronic
inflammation-induced sickness behavior may be partly attributed to
the anti-inflammatory effect of regular exercise, mediated by reduced
visceral adiposity and/or via increases in anti-inflammatory cytokines
including IL-1 receptor antagonist and IL-10 (Walsh et al., 2011a).
Recent animal research supported that the kynurenine pathway can be
impacted by exercise-induced skeletal muscle kynurenine aminotrans-
ferases which may protect against stress-induced depressive behavior
(Agudelo et al., 2014).
Importantly, recent evidence has suggested that the anti-inflam-
matory effects of exercise may be associated with clinical efficacy for
depression (Eyre et al., 2014). For example, change in IL-1β and higher
baseline levels of TNFα were associated with larger improvements in
depressive symptoms following 12 weeks of aerobic exercise among
105 depressed patients (Rethorst et al., 2013). However, exercise dose
was not significantly associated with change in cytokine levels, likely
due to the low exercise dose used in the study. Similarly, a recent
review reported no significant acute or chronic changes in any
inflammatory marker following exercise training in people with
MDD, based on in total 5 studies (Schuch et al., 2016a). Nonetheless,
among the limited available studies, significant associations have been
reported between inflammatory markers, depressive symptoms, and
exercise response (Schuch et al., 2016a).
Thus, the aims of the present study were to examine: 1) baseline
associations between IL-6 levels and depressive symptom severity; 2)
the change in IL-6 levels in response to a 12-week exercise interven-
tion; 3) if baseline IL-6 levels influence the change in depressive
symptoms by exercise; 4) if baseline depressive symptom levels
influence the change in IL-6 levels by exercise; and, 5) the association
between changes in depressive symptoms and IL-6 levels. This was
examined in 116 adults with a baseline diagnosis of mild-to-moderate
depression derived from a Swedish randomized controlled trial (RCT)
of interventions against depression, the Regassa project (Hallgren
et al., 2015).
2. Methods
2.1. Participants
Study participants were recruited from the RCT Regassa (www.
regassa.se), which evaluated the relative effectiveness of three
interventions for depression (Hallgren et al., 2015), including
physical exercise, internet-based cognitive behavioral therapy, and
treatment as usual (TAU). The study was performed at primary care
facilities located in six county councils in Sweden between the years
2011 and 2013. The study protocol has been previously described in
detail by Hallgren et al. (2015). In brief, each intervention lasted 12
weeks with assessments of depression severity at baseline (pre-
treatment) and post-treatment (3 months after baseline). An ethical
review board at the Karolinska Institutet approved the study (Dnr
2010/1779-31/4), and all participants provided informed consent
prior to participation.
Patients, 18–64 years old, who scored > 9 on the Patient Health
Questionnaire (PHQ-9) were invited to participate in the trial. The
PHQ-9 is a valid instrument for diagnosing depressive disorders and
scores above nine have a reported sensitivity and specificity of 88% for
major depression (Kroenke et al., 2001), though recent meta-analytic
evidence has somewhat questioned its specificity (Mitchell et al., 2016).
The complete MINI (Sheehan et al., 1998) was performed to confirm
diagnoses and identify psychiatric exclusion characteristics. The exclu-
sion criteria were: severe somatic illness, a primary alcohol or drug use
disorder, or a psychiatric diagnosis that required specialist treatment
(such as psychosis).
Patients in the physical exercise group were randomized to one of
three conditions: ‘light exercise’, which consisted of yoga classes (or
similar) with a focus on gentle stretching and controlled breathing;
‘moderate exercise’, an intermediate-level aerobics class; and ‘vigorous
exercise’, a higher intensity aerobics/strength-training and balance
class. Patients were requested to participate in three 60-min exercise
classes per week for 12 weeks. As previously reported (Helgadóttir
et al., 2016), heart rate (HR) within each session was monitored using
pulse watches, and average session HR significantly differed between
conditions (light: ~54% of estimated maximum heart HR (MHR,
calculated as 220 minus age); moderate: ~70% MHR; vigorous:
~76% MHR). The exercise sessions were undertaken at ‘Friskis och
Svettis’ a modern fitness center with multiple locations throughout
Sweden. Adherence was monitored through weekly face-to-face meet-
ings with a qualified personal trainer and patients who failed to attend
this meeting were contacted by the trainer with a telephone call and
encouraged to continue the exercise intervention.
Physical activity level was scored using a self-reported question-
naire (GIHAv score) (range 0–152) (Ekblom-Bak et al., 2011), where a
score of ≥21 is equivalent to the minimum physical activity level
recommended for general health by the American College of Sports
Medicine (ACSM) and the World Health Organization, while scores
≥42 are equivalent to the activity level recommended by the ACSM and
WHO to achieve additional health benefits in adults. Though some
evidence suggests limitations of self-report measures of physical
activity among people with mental illness (Soundy et al., 2014), the
GIHAv has been validated in a separate (unpublished) study in which it
corroborated objectively-measured moderate-to-vigorous and vigorous
physical activity (Hallgren et al., 2016b).
Severity of depression was assessed at baseline and post-treatment
using the Montgomery–Åsberg Depression Rating Scale (MADRS)
(Montgomery and Asberg, 1979), a 10-item scale designed to be
sensitive to changes in depressive symptoms following treatment.
Higher scores indicate more severe depression; each item yields a
score ranging from 0 to 6. Ten symptoms are rated: apparent sadness,
reported sadness, inner tension, reduced sleep, reduced appetite,
concentration difficulties, lassitude, inability to feel, pessimistic and
suicidal thoughts. Total scores range between 0 and 60.
C. Lavebratt et al. Psychiatry Research 252 (2017) 270–276
271
consistent with depression (Dantzer and Kelley, 2007). Elevated serum
or plasma levels of interleukin-6 (IL-6), IL1β, tumor necrosis factor α
(TNFα) are the most consistently replicated inflammation-related
findings in depression (Dowlati et al., 2010; Miller et al., 2009).
Indeed, Shelton and colleagues recently reported significant differences
in IL-6 between patients with Major Depressive Disorder (MDD) and
controls after adjusting for BMI (Shelton et al., 2015). Elevated IL-6
levels were also found in cerebrospinal fluid of patients with depres-
sion, and in the frontal cortex of a rat model with depression-like
behavior (Lindqvist et al., 2009; Wei et al., 2016). More recently, a
meta-analysis of 82 studies showed elevated peripheral levels of
cytokines, including IL-6, among patients with Major Depressive
Disorder compared to healthy controls (Kohler et al., 2017).
Inflammatory processes may also indirectly affect the pathophysiology
of depression by influencing neurotransmitter systems (Schwarcz et al.,
2012). For example, pro-inflammatory cytokines upregulate trypto-
phan metabolism in the kynurenine pathway generating functional
ligands (e.g, kynurenic acid and quinolinic acid) which affect NMDA
and α7 nicotinic receptors to also influence dopamine signaling
(Schwarcz et al., 2012). Activation of the kynurenine pathway has
been implicated in depressive symptoms (Schwarcz et al., 2012).
Exercise influences pro-inflammatory factors (Eyre and Baune,
2012; Eyre et al., 2013; Petersen and Pedersen, 2005), having the
potential to alter the number and function of immune cells (Walsh
et al., 2011a). Several inflammatory markers are reduced following
increased physical activity and reduced energy intake (Petersen and
Pedersen, 2005). Moreover, the protective effect of exercise on chronic
inflammation-induced sickness behavior may be partly attributed to
the anti-inflammatory effect of regular exercise, mediated by reduced
visceral adiposity and/or via increases in anti-inflammatory cytokines
including IL-1 receptor antagonist and IL-10 (Walsh et al., 2011a).
Recent animal research supported that the kynurenine pathway can be
impacted by exercise-induced skeletal muscle kynurenine aminotrans-
ferases which may protect against stress-induced depressive behavior
(Agudelo et al., 2014).
Importantly, recent evidence has suggested that the anti-inflam-
matory effects of exercise may be associated with clinical efficacy for
depression (Eyre et al., 2014). For example, change in IL-1β and higher
baseline levels of TNFα were associated with larger improvements in
depressive symptoms following 12 weeks of aerobic exercise among
105 depressed patients (Rethorst et al., 2013). However, exercise dose
was not significantly associated with change in cytokine levels, likely
due to the low exercise dose used in the study. Similarly, a recent
review reported no significant acute or chronic changes in any
inflammatory marker following exercise training in people with
MDD, based on in total 5 studies (Schuch et al., 2016a). Nonetheless,
among the limited available studies, significant associations have been
reported between inflammatory markers, depressive symptoms, and
exercise response (Schuch et al., 2016a).
Thus, the aims of the present study were to examine: 1) baseline
associations between IL-6 levels and depressive symptom severity; 2)
the change in IL-6 levels in response to a 12-week exercise interven-
tion; 3) if baseline IL-6 levels influence the change in depressive
symptoms by exercise; 4) if baseline depressive symptom levels
influence the change in IL-6 levels by exercise; and, 5) the association
between changes in depressive symptoms and IL-6 levels. This was
examined in 116 adults with a baseline diagnosis of mild-to-moderate
depression derived from a Swedish randomized controlled trial (RCT)
of interventions against depression, the Regassa project (Hallgren
et al., 2015).
2. Methods
2.1. Participants
Study participants were recruited from the RCT Regassa (www.
regassa.se), which evaluated the relative effectiveness of three
interventions for depression (Hallgren et al., 2015), including
physical exercise, internet-based cognitive behavioral therapy, and
treatment as usual (TAU). The study was performed at primary care
facilities located in six county councils in Sweden between the years
2011 and 2013. The study protocol has been previously described in
detail by Hallgren et al. (2015). In brief, each intervention lasted 12
weeks with assessments of depression severity at baseline (pre-
treatment) and post-treatment (3 months after baseline). An ethical
review board at the Karolinska Institutet approved the study (Dnr
2010/1779-31/4), and all participants provided informed consent
prior to participation.
Patients, 18–64 years old, who scored > 9 on the Patient Health
Questionnaire (PHQ-9) were invited to participate in the trial. The
PHQ-9 is a valid instrument for diagnosing depressive disorders and
scores above nine have a reported sensitivity and specificity of 88% for
major depression (Kroenke et al., 2001), though recent meta-analytic
evidence has somewhat questioned its specificity (Mitchell et al., 2016).
The complete MINI (Sheehan et al., 1998) was performed to confirm
diagnoses and identify psychiatric exclusion characteristics. The exclu-
sion criteria were: severe somatic illness, a primary alcohol or drug use
disorder, or a psychiatric diagnosis that required specialist treatment
(such as psychosis).
Patients in the physical exercise group were randomized to one of
three conditions: ‘light exercise’, which consisted of yoga classes (or
similar) with a focus on gentle stretching and controlled breathing;
‘moderate exercise’, an intermediate-level aerobics class; and ‘vigorous
exercise’, a higher intensity aerobics/strength-training and balance
class. Patients were requested to participate in three 60-min exercise
classes per week for 12 weeks. As previously reported (Helgadóttir
et al., 2016), heart rate (HR) within each session was monitored using
pulse watches, and average session HR significantly differed between
conditions (light: ~54% of estimated maximum heart HR (MHR,
calculated as 220 minus age); moderate: ~70% MHR; vigorous:
~76% MHR). The exercise sessions were undertaken at ‘Friskis och
Svettis’ a modern fitness center with multiple locations throughout
Sweden. Adherence was monitored through weekly face-to-face meet-
ings with a qualified personal trainer and patients who failed to attend
this meeting were contacted by the trainer with a telephone call and
encouraged to continue the exercise intervention.
Physical activity level was scored using a self-reported question-
naire (GIHAv score) (range 0–152) (Ekblom-Bak et al., 2011), where a
score of ≥21 is equivalent to the minimum physical activity level
recommended for general health by the American College of Sports
Medicine (ACSM) and the World Health Organization, while scores
≥42 are equivalent to the activity level recommended by the ACSM and
WHO to achieve additional health benefits in adults. Though some
evidence suggests limitations of self-report measures of physical
activity among people with mental illness (Soundy et al., 2014), the
GIHAv has been validated in a separate (unpublished) study in which it
corroborated objectively-measured moderate-to-vigorous and vigorous
physical activity (Hallgren et al., 2016b).
Severity of depression was assessed at baseline and post-treatment
using the Montgomery–Åsberg Depression Rating Scale (MADRS)
(Montgomery and Asberg, 1979), a 10-item scale designed to be
sensitive to changes in depressive symptoms following treatment.
Higher scores indicate more severe depression; each item yields a
score ranging from 0 to 6. Ten symptoms are rated: apparent sadness,
reported sadness, inner tension, reduced sleep, reduced appetite,
concentration difficulties, lassitude, inability to feel, pessimistic and
suicidal thoughts. Total scores range between 0 and 60.
C. Lavebratt et al. Psychiatry Research 252 (2017) 270–276
271
Of those 317 randomized to physical exercise treatment, 39.4%
gave blood through venipuncture at baseline, and 39.4% of those 249
who participated also at the post-treatment assessment gave blood
both at baseline and post-treatment. The blood samples were collected
after an overnight fast on a day without an exercise session within a
week before the first exercise session and within a week after the final
exercise session. Serum was immediately separated and stored at
−20C. Serum was not collected from patients who had a symptom of
infection.
2.2. IL-6 measurements
The patients who had given blood at baseline were analyzed for
serum levels of IL-6 (n=125) using ELISA according to the manufac-
turer's protocol (ELISA Kit Human IL-6 #KHC0061, Invitrogen
Corporation, Camarillo, CA). The IL-6 lower detection limit was
0.2 pg/ml. No sample had been freeze-thawed prior to analysis. The
ELISA intra-assay mean coefficient of variation (CV) from duplicate
samples was 4%, and the inter-assay CV from 5 plates was 9%. Of the
125 patients, 3 were excluded because of undetectable IL-6 levels at
baseline, and 6 patients were excluded due to being outliers ( > 3 SD
above the mean, i.e., ≥17 pg/ml) (n=3 at baseline, n=3 post-treat-
ment). Of the remaining 116 patients, post-treatment IL-6 data were
available for 89.
2.3. Statistical analyses
2.3.1. Baseline analyses (aim 1)
IL-6 values were logarithm-transformed to normalize the distribu-
tions. Pearson correlation coefficients were calculated to examine
associations between IL-6 and BMI, age, depressive symptom severity
(MADRS score), and physical activity level at baseline. How relation-
ships varied based on sex and categorized depressive symptom severity
(mild or moderate; correlational analyses for severe depression were
precluded by too few observations, n=4) was examined using stratified
models. Subsample correlational analyses were tested at α=0.01 to
account for multiple testing; p-values greater than 0.01 but less than
0.05 were considered nominally significant. All other analyses were
tested at α=0.05.
2.3.2. Changes in response to exercise (aim 2)
Last observation carried forward imputation was performed for
post-treatment IL-6 levels and depressive symptom severity for 27 and
11 patients, respectively. A 3 condition (exercise intensities) by 2 time
(baseline and post-intervention) mixed model repeated measures
ANCOVA, adjusted for baseline depressive symptom severity
(MADRS score), baseline physical activity level, number of exercise
sessions completed, and percent change in BMI, examined effects of
exercise and exercise intensities on IL-6 levels. The magnitude of
exercise effects are expressed as eta-squared; effects are classified as
small (η2 =0.04), moderate (η2
=0.25), and large (η2 =0.64) based on
established guidelines (Ferguson, 2009).
2.3.3. Prediction of change (aims 3 and 4) and associated change
(aim 5)
To examine predictive associations of change in MADRS score by
baseline IL-6 (Aim 3), multiple linear regression adjusted for age, BMI,
baseline physical activity, and number of exercise sessions completed
was used. A similar test model, adjusted for age, baseline physical
activity, number of exercise sessions, and percent change in BMI, was
applied to examine the association between baseline MADRS score and
change in IL-6 pre-post exercise (Aim 4). Likewise, the associations
between changes in IL-6 and changes in depression severity across 12
weeks of exercise (Aim 5) was assessed using multiple regression
adjusted for age, percent change in BMI, baseline physical activity
level, and number of exercise sessions completed.
3. Results
3.1. Participant Characteristics
Baseline characteristics for the 116 included participants are
presented in Table 1. At baseline, all participants were diagnosed with
Depressive Disorder. Based on suggested cut-points for the MADRS
(Snaith et al., 1986), depressive symptom severity ranged from mild
(MADRS score=8) to severe (MADRS score=38); 32.8% (n=38) re-
ported mild depression, 63.8% (n=74) reported moderate depression,
and 3.4% (n=4) reported severe depression.
3.2. Associations between baseline IL-6 and baseline characteristics
(aim 1)
In the overall sample, a positive correlation was found between BMI
and IL-6 (r(114) =0.21, p=0.022). Accordingly, IL-6 was significantly
Table 1
Baseline participant characteristics by exercise group.
Light exercise (n=48) Moderate exercise (n=36) Vigorous exercise (n=32) Total (n=116)
Age (years) 40.6 ± 11.6 45.9 ± 11.1 42.4 ± 11.4 42.7 ± 11.5
Female (%) 33 (68.8%) 22 (61.1%) 21 (65.6%) 76 (65.5%)
Number of Sessions Completed 15 ± 10 14 ± 11 15 ± 11 15 ± 11
Weight (kg) 75.6 ± 16.6 77.5 ± 15.5 76.2 ± 15.8 76.4 ± 15.9
BMI (kg/m2) 25.4 ± 4.1 26.0 ± 3.9 25.5 ± 4.0 25.6 ± 3.9
Overweight/Obese (BMI≥25; %) 25 (52.1%) 20 (55.6%) 17 (53.1%) 62 (53.5%)
Obese (BMI≥30; %) 6 (12.5%) 6 (16.7%) 5 (15.6%) 17 (14.7%)
Physical Activity Level (GIHAv)a 24.7 ± 19.0 26.8 ± 28.5 24.2 ± 21.2 25.2 ± 22.7
Physically Inactive (%) 25 (52.1%) 20 (57.1%) 17 (54.8%) 62 (54.4%)
Moderately Active (%) 13 (27.1%) 7 (20.0%) 8 (25.8%) 28 (24.6%)
Highly Active (%) 10 (20.8%) 8 (22.9%) 6 (19.4%) 24 (21.1%)
Depression (MADRS) 21.9 ± 6.3 24.4 ± 5.3 21.3 ± 6.0 22.5 ± 6.0
Mild Depression (MADRS: 7–19; %) 17 (35.4%) 7 (19.4%) 14 (43.8%) 38 (32.8%)
Moderate (MADRS: 20–34; %) 29 (60.4%) 28 (77.8%) 17 (53.1%) 74 (63.8%)
Severe (MADRS: 35+; %) 2 (4.2%) 1 (2.8%) 1 (3.1%) 4 (3.4%)
Antidepressant Use (%) 4 (8.3%) 2 (5.6%) 3 (9.4%) 9 (7.8%)
IL-6 concentration (pg/dL) 1.8 ± 1.0 1.6 ± 1.3 1.9 ± 1.3 1.8 ± 1.2
a A score of ≥21 is equivalent to the minimum physical activity level recommended for general health by the ACSM and the World Health Organization. Participants who scored < 21
were considered to be ‘physically inactive’. Scores ≥42 are equivalent to the activity level recommended by the ACSM and WHO to achieve additional health benefits in adults, and were
categorized as ‘highly active.’ Score ranges for inactivity, moderate activity and high physical activity were 0–20, 21–41, and 42–152, respectively.
C. Lavebratt et al. Psychiatry Research 252 (2017) 270–276
272
gave blood through venipuncture at baseline, and 39.4% of those 249
who participated also at the post-treatment assessment gave blood
both at baseline and post-treatment. The blood samples were collected
after an overnight fast on a day without an exercise session within a
week before the first exercise session and within a week after the final
exercise session. Serum was immediately separated and stored at
−20C. Serum was not collected from patients who had a symptom of
infection.
2.2. IL-6 measurements
The patients who had given blood at baseline were analyzed for
serum levels of IL-6 (n=125) using ELISA according to the manufac-
turer's protocol (ELISA Kit Human IL-6 #KHC0061, Invitrogen
Corporation, Camarillo, CA). The IL-6 lower detection limit was
0.2 pg/ml. No sample had been freeze-thawed prior to analysis. The
ELISA intra-assay mean coefficient of variation (CV) from duplicate
samples was 4%, and the inter-assay CV from 5 plates was 9%. Of the
125 patients, 3 were excluded because of undetectable IL-6 levels at
baseline, and 6 patients were excluded due to being outliers ( > 3 SD
above the mean, i.e., ≥17 pg/ml) (n=3 at baseline, n=3 post-treat-
ment). Of the remaining 116 patients, post-treatment IL-6 data were
available for 89.
2.3. Statistical analyses
2.3.1. Baseline analyses (aim 1)
IL-6 values were logarithm-transformed to normalize the distribu-
tions. Pearson correlation coefficients were calculated to examine
associations between IL-6 and BMI, age, depressive symptom severity
(MADRS score), and physical activity level at baseline. How relation-
ships varied based on sex and categorized depressive symptom severity
(mild or moderate; correlational analyses for severe depression were
precluded by too few observations, n=4) was examined using stratified
models. Subsample correlational analyses were tested at α=0.01 to
account for multiple testing; p-values greater than 0.01 but less than
0.05 were considered nominally significant. All other analyses were
tested at α=0.05.
2.3.2. Changes in response to exercise (aim 2)
Last observation carried forward imputation was performed for
post-treatment IL-6 levels and depressive symptom severity for 27 and
11 patients, respectively. A 3 condition (exercise intensities) by 2 time
(baseline and post-intervention) mixed model repeated measures
ANCOVA, adjusted for baseline depressive symptom severity
(MADRS score), baseline physical activity level, number of exercise
sessions completed, and percent change in BMI, examined effects of
exercise and exercise intensities on IL-6 levels. The magnitude of
exercise effects are expressed as eta-squared; effects are classified as
small (η2 =0.04), moderate (η2
=0.25), and large (η2 =0.64) based on
established guidelines (Ferguson, 2009).
2.3.3. Prediction of change (aims 3 and 4) and associated change
(aim 5)
To examine predictive associations of change in MADRS score by
baseline IL-6 (Aim 3), multiple linear regression adjusted for age, BMI,
baseline physical activity, and number of exercise sessions completed
was used. A similar test model, adjusted for age, baseline physical
activity, number of exercise sessions, and percent change in BMI, was
applied to examine the association between baseline MADRS score and
change in IL-6 pre-post exercise (Aim 4). Likewise, the associations
between changes in IL-6 and changes in depression severity across 12
weeks of exercise (Aim 5) was assessed using multiple regression
adjusted for age, percent change in BMI, baseline physical activity
level, and number of exercise sessions completed.
3. Results
3.1. Participant Characteristics
Baseline characteristics for the 116 included participants are
presented in Table 1. At baseline, all participants were diagnosed with
Depressive Disorder. Based on suggested cut-points for the MADRS
(Snaith et al., 1986), depressive symptom severity ranged from mild
(MADRS score=8) to severe (MADRS score=38); 32.8% (n=38) re-
ported mild depression, 63.8% (n=74) reported moderate depression,
and 3.4% (n=4) reported severe depression.
3.2. Associations between baseline IL-6 and baseline characteristics
(aim 1)
In the overall sample, a positive correlation was found between BMI
and IL-6 (r(114) =0.21, p=0.022). Accordingly, IL-6 was significantly
Table 1
Baseline participant characteristics by exercise group.
Light exercise (n=48) Moderate exercise (n=36) Vigorous exercise (n=32) Total (n=116)
Age (years) 40.6 ± 11.6 45.9 ± 11.1 42.4 ± 11.4 42.7 ± 11.5
Female (%) 33 (68.8%) 22 (61.1%) 21 (65.6%) 76 (65.5%)
Number of Sessions Completed 15 ± 10 14 ± 11 15 ± 11 15 ± 11
Weight (kg) 75.6 ± 16.6 77.5 ± 15.5 76.2 ± 15.8 76.4 ± 15.9
BMI (kg/m2) 25.4 ± 4.1 26.0 ± 3.9 25.5 ± 4.0 25.6 ± 3.9
Overweight/Obese (BMI≥25; %) 25 (52.1%) 20 (55.6%) 17 (53.1%) 62 (53.5%)
Obese (BMI≥30; %) 6 (12.5%) 6 (16.7%) 5 (15.6%) 17 (14.7%)
Physical Activity Level (GIHAv)a 24.7 ± 19.0 26.8 ± 28.5 24.2 ± 21.2 25.2 ± 22.7
Physically Inactive (%) 25 (52.1%) 20 (57.1%) 17 (54.8%) 62 (54.4%)
Moderately Active (%) 13 (27.1%) 7 (20.0%) 8 (25.8%) 28 (24.6%)
Highly Active (%) 10 (20.8%) 8 (22.9%) 6 (19.4%) 24 (21.1%)
Depression (MADRS) 21.9 ± 6.3 24.4 ± 5.3 21.3 ± 6.0 22.5 ± 6.0
Mild Depression (MADRS: 7–19; %) 17 (35.4%) 7 (19.4%) 14 (43.8%) 38 (32.8%)
Moderate (MADRS: 20–34; %) 29 (60.4%) 28 (77.8%) 17 (53.1%) 74 (63.8%)
Severe (MADRS: 35+; %) 2 (4.2%) 1 (2.8%) 1 (3.1%) 4 (3.4%)
Antidepressant Use (%) 4 (8.3%) 2 (5.6%) 3 (9.4%) 9 (7.8%)
IL-6 concentration (pg/dL) 1.8 ± 1.0 1.6 ± 1.3 1.9 ± 1.3 1.8 ± 1.2
a A score of ≥21 is equivalent to the minimum physical activity level recommended for general health by the ACSM and the World Health Organization. Participants who scored < 21
were considered to be ‘physically inactive’. Scores ≥42 are equivalent to the activity level recommended by the ACSM and WHO to achieve additional health benefits in adults, and were
categorized as ‘highly active.’ Score ranges for inactivity, moderate activity and high physical activity were 0–20, 21–41, and 42–152, respectively.
C. Lavebratt et al. Psychiatry Research 252 (2017) 270–276
272
End of preview
Want to access all the pages? Upload your documents or become a member.
Related Documents