The Inflammatory and Neuroanatomical Factors Involved in Post-stroke Depression

By

Kira Stefanie Bensimon

A thesis submitted in conformity with the requirements for the degree of Master of Science (MSc.)

Graduate Department of Pharmacology and Toxicology

University of Toronto

© Copyright by Kira Stefanie Bensimon 2013

The Inflammatory and Neuroanatomical Factors Involved in Post-stroke Depression

Kira Stefanie Bensimon, MSc. Candidate, 2013 Graduate Department of Pharmacology and Toxicology, the University of Toronto ABSTRACT

This cross-sectional study examined neurobiologic correlates of depression in ischemic stroke

patients. Depression severity was measured with a standardized scale (Center for Epidemiologic Studies

Depression Scale; CES-D). Eighty-two patients (53.1% male, mean (± SD) age 71.9 ± 14.2 years, mean

(± SD) National Institutes of Health Stroke Scale (NIHSS) score 4.6±4.7, mean (± SD) CES-D score

12.6 ± 10.8) were recruited. A linear regression controlling for age and stroke severity (NIHSS)

determined that the kynurenine to tryptophan ratio (β= -0.105, p=0.369) was not significantly

associated with CES-D (primary hypothesis) (overall model R2=0.069, F3,73=1.805, p=0.154).

Secondary analyses suggested one instance of cytokines favouring inflammatory states in mild

depressive symptomatology; IFN-Ɣ/IL-10 (OR, 2.17; 95% CI, 1.02-4.64, p=0.045). For the most part

however, inclusion of cytokines and neuroimaging correlates such as atrophy, lesion location and white

matter changes were non-significant. Longitudinal studies are necessary to identify the possible

neurobiologic correlates of depressive symptoms post-stroke.

ii

ACKNOWLEDGMENTS

I would first like to thank Drs. Krista Lanctôt and Nathan Herrmann for their unwavering

support throughout my entire two years at Sunnybrook. From the get go, Dr. Lanctôt has motivated me

to be confident and committed in my pursuit of a Master’s degree. At times of frustration, confusion,

anxiety (or even paranoia!) she was always only a short walk, phone call, or email away, and she

always put my worries into perspective. Her devotion to her work and easy-going, good-natured attitude

are characteristics that I truly hope to acquire in my own career, and I know that in this regard, she has

also left a lasting impression on the entire lab. Dr. Herrmann has also been a huge support for me

throughout my journey at Sunnybrook. His straightforward and candid approach to communication is a

trait that I not only admire, but also recognize within myself. His dedication to both the practice of

medicine and his research is something that I aspire to develop as I embark on the next scholastic

journey in my life, that is, medical school. Both Drs. Lanctôt and Herrmann have been extremely

encouraging in this regard as well, from countless reference letters to advice on interviews, I cannot

thank them enough for all of their help throughout the entire application process. I would like to

additionally thank Dr. Fu-Qiang Gao for his commitment to painstakingly teaching me the ins and outs

of CT analysis, certain techniques were very difficult to grasp and your patience was greatly

appreciated! I also especially wish to thank Dr. Walter Swardfager for his continued mentorship

through my Master’s experience. Thanks for taking my anxious phone calls and emails about confusing

statistical tests (at sometimes inconvenient hours) and always setting aside opportunities to give me

genuine advice for my study.

I additionally want to give a special thanks to everyone in the Neuropsychopharmacology

lab. I remember when I first started out at Sunnybrook I could not for the life of me understand why the

name of our research group was so darn long! Abby Li, Nipuni Ranepura, Russanthy Velummailum,

Marly Isen, Hao Yi, Graham Mazereeuw, Sarah Chau, Mahwesh Saleem, Randy Rovinski, and

last but definitely not least, Jordana O’Regan, thank you all so much for your companionship and

guidance throughout my time here; I have developed such warm memories of Sunnybrook because of

all of you. Jordana, our heart-to-hearts in the ‘lair’ of the lab (when we should have been recruiting,

administering MMSEs, or tracing brains, among other things!) will stay with me forever, and I know

that our friendship will endure long after I leave Sunnybrook. These past two years have been

extremely rewarding and I will undoubtedly treasure all the experiences, lessons and memories I have

collected with the Neuropsychopharmacology group along the way.

iii

TABLE OF CONTENTS

ABSTRACT ii

ACKNOWLEDGMENTS iii

1 INTRODUCTION 1

1.1 Statement of the Problem 1

1.2 Purpose of the Study and Objectives 2

1.3 Statement of Research Hypotheses and Rationale for Hypotheses 4

1.4 Review of the Literature 6

1.4.1 Prevalence 6

1.4.2 Clinical Correlates and Phenomenology of PSD 7

1.4.3 Treatments for Post-stroke Depression 10

1.4.4 Ischemic and Inflammatory Responses in Stroke 11

1.4.5 Sickness behaviour 14

1.4.6 Inflammatory Markers and Depression 15

1.4.7 IDO-mediated Mechanisms of Depression 18

1.4.8 Serotonin Hypothesis of Depression 27

1.4.9 Neuroanatomical Correlates of Post-Stroke Depression 30

2 METHODS 37

2.1 Study Design 37

2.2 Inclusion Criteria and Exclusion Criteria 38

2.3 Demographics, Medical History, and Clinical Assessments 39

2.3.1 Demographic and Medical History 39

2.3.2 Clinical Assessments 39

2.4 Blood Sampling 40

2.5 Serum and Plasma Analyses 41

2.5.1 Kynurenine Assay 41

2.5.2 Cytokine Assay 42

2.5.3 Tryptophan and LNAA Assay 43

2.6 Neuroimaging 43

2.6.1 Lesion Volume and Location Measurement 44

2.6.2 White Matter Change Analysis 46

2.6.3 Measurement of Hippocampal Thickness 47

2.6.4 Whole Brain Atrophy Measurement 49

iv

2.7 Planned Analyses 51

2.7.1 Initial Descriptive Statistics 51

2.7.2 Primary Analyses 52

2.7.3 Secondary Analyses 53

2.7.4 Exploratory Analyses 53

2.7.5 Sample Size Calculation 54

2.8 Controlling for Confounding Variables 55

3 RESULTS 56 3.1 Demographics and Clinical Characteristics 56

3.2 Assay Results 59

3.3 Primary Analyses 59

3.4 Secondary Analyses 61

3.5 Exploratory Analyses 65

4 DISCUSSION 70 4.1 The KYN Pathway’s Influence on Post-stroke Depressive Symptoms 71

4.2 Influence of Cytokines on Post-Stroke Depressive Symptoms 75

4.3 Influence of WMC, Atrophy and Lesion Volume on Post-stroke Depressive

Symptoms 76

4.4 Limitations 82

4.5 Recommendations 85

4.6 Conclusion 88

5 REFERENCES 89 LIST OF PUBLICATIONS AND ABSTRACTS 116

APPENDICES 117

Appendix 1: Sunnybrook REB Approval Letter 117

Appendix 2: Post-stroke Depression Study Patient Consent Form 120

Appendix 3: Study Roles 124

Appendix 4: Bivariate correlations between timing of assessment and serum markers 125

Appendix 5: Percent imputation of cytokines 126

Appendix 6: Pearson’s Chi-squared tests for imputed cytokine and CES-D groups 126

LIST OF TABLES

Table 1: Clinical demographic characteristics 56

v

Table 2: Vascular risk factors 57

Table 3: Lesion location and laterality 58

Table 4: Concomitant medication use 59

Table 5: Final linear regression models for cytokine as it predicts CES-D scores 62

Table 6: Final ordinal regression model for cytokines and cytokine ratios as it predicts CES-D

tertiles 64

Table 7: Results of paired samples t-tests for right and left regional ARWMC scores 69

Table 8: Spearman’s correlations between bilateral ARWMC scores and CES-D scores. 70

LIST OF FIGURES Figure 1: TRP and KYN metabolism pathway 19

Figure 2: Axial slice showing a patient with a large right hemispheric infarct (radiological

perspective: flipped) 45

Figure 3: Axial slice showing a patient with grade 2 white matter changes in bilateral frontal

and parieto-occipital regions. 47

Figure 4: Axial slice showing the measurement of left hippocampal width (radiological

perspective) by aligning the hippocampus 30° caudal to the AC-PC plane. 48

Figure 5: Axial slice showing a manual tracing of the left and right lateral ventricle, as well as

the third ventricle. 50

Figure 6: Correlation between serum K/T and CES-D scores 60

Figure 7: Bar graph of mean serum K/T levels across stroke patients in low, middle, and high

CES-D tertiles 61

Figure 8: Correlation between plasma IFN-Ɣ and CES-D scores 63

Figure 9: Correlation between plasma IFN-Ɣ/IL-10 and CES-D scores 63

Figure 10: Bar graph of mean plasma IFN-Ɣ/IL-10 levels across stroke patients in low,

middle, and high CES-D tertiles 65

Figure 11: Correlation between average hippocampal thickness and CES-D scores. 66

Figure 12: Correlation between total ventricular volume / total intracranial volume, as a

measure of whole brain atrophy, and CES-D scores 67

Figure 13: Correlation between global ARWMC scores and CES-D scores. 67

Figure 14: Correlation between lesion volume and CES-D scores 68

vi

Section 1: Introduction

1.1 Statement of the Problem

Stroke is one of the most common causes of death in industrialized countries.1 For those who survive,

it can negatively impact physical and neuropsychiatric domains leading to impairments in these areas.

While apathy, anxiety and cognitive impairment are common post-stroke sequelae,2 the onset of

depression after stroke is one of the most frequent neuropsychiatric outcomes. 3 Depression in particular

can significantly affect the stroke patient, the family and staff involved in their care, and the healthcare

system at large.3 Regarding negative consequences for the patient, post-stroke depression (PSD) has been

linked to increased mortality4, reduced functional5 and rehabilitative6 success, and poorer quality of life.3,

7 Diminished quality of life and increased depressive symptoms are also commonly experienced by

caregivers of PSD patients8. With respect to the impact of PSD on the health care system, it has been

correlated with an increased duration of inpatient stay and increased number of outpatient clinic visits.3, 9

PSD affects approximately one third of all stroke survivors10 and its prevalence surpasses the 30-

day (4.9%) and lifetime (17.1%) prevalence of depression found by the U.S. National Comorbidity

Survey.3, 11 Studies propose that the prevalence of PSD reaches a maximum at roughly 3-6 months,

diminishes by 50% at 1 year, and then it may remain elevated at approximately 20% for 2 years and

beyond.3, 12, 13 This last finding is troublesome because it demonstrates that PSD can be a chronic, non-

remitting disease in many stroke survivors. Compounding this is the issue of antidepressant treatment

efficacy. Current therapies may reduce depressive symptom severity, however they may not lead to the

remission of symptoms.14, 15 Given that functional disability and cognitive impairment are strongly

associated with PSD,16 these deficits may not improve with the use of current antidepressant therapies. 3, 14

Taking these concerns together, it is evident that more research must focus on the mechanisms involved

in the onset of PSD, in order to develop efficacious treatment strategies for this emotionally debilitating

disease.

1

2

1.2 Purpose of the Study and Objectives

It is evident that PSD is a serious health concern to the patient, their caregivers, and the entire

health system at large. Because of its high prevalence among stroke survivors and the negative

consequences it can incur, researchers have devoted much energy into developing and testing hypotheses

on the etiology of this disease. Over the years, the biological and psychosocial mechanisms of PSD have

been examined, forming multiple lines of reasoning and debate over which domain contributes more to

the development of PSD.12 However, the general consensus among experts in the field today is that these

two viewpoints are not contradictory, rather, they may interact to instigate the development of depressive

symptoms.12

The overarching objective of this study is to investigate the neurobiological correlates of PSD

through the assessment of current contemporary hypotheses, namely the inflammatory and

neuroanatomical factors which may have a role in its development. Principally, this study aims to uncover

the relationship between acute inflammatory serum biomarkers and post-stroke depressive symptom

severity in a cohort of ischemic stroke patients. In doing so, the kynurenine (KYN) to tryptophan (TRP)

ratio (K/T) will be assessed for its association with depressive symptoms. The K/T ratio is considered to

be an indicator of indolamine 2,3-dioxygenase (IDO) activity,17, 18 which is an enzyme that catalyzes the

conversion of TRP into KYN. Tryptophan is a precursor for the neurotransmitter serotonin,19, 20 and so

acute depletion of this molecule by way of increased KYN production may lead to the development of

depressive symptoms post-stroke. Furthermore, the metabolism of KYN increases the amount of

neurotoxic21 KYN metabolites, which may also have a detrimental effect on mood regulating centers of

the brain. Furthermore, IDO activation can be influenced by both elevations in pro- inflammatory

cytokines22-28 and reductions in anti-inflammatory cytokines.29, 30 Pro- versus anti-inflammatory cytokine

imbalance has been studied with respect to major depression 31-35 and post-stroke outcomes 36, 37 in the

past with positive results. Therefore as a secondary aim I will examine the relationship between pro-

3

inflammatory and anti-inflammatory cytokines and post-stroke depressive symptoms, as a possible

intermediary link between an elevated K/T ratio and depressive symptoms.

Lesion volume,38-43 white matter change (WMC), 38-48 and atrophy 39, 40, 45, 49 are other biological

correlates that have been studied in the context of PSD with mixed findings, however MDD has been

consistently linked with neurodegeneration and neuroprogression. 50-60 A comprehensive approach to

investigating atrophy, lesion volume and WMC may help to clarify their role in the context of PSD, as

these measurements are not mutually exclusive. Therefore, I additionally sought to explore the

relationship between neuroimaging findings and post-stroke depressive symptom severity, as these might

represent chronic factors involved in the development of PSD. Specifically, hippocampal thickness,

whole brain atrophy, WMC and lesion volume may be important independent correlates of depressive

symptom severity post-ischemic stroke. Finally, the fact that (1) depression has been associated with

neurodegeneration,50-60 (2) KYN metabolites are elevated in many neurodegenerative diseases and

stroke,61-66 and (3) KYN metabolites have also been implicated in the process of neurodegeneration, 21, 50,

67-77 suggest that an elevated K/T ratio may be more relevant in contributing to the development of PSD in

the presence of neurodegeneration, than either variable alone. Thus, I will also assess the role of an

interaction between the K/T ratio and any significant radiological findings from the previous exploratory

analysis, in contributing to depressive symptoms in this population.

In summary, the key objectives of this study are to explore the inflammatory and neuroanatomical

contributions and their potential interactions, to the development of PSD.

4

1.3 Statement of Research Hypotheses and Rationale for Hypotheses

Hypothesis 1: Serum levels of the K/T ratio, as a measure of IDO activity, will positively correlate with

depressive symptom severity post-ischemic stroke.

Rational: Pro-inflammatory cytokines are released during the inflammatory response following

infarction,78-80 and elevations in cytokines stimulate the synthesis of the enzyme IDO. In turn, this

increases the production of KYN from TRP, thereby increasing the K/T ratio.28, 81 Changes in the levels

of KYN and its metabolites have been associated with neuropsychiatric outcomes such as bipolar

mania,82, 83 schizophrenia,84-87 and importantly, major depression.88-91 Furthermore, our group observed a

significant positive correlation between the K/T ratio and depression scores in coronary artery disease

(CAD) patients, a disease which shares common grounds with stroke.92 TRP is the precursor for the

neurotransmitter serotonin,93 and a depletion in serotonin has been associated with major depressive

disorder (MDD) .94, 95 Additionally, acute tryptophan depletion paradigms have caused transient

depressive symptoms in recovered MDD patients.96, 97 Once IDO catalyzes the conversion of TRP to

KYN, KYN can be further degraded into excitotoxic metabolites such as quinolinic acid (QUIN) which

has been demonstrated to play a role in neurodegenerative processes.75, 98 Furthermore, this neurotoxic

metabolite may affect the emotion centres of the brain; a recent study discovered that the brains of

deceased patients with MDD and bipolar disorder had increased QUIN immunoreactivity in the prefrontal

cortex and hippocampus. 89 Thus, taking these findings together, depressive symptoms experienced post-

stroke may be the result of either acute tryptophan depletion or the production of excitotoxic KYN

metabolites that may contribute to the onset and maintenance of depression, respectively. Additionally,

changes in the K/T ratio have been correlated with depression in non-stroke patients,88 therefore it is

possible that elevations in the K/T ratio may be associated with PSD as well.

5

Hypothesis 2: Elevated pro-inflammatory cytokines IFN-Ɣ, TNF-α, IL-6, IL-18, and IL-1β and a

reduction in the anti-inflammatory cytokine IL-10, as evidenced by an elevated immunologic ratio

between these pro-inflammatory cytokines and IL-10, will correlate with depressive symptom severity

post-ischemic stroke.

Rational: IDO activation can be stimulated by both elevations in pro-inflammatory cytokines22-28 and

reductions in anti-inflammatory cytokines.29, 30 Therefore it is plausible that a change in the levels of these

cytokines post-stroke may be associated with increased IDO activation. These particular cytokines were

chosen because they have been shown to be disrupted post-stroke,99-104 or because they have been

implicated in stimulating IDO,22-30 and represent both helper T cell type 1 (IFN-Ɣ, TNF-α, IL-1β, IL-18)

and type 2 (IL-6 and IL-10) cytokines.105 Further, after immune challenge with IFN-α, levels of CSF

KYN and QUIN in hepatitis C virus (HCV) infected patients increased significantly and were

significantly correlated with depression scores.106 Pro- versus anti-inflammatory cytokine imbalance has

been studied with respect to major depression31-35 and post-stroke outcomes36, 37 in the past. With respect

to major depression, it was observed that the IL-6/IL-10 ratio was significantly elevated in major

depressive disorder (MDD) patients compared to controls,31 and multiple studies have observed a

reduction in these immunologic ratios upon antidepressant treatment.32-35 Researchers have also found

that patients with PSD had significantly higher serum ratios of TNF-α/IL-10 and IL-6/IL-10 than

controls,36 and reduced serum IL-6 and elevated serum IL-10 were significantly associated with lower

degree of patient disability.37 Additionally, IL-4 and IL-10 polymorphisms, which related to lower anti-

inflammatory cytokine production were significantly associated with PSD.107 This suggests that immune

imbalance may be more indicative of PSD symptoms than studying pro- and anti-inflammatory markers

in isolation and therefore is the impetus for the secondary hypothesis of this study.

6

1.4 Review of the Literature

1.4.1 Prevalence

One of the primary causes of death in Canada is due to stroke.108 The overall incidence of stroke

in Canada is 14.4 per 10,000, however, for patients 80 years and older this number rises to approximately

132 per 10,000. The length of stay for patients in hospital is approximately 3 weeks per indexed episode,

and 18.2% of patients die in the hospital within 28 days.108

While this mortality rate is high, there has also been much progress in the area of post-stroke

interventional and rehabilitative services that are contributing to an increased number of stroke survivors.

These patients must deal with many post-stroke functional, cognitive and emotional impairments.3 The

onset of depression after stroke is one of the most frequent neuropsychiatric outcomes and has been

linked to increased mortality rate4, reduced functional5 and rehabilitative6 success, and poorer quality of

life in stroke survivors.3, 7 In the acute phase after a stroke (approximately 1 month) the prevalence of

depressive disorders and depressive symptoms ranges from 5% to 63%.109 Furthermore, one study

estimated overall pooled prevalence of depression post-stroke was approximated at 33% for the follow-up

period between 2 weeks and 5 years post-stroke.10 A recent systematic review and meta-analysis by

Ayerbe et al. assessed fifty studies published between 1983 and 2011 and found the pooled prevalence of

depression to be 29% (95% CI 25–32) and can remain at this level for up to 10 years post-stroke.110 In

addition, they found that depression within one month post-stroke had a prevalence of 28% (95% CI 23–

34), depression at one to six months post-stroke had a 31% (95% CI 24–39) prevalence, at six months to

one year the prevalence increased to 33% (95% CI 23–43), and at more than one year the prevalence

remained elevated at 25% (95% CI 19–32).110 This surpasses the 30-day (4.9%) and lifetime (17.1%)

prevalence of depression found by the U.S. National Comorbidity Survey.3, 11 Variations in the methods

of selecting the study population, i.e. inclusion/exclusion criteria, assessment time points, and lack of an

operational definition of PSD in order to determine appropriate diagnoses, are all factors involved in the

lack of consensus on the prevalence of PSD.3, 5, 111, 112 In terms of the study population, there is a large

7

disparity between the degree of disability in inpatient versus outpatient units, which is problematic

because depression has been correlated with functional impairment in the past.5 Thus, studies exclusively

assessing inpatients or outpatients may misjudge the overall prevalence in these populations.3 With

respect to the timing of assessments, studies observing patients in the acute (weeks) phase post-stroke

may calculate higher frequencies PSD than studies with patients in the more chronic (months to years)

phases post-stroke.3 As mentioned earlier, studies suggest that the prevalence of PSD reaches a maximum

at roughly 3-6 months, diminishes by 50% at 1 year, and then it may remain elevated at approximately

20% for 2 years and beyond.3, 12, 13 The lack of consensus on the prevalence of these post-stroke sequelae

is troublesome, despite this, it is evident that depression after a stroke is a frequent and debilitating

disease that warrants further investigation.

1.4.2 Clinical Correlates and Phenomenology of PSD

The reliable diagnosis of PSD is a difficult task for clinicians, since stroke usually leads to the

development of functional disabilities and neurological deficits. In some cases the stroke, as opposed to

the underlying presence of depression, may be credited for these impairments. For instance, physical

disabilities such as paralysis may render the patient unable to participate in certain daily activities.3

However, the patient may also be experiencing anhedonia, which is the inability to enjoy activities that

were once considered pleasurable by the individual. Dysphagia may interfere with the ability to assess

disturbances in appetite and fluctuations in weight which may also manifest as somatic symptoms of

depression.3 Additionally, aphasic patients who are experiencing depressive symptoms may be unable to

express these feeling properly or entirely.3, 113

Compounding the issue of diagnosis is the even larger problem of how to define PSD. An

operational definition of this disorder is difficult due to the presence of other neuropsychiatric disorders

that may accompany it. These conditions may include anxiety, apathy, and pseudobulbar affect

(emotional incontinence), which have the potential to occur within the context of depressive symptoms or

exclusive of them.3 For instance, one study observed that apathy, defined as diminished motivation,114-116

8

occurred independently in 19.8% of post-stroke patients and occurred concurrently with depression in

20.6% of patients.117 Despite the overlap between depression and other post-stroke neuropsychiatric

sequelae, the DSM-IV-TR classifies post-stroke major depression as “mood disorder due to stroke with

major-depressive-like episode” (American Psychiatric Association 2000), while post-stroke minor

depression is associated with “research criteria” in DSM-IV. To receive the diagnosis of major

depression, patients must display at least five out of these nine symptoms: depressed mood, anhedonia,

appetite/weight change, insomnia or hypersomnia, psychomotor agitation or retardation, fatigue or loss of

energy, feelings of worthlessness or guilt, diminished concentration, or recurrent thoughts of death.

Additionally, these symptoms must be present for two weeks at minimum and depressed mood or

anhedonia must constitute one of the five symptoms present.118 To receive the diagnosis of minor

depression, patients must exhibit two to four of these nine symptoms and once again the presence of

depressed mood or anhedonia is required.118 Spalletta and colleagues119 studied 200 first-ever stroke

patients within 3 months post- stroke and observed that 25% of the cohort had major PSD and 31% had

minor PSD. In the past, minor PSD has been linked to increased mortality risk120 and diminished ability to

take part in activities of daily living.13 Therefore, physicians should take care in diagnosing both forms of

this disorder, since the consequences of either could be devastating.3

Researchers have attempted to further characterize PSD in terms of early- and late- onset of

symptoms. A study assessing 142 patients over a period of 2 years post-stroke found that symptoms such

as anxiety, loss of libido, guilt and irritability were only more frequent in the depressed group before 6

months.121 In a 3-year longitudinal study assessing 80 patients, researchers observed a 25% prevalence of

major PSD acutely and a 31% prevalence at 3 months.122 In addition, 60% of the patients diagnosed with

‘early depression’ (zero to three months) experienced remission of depressive symptoms at one-year. It

was also determined that patients who were diagnosed with early depression and still had symptoms at

one-year were more likely to experience chronic depression.122 This finding is in agreement with studies

by Wade et al.123 and Robinson et al.124 who have demonstrated that as many as half of their participants

characterized as having early- or late-onset depression, were still depressed at one-year follow-up.

9

However, Whyte and Mulsant suggest that depression developing acutely post-stroke is more likely to

spontaneously remit.12

In terms of the psychosocial correlates of PSD, functional impairment,5, 13, 122, 125-128 stroke

severity,129 personal history of depression,130, 131female gender,47, 128, 129, 132 social isolation,130, 132

neuroticism133 and cognitive impairment134, 135 have been found to moderate the risk of PSD in the past. A

meta-analysis by Hackett and Anderson, however, determined that only one half of the aforementioned

factors were significantly associated with PSD across various studies.16 These factors include functional

impairment, stroke severity, and cognitive impairment.16 Another recent meta-analysis by Ayerbe et al.

found that disability after stroke, personal history of depression pre-stroke, cognitive impairment, stroke

severity, lack of social support, and anxiety were all important predictors of PSD.110 This study also

observed that poorer quality of life, mortality and disability are significant independent outcomes of

depression.110 Moreover, the presence of cognitive impairment in the context of PSD has been associated

with diminished recovery of depressive symptoms,136 and deficits in the areas of memory, visual

perception and language have also been linked with worse long-term PSD outcomes.41, 137, 3

Numerous studies have also attempted to correlate lesion characteristics with PSD. Lesions in left

anterior and left basal ganglia regions have been associated with the development of PSD and the severity

of depressive symptoms have been correlated with the proximity of the lesion to the frontal pole.125, 138-142

The time course of this relationship may also be important; studies have shown that it is strongest at 6

months post-stroke,139 however other researchers have failed to reproduce these findings.143, 144 Although

there is much debate in this field over the true impact of lesion location of PSD, the general concept

behind this research is that certain brain circuits involved in mood regulation can be altered by a stroke,

leading to the onset of depression.3,45 This concept will be discussed in further detail elsewhere in this

thesis.

10

1.4.3 Treatments for Post-stroke Depression

Many pharmacological treatments for PSD have been tested over the past three decades. These

treatments include antidepressant therapies145-157 and psychostimulants149, 158-160, Regarding antidepressant

medications, many open-label and placebo controlled studies have determined that currently accepted

treatments for depression, such as selective serotonin reuptake inhibitors (SSRIs),147 serotonin-

norepinephrine reuptake inhibitors (SNRIs),161 and tricyclic antidepressants (TCAs)145 are effective at

lessening the symptom severity of depressive symptoms post-stroke.3 However, by DSM-IV standards,

multiple meta-analyses have found that the effects of these treatments are not significant enough to

demonstrate that depressive symptoms remit completely.3, 14, 15 Although, a randomized trial that found

nortriptyline was more efficacious at reducing depressive symptoms than either fluoxetine or placebo,153

also determined that both fluoxetine and nortriptyline taken for 3 months resulted in improvement of

executive functioning at 2 years.3, 162 This is a significant finding since some researchers believed that

smaller reductions in depressive symptoms by antidepressants may leave patients dealing with

impairments in cognition and function.14

Researchers have also assessed prophylactic treatment of depressive symptoms post-stroke. Yet the

results of these placebo-controlled prophylactic studies are not in unanimous agreement with one another.

Some studies’ findings support the use of antidepressants in patients without evidence of depressive

symptoms,163-165 while others show no significant benefit of proactive treatment in the acute post-stroke

phase.166, 167 In addition, multiple meta-analyses have examined the utility of prophylactic treatment for

PSD, however these too have produced mixed findings.168 For instance, one meta-analysis by Chen et al.

observed a significant difference in the incidence of PSD in antidepressant treated patients versus non-

treated patients. That study was then challenged by a Cochrane review that failed to find strong evidence

in support of antidepressant prophylaxis of PSD.169 This review attributes its lack of findings to the

methodological heterogeneity of studies, such as differing periods between stroke and timing of study

enrollment, wide range of medications administered and length of trial periods, and finally, a mix of

11

depression scales or classifications utilized across studies may have prevented this group from accurately

investigating treatment prophylaxis in PSD.169, 170

Concerns regarding the safety of TCAs and SSRIs for use in elderly patients with vascular disease has

prompted much investigation on their negative effects. A systematic review assessing the use of SSRIs

and TCAs in patients with vascular disease, for example, stroke and cardiac disease, observed that TCAs

were not significantly associated with increased frequency of ‘serious’ cardiovascular adverse events,

such as death due to heart failure, stroke or myocardial infarct.171 Furthermore, SSRIs were significantly

less likely to cause ‘non-serious’ cardiovascular AEs compared to TCAs.171 Conversely, in a

retrospective, clustered secondary data analysis from the National Veterans Health Administration long-

term care nursing homes (n=6,577), it was found that antidepressant medications use (OR = 1.39, P <

.0001) was associated with the odds of falling by nursing home residents.172 Furthermore, another study

found that there is an increased risk of stroke in women who use SSRIs (OR=1.45, 95% CI 1.08-1.97).173

With these conflicting findings, it may be important to explore non-pharmaceutical interventions for

treatment of PSD, such as psychoeducation,174 care coordination175 and nursing home visits,176 which have

been demonstrated to reduce depressive symptoms and improve quality of life in stroke patients.3

1.4.4 Ischemic and Inflammatory Responses in Stroke

Both in vitro and in vivo models of cerebral ischemia have shaped our understanding of the

pathophysiology of stroke.177, 178 A stroke occurs when cerebral blood flow is momentarily cut-off,

depriving brain tissue of crucial oxygen and glucose that is required for normal brain function. When this

occurs, it can lead to irreversible brain injuries.178 Abrupt cessation of blood flow to a specific area of the

brain followed by reperfusion results in a chain of inflammatory events at the cellular and molecular

level, which in turn lead to ischemic damage. First and foremost, ischemia disrupts the activity of neurons

and triggers their cell death, however it effects the glia and vascular cells as well.177 Occlusion of the

middle cerebral artery (MCAO) results in the formation of an ischemic territory, which is the total area

effected by the stroke and includes the ischemic core and the ischemic penumbra.178 The ischemic core is

12

affected quickly and incurs the most damage, since it is the area where blood flow is most limited.178 The

core of the ischemic territory becomes necrotic mainly due to lack of sufficient energy supply for

adenosine triphosphate (ATP) production.177 ATP is necessary for the neuron to maintain a gradient of

charges across its membrane; when this does not occur, sodium and calcium ions build up in the

cytoplasm, leading to the influx of water, organelle dysfunction, cell membrane deterioration, and its

eventual demise (necrotic cell death).177 The ischemic penumbra, the area surrounding the ischemic core,

is hypoxic, meaning it has limited but sufficient blood supply from collateral vasculature to prevent cells

in this region from undergoing instantaneous cell death.178 In this way, the ischemic penumbra represents

a zone of cells at risk of becoming necrotic, however its fate is determined by a number of different

factors. Because blood supply in this region is adequate to maintain the production of ATP, cells in the

ischemic penumbra do not die immediately, however they are under stress because blood flow is not at an

optimum level.178 This undue strain increases the susceptibility to certain factors that may disrupt their

equilibrium and instigate necrotic cell death. One of these critical factors is excessive production of

glutamate in extracellular spaces, leading to NMDA receptor activation and a cascade of calcium

dependent events. Eventually proteases and free-radical producing enzymes are activated, and this may

instigate necrotic cell death or apoptosis in the ischemic penumbra.177 However, the magnitude of damage

in this region depends on the severity of the infarct and the energy status of the neurons involved.177, 178

(reviewed in 179)

Hypoxic and ischemic neurons set off the inflammatory response through the release of ‘danger

signals’ that trigger immune cells.179, 180 Studies involving animals as well as humans, have demonstrated

that “innate” immune cells, for instance, macrophages181 and neutrophils,182, 183 as well as “adaptive”

immune cells such as T-lymphocytes,184 invade the ischemic territory.185 The release of “danger signals”

from injured neurons, microglia and astrocytes (local immune cells of the brain) cause innate and adaptive

immune cells to release cytokines.180 Danger signals may include components released from within the

cell after it has undergone necrosis, and these signals are collectively referred to as danger-associated

molecular patterns (DAMPS).186 DAMPS can then associate with innate immune cells such as microglia

13

and macrophages, by binding to toll-like receptors (TLRs) expressed on the surface of these cells.186 This

interaction can prime the inflammatory response and instigate a series of events which feedback to

amplify the production of cytokines, causing further cell injury and anti-inflammatory cascades.179

Therefore, it is evident that a stroke can lead to damage at both gross and cellular levels. The next section

will discuss the role and association of pro- and anti-inflammatory cytokines in acute ischemic stroke.

1.4.4.1 Cytokine Elevations Post-stroke

As illustrated above, cytokines released by innate and adaptive immune cells can wreak havoc on

neurons, microglia and astrocytes which feedback to amplify the immune response and cause further

destruction. After an ischemic stroke, a number of different pro-inflammatory cytokines are released.

Cerebrospinal fluid (CSF) and plasma of post-stroke patients has been demonstrated to contain raised

amounts of the pro-inflammatory cytokines IL-1β,99, 100 TNF-α,101 IL-699, 101-104 and IL-8.100 Researchers

have found IL-6 99, 103, 187 to correlate with larger stroke volume, stroke severity and functional

impairment103, 187 TNF-α has also been correlated with stroke severity and neurological outcome.188

Moreover, another study found that IL-6 and TNF-α were associated with risk of recurrent ischemic

stroke,189 and IL-6 has been demonstrated to remain elevated at 3 months99 and 1 year104 post stroke.

Anti-inflammatory cytokines such as IL-10 may have the opposite effect post-stroke. IL-10 production

has been demonstrated to act in a neuroprotective manner post-ischemic stroke 190 Briefly, animal and in

vitro models have demonstrated that IL-10 can significantly reduce infarct size191 and enhance neuronal

survival.192 Furthermore, it has been found that lower IL-10 is associated with clinical worsening193 and

larger infarct volume.102

Insight from ‘immunomodulation’ therapeutic strategies post-stroke, such as ‘ischemic

tolerance’, have highlighted the importance of a balance between pro- and anti-inflammatory mediators

in influencing the fate of neurons.179 Ischemic tolerance is an outcome that occurs when a sublethal

insult prevents an organ from sustaining a successive lethal insult.179 For instance, a transient, non-

injurious ischemic event in the brain will prevent the brain from incurring harm from a successive

14

injurious ischemic event.178, 179, 194 The success of ischemic tolerance relies on a balance between

immunosuppression, i.e. through interferon-β (IFN-β) mechanisms,195 and pro-inflammation,196-199

demonstrating that the equilibrium between pro- and anti-inflammatory cytokines has the potential to

have a positive or detrimental impact on the post-stroke brain, depending on their relative involvement in

the ischemic event. Yet, past research suggests that a skewed pro- to anti-inflammatory response is also

associated with depression. Myint et al. observed significant elevations in plasma IFN-Ɣ/IL-4 in

depressed subjects and upon antidepressant treatment, the level of IFN-Ɣ/IL-4 ratio decreased

significantly.200 Considering the importance of equilibrium between inflammatory mediators in ischemia

post-stroke and MDD, it seems plausible that this inflammatory balance is an important factor in the

development of PSD as well. The next part of this review will discuss this balance as it relates to sickness

behaviour.

1.4.5 Sickness behaviour

In mammals, sickness behaviour is a term used to describe a cluster of behavioural symptoms that

ensue due to acute infections or wounded tissue.57 These symptoms may include malaise, reduced pain

tolerance, fever, diminished energy, avoidance of social situations, lassitude, anorexia, poor concentration

and anxiety.201, 202 Pro-inflammatory cytokines have been intensely investigated in relation to sickness

behaviour, and much evidence points towards the involvement of TNF-α, IL-1β,203-207 and IL-6.202, 208, 209

Specifically, mice or rats injected with IL-1β or TNF-α centrally or systemically exhibit a wide range

symptoms indicative of sickness which is temporal and dose-dependent.202 These pro-inflammatory

cytokines also reduce the expression of clock gene transcripts that are important for the maintenance of

diurnal rhythms.210, 211 The role of IL-6 in cytokine-induced sickness behaviour is different than IL-1β and

TNF-α, in that its administration does not induce any behavioural symptoms,202 although

lipopolysaccharide (LPS) induced sickness behaviour is diminished in mice that produce less IL-6. 208

Therefore, IL-6 is believed to play a role in sickness behaviour by inducing the expression of other

cytokines in the brain.212

15

The severity and extent of sickness behaviour is controlled by anti-inflammatory cytokines. It is

hypothesized that they exert their effects by blocking pro-inflammatory cytokine expression and by

hindering cross-talk and communication with inflammatory pathways.205, 208 For example, LPS induced

sickness behaviour is dampened by central IL-10 injection.206 Furthermore, fever brought on by LPS

administration is more severe and drawn out in mice that produce less IL-10.212 These findings emphasize

the importance of a balance between pro- and anti-inflammatory cytokines in regulating sickness

behaviour. An aged brain also demonstrates the importance of this balancing act; older mice have

elevated expression of pro-inflammatory cytokines and reduced expression of anti-inflammatory

cytokines, for instance IL-6210 and IL-10,211 and their sickness response to LPS injection is more

pronounced and intense than in young mice.213 (Reviewed in 209)

1.4.6 Inflammatory Markers and Depression

The behaviours displayed by depressed individuals are quite similar to those exhibited in sickness

behaviour. These include sadness, fatigue, anorexia/weight loss, sleep disturbance, reduced pain

tolerance, anhedonia and difficulty concentrating.118 In patients who receive immunotherapy in the form

of IFN-α or IL-2, approximately 33% experience symptoms of depression.214 Additionally, researchers

assessing otherwise healthy MDD patients have consistently demonstrated that inflammation is

exaggerated in these individuals. For instance, patients with depression have been found to have elevated

acute phase proteins, chemokines, adhesion molecules and pro-inflammatory cytokines in their serum and

cerebrospinal fluid (CSF).214-229 In particular, cytokines and acute phase proteins associated with

depression include IL-6, CRP,215-223, 225, 228 TNF-α, IL-1β,218, 220, 223-225, 230-233 α-1-acid glycoprotein, α-1-

antichymotrypsin and haptoglobin, 219, 223, 225 and human macrophage chemoattractant protein-1, soluble

intracellular adhesion molecule-1 and E-selectin are some of the adhesion molecules and chemokines

reported to be elevated in depression.214, 226 Furthermore, multiple studies have observed a reduction in

ratios between pro- and anti-inflammatory cytokines with antidepressant treatment 32-35 and genetic

variations in IL-1β and TNF-α alleles have been demonstrated to be associated with antidepressant

16

treatment resistant depression and elevated risk of developing depression.234, 235 Patients with depressive

symptoms have significant elevations in inflammatory molecules as well.214, 215, 233, 236 Specifically, Suarez

et al. observed that higher Beck Depression Inventory (BDI) scores were correlated with elevated TNF-

alpha and IL-8 production.236

Moreover, patients burdened with diseases such as cancer,237-239 CVD240, 241 and viral infections232,

242 exhibit depressive symptoms which are also associated with inflammation.214 In particular, multiple

studies have found that inflammation after a stroke is associated with PSD. Researchers have found an

association between PSD and serum TNF-α and IL-6 36 as well as serum IL-18 post-stroke.243

Immunologic ratios of TNF-α/IL-10 and IL-6/IL-10 were also demonstrated to be significantly elevated in

patients with PSD compared to controls,36 and a study assessing IL-4 and IL-10 polymorphisms in

patients with PSD, which related to lower anti-inflammatory cytokine production, found that these alleles

were significantly associated with depression in these individuals.107 Conversely, some studies have failed

to find a relationship between the post-stroke cytokine production and depressive symptoms. One such

study by Jimenez et al. failed to find a relationship between the cytokines IL-1β, TNF-α, IL-6, as well as

C-reactive protein, although they did find that serum leptin levels were independently associated with

PSD and that patients with MDD had significant elevations in leptin levels at discharge and one month

post-stroke.244 Leptin is an adipocyte hormone that has an important role in regulating appetite and energy

equilibrium,245 however recent research has also focused on its role in neuronal plasticity246 and

protection244, 247, 248 Furthermore, leptin has also been studied for its potential to act as a biomarker for

vascular risk factor in stroke and heart attacks249, 250, and higher serum levels have been associated with

depression in the past.244, 251-253 Although these researchers failed to find a relationship between cytokines

and PSD, their finding on the relationship between leptin and PSD may represent a new avenue for

depression biomarkers post-stroke. Another study involving IL-18 in acute stroke found that serum IL-

18 was significantly elevated in alexithymic (emotionally unaware) patients but not in depressed

patients.254 In addition, a study by Ormstad et al. investigating the relationship between 13 cytokines,

depression and post-stroke fatigue, determined that depression was not associated with any of the

17

cytokines studied.255 However, post-stroke fatigue was significantly positively associated with acute

serum levels of IL-1β 6 months post-stroke and acute serum levels of IL-ra and IL-9 were negatively

correlated with post-stroke fatigue at 12 months post-stroke.256 Interestingly, IL-ra, an endogenous

antagonist to IL-1β, is involved in neuronal protection mechanisms257 and has been shown in animal

models to minimize sickness behavioural displays.256, 258 This suggests that some of the depressive

symptoms observed post-stroke may be more related to a sickness response than a true depression

response.

Although sickness behavior and depression are both associated with elevations in inflammatory

markers and mirror each other in terms of the behavioural symptoms, their differing durations suggest that

their etiologies are unique. Sickness behaviour is a motivational259 and adaptive response260 during

infection which will resolve once the body eliminates the pathogen,261 however depression does not

follow the same course.209 As mentioned earlier, about a third of patients receiving IFN-α or IL-2

immunotherapy experience symptoms of depression214 and investigation into the cause of this helped

researchers paint a picture of two forms of cytokine-induced depressive symptoms.209 One of which are

the early-onset somatic disturbances of treatment that most, if not all patients experience and these consist

of lethargy, decreased food intake, increased sensitivity to pain and disturbances in sleeping.209 The other

form of cytokine-induced depressive symptoms are psychological in nature, late-onset, and are exhibited

in as many as 50% of these patients. These symptoms involve mild disturbances in cognition, and a

depressed, anxious or irritable mood.209, 262-264 Interestingly, patients receiving immunotherapy that were

treated prophylactically with an SSRI demonstrated improvement in the aforementioned mood

disturbances without any changes in somatic depressive symptoms.262 Furthermore, patients’ depression

rating scores prior to treatment initiation, determined by the Montgomery–Asberg Depression Rating

Scale (MADRS), were significantly associated with depressive symptoms at the end of treatment,265 and

patients who experienced the psychological symptoms of depression after immunotherapy also had

significantly elevated plasma levels of cortisol and adrenocorticotropic hormone (ACTH) which are

indicative of a heightened pituitary-adrenal response.209, 266 Dantzer et al. suggest that these findings

18

indicate both physiological and psychological mechanisms in the susceptibility to developing depression

from cytokine immunotherapy.209 He also adds that “It is possible that depression represents a

maladaptive version of cytokine-induced sickness, which could occur when activation of the innate

immune response is exacerbated in intensity and/or duration...”209 This may also be the case in the

development of PSD, since stroke is associated with marked elevations in pro-inflammatory cytokines.267

The next section of this literature review will outline some of the biological mechanisms involving

inflammation-associated depression; hopefully shedding light on factors that incur ‘increased

vulnerability’ to this form of depression.

1.4.7 IDO-mediated Mechanisms of Depression

1.4.7.1 Kynurenine Pathway and its Metabolites under Normal and Inflammatory Conditions IDO is an enzyme involved in the conversion of TRP to KYN. Under normal physiologic

conditions the enzyme tryptophan 2,3-dioxygenase (TDO) is mostly responsible for this process, which

takes place in the liver.268 However, inflammatory processes trigger a shift in enzyme activity to IDO.269

This shift in metabolism occurs mainly due to elevations in pro-inflammatory cytokines during an

immune response which induce IDO.270, 271 IDO gene transcription is most potently activated by IFN-Ɣ,22,

23 however other pro-inflammatory cytokines such as IFN-α,22 IFN-β,24 IL-2,25 IL-6,26 IL-18,27 and TNF-

α28 have also been demonstrated to induce IDO activity. Anti-inflammatory cytokines, such as IL-429 and

IL-1030 can inhibit IFN-Ɣ-stimulated IDO activation. IDO is present in the lungs, kidneys, spleen, blood

and brain,272, 273 although the majority of IDO catalytic activity takes place in lymphoid tissues and

blood.274, 275 Glucocorticoids released by the adrenal glands during stress can also upregulate TDO

activity in the liver,276, 277 causing further increases in KYN.275

Under normal physiologic conditions, KYN metabolised from TRP contributes to energy stores

through a pathway that leads to ATP production. As seen in Figure 1 the steps involved in this pathway

are as follows: in the liver, an enzyme called kynurenine-3-monooxygenase (KMO) converts KYN into 3-

hydroxykynurenine (OHK), OHK is further metabolized by an enzyme called kynureninase to 3-

19

hydroxyanthranilic acid (HAA), HAA can then be completely oxidized to make ATP or catabolised by 3-

hydroxyanthranilic acid oxygenase into quinolinic acid (QUIN) which will eventually be converted to

nicotinamide adenine dinucleotide (NAD) by quinolinate phosphorybosyltransferase.278 ATP is the main

product of HAA under normal conditions.278 KYN also has the potential to undergo conversion to

kynurenic acid (KYNA) by an enzyme called kynurenine aminotransferase.275, 278

Figure 1. TRP and KYN Metabolism Pathway (Adapted from Mandi and Vecsei105 and Myint et al.275)

20

As mentioned above, the majority of TRP metabolism normally occurs in the liver,268 however a

small amount can take place elsewhere, such as the brain. In the brain, the main site of TRP metabolism is

within astrocytes and microglia.279, 280 Most KYNA production occurs in astrocytes, while QUIN

formation takes place in macrophages and microglia.281-283 QUIN is then taken up by astrocytes281 and

mainly used to produce NAD and enable glycogen storage in the brain.275, 284, 285

Under inflammatory conditions, IFN-Ɣ,22, 23 IFN-α,22 IFN-β,24 IL-2,25 IL-6,26 IL-18,27 and TNF-

α28 induce IDO activity. Pro-inflammatory cytokines can also stimulate the activity of KMO272 leading to

increased production of OHK and therefore causing QUIN levels to increase.30, 286 Anti-inflammatory

cytokines, such as IL-4 and IL-10 are capable of significantly reducing QUIN production stimulated by

TNF-α and IFN-Ɣ in human monocyte-derived macrophages.30 Until inflammation is reduced or

completely subsides, QUIN will continue to be produced.286 Picolinic acid, a metabolite of the complete

oxidation of HAA, was demonstrated to significantly increase IFN-Ɣ-dependent expression of TNF-α

mRNA,287 therefore the KYN pathway itself may have a role perpetuating QUIN production as well.275

During the inflammatory response, IDO activation may also play a pivotal role in immune

regulation and suppression.105 Its activation may suppress the immune response by reducing the

proliferative capacity and inducing apoptosis of T lymphocytes. In general, this may be carried out

through decreased TRP availability,288 increased production of oxidative and cytotoxic KYN

metabolites,289-293 and lastly by indirectly causing naïve cluster of differentiation (CD) 4 + T cells to

differentiate into regulatory T cells (Treg cells) through the production of transforming growth factor

(TGF- β).294 Recent evidence has suggested that this last mode of action is carried out via the aryl

hydrocarbon receptor (AHR). It has been demonstrated that KYN can bind to the AHR, leading to IDO

induction and the generation of (forkhead box P3) Foxp3 Tregs; TGF-β may also amplify the interaction

between KYN and AHR by upregulating AHR expression.295 The general result of IDO activation is a

dampened helper T cell type 1 (Th1) response,291 which shifts the immune response to helper T cell type 2

(Th2) activation, while also stimulating the production of Tregs. Th1 and Th2 cells produce distinct

antagonistic cytokines that aim to repress each others’ immune response and to perpetuate the production

21

of their own cytokines.105, 296 Th1 cytokines include IFN-Ɣ, TNF-α, IL-1, IL-18 and while Th2 cytokines

include IL-6, IL-4, and IL-10.105 Consequently, Tregs have a repressive effect on both the Th1 and Th2

responses, driving the immune system towards equilibrium.105, 297

Aside from the effects of KYN on immunosuppression, inflammation-mediated IDO activation

may affect neurotransmitter levels and synaptic transmission in the brain. As mentioned previously,

tryptophan is an essential amino acid that acts as a precursor for serotonin synthesis.298 Therefore,

elevated IDO activation may reduce the amount of tryptophan available for serotonin synthesis275 and

influence the rate of serotonin degradation by IDO’s ability to convert serotonin into formyl-5-

hydroxykynuramine.299 These two effects of IDO activation may compromise the synaptic transmission of

serotonin.275 Furthermore, QUIN has been demonstrated to act as an N-methyl-d-aspartate (NMDA)

receptor agonist300 and animal models have demonstrated its excitotoxic and neurodegenerative

capacities.75, 301, 302 KYNA, on the other hand, at higher than normal concentrations acts as an antagonist

to multiple ionotropic excitatory amino acid receptors. 21 At physiological concentrations, KYNA acts as

a competitive antagonist of the NMDA receptor303 and researchers have found that neurotoxicity brought

on by QUIN is diminished significantly by KYNA.301 KYNA has also been found to exhibit antagonistic

properties at the α7-nicotinic acetylcholine receptor (a7nAchR) at physiologic concentrations304 However,

Kynurenine amino-transferase, the enzyme responsible for the conversion of KYN into KYNA, has not

been demonstrated to be upregulated by cytokines. Therefore, inflammation may create an imbalance

between QUIN and KYNA, potentially increasing neurons’ susceptibility to excitotoxic

neurodegeneration. Taken together, these findings suggest that inflammation-induced IDO activation may

have a negative impact on many mechanisms of synaptic transmission. The following segment of this

literature review will highlight on the relationship between KYN, its metabolites, and depression, as well

as touch upon certain mechanisms in which toxic KYN metabolites may lead to neurodegeneration and

therefore hypothetically, depression.

22

1.4.7.2 Kynurenine and Psychiatric Disorders

Changes in the levels of KYN and its metabolites have been associated with major depression,88-

92, 305 bipolar mania,82, 83 and schizophrenia.84-87 With respect to depression, Myint et al. found the ratio of

K/T, KYNA and KYNA/KYN in depressed patients was significantly different than controls.88

Specifically, the K/T ratio was significantly elevated in depressed patients, and KYNA, as well as

KYNA/KYN levels were significantly reduced in depressed patients.88 Another study by Steiner et al.

discovered significantly elevated QUIN immunohistochemistry staining in the subgenual anterior

cingulated cortex and anterior midcingulate cortex in depressed patients compared to controls.89 KYN

was also shown to be elevated in MDD patients who attempted suicide compared to controls.306 Post-

partum depression has also been associated with elevations in the K/T ratio.307, 308 Additionally, in

adolescents with melancholic depression, researchers using magnetic resonance spectroscopic imaging

observed that levels of KYN and the HAA/KYN ratio were positively associated with increased cell

turnover, as measured by choline levels in the right caudate and left putamen, respectively.90 Choline is a

brain metabolite associated with neuronal membrane breakdown that has been associated with MDD in

the past.90, 309, 310 With respect to PSD, no study to our knowledge has examined the relationship between

the K/T ratio or KYN metabolites and depression. The only closely related study was conducted by our

group in coronary artery disease (CAD), and we found a significant association between depression

scores and the K/T ratio.92 CAD also shares common risk factors with stroke311 and a study assessing

independent predictors of CAD, using cardiac computed tomography (CTA), found that acute ischemic

stroke is independently associated with increased risk and more severe CAD in contrast to patients with

“acute chest pain at low-to-intermediate risk for acute coronary syndrome.”312 Moreover, the latest results

from studies measuring the levels of TRP, KYN and its metabolites in relation to depression have

demonstrated that disruptions in the KYN pathway may be more relevant to the symptoms of

somatization.57, 313 This has been defined in the past as a “multisomatoform disorder characterized by

medically unexplained, functional or psychosomatic symptoms”313 and within the general population it

23

has a prevalence of 4-7%.314 Maes and colleagues demonstrated this by conducting a study evaluating

KYN metabolism in depressed patients, patients with comorbid depression and somatization, patients

with just somatization, and controls.305 They found that plasma TRP was lower in patients with

somatization than depression and the KYN/KYNA ratio as well as the K/T ratio were elevated

significantly in somatization compared to depression.305 Furthermore, reduced TRP and increased

KYN/KYNA and K/T were associated with somatic symptom severity.57, 305, 313 Their group attributes

these results to the findings that: a reduction in serotonergic activity increases pain,315 KYN can also

stimulate pain and peripheral neuropathy,316 and KYNA can prevent these symptoms317 potentially by

blocking the nociceptive NMDA receptor or by stimulating the activity of the anti-nociceptive G-protein

coupled receptor-35.313, 318

In studies involving cytokine immunotherapy, the relationship among inflammation, cytokines,

and kynurenine metabolites has been well established. One study observed significant increases on the

MADRS score in hepatitis C patients treated with 24 weeks of IFN-α immunotherapy.319 They also found

increases in the K/T and KYN/KYNA ratio, which was significantly associated with total MADRS scores

over time. 319 Another study by Raison et al. found that hepatitis C patients treated for 12 weeks with

IFN-α had significantly elevated peripheral (blood) and central (CSF) concentrations of KYN.106

Elevations in CSF KYN were correlated with marked elevations in CSF QUIN and CSF KYNA.

106Furthermore, CSF QUIN and CSF KYN were associated with CSF IFN-α and depressive symptom

scores on the MADRS.106 CSF QUIN and KYN were also associated with TNF-α receptor 2 and MCP-

1.106 Of note, the concentration of TRP in the CNS depends on the concentration of large neutral amino

acids (LNAA) (tyrosine, phenylalanine, leucine, isoleucine, valine) which compete with TRP to be

transported across the blood brain barrier (BBB), and is also reliant on the central demand for TRP.275, 298

Furthermore, approximately 60% of KYN in the brain is contributed from the systemic circulation320 and

which is transported across the BBB by the large amino acid transporter.19 KYNA and QUIN, however,

enter the CNS through passive diffusion, this implies that central KYN metabolism contributes to the

24

majority of CSF QUIN and KYNA.19, 106, 321 The results of this study by Raison et al. suggest that

depression caused by IFN-α therapy is associated with KYN metabolism as well as central

inflammation.106

1.4.7.3 Kynurenine Metabolites and Neurodegeneration

KYN metabolites have been associated with neurodegenerative diseases in the past. For instance,

researchers discovered 3-4 fold increases in the levels of OHK and QUIN in the neostriatum and cortex of

Huntington’s disease patients,64 and QUIN has also been linked to AIDS-associated dementia.65 OHK was

also demonstrated to cause striatal neuron toxicity through the formation of reactive oxygen species.322

Elevations in the K/T ratio have been significantly correlated with the progression, severity and mortality

multiple neurodegenerative diseases, such as Alzheimer’s disease, Parkinson’s disease, Huntington’s

disease, as well as stroke.61-63, 323 The development of depression has also been hypothesized to have ties

with mechanisms of neuroprogression.50 Neuroprogression is a term given to the process of gross and

cellular neurodegeneration which has been hypothesized to have a link to the cause and/or consequence of

worsening depression and cognitive decline.50-54 In one case register study the rate of dementia increased

by 13% with every depressive episode leading to hospital admission.55 Furthermore, the number of

depressive episodes and length of each episode has been associated with diminished brain volumes.51, 56

For instance, an increased number of depressive episodes is linked with decreased hippocampal, basal

ganglia, orbitofrontal and subgenual volumes, and the extent of the episode has been associated with

diminished cerebral grey matter volume.56, 57 In a meta-analysis by McKinnon et al. it was determined

that compared to controls, MDD patients had reduced hippocampal volumes, and patients with ‘moderate’

(2–9 years) and chronic (>10 years) illness duration had smaller hippocampal volumes as well.58 Another

meta-analysis by Videbech et al. observed an 8% decrease in right hippocampal volume and 10%

decrease in left hippocampal volume in MDD and right hippocampal volume was negatively correlated

with the number of depressive episodes.59 In a study comparing first episode, never-treated depressed

patients to multiple episode depressed patients, researchers observed both patient types to have cognitive

25

deficits in recollection memory, which they attributed to hippocampal dysfunction.60 Furthermore, only

patients with recurrent episodes had decreased hippocampal volumes and statistical analysis demonstrated

that length of illness is significantly correlated with hippocampal volume reduction.57, 60 Taken together,

these studies suggest that the process of neuroprogression may have a role in the etiology of depression.

There is some research that implicates inflammation and oxidative stress in the development of

neuroprogressive depression as well.50, 67, 68 57 In particular, inflammation-induced IDO and KMO

activation and the production of KYN metabolites have been hypothesized to play a role in these

neurodegenerative processes,50, 67, 68 through pro-oxidative mechanisms,69-72 disruption in energy and

metabolism,73, 74 and through glutamate excitotoxicity.21, 75, 76 This last mode of action is carried out by the

previously mentioned KYN metabolite, QUIN, which has been demonstrated to act as an NMDA receptor

agonist300 and animal models have demonstrated its excitotoxic and neurodegenerative capacities.21, 75, 301,

302 Interestingly, inflammation in the systemic circulation can induce the expression of NMDA receptors,

potentially further increasing the excitotoxic effects of QUIN.313 Furthermore, KYNA production occurs

in astrocytes, while QUIN formation takes place in macrophages and microglia,281-283 and QUIN is then

taken up by astrocytes281 and mainly used to produce NAD and enable glycogen storage in the brain.275,

284, 285 However, at high concentrations QUIN inhibits the uptake of glutamate by astrocytes, which may

lead to increased synaptic concentrations of glutamate and eventually cause increased glutamatergic

transmission.324 Interestingly, observations in MDD patients have demonstrated decreased density of

astrocytes in brain regions associated with depression.76, 310 Researchers have demonstrated that the

neurodegenerative capabilities of QUIN can also occur through mechanisms other than glutamatergic

excitotoxicity.76 One study found that QUIN exposure to hippocampal nerve cells in vitro caused

extensive degeneration, acute swelling and damage to postsynaptic elements.77 Another in vitro study

measuring lipid peroxidation in rat brain homogenate, found QUIN increased lipid peroxidation in a

concentration-dependent manner.72 Whether or not these mechanisms represent acute or chronic factors

the etiology of PSD has yet to be determined. However, one study involving cardiac arrest patients found

marked reductions in hippocampal volume compared to controls 21 days after cardiac arrest.325 The

26

researchers of this study postulated that this reduction in volume signifies the hippocampus’ increased

susceptibility to global brain ischemia.325 This may imply that oxidative and neurotoxic factors can have a

relatively acute effect on brain volume and neurodegeneration. However, because of the cross-sectional

nature of this study one cannot determine whether the hypoxic (or oxidative/excitotoxicity/etc.)

consequences are acute or chronic.

KYN and its metabolites have also been studied in relation to stroke. Darlington et al. observed

an elevated K/T ratio post-stroke, found that KYNA levels were significantly elevated in patients who

died within 3 weeks of stroke, and also demonstrated that the HAA/AA ratio was significantly correlated

with lesion volume.66 As mentioned above, at physiological concentrations, KYNA acts as a competitive

antagonist of the NMDA receptor303 and has also been found to exhibit antagonistic properties at the

α7nAchR at physiologic concentrations304 One explanation the researchers give for the positive

association between KYNA levels and mortality is that excessive KYNA production may be an adaptive

physiological mechanism in response to higher than normal glutamatergic transmission that occurs due to

ischemia.66 In turn, elevations in this glutamate antagonist might cause glutamate hypofunctioning and

leave already compromised post-stroke patient vulnerable to death.66 Another research group assessing

the relationship between KYN metabolites and stroke outcomes found that the K/T ratio correlated with

stroke severity and infarct volume.326 Furthermore, worse stroke outcome at 3 months was associated with

a higher K/T ratio.326 In patients with traumatic brain injury researchers observed increased conversion of

TRP to KYN compared to controls, and decreased production of KYNA.327 The multiple findings that

depression, as well as KYN metabolites have been associated with neurodegeneration and stroke, and

KYN metabolites are associated with neuroprogression, suggests that the combined effect of an elevated

K/T ratio and neurodegeneration may be more relevant in contributing to the development of PSD, than

either variable alone. The subsequent section of the introduction will explore the relationship between

tryptophan, serotonin, and depression, as these have also been implicated in the development of MDD,

and may also have a role in the etiology of PSD.

27

1.4.8 Serotonin Hypothesis of Depression

Nearly 4 decades ago the serotonin hypothesis of depression was created, postulating that a

reduction in serotonergic transmission was an important mediator in the onset of depressive symptoms.328

This idea came from the finding that synaptic reuptake of serotonin and noradrenaline was blocked by

TCAs; researchers believed that this mechanism of improved serotonergic transmission was the key to

improvement of symptoms in depressed individuals, and the advent of SSRIs in the pharmaceutical world

gave further credence to this hypothesis.329

A combination of positron emission tomography (PET), single photon emission tomography

(SPET) and ligand imaging techniques have been used in recent years in an attempt to uncover the link

between serotonin and depression.330 Findings from those studies demonstrate that 5-H1A receptors have

diminished bindings in regions such as the frontal,331 temporal,331, 332 and limbic cortex,331 as well as the

raphe nuclei in unmedicated depressed patients,331, 332 although these results have not been consistently

replicated.333 Moreover, reduced 5-H1A binding in recovered depressed patients signifies that this may not

represent acute depression334 and binding at this receptor in the context of panic disorder is also

reduced335 raising the concern of how unique these findings are to depression relative to other mood

disorders.336 Furthermore, a study investigating 5-HT2A receptor binding observed significantly higher

binding potential in the frontal cortex, parietal cortex, and occipital cortex of recovered depressed

individuals.337 Conversely, other studies have demonstrated reduced,338 as well and no change339 in

binding density at the 5-HT2A receptor. With regards to the serotonin transporter (SERT), a study by

Malison et al. found significant decreases in binding potential at SERT in the brainstem of unipolar

MDD patients,340 but yet again, another group of researchers failed to replicate these findings; with no

change in SERT binding in the midbrain between MDD patients and controls.341 (reviewed in 336)

In PSD, there is evidence that the levels of serotonin metabolites in CSF are also altered. A study

observing the differences in CSF 5-HIAA between depressed and non-depressed stroke patients, as well

as controls, found that 5-HIAA is significantly reduced in PSD patients compared to the other two patient

28

groups.342 Another group of researchers investigated the differences in plasma and CSF serotonin

concentrations in PSD patients compared to non-depressed stroke patients.343 They observed overall

plasma and CSF serotonin concentrations in the PSD group to be lower than in the control group, as well

as a 90% reduction in plasma serotonin in the PSD group compared to a 13.3% reduction in the control

group, and also observed an 80% Reduction in CSF serotonin in the PSD and whilst the control group had

a mere 6.7% reduction in CSF serotonin.343 Additionally, a study by Mayberg et al. observed increased 5-

HT2 binding in PSD patients compared to stroke patients without depressive symptomatology, and this

was significantly associated with symptom severity as well.344

Evidence from animal and in vitro studies suggest that certain modes of serotonin

neurotransmission may be important in the development of immune-mediated depression. For instance,

researchers have demonstrated that serotonin turnover and tryptophan uptake (the precursor to serotonin

synthesis) are stimulated by LPS or pro-inflammatory cytokine administration.209, 345 Additionally,

through p38 mitogen activated kinase (MAPK) mechanisms, it was demonstrated that IL-1β and TNF-α

are both able to acutely regulate the neuronal serotonin transporter (SERT).209, 346 These researchers

consequently demonstrated that LPS administration in mice caused increased SERT activation and

behavioural despair, which did not occur in mice deficient in the interleukin-1 receptor (IL-1R).347

Furthermore, SERT knockout mice administered with LPS did not perform worse in behavioural models

of despair.347 In a separate study, researchers observed a 72% decrease in the expression of the 5-HTR1A

serotonin receptor when hepatoblastoma, myelocyte-derived and T cell leukemia-derived cell lines were

treated with IFN-α, this effect was diminished with either cessation of IFN-α treatment or administration

of a TCA (desipramine) or a serotonin reuptake inhibitor (fluoxetine).209, 348 Recent findings in humans

corroborate these results; patients treated with IFN-α demonstrated significant increases in p38 MAPK

activation, this activation was significantly more pronounced in patients that developed depression, as

assessed by the MADRS.349 These results suggest a possible role of serotonin synaptic transmission in the

development of cytokine-associated depressive symptoms and may be associated PSD in the acute stages

29

post-stroke. However, whether dysfunctional serotonergic transmission is a mediator in the development

of PSD is still a topic of debate.

1.4.8.1 Tryptophan and Depression

Tryptophan (TRP) is an essential amino acid that is used by the body to produce serotonin,19, 20

kynurenine,350 melatonin351 and the trace amine tryptamine.352 In the systemic circulation TRP exists in

two forms: albumin-bound and free, although at basal states the majority (90%) is bound to albumin,

which cannot be transported across the blood-brain barrier (BBB).353, 354 Thus, TRP transport across the

BBB depends on the amount of albumin-bound and unbound TRP,355 as well as the relative amount of

TRP compared to the concentration of large neurtal amino acids (LNAA), which compete for the binding

site on the L-type amino acid transporter at the BBB.356, 357 Once TRP crosses the BBB, it is metabolized

into serotonin in the raphe nuclei, a small cluster of neurons in the brainstem. The rate-limiting enzyme

involved in the conversion of TRP to serotonin is tryptophan hydroxylase (TPH).358, 359 (reviewed in93)

‘Tryptophan depletion’ has been used by many researchers to study the effects of reduced

serotonin activity on depressed mood. This method came from the finding by Cowen et al. that total

plasma tryptophan was significantly reduced in patients with depression compared to controls,360 and it

involves giving patients an amino acid mixture that is deficient in TRP. Reductions in TRP will reduce

the amount of TRP catalyzed by TPH, and therefore transiently reduce serotonin synthesis.361 This

method demonstrated that patients who had a depressive episode in the past and subsequently recovered,

were sensitive to the effects of tryptophan depletion, causing temporary relapse in depressive symptoms

in these individuals.96, 97 Conversely, the depressive effects of tryptophan depletion in healthy volunteers

without risk factors for developing depression were not significant.361 Cowen et al. suggest these findings

indicate that patients who have vulnerability factors to depression and have experienced depression in the

past, are more likely to develop depression after diminished serotonin activity.336 They further suggest

that depression may permanently alter the neurobiology of the serotoninergic system so that when

30

previously depressed individuals are subjected to tryptophan depletion, they display a depressed response.

(reviewed in 336)

Studies by Maes et al. and Hoes et al. measuring the excretion of xanthurenic acid (XA), a

metabolite of the KYN pathway, in depressed patients, were the first to postulate a relationship between

IDO activity, TRP depletion, and depression.362, 363 Their results demonstrated that a ‘shunt’ in KYN

metabolism by way of IDO activation, reduced TRP levels in depressed patients.57, 362, 363 This same

mechanism is plausible in the context of PSD, where stroke is characterized by marked elevations in

cytokines,99-104 leading to IDO activation,28, 274 an increase in the conversion of TRP to KYN, an elevated

K/T ratio, and a subsequent reduction in the amount of TRP available for serotonin synthesis. Studies

involving cytokine immunotherapy corroborate this theory, finding significant reductions in serum TRP

upon cytokine administration.364, 365 Furthermore, Maccay et al. demonstrated that at baseline and after

TRP depletion, patients with traumatic brain injury had increased conversion of TRP to KYN compared

to controls, and decreased production of KYNA.327 In a study assessing plasma levels of TRP in

rehabilitative stroke patients, it was found that patients on antidepressants had plasma tryptophan

concentrations approximately 70-80% of the mean value measured in patients not using

antidepressants.366 The results of those studies suggest that a reduction in TRP following a stroke may be

associated with PSD, however research is scarce on the topic. The following portion of this review will

examine the relationship between neuroimaging outcomes such as atrophy, white matter lesions, and

acute infarct volume with PSD, as this field has put much effort into determining this post-stroke

sequelae’s neuroanatomical roots.

1.4.9 Neuroanatomical Correlates of Post-Stroke Depression

Few studies have assessed the relative effects of acute infarcts, WMC, and atrophy together in

relation to PSD.39, 40, 45, 48 The first study to do so, by Vataja et al., hypothesized that an intricate

relationship between these neuroanatomical findings contribute to PSD in a similar way to their

interaction within fronto-subcortical pathways in the development of post-stroke dementia.40, 367 These

31

researchers failed to find associations between both WMC or atrophy and PSD,40, 45 while subsequent

studies assessing all three outcomes have found associations with WMC48 and atrophy39 in isolation,

without any additive effects. Similar to these aforementioned groups, this thesis aims to take a

comprehensive approach to assessing neuroanatomy as it relates to PSD, however an overview of the

literature on previously established relationships between each radiologic finding and PSD is necessary in

order to justify this approach.

1.4.9.1 Lesion Location

The Lesion Location Hypothesis was first proposed in the late 1970s by Robinson and colleagues,

and it was based on the idea that discrete lesions in the aminergic mood and emotion centres of the brain

can cause depression in otherwise healthy individuals post-stroke.138, 142, 368 The first CT study by

Robinson et al. investigating this theory found that the distance of the lesion to the frontal pole was

inversely correlated with depressive symptom severity in left hemispheric strokes .140 A subsequent study

by this group published a year later demonstrated similar results, 369 fueling the notion that left

hemisphere frontal strokes are more highly associated with PSD than any other region.44 This theory then

developed into the idea that there were left anterior lesion preferences in PSD patients, from the finding

that stroke patients with single lesions in the left anterior quadrant of the brain had elevated depression

scores compared to any other quadrant (left posterior, right anterior, right posterior); that group was also

more likely to develop MDD than the others.138 The justification put forward for those findings was that

monoaminergic (noradrenergic and serotonergic) neural circuits involved in the regulation of mood travel

from the brainstem to the frontal cortex and travel in an anterior to posterior fashion, stretching from deep

to more shallow layers of the cortex.370, 371 If fibers from this circuit were damaged by a lesion closer to

the frontal pole, a more pronounced decrease in these neurotransmitters would ensue, causing depressed

mood. However, the findings of many other studies were not in agreement with the left anterior lesion

preference of PSD. 130, 372-376 The work of Starkstein et al. added onto previous findings by demonstrating

that the contribution of lesions within the basal ganglia and thalamus were equally as important as cortical

32

strokes in their association to PSD.141, 142, 377 These observations were rationalized by the fact that the

basal ganglia is involved in five frontal-subcortical loops involved in motor control, cognitive function

and mood.378 Three of which are associated with depression, and encompass the orbitofrontal cortex

(OFC), dorsolateral prefrontal cortex (DLPFC) and anterior cingulate (ACC) regions (reviewed in379).

Furthermore, Herrmann et al. studied the contribution of basal ganglia lesions in PSD380-382 and found that

MDD patients had greater involvement of the basal ganglia in lesions than non-depressed stroke

patients.380 Other researchers attempted to correlate depression with lesion locations characterized by

specific brain regions.44 For instance, Singh et al. found that depressive symptoms were most significantly

predicted by inferior frontal lesions,125 while Kim and Choi-Kwon demonstrated that the prevalence of

PSD was higher in the frontal-temporal region than parieto-occipital regions.383 The findings from three

meta-analyses complicated the state of the literature even further, with one meta-analysis by Singh et al.

finding that methodological disparities were too vast to combine data from studies.384 While Carson et al.

found no significant preference of lesion laterality in PSD385 and Bhogal et al. observed a difference in

lesion laterality preference between in patient and community stroke survivors.386 Hospital patients with

strokes in the left hemisphere had a higher likelihood of developing depression than hospital patients with

strokes of the right hemisphere, while the opposite was found for out patients.386 Additionally, some

evidence suggests a possible contribution of timing of onset of depressive symptoms to lesion laterality,

with a meta-analysis by Robinson et al. finding that in the acute stage post-stroke (approximately 6

months) depression severity and distance of the lesion to the frontal pole in left-sided strokes are inversely

correlated.387 At this time, mixed findings in this area of PSD research demonstrate that more work must

be done to clarify the association between depression and lesion location. (reviewed in 44)

Few studies have explored the effect of lesion volume on depression post-stroke. Two studies by

Tang et al. failed to find a significant difference in lesion volume between depressed and non-depressed

stroke patients.38, 39 However, in a study by Vataja et al. it was observed that depressed patients had

significantly larger lesions in both the right and left caudate and pallidum.40 Furthermore, Nys and

colleagues found a significant difference in lesion volume between patients with moderate to severe

33

depressive symptoms, mild depressive symptom, and no depressive symptoms; patients with more severe

symptoms had significantly larger lesions than patients with mild or no symptoms.41 The findings of

Terroni et al. support these results with their observation that lesion volumes in the amygdala, ventral

anterior cingulated cortex, subgenual cortex, hippocampal subiculum, and dorsal anterior cingulated

cortex are significantly larger in patients with first major depressive episode post-stroke than non-

depressed patients.42 Moreover, another study by Zhang et al. demonstrated that depressed patients had

significantly larger infarct volumes on either the left or right side, however there was only a trending

difference in total infarct volume between depressed and non-depressed groups.43

1.4.9.2 White Matter Changes, PSD and the Vascular Depression Hypothesis

Age related white matter change (WMC) is another marker of neurodegeneration which may be

associated with the development of PSD. It is a disease of the small vessels and represents incomplete

ischemia mainly related to occlusive venous collagenosis.388 The walls of deep cerebral vessels undergo

hardening and thickening, due to vascular risk factors such as hypertension, and cause reduced blood flow

to this region, leaving the white matter vulnerable to cerebral ischemia.388-390 The damage that ensues can

cause dysfunction of axonal transmission and may result in a wide array of clinical impairments, from

rapid disruption of motor control391 to more gradual, slight changes in cognition, emotion, sensory or

motor function.392-394 In relation to emotion, a hypothesis was constructed by Alexopoulos et al. to

describe the unique cases of depression among elderly individuals.395 These cases were unique because

patients with ‘vascular depression’ have more cognitive impairment, particularly in fluency and naming,

more psychomotor retardation, poor insight, and less irritability and guilt than depressed patients without

vascular disease.396 In their seminal paper on vascular depression, Alexopoulos et al. stated: “

Cerebrovascular disease can predispose, precipitate, or perpetuate some geriatric depressive

symptoms.”395 Alexopoulos et al. classifies this as being clinically separate from non-vascular depression

due to the fact that there is increased prevalence of depression amongst patients with concomitant

vascular disease.397 Furthermore, there is lower prevalence of family history of mood disorders compared

34

to early-onset depression in the elderly,398 and lastly, WMCs in the context of depression are most evident

in the aforementioned frontal-subcortical loops,399-401 involved in mood regulation (DLPFC, ACC, and

OFC).379 With respect to late life depression, one meta-analysis found that the occurrence and severity of

white matter change is more pronounce in late life depression, with odds above 4 for white matter lesions

in late versus early onset depression.402 More recent studies using both CT403 and MRI404 have found

strong relationships between late-life depression and WMC. In particular, a study by Olesen et al. found

that in the elderly, more severe WMC is an independent correlate of MDD over 5 years.403 While results

from a 3-year follow up study indicate a causal relationship between WMC and late-life depression, with

the finding that the development of depression during the third year of study was significantly correlated

with the progression of WMC and was also a significant predictor of depression in year 3.404

Researchers have also sought to examine the relationship between WMC and depression post-

stroke, particularly with respect to the interaction between acute infarcts and the progressive, chronic

infarcts that are representative of WMC. The first study to assess this relationship in the realm of PSD

research was by Vataja et al. who examined the connection between whole brain atrophy, WMC, and

stroke lesion characteristics.40, 44 However, they observed that white matter lesions and whole brain

atrophy were not significantly different across groups with and without depression, the only significant

radiological findings were the aforementioned volume differences in the caudate, pallidum,40 and

putamen.45 Many other studies failed to find a relationship between white matter lesions and PSD.39, 41-43,

46, 47 Although, this cohort was then investigated by Pohjasvaara et al. using the BDI to rate depressive

symptom severity as opposed to the previous studies,48 which used DSM-III-R and International

Classification of diseases (ICD)-10 criteria to dichotomise clinically depressed versus non-depressed

patients40, 45 It was found that depressive symptoms were more severe amongst patients with subcortical

ischemic vascular disease, i.e. periventricular and deep white matter lesions.48 Another study by Tang et

al. demonstrated that compared to non-depressed patients, patients with depression had a greater chance

of having severe deep white matter lesions or frontal lobe infarct, and in a multivariate analysis severe

35

white matter change held as an independent predictor of depression post-stroke.38 Those results

demonstrate the importance of considering white matter lesions in the presence of acute infarcts when

assessing their association with depression, 44 as this thesis plans to consider as well.

1.4.9.3 Atrophy and Depression

As previously mentioned, researchers have extensively studied the complex relationship between

depression and atrophy. There have been many finding which indicate the role of neuroprogression in

MDD. For instance, the number of depressive episodes and length of each episode has been associated

with diminished brain volumes.51, 56 In particular, an increased number of depressive episodes are linked

with decreased hippocampal, basal ganglia, orbitofrontal and subgenual volumes, and the extent of the

episode has been associated with diminished cerebral grey matter volume.56, 57 In a meta-analysis by

McKinnon et al. it was determined that compared to controls, MDD patients had reduced hippocampal

volumes, and patients with ‘moderate’ (2–9 years) and chronic (>10 years) illness duration had smaller

hippocampal volumes as well.58 Another meta-analysis by Videbech et al. observed an 8% decrease in

right hippocampal volume and 10% decrease in left hippocampal volume in MDD and right hippocampal

volume was negatively correlated with the number of depressive episodes.59 In a study comparing first

episode, never-treated depressed patients to multiple episode depressed patients, researchers observed

both patient types to have cognitive deficits in recollection memory, which they attributed to hippocampal

dysfunction.60 Furthermore, only patients with recurrent episodes had decreased hippocampal volumes

and statistical analysis demonstrated that length of illness is significantly correlated with hippocampal

volume reduction.57, 60 With respect to PSD, existing atrophy in the context of an acute infarct may

perhaps exacerbate or instigate a neuroprogressive pathway. The results of Starkstein et al. indicate that

this may be the case, with their finding that the lateral and third ventricles in depressed patients were

significantly larger than controls.405 Although, Vataja et al. did not find any link between depression and

atrophy analyzed by cortical (frontal, parietal, and occipital lobes) and central (temporal, frontal, and

occipital horns) and medial temporal lobe (entorhinal cortex and hippocampus) regions.40, 45 More

36

recently, Tang et al. investigated the relationship between cortical atrophy and PSD, and a multivariate

analysis determined that severe frontal lobe atrophy independently predicted the presence of depression.39

Another study by Fu et al. assessing the relationship between depressive symptoms, WMC and atrophy,

observed that left inferior frontal gyrus atrophy significanty predicted depressive symptom severity.49

These positive findings suggest that location of atrophy, as opposed to lesions may be more highly

associated with PSD. The results of these studies further emphasize the importance of considering

multiple neuro-radiological measurements in the effort to determine the mechanisms of PSD, as a

comprehensive approach to studying these domains implies that they are not mutually exclusive.

In summary, this study aims to assess the relationship between IDO activation, as measured by the K/T

ratio, and depressive symptom severity. Alterations in the levels of KYN and its metabolites have been

associated with major depression88-92, 305 and the K/T ratio has been demonstrated to correlate with stroke

severity, lesion volume, and poor outcomes. 326 It is presumed that acute elevations in the K/T ratio post-

stroke will influence mood initially through a reduction in available TRP for serotonin synthesis, and

more gradually through the production of excitotoxic and neurodegenerative KYN metabolites. In

addition, previous reports have demonstrated pro-inflammatory cytokine elevations in the context of

MDD 215-225, 228, 230-233 as well as PSD. 36, 243 However, researchers have failed to find a relationship

between PSD and post-stroke pro-inflammatory cytokine elevations.244, 254255 As outlined in section 1.4.6,

the balance between pro- and anti-inflammatory cytokines may be a better marker for depression than

either type of these cytokines in isolation, as evidence by many positive findings in both MDD31-35 and

PSD. 36, 37 This study will attempt to clarify the role between changes in pro- and anti-inflammatory

cytokine production in relation to post-stroke depressive symptom severity, as an upstream link in the

pathway of IDO-mediated depression. Lastly, few studies have assessed the relative effects of acute

infarcts, WMC, and atrophy together as they relate to PSD.39, 40, 45, 48 Those have produced mixed results,

with one study finding WMC48 and atrophy39 to be significantly associated with PSD. Thus, failure to

account for pre-existing or additional brain pathology may explain the lack of consistency in previous

findings. This thesis aims to explore a comprehensive approach in assessing the association between

neuroanatomy and post-stroke depressive symptoms, and thus will assess the relationship between

depressive symptom severity and each neuroimaging measurement. In particular, this study will measure

whole brain atrophy, hippocampal atrophy, lesion volume, and WMC. Lastly, the combined effect of an

elevated K/T ratio and neurodegeneration may be more relevant in predicting the presence of PSD than

either variable alone and therefore, this study will also attempt to explore this notion as its final

exploratory aim.

Section 2: Materials and Methods

*Methods were conducted, as has been done by our Group in the past406-408

2.1 Study Design

This clinical study was cross-sectional and correlational in nature. It examined the

neuroanatomical and inflammatory correlates of severity of post-stroke depressive symptoms.

Accordingly, ischemic stroke patients were recruited, all having met World Health Organization

MONICA Project (1988) and National Institute of Neurological Disorders and Stroke (WHO-NINDS)409

criteria for stroke with visual CT-based evidence of an acute brain infarct. All patients were recruited

from either an in-patient unit or the out-patient Stroke Prevention Clinic (SPC) at Sunnybrook Health

Sciences Centre, which is an acute care hospital with regionalized stroke service. They were assessed in

terms of mood, cognition, and stroke severity. Demographic information and medical history were

obtained through hospital chart reviews and patient interviews. This study has received Research Ethics

Board approval by Sunnybrook Health Sciences Centre (Appendix 1). All patients provided written,

informed consent and each received a photocopy of the signed consent form (Appendix 2).

37

38

2.2 Inclusion Criteria and Exclusion Criteria

Inclusion criteria:

• Age: ≥ 18 years old • Gender: male or female • Language: speaks and understands English • Clinical diagnosis of stroke according to the WHO-NINDS409 criteria • Acute cerebral infarction with visual CT-based evidence

Exclusion Criteria:

• Subarachnoid or intracranial hemorrhage • Severe aphasia or dysarthria that would preclude neuropsychiatric testing • Imminently suicidal or, in the opinion of the affiliated clinician, has inadequate family monitoring for

suicidality • Impaired level of consciousness that would preclude from neuropsychiatric testing • Significant acute medical illness likely to affect neuropsychiatric symptoms, including:

o Infection o Drug or alcohol abuse o Uncontrolled diabetes o Uncontrolled anemia

o Severely disturbed liver, kidney, or lung function

o Untreated hypothyroidism

• Significant acute neurological illness likely to affect neuropsychiatric symptoms, including: o Decreased Level of

Consciousness (LOC) o Active brain tumour o Parkinson’s disease o Huntington’s disease o Multiple sclerosis

o Binswanger’s disease o Hydrocephalus o Subdural hematoma o Progressive supranuclear palsy o Severe aphasia

• Presence of a previous Axis I psychiatric diagnosis, such as mood disorder, an anxiety disorder or

schizophrenia o Major Depressive Disorder (history

of) o Schizophrenia o Bipolar disorder

o Dementia

• Concomitant use of psychotropic medication, except short-acting benzodiazepines for sedation (e.g. lorazepam)

• Current use of antidepressant pharmacotherapy

39 2.3 Demographics, Medical History and Clinical Assessments

2.3.1 Demographic and Medical History

The method for obtaining demographic information and clinical history was either through chart

review or patient interviews. Age, gender, time since stroke, marital status, current and expected living

situation, current and expected employment status and educational history were documented. Past and

current medical histories included thorough accounts of all surgical history, cardiovascular risk factors,

concomitant medications, and psychiatric history.

2.3.2 Clinical Assessments

2.3.2.1 Stroke Severity Scale

The National Institutes of Health Stroke Scale (NIHSS)410 was used to measure stroke severity,

which was carried out by clinicians at the time of patient admission, attained through chart review, or

retrieved from the chart utilizing a standardized method as has been done before.411 Stroke severity is an

established risk factor for depression and an important outcome associated with physical disability.129 The

NIHSS was used by the National Institute of Neurological and Communicative Disorders and Stroke

(NINCDS) Stroke Data Bank.412 Stroke severity, measured using the NIHSS, was demonstrated to

significantly correlate with stroke outcomes.413 In terms of functional impairment, as assessed by the Barthel

Index and the modified Rankin Scale, the NIHSS was found to significantly associate with this negative

post-stroke outcome.414

2.3.2.2 Depressive Symptom Severity Scale

Depressive symptoms were rated using the Centre for Epidemiological Studies Depression (CES-D)

scale,415 which is a patient self-report scale. It was also utilized by the NINCDS Stroke Data Bank,412 and it

was formerly validated, using the structured clinical interview in Stroke Data Bank patients, to be reliable

with previously accepted diagnostic criteria. When the CES-D was evaluated against a number of different

40 depression rating scales in aging stroke patients, it was established to retain good external and concurrent

validity.416 Previous studies have utilized this scale in the context of PSD.144, 417 Using a cut-off of ≥16 in the

post-stroke population was found to be extremely indicative of depression (sensitivity 86%, specificity 90%,

positive predictive value 80%).418 In older, non-stroke populations, this same cut-off score was sensitive

enough to detect physical, cognitive, and psychosocial impairments with had clinical relevance,419, 420 as

well as relevant and significant depressive symptoms.419, 420

2.3.2.3 Global Cognition Scale

Upon the patient giving written informed consent, the Mini-Mental State Examination (MMSE)421

was carried out. This scale was chosen to screen global cognition due to the fact that in acute care settings it

is extensively utilized, it is also a means of rapid assessment of cognition and has been validated in this

setting.422 The MMSE has previously been demonstrated to associate well with the CAMCOG (cognitive

and self-contained part of the Cambridge Examination for Mental Disorders of the Elderly) in a stroke

rehabilitation setting.423 Due to the fact that this study is exploratory and the possibility of low admissibility

of a more lengthy assessment in an acute stroke setting, this concise, although validated and clinically

relevant cognitive scale was chosen.

2.4 Blood Sampling

For inpatients, blood withdrawal was carried out by IV technicians upon their scheduled clinical

blood draws, and outpatients met with the SPC study coordinator for their blood to be drawn. Because of the

variability in blood collection time, it was not possible to have an absolute standardized collection time,

however every effort was made to conduct blood draws in the morning (approximately 7:00 a.m.). Serum

concentrations of various pro- and anti-inflammatory cytokines (TNF-α, IL-6, IL-10, IFN-γ, IL-1β and IL-

18), and kynurenine, and plasma concentrations of tryptophan and LNAA were assessed. Blood samples

utilized for cytokine and kynurenine analysis were collected in SST gel vacutainer tubes, and TRP and

LNAA were collected in EDTA vacutainer tubes.

41

After a maximum of 2 hours post-collection, the samples of blood were centrifuged at 1000g for 10

minutes. Plasma or serum sample were relocated into labeled cryovials following centrifugation and kept at

-70°C until analyzed. An extra ultrafiltration step was necessary for TRP samples to divide the free form

from protein-bound form. Here, 1.0 ccof plasma was moved into an ultra filtration device (centrifree 4104®,

Millipore) and centrifuged at 1000g for an extra 15 minutes. The device maintained all protein-bound TRP

while enabling the smaller free TRP to move across the 30,000 kDa membrane into ultrafiltrate reservoir

cups. The cups holding free TRP were then capped and held with the other samples at -70°C until ready for

batched analysis.

2.5 Serum and Plasma Analyses

2.5.1 Kynurenine Assay

KYN concentrations were calculated by the Pharmacy Quality Control and Research Group at

Sunnybrook Health Sciences Centre with high-performance liquid chromatography (HPLC) at UV detection

of 258 nm, as described elsewhere.424, 425 Personal communication with Scott Walker was carried out for a

comprehensive protocol of the KYN assay. Briefly, an equal volume of 3% perchloric acid was used for

protein precipitation. After centrifugation for 10 minutes, a 0.1 mL aliquot of the supernate was injected

directly into the high liquid chromatograph system using an autoinjector (715 Ultra WISP, Waters Canada).

The mobile phase consisted of 9% acetonitrile in 0.05 M potassium phosphate mono basic, pumped through

a reverse phase 5 μm ODS column, 250 mm × 4.6 mm (Symmetry; Waters Corporation, Milford,

Massachusetts, USA). KYN was detected using an ultraviolet detector (UV 6000; ThermoSeperation

Systems, San Jose CA) at a wavelength of 258nm. Chromatograms were recorded on a computer using the

ChromQuest Software (ThermoQuest, San Jose CA).

The standard curve was obtained by making 9 standards from a common serum stock sample.

Additional KYN was added to roughly 10 mL of serum, to yield standards with 0, 0.125, 0.250, 0.3125,

0.50, 0.750, 1.0, 1.5, and 2 mg/L of extra KYN added. Examination of these standards in duplicate produced

a linear relationship between the KYN peak area and concentration with a slope and intercept (because of

42 the endogenous kynurenine in the common serum stock). The obtain unknown concentrations of KYN in

plasma the KYN peak area in the unknown sample was divided by the slope. Replicate error, as assessed by

the coefficient of variation (CV (%) = 100 x standard deviation/mean), based on duplicate analysis of

standards between 0 mg/L of kynurenine added to 2 mg/L added ranged from 1.54% to 0.5% with

deviations from the known or expected concentrations averaging less than 7%.

2.5.2 Cytokine Assay

Cytokine measurements of TNF-α, IL-6, IL-10, IFN-γ, IL-1β and IL-18 were carried out by Dr.

Angela Panoskaltsis-Mortari who runs the Cytokine Reference Laboratory at the University of Minnesota. A

multiplex immunobead-based assay using the Human Fluorokine Multianalyte Profiling (MAP) Base Kit A

(R&D Systems, Minneapolis, MN) was utilized to assay all cytokines. All cytokine samples from this study

were assessed for sensitivity, precision, recovery, sample linearity, and specificity by the company panel.

The company website gives an overview of the methods:

“Detection is achieved through bead-based antibody-antigen sandwhich method.

Briefly, samples are incubated with colour-coded beads that are pre-coated with

analyte-specific capture anibodies for the molecule of interest. Expression levels are

determined following incubation with a biotinylated detection antibody and

streptavidin-conjugated phycoerythrin (PE). Using a Luminex® analyzer,

independent lasers determine the colour of each bead and the magnitude of the PE-

derived signal, which is directly proportional to the levels of bound analyte.”

Assay sensitivities were 0.2 pg/ml for IL-10, 1.0 pg/ml for IFN-Ɣ, 0.09 pg/ml for TNF-α, 0.057

pg/ml for IL-1β, 0.1 pg/ml for IL-6, and 12.5 pg/ml for IL-18, which are comparable to the sensitivities of

used by other studies observing stroke and depressed populations by means of standard ELISA assays.103, 426,

427 However, researchers have noted that multiplex technology can distinguish a superior range of values

compared to ELISA or Western blotting.428

43 2.5.3 Tryptophan and LNAA Assay

Dr. Simon Young, who resides at the Department of Psychiatry at McGill University, kindly

assessed concentrations of TRP and LNAA, as explained elsewhere in detail by Anderson et al.424 Through

personal communication with Simon Young, a thorough protocol of the TRP and LNAA assays was

obtained and is presented here:

“Plasma was deproteinized by adding one part of 1 M perchloric acid to three parts of

plasma, mixing and centrifuging. Tryptophan in plasma was measured by high-

performance liquid chromatography (HPLC) on a Waters Bondpack C18 reverse phase

column (Phenomenex, Torrance, CA) with fluorometric detection (Anderson and Purdy,

1979). The other large neutral amino acids (LNAA) (tyrosine, phenylalanine, leucine,

isoleucine, valine) were measured on a Beckman System Gold amino acid analyzer using

precolumn derivatization with o-phthalaldehyde and gradient reverse phase HPLC with

fluorometric detection. Aminoadipic acid was used as an internal standard.”

2.6 Neuroimaging

As a component of routine hospital procedures, patients had two to three CT scans carried out. This

study assessed the final CT scan so as to account for the complete infarct and analyze all ‘tissue at risk’. The

final scan needed to be performed at a minimum of 24 hours after the stroke. All CT scans were collected

without a contrast agent on a General Electric LightSpeed VCT series scanner (General Electric Healthcare,

Waukesha, WI). The CT images were skull stripped with the Image Edit function of Analyze 11.0 (Mayo

Clinic, Rochester, MN). The de-skulled brains were transformed using the segment function of Statistical

Parametric Mapping 5 (APM5; Wellcome Department of Imaging Neuroscience, University College

London, London, England). This function was employed to make transformation matrices that outlined the

process required to align de-skulled brains with a standard template previously aligned with Montreal

Neurological Institute (MNI) space. The Normalise: Write function of SPM5 utilized the transformation

matrices to normalize the de-skulled brains to this MNI space template. The normalized de-skulled brains

44 were assessed to make sure that they were aligned well. The scans that did not align properly with MNI

space were not used and therefore these patients were removed from the study.

2.6.1 Lesion Volume and location Measurement

Areas of infracted tissue were traced on these images using Analyze version 11.0 (Mayo Clinic,

Rochester, MN) [Figure 2]. Lesion area was obtained by tracing the infarct slice-by-slice on the CT scan,

then the area of the lesion on each slice was multiplied by slice thickness (1mm3) and summed to get the

total lesion volume. This method has been demonstrated to have high reproducibility in acute ischemic

stroke.429

45

Figure 2. Axial slice showing a patient with a large right hemispheric infarct (radiological perspective:

flipped)

46 Lesion location was determined using the methods constructed by Robinson et al.138 to test their

lesion location hypothesis of PSD.406 This included determining an anterior-posterior (AP) ratio by assessing

the distance of the most anterior border of the lesion to the frontal pole and then dividing this value by the

total length of the AP ratio in the same slice. Depending on the AP ratio, lesions were grouped into four

distinct types: ‘Anterior’ lesions are located <40% from the frontal pole, ‘Posterior’ lesions are situated

>60% from the frontal pole, ‘Extending’ lesions are <40% and >60% from the frontal pole, and

‘Intermediate’ lesions are >40% but <60% from the frontal pole.406

2.6.2 White Matter Change Analysis

WMCs were carried out with a CT and MRI-validated visual rating scale, the Age-Related White

Matter Changes (ARWMC) scale [Figure 3].430 This scale has many advantages to the extensively used

Fazekas scale,431 including a more in depth assessment of brain regions. The ARWMC scale goes beyond

the Fazekas scale in terms of this because it analyzes the brain as 10 discrete areas, as opposed to ‘deep’ and

‘periventricular’. The ARWMC regions are as follows: bilateral frontal lobes, bilateral temporal lobes,

bilateral parieto-occipital lobes, bilateral basal ganglia, and bilateral infratentorial areas. The scoring system

is as follows: 0 (no lesions), 1 (focal lesion), 2 (beginning of confluence lesions), 3 (diffuse involvement of

the entire region). However, the basal ganglia are scored by to another system: 0 (no lesions), 1 (1 focal

lesion), 2 (>1 focal lesion), 3 (confluent lesions).

47

Figure 3. Axial slice showing a patient with grade 2 white matter changes in bilateral frontal and parieto-

occipital regions.

2.6.3 Measurement of Hippocampal Thickness

Hippocampal thickness was measured using previously described methods by Gao et al. 2003.432

Briefly, as seen in Figure 4, the width of the medial temporal lobe (MTL) was used as a marker for

hippocampal thickness. Using Analyze version 11.0 (Mayo Clinic, Rochester, MN), the MTL was oriented

at an angle of approximately 30° caudal to the Anterior Commissure – Posterior Commissure (AC-PC)

48 plane. The intercollicular sulcus (ICS) was used as a landmark to obtain the best longitudinal view of the

hippocampus and parahippocampal gyrus, and the MTL width was measured midway and at its thinnest

portion between the anterior–posterior borders of the midbrain.432 Since the scans were co-registered with a

template that was calibrated to be 24% larger than normal, the measurement obtained for hippocampal

thickness was then multiplied by 0.76 to obtain the true thickness of the hippocampus. Measurements for

right and left hippocampal thickness were averaged to create one final value per patient.

Figure 4. Axial slice showing the measurement of left hippocampal width (radiological perspective) by

aligning the hippocampus 30° caudal to the AC-PC plane.

49 2.6.4 Whole Brain Atrophy Measurement

Whole brain atrophy as assessed by total ventricular volume measurement. The left and right lateral

ventricles as well as the third ventricle were manually traced using Analyze version 11.0 (Mayo Clinic,

Rochester, MN) [Figure 5]. The area of the ventricles on each slice was then multiplied by slice thickness

(1mm3) and summed to get the total volume of left and right lateral ventricles as well as the third ventricle.

Total ventricular volume was then normalized by dividing this value by the total intracranial volume (TIV).

This produces a ratio value of total ventricular volume: total intracranial volume in order to account for

varying brain volumes across subjects.

50

Figure 5. Axial slice showing a manual tracing of the left and right lateral ventricle, as well as the third

ventricle.

Aside from lesion location analysis, which was carried out by Phillip Francis,406 all neuroimaging

conducted was overseen and verified by an experienced neuroradiologist, Dr. Fu-Qiang Gao, to ensure

51 preservation of correct methodology. Briefly, training was done by Dr. Gao using a training set of 20

random images. Once those were found to be in agreement with Dr Gao’s tracings, scores, or

measurements, the remainder were done initially by the candidate and then checked by Dr. Gao.

2.7 Planned Analyses

2.7.1 Initial Descriptive Statistics

The Statistical Package for the Social Sciences (SPSS), version 20.0 was used for all statistical

analyses. Initially, all demographic variables were assessed in terms of their correlation with the CES-D.

Any variables found to be significantly associated with depression scores were then used as covariates in

latter analyses. These variable comprised of the following: age (years), gender (F/M), time between stroke

and assessment (days), vascular risk factors; cigarette smoking (Y/N), hypertension (Y/N), obesity (Y/N),

elevated cholesterol (Y/N), and diabetes (Y/N), education level (ordinal), living situation (alone or with

others), marital status (ordinal), employment status (ordinal), lesion location (ordinal), lesion laterality

(right/left), stroke severity (NIHSS score), and global cognition (MMSE) score. All of these variables were

analyzed for a significant correlation with depression scores. Continuous variables were analyzed with

spearman’s rank correlation coefficient (ρ) (Non-parametric). Pearson’s product-moment correlation

coefficient (r), was not employed because CES-D patient scores failed the Kolmogorov-Smirnov test for

normality (p>0.05). Gender, living situation, cigarette smoking, hypertension, diabetes, elevated cholesterol,

and obesity were examined using an independent-samples Student’s t-test, which compared the mean CES-

D score between the sexes, living situations, and individual vascular risk factor groups. The education level,

employment status, and marital status were investigated using a one-way analysis of variance (ANOVA),

which compared the mean CES-D score between the multiple groups.

For serum concentrations of IFN-Ɣ, TNF-α, IL-6, IL-10, IL-18, and IL-1β that were below the limit

of detectability, missing values were imputed at the above mentioned values, however there were no

undetectable sample for IL-6 or IL-18. Because of the fact that undetectable samples were imputated at the

lower limit of detectability, n the absolute values of kurtosis and skewness were more than 2 standard

52 deviation (SD) of the error. It then became imperative to log-transform the cytokine dataset in order to retain

a normally distributed dataset, as has been done by our group323, 433 and many other researchers have dealt

with this issue in similar ways243, 256, 434 in the past. Furthermore, this method should not interfere with our

objective of examining the role of increased proinflammatory cytokine concentrations in PSD, except in the

case of IL-10, an anti-inflammatory cytokine, where the goal was to detect decreased concentrations in PSD

patients. This drawback will be discussed in further detail within the discussion. For the calculation of

immunologic ratios, they were determined with raw values and log-transformed afterwards. Possible

intercorrelations between individual cytokines were also assessed.

2.7.2 Primary Analyses

Hypothesis 1: It is hypothesized that serum levels of the K/T ratio, as a measure of IDO activity, will

positively correlate with depressive symptom severity post-ischemic stroke.

To test the primary hypothesis, a simple linear regression was performed to assess the relationship between

CES-D total score and the K/T ratio. All important covariates from the initial descriptive analysis were

included in the model. Furthermore, K/T ratio was recently found to be significantly elevated in

somatization compared to depression and also associated with somatic symptom severity.57, 305, 313 In a study

involving IL-2 and/or interferon-α cancer therapy, symptoms characterized by the researchers as

‘neurovegetative’, including lack of sleep, poor appetite, and decreased energy, were significantly more

pronounced in patients with mild depressive symptoms than patients within the ‘‘absent’’ depressive

symptom range, as measured by the MADRS.435 As a result, the relationship between K/T ratio and cES-D

may not be linear. Therefore, an ordinal regression was also performed to determine if the K/T ratio

significantly predict CES-D scores as tertiles. By dividing our study population into tertiles based on CES-D

total scores, this will reveal if the K/T ratio is associated with none, mild, or moderate depressive symptoms.

For all tests, multicollinearity was assessed; tolerance statistics were calculated and examined. However for

an ordinal regression, the amount of covariates allowed in the model is determined by taking the smallest (n)

outcome category and dividing it by 10. Out of the three CES-D tertiles, the smallest outcome category has

53 an n value of 26, thus, the model can only hold 2 covariates. Therefore, a decision was made to include

NIHSS scores as the sole covariate for all subsequent ordinal regressions, since stroke severity has been

consistently identified as an important independent predictor of PSD in the past.16, 110

2.7.3 Secondary Analyses

Hypothesis 2: Elevated pro-inflammatory cytokines IFN-Ɣ, TNF-α, IL-6, IL-18, and IL-1β and a reduction

in the anti-inflammatory cytokine IL-10, as evidenced by an elevated immunologic ratio between these pro-

inflammatory cytokines and IL-10, will correlate with depressive symptom severity post-ischemic stroke.

To test my secondary hypothesis, separate simple linear regressions were performed to assess the

relationship between CES-D total score and all individual cytokines as well as their immunologic ratios with

IL-10. Since patients with depressive symptoms not meeting DSM-IV outlines for MDD have significant

elevations in inflammatory molecules,214, 236 a set of ordinal regressions were conducted to determine which

cytokines or cytokine ratios may significantly predict CES-D scores as an ordinal variable. By dividing our

study population into tertiles based on CES-D total scores, this will reveal if elevated cytokines or an

elevated pro- to anti-inflammatory cytokine ratio is associated with none, mild, or moderate depressive

symptoms. For all tests, significant clinical characteristics were included as covariates and tolerance

statistics were calculated and examined.

2.7.4 Exploratory Analyses

Lesion volume,38-43 WMC, 38-48 and atrophy 39, 40, 45, 49 have been studied in the context of PSD with

mixed findings, however MDD has been consistently linked with neurodegeneration and neuroprogression.

50-60 A comprehensive approach to investigating atrophy, lesion volume and WMC may help to clarify their

role in the context of PSD, as these measurements are not mutually exclusive. Thus, I sought to explore the

relationship between neuroimaging findings and post-stroke depressive symptom severity. Specifically,

hippocampal thickness, whole brain atrophy, WMC and lesion volume may be important independent

correlates of depressive symptom severity post-ischemic stroke. In order to assess which neuroimaging

findings were significantly associated with CES-D score, average hippocampal thickness, whole brain

54 atrophy, the global ARWMC score for WMC, and lesion volume were entered into a backward stepwise

linear regression and CES-D score was entered as the dependent variable. CES-D scores were compared

across acute lesion locations to determine if they were significantly different across groups, if so, these

locations would be included as a covariate in this analysis. The removal criterion for backwards regression

model was set to (P≤0.1). A separate backward stepwise linear regression was conducted to determine the

significance of white matter lesion location in relation to PSD. As determined by a paired samples t-test,

there was no significant difference between right and left ARWMC scores, thus, each bilateral region score

(frontal lobes, temporal lobes, parieto-occipital lobes, bilateral basal ganglia, and infratentorial) were

summed to create 5 ARWMC scores as opposed to 10 bilateral scores. These 5 regional scores were entered

into the regression model with CES-D score as the dependent variable. For all tests, significant clinical

characteristics were included as covariates and tolerance statistics were calculated and examined.

The fact that (1) depression has been associated with neurodegeneration,50-60 (2) KYN metabolites

are elevated in many neurodegenerative diseases and stroke,61-66 and (3) have also been implicated in the

process of neuroprogression,21, 50, 67-77 suggests that an elevated K/T ratio in the presence of

neurodegeneration may be more relevant in contributing to the development of PSD, than either variable

alone. In order to explore the possible joint contributions of an elevated K/T ratio and neurological

pathology in post-stroke depressive symptom severity, a simple linear regression was conducted taking into

account any significant neuroimaging correlates from the previous analysis, in conjunction with the K/T

ratio. A custom model was created taking into consideration the K/T ratio and neuroimaging correlates

separately, alongside interaction terms between these independent variables. Significant clinical

characteristics were also included and tolerance statistics were calculated and examined.

2.7.5 Sample Size Calculation

The K/T ratio has never been assessed in relation to depressive symptom severity post-ischemic

stroke. Therefore, the effect size (f 2) was approximated according to an earlier study published by our group

involving CAD patients assessed for the association between the K/T ratio and depression.92 For a multiple

linear regression the coefficient of determination (R2), based on the CAD study, was used to calculate f 2,

55

where f 2 = 𝑅2

1−𝑅2 . Subsequently, f 2 was used to determine the required sample size (n), based on the α level

(probability of type I error) set to 0.05 (α = 0.05), and the β level (probability of type II error) set to 0.2

(β=0.2). Thus, it was determined that with α = 0.05, an estimated f 2 = 0.1433, and 3 independent predictors

in the multiple linear regression model, the required sample size is 80, to obtain an observed power of 0.80.

2.8 Controlling for Confounding Variables

Many clinical and vascular risk factors have been found to be associated with stroke and depression.

Therefore, confounding variables were tested for their correlation with depressive symptom severity and any

confounders determined to be significantly associated with depressive symptom severity were included in

subsequent a priori and exploratory analyses as important covariates. It was determined that NIHSS scores

and age were significantly correlated with depression scores. In one meta-analysis stroke severity was found

to be significantly associated with PSD across various studies.16 Another recent meta-analysis by Ayerbe et

al. also found stroke severity to be an important predictor of PSD.110 Furthermore, pro- and anti-

inflammatory cytokines,228, 436, 437 as well as the K/T ratio438, 439 have been found to positively correlate with

age. Patients with a history of depression and those who are currently using antidepressant

pharmacotherapy were excluded from this study since antidepressant use has been demonstrated to alter the

ratios of pro- to anti-inflammatory cytokines, increasing the production of IL-10, 32-35 and a history

ofdepression is an established risk factor for MDD, though the mechanisms behind its development may

vary from the etiology of PSD.

*Methods were conducted, as has been done by our Group in the past406-408

Section 3: Results

3.1 Demographic and Clinical Characteristics

Eighty two patients were recruited to participate in this study. Table 1 displays the results of

independent samples t-tests, ANOVAs and Spearman’s correlations between CES-D scores and clinical

and demographic characteristics.

Table 1. Clinical demographic characteristics (n=82)

Mean ± SD or % Range *R , t, or F Value ap Value

Demographics

Age 71.9 ± 14.2 years 33-98 0.254 0.022*

Gender (% male) 53.1% -0.558 0.578

Living situation (% alone) 32.1% 1.100 0.275

Employment Status (% retired) 73.8% 1.409 0.247

Marital Status (% married) 51.9% -0.271 0.787

Education level (%>high school)

50.6% 1.528 0.192

Time since stroke, days 25 ± 35 4-155 0.105 0.353

Clinical Assessments

CESD 12.6 ± 10.8 0-47 ------------------ ------------

NIHSS 4.6±4.7 0-19 0.231 0.042*

MMSE 25.8 ± 4.2 11-30 0.034 0.762

56

57 Mean ± SD or % Range *R , t, or F Value ap Value

Plasma and Serum Markers

KYN (μmol/L) 2.98 ±1.57 0.59-8.46 -0.075 0.503

Mean ± SD or % Range *R , t, or F Value ap Value

TRP (μg/mL) 2.87±0.68 2.08-5.05 0.012 0.935

LNAA (μg/mL) 564.55±145.08 273.68-977.14 -0.145 0.314

*r , t, or F values and their corresponding ap-values reflect the results of bivariate Spearman’s correlations with CES-D scores for continuous variables, independent samples t-test between CES-D score and categorical variables, and ANOVAs between CES-D scores and ordinal variables

These descriptive statistics revealed that NIHSS score (rho=0.231, p=0.042) and age

(rho=0.254, p=0.022) were significantly associated with CES-D score, thus, these were included as

covariates in all subsequent analyses. As displayed in Table 2, there were no significant differences in

depression scores between any of the vascular risk factors (VRFs) assessed. However there a trend (t= -

1.680, p=0.097) for depression scores to be higher in the hypertensive group. Of note, hypertension

(81.5%) and hyperlipidemia (66.3%) were especially common in this patient population, and 93.8% of

this patient sample had at least one VRF.

Table 2. Vascular risk factors (VRFs) (n=82)

Vascular Risk Factors Mean ± SD or % Range * t Value ap Value

Hypertension 81.5% -1.680 0.097

Diabetes 20.3% 0.722 0.473

Hyperlipidemia 66.3% -0.212 0.833

Obesity (BMI>=30) 20.0% 0.143 0.887

Smoking 21.5% 1.564 0.122

Total number of VRFs 2.09±1.109 0-5 1.076 0.380

Abbreviations: BMI, body mass index. *t values and their corresponding ap-values reflect the results of independent samples t-test between CES-D score and categorical variables.

58

As shown in Table 3, there were no significant differences in depression scores between lesion

locations (F3,41=0.622, p=0.605) or lesion laterality (F1,56=0.371, p=0.545). An equal proportion

(28.6%) of patients has stroke in the anterior and posterior regions, while 24.5% had extending lesions

and 18.4% had intermediate lesions. Approximately two thirds (64.7%) had right-sided lesions, while

35.3% had left-sided lesions.

Table 3. Lesion location (n=45) and laterality (n=58)

% * F Value ap Value

Lesion Location

Anterior 28.6%

0.622

0.605 Intermediate 18.4%

Posterior 28.6%

Extending 24.5%

Laterality

Right 64.7%

0.371

0.545 Left 35.3%

*F values and their corresponding ap-values reflect the results of ANOVAs between CES-D scores and ordinal variables.

59 As displayed in Table 4, independent samples t-tests revealed no significant differences in mean CES-D scores between groups of patients using and not using certain medications.

Table 4. Concomitant medication use

Medication % of patients using medication *t ap-Value

Aspirin 62.5% -0.021 0.98

Anti-hypertensivesǂ 35.5% -1.65 0.103

Insulin 16.3% 1.69 0.096

Benzodiazepines 17.5% -0.059 0.95

ǂAnti-hypertensives include: Ca+2Channel Blockers, β-blockers, and diuretics. *t values and their corresponding ap-values reflect the results of independent samples t-test between CES-D score and categorical variables.

3.2 Assay Results

Plasma samples from 82 patients were gathered and sent out for testing of IL-6, IL-18, TNF-α,

IFN-Ɣ, IL-1β, and IL-10. There were 82 samples sufficient to analyze for IL-6, IL-10, and IFN-Ɣ, 62

samples for IL-18, 59 samples for TNF-α, and 54 samples for IL-1β. Samples of cytokines that were

below the assay’s limit of detectability were imputed at the lower limit of detectability for each

cytokine. The aforementioned sensitivities were 0.2 pg/ml for IL-10, 1.0 pg/ml for IFN-Ɣ, 0.09 pg/ml

for TNF-α, 0.057 pg/ml for IL-1β, 0.1 pg/ml for IL-6, and 12.5 pg/ml for IL-18, however there were no

undetectable samples for IL-6 or IL-18. Imputed cytokine levels were included in all statistical

analyses.

3.3 Primary Analyses

A simple linear regression was conducted to determine the relationship between the K/T ratio

and depression scores as a continuous variable. After controlling for age and NIHSS scores, the overall

model was not significant (R2=0.069, F3,73=1.805, p=0.15), with neither the K/T ratio (β= -0.105,

p=0.37), nor the NIHSS score (β=0.190, p=0.10, nor age (β=0.143, p=0.21) significant independent

predictors of CES-D score. A scatter plot of the relationship between serum K/T and CES-D scores is

shown in Figure 6.

60 Figure 6. Correlation between serum K/T and CES-D scores

Next, an ordinal regression was performed to determine if the K/T ratio significantly predicts

CES-D tertiles. The lowest CES-D tertile scores ranged from 0-6, the middle tertile scores ranged from

7-15, and the highest tertile scores ranged from 16-47. For an ordinal regression, the number of

covariates allowed in the model is determined by taking the smallest (n) outcome category and dividing

it by 10. Out of the three CES-D tertiles, the smallest outcome category has an n value of 26, thus, the

model can only consider 2 covariates. The NIHSS scores was chosen as the sole covariate for all

subsequent ordinal regressions, since stroke severity has been consistently identified as an important

independent predictor of PSD in the past. The ordinal regression model was conducted and

demonstrated that neither the K/T ratio (odds ratio [OR], 0.996; 95% confidence interval [CI], 0.985-

1.007, p=0.51) or NIHSS score (OR, 1.077; 95% CI, 0.987-1.177, p=0.095) were significant

independent predictors of CES-D tertile. Figure 7 displays a bar graph of mean serum K/T ratio across

CES-D tertiles.

61 Figure 7. Bar graph of mean serum K/T levels across stroke patients in low, middle, and high CES-D

tertiles

3.4 Secondary Analyses

Simple linear regressions were used to determine the relationship between each cytokine and

depression scores, as well as the IL-10 immunologic ratio and depression scores. After controlling for

age and NIHSS scores, none of the overall models were significant, nor were the cytokines or

immunologic ratios significant predictors. The results of each overall linear regression model and

individual cytokines as predictors are summarized in Table 5. A sample (IFN-Ɣ and IFN-Ɣ/IL-10) of

scatter plots showing the relationship between plasma cytokines and CES-D scores is shown in Figure

8 and 9.

62 Table 5. Final linear regression models for each cytokine as it predicts CES-D score

Cytokine Model R2 *F Df ǂβ ap-Value

IL-6 Overall 0.063 1.61 3,72 0.20

IL-6 -0.086 0.49

IL-18 Overall 0.044 0.826 3,54 0.49

IL-18 0.122 0.38

TNF-α Overall 0.010 0.17 3,52 0.92

TNF-α 0.004 0.98

IFN-Ɣ Overall 0.072 1.85 3,71 0.15

IFN-Ɣ 0.142 0.22

IL-1β Overall 0.007 0.12 3,49 0.95

IL-1β -0.043 0.76

IL-10 Overall 0.058 1.47 3,72 0.23

IL-10 -0.038 0.74

IL-6/IL-10 Overall 0.056 1.44 3,72 0.24

IL-6/IL-10 -0.006 0.96

IL-18/IL-10

Overall 0.043 0.80 3,54 0.50

IL-18/IL-10 0.118 0.40

TNF-α/IL-10 Overall 0.043 0.77 3,52 0.52

TNF-α/IL-10

0.188 0.19

IFN-Ɣ/IL-10 Overall 0.080 2.05 3,71 0.11

IFN-Ɣ/IL-10 0.167 0.15

IL-1β/IL-10 Overall 0.038 0.65 3,49 0.59

IL-1β/IL-10 0.185 0.20

*F values and their corresponding ap-values reflect the results of the overall linear regression model for each cytokine, ǂβ values and their corresponding ap-values reflect the results of the individual cytokine as a predictor in the regression

63 Figure 8. Correlation between plasma IFN-Ɣ and CES-D scores

Figure 9. Correlation between plasma IFN-Ɣ/IL-10 and CES-D scores

Next, a set of ordinal regression were performed to determine which cytokines or cytokine

ratios may significantly predict CES-D tertiles. Results of ordinal regressions for each cytokine and

cytokine ratio are displayed in Table 6. Briefly, the regression models revealed that most of the

cytokines or cytokine ratios were not significant independent predictors of CES-D scores as tertiles.

The only cytokine ratio found to be a significant predictor of mild depressive symptoms, i.e. the middle

64 CES-D tertile, was IFN-Ɣ/IL-10 (OR, 2.17; 95% CI, 1.02-4.64, p=0.045). There was a strong trend for

TNF-α/IL-10 (OR, 1.88; 95% CI, 0.98-3.60, p=0.056) to be a significant predictor of the highest CES-D

tertiles. Figure 10 presents a bar graph displaying plasma IFN-Ɣ/IL-10 across CES-D tertiles.

Table 6. Final ordinal regression model for each cytokine and cytokine ratio as it predicts CES-D tertiles

Cytokine Predictor *OR 95% CI ap-Value

IL-6 IL-6 0.86 0.30-2.45 0.78

NIHSS 1.07 0.99-1.18 0.098

IL-18 IL-18 3.37 0.48-23.90 0.22

NIHSS 1.03 0.93-1.14 0.56

TNF-α TNF-α 1.17 0.58-2.34 0.66

NIHSS 1.03 0.92-1.14 0.65

IFN-Ɣ IFN-Ɣ 1.98 0.72-5.48 0.31

NIHSS 1.07 0.98-1.17 0.11

IL-1β IL-1β 0.58 0.20-1.74 0.91

NIHSS 1.01 0.90-1.12 0.34

IL-10 IL-10 0.65 0.33-1.30 0.23

NIHSS 1.08 0.99-1.18 0.10

IL-6/IL-10 IL-6/IL-10 1.29 0.72-2.31 0.40

NIHSS 1.07 0.98-1.17 0.11

IL-18/IL-10 IL-18/IL-10 1.97 0.90-4.32 0.091

NIHSS 1.02 0.92-1.13 0.77

TNF-α/IL-10 TNF-α/IL-10 1.88 0.98-3.60 0.056

NIHSS

1.03 0.93-1.15 0.59

65

Cytokine Predictor *OR 95% CI ap-Value

IFN-Ɣ/IL-10 IFN-Ɣ/IL-10 2.17 1.02-4.64 0.045**

NIHSS 1.07 0.98-1.17 0.12

IL-1β/IL-10 IL-1β/IL-10 1.70 0.70-4.11 0.24

NIHSS 1.01 0.90-1.12 0.91

*OR values and their corresponding ap-values reflect the results of individual cytokines and the NIHSS score as a predictors in each ordinal regression model

Figure 10. Bar graph of mean plasma IFN-Ɣ/IL-10 levels across stroke patients in low, middle, and

high CES-D tertiles

3.5 Exploratory Analyses

In order to explore the relationship between neuroimaging findings and CES-D score, average

hippocampal thickness, whole brain atrophy, the global ARWMC score for WMC, and lesion volume

were entered into a backward stepwise linear regression and CES-D score was entered as the dependent

variable. NIHSS score and age were included as covariates. The initial descriptive statistics did not

reveal a significant difference in CES-D scores across lesion location or laterality, therefore this was not

entered into the model as a covariate. The removal criterion for backwards regression model was set to

(P≤0.1). The regression analysis performed 7 iterations, and all predictors were removed from the final

*

*

66 model. Global ARWMC score was excluded first (adjusted R2= 0.001, β= -0.15, p=0.92), followed by

age (adjusted R2= 0.020, β= 0.101, p=0.51), lesion volume (adjusted R2= 0.030, β= -0.127, p=0.35),

NIHSS score (adjusted R2= 0.032, β= 0.145, p=0.29), whole brain atrophy (adjusted R2= 0.029, β=

0.203, p=0.18), and lastly, average hippocampal thickness (adjusted R2= 0.015, β= 0.179, p=0.18).

Figure 11-14 display scatter plots indicating the relationship between CES-D scores and the

neuroimaging outcomes.

Figure 11. Correlation between average hippocampal thickness and CES-D scores.

67 Figure 12. Correlation between total ventricular volume / total intracranial volume, as a

measure of whole brain atrophy, and CES-D scores

Figure 13. Correlation between global ARWMC scores and CES-D scores.

68 Figure 14. Correlation between lesion volume and CES-D scores

A separate backward stepwise linear regression was conducted to determine the significance of

white matter lesion location in relation to PSD. Because there was no significant differences in the

mean ARWMC scores for right and left regions, (Table 7) each bilateral region score (frontal lobes,

temporal lobes, parieto-occipital lobes, bilateral basal ganglia, and infratentorial) were summed to

create 5 ARWMC scores as opposed to 10 bilateral scores. These 5 regional scores were entered into

the regression model with CES-D score as the dependent variable. Controlling for NIHSS and age, the

regression analysis performed 8 iterations, and all predictors were removed from the final model. Total

frontal ARWMC score was excluded first (adjusted R2= -0.021, β= -0.061, p=0.69), followed by total

temporal ARWMC score (adjusted R2= -0.007, β= 0.059, p=0.640), age (adjusted R2= 0.006, β= 0.088,

p=0.50), total basal ganglia ARWMC score (adjusted R2= 0.015, β= -0.138, p=0.30), total infratentorial

ARWMC score (adjusted R2= 0.013, β= -0.127, p=0.30), total parieto-occipital ARWMC score

(adjusted R2= 0.012, β= 0.119, p=0.34), and lastly, NIHSS score (adjusted R2= 0.013, β= 0.167,

p=0.18). Table 8 displays the results of Spearman’s correlations between each bilateral ARWMC score

and CES-D scores.

69 Table 7. Results of paired samples t-tests for right and left regional ARWMC scores

Right and Left

ARWMC Regions

Mean ± SD *t dF ap-Value

Right Frontal 1.8±1.1 -0.60 66 0.55

Left Frontal 1.9±1.2

Right PO 1.2±1.4 0.77 66 0.45

Left PO 1.1±1.4

Right Temporal 0.4±0.9 0.90 66 0.37

Left Temporal 0.5±1.0

Right Basal Ganglia 1.1±1.1 -0.68 66 0.50

Left Basal Ganglia 1.0±1.1

Right Infratentorial 0.3±0.8 -0.67 66 0.50

Left Infratentorial 0.2±0.7

*t values and their corresponding ap-values reflect the results of paired samples t-tests between the means of right and left regional ARWMC scores.

Table 8. Spearman’s correlations between bilateral ARWMC scores and CES-D scores.

ARWMC Region *R ap-Value

Left Frontal -0.053 0.67

Right Frontal 0.171 0.17

Left Parieto-occipital 0.087 0.49

Right Parieto-occipital 0.182 0.14

Left Temporal -0.021 0.86

Right Temporal -0.032 0.80

Left Basal ganglia -0.095 0.45

Right Basal ganglia 0.063 0.61

Left Infratentorial -0.114 0.36

Right Infratentorial -0.008 0.95

*r values and their corresponding ap-values reflect the results of bivariate Spearman’s correlations with CES-D scores

Because no significant neuroimaging predictors emerged from these regression models, the

final exploratory analysis of assessing the interaction between elevated K/T ratio and neurological

pathology in post-stroke depression severity was not completed.

4. Discussion

Serum K/T ratio did not correlate with depressive symptom severity and it was not significantly

different across CES-D tertiles, indicating none, mild or moderate-severe depressive symptom severity.

Similarly, secondary analyses revealed no significant correlations between any plasma cytokines or IL-

10 immunologic ratios and depressive symptoms severity, except plasma IFN-Ɣ/IL-10 was found to be

a significant predictor of mild depressive symptoms. Additionally, exploratory analyses demonstrated

that none of the neuroimaging measurements or regional ARWMC scores were associated with

depressive

70

71 symptoms. Finally, stroke severity, as measured by the NIHSS, and age were the only variables

significantly associated with depressive symptoms in this study.

4.1 The KYN Pathway’s Influence on Post-stroke Depressive Symptoms

This study aimed to assess the relationship between depressive symptoms and the K/T ratio in

ischemic stroke patients. After controlling for all important covariates, including age and NIHSS scores,

the relationship between CES-D scores and K/T ratio was not significant. Our patient sample was

divided into tertiles based on the severity of depressive symptoms, to determine if an elevated K/T ratio

was associated with none, mild or severe depressive symptoms. However, the K/T ratio was not

significantly different across CES-D groups. Changes in the levels of KYN and its metabolites have

been associated with neuropsychiatric outcomes such as MDD,88-92, 305 bipolar mania,82, 83 and

schizophrenia.84-87. With respect to MDD, researchers have observed both positive,88-92 and negative

findings.88, 305, 362, 363, 440-444 Regarding negative findings, these researchers did not find any difference in

the levels of KYN,440, 441, 444 TRP,441 or xanthurenic acid (XA),362, 363, 442, 443 another KYN metabolite

produced by KMO, between depressed and non-depressed patients. Furthermore, two important meta-

analyses by Hackett and Anderson and Ayerbe et al. found that functional disability and stroke severity

were amongst the most significant correlates of PSD 16, 110 While the studies assessed in those analyses

did not include immunological correlates, in the context of PSD it is possible that immunological

contributions may only play a minor role in the development of depression, and these are masked by the

overwhelmingly strong relationship between PSD, stroke severity, and functional disability.16, 110 Our

finding that NIHSS scores significantly correlated with CES-D scores adds credence to this hypothesis.

However, researchers today believe the etiology of PSD stems from the joint effects of biological and

psychosocial mechanisms,12 and the CES-D may also not be the most appropriate scale, however this

will be explored in depth in the limitations section below. In addition, in the context of coronary artery

disease (CAD), we previously reported a significant positive association between the K/T ratio and

degree of depressive symptoms (measured by the CES-D).92 The mean K/T ratio in our CAD

72 population was much lower (25.35±10.73μmol/L) than that of the present population

(70.43±38.35μmol/L), in agreement with means reported in other CAD 445 and stroke66, 323 populations.

The results in stroke suggest IDO activation is above and beyond the levels associated with CES-D

scores in CAD. Furthermore, elevations in the K/T ratio have been significantly correlated with the

progression, severity and mortality of multiple neurological and inflammation-associated diseases, such

as Alzheimer’s disease, Parkinson’s disease, Huntington’s disease, traumatic brain injury, cancer,

malaria, HIV, rheumatoid arthritis, as well as stroke.61-63, 66, 326, 327, 446-450 With respect to stroke, our

group previously found the K/T ratio to significantly correlate with post-stroke cognitive impairment

(PSCI).323 Although units and measurement techniques varied between these studies, it seems as though

the levels of the of the K/T ratio were substantially elevated in most of these groups61, 62, 449, 450

compared to our PSCI323 and PSD populations. This may imply that in stroke patients, extreme

elevations in the K/T ratio make it difficult to decipher the relationship between a high K/T ratio and

depression against the back-drop of other post-stroke sequelae,408 since the K/T ratio and KYN

metabolites have been shown to strongly correlate with mortality66 and worse outcome post-stroke.326

Furthermore, descriptive statistics revealed that insulin use was a near significant predictor of

CES-D groups, and thus a post-hoc analysis was conducted controlling for insulin use as a covariate in

the a priori hypothesis. It is interesting that insulin use was nearly significantly different between CES-

D groups, given the established relationship between diabetes and depression.451 Furthermore, a study

by Ormstad and colleagues identified serum glucose levels to be a significant predictor of post-stroke

fatigue at 6 months post-stroke.256 However, this post-hoc analysis still revealed no significant

relationship and it would be worthwhile for future research to attempt to characterize the relationship

between insulin use, glucose levels and post-stroke depressive symptomatology further.

As discussed in the methods section 2.8.4, the necessary sample size calculated with an α-level

of 0.05, the power set to 80%, and effect size of 0.1433 determined from our CAD population,92 was 80

participants. This is 2 patients less than the amount of patients included in our a priori hypothesis

73 (n=82), however this sample size did not generate a significant relationship between the K/T ratio and

depression. One reason for this could be because that this study may not have had the power to detect

the primary relationship; a post-hoc statistical power calculation revealed that the observed power to

detect a statistically significant correlation between the K/T ratio and depression scores was

approximately 51.0%, which is well below the initial sample size calculation accounting for a power of

80%.

Furthermore, as outlined in section 1.4.8.1, new research suggests that IDO activation may

actually have a positive role in curbing the inflammatory milieu associated with stroke, and thus,

dampening the relationship between cytokines, KYN, and PSD. Animal models of cerebral ischemia

have demonstrated the neuroprotective effects of KYN,452, 453 and IDO activation may suppress the

immune response by reducing the proliferative capacity and inducing apoptosis of T lymphocytes. In

general, this may be carried out through decreased TRP availability,288 increased production of

oxidative and cytotoxic KYN metabolites,289-293 and lastly by indirectly causing naïve cluster of

differentiation (CD) 4 + T cells to differentiate into Treg cells through the production of TGF- β.294 This

last mode of action may be carried out via the AHR. It has been demonstrated that KYN can bind to the

AHR, leading to IDO induction and the generation of Foxp3 Tregs; TGF-β may also amplify the

interaction between KYN and AHR by upregulating AHR expression.295 The general result of IDO

activation is a dampened Th1 response,291 which shifts the immune response to Th2 activation, while

also stimulating the production of Tregs. Consequently, Tregs have a repressive effect on the Th1 and

Th2 responses, driving the immune system towards equilibrium.105, 297 Taking these novel findings into

consideration, it seems plausible that the immunosuppressive effects of KYN metabolites may account

for the lack of a relationship between inflammation-associated IDO activation and PSD in this study.

This may be tested in animal studies in the future by looking at the immunosuppressive effects of

kynurenine, quinolinic acid and 3-hydroxykynurenine and analyzing responses of animals to

behavioural paradigms of depression.

74 The level of free TRP was also not significantly associated with CES-D scores. As mentioned

in section 1.4.9.1, TRP is the precursor for the neurotransmitter serotonin,93 and a depletion in serotonin

has been associated with major depressive disorder (MDD) in the past.94, 95 Additionally, acute TRP

depletion paradigms have caused transient depressive symptoms in recovered MDD patients.96, 97

However, these symptoms were only present in patients with a personal history of depression who had

recovered, while this study excluded patients with a history of depression due to the fact that PSD is

hypothesized to have a different etiology than depression in the non-stroke population. Therefore,

although it was hypothesized that acute changes in mood post-stroke are due to TRP depletion as a

result of IDO activation, it seems more likely that disruptions in mood are the result of cytokine

imbalance and may be more closely related to sickness behaviour. A discussed in section 1.4.6, one

researcher discovered that post-stroke fatigue, not depression, was found to significantly correlate with

pro-inflammatory cytokines.256 Furthermore, sickness behaviour is a motivational259 and adaptive

response260 during infection which will resolve once the body eliminates the pathogen,261 while

depression does not follow the same course.209 Only about a third of patients receiving IFN-α or IL-2

immunotherapy experience symptoms of depression,214 prompting researchers to characterize two forms

of cytokine-induced depressive symptoms.209 One of which are the early-onset somatic disturbances of

treatment that most, if not all patients experience,209 while the other form of cytokine-induced

depressive symptoms are psychological in nature, late-onset, and are exhibited in as many as 50% of

these patients.209, 262-264 Furthermore, SSRI prophylaxis in these patients improved psychological mood

disturbances without influencing the somatic depressive symptoms.262 Dantzer et al. suggests that “It is

possible that depression represents a maladaptive version of cytokine-induced sickness, which could

occur when activation of the innate immune response is exacerbated in intensity and/or duration...”209

These findings shed light on the results of the current study and may imply that acute depressive

symptoms observed post-stroke are more closely related to early onset sickness behaviour than a true

depression response. While the chronic mood disturbances after stroke, similar to the ‘late-onset,

75 psychological’ symptoms of immunotherapy, may be a result of increases in neurotoxic KYN

metabolites wreaking havoc on emotion-regulating centres of the brain.

4.2 Influence Cytokines on Post-stroke Depressive Symptoms

As another possible mechanism, or as an upstream link in the pathway of IDO-mediated

depression, the levels of pro- and anti-inflammatory cytokines were assessed in relation to PSD. This

study failed to find a relationship between elevations in pro-inflammatory cytokines or reduction in IL-

10 and CES-D scores in isolation. This is in contrast to previous reports of pro-inflammatory cytokine

elevations in the context of MDD215-225, 228, 230-233 as well as PSD.36, 243 However, certain researchers

have failed to find a relationship between PSD and post-stroke pro-inflammatory cytokine

elevations.244, 254255 As outlined in section 1.4.6, the balance between pro- and anti-inflammatory

cytokines may be more highly associated with depression than either type of these cytokines in

isolation, as evidence by many positive findings in both MDD31-35 and PSD. 36, 37 In this study, the only

cytokine ratio found to be a significant correlate of CES-D tertiles was IFN-Ɣ/IL-10, however there was

a strong trend for TNF-α/IL-10 to be a significant correlate of CES-D tertiles as well. This finding may

be related to the aforementioned influence of IDO activation on immunosuppression. Since Tregs

(possibly stimulated by KYN)295 are capable of producing IL-10454-456 and IL-10 production in

general,190 as well as through Tregs production457 has been demonstrated to act in a neuroprotective

manner post-ischemic stroke. Briefly, animal and in vitro models have demonstrated that IL-10 can

significantly reduce infarct size191 and enhance neuronal survival.192 Therefore, the immunologic ratio

between pro-inflammatory cytokines and IL-10 may broadly signify immune imbalance between Th1

and Th2 immune responses, which would be detrimental to the post-stroke brain. Yet, past research

suggests that a skewed Th1/Th2 response is also associated with depression. Myint et al. observed

significant elevations in plasma IFN-Ɣ/IL-4 in depressed subjects and upon antidepressant treatment,

the level of TGF-β1 increased and IFN-Ɣ/IL-4 ratio decreased significantly.200 In addition, the ordinal

regression between IFN-Ɣ/IL-10 and CES-D tertiles revealed that an elevation in this immunologic

76 ratio was a significant predictor of the middle CES-D tertile, while TNF-α/IL-10 was trending to be a

significant predictor of the highest CES-D tertile. In relation to PSD, the CES-D is a widely used and

validated measure for quick screening of the presence and severity of depressive symptoms. 3 In

patients who have suffered from a stroke, a cut-off score of 16 or greater has a sensitivity ranging from

86-100% and specificity of 73-90%. 418, 458 In the present study, CES-D total scores were divided into

tertiles; the lowest tertile ranged from 0-6, the middle tertile ranged from 7-15, and the highest tertile

ranged from 16-47. Therefore, in this patient sample the highest tertile begins at the same cut-off

previously found to have high sensitivity and moderate specificity in the context of PSD. Perhaps if

more samples for TNF-α were assayed, the relationship between TNF-α/IL-10 and the highest CES-D

tertile would become significant. This may be a limitation of the study that will be discussed in further

detail below. Although a score between 7-15 was not previously validated for the detection of PSD,

considering the CES-D as a depressive symptom severity continuum,415 it may indicate that a score

between 7-15 represents mild depressive symptom severity. Depressive symptom severity, as opposed

to categorizing patients into depressed and non-depressed groups, has also been correlated with

elevations in inflammatory molecules by many researchers.214, 215, 233, 236 For instance, Suarez et al.

observed that higher BDI scores were correlated with elevated TNF-alpha and IL-8

production,236 indicating that a linear relationship exists between immune imbalance and post-stroke

depressive symptom severity, as opposed to solely categorical associations. The results of this

exploratory analysis are interesting, however they should be interpreted cautiously, since few groups

have assessed the relationship between mild depressive symptoms and immune imbalance and more

research is necessary in order to determine the significance of these findings in the context of PSD.

4.3 Influence of WMC, Atrophy and Lesion Volume on Post-stroke Depressive Symptoms

As mentioned in section 1.5, few studies have assessed the interaction between acute infarcts,

WMC, and atrophy in relation to PSD, and all have failed to find joint associations between these three

radiological finding and PSD39, 40, 45, 48 This thesis also failed to find a relationship between all of the

neurological measurements assessed, including average hippocampal thickness, whole brain atrophy,

77 global ARWMC score for WMC, and lesion volume, and CES-D scores. There was also no association

between any of the regional ARWMC scores and depressive symptom severity. Furthermore, because

none of the aforementioned neuroimaging measurements were correlated with depressive symptom

severity, the final exploratory analysis incorporating neuroanatomy with the K/T ratio was not pursued.

As outlined in section 1.5.1, lesion location and its association with PSD has been a highly

debated topic. This study failed to find a relationship between lesion location (classified as anterior,

intermediate, posterior and extending, as well as laterality) and depression scores. This is in agreement

with multiple studies failing to find a relationship between left anterior lesions and PSD. 130, 372-376

Research by Robinson et al. assessing the timing of depressive symptom development is also not in

agreement with the findings of this study, since they found that in the acute stage post-stroke

(approximately 6 months) depression severity and distance of the lesion to the frontal pole in left-sided

strokes are inversely correlated.387 However, in this study timing of assessment post-stroke ranged from

4-155 days; well within the ‘acute stage’ by the Robinson et al. classification and not significantly

correlated with CES-D scores. The work of Starkstein et al.141, 142, 377 and Herrmann et al.380

demonstrated that lesions within the basal ganglia and thalamus were also associated with depression

post-stroke. Unfortunately the method of assessing lesion location for this thesis did not dichotomize

basal ganglia versus cortical strokes, as it was adopted from the Robinson et al. hypothesis of

classifying lesions based on the 4 quadrants of the brain.138 This may be a limitation of the study

considering the positive associations these authors found between basal ganglia lesions and PSD,

however the mixed and inconclusive findings of meta-analyses384-386 indicate that more research still

needs to be done to elucidate the relationship between all classifications of lesion location and PSD.

As previously discussed, most research supports the involvement of lesion volume in PSD, 40-43

however, like this thesis, other researchers have failed to find a relationship.38, 39 The variation in

findings may be due to methodological differences between this thesis and the other studies that have

assessed lesion volume in the past. This thesis, as described in the methods section, measured lesion

volume by manually tracing lesion area for each slice on the CT scan, multiplying the area on each slice

78 by slice thickness (1mm3), and then summing to obtain the total lesion volume. This particular method

has been found to have very good reproducibility in acute ischemic stroke.429 Studies have used many

different methods, including similar methods to the one used here, however they varied by

neuroimaging technique, as one study assessed on DWI,43 while another assessed on both CT and MRI

images.41 Furthermore, a study by Vataja et al. estimated lesion volume by grouping lesion diameters

into 4 categories and radii were then used in the equation for volume of a ball.40 Although, considering

the differences in measurement techniques between studies, the mean ± SD (21.11±47.02 cm3) lesion

volume of this study was quite similar to the volumes measured by these previous researchers.38, 40, 41, 43

This may indicate that the difference in methodology between studies is most likely not the reason for

variable findings between groups, and that more research should be conducted in the field to determine

the importance of lesion volume in the development of PSD.

This thesis did not find a relationship between WMC and post-stroke depressive symptoms.

Both the global and regional ARWMC scores did not significantly correlate with CES-D scores. As

previously discussed in section 1.5.2, neurodegeneration in the form of age-related WMC is highly

associated with depression among the elderly.402-404 However, in relation to PSD the evidence is less

consistent; with multiple studies failing to find a relationship with WMC.39-43, 45-47 Only two studies

have found a relationship between WMC and PSD,38, 48 one of which by Pohjasvaara et al. used a

depression symptom severity scale (BDI) to measure depressive symptoms,48 like the present study.

Pohjasvaara et al. used the Fazekas scale431 to measure WMC, while this study employed the ARWMC

scale,430 which may account for the difference in findings. Yet, the Fazekas scale was observed to have

reduced inter-rater reliability in assessing less severe white matter disease459 and is only suitable for

MRI, while the ARWMC scale was developed to target these reliability issues44 and is also designed for

use in both CT and MRI.430 These variations in findings and methodology make it difficult to decipher

the true relationship, if any, between PSD and WMC. Although a clear association between late-life

depression and WMC exists, depression in the context of stroke may not have its origins in pre-existing

WMC.

79

Additionally, this thesis found that whole brain atrophy was also not significantly correlated

with depression scores. Frontal lobe atrophy has been found to significantly correlate with late-life

depression,460 and was proposed by one study as a risk factor for PSD.405 Section 1.4.8.3 covers the

relationship established between depression and atrophy in the non-stroke population and how the

neuroprogression hypothesis of depression plays into these findings. However, in section 1.5.3 it is

made evident that atrophy in relation to PSD may be more complex. This thesis hypothesized that

atrophy in the context of an acute infarct will be associated with depressive symptoms following a

stroke, however because of the cross-sectional nature of the study one cannot assume that atrophy pre-

existed the stroke lesion. This thesis had two measures of atrophy; hippocampal and whole brain. Three

studies assessing atrophy and PSD demonstrated a relationship between atrophy and PSD,39, 49, 405 while

two studies failed to find a relationship.40, 45 All three of these positive studies used different methods to

assess atrophy; one measured atrophy by visual rating, another measured the ventricle to brain ratio,

and the third measured atrophy using a novel quantitative hybrid warping method. In contrast, both of

the negative studies used a visual rating scale to assess atrophy. Furthermore, this study assessed whole

brain atrophy by manually tracing the ventricles to obtain total ventricular volume and then normalizing

these values by dividing this value by the total intracranial volume (TIV) to produce a ratio value of

total ventricular volume: total intracranial volume. A limitation of this thesis is the variation in

methodology between it, and the aforementioned positive and negative studies, making it difficult to

compare findings between groups. Another limitation is that fact that this thesis measured hippocampal

thickness on CT as opposed to hippocampal volume, which is only possible on MRI. This may also

account for the negative findings; which will be discussed in the limitations section below. The

reasoning behind including hippocampal thickness as a neuroimaging outcome was because of the

findings in animal studies that demonstrate kynurenine metabolites (mainly quinolinic acid) to be

especially toxic to the hippocampus, causing targeted neurodegeneration in this area specifically.301

Furthermore, administration of kynurenic acid or blockade of the enzyme that indirectly leads to

quinolinic acid production (kynurenine-3-monooxygenase), reduces QUIN-induced neurotoxicity.301

80 The hippocampus is also especially vulnerable to hypoxia, one study involving cardiac arrest patients

found marked reductions in hippocampal volume compared to controls 21 days after cardiac arrest.325

The researchers of this study postulated that this reduction in volume signifies the hippocampus’

increased susceptibility to global brain ischemia.325 This may imply that oxidative and neurotoxic

factors can have a relatively acute effect on brain volume and neurodegeneration. However, because of

the cross-sectional nature of this study one cannot determine whether the hypoxic (or

oxidative/excitotoxicity/etc.) consequences are acute or chronic. Finally, a recent meta-analysis found

that smaller amygdala volume in stroke patients was significantly associated with depression, however

whole brain atrophy was not.461 This suggests that PSD research in the area of neurimaging should

focus more attention on measuring discrete regional differences in brain volume, as this method may be

better at determining the role of atrophy in PSD.

As previously mentioned in the discussion, physical disability, as measured by stroke severity

or functional impairment, has been found by two important meta-analyses to be one of the most

significant correlates of PSD.16, 110 As suggested for the relationship between K/T and PSD, the same

may be true for other biological measurements in relation to PSD, such as the neuroimaging outcomes

measured in this thesis. Thus, it is likely that they too, only play a minor role in the development of

depression and these are overwhelmed by the strong relationship between PSD, stroke severity, and

functional disability.16, 110 Our finding that NIHSS scores significantly correlated with CES-D scores

adds credibility to this notion. Whyte and Mulsant believe the etiology of PSD stems from the joint

effects of biological and psychosocial mechanisms,12 however this thesis only found strong correlations

between the NIHSS and CES-D. Many of the PSD studies mentioned in section 1.5 assessed either

functional impairment or stroke severity alongside neuroimaging findings.38, 39, 43, 46, 47, 49, 125, 462 The

results of these studies demonstrated that physical disability, as measured by either one of these

psychosocial outcomes, is significantly correlated with depression.39, 46, 47, 49, 125, 462 Furthermore, two of

these studies found that after controlling for these outcomes, the effect of the particular neuroimaging

measurement on depression was negated.47, 49, 406 Therefore, this may imply that the effects of either

81 stroke severity or functional impairment are strong enough to overrule any small neurobiological

contributions in the development of PSD.44

As mentioned earlier, because none of the aforementioned neuroimaging measurements were

correlated with depressive symptom severity, the final exploratory analysis assessing the joint effects of

neuroanatomy and the K/T ratio on PSD was not pursued. Although we were not able to explore this

aim, it is still worthwhile to briefly discuss the motivation behind it. The idea behind this exploratory

analysis was fueled by the fact that depression (both PSD and MDD) have been found to significantly

correlate with neurodegeneration in the form of atrophy,51, 56-5939, 49, 60, 405, 460 WMC,38, 48, 402-404 and acute

stroke lesion volume.40-43 In addition, altered IDO activity has been significantly correlated with the

progression, severity and mortality multiple neurodegenerative diseases, such as Alzheimer’s disease,

Parkinson’s disease, Huntington’s disease, as well as stroke and traumatic brain injury.61-63, 323, 327 KYN

and its metabolites have also been demonstrated to cause neurodegeneration in animal and in vitro

models. In particular, inflammation-induced IDO and KMO activation and the production of KYN

metabolites have been hypothesized to play a role in neuroprogression,50, 67, 68 through pro-oxidative

mechanisms,69-72 disruption in energy and metabolism,73, 74 and through glutamate excitotoxicity.21, 75, 76

Neuroprogression is a term given to the process of gross and cellular neurodegeneration which has been

hypothesized to have a link to the cause and/or consequence of worsening depression and cognitive

decline.50-54 It then follows that an elevated K/T ratio in the presence of neurodegeneration may be more

relevant in contributing to the development of PSD than either variable alone, as the aforementioned

evidence suggests that these biological outcomes are not mutually exclusive. However, the results of

this thesis do not support this notion, since there was no significant correlation between the K/T ratio,

most of the serum cytokines, or any of the neuroimaging outcomes and depression scores. This is in

agreement with two recent studies published by Ormstad et al. who assessed the correlations between

12 acute serum cytokine levels and depression, as well as infarct volume and laterality.255, 434 They

found that serum inflammatory markers, lesion volume and laterality were not associated with

82 depression,255 nor were the various measured cytokines significantly correlated with lesion volume or

laterality.434

4.4 Limitations

The first limitation of this thesis is the large variation in time of assessment post-stroke, which

ranged from 4-155 days. This was not found to be a significant correlate of CES-D scores in the initial

descriptive analysis, and recent evidence suggests that IDO activity remains elevated at least one year

following severe brain injury.327 However, a study by Noonan et al. demonstrated that significantly

elevated C-reactive protein (CRP) and white cell count levels at 18 months post-stroke were not

associated with depression scores.463 Thus, in the present study, the broad range of time between date of

stroke and assessment may have eliminated or dampened the association between the K/T ratio, as well

as cytokine concentrations and depression scores, respectively. Yet, as seen in Appendix 4, bivariate

Spearman’s correlations revealed no significant associations between time post-stroke and any of the

serum biomarkers.

Secondly, the gold standard for the diagnosis of post-stroke depression is the Structural Clinical

Interview Diagnostic (SCID) and Statistical Manual (DSM-IV) criteria.3 Thus it could be argued that

using the CES-D to screen for the presence of depressive symptoms does not capture or identify a true

depressive episode with clinical relevance. However, as mentioned previously, in patients who have

suffered from a stroke, a cut-off score of 16 or greater has a sensitivity ranging from 86-100% and

specificity of 73-90%. 418, 458 In the present thesis, 31.7% of patients had a score of ≥16, which is quite

similar to a very recent meta-analysis by Ayerbe et al. that found the pooled prevalence of depression

observed at any period to be approximately 29%, and 33% between 1 and 6 months post stroke.110

Furthermore, proper administration of SCID depends on sound clinical objectivity. In a study such as

this one, the more researchers involved in administering a diagnostic test, the more likely the reliability

of scores will be decreased. Conversely, the CES-D is a self-report screening scale, and researchers

have suggested that depression screens may prevent patients from accurately lining up the emotions

83 they feel with descriptions of emotions in the assessment.464 Furthermore, there is scarce research on

how adults perceive depression and how they interpret this paradigm into context with their personal

emotions,465, 466 although some findings demonstrate that patients have trouble deciphering between

“reactions to adversity” and depressed mood.464 However the main goal of this thesis was to

characterize the relationship between the K/T ratio and depressive symptom severity, as it was

hypothesized that their relationship was linear. Furthermore, K/T ratio was recently found to be

significantly elevated in somatization compared to depression and also associated with somatic

symptom severity.57, 305, 313 In a study involving IL-2 and/or interferon-α cancer therapy, symptoms

characterized by the researchers as ‘neurovegetative’, including lack of sleep, poor appetite, and

decreased energy, were significantly more pronounced in patients with mild depressive symptoms than

patients within the ‘‘absent’’ depressive symptom range, as measured by the MADRS.435 Thus, a

depression scale was necessary in order to determine if mild, moderate, or severe depressive symptoms

were associated with an elevated K/T ratio; something that a categorical diagnosis of depressed versus

non-depressed would not have been able to detect as well.

Additionally, peripheral concentrations of cytokines and plasma KYN concentrations may not

be representative of central concentrations. Firstly, although KYN can cross the blood—brain barrier,19,

467 peripheral concentrations of KYN may be much lower than central concentrations and thus, may fail

to show strong correlations with depression. A study by Raison et al. involving IFN-α treatment

observed large differences between peripheral and central KYN.106 However, this study also found

significantly increased peripheral blood KYN was highly correlated with cerebral spinal fluid (CSF)

KYN levels, and these levels were significantly associated with CSF levels of KYN metabolites.106

Furthermore, it was demonstrated in bivariate correlations that CSF KYN and CSF QUIN significantly

correlated with depression scores, yet in a multiple linear regression taking into account CSF TRP,

KYN, QUIN, KYNA, IFN-α, IL-6, sIL-6R, sTNFR2, MCP-1 and history of depression, the only KYN

metabolite that emerged as a significant predictor of depression scores was CSF QUIN.106 Although

CSF KYN was not directly significantly associated with depression scores, CSF QUIN was found to

84 correlate with peripheral KYN concentrations106 and it is hypothesized that K/T ratio is correlated with

depression scores in part due to toxic KYN metabolites, such as QUIN, wrecking havoc on emotional

neuronal circuits of the brain. In terms of peripheral levels of cytokines and their representation of

central concentrations, the same holds true. Elevations in cytokines have been consistently associated

with depression in the non-stroke population 215-225, 228, 230-233 and many studies have found them to be

elevated in the context of PSD,36, 243 as well as stroke outcomes such as disability37 and fatigue.256

Associations between peripheral and central concentrations of cytokines have been documented,99 and

cytokines have been demonstrated to cross the BBB, and they can also cross back from the CSF into

peripheral circulation (reviewed in468), however they should not be considered one in the same.469, 470

Thus the results of this study were also limited by the fact that we did not measure intrathecal

production of inflammatory markers, as these may be more representative of concentrations in the

brain.

On the same note, imputation of cytokine values at the lower limit of detectability may limit the

utility of this study. Assay sensitivities were 0.2 pg/ml for IL-10, 1.0 pg/ml for IFN-Ɣ, 0.09 pg/ml for

TNF-α, 0.057 pg/ml for IL-1β, 0.1 pg/ml for IL-6, and 12.5 pg/ml for IL-18, which are comparable to

the sensitivities of used by other studies observing stroke and depressed populations by means of

standard ELISA assays,103, 426, 427 however there were no undetectable samples for IL-6 or IL-18. While,

imputing at the lower limit of detectability may create data that stray from the true mean, the key aim of

this study was to examine increased levels of cytokines, as it was hypothesized that pro-inflammatory

cytokines will be increased in depressed patients of this population.407 As such, we did not think this

method would interfere with our objective of examining the role of increased cytokine production.

Nonetheless, IL-10, an anti-inflammatory cytokine, was imputed at the lower limit of detectability,

when our goal was to detect decreased concentrations in PSD patients.407 For IL-10, since values may

be even lower than the limit of detectability, imputing them would artificially inflate concentrations.

Imputation of values may warp the linear relationship between CES-D scores and cytokine levels and

cause the researcher to see less of a correlation between CES-D scores and IL-10,407 as well as for the

85 IL-10 immunologic ratios. A significant relationship was observed between IFN-Ɣ/IL-10 and the

middle CES-D tertile, however had there been less imputations for IL-10, the trend between TNF-α/IL-

10 and CES-D tertiles may have become significant. Appendix 5 displays the percentage of values

imputed for each cytokine and Appendix 6 shows the result of Chi-squared tests between depressed

and non-depressed groups (CES-D≥16 and <16) and imputed versus non-imputed cytokines,

demonstrating that there are no significant differences in imputation of any cytokines between groups.

Lastly, the use of CT neuroimaging for analyzing the radiological correlates of PSD may be

considered another limitation of this thesis. In terms of WMC scoring, the ARWMC scale was shown to

identify an increased number of small white matter lesions on MRI, while bigger infarcts were

identified equivalently or better by CT.430 MRI is better at detecting the subtleties of WMC than CT and

is less susceptible to artefacts such as bone hardening, especially near the medial temporal lobe.403, 471

However, CT is more apt at distinguishing clinically significant lesions,472 is equivalent to MRI in

measuring atrophy,471 and possibly more appropriate for use in geriatric populations because it is not as

easily affected by motion artefacts.403 Furthermore, CT is the most extensively utilized neuroimaging

device in the world403 and thus, this study may be more clinically applicable in the context of stroke.

4.5 Recommendations

The most important recommendation for future studies assessing the relationship between

peripheral inflammatory markers and PSD, would be to track these markers and depressive symptom

severity longitudinally. Although the K/T ratio was not significantly correlated with depressive

symptom severity in this study, it would be interesting to see if an initial increase in the K/T ratio

confers greater vulnerability to depression at 6, 12, 18 months and beyond. The aforementioned

observation that the K/T ratio is initially much higher in post-stroke populations than in other

neurological diseases adds legitimacy to this recommendation. Furthermore, studies propose that the

prevalence of PSD reaches a maximum at roughly 3-6 months, diminishes by 50% at 1 year, and then it

may remain elevated at approximately 20% for 2 years and beyond.3, 12, 13 This last finding is

86 troublesome because it demonstrates that PSD can be a chronic, non-remitting disease in many stroke

survivors. It was hypothesized that one of the ways in which an increased K/T ratio contributes to PSD

is through an increase in neurotoxic KYN metabolites which may cause neurodegeneration of the post-

stroke brain. If this is the case, it seems more likely that the effects of these toxic KYN metabolites

would produce a delayed effect, and may possibly contribute to the chronic, non-remitting cases of

PSD.

Following on the recommendation for longitudinal studies, it would also be important for future

studies to incorporate measures of functional outcome, as this would help decipher if the psychosocial

impacts on depression stand compared to the biological contributions of inflammation. This may also

help uncover the complex interaction between psychosocial stressors, inflammatory markers and

depression. A recent meta-analysis revealed elevated levels of pro-inflammatory cytokines were

associated with these stressors473 and elevated levels of IL-6, in particular, have been associated with

childhood adversity as well.470, 474 In the present study a significant correlation was demonstrated

between NIHSS scores and the CES-D. Although disability, as measured by the NIHSS, has ties with

functional outcome, the NIHSS is a measure of stroke severity, which strictly speaking is not a

psychosocial outcome.406 Future studies might assess activities of daily living, using the Barthel

Index,475 or the modified Rankin Scale476, 477 as these scales typify the psychosocial outcome of

functional dependence and will help to further characterize its place in the development of depression

post-stroke.

In MDD, genetic variations in IL-1β and TNF-α alleles have been demonstrated to be

associated with antidepressant treatment resistant depression and elevated risk of developing

depression.234, 235 While in PSD, IL-4 and IL-10 polymorphisms related to lower anti-inflammatory

cytokine production were found to be significantly associated with depression.107 It would be

worthwhile for future studies to assess genetic variations in pro- and anti-inflammatory cytokine alleles

as they relate not only PSD, but also to increased activation of IDO as measured by the K/T ratio. This

87 may help to reveal populations that are more vulnerable to the effects of inflammation-associated IDO

activation. Furthermore, a neurotrophin called brain derived neurotrophic factor (BDNF) is a factor that

has been implicated in adult neuronal growth and researchers have demonstrated its impaired

production in the context of depression,478-481 and in chronic482 or acute483 stress. In addition, reductions

in peripheral BDNF concentrations have been associated with diminished hippocampal volumes in

antidepressant-naive first-episode MDD,484 while a polymorphism in the BDNF allele called

BDNFVal66Met, associated with a reduction in its release, was also found to be associated with reduced

hippocampal volume485 (reviewed in470). In relation to PSD, a very recent study by Kim et al. found that

BDNF methylation status was positively correlated with the presence of PSD at baseline and follow-up,

and the BDNFVal66Met polymorphism was associated with depression at baseline.486 Furthermore, a

recent meta-analysis found a significant reduction of BDNF levels in PSD.461 Future studies may find it

meaningful to analyze the association between serum levels of BDNF and hippocampal atrophy in

depressed stroke patients.

Finally, future studies should assess the relationship between PSD and the potent inflammatory

cytokine, IL-17. This is a cytokine produced by ƔƍT cells487 and Th17 cells,488 that has been implicated

in the process of post-stroke neurodegeneration (reviewed in185). IL-10 may play a role in curbing

detrimental IL-17 expression via intracellular suppressors of cytokine signalling (SOCS) proteins,

which link up with cytokine receptor complexes and inhibit intracellular cascades.489-491 Furthermore,

knockdown of one particular SOCS protein (SOCS3) lead to larger infarct volumes and worse

neurological outcomes.490 As mentioned above, IDO may act to suppress immune responses via KYN

binding to the AHR in dendritic cells and causing the generation of Tregs.294, 295 Tregs are also capable

of producing IL-10454-456 and thus KYN binding to the AHR may shift cytokine production to IL-22 (IL-

10 superfamily cytokine) production, as opposed to IL-17 production.492 In this way, KYN may actually

have a positive effect on preventing neurodegeneration caused by IL-17. It would be interesting to

measure the levels ofIL-17 in patients with PSD, and even more intriguing would be to assess the ratios

88 of IL-17/IL-10 and IL-17/KYN in the context of PSD. These immunologic ratios may be more

representative of neurodegeneration than the ones assessed in this thesis.

4.6 Conclusion

In conclusion, this thesis sought to investigate the neurobiological mechanisms of PSD through

the assessment of current contemporary hypotheses, namely the inflammatory and neuroanatomical

factors which may have a role in its development. Principally, this study aimed to uncover the

relationship between acute inflammatory serum biomarkers and post-stroke depressive symptom

severity. The K/T ratio was not significantly associated with depression scores. Similarly, while IFN-

Ɣ/IL-10 ratio was found to be a significant predictor of mild depressive symptoms, this was found in

exploratory analyses, and no relationship was found with other ratios. As such, the finding lacked

consistency. Furthermore, this thesis explored the relationship between neuroimaging findings and post-

stroke depressive symptom severity, as it was hypothesized that they represent chronic factors involved

in the development of PSD. Yet, none of the neuroimaging measurements of this study were associated

with depression. The fact that stroke severity was significantly correlated with depression may indicate

that psychosocial mechanisms override the effects of neurobiology in the etiology of PSD. With respect

to immunologic ratios, longitudinal studies incorporating a measure of functional dependence are

necessary to further characterize the relationship between psychosocial factors and immune imbalance

within the realm of PSD research.

5. References

1. Lloyd-Jones D, Adams RJ, Brown TM, Carnethon M, Dai S, De Simone G, et al. Executive summary: Heart disease and stroke statistics--2010 update: A report from the american heart association. Circulation. 2010;121:948-954

2. Carota A, Staub F, Bogousslavsky J. Emotions, behaviours and mood changes in stroke. Curr Opin Neurol. 2002;15:57-69

3. Rothenburg L, Herrmann N, Lanctot K. Post-stroke depression. Geriatrics and Aging. 2007:415-424

4. House A, Knapp P, Bamford J, Vail A. Mortality at 12 and 24 months after stroke may be associated with depressive symptoms at 1 month. Stroke; a journal of cerebral circulation. 2001;32:696-701

5. Herrmann N, Black SE, Lawrence J, Szekely C, Szalai JP. The sunnybrook stroke study: A prospective study of depressive symptoms and functional outcome. Stroke; a journal of cerebral circulation. 1998;29:618-624

6. Gillen R, Tennen H, McKee TE, Gernert-Dott P, Affleck G. Depressive symptoms and history of depression predict rehabilitation efficiency in stroke patients. Archives of physical medicine and rehabilitation. 2001;82:1645-1649

7. Kong KH, Yang SY. Health-related quality of life among chronic stroke survivors attending a rehabilitation clinic. Singapore Med J. 2006;47:213-218

8. Choi-Kwon S, Kim HS, Kwon SU, Kim JS. Factors affecting the burden on caregivers of stroke survivors in south korea. Archives of physical medicine and rehabilitation. 2005;86:1043-1048

9. Ghose SS, Williams LS, Swindle RW. Depression and other mental health diagnoses after stroke increase inpatient and outpatient medical utilization three years poststroke. Med Care. 2005;43:1259-1264

10. Hackett ML, Yapa C, Parag V, Anderson CS. Frequency of depression after stroke: A systematic review of observational studies. Stroke; a journal of cerebral circulation. 2005;36:1330-1340

11. Blazer DG, Kessler RC, McGonagle KA, Swartz MS. The prevalence and distribution of major depression in a national community sample: The national comorbidity survey. The American journal of psychiatry. 1994;151:979-986

12. Whyte EM, Mulsant BH. Post stroke depression: Epidemiology, pathophysiology, and biological treatment. Biological psychiatry. 2002;52:253-264

13. Whyte EM, Mulsant BH, Vanderbilt J, Dodge HH, Ganguli M. Depression after stroke: A prospective epidemiological study. Journal of the American Geriatrics Society. 2004;52:774-778

14. Hackett ML, Anderson CS, House AO. Management of depression after stroke: A systematic review of pharmacological therapies. Stroke; a journal of cerebral circulation. 2005;36:1098-1103

15. Chen Y, Guo JJ, Zhan S, Patel NC. Treatment effects of antidepressants in patients with post-stroke depression: A meta-analysis. Ann Pharmacother. 2006;40:2115-2122

16. Hackett ML, Anderson CS. Predictors of depression after stroke: A systematic review of observational studies. Stroke; a journal of cerebral circulation. 2005;36:2296-2301

17. Werner-Felmayer G, Werner ER, Fuchs D, Hausen A, Reibnegger G, Wachter H. Characteristics of interferon induced tryptophan metabolism in human cells in vitro. Biochim Biophys Acta. 1989;1012:140-147

18. Fuchs D, Moller AA, Reibnegger G, Stockle E, Werner ER, Wachter H. Decreased serum tryptophan in patients with hiv-1 infection correlates with increased serum neopterin and with neurologic/psychiatric symptoms. J Acquir Immune Defic Syndr. 1990;3:873-876

89

90 19. Fukui S, Schwarcz R, Rapoport SI, Takada Y, Smith QR. Blood-brain barrier transport of

kynurenines: Implications for brain synthesis and metabolism. Journal of neurochemistry. 1991;56:2007-2017

20. McKinney J, Knappskog PM, Haavik J. Different properties of the central and peripheral forms of human tryptophan hydroxylase. Journal of neurochemistry. 2005;92:311-320

21. Perkins MN, Stone TW. An iontophoretic investigation of the actions of convulsant kynurenines and their interaction with the endogenous excitant quinolinic acid. Brain research. 1982;247:184-187

22. Pemberton LA, Kerr SJ, Smythe G, Brew BJ. Quinolinic acid production by macrophages stimulated with ifn-gamma, tnf-alpha, and ifn-alpha. Journal of interferon & cytokine research : the official journal of the International Society for Interferon and Cytokine Research. 1997;17:589-595

23. Shopsin B. Enhancement of the antidepressant response to l-tryptophan by a liver pyrrolase inhibitor: A rational treatment approach. Neuropsychobiology. 1978;4:188-192

24. Guillemin GJ, Kerr SJ, Pemberton LA, Smith DG, Smythe GA, Armati PJ, et al. Ifn-beta1b induces kynurenine pathway metabolism in human macrophages: Potential implications for multiple sclerosis treatment. Journal of interferon & cytokine research : the official journal of the International Society for Interferon and Cytokine Research. 2001;21:1097-1101

25. Carlin JM, Borden EC, Sondel PM, Byrne GI. Biologic-response-modifier-induced indoleamine 2,3-dioxygenase activity in human peripheral blood mononuclear cell cultures. Journal of immunology (Baltimore, Md. : 1950). 1987;139:2414-2418

26. Fujigaki H, Saito K, Fujigaki S, Takemura M, Sudo K, Ishiguro H, et al. The signal transducer and activator of transcription 1alpha and interferon regulatory factor 1 are not essential for the induction of indoleamine 2,3-dioxygenase by lipopolysaccharide: Involvement of p38 mitogen-activated protein kinase and nuclear factor-kappab pathways, and synergistic effect of several proinflammatory cytokines. Journal of biochemistry. 2006;139:655-662

27. Liebau C, Baltzer AW, Schmidt S, Roesel C, Karreman C, Prisack JB, et al. Interleukin-12 and interleukin-18 induce indoleamine 2,3-dioxygenase (ido) activity in human osteosarcoma cell lines independently from interferon-gamma. Anticancer research. 2002;22:931-936

28. Fujigaki S, Saito K, Sekikawa K, Tone S, Takikawa O, Fujii H, et al. Lipopolysaccharide induction of indoleamine 2,3-dioxygenase is mediated dominantly by an ifn-gamma-independent mechanism. European journal of immunology. 2001;31:2313-2318

29. Musso T, Gusella GL, Brooks A, Longo DL, Varesio L. Interleukin-4 inhibits indoleamine 2,3-dioxygenase expression in human monocytes. Blood. 1994;83:1408-1411

30. Chiarugi A, Calvani M, Meli E, Traggiai E, Moroni F. Synthesis and release of neurotoxic kynurenine metabolites by human monocyte-derived macrophages. Journal of neuroimmunology. 2001;120:190-198

31. Dhabhar FS, Burke HM, Epel ES, Mellon SH, Rosser R, Reus VI, et al. Low serum il-10 concentrations and loss of regulatory association between il-6 and il-10 in adults with major depression. Journal of psychiatric research. 2009;43:962-969

32. Kubera M, Kenis G, Bosmans E, Scharpe S, Maes M. Effects of serotonin and serotonergic agonists and antagonists on the production of interferon-gamma and interleukin-10. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology. 2000;23:89-98

33. Kubera M, Lin AH, Kenis G, Bosmans E, van Bockstaele D, Maes M. Anti-inflammatory effects of antidepressants through suppression of the interferon-gamma/interleukin-10 production ratio. J Clin Psychopharmacol. 2001;21:199-206

34. Maes M, Song C, Lin AH, Bonaccorso S, Kenis G, De Jongh R, et al. Negative immunoregulatory effects of antidepressants: Inhibition of interferon-gamma and stimulation of interleukin-10 secretion. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology. 1999;20:370-379

91 35. Kenis G, Maes M. Effects of antidepressants on the production of cytokines. The international

journal of neuropsychopharmacology / official scientific journal of the Collegium Internationale Neuropsychopharmacologicum (CINP). 2002;5:401-412

36. Su JA, Chou SY, Tsai CS, Hung TH. Cytokine changes in the pathophysiology of poststroke depression. General hospital psychiatry. 2012;34:35-39

37. Basic Kes V, Simundic AM, Nikolac N, Topic E, Demarin V. Pro-inflammatory and anti-inflammatory cytokines in acute ischemic stroke and their relation to early neurological deficit and stroke outcome. Clinical biochemistry. 2008;41:1330-1334

38. Tang WK, Chen YK, Lu JY, Chu WC, Mok VC, Ungvari GS, et al. White matter hyperintensities in post-stroke depression: A case control study. Journal of neurology, neurosurgery, and psychiatry. 2010;81:1312-1315

39. Tang WK, Chen YK, Lu JY, Mok VC, Chu WC, Ungvari GS, et al. Frontal lobe atrophy in depression after stroke. Stroke research and treatment. 2013;2013:424769

40. Vataja R, Pohjasvaara T, Leppavuori A, Mantyla R, Aronen HJ, Salonen O, et al. Magnetic resonance imaging correlates of depression after ischemic stroke. Archives of general psychiatry. 2001;58:925-931

41. Nys GM, van Zandvoort MJ, van der Worp HB, de Haan EH, de Kort PL, Kappelle LJ. Early depressive symptoms after stroke: Neuropsychological correlates and lesion characteristics. J Neurol Sci. 2005;228:27-33

42. Terroni L, Amaro E, Iosifescu DV, Tinone G, Sato JR, Leite CC, et al. Stroke lesion in cortical neural circuits and post-stroke incidence of major depressive episode: A 4-month prospective study. The world journal of biological psychiatry : the official journal of the World Federation of Societies of Biological Psychiatry. 2011;12:539-548

43. Zhang T, Jing X, Zhao X, Wang C, Liu Z, Zhou Y, et al. A prospective cohort study of lesion location and its relation to post-stroke depression among chinese patients. Journal of affective disorders. 2012;136:e83-87

44. Francis PL, Herrmann N, Tennen G, Lanctôt KL. A brief history of poststroke depression neuroimaging. Aging Health. 2009;5:79-88

45. Vataja R, Leppavuori A, Pohjasvaara T, Mantyla R, Aronen HJ, Salonen O, et al. Poststroke depression and lesion location revisited. The Journal of neuropsychiatry and clinical neurosciences. 2004;16:156-162

46. Aben I, Lodder J, Honig A, Lousberg R, Boreas A, Verhey F. Focal or generalized vascular brain damage and vulnerability to depression after stroke: A 1-year prospective follow-up study. International psychogeriatrics / IPA. 2006;18:19-35

47. Provinciali L, Paolucci S, Torta R, Toso V, Gobbi B, Gandolfo C. Depression after first-ever ischemic stroke: The prognostic role of neuroanatomic subtypes in clinical practice. Cerebrovascular diseases (Basel, Switzerland). 2008;26:592-599

48. Pohjasvaara T, Mantyla R, Ylikoski R, Kaste M, Erkinjuntti T. Clinical features of mri-defined subcortical vascular disease. Alzheimer disease and associated disorders. 2003;17:236-242

49. Fu JH, Wong K, Mok V, Hu X, Xiong Y, Chen Y, et al. Neuroimaging predictors for depressive symptoms in cerebral small vessel disease. International journal of geriatric psychiatry. 2010;25:1039-1043

50. Maes M, Yirmyia R, Noraberg J, Brene S, Hibbeln J, Perini G, et al. The inflammatory & neurodegenerative (i&nd) hypothesis of depression: Leads for future research and new drug developments in depression. Metabolic brain disease. 2009;24:27-53

51. Moylan S, Maes M, Wray NR, Berk M. The neuroprogressive nature of major depressive disorder: Pathways to disease evolution and resistance, and therapeutic implications. Molecular psychiatry. 2013;18:595-606

52. Leonard BE, Myint A. Inflammation and depression: Is there a causal connection with dementia? Neurotoxicity research. 2006;10:149-160

92 53. Song C, Wang H. Cytokines mediated inflammation and decreased neurogenesis in animal

models of depression. Progress in neuro-psychopharmacology & biological psychiatry. 2011;35:760-768

54. Catena-Dell'Osso M, Bellantuono C, Consoli G, Baroni S, Rotella F, Marazziti D. Inflammatory and neurodegenerative pathways in depression: A new avenue for antidepressant development? Current medicinal chemistry. 2011;18:245-255

55. Kessing LV, Andersen PK. Does the risk of developing dementia increase with the number of episodes in patients with depressive disorder and in patients with bipolar disorder? Journal of neurology, neurosurgery, and psychiatry. 2004;75:1662-1666

56. Lampe IK, Hulshoff Pol HE, Janssen J, Schnack HG, Kahn RS, Heeren TJ. Association of depression duration with reduction of global cerebral gray matter volume in female patients with recurrent major depressive disorder. The American journal of psychiatry. 2003;160:2052-2054

57. Maes M, Berk M, Goehler L, Song C, Anderson G, Galecki P, et al. Depression and sickness behavior are janus-faced responses to shared inflammatory pathways. BMC Med. 2012;10:66

58. McKinnon MC, Yucel K, Nazarov A, MacQueen GM. A meta-analysis examining clinical predictors of hippocampal volume in patients with major depressive disorder. Journal of psychiatry & neuroscience : JPN. 2009;34:41-54

59. Videbech P, Ravnkilde B. Hippocampal volume and depression: A meta-analysis of mri studies. The American journal of psychiatry. 2004;161:1957-1966

60. MacQueen GM, Campbell S, McEwen BS, Macdonald K, Amano S, Joffe RT, et al. Course of illness, hippocampal function, and hippocampal volume in major depression. Proceedings of the National Academy of Sciences of the United States of America. 2003;100:1387-1392

61. Widner B, Leblhuber F, Fuchs D. Increased neopterin production and tryptophan degradation in advanced parkinson's disease. Journal of neural transmission (Vienna, Austria : 1996). 2002;109:181-189

62. Widner B, Leblhuber F, Walli J, Tilz GP, Demel U, Fuchs D. Tryptophan degradation and immune activation in alzheimer's disease. Journal of neural transmission (Vienna, Austria : 1996). 2000;107:343-353

63. Forrest CM, Mackay GM, Stoy N, Spiden SL, Taylor R, Stone TW, et al. Blood levels of kynurenines, interleukin-23 and soluble human leucocyte antigen-g at different stages of huntington's disease. Journal of neurochemistry. 2010;112:112-122

64. Guidetti P, Luthi-Carter RE, Augood SJ, Schwarcz R. Neostriatal and cortical quinolinate levels are increased in early grade huntington's disease. Neurobiology of disease. 2004;17:455-461

65. Guillemin GJ, Kerr SJ, Brew BJ. Involvement of quinolinic acid in aids dementia complex. Neurotoxicity research. 2005;7:103-123

66. Darlington LG, Mackay GM, Forrest CM, Stoy N, George C, Stone TW. Altered kynurenine metabolism correlates with infarct volume in stroke. The European journal of neuroscience. 2007;26:2211-2221

67. Leonard B, Maes M. Mechanistic explanations how cell-mediated immune activation, inflammation and oxidative and nitrosative stress pathways and their sequels and concomitants play a role in the pathophysiology of unipolar depression. Neuroscience and biobehavioral reviews. 2012;36:764-785

68. Berk M, Kapczinski F, Andreazza AC, Dean OM, Giorlando F, Maes M, et al. Pathways underlying neuroprogression in bipolar disorder: Focus on inflammation, oxidative stress and neurotrophic factors. Neuroscience and biobehavioral reviews. 2011;35:804-817

69. Smith AJ, Smith RA, Stone TW. 5-hydroxyanthranilic acid, a tryptophan metabolite, generates oxidative stress and neuronal death via p38 activation in cultured cerebellar granule neurones. Neurotoxicity research. 2009;15:303-310

70. Guidetti P, Schwarcz R. 3-hydroxykynurenine potentiates quinolinate but not nmda toxicity in the rat striatum. The European journal of neuroscience. 1999;11:3857-3863

93 71. Goldstein LE, Leopold MC, Huang X, Atwood CS, Saunders AJ, Hartshorn M, et al. 3-

hydroxykynurenine and 3-hydroxyanthranilic acid generate hydrogen peroxide and promote alpha-crystallin cross-linking by metal ion reduction. Biochemistry. 2000;39:7266-7275

72. Rios C, Santamaria A. Quinolinic acid is a potent lipid peroxidant in rat brain homogenates. Neurochemical research. 1991;16:1139-1143

73. Schuck PF, Tonin A, da Costa Ferreira G, Viegas CM, Latini A, Duval Wannmacher CM, et al. Kynurenines impair energy metabolism in rat cerebral cortex. Cellular and molecular neurobiology. 2007;27:147-160

74. Baran H, Staniek K, Kepplinger B, Stur J, Draxler M, Nohl H. Kynurenines and the respiratory parameters on rat heart mitochondria. Life sciences. 2003;72:1103-1115

75. Schwarcz R, Whetsell WO, Jr., Mangano RM. Quinolinic acid: An endogenous metabolite that produces axon-sparing lesions in rat brain. Science (New York, N.Y.). 1983;219:316-318

76. Maes M, Leonard BE, Myint AM, Kubera M, Verkerk R. The new '5-ht' hypothesis of depression: Cell-mediated immune activation induces indoleamine 2,3-dioxygenase, which leads to lower plasma tryptophan and an increased synthesis of detrimental tryptophan catabolites (trycats), both of which contribute to the onset of depression. Progress in neuro-psychopharmacology & biological psychiatry. 2011;35:702-721

77. Khaspekov L, Kida E, Victorov I, Mossakowski MJ. Neurotoxic effect induced by quinolinic acid in dissociated cell culture of mouse hippocampus. Journal of neuroscience research. 1989;22:150-157

78. Barone FC, Feuerstein GZ. Inflammatory mediators and stroke: New opportunities for novel therapeutics. Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism. 1999;19:819-834

79. Emsley HC, Smith CJ, Gavin CM, Georgiou RF, Vail A, Barberan EM, et al. An early and sustained peripheral inflammatory response in acute ischaemic stroke: Relationships with infection and atherosclerosis. Journal of neuroimmunology. 2003;139:93-101

80. Lindsberg PJ, Grau AJ. Inflammation and infections as risk factors for ischemic stroke. Stroke; a journal of cerebral circulation. 2003;34:2518-2532

81. Babcock TA, Carlin JM. Transcriptional activation of indoleamine dioxygenase by interleukin 1 and tumor necrosis factor alpha in interferon-treated epithelial cells. Cytokine. 2000;12:588-594

82. Myint AM, Kim YK, Verkerk R, Park SH, Scharpe S, Steinbusch HW, et al. Tryptophan breakdown pathway in bipolar mania. Journal of affective disorders. 2007;102:65-72

83. Miller CL, Llenos IC, Dulay JR, Weis S. Upregulation of the initiating step of the kynurenine pathway in postmortem anterior cingulate cortex from individuals with schizophrenia and bipolar disorder. Brain research. 2006;1073-1074:25-37

84. Schwarcz R, Rassoulpour A, Wu HQ, Medoff D, Tamminga CA, Roberts RC. Increased cortical kynurenate content in schizophrenia. Biological psychiatry. 2001;50:521-530

85. Erhardt S, Blennow K, Nordin C, Skogh E, Lindstrom LH, Engberg G. Kynurenic acid levels are elevated in the cerebrospinal fluid of patients with schizophrenia. Neuroscience letters. 2001;313:96-98

86. Condray R, Dougherty GG, Jr., Keshavan MS, Reddy RD, Haas GL, Montrose DM, et al. 3-hydroxykynurenine and clinical symptoms in first-episode neuroleptic-naive patients with schizophrenia. The international journal of neuropsychopharmacology / official scientific journal of the Collegium Internationale Neuropsychopharmacologicum (CINP). 2011;14:756-767

87. Myint AM, Schwarz MJ, Verkerk R, Mueller HH, Zach J, Scharpe S, et al. Reversal of imbalance between kynurenic acid and 3-hydroxykynurenine by antipsychotics in medication-naive and medication-free schizophrenic patients. Brain, behavior, and immunity. 2011;25:1576-1581

94 88. Myint AM, Kim YK, Verkerk R, Scharpe S, Steinbusch H, Leonard B. Kynurenine pathway in

major depression: Evidence of impaired neuroprotection. Journal of affective disorders. 2007;98:143-151

89. Steiner J, Walter M, Gos T, Guillemin GJ, Bernstein HG, Sarnyai Z, et al. Severe depression is associated with increased microglial quinolinic acid in subregions of the anterior cingulate gyrus: Evidence for an immune-modulated glutamatergic neurotransmission? Journal of neuroinflammation. 2011;8:94

90. Gabbay V, Liebes L, Katz Y, Liu S, Mendoza S, Babb JS, et al. The kynurenine pathway in adolescent depression: Preliminary findings from a proton mr spectroscopy study. Progress in neuro-psychopharmacology & biological psychiatry. 2010;34:37-44

91. Riedel WJ, Klaassen T, Schmitt JA. Tryptophan, mood, and cognitive function. Brain, behavior, and immunity. 2002;16:581-589

92. Swardfager W, Herrmann N, Dowlati Y, Oh PI, Kiss A, Walker SE, et al. Indoleamine 2,3-dioxygenase activation and depressive symptoms in patients with coronary artery disease. Psychoneuroendocrinology. 2009;34:1560-1566

93. Ruddick JP, Evans AK, Nutt DJ, Lightman SL, Rook GA, Lowry CA. Tryptophan metabolism in the central nervous system: Medical implications. Expert reviews in molecular medicine. 2006;8:1-27

94. Barton DA, Esler MD, Dawood T, Lambert EA, Haikerwal D, Brenchley C, et al. Elevated brain serotonin turnover in patients with depression: Effect of genotype and therapy. Archives of general psychiatry. 2008;65:38-46

95. Hou C, Jia F, Liu Y, Li L. Csf serotonin, 5-hydroxyindolacetic acid and neuropeptide y levels in severe major depressive disorder. Brain research. 2006;1095:154-158

96. Ruhe HG, Mason NS, Schene AH. Mood is indirectly related to serotonin, norepinephrine and dopamine levels in humans: A meta-analysis of monoamine depletion studies. Molecular psychiatry. 2007;12:331-359

97. Smith KA, Fairburn CG, Cowen PJ. Relapse of depression after rapid depletion of tryptophan. Lancet. 1997;349:915-919

98. Guillemin GJ. Quinolinic acid, the inescapable neurotoxin. The FEBS journal. 2012;279:1356-1365

99. Tarkowski E, Rosengren L, Blomstrand C, Wikkelso C, Jensen C, Ekholm S, et al. Early intrathecal production of interleukin-6 predicts the size of brain lesion in stroke. Stroke; a journal of cerebral circulation. 1995;26:1393-1398

100. Kostulas N, Pelidou SH, Kivisakk P, Kostulas V, Link H. Increased il-1beta, il-8, and il-17 mrna expression in blood mononuclear cells observed in a prospective ischemic stroke study. Stroke; a journal of cerebral circulation. 1999;30:2174-2179

101. Vila N, Castillo J, Davalos A, Chamorro A. Proinflammatory cytokines and early neurological worsening in ischemic stroke. Stroke; a journal of cerebral circulation. 2000;31:2325-2329

102. Perini F, Morra M, Alecci M, Galloni E, Marchi M, Toso V. Temporal profile of serum anti-inflammatory and pro-inflammatory interleukins in acute ischemic stroke patients. Neurological sciences : official journal of the Italian Neurological Society and of the Italian Society of Clinical Neurophysiology. 2001;22:289-296

103. Waje-Andreassen U, Krakenes J, Ulvestad E, Thomassen L, Myhr KM, Aarseth J, et al. Il-6: An early marker for outcome in acute ischemic stroke. Acta neurologica Scandinavica. 2005;111:360-365

104. Beamer NB, Coull BM, Clark WM, Briley DP, Wynn M, Sexton G. Persistent inflammatory response in stroke survivors. Neurology. 1998;50:1722-1728

105. Mandi Y, Vecsei L. The kynurenine system and immunoregulation. Journal of neural transmission (Vienna, Austria : 1996). 2012;119:197-209

106. Raison CL, Dantzer R, Kelley KW, Lawson MA, Woolwine BJ, Vogt G, et al. Csf concentrations of brain tryptophan and kynurenines during immune stimulation with ifn-alpha: Relationship to cns immune responses and depression. Molecular psychiatry. 2010;15:393-403

95 107. Kim JM, Stewart R, Kim SW, Shin IS, Kim JT, Park MS, et al. Associations of cytokine gene

polymorphisms with post-stroke depression. The world journal of biological psychiatry : the official journal of the World Federation of Societies of Biological Psychiatry. 2012;13:579-587

108. Johansen HL, Wielgosz AT, Nguyen K, Fry RN. Incidence, comorbidity, case fatality and readmission of hospitalized stroke patients in canada. Can J Cardiol. 2006;22:65-71

109. Gottlieb D, Salagnik I, Kipnis M, Brill S. Post stroke depression, first year post stroke, in middle band patients. International journal of geriatric psychiatry. 2002;17:486-487

110. Ayerbe L, Ayis S, Wolfe CD, Rudd AG. Natural history, predictors and outcomes of depression after stroke: Systematic review and meta-analysis. The British journal of psychiatry : the journal of mental science. 2013;202:14-21

111. Ramasubbu R, Kennedy SH. Factors complicating the diagnosis of depression in cerebrovascular disease, part i--phenomenological and nosological issues. Canadian journal of psychiatry. Revue canadienne de psychiatrie. 1994;39:596-600

112. Primeau F. Post-stroke depression: A critical review of the literature. Canadian journal of psychiatry. Revue canadienne de psychiatrie. 1988;33:757-765

113. Cohen-Cole SA, Stoudemire A. Major depression and physical illness. Special considerations in diagnosis and biologic treatment. Psychiatr Clin North Am. 1987;10:1-17

114. Okada K, Kobayashi S, Yamagata S, Takahashi K, Yamaguchi S. Poststroke apathy and regional cerebral blood flow. Stroke; a journal of cerebral circulation. 1997;28:2437-2441

115. Yamagata S, Yamaguchi S, Kobayashi S. Impaired novelty processing in apathy after subcortical stroke. Stroke; a journal of cerebral circulation. 2004;35:1935-1940

116. Starkstein SE, Fedoroff JP, Price TR, Leiguarda R, Robinson RG. Apathy following cerebrovascular lesions. Stroke; a journal of cerebral circulation. 1993;24:1625-1630

117. Hama S, Yamashita H, Shigenobu M, Watanabe A, Kurisu K, Yamawaki S, et al. Post-stroke affective or apathetic depression and lesion location: Left frontal lobe and bilateral basal ganglia. European archives of psychiatry and clinical neuroscience. 2007;257:149-152

118. Diagnostic and statistical manual of mental disorders dsm-iv-tr fourth edition (text revision). Amer Psychiatric Pub; 2000.

119. Spalletta G, Ripa A, Caltagirone C. Symptom profile of dsm-iv major and minor depressive disorders in first-ever stroke patients. The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry. 2005;13:108-115

120. Morris PL, Robinson RG, Andrzejewski P, Samuels J, Price TR. Association of depression with 10-year poststroke mortality. The American journal of psychiatry. 1993;150:124-129

121. Paradiso S, Ohkubo T, Robinson RG. Vegetative and psychological symptoms associated with depressed mood over the first two years after stroke. International journal of psychiatry in medicine. 1997;27:137-157

122. Astrom M, Adolfsson R, Asplund K. Major depression in stroke patients. A 3-year longitudinal study. Stroke; a journal of cerebral circulation. 1993;24:976-982

123. Wade DT, Legh-Smith J, Hewer RA. Depressed mood after stroke. A community study of its frequency. The British journal of psychiatry : the journal of mental science. 1987;151:200-205

124. Robinson RG, Starr LB, Price TR. A two year longitudinal study of mood disorders following stroke. Prevalence and duration at six months follow-up. The British journal of psychiatry : the journal of mental science. 1984;144:256-262

125. Singh A, Black SE, Herrmann N, Leibovitch FS, Ebert PL, Lawrence J, et al. Functional and neuroanatomic correlations in poststroke depression: The sunnybrook stroke study. Stroke; a journal of cerebral circulation. 2000;31:637-644

126. Cully JA, Gfeller JD, Heise RA, Ross MJ, Teal CR, Kunik ME. Geriatric depression, medical diagnosis, and functional recovery during acute rehabilitation. Archives of physical medicine and rehabilitation. 2005;86:2256-2260

127. Sharpe M, Hawton K, Seagroatt V, Bamford J, House A, Molyneux A, et al. Depressive disorders in long-term survivors of stroke. Associations with demographic and social factors,

96

functional status, and brain lesion volume. The British journal of psychiatry : the journal of mental science. 1994;164:380-386

128. Paolucci S, Antonucci G, Grasso MG, Morelli D, Troisi E, Coiro P, et al. Post-stroke depression, antidepressant treatment and rehabilitation results. A case-control study. Cerebrovascular diseases (Basel, Switzerland). 2001;12:264-271

129. Kotila M, Numminen H, Waltimo O, Kaste M. Depression after stroke: Results of the finnstroke study. Stroke; a journal of cerebral circulation. 1998;29:368-372

130. Andersen G, Vestergaard K, Ingemann-Nielsen M, Lauritzen L. Risk factors for post-stroke depression. Acta psychiatrica Scandinavica. 1995;92:193-198

131. Pohjasvaara T, Leppavuori A, Siira I, Vataja R, Kaste M, Erkinjuntti T. Frequency and clinical determinants of poststroke depression. Stroke; a journal of cerebral circulation. 1998;29:2311-2317

132. Paradiso S, Robinson RG. Gender differences in poststroke depression. The Journal of neuropsychiatry and clinical neurosciences. 1998;10:41-47

133. Aben I, Denollet J, Lousberg R, Verhey F, Wojciechowski F, Honig A. Personality and vulnerability to depression in stroke patients: A 1-year prospective follow-up study. Stroke; a journal of cerebral circulation. 2002;33:2391-2395

134. Robinson RG, Bolla-Wilson K, Kaplan E, Lipsey JR, Price TR. Depression influences intellectual impairment in stroke patients. The British journal of psychiatry : the journal of mental science. 1986;148:541-547

135. Vataja R, Pohjasvaara T, Mantyla R, Ylikoski R, Leskela M, Kalska H, et al. Depression-executive dysfunction syndrome in stroke patients. The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry. 2005;13:99-107

136. Downhill JE, Jr., Robinson RG. Longitudinal assessment of depression and cognitive impairment following stroke. J Nerv Ment Dis. 1994;182:425-431

137. Nys GM, van Zandvoort MJ, van der Worp HB, de Haan EH, de Kort PL, Jansen BP, et al. Early cognitive impairment predicts long-term depressive symptoms and quality of life after stroke. J Neurol Sci. 2006;247:149-156

138. Robinson RG, Kubos KL, Starr LB, Rao K, Price TR. Mood disorders in stroke patients. Importance of location of lesion. Brain : a journal of neurology. 1984;107 ( Pt 1):81-93

139. Robinson RG, Starr LB, Lipsey JR, Rao K, Price TR. A two-year longitudinal study of post-stroke mood disorders: Dynamic changes in associated variables over the first six months of follow-up. Stroke; a journal of cerebral circulation. 1984;15:510-517

140. Robinson RG, Szetela B. Mood change following left hemispheric brain injury. Ann Neurol. 1981;9:447-453

141. Starkstein SE, Robinson RG, Price TR. Comparison of cortical and subcortical lesions in the production of poststroke mood disorders. Brain : a journal of neurology. 1987;110 ( Pt 4):1045-1059

142. Starkstein SE, Robinson RG, Berthier ML, Parikh RM, Price TR. Differential mood changes following basal ganglia vs thalamic lesions. Archives of neurology. 1988;45:725-730

143. Schwartz JA, Speed NM, Brunberg JA, Brewer TL, Brown M, Greden JF. Depression in stroke rehabilitation. Biological psychiatry. 1993;33:694-699

144. Morris PL, Robinson RG, de Carvalho ML, Albert P, Wells JC, Samuels JF, et al. Lesion characteristics and depressed mood in the stroke data bank study. The Journal of neuropsychiatry and clinical neurosciences. 1996;8:153-159

145. Lipsey JR, Robinson RG, Pearlson GD, Rao K, Price TR. Nortriptyline treatment of post-stroke depression: A double-blind study. Lancet. 1984;1:297-300

146. Reding MJ, Orto LA, Winter SW, Fortuna IM, Di Ponte P, McDowell FH. Antidepressant therapy after stroke. A double-blind trial. Archives of neurology. 1986;43:763-765

97 147. Andersen G, Vestergaard K, Lauritzen L. Effective treatment of poststroke depression with the

selective serotonin reuptake inhibitor citalopram. Stroke; a journal of cerebral circulation. 1994;25:1099-1104

148. Lauritzen L, Bendsen BB, Vilmar T, Bendsen EB, Lunde M, Bech P. Post-stroke depression: Combined treatment with imipramine or desipramine and mianserin. A controlled clinical study. Psychopharmacology. 1994;114:119-122

149. Lazarus LW, Moberg PJ, Langsley PR, Lingam VR. Methylphenidate and nortriptyline in the treatment of poststroke depression: A retrospective comparison. Archives of physical medicine and rehabilitation. 1994;75:403-406

150. Dam M, Tonin P, De Boni A, Pizzolato G, Casson S, Ermani M, et al. Effects of fluoxetine and maprotiline on functional recovery in poststroke hemiplegic patients undergoing rehabilitation therapy. Stroke; a journal of cerebral circulation. 1996;27:1211-1214

151. Stamenkovic M, Schindler S, Kasper S. [therapy of post-stroke depression with fluoxetine. A pilot project]. Nervenarzt. 1996;67:62-67

152. Dahmen N, Marx J, Hopf HC, Tettenborn B, Roder R. Therapy of early poststroke depression with venlafaxine: Safety, tolerability, and efficacy as determined in an open, uncontrolled clinical trial. Stroke; a journal of cerebral circulation. 1999;30:691-692

153. Robinson RG, Schultz SK, Castillo C, Kopel T, Kosier JT, Newman RM, et al. Nortriptyline versus fluoxetine in the treatment of depression and in short-term recovery after stroke: A placebo-controlled, double-blind study. The American journal of psychiatry. 2000;157:351-359

154. Wiart L, Petit H, Joseph PA, Mazaux JM, Barat M. Fluoxetine in early poststroke depression: A double-blind placebo-controlled study. Stroke; a journal of cerebral circulation. 2000;31:1829-1832

155. Jorge RE, Robinson RG, Arndt S, Starkstein S. Mortality and poststroke depression: A placebo-controlled trial of antidepressants. The American journal of psychiatry. 2003;160:1823-1829

156. Murray V, von Arbin M, Bartfai A, Berggren AL, Landtblom AM, Lundmark J, et al. Double-blind comparison of sertraline and placebo in stroke patients with minor depression and less severe major depression. The Journal of clinical psychiatry. 2005;66:708-716

157. Yamakawa Y, Satoh S, Sawa S, Ohta H, Asada T. Efficacy of milnacipran on poststroke depression on inpatient rehabilitation. Psychiatry and clinical neurosciences. 2005;59:705-710

158. Johnson ML, Roberts MD, Ross AR, Witten CM. Methylphenidate in stroke patients with depression. Am J Phys Med Rehabil. 1992;71:239-241

159. Lazarus LW, Winemiller DR, Lingam VR, Neyman I, Hartman C, Abassian M, et al. Efficacy and side effects of methylphenidate for poststroke depression. The Journal of clinical psychiatry. 1992;53:447-449

160. Lingam VR, Lazarus LW, Groves L, Oh SH. Methylphenidate in treating poststroke depression. The Journal of clinical psychiatry. 1988;49:151-153

161. Kimura M, Kanetani K, Imai R, Suzuki H, Isayama K, Endo S. Therapeutic effects of milnacipran, a serotonin and noradrenaline reuptake inhibitor, on post-stroke depression. Int Clin Psychopharmacol. 2002;17:121-125

162. Narushima K, Paradiso S, Moser DJ, Jorge R, Robinson RG. Effect of antidepressant therapy on executive function after stroke. The British journal of psychiatry : the journal of mental science. 2007;190:260-265

163. Rasmussen A, Christensen J, Clemmensen PM, Dalsgaard NJ, Dam H, Hindberg I, et al. Platelet serotonin transporter in stroke patients. Acta neurologica Scandinavica. 2003;107:150-153

164. Niedermaier N, Bohrer E, Schulte K, Schlattmann P, Heuser I. Prevention and treatment of poststroke depression with mirtazapine in patients with acute stroke. The Journal of clinical psychiatry. 2004;65:1619-1623

165. Robinson RG, Jorge RE, Moser DJ, Acion L, Solodkin A, Small SL, et al. Escitalopram and problem-solving therapy for prevention of poststroke depression: A randomized controlled trial. JAMA. 2008;299:2391-2400

98 166. Palomaki H, Kaste M, Berg A, Lonnqvist R, Lonnqvist J, Lehtihalmes M, et al. Prevention of

poststroke depression: 1 year randomised placebo controlled double blind trial of mianserin with 6 month follow up after therapy. Journal of neurology, neurosurgery, and psychiatry. 1999;66:490-494

167. Almeida OP, Waterreus A, Hankey GJ. Preventing depression after stroke: Results from a randomized placebo-controlled trial. The Journal of clinical psychiatry. 2006;67:1104-1109

168. Chen Y, Patel NC, Guo JJ, Zhan S. Antidepressant prophylaxis for poststroke depression: A meta-analysis. Int Clin Psychopharmacol. 2007;22:159-166

169. Hackett ML, Anderson CS, House A, Xia J. Interventions for treating depression after stroke. Cochrane Database Syst Rev. 2008:CD003437

170. Starkstein SE, Mizrahi R, Power BD. Antidepressant therapy in post-stroke depression. Expert opinion on pharmacotherapy. 2008;9:1291-1298

171. Swenson JR, Doucette S, Fergusson D. Adverse cardiovascular events in antidepressant trials involving high-risk patients: A systematic review of randomized trials. Canadian journal of psychiatry. Revue canadienne de psychiatrie. 2006;51:923-929

172. French DD, Werner DC, Campbell RR, Powell-Cope GM, Nelson AL, Rubenstein LZ, et al. A multivariate fall risk assessment model for vha nursing homes using the minimum data set. J Am Med Dir Assoc. 2007;8:115-122

173. Smoller JW, Allison M, Cochrane BB, Curb JD, Perlis RH, Robinson JG, et al. Antidepressant use and risk of incident cardiovascular morbidity and mortality among postmenopausal women in the women's health initiative study. Arch Intern Med. 2009;169:2128-2139

174. Clark MS, Rubenach S, Winsor A. A randomized controlled trial of an education and counselling intervention for families after stroke. Clin Rehabil. 2003;17:703-712

175. Claiborne N. Effectiveness of a care coordination model for stroke survivors: A randomized study. Health Soc Work. 2006;31:87-96

176. Burton C, Gibbon B. Expanding the role of the stroke nurse: A pragmatic clinical trial. J Adv Nurs. 2005;52:640-650

177. Lipton P. Ischemic cell death in brain neurons. Physiological reviews. 1999;79:1431-1568 178. Moskowitz MA, Lo EH, Iadecola C. The science of stroke: Mechanisms in search of

treatments. Neuron. 2010;67:181-198 179. Iadecola C, Anrather J. The immunology of stroke: From mechanisms to translation. Nat Med.

2011;17:796-808 180. Kono H, Rock KL. How dying cells alert the immune system to danger. Nature reviews.

Immunology. 2008;8:279-289 181. Konoeda F, Shichita T, Yoshida H, Sugiyama Y, Muto G, Hasegawa E, et al. Therapeutic effect

of il-12/23 and their signaling pathway blockade on brain ischemia model. Biochemical and biophysical research communications. 2010;402:500-506

182. Brea D, Blanco M, Sobrino T, Ramos-Cabrer P, Castillo J. [the levels of expression of toll-like receptors 2 and 4 in neutrophils are associated with the prognosis of ischaemic stroke patients]. Rev Neurol. 2011;52:12-19

183. McColl BW, Rothwell NJ, Allan SM. Systemic inflammatory stimulus potentiates the acute phase and cxc chemokine responses to experimental stroke and exacerbates brain damage via interleukin-1- and neutrophil-dependent mechanisms. The Journal of neuroscience : the official journal of the Society for Neuroscience. 2007;27:4403-4412

184. Yilmaz G, Arumugam TV, Stokes KY, Granger DN. Role of t lymphocytes and interferon-gamma in ischemic stroke. Circulation. 2006;113:2105-2112

185. Swardfager W, Winer DA, Herrmann N, Winer S, Lanctot KL. Interleukin-17 in post-stroke neurodegeneration. Neuroscience and biobehavioral reviews. 2013;37:436-447

186. Marsh BJ, Williams-Karnesky RL, Stenzel-Poore MP. Toll-like receptor signaling in endogenous neuroprotection and stroke. Neuroscience. 2009;158:1007-1020

187. Smith CJ, Emsley HC, Gavin CM, Georgiou RF, Vail A, Barberan EM, et al. Peak plasma interleukin-6 and other peripheral markers of inflammation in the first week of ischaemic stroke

99

correlate with brain infarct volume, stroke severity and long-term outcome. BMC neurology. 2004;4:2

188. Zaremba J, Losy J. Early tnf-alpha levels correlate with ischaemic stroke severity. Acta neurologica Scandinavica. 2001;104:288-295

189. Welsh P, Lowe GD, Chalmers J, Campbell DJ, Rumley A, Neal BC, et al. Associations of proinflammatory cytokines with the risk of recurrent stroke. Stroke; a journal of cerebral circulation. 2008;39:2226-2230

190. Spera PA, Ellison JA, Feuerstein GZ, Barone FC. Il-10 reduces rat brain injury following focal stroke. Neurosci Lett. 1998;251:189-192

191. de Bilbao F, Arsenijevic D, Moll T, Garcia-Gabay I, Vallet P, Langhans W, et al. In vivo over-expression of interleukin-10 increases resistance to focal brain ischemia in mice. Journal of neurochemistry. 2009;110:12-22

192. Kipnis J, Avidan H, Caspi RR, Schwartz M. Dual effect of cd4+cd25+ regulatory t cells in neurodegeneration: A dialogue with microglia. Proceedings of the National Academy of Sciences of the United States of America. 2004;101 Suppl 2:14663-14669

193. Vila N, Castillo J, Davalos A, Esteve A, Planas AM, Chamorro A. Levels of anti-inflammatory cytokines and neurological worsening in acute ischemic stroke. Stroke; a journal of cerebral circulation. 2003;34:671-675

194. Jensen HA, Loukogeorgakis S, Yannopoulos F, Rimpilainen E, Petzold A, Tuominen H, et al. Remote ischemic preconditioning protects the brain against injury after hypothermic circulatory arrest. Circulation. 2011;123:714-721

195. Marsh B, Stevens SL, Packard AE, Gopalan B, Hunter B, Leung PY, et al. Systemic lipopolysaccharide protects the brain from ischemic injury by reprogramming the response of the brain to stroke: A critical role for irf3. The Journal of neuroscience : the official journal of the Society for Neuroscience. 2009;29:9839-9849

196. Kunz A, Park L, Abe T, Gallo EF, Anrather J, Zhou P, et al. Neurovascular protection by ischemic tolerance: Role of nitric oxide and reactive oxygen species. The Journal of neuroscience : the official journal of the Society for Neuroscience. 2007;27:7083-7093

197. Blondeau N, Widmann C, Lazdunski M, Heurteaux C. Activation of the nuclear factor-kappab is a key event in brain tolerance. The Journal of neuroscience : the official journal of the Society for Neuroscience. 2001;21:4668-4677

198. Ohtsuki T, Ruetzler CA, Tasaki K, Hallenbeck JM. Interleukin-1 mediates induction of tolerance to global ischemia in gerbil hippocampal ca1 neurons. Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism. 1996;16:1137-1142

199. Pradillo JM, Romera C, Hurtado O, Cardenas A, Moro MA, Leza JC, et al. Tnfr1 upregulation mediates tolerance after brain ischemic preconditioning. Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism. 2005;25:193-203

200. Myint AM, Leonard BE, Steinbusch HW, Kim YK. Th1, th2, and th3 cytokine alterations in major depression. Journal of affective disorders. 2005;88:167-173

201. Dantzer R. Cytokine-induced sickness behavior: Mechanisms and implications. Annals of the New York Academy of Sciences. 2001;933:222-234

202. Dantzer R. Cytokine-induced sickness behavior: Where do we stand? Brain, behavior, and immunity. 2001;15:7-24

203. van Dam AM, Brouns M, Louisse S, Berkenbosch F. Appearance of interleukin-1 in macrophages and in ramified microglia in the brain of endotoxin-treated rats: A pathway for the induction of non-specific symptoms of sickness? Brain research. 1992;588:291-296

204. Quan N, Stern EL, Whiteside MB, Herkenham M. Induction of pro-inflammatory cytokine mrnas in the brain after peripheral injection of subseptic doses of lipopolysaccharide in the rat. Journal of neuroimmunology. 1999;93:72-80

100 205. Heyen JR, Ye S, Finck BN, Johnson RW. Interleukin (il)-10 inhibits il-6 production in

microglia by preventing activation of nf-kappab. Brain research. Molecular brain research. 2000;77:138-147

206. Bluthe RM, Castanon N, Pousset F, Bristow A, Ball C, Lestage J, et al. Central injection of il-10 antagonizes the behavioural effects of lipopolysaccharide in rats. Psychoneuroendocrinology. 1999;24:301-311

207. Laye S, Parnet P, Goujon E, Dantzer R. Peripheral administration of lipopolysaccharide induces the expression of cytokine transcripts in the brain and pituitary of mice. Brain research. Molecular brain research. 1994;27:157-162

208. Strle K, McCusker RH, Tran L, King A, Johnson RW, Freund GG, et al. Novel activity of an anti-inflammatory cytokine: Il-10 prevents tnfalpha-induced resistance to igf-i in myoblasts. Journal of neuroimmunology. 2007;188:48-55

209. Dantzer R, O'Connor JC, Freund GG, Johnson RW, Kelley KW. From inflammation to sickness and depression: When the immune system subjugates the brain. Nat Rev Neurosci. 2008;9:46-56

210. Ye SM, Johnson RW. Increased interleukin-6 expression by microglia from brain of aged mice. Journal of neuroimmunology. 1999;93:139-148

211. Ye SM, Johnson RW. An age-related decline in interleukin-10 may contribute to the increased expression of interleukin-6 in brain of aged mice. Neuroimmunomodulation. 2001;9:183-192

212. Leon LR, Kozak W, Rudolph K, Kluger MJ. An antipyretic role for interleukin-10 in lps fever in mice. Am J Physiol. 1999;276:R81-89

213. Huang Y, Henry CJ, Dantzer R, Johnson RW, Godbout JP. Exaggerated sickness behavior and brain proinflammatory cytokine expression in aged mice in response to intracerebroventricular lipopolysaccharide. Neurobiology of aging. 2008;29:1744-1753

214. Raison CL, Capuron L, Miller AH. Cytokines sing the blues: Inflammation and the pathogenesis of depression. Trends Immunol. 2006;27:24-31

215. Alesci S, Martinez PE, Kelkar S, Ilias I, Ronsaville DS, Listwak SJ, et al. Major depression is associated with significant diurnal elevations in plasma interleukin-6 levels, a shift of its circadian rhythm, and loss of physiological complexity in its secretion: Clinical implications. The Journal of clinical endocrinology and metabolism. 2005;90:2522-2530

216. Miller GE, Stetler CA, Carney RM, Freedland KE, Banks WA. Clinical depression and inflammatory risk markers for coronary heart disease. The American journal of cardiology. 2002;90:1279-1283

217. Bouhuys AL, Flentge F, Oldehinkel AJ, van den Berg MD. Potential psychosocial mechanisms linking depression to immune function in elderly subjects. Psychiatry research. 2004;127:237-245

218. Kahl KG, Rudolf S, Stoeckelhuber BM, Dibbelt L, Gehl HB, Markhof K, et al. Bone mineral density, markers of bone turnover, and cytokines in young women with borderline personality disorder with and without comorbid major depressive disorder. The American journal of psychiatry. 2005;162:168-174

219. Tiemeier H, Hofman A, van Tuijl HR, Kiliaan AJ, Meijer J, Breteler MM. Inflammatory proteins and depression in the elderly. Epidemiology. 2003;14:103-107

220. Schlatter J, Ortuno F, Cervera-Enguix S. Monocytic parameters in patients with dysthymia versus major depression. Journal of affective disorders. 2004;78:243-247

221. Danner M, Kasl SV, Abramson JL, Vaccarino V. Association between depression and elevated c-reactive protein. Psychosomatic medicine. 2003;65:347-356

222. Ford DE, Erlinger TP. Depression and c-reactive protein in us adults: Data from the third national health and nutrition examination survey. Arch Intern Med. 2004;164:1010-1014

223. Maes M. Major depression and activation of the inflammatory response system. Advances in experimental medicine and biology. 1999;461:25-46

224. Levine J, Barak Y, Chengappa KN, Rapoport A, Rebey M, Barak V. Cerebrospinal cytokine levels in patients with acute depression. Neuropsychobiology. 1999;40:171-176

101 225. Lanquillon S, Krieg JC, Bening-Abu-Shach U, Vedder H. Cytokine production and treatment

response in major depressive disorder. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology. 2000;22:370-379

226. Rajagopalan S, Brook R, Rubenfire M, Pitt E, Young E, Pitt B. Abnormal brachial artery flow-mediated vasodilation in young adults with major depression. The American journal of cardiology. 2001;88:196-198, A197

227. Sluzewska A, Sobieska M, Rybakowski JK. Changes in acute-phase proteins during lithium potentiation of antidepressants in refractory depression. Neuropsychobiology. 1997;35:123-127

228. Maes M, Bosmans E, De Jongh R, Kenis G, Vandoolaeghe E, Neels H. Increased serum il-6 and il-1 receptor antagonist concentrations in major depression and treatment resistant depression. Cytokine. 1997;9:853-858

229. Mikova O, Yakimova R, Bosmans E, Kenis G, Maes M. Increased serum tumor necrosis factor alpha concentrations in major depression and multiple sclerosis. Eur Neuropsychopharmacol. 2001;11:203-208

230. Tuglu C, Kara SH, Caliyurt O, Vardar E, Abay E. Increased serum tumor necrosis factor-alpha levels and treatment response in major depressive disorder. Psychopharmacology. 2003;170:429-433

231. Hestad KA, Tonseth S, Stoen CD, Ueland T, Aukrust P. Raised plasma levels of tumor necrosis factor alpha in patients with depression: Normalization during electroconvulsive therapy. J ECT. 2003;19:183-188

232. Owen BM, Eccleston D, Ferrier IN, Young AH. Raised levels of plasma interleukin-1beta in major and postviral depression. Acta psychiatrica Scandinavica. 2001;103:226-228

233. Thomas AJ, Davis S, Morris C, Jackson E, Harrison R, O'Brien JT. Increase in interleukin-1beta in late-life depression. The American journal of psychiatry. 2005;162:175-177

234. Yu YW, Chen TJ, Hong CJ, Chen HM, Tsai SJ. Association study of the interleukin-1 beta (c-511t) genetic polymorphism with major depressive disorder, associated symptomatology, and antidepressant response. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology. 2003;28:1182-1185

235. Jun TY, Pae CU, Hoon H, Chae JH, Bahk WM, Kim KS, et al. Possible association between -g308a tumour necrosis factor-alpha gene polymorphism and major depressive disorder in the korean population. Psychiatric genetics. 2003;13:179-181

236. Suarez EC, Lewis JG, Krishnan RR, Young KH. Enhanced expression of cytokines and chemokines by blood monocytes to in vitro lipopolysaccharide stimulation are associated with hostility and severity of depressive symptoms in healthy women. Psychoneuroendocrinology. 2004;29:1119-1128

237. Musselman DL, Miller AH, Porter MR, Manatunga A, Gao F, Penna S, et al. Higher than normal plasma interleukin-6 concentrations in cancer patients with depression: Preliminary findings. The American journal of psychiatry. 2001;158:1252-1257

238. Bower JE, Ganz PA, Aziz N, Fahey JL. Fatigue and proinflammatory cytokine activity in breast cancer survivors. Psychosomatic medicine. 2002;64:604-611

239. Meyers CA, Albitar M, Estey E. Cognitive impairment, fatigue, and cytokine levels in patients with acute myelogenous leukemia or myelodysplastic syndrome. Cancer. 2005;104:788-793

240. Lesperance F, Frasure-Smith N, Theroux P, Irwin M. The association between major depression and levels of soluble intercellular adhesion molecule 1, interleukin-6, and c-reactive protein in patients with recent acute coronary syndromes. The American journal of psychiatry. 2004;161:271-277

241. Miller GE, Freedland KE, Duntley S, Carney RM. Relation of depressive symptoms to c-reactive protein and pathogen burden (cytomegalovirus, herpes simplex virus, epstein-barr virus) in patients with earlier acute coronary syndromes. The American journal of cardiology. 2005;95:317-321

102 242. van der Ven A, van Diest R, Hamulyak K, Maes M, Bruggeman C, Appels A. Herpes viruses,

cytokines, and altered hemostasis in vital exhaustion. Psychosomatic medicine. 2003;65:194-200

243. Yang L, Zhang Z, Sun D, Xu Z, Zhang X, Li L. The serum interleukin-18 is a potential marker for development of post-stroke depression. Neurol Res. 2010;32:340-346

244. Jimenez I, Sobrino T, Rodriguez-Yanez M, Pouso M, Cristobo I, Sabucedo M, et al. High serum levels of leptin are associated with post-stroke depression. Psychological medicine. 2009;39:1201-1209

245. Banks WA. The many lives of leptin. Peptides. 2004;25:331-338 246. Shanley LJ, Irving AJ, Harvey J. Leptin enhances nmda receptor function and modulates

hippocampal synaptic plasticity. The Journal of neuroscience : the official journal of the Society for Neuroscience. 2001;21:RC186

247. Zhang F, Wang S, Signore AP, Chen J. Neuroprotective effects of leptin against ischemic injury induced by oxygen-glucose deprivation and transient cerebral ischemia. Stroke; a journal of cerebral circulation. 2007;38:2329-2336

248. Signore AP, Zhang F, Weng Z, Gao Y, Chen J. Leptin neuroprotection in the cns: Mechanisms and therapeutic potentials. Journal of neurochemistry. 2008;106:1977-1990

249. Soderberg S, Ahren B, Jansson JH, Johnson O, Hallmans G, Asplund K, et al. Leptin is associated with increased risk of myocardial infarction. Journal of internal medicine. 1999;246:409-418

250. Sierra-Johnson J, Romero-Corral A, Lopez-Jimenez F, Gami AS, Sert Kuniyoshi FH, Wolk R, et al. Relation of increased leptin concentrations to history of myocardial infarction and stroke in the united states population. The American journal of cardiology. 2007;100:234-239

251. Rubin RT, Rhodes ME, Czambel RK. Sexual diergism of baseline plasma leptin and leptin suppression by arginine vasopressin in major depressives and matched controls. Psychiatry research. 2002;113:255-268

252. Esel E, Ozsoy S, Tutus A, Sofuoglu S, Kartalci S, Bayram F, et al. Effects of antidepressant treatment and of gender on serum leptin levels in patients with major depression. Progress in neuro-psychopharmacology & biological psychiatry. 2005;29:565-570

253. Gecici O, Kuloglu M, Atmaca M, Tezcan AE, Tunckol H, Emul HM, et al. High serum leptin levels in depressive disorders with atypical features. Psychiatry and clinical neurosciences. 2005;59:736-738

254. Bossu P, Salani F, Cacciari C, Picchetto L, Cao M, Bizzoni F, et al. Disease outcome, alexithymia and depression are differently associated with serum il-18 levels in acute stroke. Current neurovascular research. 2009;6:163-170

255. Ormstad H, Aass HCD, Amthor K, Lund-Sorensen N, Sandvik L. Serum levels of cytokines, glucose, and hemoglobin as possible predictors of poststroke depression, and association with poststroke fatigue. International Journal of Neuroscience. 2012:1-9

256. Ormstad H, Aass HC, Amthor KF, Lund-Sorensen N, Sandvik L. Serum cytokine and glucose levels as predictors of poststroke fatigue in acute ischemic stroke patients. Journal of neurology. 2011;258:670-676

257. Loddick SA, Wong ML, Bongiorno PB, Gold PW, Licinio J, Rothwell NJ. Endogenous interleukin-1 receptor antagonist is neuroprotective. Biochemical and biophysical research communications. 1997;234:211-215

258. Dantzer R. Cytokine-induced sickness behaviour: A neuroimmune response to activation of innate immunity. European journal of pharmacology. 2004;500:399-411

259. Holmes JE, Miller NE. Effects of bacterial endotoxin on water intake, food intake, and body temperature in the albino rat. The Journal of experimental medicine. 1963;118:649-658

260. Hart BL. Biological basis of the behavior of sick animals. Neuroscience and biobehavioral reviews. 1988;12:123-137

261. Johnson RW. The concept of sickness behavior: A brief chronological account of four key discoveries. Vet Immunol Immunopathol. 2002;87:443-450

103 262. Capuron L, Gumnick JF, Musselman DL, Lawson DH, Reemsnyder A, Nemeroff CB, et al.

Neurobehavioral effects of interferon-alpha in cancer patients: Phenomenology and paroxetine responsiveness of symptom dimensions. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology. 2002;26:643-652

263. Capuron L, Ravaud A, Dantzer R. Early depressive symptoms in cancer patients receiving interleukin 2 and/or interferon alfa-2b therapy. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2000;18:2143-2151

264. Constant A, Castera L, Dantzer R, Couzigou P, de Ledinghen V, Demotes-Mainard J, et al. Mood alterations during interferon-alfa therapy in patients with chronic hepatitis c: Evidence for an overlap between manic/hypomanic and depressive symptoms. The Journal of clinical psychiatry. 2005;66:1050-1057

265. Capuron L, Ravaud A. Prediction of the depressive effects of interferon alfa therapy by the patient's initial affective state. The New England journal of medicine. 1999;340:1370

266. Capuron L, Raison CL, Musselman DL, Lawson DH, Nemeroff CB, Miller AH. Association of exaggerated hpa axis response to the initial injection of interferon-alpha with development of depression during interferon-alpha therapy. The American journal of psychiatry. 2003;160:1342-1345

267. Lambertsen KL, Biber K, Finsen B. Inflammatory cytokines in experimental and human stroke. Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism. 2012;32:1677-1698

268. Watanabe Y, Fujiwara M, Yoshida R, Hayaishi O. Stereospecificity of hepatic l-tryptophan 2,3-dioxygenase. The Biochemical journal. 1980;189:393-405

269. Moffett JR, Blinder KL, Venkateshan CN, Namboodiri MA. Differential effects of kynurenine and tryptophan treatment on quinolinate immunoreactivity in rat lymphoid and non-lymphoid organs. Cell and tissue research. 1998;293:525-534

270. Saito K, Crowley JS, Markey SP, Heyes MP. A mechanism for increased quinolinic acid formation following acute systemic immune stimulation. The Journal of biological chemistry. 1993;268:15496-15503

271. Saito K, Markey SP, Heyes MP. Effects of immune activation on quinolinic acid and neuroactive kynurenines in the mouse. Neuroscience. 1992;51:25-39

272. Mellor AL, Munn DH. Tryptophan catabolism and t-cell tolerance: Immunosuppression by starvation? Immunology today. 1999;20:469-473

273. Heyes MP, Saito K, Major EO, Milstien S, Markey SP, Vickers JH. A mechanism of quinolinic acid formation by brain in inflammatory neurological disease. Attenuation of synthesis from l-tryptophan by 6-chlorotryptophan and 4-chloro-3-hydroxyanthranilate. Brain : a journal of neurology. 1993;116 ( Pt 6):1425-1450

274. Moffett JR, Namboodiri MA. Tryptophan and the immune response. Immunology and cell biology. 2003;81:247-265

275. Myint AM. Kynurenines: From the perspective of major psychiatric disorders. The FEBS journal. 2012;279:1375-1385

276. Knox WE. Two mechanisms which increase in vivo the liver tryptophan peroxidase activity: Specific enzyme adaptation and stimulation of the pituitary adrenal system. British journal of experimental pathology. 1951;32:462-469

277. Salter M, Pogson CI. The role of tryptophan 2,3-dioxygenase in the hormonal control of tryptophan metabolism in isolated rat liver cells. Effects of glucocorticoids and experimental diabetes. The Biochemical journal. 1985;229:499-504

278. Leklem JE. Quantitative aspects of tryptophan metabolism in humans and other species: A review. The American journal of clinical nutrition. 1971;24:659-672

279. Grant RS, Kapoor V. Murine glial cells regenerate nad, after peroxide-induced depletion, using either nicotinic acid, nicotinamide, or quinolinic acid as substrates. Journal of neurochemistry. 1998;70:1759-1763

104 280. Grant RS, Naif H, Espinosa M, Kapoor V. Ido induction in ifn-gamma activated astroglia: A

role in improving cell viability during oxidative stress. Redox report : communications in free radical research. 2000;5:101-104

281. Guillemin GJ, Kerr SJ, Smythe GA, Smith DG, Kapoor V, Armati PJ, et al. Kynurenine pathway metabolism in human astrocytes: A paradox for neuronal protection. Journal of neurochemistry. 2001;78:842-853

282. Guillemin GJ, Smith DG, Kerr SJ, Smythe GA, Kapoor V, Armati PJ, et al. Characterisation of kynurenine pathway metabolism in human astrocytes and implications in neuropathogenesis. Redox report : communications in free radical research. 2000;5:108-111

283. Guillemin GJ, Smythe G, Takikawa O, Brew BJ. Expression of indoleamine 2,3-dioxygenase and production of quinolinic acid by human microglia, astrocytes, and neurons. Glia. 2005;49:15-23

284. Lardy HA. The role of tryptophan metabolites in regulating gluconeogenesis. The American journal of clinical nutrition. 1971;24:764-765

285. Yuskaitis CJ, Jope RS. Glycogen synthase kinase-3 regulates microglial migration, inflammation, and inflammation-induced neurotoxicity. Cellular signalling. 2009;21:264-273

286. Heyes MP, Lackner A. Increased cerebrospinal fluid quinolinic acid, kynurenic acid, and l-kynurenine in acute septicemia. Journal of neurochemistry. 1990;55:338-341

287. Melillo G, Cox GW, Radzioch D, Varesio L. Picolinic acid, a catabolite of l-tryptophan, is a costimulus for the induction of reactive nitrogen intermediate production in murine macrophages. Journal of immunology (Baltimore, Md. : 1950). 1993;150:4031-4040

288. Lee GK, Park HJ, Macleod M, Chandler P, Munn DH, Mellor AL. Tryptophan deprivation sensitizes activated t cells to apoptosis prior to cell division. Immunology. 2002;107:452-460

289. Frumento G, Rotondo R, Tonetti M, Damonte G, Benatti U, Ferrara GB. Tryptophan-derived catabolites are responsible for inhibition of t and natural killer cell proliferation induced by indoleamine 2,3-dioxygenase. The Journal of experimental medicine. 2002;196:459-468

290. Terness P, Bauer TM, Rose L, Dufter C, Watzlik A, Simon H, et al. Inhibition of allogeneic t cell proliferation by indoleamine 2,3-dioxygenase-expressing dendritic cells: Mediation of suppression by tryptophan metabolites. The Journal of experimental medicine. 2002;196:447-457

291. Fallarino F, Grohmann U, Vacca C, Bianchi R, Orabona C, Spreca A, et al. T cell apoptosis by tryptophan catabolism. Cell Death Differ. 2002;9:1069-1077

292. Zaher SS, Germain C, Fu H, Larkin DF, George AJ. 3-hydroxykynurenine suppresses cd4+ t-cell proliferation, induces t-regulatory-cell development, and prolongs corneal allograft survival. Invest Ophthalmol Vis Sci.52:2640-2648

293. Hayashi T, Mo JH, Gong X, Rossetto C, Jang A, Beck L, et al. 3-hydroxyanthranilic acid inhibits pdk1 activation and suppresses experimental asthma by inducing t cell apoptosis. Proceedings of the National Academy of Sciences of the United States of America. 2007;104:18619-18624

294. Fallarino F, Grohmann U, You S, McGrath BC, Cavener DR, Vacca C, et al. The combined effects of tryptophan starvation and tryptophan catabolites down-regulate t cell receptor zeta-chain and induce a regulatory phenotype in naive t cells. Journal of immunology (Baltimore, Md. : 1950). 2006;176:6752-6761

295. Mezrich JD, Fechner JH, Zhang X, Johnson BP, Burlingham WJ, Bradfield CA. An interaction between kynurenine and the aryl hydrocarbon receptor can generate regulatory t cells. Journal of immunology (Baltimore, Md. : 1950). 2010;185:3190-3198

296. Harber M, Sundstedt A, Wraith D. The role of cytokines in immunological tolerance: Potential for therapy. Expert reviews in molecular medicine. 2000;2:1-20

297. Vecsei L, Szalardy L, Fulop F, Toldi J. Kynurenines in the cns: Recent advances and new questions. Nature reviews. Drug discovery. 2012;12:64-82

298. Fernstrom JD. Effects on the diet on brain neurotransmitters. Metabolism: clinical and experimental. 1977;26:207-223

105 299. Pertz H, Back W. [synthesis and resolution of chiral ring-opened serotonin analogs of the 5-

hydroxykynuramine type]. Pharmaceutica acta Helvetiae. 1988;63:128-131 300. Bender DA, McCreanor GM. Kynurenine hydroxylase: A potential rate-limiting enzyme in

tryptophan metabolism. Biochemical Society transactions. 1985;13:441-443 301. Wu HQ, Guidetti P, Goodman JH, Varasi M, Ceresoli-Borroni G, Speciale C, et al.

Kynurenergic manipulations influence excitatory synaptic function and excitotoxic vulnerability in the rat hippocampus in vivo. Neuroscience. 2000;97:243-251

302. Santamaria A, Jimenez-Capdeville ME, Camacho A, Rodriguez-Martinez E, Flores A, Galvan-Arzate S. In vivo hydroxyl radical formation after quinolinic acid infusion into rat corpus striatum. Neuroreport. 2001;12:2693-2696

303. Kessler M, Terramani T, Lynch G, Baudry M. A glycine site associated with n-methyl-d-aspartic acid receptors: Characterization and identification of a new class of antagonists. Journal of neurochemistry. 1989;52:1319-1328

304. Carpenedo R, Pittaluga A, Cozzi A, Attucci S, Galli A, Raiteri M, et al. Presynaptic kynurenate-sensitive receptors inhibit glutamate release. The European journal of neuroscience. 2001;13:2141-2147

305. Maes M, Galecki P, Verkerk R, Rief W. Somatization, but not depression, is characterized by disorders in the tryptophan catabolite (trycat) pathway, indicating increased indoleamine 2,3-dioxygenase and lowered kynurenine aminotransferase activity. Neuro endocrinology letters. 2011;32:264-273

306. Sublette ME, Galfalvy HC, Fuchs D, Lapidus M, Grunebaum MF, Oquendo MA, et al. Plasma kynurenine levels are elevated in suicide attempters with major depressive disorder. Brain, behavior, and immunity. 2011;25:1272-1278

307. Kohl C, Walch T, Huber R, Kemmler G, Neurauter G, Fuchs D, et al. Measurement of tryptophan, kynurenine and neopterin in women with and without postpartum blues. Journal of affective disorders. 2005;86:135-142

308. Maes M, Verkerk R, Bonaccorso S, Ombelet W, Bosmans E, Scharpe S. Depressive and anxiety symptoms in the early puerperium are related to increased degradation of tryptophan into kynurenine, a phenomenon which is related to immune activation. Life sciences. 2002;71:1837-1848

309. Yildiz-Yesiloglu A, Ankerst DP. Review of 1h magnetic resonance spectroscopy findings in major depressive disorder: A meta-analysis. Psychiatry research. 2006;147:1-25

310. Hercher C, Turecki G, Mechawar N. Through the looking glass: Examining neuroanatomical evidence for cellular alterations in major depression. Journal of psychiatric research. 2009;43:947-961

311. Adams RJ, Chimowitz MI, Alpert JS, Awad IA, Cerqueria MD, Fayad P, et al. Coronary risk evaluation in patients with transient ischemic attack and ischemic stroke: A scientific statement for healthcare professionals from the stroke council and the council on clinical cardiology of the american heart association/american stroke association. Stroke; a journal of cerebral circulation. 2003;34:2310-2322

312. Jensen JK, Medina HM, Norgaard BL, Ovrehus KA, Jensen JM, Nielsen LH, et al. Association of ischemic stroke to coronary artery disease using computed tomography coronary angiography. International journal of cardiology. 2012;160:171-174

313. Anderson G, Maes M, Berk M. Biological underpinnings of the commonalities in depression, somatization, and chronic fatigue syndrome. Medical hypotheses. 2012;78:752-756

314. Rief W, Hessel A, Braehler E. Somatization symptoms and hypochondriacal features in the general population. Psychosomatic medicine. 2001;63:595-602

315. Harvey JA, Schlosberg AJ, Yunger LM. Behavioral correlates of serotonin depletion. Federation proceedings. 1975;34:1796-1801

316. Stone TW, Darlington LG. Endogenous kynurenines as targets for drug discovery and development. Nature reviews. Drug discovery. 2002;1:609-620

106 317. Kaszaki J, Palasthy Z, Erczes D, Racz A, Torday C, Varga G, et al. Kynurenic acid inhibits

intestinal hypermotility and xanthine oxidase activity during experimental colon obstruction in dogs. Neurogastroenterology and motility : the official journal of the European Gastrointestinal Motility Society. 2008;20:53-62

318. Cosi C, Mannaioni G, Cozzi A, Carla V, Sili M, Cavone L, et al. G-protein coupled receptor 35 (gpr35) activation and inflammatory pain: Studies on the antinociceptive effects of kynurenic acid and zaprinast. Neuropharmacology. 2011;60:1227-1231

319. Wichers MC, Koek GH, Robaeys G, Verkerk R, Scharpe S, Maes M. Ido and interferon-alpha-induced depressive symptoms: A shift in hypothesis from tryptophan depletion to neurotoxicity. Molecular psychiatry. 2005;10:538-544

320. Gal EM, Sherman AD. L-kynurenine: Its synthesis and possible regulatory function in brain. Neurochemical research. 1980;5:223-239

321. Schwarcz R, Pellicciari R. Manipulation of brain kynurenines: Glial targets, neuronal effects, and clinical opportunities. The Journal of pharmacology and experimental therapeutics. 2002;303:1-10

322. Okuda S, Nishiyama N, Saito H, Katsuki H. 3-hydroxykynurenine, an endogenous oxidative stress generator, causes neuronal cell death with apoptotic features and region selectivity. Journal of neurochemistry. 1998;70:299-307

323. Gold AB, Herrmann N, Swardfager W, Black SE, Aviv RI, Tennen G, et al. The relationship between indoleamine 2,3-dioxygenase activity and post-stroke cognitive impairment. Journal of neuroinflammation. 2011;8:17

324. Tavares RG, Tasca CI, Santos CE, Alves LB, Porciuncula LO, Emanuelli T, et al. Quinolinic acid stimulates synaptosomal glutamate release and inhibits glutamate uptake into astrocytes. Neurochemistry international. 2002;40:621-627

325. Fujioka M, Nishio K, Miyamoto S, Hiramatsu KI, Sakaki T, Okuchi K, et al. Hippocampal damage in the human brain after cardiac arrest. Cerebrovascular diseases (Basel, Switzerland). 2000;10:2-7

326. Brouns R, Verkerk R, Aerts T, De Surgeloose D, Wauters A, Scharpe S, et al. The role of tryptophan catabolism along the kynurenine pathway in acute ischemic stroke. Neurochemical research. 2010;35:1315-1322

327. Mackay GM, Forrest CM, Stoy N, Christofides J, Egerton M, Stone TW, et al. Tryptophan metabolism and oxidative stress in patients with chronic brain injury. Eur J Neurol. 2006;13:30-42

328. Coppen A. The biochemistry of affective disorders. The British journal of psychiatry : the journal of mental science. 1967;113:1237-1264

329. Cowen PJ. A role for 5-ht in the action of antidepressant drugs. Pharmacology & therapeutics. 1990;46:43-51

330. Grasby PM. Imaging the neurochemical brain in health and disease. Clinical medicine (London, England). 2002;2:67-73

331. Sargent PA, Kjaer KH, Bench CJ, Rabiner EA, Messa C, Meyer J, et al. Brain serotonin1a receptor binding measured by positron emission tomography with [11c]way-100635: Effects of depression and antidepressant treatment. Archives of general psychiatry. 2000;57:174-180

332. Drevets WC, Thase ME, Moses-Kolko EL, Price J, Frank E, Kupfer DJ, et al. Serotonin-1a receptor imaging in recurrent depression: Replication and literature review. Nuclear medicine and biology. 2007;34:865-877

333. Parsey RV, Oquendo MA, Ogden RT, Olvet DM, Simpson N, Huang YY, et al. Altered serotonin 1a binding in major depression: A [carbonyl-c-11]way100635 positron emission tomography study. Biological psychiatry. 2006;59:106-113

334. Bhagwagar Z, Rabiner EA, Sargent PA, Grasby PM, Cowen PJ. Persistent reduction in brain serotonin1a receptor binding in recovered depressed men measured by positron emission tomography with [11c]way-100635. Molecular psychiatry. 2004;9:386-392

107 335. Neumeister A, Bain E, Nugent AC, Carson RE, Bonne O, Luckenbaugh DA, et al. Reduced

serotonin type 1a receptor binding in panic disorder. The Journal of neuroscience : the official journal of the Society for Neuroscience. 2004;24:589-591

336. Cowen PJ. Serotonin and depression: Pathophysiological mechanism or marketing myth? Trends in pharmacological sciences. 2008;29:433-436

337. Bhagwagar Z, Hinz R, Taylor M, Fancy S, Cowen P, Grasby P. Increased 5-ht(2a) receptor binding in euthymic, medication-free patients recovered from depression: A positron emission study with [(11)c]mdl 100,907. The American journal of psychiatry. 2006;163:1580-1587

338. Mintun MA, Sheline YI, Moerlein SM, Vlassenko AG, Huang Y, Snyder AZ. Decreased hippocampal 5-ht2a receptor binding in major depressive disorder: In vivo measurement with [18f]altanserin positron emission tomography. Biological psychiatry. 2004;55:217-224

339. Meyer JH, Kapur S, Houle S, DaSilva J, Owczarek B, Brown GM, et al. Prefrontal cortex 5-ht2 receptors in depression: An [18f]setoperone pet imaging study. The American journal of psychiatry. 1999;156:1029-1034

340. Malison RT, Price LH, Berman R, van Dyck CH, Pelton GH, Carpenter L, et al. Reduced brain serotonin transporter availability in major depression as measured by [123i]-2 beta-carbomethoxy-3 beta-(4-iodophenyl)tropane and single photon emission computed tomography. Biological psychiatry. 1998;44:1090-1098

341. Herold N, Uebelhack K, Franke L, Amthauer H, Luedemann L, Bruhn H, et al. Imaging of serotonin transporters and its blockade by citalopram in patients with major depression using a novel spect ligand [123i]-adam. Journal of neural transmission (Vienna, Austria : 1996). 2006;113:659-670

342. Bryer JB, Starkstein SE, Votypka V, Parikh RM, Price TR, Robinson RG. Reduction of csf monoamine metabolites in poststroke depression: A preliminary report. The Journal of neuropsychiatry and clinical neurosciences. 1992;4:440-442

343. Gao HQ, Zhu HY, Zhang YQ, Wang LX. Reduction of cerebrospinal fluid and plasma serotonin in patients with post-stroke depression: A preliminary report. Clinical and investigative medicine. Medecine clinique et experimentale. 2008;31:E351-356

344. Mayberg HS, Robinson RG, Wong DF, Parikh R, Bolduc P, Starkstein SE, et al. Pet imaging of cortical s2 serotonin receptors after stroke: Lateralized changes and relationship to depression. The American journal of psychiatry. 1988;145:937-943

345. Dunn AJ, Swiergiel AH, de Beaurepaire R. Cytokines as mediators of depression: What can we learn from animal studies? Neuroscience and biobehavioral reviews. 2005;29:891-909

346. Zhu CB, Blakely RD, Hewlett WA. The proinflammatory cytokines interleukin-1beta and tumor necrosis factor-alpha activate serotonin transporters. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology. 2006;31:2121-2131

347. Zhu CB, Lindler KM, Owens AW, Daws LC, Blakely RD, Hewlett WA. Interleukin-1 receptor activation by systemic lipopolysaccharide induces behavioral despair linked to mapk regulation of cns serotonin transporters. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology. 2010;35:2510-2520

348. Cai W, Khaoustov VI, Xie Q, Pan T, Le W, Yoffe B. Interferon-alpha-induced modulation of glucocorticoid and serotonin receptors as a mechanism of depression. Journal of hepatology. 2005;42:880-887

349. Felger JC, Alagbe O, Pace TW, Woolwine BJ, Hu F, Raison CL, et al. Early activation of p38 mitogen activated protein kinase is associated with interferon-alpha-induced depression and fatigue. Brain, behavior, and immunity. 2011;25:1094-1098

350. Yao JK, Dougherty GG, Jr., Reddy RD, Keshavan MS, Montrose DM, Matson WR, et al. Altered interactions of tryptophan metabolites in first-episode neuroleptic-naive patients with schizophrenia. Molecular psychiatry. 2010;15:938-953

351. Klein DC, Weller JL. Indole metabolism in the pineal gland: A circadian rhythm in n-acetyltransferase. Science (New York, N.Y.). 1970;169:1093-1095

108 352. Lindemann L, Hoener MC. A renaissance in trace amines inspired by a novel gpcr family.

Trends in pharmacological sciences. 2005;26:274-281 353. McMenamy RH. Binding of indole analogues to human serum albumin. Effects of fatty acids.

The Journal of biological chemistry. 1965;240:4235-4243 354. Madras BK, Cohen EL, Messing R, Munro HN, Wurtman RJ. Relevance of free tryptophan in

serum to tissue tryptophan concentrations. Metabolism: clinical and experimental. 1974;23:1107-1116

355. Pardridge WM, Fierer G. Transport of tryptophan into brain from the circulating, albumin-bound pool in rats and in rabbits. Journal of neurochemistry. 1990;54:971-976

356. Hargreaves KM, Pardridge WM. Neutral amino acid transport at the human blood-brain barrier. The Journal of biological chemistry. 1988;263:19392-19397

357. Fernstrom JD. Branched-chain amino acids and brain function. The Journal of nutrition. 2005;135:1539S-1546S

358. Fernstrom JD. Role of precursor availability in control of monoamine biosynthesis in brain. Physiological reviews. 1983;63:484-546

359. Fernstrom JD, Wurtman RJ. Brain serotonin content: Physiological dependence on plasma tryptophan levels. Science (New York, N.Y.). 1971;173:149-152

360. Cowen PJ, Parry-Billings M, Newsholme EA. Decreased plasma tryptophan levels in major depression. Journal of affective disorders. 1989;16:27-31

361. Young SN, Smith SE, Pihl RO, Ervin FR. Tryptophan depletion causes a rapid lowering of mood in normal males. Psychopharmacology. 1985;87:173-177

362. Hoes MJ. The clinical significance of an elevated excretion of xanthurenic acid in psychiatric patients. Acta psychiatrica Belgica. 1979;79:638-646

363. Maes M, De Ruyter M, Suy E. The renal excretion of xanthurenic acid following l-tryptophan loading in depressed patients. Human Psychopharmacology: Clinical and Experimental. 1987;2:231-235

364. Brown RR, Lee CM, Kohler PC, Hank JA, Storer BE, Sondel PM. Altered tryptophan and neopterin metabolism in cancer patients treated with recombinant interleukin 2. Cancer research. 1989;49:4941-4944

365. Capuron L, Ravaud A, Neveu PJ, Miller AH, Maes M, Dantzer R. Association between decreased serum tryptophan concentrations and depressive symptoms in cancer patients undergoing cytokine therapy. Molecular psychiatry. 2002;7:468-473

366. Aquilani R, Verri M, Iadarola P, Arcidiaco P, Boschi F, Dossena M, et al. Plasma precursors of brain catecholaminergic and serotonergic neurotransmitters in rehabilitation patients with ischemic stroke. Archives of physical medicine and rehabilitation. 2004;85:779-784

367. Pohjasvaara T, Mantyla R, Salonen O, Aronen HJ, Ylikoski R, Hietanen M, et al. How complex interactions of ischemic brain infarcts, white matter lesions, and atrophy relate to poststroke dementia. Archives of neurology. 2000;57:1295-1300

368. Robinson RG, Shoemaker WJ, Schlumpf M, Valk T, Bloom FE. Effect of experimental cerebral infarction in rat brain on catecholamines and behaviour. Nature. 1975;255:332-334

369. Robinson RG, Price TR. Post-stroke depressive disorders: A follow-up study of 103 patients. Stroke; a journal of cerebral circulation. 1982;13:635-641

370. Morrison JH, Molliver ME, Grzanna R. Noradrenergic innervation of cerebral cortex: Widespread effects of local cortical lesions. Science (New York, N.Y.). 1979;205:313-316

371. Palkovits M, Saavedra JM, Jacoboqitz DM, Kizer JS, Zaborszky L, Brownstein MJ. Serotonergic innervation of the forebrain: Effect of lesions on serotonin and tryptophan hydroxylase levels. Brain research. 1977;130:121-134

372. Sinyor D, Jacques P, Kaloupek DG, Becker R, Goldenberg M, Coopersmith H. Poststroke depression and lesion location. An attempted replication. Brain : a journal of neurology. 1986;109 ( Pt 3):537-546

373. Eastwood MR, Rifat SL, Nobbs H, Ruderman J. Mood disorder following cerebrovascular accident. The British journal of psychiatry : the journal of mental science. 1989;154:195-200

109 374. House A, Dennis M, Warlow C, Hawton K, Molyneux A. Mood disorders after stroke and their

relation to lesion location. A ct scan study. Brain : a journal of neurology. 1990;113 ( Pt 4):1113-1129

375. Dam H, Pedersen HE, Ahlgren P. Depression among patients with stroke. Acta psychiatrica Scandinavica. 1989;80:118-124

376. MacHale SM, O'Rourke SJ, Wardlaw JM, Dennis MS. Depression and its relation to lesion location after stroke. Journal of neurology, neurosurgery, and psychiatry. 1998;64:371-374

377. Starkstein SE, Robinson RG, Berthier ML, Price TR. Depressive disorders following posterior circulation as compared with middle cerebral artery infarcts. Brain : a journal of neurology. 1988;111 ( Pt 2):375-387

378. Tekin S, Cummings JL. Frontal-subcortical neuronal circuits and clinical neuropsychiatry: An update. Journal of psychosomatic research. 2002;53:647-654

379. Drevets WC. Functional anatomical abnormalities in limbic and prefrontal cortical structures in major depression. Progress in brain research. 2000;126:413-431

380. Herrmann M, Bartels C, Schumacher M, Wallesch CW. Poststroke depression. Is there a pathoanatomic correlate for depression in the postacute stage of stroke? Stroke; a journal of cerebral circulation. 1995;26:850-856

381. Herrmann M, Bartels C, Wallesch CW. Depression in acute and chronic aphasia: Symptoms, pathoanatomical-clinical correlations and functional implications. Journal of neurology, neurosurgery, and psychiatry. 1993;56:672-678

382. Beblo T, Wallesch CW, Herrmann M. The crucial role of frontostriatal circuits for depressive disorders in the postacute stage after stroke. Neuropsychiatry, neuropsychology, and behavioral neurology. 1999;12:236-246

383. Kim JS, Choi-Kwon S. Poststroke depression and emotional incontinence: Correlation with lesion location. Neurology. 2000;54:1805-1810

384. Singh A, Herrmann N, Black SE. The importance of lesion location in poststroke depression: A critical review. Canadian journal of psychiatry. Revue canadienne de psychiatrie. 1998;43:921-927

385. Carson AJ, MacHale S, Allen K, Lawrie SM, Dennis M, House A, et al. Depression after stroke and lesion location: A systematic review. Lancet. 2000;356:122-126

386. Bhogal SK, Teasell R, Foley N, Speechley M. Lesion location and poststroke depression: Systematic review of the methodological limitations in the literature. Stroke; a journal of cerebral circulation. 2004;35:794-802

387. Narushima K, Kosier JT, Robinson RG. A reappraisal of poststroke depression, intra- and inter-hemispheric lesion location using meta-analysis. The Journal of neuropsychiatry and clinical neurosciences. 2003;15:422-430

388. Moody DM, Brown WR, Challa VR, Anderson RL. Periventricular venous collagenosis: Association with leukoaraiosis. Radiology. 1995;194:469-476

389. Pantoni L, Garcia JH. Pathogenesis of leukoaraiosis: A review. Stroke; a journal of cerebral circulation. 1997;28:652-659

390. Pantoni L, Garcia JH, Gutierrez JA. Cerebral white matter is highly vulnerable to ischemia. Stroke; a journal of cerebral circulation. 1996;27:1641-1646; discussion 1647

391. Hamner MA, Moller T, Ransom BR. Anaerobic function of cns white matter declines with age. Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism. 2011;31:996-1002

392. Matute C, Ransom BR. Roles of white matter in central nervous system pathophysiologies. ASN neuro. 2012;4

393. Filley CM. White matter and behavioral neurology. Annals of the New York Academy of Sciences. 2005;1064:162-183

394. Desmond DW. Cognition and white matter lesions. Cerebrovascular diseases (Basel, Switzerland). 2002;13 Suppl 2:53-57

110 395. Alexopoulos GS, Meyers BS, Young RC, Campbell S, Silbersweig D, Charlson M. 'Vascular

depression' hypothesis. Archives of general psychiatry. 1997;54:915-922 396. Alexopoulos GS, Meyers BS, Young RC, Kakuma T, Silbersweig D, Charlson M. Clinically

defined vascular depression. The American journal of psychiatry. 1997;154:562-565 397. Luber MP, Hollenberg JP, Williams-Russo P, DiDomenico TN, Meyers BS, Alexopoulos GS,

et al. Diagnosis, treatment, comorbidity, and resource utilization of depressed patients in a general medical practice. International journal of psychiatry in medicine. 2000;30:1-13

398. Baron M, Mendlewicz J, Klotz J. Age-of-onset and genetic transmission in affective disorders. Acta psychiatrica Scandinavica. 1981;64:373-380

399. Coffey CE, Figiel GS, Djang WT, Cress M, Saunders WB, Weiner RD. Leukoencephalopathy in elderly depressed patients referred for ect. Biological psychiatry. 1988;24:143-161

400. Coffey CE, Figiel GS, Djang WT, Saunders WB, Weiner RD. White matter hyperintensity on magnetic resonance imaging: Clinical and neuroanatomic correlates in the depressed elderly. The Journal of neuropsychiatry and clinical neurosciences. 1989;1:135-144

401. Coffey CE, Figiel GS, Djang WT, Weiner RD. Subcortical hyperintensity on magnetic resonance imaging: A comparison of normal and depressed elderly subjects. The American journal of psychiatry. 1990;147:187-189

402. Herrmann LL, Le Masurier M, Ebmeier KP. White matter hyperintensities in late life depression: A systematic review. Journal of neurology, neurosurgery, and psychiatry. 2008;79:619-624

403. Olesen PJ, Gustafson DR, Simoni M, Pantoni L, Ostling S, Guo X, et al. Temporal lobe atrophy and white matter lesions are related to major depression over 5 years in the elderly. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology. 2010;35:2638-2645

404. Firbank MJ, Teodorczuk A, van der Flier WM, Gouw AA, Wallin A, Erkinjuntti T, et al. Relationship between progression of brain white matter changes and late-life depression: 3-year results from the ladis study. The British journal of psychiatry : the journal of mental science. 2012;201:40-45

405. Starkstein SE, Robinson RG, Price TR. Comparison of patients with and without poststroke major depression matched for size and location of lesion. Archives of general psychiatry. 1988;45:247-252

406. Francis PL. Neuroanatomical correlates of depressive symptoms following acute ischemic stroke. Graduate Department of Pharmacology and Toxicology. 2011;M.Sc.

407. Wong A. Neuropathology of post-stroke depression: Possible role of inflammatory molecules and indolamine 2,3-dioxygenase. Department of Pharmacology and Toxicology. 2010;M.Sc.

408. Rothenburg L. Exploring the biological correlates of post-stroke depression. Department of Pharmaceutical Sciences. 2007;M.Sc.

409. Foulkes MA, Wolf PA, Price TR, Mohr JP, Hier DB. The stroke data bank: Design, methods, and baseline characteristics. Stroke; a journal of cerebral circulation. 1988;19:547-554

410. Brott T, Adams HP, Jr., Olinger CP, Marler JR, Barsan WG, Biller J, et al. Measurements of acute cerebral infarction: A clinical examination scale. Stroke; a journal of cerebral circulation. 1989;20:864-870

411. Williams LS, Yilmaz EY, Lopez-Yunez AM. Retrospective assessment of initial stroke severity with the nih stroke scale. Stroke. 2000;31:858-862

412. Kunitz SC, Gross CR, Heyman A, Kase CS, Mohr JP, Price TR, et al. The pilot stroke data bank: Definition, design, and data. Stroke; a journal of cerebral circulation. 1984;15:740-746

413. Adams HP, Jr., Davis PH, Leira EC, Chang KC, Bendixen BH, Clarke WR, et al. Baseline nih stroke scale score strongly predicts outcome after stroke: A report of the trial of org 10172 in acute stroke treatment (toast). Neurology. 1999;53:126-131

414. Li CC, Chen YM, Tsay SL, Hu GC, Lin KC. Predicting functional outcomes in patients suffering from ischaemic stroke using initial admission variables and physiological data: A

111

comparison between tree model and multivariate regression analysis. Disability and rehabilitation. 2010;32:2088-2096

415. Radloff L. The ces-d scale: A self-report depression scaled for research in the general population. Applied Psychological Measurment. 1977;1:385-401

416. Agrell B, Dehlin O. Comparison of six depression rating scales in geriatric stroke patients. Stroke; a journal of cerebral circulation. 1989;20:1190-1194

417. Sato Y, Kaji M, Saruwatari N, Oizumi K. Hemiosteoporosis following stroke: Importance of pathophysiologic understanding and histologic evidence. Stroke; a journal of cerebral circulation. 1999;30:1978-1979

418. Parikh RM, Eden DT, Price TR, Robinson RG. The sensitivity and specificity of the center for epidemiologic studies depression scale in screening for post-stroke depression. International journal of psychiatry in medicine. 1988;18:169-181

419. Raji MA, Reyes-Ortiz CA, Kuo YF, Markides KS, Ottenbacher KJ. Depressive symptoms and cognitive change in older mexican americans. Journal of geriatric psychiatry and neurology. 2007;20:145-152

420. Callahan CM, Wolinsky FD, Stump TE, Nienaber NA, Hui SL, Tierney WM. Mortality, symptoms, and functional impairment in late-life depression. Journal of general internal medicine. 1998;13:746-752

421. Folstein MF, Folstein SE, McHugh PR. "Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. Journal of psychiatric research. 1975;12:189-198

422. Tombaugh TN, McIntyre NJ. The mini-mental state examination: A comprehensive review. Journal of the American Geriatrics Society. 1992;40:922-935

423. te Winkel-Witlox AC, Post MW, Visser-Meily JM, Lindeman E. Efficient screening of cognitive dysfunction in stroke patients: Comparison between the camcog and the r-camcog, mini mental state examination and functional independence measure-cognition score. Disability and rehabilitation. 2008;30:1386-1391

424. Anderson GM, Purdy WC. Liquid chromatographic-fluorometric system for the determination of indoles in physiological samples. Analytical chemistry. 1979;51:283-286

425. Forrest CM, Mackay GM, Stoy N, Egerton M, Christofides J, Stone TW, et al. Tryptophan loading induces oxidative stress. Free Radic Res. 2004;38:1167-1171

426. Marques-Deak AH, Neto FL, Dominguez WV, Solis AC, Kurcgant D, Sato F, et al. Cytokine profiles in women with different subtypes of major depressive disorder. Journal of psychiatric research. 2007;41:152-159

427. Zaremba J, Losy J. Interleukin-18 in acute ischaemic stroke patients. Neurological sciences : official journal of the Italian Neurological Society and of the Italian Society of Clinical Neurophysiology. 2003;24:117-124

428. Bobrowski WF, McDuffie JE, Sobocinski G, Chupka J, Olle E, Bowman A, et al. Comparative methods for multiplex analysis of cytokine protein expression in plasma of lipopolysaccharide-treated mice. Cytokine. 2005;32:194-198

429. van der Worp HB, Claus SP, Bar PR, Ramos LM, Algra A, van Gijn J, et al. Reproducibility of measurements of cerebral infarct volume on ct scans. Stroke; a journal of cerebral circulation. 2001;32:424-430

430. Wahlund LO, Barkhof F, Fazekas F, Bronge L, Augustin M, Sjogren M, et al. A new rating scale for age-related white matter changes applicable to mri and ct. Stroke; a journal of cerebral circulation. 2001;32:1318-1322

431. F Fazekas JC, A Alavi, HI Hurtig and RA Zimmerman. Mr signal abnormalities at 1.5 t in alzheimer's dementia and normal aging. American Journal of Roentgenology. 1987;149:351-356

432. Gao FQ, Black SE, Leibovitch FS, Callen DJ, Lobaugh NJ, Szalai JP. A reliable mr measurement of medial temporal lobe width from the sunnybrook dementia study. Neurobiology of aging. 2003;24:49-56

112 433. Rothenburg LS, Herrmann N, Swardfager W, Black SE, Tennen G, Kiss A, et al. The

relationship between inflammatory markers and post stroke cognitive impairment. Journal of geriatric psychiatry and neurology. 2010;23:199-205

434. Ormstad H, Aass HC, Lund-Sorensen N, Amthor KF, Sandvik L. Serum levels of cytokines and c-reactive protein in acute ischemic stroke patients, and their relationship to stroke lateralization, type, and infarct volume. Journal of neurology. 2011;258:677-685

435. Capuron L, Ravaud A, Miller AH, Dantzer R. Baseline mood and psychosocial characteristics of patients developing depressive symptoms during interleukin-2 and/or interferon-alpha cancer therapy. Brain, behavior, and immunity. 2004;18:205-213

436. Vazquez-Del Mercado M, Garcia-Gonzalez A, Munoz-Valle JF, Garcia-Iglesias T, Martinez-Bonilla G, Bernard-Medina G, et al. Interleukin 1beta (il-1beta), il-10, tumor necrosis factor-alpha, and cellular proliferation index in peripheral blood mononuclear cells in patients with ankylosing spondylitis. The Journal of rheumatology. 2002;29:522-526

437. Hulthe J, McPheat W, Samnegard A, Tornvall P, Hamsten A, Eriksson P. Plasma interleukin (il)-18 concentrations is elevated in patients with previous myocardial infarction and related to severity of coronary atherosclerosis independently of c-reactive protein and il-6. Atherosclerosis. 2006;188:450-454

438. Frick B, Schroecksnadel K, Neurauter G, Leblhuber F, Fuchs D. Increasing production of homocysteine and neopterin and degradation of tryptophan with older age. Clinical biochemistry. 2004;37:684-687

439. Pertovaara M, Raitala A, Lehtimaki T, Karhunen PJ, Oja SS, Jylha M, et al. Indoleamine 2,3-dioxygenase activity in nonagenarians is markedly increased and predicts mortality. Mechanisms of ageing and development. 2006;127:497-499

440. Wood K, Harwood J, Coppen A. The effect of antidepressant drugs on plasma kynurenine in depressed patients. Psychopharmacology. 1978;59:263-266

441. Moller SE, Kirk L, Honore P. Tryptophan tolerance and metabolism in endogenous depression. Psychopharmacology. 1982;76:79-83

442. Maes M, De Ruyter M, Suy E. Xanthurenic acid flow in 24-hour urine following l-tryptophan loading in depressive patients. Acta psychiatrica Belgica. 1986;86:120-130

443. Cazzullo CL, Mangoni A, Mascherpa G. Tryptophan metabolism in affective psychoses. The British journal of psychiatry : the journal of mental science. 1966;112:157-162

444. Frazer A, Pandey GN, Mendels J. Metabolism of tryptophan in depressive disease. Archives of general psychiatry. 1973;29:528-535

445. Wirleitner B, Rudzite V, Neurauter G, Murr C, Kalnins U, Erglis A, et al. Immune activation and degradation of tryptophan in coronary heart disease. Eur J Clin Invest. 2003;33:550-554

446. Suzuki Y, Suda T, Furuhashi K, Suzuki M, Fujie M, Hahimoto D, et al. Increased serum kynurenine/tryptophan ratio correlates with disease progression in lung cancer. Lung cancer (Amsterdam, Netherlands). 2010;67:361-365

447. Giusti RM, Maloney EM, Hanchard B, Morgan OS, Steinberg SM, Wachter H, et al. Differential patterns of serum biomarkers of immune activation in human t-cell lymphotropic virus type i-associated myelopathy/tropical spastic paraparesis, and adult t-cell leukemia/lymphoma. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 1996;5:699-704

448. Medana IM, Day NP, Salahifar-Sabet H, Stocker R, Smythe G, Bwanaisa L, et al. Metabolites of the kynurenine pathway of tryptophan metabolism in the cerebrospinal fluid of malawian children with malaria. The Journal of infectious diseases. 2003;188:844-849

449. Schroecksnadel K, Winkler C, Duftner C, Wirleitner B, Schirmer M, Fuchs D. Tryptophan degradation increases with stage in patients with rheumatoid arthritis. Clinical rheumatology. 2006;25:334-337

113 450. Zangerle R, Widner B, Quirchmair G, Neurauter G, Sarcletti M, Fuchs D. Effective

antiretroviral therapy reduces degradation of tryptophan in patients with hiv-1 infection. Clinical immunology (Orlando, Fla.). 2002;104:242-247

451. Molife C. Is depression a modifiable risk factor for diabetes burden? Journal of primary care & community health. 2010;1:55-61

452. Nozaki K, Beal MF. Neuroprotective effects of l-kynurenine on hypoxia-ischemia and nmda lesions in neonatal rats. Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism. 1992;12:400-407

453. Gigler G, Szenasi G, Simo A, Levay G, Harsing LG, Jr., Sas K, et al. Neuroprotective effect of l-kynurenine sulfate administered before focal cerebral ischemia in mice and global cerebral ischemia in gerbils. European journal of pharmacology. 2007;564:116-122

454. von Boehmer H. Mechanisms of suppression by suppressor t cells. Nat Immunol. 2005;6:338-344

455. Klein L, Khazaie K, von Boehmer H. In vivo dynamics of antigen-specific regulatory t cells not predicted from behavior in vitro. Proceedings of the National Academy of Sciences of the United States of America. 2003;100:8886-8891

456. Annacker O, Pimenta-Araujo R, Burlen-Defranoux O, Barbosa TC, Cumano A, Bandeira A. Cd25+ cd4+ t cells regulate the expansion of peripheral cd4 t cells through the production of il-10. Journal of immunology (Baltimore, Md. : 1950). 2001;166:3008-3018

457. Liesz A, Suri-Payer E, Veltkamp C, Doerr H, Sommer C, Rivest S, et al. Regulatory t cells are key cerebroprotective immunomodulators in acute experimental stroke. Nat Med. 2009;15:192-199

458. Shinar D, Gross CR, Price TR, Banko M, Bolduc PL, Robinson RG. Screening for depression in stroke patients: The reliability and validity of the center for epidemiologic studies depression scale. Stroke; a journal of cerebral circulation. 1986;17:241-245

459. Wardlaw JM, Ferguson KJ, Graham C. White matter hyperintensities and rating scales-observer reliability varies with lesion load. Journal of neurology. 2004;251:584-590

460. Kumar A, Bilker W, Jin Z, Udupa J. Atrophy and high intensity lesions: Complementary neurobiological mechanisms in late-life major depression. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology. 2000;22:264-274

461. Noonan K, Carey LM, Crewther SG. Meta-analyses indicate associations between neuroendocrine activation, deactivation in neurotrophic and neuroimaging markers in depression after stroke. Journal of stroke and cerebrovascular diseases : the official journal of National Stroke Association. 2012

462. Carota A, Berney A, Aybek S, Iaria G, Staub F, Ghika-Schmid F, et al. A prospective study of predictors of poststroke depression. Neurology. 2005;64:428-433

463. Noonan K, Carey, Pascoe M, Linden, Crewther S. Sustained inflammation 1.5 years post-stroke is not associated with depression in elderly stroke survivors. Clinical Interventions in Aging. 2013:69

464. Christensen H, Batterham PJ, Grant JB, Griffiths KM, Mackinnon AJ. A population study comparing screening performance of prototypes for depression and anxiety with standard scales. BMC medical research methodology. 2011;11:154

465. Brown C, Battista DR, Sereika SM, Bruehlman RD, Dunbar-Jacob J, Thase ME. Primary care patients' personal illness models for depression: Relationship to coping behavior and functional disability. General hospital psychiatry. 2007;29:492-500

466. Cornford CS, Hill A, Reilly J. How patients with depressive symptoms view their condition: A qualitative study. Family practice. 2007;24:358-364

467. Owe-Young R, Webster NL, Mukhtar M, Pomerantz RJ, Smythe G, Walker D, et al. Kynurenine pathway metabolism in human blood-brain-barrier cells: Implications for immune tolerance and neurotoxicity. Journal of neurochemistry. 2008;105:1346-1357

468. Erickson MA, Dohi K, Banks WA. Neuroinflammation: A common pathway in cns diseases as mediated at the blood-brain barrier. Neuroimmunomodulation. 2012;19:121-130

114 469. De Simoni MG, Milia P, Barba M, De Luigi A, Parnetti L, Gallai V. The inflammatory

response in cerebral ischemia: Focus on cytokines in stroke patients. Clinical and experimental hypertension (New York, N.Y. : 1993). 2002;24:535-542

470. Audet MC, Anisman H. Interplay between pro-inflammatory cytokines and growth factors in depressive illnesses. Frontiers in cellular neuroscience. 2013;7:68

471. Frisoni GB. Structural imaging in the clinical diagnosis of alzheimer's disease: Problems and tools. Journal of neurology, neurosurgery, and psychiatry. 2001;70:711-718

472. Lopez OL, Becker JT, Jungreis CA, Rezek D, Estol C, Boller F, et al. Computed tomography--but not magnetic resonance imaging--identified periventricular white-matter lesions predict symptomatic cerebrovascular disease in probable alzheimer's disease. Archives of neurology. 1995;52:659-664

473. Steptoe A, Hamer M, Chida Y. The effects of acute psychological stress on circulating inflammatory factors in humans: A review and meta-analysis. Brain, behavior, and immunity. 2007;21:901-912

474. Carpenter LL, Gawuga CE, Tyrka AR, Lee JK, Anderson GM, Price LH. Association between plasma il-6 response to acute stress and early-life adversity in healthy adults. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology. 2010;35:2617-2623

475. Mahoney FI, Barthel DW. Functional evaluation: The barthel index. Maryland state medical journal. 1965;14:61-65

476. Rankin J. Cerebral vascular accidents in patients over the age of 60. Ii. Prognosis. Scottish medical journal. 1957;2:200-215

477. Farrell B, Godwin J, Richards S, Warlow C. The united kingdom transient ischaemic attack (uk-tia) aspirin trial: Final results. Journal of neurology, neurosurgery, and psychiatry. 1991;54:1044-1054

478. Karege F, Perret G, Bondolfi G, Schwald M, Bertschy G, Aubry JM. Decreased serum brain-derived neurotrophic factor levels in major depressed patients. Psychiatry research. 2002;109:143-148

479. Shimizu E, Hashimoto K, Okamura N, Koike K, Komatsu N, Kumakiri C, et al. Alterations of serum levels of brain-derived neurotrophic factor (bdnf) in depressed patients with or without antidepressants. Biological psychiatry. 2003;54:70-75

480. Molendijk ML, Bus BA, Spinhoven P, Penninx BW, Kenis G, Prickaerts J, et al. Serum levels of brain-derived neurotrophic factor in major depressive disorder: State-trait issues, clinical features and pharmacological treatment. Molecular psychiatry. 2011;16:1088-1095

481. Cattaneo A, Gennarelli M, Uher R, Breen G, Farmer A, Aitchison KJ, et al. Candidate genes expression profile associated with antidepressants response in the gendep study: Differentiating between baseline 'predictors' and longitudinal 'targets'. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology. 2013;38:377-385

482. Mitoma M, Yoshimura R, Sugita A, Umene W, Hori H, Nakano H, et al. Stress at work alters serum brain-derived neurotrophic factor (bdnf) levels and plasma 3-methoxy-4-hydroxyphenylglycol (mhpg) levels in healthy volunteers: Bdnf and mhpg as possible biological markers of mental stress? Progress in neuro-psychopharmacology & biological psychiatry. 2008;32:679-685

483. Meng D, Wu T, Rao U, North CS, Xiao H, Javors MA, et al. Serum npy and bndf response to a behavioral stressor in alcohol-dependent and healthy control participants. Psychopharmacology. 2011;218:59-67

484. Eker C, Kitis O, Taneli F, Eker OD, Ozan E, Yucel K, et al. Correlation of serum bdnf levels with hippocampal volumes in first episode, medication-free depressed patients. European archives of psychiatry and clinical neuroscience. 2010;260:527-533

485. Molendijk ML, van Tol MJ, Penninx BW, van der Wee NJ, Aleman A, Veltman DJ, et al. Bdnf val66met affects hippocampal volume and emotion-related hippocampal memory activity. Translational psychiatry. 2012;2:e74

115 486. Kim JM, Stewart R, Kang HJ, Kim SY, Kim SW, Shin IS, et al. A longitudinal study of bdnf

promoter methylation and genotype with poststroke depression. Journal of affective disorders. 2013

487. Korn T, Petermann F. Development and function of interleukin 17-producing gammadelta t cells. Annals of the New York Academy of Sciences. 2012;1247:34-45

488. Korn T, Bettelli E, Oukka M, Kuchroo VK. Il-17 and th17 cells. Annual review of immunology. 2009;27:485-517

489. Bates S, Read SJ, Harrison DC, Topp S, Morrow R, Gale D, et al. Characterisation of gene expression changes following permanent mcao in the rat using subtractive hybridisation. Brain research. Molecular brain research. 2001;93:70-80

490. Raghavendra Rao VL, Bowen KK, Dhodda VK, Song G, Franklin JL, Gavva NR, et al. Gene expression analysis of spontaneously hypertensive rat cerebral cortex following transient focal cerebral ischemia. Journal of neurochemistry. 2002;83:1072-1086

491. Ma X, Reynolds SL, Baker BJ, Li X, Benveniste EN, Qin H. Il-17 enhancement of the il-6 signaling cascade in astrocytes. Journal of immunology (Baltimore, Md. : 1950). 2010;184:4898-4906

492. Martin B, Hirota K, Cua DJ, Stockinger B, Veldhoen M. Interleukin-17-producing gammadelta t cells selectively expand in response to pathogen products and environmental signals. Immunity. 2009;31:321-330

List of Publications and Abstracts

Papers in Preparation Bensimon K, Herrmann N, Ranepura N, Lanctôt KL. Elevated IFN-g, TNF-a, IL-18 and a reduction in the anti-inflammatory cytokine IL-10 is associated with post-stroke depression.

Bensimon K, Herrmann N, Gao FQ Ranepura N, Lanctôt KL. The neuroanatomical correlates of post-stroke depression.

Published Papers Elterman DS, Barkin J, Radomski SB, Fleshner NE, Liu B, Bensimon K, Arora S, Robinette

M, Finelli A. Results of high intensity focused ultrasound treatment of prostate cancer: early Canadian experience at a single center. Can J Urol. 2011 Dec; 18(6):6037-42.

Published Abstracts Elterman D, Barkin J, Fleshner N, Bensimon K, Liu B, Arora S, Robinette M, Finelli A, Klotz

L, and Radomski S. Investigational study for the treatment of localized (T1c/T2a) prostate cancer with high intensity focused ultrasound in Canada. Can Urol Assoc J. 2009 June; 3 (3 Suppl 1): S21-S57.

Elterman D, Fleshner N, Barkin J, Liu B, Bensimon K, Arora S, Robinette M, Finelli A, Klotz L, and Radomski S. Preliminary results of high-intensity focused ultrasound treatment of prostate cancer: early Canadian experience. Can Urol Assoc J. 2009 June; 3 (3 Suppl 1): S5–S20.

Presentations Bensimon K, Herrmann N, Ranepura N, Black S, Swartz R, Gladstone D, Gao FQ, Bayley

M, Lanctôt KL. The neurotrophic effects of lithium carbonate: a feasibility study. Poster presentation at Visions in Pharmacology, University of Toronto, June 6, 2012.

Tan DHS, Blitz S, Bensimon K, Collins M, Walmsley S. Serologic response to syphilis treatment among HIV-infected adults. Poster presentation at the 20th Canadian Conference on HIV/AIDS Research, Toronto, April 14-17, 2011.

Bensimon K, Tan DHS, Blitz S, Collins M, Walmsley S. Retrospective study of the serologic response to syphilis treatment among HIV-infected adults. Oral Presentation at the Ontario HIV Treatment Network Conference, Toronto, November, 2010.

Herschorn S, Radomski S, Liu B, Bensimon K, Gibbons E. Clinical and urodynamic findings

in post-radical prostatectomy incontinence. Poster Presentation at the 24th Annual European Urologic Association Congress, Stockholm, Sweden, March 20, 2009.

116

RE

".ikt

4!,i,!;'

, SunnvhroolcCEi\i-',tri.t,, r.ion.u, .i* r*u

gtrX I o ?!lt?Research Ethics Board (REB)

RENEWAL FORM

The Renewal Form is an application for continuing ethics approval and must be submitted forreview and approval prior to the study's expiry date. Ethics approval expires each subsequentyear from the day REB approval was initially granted unless otherwise indicated by theSunnybrook REB. Failure to submit this form prior to the expiry date signifies that the studydoes not have REB approval and all research activities must be suspended. Conductingresearch without REB approval may result in a notice of non-compliance involving correctiveaction, up to and including, termination of the research study.

Principal Investigator (Pl): Dr. Krista Lanctot

REB Project ldentification Number (PlN): 380-2004

Full Study Title: Post-Stroke Depression: The Role of Cytokine-Serotonin Interactions inTreatment Response

1. Date of initial Sunn08-Dec-2004

2.Type of REB review requested. (Final decision rests with the REB Chair.)

X Delegated Review E futt Board Review

3. ls this an Industry-Sponsored/Supported study?

! VfS (lf YES, complete the table below.) X ruO (lf NO, proceed to question 4.)

4. ls this study open for enrollment af Sunnybrook? | YES X NO

lf YES, attach a copy of the current Informed Consent Form(s).

I nvoici n g I nfo rmation for I nd ustry-S ponsored/S u pported Stud iesA fee of $500 Cdn is invoiced for all Industry-Sponsored/Supported Studies applying forcontinuino ethics aooroval.Invoice to the Followinq Companv:Contact Name:Telephone: E-mai l :Street Address: Suite:Citv: Province/State:Country: Postal/Zip Code.

Version Date: 25 March 2009 Page 1 of 3

Sunnybrook Health $ciences Centre REB - 2075 Bayview Avenue, Room C819, Toronto ON M4N 3MsTel: 416-480-6100 ext. 4?76 or 88144 Fax: 416-480-5385 Homepage: h-ttn:ilsunnvbroo!.glrllgqearchl?paqe=reohome

we had Planned to recruit' We are in

the process of completing the database.

R lJnur at Sun ook

Were consented100107107nlanla

CompteteO stuOy treatme nlaWithdrew congent 3

nla

ffichartrevie nla

6. Have all Serious Adverse Events (SAEs) experienced by a Sunnybrook participant been

reported to the REB?

f] VES I NO, will submit immediately f NO SAEs have occurred

7. In the opinion of the pl, is there a concern or trend in the SAEs that have occurred with

Su n nybrook ParticiPants?

Have all significant protocol deviations/violations been reported to the REB?

I VES I NO, will submit immediately f NO srgnificant deviations/violations to report

Since the last REB approval, is there any new ethical or scientif ic information outside of a

protocol amendment that would be relevant to the continuing review of this study?

EvrsXruo

8.

9 .

10. Since the last REB approval, is there any change in the conflict of interest information

provided to the REB for'any of the investigators, study staff or members of their immediate

fami ly?IvrsXruo

lf YES, provide details.-Version

Date: 2S march 2009 ' pagp Zof 3

Sunnybrook Health $ciences Centre RES - 2075 Bayview Avenue, Room C819, Toronto ON M4N 3M5

Tel: 416-489-6100 ext. 4276 or 88144 Fax: 416-480-5385 Homepage: http://sqnnvbrook.calresearchl?paseireohome

I Yes E ruo X xo SAEs have occurred

l f YES, provide details and action taken.

1. Person completinq this formTitle. Miss First Name: Russanthv Last Name: VelummailumDepUDiv:Neu roosvch p ha rmacoloov

Institution. Sunnybrook Research Institute

Full Address: 2075 Bayview Avenue, Toronto, ON M4N3M5

Room Number: FG-05

Telephone'. 41 6-480-6 1 00 Extension. 3185 E-mai l :rvelumm@sri. utoronto.ca

I

Version Date: 25 March 2009 Page 3 of 3

$unnybrook Health Sciences Centre RES - 2075 Bayview,Avenue, Room C819, Toronto ON M4N 3M5Tel: 416-480-8100 ext. 4276 or 88144 Fax: 416-4S0-538$ Homepage: http: i lsunnvbrook.calresearchl?paoe=reohome

I assume full responsibility for the scientific and ethical conduct of this study and agree toconduct this study in compliance with the Tri-Council Policy Statement: Ethical Conduct forResearch Involving Human Subjects (TCPS), Personal Health Information Protection Act(PHIPA) and any other relevant regulations or guidelines. I certify that all researchers andpersonnel involved in this study at this institution are appropriately qualified and trained tofulfi l l their role in this study.

---Wf,"i,re of Principal Investi

Research Ethics Office Use Only

The Sunnybrook REB has reviewed the informationhas obtained ethics approval by way of:

D, Delegated Review

I futt Board Review --' Date of Full Board meeting:

provided and confirms that this study

This study is only for the following period:

Phi l

Version Date: 02/MAR/2009

Stroke and Depression Study

Page 1 of 4

Stroke and Depression: The Role of Cytokine - Serotonin Interactions

Patient Information and Consent Investigators: Dr. K.L. Lanctôt Sunnybrook Health Sciences Centre

Dr. N. Herrmann Sunnybrook Health Sciences Centre Dr. S.E. Black Sunnybrook Health Sciences Centre Dr. D. Gladstone Sunnybrook Health Sciences Centre Dr. J. Ween Baycrest Dr. W. Goldstein York Central Hospital Dr. M. Waldman St. John’s Rehab

1. Information for subject: You are being asked to participate in a study conducted at Sunnybrook Health Sciences Centre under the supervision of the above investigators. Participation is voluntary and will involve the following:

2. Description and purpose of the trial: The purpose of this study is to evaluate determinants of the development of depressive and cognitive (memory and thinking) symptoms after a stroke. Both symptoms are common post-stroke, and may be related to levels of serotonin (an important brain chemical involved in regulating mood and thinking). We are interested in assessing the relationship between different types of cytokines (naturally produced inflammatory chemicals) and serotonin, and the role they play in any depressive or cognitive symptoms that you may or may not have. In addition, we are also interested in the impact of cytokines and other chemicals related to serotonin production on the size of the hippocampus (an important brain structure involved in regulating mood and thinking).

3. Study Details: Participation in this study will involve three separate visits. The first visit will occur soon (less than 4 weeks) after you experienced a stroke, and the second and third visits will be scheduled 6- and 12- weeks after the first. Each visit will last approximately 1½ hours, but can be broken up into several shorter (20-30 minute) assessments if needed. Each visit will involve the following:

Sunnybrook Health Sciences Centre 2075 Bayview Ave., Suite FG-05 Toronto, ON M4N 3M5

University of Toronto Faculty of Medicine 1 King’s College Circle, Toronto, ON M5S 1A8

Version Date: 02/MAR/2009

Stroke and Depression Study

Page 2 of 4

a) Assessments: The study coordinator will first meet with you and review your medical chart in order to assess your eligibility for the study. If you are eligible to participate, the study coordinator will then interview you using standard questionnaires that assess your mood, cognition and physical functioning. Certain details (e.g. medical history, demographic characteristics, current medications and details of your stroke) will be copied from your medical chart. Any CT and MRI scans conducted clinically during your hospital stay will also be analyzed to determine the characteristics of your stroke. Lastly, if you report experiencing significant depressive symptoms, the study physician will meet with you for further assessment and treatment, if necessary. This study will not interfere with your selection of treatment choice. However, information will be collected regarding your response to treatment through the questionnaires described above.

b) Blood Draw: A sample of blood will be drawn in order to measure levels of certain signaling molecules related to the serotonin and inflammatory systems (called cytokines, kynurenines and tryptophan). A total of 31 mL (2 tablespoons) of blood will be drawn.

c) Cheek Swab (1st Visit Only): A sample of skin cells from the inside of your cheek will be taken using a sterile cotton-tipped swab. The DNA inside these cells will be used for us to determine which forms of certain genes (“polymorphisms”) you have that are related to the cytokine, kynurenine and serotonin pathways. Your sample will be identified only by a unique number and will be destroyed once the genetic tests are complete.

d) MRI (Optional, 3rd Visit Only): MRI scanning is a method of making pictures of the brain using magnetic waves. This gives us information about the structure of the brain, including the hippocampus. The scan will take less than one hour. The scan will involve lying still with a simple device placed around your head/neck. The MRI machine looks like a long narrow tube. Even though the tube is open, some people feel confined in small places. If this bothers you, please notify us. You may end the scan at any time by telling the MRI staff, who will be able to see you, hear you, and communicate with you at all times. When MRI pictures are taken, it is normal for the MRI machine to make noises (banging and clicking). You will be asked to wear earplugs or headphones for your comfort during the exam. The scans performed in this study are not optimized to find clinical abnormalities. However, on occasion, a member of the research team may notice a finding on a scan that seems abnormal, in which case we will consult a neurologist as to whether the abnormality merits further investigation. If this is done, a member of the research team will contact you. The decision as to whether to proceed with further clinical examination or treatment will be yours; with your permission, we would forward your results to your family physician for further follow-up.

4. Benefits: You will not benefit directly from participation in this study.

5. Risks: When your blood is drawn, there may be some discomfort and/or bruising, however these are expected to be very mild.

The mouth swab is simple and painless.

Version Date: 02/MAR/2009

Stroke and Depression Study

Page 3 of 4

If you consent to the MRI—During the MRI procedure, you may be bothered by feelings of confinement (claustrophobia), and by clicking and banging noises made by the magnet. Because MRI uses strong magnetic fields, you may not participate in MRI scans if you have a pacemaker, an implanted defibrillator or certain other implanted electronic or metallic devices. An interview will be conducted prior to your MRI to advise the MRI staff if you have had brain surgery for a cerebral aneurysm, or if you have implanted medical or metallic devices, shrapnel, or other metal, such as metal in your eye. There is a small chance that we may observe something abnormal on your MRI. If this is the case, we will inform you, which may cause you anxiety, and suggest the need for further tests.

If you consent to the contrast medicine during MRI—The Omniscan injection is very safe. Rarely, the contrast medicine may leak out of the vein during the injection. If this does occur, there will be some swelling and tenderness, but this does not harm the tissue and the Omniscan is cleared naturally over the next day. Although uncommon, the Omniscan injection may cause a feeling of coolness in the arm or mild nausea in about 1 in 200 patients. Allergic reactions can occur, but they are extremely rare. These can include spasm of the airways, low blood pressure and asthma and only happen in about one in 10,000 cases. Severe reactions have been reported to occur at a rate of about one in 32,500.

6. Alternative Treatments: You are eligible to receive treatment for your stroke and any depressive symptoms you may have even if you choose not to participate in this study. Participation in this study will not affect your treatment in any way.

7. Costs: You will be given a $20.00 honorarium each time you visit Sunnybrook for the purposes of this study. If you participate in the MRI substudy, you will be given a $40.00 honorarium for your third visit. You will incur no costs as a result of participation in this study.

8. Participation/Termination: Your participation in this study is voluntary. Thus, if you do not wish to take part in this study or wish to withdraw at any time after commencing the study, your care will not be affected in any way.

You may be withdrawn from the study, at any point, if the investigator of this study considers it to be in your best interest. You may withdraw your consent at any point during the study.

9. Confidentiality: Your identity in this study will be treated as confidential. Certain Sunnybrook research staff, the Sunnybrook Research Ethics Board, and other agencies as required by law, may need to review your medical chart. We will have access to your medical chart for information on: blood pressure, heart rate, prescribed drugs, depressive symptoms and health status for 1 year. On all data collected for this study, you will be identified only by a unique number. If you disclose the intention to harm yourself or others, this information may not be kept confidential, as required by law.

10. Contacts: If you or your substitute decision maker have any questions about this study or for more information, you may contact the Study Co-ordinators: Philip Francis or Amy Wong (416-480-6100 x3185), Dr. Krista L. Lanctôt (416-480-6100 x2241) or Dr. Nathan Herrmann (416-480-6100 x6133).

If you have any questions about your rights as a research subject, you may contact Dr. Philip Hébert, the Chair of the Sunnybrook Research Ethics Board, at 416-480-4276.

Version Date: 02/MAR/2009

Stroke and Depression Study

Page 4 of 4

Stroke and Depression Study

Consent to Participate in this Study:

I, (patient’s name) __________________________________ have read the above information and fully understand the nature and the purpose of the study in which I have been asked to take part. The explanation I have been given has mentioned both the possible risks and benefits of the study. I understand that I will be free to withdraw from the study at any time without affecting my subsequent treatment by my doctor in any way. I voluntarily consent to participate in this study.

Consent to MRI Substudy

By placing my initials in the “no” box, I am stating that I do not consent to the MRI substudy.

By placing my initials in the “yes” box, I am stating that I do consent to the MRI substudy. However, I may still choose at any time not to complete the MRI scan, and still participate in the main study.

Consent to Contrast Injection

By placing my initials in the “no” box, I am stating that I do not consent to receiving the contrast medicine injection. However, if I placed my initials in the “yes” box above, I may still complete the MRI scan without the contrast medicine injection.

By placing my initials in the “yes” box, I am agreeing to receive the contrast medicine injection. However, I may choose at any time not to receive the contrast medicine, but still complete the MRI scan and/or participate in the main study.

_________________________________ Name of Patient (typed or printed) _________________________________ _____________________ Signature of Patient Date _________________________________ Name of Investigator (typed or printed) _________________________________ _____________________ Signature of the Investigator Date

Yes

No

Yes

No

124

APPENDIX 3: Study Roles

Study Concept and Design: Drs. Krista Lanctot, Nathan Herrmann

Data Acquisition: Amy Wong, Lana Rothenburg

Database Management: Kira Bensimon, Nipuni Ranepura, Russanthy Velummailum

Cytokine Assays: Dr. Angela Panoskaltsis-Mortari, University of Minnesota

Kynurenine Assay: Mr. Scott Walker, Sunnybrook Health Sciences Centre

Tryptophan Assay: Dr. Simon Young, McGill University

Lesion volume, Whole Brain Atrophy, Hippocampal Thickness, ARWMC scoring: Kira

Bensimon

Lesion Location and Laterality: Philip Lennox Francis

Data Analysis and Interpretation: Kira Bensimon, Drs. Krista Lanctot, Nathan Herrmann

Funding: A Heart and Stroke Foundation grant to: Drs. Krista Lanctot (PI), Nathan Herrmann

(Co-PI), Sandra Black, Demetrios Sahlas, David Gladstone

Study Supervision: Drs. Krista Lanctot, Nathan Herrmann

125

Appendix 4: Bivariate correlations between timing of assessment post-stroke and serum biomarkers

Serum Biomarker Raw Mean±SD *R ap-Value

K/T 70.4±38.4 0.158 0.158

IL-6 6.6±9.1 -0.144 0.202

IL-18 287.6±170.1 0.001 0.993

TNF-α 2.5±3.3 0.039 0.774

IFN-Ɣ 0.8±0.7 -0.071 0.533

IL-1β 0.4±0.5 0.121 0.389

IL-10 3.2±15.8 -0.068 0.547

IL-6/IL-10 26.3±76.1 -0.021 0.855

IL-18/IL-10 1054.9±1062.8 0.028 0.833

TNF-α/IL-10 8.4±10.7 -0.001 0.995

IFN-Ɣ/IL-10 2.7±3.7 0.061 0.592

IL-1β/IL-10 1.3±1.5 -0.002 0.990

*r values and their corresponding ap-values reflect the results of bivariate Spearman’s correlations with CES-D scores, units for cytokines are expressed in pg/ml and the K/T ratio is expressed in umol/L.

126

Appendix 5: Percent imputation of each cytokine

Cytokine Percent Imputed

IL-6 0%

IL-18 0%

TNF-α 27.1%

IFN-Ɣ 23.8%

IL-1β 5.6%

IL-10 58.0%

Appendix 6: Percent of cytokines imputed in each cytokine group and Pearson’s Chi-squared test between imputed and non-imputed cytokines and CES-D groups

Cytokine

CES-D ≥ 16 CES-D < 16 *Χ2 ap-Value

IL-6 None imputed

IL-18 None imputed

TNF-α 11.8% 33.3% 2.848 0.091

IFN-Ɣ 24.0% 23.6% 0.001 0.972

IL-1β 6.7% 5.1% 0.049 0.825

IL-10 57.7% 58.2% 0.002 0.967

* Χ2 values and their corresponding ap-values reflect the results of Pearson’s Chi-squared tests between imputed and non-imputed cytokines and CES-D groups.