Downloaded from UvA-DARE, the Institutional Repository of the University of Amsterdam (UvA)
Final published version (publisher's pdf)
SOURCE, OR PART OF THE FOLLOWING SOURCE:
Title Dose-escalation in the picture : pharmacological and imaging studies in depression
Faculty Faculty of Medicine
FULL BIBLIOGRAPHIC DETAILS:
It is not permitted to download or to forward/distribute the text or part of it without the consent of the copyright holder
(usually the author), other then for strictly personal, individual use.
UvA-DARE is a service provided by the Library of the University of Amsterdam (http://dare.uva.nl)
Dose-escalation in the picture
Pharmacological and Imaging studies in depression
Dose-escalation in the picture
Dose-escalation in the picture
Omslag Ruhe.indd 117-9-2008 15:56:58
DOSE-ESCALATION IN THE PICTURE
PHARMACOLOGICAL AND IMAGING STUDIES IN DEPRESSION
The work described in this thesis was mostly performed at the Academic Medical Center, Program
for Mood Disorders in the Netherlands. The studies were conducted in close collaboration with
the departments of Nuclear Medicine, General Practice, Pharmacology and Pharmacotherapy,
Clinical Epidemiology, Biostatistics and Bioinformatics and Radiology.
Most of the studies in this thesis were financially supported by grants from the Academic Medical
Center (SFA.07.012), the Netherlands Organisation for Health Research and Development
(ZonMw), program Mental Health, education of investigators in mental health (OOG; #100-002-
002), and from the Dutch Brain Foundation (14F06.45).
Additional support for the completion of this thesis was provided by Amicale Facel Holland.
Dose-Escalation in the Picture. Pharmacological and Imaging studies in depression
Thesis, University of Amsterdam, The Netherlands. With a summary in Dutch.
Copyright © 2008, Henricus G. Ruhé, Amstelveen, The Netherlands H.G.Ruhe@AMC.UvA.NL
All rights reserved. No part of this thesis may be reproduced, stored or transmitted in any form or
by any means, without prior permission of the author.
Layout:ProBook with FLE technology by GDS Cross Media Group BV, Nieuw Vennep, The
Cover: Inge E. Kos, Medical Photography and Illustration Academic Medical Center, The
Printed by : PrintPartners Ipskamp BV, Enschede, The Netherlands
Dose-Escalation in the Picture
Pharmacological and Imaging studies
ter verkrijging van de graad van doctor
aan de Universiteit van Amsterdam
op gezag van de Rector Magnificus
prof. dr. D.C. van den Boom
ten overstaan van een door het college voor promoties
in het openbaar te verdedigen in de Aula der Universiteit
op woensdag 5 november 2008, te 10:00 uur
Henricus Gerardus Ruhé
geboren te Amsterdam
Promotor: Prof. dr. A.H. Schene
Co-promotor:Dr. J. Booij
Overige leden:Prof. dr. D.A.J.P. Denys
Prof. dr. R. Dierckx
Prof. dr. B.L.F. van Eck-Smit
Prof. dr. C.G. Kruse
Prof. dr. W.A. Nolen
Faculteit der Geneeskunde
‘The brain is not a chemical factory but an extremely complicated survival machine’
Arvid Carlsson, Nobel lecture 2000
Part I. General introduction9
1. General introduction and outline of the thesis11
2. Design and methods of the DELPHI-study 31
Part II. Studies to guide clinical treatment of Major Depressive
3. Clinical use of the Hamilton Depression Rating Scale: Is
increased efficiency possible? A post hoc comparison of HDRS,
Maier and Bech subscales, CGI and SCL-90 scores.
Comprehensive Psychiatry 2005; 46: 417-427
4. Dose-escalation for insufficient response to a standard dose of
a selective serotonin reuptake inhibitor in major depressive
disorder: a systematic review.
British Journal of Psychiatry 2006; 189: 309-316
5. Switching antidepressants after a first selective serotonin
reuptake inhibitor in major depressive disorder:
a systematic review
Journal of Clinical Psychiatry 2006; 67: 1836-1855
Part III. Serotonin in the etiology of Major Depressive Disorder 99
6. Mood is indirectly related to serotonin, norepinephrine and
dopamine levels in humans: a meta-analysis of monoamine
Molecular Psychiatry 2007; 12: 331-359
7. Serotonin transporter binding with [123I]β-CIT SPECT in major
depressive disorder versus controls: effect of season and
Part IV. Neurobiological effects of paroxetine in the treatment
of Major Depressive Disorder155
8. Serotonin transporter gene promoter polymorphisms modify
the association between paroxetine serotonin transporter
occupancy and clinical response in major depressive disorder
Pharmacogenetics and Genomics 2008; in press
9. Evidence why paroxetine dose-escalation is not effective in
major depressive disorder: a randomized-controlled trial with
assessment of serotonin transporter occupancy
Neuropsychopharmacology 2008; in press
10.Successful pharmacological treatment of major depressive
disorder attenuates amygdala activation to negative facial
expressions. An fMRI study
11. Effect of the selective serotonin reuptake inhibitor paroxetine
on platelet function is modified by a SLC6A4 serotonin
Journal of Thrombosis and Haemostasis 2008; in press
Part V. Summary and general discussion231
12. Summary, conclusions and general discussion233
13. Samenvatting en conclusies253
Part VI. Abbreviations, color figures, publications,
acknowledgement and curriculum vitae269
GENERAL INTRODUCTION AND
OUTLINE OF THE THESIS
General introduction and outline
The present thesis concerns the pharmacological treatment of Major Depressive Disorder (MDD).
It especially focuses on what to do when patients do not respond to a standard dose of an
In this thesis, the results of studies that were performed since 2001, while working at the
Program for Mood Disorders of the department of Psychiatry of the Academic Medical Center will
be reported. These studies first aimed at systematically reviewing the existing literature about
treatment strategies for non-response to the first antidepressant (included in Part II). These
reviews identified an important gap between equivocal evidence for dose-escalation and the firm
recommendation of this strategy in clinical guidelines. This was the starting point for the
experimental studies to further investigate the clinical efficacy and the mechanism behind dose-
escalation, described in Part IV. Meanwhile, the findings, and further reading of conflicting results
of others raised questions about the pathophysiological origin of MDD. Therefore, the generally
known serotonin hypothesis for MDD, will be addressed in two additional studies (Part III). Part V
of the thesis comprises the general discussion and will address directions for future research.
In this introduction chapter, the reader will be introduced to some backgrounds of MDD, its
pharmacological treatment, the monoamine hypothesis about the etiology of MDD, and briefly on
the mechanisms of action of antidepressant drugs. After a summary, the separate studies of the
thesiswill be introduced. In chapter 2, the methods of the Dose-Escalation Legitimate?
Pharmacology and Imaging studies in depression (DELPHI-study) will be described. However, let’s
start with some history.
MELANCHOLY, the subject of our present discourse, is either in disposition, or habit. In disposition
is that transitory melancholy which comes and goes upon every small occasion of sorrow, need,
sickness, trouble, fear, grief, passion or perturbation of the mind, any manner of care, discontent or
thought, which causeth anguish, dullness, heaviness and vexation of spirit, any wayes opposite to
pleasure, mirth, joy, delight, causing forwardness in us, or a dislike. In which equivocal and improper
sense, we call him melancholy, that is dull, sad, sowr, lumpish, ill disposed, solitary, any way moved or
displeased. And from these melancholy dispositions, no man living is free. […] Melancholy, in this
sense, is the character of mortality. […]
It falleth out oftentimes that these dispositions become habits, and many affects contemned
make a disease. […]for that which is but a flea-biting to one, causeth unsufferable torment to
another; and which one by his singular moderation and well composed carriage can happily
overcome, a second is no whit able to sustain; but upon every small occasion of mis-conceived abuse,
injury, grief, disgrace, loss, cross, rumour, etc. (if solitary, or idle) yields so far to passion, that his
complexion is altered, his digestion is hindered, his sleep gone, his spirits obscured, and his heart
heavy, his hypocondries mis-affected; wind, crudity, on a sudden overtake him, and he himself
overcome with melancholy. […] If any discontent seise upon a patient, in an instant, all other
perturbations will set upon him; and then, like a lame dog or broken-winged goose, he droops, and
pines away, and is brought at last to that ill habit or malady of melancholy it self. […]
But all these melancholy fits, […] displeasing, violent and tyrannizing over those whom they
seise on for the time–yet these fits, I say, or men affected, are but improperly so called, because they
continue not, but come and go, as by some objects they are moved. This melancholy, of which we are
to treat, is an habit, morbus sonticus, or chronicus, a chronick or continuate disease, a settled
humour, not errant, but fixed; and as it was long increasing, so, now being (pleasant or painful)
grown to an habit, it will hardly be removed. […]
INVETERATE melancholy, howsoever it may seem to be a continuate, inexorable disease, hard to
be cured, accompanying them to their graves most part, yet many times it may be helped, even that
which is most violent, or at least it may be mitigated and much eased. Nil desperandum. It may be
hard to cure, but not impossible for him that is, most grieviously affected, if he be but willing to be
Burton’s work The anatomy of melancholy* (first published in 1621) was a best-seller.1 It
describes many aspects of the disease that we currently recognize as depression:2 the sometimes
difficult distinction of the illness melancholy (depression) from ‘normal’ sadness, pain or sorrow
after misery in life, the relation between stressful life-events and the development of the illness,
the interaction of life-events and coping styles to develop the illness or remain well, the clinical
symptoms, the enormous impact of the illness on a depressed person’s life, social dysfunction and
the tendency of recurrence and chronicity.1
Nowadays, MDD is one of the most prevalent and disabling illnesses in psychiatry, after
ischemic heart disease the second most common cause of disability worldwide and expected to
be the world’s second cause of disability by 2030, behind HIV/AIDS.3 MDD is diagnosed by the
occurrence of a cluster of different symptoms affecting mood, pleasure, attention, activities, vital
somatic functions (eating, sleep), and (ruminative) thoughts over oneself, guilt or suicide over a
prolonged period of time. For clinical and scientific uniformity, MDD is currently defined by
criteria specified in the Diagnostic and Statistical Manual (4th edition; DSM-IV),4 which is a non-
Epidemiology of MDD
The 12-month prevalence of MDD is estimated to be 5.8% in the Netherlands,5 and 5.3% in the
United States.6 Prevalence rates are two times higher in women,5;7-10 and higher in widowed,
divorced, unemployed or disabled people, people with somatic disease and first degree relatives
of patients with MDD.6;9 In the general Dutch population, the median duration of MDD is 3
months, with 63% of new MDD episodes recovering within 6 months (regardless whether
treatment is offered or the natural course of the disease is awaited), however, this prognosis is
less favourable for people who seek help.11 Estimations of the lifetime prevalence are 15.4% in the
Netherlands,5 >16% in Europe,8 and 13.2%-16.2%6 in the USA, with the same 2:1 female-male
distribution. Most epidemiological studies indicate that nowadays 51-69% of depressed patients
seek or receive treatment,8;9 which has increased over the last decade relative to the nineties of
the previous century.10 However, only 50% of patients who are treated for MDD meet diagnostic
criteria of MDD or meet indication criteria for treatment,10;12 and it is estimated that only 21.7% of
MDD-patients in the USA receive adequate treatment.9 Therefore, depending on one’s view, MDD
is frequently over-treated, under-treated or mis-treated.
MDD is a recurrent and potentially chronic disease. After the first episode, 50-80% of patients
have a recurrence within 10 year.13;14 Of incident episodes of MDD 15-20% will become a chronic
depression (defined as a duration of ≥2 years), although higher percentages (46%) have been
MDD is associated with high direct treatment costs as well as indirect costs of loss of
productivity and quality of life.8;17 In the Netherlands, treatment costs of MDD (€250 million) in
1994 were estimated to comprise 1% of the total healthcare budget, and 13.6% of the total mental
health budget.18 In 2003, MDD treatment costs in the Netherlands had increased to €660 million,
still consuming 1.1% of the total healthcare budget.19 In 1999 10% of newly assigned disability-
payments (WAO) in the Netherlands was due to MDD.20
Treatment of MDD
In order to treat major depressive disorder effectively, many national clinical guidelines were
developed.13;21-30 Besides various forms of psychotherapy, pharmacotherapy is often applied as
monotherapy or in combination with psychotherapy. Antidepressants are effective drugs for
MDD when compared to placebo, although effect sizes are moderate,24;26;31-33 and some critics
claim that effects are exaggerated by pharmaceutical companies.34-37 Antidepressants are
prescribed for MDD, but also for other psychiatric disorders, and are of considerable financial
interest. In 2006, approximately 780.000 inhabitants (4.7%) of the Netherlands used
*Melancholy is derived from the Greek Μελαγχολια (melancholia), which refers to the black choler (bile) that was seen as
the material cause of the illness those days.
General introduction and outline
antidepressants, mostly Selective Serotonin Reuptake Inhibitors (SSRIs).
Annual prescriptions of antidepressants increased over the last decade by ~6% every year. The
costs of prescribed antidepressants were €156 million in 2006,38 again, most costs represent the
prescription of SSRIs.
A range of antidepressants have been developed, all aiming at the stimulation of serotonergic
and/or noradrenergic neurotransmission.39;40 Four classes of antidepressants exist: SSRIs, Tricyclic
antidepressants (TCAs), Monoamine-oxidase inhibitors (MAOIs) and a miscellaneous group. TCAs
and SSRIs, but also ‘dual action’ antidepressants block the transporters at the pre-synaptic nerve
ending, which causes inhibition of the reuptake of either serotonin (5-HT), noradrenaline (NA; =
norepinephrine), or both. The classic MAOIs irreversibly inhibit the enzymes mono-amine oxidase
A and B, which degradeserotonin, noradrenaline and dopamine. The miscellaneous group
harbours ‘dual action’ antidepressants (e.g. venlafaxine or duloxetine), and antidepressants with
pre- and post-synaptic targets different from serotonin or noradrenaline transporters.39;40
By nature of their largely similar, but also slightly distinct pharmacological affinities for
transporters and receptors, antidepressants have (almost) comparable efficacy,32;41-43 but are
different in their adverse effects. Because SSRIs are tolerated better than TCAs,32;44;45 SSRIs have
become the antidepressants of first choicein most countries.38 Within their class, different SSRIs
also have comparable efficacy, but different side effect profiles.46;47
Achieving remission is the aim of treatment,because persistence of (residual) symptoms
increases the risk of a future relapse or recurrence and is associated with ongoing psychosocial
dysfunction.48;49 If remission is achieved, patients are advised to continue their antidepressants
(at least 6 months) in order to prevent relapse,21;50 although guidelines’ recommendations are
inconsistent.13;21;23-25 Nevertheless, only 50% of the patients respond (defined as a 50% decrease in
depression severity)40 to the initial antidepressant, and remission rates (defined as a depression
severity score below a certain cut-off) are lower (28-35%).28;29;51 Therefore, clinicians who treat
MDD are often confronted with non-response, for which evidence-based approaches are
Insufficient response to antidepressants: A challenge for the clinician.
The clinician faces several dilemmas when a patient shows insufficient response to an
antidepressant. First, how does one ascertain clinical improvement?
A variety of measurement scales is available to measure depression severity and improvement
of therapy.53 Psychopathology rating scales with clear cut-off scores provide a more direct
representation of the effects of treatment than quality of life scales or a functional outcome.
Thus, the use of a symptom-rating scale is highly recommended to assess treatment effects,
either clinician-rated (e.g. the Hamilton Depression Rating Scale (HDRS),54 Montgomery Åsberg
Depression Rating Scale (MADRS)55 or Inventory for Depressive Symptoms (IDS-C))56 or self-
rated(Inventory for Depressive Symptoms (IDS-SR)56, Quick-IDS (QIDS-SR)57 or Beck Depression
Inventory (BDI)).58 Despite criticism on internal validity, selection and rating of items, and
sensitivity to change,59-64 most scales correlate with each other.57;65 Frank et al.66 proposed
tentative definitions for response (≥50% decrease of pretreatment rating scale scores) and
remission (e.g. HDRS ≤7, MADRS ≤7, IDS-SR ≤13, BDI ≤9). Therefore, measurements should be
routinely performed before the start of treatment and at consecutive critical decision points.67;68
However, scales are not routinely implemented in daily practice. Apparently, the current
challenge of implementation is to persuade clinicians to measure MDD severity repeatedly with
The next dilemma is when to measure the effects of the antidepressants. Although a recent
meta-analysis indicated that the effects of antidepressants on depressive symptoms can be
observed within the first weeks of treatment,69 patients may need 6-10 weeks to achieve
substantial improvement.51;70 Therefore, the initiated treatment should not be changed too early,
neither too late. Changing a regimen early may forestall a potentially late effect of the current
treatment, but may also offer an increased chance of response. Contrary, unchanged treatment
may increase despair due to the failure of response,but may prevent the (possible premature)
start of a more aggressive treatment with increased side effects. Over the last decades, the time
to declare that an antidepressant trial failed, increased from 3 weeks71 to 6-8 weeks.70;72
Interestingly, recommendations on this lag-time were never derived from studies primarily
designed to quantify the timing of the determination of response. Current guidelines for
depression differentially recommend a minimal duration of antidepressant therapy between 4
and 6 weeks.13;21;23;25;26;28;29;67;73 At these time points (‘critical decision points’)67 a next treatment-
strategy must be considered.
A final dilemma is to adequately re-evaluate the diagnosis, without the (counter-) transference
of helplessness. Critical decision points are important for the reassessment of somatic or
psychiatric co-morbidity, re-evaluation of persistent psychosocial problems and treatment
adherence. However, when none of these issues apply, the patient may be accused of ‘a failure to
cooperate’, or blamed for the non-response by ‘having a personality disorder’ or ‘not being
motivated for treatment’. Although these reproaches occasionally may be justified, they could
also indicate the patient’s and/or clinician’s helplessness, which is often transferred from a
depressed person to the persons surrounding him or her. Most depressed patients want to
improve, but are indeed disabled in their coping-styles, their motivation and their problem solving
capacities. As such, routinely administered behavioural interventions (registration, planning and
gradual reactivation)74 are valuable in the pharmacological treatment of MDD, also deferring
Insufficient response to antidepressants: What to do when the miracle doesn’t
Five pharmacological strategies for insufficient response to an antidepressant can be distin
guished: 1. prolongation, 2. dose-escalation, 3. switching, 4. augmentation and 5. combina tion, all
regarding the antidepressant initially given. Of course, a sixth strategy could be the addition of
psychotherapy to the monotherapy of pharmacotherapy,75;76 but as this is not a topic of this
thesis, only the first 5 strategies are described.
1 Prolongation of the trial for another 2-4 weeks. This strategy is applicable only in case of doubt
of the nature of the non-response (e.g. amendable psychosocial stressors, non-adherance),
but is not favourable given the relative importance assigned to critical decision points.
2 Dose-escalation. With this strategy the dose of the prescribed antidepressant is increased to
the maximal tolerable dose. Dose-escalation assumes a dose-response relationship, which is
equivocal for SSRIs. The advantage of dose-escalation is that it is easy and quick to apply.
3 Switching of the antidepressant. With this strategy the prescribed antidepressant is
discontinued (eventually after tapering and a wash-out period) and the antidepressant is
switched to another antidepressant either belonging to the same class, or to a different class
compared to the initial antidepressant. The obvious advantage of switching is that it facilitates
a new pharmacological approach, possibly targeting different neurotransmitter systems. Its
disadvantage is that it requires some time, and that one may unwontedly discard achieved
improvement when the chosen antidepressant appears to be a less effective one for this
4 Augmentation of the prescribed antidepressant. With this strategy a drug is added that has no
strong antidepressant effects by itself, but is known to increase the effects of antidepressants
(e.g. lithium). The advantage of augmentation is that the effect of the antidepressant initially
prescribed is maintained. Furthermore augmentation might result in faster responses than
switching (as tapering is not required, nor a wash-out). Disadvantage is the risk of poly-
pharmacy and the related interactions.
5 Combination of antidepressants. With this strategy another antidepressant drug, mostly with
different pharmacological properties compared to the antidepressant already prescribed, is
added to the initial antidepressant. This strategy has the same advantages and disadvantages
as augmentation strategies, but the risks of severe interactions are higher.
General introduction and outline
Neurotransmission, the serotonergic system and neuronal networks
The brain passes information through electrical signals in neurons. Synapses form the functional
contacts between neurons and mainly communicate through neurotransmitter secretion.
Neurotransmitters are stored in synaptic vesicles. The vesicles are released into the synaptic cleft
in response to presynaptic depolarisation. After the release of neurotransmitters in the synapse,
they are either metabolized or transported back into the terminal to be used again.77
Neurotransmitters bind to specific receptor proteins on the membrane of the pre- and
postsynaptic neurons. Postsynaptically, receptors either induce a change in postsynaptic ion
channels (ionotropic receptors), causing a depolarisation, or they activate G-protein mediated
complexes (metabotropic receptors), which activate one or more metabolic steps (‘second
messengers’). Activation of metabotropic receptors is generally responsible for forming enzymes
that regulate gene expression, neurotransmitter synthesis, receptors and neuroplasticity. Which
of the mechanisms is used depends on the particular postsynaptic receptor type.78 Three major
‘monoamine’ neurotransmitters are associated with psychiatric disorders: serotonin,
noradrenaline and dopamine, of which the latter is traditionally related to psychotic disorders.
SSRIs target the serotonergic system and increase serotonergic neurotransmission. From here,
the serotonergic system and the neuronal networks that are believed to be involved in MDD will
The serotonin neurons of the brain start in the raphe nuclei in the midbrain and project to the
neocortex, basal ganglia, temporolimbic zones, hypothalamus, cerebellum and the brain stem
(Figure 1.1).80;81 Serotonin is involved in several functions: sleep and wakefulness, appetite, nausea,
migraine, headaches and regulation of mood.77 The serotonin receptors comprise of 5-HT1A, 1B,
1C, 1D, 1E, 1F, 5-HT2A, 2B, 2C, 5-HT4, 5-HT5, 5-HT6 and 5-HT7 (metabotropic) and the 5HT3 (ionotropic)
receptors.78 Serotonin plays an important role in brain development via regulation of neurite
outgrowth, synaptogenesis and cell survival.82;83 Serotonin that is released into the synaptic cleft
is either taken up back into the presynaptic nerve ending by the serotonin transporter (SERT), or
degraded by MAO-A.77;78
The serotonin (and noradrenaline) deficiency hypothesis
In the late fifties of the previous century, TCAs appeared to be effective in treating MDD by
increasing serotonergic and noradrenergic neurotransmission. This discovery led to the
monoamine hypothesis: MDD might etiologically be explained by a deficiency in monoamine
neurotransmitters: serotonin ornoradrenaline. Since then, the working mechanism of AD is
believed to be by (1) increased neurotransmission by increased synaptic levels of serotonin,
noradrenaline and/or (2) specific agonistic effects on serotonin or noradrenaline (sub-)receptors.
Figure 1.1. Serotonergic pathways through the human brain.
Serotonergic system. From the raphe nuclei in the midbrain, neurons project to the neocortex, basal ganglia, temporolimbic
zones, hypothalamus, cerebellum and the brain stem.
Depletion of the available serotonin andnoradrenaline is used as a model to test the
involvement of monoaminergic systems in MDD. Serotonin depletion can be achieved by rapidly
lowering the essential amino-acid tryptophan which cannot be synthesized by the body and must
be ingested to enable formation of serotonin. To achieve depletion, a tryptophan free amino-acid
mixture is administered (acute tryptophan depletion).84 Depletion ofnoradrenaline and
dopamineoccurs simultaneously, and uses the same concept (acute depletion of the essential
amino-acids phenylalanine and tyrosine).85 As an alternative to induce a state of depletion,
enzyme-blocking agents decrease the production of the monoamines. Para-chlorophenylalanine
blocks serotonin synthesis,86 and Alpha-methyl-para-tyrosine blocks noradrenaline and dopamine
Since 1975 an increasing number of depletion studies have been conducted, with different
effects in different study-populations. In general in healthy controls no clear mood-effects were
found, unless they had relatives with MDD. In remitted MDD patients who used antidepressants
(or shortly after tapering) approximately 50% of the patients experienced a relapse after
depletion. However this occurred only after depletion of the monoamine that their
antidepressant targeted. In depressed patients no consistent deterioration of the mood effects
Thus, the monoamine-deficiency theory, in its purest form, states that depression can be
cured by the increase of serotonergic and/or noradrenergic neurotransmission. However, the
reverse train of thought, that depression is bio-etiologically caused by a deficiency of monoamines
(e.g. serotonin and/or noradrenaline) has attractive face-validity, but probably is an untenable,
superficial simplification.75;94 Therefore, the current, less pertinent view is that the monoamine
hypothesis only partially explains MDD and the response to AD.95-98
The limbic-cortical dysregulation hypothesis
In a more multidimensional, systems-level model, MDD can be viewed as a disorder affecting
discrete but functionally integrated pathways; neural networks, which can be identified by
neuroimaging techniques.† In such a network, dysfunction in one or more of the elements (e.g.
after cognitive or somatic stress), will initially be tried to be influenced (or compensated) by
other, remaining parts of the network, that try to maintain homeostatic emotional control.
Therefore, results from neuroimaging studies investigating differences between healthy controls
and MDD patients, must be considered as the identification of regions to be either etiologically
abnormal or regions involved in (mal-)adaptive compensatory processes.99
Because MDD is an affective disorder, the neurobiology of emotion processes is likely
involved. For the processing of emotions two systems are important: a ventral system (consisting
of amygdala, insula, ventral striatum, ventral anterior cingulate gyrus, and ventral prefrontal
cortex) and a dorsal system (consisting of hippocampus, dorsal anterior cingulate gyrus, and
dorsal prefrontal cortex). The ventral system serves to identify the emotional significance of a
stimulus, the production of mood states, and automatic regulation of emotional responses, while
the dorsal system serves to effortfully regulate mood states and subsequent behaviour.100
Initial lesion-deficit studies, early Positron Emission Tomography (PET) studies (measuring
regional resting state glucose metabolism or blood flow) and later functional Magnetic
Resonance Imaging (fMRI)-studies identified several brain regions to be affected by MDD: the
limbic structures (amygdala, hippocampus, hypothalamus and brainstem), the subcortical (basal
ganglia and thalamus) and the cortical (dorsolateralprefrontal, ventrolateral prefrontal and
orbitofrontalcortex (DLPFC, VLPFC, OFC respectively) structures. In these, the most consistent
finding is a hypoactivity of the (dorsal) frontal lobe, while often a hyperactivity in the (ventral)
VLPFCand OFC is found.99;101 Additionally, an increased activity of the (rostral, subgenual) anterior
cingulated gyrus99;101 and the amygdala, anterior insula, and ventral striatum was found, although
less consistently.101 Furthermore, fMRI studies point to a increased sensitivity of the amygdala for
† Single Photon Emission Computed Tomography (SPECT) and Positron Emission Tomography (PET) use a variety of
radioligands with various half-lives to quantify different targets (transporters, receptors, or blood-flow or metabolism).
Functional Magnetic Resonance Imaging (fMRI) is a technique to image brain activity (hemodynamic response) related to
a specific task or stimulus.
General introduction and outline
emotional stimuli, and a bias to interpret these stimuli in a negative context.101 Interestingly,
induction of sad mood in healthy volunteers also produces increased blood flow in the insula,
subgenual anterior cingulate gyrus and decreased blood flow in the dorsomedial prefrontal
The above abnormalities were combined in the limbic-cortical dysregulation model, which
includes a dorsal neocortical hypofunction, which results in ventral (para)limbic hyperactivity,
with a reciprocity in this inverse relationship.103;104
Which effects in the brain cause antidepressants to be antidepressants?
The various treatment forms available to relieve MDD and their moderate efficacy poses the
question: How do antidepressants bring about their therapeutic effects? For this question, three
levels of action must be distinguished: 1. direct neurotransmission effects, 2. second messenger
effects, and 3. change in neuronal networks. For levels 1. and 2. see Figure 1.2. At all three levels,
there appears to be a time dependent differentiation of the effects that occur, e.g. not all effects
are the same in the consecutive weeks after the initiation of treatment. This may explain why it
sometimes takes several weeks before the therapeutic effects become apparent. Because most
of the recent research was done with SSRIs and SNRIs, the effects of these drugs are described
hereafter, unless indicated differently.
Direct effects of antidepressants on neurotransmission
When an SSRI is ingested, the blockade of the target transporter (i.e. the serotonin transporter
(SERT)) occurs within several minutes.105;106 Therefore, the antidepressant effects cannot only be
based on increased neurotransmission by decreased serotonin reuptake, as these effects take
much longer than this direct pharmacological effect. Further research revealed that after two
days of treatment, the firing-rate of SERT-containing neurons in rats decreased, but that this
firing-rate was restored within 2 weeks of continued treatment. This was attributed to
somatodendritic 5-HT1A autoreceptors, which normally have a negative feedback on the neuron’s
firing rate. These 5-HT1A autoreceptors appeared to desensitize.105 Microdialysis-experiments in
rats showed that the restored firing-rate after 14 days was responsible for a 6-fold increase in
intrasynaptic serotonin level, while after the acute blockade of the SERT this increase of the
serotonin level was only small and transient.107 Furthermore, in humans, serotonin autoreceptors
(5-HT1B/1D) (and also noradrenergic α2 autoreceptors) in the synapse normally inhibit serotonin
release by feedback-mechanisms as well. Prolonged treatment with SSRIs again desensitize these
receptors, resulting in increased serotonin release in the synaptic cleft.105;108 As such,
desensitisation of HT1A autoreceptors in the raphe nuclei in the midbrain may have effects on
serotonergic neurotransmission in critical brain areas where these serotonergic neurons project
to.109 Finally, in rats, after prolonged administration of SSRIs the SERT itself is downregulated by
80-90%.110-112 This is probably caused by trafficking and internalization of the SERTs instead of
altered SERT-gene-regulation, because mRNA expression in the cells studiedwas unaltered by the
Other antidepressants have rather different effects. TCAs (except clomipramine) do not
change the pre-synaptic serotonin-containing neurons, but appear to sensitize the postsynaptic 5-
HT receptors for serotonin.105 Along with serotonin, the responsiveness for noradrenaline was
also found to be enhanced, likely due to enhanced α-adrenoreceptor-mediated transmission.
MAOIs also increase serotonergic transmission by desensitisation of 5-HT1A autoreceptors, but do
not desensitize other 5-HT autoreceptors, and desensitize noradrenergic α2 autoreceptors, which
indirectly enhances serotonergic transmission.
Figure 1.2. Transporters, receptors and second messenger systems involved in the effects of antidepressants.
Figure adapted from Belmaker and Agam.79 The left half of the presynaptic neuron represents a serotonergic neuron, the
right half a norepinephrinergic neuron. For color figure see page 277.
In the presynaptic neuron, serotonin is synthesized from tryptophan by tryptophan hydroxylase and stored in vesicles.
Likewise, norepinephrine is synthesized from tyrosine by tyrosine hydroxylase. These vesicles merge with the cell membrane
when the neuron is depolarized, thereby releasing their contents into the synaptic cleft.
After release, serotonin and norepinephrine are transported back into the presynaptic neuron by serotonine and
norerepinephrine transporters. Furthermore, serotonin and norepinephrine are catabolized by the monoamine-oxidase A
(MAO-A) enzyme. In the synaptic cleft, serotonin and norepinephrine affect both the pre-and post-synaptic neuron. The pre-
synaptic 5-HT1A and 5-HT1B auto-receptors decrease serotonin release by inhibitory feedback; the α2-adrenergic receptor
does the same for the release of norepinephrine.
Post-synaptically, serotonin and norepinephrine bind to G-protein-coupled monoamine receptors (MARs): the cyclicAMP
(cAMP)-coupled receptor, which activates protein kinase A (PKA), and the Phosphatidylinositol (PI)-coupled receptor, which
activates phospholipase C (PLC) which thereafter form inositol triphosphate (IP3) and diacylglycerol (DAG). IP3 and DAG
activate protein kinase C (PKC). Both PKA and PKC finally activate cAMP responsive element binding (CREB) protein, which
stimulates DNA transcription. For example, this might result in the production of brain derived neurotrophic factor (BDNF).
General introduction and outline
Second messenge reffects of antidepressants
The secondary effects of antidepressants can be divided in post-synaptic effects on thesensitivity
and availability of receptors, and the effects caused by the activation of post-synaptic receptors.79
Effects on post-synaptic receptors include a decrease of 5-HT1A and 5-HT2Areceptor density,113 and
by using indirect measures (e.g. growth-hormone release), it is important that Meyer et al. only
demonstrated 10% in-vivo 5-HT2A receptor downregulation in MDD patients between 20-30 years,
but not in older patients.116 The desensitisation of post-synaptic receptors might explain the
tolerance to adverse effects that occurs after some days of exposure to antidepressants.109
Interestingly, in rodents, the activation of post-synaptic 5-HT1A receptors in the hippocampus
stimulates neurogenesis, which appears to be required for therapeutic effects of
After the activation of the G-protein metabotropic receptors (either by serotonin or
noradrenaline), second messenger systems mediate signals to the neuron’s nucleus, where cAMP
responsive element binding protein (CREB) regulates CREB-directed gene transcription.79;118 Most
evidence indicates that CREB is upregulated by chronic antidepressant use, but not for all
antidepressants, and results of studies appear to be biased by the cell type investigated, the brain
region where these cells originate from and the timing after the first antidepressant exposure.118
One of the genes that is (positively) influenced by CREB is the brain derived neurotrophic factor
(BDNF) gene.119 BDNF is a plethoric growth factor, which regulates neuronal survival, migration,
differentiation, axonal and dendritic growth and synapse formation. The genomic structure of
BDNF is complex, which facilitates differential activation by diverse and variable stimuli, which can
be different in different brain regions and even in different parts of the cell.83
114 and 5-HT2C/2B responsivity.113;115 Because these effects are either investigated in animals or
Changes in neuronal networks
At the neuro-anatomical level, the novel neuroimaging techniques have ‘opened’ the brain to
study the in-vivo effects of psychiatric treatment.104 In MDD, early PET studies found time
dependent changes during the treatment with fluoxetine.120 After 1 week, glucose metabolism
increased in the hippocampus and brainstem, and decreased in posterior cingulate gyrus, striatum
and thalamus compared to the pre-treatment scan. After 6 weeks, patients who responded to the
treatment had a decrease in metabolism in the subgenual cingulate gyrus, the hippocampus, the
pallidum and insula and an increase in the anterior and posterior cingulate gyrus, prefrontal and
parietal cortex. However, the non-responders showed a persistent, unchanged pattern of change
as seen in week 1. The changes in prefrontal cortex and subgenual cingulate gyrus correlated best
with symptomatic improvement.99;101;120 These findings were replicated in patients treated with
paroxetine.121 Moreover, patients who responded while on placebo-treatment showed somehow
similar but also distinct‡ changes in brain metabolism compared to fluoxetine responders.99;122 This
indicates that in neuroimaging studies, response andtreatment effects may coincide, but both
may also have their specific, distinguishable effects.
In recent fMRI studies, the increased activation of the amygdala to negative (sad, fearful,
angry) and happy faces have been investigated after treatment of SSRIs (fluoxetine,123;124
sertraline125), venlafaxine126;127 and bupropion.128 These studies rather consistently found
decreased activation of the amygdala, insula and increased cortical activity after treatment.
However,most patients in these studies were treatment responders at the end of the study. In
contrast with previous PET-studies, none of the fMRI studies used a placebo-comparison
measured twice, so no distinction between specific drug and response effects in fMRI can be
‡Similar changes included increased metabolism in frontal, parietal and posterior cingulate, and decreased metabolism in
subgenual cingulate gyrus. Distinct changes included no changes in subcortical brainstem, hippocampal, and caudate
metabolism in placebo-responders.
(Pharmaco-)genetic effects relevant for antidepressants
Several genetic polymorphisms have been investigated in relation to treatment response to
SSRIs. The polymorphism studied most is the SERT promoter gene (5-HTTLPR), for which a long
(L) and a short (S) variant were identified, with a recently discovered functional tri-allelic variant
(rs25531).129 The 5-HTTLPR is associated with the transcriptional activity of the SERT gene.130 Cells
homozygous for the L-allele produce higher concentrations of SERT mRNA, and the rate of
serotonin uptake by the transporter is >2-fold higher than in cells containing one or two copies of
the S-allele. A meta-analysis of 15 studies showed a pooled association between the 5-HTTLPR-
polymorphism and SSRI efficacy,131 with MDD patients with at least one L-allele having higher
response rates to SSRIs. However, in a large sample of patients treated with citalopram (an SSRI),
treatment response was not associated with the tri-allelic 5-HTTLPR-polymorphism.132;133
Furthermore, individuals carrying the S-allele experience increased adverse events after SSRI
treatment,132;134 have elevated risk of depression in relation to life events,135 but alsoshow
increased amygdala reactivity to fearful stimuli.136;137 A large MRI-study in healthy controls showed
associations of the 5-HTTLPR S/S polymorphism with unfavourable alterations in anatomy and
function of the amygdala-cingulate feedback circuit.138 Thesefindings strongly argue for an
important role of the 5-HTTLPR-polymorphism in the development and functioning of emotional
networks involved in MDD. Other pharmacogenetic associations with clinical response have been
investigated, but will not be further addressed in this thesis.
Summary and questions addressed in this thesis
Major Depressive Disorder is a prevalent and disabling illness, which potentially recurs and may
become a chronic disease. It is the second most common cause of disability worldwide, and has a
12-month prevalence of ±5.5%, with lifetime prevalences of 12-14% in males and 22-24% in females.3;5
MDD is associated with high direct treatment costs as well as indirect costs of loss of productivity
and quality of life. Besides various forms of psychotherapy, pharmacotherapy with antidepressant
drugs is often applied.13;21-25
Antidepressants are grouped in four different classes: Selective serotonin reuptake inhibitors
(SSRIs), tricyclic antidepressants (TCAs), monoamine oxidase inhibitors (MAOIs) and a
miscellaneous group including so-called dual action antidepressants. SSRIs, dual-action
antidepressants and TCAs are used most often.39 Most antidepressants increase serotonergic and/
or noradrenergic neurotransmission. When considered more precisely, different classes of
antidepressants appear to have distinct additional pre- and postsynaptic effects.105;106 Secondary
to increased serotonergic and/or noradrenergic neurotransmission, complex second messenger
pathways are activated, which are only partly understood.79 Macroscopically, presumably relevant
changes in neuronal networks following antidepressant treatment have been identified.99;101;104
Nevertheless, which changes are required and specific for the improvement of symptomsremains
an enigma. Furthermore, the effects of increased neurotransmission might be affected by the
genetic make-up of patients (i.e. the 5-HTTLPR polymorphism129). Genetic polymorphisms appear
to be associated with treatment effects,131 but might also influence the development and
connectivity of the neuronal networks.138
Despite comparable efficacy of antidepressants, only 50% of the MDD-patients respond to the
first antidepressant trial given, while fewer achieve full remission of symptoms.48;51 Therefore,
insufficient response to the first antidepressant is a relevant clinical problem that challenges
The clinician’s first dilemma is how to adequately and efficiently measure the changes in
symptom severity, for which the further development and implementation of short, easy to use
and valid clinician rated questionnaires would improve clinical decision making. The second
dilemma is to time the change of the treatment initiallystarted: this should not be changed too
early, neither too late.70 The third dilemma is to balance helplessness, impaired functioning by the
disease and (counter-) transference by ‘blaming the victim’, i.e. reproaching the patient of the
moderate efficacy of antidepressants as ‘being unmotivated’ or ‘having a personality disorder’.
General introduction and outline
When ‘the miracle doesn’t happen’, 5 strategies for non-response are possible options:
prolongation of the initial trial, dose-escalation, switching, augmentation and combination of
antidepressants. For all these options the current evidence is either equivocal or fragmentary. In
order to develop evidence based treatment algorithms for the 50% of MDD-patients who turn out
to be non-responsive to the first antidepressant trial, systematic overviews of the literature are
required and gaps in the evidence-base need to be addressed by new research-projects.
More fundamentally, the perceived delayed onset of symptom improvement with
antidepressants, their moderated efficacy, and their extensive effects in the brain beyond an
increase in serotonergic or noradrenergic neurotransmission, summons the question what is the
etiopathogenetic model for MDD. Or, more specifically, what is the evidence that corroborates
the corner-stone of most antidepressant’s action: the monoamine hypothesis and the
involvement of the serotonergic system as a causal explanation for MDD.
Research projects underlying this thesis
This thesis incorporates three research projects that were initiated and conducted by the Program
for Mood Disorders of the Academic Medical Center (AMC). One additional project was initiated
by the department of vascular medicine of the AMC.
First, a guideline-project was initiated with a grant from the Academic Medical Center (No
SFA.07.012). Aim of this guideline project was to develop an evidence based guideline for non-
response after a first SSRI.
Second, as a result of this guideline project, two grants were obtained from the Netherlands
Organization for Health Research and Development (ZonMw), program Mental Health, education
of investigators in mental health (Geestkracht-OOG; projects #100-002-001 and #100-002-002).
These grants were applied to initiate the DELPHI-study (Dose-Escalation Legitimate?
Pharmacology and Imaging studies in depression). The DELPHI-study was set up as a
methodologically sound trial to study clinical effectiveness of dose-escalation as a strategy for
non-response. As a novel extension to existent dose-escalation trials, we aimed to investigate the
molecular neurobiological target of dose-escalation: the occupancy of the serotonin transporter
(SERT) by paroxetine (a SSRI). This was done by the acquisition of two or three single photon
emission computed tomography (SPECT) scans in a subgroup of drug-free patients of the total
A third project was started as an extension to the DELPHI-SPECT study. This third project could
be planned after a grant from the Dutch Brain Foundation (Hersenstichting) was obtained
(project #14F06.45). The DELPHI-fMRI aimed to investigate the neurobiological changes in brain
activation by treatment with paroxetine in a functional Magnetic Resonance Imaging (fMRI)
study. This study was superimposed on the SPECT-imaging of the last 22 drug-free patients
participating in DELPHI.
The final fourth research project in this thesis was initiated after an almost fatal bleeding
complication, which occurred in a patient who underwent surgery, and lost 12 litres of blood. This
complication was afterwards attributed to fluoxetine use. This project was funded internally by
mutual contributions of both participating AMC-departments.
The questions addressed in this thesis
1 Is a short, easy to use clinician rated questionnaires as effective and precise as the routine
Hamilton depression rating scale (HDRS)?
The HDRS54 is most frequently used as the golden-standardin clinical trials, but is probably too
extensive to use in clinical practice.61 To facilitate clinical measurements of depression severity
and objectivity for critical decision points, we reanalyzed the treatment-outcomes of two
antidepressant-psychotherapy trials, which were performed by the Mentrum research
group.139;140 We therefore investigated whether the effect-sizes of two 6-item subscales of the
HDRS (17-items) – the Maier141 and Bech59 subscales – were comparable to the original HDRS in
the measurement of depression severity and the sensitivity to measure changes.
Furthermore, we investigated whether this comparability was stable across the full range of
response to treatment, and across different treatments and for different baseline severity of
depression. We also determined cut-off points for remission for these subscales compared to
conventional HDRS definitions.66;142;143 See chapter 3.
2 What is the evidence for dose-escalation as a strategy for non-response to a first SSRI?
For this question, we performed a systematic review of the evidence for the dose response
relationship for SSRIs in MDD. See chapter 4.
3 What is the evidence for switching antidepressants as a strategy for non-response to a first
For this question, we performed a systematic review of the evidence for switching after
failure of a first SSRI in MDD. Part of this study was a meta-analysis of three switch-studies.
See chapter 5.
4 Does the depletion of monoamine (5-HT and NA/DA) systems lower mood in humans, and is
this lowering of mood different across different populations?
For this question, we performed a systematic review of monoamine depletion studies
reporting mood effects of depletion. As an extension of previous systematic reviews of
monoamine depletion studies,88-93 we aimed to pool the results ofthe small-sized depletion
studies, because they might not have detected small differences by a lack of power, and
pooling would quantify the balance of positive versus negative studies. Therefore, we applied
a pooling technique (modified from conventional meta-analyses of randomized controlled
trial data and including an adjustment for small sample bias) to handle the statistically paired
cross-over designs of these studies in formal, stratified meta-analyses.144 See chapter 6.
5 Do MDD-patients and healthy controls differ in the number of central serotonin transporters,
and is the amount of available SERTs correlated with depression severity?
Despite the fact that the working-mechanism of antidepressants supports the monoamine
deficiency theory, the pathogenesis of MDD remains unclear.95 Therefore, differences in SERT
availability in patients and healthy controls have been studied previously, with conflicting
results. Additionally, significant effects on SERT availability have been reported for gender,145
smoking behavior,145 aging146;147 and season of scanning.148 Therefore, we analyzed the
baseline SPECT-scans of the DELPHI-SPECT participants versus age and sex-matched healthy
controls. Because our sample size was large, we were able to properly account for potential
confounders and possible interactions, of which the multivariate effects are reported. See
6 Does a common genetic polymorphism of the promoter region of the serotonin transporter
gene (SLC6A4) modify the association between the SERT occupancy by paroxetine and the
We performed this study because SSRI-response is likely associated with 5-HTTLPR
polymorphisms, 5-HTTLPR polymorphisms might influence SERT availability (the target for
SSRIs), and it is unclear how occupancy of the available SERTs is related to clinical response.
Thus, we aimed to investigate the paroxetine treatment by genotype interaction regarding
clinical response on the molecular level of SERT occupancy. We quantified the relation
between SERT occupancy and clinical response, and studied how the 5-HTTLPR -
polymorphism affected this SERT occupancy-response relationship. We performed this study
in the open phase of the DELPHI-SPECT study, when patients were treated with paroxetine 20
mg/day for 6 weeks. See chapter 8.
7 Is dose-escalation of paroxetine an effective clinical strategy for non-response in MDD?
The systematic review of dose-escalation (chapter 4), identified methodological flaws in
previous dose-escalation trials. In this study we reevaluated the clinical efficacy of dose-
escalation of paroxetine without these flaws, and, considering the molecular target of SSRIs,
we also tested whether paroxetine dose-escalation increasedSERT occupancy more than
placebo dose-escalation. We therefore performed a 6 week, multicenter, randomized study in
depressed patients not responding to 6 weeks of paroxetine at 20 mg/day. As a novel
extension to previous clinical trials, and in order to elucidate the neurobiological basis for an
General introduction and outline
expected lack of benefit of dose-escalation, we included a SPECT imaging approach.
Herewith, we quantified whether paroxetine dose-escalation increased SERT occupancy more
than placebo dose-escalation. This enabled us to relate clinical findings to the neurobiological
correlate of SERT occupancy. See chapter 9.
8 Does treatment with paroxetine normalize amygdala hyperactivation in MDD?
We initiated this study after the first reports of attenuated amygdala activation after
treatment with sertraline.125 Thereafter several groups replicated a baseline hyperactivation of
the amygdala but less consistently reported the attenuation of amygdala-hyperactivation
after treatment.123;126;128 We therefore investigated whether: activation of the amygdala by
(negative) facial expressions differed from healthy controls, this activation of the amygdala
changed after 6 and 12 weeks of treatment with paroxetine, the activationof the amygdala
and other brain areas merely changedby paroxetine treatment or in relation with clinical
response, and whether dose-escalation of paroxetine in week 6 non-responders affected
activations, compared to placebo-dose-escalation. For this study, we performed an fMRI study
in 22MDD patients who participated in the DELPHI-fMRI study. Patients were treated with
paroxetine (20 mg/day followed by a randomized dose-escalation for non-responders) and
were scanned at baseline, 6 weeks and 12 weeks of treatment. We obtained a baseline scan
for 21 matched controls, to contrast baseline amygdala activation in MDD-patients. See
9 What are the changes in hemostasis and blood platelet parameters when patients are treated
with paroxetine, and are these changes modified by dose-escalation or a genetic
polymorphism of the promoter region of the serotonin transporter gene?
In this study, we evaluated the effects of standard and increasing dosages of paroxetine on
the bleeding tendency and hemostatic functions of platelets in patients who were drug-free
before the start of paroxetine. In addition, we assessed whether these effects are modified by
the 5-HTTLPR polymorphism. See chapter 11.
1. Burton R. The anatomy of melancholy, what it is, with all the kinds, causes, symptoms, prognostics, and several cures of it. In
three partitions. With their several sections, members, and subsections, philosophically, medecinally, historically opened and
cut up. 6th ed. London: J. Cuthell; J. Nunn; Longman and Co.; J. Mawman; Baldwin and Co.; Lackington and Co.; Black and
Co.; S. Bagster; Edwards and Knibb; Ogle, Duncan, and Co. R. Saunders; J. Sheldon; Wilsons, York; 1651:1-461.
Schene AH, Huyser J, Ruhe HG. Validiteit van de diagnose depressie. In: Koerselman F, Vergouwen AC, eds. De
gecompliceerde kliniek 2005: over de geldigheid van diagnoses. Amsterdam: Benecke; 2005.
Mathers CD, Loncar D. Projections of global mortality and burden of disease from 2002 to 2030. PLoS Med. 2006; 3: e442.
American Psychiatric Association. Diagnostic and statistical manual of mental disorders, fourth edition. 4 ed. Washington
DC: American Psychiatric Association; 1994.
Bijl RV, Zessen Gv, Ravelli A. Psychiatrische morbiditeit onder volwassenen in Nederland: het NEMESIS-onderzoek. II.
Prevalentie van psychiatrische stoornissen. NedTijdschrGeneesk. 1997; 141: 2453-2460.
Hasin DS, Goodwin RD, Stinson FS, Grant BF. Epidemiology of major depressive disorder: results from the National
Epidemiologic Survey on Alcoholism and Related Conditions. Arch Gen Psychiatry. 2005; 62: 1097-1106.
Kessler RC, McGonagle KA, Zhao S, Nelson CB, Hughes M, Eshleman S et al. Lifetime and 12-month prevalence of DSM-III-
R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Arch Gen Psychiatry. 1994; 51:
Lepine JP, Gastpar M, Mendlewicz J, Tylee A. Depression in the community: the first pan-European study DEPRES
(Depression Research in European Society). Int Clin Psychopharmacol. 1997; 12: 19-29.
Kessler RC, Berglund P, Demler O, Jin R, Koretz D, Merikangas KR et al. The epidemiology of major depressive disorder:
results from the National Comorbidity Survey Replication (NCS-R). JAMA. 2003; 289: 3095-3105.
10. Kessler RC, Demler O, Frank RG, Olfson M, Pincus HA, Walters EE et al. Prevalence and treatment of mental disorders,
1990 to 2003. N Engl J Med. 2005; 352: 2515-2523.
11. Spijker J, de Graaf R, Bijl RV, Beekman AT, Ormel J, Nolen WA.Duration of major depressive episodes in the general
population: results from The Netherlands Mental Health Survey and Incidence Study (NEMESIS). Br J Psychiatry. 2002; 181:
12. Kendrick T, King F, Albertella L, Smith PW. GP treatment decisions for patients with depression: an observational study.
Br J Gen Pract. 2005; 55: 280-286.
13. American Psychiatric Association. Practice guideline for the treatment of patients with major depressive disorder
(revision). American Psychiatric Association. Am J Psychiatry. 2000; 157: 1-45.
14. Beekman AT, Ormel J. Depressie. In: Jong Ad, Brink Wvd, Ormel J, Wiersma D, eds. Handboek psychiatrische
epidemiologie. 1 ed. Maarssen: Elsevier/ De Tijdstroom; 1999:300-328.
15. Keller MB, Lavori PW, Mueller TI, Endicott J, Coryell W, Hirschfeld RM et al. Time to recovery, chronicity, and levels of
psychopathology in major depression. A 5-year prospective follow-up of 431 subjects. Arch Gen Psychiatry. 1992; 49: 809-
16. Wells KB, Burnam MA, Rogers W, Hays R, Camp P. The course of depression in adult outpatients. Results from the
Medical Outcomes Study. Arch Gen Psychiatry. 1992; 49: 788-794.
17. Greenberg PE, Stiglin LE, Finkelstein SN, Berndt ER. Depression: a neglected major illness. J Clin Psychiatry. 1993; 54: 419-
18. Ruysbroek JM. Aan welke ziekten en aandoeningen wordt het geld besteedt? 1.2 (25 juni 2001). 2001. Bilthoven, RIVM.
Volksgezondheid Toekomst Verkenning, Nationaal Kompas Volksgezondheid.
19. Slobbe LC, Kommer GJ, Smit JM, Groen J, Meerding WJ, Polder JJ. Kosten van ziekten in Nederland 2003. Zorg voor
euro's. [RIVM rapport# 270751010]. 2008. Bilthoven, RIVM. Volksgezondheid Toekomst Verkenning, Nationaal Kompas
20. Ziektendiagnosen bij uitkeringen voor arbeidsongeschiktheid. Statistische informatie over medische classificaties in
WAO, WAZ en Wajong 1999.2001. Amsterdam, Landelijk instituut sociale verzekeringen.
21. Richtlijn Depressie. Multidisciplinaire Richtlijn Depressie. Richtlijn voor de diagnostiek en behandeling van volwassen
cliënten met een depressie. Brandt-Dominicus JC.2005. Utrecht, Trimbos-instituut, in opdracht van de Landelijke
Stuurgroep Multidisciplinaire Richtlijnontwikkeling in de GGZ.
22. Marwijk HWv, Grundmeijer HG, Bijl D, Gelderen MGv, Haan Md, Weel-Baumgarten EMv et al. NHG-Standaard Depressieve
stoornis (depressie) (eerste herziening). Huisarts Wet. 2003; 46: 614-623.
23. NICE. Clinical Guideline 23. Depression: management of depression in primary and secondary care. December 2004. 23, 1-
63. 2004. London, National Institute for Clinical Excellence.
24. Kennedy SH, Lam RW, Cohen NL, Ravindran AV, CANMAT Depression Work Group. Clinical guidelines for the treatment of
depressive disorders. IV. Medications and other biological treatments. Can J Psychiatry. 2001; 46 Suppl 1: 38S-58S.
25. Anderson IM, Nutt DJ, Deakin JFW. Evidence-based guidelines for treating depressive disorders with antidepressants: A
revision of the 1993 British Association for Psychopharmacology guidelines. J Psychopharmacol. 2000; 14: 3-20.
26. Mulrow CD, Williams JW, Jr., Trivedi M, Chiquette E, Aguilar C, Cornell JE et al. Evidence report on: Treatment of
depression - Newer pharmacotherapies. Psychopharmacol Bull. 1999; 34: 409-795.
27. Centraal Begeleidingsorgaan voor de Intercollegiale Toetsing. Consensusbijeenkomst depressie bij volwassenen: vrijdag 16
september 1994 Utrecht. Utrecht: Centraal Begeleidingsorgaan voor de Intercollegiale Toetsing; 1994.
28. Depression Guideline Panel. Depression in Primary Care: Volume 1. Detection and Diagnosis. Clinical Practice Guideline,
Number 5. AHCPR Publication No. 93-0550. 1993. Rockville, MD, U.S. Department of Health and Human Services, Public
Health Service, Agency for Health Care Policy and Research.
29. Depression Guideline Panel. Depression in Primary Care: Volume 2. Treatment of Major Depression. Clinical Practice
Guideline, Number 5. AHCPR Publication No. 93-0551. 1993. Rockville, MD, U.S. Department of Health and Human
Services, Public Health Service, Agency for Health Care Policy and Research.
30. Jonghe Fde, Swinkels JA, Vergouwen AC, Brink Gvan de. Antidepressiva 2008. Een leidraad voor het rationeel omgaan met
antidepressiva bij de behandeling van patiënten met een depressie. Amsterdam: Benecke NL; 2008.
31. Anderson IM, Tomenson BM. The efficacy of selective serotonin re-uptake inhibitors in depression: A meta-analysis of
studies against tricyclic antidepressants. J Psychopharmacol. 1994; 8: 238-249.
32. Anderson IM. Meta-analytical studies on new antidepressants. Br Med Bull. 2001; 57: 161-178.
33. Anderson IM. Drug treatment of depression: Reflections on the evidence. AdvPsychiatrTreat. 2003; 9: 11-20.
34. Kirsch I, Deacon BJ, Huedo-Medina TB, Scoboria A, Moore TJ, Johnson BT. Initial severity and antidepressant benefits: a
meta-analysis of data submitted to the Food and Drug Administration. PLoS Med. 2008; 5: e45.
35. Turner EH, Matthews AM, Linardatos E, Tell RA, Rosenthal R. Selective publication of antidepressant trials and its
influence on apparent efficacy. N Engl J Med. 2008; 358: 252-260.
36. Barbui C, Furukawa TA, Cipriani A. Effectiveness of paroxetine in the treatment of acute major depression in adults: a
systematic re-examination of published and unpublished data from randomized trials. CMAJ. 2008; 178: 296-305.
37. Moncrieff J, Cohen D. Do antidepressants cure or create abnormal brain states? PLoS Med. 2006; 3: e240.
38. Stichting Farmaceutische Kengetallen. Data en feiten 2007.2007. Den Haag, Stichting Farmaceutische Kengetallen.
39. Ruhe HG, Rijswijk Evan, Verkes RJ. Biologische interventies bij depressie. In: Huyser J, Schene AH, Sabbe B, Spinhoven Ph,
eds. Handboek depressieve stoornissen. Utrecht: De Tijdstroom; 2008: 201-218.
40. Thase ME, Rush AJ. Treatment-Resistant Depression. In: Bloom FE, Kupfer DJ, eds. Psychopharmacology: The Fourth
Generation of Progress. New York: Raven Press Ltd.; 1995:1081-1097.
41. Freemantle N, Anderson IM, Young P. Predictive value of pharmacological activity for the relative efficacy of
antidepressant drugs. Meta-regression analysis. Br J Psychiatry. 2000; 177: 292-302.
42. Smith D, Dempster C, Glanville J, Freemantle N, Anderson I. Efficacy and tolerability of venlafaxine compared with
selective serotonin reuptake inhibitors and other antidepressants: a meta-analysis. Br J Psychiatry. 2002; 180: 396-404.
43. Papakostas GI, Thase ME, Fava M, Nelson JC, Shelton RC. Are antidepressant drugs that combine serotonergic and
noradrenergic mechanisms of action more effective than the selective serotonin reuptake inhibitors in treating major
depressive disorder? A meta-analysis of studies of newer agents. Biol Psychiatry. 2007; 62: 1217-1227.
General introduction and outline
44. Arroll B, MacGillivray S, Ogston S, Reid I, Sullivan F, Williams B et al. Efficacy and tolerability of tricyclic antidepressants
and SSRIs compared with placebo for treatment of depression in primary care: a meta-analysis. Ann Fam Med. 2005; 3:
45. Anderson IM. Selective serotonin reuptake inhibitors versus tricyclic antidepressants: a meta-analysis of efficacy and
tolerability. J Affect Disord. 2000; 58: 19-36.
46. Kroenke K, West SL, Swindle R, Gilsenan A, Eckert GJ, Dolor R et al. Similar effectiveness of paroxetine, fluoxetine, and
sertraline in primary care: a randomized trial. JAMA. 2001; 286: 2947-2955.
47. Simon G. Choosing a first-line antidepressant: equal on average does not mean equal for everyone. JAMA. 2001; 286:
48. Thase ME. Evaluating antidepressant therapies: remission as the optimal outcome. J Clin Psychiatry. 2003; 64 Suppl 13: 18-
49. McIntyre RS, O'Donovan C. The human cost of not achieving full remission in depression. Can J Psychiatry. 2004; 49: 10S-
50. Geddes JR, Carney SM, Davies C, Furukawa TA, Kupfer DJ, Frank E et al. Relapse prevention with antidepressant drug
treatment in depressive disorders: a systematic review. Lancet. 2003; 361: 653-661.
51. Trivedi MH, Rush AJ, Wisniewski SR, Nierenberg AA, Warden D, Ritz L et al. Evaluation of outcomes with citalopram for
depression using measurement-based care in STAR*D: implications for clinical practice. Am J Psychiatry. 2006; 163: 28-40.
52. Rush AJ, Fava M, Wisniewski SR, Lavori PW, Trivedi MH, Sackeim HA et al. Sequenced treatment alternatives to relieve
depression (STAR*D): rationale and design. Control Clin Trials. 2004; 25: 119-142.
53. Yonkers KA, Samson J. Mood disorders measures. In: Rush AJ, Pincus HA, First MB, et al., eds. Handbook of psychiatric
measures. 1st ed. Washington DC: American Psychiatric Association; 2000:515-548.
54. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry. 1960; 23: 56-61.
55. Montgomery SA, Asberg M. A new depression scale designed to be sensitive to change. Br J Psychiatry. 1979; 134: 382-
56. Rush AJ, Giles DE, Schlesser MA, Fulton CL, Weissenburger J, Burns CT. The Inventory for Depression Symptomatology
(IDS): preliminary findings. Psychiatry Res. 1986; 18: 65-87.
57. Rush AJ, Trivedi MH, Ibrahim HM, Carmody TJ, Arnow B, Klein DN et al. The 16-Item Quick Inventory of Depressive
Symptomatology (QIDS), clinician rating (QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients with
chronic major depression. Biol Psychiatry. 2003; 54: 573-583.
58. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch GenPsychiatry. 1961; 4:
561 571: 561.
59. Bech P, Gram LF, Dein E, Jacobsen O, Vitger J, Bolwig TG. Quantitative rating of depressive states. Acta Psychiatr Scand.
1975; 51: 161-170.
60. Jonghe F de. Leidraad voor het scoren van de Hamilton Depression Rating Scale: HDRS leidraad. Amsterdam: Benecke
61. Faries D, Herrera J, Rayamajhi J, DeBrota D, Demitrack M, Potter WZ. The responsiveness of the Hamilton Depression
Rating Scale. J Psychiatr Res. 2000; 34: 3-10.
62. Gibbons RD, Clark DC, Kupfer DJ. Exactly what does the Hamilton Depression Rating Scale measure? J Psychiatr Res. 1993;
63. Moeller H-J, Mueller H, Volz HP. How to assess the onset of antidepressant effect: Comparison of global ratings and
findings based on depression scales. Pharmacopsychiatry. 1996; 29: 57-62.
64. Santor DA, Coyne JC. Examining symptom expression as a function of symptom severity: Item performance on the
Hamilton Rating Scale for Depression. Psychol Assess. 2001; 13: 127-139.
65. Vittengl JR, Clark LA, Kraft D, Jarrett RB. Multiple measures, methods, and moments: a factor-analytic investigation of
change in depressive symptoms during acute-phase cognitive therapy for depression. Psychol Med. 2005; 35: 693-704.
66. Frank E, Prien RF, Jarrett RB, Keller MB, Kupfer DJ, Lavori PW et al. Conceptualization and rationale for consensus
definitions of terms in major depressive disorder. Remission, recovery, relapse, and recurrence. Arch Gen Psychiatry. 1991;
67. Crismon ML, Trivedi M, Pigott TA, Rush AJ, Hirschfeld RMA, Kahn DA et al. The Texas medication algorithm project:
Report of the Texas consensus conference panel on medication treatment of major depressive disorder. J Clin Psychiatry.
1999; 60: 142-156.
68. Trivedi MH. Using treatment algorithms to bring patients to remission. J Clin Psychiatry. 2003; 64 Suppl 2: 8-13.
69. Taylor MJ, Freemantle N, Geddes JR, Bhagwagar Z. Early onset of selective serotonin reuptake inhibitor antidepressant
action: systematic review and meta-analysis. Arch Gen Psychiatry. 2006; 63: 1217-1223.
70. Quitkin FM, Petkova E, McGrath PJ, Taylor B, Beasley C, Stewart J et al. When should a trial of fluoxetine for major
depression be declared failed? Am J Psychiatry. 2003; 160: 734-740.
71. Katz MM, Koslow SH, Maas JW, Frazer A, Bowden CL, Casper R et al. The timing, specificity and clinical prediction of
tricyclic drug effects in depression. Psychol Med. 1987; 17: 297-309.
72. Quitkin FM, McGrath PJ, Stewart JW, Ocepek-Welikson K, Taylor BP, Nunes E et al. Chronological milestones to guide
drug change. When should clinicians switch antidepressants? Arch Gen Psychiatry. 1996; 53: 785-792.
73. Kennedy N, McDonough M. Pharmacological management of treatment resistant depression: A clinical review.
IrJPsycholMed. 2003; 20: 18-23.
74. Boer NAde. Verpleegkundige interventies bij depressie. 1 ed. Amsterdam: Programma Stemmingstoornissen AMC/De
75. Friedman MA, Detweiler-Bedell JB, Leventhal HE, Horne R, Keitner GI, Miller IW. Combined psychotherapy and
pharmacotherapy for the treatment of major depressive disorder. Clin Psychol Sci Prac. 2004; 11: 47-68.
76. Pampallona S, Bollini P, Tibaldi G, Kupelnick B, Munizza C. Combined pharmacotherapy and psychological treatment for
depression: a systematic review. Arch Gen Psychiatry. 2004; 61: 714-719.
77. Siegel GJ, Agranoff BW, Albers RW, Fisher SK, Uhler MD. Basic neurochemistry, molecular, cellular, and medical aspects. 6
ed. Philadelphia: Lippincott, Williams & Wilkins; 1999.
78. Purves D, Augustine G, Fitzpatrick D, Katz L, LaMantia A. Neuroscience. 2 ed. Sunderland: Sinauer Associates, Inc.; 2001.
79. Belmaker RH, Agam G. Major depressive disorder. N Engl J Med. 2008; 358: 55-68.
80. Andreasen NC. Brave new brain: conquering mental illness in the era of the genome [Schitterend nieuw brein: psychiatrie in
het tijdperk van het genoom]. 1 ed. Amsterdam: Nieuwezijds B.V.; 2002.
81. Marieb E. Human anatomy & Physiology. 5 ed. Redwood City: The Benjamin/ Cummings Publishing Company, Inc.; 1999.
82. Gaspar P, Cases O, Maroteaux L. The developmental role of serotonin: news from mouse molecular genetics. Nat Rev
Neurosci. 2003; 4: 1002-1012.
83. Martinowich K, Lu B. Interaction between BDNF and serotonin: role in mood disorders. Neuropsychopharmacology. 2008;
84. Young SN, Smith SE, Pihl RO, Ervin FR. Tryptophan depletion causes a rapid lowering of mood in normal males.
Psychopharmacology (Berl). 1985; 87: 173-177.
85. Moja EA, Lucini V, Benedetti F, Lucca A. Decrease in plasma phenylalanine and tyrosine after phenylalanine-tyrosine free
amino acid solutions in man. Life Sci. 1996; 58: 2389-2395.
86. Shopsin B, Gershon S, Goldstein M, Friedman E, Wilk S. Use of synthesis inhibitors in defining a role for biogenic amines
during imipramine treatment in depressed patients. Psychopharmacol Commun. 1975; 1: 239-249.
87. Delgado PL, Miller HL, Salomon RM, Licinio J, Heninger GR, Gelenberg AJ et al. Monoamines and the mechanism of
antidepressant action: effects of catecholamine depletion on mood of patients treated with antidepressants.
Psychopharmacol Bull. 1993; 29: 389-396.
88. Bell C, Abrams J, Nutt D. Tryptophan depletion and its implications for psychiatry. Br J Psychiatry. 2001; 178: 399-405.
89. Booij L, Van der Does AJ, Riedel WJ. Monoamine depletion in psychiatric and healthy populations: review. Mol Psychiatry.
2003; 8: 951-973.
90. Van der Does AJ. The effects of tryptophan depletion on mood and psychiatric symptoms. J Affect Disord. 2001; 64: 107-
91. Moore P, Landolt HP, Seifritz E, Clark C, Bhatti T, Kelsoe J et al. Clinical and physiological consequences of rapid
tryptophan depletion. Neuropsychopharmacology. 2000; 23: 601-622.
92. Booij L, Van der DW, Benkelfat C, Bremner JD, Cowen PJ, Fava M et al. Predictors of mood response to acute tryptophan
depletion. A reanalysis. Neuropsychopharmacology. 2002; 27: 852-861.
93. Van der Does AJ. The mood-lowering effect of tryptophan depletion: possible explanation for discrepant findings. Arch
Gen Psychiatry. 2001; 58: 200-202.
94. Damasio AR. Descartes' error - emotion, reason and the human brain [De vergissing van Descartes - gevoel, verstand en het
menselijk brein]. 2nd ed. Amsterdam: Wereldbibliotheek; 1994.
95. Owens MJ. Selectivity of antidepressants: from the monoamine hypothesis of depression to the SSRI revolution and
beyond. J Clin Psychiatry. 2004; 65 Suppl 4: 5-10.
96. Burke WJ. Selective versus multi-transmitter antidepressants: are two mechanisms better than one? J Clin Psychiatry.
2004; 65 Suppl 4: 37-45.
97. Fava M, Kendler KS. Major depressive disorder. Neuron. 2000; 28: 335-341.
98. Delgado PL. How antidepressants help depression: mechanisms of action and clinical response. J Clin Psychiatry. 2004; 65
Suppl 4: 25-30.
99. Mayberg HS. Modulating dysfunctional limbic-cortical circuits in depression: towards development of brain-based
algorithms for diagnosis and optimised treatment. Br Med Bull. 2003; 65: 193-207.
100. Phillips ML, Drevets WC, Rauch SL, Lane R. Neurobiology of emotion perception I: The neural basis of normal emotion
perception. Biol Psychiatry. 2003; 54: 504-514.
101. Phillips ML, Drevets WC, Rauch SL, Lane R. Neurobiology of emotion perception II: Implications for major psychiatric
disorders. Biol Psychiatry. 2003; 54: 515-528.
102. Mayberg HS, Liotti M, Brannan SK, McGinnis S, Mahurin RK, Jerabek PA et al. Reciprocal limbic-cortical function and
negative mood: converging PET findings in depression and normal sadness. Am J Psychiatry. 1999; 156: 675-682.
103. Mayberg HS. Limbic-cortical dysregulation: a proposed model of depression. J Neuropsychiatry Clin Neurosci. 1997; 9: 471-
104. Malhi GS, Lagopoulos J. Making sense of neuroimaging in psychiatry. Acta Psychiatr Scand. 2008; 117: 100-117.
105. Blier P, De Montigny C. Current advances and trends in the treatment of depression. Trends Pharmacol Sci. 1994; 15: 220-
106. Nemeroff CB, Owens MJ. Neuropharmacology of paroxetine. Psychopharmacol Bull. 2003; 37 Suppl 1: 8-18.
107. Bel N, Artigas F. Chronic treatment with fluvoxamine increases extracellular serotonin in frontal cortex but not in raphe
nuclei. Synapse. 1993; 15: 243-245.
108. Davidson C, Stamford JA. Effect of chronic paroxetine treatment on 5-HT1B and 5-HT1D autoreceptors in rat dorsal raphe
nucleus. Neurochem Int. 2000; 36: 91-96.
109. Vaswani M, Linda FK, Ramesh S. Role of selective serotonin reuptake inhibitors in psychiatric disorders: a comprehensive
review. Prog Neuropsychopharmacol Biol Psychiatry. 2003; 27: 85-102.
110. Benmansour S, Cecchi M, Morilak DA, Gerhardt GA, Javors MA, Gould GG et al. Effects of chronic antidepressant
treatments on serotonin transporter function, density, and mRNA level. J Neurosci. 1999; 19: 10494-10501.
General introduction and outline
111. Benmansour S, Owens WA, Cecchi M, Morilak DA, Frazer A. Serotonin clearance in vivo is altered to a greater extent by
antidepressant-induced downregulation of the serotonin transporter than by acute blockade of this transporter.
JNeurosci. 2002; 22: 6766-6772.
112. Pineyro G, Blier P, Dennis T, De Montigny C. Desensitization of the neuronal 5-HT carrier following its long-term blockade.
J Neurosci. 1994; 14: 3036-3047.
113. Maj J, Bijak M, Dziedzicka-Wasylewska M, Rogoz R, Rogoz Z, Skuza G et al. The effects of paroxetine given repeatedly on
the 5-HT receptor subpopulations in the rat brain. Psychopharmacology (Berl). 1996; 127: 73-82.
114. Whale R, Clifford EM, Bhagwagar Z, Cowen PJ. Decreased sensitivity of 5-HT1D receptors in melancholic depression. Br J
Psychiatry. 2001; 178: 454-457.
115. Kennett GA, Lightowler S, de B, V, Stevens NC, Wood MD, Tulloch IF et al. Effect of chronic administration of selective 5-
hydroxytryptamine and noradrenaline uptake inhibitors on a putative index of 5-HT2C/2B receptor function.
Neuropharmacology. 1994; 33: 1581-1588.
116. Meyer JH, Kapur S, Eisfeld B, Brown GM, Houle S, DaSilva J et al. The effect of paroxetine on 5-HT(2A) receptors in
depression: an [18F]setoperone PET imaging study. Am J Psychiatry. 2001; 158: 78-85.
117. Djavadian RL. Serotonin and neurogenesis in the hippocampal dentate gyrus of adult mammals. Acta Neurobiol Exp
(Wars). 2004; 64: 189-200.
118. Blendy JA. The role of CREB in depression and antidepressant treatment. Biol Psychiatry. 2006; 59: 1144-1150.
119. Angelucci F, Brene S, Mathe AA. BDNF in schizophrenia, depression and corresponding animal models. Mol Psychiatry.
2005; 10: 345-352.
120. Mayberg HS, Brannan SK, Tekell JL, Silva JA, Mahurin RK, McGinnis S et al. Regional metabolic effects of fluoxetine in
major depression: serial changes and relationship to clinical response. Biol Psychiatry. 2000; 48: 830-843.
121. Kennedy SH, Evans KR, Kruger S, Mayberg HS, Meyer JH, McCann S et al. Changes in regional brain glucose metabolism
measured with positron emission tomography after paroxetine treatment of major depression. Am J Psychiatry. 2001;
122. Mayberg HS, Silva JA, Brannan SK, Tekell JL, Mahurin RK, McGinnis S et al. The functional neuroanatomy of the placebo
effect. Am J Psychiatry. 2002; 159: 728-737.
123. Fu CH, Williams SC, Cleare AJ, Brammer MJ, Walsh ND, Kim J et al. Attenuation of the neural response to sad faces in
major depression by antidepressant treatment: a prospective, event-related functional magnetic resonance imaging
study. Arch Gen Psychiatry. 2004; 61: 877-889.
124. Fu CH, Williams SC, Brammer MJ, Suckling J, Kim J, Cleare AJ et al. Neural responses to happy facial expressions in major
depression following antidepressant treatment. Am J Psychiatry. 2007; 164: 599-607.
125. Sheline YI, Barch DM, Donnelly JM, Ollinger JM, Snyder AZ, Mintun MA. Increased amygdala response to masked
emotional faces in depressed subjects resolves with antidepressant treatment: an fMRI study. Biol Psychiatry. 2001; 50:
126. Davidson RJ, Irwin W, Anderle MJ, Kalin NH. The neural substrates of affective processing in depressed patients treated
with venlafaxine. Am J Psychiatry. 2003; 160: 64-75.
127. Schaefer HS, Putnam KM, Benca RM, Davidson RJ. Event-related functional magnetic resonance imaging measures of
neural activity to positive social stimuli in pre- and post-treatment depression. Biol Psychiatry. 2006; 60: 974-986.
128. Robertson B, Wang L, Diaz MT, Aiello M, Gersing K, Beyer J et al. Effect of bupropion extended release on negative
emotion processing in major depressive disorder: a pilot functional magnetic resonance imaging study. J Clin Psychiatry.
2007; 68: 261-267.
129. Hu XZ, Lipsky RH, Zhu G, Akhtar LA, Taubman J, Greenberg BD et al. Serotonin transporter promoter gain-of-function
genotypes are linked to obsessive-compulsive disorder. Am J Hum Genet. 2006; 78: 815-826.
130. Lesch KP, Gutknecht L. Pharmacogenetics of the serotonin transporter. Prog Neuropsychopharmacol Biol Psychiatry.
2005; 29: 1062-1073.
131. Serretti A, Kato M, De Ronchi D, Kinoshita T. Meta-analysis of serotonin transporter gene promoter polymorphism (5-
HTTLPR) association with selective serotonin reuptake inhibitor efficacy in depressed patients. Mol Psychiatry. 2007; 12:
132. Hu XZ, Rush AJ, Charney D, Wilson AF, Sorant AJ, Papanicolaou GJ et al. Association between a functional serotonin
transporter promoter polymorphism and citalopram treatment in adult outpatients with major depression. Arch Gen
Psychiatry. 2007; 64: 783-792.
133. Kraft JB, Peters EJ, Slager SL, Jenkins GD, Reinalda MS, McGrath PJ et al. Analysis of association between the serotonin
transporter and antidepressant response in a large clinical sample. Biol Psychiatry. 2007; 61: 734-742.
134. Murphy DL, Lerner A, Rudnick G, Lesch KP. Serotonin transporter: gene, genetic disorders, and pharmacogenetics. Mol
Interv. 2004; 4: 109-123.
135. Caspi A, Sugden K, Moffitt TE, Taylor A, Craig IW, Harrington H et al. Influence of life stress on depression: moderation by
a polymorphism in the 5-HTT gene. Science. 2003; 301: 386-389.
136. Hariri AR, Mattay VS, Tessitore A, Kolachana B, Fera F, Goldman D et al. Serotonin transporter genetic variation and the
response of the human amygdala. Science. 2002; 297: 400-403.
137. Hariri AR, Drabant EM, Munoz KE, Kolachana BS, Mattay VS, Egan MF et al. A susceptibility gene for affective disorders
and the response of the human amygdala. Arch Gen Psychiatry. 2005; 62: 146-152.
138. Pezawas L, Meyer-Lindenberg A, Drabant EM, Verchinski BA, Munoz KE, Kolachana BS et al. 5-HTTLPR polymorphism
impacts human cingulate-amygdala interactions: a genetic susceptibility mechanism for depression. Nat Neurosci. 2005;
139. Jonghe F de, Kool S, Aalst G van, Dekker J, Peen J. Combining psychotherapy and antidepressants in the treatment of
depression. J Affect Disord. 2001; 64: 217-229.
140. Dekker J, Molenaar PJ, Kool S, Aalst G van, Peen J, Jonghe F de. Dose-effect relations in time-limited combined psycho-
pharmacological treatment for depression. Psychol Med. 2005; 35: 47-58.
141. Maier W, Philipp M. Improving the assessment of severity of depressive states: a reduction of the Hamilton Depression
Scale. Pharmacopsychiatry. 1985; 18: 114-115.
142. Prien RF, Carpenter LL, Kupfer DJ. The definition and operational criteria for treatment outcome of major depressive
disorder. A review of the current research literature. Arch Gen Psychiatry. 1991; 48: 796-800.
143. Ballenger JC. Clinical guidelines for establishing remission in patients with depression and anxiety. J Clin Psychiatry. 1999;
60 Suppl 22: 29-34.
144. The Cochrane Collaboration. RevMan 4.2 User Guide, version 4.2 for Windows. 4.2 ed. Oxford, UK: The Cochrane
145. Staley JK, Krishnan-Sarin S, Zoghbi S, Tamagnan G, Fujita M, Seibyl JP et al. Sex differences in [123I]beta-CIT SPECT
measures of dopamine and serotonin transporter availability in healthy smokers and nonsmokers. Synapse. 2001; 41: 275-
146. Pirker W, Asenbaum S, Hauk M, Kandlhofer S, Tauscher J, Willeit M et al. Imaging serotonin and dopamine transporters
with 123I-beta-CIT SPECT: binding kinetics and effects of normal aging. J Nucl Med. 2000; 41: 36-44.
147. van Dyck CH, Malison RT, Seibyl JP, Laruelle M, Klumpp H, Zoghbi SS et al. Age-related decline in central serotonin
transporter availability with [123I]beta-CIT SPECT. Neurobiol Aging. 2000; 21: 497-501.
148. Buchert R, Schulze O, Wilke F, Berding G, Thomasius R, Petersen K et al. Is correction for age necessary in SPECT or PET of
the central serotonin transporter in young, healthy adults? J Nucl Med. 2006; 47: 38-42.
DESIGN AND METHODS OF THE DELPHI-STUDY
Design of the DELPHI study
DELPHI is an acronym for Dose-Escalation Legitimate? Pharmacology and Imaging studies in
depression. The DELPHI-study is the backbone of the studies described in Part IV of this thesis.
Because the DELPHI study in fact comprised three research projects, with slightly different
patient populations and different numbers of participants, this chapter will describe the design of
the DELPHI-study and indicate how the nested sub-studies relate to the major DELPHI-study.
Furthermore the patient disposition over these projects will be given.
Objectives, rationale and research questions
The major aims of the DELPHI-study were toinvestigate the clinical efficacy of and the
neurobiological mechanism behind dose-escalation. In order to determine the neurobiological
effects of dose-escalation, we intended to study different biological measures during treatment:
paroxetine serum concentrations (PSC), serotonin transporter (SERT) occupancy, functional
activations in the cortico-limbic-network, awakening cortisol changes and changes in ω-3/ω-6 poly
unsaturated fatty acids (PUFAs). Furthermore, we planned to study the effects of genetic
polymorphisms on outcomes (i.e. the serotonin transporter gene promoter region (5-HTTLPR)).
Results related to some of these neurobiological measures are described in this thesis.
As a study drug, we choseparoxetine for three reasons. First, paroxetine is the selective
serotonin reuptake inhibitor (SSRI) which is prescribed most frequently in the Netherlands.1
Second, paroxetine is a potent inhibitor of the cytochrome P450 2D6 sub-enzyme, which also is
the enzyme that is responsible for its metabolism. Therefore paroxetine inhibits its own
metabolism, which causes an exponential rise in blood serum concentrations after dose-
escalation.2 Third, when we initiated this study, Gilmor et al. reported noradrenergic reuptake
inhibition by paroxetine, suggesting that paroxetine, like venlafaxine and duloxetine, in fact was a
‘dual action’ antidepressant, especially at higher doses.3
Our main research-questions were:
1 Is a 6-week true dose-escalation of paroxetine (up to 30-50 mg/day) in patients non-responsive
to a 6-week trial with standard dose (20 mg/day) more effective than placebo dose-escalation?
2 Does a 6 week true dose-escalation of paroxetine increase SERT occupancy more than a
3 What is the relation between SERT occupancy and clinical response to paroxetine (either at a
standard dose or after dose-escalation)? Is this relation modified by other neurobiological
parameters (i.e. genetic polymorphisms)?
4 Which changes in the functional activation of the cortico-limbic brain regions correlate with
clinical response to paroxetine? Is there a common and/or differential effect of paroxetine
exposure and/or clinical response?
5 Does a 6 week true dose-escalation of paroxetine generate additional changes (e.g. in
noradrenergic reuptake inhibition, cortisol awakening responses, or ω-3/ω-6 PUFAs) compared
to a placebo dose-escalation, and are these changes related to clinical response?
Only questions 1-4 will be discussed in this thesis, and the methods used for these questions will
briefly be discussed.However, for specific technical we refer to the method-sections in the
Design and interventions
The DELPHI-study was set up as a randomized clinical trial (ISRCTN register nr. ISRCTN44111488;
http://www.trialregister.nl/trialreg/admin/rctview.asp?TC=193). The study was approved by the
Academic Medical Center (AMC) medical ethical committee (03/120, 03/287 and addendum to 03/
120), and all participants provided written informed consent.
Between October 2003 and February 2007 patients were recruited from primary care, our
AMC Program for Mood Disorders, and public psychiatric settings. Patients were treated by their
referring physician or were referred to our outpatient department. In DELPHI two nested sub-
studies were embedded, the DELPHI-SPECT and the DELPHI-fMRI (Figure 2.1). These sub-studies
had slightly different in- and exclusion criteria (see below).
All eligible patients were treated open-label with paroxetine 20 mg/day for 6 weeks. When
severe adverse effects occurred, dosages were reduced to 10 mg/day and again increased to 20
mg/day after one week. After 6 weeks of treatment, patients who responded (defined as ≥50%
reduction of the pretreatment Hamilton depression rating scale (HDRS)) continued paroxetine 20
mg/day. Treatment non-responders were randomized to a true dose-escalation or a placebo dose-
escalation, added to paroxetine 20 mg/day in a double blind design. Dose-escalation was provided
in blue capsules containing 10 mg paroxetine or placebo. Randomization was stratified for
treatment setting (SPECT/fMRI-group, outpatient department AMC, primary care, and public
psychiatry), gender and age. Within strata, we applied a minimization method to achieve a
balanced distribution. We concealed allocation by using an independently operated computer
Dose-escalation consisted of incremental steps of one capsule every 5 days towards a
maximum of 50 mg/day (20 mg + 3 capsules). Patients were allowed to increase at a slower pace
(e.g. by 7 days) or stop further escalation (e.g. 20 mg + 2 capsules) according to adverse effects.4
No dosage adjustments were allowed during the last 3 weeks of the study. We checked
adherence by pill-counts and anamnesis.5
Figure 2.1. Design of the DELPHI-study.
DELPHI-SPECT and DELPHI-fMRI represent sub-studies nested within the total study. For these 2 sub-studies drug-free
patients were recruited and treated in the Academic Medical Center. For color figure see page 280.
?? ???????? ?????
Design of the DELPHI study
Figure 2.2 summarizes patient disposition in the DELPHI-study and the DELPHI-SPECT/fMRI
General in- and exclusion criteria
Eligible patients met the following inclusion criteria: Age between 18 and 70 years,major
depressive disorder (MDD) determined by the structured clinical interview for DSM-IV (SCID),6
and a HDRS (17 items)7 score above 18. All participants were either drug-free and/or had
undergone no more than one antidepressant treatment (other than paroxetine) at an effective
dose for ≥ 6 weeks for the present MDD-episode. By the latter criterion, we avoided treatment
resistance as potential bias for inefficacy of dose-escalation. Exclusion criteria, apart from
pregnancy (or wish to become pregnant), were bipolar disorder, psychotic features, neurological
cognitive impairments (i.e. dementia), primary anxiety and/or substance abuse disorders and
acute, severe suicidal ideation. Contrary, we allowed secondary co-morbid anxiety and/or
substance abuse to increase applicability of our findings. In total, 278 patients were referred, of
whom 107 patients started treatment with paroxetine 20 mg/day. Twenty patients withdrew from
the study before week 6, and 27 patients were responder by then. Sixty non-responders were
Figure 2.2. Patient enrolment and disposition over sub-studies.
The patients in DELPHI-SPECT and DELPHI-fMRI sub-studies are combined (left). For disposition of patients randomized see
???????? ????????? ???????
????????? ??????????? ????????
???????? ??????? ???????????
??????? ??? ???????? ???????
?????????? ?? ???????? ???????
??????????????? ?? ??????????? ???????
???????? ????? ??????? ??????
???????? ?????????? ????????
????????????????? ????? ???????
?????????? ???????????????????????????? ?????????????????
??????? ??????? ??????
??????? ??????? ??????
????? ??????? ??????
????????? ????? ??????????????
??????? ????????????? ?????
??????? ??????? ??????
??????? ????????????? ?????
??????? ??????? ??????
??????? ????? ????????? ??????
?? ?????????????? ??????????? ???????
? ?????????? ???????? ????????????
? ???????? ??? ???????? ???????
?? ?????????????? ??????????? ???????
? ?????????? ???????? ????????????
? ?????????? ?????????? ???????
???????? ??? ??????????? ???????? ???????
? ?? ??????????? ??????????????????
? ?? ???? ????????????????????? ???????
? ?? ??????????? ???????
? ?? ???? ?????????? ???????
???????? ????? ???????
DELPHI-SPECT and fMRI
Patients who were drug-free (for >4 weeks and ≥5 half-lives of a previous antidepressant) were
asked to additionally participate in the neuroimaging sub-studies. Patients were initially asked to
participate in DELPHI-SPECT only, but from august 2005 onwards, when DELPHI-fMRI was
approved, we asked them for both DELPHI-SPECT and DELPHI fMRI. Participation in only one of
the neuroimaging studies was also possible. We limited age to 25-55 years to reduce variability in
SERT-measurements by age.8 For participation in the fMRIs, patients had to be free of metal
objects in their body. None of the included patients reported past or present use of 3,4-
methylenedioxymethamphetamine. We treated DELPHI-SPECT and DELPHI-fMRI patients at
theoutpatient department of the Program for Mood Disorders. We supplied medication in
We included 51 patients in the DELPHI-SPECT study, of whom 33 non-responders were
randomized. For DELPHI-fMRI we included 22 patients, of whom 16 were randomized. Twenty
patients participated both in the DELPHI-SPECT and DELPHI-fMRI. Unfortunately not all
(repeated) scans were analyzable adequately, the reason why in chapters 8-10 different numbers
of patients are described.
We recruited 53 healthy controls as reference for the study-entry scans. We individually matched
each patient in DELPHI-SPECT and DELPHI-fMRI by gender and age (±2.5 years). Healthy controls
were in good physical health, and had never used psychotropic medication. Exclusion criteria
were current or lifetime psychiatric disorder(s) according to the SCID (including abuse or
addiction disorders), a Beck Depression Inventory (BDI) score >9, alcohol use >4 units per day (last
month) or a 1st-degree relative with psychiatric disorder(s). We allowed healthy controls to have
incidentally used illicit drugs, unless criteria for a DSM-IV disorder was met, but we prohibited illicit
drug use the month prior to scanning. Twenty healthy controls also participated both in the
DELPHI-SPECT and DELPHI-fMRI.
Time points and questionnaires
We administered the HDRS17,7 Inventory for depressive symptoms (IDS-SR30),9 the occurrence of
adverse effects and health-related quality of life (MOS-SF36)10 at study-entry, randomization (T0),
and 6 weeks after randomization (T1). Adverse effects and depressive symptoms were also
monitored in the weeks 1, 2 and 4 after the initial start of treatment, and after randomization,
using the Maier and Bech subscales11 and IDS-SR30
administered clinician-rated questionnaires. Agreement between raters was good (intraclass
correlation coefficient = 0.98). Raters and patients were blinded for treatment.
9 (Figure 2.1). Three trained investigators
Definition of primary and secondary outcomes
Primary clinical outcomes were HDRS17-scores and the proportion of patients achieving response
(≥50% decrease in HDRS17) or remission (HDRS17 ≤7). Secondary outcomes were total and specific
(adverse effects / inefficacy) dropout rates, the Maier and Bech subscales and IDS-SR30-scores,
the occurrence of adverse effects and health-related quality of life.
Paroxetine serum concentrations: See chapters 8 and 9 for details.
SERT-gene promoter polymorphism: See chapters 8 and 11 for details.
Saliva and additional blood specimens
At study-entry, at T0, and T1, we collected two saliva specimens at awakening and 30 minutes
thereafter, to determine salivary cortisol, dehydroepiandrosterone-sulphate, and α-amylase.12-15
Furthermore, we collected blood-specimens to quantify [3H]Noradrenaline and [3H]Serotonine
uptake in ex-vivo models.3;16;17 Finally we collected blood-specimens to determine levels of platelet
and plasma ω-3/ω-6 PUFAs.18 These neurobiological measurements remain to be analysed in
different papers not in this thesis.
Design of the DELPHI study
Measurement of SERT occupancy
We performed SPECT imaging at study-entry, T0 and T1 (Figure 2.1), between 2 to 10 pm according
to previously described procedures.19 We made all scans 230 ±18 (SD) minutes after intravenous
injection of approximately 100 MBq [123I]β-CIT, when the radioligand is at equilibrium for SERT
binding in brain areas expressing high densities of SERTs.20 We performed SPECT imaging using a
12-detector single slice brain-dedicated scanner (Neurofocus 810, Strichmann Medical Equipment;
Cleveland, OH). After attenuation correction and reconstruction in 3D mode (http://
www.neurophysics.com), we defined regions of interest (RoIs) for midbrain, diencephalon
(regions rich of SERT) and cerebellum (as a reference) by using validated templates (Figure 2.3).19
For further details of SPECT-procedures see chapters 7-10.
Magnetic resonance imaging of functional activation of thecortico-limbic-network
We acquired fMRI scans at study-entry, T0 and T1 (Figure 2.1), between 2 to 10 pm. FMRI-sessions
lasted 50-60 minutes, each including a cognitive task (Tower of London),21 a structural scan and a
facial expression task, reported in this thesis.22;23 We used a 3Tesla Intera MRI scanner (Philips,
Eindhoven, NL), with a 6 channel head-coil for radiofrequency reception. Two magnet compatible
response boxes were used to record subject’s performance and reaction times. For further details
of fMRI-settings and the parameters of the faces paradigm: see chapter 10.
Power and interim analysis
For the randomization-phase of the total DELPHI-study, we performed a-priori power-calculations
for two co-primary endpoints. We planned an interim analysis after SPECT data had been collected
on at least 30 patients. Stopping criteriawere predetermined using the O'Brien and Fleming24
approach, and were p<0.0026 in case of superiority and p>0.50 for futility. See chapter 9 for
Figure 2.3. Regions of Interest (RoI) for Midbrain, Cerebellum and Diencephalon.
Example of SPECT images after 3D reconstruction. Templates with fixed RoIs are shown. For color figure see page 281.
A. Midbrain (circle) and cerebellum. B. Striatum (for demarcation midbrain-diencephalon) and diencephalon (circle). RoIs
were positioned by hand and situated based on anatomy and maximum concentration of activity/ml in the RoI.
Stichting Farmaceutische Kengetallen. Data en feiten 2007.2007. Den Haag, Stichting Farmaceutische Kengetallen.
Anonymous. Paxil® Prescribing information and medication guide PXL:47PI (available at http://us.gsk.com/products/
assets/us_paxil.pdf). 1-44. 2008. Research Triangle Park, North Carolina, GlaxoSmithKline.
Gilmor ML, Owens MJ, Nemeroff CB. Inhibition of norepinephrine uptake in patients with major depression treated with
paroxetine. Am J Psychiatry. 2002; 159: 1702-1710.
Baker CB, Woods SW. Is there a SSRI dose response in treating major depression? The case for re-analysis of current data
and for enhancing future study design. Depress Anxiety. 2003; 17: 10-18.
Saunders K, Simon G, Bush T, Grothaus L. Assessing the feasibility of using computerized pharmacy refill data to monitor
antidepressant treatment on a population basis: a comparison of automated and self-report data. J Clin Epidemiol. 1998;
First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical Interview for DSM-IV Axis I Disorders Patient Edition
(SCID-I/P version 2.0). Translated in Dutch by. Groenestijn MAC, Akkerhuis GW, Kupka RW, Schneider N, and Nolen
WA.1999. Lisse, the Netherlands, Swets & Zeitlinger B.V.
Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry. 1960; 23: 56-61.
van Dyck CH, Malison RT, Seibyl JP, Laruelle M, Klumpp H, Zoghbi SS et al. Age-related decline in central serotonin
transporter availability with [123I]β-CIT SPECT. Neurobiol Aging. 2000; 21: 497-501.
Rush AJ, Gullion CM, Basco MR, Jarrett RB, Trivedi MH. The Inventory of Depressive Symptomatology (IDS):
psychometric properties. Psychol Med. 1996; 26: 477-486.
10. Aaronson NK, Muller M, Cohen PD, Essink-Bot ML, Fekkes M, Sanderman R et al. Translation, validation, and norming of
the Dutch language version of the SF-36 Health Survey in community and chronic disease populations. J Clin Epidemiol.
1998; 51: 1055-1068.
11. Ruhe HG, Dekker JJ, Peen J, Holman R, Jonghe F de. Clinical use of the Hamilton Depression Rating Scale: is increased
efficiency possible? A post hoc comparison of Hamilton Depression Rating Scale, Maier and Bech subscales, Clinical
Global Impression, and Symptom Checklist-90 scores.Compr Psychiatry. 2005; 46: 417-427.
12. Nater UM, Rohleder N, Schlotz W, Ehlert U, Kirschbaum C. Determinants of the diurnal course of salivary alpha-amylase.
Psychoneuroendocrinology. 2007; 32: 392-401.
13. Pruessner JC, Wolf OT, Hellhammer DH, Buske-Kirschbaum A, von Auer K, Jobst S et al. Free cortisol levels after
awakening: a reliable biological marker for the assessment of adrenocortical activity. Life Sci. 1997; 61: 2539-2549.
14. Rohleder N, Nater UM, Wolf JM, Ehlert U, Kirschbaum C. Psychosocial stress-induced activation of salivary alpha-amylase:
an indicator of sympathetic activity? Ann N Y Acad Sci. 2004; 1032: 258-263.
15. Schmidt-Reinwald A, Pruessner JC, Hellhammer DH, Federenko I, Rohleder N, Schurmeyer TH et al. The cortisol response
to awakening in relation to different challenge tests and a 12-hour cortisol rhythm. Life Sci. 1999; 64: 1653-1660.
16. Owens MJ, Morgan WN, Plott SJ, Nemeroff CB. Neurotransmitter receptor and transporter binding profile of
antidepressants and their metabolites. J Pharmacol Exp Ther. 1997; 283: 1305-1322.
17. Owens MJ, Krulewicz S, Simon JS, Sheehan DV, Thase ME, Carpenter DJ et al. Estimates of Serotonin and Norepinephrine
Transporter Inhibition in Depressed Patients Treated with Paroxetine or Venlafaxine. Neuropsychopharmacology. 2008.
18. Dacremont G, Cocquyt G, Vincent G. Measurement of very long-chain fatty acids, phytanic and pristanic acid in plasma
and cultured fibroblasts by gas chromatography. J Inherit Metab Dis. 1995; 18 Suppl 1: 76-83.
19. Win MM de, Habraken JB, Reneman L, Brink W van den, Heeten GJ den, Booij J. Validation of [123I]β-CIT SPECT to assess
serotonin transporters in vivo in humans: a double-blind, placebo-controlled, crossover study with the selective serotonin
reuptake inhibitor citalopram. Neuropsychopharmacology. 2005; 30: 996-1005.
20. Pirker W, Asenbaum S, Hauk M, Kandlhofer S, Tauscher J, Willeit M et al. Imaging serotonin and dopamine transporters
with [123I]β-CIT SPECT: binding kinetics and effects of normal aging. J Nucl Med. 2000; 41: 36-44.
21. Heuvel OAvan den, Veltman DJ, Groenewegen HJ, Cath DC, Balkom AJ van, Hartskamp Jvan et al. Frontal-striatal
dysfunction during planning in obsessive-compulsive disorder. Arch Gen Psychiatry. 2005; 62: 301-309.
22. Hariri AR, Bookheimer SY, Mazziotta JC. Modulating emotional responses: effects of a neocortical network on the limbic
system. Neuroreport. 2000; 11: 43-48.
23. Wolfensberger SPA, Veltman DJ, Hoogendijk WJG, Boomsma DI, Geus EJC de. Amygdala responses to emotional faces in
twins discordant or concordant for the risk for anxiety and depression. Neuroimage. 2008; E-pub: doi: 10.1016/
24. O'Brien PC, Fleming TR. A multiple testing procedure for clinical trials. Biometrics. 1979; 35: 549-556.
Design of the DELPHI study
STUDIES TO GUIDE CLINICAL TREATMENT
OF MAJOR DEPRESSIVE DISORDER
Studies to guide clinical treatment of Major Depressive Disorder
CLINICAL USE OF THE HAMILTON DEPRESSION
RATING SCALE: IS
A POST HOC COMPARISON OF HDRS, MAIER AND
BECH SUBSCALES, CGI AND SCL-90 SCORES.
Comprehensive Psychiatry 2005; 46: 417-427
Henricus G. Ruhé1,2
Jack J. Dekker1,3
Frans de Jonghe1,2
1 Mentrum, Amsterdam, The Netherlands
2 Department of Psychiatry, Academic Medical Center, University of Amsterdam, Amsterdam, The
3 Department of Clinical Psychology, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands
4 Department of Clinical Epidemiology and Biostatistics, Academic Medical Center, University of
Amsterdam, Amsterdam, The Netherlands
The 17-item Hamilton Depression Scale (HDRS) is used as a semi gold standard in research. In
treatment guidelines, the HDRS-measurements serve to determine response, remission and guide
clinical decision-making for non-responders. However, its use in clinical practice is limited, possibly
because the HDRS is time-consuming. Additionally, the multidimensional HDRS is criticized for not
measuring a unidimensional aspect as depression severity. The Maier and the Bech, two 6-item
severity subscales extracted from the HDRS, are relatively unknown.
To investigate whether the measurements obtained with Maier and Bech subscales are
comparable with the original HDRS-measurements.
Data from two randomized controlled trials in 482 male and female patients, diagnosed with a
major depression (with or without dysthymia) according to DSM-III-R, of whom 219 participated in
the trials, were reanalyzed. A standardized stepwise psychopharmacological treatment was
compared with a combination of pharmacotherapy with Short Psychodynamic Supportive
Psychotherapy in a psychiatric outpatient department. Outcome measures were internal
consistency and concurrent validity of HDRS, Maier, Bech, Clinical Global Impression (CGI) scales
and Symptom CheckList depression subscale. Effect sizes of HDRS, Maier and Bech were used to
compare measured treatment effects for the randomized subjects participating in the trials. Item
Response Theory was used to obtain conversion tables for the HDRS, Maier, Bech and SCL-90
We found moderate internal consistence (Cronbach α ≈ 0.6-0.7) and high correlations of the Maier
and Bech subscales with overall HDRS-scores. Overall there were no clinically relevant differences
in effect sizes between Maier, Bech and HDRS, although some differences were statistically
significant. Receiver operating characteristic curves showed no difference between Maier and
Bech to define remission, but showed the CGI ratings to be unreliable. A cut-off ≤4 corresponded
with a HDRS ≤ 7 criterion in both subscales.
In clinical practice, both Maier and Bech scales can be used as equivalents of the HDRS, but will be
Clinical use of HDRS subscales
Major depressive disorder is a severe, disabling illness, expected to be the world’s second health-
problem in 2020.1 Depression is associated with high costs, regarding direct treatment and
indirect costs of loss of productivity and quality of life.2 Several clinical guidelines were developed
to guide the treatment of this disorder, both psychotherapy and pharmacotherapy (or in
combination) appear effective.3-10
The use of self-report or clinician-rated symptom-scales is recommended to assess severity
and response to treatment.8;11;12 Some experts claim clinician-rated symptom scales to have a
larger validity and reliability than self-reporting scales, especially in patients with cognitive
impairment, and more severe or psychotic depressions.11;13;14 Specific symptom-scales are more
reliable than global rating scales.11;13;15 Especially, rating scales can be used to objectively
determine specific cut-off points for response and remission.12;16;17
In most clinical trials the Hamilton Depression Rating Scale (HDRS)18;19 – a clinician rated
symptom-scale – is used as a standard to determine severity and response.5;8;11;15;20-23 Many versions
of the HDRS exist, with the number of items usually varying between 17 and 24,11;18;19;22 however up
to 36 items have been described.23 Longer versions were especially developed to cover reverse
neurovegetative (atypical) symptoms.23 The Clinical Global Impression (CGI)24 – a clinician-rated
global scale – is also frequently used.5;8;15;25 In clinical practice, although recommended, rating-
scales are not used routinely. Explanations for this discrepancy could be ignorance of existing
scales, a strong belief in one’s clinical judgment, an unsystematic approach of depression, but also
the amount of time needed for rating-scales (e.g. 15-20 minutes for the HDRS11) and the necessity
The HDRS is criticized as being sensitive to somatic symptoms (e.g. somatic illness or side-
effects of drugs),11;15;27;28 for not rating all 9 DSM-IV domains, its unequal weightings of different
symptoms and for the multidimensionality of the HDRS total score.13;21;29-31 Multidimensionality is
important to cover the maximum range of clinical features of major depressive disorder, but does
not necessarily measures depression severity. Multidimensional scales can be misleading when
measurement of severity and treatment response is concerned,13;21;28 especially when the
measured depressive symptoms do not change proportionally with depression severity.Finally,
some reports emphasize that the HDRS systematically favors (sedative) Tricyclic Antidepressants
(TCAs) above Selective Serotonin Reuptake Inhibitors (SSRIs).27;32-35 Sleep and somatic items may
appear to be 'improved' by side-effects of TCAs, but worsened by side-effects (e.g. insomnia,
gastrointestinal complaints, agitation) of SSRIs.
In order to overcome the problems of the multidimensional HDRS mentioned above, a more
unidimensional subscale from the HDRS covering core-symptoms of severity is desired. Also, from
a clinical point of view, fewer items will be less time consuming for application by busy clinicians.
However, for the purpose of reference, subscale scores must remain anchored to the original
HDRS. To identify shorter unidimensional subscales, Maier et al.28 used Rasch- and Mokken-
analyses and Gibbons et al.29 used Factor-analysis. Bech and collegues developed another 6-item
subscale. This scale initially emerged from an analysis with experienced psychiatrists as a validity
criterium,36 and was validated psychometrically thereafter using Rasch-analyses.37;38 This Bech
subscale was combined with four items of the Cronholm-Ottosson Depression Scale to form the
Bech-Rafaelsen Melancholia Scale.39 Santor and Coyne examined the score-performances of
individual HDRS-items as a function of depression severity with a nonparametric Item Response
Theory (IRT) approach, retaining 14 items.21 These 14 items included all 6 items of the Maier
subscale and all 8 items of the Gibbons subscale. However one item from the Bech subscale (13,
somatic symptoms) was not included.
In a meta-analysis of individual patient-data, Faries et al. evaluated the responsiveness of total
HDRS and subscale scores in TCA and SSRI pharmacotherapy trials, finding a maximal sensitivity
for the Maier subscale.40 In a similar reanalysis, Entsuah et al. found larger effect sizes for the
Bech, Maier and Gibbons subscales compared to the HDRS in trials comparing SSRIs or
venlafaxine.41 O'Sullivan et al. found comparable sensitivity to detect changes for the six-item
Bech subscale compared to the 17-item HDRS.20 Hooper et al. found equal sensitivity to change
during treatment for the 6-item Bech subscale compared to the HDRS 17 item version.26 Möller
and Bech et al. used the Bech subscale to reexamine treatment efficacy of SSRIs and mirtazapine
(versus TCAs or placebo).32;42-44 The latter publications did not provide data for the Maier subscale.
In this paper we describe a secondary analysis of our trial data, in order to answer the
1) Are the Maier, Bech and HDRS comparable in the measurement of depression severity and
the sensitivity to measure changes in severity? 2) Is this comparability stable across the full range
of response to treatment (e.g. non-response, partial and full response), across different
treatments and different baseline severity of depression? and 3) What are clinical cut-off points
for the subscales to determine remission compared to conventional definitions.12;16;17
We hypothesized that the differences between Maier, Bech and HDRS-scales would be small
and that there would be no apparent effect modification across neither treatments nor baseline
severity. In contrast, we hypothesized that for non-responders and partial responders the effect
sizes would be smaller than for responders. This would additionally prove the hypothesis of
sensitivity to change.
In the present analyses we use data from two published randomized controlled trials conducted
between 1993 and 1998 which were published or accepted for publication.45;46 The first trial aimed
at efficacy and effectiveness of pharmacotherapy versus the combination of pharmacotherapy
with Short Psychodynamic Supportive Psychotherapy (SPSP)47-50 (16 sessions).45 The second trial
investigated efficacy and effectiveness of a combination of pharmacotherapy with 8 versus 16
sessions of SPSP.46 Pharmacotherapy in both trials consisted of three successive steps in case of
intolerance or inefficacy. Both trials started with fluoxetine (20 mg/ day), when this was
unsuccessful (CGI-I >2, only 'minimally improved' or worse) after 6 weeks amitriptyline (≥150 mg/
day, dependent of plasma-levels) was initiated in trial 1 and nortriptyline (≥150 mg/day, dependent
of plasma-levels) in trial 2. If again unsuccessful after 6 weeks, moclobemide (300-600 mg/day)
was started in trial 1 and mirtazapine (30-45 mg/day) in trial 2.
Inclusion criteria for participation in the trials were age between 18 and 60 years, DSM-III-R
defined Major Depression (with or without dysthymia) assessed in a structured clinical interview,
a 17-item HDRS baseline score of at least 14 points and written informed consent. Patients were
excluded in case of psycho-organic or psychotic or dissociative disorders, drug abuse, or when the
patient was considered to be too unreliable to participate in a clinical trial. Other axis 1 co-
morbidity was not excluded. Further exclusion criteria were if there was a serious communicative
or practical problem (e.g. language barrier or the patient will soon leave the country), if there was
a contraindication for one of the antidepressants used, if the patient was adequately treated with
antidepressants during the present depressive episode, if the patient used other psychotropic
medication, or if the patient was or planned to become pregnant. Additional exclusion criteria
were of the usual kind in drug research: “too ill” (e.g. antidepressants must be started
immediately) and/or “too suicidal” (e.g. hospitalization is unavoidable) to participate in a clinical
trial. The study was approved by the medical ethics committee. After complete description of the
study to the subjects, written informed consent was obtained.
Of 3226 newly registered outpatients, 988 patients had a depressive disorder. By initial
screening 503 of these 988 patients were excluded by the above exclusion-criteria leaving 485
subjects (including patients that later refused to participate or had a HDRS below 14; further
referred to as the diagnostic sample). To enter the trials, a second exclusion check was performed
by a psychiatrist (excluding 73 patients), and 142 subjects with a HDRS-17 <14 were excluded,
Clinical use of HDRS subscales
leaving 270 patients for randomization. After randomization 51 patients refused participation,
leaving 219 patients who started the proposed therapy (further referred to as the per protocol
In this manuscript we used the diagnostic sample for most cross-sectional analyses, and the
randomized patients in the per protocol sample for analyses of sensitivity of response-data. For
non-completers, the last observation was carried forward (LOCF).
Primary outcome-measures were the 17-item HDRS,18;19 the Maier subscale of the HDRS
(containing items 1,2,7-10),28 the Bech subscale of the HDRS (items 1,2,7,8,10 and 13),37 the Clinical
Global Impression Severity (CGI-S) and Improvement (CGI-I) scale,24 and the Symptoms check list,
90 items (SCL-90) depression subscale (SCL-90dep).51;52 Thus, three levels of information were
obtained: data from 1) an independent, trained, supervised and blinded research-assistant (HDRS-
17 & Maier, Bech), 2) the treating clinician (CGI-S/I) and 3) the patient (SCL-90dep). The HDRS was
administered using a semi-structured interview.53 Before participating in the study, the reliability
of the HDRS-assessments was established. During the study, in order to avoid slippage,
audiotaped assessments were discussed monthly.
In the analyses of treatment-efficacy, response was defined as a ≥50% HDRS-score reduction,
partial response as ≥ 20-50% reduction in HDRS-score and remission as a HDRS score of 7 points or
Cronbach-α coefficients and mean inter-item correlations were used to express internal
consistency. To check whether the increased number of items in the HDRS accounted for a higher
Cronbach-α coefficient than in the subscales (with only 6 items), we applied the Spearman-Brown
formula.55 Next we calculated concurrent validity as Pearson correlation coefficients between
total HDRS, Maier and Bech subscale scores and SCL-90dep scores. Linear regression models
calculated variance of HDRS-scores explained by the subscales.56 These analyses were performed
in our diagnostic sample. Concurrent validity between CGI-S/I and HDRS subscale-ratings was
determined also, however, to avoid low correlations due to limited dispersion, this was done for
the last observation in the per protocol sample. The CGI improvement scale was compared with
changes expressed as percentages of the baseline score.
In order to compare differences in sensitivity to measure treatment effects (also referred to
as responsiveness), in data from the per protocol sample effect sizes (E-S) for HDRS, Maier and
Bech subscales were calculated per subject as the within-subject changes in scale-scores divided
by the pooled standard deviation of the mean change in scale score
differences in effect sizes could be tested and 95% confidence intervals (95% CI) could be
calculated. Differences in E-S between the scales were tested by paired T-tests. In order to
determine significant effect-modification, the above analyses were repeated while data were
stratified. For stratification we used initial HDRS scores of at least 19 for severe depression,11
criteria for response as described above and treatment-condition. Differences in E-S between
strata were tested by Analysis of variance-models (ANOVAs).
The Partial Credit Item Response Theory (IRT) model57 was used to estimate the relationships
between total scores on the HDRS and total scores on the Maier and Bech subscales of the HDRS.
The scores were those obtained at exit (per protocol sample). The computer program OPLM58
was used to obtain a set of weights for each item in the HDRS using conditional maximum
likelihood methods. The same software and the item weights were used to obtain estimates of
the latent trait associated with each score on the HDRS, the Maier subscale and the Bech
subscale. The total scores for the pairs of scales were equated by matching the total scores for
which the latent trait scores were most similar.59 These methods are very similar to those used in a
previous publication about the Quick Inventory for Depressive Symptomatology IDS.60 The range
of SCL scores associated with each HDRS score was obtained directly from the original data.
.20 In this way,
Chapter Download full-text
Finally, Receiver operator characteristic (ROC) curves were constructed to summarize validity
of cut-off points. Differences in areas under the curve (AUC) were tested with attention for
interrelation (because we studied these tests within the same subjects) as described by Hanley.61
For all data analysis except the IRT analysis, SPSS for Windows version 10.1 was used.62 For all tests
two-tailed significance levels were applied.
Table 3.1. Studied populations.*
Per protocol sample (n= 219)‡
Trial I Trial II Combined
I + II
Sex (% female) 60.3 63.2 61.1 60.0 68.9 63.0
Age 35.3 ±9.9 34.9 ±8.2 34 ±9.4 38.1 ±10.5 36.2 ±10.5 35.5 ±9.7
Marital status (%):
Educational level (%):
Duration of episode
< 1 year
> 2 years
Psychiatric treatment during this
95 (20.2) 13 (23.6) 9 (12.9) 5 (11.4) 10 (22.7) 37 (17.4)
≤3 mths before (%)
Depressive episodes (prev. 5
3 or more
Baseline scores of rating scales
HDRS-17 17.1 ±6.5 21.0 ±4.8 20 ±4.9 19.4 ±3.8 20.3 ±4.4 20.2 ±4.6§
Maier 9.2 ±3.6 11.0 ±2.9 10.9 ±2.8 10.7 ±2.3 10.9 ±2.9
9.4 ±3.7 11.5 ±2.8 11.2 ±2.7 10.7 ±2.3 11.0 ±3.0
4.7 ±0.7 4.8 ±0.6 4.7 ±0.7 4.5 ±0.7 4.6 ±0.6 4.7 ±0.7
SCL-90 depression subscale
* Data represent means (±SD) unless indicated. Denominators of percentages vary due to missing values.
† Total diagnostic sample.
‡ No significant differences between treatment-groups (per protocol sample) (ANOVA or χ2).
§ Significant differences (p< 0.05) between in- and excluded patients (indep. T-Test).
¶ n= 241
References to studies: Trial I45 and Trial II46
45.9 ±11.8 48.7 ±11.7 47.8 ±9.8 49.3 ±8.7 52.0 ±10.1