Opinion

Stuck in a rut: rethinking depression

and its treatment
Paul E. Holtzheimer and Helen S. Mayberg
Department of Psychiatry and Behavioral Sciences, Emory University School of Medicine, 101 Woodruff Circle NE, Suite 4000,
Atlanta, GA 30322, USA

The current definition of major depressive disorder (MDD) resistant depression (TRD). We then present a brief history
emerged from efforts to create reliable diagnostic criteria of antidepressant treatment development and emphasize
for clinical and research use. However, despite decades of that, after several decades of research, antidepressant effi-
research, the neurobiology of MDD is largely unknown, cacy has not substantially improved. We next consider the
and treatments are no more effective today than they most commonly posited explanation for this, namely the
were 50–70 years ago. Here, we propose that the current clinical heterogeneity of the disorder, which suggests differ-
conception of depression is misguiding basic and clinical ent biological subtypes of depression. We conclude that the
research. Redefinition is necessary and could include a current construct of depression, as defined for the purposes
focus on a more narrowly defined set of core symptoms. of clinical and basic research, has lost its utility. We note
However, we conclude that depression is better defined that focusing investigation towards specific symptoms con-
as the tendency to enter into, and inability to disengage sidered core to the illness might be more fruitful in elucidat-
from, a negative mood state rather than the mood state ing the neurobiology of the disease and developing improved
per se. We also discuss the implications of this revised treatments. Moreover, we propose that the illness ‘depres-
definition for future clinical and basic research. sion’ is better defined as the inability to disengage from a
negative mood state and the tendency to re-enter this state
Introduction inappropriately: that is, the tendency to get ‘stuck in a rut’.
These propositions are supported by emerging data on deep
‘If a fright or despondency lasts for a long time,
brain stimulation (DBS) in the treatment and study of
it is a melancholic affection’
depression.
Hippocrates, Aphorisms, Section 6, no. 23
The extent and cost of depression
Depression has been recognized as a distinct and serious The US lifetime prevalence of MDD is 17%, with twice as
condition for more than 2000 years. Although definitions many women affected as men [2]. Median age of symptom
have varied, the core construct of depression has remained onset is 30 years, although the disorder can arise at almost
intact: a condition of despondency and anergia akin to any age [2]. MDD is moderately heritable (ranging from
normal sadness, but crucially different in its severity, 30% to 40%) [3], supporting a role for both genetic and
duration and/or lack of a triggering event. In modern environmental risk factors. MDD is the leading cause of
psychiatry, major depressive disorder (MDD; Glossary) years lost owing to disability worldwide and the third
is broadly defined as a syndrome including episodes of overall contributor to the worldwide burden of disease
persistent negative mood (feeling down, sad, blue, etc.) (but projected to be the biggest contributor by 2030) [4].
or anhedonia (a loss of interest in pleasurable activities), Several treatments are available for depression, includ-
plus a combination of additional emotional, psychological ing medications, psychotherapy and various somatic treat-
and somatic symptoms (Box 1) [1]. This constellation of ments, such as electroconvulsive therapy (ECT), vagus
symptoms must persist for at least two weeks and not be nerve stimulation (VNS) and transcranial magnetic stim-
secondary to an obvious substance, medical condition, or ulation (TMS). Unfortunately, up to two-thirds of patients
other psychiatric disorder; neither can the symptoms be remain symptomatic following first-line treatment [5] and
better explained by normal grief. This definition emerged a third fail to achieve remission (defined as full resolution
from efforts to develop reliable diagnostic criteria for use in of depressive symptoms) after four established treatments
clinical and research settings, culminating in the Diagnos- [6]; approximately 10–20% of depressed patients can show
tic and Statistical Manual of the American Psychiatric little if any improvement despite multiple, often aggressive
Association [currently in its 4th edition (DSM-IV-TR)] [1]. treatments. Thus, a conservative estimate places the US
Here, we describe the inadequacy of available antide- prevalence of TRD at 1–3%. Importantly, this estimate
pressant treatments and conclude that a shift is needed in does not include the prevalence of dysthymia or treat-
the definition of depression and the research agenda going ment-resistant bipolar depression.
forward. We provide a brief overview of the epidemiology of Even among those patients who achieve remission with
depression, including the high prevalence of treatment- antidepressant treatment, recurrence is the rule rather
than the exception: the probability of another depressive
Corresponding author: Holtzheimer, P.E. (pholtzh@emory.edu). episode is >60% after the first, >70% after the second, and
0166-2236/$ – see front matter ß 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.tins.2010.10.004 Trends in Neurosciences, January 2011, Vol. 34, No. 1 1
Opinion Trends in Neurosciences January 2011, Vol. 34, No. 1

Glossary >90% after the third [1]. Relapse rates may be lower
among patients maintained on antidepressant medica-
Atypical depression: a clinical subtype of depression defined by normal to
increased mood reactivity together with increased sleep, increased appetite, tions [7,8] and those receiving psychotherapy [9,10]. How-
rejection sensitivity and a heavy, leaden feeling in the limbs. ever, in a recent, large, multi-center trial assessing the
Bipolar disorder: a mood disorder characterized by periods of euthymia
(normal mood state), major depressive episodes, manic or hypomanic
efficacy of sequential treatments for MDD, the six-month
episodes, and/or mixed episodes (occurrence of a major depressive and manic relapse rate was especially high (between 34% and 83%)
episode concurrently). [6]. Notably higher relapse rates were seen in patients with
Chlorpromazine: an antipsychotic medication with a primary mechanism of
action of blocking dopaminergic receptors (although it also has effects on other
greater treatment resistance (even if remission was
neurotransmitter systems); chlorpromazine and other antipsychotics are achieved).
primarily used in the treatment of psychotic disorders (e.g. schizophrenia),
but have also shown efficacy as augmentation medications for treatment-
resistant depression.
A century of antidepressant treatment research:
Cognitive behavioral therapy (CBT): a form of psychotherapy that attempts to limited progress
correct cognitive distortions and dysfunctional behaviors that are posited to In the absence of effective pharmacological treatments,
cause, or contribute to, emotional illness.
Deep brain stimulation (DBS): a neurosurgical treatment in which one or more attempts to treat malignant psychiatric illness have often
electrodes are inserted into a specific brain structure and a focal electrical pursued an aggressive course (Figure 1). This began with
current is delivered to the surrounding brain tissue; stimulation parameters are an early attempt to treat psychosis with cortical resection
typically controlled via a computer/battery pack implanted subcutaneously
(often in the chest wall). [11] and peaked with the mid-20th-century popularity of
Dysthymia: a mood disorder characterized by two to four depressive prefrontal leucotomy (the surgical ablation of prefrontal
symptoms occurring nearly every day for more than two years; although
white matter tracts) for multiple neuropsychiatric indica-
associated with fewer depressive symptoms than is an MDE, dysthymia is
equally distressing and disabling owing to its chronic nature. tions [12]. These treatments were based on a rudimentary
Electroconvulsive therapy (ECT): a treatment that involves the transcranial understanding of the neural circuitry of emotional and/or
delivery of electricity to induce a generalized seizure under general anesthesia;
a treatment course will typically include two to three treatments per week over
behavioral regulation derived from animal and human
three to four weeks. brain injury data [13] and the hypothesis that severing
Lithium: a psychotropic medication used in the treatment of bipolar disorder white matter connections between cortical and subcorti-
where it has shown antimanic and antidepressant properties, as well as the
ability to prevent mood episodes over time; lithium is also an accepted
cal structures would diminish aberrant behavior. Modern
augmentation medication for MDD; the primary mechanism of action for ablative strategies include subcaudate tractotomy, ante-
lithium is unknown. rior capsulotomy, anterior cingulotomy and limbic leucot-
Major depressive disorder (MDD): a mood disorder characterized only by major
depressive episodes; also referred to as unipolar depression (to distinguish it
omy. These approaches involve more focal disruption of
from depression in the context of bipolar disorder). white matter tracts using stereotactic neurosurgical tech-
Major depressive episode (MDE): a mood episode defined by the presence of niques. Observational data suggest efficacy in 25–50% of
depressed mood or anhedonia, plus a combination of additional emotional,
cognitive, behavioral and/or somatic symptoms (Box 1). patients across a variety of treatment-refractory psychi-
Manic episode: a mood episode defined by elevated or irritable mood plus a atric disorders, including depression and obsessive-com-
combination of additional symptoms (grandiosity, racing thoughts, de- pulsive disorder (OCD) [14]. Side effects can include
creased need for sleep, increased rate and volume of speech, increased
goal-directed activity, and/or increased participation in pleasurable but risky cognitive detriments, personality changes and seizures
activities). [14].
Monoamine oxidase inhibitor (MAOI): a class of antidepressant medications
Several non-surgical, but still relatively aggressive,
with a primary mechanism of action of inhibiting the enzyme monoamine
oxidase, leading to an increase in available monoaminergic neurotransmitters treatments were also used during the early 20th century
(serotonin, norepinephrine and dopamine). for severe psychiatric illness, including malarial pyrother-
Melancholic depression: a subtype of depression associated with profound
anhedonia, severe weight loss, significant psychomotor agitation or retarda-
apy [15], camphor-induced seizures [16] and hypoglycemic
tion, severe insomnia with early morning awakening, excessive guilt and coma [17]. In general, these treatments were developed to
diurnal variation (mood is often somewhat improved later in the day). treat ‘schizophrenia’, although formal, reliable diagnostic
Monoaminergic hypothesis: a classic neurobiological hypothesis that pro-
poses depression to arise from dysfunction of the monoaminergic neuro-
criteria were not used. Also, although these treatments
transmitters (i.e. serotonin, norepinephrine and dopamine). were often associated with convulsions, this was not nec-
Subcallosal cingulate (SCC): a ventral portion of the anterior cingulate gyrus essarily the presumed mechanism of action at the time
that is one target for DBS as a treatment for TRD.
Serotonin norepinephrine reuptake inhibitor (SNRI): a class of antidepressant
(except for camphor-induced seizures). Developed within
medications with a primary mechanism of action of blocking the reuptake of this context, ECT was introduced in 1938. Over time, ECT
serotonin and norepinephrine via presynaptic transporters, thereby increasing has become established as the most effective acute treat-
the availability of both neurotransmitters.
Selective serotonin reuptake inhibitor (SSRI): a class of antidepressant ment for a depressive episode, with remission rates of
medications with a primary mechanism of action of blocking the reuptake of >50% [18–20]. However, cognitive side effects can be sig-
serotonin via the presynaptic transporter, thereby increasing the availability of nificant and occasionally persistent [21] and relapse
serotonin.
Tricyclic antidepressants (TCA): a class of antidepressant medications with a remains a significant problem [18].
primary mechanism of action of blocking the reuptake of norepinephrine (and, For many years, the treatments discussed above
to a lesser degree, serotonin) via presynaptic transporters, thereby increasing
remained the only ones available for patients with severe
the availability of these neurotransmitters.
Transcranial magnetic stimulation (TMS): a non-invasive technique that uses a psychiatric illness. However, during the mid-20th century,
rapidly changing magnetic field to depolarize cortical neurons; TMS can either an increase in psychopharmacological research occurred.
be delivered as a single pulse or in a repetitive train (rTMS).
Treatment-resistant depression (TRD): a depressive episode (in the context of
Contemporary with the discovery of the antimanic proper-
either MDD or bipolar disorder) that has not responded to adequate ties of lithium [22,23] (now a standard first-line treatment
antidepressant treatment. for bipolar disorder) and the antipsychotic and antimanic
Vagus nerve stimulation (VNS): a technique that involves implanting an
electrode on a vagus nerve and stimulating via a subcutaneous computer/
properties of chlorpromazine [24,25] (a treatment for schizo-
battery back. phrenia), the first antidepressant medications were identi-
fied during the mid-1950 s. These included the tricyclic

2
Opinion Trends in Neurosciences January 2011, Vol. 34, No. 1

Box 1. DSM-IV-TR diagnostic criteria for a major depressive episode

To be diagnosed as suffering from a major depressive episode, a patient must have five or more of the following symptoms for a duration of two
or more weeks. Furthermore, one of these five symptoms must consist of depressed mood or anhedonia [1].

Mood/Thought Somatic/Vegetative

Depressed mood Appetite/weight changes

Anhedonia Sleep disturbances

Worthlessness or guilt Changes in psychomotor activity

Poor concentration, indecisiveness Fatigue

Thoughts of death, suicidal ideation

compound imipramine [26] and the monoamine oxidase mechanisms in mood disorders, including depression [45–
inhibitor (MAOI) iproniazid [27,28], leading to the ‘mono- 47]. It is encouraging to see this expansion of psychophar-
amine hypothesis’ of depression [29–31]. Research in de- macological research beyond a limited focus on the mono-
pression then largely shifted to delineating the aminergic systems. Unfortunately, this work has yet to yield
neurochemical bases of the disorder and helped establish treatments that substantially change the overall response
a role for serotonin, norepinephrine and dopamine dysfunc- rates and course of illness for patients with depression [48].
tion in the pathophysiology and treatment of depression. In conjunction with the development of somatic treat-
Medications modulating monoaminergic neurotransmis- ments, formal psychotherapeutic approaches to depression
sion have consistently shown response rates (defined as a have evolved (Figure 1). Psychoanalysis was developed
50% decrease in depression severity from baseline) of during the late 19th century [49], and the 20th century
60%. saw a dramatic increase in types of psychotherapy, many
Other neuromodulatory systems have also been explored influenced by advances in behavioral psychology. The
as potential drug targets. Abnormalities of the hypothalam- most-studied psychotherapies for depression include cog-
ic–pituitary–adrenal (HPA) axis have been identified in nitive behavioral therapy (CBT) [50] and interpersonal
depressed patients, which is consistent with findings that psychotherapy [51], with reported efficacy similar to that
trauma and psychosocial stress predisposes an individual to of antidepressant medications used as monotherapy (e.g.
depression and suggests that the stress hormone axis is a response rates of 60%).
target for antidepressant medications [32,33]. Abnormali- More recently, treatments have been developed to tar-
ties within the hypothalamic–pituitary–thyroid axis have get medication-resistant depression specifically. VNS was
also been linked with depression [34] and thyroid hormone approved by the US Food and Drug Administration (FDA)
augmentation is an accepted strategy for TRD, although in 2005 for the treatment of depression not responding to at
with limited efficacy [35]. Dysfunction of glutamate and g- least four antidepressant treatments. VNS does not have
aminobutyric acid (GABA) systems have been found in acute antidepressant efficacy, but has a 30% response rate
depressed patients and certain animal models of depression, (15% remission rate) with 12 months of open-label treat-
and these systems have been suggested as targets for fur- ment [52,53]. Although relapse rates over one to two years
ther treatment development [36]. The NMDA glutamate can be as high as 56% [54,55], patients treated with VNS
receptor antagonist ketamine, for instance, has acute anti- appear to relapse less often than do patients receiving only
depressant effects in severely depressed patients, although ‘treatment-as-usual’ [54,56,57]. Repetitive TMS was FDA
relapse occurs relatively quickly [37–39]; and riluzole (an approved in 2008 for depression not responding to one
agent that appears to inhibit glutamate release, increase antidepressant medication. TMS has consistently demon-
glutamate reuptake and block glutamate receptors) has strated statistically significant antidepressant effects [58–
antidepressant efficacy [40]. More recently, the antimus- 60]. However, response rates are generally <30%, efficacy
carinic agent, scopolamine, has been shown to have acute appears limited in patients failing more than one adequate
antidepressant effects [41]. The opioid system has also been antidepressant treatment [61] and long-term relapse rates
proposed as a target of antidepressant treatment [42]. In- following the acute course are unknown but likely to be
flammatory processes have been implicated in depression, similar to those seen with ECT [62].
and anti-inflammatory therapies might have antidepres- More invasive antidepressant treatments have also
sant efficacy [43]. Nociceptive neurokinins might also have a been revisited in the past few years. Subcallosal cingulate
role in the neurobiology of depression [44]. Finally, an (SCC) DBS was associated with a 60% response rate in
increasing number of studies are focusing on the dysregula- TRD patients following six months of chronic, high-fre-
tion of second messenger systems, gene transcription, vari- quency stimulation in an uncontrolled, open-label study
ous neurotrophic factors, cell turnover and epigenetic [63]; these effects were largely maintained over an addi-

3
()TD$FIG][ Opinion Trends in Neurosciences January 2011, Vol. 34, No. 1

1880: Joseph Breuer treats Anna O. by talking

1891: Gottlieb Burckhardt, bilateral cortical with her
resections in 6 psychotic, aggressive patients
1890s: Sigmund Freud introduces
psychoanalysis

1910: Ludwig Puusepp, severing of frontal- 1900s: Phenobarbital (and other sedative-
parietal fibers in 3 “bipolar” patients hypnotics) introduced

1927: Manfred Sakel, hypoglycemic coma 1920’s-1940’s: Early behavioral therapies

1935: Egas Moniz, prefrontal leucotomy

(1949 Nobel Prize in Medicine)

1938: Ugo Cerletti, ECT

1945-1955: >50,000 prefrontal lobotomies in

U.S. 1949: John Cade, lithium as an antimanic
treatment

1953-1960: Chlorpromazine, imipramineand

iproniazid introduced

1950’s-1990’s: Formalization and

dissemination of cognitive-behavioral and
similar therapies

2005: First report of DBS for TRD

2005: VNS approved for TRD

2008: TMS approved for TRD

TRENDS in Neurosciences

Figure 1. Milestones in the development of psychiatric treatments. Treatment categories: psychotherapy (purple), medication (green), other somatic interventions (pink).

tional six months [63]. DBS of other targets might also resistant patients; however, these data are also prelimi-
have antidepressant efficacy. These include the ventral nary.
anterior internal capsule/ventral striatum (40% response
rate following six months of chronic stimulation, with a One syndrome, many faces: clinical heterogeneity in
53% response rate at last follow-up) [64], nucleus accum- depression
bens (50% response rate following six months of chronic A common explanation for the limited efficacy of antide-
stimulation) [65], inferior thalamic peduncle [66] and habe- pressant treatments is that ‘depression’ probably refers to
nula [67]. Aside from DBS, seven months of treatment with multiple diseases, each with a distinct neurobiology. In
epidural stimulation of bilateral medial and dorsolateral support of this, the clinical heterogeneity of depression is
prefrontal cortices was associated with remission in three frequently cited: age of onset and family history can vary
out of five TRD patients [68]. It should be emphasized that, widely, and presenting symptoms are pleomorphic. For
despite encouraging early results with these various sur- example, although it is most common for patients to present
gical interventions, these data are preliminary and based with decreased sleep and appetite, the reverse can occur and
on small, open-label, uncontrolled studies. is often associated with ‘atypical depression’, in which
In sum, currently available treatments for depression normal mood reactivity and hedonic responses are generally
are modestly effective, but treatment resistance and recur- maintained. Some patients can have a more classic ‘melan-
rence remain significant problems. Importantly, current cholic’ presentation associated with severe anhedonia, ab-
medications are no more effective than those introduced sent mood reactivity, profound psychomotor retardation or
over 50 years ago. Recently approved somatic treatments agitation, and pronounced early morning awakening
(VNS and rTMS) have shown limited short- or long-term (among other symptoms). Co-morbid anxiety and somatic
efficacy. The most effective treatment for a major depres- complaints are common in patients with depression, and
sive episode (MDE) is still ECT, although relapse remains often resolve with successful treatment, although these
a significant problem. Recent data suggest focal neuromo- symptoms are not part of the formal diagnostic criteria;
dulation of various brain regions (via DBS or direct cortical conversely, many patients with a primary anxiety disorder
stimulation) has some benefit in extremely treatment- diagnosis (e.g. generalized anxiety disorder, OCD and/or

4
Opinion Trends in Neurosciences January 2011, Vol. 34, No. 1

post-traumatic stress disorder) frequently present with ously tested animal models. Yet, no animal model mirrors
clinically significant depressive symptoms that often meet the full syndrome of a DSM-defined depressive episode or
criteria for MDD. Patients with bipolar disorders can pres- the phenomenological heterogeneity of MDD. Crucially,
ent with depressive episodes that are phenomenologically current animal models do not have face validity for impor-
indistinguishable from those presenting with MDD, yet the tant aspects of MDD, including a relapsing course, tachy-
primary diagnoses are viewed as distinct. phylaxis and the potential transformation (within a
It is commonly presumed that the clinical heterogeneity patient) from a treatment-responsive to a treatment-resis-
of depression reflects biological heterogeneity, such that tant illness.
specific interventions might be needed for specific (clinical)
subtypes of patients with depression. This view has led to Rethinking depression and its treatment
attempts to define clinically significant depressive sub- Working within an accepted model and definition of de-
types based on phenomenological differences. Early appar- pression, the development of DBS of the SCC for TRD was
ent success was seen with the recognition that patients based on the hypothesis that Brodmann Area 25 (BA25)
with atypical depression might preferentially respond to served as a key node in a distributed network of brain
MAOIs versus tricyclic antidepressants (TCAs) [69]; how- regions involved in mood regulation [80,81]. Activity in
ever, many patients with atypical depression show a good BA25 was noted to increase with sad mood, and a decrease
response to a TCA, thus limiting the utility of this subtype in BA25 activity was a consistent finding associated with
for clinical and research purposes [70]. Similarly, it is successful antidepressant response to several somatic
generally accepted that patients with bipolar depression treatments [80]. According to this hypothesis, the many
should be initially treated with a ‘mood stabilizer’, such as symptoms of depression were due to abnormalities in
lithium or an anticonvulsant [71,72]. However, compared several discrete, but interconnected neural systems, and
with typical mood stabilizers, monotherapy with a typical BA25 was a critical region that had direct connections to
antidepressant medication might be as or more effective in most of these circuits.
achieving and maintaining euthymia in bipolar disorder, Proof-of-principle testing helped confirm the hypothesis
at least in patients with bipolar II disorder (consisting of that directly modulating BA25 activity and white matter
depressive and milder hypomanic episodes, but not more tracts connecting this brain region to other nodes in a
severe manic episodes) [73,74]. Furthermore, it is not clear presumed depression network would have antidepressant
that the neurobiology of unipolar versus bipolar depression effects in TRD patients [63,79]. However, only 60% of
is significantly different [75]. Therefore, the definition of patients responded. With DBS for patients with severe,
depressive subtypes based on clinical criteria has not led to treatment-resistant movement disorders [e.g. Parkinson’s
improved overall response rates; neither has this helped disease (PD)], most patients show clinically significant
identify biologically distinct subtypes to guide further improvement (of motor symptoms) with chronic stimula-
treatment development. tion [82–84].
Variability also exists in antidepressant treatment re- The crucial question then becomes: if BA25 is a critical
sponse across patients with depression, and even within node in a depression network, whose activity and connec-
clinically defined subtypes. Some patients fail to respond to tivity with other brain regions is specifically dysfunctional
one medication, but will show an excellent response to a in TRD, why were response and remission rates not
second (sometimes similar) medication [76]. Others might higher? One simple answer is, as above: depression is a
not respond well to medications, but achieve clinical re- complex disorder with multiple neurobiological etiologies;
mission with psychotherapy [77], and vice versa; many thus, only patients with ‘BA25-dependent’ TRD might
patients might need both to achieve remission [78]. Final- respond to this treatment. We reject this possibility for
ly, it is acknowledged, although less well studied, that several reasons: (i) Several decades of neurobiological
some patients can do well with a particular medication research have failed to identify clinically meaningful bio-
early in their illness, but either relapse during mainte- logical subtypes of ‘depression’ as currently defined; (ii)
nance treatment or show limited response to the same most TRD patients have recurrent illness [i.e. they previ-
medication in a later episode. To this point, most patients ously remitted with adequate (standard) treatment]. Thus,
enrolled in studies of TRD have recurrent depression, at an earlier stage of illness, TRD patients show a response
implying that they achieved remission during one or more to the same biological interventions as do non-TRD
previous depressive episodes [53,64,79]. patients. TRD then cannot be simply defined as a ‘non-
However, few studies (clinical or basic) have empha- monoaminergic’ depression. Furthermore, this highlights
sized clinical subtypes. In human studies, patients that the recurrence and potential malignant transformation of
meet DSM criteria for MDD are generally included for the disorder over time, and the limits of current treatments
treatment and neurobiological studies regardless of to address this; and (iii) most TRD patients enrolled in the
phenomenological heterogeneity; post-hoc analyses of sub- SCC DBS study showed at least some (>30%) improve-
types often include too few patients to draw meaningful ment, and these effects persisted over time [63], suggesting
conclusions. In animal studies, models are often judged by that modulation of a BA25-associated network leads to
how closely they approximate the full syndrome of depres- sustained positive effects on mood regulation in most
sion (while necessarily focusing on symptoms that can be patients. Thus, we conclude: that BA25 and connected
reproduced in an animal). Additionally, new animal mod- brain regions are involved in the regulation of negative
els might only become accepted to the extent that they mood, and that modulation of this network via focal
reproduce behavioral effects with treatment seen in previ- stimulation is likely to impact mood regulation in a poten-

5
Opinion [()TD$FIG] Trends in Neurosciences January 2011, Vol. 34, No. 1

tially beneficial way in most patients. However, although

this beneficial mood effect is necessary for the full syn-
“Normal”
drome of depressive symptoms to resolve, it might not in state
itself be sufficient to achieve complete remission in all
patients.
With this supposition in mind, we propose another an-

Healthy
Stress

MDD
swer: the syndrome of depression, as currently defined, is too (Trauma, Loss, Illness)
broad to serve as a useful construct for treatment develop-
ment, especially for more focal treatments. By analogy, PD
is recognized as a complex neurological syndrome (with
motor, sleep, cognitive, mood and autonomic abnormalities), “Down”
and it is recognized that certain treatments might improve a state
subset of symptoms but not affect, or even worsen, others.
For example, subthalamic DBS can improve the core motor
symptoms of PD [82], but might not significantly affect “Depression”
postural instability, mood or cognition [85,86]; in addition,
negative effects on mood and cognition have been reported
[87,88]. A similar approach to depression would identify TRENDS in Neurosciences

specific symptoms or subsets of symptoms (e.g. depressed Figure 2. Proposed model of depression: ‘stuck in a rut’. In response to a stressor,
mood and/or anhedonia versus sleep versus anxiety) as there is a mood reaction (i.e. a shift from the normal, euthymic state to a ‘down’
targets for treatment; the efficacy of a particular treatment state) that includes the symptoms associated with the syndrome of depression,
grief, sickness behavior, and so on. The return to the normal state is achieved
could then be gauged by its effect on these symptoms, rather through mood regulation. In healthy, nondepressed individuals, the tendency to
than using improvement in the entire syndrome as the enter the down state is relatively low and, when it occurs, the return to normal is
relatively quick and complete (green arrows). In individuals with MDD, the
primary efficacy measure. A version of this approach has
pressure to enter the down state is relatively high (and might even occur in the
recently been suggested (using a dimensional construct for absence of a clear stressor), but the ability to return to normal is impaired (blue
the depressive syndrome) as a way to develop novel antide- arrows). During a depressive episode, the down state becomes more established,
and the individual might have greater difficulty returning to the normal state
pressant treatment strategies [89]. Although such without external intervention (e.g. psychotherapy, medication, etc.). This model
approaches should not diminish the goal of achieving full emphasizes that the down state itself is not abnormal. Instead, it is the tendency to
symptomatic and functional recovery, they might be more enter and get stuck in this state that defines depression. Thus, the neurobiology of
depression should be that of mood reaction and regulation rather than the mood
useful in developing specific treatment targets for clinical state per se.
and basic research studies.
However, it is our opinion that a redefinition of depres-
sion should go further. We recognize that the observed White syndrome, sick sinus syndrome, etc.), the actual
symptoms of ‘depression’ track together in presentation arrhythmia (e.g. tachycardia) is the symptomatic presen-
and improvement with treatment, including domains of tation that might need acute treatment, but treatment of
mood (e.g. sadness, irritability or anxiety), cognition (e.g. the underlying disease requires addressing the tendency of
guilt, low self-esteem, poor concentration and/or atten- the system to enter an aberrant rhythm that it cannot exit.
tion, decreased and/or increased psychomotor speed, Similarly, we conceptualize the depressive state as an
hopelessness and suicidal ideation), and somatic state aberrant neural rhythm, but the depressive disorder as
(e.g. sleep, appetite and libido disturbances, somatic anxi- the tendency of the brain to go into, and stay in, that
ety, motoric retardation and agitation). Nevertheless, we rhythm inappropriately.
conclude that this depressive state, as currently defined, is Such a view has important implications for treatment
not abnormal per se, but reflects an etiologically non- development. With cardiac arrhythmias, several diverse
specific response to stress/ distress (e.g. physical or emo- interventions with different mechanisms of action (e.g.
tional stress, infection, inflammation, etc.) that includes a most medications, cardioversion and time [i.e. spontaneous
stereotypical set of emotional, cognitive–behavioral and recovery]) can return the system to the physiological state,
somatic responses. This is supported by the currently but others (e.g. ablation, pacing and certain medications)
accepted clinical criteria, where the diagnosis of a MDE might be needed to help maintain the ‘normal’ state over
can only be made if the presentation is not clearly due to a time. Likewise, diverse ‘antidepressant’ treatments (e.g.
state caused by withdrawal from an addictive substance, various medications, psychotherapy, ECT or TMS) can
effects of a medical illness, effects of a medication or shift a patient out of the depressive state but might not
normal bereavement, highlighting the non-specific nature necessarily prevent re-entry into that state. This is espe-
of the phenomenology. cially true of treatments with purported acute ‘antidepres-
We instead propose that the primary abnormality of sant’ effects (e.g. sleep deprivation [90], ketamine [38] and
depression is not the depressive state itself, but rather the scopolamine [41]), where a positive mood and/or behavioral
inability to regulate that state appropriately (Figure 2): change is seen within minutes to hours, but effects are
depressed patients become ‘depressed’ without a trigger, short lived (with return of depressive symptoms within
become more severely ‘depressed’ following a trigger (e.g. hours or days). Other treatments, with mechanisms of
death of a loved one) than is considered normal, and/or action distinct from those of more acute treatments, might
remain ‘depressed’ for longer than would be expected. By be needed to prevent the abnormal recurrence of the
analogy, with cardiac arrhythmias (e.g. Wolff-Parkinson- depressive state. For example, certain psychotherapies

6
Opinion Trends in Neurosciences January 2011, Vol. 34, No. 1

Box 2. Future research directions Summary

Accurately and precisely defining depression is a non-
 Further investigate the neurobiological bases for normal and
abnormal mood regulation (rather than the depressed state per se) trivial problem of critical importance. Current models
 Further investigate the neurobiology of treatment resistance and have outlasted their utility as a basis for treatment
depressive relapse development, a conclusion primarily based on the obser-
 Explore strategies to redefine the key endpoints for clinical trials, vation that antidepressant treatments today are no more
recognizing that specific treatments might only target a subset of
effective than those available 50–70 years ago, despite
depressive symptoms
 Develop animal models with face validity for key elements of intensive neurobiological investigation. Although a full
MDD, such as relapse, tachyphylaxis and treatment resistance discussion is beyond the scope of this article, similar
issues have been described in the study and treatment of
other psychiatric disorders, including schizophrenia and
represent positive steps in this direction [9,10,91–93]. It OCD [102–105], highlighting the challenges of a fully
has been suggested that such treatments work by strength- phenomenologically based nosology.
ening the cognitive cortical–subcortical systems responsi- The limitations of the current definition of depression
ble for regulating mood [94–96]). for clinical and basic science research purposes are es-
In light of our proposed redefinition of depression, we sentially twofold: (i) focusing on the full syndrome of
suggest several directions to take with future research on depression as the target of treatment; and (ii) emphasiz-
MDD (Box 2). One important focus of such research should ing the acute resolution of symptoms rather than the
be on identifying interventions that not only alter the maintenance of normal mood regulation over time. As
underlying state, but also prevent the brain from entering an alternative, we propose the following main arguments.
or remaining in that state at inappropriate times. Al- First, treatment development should be guided by the
though preliminary, emerging data for DBS suggest that recognition that different depressive symptoms might
such a treatment intervention is possible. With SCC DBS, need to be targeted with different treatments. Second,
there are immediate changes in mood with initial acute the goal of antidepressant treatment should ultimately be
stimulation that might or might not persist, with resolu- the maintenance of normal mood regulation over the long
tion of the full syndrome of the depressive episode requir- term. Preliminary data from DBS trials suggest that the
ing weeks to months [63,79]. However, crucially, once latter point is at least a possibility with such a treatment
patients are well, they appear to stay well over time: regime. However, this is an invasive treatment that, if
although the published follow-up data are currently limit- validated, will only be appropriate for a select few
ed to one year of treatment, it is notable that >70% of patients with the most severe forms of TRD. Still, by
patients responding to SCC DBS at six months were still building on the knowledge gained from such treatments,
responders at 12 months [63]. Although not described in it is hoped that the neuroscience community will refocus
detail, similar results can be seen with ventral capsule/ efforts on developing less-invasive interventions for MDD
ventral striatal DBS for TRD [64]. Furthermore, if stimu- that achieve the same degree of short- and long-term
lation ceases (e.g. the device is turned off or the battery effectiveness.
depletes), patients usually return to the depressive state
[63,64,97]. Disclaimer statement
It is hoped that less invasive strategies to help prevent P.E.H. has received grant funding from the Dana Founda-
patients with depression from becoming ‘stuck in a rut’ tion, Greenwall Foundation, NARSAD, NIMH (K23
can be developed, which do not require brain surgery. MH077869), National Institutes of Health Loan Repay-
However, it is our opinion that such strategies are likely ment Program, Northstar, Inc., Stanley Medical Research
to differ from the treatments currently under develop- Institute, and Woodruff Foundation; he has received con-
ment that focus primarily on shifting patients out of the sulting fees from AvaCat Consulting, St Jude Medical
depressive state. An important focus for future clinical Neuromodulation and Oppenheimer & Co. H.S.M. has a
and basic research in light of our proposed redefinition of consulting agreement with St Jude Medical Inc., which has
depression would be the regulation of mood states, rather licensed her intellectual property to develop SCC DBS for
than the states themselves. Several imaging studies have the treatment of severe depression. H.S.M. has received
implicated abnormal emotional processing (not just ab- grant funding from the Dana Foundation, NARSAD, Na-
normal mood state) in the pathophysiology of depression tional Institute of Mental Health (NIMH), Stanley Medical
[94,96,98]. Furthermore, as discussed above, recent data Research Institute, and Woodruff Foundation.
suggest that cognitive–behaviorally based and chronobio-
logical therapies (which can be effective in TRD patients References
1 APA (2000) Diagnostic and Statistical Manual of Mental Disorders
[77,99] and might decrease recurrence risk [9,10,92,93]) (DSM-IV-TR; 4th edn), American Psychiatric Association
function via specific neurobiological routes related to 2 Kessler, R.C. et al. (2005) Lifetime prevalence and age-of-onset
mood regulation [96,100,101]. With respect to the devel- distributions of DSM-IV disorders in the National Comorbidity
opment of future animal models, an emphasis on symp- Survey Replication. Arch. Gen. Psychiatry 62, 593–602
3 Sullivan, P.F. et al. (2000) Genetic epidemiology of major depression:
tom breakthrough and/or recurrence is needed, given
review and meta-analysis. Am. J. Psychiatry 157, 1552–1562
that there are currently no accepted animal models of 4 WHO (2008) The Global Burden of Disease: 2004 Update, WHO.
TRD or highly recurrent depression. Such models are 5 Trivedi, M.H. et al. (2006) Evaluation of outcomes with citalopram for
needed to better identify and develop novel treatment depression using measurement-based care in STAR*D: implications
targets. for clinical practice. Am. J. Psychiatry 163, 28–40

7
Opinion Trends in Neurosciences January 2011, Vol. 34, No. 1

6 Rush, A.J. et al. (2006) Acute and longer-term outcomes in depressed 34 Musselman, D.L. and Nemeroff, C.B. (1996) Depression and endocrine
outpatients requiring one or several treatment steps: a STAR*D disorders: focus on the thyroid and adrenal system.. Br. J. Psychiatry
report. Am. J. Psychiatry 163, 1905–1917 168 (Suppl. 30), 123–128
7 Gelenberg, A.J. et al. (2003) Randomized, placebo-controlled trial of 35 Rush, A.J. et al. (2004) Sequenced treatment alternatives to relieve
nefazodone maintenance treatment in preventing recurrence in depression (STAR*D): rationale and design. Control Clin. Trials 25,
chronic depression. Biol. Psychiatry 54, 806–817 119–142
8 Keller, M.B. et al. (2007) The Prevention of Recurrent Episodes of 36 Krystal, J.H. et al. (2002) Glutamate and GABA systems as targets for
Depression with Venlafaxine for Two Years (PREVENT) Study: novel antidepressant and mood-stabilizing treatments. Mol.
outcomes from the 2-year and combined maintenance phases. J. Psychiatry 7 (Suppl. 1), S71–S80
Clin. Psychiatry 68, 1246–1256 37 Berman, R.M. et al. (2000) Antidepressant effects of ketamine in
9 Hollon, S.D. et al. (2005) Prevention of relapse following cognitive depressed patients. Biol. Psychiatry 47, 351–354
therapy vs medications in moderate to severe depression. Arch. Gen. 38 Zarate, C.A., Jr et al. (2006) A randomized trial of an N-methyl-D-
Psychiatry 62, 417–422 aspartate antagonist in treatment-resistant major depression. Arch.
10 Fava, G.A. et al. (2004) Six-year outcome of cognitive behavior therapy Gen. Psychiatry 63, 856–864
for prevention of recurrent depression. Am. J. Psychiatry 161, 1872– 39 aan het Rot, M. et al. (2010) Safety and efficacy of repeated-dose
1876 intravenous ketamine for treatment-resistant depression. Biol.
11 Burckhardt, G. (1891) On cortical resection as a contribution to the Psychiatry 67, 139–145
operative treatment of psychosis. Psychiatrie psychischgerichtliche 40 Sanacora, G. et al. (2007) Preliminary evidence of riluzole efficacy in
Medizin 47, 463–548 antidepressant-treated patients with residual depressive symptoms.
12 Jasper, H.H. (1995) A historical perspective. The rise and fall of Biol. Psychiatry 61, 822–825
prefrontal lobotomy. Adv. Neurol. 66, 97–114 41 Drevets, W.C. and Furey, M.L. (2010) Replication of scopolamine’s
13 Papez, J.W. (1937) A proposed mechanism of emotion. Arch. Neurol. antidepressant efficacy in major depressive disorder: a randomized,
Psychiatry 38, 725–743 placebo-controlled clinical trial. Biol. Psychiatry 67, 432–438
14 Sachdev, P. and Sachdev, J. (1997) Sixty years of psychosurgery: 42 Hegadoren, K.M. et al. (2009) The role of beta-endorphin in the
its present status and its future. Aust. N.Z. J. Psychiatry 31, 457– pathophysiology of major depression. Neuropeptides 43, 341–353
464 43 Raison, C.L. et al. (2006) Cytokines sing the blues: inflammation and
15 Epstein, N.N. (1936) Artificial fever as a therapeutic procedure. Cal. the pathogenesis of depression. Trends Immunol. 27, 24–31
West. Med. 44, 357–358 44 Bondy, B. et al. (2003) Substance P serum levels are increased in
16 Meduna, L.J. (1935) Die Konvulsionstherapie der Schizophrenie. major depression: preliminary results. Biol. Psychiatry 53, 538–542
Psychr. Neurol. Wochenschr. 37, 317–319 45 Krishnan, V. and Nestler, E.J. (2008) The molecular neurobiology of
17 Sakel, M. (1937) The origin and nature of the hypoglycemic therapy of depression. Nature 455, 894–902
the psychoses. Bull. N.Y. Acad. Med. 13, 97–109 46 Hunsberger, J.G. et al. (2009) Cellular mechanisms underlying
18 Kellner, C.H. et al. (2006) Continuation electroconvulsive therapy vs affective resiliency: the role of glucocorticoid receptor- and
pharmacotherapy for relapse prevention in major depression: a multisite mitochondrially-mediated plasticity. Brain Res. 1293, 76–84
study from the Consortium for Research in Electroconvulsive Therapy 47 Li, N. et al. (2010) mTOR-dependent synapse formation underlies the
(CORE). Arch. Gen. Psychiatry 63, 1337–1344 rapid antidepressant effects of NMDA antagonists. Science 329, 959–
19 Sackeim, H.A. et al. (2009) Effect of concomitant pharmacotherapy on 964
electroconvulsive therapy outcomes: short-term efficacy and adverse 48 Holtzheimer, P.E., III and Nemeroff, C.B. (2006) Emerging treatments
effects. Arch. Gen. Psychiatry 66, 729–737 for depression. Expert Opin. Pharmacother. 7, 2323–2339
20 The UK ECT Review Group (2003) Efficacy and safety of 49 Freud, S. and Breuer, J. (1895) Studien über Hysterie, Deuticke.
electroconvulsive therapy in depressive disorders: a systematic 50 Beck, A.T. (2005) The current state of cognitive therapy: a 40-year
review and meta-analysis. Lancet 361, 799–808 retrospective. Arch. Gen. Psychiatry 62, 953–959
21 Sackeim, H.A. et al. (2007) The cognitive effects of electroconvulsive 51 de Mello, M.F. et al. (2005) A systematic review of research findings on
therapy in community settings. Neuropsychopharmacology 32, 244– the efficacy of interpersonal therapy for depressive disorders. Eur.
254 Arch. Psychiatry Clin. Neurosci. 255, 75–82
22 Cade, J.F.J. (1949) Lithium salts in the treatment of psychotic 52 Rush, A.J. et al. (2005) Effects of 12 months of vagus nerve stimulation
excitement. Med. J. Aust. 2, 349–352 in treatment-resistant depression, a naturalistic study. Biol. Psychiatry
23 Schou, M. et al. (1954) The treatment of manic psychoses by the 58, 355–363
administration of lithium salts. J. Neurol. Neurosurg. Psychiatry 17, 53 Rush, A.J. et al. (2005) Vagus nerve stimulation for treatment-
250–260 resistant depression: a randomized, controlled acute phase trial.
24 Bower, W.H. (1954) Chlorpromazine in psychiatric illness. N. Engl. J. Biol. Psychiatry 58, 347–354
Med. 251, 689–692 54 Nahas, Z. et al. (2005) Two-Year Outcome of Vagus Nerve Stimulation
25 Winkelman, N.W., Jr (1954) Chlorpromazine in the treatment of (VNS) for treatment of major depressive episodes. J. Clin. Psychiatry
neuropsychiatric disorders. J. Am. Med. Assoc. 155, 18–21 66, 1097–1104
26 Kuhn, R. (1958) The treatment of depressive states with G 22355 55 Schlaepfer, T.E. et al. (2008) Vagus nerve stimulation for
(imipramine hydrochloride). Am. J. Psychiatry 115, 459–464 depression: efficacy and safety in a European study. Psychol.
27 Kiloh, L.G. et al. (1960) A controlled trial of iproniazid in the Med. 38, 651–661
treatment of endogenous depression. J. Mental Sci. 106, 1139–1144 56 Dunner, D.L. et al. (2006) Prospective, long-term, multicenter study of
28 Bailey, S.D. et al. (1959) Comparison of iproniazid with other amine the naturalistic outcomes of patients with treatment-resistant
oxidase inhibitors, including W-1544, JB-516, RO 4-1018, and RO 5- depression. J. Clin. Psychiatry 67, 688–695
0700. Ann. N.Y. Acad. Sci. 80, 652–668 57 Sackeim, H.A. et al. (2007) Durability of antidepressant response to
29 Schildkraut, J.J. (1965) The catecholamine hypothesis of affective vagus nerve stimulation (VNSTM). Int. J. Neuropsychopharmacol. 10,
disorders: a review of supporting evidence. Am. J. Psychiatry 122, 817–826
509–522 58 Holtzheimer, P.E., III et al. (2001) A meta-analysis of repetitive
30 Bunney, W.E., Jr and Davis, J.M. (1965) Norepinephrine in transcranial magnetic stimulation in the treatment of depression.
depressive reactions. A review. Arch. Gen. Psychiatry 13, 483–494 Psychopharmacol. Bull. 35, 149–169
31 Prange, A.J., Jr (1964) The pharmacology and biochemistry of 59 O’Reardon, J.P. et al. (2007) Efficacy and safety of transcranial
depression. Dis. Nerv. Syst. 25, 217–221 magnetic stimulation in the acute treatment of major depression: a
32 Arborelius, L. et al. (1999) The role of corticotropin-releasing factor in multisite randomized controlled trial. Biol. Psychiatry 62, 1208–
depression and anxiety disorders. J. Endocrinol. 160, 1–12 1216
33 Heim, C. et al. (2000) Pituitary-adrenal and autonomic responses to 60 George, M.S. et al. (2010) Daily left prefrontal transcranial magnetic
stress in women after sexual and physical abuse in childhood. J. Am. stimulation therapy for major depressive disorder: a sham-controlled
Med. Assoc. 284, 592–597 randomized trial. Arch. Gen. Psychiatry 67, 507–516

8
Opinion Trends in Neurosciences January 2011, Vol. 34, No. 1

61 Lisanby, S.H. et al. (2009) Daily left prefrontal repetitive transcranial 83 Lyons, K.E. and Pahwa, R. (2008) Deep brain stimulation and tremor.
magnetic stimulation in the acute treatment of major depression: Neurotherapeutics 5, 331–338
clinical predictors of outcome in a multisite, randomized controlled 84 Ostrem, J.L. and Starr, P.A. (2008) Treatment of dystonia with deep
clinical trial. Neuropsychopharmacology 34, 522–534 brain stimulation. Neurotherapeutics 5, 320–330
62 Dannon, P.N. et al. (2002) Three and six-month outcome following 85 Moro, E. et al. (2010) Unilateral pedunculopontine stimulation
courses of either ECT or rTMS in a population of severely improves falls in Parkinson’s disease. Brain 133, 215–224
depressed individuals – preliminary report. Biol. Psychiatry 51, 86 Witt, K. et al. (2008) Neuropsychological and psychiatric changes after
687–690 deep brain stimulation for Parkinson’s disease: a randomised,
63 Lozano, A.M. et al. (2008) Subcallosal cingulate gyrus deep brain multicentre study. Lancet Neurol. 7, 605–614
stimulation for treatment-resistant depression. Biol. Psychiatry 64, 87 Zahodne, L.B. et al. (2009) Cognitive declines one year after unilateral
461–467 deep brain stimulation surgery in Parkinson’s disease: a controlled
64 Malone, D.A., Jr et al. (2009) Deep brain stimulation of the ventral study using reliable change. Clin. Neuropsychol. 23, 385–405
capsule/ventral striatum for treatment-resistant depression. Biol. 88 Voon, V. et al. (2008) A multicentre study on suicide outcomes
Psychiatry 65, 267–275 following subthalamic stimulation for Parkinson’s disease. Brain
65 Bewernick, B.H. et al. (2010) Nucleus accumbens deep brain 131, 2720–2728
stimulation decreases ratings of depression and anxiety in 89 Katz, M.M. et al. (2010) Rethinking depression and the actions of
treatment-resistant depression. Biol. Psychiatry 67, 110–116 antidepressants: uncovering the links between the neural and
66 Jimenez, F. et al. (2005) A patient with a resistant major depression behavioral elements. J. Affect. Disord. 120, 16–23
disorder treated with deep brain stimulation in the inferior thalamic 90 Wu, J.C. and Bunney, W.E. (1990) The biological basis of an
peduncle. Neurosurgery 57, 585–593 antidepressant response to sleep deprivation and relapse: review
67 Sartorius, A. et al. (2010) Remission of major depression under deep and hypothesis. Am. J. Psychiatry 147, 14–21
brain stimulation of the lateral habenula in a therapy-refractory 91 Frank, E. et al. (1991) Efficacy of interpersonal psychotherapy as a
patient. Biol. Psychiatry 67, e9–e11 maintenance treatment of recurrent depression. Contributing factors.
68 Nahas, Z. et al. (2010) Bilateral epidural prefrontal cortical Arch. Gen. Psychiatry 48, 1053–1059
stimulation for treatment-resistant depression. Biol. Psychiatry 67, 92 Frank, E. et al. (2000) Interpersonal and social rhythm therapy:
101–109 managing the chaos of bipolar disorder. Biol. Psychiatry 48, 593–
69 Quitkin, F.M. et al. (1993) Columbia atypical depression. A subgroup 604
of depressives with better response to MAOI than to tricyclic 93 Frank, E. et al. (2005) Two-year outcomes for interpersonal and social
antidepressants or placebo. Br. J. Psychiatry 163 (Suppl. 21), 30–34 rhythm therapy in individuals with bipolar I disorder. Arch. Gen.
70 Thase, M.E. (2009) Atypical depression: useful concept, but it’s time to Psychiatry 62, 996–1004
revise the DSM-IV criteria. Neuropsychopharmacology 34, 2633–2641 94 Siegle, G.J. et al. (2002) Can’t shake that feeling: event-related
71 APA (2002) Practice guideline for the treatment of patients with fMRI assessment of sustained amygdala activity in response to
bipolar disorder (revision). Am. J. Psychiatry 159, (Suppl.) emotional information in depressed individuals. Biol. Psychiatry
72 Baldessarini, R.J. et al. (2010) Bipolar depression: overview and 51, 693–707
commentary. Harv. Rev. Psychiatry 18, 143–157 95 Siegle, G.J. et al. (2007) Increased amygdala and decreased
73 Agosti, V. and Stewart, J.W. (2007) Efficacy and safety of dorsolateral prefrontal BOLD responses in unipolar depression:
antidepressant monotherapy in the treatment of bipolar-II related and independent features. Biol. Psychiatry 61, 198–209
depression. Int. Clin. Psychopharmacol. 22, 309–311 96 Segal, Z.V. et al. (2006) Cognitive reactivity to sad mood provocation
74 Amsterdam, J.D. and Shults, J. (2010) Efficacy and safety of long- and the prediction of depressive relapse. Arch. Gen. Psychiatry 63,
term fluoxetine versus lithium monotherapy of bipolar II disorder: 749–755
a randomized, double-blind, placebo-substitution study. Am. J. 97 Schlaepfer, T.E. et al. (2008) Deep brain stimulation to reward
Psychiatry 167, 792–800 circuitry alleviates anhedonia in refractory major depression.
75 Savitz, J. and Drevets, W.C. (2009) Bipolar and major depressive Neuropsychopharmacology 33, 368–377
disorder: neuroimaging the developmental-degenerative divide. 98 Elliott, R. et al. (2002) The neural basis of mood-congruent processing
Neurosci. Biobehav. Rev. 33, 699–771 biases in depression. Arch. Gen. Psychiatry 59, 597–604
76 Rush, A.J. et al. (2006) Bupropion-SR, sertraline, or venlafaxine-XR 99 Fava, G.A. et al. (1997) Cognitive-behavioral management of drug-
after failure of SSRIs for depression. N. Engl. J. Med. 354, 1231–1242 resistant major depressive disorder. J. Clin. Psychiatry 58, 278–282
77 Thase, M.E. et al. (2007) Cognitive therapy versus medication in quiz 283–284
augmentation and switch strategies as second-step treatments: a 100 Siegle, G.J. et al. (2006) Use of FMRI to predict recovery from unipolar
STAR*D report. Am. J. Psychiatry 164, 739–752 depression with cognitive behavior therapy. Am. J. Psychiatry 163,
78 Keller, M.B. et al. (2000) A comparison of nefazodone, the cognitive 735–738
behavioral-analysis system of psychotherapy, and their combination for 101 Farb, N.A. et al. (2010) Minding one’s emotions: mindfulness training
the treatment of chronic depression. N. Engl. J. Med. 342, 1462–1470 alters the neural expression of sadness. Emotion 10, 25–33
79 Mayberg, H.S. et al. (2005) Deep brain stimulation for treatment- 102 Carpenter, W.T. et al. (2009) The psychoses: cluster 3 of the proposed
resistant depression. Neuron 45, 651–660 meta-structure for DSM-V and ICD-11. Psychol. Med. 39, 2025–2042
80 Mayberg, H.S. (2009) Targeted electrode-based modulation of neural 103 Conn, P.J. et al. (2008) Schizophrenia: moving beyond monoamine
circuits for depression. J. Clin. Invest. 119, 717–725 antagonists. Mol. Interv. 8, 99–107
81 Price, J.L. and Drevets, W.C. (2010) Neurocircuitry of mood disorders. 104 Starcevic, V. and Brakoulias, V. (2008) Symptom subtypes of
Neuropsychopharmacology 35, 192–216 obsessive-compulsive disorder: are they relevant for treatment?
82 Moro, E. et al. (2010) Long-term results of a multicenter study on Aust. N.Z. J. Psychiatry 42, 651–661
subthalamic and pallidal stimulation in Parkinson’s disease. Mov. 105 Mataix-Cols, D. et al. (2005) A multidimensional model of obsessive-
Disord. 25, 578–586 compulsive disorder. Am. J. Psychiatry 162, 228–238