P

STAa

Bpf

Mw7

R1wln

Ccba

Kn

Dwtp((cccrt

wbaspcas

F

A

R

0d

RIORITY COMMUNICATION

ubcallosal Cingulate Gyrus Deep Brain Stimulation forreatment-Resistant Depression

ndres M. Lozano, Helen S. Mayberg, Peter Giacobbe, Clement Hamani, R. Cameron Craddock,nd Sydney H. Kennedy

ackground: A preliminary report in six patients suggested that deep brain stimulation (DBS) of the subcallosal cingulate gyrus (SCG) mayrovide benefit in treatment-resistant depression (TRD). We now report the results of these and an additional 14 patients with extended

ollow-up.

ethods: Twenty patients with TRD underwent serial assessments before and after SCG DBS. We determined the percentage of patientsho achieved a response (50% or greater reduction in the 17-item Hamilton Rating Scale for Depression [HRSD-17]) or remission (scores of

or less) after surgery. We also examined changes in brain metabolism associated with DBS, using positron emission tomography.

esults: There were both early and progressive benefits to DBS. One month after surgery, 35% of patients met criteria for response with0% of patients in remission. Six months after surgery, 60% of patients were responders and 35% met criteria for remission, benefits thatere largely maintained at 12 months. Deep brain stimulation therapy was associated with specific changes in the metabolic activity

ocalized to cortical and limbic circuits implicated in the pathogenesis of depression. The number of serious adverse effects was small witho patient experiencing permanent deficits.

onclusions: This study suggests that DBS is relatively safe and provides significant improvement in patients with TRD. Subcallosalingulate gyrus DBS likely acts by modulating brain networks whose dysfunction leads to depression. The procedure is well tolerated andenefits are sustained for at least 1 year. A careful double-blind appraisal is required before the procedure can be recommended for use on

wider scale.

ey Words: Brain imaging, deep brain stimulation, depression,euromodulation, subcallosal cingulate gyrus

epression is a common disorder and a leading cause ofyears lived with disability worldwide (1–3). While treat-ment is often effective, a substantial subgroup of patients

ith depression, approximately 10% to 20% of sufferers, con-inue to be severely disabled despite adequate trials of antide-ressant drugs, psychotherapy, and electroconvulsive therapyECT) (4). For these patients with treatment-resistant depressionTRD), the illness not only affects quality of life but is also a majorause of suicide and contributes to increased mortality in theontext of comorbid disorders including diabetes and cardiovas-ular disease (5–7). Recent advances in understanding the neu-obiology and brain circuitry of depression are driving novelreatment approaches for patients with TRD (8).

Functional imaging has shown that depression is associatedith increased activity in the subcallosal cingulate gyrus (SCG), arain area involved in mood regulation (9 –11). That this over-ctivity is important in the pathophysiology of depression isupported by the finding that disparate interventions includingharmacotherapy, transcranial magnetic stimulation, and electro-onvulsive therapy ameliorate the clinical features of depressionnd alter the activity of the SCG (12,13). These observations haveparked interest in the possibility of directly modulating the

rom the Division of Neurosurgery (AML, CH), Department of Surgery, andDepartment of Psychiatry (HSM, PG, SHK), University of Toronto, To-ronto, Canada; and the Department of Psychiatry (HSM, RCC), EmoryUniversity, Atlanta, Georgia.

ddress reprint requests to Andres M. Lozano, M.D., Ph.D., Professor and RRTasker Chair, Division of Neurosurgery, Toronto Western Hospital, Room4-447, 399 Bathurst Street, Toronto, Ontario M5T2S8, Canada; E-mail:lozano@uhnres.utoronto.ca.

eceived April 4, 2008; revised May 26, 2008; accepted May 31, 2008.

006-3223/08/$34.00oi:10.1016/j.biopsych.2008.05.034

output of the SGC and, consequently, its downstream targets inTRD. Deep brain stimulation (DBS) represents an adjustable andreversible method of focally modulating the activity of dysfunc-tional brain circuits with electrical stimulation. Leveraging ourexperience with DBS in treating patients with Parkinson disease(14 –16), we implanted DBS electrodes in the SCG in six TRDpatients and showed improvements in symptoms of depressionwere associated with decreased activity in SCG (17).

We now report the clinical effects, safety, and the changes inthe activity of brain circuitry in the 6 initial patients and anadditional 14 patients with TRD receiving DBS of the SCG for alonger time, 12 months.

Methods and Materials

PatientsTwenty patients with TRD received DBS of the SCG between

May 2003 and November 2006 (Table 1). Referrals came fromhospital and community psychiatrists who were aware of theprotocol and were not directly involved in its implementation. Allpatients met criteria for major depressive disorder (MDD), werein a current major depressive episode (MDE) as determined bythe Structured Clinical Interview for DSM-IV Axis I Disorders-Patient Edition (SCID-I/P) (18), and had a minimum score of 20on the 17-item Hamilton Rating Scale for Depression (HRSD-17)(19). Inclusion and exclusion criteria have been previouslydescribed (17). Essentially, inclusion criteria required a durationof at least 1 year for the current MDE and treatment resistance asdefined as failure to respond to a minimum of four differenttreatments, including antidepressant pharmacotherapy of suffi-cient dose and duration, evidence-based psychotherapy, andelectroconvulsive therapy. Exclusion criteria included comorbidAxis I psychiatric conditions, a cluster B Axis II diagnosis asdetermined by the Structured Clinical Interview for DSM-IV AxisII Personality Disorders (SCID-II) (20), suicidal behavior within

the past year or a score of 3 or more on the HRSD-17 suicide

BIOL PSYCHIATRY 2008;64:461–467© 2008 Published by Society of Biological Psychiatry

ictBS

A

rbTwdscO(Cclcpaescf

S

a(Ewnn

TT

NAALNPRRMAE

E

BBB

(

ts

462 BIOL PSYCHIATRY 2008;64:461–467 A.M. Lozano et al.

w

tem, and concurrent neurological or medical conditions thatould interfere with the treatment. The study was approved byhe Institutional Review Board of the University Health Network,aycrest Centre, and the Centre for Addiction and Mental Health.igned informed consent was obtained from each patient.

ssessmentPsychiatric assessments and stimulator adjustments (if war-

anted) were performed at 1, 2, and 4 weeks after surgery;iweekly for 3 months; and then monthly for up to 12 months.he primary outcome measure was the percentage of patientsho achieved a 50% or greater reduction in the severity ofepression, as measured by HRSD-17 scores (defined as “re-ponse”) with a secondary outcome as those who achievedlinical “remission” (defined as a HRSD-17 score of 7 or less).ther standardized ratings included the Beck Anxiety Inventory

BAI) (21), the Beck Depression Inventory (BDI) (22), and thelinical Global Impression for Severity (CGI-S) scale. Neuropsy-hological testing was administered to establish baseline intel-ectual and cognitive function prior to surgery and to monitor forhanges after 3, 6, and 12 months of continuous stimulation. Areliminary report on the neuropsychological battery utilizednd results obtained in a subgroup of these patients is describedlsewhere (23). Positron emission tomography (PET) scan mea-ures of regional glucose metabolism were used to quantifyhanges in brain activity with chronic stimulation, expanding onindings reported in the initial cohort of these patients (17).

urgical ProcedureA Leksell stereotactic frame (Elekta, Stockholm, Sweden) was

pplied using local anesthesia. Magnetic resonance imagingMRI) was obtained using a 1.5 T scanner (Signa, Generallectric, Milwaukee, Wisconsin). Images of 2 mm thicknessithout overlap were reconstructed in three dimensions usingeuronavigation software (Stealth, Medtronic, Minneapolis, Min-

able 1. Demographic Characteristics and History in 20 Subjects withreatment-Resistant Depression

Patients Male Female

umber of Patients 20 9 11ge at Enrollment (years) 47.4 � 10.4 49.6 � 14.2 45.3 � 5.6ge of Onset of MDD (years) 27.1 � 8.3 24.4 � 9.2 29.2 � 7.3ength of Current Episode (years) 6.9 � 5.6 6.8 � 6.0 7.0 � 5.5umber of Lifetime MDE (n) 3.9 � 3.1 3.6 � 2.6 4.1 � 3.5ositive Family History MDD (n) 14/20 6/9 8/11eceived ECT (n) 17/20 8/9 9/11eceived Psychotherapy 20/20 9/9 11/11elancholic Subtype (n) 13/20 7/9 6/11typical Subtype (n) 7/20 2/9 5/11mployment Status Subjects

unemployed preoperative (n)18/20 7/9 11/11

mployment Status Subjectsunemployed 12 monthspostoperative (n)

12/20 5/9 7/11

aseline HRSD-17 24.4 � 3.5 24.4 � 3.9 24.3 � 3.3aseline BAI 14.1 � 9.2 10.2 � 6.6 16.5 � 10.2aseline BDI 27.5 � 9.2 20.6 � 6.7 31.9 � 7.9

Values � standard deviation. T test of gender difference in baseline BDIt � 2.65, p � .05). All other comparisons p �.05.

BAI, Beck Anxiety Inventory; BDI, Beck Depression Inventory; ECT, elec-roconvulsive therapy; HRSD-17, 17-item Hamilton Rating Scale for Depres-ion; MDD, major depressive disorder; MDE, major depressive episode.

esota). The SCG was identified by direct visualization on

ww.sobp.org/journal

coronal images and target points spanning the entire verticalheight of the gyrus were chosen. Bilateral burr holes were placed1 cm anterior to the coronal suture and 2 cm from the midlineunder local anesthesia. Microelectrode recordings were used toidentify the upper and lower cortical banks and the interveningwhite matter of the SCG. Quadripolar DBS electrodes (Medtronic3387, Medtronic) were implanted with the most distal contact(contact 0 on the right, contact 4 on the left) adjacent to theventral bank of gray matter, the two central contacts (contacts 1and 2 right, contacts 5 and 6 left) in white matter, and theuppermost contacts (contact 3 right, contact 7 left) adjacent to theupper bank of gray matter of the SCG gyrus using fluoroscopicguidance and confirmed on postoperative MRI. A dual channelprogrammable internal pulse generator (Kinetra, Medtronic) wasimplanted subcutaneously in the infraclavicular area under gen-eral anesthesia. Patients were discharged on postoperative days2 to 5.

Stimulation SettingsBlinded stimulation through the implanted DBS electrodes

was applied intraoperatively and in the immediate postoperativeperiod. The acute response to stimulation was noted at eachcontact and each setting with the patient blinded to the stimula-tor settings or changes. The selection of electrode contacts for theinitial settings for chronic stimulation was made on the basis ofthe observation of acute behavioral effects with stimulation.These included calmness, improved mood, and increased inter-est and motivation and were achieved using 3 V to 6 V. As partof dose finding, stimulation intensity was increased to look foradverse effects. In two patients, mental slowing occurred at highsettings of 8 V to 10 V, particularly at the higher contacts (3 and7). In patients who showed little or no acute behavioral changes,we started stimulation at contacts 1 and 5 set at 3.5 V, 90microseconds, and 130 Hz. Adjustments to the neurostimulator atfollow-up visits were made if the patient was failing to showimprovement (less than a 10% reduction in the HRSD-17 scores)or developed adverse effects, for example increased anxiety.Changes consisted of choosing different contact pairs for stimu-lation or adjusting the voltage. Patients received continuousmonopolar stimulation at settings that ranged from 3.5 V to 5.0 Vwith the pulse width set at 90 microseconds and the frequency at130 Hz with the pulse generator case set as the positive terminal.Notably, patients reported inability to discern if the contacts wereon or off during these subsequent sessions and remained blindedas to which contact was being stimulated and the parametersettings.

Statistical AnalysisA reduction from baseline to end point in total HRSD-17 score

of 50% or greater (i.e., responders) was chosen as the primarymeasure of effectiveness. We also determined the proportionachieving a score of 7 or less at end point on the HRSD-17 (i.e.,remitters). We further examined changes in the subcomponentsof the HRSD-17 (mood, anxiety, somatic, and sleep) (24).Pairwise comparisons between the results of the baseline evalu-ation and the results of the follow-up evaluations were madewith the Student t test. To evaluate long-term effectiveness, datawere examined at 6 and 12 months after surgery. Data from earlyterminators were analyzed in a last observation carried forwardmanner.

PET Scan Acquisition and Data AnalysisWe have previously characterized specific abnormalities in

regional cerebral blood flow in patients with TRD (compared

wwos6buismcT

R

P

TmordoOsbr

4aawesctdmT

Fbtbss

A.M. Lozano et al. BIOL PSYCHIATRY 2008;64:461–467 463

ith healthy age-matched control subjects), as well as changesith SCG DBS (17). We sought to confirm and expand on thesebservations in the current study. Positron emission tomographycans were used to establish changes in brain activity after 3 and

months of chronic stimulation relative to the presurgicalaseline. In the initial report, regional blood flow was measuredsing 150-water PET (as previously described). Due to technicalssues including installation of a new PET scanner and lack ofubsequent availability of the blood flow tracer, regional glucoseetabolism was measured instead using 18F-fluorodeoxyglu-

ose (18FDG) in the last 11 patients (3 patients were not imaged).he PET protocol and analysis are described in Supplement 1.

esults

rimary Outcome MeasuresThe clinical and demographic features of 20 patients with

RD undergoing SCG DBS are shown in Table 1. We treated 9en and 11 women with SCG DBS. All had failed multiple trialsf pharmacotherapy and psychotherapy and all but three (whoefused) also received a course of electroconvulsive therapyuring the current episode without response. The mean durationf the current major depressive episode was 6.9 years (SD 5.6).ne patient who initially was diagnosed with unipolar depres-

ion had more accurately, in retrospect, previously undiagnosedipolar II disorder. All were required to maintain a stable dosingegimen in the 4 weeks prior to surgery.

The mean number of medications at the time of surgery was.2 and the median was 4. Eleven patients were receiving twontidepressants from distinct classes augmented with an atypicalntipsychotic or lithium and a benzodiazepine. Five patientsere receiving one antidepressant combined with a benzodiaz-pine or antipsychotic agent, two were receiving an antidepres-ant alone, and two had discontinued all antidepressants butontinued to receive a benzodiazepine. So as to not confoundhe effects of DBS, we attempted not to change medicationsuring the course of the study, and no new antidepressantedications were prescribed in the first 6 months after surgery.he emergence of dose-related, medication-specific adverse

0

5

10

15

20

25

Base 1 2 3 4 5 6 7 8 9 10 11 12

Tota

l HA

M-D

17

Sco

re

All Subjects

All time points

p < 0.001

Change in HRSD-17 Score From Baseline

Months Post-DBS

igure 1. Mean HRSD-17 score for 20 patients with TRD receiving SGC DBS ataseline and at subsequent visits over a 12-month period. The values at all

ime points from 1 to 12 months are significantly different compared withaseline p � .001. Error bars indicate standard deviation. DBS, deep braintimulation; HRSD-17, 17-item Hamilton Rating Scale for Depression; SGC,

ubcallosal cingulate gyrus; TRD, treatment-resistant depression.

effects or clinical improvement led to reductions in the numberor dose of medications. Minor medication changes occurred overtime as described in Supplement 2.

The mean total HRSD-17 score in the 20 patients was signif-icantly improved at all time points examined 1 month or longerafter DBS than at baseline (Figure 1). After 1 week of stimulation,40% of patients were responders and one patient was in remis-sion. Some of this initial benefit may be related to a microlesioneffect as is commonly observed in Parkinson’s disease where themere insertion of a DBS electrode can produce transient clinicalbenefits. Consistent with this notion, the response rate fell to 30%with one patient in remission 2 weeks after surgery.

We observed a progressive improvement with chronic DBS.From 2 weeks to 6 months after surgery, an increasing proportionof patients improved, reaching a plateau at 6 months when 60%of patients met response criteria and 35% achieved remission(Figure 2). These benefits were largely maintained for theduration of the study. At 12 months, 55% of patients wereresponders and 35% achieved or were within 1 point of remis-sion (scoring 8 or less on the HRSD-17 scale). Eight (72.7%) of 11patients who met criteria for response at 6 months also metcriteria for response at 12 months, while 3 (33%) of 9 patientswho were not deemed responders at 6 months obtained re-sponse status at 12 months (Fisher exact test p � .08). There wasno difference in response rates at 12 months post-DBS betweenthose that had previously received ECT and those that had not(9/17 vs. 2/3, Fisher exact test p � .58). There was no significantloss of effect requiring dose adjustments over time, an observa-tion that mimics the relative stability of DBS stimulation param-eters in patients with other disorders including Parkinson’sdisease and dystonia.

Deep brain stimulation was associated with global improve-ments in depressive symptomatology as measured by the mood,anxiety, somatic, and sleep subcomponents of the HRSD scale(Table 2). Improvements in each of the symptom clusters in-creased with time after the initiation of stimulation. The maximalimprovement in the mood component occurred by 3 months.Longer times were required to reach maximal improvements inanxiety, sleep, and somatic symptoms (Table 2). An analysis ofthe rates of symptom cluster improvement in responders versus

0

10

20

30

40

50

60

70

Sx 1 2 3 4 5 6 7 8 9 10 11 12

Recentage of Responders

Percentage of Remitters

Percent of Subjects Meeting Criteria for Response or Remission by Time

Months Post-DBS

Per

cent

of S

ubje

cts

Figure 2. Patients meeting response or remission criteria after DBS. Theproportion of patients responding or reaching remission increased overtime to plateau from 6 to 12 months. DBS, deep brain stimulation.

nonresponders at 12 months suggested that responders achieved

www.sobp.org/journal

ga1bsttdnc

S

mpStpisH.iSo(att

T

T

M

A

S

S

B

B

C

mae

cal GlR

464 BIOL PSYCHIATRY 2008;64:461–467 A.M. Lozano et al.

w

reater symptom improvement in the mood cluster of symptomst months 3, 6, and 12 and in the somatic cluster of symptoms atand 12 months compared with nonresponders. There were noaseline differences in total HRSD-17 score or any of theymptom cluster scores between patients who achieved or failedo achieve response status at 12 months. The different responsesimes of the various depression components suggest that theistinct symptoms of depression may be supported by differenteural substrates, a hypothesis further supported by the timeourse of PET changes described below.

econdary MeasuresWe sought to validate the improvements in depression as

easured with the HRSD-17 ratings obtained by examining theatients with the patient’s self-rating of their response to DBS.elf-reported depressive symptoms at 12 months as measured byhe BDI demonstrated a robust positive correlation with thehysician-rated HRSD-17 at 12 months (r � .73, p � .001),ndicating a congruency in the clinical and self-reported impres-ions of improvement. Additionally, the total BAI score andRSD-17 scores at 12 months were significantly correlated (r �

58, p � .01). Consistent with these observations, patientsmproved from a mean baseline Clinical Global Impression-everity rating of 5.1 (markedly to severely ill) to a mean ratingf 3.1 at 6 months and 3.2 at 12 months (mildly to moderately ill)Table 2). Improvements in CGI-S scores were seen at 1 monthnd remained statistically significant for the entire 12 months ofhe trial. Further, 6 of 17 patients who were not working due to

able 2. HRSD-17 Scores and Subcomponent Scores at Baseline and 1, 3, 6

Baseline(Mean and SD)

1 Month Mean and SDand p Value

3

otal HRSD-17 24.4 � 3.5 15.4 � 5.7t � 6.19

p � .0001ood Subscore 11.9 � 1.5 7.1 � 3.3(items 1, 2, 3, 7, 8) t � 5.36

p � .0001nxiety Subscore 3.8 � 2.4 2.1 � 1.9(items 9, 10, 11, 15, 17) t � 2.97

p � .01leep Subscore 3.6 � 2.0 1.9 � 1.8

(items 4, 5, 6) t � 3.85p � .005

omatic Subscore 4.9 � .8 4.2 � 1.3(items 12, 13, 14, 16) t � 2.16

p � .05AI 14.1 � 9.2 13.1 � 9.4

t � .29p � .78

DI 27.5 � 9.2 22.7 � 10.9t � 1.79

p � .10GI-Severity 5.1 � .7 4.0 � 1.2

t � 4.52p � .0005

The HRSD-17 contains 17 items that are assigned to various symptom cluood, guilt, suicide, work and interests, and psychomotor retardation com

sis, and insight loss comprised the anxiety subscore. The insomnia and snergy, appetite, and weight loss, respectively.

BAI, Beck Anxiety Inventory; BDI, Beck Depression Inventory; CGI, Cliniating Scale for Depression; TRD, treatment-resistant depression.

heir illness at enrollment returned to employment after being

ww.sobp.org/journal

unemployed for 2 to 7 years, indicating that DBS also led tosignificant social reintegration.

PET Measures of Regional Glucose Metabolism: Effectsof Stimulation

Of the 11 patients undergoing 18FDG PET scans, there were8 responders and 3 nonresponders. As only two nonresponderscompleted all three PET scans, results are presented only for theeight responders. The response to treatment with DBS wasaccompanied by widespread changes in metabolic activity inlimbic and cortical areas as measured with 18FDG PET (Figure3). Areas of significant change included decreases in orbital(Brodmann area [BA] 11), medial frontal cortex (BA 10/9/8), andinsula and increases in lateral prefrontal cortex (BA 11/47, BA46/10/9), parietal (BA 40), anterior midcingulate (BA 24), andposterior cingulate areas (BA 23) (Supplement 3). The medialand orbital frontal decreases were seen by 3 months and weresustained at 6 months (Supplement 3, Figure 3). A differentpattern was seen in both ventrolateral and dorsolateral prefron-tal, parietal, midcingulate, and posterior cingulate regions, wherechanges were only significant at 6 months (Supplement 3). Thetopography of the brain areas affected with DBS is specific and isconsistent with the known connectivity of the subcallosal cingu-late gyrus. The pattern of metabolic changes seen here generallyreplicates those seen in the original study of blood flow PET(Figure 3). Taken together, these results indicate that SCG DBSproduces striking changes in cognitive and limbic brain areasand they provide a biological basis for the observed improve-

12 Months (n � 20) after DBS for TRD

th Mean and SDnd p Value

6 Month Mean and SD andp Value

12 Month Mean and SDand p Value

2.5 � 7.2 11.8 � 5.9 12.6 � 6.3t � 6.49 t � 9.70 t � 8.70

p � .0001 p � .0001 p � .00015.2 � 3.6 5.1 � 3.2 5.9 � 3.7t � 7.28 t � 9.21 t � 6.62

p � .0001 p � .0001 p � .00012.0 � 1.6 1.8 � 1.9 1.6 � 1.9t � 2.68 t � 3.32 t � 3.80

p � .05 p � .01 p � .011.8 � 1.8 1.7 � 1.7 2.0 � 1.3t � 3.12 t � 3.91 t � 3.28

p � .01 p � .001 p � .0053.4 � 1.8 3.2 � 1.2 3.0 � 1.4t � 3.13 t � 4.98 t � 5.18

p � .01 p � .0001 p � .00012.8 � 10.2 12.5 � 10.6 12.9 � 10.6t � .17 t � .54 t � .53

p � 0.87 p � .60 p � .602.3 � 11.8 20.9 � 11.3 22.6 � 11.5t � .95 t � 1.83 t � .47

p � .36 p � .09 p � .653.4 � 1.2 3.1 � 1.4 3.2 � 1.4t � 6.03 t � 7.12 t � 6.62

p � .0001 p � .0001 p � .0001

. As defined by Shafer (24), items from the HRSD-17 representing depressedthe mood subscore. Psychic and somatic anxiety, agitation, hypochondri-

c subscores were derived from the three HRSD-17 sleep items and libido,

obal Impression; DBS, deep brain stimulation; HRSD-17, 17-item Hamilton

, and

Mona

1

1

2

stersprisedomati

ments in depression in these patients.

ttrodmsioa

A

pibpwcbfaw

Fosdfsc ponst whit

TF

A

WRWPWPPN

A.M. Lozano et al. BIOL PSYCHIATRY 2008;64:461–467 465

Of interest, opposite metabolic changes were present at thearget of stimulation, the subcallosal cingulate white matter, andhe immediately adjacent gray matter. In contrast to our previouseport using blood flow where only decreased SCG activity wasbserved, we noted focal increases in metabolism in the imme-iate vicinity of the stimulating electrode with decreases inetabolism in the adjacent caudal subcallosal gray matter con-

istent with the previous blood flow results (Figure 3, sagittalmage, left). These results suggest that there is a direct activationf the white matter at target that can lead to either metabolicctivation or inhibition in distinct remote brain areas.

dverse EventsTable 3 shows the adverse events in the 20 patients. Four

atients had wound infections. In three cases, this occurred earlyn the series when electrodes were externalized for several daysefore the internal pulse generator was inserted. These threeatients had their hardware removed. In one patient, the hard-are was reimplanted after a 6-month delay with recapture of the

linical benefit. In the two others who did not receive significantenefit, the explanted hardware was not replaced. All patientsrom patient 6 on had electrode and pulse generator inserted in

single surgery. One patient had superficial scalp cellulitis 2eeks after surgery that responded to antibiotics. One patient

igure 3. Changes in regional activity in TRD patients treated with chronic SCf treatment using 18F-fluoro-deoxyglucose PET (FDG, n�8) or 150 wateignificance threshold p�.005. Note similar pattern of relative glucose morsolateral prefrontal (dF) and posterior cingulate (PCC) with both tracers

ocal glucose metabolic increases restricted to the adjacent white matter (Scans. Abbreviations: 18FDG, 18F-fluoro-deoxyglucose; DBS, deep brain singulate cortex; oF, orbital frontal; ofWM, orbital frontal white matter; p,omography; SCG, subcallosal cingulate gyrus; SCWM, subcallosal cingulate

able 3. Adverse Effects in 20 Treatment-Resistant Depression Patientsollowed for 12 Months

dverse Effect Number of Patients

ound Infection and Hardware removal 3einsertion of DBS Hardware 1ound Infection Managed with Antibiotics Alone 1

erioperative Seizure 1orsening Mood/Irritability 2

erioperative Headache 4ain at Pulse Generator Site 1o Adverse Effects 7

DBS, deep brain stimulation.

experienced a generalized seizure the evening of surgery. Hewas treated with phenytoin for 3 months with no further seizures.Four patients reported headache or pain at the site of the pulsegenerator implant in the immediate postoperative period. Thebehavioral aspects of this pain included crying and despair. Suchresponses were unanticipated, as they are rarely seen in otherpatient populations having similar DBS surgery for other diag-noses. In seven patients, there were no emergent adverse effects.Finally, neuropsychological testing confirmed the absence of anycognitive adverse effects (23). There were no cases of hypomaniawith stimulation in these patients.

Discussion

Impact of DBS on TRDWe found that SCG DBS produced robust improvements in

depression in patients with TRD. Benefits were seen acrossmultiple domains of depression as reflected by improvements ineach of the mood, anxiety, somatic, and sleep subcomponents ofthe HRSD-17 scale. The maximal benefits were delayed andprogressive. They reached a plateau at 6 months and weregenerally sustained up to last time point measured at 12 months.Patients who showed a benefit at 1 month were likely to maintainresponse 6 and 12 months later, indicating that the relapse ratewas low among responders. Overall, patients were markedlyimproved with surgery and the benefits were sustained overtime, leading in certain cases to reintegration into their family andsocial activities and a return to their past work activities. Theseresults, particularly in this treatment-resistant patient population,are striking.

Mechanism of ActionThe SCG is synaptically connected with cortical and subcor-

tical structures implicated in the pathophysiology of majordepression including pregenual anterior cingulate, midcingulate,and posterior cingulate; ventromedial and dorsomedial frontalcortex; and the hypothalamus, amygdala, hippocampus, nucleus

S. Changes relative to baseline in DBS responders studied at 3 and 6 monthsd flow PET (n�3). Increases in activity shown in red; decreases in blue;lism and blood flow changes in orbital frontal (oF), medial frontal (mF),

ordance is seen in the subcallosal cingulate gyrus target region (SCG) with) in the FDG scans and more widespread SCG decreases in the blood flow

ation; dF, dorsal frontal; ht, hypothalamus; mF, medial frontal; MCC, mid; PCu, precuneus; PCC, posterior cingulate cortex; PET, positron emissione matter; vF, ventral frontal.

G DBr blooetabo. DiscCWM

timul

accumbens, and periaqueductal areas. The SCG is thus in a

www.sobp.org/journal

poairtOta(tstooetb

wsptancsiihua

R

pmcseabhdccncasiesbiSc

L

ocp

466 BIOL PSYCHIATRY 2008;64:461–467 A.M. Lozano et al.

w

osition to influence neural elements that underlie major aspectsf depression including mood state, cognitive processes, sleepnd circadian rhythm, appetite, cortisol regulation, pain process-ng, reward mechanisms, and emotional reactivity. Our PETesults show that DBS produces large changes in brain activityhat are distributed along these downstream targets of the SCG.f interest, distinct therapies including antidepressant medica-

ions, cognitive behavioral therapy, and electroconvulsive ther-py produce similar changes in many of these same brain regions25). This suggests that SCG DBS together with other therapieshat also ameliorate depression exert their effects on a commonet of neural substrates. These observations support the idea thathe changes in the clinical attributes of depression we havebserved with DBS are a consequence of changes in the activityf these brain regions. We cannot be certain, however, to whatxtent these changes in brain activity are necessary or sufficiento produce the clinical benefit or whether they are merelyiological markers of the state of improvement.

Despite the clinical and metabolic changes in brain circuits,e do not know the precise mechanism of action through which

timulation exerts its effects at the cellular level. The proposedhysiologic effects of DBS include axonal excitation, depolariza-ion blockade, synaptic release of neurotransmitters, jamming ofbnormal neuronal firing, and disruption of pathological neuro-al synchrony (26). Both anterograde and retrograde effects mayontribute to the observed clinical effects. Our results are con-istent with the notion that SCG DBS drives focal activity at themmediate target, which, in turn, leads to inhibition or excitationn adjacent and remote areas to which it is connected. Weypothesize that DBS disrupts pathological activity in the circuitsnderlying depression and allows more normal cerebral functionnd behavior.

esponders Versus NonrespondersOne of the greatest challenges is to understand why some

atients with TRD respond to SCG DBS and others do not. Theajority of patients in the study receiving DBS met response

riteria and most others, indeed all but two patients, showedome drop in HRSD-17 scores; no patient experienced a wors-ning of symptoms. The reasons for the variability in responsecross this patient population with TRD are not well understoodut likely are related to the heterogeneity of MDD. To date, weave not identified demographic, clinical, or image-based pre-ictors of response. There were too few nonresponders toharacterize responder/nonresponder differences in the PEThanges, especially since there is a strong partial response inonresponders. This may change with a larger sample. We alsoonsidered the possibility that the response to SCG DBS could be

function of the location of electrode placement and thepecificity of the neural structures being stimulated. Our prelim-nary analysis suggests that there are no significant differences inlectrode location within the SCG in responders and nonre-ponders. We are using diffusion tensor imaging and voxel-ased morphometry to study whether neuroanatomic variationsncluding the position of axonal afferents and efferents within theCG or differences in gray matter density across subjects couldontribute to the variability (27).

imitationsA limitation of this study is the open label assessment of

utcomes. While we do not know the relative magnitude of itsontribution, there are several observations that argue against a

lacebo response playing a major role. First, the response was

ww.sobp.org/journal

progressive over time. A different pattern with an initial largeeffect followed by a decay would be more in keeping with aplacebo response. Second, we have noted that for each patient,the effects vary with which electrode contacts and electricalstimulation settings are chosen, arguing for the specificity ofstimulation. Third, in two patients, one planned and anotherunplanned, blinded discontinuation of stimulation was associ-ated with a recurrence of depression within 1 week and arecapture of benefit within 1 week of reintroduction of thestimulation. Finally, examination of placebo response rates inpatients with chronic and severe depression, particularly thosewith melancholia, show a reduced propensity for placebo-mediated improvements (28,29), including a reported placeboresponse rate to vagus nerve stimulation in a comparably treat-ment-resistant patient population of 10% at 10 weeks (30). Takentogether, the results suggest that the improvements are highlydependent on the delivery of effective stimulation to the SCG.

Conclusion

We conclude that SCG DBS improves many of the symptomsof severe depression in patients who have failed to respond toconventional treatments. Improvements are seen within 1 monthand last for at least 1 year. The procedure is well tolerated and nopatient suffered a permanent serious adverse effect. In contrast topreviously utilized ablative neurosurgical procedures for depres-sion, DBS is adjustable and stimulation is reversible. Thesefeatures increase safety and may offer advantages for both theefficacy of the therapy and its acceptance in the patient, medical,and psychiatric community.

We have shown that DBS applied to the SCG producesstriking changes in the metabolic activity of the brain circuits thatmediate depression. Our results are consistent with the notionthat changes in activity in these selected brain circuits areresponsible for the behavioral changes and clinical improve-ments we have seen in this treatment-resistant population. Deepbrain stimulation in treatment-resistant patients requires a dedi-cated multidisciplinary team of neurosurgeons, psychiatrists, andsupport staff. While these results, particularly in this difficult-to-treat population, are promising, they represent an initial step anda blinded evaluation of DBS in TRD is required before it could beadopted on a wider scale.

AML is a Tier 1 Canada Research Chair in Neuroscience andholds the RR Tasker Chair in Functional Neurosurgery. This workwas supported in part by a Distinguished Investigator Awardfrom National Alliance for Research in Schizophrenia andDepression (HSM).

We are grateful to Dr. McNeely for assistance with neuropsy-chological testing, Drs. Dostrovsky and Hutchison for theirongoing collaboration in intraoperative neurophysiology, andMs. Sakina Rizvi for assistance in study coordination.

Andres Lozano: Consultant, Advanced Neuromodulation Sys-tems and Medtronic, and Intellectual property holder and Licen-sor in the field of Deep Brain Stimulation. Helen Mayberg:Consultant, Licensing Fees from Intellectual Property: AdvancedNeuromodulation Systems. Sidney H. Kennedy: Speaker’s Bu-reau: Dr. Kennedy has received honoraria from Biovail, Eli Lilly,GlaxoSmithKline, Janssen-Ortho, Lundbeck, Organon, Servier,and Wyeth. Consultant-Advisory Boards: Dr. Kennedy has re-ceived consultant fees from Advanced Neuromodulation SystemsInc., Biovail, Boehringer Ingelheim, Eli Lilly, GlaxoSmithKline,

Janssen-Ortho, Lundbeck, Organon, Pfizer, Servier, and Wyeth.

GAbCcf

o

1

1

1

1

A.M. Lozano et al. BIOL PSYCHIATRY 2008;64:461–467 467

rant Funding: Dr. Kennedy has received grant funding fromstraZeneca, Eli Lilly, GlaxoSmithKline, Janssen Ortho, Lund-eck, and Merck Frosst. Drs. Clement Hamani and Cameronraddock reported no biomedical financial interest or potentialonflicts of interest. Dr. Peter Giacobbe has received postdoctoralellowship financial support from Advanced Modulation Systems.

Supplementary material cited in this article is availablenline.

1. The World Health Organization(2004): Annex Table 3: Burden of diseasein DALYs by cause, sex, and mortality stratum in WHO regions, estimatesfor 2002. In: The World Health Report: Changing History. Geneva: WorldHealth Organization.

2. Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE(2005): Lifetime prevalence and age-of-onset distributions of DSM-IVdisorders in the National Comorbidity Survey Replication. Arch Gen Psy-chiatry 62:593– 602.

3. Kessler RC, Chiu WT, Demler O, Merikangas KR, Walters EE (2005): Preva-lence, severity, and comorbidity of 12-month DSM-IV disorders in the Na-tional Comorbidity Survey Replication. Arch Gen Psychiatry 62:617–627.

4. Depression Guideline Panel, editor(1993): Depression in Primary Care:Volume 2 Treatment of Major Depression. Rockville, MD: U.S. Departmentof Health and Human Services, Public Health Service, Agency for HealthCare Policy and Research.

5. Conwell Y, Brent D (1995): Suicide and aging. I: Patterns of psychiatricdiagnosis. Int Psychogeriatr 7:149 –164.

6. Kochanek KD, Murphy SL, Anderson RN, Scott C (2004): Deaths: Finaldata for 2002. Natl Vital Stat Rep 53:1–115.

7. Weissman MM, Bland RC, Canino GJ, Greenwald S, Hwu HG, Joyce PR,et al. (1999): Prevalence of suicide ideation and suicide attempts innine countries. Psychol Med 29:9 –17.

8. Dougherty DD, Rauch SL (2007): Somatic therapies for treatment-resis-tant depression: New neurotherapeutic interventions. Psychiatr ClinNorth Am 30:31–37.

9. Mayberg HS, Liotti M, Brannan SK, McGinnis S, Mahurin RK, Jerabek PA,et al. (1999): Reciprocal limbic-cortical function and negative mood:Converging PET findings in depression and normal sadness. Am J Psy-chiatry 156:675– 682.

0. Mayberg HS (2003): Modulating dysfunctional limbic-cortical circuits indepression: Towards development of brain-based algorithms for diag-nosis and optimised treatment. Br Med Bull 65:193–207.

1. Seminowicz DA, Mayberg HS, McIntosh AR, Goldapple K, Kennedy S,Segal Z, et al. (2004): Limbic-frontal circuitry in major depression: A pathmodeling metanalysis. Neuroimage 22:409 – 418.

2. Agid Y, Buzsaki G, Diamond DM, Frackowiak R, Giedd J, Girault JA, et al.(2007): How can drug discovery for psychiatric disorders be improved?Nat Rev Drug Discov 6:189 –201.

3. Ressler KJ, Mayberg HS (2007): Targeting abnormal neural circuits in

mood and anxiety disorders: From the laboratory to the clinic. NatNeurosci 10:1116 –1124.

14. Fine J, Duff J, Chen R, Chir B, Hutchison W, Lozano AM, et al. (2000):Long-term follow-up of unilateral pallidotomy in advanced Parkinson’sdisease. N Engl J Med 342:1708 –1714.

15. Lang AE, Lozano AM (1998): Parkinson’s disease. Second of two parts.N Engl J Med 339:1130 –1143.

16. Lang AE, Lozano AM (1998): Parkinson’s disease. First of two parts. N EnglJ Med 339:1044 –1053.

17. Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C,et al. (2005): Deep brain stimulation for treatment-resistant depression.Neuron 45:651– 660.

18. First MB, Spitzer RL, Gibbon M, Williams JBW (2002): Structured ClinicalInterview for DSM-IV-TR Axis I Disorders, Research Version, Patient Edition.(SCID-I/P). New York: Biometrics Research, New York State PsychiatricInstitute.

19. Hamilton MA (1960): Rating scale for depression. J Neurol NeurosurgPsychiatry 23:56 – 62.

20. First MB, Spitzer RL, Gibbon M, Williams JBW (1997): Structured ClinicalInterview for DSM-IV Personality Disorders, (SCID-II). Washington, DC:American Psychiatric Press, Inc.

21. Beck AT, Steer RA (1993): Beck Anxiety Inventory Manual. San Antonio, TX:Psychological Corporation.

22. Beck AT, Steer RA, Brown GK (1996): Beck Depression Inventory Manual,2nd ed. San Antonio, TX: Psychological Corporation.

23. McNeely HE, Mayberg HS, Lozano AM, Kennedy SH (2008): Neuropsy-chological impact of Cg25 deep brain stimulation for treatment resis-tant depression: Preliminary results over 12 months. J Nerv Ment Disease196:405– 410.

24. Shafer AB (2006): Meta-analysis of the factor structures of four depres-sion questionnaires: Beck, CES-D, Hamilton, and Zung. J Clin Psychol62:123–146.

25. Goldapple K, Segal Z, Garson C, Lau M, Bieling P, Kennedy S, et al. (2004):Modulation of cortical-limbic pathways in major depression: Treat-ment-specific effects of cognitive behavior therapy. Arch Gen Psychiatry61:34 – 41.

26. Lozano AM, Dostrovsky J, Chen R, Ashby P (2002): Deep brain stimula-tion for Parkinson’s disease: Disrupting the disruption. Lancet Neurol1:225–231.

27. Johansen-Berg H, Gutman DA, Behrens TE, Matthews PM, RushworthMF, Katz E, et al. (2008): Anatomical connectivity of the subgenual cin-gulate region targeted with deep brain stimulation for treatment-resis-tant depression. Cereb Cortex 18:1374 –1383.

28. Brown WA (2007): Treatment response in melancholia. Acta PsychiatrScand Suppl 433:125–129.

29. Peselow ED, Sanfilipo MP, Difiglia C, Fieve RR (1992): Melancholic/en-dogenous depression and response to somatic treatment and placebo.Am J Psychiatry 149:1324 –1334.

30. Rush AJ, Marangell LB, Sackeim HA, George MS, Brannan SK, Davis SM,et al. (2005): Vagus nerve stimulation for treatment-resistant depres-sion: A randomized, controlled acute phase trial. Biol Psychiatry 58:347–

354.

www.sobp.org/journal

Lozano et al.

Supplement 1 PET data acquisition and analysis Regional cerebral glucose metabolism (rCMGlc) was acquired preoperatively and after 3 and 6 months of chronic DBS using standard methods (1). As with the original blood flow scans performed in the first report of this procedure (2), all scans were acquired with subjects resting, with eyes closed and no explicit cognitive or motor instructions. Wakefulness was monitored every 10 minutes during the 40 minute glucose update period. For each scan, a 5 mCi (185-Mbq) FDG dose was injected intravenously, with scans acquired on a Siemens-Biograph HiRez XVI PET camera operating in 3D mode (16.2 cm field of view, 81 parallel slices; 2.0 mm center-to-center inter-slice distance). Raw images were attenuation corrected, reconstructed and smoothed to a final in plane resolution of 6.8 mm full width at half maximum (FWHM). Absolute glucose metabolism was not quantified. Statistical analyses were performed using SPM5 (Wellcome Department of Cognitive Neurology, London, UK) implemented in Matlab (version 5.3, Mathworks Inc., Sherborn, MA). Scans at all 3 time points were available for 10 subjects. PET scans were first realigned, and then spatially normalized into standard three-dimensional space relative to the anterior commissure using the MNI ICBM 152 stereotactic template within SPM5. The images were then smoothed using a 6mm FWHM Gaussian kernel and corrected for differences in the whole-brain global mean, and to a final in-plane resolution of 9.1 mm at FWHM. Effects of DBS on glucose metabolism (Baseline, 3 months, 6 months) were assessed using a general linear model (F-test: peak voxel threshold p=.01; cluster threshold 50 voxels (voxel=2mm3) (3). Post-hoc t-tests were used to evaluate how regional metabolism changed over time (3 month vs Baseline, 6 month versus baseline, 6 vs 3 months). Primary analyses focused on the 8 DBS responders (Figure 3). For all analyses, t-maps are reported at a significance level of p<.005 uncorrected (t>3.06, df=11.9). Brain locations are reported as x, y, z coordinates in MNI space with approximate Brodmann areas (BA) identified by mathematical transformation of SPM5 coordinates into Tailarach space (http://www.mrc-cbu.cam.ac.uk/Imaging/) (Table 3). 1. Goldapple K, Segal Z, Garson C, Lau M, Bieling P, Kennedy S, Mayberg HS.

(2004): Modulation of cortical-limbic pathways in major depression: treatment-specific effects of cognitive behavior therapy. Archives of general psychiatry 61:34-41.

2. Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, et al. (2005): Deep brain stimulation for treatment-resistant depression. Neuron 45:651-660.

3. Worsley KJ, Marrett S, Neelin P, Vandal AC, Friston KJ, Evans AC (1996): A unified statistical approach for determining significant signals in images of cerebral activation. Hum Brain Mapp 4:58-73.

Lozano et al. Supplement 2

Medications in each of 20 individual treatment resistant depression patients at baseline and at 3, 6 and 12 months after Subcallosal cingulate Gyrus Deep Brain Stimulation.

PT Baseline Medications Month 3 Month 6 Month 12 HDRS17 Baseline Res/Rem 6 mo Res/Rem 12 mo1 Bupropion 550mg 450mg (at wk10) 450mg 450mg 29 83% (5)* 17% (24)

Citalopram 120mg No changes No changes No changes remitter# noDexedrine 45mg 30mg (at wk10) 30mg 30mgPerphenazine 6mg No changes No changes No changes

2 Venlafaxine 150mg No changes No changes No changes 22 32% (15) 32% (15)Olanzapine 10mg No changes No changes No changes no no

3 Paroxetine 40mg bid No changes No changes No changes 26 65% (9) 46% (14)Bupropion 100mg No changes No changes No changes responder noRisperidone 2mg bid No changes No changes No changesClonazepam 2mg tid No changes No changes No changesLorazepam 2mg tid No changes No changes No changes

4 Bupropion SR 30mg No changes No changes No changes 24 8% (22) 8% (22)Paroxetine 30mg No changes No changes No changes no noModafanil 200mg No changes No changes No changesLamotrigine No changes No changes No changesTrazodone 150mg No changes No changes No changesZopiclone 7.5mg No changes No changes No changes

5 Bupropion 300 mg No changes No changes 150mg 26 65% (9) 62% (10)Citalopram 80mg No changes No changes No changes responder responderLorazepam 0mg 0mg 1-2mgMethylphenidate 0mg 0mg 10mgQuetiapine 300mg No changes no changes 450mg

6 Dalmane 60 mg discontinued 25 44% (14) 68% (8)Methylphenidate 40 mg discontinued respoderMirtazapine 60 mg 30mg discontinuedNozinan 20 mg No changes No changes discontinuedQuetiapine 900mg 700mg (at wk8) 500mg 600mgVenlafaxine 300mg No changes 375mg (mth4) 225mg (mth10)

7 Mirtazapine 30mg discontinued 34 56% (15) 65% (12)Risperidone 1mg No changes No changes No changes responder responderBupropion 150mg bid 150mg 150mg bid (mth5) No changesLamotrigine 300mg No changes 250mg No changesMethylphenidate-XR 54mg 18mg discontinued discontinuedDextroamphetamine sulfate 15mg No Changes No Changes No ChangesContin 240mg bid - codeine No Changes No Changes No ChangesEltroxin 0.15mg - thyroid No Changes No Changes No ChangesSythroid 0.75mg - thyroid No Changes No Changes No ChangesAndrogel 5 mg - testosterone therapy No Changes No Changes No ChangesClonazepam 0.5mg prn 1.5mg prn (mth 2) 2mg prn (mth 4) 2mg prnAtomoxetine started 40-80mg mth 4 160mg No changes

8 Venlafaxine 375mg 150mg (wk 1) No changes No changes 24 42% (14) 58% (10)Fluoxetine 60mg No Changes No Changes No Changes no responderLithium 1500mg No Changes No Changes No ChangesClonazepam 1mg No changes No changes 0.5mg (mth 11)Ritalin 20mg bid No Changes No Changes No ChangesPropanolol 20mg bid No Changes No Changes No ChangesOlanzapine 2.5mg 5mg No changes 7.5mgDilantin 200 mg bid started following a post-op seizure and discontinued at 3 monthsSenokot 8.6mg bid prn No changes No changes No changesProdiem 1 caplet prn No changes No changes No changesCialis 20mg prn No changes No changes No changes

9 Phenylzine 30mg bid no changes 30mg qam and 15mg noon 30mg qam and 15mg noon 24 37.5 (15) 54% (11)Quetiapine 200 mg 75mg 75mg 150mg no responderLithium 900 mg No changes No changes No changesLorazepam 2mg No changes No changes No changesolanzapione 10g prn No changes No changes No changesl-thyroxine 0.5mg No changes No changes No changes

PT Baseline Medications Month 3 Month 6 Month 12 HDRS17 Baseline Res/Rem 6 mo Res/Rem 12 mo10 Escitalopram 20mg No changes No changes 30mg 29 17% (24) 34% (19)

Mirtazapine 30mg No changes No changes No changes no noRisperidone 2 mg No changes No changes No changesDiazepam 5mg No changes No changes No changesL-Thyroxine 0,088mg No changes No changes No changesAmantidine 100mg No changes No changes No changesPhenobarbital 200mg prn No changes No changes No changesLisinopril-Hydrochlorothiazide 1 tab No changes No changes No changesMetoprolol 200mg No changes No changes No changesMetformin 1000mg No changes No changes No changesFerrous Gluconate 300mg No changes No changes No changesDiane 35 1 tablet No changes No changes No changes

11 clonazepam 0.5mg prn 4mg prn 3mg prn 2mg prn 21 67% (7) 62% (8)quetiapine 100mg 100mg (started wk10) discontinued remitter responderbuspirone 0 mg 20mgtid 0 mgmodafanil 0 mg 50mg 0mgescitalopram 0mg 5mg 0mg

12 Parnate 100mg No Changes No changes No changes 22 68% (7) 68% (7)Ritalin No changes No changes No changes remitter remitterModafanil No changes No changes No changes

13 Venlafaxine 150mg tid No changes No changes No changes 23 39% (14) 30% (16)Lithium 600mg No changes No changes No changes no noZopiclone 7.5mg No changes No changes No changesOlanzapine 5mg tid No changes discontinued (5mths)Mirtazapine 60mg No changes No changes discontinuedSeroquel 50mg tid No changes No changes 50mg bid (mth11)Lorazepam 1mg tid No changes No changes No changesDomperidone 10mg tid No changes No changes No changesLipitor 10mg No changes No changes No changes

14 Quetiapine 175mg No changes No changes No changes 22 59% (9) 77% (5)Clonazepam 3mg No changes 2mg No changes responder remitterVenlafaxine 300mg No changes No changes No changesBupropion 150mg No changes No changes No changesZopiclone 7.5mg No changes No changes No changes

15 Escitalopram 40mg 20mg (wk9) No changes No changes 23 74% (6) 87% (3)Modafanil 300mg 200mg (wk9) 200mg qam and 300mg qhs 300mg bid remitter remitterQuetiapine 100mg No changes No changes 150mg (mth 11)Trazadone 300mg No changes No changes No changesZopiclone 7.5-15mg prn No changes No changes No changesAldactone (alopecia) No changes No changes No changes

16 Clonazepam 1-3mg prn No changes No changes No changes 20 80% (4) 40% (12)Zopiclone 2mg prn No changes No changes No changes remitter noQuetiapine 25mg prn discontinued

17 Fluoxetine 60mg 40mg No changes No changes 21 52% (10) 33% (14)Pramipexole 0.25mg No changes No changes No changes responder noDexedrin 0mg started mth 4 No changesFerrous Gluconate 1 tablet No changes No changes No changesZopiclone 7.5mg No changes No changes No changesClonazepam 1mg No changes No changes 0.5mg (mth 10)L-thyroxine 0.088mg No changes No changes No changes

18 Clomipramine 175mg No changes No changes No changes 20 65% (7) 60% (8)Seroquel 50mg discontinued discontinued 50mg remitter responder-remitterTryptophan 3000mg discontinued

19 Citalopram 60mg No changes No changes 40mg 26 85% (4) 69 (8)Norvasc 10mg No changes No changes No changes remitter responder-remitterLorazepam 1mg No changes No changes No changesHydrazide 25mg No changes No changes No changesInhibace 5mg bid No changes No changes No changesLipitor 20mg No changes No changes No changes

20 Trazadone 50mg No changes No changes No changes 26 23% (20) 0% (26)Escitalopram 20mg No changes discontinued no noEndocet 6 tablets No changes No changes No changesLorazepam 2mg No changes No changes No changes

*The percentage decrease and the raw score (in parentheses) in the HDRS-17 score after 6 or 12 months of SCG DBS#Responder=Hamilton Depression score decrease greater than 50%; Remitter=absolute Hamilton Depression Score <8

Lozano et al.

Medication Classes

pt SNRI/SSRI 2nd AD Augm/Atypical BZD/Hypnotic Stimulant Comments1 Citalopram Bupropion Perphenizine Dexedrine2 Venlafaxine Olanzapine3 Paroxetine Bupropion Risperidone Clonazepam

4 Paroxetine Bupropion LamotrigineTrazodoneZopiclone Modafanil

5 Citalopram Bupropion Quetiapine Lorazepam Methylphenidate

6 Venlafaxine MirtazapineQuetiapine

Nozinan Flurazepam Methylphenidate

7MirtazapineBupropion

RisperidoneLamotrigine Clonazepam

AtomoxetineDextroamphetamine

Methylphenidate

8VenlafaxineFluoxetine 1

LithiumOlanzapine Clonazepam Methylphenidate phenytoin prophylaxis 3 month post-surgery

9 MAOI

QuetiapineLithium

Olanzapine Lorazepam

10 Escitalopram MirtazapineRisperidone

Phenobarbital Diazepam11 Clonazepam brief trials of buspirone, escitalopram, nortriptyline d/c 2/5212 Parnate self-medicated with Methylphenidate and Modafanil

13 Venlafaxine MirtazapineLithium

QuetiapineZopiclone

Lorazepam

14 Venlafaxine Bupropion QuetiapineClonazepam

Zopiclone15 Escitalopram Trazodone Quetiapine Zopiclone Modafanil

16Clonazepam

Zopiclone

17 Fluoxetine PramipexoleZopiclone

Clonazepam18 Clomiprimine19 Citalopram Lorazepam

20 EscitalopramTrazodoneLorazepam

Medication decrease

Lozano et al. Supplement 3

3 months DBS vs Baseline 6 months DBS vs Baseline

Region Brodmann side cluster T value* coordinates region Brodmann side cluster T value* coordinatesarea size (v3) x y z area size x y z

INCREASES INCREASESsubgenual cingulate WM 25/32 R 7 3.69 4 27 -6 subgenual cingulate WM 25/32 L 4.08 -8 31 -7medial orbital frontal WM 11 L 38 4.13 -14 40 -17 medial orbital frontal WM 11 L 165 8.07 -12 34 -20

32/10 R 15 3.66 18 45 11 11 R 799 5.39 18 28 -17posterior cingulate 23/31 L 534 5.1 -4 -52 12

23/31 L 33 4.15 -20 -52 1423/31 L 18 4.31 -10 -30 29

rostral Caudate R 4 3.25 18 21 3 rostral Caudate R 4 3.32 10 19 -1hippocampus R 30 3.78 20 -45 -10 hippocampus R 10 4.14 22 -41 -6

parahippocampus R 13 3.36 34 -24 -14lateral orbital frontal 11 L 19 3.74 -36 50 -19 lateral orbital frontal 11 L 180 5.36 -38 42 -16

11 R 21 3.72 30 48 -12ventrolateral prefrontal 45 R 14 3.51 44 18 6 ventrolateral prefrontal 44/45 R 83 4.65 50 14 7

dorsolateral prefrontal 8 L 21 4.64 -30 20 49medial premotor 6 R 54 4.19 8 -8 69 dorsolateral premotor 6 L 54 4.26 -30 -5 57

6 R 545 5.09 2 -28 68 6 R 6 3.44 48 -5 11mid insula R 55 3.74 34 -2 -3 mid insula R 12 3.53 32 8 11post insula/inf parietal 40/41 R 6439 9.62 34 -17 16 post insula/inf parietal 40/41 R 15 4.37 36 -34 11Parietal 40 L 34 4.28 -30 -39 37 Parietal 40 R 89 6.6 34 -17 16

40 L 780 5.33 -32 -72 42 40 R 3935 6.87 28 -45 394 L 37 3.83 -36 -22 64 40 R 442 6.06 24 -15 502 L 151 4.96 -53 -17 47 2 L 22 3.95 -53 -17 47

PreCuneus 7 L 5.04 -22 -65 22 PreCuneus 7 L 23 3.51 -8 -21 45lateral temporal-occipital 21 R 82 6 65 -20 -17 lateral temporal-occipital 21 R 1496 9.39 65 -20 -17

21 R 94 5.58 57 1 -14 21 R 22 4.56 59 1 -1420 R 353 4.26 40 -21 -26 21 R 109 4.37 50 -5 -25

21/39 R 5.74 61 -57 19 21/37 R 12 3.51 44 -50 8

DECREASES DECREASESmedial orbital frontal 11 L 42 -4.01 -4 61 -15 medial orbital frontal 11 L 39 -4.77 -2 61 -17ventromedial frontal 10 L 18 -3.94 -6 57 5 ventromedial frontal 10 R 1017 -6.88 6 59 23ventral frontal pole 10 L 24 -3.64 -24 58 -5

10 R 47 -4.23 20 63 13dorsal frontal pole 10 L 475 -8.03 -28 52 23 dorsal frontal pole 10 L 391 -9.63 -30 52 25

9 R 37 -3.96 28 46 27dorsomedial frontal 8 R 93 -4.39 8 41 44 dorsomedial frontal 9/8 R 8 -3.56 8 45 38

8 L 931 -6.91 -8 41 40 8 L 6 -3.37 -20 43 44rostral anterior cingulate 24 L 7 -3.72 -8 41 3anterior mid-cingulate 24 L 45 -5.2 -4 -16 36 anterior mid cingulate 24 L 92 -3.68 -12 12 47

32 R 18 -4.09 8 12 42 24/32 R 11 -3.46 10 13 34dorsal caudate L 38 -5.72 -14 7 14 dorsal caudate L 30 -4.85 -12 5 13dorsolateral prefrontal 46 L 18 -4.34 -42 32 13 dorsolateral prefrontal 44 L 26 -3.69 -65 3 22

9 L 13 -3.66 -44 15 38 9/44 L 2 -3.34 -34 17 278 L 39 -3.93 -40 18 47 9 R 145 -5.58 30 44 298 R 28 -3.74 46 20 43 8 L 16 -3.76 -42 18 49

dorsolateral premotor 6 L 11 -3.89 -40 -2 39dorsal pons R 7 -3.75 4 -34 -10

*SPM t maps. Significance threshold: p < 0.005 uncorrected (t > 3.06, df=11.9) **conversion MNI coordinates to Talairach coordinates (http://www.mrc-cbu.cam.ac.uk/Imaging/)