Long-term results of Deep Brain Stimulation of the Anterior Cingulate

Cortex for Neuropathic Pain

Running title: Follow-up of ACC DBS

Sandra GJ BOCCARD1, PhD; Simon J PRANGNELL1, DClinPsych; Laurie PYCROFT1, MSc; Binith

CHEERAN1; Liz MOIR1, RGN; Erlick AC PEREIRA 2, DM; James J FITZGERALD1, PhD;

Alexander L GREEN1, MD; Tipu Z AZIZ1, F.Med.Sci

1 Oxford Functional Neurosurgery and Experimental Neurology Group, Nuffield Department of

Clinical Neurosciences and Surgery, University of Oxford, UK2 Academic Neurosurgery Unit, St George’s, University of London, UK

Corresponding author: Mrs. Sandra GJ BOCCARD-BINET, PhD, Nuffield Department of Clinical

Neurosciences, Level 6, West Wing, John Radcliffe Hospital, Oxford, OX3 9DU, UK TEL: +44 1865

231845 FAX: +44 1865 231885 sandra.boccard@ndcn.ox.ac.uk

1

Abstract

Background - Deep Brain Stimulation of the Anterior Cingulate Cortex is a recent technique that has

shown some promising short-term results in patients with chronic refractory neuropathic pain. Three

years after the first case-series, we assessed its efficacy on a larger cohort, with longer follow-up.

Methods – 24 patients (19 males; 49.1 years) with neuropathic pain underwent bilateral ACC DBS.

Patient reported outcome measures were collected pre- and post-surgery, using the Numerical Rating

Scale (NRS), Short-Form 36 quality of life (SF-36), McGill pain (MPQ) and EuroQol-5D

questionnaires.

Results – 22 patients after a trial week were fully internalized and 12 had a mean follow-up of 38.9

months. Six months post-surgery the mean NRS score dropped from 8.0 to 4.27 (P=.004). There was a

significant improvement in the MPQ (mean -36%; P=.021) and EQ-5D score significantly decreased

(mean -21%; P=.036). The PF domain of SF-36 was significantly improved of (mean +54.2%; P=.01).

Furthermore, in 83% of these patients: at 6 months NRS was improved by 60% (P<.001) and MPQ

decreased by 47% (P<.01). After 1 year, NRS decreased by 43% (P< .01), EQ-5D was significantly

reduced (mean -30.8; P=.05) and significant improvements were also observed for different domains

of the SF-36. At longer follow-ups, efficacy was sustained up to 42 months in some patients, with a

NRS as low as 3.

Conclusions – Follow-up results confirm that ACC DBS alleviates chronic neuropathic pain refractory

to pharmacotherapy and improves quality of life in a significant number of patients.

Key words: Deep Brain Stimulation, Anterior Cingulate Cortex, Neuropathic pain

2

INTRODUCTION

Neuropathic pain is defined by the International Association for the Study of Pain (IASP) as ”pain

arising as a direct consequence of a lesion or disease affecting the somatosensory nervous system” 1.

Chronic neuropathic pain generated by abnormal activity of the nervous system is particularly

widespread and debilitating. In the U.S. alone, more than 116 million adults suffer from chronic pain,

a condition still difficult to treat with an estimated annual cost of over $500 billion in medical

treatment and loss of productivity 2-4. Deep Brain Stimulation (DBS) has been used to relieve

pharmaco-resistant neuropathic pain for more than sixty years5-9. In spite of its increasing use the

mechanisms of DBS for pain are still unclear and its efficacy varies from one patient to another 6.

Many brain areas have been targeted, notably the periaqueductal/periventricular gray matter

(PAG/PVG) 10 and the ventral posteromedial and lateral nuclei of the sensory thalamus (VPM/VPL) 11.

Stimulation of these structures has been effective on the sensory component of pain12; however, pain is

a complex sensation consisting of at least three components: sensory-discriminative (pain intensity),

affective (pain unpleasantness) and cognitive, generating an integrated experience13.

Not all patients respond to stimulation along the primary sensory pathways and therefore alternative

targets have been sought. For instance, it may be that some patients’ pain might be improved by the

stimulation of other regions such as the anterior cingulate cortex (ACC) a critical structure of the

affective component14. The ACC has been shown to be involved in many motor and psychological

functions15, and has also been demonstrated as playing major roles in pain16-20 and empathy 21.

Historically this part of the brain has been neurosurgically lesioned to relieve pain in terminal cancer,

with the first such procedure performed in 194822. Subsequently, this technique was promoted by

Ballantine in Massachusetts23. We published a review of eight studies for pain of malignant origin,

mentioning variable efficacy of the procedure from 32-83%24. Bilateral anterior cingulotomy has also

been used to relieve refractory pain of noncancerous origin25, 26.

Three years after assessing the first case series of ACC DBS for pain, we present here a larger cohort

with long term results.

3

METHODS

Patients

Patients were referred by clinicians nationally to a single center multi-disciplinary team consisting of

pain specialists, neuropsychologists and neurosurgeons whose clinical protocol is detailed elsewhere 6.

Neuropsychological evaluation excluded psychiatric disorders and ensured minimal cognitive

impairment. A definable organic origin for pain was sought, with the patient refractory to or poorly

tolerant of pharmacological treatments. In one case, the patient's pain was of unknown cause but was

clearly neuropathic in character, and after careful assessment underlying psychiatric morbidity was felt

extremely unlikely. Eligible patients had neuropathic pain refractory to pharmacotherapy for at least

two years together with absence of surgical contraindications such as coagulopathy. Informed consent

was obtained from all patients prior to surgery. Some of the patients had failed surgical treatments

such as DBS of PVG/PAG and/or sensory thalamus (patients A, C, E, H, I, J, L, P and R). For other

patients, widespread, poorly localized pain (patients B, D, F, G, K, M, N, O, Q, S, T, U, V, W and X)

led to ACC DBS being offered in the first instance.

Surgical Methods

Our surgical technique for deep brain stimulation has been previously described 27,28. Briefly, a

stereotactic frame was applied to the patient’s head and a CT scan was performed then fused with a

preoperatively obtained MRI and used to calculate target coordinates and lead trajectory. The ACC

target used was 20mm posterior to the anterior tip of the frontal horns of the lateral ventricles

(Figure1). The contacts were mostly in the cingulum bundle white matter, with the deepest contact in

the corpus callosum. The lead was a Medtronic model 3387 depth lead (Medtronic Inc., Minneapolis,

USA). A post-operative CT scan, fused to the pre-operative MRI, confirmed electrode position. Leads

were externalized via temporary extensions. The implanted pulse generator (IPG) was introduced

under the skin of the patient’s upper chest after a one-week trial period of stimulation on the ward.

4

Pain assessments

Pain and health-related quality of life measures were assessed before surgery and during follow-up.

The Numerical Rating Scale (NRS) and McGill pain questionnaire (MPQ) were used29. The NRS

extends from `no pain’ (0) to `the worst pain you can imagine’ (10), and the MPQ gives additional

qualitative information in domains of ‘sensory’, ‘affective’, ‘evaluative’ and ‘miscellaneous’ pain

severity. Patients recorded their NRS scores during their good, bad, and average days; pre-surgery and

post-surgery. The NRS scores were reviewed to ensure consistency and the mean was then calculated.

The MPQ was also completed before and after surgery and analyzed using the ranked pain rating

index (PRI(R))29.

Patients also completed short-form 36 (SF-36®) and Euroqol 5 domain (EQ-5D) quality of life

questionnaires alongside the pain questionnaires. SF-36® responses were regrouped into 8 domains of

physical functioning (PF), role - physical (RP), bodily pain (BP), general health (GH), vitality (VT),

social functioning (SF), role - emotional (RE), and mental health (MH). Results were scored using

online tools30. Norm-based scores allowed comparison between studies. The SF-36® scale ranged

from a score of 0, an extreme of dysfunction or symptom severity, to 100, optimal function. The health

state of patients was evaluated by EQ-5D. Its two sections evaluate firstly the health state in five

dimensions (mobility, self-care, usual activities, pain and anxiety) and secondly on a ‘health’ NRS,

with 0 being the worst state they can imagine and 100 the best. EQ-5D scores were calculated as

detailed elsewhere31.

As pain assessments were repeated measures throughout the follow-up, pre- and postoperative scores

were compared using a general linear mixed model, with a P value of <.05 taken as statistically

significant.

Neuropsychometry

Complete pre- and post-operative psychometric assessment data was available for eight out of the

twenty four patients featured in this paper (G, M, N, O, Q, R, T, and U). Assessment 1 was completed

as a baseline measure prior to surgery and assessment 2 was completed on average 21.88 months

following surgery (range 7 – 58 months, SD = 15.90). Patients’ scores were standardized according to

normative data provided for each test. In line with standard neuropsychological practice, pre- and

post-operative score differences of between 1 and 2 standard deviations were regarded as indicative of

mild change and differences of more than 2 standard deviations as indicative of significant change.

The assessment battery covered the following cognitive domains:

Reasoning: Wechsler Adult Intelligence Scale - Revised 32/ Wechsler Adult Intelligence Scale-IV

(WAIS-IV; 33) Matrix Reasoning and Similarities subtests

5

Verbal memory: Adult Memory and Information Processing Battery List & Story Learning Tasks 34/ California Verbal Learning Test-II 35 & Wechsler Memory Scale-IV (WMS-IV; 36) Logical

Memory I & II

Visual memory: Rey Complex Figure Recall

Visuospatial processing: Rey Complex Figure Copy 37; Visual Object and Space Perception

Battery 38 Screening, Dot Counting and Position Discrimination subtests.

Attention and executive function: WAIS-IV Digit Span subtest; Letter fluency 39; Delis-Kaplan

Executive Function System 40 Colour-Word Interference subtest / The Stroop Task 41

Psychomotor speed: Symbol Digit Modalities Test 42

Semantic fluency: Category Fluency 39

Mood: Hospital Anxiety and Depression Scale 43

Stimulation settings

Before implantation of the pulse generator, the effect of stimulation was trialed for a week with the

following settings: 4 to 6.5 V, 130 Hz and 450 µs. The deepest contact was used as the cathode, and

the most superficial as the anode.

Results

Patients

As presented in Figure 2, 24 patients were treated: 19 males and 5 females, with a mean age of 49.1

years [21-72]. Pain etiologies included failed back surgery syndrome (6 patients), post stroke pain (9

patients), brachial plexus injury (3 patients), spinal cord injury (2 patients), head injury (1 patient),

road traffic accident (2 patients), and pain of unknown cause (1 patient). Pain was located in the whole

body (4 patients), hemi-body (6 patients), back (3 patients), chest (2 patients), arms/hands (7 patients),

or legs/feet (3 patients). The overall mean NRS score was 8.2 (range 5-10) before surgery.

Figure 2: Patients Demographic Characteristics and Preoperative Pain Data

NRS: Numerical Rating Scale

6

Date

A m 49 9.0Bilateral

PVG Back FBSS Feb-07 Both / IPG Removed No pain relief Mar-09

B m 49 6.7 / Whole body FBSS Jan-09 Both / DBS ON / Jan-11

C m 50 6.0Bilateral

PVG Right hemibody Post Stroke Apr-09 Left Infection DBS Removed No pain relief Aug-09

D m 72 8.0 / Right hemibody Post Stroke Jun-09 Left /IPG Not

implantedNo pain relief before IPG

implantation Jun-09

E f 42 8.0Bilateral

PVG-VPL Right Leg Post Stroke Jul-09 Both / DBS RemovedUnable to comply with

treatment Sep-09

Jul-10 Both Infection DBS Removed Feb-12Feb-13 Both Infection DBS Removed No pain relief Mar-13

G m 51 8.0 / Right leg FBSS Oct-10 Both Right wire Broken DBS ON / Mar-16

H m 50 9.0Bilateral

PVG-VPL Left arm Brachial plexus injury Apr-11 Right Infection DBS Removed Infection Oct-11

I m 33 10.0Bilateral

PVG-VPL Left arm Brachial plexus injury Sep-11 Both / DBS Removed Insufficient pain relief May-12

J m 52 10.0Bilateral

PVG-VPL Chest Unknown Nov-11 Both / DBS Removed Insufficient pain relief May-12

K f 46 10.0 / Whole spine FBSS Mar-12 Both Seizures DBS OFF May-16

L f 58 9.0 Bilateral VPL Right hemibody Head injury May-12 Both /IPG Not

implantedNo pain relief before IPG

implantation /

M m 53 8.0 / Right hand Post Stroke May-12 Both Infection + Eczema DBS Removed / May-14

N m 49 10.0 / Right hemibody FBSS Jun-12 Both Seizures DBS ON / Jan-16

O m 63 9.0 / Right hemibody Post Stroke Sep-12 Both / DBS ON / Jun-16

P m 46 6.7Bilateral

PVG-VPL Right arm Brachial plexus injury Nov-12 Both / DBS ON / Jun-16

Q m 58 6.0 / Left arm, chest Post Stroke Feb-13 Both Seizures DBS ONPain free for 1 year before pain

returned Jun-16

R m 48 5.0 Right PVG Right hemibody Post Stroke Mar-13 Both / DBS ON / May-16

S m 71 6.0 / Left arm Post Stroke Mar-13 Both Absence seizures DBS OFF / Dec-15

T m 56 10.0 / Face Post Stroke Jul-13 Both / DBS OFFGood pain relief but unable to

manage device May-16

U m 39 9.0 / Whole body FBSS Nov-13 Both / DBS ON Remains in severe pain Jul-16

V f 21 8.0 / Back and leg RTA Nov-13 Both / DBS ON / Jun-16

W f 37 7.0 / Whole body RTA Nov-14 Both1 lead broken during

stage 2 DBS Removed Infection Dec-14

X m 42 7.0 / Whole body SCI Nov-14 Both / DBS OFF / Dec-15

Comments

SCI - TetraplegiaBoth arms/8.344mF

Date of surgery Complications

Follow-up course

Location Origin Current StatePatient Sex Age at

surgeryNRS score

Pre OpPrevious

DBSPain Side

of electrode(s)

Table 1 (supplemental data): Patients Demographic Characteristics and Preoperative Pain Data

RTA, Road Traffic Accident; FBSS, Failed Back Surgery Syndrome; SCI, Spinal Cord Injury; PVG,

Peri-Ventricular Grey area; VPL, Ventral Postero-Lateral nucleus of the thalamus.

Outcome

Of the 24 patients treated, 22 (91.6%) went on to full implantation after the trial week. Only patients D

and L were not fully implanted as they did not report any pain relief during the trial week. Among the

fully implanted patients, 10 patients, mostly from the earliest cohort, did not have follow up data for

the following reasons; DBS removed for infection (C, F, H, W); diminished pain relief after a trial

period comprised between 6 and 25 months (J, I, A); further strokes rendering the implant ineffective,

and two refused to participate in the study (S, X).

7

Mean follow-up time was 38.9 months (range 24-65) for the 12 patients whose results were included

for analysis. Table 2 summarizes their pain and quality-of-life.

PatientDBS

statusFollow-up,

mo VAS Sensory Affective Evaluative Misc.MPQ Total PF RP BP GH VT SF RE MH

SF-36 Total EQ-5d

Health State

B / Pre-op 6.7 28 1 4 10 43 15 28 29 28 49 19 24 50 242 5 20ON 6 3.0 25 0 0 0 25 15 28 33 20 51 19 24 57 247 5 50ON 12 5.0 N/A N/A N/A N/A N/A 19 28 33 24 51 25 24 60 264 N/A N/A

G / Pre-op 8.0 6 6 5 10 27 15 28 25 22 23 14 24 21 172 8 15ON 6 8.0 23 1 3 5 32 19 28 25 34 33 19 24 39 220 7 25ON 12 7.0 17 9 4 2 32 15 28 29 37 33 25 24 37 227 6 25ON 24 8.0 19 3 3 11 36 15 28 25 27 23 14 24 21 176 7 25ON 36 8.0 15 5 4 11 35 15 28 29 29 28 19 24 32 204 7 20ON 60 8.0 19 9 5 12 53 17 35 20 31 23 14 34 12 186 8 40

K / Pre-op 10.0 22 4 5 5 36 15 28 20 55 33 25 55 35 265 6 77ON 6 3.0 15 1 1 4 21 49 35 33 53 49 46 45 48 359 4 78ON 12 3.5 6 0 3 2 11 28 38 56 51 46 34 60 355 1 95ON 24 5.5 17 3 4 7 31 45 28 38 58 49 35 55 50 358 2 83OFF 36 9.5 9 2 4 2 17 26 28 20 39 40 14 55 37 258 5 10OFF 42 10.0 33 14 5 12 64 17 28 20 29 25 14 24 37 194 8 5

M / Pre-op 8.0 18 1 5 7 31 24 28 20 20 40 30 24 39 223 6 20ON 6 10.0 12 3 5 8 28 22 28 20 17 30 25 24 21 186 6 20ON 12 17 3 5 6 31 22 28 20 17 30 19 24 23 183 6 10

N / Pre-op 10.0 6 4 4 6 20 17 56 24 38 37 30 55 37 295 7 90ON 6 0.0 1 1 0 3 5 17 56 63 44 44 46 55 50 376 4 60ON 12 0.0 0 0 0 0 0 22 28 63 37 47 35 24 48 303 4 90ON 36 8.0 15 7 4 9 35 19 28 25 31 51 14 24 50 243 7 40

O / Pre-op 9.0 19 6 5 10 40 15 28 20 27 23 19 24 30 185 6 20ON 6 5.0 11 0 3 3 17 24 28 25 29 25 25 24 35 214 6 15ON 12 8.0 25 2 4 11 42 24 28 25 20 25 25 24 44 214 6 20ON 36 9.0 22 9 5 7 43 22 28 20 17 23 19 24 21 173 7 10ON 42 8.0 26 7 5 6 44 17 35 20 24 28 19 24 39 206 7 10

P / Pre-op 6.7 29 0 1 5 35 22 28 38 55 63 52 55 50 363 6 100ON 6 8.0 18 17 0 9 44 39 28 38 45 42 35 24 35 285 5 70ON 12 9.5 33 8 4 17 62 36 28 20 45 30 25 24 10 217 6 20ON 36 9.5 42 12 4 14 72 32 28 24 49 30 19 24 19 224 6 95ON 42 10.0 30 9 5 10 54 30 28 20 46 33 19 24 28 227 7 90

Q / Pre-op 6.0 5 3 6 9 23 17 28 20 37 23 14 24 32 195 7 30ON 6 1.0 0 0 0 0 0 36 28 25 42 35 14 24 35 238 2 70ON 12 3.0 5 2 2 5 14 28 35 38 41 30 25 55 35 285 6 60ON 36 3.0 1 0 2 0 3 19 28 38 29 33 25 24 26 220 4 40ON 42 3.0 4 1 2 0 7 24 28 33 36 47 25 24 44 259 5 52

R / Pre-op 5.0 15 16 8 18 57 28 28 29 42 33 35 55 35 284 5 25ON 6 1.0 14 8 2 6 30 34 28 20 49 30 25 24 26 234 4 15ON 12 3.0 32 6 3 8 49 30 28 38 46 40 30 34 37 282 4 35ON 36 6.0 22 5 0 0 27 40 28 33 39 33 19 24 35 251 4 50ON 42 7.0 7 31 13 3 9 N/AN/AN/AN/AN/AN/AN/AN/A N/A N/A 50

T / Pre-op 10.0 30 4 5 7 46 15 28 20 44 30 14 34 30 215 9 20ON 6 0.0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/AOFF 24 8.0 N/A N/A N/A N/A N/A 15 20 17 42 19 45 44 202 8 N/AOFF 36 8.0 28 8 4 14 54 15 28 29 22 40 25 24 37 219 5 0

U / Pre-op 9.0 32 0 5 8 45 34 28 20 62 56 25 55 57 337 6 30ON 6 8.0 18 4 5 6 33 43 28 29 31 59 25 34 53 301 5 40ON 24 8.0 18 2 4 8 32 26 28 24 56 59 14 34 57 297 6 80ON 36 8.5 18 7 4 12 41 28 28 20 46 40 14 24 48 247 5 50

V / Pre-op 8.0 32 9 5 13 57 15 28 20 37 47 19 24 23 213 9 N/AON 24 7.0 25 4 3 4 36 15 28 25 22 28 14 24 16 172 8 35ON 36 5.0 16 9 3 3 31 15 42 29 31 44 19 55 52 289 7 35

Table 2: Pre- and postoperative data: individual pain assessment scores of each patient

preoperatively and throughout follow-up

NRS, Numerical Rating Scale; Misc., miscellaneous; MPQ, McGill Pain Questionnaire; PF, physical

functioning; RP, role-physical; BP, bodily pain; GH, general health; VT, vitality; SF, social

functioning; RE, role emotional; MH, mental health; Health State self-reported preoperatively and at

follow-up. N/A, not available. Scores in grey font are scores when stimulator is OFF.

8

Pain relief

As shown in Figure 3, six months post-surgery, mean NRS dropped significantly from 8 (range 5-10)

to 4.3 (mean -45%, range -25% to +100%, P = .004). A significant improvement of MPQ was

observed (mean -36%, range -26% to +100%, P = .021), and particularly on the evaluative component

(mean -62%, range +0% to +100%, P < .001). Moreover, EQ-5D scores significantly decreased (mean

-21%, range +0% to +71%, P = .036).

A statistically significant improvement of the PF domain of SF-36 quality of life survey was observed

(mean +54.2%, range -8% to +227%, P = .01).

Among the 12 patients with available follow-up data, 10 patients (83.3%) reported substantial pain

relief (all but patients M and P). In these patients, NRS was improved by 60.3% at 6 months (range,

0% to -100%; P < .001) and by 43.4% at 1 year (range, -11% to -100%; P < .01). MPQ total score

decreased by 46.5% at 6 months (range, +19% to -100%; P < .01), particularly the Evaluative

subscore at 6 months (mean -65%; range, 0% to -100%; P <0.001) and 1 year (mean -46.7%; range, -

20% to -100%; P = .017) and the Miscellaneous subscore (mean -58.6%; range, -20% to -100%; P

= .014) subscores. Statistically significant improvements were also observed for the PF (Physical

Functioning) at 6 months and the BP (bodily pain) domains of the SF-36 quality of life survey 1 year

post-operatively: mean +59 (range, 0% to 227%; P = .035) and mean +61% (range, 14% to 163%; P =

.014), respectively. EQ-5D score was also significantly reduced after 1 year (mean -30.8; range, 0 to -

83%; P = .05). At longer follow-up there was some loss of efficacy. Nevertheless, pain relief was

sustained up to 24 months (patient K), 36 months (patient V) and 42 months (patient Q), with a NRS

score as low as 3. Furthermore, in two patients stimulation had to be turned OFF because of seizure

occurrence despite pain relief.

9

Figure 3: Anterior cingulate cortex DBS Patients’ outcomes

A. NRS (Numerical Rating Scale), B. SF-36 (Short-Form-36), C. MPQ (McGill Pain Questionnaire),

D. EQ-5D (EuroQol questionnaire), E. Health State self-reported, before (pre-op) and after surgery

(from 5 to 36 months post-surgery).

10

Aetiologies

Figure 4 shows patients’ distribution according to their pain aetiology. Only FBSS and Stroke

groups had enough patients with available follow-up to run reliable analyses and statistics.

FBSS patients’ mean NRS significantly dropped from 8.7 to 4.4 (p=0.025) 6 months post-

operatively. The Evaluative and Miscellaneous components of the MPQ score, were

significantly improved at 6 months and 1 year respectively, decreasing from 4.6 to 1.8

(p=0.02) and 7.8 to 1.3 (p=0.008). One year post-operatively, the MPQ total score went from

34.2 to 14.3 (p=0.063) and the EQ-5D from 6.4 to 3.7 (p=0.061). Regarding the Stroke

patients, MPQ Total score 39.4 to 34.0 (p=0.028) 6 months after surgery.

Comparing both groups, the FBSS one was significantly better than the Stroke one for 3

subscores of the SF-36 test, namely BP (p=0.032), VT (p=0.032) and MH (p=0.016) with SF-

36 total scores being respectively 300.8 and 217.9 (p=0.063).

Figure 4: Patients’ aetiology and post-operative data

NRS, Numerical Rating Scale; FBSS, Failed Back Surgery Syndrome; SCI, Spinal Cord Injury; BPI,

Brachial Plexus Injury

Affective component of pain

As previously reported28, several patients describe the ACC DBS induced pain relief as a decrease of

the perceived unpleasantness of pain, more than a reduction of the sensation itself. They described

their pain as being separate to them and stated that when the stimulator is on, their pain is “not

distressing”, "not particularly bothersome" or again "not worrying anymore".

Cognitive Outcomes following Surgery

Patients’ cognitive outcome are summarized in Table 3. No consistent pattern of decline or

improvement emerged from comparison of individual patient’s pre and post-operative test scores. For

the majority of patients their postoperative test scores were consistent with those obtained during

preoperative testing. It was remarkable that a number of patients presented with improvements in

selected aspects of cognitive function. For example, patient O’s and T’s verbal memory scores were

stronger following surgery. Patients G, O and U demonstrated improved performance on the Stroop

task (a measure of verbal inhibition) whilst M and R achieved stronger semantic fluency scores.

Based on a review of SF-36 scores and of clinical history taken at postoperative assessment, there did

11

not appear to be any association between changes (positive or negative) in cognitive function and

quality of life.

Side-effects

Despite the efficacy in pain relief, recurrent stereotyped events occurred in several patients on long

term follow up (Figure 4). Following review of medical notes and telephone interviews using a

validated questionnaire, 6 patients were identified to have suffered recurrent stereotyped events

suggestive of seizures. In one, events were felt to relate to alcohol withdrawal, and specialist review

was declined by the patient. In another, events had ceased following explantation (following

infection), and follow up has been declined by the patient. Four patients were identified to be suffering

ongoing events, and assessed by DBS Neurology service with clinical evaluation and EEG video

telemetry in 2015. Two patients had stimulation induced seizures, ceasing with cessation of

stimulation and resultant loss of pain relief. Nevertheless, 2 patients developed de novo stimulation

induced epilepsy (seizures continue despite cessation of stimulation). This phenomenon, unseen before

with DBS, but observed with anterior cingulotomy26, is under investigation.

12

Patient Time to Post ACC DBS Testing

Reasoning Verbal Memory

Visual Memory

Visuospatial Attention / Executive Function

Psychomotor Speed

Semantic Fluency

Mood

G 54 ↓(FAS)↑ (stroop)

M 7 ↓↓ (stroop) ↓ ↑N 15 ↑↑O 23 ↑ ↑ (stroop) ↓Q 12 ↑ ↓ ↓R 37 ↑ (FAS) ↑↑ ↑T 13 ↑U 14 ↑↑ ↑↑ (stroop) ↓

Overall ↓ 0/8 0/8 0/8 0/8 1/8 Stroop1/8 FAS

2/8 3/8 0/8

Overall ↑ 1/8 2/8 1/8 0/8 3/8 Stroop1/8 FAS

1/8 2/8 1/8

Table 3. Summary of neuropsychological outcomes following surgery. ↑ = mild improvement; ↑↑ = significant improvement; ↓ = mild decline; ↓↓ = significant decline.

13

DiscussionBackground and rationale

Chronic pain is an integrated experience, with sensory, affective, and cognitive aspects. The Anterior

Cingulate Cortex (ACC) has been demonstrated to be a key structure of the affective component14.

Anatomically, the ACC is part of the cingulate gyrus and consists of Brodmann’s areas 24, 25, 32 and

33. Electrodes of this case-series were implanted in Brodmann’s area 24 (dACC). It contains a

prominent layer V and a thin layer IV. Several tracts connect the ACC to the limbic system (amygdala,

insula, hypothalamus, ventral striatum and ventromedial prefrontal cortex), to the motor system

(premotor and motor cortices, spinal cord) and to cognition involved areas (cingulate motor areas in

the cingulate sulcus and nociceptive cortex)15. It receives thalamic inputs from the midline nuclei

(anteromedial, paraventricular, parataenial, paracentral, central and centrolateral, reuniens,

parafascicular, limitans, mediodorsal, and ventral anterior nuclei).

Even though they are part of the same structure, the Anterior and Posterior Cingulate Cortex (PCC)

present very different cytoarchitectures, connections, and functions. A dichotomy of the cingulate

gyrus has been proposed by Vogt 20.

As part of the brain’s limbic system44, lesions in the ACC are known to induce emotional instability,

apathy, and akinetic mutism45, due to its role in many emotional, cognitive and motor functions15, 45-47.

Many studies in animals also demonstrate the role played by ACC in pain 48-51. The obvious link

between ACC and pain led Foltz and White to perform cingulate lesions by electrocoagulation to

relieve intractable pain23, 25. They described the effect of cingulotomy as following:

‘The perception of pain as such does not appear to be modified, but the patient’s total reaction to pain

and the threat to existence that it represents is modified markedly. Most of the patients stated that they

continued to have pain but it was ‘not particularly bothersome,’ ‘doesn’t worry me…’ 25.

Similarly, as previously mentioned in our first case-series paper, some patients described that they felt

as if their pain was separate from them physically and that they did not think about it anymore,

although they could still sense it28.

A question arises: is the effect of cingulotomy on responses to noxious stimuli due to disruption of

axons of passage? Positron emission tomography (PET) and fMRI studies showed activation of the

ACC in acute or chronic painful situations52-55 Interestingly, PET and magnetoencephalography (MEG)

studies showed activity in the rostral and dorsal ACC on pain patients treated using PAG, thalamic or

dual PAG and VP DBS56-58. More recently, we presented long-term changes in ACC activity in

addition to PAG activation with ACC DBS59.

The rationale for stimulating the ACC in the case-series presented here arises both from the above

cited neuroimaging studies and from historical series of cingulotomy to relieve cancer pain. In 2007

Spooner and colleagues reported the first dorsal ACC DBS to relieve a patient with cervical spinal

cord injury suffering from whole body pain, but no long-term follow-up was recorded as the patient

died one year post-surgery60 .

14

Neuromodulation of the ACC and surrounding areas has been performed to treat other disorders,

primarily psychiatric and affective ones. The subcallosal cingulate was targeted for OCD and

depression61, 62, and DBS of the anterior limb of internal capsule (ALIC) was FDA approved for OCD.

Furthermore, in central post stroke pain, the stimulation of the ALIC/Nucleus Accumbens has been

described as a potential target for the affective component63.

In our case-series, electrodes were implanted mostly in the white matter, 20 mm posterior to the

anterior tip of the frontal horns of the lateral ventricles, with the deepest contacts in the corpus

callosum and the upper electrodes in the cingulum bundle. Our patients preferred stimulation settings

with the deepest contacts as active during the trial week where many settings were tried, suggesting a

role of the corpus callosum in the relief of pain. Another interesting observation was that it usually

took a few days for any clinical improvement to be observed in this cohort, in contrast with PAG or

VP DBS where the relief was achieved instantaneously6. With increased experience patient selection

tended towards those without particular pain localization, such as stroke pain patients and those with

significantly impaired quality of life.

The variability of efficacy between patients, led to a tractography study to investigate the difference in

connectivity between good and bad outcome patients, and found on a small cohort in whom strong

connectivity to the precuneus was correlated with an unsuccessful outcome64. This result, if confirmed

on a larger cohort, may help in defining the best electrode placement in future patients.

Analysis of neuropsychometric data did not identify any consistent patterns of cognitive change

following ACC DBS. There was some variability between patients, with some demonstrating

improvements in areas such as verbal memory or executive function, yet others demonstrating declines

in category fluency or psychomotor speed. Previous reports had indicated that DBS may be associated

with reductions in letter and semantic fluency (Gray et al) yet but this effect was not observed here. It

was notable that two patients demonstrated improvements in verbal fluency. Consistent with the Gray

et al. study cognitive changes did not appear to have had an appreciable impact on quality of life.

Evaluation of pain relief

Pain has a major impact on patients’ life inducing disability, employment issues as well as social and

family isolation. Evaluating patients’ psychological factors and quality of life is therefore more

informative than the NRS to estimate the benefits of DBS. Improvement of quality of life induced by a

reduction of pain was previously shown by Gray et al. in a longitudinal study investigating DBS for

neuropathic pain65. Stimulation of conventional targets of DBS for pain e.g. thalamus and PVG

reduces the intensity of pain. In contrast, stimulating the ACC aims to decrease its affective component

and most of the patients accordingly stated they continued to have pain but it was "not distressing,"

"not particularly bothersome," "not worrying anymore," etc. Yet, pain assessments fail at adequately

measuring this effect on the affective component. Pain is a very personal experience, and every patient

15

is different, but standard pain assessments are not ideal to describe or measure the pain intensity or the

way it is sensed. For instance, when asking to rank their pain for the NRS, some patients may take into

account the pain intensity, and some other ones the induced distress only. Also, there is a substantial

difference between the results of the assessment forms and what patients verbally report to clinicians.

This may result from several factors notably the difference between what patients feel, what they say

during clinical appointments, and what they write while on their own. Fear of losing their disability

benefits may lead to some filling their forms inaccurately, ranking their pain and quality of life much

worse than in reality. Some others may feel that they do not want to disappoint the care team and

pretend that the surgery was more effective than in reality, yet feel free to report it more accurately in

the forms. Moreover, each patient’s experience of pain is different and averaging scores may not

reflect the effectiveness of the procedure. In the initial case-series, as we had fewer patients, we

decided to present individual results, with each patient as a case-report.

Key results

All patients but two transitioned from externalized to fully internalized systems. Mean pre-surgery

pain as reported on the NRS was 8 and dropped below 4 for six patients, with two patients rendered

pain free. A significant improvement in quality of life (MPQ and EQ-5D) was observed. Moreover,

statistically significant improvements were gained in some domains of SF-36 quality of life. Long-

term efficacy was sustained up to 42 months in some patients.

Conclusion

Patients tend to describe that some pain is still present, but that it is less bothersome. Consequently,

little or no improvement of the pain NRS score does not necessarily equate to poor therapeutic

efficacy. Furthermore, as the primary effect is not pain alleviation, patient satisfaction may appear

lower. Nevertheless, a demonstrably improved performance in quality of life was seen in this cohort.

The main limitations are those of an open-label study and randomization would be desirable, as

described in the recent study protocol of DBS for thalamic pain syndrome12. Moreover, pain

assessments targeting the affective component of pain may be preferable to evaluate improvements.

Patients of this study were patients suffering from intractable pain that prevented them from living a

normal life. DBS of the Anterior Cingulate Cortex improved quality of life, by lessening the impact of

pain upon them. We demonstrate here that DBS of the ACC can significantly alleviate the suffering of

patients with otherwise treatment resistant chronic pain. This provides a promising avenue for patients

for whom other treatments including PAG and VP DBS are ineffective8. However, long term

stimulation does appear to have an increased risk of seizures/epilepsy which may relate to the pattern

of stimulation which is subject to current investigation.

Acknowledgments

16

The authors thank Dr Carlo Chiorri, PhD, Lecturer in Psychometrics at the University of Genoa, Italy

(DISFOR, Department of Educational Sciences, Psychology Unit) for assistance in statistical analyses.

The research was supported by the Norman Collisson Foundation, Charles Wolfson Charitable Trust

and EPSRC and by the National Institute for Health Research (NIHR) Oxford Biomedical Research

Centre based at Oxford University Hospitals NHS Trust and University of Oxford. The views

expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department

of Health.

17

Figures legends

Figure1: Views of bilateral electrodes in the Anterior Cingulate Cortex.

A. Preoperative trajectory planning. B. Coronal view on a post-operative fused MR/CT images. C.

Sagittal view on a post-operative fused MR/CT images.

Figure 3: Anterior cingulate cortex DBS Patients’ outcomes

18

A. NRS (Numerical Rating Scale), B. SF-36 (Short-Form-36), C. MPQ (McGill Pain Questionnaire),

D. EQ-5D (EuroQol questionnaire), E. Health State self-reported, before (pre-op) and after surgery

(from 5 to 36 months post-surgery).

19

References

1. IASP. Taxonomy. http://www.iasp-pain.org/. 2014.2. Scholten PM, Harden RN. Assessing and Treating Patients With Neuropathic Pain. PM

& R : the journal of injury, function, and rehabilitation. Nov 2015;7(11 Suppl):S257-269.

3. Eisenberg E, McNicol ED, Carr DB. Efficacy and safety of opioid agonists in the treatment of neuropathic pain of nonmalignant origin: systematic review and meta-analysis of randomized controlled trials. JAMA. Jun 22 2005;293(24):3043-3052.

4. Pizzo PA, Clark NM. Relieving Pain in America: A Blueprint for Transforming Prevention, Care, Education, and Research. Washington (DC)2011.

5. Heath RG. Psychiatry. Annu Rev Med. 1954;5:223-236.6. Boccard SG, Pereira EA, Moir L, Aziz TZ, Green AL. Long-term outcomes of deep

brain stimulation for neuropathic pain. Neurosurgery. Feb 2013;72(2):221-230; discussion 231.

7. Kumar K, Toth C, Nath RK. Deep brain stimulation for intractable pain: a 15-year experience. Neurosurgery. Apr 1997;40(4):736-746; discussion 746-737.

8. Levy RM. Deep brain stimulation for the treatment of intractable pain. Neurosurg Clin N Am. Jul 2003;14(3):389-399, vi.

9. Boccard SG, Pereira EA, Aziz TZ. Deep brain stimulation for chronic pain. J Clin Neurosci. Oct 2015;22(10):1537-1543.

10. Hosobuchi Y, Adams JE, Linchitz R. Pain relief by electrical stimulation of the central gray matter in humans and its reversal by naloxone. Science. Jul 8 1977;197(4299):183-186.

11. Mazars G, Merienne L, Ciolocca C. [Intermittent analgesic thalamic stimulation. Preliminary note]. Rev Neurol (Paris). Apr 1973;128(4):273-279.

12. Pereira EA, Boccard SG, Linhares P, et al. Thalamic deep brain stimulation for neuropathic pain after amputation or brachial plexus avulsion. Neurosurg Focus. Sep 2013;35(3):E7.

13. Melzack R. From the gate to the neuromatrix. Pain. Aug 1999;Suppl 6:S121-126.14. Albe-Fessard D, Berkley KJ, Kruger L, Ralston HJ, 3rd, Willis WD, Jr. Diencephalic

mechanisms of pain sensation. Brain Res. Aug 1985;356(3):217-296.15. Devinsky O, Morrell MJ, Vogt BA. Contributions of anterior cingulate cortex to

behaviour. Brain. Feb 1995;118 ( Pt 1):279-306.16. Iwata K, Kamo H, Ogawa A, et al. Anterior cingulate cortical neuronal activity during

perception of noxious thermal stimuli in monkeys. J Neurophysiol. Sep 2005;94(3):1980-1991.

17. Koyama T, Kato K, Mikami A. During pain-avoidance neurons activated in the macaque anterior cingulate and caudate. Neurosci Lett. Mar 31 2000;283(1):17-20.

18. LaGraize SC, Labuda CJ, Rutledge MA, Jackson RL, Fuchs PN. Differential effect of anterior cingulate cortex lesion on mechanical hypersensitivity and escape/avoidance behavior in an animal model of neuropathic pain. Exp Neurol. Jul 2004;188(1):139-148.

19. Lenz FA, Rios M, Zirh A, Chau D, Krauss G, Lesser RP. Painful stimuli evoke potentials recorded over the human anterior cingulate gyrus. J Neurophysiol. Apr 1998;79(4):2231-2234.

20. Vogt BA. Pain and emotion interactions in subregions of the cingulate gyrus. Nat Rev Neurosci. Jul 2005;6(7):533-544.

20

21. Singer T, Seymour B, O'Doherty J, Kaube H, Dolan RJ, Frith CD. Empathy for pain involves the affective but not sensory components of pain. Science. Feb 20 2004;303(5661):1157-1162.

22. Whitty CW, Duffield JE, Tov PM, Cairns H. Anterior cingulectomy in the treatment of mental disease. Lancet. Mar 8 1952;1(6706):475-481.

23. Ballantine HT, Jr., Cassidy WL, Flanagan NB, Marino R, Jr. Stereotaxic anterior cingulotomy for neuropsychiatric illness and intractable pain. J Neurosurg. May 1967;26(5):488-495.

24. Viswanathan A, Harsh V, Pereira EA, Aziz TZ. Cingulotomy for medically refractory cancer pain. Neurosurg Focus. Sep 2013;35(3):E1.

25. Foltz EL, White LE, Jr. Pain "relief" by frontal cingulumotomy. J Neurosurg. Feb 1962;19:89-100.

26. Wilkinson HA, Davidson KM, Davidson RI. Bilateral anterior cingulotomy for chronic noncancer pain. Neurosurgery. Nov 1999;45(5):1129-1134; discussion 1134-1126.

27. Boccard SG, Pereira EA, Moir L, et al. Deep brain stimulation of the anterior cingulate cortex: targeting the affective component of chronic pain. NeuroReport. Jan 22 2014a;25(2):83-88.

28. Boccard SG, Fitzgerald JJ, Pereira EA, et al. Targeting the Affective Component of Chronic Pain: A Case Series of Deep Brain Stimulation of the Anterior Cingulate Cortex. Neurosurgery. Apr 15 2014b.

29. Melzack R. The McGill Pain Questionnaire: major properties and scoring methods. Pain. Sep 1975;1(3):277-299.

30. SF-36. SF-36® Health Survey Scoring Demonstration website. http://www.sf-36.org/demos/sf-36.html. 2011, 2012.

31. Kind P, Dolan P, Gudex C, Williams A. Variations in population health status: results from a United Kingdom national questionnaire survey. BMJ. Mar 07 1998;316(7133):736-741.

32. Wechsler D. The Wechsler Adult Intelligence Scale. Revised (WAIS-R) ed. New York, USA: Manual. Psychological Corporation; 1981.

33. Wechsler D. The Wechsler Adult Intelligence Scale – Fourth Edition. Fourth ed. London, UK: Pearson Assessment; 2008.

34. Coughlan AKH, S. E. The Adult Memory and Information Processing Test Battery Manual. Leeds, UK: A. K Coughlan / St James’ University Hospital; 1985.

35. Delis DC, Kaplan, E., Kramer, J.H. & Ober, B.A. California verbal learning test. Second ed. San Antonio, USA: The Psychological Corporation; 2000.

36. Wechsler D. The Wechsler Memory Scale. Fourth ed. London, UK: Pearson Assessment; 2009.

37. Meyers JEM, K. R. Rey Complex Figure Test and Recognition Trial. Odessa, USA: Psychological Assessment Resources Inc; 1995.

38. Corporation TVT. The Visual Object and Space Perception Battery. Bury St Edmonds, UK: TVTC; 1991.

39. Strauss E, Sherman, E.M.S. & Spreen, O. A compendium of neuropsychological tests. Third ed. Oxford, UK: Oxford University Press; 2006.

40. Delis DC, Kaplan, E., & Kramer, J.H. Delis Kaplan Executive Function System. San Antonio, USA: The Psychological Corporation; 2001.

41. Trenerry MR, Crosson, B., DeBoe, J., & Leber, W. R. Stroop: Neuropsychological screening test manual. Odessa, USA: Psychological Assessment Resources Inc; 1989.

42. Smith A. Symbol Digit Modalities Test: Manual. Los Angeles, USA: Western Psychological Service; 1983.

21

43. Zigmund AS, & Snaith, R.P. The hospital anxiety and depression scale. Acta Psychiatrica Scandinavica. 1983;67:361 – 370.

44. Papez JW. A proposed mechanism of emotion. Arch. Neurol. Psychiatry. 1937;38:725–743.

45. Paus T. Primate anterior cingulate cortex: where motor control, drive and cognition interface. Nat Rev Neurosci. Jun 2001;2(6):417-424.

46. Matsumoto K, Tanaka K. The role of the medial prefrontal cortex in achieving goals. Curr Opin Neurobiol. Apr 2004;14(2):178-185.

47. Rushworth MF, Noonan MP, Boorman ED, Walton ME, Behrens TE. Frontal cortex and reward-guided learning and decision-making. Neuron. Jun 23 2011;70(6):1054-1069.

48. Vogt BA, Rosene DL, Pandya DN. Thalamic and cortical afferents differentiate anterior from posterior cingulate cortex in the monkey. Science. Apr 13 1979;204(4389):205-207.

49. Vogt LJ, Vogt BA, Sikes RW. Limbic thalamus in rabbit: architecture, projections to cingulate cortex and distribution of muscarinic acetylcholine, GABAA, and opioid receptors. J Comp Neurol. May 8 1992;319(2):205-217.

50. Gabriel M, Sparenborg S, Kubota Y. Anterior and medial thalamic lesions, discriminative avoidance learning, and cingulate cortical neuronal activity in rabbits. Exp Brain Res. 1989;76(2):441-457.

51. Gabriel M, Kubota Y, Sparenborg S, Straube K, Vogt BA. Effects of cingulate cortical lesions on avoidance learning and training-induced unit activity in rabbits. Exp Brain Res. 1991;86(3):585-600.

52. Jones AK, Brown WD, Friston KJ, Qi LY, Frackowiak RS. Cortical and subcortical localization of response to pain in man using positron emission tomography. Proc Biol Sci. Apr 22 1991;244(1309):39-44.

53. Talbot JD, Marrett S, Evans AC, Meyer E, Bushnell MC, Duncan GH. Multiple representations of pain in human cerebral cortex. Science. Mar 15 1991;251(4999):1355-1358.

54. Casey KL, Minoshima S, Berger KL, Koeppe RA, Morrow TJ, Frey KA. Positron emission tomographic analysis of cerebral structures activated specifically by repetitive noxious heat stimuli. J Neurophysiol. Feb 1994;71(2):802-807.

55. Derbyshire SW, Jones AK, Devani P, et al. Cerebral responses to pain in patients with atypical facial pain measured by positron emission tomography. J Neurol Neurosurg Psychiatry. Oct 1994;57(10):1166-1172.

56. Pereira EA, Green AL, Bradley KM, et al. Regional cerebral perfusion differences between periventricular grey, thalamic and dual target deep brain stimulation for chronic neuropathic pain. Stereotact Funct Neurosurg. 2007;85(4):175-183.

57. Ray N, Kringelbach M, Jenkinson N, et al. Using magnetoencephalography to investigate brain activity during high frequency deep brain stimulation in a cluster headache patient. Biomed Imaging Interv J. Jan 2007;3(1):e25.

58. Kringelbach ML, Jenkinson N, Green AL, et al. Deep brain stimulation for chronic pain investigated with magnetoencephalography. NeuroReport. Feb 12 2007;18(3):223-228.

59. Mohseni HR, Smith PP, Parsons CE, et al. MEG can map short and long-term changes in brain activity following deep brain stimulation for chronic pain. PLoS One. 2012;7(6):e37993.

60. Spooner J, Yu H, Kao C, Sillay K, Konrad P. Neuromodulation of the cingulum for neuropathic pain after spinal cord injury. Case report. J Neurosurg. Jul 2007;107(1):169-172.

22

61. Lozano AM, Mayberg HS, Giacobbe P, Hamani C, Craddock RC, Kennedy SH. Subcallosal cingulate gyrus deep brain stimulation for treatment-resistant depression. Biol Psychiatry. Sep 15 2008;64(6):461-467.

62. Mayberg HS, Lozano AM, Voon V, et al. Deep brain stimulation for treatment-resistant depression. Neuron. Mar 3 2005;45(5):651-660.

63. Mallory GW, Abulseoud O, Hwang SC, et al. The nucleus accumbens as a potential target for central poststroke pain. Mayo Clin Proc. Oct 2012;87(10):1025-1031.

64. Boccard SG, Fernandes H, Jbabdi S, et al. A tractography study of Deep Brain Stimulation of the Anterior Cingulate Cortex in chronic pain: a key to improve the targeting. World Neurosurg. Sep 3 2015.

65. Gray AM, Pounds-Cornish E, Eccles FJ, Aziz TZ, Green AL, Scott RB. Deep brain stimulation as a treatment for neuropathic pain: a longitudinal study addressing neuropsychological outcomes. J Pain. Mar 2014;15(3):283-292.

23