Handbook of Clinical Neurology, Vol.

116 (3rd series)

Chapter 9

Deep brain stimulation for Parkinson’s disease – patient selection

INTRODUCTION DBS was proposed for these targets with the idea of mim-
icking pallidotomy or subthalamotomy, known to improve
About a third of deep brain stimulation (DBS) failures
the main parkinsonian motor features in humans and mon-
can be ascribed to an inappropriate indication for surgery
keys treated with 1-methyl-4-phenyl-1,2,3,6-tetrahydro-
(Okun et al., 2005). Thus, careful patient selection is a
pyridine (MPTP) respectively. Patient with early-onset
crucial determinant of consistently favorable outcomes
PD who show a good response to levodopa but develop
following DBS (Bronstein et al., 2011). Patient selection
intractable motor complications would represent the most
relies on the individual risk–benefit ratio, which is related suitable candidates for DBS. However, patients with PD
to the individual features of Parkinson’s disease (PD) and
suffer from symptoms related to lesions outside the sub-
neurosurgical contraindications. Given the complexity
stantia nigra, which are partially responsive or unrespon-
of symptoms in most patients with advanced PD, the
sive to dopaminergic drugs and are responsible for the
selection process should be implemented in specific
progressive symptoms that compromise quality of life.
centers, and should include a movement disorder
Therefore, an important part of the selection process
neurologist, a neurosurgeon, a neuropsychologist, a psy-
for DBS is to evaluate to which extent these levodopa-
chiatrist, a neuroradiologist, and, in some cases, a neuro-
unresponsive symptoms might jeopardize the outcome
physiologist, an internist, and a PD nurse – all experts in of DBS surgery.
DBS. A computerized DBS screening module and elec-
tronic decision-making tool (Moro et al., 2009), which
can assist neurologists in deciding which patients with CONTRAINDICATIONS FOR
PD should be referred for DBS evaluation, has been NEUROSURGERY
developed, but the traditional triage by a neurologist spe- Is there an age limit?
cialized in movement disorders remains the gold stan-
dard (Oyama et al., 2012). The major concerns regarding age have been the associ-
ated comorbidities, cognitive dysfunction (Saint-Cyr
et al., 2000), higher incidence of levodopa-resistant symp-
THE IDEAL PARKINSON’S DISEASE
toms such as dysarthria, dysphagia, postural instability,
CANDIDATE FOR DEEP BRAIN
and gait difficulties (Russmann et al., 2004; Derost
STIMULATION
et al., 2007; Ory-Magne et al., 2007), and higher overall
In 1987, high-frequency stimulation was applied within risk of surgical complications (Sansur et al., 2007;
the thalamus to improve tremors of different causes, Vesper et al., 2007; Hu et al., 2010; Merola et al., 2011).
because it mimicked the effect of thalamotomy, with Uncontrolled studies of the influence of age at the time
the added advantages of reversibility and adjustability. of surgery on postoperative outcome after DBS of the
Advancement in the knowledge of basal ganglia patho- STN or GPi have given inconsistent results, although
physiology in the subsequent decade showed that selec- more studies and expert opinion have shown a superior
tive nigrostriatal dopaminergic lesions in animal models outcome in younger patients than in older patients
of PD led to increased activity of the output of the basal (Russmann et al., 2004; Lang et al., 2006; Derost et al.,
ganglia, namely the subthalamic nucleus (STN) and the 2007; Tsai et al., 2009). Unlike STN DBS, age may be less
internal part of the globus pallidus (GPi). Therefore, critical when considering stimulation of the ventral

*Correspondence to: Professor Pierre Pollak, Chef de Service, Service de Neurologie, Hôpitaux Universitaires de Genève, 4 Rue
Gabrielle-Perret-Gentil, 1211 Genève 14, Switzerland. Tel: þ41-22-372 83 02, Fax: þ41-22-372 83 33, E-mail: [email protected]
98 P. POLLAK
intermediate nucleus of the thalamus (Vim) in elderly The second most frequent exclusion was abnormalities
patients suffering mainly from disabling tremor. More- in MRI scans (9%) (Lopiano et al., 2001). MRI may also
over, Vim DBS can be performed unilaterally, and show silent vascular malformations, cortical atrophy,
patients suffering from the tremor-dominant type of hydrocephalus, or large cysts with brain distortion. More
PD often have a more benign disease course. often than not these discoveries lead the neurosurgeon to
The value of age as an independent outcome predic- adapt the implantation trajectory. Severe cortical atro-
tor for DBS has been debated (Lang et al., 2006). phy increases the risk of postoperative subdural hemato-
Although no specific age cutoff has been defined in mas. Such abnormalities, in addition with other negative
many clinical DBS studies, most have excluded patients predictive factors, can bias a decision away from
older than 75 years. At our institution, we tend to con- surgery.
sider age under 75 years to be a relative contraindication
and age above 80 years to be an absolute contraindica- Other presurgical issues
tion, although some rare exceptions to this rule may be
those patients with a biological age that seems younger In the weeks before surgery, the neurologist checks for
than their chronological age. In a recent large controlled the absence of scalp infection and for the use of any
study, one-quarter to one-third of the patients with STN drugs that favor bleeding, such as aspirin and other anti-
or GPi DBS were aged 70 years or more (Weaver et al., platelet drugs, nonsteroidal anti-inflammatory agents
2009; Follett et al., 2010). drugs, anticoagulants, and lisuride. Because of the risk
However, it remains intuitive that elderly patients of intraoperative and postoperative psychosis and con-
are particularly fragile, have less cognitive reserve, a fusion, especially in patients with advanced PD and high
higher incidence of levodopa-resistant symptoms, and doses of antiparkinsonian drugs, it seems wise to sim-
life expectancy is shorter, thereby influencing the plify the pharmacotherapy. Levodopa is favored; minor
risk–benefit ratio. drugs are generally stopped; and the dosages of dopa-
mine agonist drugs are decreased a few days or weeks
Comorbidities before the date of DBS, based on pharmacokinetics. This
can lead to a worsening of PD, which should be explained
All DBS centers exclude patients with serious systemic to and accepted by the patient.
comorbidities, and it is therefore not surprising that
almost no systematic studies are available that specifi-
PREVIOUS THERAPIES FOR
cally correlate complication rates or outcome with pre-
PARKINSON’S DISEASE
surgical comorbidity. One study showed a greater than
10-fold increase in cerebral hemorrhage in patients with Medical therapy
preoperative vascular hypertension (Gorgulho et al.,
The inherent risk of surgery, far greater than that of any
2005), emphasizing the importance of taking into
pharmacological treatment, necessitates that candidates
account general health problems when estimating the
for DBS should have tried the main antiparkinsonian
risk–benefit ratio of surgical candidates.
drugs. A trial of levodopa at the highest tolerated dose
Although formal studies are lacking, serious comor-
(ideally at least 800 mg daily for 3 months), as well as
bidities such as evolving malignancies with markedly
a dopamine agonist, is mandatory before consideration
reduced life expectancy, unstable heart disease,
of surgery. Levodopa is considered the gold standard
disabling cerebrovascular disease, active infection, and
treatment for PD, and in some patients the antiparkinso-
all other serious neurological and medical comorbidities
nian effects may be delayed with the risk of giving up the
that would jeopardize the benefit of DBS or increase sur-
drug before achieving benefit. The main risks of increas-
gical risk should be regarded as contraindications to
ing levodopa dosage are motor fluctuations and worsen-
DBS (Lang et al., 2006).
ing of dyskinesia, whereas those of dopamine agonist
drugs are psychosis and impulse control disorders.
Brain magnetic resonance imaging
Apart from DBS, there are currently two available
Preoperative screening by magnetic resonance imaging strategies for reducing motor complications in patients
(MRI) allows for identification of structural lesions that with advanced PD: continuous subcutaneous apomor-
may increase the risk of surgery or decrease its benefit phine infusion and continuous intestinal levodopa/carbi-
(such as relevant white-matter lesions) and evidence of dopa pump infusion (ILI). Despite very few comparative
atypical parkinsonism that also will likely affect the out- studies that help to select the best treatment for individ-
come. One study evaluated exclusion criteria retrospec- ual patients, the three procedures have a slightly differ-
tively in 98 patients hospitalized for surgical screening. ent risk–benefit profile. DBS is best for younger
Some 30% of the subjects were not suitable for surgery. patients. Patients over 75–80 years are usually not good
 DEEP BRAIN STIMULATION FOR PARKINSON’S DISEASE – PATIENT SELECTION 99
candidates for DBS. DBS carries the risk of worsening FEATURES OF PARKINSON’S DISEASE
dysarthria and freezing of gait or instability. Therefore,
in patients in whom dysarthria or poor postural stability
Disease severity
during on-motor periods is the main cause of disability, There is no consensus on a specific severity level and/or
consideration should be given to infusion procedures, cutoff as a determinant of outcome from DBS. Patients
which are reversible, before DBS. As levodopa is the with the most severe motor score may show dramatic
antiparkinsonian drug least amenable to inducing psy- improvement, provided they had the other clinical cri-
chic adverse effects, ILI is suggested to be the preferred teria for surgery. Moreover, disease severity that leads
procedure for patients with cognitive impairment or to disability is influenced by individual factors, such
psychosis, although device management may be a prob- as professional status and social function, and should
lem. In patients with impulse control disorders, DBS be considered.
(Lhommee et al., 2012) or ILI is preferable because
these approaches allow for dopamine agonist drug
Disease duration and timing of surgery
arrest. Although surgery and infusion therapies are
able to improve severe tremor or dyskinesias, DBS is Disease duration has not been a primary factor in select-
considered by far the most effective treatment for ing patients for DBS. There is currently no evidence of a
drug-refractory tremor, or intractable dyskinesias neuroprotective effect of DBS to provide a rationale for
(Volkmann et al., 2013). surgery before the development of levodopa-induced
Patients need to be advised by an experienced neurol- motor complications. Historically, patients with PD
ogist for all therapies, to include the expected benefits, who undergo DBS have had a disease duration of
possible risks, and technical and handling problems. 10–15 years (Limousin et al., 1998), but with wide varia-
Ultimately, it is the patients who give informed consent tion. Recent studies suggest that DBS may be advocated
for one of the therapies, and their decision will often take in patients at an earlier stage of PD. However, operating
into account lifestyle issues that are seldom considered on patients earlier than 5 years following diagnosis
in clinical studies, for instance whether the patient would lead to the inclusion of patients with atypical par-
accepts an implantable system and brain surgery, which kinsonism, hence with only a short-term potential benefit
involves a higher one-time risk, or would prefer an infu- from DBS. STN DBS done after a mean duration of
sion system that requires more nursing care and visible disease of 7.5 years, and within the first 3 years after
components (Volkmann et al., 2013). However, a com- the development of levodopa-induced motor complica-
parison of the effects of DBS plus medical therapy ver- tions, was shown to be superior to medical therapy
sus medical therapy, in a population of patients whose (Schuepbach et al., 2013). However, serious adverse
treatment could have included apomorphine (one-third events related to surgical implantation or to the neuro-
of patients), showed that DBS was more beneficial than stimulation device occurred in 17.7% of patients, and
medical therapy (Williams et al., 2010). almost all studies report the risk of surgery-induced
severe adverse effects, especially symptomatic intracra-
nial bleeding in 1–2% of patients. A well informed
Surgical treatment
patient will finally decide when their disease-related dis-
DBS can be performed safely and successfully in patients ability warrants acceptance of the risks of surgery.
who have undergone previous ablative or DBS procedures
(Moro et al., 2000, 2002; Mogilner et al., 2002; Fraix et al.,
Levodopa responsiveness
2005). Thalamic surgery mainly improves tremor. Patients
with PD who develop long-term levodopa-induced motor So far, response to levodopa has been widely accepted as
complications after Vim thalamic surgery can be consid- the best outcome predictor for response to DBS. Several
ered for STN or GPi DBS. Bilateral STN DBS may be effi- reports have shown a good correlation of improvement in
cacious in improving motor fluctuations in patients who parkinsonism with levodopa or STN stimulation (Charles
have had prior unilateral or bilateral pallidal surgery, pro- et al., 2002; Welter et al., 2002). It is thus mandatory to
vided these patients still fulfill the criteria for DBS. evaluate carefully the beneficial and adverse motor
As patients having GPi or STN DBS are not managed effects of levodopa in each surgical candidate in order
in the same way, it is advocated to propose either contra- to explain to the patient what the signs and symptoms
lateral pallidal DBS or bilateral STN DBS in patients are that can potentially be improved by DBS. This corre-
with previous pallidal surgery. In patients with off-period lation is less clear with GPi DBS. Most centers use a for-
dyskinesias after neurotransplantation, GPi DBS contra- mal levodopa test, and a 30% improvement in the Unified
lateral to the most disabling dyskinesias may be consid- Parkinson Disease Rating Scale motor score has been con-
ered (Herzog et al., 2008a). sidered the minimal percentage below which DBS should
100 P. POLLAK
not be performed. However, this percentage should not be surgery are not improved by GPi or STN DBS. DBS
considered an absolute and sole criterion. The type and should not be recommended to patients who, at their
quality of improved symptoms are also of paramount best, remain disabled by freezing of gait or risk falling
importance (see below). In addition, severe tremor resis- during the pull test (Visser et al., 2008). At 5-year
tance to levodopa therapy is considered an exception to follow-up after bilateral STN DBS, off- and on-period
this (Zaidel et al., 2010). Large-amplitude parkinsonian akinesia, gait and instability worsened over time. During
tremor is sometimes improved by huge doses of levodopa the off-condition, these symptoms remained better than
(e.g., 500 mg for a single challenge), not tolerated during in the presurgical condition, whereas this was not the
chronic use. case during the on-condition (Krack et al., 2003). This
The percentage of levodopa-induced improvement means that surgery-induced benefit is partly maintained
may be difficult to assess, because classical motor tests on the levodopa-responsive part of these symptoms,
do not measure the worst-off and best-on motor states. which continue to evolve independently of both levodopa
Off-period motor signs are generally evaluated in defined and STN DBS. Moreover, in patients with levodopa-
off conditions (for example, in the morning after 12 hours responsive freezing of gait before surgery, STN stimu-
without dopaminergic treatment). In this condition the lation improved freezing of gait in most patients,
off-period motor signs may not be at their worst, because although it was not always as effective as levodopa in
some patients experience a major sleep benefit or a improving gait impairments. In addition, surgery can
rebound off-period in the afternoon that can be much induce gait problems in some patients (Ferraye et al.,
more disabling than an early-morning off-period. Even 2008; van Nuenen et al., 2008). DBS of the pedunculo-
more important is the evaluation of the best-on motor pontine nucleus (PPN) may partly improve on-period
state, which has been shown to be one of the best predic- gait, and reduce falls (Ferraye et al., 2010; Moro et al.,
tors of outcome (Fraix et al., 2006). Patients having no 2010). More data are needed before any firm conclusion
complaints related to parkinsonism at the peak effect can be drawn with regard to this investigational target.
of levodopa, who may claim to feel like normal (even Larger controlled studies should be done to compare
for less than 1 hour), can be considered the best candidates the symptoms that respond to STN versus GPi DBS, and
for surgery, whereas at other times they may exhibit to determine whether there are specific features that are
severe motor off-periods or dyskinesias. Therefore, to more resistant to one target than to the other.
ascertain the best-on motor condition it is important to
use a supratherapeutic dose of levodopa. DYSARTHRIA
The severity and type of dyskinesias are not determin-
Levodopa can improve or worsen parkinsonian speech.
ing factors with regard to surgery, as STN or GPi DBS
In the same way DBS, regardless of target, induces var-
greatly improves all types of dyskinesia (Follett et al.,
iable effects on dysarthria. Most studies have reported
2010). However, GPi DBS has a direct effect on dyskine-
limited or no improvement of preoperative speech dys-
sia, whereas STN DBS may induce or aggravate dyskine-
function after STN or GPi DBS (Limousin et al., 1998;
sia during the postoperative period. It seems that, during
Burchiel et al., 1999; Dromey et al., 2000; Krack et al.,
this period, patients with severe preoperative dyskinesias
2003; Deuschl et al., 2006). Rare exceptions have been
are more at risk of developing STN stimulation-induced
reported in patients with preoperative levodopa-sensitive
dyskinesia, necessitating an initial careful progressive
dysarthria (Gentil et al., 2003). Thus, a patient with
increase in voltage (Limousin et al., 1996).
severe dysarthria before surgery should not expect
improvement from DBS, and speech may even be wors-
Levodopa-unresponsive motor symptoms
ened. This has been well studied after STN DBS, and
These symptoms should be assessed during a levodopa reported to be related to the stimulation of the cerebel-
challenge at a time when the patient and examiner agree lothalamic tract located just posteromedially to the
that the peak of optimal benefit has been obtained. There STN, more frequently for left STN DBS (Tripoliti
is a consensus that the symptoms that remain at the best- et al., 2008, 2011).
on motor period are likely not improved by DBS (Lang
et al., 2006), or sometimes worsened. The only exemption DYSPHAGIA
to this rule is severe amplitude tremor, for which Vim and
Dysphagia is typical of the very advanced stage of PD,
STN DBS are more effective than dopaminergic drugs.
generally after the age of DBS indication. DBS does
not preclude the development of this symptom years
GAIT AND INSTABILITY
after surgery (Merola et al., 2011). DBS does not usually
Several studies have indicated that patients with great influence deglutition (Lengerer et al., 2012). However,
difficulty in gait and risk of falls at their best before rare case reports have shown a possible worsening
 DEEP BRAIN STIMULATION FOR PARKINSON’S DISEASE – PATIENT SELECTION 101
(Ghika et al., 1999). Therefore, the presence of preoper- evaluating attention and executive functions. A cutoff
ative dysphagia indicates that surgery may be unsuitable. score of 130 usually distinguishes demented from nonde-
mented patients. In the case of borderline scores or com-
Other levodopa-unresponsive symptoms plaints of relevant cognitive deterioration, it is useful to
repeat the evaluation after 6–12 months to ascertain that
COGNITION
cognitive function is stable. A progressive worsening
Dementia, as defined by DSM-IV, is the most frequent usually heralds dementia. Therefore, it is wise to know
exclusion criterion for DBS surgery, because dementia the patient’s cognitive performance within at least 1 year
may be worsened and patients will not have the leisure before referral for DBS. It also must be ascertained that
to take advantage of the surgery-induced benefit of cognitive dysfunction is not related to treatable causes
motor function. Almost all patients with PD exhibit some such as depression, or antiparkinsonian medication,
cognitive dysfunction, especially in the executive especially anticholinergics.
domain. The issue is to know the extent to which this dys-
function can jeopardize the outcome of DBS. Any brain
PSYCHIATRIC AND BEHAVIORAL ISSUES
surgery is invasive and patients usually receive general
anesthesia, at least for insertion of the pulse generator. PD is a neuropsychiatric disorder (Weintraub and
Therefore, some cognitive reserve is needed to prevent Burn, 2011). Psychiatric symptoms that are disease-
postoperative deterioration. Although transient post- related must be distinguished from the psychiatric
operative cognitive side-effects are relatively common side-effects of medication, especially impulse control
(Krack et al., 2003), DBS surgery has relatively few per- disorders and dopamine dysregulation syndrome
manent cognitive side-effects in well selected young and (Evans et al., 2009). In contrast to GPi DBS, STN DBS
nondemented patients, even when performed bilaterally allows for and requires major manipulations in dopami-
(Ardouin et al., 1999; Fields and Troster, 2000; Alegret nergic medication, making the management of postoper-
et al., 2001; Witt et al., 2008; Williams et al., 2010). Small ative patients more difficult than for other targets. Vim
decrements in some areas of information processing DBS seems to be much safer with regard to behavioral
were reported in a group of patients whose upper quar- side-effects. Worsening of depression after STN but
tile age was 70 years or more (Weaver et al., 2009). Some not GPi DBS was reported in a study that randomized
studies have reported that elderly patients, or those with the two targets (Follett et al., 2010), but this finding is
borderline preoperative global cognitive scores, are at debatable and unreplicated (Okun et al., 2009).
risk of permanent postoperative cognitive deterioration It has been shown that patients with a history of pre-
(Vingerhoets et al., 1999; Saint-Cyr et al., 2000; Krack operative psychiatric disorders are most prone to
et al., 2003) or failure of improvement in quality of life develop postoperative psychiatric complications
(Witt et al., 2011). A randomized study found that GPi (Houeto et al., 2002). Surgery is generally deferred in
DBS may be safer than STN DBS, with more decline patients with an unstable psychiatric condition at the
of one component of processing speed (visuomotor) time of surgery, until the symptoms have been managed
after STN DBS (Follett et al., 2010). In another random- adequately. This is particularly true for depression and
ized study, a worsening of letter verbal fluency was seen psychosis. Ongoing severe depression with suicidal ide-
in unilateral STN but not GPi DBS, even in the STN DBS ation should be considered an absolute contraindication
off-state, suggesting a surgical rather than a stimulation- to surgery because of the risk of suicide (Burkhard et al.,
induced effect (Okun et al., 2009). In patients with 2004; Funkiewiez et al., 2004). The reported increased
contraindications for STN DBS because of levodopa- rate of suicide in patients with PD who have undergone
resistant axial motor signs or cognitive decline, neuro- STN DBS underscores the need for a comprehensive pre-
psychological performance remained unchanged operative psychiatric assessment and treatment of
6 months after bilateral GPi DBS (Rouaud et al., 2010). depression, as well as for careful and detailed postoper-
There is no consensus on the type of testing and level ative follow-up (Voon et al., 2008).
of performance that would exclude patients from receiv- Although psychiatric symptoms that are disease-
ing DBS. A thorough screening for cognitive deficits is related, such as apathy, can be exacerbated following
mandatory. Neuropsychological evaluation with a spe- STN DBS in the context of a dopaminergic withdrawal
cial emphasis on memory and executive function is syndrome (Krack et al., 2003), symptoms within the spec-
highly recommended (Voon et al., 2006). Several groups trum of the dopamine dysregulation syndrome are
use the Mattis Dementia Rating Scale (MDRS) (range improved or even disappear if dopaminergic treatment
0–144) for evaluation of overall cognitive function. This can be greatly reduced after surgery (Lhommee et al.,
is an appropriate tool in degenerative diseases involving 2012). Therefore, STN DBS in PD can be considered as
subcortical structures, given the inclusion of tests a good indication for patients with disabling psychic
102 P. POLLAK
fluctuations, dopaminergic treatment abuse, and drug- can result in a relative improvement of autonomic func-
induced behavioral addictions. However, some favorable tion after surgery (Ludwig et al., 2007). One study
dopamine agonist-induced effects, such as creativity or showed that STN DBS led to a significant enhancement
increased motivation, can disappear within the first year of afferent urinary bladder information processing and
after surgery. The patients must be aware of this risk. can thus improve urinary symptoms (Herzog et al.,
However, resumption of a dopamine agonist drug and 2008b). Constipation can be improved. However, STN
adjustment of its dose to control the re-emergence DBS does not prevent the development of urinary incon-
of other deleterious hyperdopaminergic behaviors tinence or orthostatic blood pressure in the long term
(hypersexuality or pathological gambling) usually enable (Merola et al., 2011). Preoperative dysautonomic fea-
the treatment of postoperative apathy (Thobois et al., tures are not a contraindication to DBS, and may be
2013). Patients suffering from both PD and obsessive– moderately improved.
compulsive disorder can be good candidates for STN
DBS, as this surgery has been shown to be beneficial in PAIN
both disorders (Mallet et al., 2002).
Painful sensations are common in PD. In many patients,
SLEEP PROBLEMS such sensations correspond to neuropathic pain and could
be related to central alterations of pain processing. Bilat-
Patients with PD often report increased daily somno- eral STN DBS was reported to raise pain thresholds and
lence and night sleep abnormalities, which are both dis- restore functioning of the discriminative pain system
ease and drug related. Are these symptoms relevant in (Dellapina et al., 2012). Thus, STN DBS can be performed
terms of modulating the selection process for DBS? in patients with severe pain, as confirmed by Witjas and
Data are available mainly for STN DBS. colleagues (2007), who showed that, among nonmotor
Several studies have suggested that chronic STN DBS symptoms, pain/sensory fluctuations had the best
may improve sleep quality through increased nocturnal response to STN DBS, in the same way as akathisia.
mobility and reduction of sleep fragmentation. More
precisely, STN stimulation greatly reduces night-time WEIGHT
akinesia and suppresses axial and early morning dysto-
nia, but does not alleviate periodic leg movements or Many studies have described a rapid and marked weight
rapid-eye-movement (REM) sleep behavior disorders. gain, beginning just after DBS in the majority of
Total sleep time increases, wakefulness after sleep onset patients, with women gaining more fat than men
decreases, and sleep quality improves (Arnulf et al., (Montaurier et al., 2007). This weight gain sometimes
2000b; Chahine et al., 2011). These changes in sleep were induces obesity and can have metabolic repercussions.
related to improvements in functioning, specifically It occurs whatever target is stimulated, unilaterally or
those affected by bradykinesia (Lyons and Pahwa, bilaterally, but some reports have shown a greater gain
2006). Decreasing dopamine agonist drugs after STN after STN DBS (Sauleau et al., 2009). The pathophysio-
DBS may unmask a restless legs syndrome (RLS) in logical mechanisms responsible for the weight gain are
some patients. Vim DBS does not modify sleep quality multifactorial.
or architecture (Arnulf et al., 2000a). Stimulating the It is not known whether patients who are obese before
PPN, a structure with a REM-promoting action, nearly surgery are at greater risk of weight gain. Therefore,
doubled nocturnal REM sleep without significant DBS can be proposed to them, with an even stricter need
changes in other sleep states (Lim et al., 2009). for proactive nutritional counseling than for other
On the whole, patients with nocturnal sleep distur- patients.
bances can be improved by STN DBS, but RLS may be
aggravated in the occasional patient. PERSONAL, PROFESSIONAL, AND
SOCIAL ISSUES
DYSAUTONOMIA
Disability depends to some extent on individual social
Dysfunctions of the autonomic nervous system are com- and personal status, and patient selection must be indi-
mon in PD. It has been reported that, in contrast to levo- vidualized according to each patient’s professional activ-
dopa, STN DBS has only minor effects on autonomic ities, interpersonal relationships, and expectations of
functions or can stabilize blood pressure (Stemper functionality. For example, a patient who is married
et al., 2006). Drenching sweats were reported to be and has a dedicated husband or wife constantly present
improved (Witjas et al., 2007). As less pharmacotherapy may more easily accept an occasional off-period with
is needed after STN stimulation, reduced drug intake loss of autonomy than a patient with the same motor
 DEEP BRAIN STIMULATION FOR PARKINSON’S DISEASE – PATIENT SELECTION 103
features living on his or her own. Patients who are work- with Parkinson’s disease and essential tremor. J Sleep
ing may have to take increasingly high doses of medica- Res 9: 55–62.
tion to be able to pursue their physical activities at the Arnulf I, Bejjani BP, Garma L et al. (2000b). Improvement of
same pace as a nondisabled person, thereby exacerbating sleep architecture in PD with subthalamic nucleus stimula-
tion. Neurology 55: 1732–1734.
a vicious circle with increasingly severe dyskinesia,
Bronstein JM, Tagliati M, Alterman RL et al. (2011). Deep
rebound off-periods, dopamine dependency, and psychi-
brain stimulation for Parkinson disease: an expert consen-
atric side-effects. On the other hand, surgery may allow sus and review of key issues. Arch Neurol 68: 165.
some patients to resume or pursue their professional Burchiel KJ, Anderson VC, Favre J et al. (1999). Comparison
activities, and, in such patients, DBS should be encour- of pallidal and subthalamic nucleus deep brain stimulation
aged early after the onset of motor complications for advanced Parkinson’s disease: results of a randomized,
(Schuepbach et al., 2013). blinded pilot study. Neurosurgery 45: 1375–1382.
Patients’ expectations are not always realistic, and Burkhard PR, Vingerhoets FJ, Berney A et al. (2004). Suicide
each neurologist is responsible, after careful assessment after successful deep brain stimulation for movement dis-
of the individual risk–benefit ratio, for providing a real- orders. Neurology 63: 2170–2172.
istic explanation of the potential restorative benefits of Chahine LM, Ahmed A, Sun Z (2011). Effects of STN DBS for
Parkinson’s disease on restless legs syndrome and other
surgery. Surgery and postoperative adaptation to drugs
sleep-related measures. Parkinsonism Relat Disord 17:
and stimulation parameters require a great deal of
208–211.
patient cooperation. Any severe personality disorder Charles PD, Van Blercom N, Krack P et al. (2002). Predictors
that prevents a minimum of cooperation is likely to be of effective bilateral subthalamic nucleus stimulation for
a contraindication to this type of elective surgery. PD. Neurology 59: 932–934.
Patients with severe anxiety may be unable to cope Dellapina E, Ory-Magne F, Regragui W et al. (2012). Effect of
with the stress of surgery performed under local anes- subthalamic deep brain stimulation on pain in Parkinson’s
thesia, which makes them uncooperative. Therefore, disease. Pain 153: 2267–2273.
electrode implantation under general anesthesia should Derost PP, Ouchchane L, Morand D et al. (2007). Is DBS-STN
be considered. appropriate to treat severe Parkinson disease in an elderly
population? Neurology 68: 1345–1355.
Deuschl G, Schade-Brittinger C, Krack P et al. (2006). A ran-
CONCLUSION domized trial of deep-brain stimulation for Parkinson’s dis-
ease. N Engl J Med 355: 896–908.
An experienced multidisciplinary team is mandatory for Dromey C, Kumar R, Lang AE et al. (2000). An investigation
correct selection of patients with PD as candidates for of the effects of subthalamic nucleus stimulation on acous-
DBS. Patients should have medically intractable motor tic measures of voice. Mov Disord 15: 1132–1138.
fluctuations or dyskinesia, or intractable tremor. Many Evans AH, Strafella AP, Weintraub D et al. (2009). Impulsive
factors have to be taken into account before advocating and compulsive behaviors in Parkinson’s disease. Mov
Disord 24: 1561–1570.
surgery to a given patient. The most important are age,
Ferraye MU, Debu B, Fraix V et al. (2008). Effects of subtha-
levodopa responsiveness, type and severity of levodopa- lamic nucleus stimulation and levodopa on freezing of gait
unresponsive symptoms, previous medical and surgical in Parkinson disease. Neurology 70: 1431–1437.
treatments, comorbid disorders, cognitive and psychiat- Ferraye MU, Debu B, Fraix V et al. (2010). Effects of pedun-
ric states, and brain MRI findings. Eventually, the con- culopontine nucleus area stimulation on gait disorders in
sent for DBS is given by well informed patients and their Parkinson’s disease. Brain 133: 205–214.
decision takes into account personal, professional, and Fields JA, Troster AI (2000). Cognitive outcomes after deep
social issues that are seldom considered in evidence- brain stimulation for Parkinson’s disease: a review of initial
based medicine. studies and recommendations for future research. Brain
Cogn 42: 268–293.
REFERENCES Follett KA, Weaver FM, Stern M et al. (2010). Pallidal versus
subthalamic deep-brain stimulation for Parkinson’s dis-
Alegret M, Junque C, Valldeoriola F et al. (2001). Effects of ease. N Engl J Med 362: 2077–2091.
bilateral subthalamic stimulation on cognitive function in Fraix V, Pollak P, Moro E et al. (2005). Subthalamic nucleus
Parkinson disease. Arch Neurol 58: 1223–1227. stimulation in tremor dominant parkinsonian patients with
Ardouin C, Pillon B, Peiffer E et al. (1999). Bilateral subtha- previous thalamic surgery. J Neurol Neurosurg Psychiatry
lamic or pallidal stimulation for Parkinson’s disease affects 76: 246–248.
neither memory nor executive functions: a consecutive Fraix V, Houeto JL, Lagrange C et al. (2006). Clinical and eco-
series of 62 patients. Ann Neurol 46: 217–223. nomic results of bilateral subthalamic nucleus stimulation
Arnulf I, Bejjani BP, Garma L et al. (2000a). Effect of low and in Parkinson’s disease. J Neurol Neurosurg Psychiatry 77:
high frequency thalamic stimulation on sleep in patients 443–449.
104 P. POLLAK
Funkiewiez A, Ardouin C, Caputo E et al. (2004). Long term Lyons KE, Pahwa R (2006). Effects of bilateral subthalamic
effects of bilateral subthalamic nucleus stimulation on nucleus stimulation on sleep, daytime sleepiness, and early
cognitive function, mood, and behaviour in Parkinson’s morning dystonia in patients with Parkinson disease.
disease. J Neurol Neurosurg Psychiatry 75: 834–839. J Neurosurg 104: 502–505.
Gentil M, Pinto S, Pollak P et al. (2003). Effect of bilateral Mallet L, Mesnage V, Houeto JL et al. (2002). Compulsions,
stimulation of the subthalamic nucleus on parkinsonian Parkinson’s disease, and stimulation. Lancet 360:
dysarthria. Brain Lang 85: 190–196. 1302–1304.
Ghika J, Ghika-Schmid F, Fankhauser H et al. (1999). Bilateral Merola A, Zibetti M, Angrisano S et al. (2011). Parkinson’s
contemporaneous posteroventral pallidotomy for the treat- disease progression at 30 years: a study of subthalamic deep
ment of Parkinson’s disease: neuropsychological and neu- brain-stimulated patients. Brain 134: 2074–2084.
rological side effects. Report of four cases and review of the Mogilner AY, Sterio D, Rezai AR et al. (2002). Subthalamic
literature. J Neurosurg 91: 313–321. nucleus stimulation in patients with a prior pallidotomy.
Gorgulho A, De Salles AA, Frighetto L et al. (2005). Incidence J Neurosurg 96: 660–665.
of hemorrhage associated with electrophysiological studies Montaurier C, Morio B, Bannier S et al. (2007). Mechanisms
performed using macroelectrodes and microelectrodes in of body weight gain in patients with Parkinson’s disease
functional neurosurgery. J Neurosurg 102: 888–896. after subthalamic stimulation. Brain 130: 1808–1818.
Herzog J, Pogarell O, Pinsker MO et al. (2008a). Deep brain Moro E, Esselink RA, Van Blercom N et al. (2000). Bilateral
stimulation in Parkinson’s disease following fetal nigral subthalamic nucleus stimulation in a parkinsonian patient
transplantation. Mov Disord 23: 1293–1296. with previous unilateral pallidotomy and thalamotomy.
Herzog J, Weiss PH, Assmus A et al. (2008b). Improved sensory Mov Disord 15: 753–755.
gating of urinary bladder afferents in Parkinson’s disease Moro E, Esselink RJ, Benabid AL et al. (2002). Response to
following subthalamic stimulation. Brain 131: 132–145. levodopa in parkinsonian patients with bilateral subthala-
Houeto JL, Mesnage V, Malle L et al. (2002). Behavioural dis- mic nucleus stimulation. Brain 125: 2408–2417.
orders, Parkinson’s disease and subthalamic stimulation. Moro E, Allert N, Eleopra R et al. (2009). A decision tool to
J Neurol Neurosurg Psychiatry 72: 701–707. support appropriate referral for deep brain stimulation in
Hu X, Jiang X, Zhou X et al. (2010). Avoidance and manage- Parkinson’s disease. J Neurol 256: 83–88.
ment of surgical and hardware-related complications of Moro E, Hamani C, Poon YY et al. (2010). Unilateral pedun-
deep brain stimulation. Stereotact Funct Neurosurg 88: culopontine stimulation improves falls in Parkinson’s
296–303. disease. Brain 133: 215–224.
Krack P, Batir A, Van Blercom N et al. (2003). Five-year Okun MS, Tagliati M, Pourfar M et al. (2005). Management of
follow-up of bilateral stimulation of the subthalamic referred deep brain stimulation failures: a retrospective
nucleus in advanced Parkinson’s disease. N Engl J Med analysis from 2 movement disorders centers. Arch
349: 1925–1934. Neurol 62: 1250–1255.
Lang AE, Houeto JL, Krack P et al. (2006). Deep brain stim- Okun MS, Fernandez HH, Wu SS et al. (2009). Cognition and
ulation: preoperative issues. Mov Disord 21: S171–S196. mood in Parkinson’s disease in subthalamic nucleus versus
Lengerer S, Kipping J, Rommel N et al. (2012). Deep-brain- globus pallidus interna deep brain stimulation: the
stimulation does not impair deglutition in Parkinson’s COMPARE trial. Ann Neurol 65: 586–595.
disease. Parkinsonism Relat Disord 18: 847–853. Ory-Magne F, Brefel-Courbon C, Simonetta-Moreau M et al.
Lhommee E, Klinger H, Thobois S et al. (2012). Subthalamic (2007). Does ageing influence deep brain stimulation out-
stimulation in Parkinson’s disease: restoring the balance of comes in Parkinson’s disease? Mov Disord 22: 1457–1463.
motivated behaviours. Brain 135: 1463–1477. Oyama G, Rodriguez RL, Jones JD et al. (2012). Selection of
Lim AS, Moro E, Lozano AM et al. (2009). Selective enhance- deep brain stimulation candidates in private neurology
ment of rapid eye movement sleep by deep brain stimula- practices: referral may be simpler than a computerized tri-
tion of the human pons. Ann Neurol 66: 110–114. age system. Neuromodulation 15: 246–250.
Limousin P, Pollak P, Hoffmann D et al. (1996). Abnormal Rouaud T, Dondaine T, Drapier S et al. (2010). Pallidal
involuntary movements induced by subthalamic nucleus stimulation in advanced Parkinson’s patients with contrain-
stimulation in parkinsonian patients. Mov Disord 11: dications for subthalamic stimulation. Mov Disord 25:
231–235. 1839–1846.
Limousin P, Krack P, Pollak P et al. (1998). Electrical stimu- Russmann H, Ghika J, Villemure JG et al. (2004). Subthalamic
lation of the subthalamic nucleus in advanced Parkinson’s nucleus deep brain stimulation in Parkinson disease
disease. N Engl J Med 339: 1105–1111. patients over age 70 years. Neurology 63: 1952–1954.
Lopiano L, Rizzone M, Perozzo P et al. (2001). Deep brain Saint-Cyr JA, Trepanier LL, Kumar R et al. (2000).
stimulation of the subthalamic nucleus: selection of Neuropsychological consequences of chronic bilateral
patients and clinical results. Neurol Sci 22: 67–68. stimulation of the subthalamic nucleus in Parkinson’s
Ludwig J, Remien P, Guballa C et al. (2007). Effects of sub- disease. Brain 123: 2091–2108.
thalamic nucleus stimulation and levodopa on the auto- Sansur CA, Frysinger RC, Pouratian N et al. (2007). Incidence
nomic nervous system in Parkinson’s disease. J Neurol of symptomatic hemorrhage after stereotactic electrode
Neurosurg Psychiatry 78: 742–745. placement. J Neurosurg 107: 998–1003.
 DEEP BRAIN STIMULATION FOR PARKINSON’S DISEASE – PATIENT SELECTION 105
Sauleau P, Leray E, Rouaud T et al. (2009). Comparison of Volkmann J, Albanese A, Antonini A et al. (2013). Selecting
weight gain and energy intake after subthalamic versus deep brain stimulation or infusion therapies in advanced
pallidal stimulation in Parkinson’s disease. Mov Disord Parkinson’s disease: an evidence-based review. J Neurol
24: 2149–2155. [Epub ahead of print].
Schuepbach WM, Rau J, Knudsen K et al. (2013). Voon V, Kubu C, Krack P et al. (2006). Deep brain stimula-
Neurostimulation for Parkinson’s disease with early motor tion: neuropsychological and neuropsychiatric issues.
complications. N Engl J Med 368: 610–622. Mov Disord 21: S305–S327.
Stemper B, Beric A, Welsch G et al. (2006). Deep brain stim- Voon V, Krack P, Lang AE et al. (2008). A multicentre study
ulation improves orthostatic regulation of patients with on suicide outcomes following subthalamic stimulation for
Parkinson disease. Neurology 67: 1781–1785. Parkinson’s disease. Brain 131: 2720–2728.
Thobois S, Lhommee E, Klinger H et al. (2013). Parkinsonian Weaver FM, Follett K, Stern M et al. (2009). Bilateral deep
apathy responds to dopaminergic stimulation of D2/D3 brain stimulation vs best medical therapy for patients with
receptors with piribedil. Brain 136: 1568–1577. advanced Parkinson disease: a randomized controlled trial.
Tripoliti E, Zrinzo L, Martinez-Torres I et al. (2008). Effects of JAMA 301: 63–73.
contact location and voltage amplitude on speech and Weintraub D, Burn DJ (2011). Parkinson’s disease: the
movement in bilateral subthalamic nucleus deep brain quintessential neuropsychiatric disorder. Mov Disord 26:
stimulation. Mov Disord 23: 2377–2383. 1022–1031.
Tripoliti E, Zrinzo L, Martinez-Torres I et al. (2011). Effects of Welter ML, Houeto JL, Tezenas du Montcel S et al. (2002).
subthalamic stimulation on speech of consecutive patients Clinical predictive factors of subthalamic stimulation in
with Parkinson disease. Neurology 76: 80–86. Parkinson’s disease. Brain 125: 575–583.
Tsai ST, Lin SH, Chou YC et al. (2009). Prognostic factors of Williams A, Gill S, Varma T et al. (2010). Deep brain stimula-
subthalamic stimulation in Parkinson’s disease: a compar- tion plus best medical therapy versus best medical therapy
ative study between short- and long-term effects. Stereotact alone for advanced Parkinson’s disease (PD SURG trial):
Funct Neurosurg 87: 241–248. a randomised, open-label trial. Lancet Neurol 9: 581–591.
van Nuenen BF, Esselink RA, Munneke M et al. (2008). Witjas T, Kaphan E, Regis J et al. (2007). Effects of chronic
Postoperative gait deterioration after bilateral subthalamic subthalamic stimulation on nonmotor fluctuations in
nucleus stimulation in Parkinson’s disease. Mov Disord 23: Parkinson’s disease. Mov Disord 22: 1729–1734.
2404–2406. Witt K, Daniels C, Reiff J et al. (2008). Neuropsychological
Vesper J, Haak S, Ostertag C et al. (2007). Subthalamic and psychiatric changes after deep brain stimulation for
nucleus deep brain stimulation in elderly patients – analysis Parkinson’s disease: a randomised, multicentre study.
of outcome and complications. BMC Neurol 7: 7. Lancet Neurol 7: 605–614.
Vingerhoets G, van der Linden C, Lannoo E et al. (1999). Witt K, Daniels C, Krack P et al. (2011). Negative impact of
Cognitive outcome after unilateral pallidal stimulation in borderline global cognitive scores on quality of life after
Parkinson’s disease. J Neurol Neurosurg Psychiatry 66: subthalamic nucleus stimulation in Parkinson’s disease.
297–304. J Neurol Sci 310: 261–266.
Visser JE, Allum JH, Carpenter MG et al. (2008). Subthalamic Zaidel A, Bergman H, Ritov Y et al. (2010). Levodopa and
nucleus stimulation and levodopa-resistant postural insta- subthalamic deep brain stimulation responses are not con-
bility in Parkinson’s disease. J Neurol 255: 205–210. gruent. Mov Disord 25: 2379–2386.