Surgical Management of Parkinson's Disease in the Elderly

ABSTRACT

Background

Deep Brain Stimulation (DBS) is an increasingly popular therapy for Parkinson's Disease (PD). Despite the experience gained over time with DBS of either the subthalamus or the globus pallidus pars interna, there is still no consensus regarding the age limit for DBS indication.

Objectives

This narrative review of the literature discusses the issues of age and DBS, emphasizing the critical need for good quality evidence to support the surgical management of elderly patients with PD.

Methods

We searched PubMed using the terms Parkinson's Disease; Parkinson's Disease therapy; deep brain stimulation; antiparkinsonian agents therapeutic use; age factors; aged; aged, 80 and over; middle aged; treatment outcome; and risk assessments.

Results

We identified several limitations of the available evidence, such as under‐representation of older patients in DBS studies, small sample sizes in studies with older participants, heterogeneity of outcomes, and conflicting results.

Conclusions

Despite preliminary suggestions that age might affect the outcomes of DBS, the evidence to support the hypothesis of age as an independent predictor of DBS outcomes is limited and results are controversial. Ultimately, finding an age‐independent biomarker predicting DBS outcome is the final goal to expand this powerful treatment to all patients age in an effective and safe manner.

Parkinson's disease (PD) is one of the most prevalent neurodegenerative disorders, with a prediction of overtaking Alzheimer's disease prevalence for the years to come. ¹ , ² Over the past three decades, PD earned one the most outstanding revolutionary symptomatic treatments: deep brain stimulation (DBS). There is ample evidence that DBS provides a substantial improvement in the quality of life (QoL) of PD patients and, not surprisingly, has become one of the most common pursued treatment. ³ DBS of either the Subthalamic nucleus (STN) or Globus Pallidus pars Interna (GPi) is able to improve most cardinal signs of the disease and—to a greater extent—motor fluctuations and levodopa‐induced dyskinesias. ⁴ , ⁵ On the other hand, albeit explored by different research groups throughout the years, we still lack clear evidence for a neuroprotective and disease modifying effect of DBS.

The more DBS is used, the more movement disorders specialists understand its limitations, side effects, and contraindications. Experts in DBS follow a standard indication protocol, and a rigorous pre‐surgical checklist that needs to be fulfilled. The main DBS indications for PD are the occurrence of disabling motor fluctuations, l‐dopa‐induced dyskinesias, or l‐dopa‐resistant tremor, notwithstanding an optimized treatment (Table 1). In spite of the experience gained over time with DBS therapy of either STN or GPi, the age limit for the procedure is still a matter of debate, with different centers using different cut‐offs, or no‐cut‐off. ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ The same applies to clinical trials, where some studies considered the age of 75 as cut‐off, ¹¹ , ¹² others the age of 80, ¹³ and others had no specified maximum age. ¹⁴ , ¹⁵ , ¹⁶ Considering the lack of evidence for a given age cut‐off, the explanation for the choice of age limit is unclear. ¹⁰

TABLE 1.

The selection of advanced treatments for PD patients involves four steps

STEP 1: Eligibility for Advanced Treatment
Inclusion criteria Established diagnosis of PD Disease duration longer than 4 years* Motor fluctuations and/or dyskinesias Quality of life and social functioning influenced by levodopa‐responsive signs Oral therapy optimized but ineffective, e.g.: Off time more than 25% daytime Severe dyskinesias Severe off periods (pain, dystonia, panic attack) Unpredictable off periods		Exclusion criteria Short life span Major systemic conditions limiting the quality of life more than PD Uncontrolled depression Uncontrolled psychosis** Dementia
		Supportive criteria Excellent l‐dopa response Patient's motivation Lack of excessive expectations Supportive entourage
step 2: Treatment Selection
DBS	Ablation	LCIG	CSAI
Supportive criteria l‐dopa‐resistant tremor Brittle response to l‐dopa Autonomy needs (eg traveling) Esthetic considerations (ie no external device) Nocturnal akinesia Specific exclusion criteria Suicide risk Depression/apathy High surgical risk (eg severe atrophy or diffuse vascular white matter changes at brain MRI)	Supportive criteria l‐dopa‐resistant tremor Brittle response to l‐dopa Asymmetric signs Overall frailty Poor cooperation/support for programming Esthetic consideration (eg baldness) Specific exclusion criteria Bilateral/midline bothersome symptoms Long life span	Supportive criteria Need to simplify oral treatment DDS? Swallowing impairment Patient's choice to avoid neurosurgery Specific exclusion criteria Lack of on state without troublesome dyskinesias Nocturnal akinesia Camptocormia Pre‐existing neuropathy? Allergy to carbidopa Contradictions to PEG***	Supportive criteria Depression/apathy Need to simplify oral treatment DDS? Patient's choice to avoid neurosurgery Need to obtain rapid relief Specific exclusion criteria Age > 65–70 years Nocturnal akinesia Psychosis ICD Allergy to morphine Uncontrolled asthma Severe dysautonomia
step 3: Target Selection (In Case Of Dbs Or Ablation)
STN (DBS)	GPi (DBS/ablation)	Vim (DBS/ ablation)
Supportive criteria Young age Severe tremor Need to reduce medications Nocturnal akinesia Specific exclusion criteria Age > 70 years Depression/apathy MCI with progressive features (eg multi‐domain, visuo‐spatial) Axial impairment (balance, speech or gait problems)	Supportive criteria Severe dyskinesias L‐dopa resistant dystonia L‐dopa responsive non‐motor symptoms DDS? Specific exclusion criteria Age > 75–80 years?	Supportive criteria Severe tremor Absence of other PD‐related issues Specific exclusion criteria Age > 80–85 years? Other PD‐related bothersome symptoms
step 4: When?
As soon as symptoms and signs improvable with the advanced therapy under consideration are affecting the quality of life and/or social functional of the patient in spite of reasonable attempts with conservative treatments (eg drugs).

^*A disease duration cut‐off is no longer mandatory but still useful to ascertain PD diagnosis.

^**Eg non‐visual hallucinations or not easily treatable with low doses of neuroleptics,

^***Abdominal/dermatological active diseases, previous GI surgery, etc.

Abbreviations: CSAI, continuous subcutaneous apomorphine infusion; DDS, dopamine dysregulation syndrome; GPi, globus pallidus pars interna; ICD, impulse control disorder; LCIG, l‐dopa‐carbidopa intestinal gel; MCI, mild cognitive impairment; MRI, magnetic resonance imaging; PD, Parkinson's disease; PEG, percutaneous gastrostomy; STN, subthalamic nucleus; Vim, ventralis intermedius nucleus of the thalamus.

According to a recent meta‐analysis of clinical trials comparing the effects of STN and GPi DBS in patients with PD, the majority of patients included in the studies were in the early sixties. ¹⁷ A recent systematic review focused on QoL found that in four studies higher age was associated with lower improvement in QoL domains in the short‐term follow‐up, whereas in other six studies there was no correlation between age and postoperative QoL. ¹⁸

Since the success rate of DBS depends on a precise indication, ¹⁹ and that age is one vital element to be considered, this topic reunites most of the specialist's concerns when selecting patients for DBS therapy. Our aim was to review the literature covering the issues of age and DBS. We searched PubMed using the terms Parkinson Disease; Parkinson Disease therapy; deep brain stimulation; antiparkinsonian agents therapeutic use; age factors; aged; aged, 80 and over; middle aged; treatment outcome; and risk assessments. Herein, we present a narrative review emphasizing the critical need for good quality evidence to support the decision‐making process tailored to elderly patients with PD.

DBS Selection Process in the Elderly

PD has one of the most challenging therapeutic management in neurology, mainly due to the variability of its motor and non‐motor symptoms, which also presents differently depending on patient's age. For instance, in the elderly population resting tremor is observed more frequently as the initial symptom as compared to young patients. ²⁰ , ²¹ However, as disease progresses, older patients are more likely to experience axial symptoms, such as gait and postural impairment, ²¹ as well as to develop psychotic symptoms. ²⁰ , ²²

The task also requires a full understanding of PD pathophysiology, patient's lifestyle, age‐related brain changes, and comorbidities. Brain aging with the associated neuronal loss and brain atrophy may influence the effects of DBS in the elderly population. Kempster et al.²³ demonstrated that age of onset and total disease duration predict the inexorable progression of PD, particularly with respect to the development of four relevant milestones: visual hallucinations, recurrent falls, necessity of long‐term care facilities, and dementia. These milestones directly impact on DBS decision.

With age, several brain functions decline, including working and episodic memory, decision‐making, and executive function. ²⁴ Age‐related changes also modify anatomy, possibly reducing the relative distance between intended target of stimulation and surrounding structures. ²⁵ Other neurodegenerative co‐pathologies, Alzheimer's disease in particular, are also present and further influence DBS outcome. ²⁶ Neuronal loss and dendritic regression are the main contributors to the age‐related brain atrophy, although the impact of this histological changes on DBS response is not known at the moment. ²⁷

Finally, concomitant gerontologic conditions of PD patients also interfere with the selection of DBS candidates. Frailty and sarcopenia can directly contribute to a worse performance of PD patients. ²⁸ An indirect effect of these conditions is also possible, as – for instance – body mass index reduction requires adjustments of l‐dopa dose. ²⁹

Response to Levodopa

An excellent response to l‐dopa has been historically considered as one of the most important predictors of good motor outcome after DBS. ³⁰ The relationship between age and l‐dopa responsiveness has received attention in the literature. ⁸ , ³¹ Studies have demonstrated that age has a more significant detrimental effect on l‐dopa responsiveness than disease duration, most likely reflecting the more widespread pathology of late‐onset versus early‐onset PD. ³² Thus, an elderly patient with a short disease duration may have a poorer response to l‐dopa than a younger patient with prolonged disease. ³¹ , ³³

PD management has undergone a profound change in recent years, ultimately resulting in a higher number of DBS referrals. For decades, l‐dopa therapy has been delayed to avoid (or rather postpone) the development of l‐dopa‐induced dyskinesias and motor fluctuations, in favor of dopamine agonists and monoamine oxidade‐B (MAO‐B) inhibitors as initial therapies. ¹⁵ More recently, based on evidence that motor complications are related to PD intrinsic mechanisms, disease severity and duration, ³⁴ , ³⁵ , ³⁶ l‐dopa sparing agents have been progressively abandoned especially in North America, and this is even more relevant in older patients who are less likely to tolerate them. ³⁶ Beyond that, a recent meta‐analysis identified the synergistic effect of l‐dopa and DBS on motor improvement up to 5 years after surgery, ³⁷ undoubtedly another meaningful fact to be taken into account.

In the light of this current approach for PD management, and since older patients have a diminished response to l‐dopa, thus requiring higher doses, it is conceivable that the elderly PD population will soon enrich more and more DBS referrals (Fig. 1).

FIG. 1.

The medical treatment of Parkinson's disease has changed over the years, with the early adoption of levodopa and less emphasis of the impact of fluctuations and dyskinesias on quality of life. This figure (modified from ³⁸ ) illustrates two possible scenarios depending on the age at onset. In scenario 1 (young‐onset PD, starting at age 40 in this example) motor complications are expected 7–8 years after disease onset, at an age when choosing advanced therapy is relatively straightforward (STN DBS or CSAI). In scenario 2 (typical onset, starting at age 65 in this example) develop motor complications will begin at an age with greater surgical risks, thus challenging the choice of the best surgical option. Abbreviations: CSAI, Continuous subcutaneous apomorphine infusion; DBS, Deep brain stimulation; GPi, Globus pallidus pars interna; iCOMT, Inhibitors of catechol‐O‐methyltransferase; LCIG, Levodopa‐carbidopa intestinal gel; STN, Subthalamic nucleus.

Disease Duration

The past view of DBS as an option when all other conservative measures have failed has led to inferior result of the neurostimulation in patients with advanced disease, as elegantly shown by the EARLYSTIM trial. ³⁹ In this trial, patients (mean age: 52.6 years, mean disease duration: 7.5 years) were randomized to STN DBS or “best medical therapy” as soon as bothersome fluctuations started. The STN DBS group was found to have an immediate and sustained better outcome in many metrics, including QoL (primary endpoint) after a 2‐year follow‐up. ³⁹

The EARLYSTIM approach has been found to perform well in patients with short disease duration and at a relatively young age. However, the effects of early DBS in patients with short disease duration but a decade or more older is presently unknown.

Target Selection

There is no evidence of the superiority of either STN or GPi in terms of improvement of motor complications and motor scores, particularly when groups are analyzed under the effect of both medication and stimulation. ³⁰ , ⁴⁰

Undoubtedly, STN is the preferred target in the young population, mainly due to the possibility of a significant drug reduction after surgery. ⁴⁰ Given the aforementioned age bias of DBS, STN is the most frequent target used for PD. STN has however a higher rate of neuropsychiatric side effects than GPi, ³² and a higher chance of inducing balance and gait disorders in vulnerable patients. ³¹ , ³³

On the other hand, GPi stimulation does not produce a substantial decrease in drug regimen and requires a greater amplitude of stimulation, which can quickly drain the battery level. ¹⁵ GPi DBS has however a better efficacy to treat dyskinesias than STN and it seems safer in terms of axial motor function, mood, and cognition. ⁴¹

The Ventral intermediate (Vim) nucleus of the thalamus is the best target to treat different types of tremors. In PD, it is reserved for patients with a tremor‐dominant phenotype that cannot undergo STN or GPi due to contraindications such as cognitive decline. ⁴² Interestingly, since PD tremor usually requires high doses of l‐dopa, a medication reduction is indeed possible after Vim DBS.

There is lack of evidence in the literature to support the choice between STN or GPi DBS in elderly patients, as the clinical trials comparing these targets have typically recruited a younger population (Table 2). Target decision is therefore based on the specificities of each patient and personal experience of the treating physician. ³³ Overall, while the impact of advance age can play a role in post‐operative deterioration after STN DBS, less concerning is the effect of age after GPi and – especially – Vim surgery (Table 1).

TABLE 2.

The age of PD patients enrolled in randomized controlled trials of GPi vs. STN DBS

Study	No. of centers (country)	Design	Arms	Duration (months)	N*	GPi: mean age (range)	STN: mean age (range)	Note
Burchiel et al., 1999 43	1 (US)	Open	STN vs. GPI (1:1)	12	10	46.5 (35.5–57.5)	62.8 (50.8–74.8)	STN patients older than GPi (P = 0.06)
Anderson et al., 2005 13	1 (US)	Double‐blind	STN vs. GPI (1:1)	12	23	54 (42–66)	61 (52–70)	Age > 80 exclusion criteria.
Nakamura et al., 2007 44	1 (US)	Single‐blind, vs. HC	STN vs. GPI (1:1)	3–7	33	59.6 (57.3–61.9)	59.9 (57.7–62.1)	Unilateral
Rothlind et al., 2007 45	2 (US)	Open	STN vs. GPI (NA)	15**	42***	60.2 (51.4–69.0)	61.4 (51.3–71.5)	Unilateral DBS. 14 GPi and 15 STN underwent a contralateral surgery
Okun et al., 2009 12	1 (US)	Double‐blind	STN vs. GPI (1:1)	7	52	60.2 (54–66.4)	59.8 (49.8–69.8)	Unilateral implants. Age > 75 exclusion criteria.
Follett et al., 2010 16	13 (US)	Double‐blind	STN vs. GPI (1:1)	24	299	61.8 (53.1–70.5)	61.9 (53.2–70.6)	Extension of Weaver et al., 2009
Robertson et al., 2011 46	1 (US)	Double‐blind, vs. HC	STN vs. GPI (1:1)	6	27	65.5 (56.9–74.1)	63.8 (57.5–70.1)	Part of the cohort in Follett et al., 2010
Rocchi et al., 2012 47	1 (US)	Open, vs. HC and non‐DBS PD	STN vs. GPI (1:1)	6	29	61.1 (52.7–69.5)	61.4 (55.9–66.9)	22 of 29 patients from the cohort in Follett et al., 2010
Weaver et al., 2012 32	13 (US)	Double‐blind	STN vs. GPI (1:1)	36	198	60.4 (52.1–68.7)	60.7 (51.8–69.6)	Longer term follow‐up of Follett et al., 2010
St George et al., 2012 48	2 (US)	Double‐blind, vs. HC and non‐DBS PD	STN vs. GPI (1:1)	6	24	62.8 (54.7–70.9)	61.6 (55.8–67.4)	Part of the cohort in Follett et al., 2010
Odekerken et al., 2013 15	5 (NL)	Open	STN vs. GPI (1:1)	12	128	59.1 (51.3–66.9)	60.9 (53.3–68.5)
Weintraub et al., 2013 49	13 (US)	Double‐blind, vs. BMT	STN vs. GPI (1:1)	6	108	NA	NA	Same cohort of Weaver et al., 2009 and Weaver et al., 2012
Okun et al., 2014 50	1 (US)	Open	STN vs. GPI (1:1)	12	30	60.1 (54.6–65.6)	58 (47.3–68.7)	Unilateral, same cohort of Okun et al, 2009. 6 STN and 3 GPi proceeded to bilateral DBS
St George et al., 2014 51	1 (US)	Double‐blind, vs. HC	STN vs. GPI (1:1)	6	28	61.2 (51.9–70.5)	61 (55.3–66.7)
Odekerken et al., 2015 52	5 (NL)	Open	STN vs. GPI (1:1)	12	121	59.2 (51.5–66.9)	60.3 (52.9–67.7)	Part of same cohort of Odekerken et al., 2013
St George et al., 2015 53	1 (US)	Double‐blind, vs. HC	STN vs. GPI (1:1)	6	21	62.8 (54.6–71)	62 (56.3–67.7)
Odekerken et al., 2016 30	5 (NL)	Open	STN vs. GPI (1:1)	36	90	59.1 (51.3–66.9)	60.9 (53.3–68.5)	Part of same cohort of Odekerken et al., 2013

^*Intention to treat.

^**Patients were implanted unilaterally for the first 6 months.

^***23 GPi and 19 STN.

Abbreviations: BMT, best medical treatment; GPi, globus pallidus pars interna; HC, matched healthy controls; NA, not available; NL, The Netherlands; STN, subthalamic nucleus; US, United States of America.

Surgical Complications and Comorbidities

Although the rate of complications related to DBS is low, patients can face them in all stages of the surgical treatment. There are intraoperative complications—eg difficult airway management, confusion, uncontrolled hypertension, pulmonary embolism—which may occur in up to 16% of the patients. ³³ Usually, these complications are manageable and an undesired outcome is rare. The most serious are stroke, intracranial hemorrhage, seizures, and, more rarely, pneumocephalus. ⁵⁴ If we consider that the elderly population is prone to more comorbidities and surgical risks, it is feasible to conclude that these patients have a greater chance of a poor result. ⁵⁵ However, a retrospective analysis of 1757 DBS patients, in which 7.3% were older than 75 years, did not demonstrate a higher rate of surgical complications in this age‐range compared to the others. ⁶ However, this study received criticism for suggesting that the therapeutic window for DBS should be expanded beyond the traditional cut‐offs, as it did not evaluate aspects that are relevant for these patients' age group, ie QoL, axial motor symptoms and cognition. ⁵⁶

Comorbidities in the aging PD population can increase the risk of post‐surgical complications and therefore also represent relative or absolute contraindications to DBS (Table 1). ⁵⁴ An approach that involves a multidisciplinary team is needed to achieve the best possible management of these comorbidities before DBS implantation. Most of these conditions have indeed treatment options (eg hypertension, diabetes mellitus, etc.), although their long‐term effects, such as chronic brain vasculopathy, still increase the surgical risks and hamper the response to stimulation. There are concerns about the use of anticoagulants and antiplatelet therapy, once there is paucity of studies that evaluated this issue. The available recommendations are similar to other surgeries, ie discontinuing these drugs prior to DBS surgery. ⁵⁷ Regarding cardiac pacemakers, there is some weak evidence from small studies suggesting that both devices do not appear to interfere with each other. ⁵⁸

Effect of Cognition

One important concern when selecting candidates for DBS is patient's cognitive profile, particularly when selecting older patients, as age has been pointed as one of the predictors for cognitive decline after STN DBS. ⁵⁹ However, the influence of age on cognitive impairment after DBS is still unclear, mainly due conflicting findings in different studies.

Results of a clinical trial aiming to evaluate the neuropsychological impact of STN and GPi DBS in PD suggested that STN DBS had a greater negative impact in cognitive tests (Stroop word reading, Stroop color naming and Trail Making Test part B) after a 12 month‐follow‐up, and found that age and semantic verbal fluency at baseline were the only predictors of cognitive decline. ⁵² Other small studies have also suggested that older age is a predictor of impaired verbal fluency after surgery. ⁶⁰

Contrarily, a long‐term follow up study observed that age at surgery had no correlation with the rate of cognitive decline after STN DBS, found to be 1.6 points/year in the Mattis Dementia Rating Scale. ⁶⁰ Instead, disease duration positively correlated with conversion to dementia. ⁶¹ Similarly, the results of a meta‐analysis of 28 cohorts (with a total of 612 patients) reporting neuropsychological results after STN DBS found no association between the postoperative changes in verbal fluency and patient age, disease duration, stimulation parameters, or change in dopaminergic dose after surgery. ⁴² These contradictory results clearly indicate the need for more predictive studies and, for the time being, a very careful assessment of patients' cognitive reserve when considering DBS in the elderly.

Outcomes of DBS in the Elderly with PD

Few studies exploring the effects of DBS (targeting STN in particular) in elderly patients have been carried out, mostly involving small cohorts and generally reporting conflicting findings due to the variability of outcomes under scrutiny. In fact, the most relevant DBS publications on approved targets to this date have enrolled relatively young patients (Table 3). Furthermore, studies enrolling old patients are inevitably biased, likely reflecting a small percentage of people that have passed criteria for l‐dopa responsiveness, brain atrophy, lack of comorbidity, etc.

TABLE 3.

The age of PD patients enrolled in randomized controlled trials of approved DBS targets

Study	N. of Centers (Country)	Blinding	Arms	Duration (Months)	N*	Mean Age (Range)	Note
Deuschl et al., 2006 11	10 (Germany and Austria)	Open	STN vs. BMT (1:1)	6	156	60.5 (53.1–67.9)	Age > 75 exclusion criteria
Schupbach et al., 2007 62	1 (France)	Open	STN vs. BMT (1:1)	18	20	48.4 (45.1–51.7)	Age > 55 exclusion criteria
Witt et al., 2008 63	10 (Germany and Austria)	Open	STN vs. BMT (1:1)	6	123**	60.2 (52.3–68.1)	Ancillary protocol of Deuschl et al, 2006
Weaver et al., 2009 14	13 (US)	Single‐blind	STN or GPi vs. BMT (1:1:1 ratio)	6	255	62.4 (37–83)	Age > 69 in 25% of patients.
Williams et al., 2010 64	13 (UK)	Open	STN or GPi*** vs. BMT (1:1)	12	366	59 (37–79)
Okun et al., 2012 65	15 (US)	Open	STN vs. SHAM stimulation (3:1)	12	136	60.6 (52.3–68.9)
Schuepbach et al., 2013 39	8 (France) and 9 (Germany)	Open	STN vs. BMT (1:1)	24	251	52.9 (46.3–59.5)	EARLYSTIM trial (see text)
Witt et al., 2013 66	1 (Germany)	Retrospective	STN vs. BMT (1:1)	6	62	59.8 (52.3–67.3)
Charles et al., 2014 67	1 (US)	Single‐blinded	STN vs. BMT (1:1)	24	30	60 (51–74)	Inclusion criteria: age 50–75
Troster et al., 2017 68	15 (US)	Open	STN vs. SHAM stim (3:1)	12	136	60.6 (41–78)	Secondary analysis of Okun et al., 2012
Hacker et al., 201869	1 (US)	Single‐blinded	STN vs. BMT (1:1)	24	30	60.5 (50.2–69.5)	Secondary analysis of Charles et al., 2014
Lhommée et al., 2018 70	8 (France) and 9 (Germany)	Open	STN vs. BMT (1:1)	24	251	52.9 (46.3–59.5)	Secondary endpoints of Schuepbach et al., 2013

^*Intention to treat.

^**Per‐protocol.

^***The surgical target (either STN or GPi) was not randomized but decided by each team according to their best clinical judgment at the time of patient's recruitment.

Abbreviations: BMT, best medical treatment; GPi, globus pallidus pars interna; NA, not available; STN, subthalamic nucleus; UK, United Kingdom; US, United States of America.

One of the first studies raising awareness on the limitations of DBS in elderly was published in 2004. ³³ The authors evaluated the outcomes of DBS in a sample of 52 consecutive patients, divided in three groups: under 60, 60 to 70, and over 70 years of age. They observed that the oldest group had similar improvement of dyskinesia, motor fluctuations, and motor scores in the off‐medication state, albeit a worsening was found in the on‐medication motor scores, axial subscores, and activities of daily living (ADL). ³³ Other groups have agreed that in older patients with poorer DBS outcomes, the axial subscores (i.e. gait, balance and posture) tend to decline, perhaps suggesting a predictive value for axial motor scores at baseline. ⁷¹

A study compared the effects of STN DBS in patients younger than 70 (N = 56) versus 70 and older (N = 16), with comparable disease severity but lower l‐dopa response in the latter group. ⁷² Despite a great improvement of tremor and Levodopa equivalent daily dose (LEDD) reduction in both groups, the elderly group had a suboptimal improvement of rigidity, akinesia and ADL. The improvement in axial symptoms was not statistically different between the two groups, however, dysarthria was more frequent in old patients. ⁷² Another small study evaluating the short and long‐term outcomes of DBS in 37 patients over 70 found that patients had a positive outcome with respect to motor scores and LEDD reduction; ⁷³ however, this study had a high heterogeneity in terms of targets implanted and duration of follow‐up. Also, it failed to assess important outcomes such as QoL, ADL, and axial subscores. ⁷³

A larger study including 151 patients compared the response to STN DBS in two different age groups: younger (N = 114) and older (N = 37) than 70 years of age. ⁷⁴ The authors found similar improvement in the motor scores measured by the Unified PD Rating Scale (UPDRS) part III, but the older group had a more substantial reduction of LEDD and number of daily medication doses. ⁷⁴ Once again, the interpretation of these findings are limited by the lack of assessment of more patient‐relevant outcomes.

Patient‐Centered Outcomes

When QoL is taken into consideration, results of studies are less controversial. A prospective open‐label study conducted by the Non‐Motor PD study group within the International Parkinson's and Movement Disorder Society evaluated the effects of STN DBS in 120 patients according to the age groups (≤59, 60–69, and ≥70 years) with comparable disease duration and disease severity at baseline. ⁶⁰ The authors reported that despite of a significant improvement in QoL measured by the PDQ‐8 in all age groups, the group aged 70 and older had a lower effect size (0.42) in comparison to 0.83 and 0.59 in the groups younger than 59 and 60 to 69, respectively. ⁶⁰ The same reduced effect size with older age was seen for motor benefit and ADL. ⁶⁰

Other small studies have addressed postoperative QoL in elderly patients. A retrospective study evaluated the response to STN DBS up to 4 years in a group of 30 patients with an average age of 77.5 years (range 75.0–84.5) and mean disease duration of 11.8 years. ¹⁸ Authors reported that there was an improvement of motor scores and motor complications, but no improvement in QoL and axial subscores. ⁷⁵ Considering a different age cut‐off of 65 years, another small study with 20 subjects in each age group also found that improvement in QoL (measured with the PDQ‐39) was not significantly different between age groups 1 and 2 years after STN DBS. ⁷⁶ The authors also addressed caregiver burden using the Zarit care burden interview, which had comparable improvement in both groups. ⁷⁶

Effect on Survival

In addition to QoL, survival is another concern when managing elderly patients with PD. A study evaluating the 10 year‐survival rates in a single‐center cohort of 320 patients with PD after DBS found a survival probability of 51%, with a mean age of 73 years at the time of death. The main predictors of death were age at the time of DBS implantation (Hazard ratio (HR) 1.02) and male sex (HR 1.42), independently from the number of medical comorbidities. ⁷¹ Similarly, a retrospective study on 79 PD patients with STN DBS found a 70% survival after 10 years, with an age at surgery above 60 years increasing the mortality by 2.3 fold. ⁷⁷

Alternatives to DBS Therapy in the Elderly

DBS is not the only advanced option for PD patients, and this should be emphasized especially when evaluating frail patients, including the elderly. Other alternatives could be contemplated, but there is no evidence of superiority versus DBS, neither the role of advanced age in selecting these alternative options (Table 1).

One option is l‐dopa‐carbidopa intestinal gel (LCIG), which provides a more regular plasmatic level of l‐dopa, thus improving motor fluctuations and dyskinesias. ⁷⁸ Being delivered through a percutaneous gastrostomy, LCIG is minimally invasive and can be an option for patients with contraindications for brain surgery or not willing to undergo major surgery. ⁷⁹ LCIG major complications are abdominal pain, peritonitis, neuropathy, tube‐related issues. In the elderly, special attention should be given to the nutritional status, as LCIG can induce vitamin B6 or B12 deficiency resulting in neuropathy and cognitive decline. No studies have, however, specifically addressed the effect of advanced age on LCIG outcome and safety, mean age of enrolled patients being 65 years. ⁷⁹

Surgical ablation of the Vim (thalamotomy) or GPi (pallidotomy) are other alternatives for frail PD patients, and have been considered reasonable solutions for patients with contraindication to DBS experiencing severe tremor or dyskinesias (Table 1). ⁸⁰ Indeed, LCIG might not be a good option in these cases as tremor is well‐known to be often l‐dopa‐resistant and also because troublesome dyskinesias might not be well controlled especially in the long term. ⁸¹

Ablations can be performed with radiofrequency (over 40 years of experience and widely available), radiations (gamma knife radiosurgery) or focused ultrasound (FUS). FUS is an attractive option given the scalpel‐free surgery, with virtually no risk of infection or stroke. ⁸² Gamma knife radiosurgery is also scalpel‐free but mainly useful in patients not able to temporarily discontinue anticoagulation therapy. The major limitation of the surgical ablation is the potential risk of permanent deficits, the limitation to unilateral procedures, and specific technical issues. For example, FUS cannot be performed in patients with unfavorable cranial thickness or contraindications to magnetic resonance imaging. ⁸³ , ⁸⁴

Conclusions

Choosing the right PD treatment is a delicate task, which is even more complicated in old patients. There are many variables to be pondered and different demands to be attended. It is relatively easy to select young patients for DBS and identify the best target, as this is the STN most of the time. More complex is the scenario later on due to the limited scientific evidence and lack of head‐to‐head randomized trials.

The gold standard of PD treatment in the geriatric population is identifying potential benefits while mitigating surgical risks. It is pivotal to this equation to consider the needs and expectations of patients and their families as well as the awareness and understanding of all the possible options, not just DBS.

Despite of preliminary suggestions that age might affect the outcomes of DBS, to date the evidence to support the hypothesis of age as an independent predictor of DBS outcomes is limited and results are controversial. ⁸⁵ In most studies the overall samples are small and the elderly population is underrepresented (Tables 2 and 3), therefore limiting the power of comparisons. Also, the different age cut‐off used in the studies, the different targets implanted (with very little attention to GPi), and the different outcomes measured preclude any definitive answers. Although most of the studies concluded that DBS is a safe choice in patients over 70 years old, there are many unanswered questions as outlined in the present review.

Improvement of QoL should be the goal of any given therapy for PD, and despite the well documented improvement of QoL after DBS, there is a debate regarding the magnitude of effect and domains affected by DBS in elderly PD patients. ¹⁸ Beyond age, other factors such as l‐dopa response, severity of disease and QoL at baseline, as well as severity of motor fluctuations and dyskinesia may play a role, and to date there is no single biomarkers to predict DBS outcomes. ⁷⁵ Ultimately, finding an age‐independent biomarker predicting DBS outcome is the final goal to expand this powerful treatment to all patients age in an effective and safe manner.

Author Roles

(1) Research Project: A. Conception, B. Organization, C. Execution; (2) Manuscript Preparation: A. Writing of the First Draft, B. Review and Critique.

P.A.: 1A, 1B, 1C, 2A

C.C.A.: 1A, 1B, 1C, 2A, 2B

A.F.: 1A, 1B, 1C, 2A, 2B

Disclosures

Ethical Compliance Statement

The authors confirm that the approval of an institutional review board was not required for this work. No informed consent was required. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this work is consistent with those guidelines.

Funding Sources and Conflicts of Interest

No specific funding was received for this work and the authors have e no conflicts of interest relevant to this work.

Financial Disclosures for Previous 12 Months

Dr. Alfonso Fasano received honoraria from Abbott, Abbvie, Boston Scientific, Brainlab, Ceregate, and Medtronic.