From Journal of Neuroscience Nursing



Stimulation of the Subthalamic Nucleus for the Treatment of Parkinson's Disease: Postoperative Management, Program-ming and Rehabilitation



R. Malcolm Stewart, J. Michael Desaloms, Manjit K. Sanghera

Abstract and Introduction

Abstract

High-frequency stimulation of the subthalamic nucleus is a neurosurgical procedure for the alleviation of motor symptoms of Parkinson's disease and debilitating medication-induced dyskinesias. Stimulation is achieved with electrodes implanted stereotactically in the subthalamic nucleus by a neurosurgeon specializing in stereo-tactic surgery and a team composed of an anesthesiologist, a neurophysiologist, certified nurses and nurse practitioners and, at some centers, a neurologist. The teamwork continues in the recovery room and the inten-sive care unit, where the patient may experience transient adverse behavioral effects. Two weeks after sur-gery, the neurostimulator is activated and programmed. The medications also are adjusted to complement stimulation to maximize the therapeutic effects and minimize the stimulation-induced side effects. For those patients who are deconditioned or have major speech, gait, or balance problems, rehabilitation therapy is em-ployed.

Introduction

Parkinson's disease (PD) is a neurodegenerative disorder affecting approximately 1% of the population in the United States. The disease is characterized by resting tremors, rigidity, bradykinesia, postural instability, and gait disability. Many of these symptoms improve with levo-dopa (L-dopa) and dopamine agonist therapy. However, as the disease progresses, the medication becomes less effective or produces debilitating side ef-fects, such as motor fluctuations and dyskinesias. The failure of medical therapy to provide long-lasting relief of symptoms and concomitant improvements in neuroimaging, intraoperative neurophysiological mapping, and neurosurgical stereotactic technique have led to surgical approaches to the treatment of PD.

High-frequency stimulation of the subthalamic nucleus (STN) is one neurosurgical procedure currently avail-able for the alleviation of motor symptoms of PD and the debilitating medication-induced dyskinesias that of-ten accompany the disease. Stimulation of the STN is achieved by implanting deep brain stimulation (DBS) electrodes precisely in the nucleus.

The success of this treatment is directly related to the expertise of a team of professionals involved in the pa-tient's care. This team comprises a movement disorder specialist who diagnoses PD and selects the appro-priate patient for surgery, a neurosurgeon specializing in stereotactic surgery, and a team composed of an anesthesiologist, a neurophysiologist, and advanced practice nurses. Postoperatively, the teamwork continues in the recovery room and in the intensive care unit, where the patient may experience transient adverse be-havioral effects, such as hallucinations.

At Presbyterian Hospital of Dallas, DBS electrodes have been implanted in the STN of 78 patients. A previous article (Sanghera, Desaloms, & Stewart, 2004) reported on the criteria for patient selection, preoperative pa-tient preparation, localization of the STN, and the surgical implantation of the DBS electrodes in the STN. This article reports on the postoperative well-being of the patient, programming of the neurostimulator, complica-tions resulting from the surgery, and the occurrence of hardware-related problems.



Postoperative Management

Although most patients who undergo the procedure may go home 2 days after surgery, those who are older and those with more advanced disease may require a longer inpatient stay. In addition, those patients who experience confusion, hallucination, and/or disorientation may be prescribed quetiapine and kept under ob-servation for several days. When the patient goes home, hydrocodone (Vicodin 50 mg/500 mg) is prescribed for pain, to be taken on an as-needed basis. Darvocet-N 100 mg is prescribed if the patient is allergic to co-deine. If a patient develops narcotic-related confusion, acetominophen alone may be used.

Some patients may be tempted to pick at their suture wounds compulsively, increasing their risk of intracranial infection and possibly necessitating the removal of the implanted DBS leads. Quetiapine can be prescribed to reduce this compulsive behavior.

Patients are encouraged to continue with their presurgery diet and increase their exercise regime to reduce deconditioning and avoid significant weight gain, which has been reported to follow DBS surgery (Barichella et al., 2003; Macia et al., 2004). If patients experience postoperative dyskinesias, anti-Parkinson's medication that may be restarted after surgery at the discretion of the neurologist may be reduced or taken with food to slow down the absorption process. Some patients also may be encouraged to enter a rehabilitation program after the wounds have healed to improve their functional ability.

Following removal of the sutures, which can occur in the physician's office, the neurologist programs the neu-rostimulator on an outpatient basis. For those patients in rehabilitation, neurostimulator programming will be-gin in the hospital. In addition, patients experiencing speech problems (e.g., an abnormally soft voice) may undergo the Lee Silverman Voice Treatment (LSVT) therapy on an outpatient basis (Ramig, Fox, & Sapir, 2004).



Programming the Neurostimulator

The neurostimulator, also known as the implantable pulse generator (IPG), is a pacemaker-like device pow-ered by a small battery that is implanted subcutaneously near the clavicle. It contains a computer chip that can be programmed to send electrical pulses to the electrode contacts to control the PD symptoms. Currently, two types of neurostimulators are available (Fig 1). The Soletra accommodates one DBS lead and one extension (Fig 1A). Thus, two Soletra neurostimulators are required for bilateral therapy. The newer Kinetra dual-channel neurostimulator (Fig 1B) provides bilateral neurostimulation from a single neurostimulator and ap-pears to be the system of choice with neurosurgeons because it requires fewer incisions and reduces the du-ration of the surgery. However, the Kinetra is somewhat heavier and larger than the Soletra and is best suited for larger patients. Utimately, the choice of the neurostimulator rests with the patient and the neurosur-geon.

The primary goal of programming is to set the stimulation parameters to optimize symptom management with minimum adverse effects. Programming usually is conducted after the patient has been off his or her Parkin-son's medication for approximately 12 hours. Ideally, the person who does the programming also should be able to change medication based upon the patient's response to the stimulation.

At the start of programming, the function of each contact is tested by measuring the current flow and imped-ance between (a) each contact as the cathode (-) and case (of the neurostimulator) as the anode (+), and (b) between each of the four contacts as the cathode and/or the anode. A current flow between a contact and case, or between contacts, of < 7 µA with an electrode impedance of > 2000 ohms, would suggest an open or an incomplete circuit. A short circuit, on the other hand, has a low electrode impedance (< 50 ohms), perhaps due to damage to the electrode insulation, leading to a high current flow (> 500 µA) that may cause tissue damage. The presence of either an open or a short circuit renders those particular contact(s) ineffective for stimulation (Table 1).

Once the viability of each contact has been established, a stimulation profile is performed to determine the clinical response to stimulation of each of the four electrode contacts (Fig 2A). In turn, each contact is set as the cathode and the case as the anode. This is monopolar stimulation (Fig 2B). With the pulse width at 60 µsec and the frequency at 130 Hz, the amplitude of stimulation is gradually increased from 0.1 to 3.7 V, in in-crements of 0.1-0.3 V. The stimulation threshold is determined for the best overall beneficial motor effect, as well as for the least acute adverse effects. Adverse effects that may occur include paraesthesias, muscle con-tractions, double vision, and mood disturbances (e.g., anger or depression). These transient occurrences re-flect stimulus spread to the medial lemniscus, internal capsule, ocular motor nerve nuclear complex, and the limbic portion of the STN, respectively. Stimulation is kept below 3.7 V because any voltage above this value leads to a doubling of the battery drain. To reduce voltage to below 3.7 V, the pulse width can be increased to 90 µsec or higher, and the frequency can be increased to 185 Hz. The electrode contacts whose stimulation results in the greatest beneficial effect with the fewest side effects is selected for chronic stimulation.

After selecting the contact(s) with the best monopolar stimulation profile, bipolar stimulation may be used to reduce the stimulus spread if an adverse effect is induced by the more widespread monopolar stimulation. Bi-polar stimulation induces current flow between the contacts. Each contact can be programmed to be either an anode (+) or a cathode (-). This type of stimulation creates a narrow, more focused current field with an in-tense effect around the cathode (Fig 2C), in contrast with monopolar stimulation in which current radiates in all directions (Fig 2B). The stimulus spread with bipolar simulation is more trapezoidal and flows from the cath-ode (-) to the anode (+). Sometimes reversing the cathode and the anode can improve the response because different anatomical structures may be stimulated.

Initial programming may take 2-3 hours to complete. Since stable therapeutic changes with STN stimulation take some time to develop, the patient should be asked to remain in the office for at least 1-2 hours after pro-gramming is completed to allow for observation of any delayed effects and for any needed fine adjustments to be made. Tremors and balance response to stimulation occur within minutes to hours, but improvement in ri-gidity and bradykinesia may take up to 36 hours (Temperli et al., 2003). Importantly, there may be a delay of 12-18 hours before dyskinesias occur. The patient should be advised of this complication and shown how to use a magnet to turn the neurostimulator on and off, if necessary.

It usually is not possible to alleviate completely all Parkinson's symptoms with the stimulator alone; therefore, medications often are used to ensure additional improvement. It is optimal to have the same person adjust the stimulator and the medication. In general, frequent adjustments (1-2 times a month) over a 3-month period are necessary to achieve good results. As adjustments are made to the stimulation parameters, doses of medica-tion also may have to be altered. After 3 months, the patient may come in for fine adjustments on an as-needed basis.

With extended usage, the intrinsic voltage of the neurostimulator will decrease. When the voltage of the bat-tery falls below 3.6 V, the battery can be replaced on an outpatient basis using general anesthesia or intrave-nous sedation. The Soletra gives very little indication of a battery failure; often, the sudden appearance of ad-ditional Parkinsonian symptoms is the only indication the battery has failed. Many patients purchase the Sole-tra Access Review, a handheld device that allows the patient to check the on/off status of their neurostimulator as well as whether the battery charge is sufficient. The Access Review is much easier to use than the magnet in turning the neurostimulator on and off, but it does not allow any patient control of the stimulation parame-ters. Patients who are implanted with the Kinetra neurostimulator may purchase the Kinetra Access Therapy Controller to performs that same functions as does the Access Review, as well as allowing the patient to monitor the battery charge of the neurostimulator system. In addition, the Access Therapy Controller allows the patient to fine-tune the therapy by altering the stimulation parameters within limits set by the neurologist. In this way, the patient can control adverse side effects and maximize the effectiveness of the stimula-tion.

Patients might decide to turn off the neurostimulator at night to preserve the life of the battery. Although this practice is acceptable for essential tremor, it is not recommended for cases of PD because some Parkinson-ian symptoms, such as rigidity, respond only to continuous stimulation; by turning off the neurostimulator at night, the patient may experience more difficulties in the morning.



Adverse Effects

Complications may arise at any time before or during surgery and up to several years afterward (Table 2). It is essential to recognize the possibility of complications and minimize them as much as possible. Patients se-lected inappropriately for DBS surgery may have little or no beneficial response (Sanghera et al., 2004). Sur-gery may exacerbate dementia, memory disorders, severe depression and psychiatric disorders in some DBS patients (Trepanier et al., 2000). Those diagnosed with atypical Parkinsonism, diabetes, and cardiovascular disease also may have a poor surgical outcome.

Other complications relate to the surgical procedure itself. Most of these complications are transient, occurring during the first few days after the surgery, but others may be permanent. In the longest follow-up study pub-lished, Krack et al., (2003) reported on the effect of bilateral stimulation of the STN in 49 PD patients 5 years after the surgery. They found that the most common transient adverse effects patients experienced from sur-gery were delirium (24%); asymptomatic bleeding detected only on the MRI (16%), problems with wound heal-ing (8%), confusion (6%), hemiballism (2%) and seizures (2%). Other serious but transient effects reported were skin infection, head trauma due to falls, pulmonary embolism, hemiparesis secondary to hemorrhage, subdural hematomas, venous infarction, cerebrospinal fluid leak and skin erosion (1%-2%) (Deep Brain Stimu-lation for Parkinson's Disease Study Group, 2001; Krack et al.; Umemura et al., 2003).

Long-term adverse effects seen with STN stimulation may be related to incorrect placement of the DBS elec-trodes. In other situations, stimulation may be ineffective. In either case, there is no alternative but to remove and accurately replace the electrodes in the STN during a second surgery. Adverse effects also may be due to excessive stimulation that causes current to spread to adjacent structures. Decreasing the intensity of stimulation will reverse any adverse effects associated with stimulus spread. On the other hand, the presence of disabling dyskinesia with excessive STN stimulation indicates correct electrode placement. Often, there may be only a very narrow division between parameters that alleviate motor symptoms and those that induce dyskinesia. In addition, not all motor symptoms may be alleviated by stimulation. In such cases, careful ad-justment of stimulation parameters and medications are needed for a favorable outcome. The expertise of the programmer in balancing stimulation and medication is one of the most important aspects of achieving a fa-vorable outcome.

Other commonly reported adverse effects related to stimulation include dysarthria, diplopia, and paresthesia (reflecting stimulus spread to the internal capsule, ocular motor nerve complex, and medial lemniscus respec-tively). Significant weight gain also has been reported following DBS surgery. Although the exact cause of this weight gain is not known, it could be due to a reduction in the energy expenditure as dyskinesias and tremors lessen (Barichella et al., 2003; Gironell, Pascula-Sedano, Oterin, & Kulisevsky, 2002). It also has been sug-gested that postoperative weight gain could be a result of a reduction in dopamine therapy that frequently pro-duces nausea (Ahlskog, 2001). A regional effect of STN DBS on the hypothalamic satiety center also may be involved (Macia et al., 2004).

Cognitive evaluations conducted 1 year after DBS have shown changes in patients' verbal fluency (distur-bance in language or aphasia), global cognitive abilities, memory, attention, and executive frontal lobe func-tions (Saint-Cyr, Trepanier, Kumar, Lozano, & Lang, 2000). However, only decline in verbal fluency was found to be permanent and consistent across all studies (Woods, Fields, & Troster, 2002); one study found verbal fluency declined in 16% of subjects (Ardouin et al., 1999). Deficits in verbal fluency reflect changes in frontal lobe function. Dysarthria—a mechanical speech problem—also can occur through involvement of corticobul-bar projections. Dysarthria may respond to a reduction in DBS stimulation as well as speech therapy.

Most studies have shown that STN stimulation results in a reduction in mood disturbances, such as depres-sion and anxiety. This finding most likely is related to easing of PD motor symptoms (Higginson, Fields, & Troster, 2001). Conversely, an increase in mood swings or depression may be present in patients who are disappointed that they did not attain the level of beneficial effects they expected. Studies also have reported an increased level of anxiety and depression, episodes of mania/hypomania, apathy, hallucination and de-mentia with STN stimulation (Krack et al., 2003). Sometimes, a patient will exhibit acute depression or anger when certain contacts are used. This behavior immediately improves when the stimulation is changed to a dif-ferent contact. In the 5-year follow-up study conducted by Krack et al. (2003), 6% of the patients displayed hallucinations, 6% displayed apathy, and 10% developed dementia. Whether these conditions are related to surgical treatment or disease progression is not known. The most conservative interpretation of these findings is that DBS is fairly effective in providing symptomatic relief for the motor disturbances of PD, but does not slow down the progression of the nonmotor symptoms. Death rate, including suicide attributed to DBS surgery ranges from 1.8% (Umemura et al., 2003) to 6% (Krack et al., 2003). However, the general consensus is that STN stimulation does not change overall cognition among PD patients. In patients with chronic psychosis (hal-lucinations) following surgery, quetiapine may be needed. Postoperative neuropsychological evaluation can be useful in recognizing some of these complications. In addition, preoperative neuropsychological evaluation is used to recognize patients at risk for developing postoperative psychosis and may rule them out as candi-dates for the DBS procedure.

Postoperative complications resulting from implanted hardware also may arise (Joint, Nandi, Parking, Greg-ory, & Aziz, 2002; Kondziolka, Whiting, Germanwala, & Oh, 2002; Oh, Abosch, Kim, Lang, & Lozano, 2002). The extracranial wires from the neurostimulator may fracture because of inadvertent torque placed on the wire during implantation and suturing of the neurostimulator; or the neurostimulator may migrate postoperatively in its pocket, and the wires may twist and break. The reported overall infection rate of the DBS system is 2% to 10% (Umemura et al., 2003). If an intracranial infection occurs, the system most likely will have to be removed because treatment with antibiotics alone is generally ineffective (Umemura et al., 2003). Nonetheless, the au-thors have had one patient with an intracranial infection from wound picking who was treated successfully with antibiotics; no hardware needed to be removed.

Drifting of the implanted DBS electrode also may occur if the electrode is not anchored properly in the burr hole. There also may be a malfunction of the electrode leads or extension due to breakage or a short circuit. These problems can be diagnosed by using X ray examination, observing the presence of electrical signals along the implanted stimulating pathway, or measuring the circuit impedance (which would be less than 50 ohms). Under these conditions, replacement of the leads or extension is necessary.

Complications also may occur due to the environment. The neurostimulator may inadvertently be turned off by external magnetic devices, such as used in security scanners, refrigerators, or older computers. The im-planted system also may be affected by cardiac pacemakers, defibrillators, ultrasonic dental cleaning equip-ment, electrocautery, and radiation therapy (Nutt, Anderson, Peacock, Hammerstad, & Burchiel, 2001). While a head MRI may be performed with the system turned off, a chest or thoracic spine MRI is prohibited even when the neurostimulator is turned off because they generate larger magnetic fields that can heat and dam-age the implanted electrodes. In addition, the use of diathermy can have an adverse effect because diather-mal energy transfer through the implanted system can cause tissue damage that can result in severe injury and death. If the neurostimulators are placed too close to each other, the programming of one neurostimulator can affect the programming of the other. In this case, an aluminum pie plate can be placed perpendicular to the chest between the two neurostimulators to reduce energy transfer during the programming session. Once the programming is complete, there is no cross talk between the neurostimulators. The presence of a DBS electrode does not preclude the subsequent implantation of a cardiac pacemaker. The converse is also true, as long as CT is used for brain imaging.



Rehabilitation

The goal of rehabilitation is to restore the patient's motor function and have the patient achieve the highest level of functional independence. Generally, patients with PD are severely deconditioned. The authors' pro-gram endeavors to optimize the benefit of surgery by employing allied health modalities, including speech, physical, occupational, and recreational therapies. It is felt that an optimization of response is obtained by combined modality therapy. Depending on the needs of the patient, therapy may be conducted for several weeks on an outpatient basis or for 10-14 inpatient days following surgery. Generally, a physical medicine consultation is obtained when the patient has completed the surgery to help in planning the postsurgical ther-apy regimen.



Family and Caregivers

Stress is a major issue for patients and their families. Stress can be caused by the physical, emotional, behav-ioral, or financial challenges of living with PD. Stress can change the relationships between patients and care-givers. Some patients will accept the need to slow down, but many are resentful of the loss of their independ-ence. Poor adjustments and unrealistic expectations can negatively affect marriages and other relationships. A successful surgery often can reduce this stress, but newfound patient independence also can adversely af-fect relationships and may even lead to divorce. Other practical issues that should be addressed include loss of job and income, sexuality changes, self-esteem changes, and caregiver stress. Referral to a counselor, neuropsychiatric nurse practitioner, or psychiatrist might be necessary.

Another issue that should be addressed is the expectation of what the surgery will accomplish. While medical personnnel understand the complex nature of the surgery and have realistic expectations regarding outcomes for the surgery, patients and their families usually do not have a complete comprehension of the procedures. Patience, encouragement, and continued education is necessary to help them understand what to expect. Pa-tients often find it helpful to have the opportunity to discuss the procedure and its aftermath with other patients who have undergone DBS. Age- and needs-appropriate support groups also are useful.