Cerebral: surface


Chapter 2

Stimulation of the brain cortex has emerged as one of the most promising avenues for palliation of neurological and psychiatric conditions. Cortical stimulation (CS) can be administered either noninvasively (NICS) ( Box 2.1 ) or invasively (ICS) [ , ].

Box 2.1
Common modalities of CS [ , ].

  • 1

    Transcranial Magnetic Stimulation (TMS) : TMS in a repetitive mode (rTMS) is a noninvasive technique with a benign side effect profile and is easy to administer. Standard TMS consists of a brief, high-intensity current pulse in the coil of a wire which produces a magnetic field that can reach up to 2 T and lasts for approximately 100 ms. Parameters that modulate the geometry of the induced electric field include the type and orientation of the coil and the waveform of the magnetic pulse as well as pulse intensity. Patients are awake, alert, and sit in a comfortable chair during treatment sessions, which are approximately 20–30 min in duration. Cardiac or neural pacemakers, metal clips in the brain and previous seizures are a contraindication. Side effects are transient and include scalp discomfort, headache, sleepiness, dizziness, or an increase in pain. Ear plugs are recommended to avoid hearing damage. The target (e.g., the primary motor cortex—M1—or the dorsolateral prefrontal cortex—DLPFC) is located either with conventional craniometric techniques or image-guided neuronavigation. The motor threshold of the hand area of the unaffected hemisphere is often assessed to calculate the intensity of therapeutic stimulation, but its significance is moot when the target is nonmotor or outside the hand area in M1. Effects can be immediate or build up over several sessions (days). After effects, that is, symptom relief beyond the actual period of stimulation, are highly variable, either being absent or lasting up to 2 weeks; they are unassociated with entrainment of specific frequencies. A hand-held machine for home use is available, but expensive.

  • 2

    Transcranial Direct Current Stimulation (tDCS) : The benefits of tDCS are its low-cost, ease of use, safety, tolerability, and portability. Patients are awake, alert, and sit in a comfortable chair during treatment sessions. The power supply is 10 mA delivered via two surface conductive electrodes covered with saline or gel soaked sponges to reduce impedance. The sizes are 5 × 7 cm (but can be larger or smaller). The anode carries the positive charge and is considered excitatory, the cathode conveys the negative charge and is considered inhibitory; however, subtle variations in technique may have substantial effects. Various parameters of stimulation in different settings have been used, with intensities up to 5 mA (generally 2 mA), in single to repeated sessions on consecutive days, and a duration of 5′-30’. The safe current density delivered is 0.029–0.08 mA/cm2. The target is generally located by using the EEG 10–20 system (e.g., M1: C3/4); the reference is placed on the contralateral supraorbital area or shoulder. However, given the large electrodes used, tDCS is expected to engage nearby areas (e.g., both M1 and S1). In psychiatric conditions, one electrode is placed over the left DLPFC, and the second electrode is placed either over the right DLPFC or over the left supraorbital region. A more focal high-definition tDCS is under study. Side effects include redness under the electrode, tingling, lightheadedness, headache, fatigue, insomnia, anxiety, confusion, nausea, among others: these are generally transient. Exclusion criteria include prior head trauma, pregnancy, and epilepsy.

  • 3

    Invasive Cortical Stimulation (ICS) : ICS requires a neurosurgical approach to place one or more stimulating paddle electrodes on the dura mater overlying the target area (extradural ICS, e-ICS)—or, infrequently, underneath it (subdural ICS, s-ICS). No stimulation leads specifically designed for ICS are available at the present time: quadripolar electrode arrays (5 mm contacts with 1 cm spacing), singly or in combination (for wider coverage), are borrowed from neurosurgical spinal cord stimulation; 8, 16, and even 32 contact paddles are also used. Both constant current or constant voltage stimulators are used. In current-controlled devices, output (measured in Amperes) is kept constant while the voltage is adjusted to keep the current intensity at the desired level. Conversely, in voltage-controlled devices, the output (measured in Volts) is kept constant while the output current is adjusted according to the tissue impedance. The patient either lies down supine or in lateral decubitus (Park Bench). Electrodes are inserted via one or two burr holes or via a flap craniotomy ( Fig. 2.1 ), either under local anesthesia with conscious sedation(neuroleptoanalgesia) or general anesthesia (e.g., TIVA). Clinical outcomes are similar, but burr holes are less traumatic to the patient. If M1 is elected as a target, paddles can be positioned parallel to the central sulcus or perpendicularly, with similar outcomes, at least for chronic pain. ICS does not require stereotactic frames or microelectrode recording. Pulse generators are implanted in the subclavicular region, similar to DBS procedures. Devices that allow sequential testing of multiple contact configurations at very short intervals are available, but are bound to miss effective configurations for some conditions (e.g., pain, tinnitus), given the lack of a straightforward stimulus–effect relationship; also, after effects are not factored in. ICS allows for stimulation with increased spatial specificity than NICS. Absolute exclusion criteria are similar to other neuromodulation techniques and include major depression accompanied by suicidal thoughts or gestures (unless specifically targeted for palliation), major psychosocial stressors (job dissatisfaction, marital problems …), major personality disorders, dementia, substance abuse, poor motivation, malingering. Although referred to as invasive, actually ICS is by far safer than deep brain stimulation (DBS). Complications of DBS can be serious, for example, venous infarction caused by the transection or coagulation of large draining veins at the site of the burr hole, deep infarcts, subdural hemorrhage, intracerebral bleeding, brain abscesses, cognitive and neuropsychiatric disturbances, and suicide. Mortality, although rare, is nontrivial (average: 0.4%). e-ICS, the commonest form of ICS, virtually eliminates the possibility of intracranial hemorrhage; extradural hematomas are exceptional and easily manageable (s-ICS, however, carries morbidity and mortality similar to DBS, with potentially more frequent cerebrospinal fluid—CSF—leaks). Postoperative MRI, including high-field (3T) MRI, can be performed safely [ ].

    Figure 2.1, Methods of implantation of stimulating paddles for invasive cortical stimulation: (A): single burr hole; (B): two burr holes; (C): flap craniotomy; (D): subdural approach.

The two commonest modalities of NICS are transcranial magnetic stimulation (TMS), especially in its repetitive configuration (rTMS) and transcranial electrical stimulation (TES). TES can be administered as constant current stimulation (tDCS) or charge-balanced, alternating sine-wave currents that specifically match the frequency of pathologic rhythms (tACS) [ ].

NICS—as tDCS—is the oldest form of CS and harkens back to Giovanni Aldini's work at the turn of the 19th century, as he administered electrical current to the head of depressed individuals with Volta's Pila [ , ]. The first modern reports of the use of CS for therapeutic purposes appeared in the last decade of the 20th century [ , ]. As for ICS, the first application was reported in 1991 (central pain), followed by Parkinson’s disease in 2000, stroke rehabilitation in 2003, tinnitus in 2004, the vegetative state in 2008, and depression in 2010 ([ , ] for full details). Given the ease of administration, NICS has been studied far more extensively than ICS and this led to regulatory approval of its use for several conditions: in the US, TMS is FDA-cleared as a treatment for major depression (2008), for pain associated with certain migraine headaches (2013) and for obsessive-compulsive disorder (2018); in 2018 the FDA approved ultrafast (intermittent theta-burst stimulation) TMS for depression.

ICS versus NICS

Chronic neuropathic pain

Central pain (CP)

rTMS is not superior to ICS, at whatever frequency (0.2–20Hz) or number of pulses studied. The effect is modest (once sham relief is subtracted), with great individual variability, and generally transient (minutes to days; weeks or more if placebo effect is discounted). The suggestion that repeated weekly sessions can relieve the pain on an extended basis in some patients is supported by long-term studies, but globally most patients will not derive an enduring benefit and a placebo effect most likely accounts for those who do. No predictor of efficacy has been established, including sensory thresholds, location of lesion, gender or age of patients, and duration of pain. rTMS appears to predict short-term efficacy of ICS; however, a negative trial should not be grounds for not implanting a surgical paddle. Dorsolateral prefrontal rTMS is poorly effective or not effective for CP. tDCS appears to be even less effective than rTMS. A confounder in all studies is that patients are always kept on their previous drug regimen and several of these drugs can alter cortical excitability. Other confounders are conscious expectancy of the patient and unconscious pavlovian conditioning, all of which can seriously bias evaluation of efficacy (see review in Ref. [ ]). As for ICS, >50% of all reported patients with CP of brain origin have been relieved >40% on a VAS scale at long term (but sometimes not on the McGill Pain Questionnaire, which is less focused on measuring intensity). However, there are a few patients who are satisfied with 20%–40% relief, at least for some time, due to improved functional ability and quality of life; conversely, others may sometimes report high levels of pain reduction, yet fail to demonstrate any functional improvement. Some patients with CP may be relieved, but the number of treated cases remains small for any strong conclusions. Concurrent drugs can be reduced or even stopped in several cases. Evoked and spontaneous pains respond equally well; nonpainful paresthesias are refractory with current protocols (see review in Ref. [ ]). When analgesia fades off, repositioning of the electrode or intensive reprogramming may restore benefit in some cases, although at a lower level than before. Importantly, granulations and fibrosis around the contacts have been found in some failures: curettage may restore benefit. Currently, the best indicators of a successful implantation include a patient's GABA signature (i.e., drug dissection with, e.g., the propofol test) and preoperative rTMS. There is no synergistic effect from combining ICS and DBS and this combination is actually strongly discouraged. Despite claims to the contrary, DBS per se does not work for the vast majority of CP patients (see full discussion in Ref. [ ]).

Peripheral pains

The best outcomes have been reported for trigeminal Poisson–Nernst–Planck (PNP) (≈70% of responders), for which ICS is superior to DBS, plexopathies, and phantom and stump pains, with results similar to DBS. NICS results are not as good. Other PNP can also respond (e.g., CRPS) [ ].

Movement disorders

Parkinson's disease (PD) and parkinsonism

Currently, e-ICS of the primary motor cortex (M1) represents an alternative neurosurgical treatment for PD when DBS cannot be employed or patients are unresponsive or do not fulfill DBS inclusion criteria (roughly half of all parkinsonians) or simply refuse it. M1 e-ICS achieves a significant and sustained improvement in motor symptoms, with a remarkable effect on axial symptoms, freezing, l -Dopa-induced dyskinesias, and quality of life (DBS is poorly effective on postural instability, freezing and nonmotor symptoms, but has a superior effect on distal symptoms). Although rigidity and, less so, tremor may improve within several minutes of stimulation (unlike DBS), the full effect on bradykinesia, gait, and axial symptoms only grows with time (days to weeks) [ , ]. Unilateral e-ICS induces clinical effects bilaterally and is more cost-effective than DBS. M1 e-ICS may also benefit Multiple System Atrophy-Associated Parkinsonism [ , ] and pure akinesia [ ]. The surgical technique involves the implantation of a paddle electrode placed extradurally parallel to the motor strip overlying the M1 hand knob or the upper limb area [ , ]. ICS is generally performed contralaterally to the most affected side, but on occasion ipsilaterally or bilaterally. Stimulation is generally continuous, low frequency (<50Hz), although higher frequencies have also been reported, always subthreshold for movements and sensory feelings, in almost all cases with bipolar configuration, through the most distal contacts. In M1 ICS, the clinical changes are delayed for several days after switching the stimulator ON or OFF or after modifying the parameters of stimulation, whereas they are durable (weeks to months) after stopping long-term stimulation (After-Effects). This points to ICS-induced plastic rearrangements, which makes the parameters search stage more labor-intensive than DBS: at least 1 month must elapse before a fresh change of parameters [ , ]. Failures of ICS could be put down to individual variations in the organization of M1, that is, mosaicism, inverted disposition of M1 functional areas, wide somatotopic variability of individual distal arm representations, not only between individuals, but also between hemispheres in the same individual [ , ]. Older patients exhibit less distinctive cortical representations and decreased basal interhemispheric inhibition, which may lead to suboptimal results, but also cognitive disruption that impacts on final outcome [ ].

NICS only modestly allays PD (average: 20%) with multiple sessions over a longer duration of time: as such, it is inferior to ICS [ ]. Interestingly, in rTMS studies, the shape of the hand knob in M1 determines the physiologic response to coil orientation: the inverted Ω shape responds preferentially to a 45 degrees coil angle, the ε shape to a 90 degrees coil angle [ ]; it remains to be seen how the hand knob shape bears on M1 ICS.

Fixed dystonia

Promising early results have been published (reviewed in Ref. [ ]), but no definitive conclusion can be drawn presently.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here