Motor Cortex Stimulation for Pain and Movement Disorders

Introduction

Historically, chronic stimulation of the primary motor cortex (MC) to treat pain and a variety of movement disorders spans over 50 years, but only within the last 15 years has motor cortex stimulation (MCS) grown. Although there is much debate, MCS is increasingly performed in a number of centers around the world. Prior to electrical stimulation, Bucy and others, ^, in the 1930s and 1940s, used direct extirpation of M1 to treat Parkinson disease (PD) and other disorders.

Heath, in 1963, discusses the possibility of stimulation of the septal area in hopes of activating pleasure centers to alleviate pain, which he investigated with externally implanted electrodes and the patient controlling the stimulator (self-stimulation). It was not until the 1970s that Mundinger ^, used stimulation in the sensory thalamus and zona incerta to treat spasmodic torticollis, facial dyskinesia, and athetosis, with some good results. Removal of the sensory cortex (S1) to treat pain was tried by various investigators. In patients with “burning” pain, limited success led to resection of the precentral gyrus, which could completely ablate the pain. In 1955 White and Sweet demonstrated only 13% relief with postcentral resection.

In 1985 and then 1991, Tsubokawa and colleagues reported chronic stimulation of the M1 region to treat postthalamic stroke pain. ^, Woolsey reported inhibition of tremor and rigidity in PD patients with precentral stimulation, although permanent implantation of electrodes was not tried. Nguyen and colleagues, in 1998 reported the first use of permanent electrodes for the stimulation in the M1 region to treat PD.

A Medline search for MCS literature published between 1991 and March 2019 revealed 893 cases of treatment with MCS, for pain (686 cases) ^, ; stroke rehabilitation (19 cases ^{, , ,} ); movement disorders (272 cases; however, 22 of these cases involved mixed MCS and deep brain stimulation [DBS]) ^, and multiple sclerosis (1 case). For most cases, we were able to remove patients included in multiple studies, but in some multicenter studies in which investigators in the different centers published data in both a large study group paper and their own papers, this was difficult to do.

For treating movement disorders, stimulation in the M1 region of the cortex might seem obvious, but it is well known that high levels of stimulation will cause motor activation, which offers no long-term benefit as a side effect. However, subthreshold stimulation avoids this complication, allowing stimulation to affect, in some sense, the final common link between deeper circuitry coordinating movement and the spinal cord itself. Because this region is one of the key areas where the pyramidal and extrapyramidal systems interact, disorders of movement might be expected to respond to some type of stimulation of circuitry here. Given the aforementioned studies, this seems to be the case. On the pain side, the cortex is likely to be integral in perception of pain, as complete lesioning of pathways in the periphery and spinal cord does not always reduce the pain at all, leaving the patient “sensing” pain as if from those very areas. In addition, strokes in the thalamus can create a sense of pain in one or more of the thalamic projection regions within the areas damaged by the stroke, despite absence of injury more distal in the nervous system. Accordingly, electrical modulation of the cortex is worth exploring as an approach to bring about clinical benefit.

As of the writing of this chapter, the most extensive experience with MCS is that of Tsubokawa and Katayama ^{, ,} and the group at Nihon University School of Medicine, who not only explored MCS as a stand-alone therapy but also compared DBS and spinal cord stimulation (SCS) as separate therapies or in combination for the treatment of pain. In these studies, 59% of patients who underwent DBS or MCS for poststroke involuntary movements had some benefit, and 19% of patients who underwent MCS for treatment of pain also demonstrated improvement in their movement problems. Some syndromes did not show significant benefit with MCS. Of 5 patients treated with MCS for phantom limb pain, only 1 (20%) demonstrated better pain relief than with DBS in the ventrocaudal (VC) nucleus of the thalamus (6 of 10 patients [60%]), SCS (6 of 19 patients [32%]), or a combination thereof. Reviewing Tsubokawa’s initial study from 1998 and a second report on the same patients in 1993 demonstrated that ∼50% (5 of 11) of patients had “excellent” outcomes (defined as 100% improvement) at 2 years as compared with 6 of 12 in the first year. At first glance this would indicate that initial benefit predicts long-term benefit, yet 3 of 12 patients who had improvement defined as “good” (60%–80% improvement) at 1 year experienced a decrease from their initial improvement to either “fair” (40%–60% improvement) or “poor” (less than 40% improvement). These initial results, nonetheless, helped spur further investigation into MCS for the treatment of pain. Katayama et al. in 1998 reported initial results showing that 23 of 31 poststroke patients had satisfactory pain control, although at 2 years pain control was satisfactory in only 8 of these initial 23 patients. In all of the prior studies, electrodes were placed epidurally. Four groups ^{, ,} subsequently described subdural placement of the MCS electrode.

Stimulus “on” times and waveform parameters used to treat chronic pain issues vary widely. The range of stimulation amplitudes is from 0.5 V to 10.5 V with a mean of 3.8 V ± 2.2 V, and the range of stimulation frequencies used is from 5 Hz to 210 Hz with a mean of 51.1 Hz ± 35.7 Hz. Pulse width varied from 1 to 500 microseconds with a mean of 251.2 ± 141.1 microseconds. ^a During programming, the most common adverse effects consisted of short generalized seizures, occurring during initial testing phases typically. The second and third most likely complications were hardware failures and infections.

a References , , , , , , , , , , , , , , , , .

Multiple centers in Italy have now investigated the use of MCS in the treatment of movement disorders, primarily PD. ^b

b References , , , , , , , .

Initial results demonstrated a strong beneficial effect on the most disabling symptoms of PD, including tremor, bradykinesia, and rigidity. The Italian Neurosurgical Society Study Group ^{, ,} investigated 16 patients; only 10 with a follow-up of 3 to 30 months were ultimately analyzed. All but one patient in this study were offered MCS owing to ineligibility for DBS because of age, MRI findings, or psychological problems. The mean length of disease in the study population was 12.4 years. The results of the study were that three patients ended up with less than a 25% improvement, six demonstrated a 25% to 50% improvement, and one demonstrated a greater than 50% improvement on the Unified Parkinson’s Disease Rating Scale (UPDRS) and the Hoehn and Yahr Scale. Long-term stimulation parameters included a voltage range of 2 to 6 V, a pulse width range of 60 to 210 microseconds, and a frequency range of 25 to 80 Hz. At 36 months there was a statistically significant improvement in the UPDRS III (motor subscale) of 19.1%. Overall there was no significant reduction in medications. We performed a prospective investigation in four patients with PD that demonstrated an initial benefit in all patients, but with a return to baseline by the 1-year postimplant time period ( Table 129.1 ). ^,

Since 1991, surgical complications have included seizures in 19 patients ^c

c References , , , , , , , , .

(two of which were intraoperative ), a significant hemorrhage that included postoperative vegetative state (one patient), death (one patient), aphasia (one patient), two epidural hematomas, ^, and a subdural hematoma requiring evacuation. Postoperatively, infection was the most common reason for device failure, requiring either a reoperation or removal of the system. The most common significant programming side effect was seizures, which occurred in 17 patients. ^d

d References , , , , , , , .

Extradural MCS has advantages over DBS in that microelectrode recording (MER) is often performed with the patient awake, although more and more centers have abandoned MER in favor of image-based DBS with the patient asleep. Even when compared with DBS performed using general anesthesia, MCS does not require stereotactic techniques or frames, does not require the expertise of MERs, and virtually eliminates the possibility of creating a clinically significant hemorrhage. This alone makes it an important alternative or even preliminary procedure to consider as compared with DBS in some cases, particularly in patients who are not good candidates for DBS because of other medical conditions, particularly in chronic severe refractory pain syndromes of the face or upper extremity, as these are coded by regions within M1 on the more accessible lateral convexity of the cortex. Although MCS might be considered for movement disorders, it does not yet have any class I evidence, whereas DBS does.