Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Deep Brain Stimulation for Chronic Pain
Introduction and History
Interest in deep brain stimulation (DBS) for chronic pain has waxed and waned over the past half century. Stimulation of the septal region of the brain for pain relief was first attempted in the 1950s. Chronic stimulation of a contemporary target, the sensory thalamus, was first described in the early 1970s for anesthesia dolorosa, inspired by observations made during thalamic ablation. DBS for chronic pain was performed more than 1000 times between the early 1970s and the late 1980s, and at this time, it was largely considered safe and effective. However, the procedure was almost entirely abandoned in the United States in 1989 after the US Food and Drug Administration relegated DBS devices to investigational status until clinical trials could demonstrate their safety and efficacy. In the 1990s, two industry-sponsored prospective trials of DBS for pain designed to address this requirement both demonstrated disappointing results, with less than one-fifth of patients having a 50% reduction in pain after 2 years. However, it is noteworthy that these trials were an amalgam of prospective case series that were not randomized and not case-controlled, and suffered from both poor enrollment and high attrition. , A short time later, in the late 1990s, DBS hardware was approved for movement disorders, giving surgeons an avenue to perform DBS for chronic pain on an off-label basis. Since that time, interest in DBS for chronic pain has once again grown, especially as the need for developing more effective therapies for chronic pain becomes increasingly apparent. Furthermore, the widespread use of DBS for movement disorders together with the advent of magnetic resonance imaging (MRI), intraoperative computed tomography (CT), improved targeting modalities, and new DBS hardware design has refined the technical approach to lead placement, making it safer and more accurate. ,
Presently, DBS for intractable neuropathic pain continues to be performed in Canada and some European countries, where it is approved by health regulatory agencies, as well as in a handful of centers in the United States, where it is performed on an off-label basis. There is growing contemporary experience with this procedure, with several groups continuing to refine their approach to DBS for neuropathic pain, examining its efficacy by indication and by target, and modernizing the technical approach to DBS to follow current standards in movement-disorder surgery.
Deep Brain Stimulation Targets
The three most common contemporary targets used in DBS for chronic pain are the sensory thalamus ( Fig. 117.1A ), the periventricular and periaqueductal gray matter (PVG/PAG) (see Fig. 117.1B ), and the rostral anterior cingulate cortex (ACC). Targeting the sensory thalamus with low frequency stimulation produces paresthesias in the contralateral face or body that can provide pain relief for patients with concordant pain in that distribution. The somatotopy of the sensory thalamus displays a medial-lateral organization, with the head of the homunculus being medial and the feet lateral, and has been confirmed in human microelectrode recording studies. Another target, the periaqueductal gray matter, is known to modulate nociception, and DBS of PAG/PVG was originally developed from rodent experiments that were later translated to human patients. With stimulation of PAG/PVG, which also has a somatotopic organization, patients may experience a warm sensation that alleviates pain in the corresponding distribution. The PAG/PVG exhibits a rostrocaudally inverted sensory homunculus (“Trendelenburg man”), with the representation of the feet found shallower along the trajectory, and the arm then the face found successively deeper. Finally, ACC stimulation, much like a cingulotomy, is thought to provide pain relief by modulating the emotional valence of pain.
Between PAG/PVG and sensory thalamus, there is no widespread consensus as to which target is best. While some authors have suggested that the sensory thalamus is superior to PAG/PVG for neuropathic pain, the surgeons at Radcliffe Hospital in Oxford, who have published the largest contemporary experience in DBS for chronic pain, report utilizing a trial of both sensory thalamus and PAG/PVG in each patient before selecting a final target based on trial stimulation results. Anterior cingulate stimulation is generally reserved for patients who fail sensory thalamic and/or PAG/PVG stimulation or who have poorly localized or widespread pain.
Patient Selection
DBS has been utilized for a wide variety of pain syndromes including anesthesia dolorosa, cancer pain, complex regional pain syndrome, failed back surgery syndrome, postherpetic pain, post-brachial plexus injury pain, phantom limb pain, post-stroke pain, and pain following spinal cord injury, among others. Importantly, symptom history may be a more relevant determinant than etiology in patient selection, with DBS being appropriate for patients with pain characterized by “hyperalgesia, allodynia, and hyperpathia.” Consistently, the Radcliffe Hospital group report that more than 80% of patients with a “burning” pain have found benefit from DBS. Prior to consideration for DBS for chronic pain, patients should have been thoroughly evaluated by a multidisciplinary team including a pain specialist, a neuropsychologist, a neurologist, and a neurosurgeon. The patient should have previously attempted and failed several appropriate nonsurgical therapies, including medication and interventional procedures.
Operative Technique
DBS lead placement for chronic pain targeting the sensory thalamus and PAG/PVG is performed with the patient awake, using indirect targeting for initial lead placement, followed by intraoperative macroelectrode stimulation testing to map the somatotopy of the target to confirm stimulation-evoked paresthesias (ventral posteromedial nucleus [VPM]/ventral posterolateral nucleus [VPL]) or stimulation-evoked warm sensation (PAG/PVG) over the painful area. At the Barrow, we obtain the following MRI sequences of the patient’s head in the days prior to surgery, all with 2 mm slice thickness: proton density, T2 fast spin echo, and T1 spoiled gradient echo with and without gadolinium enhancement. Trajectory planning is performed on Medtronic Frame-Link software prior to surgery. The target for the VPM thalamus is at the level of the AC–PC plane, 6 to 8 mm posterior to the midcommissural point, and 10 to 12 mm lateral to midline. This places the medial-lateral coordinate of the VPM approximately halfway between the wall of the third ventricle and the internal capsule. The VPL is targeted slightly more laterally, 2 to 3 mm medial to the internal capsule for the arm area and 1 to 2 mm medial to the internal capsule for the leg area. The anatomical boundaries of the VPM/VPL thalamus are the centromedian and parafascicular nuclei medially, the internal capsule laterally, the thalamic fasciculus, zona incerta, and subthalamic nucleus inferiorly, the thalamic nucleus ventralis intermedius anteriorly, and the pulvinar thalamic nucleus posteriorly. The target for PAG/PVG is within 1 cm of the AC–PC plane inferiorly, usually targeted at the level of the superior colliculus 2 to 3 mm ventrolateral to the cerebral aqueduct. The anatomical boundaries of the PAG/PVG in the midbrain include the medial lemniscus laterally, superior colliculus inferoposteriorly, and the red nucleus inferoanteriorly. Skull entry points are chosen to avoid cortical blood vessels, generally within 1 cm anterior to the coronal suture, with an extraventricular trajectory that is at least 4 mm away from the wall of the lateral ventricle. We recommend trajectory planning in advance of the day of surgery until the surgeon becomes quite comfortable with these targets.
On the day of surgery, a Leksell frame is placed utilizing local anesthetic and conscious sedation, then a fiducial box is placed before the patient undergoes an intraoperative volumetric CT scan. This scan is then uploaded to the Frame-link station and merged to the preoperative MRI scan. Image co-registration brings the preoperative MRI into stereotactic space. After carefully checking the merged images, Leksell coordinates for the appropriate targets may be obtained. If necessary, a small area of hair is clipped at the approximate entry-point. Following sterile prep and drape, a curvilinear incision is made surrounding the entry point. A perforator drill is used to create a burr hole, and a locking cap is attached. The dura is opened and coagulated, and subsequently the pial surface is opened and coagulated. A guide cannula is then placed, and the DBS electrode is slowly advanced to target. Test stimulation is performed from 5 mm above to 3 mm past the target, depending on the patient’s anatomy, with an attempt to elicit sensation in the desired somatotopic distribution of the pain, as well as to evaluate for unexpected adverse effects of stimulation. The distal two contacts are used in bipolar configuration, with frequency usually at 20 to 30 Hz and pulse width of 150 to 200 microseconds, and amplitudes from 1 to 4 V. The electrode is advanced, as necessary, to achieve appropriate somatotopic coverage. During implantation of the PAG lead, the patient will often report a warm, pleasant sensation in a rough somatotopic distribution. Once an acceptable location is confirmed, the lead is secured and attached to a trialing extension lead, which is tunneled out of the skin through a stab incision for connection to an external pulse generator. Patients may be admitted for an extended inpatient stay or discharged home for the trialing period, which typically lasts 2 to 4 weeks. After a successful trial, the patient is brought back to the operating room for internalization and placement of an infraclavicular internal pulse generator.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here