Medial Thalamotomy

Stereotactic thalamic neuroablative surgery for pain is relatively safe with respect to deep brainstem structures, and because of the wide involvement of many thalamic nuclei in pain processing, it has been considered a part of the pain surgery armamentarium. The first structure targeted for neuroablation was the ventral caudal (Vc) nucleus, as defined by Hassler ; however, it was soon recognized that neuroablation of the Vc nucleus was associated with significant deafferentation pain phenomena. The work of Mark and colleagues led to the belief that targeting the medial thalamic nuclei was more effective in managing pain. Nuclear targets for neuroablative medial thalamotomy are (1) the centralis lateralis, (2) centrum medianum, and (3) parafascicularis. Several pain syndromes, including cancer pain, central and peripheral deafferentation pain, spinal cord injury, malignancy, arthritis, and the neurogenic pains associated with Parkinson’s disease, have been successfully treated by medial thalamotomy. Frank and coauthors reported the overall success rate of medial thalamotomy to be 52%, with cancer pain being the main condition treated. Jeanmonod and coworkers and Young and colleagues used radiofrequency and Gamma Knife treatment, respectively, and reported a 60% success rate in achieving pain control. The ideal target lying between the three main medial thalamic nuclei (listed above) has yet to be determined, although the centrum medianum nucleus is the most frequently targeted. DBS of the medial nuclei does not usually produce a conscious sensory response, and lesioning does not induce sensory loss. The published literature on medial thalamotomy is inconsistent regarding the target, guidance technique, patient population, and lesioning method used. Therefore, the actual success rate of medial thalamotomy is difficult to assess. However, in general, the procedure is considered to be effective in treating nociceptive pain, with recent data pointing to some success in relieving neuropathic pain.

Stereotactic Cingulotomy

Cingulotomy refers to stereotactic lesioning of the anterior cingulate gyrus. Le Beau performed the first open cingulectomy to treat intractable pain in 1954. It is believed that cingulotomy causes relief by altering a patient’s emotional reaction to painful stimuli through interruption of the Papez circuit and increasing tolerance to the subjective and emotional feelings of pain. Cingulotomy is performed with standard stereotactic protocols, usually under general anesthesia. Bilateral lesions are made in the anterior aspect of the cingulate gyrus, and success of the procedure is directly related to the extent of ablation of the cingulum ( Fig. 22.1 ). A suitable stereotactic cingulotomy candidate is a terminally ill patient with widespread metastatic disease that has extended to the musculoskeletal system, where intrathecal or intraventricular administration of opiates is difficult. Emotional factors associated with pain would favor selection of a stereotactic cingulotomy procedure. Of note, stereotactic cingulotomy has been used to treat nonmalignant pain with a success rate of approximately 25%. Stereotactic cingulotomy involves the ablation of sufficient anterior cingulate gyrus volume, which is usually achieved by producing at least two lesions with a wide–surface area, noninsulated tip electrode. The procedure is generally safe with few and minor side effects. Pillay and Hassenbusch reported on a series of 12 patients in which 7 had satisfactory pain relief. Cingulotomy is rarely used today, mainly because of its narrow indication, advances in the medical management of terminal cancer patients, and the widespread use of neuroaugmentive procedures.

Diagrammatic representation of cerebral neuromodulation and neuroablation procedures and spinal neuroablation procedures. DREZ, dorsal root entry zone.

(Adapted with permission from © Springer-Verlag 2007, Raslan AM, McCartney S, Burchiel KJ. Management of chronic severe pain: cerebral neuromodulatory and neuroablative approaches. Acta Neurochir Suppl . 2007;97:17-26; and Raslan AM, McCartney S, Burchiel KJ. Management of chronic severe pain: spinal neuromodulatory and neuroablative approaches. Acta Neurochir Suppl . 2007;97:33-41. With kind permission from Springer Science and Business Media.)

Caudalis Dorsal Root Entry Zone (Brainstem Level)

Following the introduction of stereotaxis in the 1960s, the use of open ablative brain and brainstem surgery was almost abandoned. Siqueira first reported performance of the caudalis DREZ procedure in two patients. Gorecki, Nashold, and colleagues at Duke University later adopted the technique and expanded its indications. In the caudalis DREZ procedure, the caudal portion of the spinal trigeminal nucleus, along with the overlying trigeminal tract, is destroyed. Similar to spinal DREZ surgery, the objective is to destroy the cells of second-order neurons thought to be hyperactive in trigeminal deafferentation pain, thereby achieving pain relief (see Fig. 22.1 ). The main indications for the caudalis DREZ procedure are ophthalmic post-herpetic neuralgia and trigeminal anesthesia dolorosa. In cases of neuropathic facial pain in which all other medical and surgical modalities are ineffective, the caudalis DREZ procedure may represent a last resort. The procedure is rarely performed, and potential risks include ipsilateral limb ataxia and weakness.

Anterolateral Cordotomy

Anterolateral cordotomy refers to lesioning, sectioning, or other disruption of the lateral spinothalamic tract (LST), which is located in the anterolateral quadrant of the spinal cord. The procedure was historically performed in the upper thoracic spine via an open posterior approach and, less commonly, high in the cervical spine. The spinal cord anterolateral ascending pain transmission system carries information about pain and temperature from one side of the body. The tract is formed by the central processes of nociceptive neurons in the dorsal horn that cross the spinal cord in the anterior commissure, ascend in the anterolateral column to the brainstem, and relay in the thalamus. Lesions of the anterolateral tract produce a contralateral deficit in pain and temperature sensation two to five segments below the level of the cordotomy. Fibers in the LST have a somatotopic arrangement, with the sacral segments arranged posterolaterally and the cervical segments anteromedially. The corticospinal (pyramidal) tract lies posterior to the LST with white matter in between. The ventral spinocerebellar tract overlies the LST, and a lesion that damages the spinocerebellar tract may cause ipsilateral ataxia of the arm. Autonomic pathways for vasomotor and genitourinary control and reticulospinal fibers that subserve ipsilateral automatic respiration are also part of the anterolateral quadrant of the spinal cord. A patient with hemibody somatic cancer pain localized caudal to the cervical and upper thoracic area represents the best candidate for a cordotomy procedure.

From the beginning of the 20th century until the late 1960s to early 1970s, cordotomy was an open procedure undertaken at the mid to high thoracic levels since these sites largely avoided the complications of upper limb ataxia and sleep apnea. Introduction of the minimally invasive percutaneous approach for cordotomy by Mullan, Rosomoff, and their colleagues mitigated some of the neurologic risks and made it possible for the procedure to be performed on patients in poor general health. In the mid-1980s and early 1990s, advances in opioid pharmacology, as well as the introduction of reversible and testable neuroaugmentive techniques, reduced the perceived need for spinal destructive procedures for pain control and led to a major reduction in the number of cordotomies performed by neurosurgeons worldwide. However, these neuroaugmentive procedures were expensive, particularly given the short life expectancy of many of the candidates, and were not uniformly effective.

Kanpolat and coworkers first introduced the concept of computed tomography (CT)-guided cordotomy, which allowed a safer, selective, and more effective procedure. In 1995, Fenstermaker and associates performed anterior CT-guided lower cervical cordotomy through the disk space to avoid sleep apnea (a modification of Gildenberg and colleagues’ anterior low cervical percutaneous cordotomy). CT-guided cordotomy is typically performed as a percutaneous procedure via a lateral approach to the spinal cord at the level of C2. However, the anterior cervical transdiscal approach can also be used, and in a recent clinical study this approach was used to control cancer pain in six of eight patients with pulmonary-pleural malignancy while avoiding sleep apnea.

Today, the CT-guided cordotomy procedure involves lumbar puncture and injection of a water-soluble dye into the patient’s intrathecal space. After 30 minutes, a cervical CT scan is performed. This and subsequent scans are used to direct the cordotomy electrode into the anterolateral quadrant of the ipsilateral spinal cord. The electrode is insulated throughout the entire shaft except the tip (2 mm in length and 0.3 to 0.4 mm in diameter). After measurement of the skin-dura distance and local anesthesia of the lateral cervical region, an electrode is introduced from the lateral side of the neck opposite the C2 foramen into the anterolateral quadrant of the spinal cord. To ensure complete entry into the spinothalamic tract (while avoiding the corticospinal tract), electrophysiologic testing is essential. Radiofrequency lesions are performed until adequate hypoesthesia is achieved in the contralateral hemibody or at least in the region of pain. CT-guided cordotomy has a higher success rate than do more traditional approaches, as well as fewer complications. Control of cancer pain is reportedly achieved in more than 95% of cases. Procedural complications may include weakness, hypotension, dysesthesia, mirror-image pain, ataxia, incontinence, and sleep apnea. However, contemporary CT-guided cordotomy complications tend to be both minor and transient.

A recent evidence-based review concluded that the case for cordotomy is somewhat unique among all cancer pain procedures in that it has the most supportive evidence. In that review the GRADE system of recommendation was used, and a recommendation for cordotomy was given. The GRADE system produces recommendations that are independent of the level of evidence.

Trigeminal Tractotomy-Nucleotomy (Spinal Level)

Sensory information from the 5th, 7th, 9th, and 10th cranial nerves is carried by the trigeminal tract and branches into the trigeminal tract spinal nucleus and extended caudally into the spinal cord to C2. The trigeminal tract is considered a target for surgically treating facial pain, and the history of procedures directed to this target is similar to cordotomy in that initial open procedures evolved toward less invasive stereotactic operations. Crue and colleagues and Hitchcock developed a stereotactic technique to lesion the trigeminal tract and nucleus via radiofrequency that was named trigeminal nucleotomy. As with CT-guided cordotomy, CT is used when performing the trigeminal tractotomy-nucleotomy (TR-NC) procedure today. Indications include anesthesia dolorosa, post-herpetic neuralgia, neuropathic facial pain, facial cancer pain, and either glossopharyngeal or geniculate neuralgia. The procedure can be considered, in some ways, a mini–caudalis DREZ procedure. The nucleus caudalis DREZ operation involves the same concept as the TR-NC procedure but includes destruction of the substantia gelatinosa (Rexed laminae II and III) of the nucleus caudalis. Pain relief from TR-NC is reported to be complete or satisfactory in 80% of cases. Complications include ataxia from injury to the spinocerebellar tract (usually temporary) and contralateral hypoalgesia if the spinothalamic tract is included in the lesion.

ExtralemNiscal Myelotomy

The extralemniscal myelotomy (ELM) procedure was first described by Hitchcock, who initially aimed to destroy the decussating fibers of the spinothalamic tract in the anterior commissure of the spinal cord to control pain in the neck and both arms. ELM was achieved by creating a lesion in the central medullary region at the cervicomedullary junction. Unexpectedly, it was noted that the ELM procedure also seemed to control pain caudal to the level of the lesion. Schvarcz added the term “extralemniscal” to “myelotomy” because of the contention that the lesion incorporated an ascending polysynaptic nociceptive pathway. Subsequently, the presence of such a tract has been confirmed anatomically. Several authors have now presented reports of midline “punctuate” ELM via open procedures to interrupt this pathway at various spinal cord levels. The polysynaptic ascending pathway is thought to carry visceral nociceptive information and lies deep to the midline dorsal column. The concept of CT guidance, previously applied to cordotomy and TR-NC procedures, has also been applied to ELM by Kanpolat and colleagues, thus developing the image-guided ELM procedure used today.

ELM is currently conceived as a pain control procedure for pain of visceral origin, including patients with pelvic malignancy, or for cancer pain in the lower part of the trunk and lower extremities with a predominant visceral pain component. The procedure appears to be safe; however, pain relief results are not as good as those achieved with cordotomy and TR-NC procedures.

Dorsal Root Entry Zone Lesions

With introduction of the gate control theory in the 1960s, attention was drawn to the spinal dorsal horn as the initial physiologic substrate for pain modulation. The dorsal horn and DREZ were then reconsidered as targets for both neuromodulation (spinal cord stimulation) and neuroablation. In 1972, Sindou first attempted cervical DREZ destruction for neuropathic deafferentation pain in the upper extremity secondary to brachial plexus avulsion. Nashold and associates soon followed and introduced the use of radiofrequency lesions to perform DREZ disruption. Laser and ultrasound have also been used to damage the DREZ. When large-fiber afferents (touch, position sense) in peripheral nerves or dorsal roots are altered, there is a reduction in the inhibitory control of the dorsal horn. This situation is presumed to result in excessive firing of the dorsal horn neurons, which is thought to be the cause of deafferentation pain and hence able to be controlled by DREZ lesioning. The technical details of the procedure and its variants are beyond the scope of this chapter, but DREZ lesioning is performed as an open surgical procedure under general anesthesia and oftentimes accompanied by intraoperative neurophysiologic monitoring. Surgical candidates are patients with brachial plexus avulsion, Pancoast’s tumor with brachial plexus invasion combined with a good general condition and reasonable life expectancy, pain caused by spinal cord or cauda equina lesions, and pain accompanying spasticity after plexus or cord injury. A general prerequisite for the DREZ procedure is a lack of functional use of the limb where the DREZ procedure is performed since complete sensory denervation of the limb will render it functionless even if there is residual motor power. When patients are carefully selected and the lesions accurately performed, the success rate can be as high as 90% (with follow-up success rates reported for up to 4 years). Complications and side effects include cerebrospinal fluid (CSF) fistula, meningitis, ataxia, increased neurologic deficits, and dysesthesias.

Anterolateral Cordotomy

Anterolateral cordotomy refers to lesioning, sectioning, or other disruption of the lateral spinothalamic tract (LST), which is located in the anterolateral quadrant of the spinal cord. The procedure was historically performed in the upper thoracic spine via an open posterior approach and, less commonly, high in the cervical spine. The spinal cord anterolateral ascending pain transmission system carries information about pain and temperature from one side of the body. The tract is formed by the central processes of nociceptive neurons in the dorsal horn that cross the spinal cord in the anterior commissure, ascend in the anterolateral column to the brainstem, and relay in the thalamus. Lesions of the anterolateral tract produce a contralateral deficit in pain and temperature sensation two to five segments below the level of the cordotomy. Fibers in the LST have a somatotopic arrangement, with the sacral segments arranged posterolaterally and the cervical segments anteromedially. The corticospinal (pyramidal) tract lies posterior to the LST with white matter in between. The ventral spinocerebellar tract overlies the LST, and a lesion that damages the spinocerebellar tract may cause ipsilateral ataxia of the arm. Autonomic pathways for vasomotor and genitourinary control and reticulospinal fibers that subserve ipsilateral automatic respiration are also part of the anterolateral quadrant of the spinal cord. A patient with hemibody somatic cancer pain localized caudal to the cervical and upper thoracic area represents the best candidate for a cordotomy procedure.

From the beginning of the 20th century until the late 1960s to early 1970s, cordotomy was an open procedure undertaken at the mid to high thoracic levels since these sites largely avoided the complications of upper limb ataxia and sleep apnea. Introduction of the minimally invasive percutaneous approach for cordotomy by Mullan, Rosomoff, and their colleagues mitigated some of the neurologic risks and made it possible for the procedure to be performed on patients in poor general health. In the mid-1980s and early 1990s, advances in opioid pharmacology, as well as the introduction of reversible and testable neuroaugmentive techniques, reduced the perceived need for spinal destructive procedures for pain control and led to a major reduction in the number of cordotomies performed by neurosurgeons worldwide. However, these neuroaugmentive procedures were expensive, particularly given the short life expectancy of many of the candidates, and were not uniformly effective.

Kanpolat and coworkers first introduced the concept of computed tomography (CT)-guided cordotomy, which allowed a safer, selective, and more effective procedure. In 1995, Fenstermaker and associates performed anterior CT-guided lower cervical cordotomy through the disk space to avoid sleep apnea (a modification of Gildenberg and colleagues’ anterior low cervical percutaneous cordotomy). CT-guided cordotomy is typically performed as a percutaneous procedure via a lateral approach to the spinal cord at the level of C2. However, the anterior cervical transdiscal approach can also be used, and in a recent clinical study this approach was used to control cancer pain in six of eight patients with pulmonary-pleural malignancy while avoiding sleep apnea.

Today, the CT-guided cordotomy procedure involves lumbar puncture and injection of a water-soluble dye into the patient’s intrathecal space. After 30 minutes, a cervical CT scan is performed. This and subsequent scans are used to direct the cordotomy electrode into the anterolateral quadrant of the ipsilateral spinal cord. The electrode is insulated throughout the entire shaft except the tip (2 mm in length and 0.3 to 0.4 mm in diameter). After measurement of the skin-dura distance and local anesthesia of the lateral cervical region, an electrode is introduced from the lateral side of the neck opposite the C2 foramen into the anterolateral quadrant of the spinal cord. To ensure complete entry into the spinothalamic tract (while avoiding the corticospinal tract), electrophysiologic testing is essential. Radiofrequency lesions are performed until adequate hypoesthesia is achieved in the contralateral hemibody or at least in the region of pain. CT-guided cordotomy has a higher success rate than do more traditional approaches, as well as fewer complications. Control of cancer pain is reportedly achieved in more than 95% of cases. Procedural complications may include weakness, hypotension, dysesthesia, mirror-image pain, ataxia, incontinence, and sleep apnea. However, contemporary CT-guided cordotomy complications tend to be both minor and transient.

A recent evidence-based review concluded that the case for cordotomy is somewhat unique among all cancer pain procedures in that it has the most supportive evidence. In that review the GRADE system of recommendation was used, and a recommendation for cordotomy was given. The GRADE system produces recommendations that are independent of the level of evidence.

Trigeminal Tractotomy-Nucleotomy (Spinal Level)

Sensory information from the 5th, 7th, 9th, and 10th cranial nerves is carried by the trigeminal tract and branches into the trigeminal tract spinal nucleus and extended caudally into the spinal cord to C2. The trigeminal tract is considered a target for surgically treating facial pain, and the history of procedures directed to this target is similar to cordotomy in that initial open procedures evolved toward less invasive stereotactic operations. Crue and colleagues and Hitchcock developed a stereotactic technique to lesion the trigeminal tract and nucleus via radiofrequency that was named trigeminal nucleotomy. As with CT-guided cordotomy, CT is used when performing the trigeminal tractotomy-nucleotomy (TR-NC) procedure today. Indications include anesthesia dolorosa, post-herpetic neuralgia, neuropathic facial pain, facial cancer pain, and either glossopharyngeal or geniculate neuralgia. The procedure can be considered, in some ways, a mini–caudalis DREZ procedure. The nucleus caudalis DREZ operation involves the same concept as the TR-NC procedure but includes destruction of the substantia gelatinosa (Rexed laminae II and III) of the nucleus caudalis. Pain relief from TR-NC is reported to be complete or satisfactory in 80% of cases. Complications include ataxia from injury to the spinocerebellar tract (usually temporary) and contralateral hypoalgesia if the spinothalamic tract is included in the lesion.

ExtralemNiscal Myelotomy

The extralemniscal myelotomy (ELM) procedure was first described by Hitchcock, who initially aimed to destroy the decussating fibers of the spinothalamic tract in the anterior commissure of the spinal cord to control pain in the neck and both arms. ELM was achieved by creating a lesion in the central medullary region at the cervicomedullary junction. Unexpectedly, it was noted that the ELM procedure also seemed to control pain caudal to the level of the lesion. Schvarcz added the term “extralemniscal” to “myelotomy” because of the contention that the lesion incorporated an ascending polysynaptic nociceptive pathway. Subsequently, the presence of such a tract has been confirmed anatomically. Several authors have now presented reports of midline “punctuate” ELM via open procedures to interrupt this pathway at various spinal cord levels. The polysynaptic ascending pathway is thought to carry visceral nociceptive information and lies deep to the midline dorsal column. The concept of CT guidance, previously applied to cordotomy and TR-NC procedures, has also been applied to ELM by Kanpolat and colleagues, thus developing the image-guided ELM procedure used today.

ELM is currently conceived as a pain control procedure for pain of visceral origin, including patients with pelvic malignancy, or for cancer pain in the lower part of the trunk and lower extremities with a predominant visceral pain component. The procedure appears to be safe; however, pain relief results are not as good as those achieved with cordotomy and TR-NC procedures.

Dorsal Root Entry Zone Lesions

With introduction of the gate control theory in the 1960s, attention was drawn to the spinal dorsal horn as the initial physiologic substrate for pain modulation. The dorsal horn and DREZ were then reconsidered as targets for both neuromodulation (spinal cord stimulation) and neuroablation. In 1972, Sindou first attempted cervical DREZ destruction for neuropathic deafferentation pain in the upper extremity secondary to brachial plexus avulsion. Nashold and associates soon followed and introduced the use of radiofrequency lesions to perform DREZ disruption. Laser and ultrasound have also been used to damage the DREZ. When large-fiber afferents (touch, position sense) in peripheral nerves or dorsal roots are altered, there is a reduction in the inhibitory control of the dorsal horn. This situation is presumed to result in excessive firing of the dorsal horn neurons, which is thought to be the cause of deafferentation pain and hence able to be controlled by DREZ lesioning. The technical details of the procedure and its variants are beyond the scope of this chapter, but DREZ lesioning is performed as an open surgical procedure under general anesthesia and oftentimes accompanied by intraoperative neurophysiologic monitoring. Surgical candidates are patients with brachial plexus avulsion, Pancoast’s tumor with brachial plexus invasion combined with a good general condition and reasonable life expectancy, pain caused by spinal cord or cauda equina lesions, and pain accompanying spasticity after plexus or cord injury. A general prerequisite for the DREZ procedure is a lack of functional use of the limb where the DREZ procedure is performed since complete sensory denervation of the limb will render it functionless even if there is residual motor power. When patients are carefully selected and the lesions accurately performed, the success rate can be as high as 90% (with follow-up success rates reported for up to 4 years). Complications and side effects include cerebrospinal fluid (CSF) fistula, meningitis, ataxia, increased neurologic deficits, and dysesthesias.