Peripheral neurectomy involves cutting a nerve distal to the dorsal root, external to the spine or cranium. The role of peripheral neurectomy in the management of pain due to cancer is very limited.¹⁵ The theoretical advantages are the ease of surgery, complete denervation in the distribution of the cut nerve, and, that a preoperative nerve block with a local anesthetic will accurately reveal the extent of denervation and whether or not pain relief can be expected from the operation. The disadvantages of a peripheral neurectomy include the loss of all sensory modalities in the denervated region, the loss of motor function if the nerve to be cut is mixed, the re-innervation that will occur because of both peripheral and central sprouting, and the potential for growth of the neoplasm to go beyond the denervated region and lead to pain recurrence.

Surgical intercostal neurectomy may be useful in a patient with chest wall neoplasm too frail to undergo dorsal rhizotomy. The loss of intercostal muscle function in the denervated region is usually inconsequential. Denervation of abdominal segments may result in muscle weakness that produces an asymmetrical abdominal protuberance that the patient may find painful and embarrassing. Neurectomies in the extremities are usually unwise because of the paresis that ensues. Cranial rhizotomies are usually a better choice than neurectomies for pain due to cancer. Peripheral neurectomies can also lead to a painful neuroma or denervation pain; both can be difficult to manage.

Sympathectomy involves excision or cauterization of the paravertebral sympathetic chain and associated ganglia or their peripheral branches. Surgical sympathectomy plays a limited role in the management of pain secondary to cancer. Nociceptive afferents from the viscera travel in the sympathetic chains and vagus nerve and may be surgically interrupted. Splanchniectomy may be part of an abdominal operation and can provide relief of visceral-mediated pain. However, this is a formidable surgical procedure by itself that can usually be much more easily accomplished with neurolytic agents that are percutaneously administered.

Some of the pains associated with cancer are not nociceptive but arise from nerve injury by the neoplasm, surgery, or radiation therapy. This component of neuropathic pain may respond to sympathectomy. Therefore, if the patient’s pain persists after somatic blocks or ablative surgery, sympathetic block is warranted to ascertain if the pain can be eliminated. If so, a sympathectomy should be considered.¹⁶ In addition to the traditional older methods for an open sympathectomy, there are now endoscopic and percutaneous methods of achieving sympathectomy. Neurolytic sympathetic blocks are probably a better choice than open sympathectomy in the cancer pain patient.

A rhizotomy is a surgical procedure to cut nerve roots. Because most peripheral nociceptive impulses are conveyed through a mixed somatic nerve to the spinal cord via the dorsal root, dorsal rhizotomy was the most commonly performed operation for pain relief in the first 75 years of the 20th century. As better clinical trials with longer follow-ups were reported, the popularity of dorsal rhizotomy waned. This was initially because of the development of cordotomy and then the use of intraspinal opioids. The advantages of dorsal rhizotomy include: All modalities of sensation but no motor function is lost in the denervated area, large areas can be denervated by cutting many adjacent dorsal roots, and nerve blocks can be used to determine the required extent of denervation. The disadvantages include the fact that one cannot totally denervate an extremity without losing useful function, a laminectomy is required for each root to be cut, and sprouting, both peripheral and central, may cause pain relief to be temporary. This operation is particularly useful for thoracic and abdominal wall pain but it will not interrupt or attenuate visceral pain. From a practical perspective, it is always necessary to section two roots above and two roots below the region of pain to get reliable pain relief that may last for the duration of the patient’s life. This can lead to the need for very extensive laminectomies if an open procedure is contemplated.

Percutaneous dorsal rhizotomies have particular utility in the patient with pain due to cancer. This strategy does put at risk the artery that accompanies the dorsal root; particularly in the lower thoracic region this can lead to spinal cord infarction. Sacral rhizotomy via a laminectomy at S2 and epidural ligation of all the nerve roots in the cauda equina can effectively alleviate pelvic pain in the patient who has already had urinary and fecal diversionary procedures.¹⁷

It is unclear whether or not resection of the dorsal root ganglion adds to the likelihood of success for dorsal rhizotomy.¹⁸ Most of this controversy relates to patients with lumbar rhizotomy for failed back surgery syndrome and cannot be extrapolated to patients with pain due to cancer. Ganglionectomy is frequently followed by a transient dysesthetic pain syndrome in the denervated area and patients should be warned of this. Ganglionectomy is an extradural procedure and the risk of cerebrospinal fluid (CSF) leak is far less than after intradural rhizotomy.

Cranial rhizotomy may be useful in patients with head and neck cancer that has not been controlled by other means. The traditional procedure has been a posterior fossa approach to the trigeminal, nervus intermedius, and glossopharyngeal and upper vagal nerves accompanied by section of the dorsal roots of C2 and C3 (there is no dorsal root of C1). Such an operation produces complete anesthesia of the face and pharynx and can only be done unilaterally. It can provide significant pain relief (75% success rate) in appropriately selected patients for the duration of the cancer patient’s life. Unfortunately, many patients with advanced head and neck cancers have extensive tumor involvement which overlaps dermatomal segments that reduces the likelihood of successful pain relief with single nerve cranial rhizotomy.

A cordotomy involves an incision or Radio Frequency (RF) lesion made in the anterolateral quadrant of the spinal cord to interrupt the axons of the spinothalamic tract (Fig. 21.2). Cordotomy was the most commonly performed operation for cancer pain before the advent of intraspinal opioids. The advantages of cordotomy are that only pain and temperature sensation are abolished contralateral to the lesion in the spinal cord, commencing three segments below the level incised. This means that the level of the operation can be tailored to the region of pain. Motor loss is not expected and touch and position sense remain intact. The disadvantages include the tendency of the pain relief to fade after a year or so, the fact that bilateral lesions can lead to impairment of bladder function, and that bilateral high cervical cordotomy can lead to Ondine Curse, the failure to breath when the patient is asleep.¹⁹ Cordotomy can be performed at the upper thoracic level for lower half of the body pains and at the high cervical level for pains below C5. Percutaneous cordotomy is performed at the C1-2 level and offers the possibility of surgical pain relief in a patient too debilitated to withstand an open procedure.²⁰ A computed to-mography (CT)-guided method has been developed that lowers the risks of an inappropriate lesion site. For most cancer pain patients, percutaneous C1-2 cordotomy is the procedure of choice, although bilateral lesions do have significant risks to both bladder function and loss of CO2 responsiveness. The quality of pain relief is identical with both the open and percutaneous operations. Raslan²¹ described the use of percutaneous CT-guided radiofrequency ablation of upper cervical spinal cord pain pathways in the treatment of cancer-related pain. Fifty-one patients with cancer-related body or face pain were treated with computed tomography-guided radiofrequency ablation of the spinothalamic tract or trigeminal tract nucleus in the upper cervical region of the spinal cord. Forty-one patients underwent a unilateral cervical cordotomy, and ten patients underwent a trigeminal tractotomy-nucleotomy (TR-NC). Patients with somatic pain on one side of the body reaching to the midline and below dermatome C5 and patients with visceral pain restricted to one side of the body not reaching the midline were offered cervical cordotomy. Patients with somatic pain involving one side of the face, tongue, or inner mouth were offered the option of TR-NC. Forty-one patients underwent cervical cordotomy at C1-2 level, and ten patients underwent TR-NC also at the C1-2 level. The results from these procedures are shown in Table 21.2.

Postoperatively, 98% of the patients were Grade I or II, which was categorized as a successful procedure. At 1 month postoperatively, 98% of the patients remained Grade I or II, which decreased to 80% at 6 months follow-up. One patient, who was Grade III at 1-month follow-up, dropped to Grade IV at 3 months follow-up and died before the 6-month follow-up contact. Despite two patients having only one lesion performed and the procedure halted, results in terms of pain relief in the first 3 months were not affected and most likely reflect the effect of a single lesion. No complications related to the procedures used were observed. No change in motor strength was recorded in any patient. No patient developed sleep apnea. There were no permanent or serious
complications such as respiratory depression and/or weakness in this series. Complications were transient and not severe. Hypotension occurred in three cases (two cordotomies and one TR-NC). All patients responded to treatment with 24-hour parenteral fluid administration and were discharged without event. Headache was reported in four patients (three cordotomies and one TR-NC). All were treated with analgesics and fluids for 48 hours and discharged without event. Dysesthesia occurred in two patients undergoing cordotomy and persisted in one patient for 3 days and in the other for 2 weeks, which completely resolved by the first month follow-up assessment.

Most other large series reveal an 80% success rate in cancer patients, although those who survive for several years may lose the cordotomy effect. Failure to get initial pain relief almost always means that the cordotomy lesion was not large enough or that it was not far enough above the source of pain. Midline or bilateral pain from the pelvis is better treated with a midline myelotomy than bilateral cordotomies.

Myelotomy for pain is an operation to incise the spinal cord in the dorsal midline so as to transect the decussating spinothalamic fibers and/or damage the visceral afferent tract that ascends in the deep medial dorsal columns. On the basis of these novel pathways, midline myelotomy (either commissural or punctate), has been developed as treatment of visceral pain. Midline myelotomy has the advantage of producing bilateral pain relief from a single incision in the spinal cord. At first, this operation was thought to provide pain relief by sectioning the decussating spinothalamic fibers. However, the studies of Willis et al. and the clinical observations of Hitchcock have indicated that there is a visceral afferent tract that travels in the deep midline of the dorsal columns.²² The axons of this pathway ascend in the ipsilateral dorsal column and terminate in the nucleus gracilis from where the nociceptive input is transmitted to the contralateral ventral posterolateral thalamus. This tract probably explains the observation that pain relief can be found in regions of the body below the level of pain and temperature loss. A midline myelotomy, whether longitudinal or punctate, has a very low risk of impairing motor function, although some proprioceptive losses in the lower extremities are not rare after this procedure. The procedure requires a low thoracic laminectomy and the duration of pain relief may not be more than a year in many patients.

Hwang et al.²³ reported on the use of a punctate midline myelotomy at the T3 vertebral level in a small number of patients with visceral pain associated with hepatobiliary or pancreatic cancer. Although only six patients were studied, their results suggest some use of this technique in the management of severe pain for these particular cancers. In an experimental animal model, Chang²⁴ reported on the use of a high cervical punctate midline myelotomy for intractable visceral pain due to abdominal or pelvic cancers. Similarly, Hitchcock²⁵ described a high cervical midline myelotomy that produced pain relief without sensory loss, but few have replicated his findings.¹² Nauta et al.²⁶ performed a punctate midline myelotomy in six cancer patients with intractable refractory visceral pain and noted good pain relief with significant opioid reductions postprocedure.

Dorsal root entry zone (DREZ) lesions require destroying the central projections of nociceptive afferents where they enter the spinal cord. DREZ were first performed by Nashold in 1975, although Sindou described sectioning the lateral root fibers going to the Lissauer tract in 1972. In North America, the Nashold strategy has been the most commonly utilized. This operation has the advantages of producing complete sensory loss in the segments incised, has a long-lasting effect, does not produce any motor loss, and is most effective for neuropathic pain states. Its disadvantages include requiring a laminectomy for each level to be lesioned, risk to spinal cord function with ipsilateral motor and dorsal column function loss (about 10% of patients), and being a major operation with significant surgical stress. The lesion in the root entry zone can be made with a radiofrequency electrode, a laser, or a knife. In the management of pain due to cancer, it is not clear that this operation is any better than dorsal rhizotomy, which certainly has less risk of spinal cord dysfunction.

An analogous operation can be performed on the descending trigeminal nucleus and root entry zone in the
medulla to treat chronic pain in the face and pharynx; this is much more commonly used for neuropathic pains than those due to cancer.²⁷ Only pain and temperature sensation is lost; corneal anesthesia does not ensue. The operation can be performed bilaterally if necessary.

This is an open operation with direct visualization of the brain stem and the localization of a lesion in the spinothalamic tract by anatomic features. This permits a higher section of the spinothalamic tract and a higher level of analgesia that includes the cervical region. The operation has been most frequently performed for cancer pain involving the shoulder and neck on the opposite side of the body. The success rate has been around 50% with a 15% complication rate, and the pain relief often does not last for more than a year.²⁸ This is a major surgical procedure and is rarely performed today; stereotactic surgical procedures and aggressive use of opioids have superseded this operation.

Open mesencephalic tractotomy had a much-too-high neurologic complication rate; in 1962 Spiegel and Wycis described stereotactic mesencephalic tractotomy. This operation has been widely used and is known to yield a 50% to 80% success rate; however, most patients will again report pain after a year or so.²⁹ The operative mortality is 5% to 10%, and about 20% of patients complain of transient dysesthesias or oculomotor disturbances. This operation is aimed at sectioning the spinothalamic fibers and adjacent spinoreticular fibers. A limited number of neurosurgeons are comfortable performing stereotactic lesions for pain. The surgical stress and morbidity are low; this operation does have utility for cancer pain patients. The pain is relieved on the opposite side of the body from the tractotomy and bilateral lesions can be made for bilateral pain. Head and neck as well as body pains can be controlled by mesencephalic tractotomy, as the trigeminal and spinal noxious afferent thalamic fibers are adjacent to each other.

This operation destroys a target area in the thalamus. A panoply of thalamic sites have been the target for stereotactic relief of pain. The results have been highly variable, certainly in part because of the different targets and lesion sizes (Table 21.3). The major issue with all of the sites has been the transient pain relief; good results do not often last more than 1 year. Producing a lesion in the thalamus by radiofrequency current, cryoprobe, or radiosurgery (Gamma Knife) may have significant side effects if the lesion is not precisely located in the desired area. Modern magnetic resonance imaging (MRI) scanning and neuronavigation equipment have reduced the risk of target errors.

Lesions in the ventrocaudal nuclei produce the loss of all sensory modalities as well as pain and for this reason they are not optimal for most patients. Lesions in the medial thalamus or pulvinar have been more useful as they only produce pain relief without sensory loss. Lesions in the anterior nuclei seem to resemble cingulumotomy in producing abatement of suffering rather than pain. Many other regions have been targeted, but the reported case series are small and results not reliable. Thalamotomy utilizing the Gamma Knife to avoid any surgical procedure is probably going to be used more in coming years.³⁰

This operation has rarely been performed and the true long-term results are unknown. It does not have a place in the current neurosurgical armamentarium.

Bilateral lesions are made in the mesial frontal lobes to interrupt the cingulum bundles. This operation was first performed by Foltz and White in 1962.²⁹ Large series of patients have been carefully studied and the clinical results are well established. The principle of this operation is to disconnect the limbic structures that process affective states from the spinothalamic afferents that transmit pain sensations. The short-term success rate for patients with cancer is about 80%; good data on long-term outcomes is sparse but we know there is some drop off.³⁰ Cingulumotomy does not alter sensation in any way. Patients report that they still have pain but that it does not cause them to suffer. It clearly does not alter cognition or judgment. Originally performed through bilateral frontal burr holes with radiofrequency lesions, it is now possible to perform cingulumotomy using stereotactic radiosurgery (Gamma Knife) and avoid any surgical procedure.

Shortly after World War II, resections of the postcentral and even precentral gyri were reported to be useful in the management of pain on the opposite side of the body. Follow-up reports and larger case series did not reveal many long-term good results; this operation is not performed today in the pursuit of pain relief.³¹

The recognition that some cancers were endocrine dependent led to the idea that hypophysectomy might remove the trophic hormones and retard tumor growth and dissemination. The initial procedures were performed subfrontally
via an open craniotomy. Then, innovative techniques such as transecting the pituitary stalk while preserving the gland, the use of stereotaxis to make a lesion within the pituitary fossa, and the injection of alcohol via a transsphenoidal route broadened the use of this operation. It was also recognized that endocrine sensitivity was not required for the patient to obtain pain relief; nor was tumor regression. Pain relief may start very rapidly after pituitary destruction. Significant complications including partial diabetes insipidus in 75.5%, and CSF leaks in 11.3% after transsphenoidal microsurgical hypophysectomy were reported in one series.³² Stereotactic radiosurgery (Gamma Knife) has greatly reduced the morbidity associated with pituitary operations. Approximately 70% of patients with disseminated cancer will gain significant pain relief for 3 months up to 1 year from this operation, no matter how the lesion in the gland is made.³² Hormone replacement therapy is usually required. The mechanism of pain relief is not understood. Pituitary ablation may have some utility for the disseminated cancer pain patient who has failed opioid therapy by oral and intrathecal routes.³³

Historically, a large number of agents have been used for chemical neurolysis. Some of these agents have included hypertonic cold saline solution, chlorocresol, ammonium salt solutions, osmic acid, and glycerol. Osmic acid is an aqueous solution of osmium tetroxide, which is an oxidizing agent with a high affinity for lipids and coagulate proteins. Following contact with organic substances, osmic acid is rapidly reduced to stable lower oxides and becomes inert. Osmic acid in a 1% aqueous solution has been used successfully for chemical synovectomy in patients with chronic knee synovitis.³⁴ Glycerol is a neurolytic agent found endogenously, and may act as a membrane stabilizing agent by inhibition of ectopic impulse generation at the site of nerve injury.³⁵ The neurophysiologic mechanism of glycerol neurolysis appears to be a nonspecific conduction block of large and small fibers, which is established within minutes of its application. Anhydrous glycerol application on normal nerves showed a rapid loss of C-fiber conduction within 5 minutes of application; after 10 to 30 minutes, a complete conduction block in all fiber types was produced.³⁵ Glycerol may damage thin unmyelinated fibers (C fibers), small-diameter (Aδ fibers), and large-diameter (Aβ) myelinated fibers. Morphological studies have demonstrated glycerolinduced damage of large-diameter myelinated fibers,³⁶ as well as damage of both small myelinated/unmyelinated and large myelinated fibers.³⁷, ³⁸ Electrophysiologic studies have suggested nonselective damage of small myelinated/unmyelinated and large myelinated fibers³⁵ and more selective damage of large myelinated fibers.³⁸ Absolute glycerol in volumes not exceeding 0.3 mL has been used in the treatment of trigeminal neuraglia.³⁹ Most of these agents are not easily available. Clinicians now use phenol or alcohol for chemical neurolysis.

Phenol (carbolic acid) is a neurolytic agent with local anesthetic properties. It is soluble in water to a maximum ratio of 1:15; consequently, at room temperature, a concentration of 6.7% cannot be exceeded for aqueous phenol. Commonly, 1-g amounts of phenol crystals are sterilized and stored in light-opaque 20-mL containers. Clinicians can add either water for injection to produce aqueous phenol or glycerin to produce hyperbaric solutions. Solutions of phenol are prepared by dissolving phenol crystals in saline, distilled sterile water, glycerin, or radiocontrast agents. When injected near motor nerves, phenol can produce flaccid paralysis and might also cause systemic complications, such as nausea and vomiting, central nervous system (CNS) stimulation, cardiovascular depression, and cardiac arrhythmias.⁴⁰ The ideal concentration of phenol for neurolytic treatment is not precisely defined but varies from 3% to 10%. Motor weakness, anterior and posterior root damage, and direct cord injury were noted in primates following epidural administration of 0.5 mL of 3% phenol.⁴¹ A concentration of 12% produces the maximal axonotomic effect.⁴² Concentrations >5% cause protein denaturation.

Alcohol (ethyl alcohol) concentrations greater than 35% are needed to produce neurolysis.⁴³ Its administration is associated with a burning sensation. Ethyl alcohol is commercially available as a colorless solution and can be easily injected through small-bore needles. A concentration of 40% alcohol seems to be equivalent to 3% phenol as a neurolytic agent.⁴⁴ Neurolysis is achieved by dehydration, with extraction of cholesterol, phospholipids, and cerebrosides, and precipitation of mucoproteins. This results in sclerosis of nerve fibers and the myelin sheath. There is nonselective destruction of nervous tissue by precipitation of cell membrane proteins and the extraction of lipid compounds, resulting in demyelination and subsequent Wallerian degeneration. By contrast with the action of phenol, alcohol often spares the Schwann cell tube which implies that the axon may regenerate over its former course. Chua and Kong⁴⁵ reported on the effect of alcohol neurolysis of the sciatic nerve for the treatment of hemiplegic knee flexor spasticity. Eight patients underwent neurolysis with 50% to 100% alcohol (mean concentration was 66.3%). The authors noted that all patients had improvement in terms of reduction of spasticity with minimal side effects (no patient had associated painful dysesthesias with injection) and beneficial effects lasted at least 6 months.

Thermal ablation involves energy sources with either heat (radiofrequency, laser, microwave) or cold (cryoablation).
Radiofrequency (RF) ablation implies the use of coagulation induction from all electromagnetic energy sources with frequencies less than 900 kHz; most devices function in the range 375 to 500 kHz.⁴⁶

Stay updated, free articles. Join our Telegram channel

Join