Interventional Therapies for Cancer Pain Management: Important Adjuvants to Systemic Analgesics

JN058_Jrnl_50806Swarc.qxd 10/3/07 2:16 PM Page 851

851

Original Article

Anthony Eidelman, MD; Traci White, MD; and Robert A. Swarm, MD, St. Louis, Missouri

Key Words functional status, diminished appetite, insomnia, depres- Cancer pain, pain management, chronic pain, palliative care sion, anxiety, and loss of control or hope.1 Without adequate pain control, those with cancer not only Abstract experience the anguish of poorly controlled pain but also Optimized use of systemic analgesics fails to adequately control have greatly diminished quality of life and may even pain in some patients with cancer. Commonly used analgesics, have reduced life expectancy. Moreover, in the past 2 including opioids, nonopioids (acetaminophen and non-steroidal decades, as the mortality rates from certain malignan- anti-inflammatory drugs), and adjuvant analgesics (anticonvulsants and antidepressants), have limited analgesic efficacy, and their use cies have declined, the population of cancer survivors is often associated with adverse effects. Without adequate pain facing the hardship of chronic pain has increased. Cancer control, patients with cancer not only experience the anguish of patients who have poorly controlled pain during the poorly controlled pain but also have greatly diminished quality of acute disease phase are at increased risk for developing life and may even have reduced life expectancy. Interventional pain chronic pain.2 Fortunately, multiple modalities exist to therapies are a diverse set of procedural techniques for controlling pain that may be useful when systemic analgesics fail to provide manage cancer pain, including direct antineoplastic ther- adequate control of cancer pain or when the adverse effects of apies, systemic analgesics, interventional pain proce- systemic analgesics cannot be managed reasonably. Commonly used dures, and cognitive behavioral therapies. In most cases, interventional therapies for cancer pain include neurolytic neural the management of cancer pain begins with readily avail- blockade, spinal administration of analgesics, and vertebroplasty. able analgesics, as outlined by the World Health Compared with systemic analgesics, which generally have broad 3 indications for control of pain, individual interventional therapies Organization’s analgesic ladder (see steps 1–3 in Table 1). generally have specific, narrow indications. When appropriately Although systemic analgesics are effective in most can- selected and implemented, interventional pain therapies are cer patients, up to 21% of patients continue to have in- important components of broad, multimodal cancer pain manage- adequately controlled pain despite undergoing optimized ment that significantly increases the proportion of patients able treatment with systemic analgesics.4,5 When systemic to experience adequate pain control. (JNCCN 2007;5:851–858) analgesics are inadequate or when the risk/benefit ratio is favorable, various interventional procedures, includ- Inadequately controlled pain produces immense phys- ing peripheral nerve blocks, spinal analgesics, and ver- ical and psychological suffering, including decreased tebroplasty, may play an important role in cancer pain management (see step 4 in Table 1).

From the Division of Pain Management, Department of In certain circumstances, including vertebroplasty Anesthesiology, Washington University School of Medicine, for compression fracture or celiac plexus block for pan- St. Louis, Missouri. Submitted April 18, 2007; accepted for publication May 16, 2007. creatic cancer, intervention should not be delayed for ex- The authors have no financial interest, arrangement, or affiliation tensive trials of conservative therapy. Unnecessarily with the manufacturers of any products discussed in the article or their competitors. delaying interventional techniques may result in diffi- Correspondence: Anthony Eidelman, MD, Division of Pain cultly accomplishing specific interventions when ad- Management, Department of Anesthesiology, Washington University School of Medicine, Campus Box 8054, 660 South Euclid vanced care is necessary but the patient’s performance Avenue, St. Louis, MO 63110. E-mail: [email protected] status has declined.

852 Original Article

Eidelman et al.

Table 1 Modified Analgesic Ladder for Cancer has a narrow risk/benefit ratio and the potential to Pain Management cause significant untoward effects; therefore, neural destruction is generally considered when conserva- Step 1. Mild pain Use nonopioid analgesics (nonsteroidal anti-inflammatory tive modalities are exhausted. Neurolytic blockade of drugs or acetaminophen) somatic nerves may be complicated by postneurolysis Step 2. Moderate pain Use a “weak” opioid (codeine, neuritic pain or the development of deafferentation hydrocodone, oxycodone) pain, which may typically develop several weeks to along with nonopioid agents months after neurolysis. Concern for neuritic and deaf- Step 3. Severe pain Use a “strong” opioid (morphine, hydromorphone, oxycodone) ferentation pain generally restricts use of neurolytic along with nonopioid agents blockade of somatic nerves to patients with advanced Step 4. Pain not Interventional pain therapies malignancies or those with short life expectancy. controlled by systemic including neurolytic injections, Interventional pain procedures should be per- analgesics* vertebroplasty, and spinal formed by clinicians with the skill, training, and ex- analgesics (with continuation of systemic analgesics when pertise to safely perform these procedures and manage needed) potential complications. This is especially important in cancer pain management because unique chal- Adjuvant drugs, including antidepressants, corticosteroids, lenges are present when these interventions are per- or anticonvulsants, are recommended for each step of the ladder, when needed. formed in the setting of malignancy. Extensive *Earlier use advocated when risk/benefit ratio is favorable. surgery, radiation therapy, or tumor mass may alter anatomy, necessitating technique modification. Special attention to technique is required because When considering the use of interventional pain this population may have increased risk for infection therapies for cancer pain, an accurate clinical and ra- because of immunosuppression from chemotherapy, diologic assessment of the specific pain in question is steroids, or general debilitation. Furthermore, cancer essential to guide technique selection and implemen- patients are often opioid-tolerant and therefore, if tation.6 Localized cancer pain may be controlled by procedural sedation is required, significantly greater destroying peripheral nerves or plexuses using chem- doses may be required. ical or thermal modalities; however, diffuse, general- The goals of interventional pain therapies include ized cancer pain is unlikely to be controlled with 1) reduced pain intensity, 2) decreased requirement of selective peripheral neurolysis. Spinal analgesics, es- systemic analgesics to allow reduction in analgesic as- pecially when based on epidural or subarachnoid ad- sociated adverse effects, and 3) improved quality of ministration of combination analgesics involving life and functional status. Some studies suggest that, opioid, local anesthetic, or clonidine, are an especially through improving patients’ dietary intake and general potent tool for managing appropriate cancer-related function, interventional pain procedures may increase pain. Vertebroplasty plays a unique role in managing the survival of patients with cancer pain.7,8 However, pain from pathologic vertebral compression fractures other trials do not support this finding.9 Although an caused by osteoporosis, metastasis, or myeloma. association between pain control and survival is a Because many interventional therapies have a spe- possibility that needs further evaluation, the urgent cific anatomic site of action, whereas patients with need of cancer patients with inadequate pain control metastatic cancer may have multiple anatomic sites of from systemic analgesics should lead to more consis- pain,6 interventional pain procedures may not elimi- tent and timely use of interventional pain therapies. nate all pain but are best used as a component of mul- Furthermore, opioid-induced hyperalgesia may be seen timodal management of cancer pain. in cancer patients, especially during rapid dose esca- In the context of interventional cancer pain man- lation. In extreme cases, a paradoxical increase in pain agement, neuroablation is often accomplished through may occur in response to increasing doses of opioid. chemical neurolysis, also known as neurolytic neural With opioid-induced hyperalgesia, increased sensitiv- blockade. Nerve blocks with local anesthetic are of- ity to pain stimuli and pain that becomes more dif- ten performed before chemical neurolysis for both di- fuse seem to occur, extending beyond the distribution agnostic and prognostic purposes. Neurolytic blockade of pre-existing pain.10 Interventional therapies may

Original Article 853

Interventional Therapies for Cancer Pain Management

be especially indicated in management of cancer pain pain, with similar results for other upper abdominal when opioid-induced hyperalgesia is present. malignancies. After NCBP, patients often experience a significant improvement in appetite. One study7 found that patients undergoing NCPB experienced Sympathetic Blocks for Visceral Cancer Pain not only improved pain control but also improved Cancer patients often experience visceral pain aris- mood and longer survival compared with those un- ing from the deep organs of the thorax, abdomen, and dergoing saline (placebo) celiac block. However, a pelvis. Visceral pain tends to be vague and poorly lo- subsequent study9 failed to confirm improvement other calized compared with somatic pain.6 Afferent visceral than pain relief. Although effect on survival remains pain nerve fibers pass through the paravertebral sym- a topic for further investigation, NCPB is effective for pathetic ganglia, and therefore blockade of visceral pain management in upper abdominal malignancy. sympathetic innervation can be an effective tool in However, successful NCPB is less likely in patients controlling visceral cancer pain. with extensive disease involving somatic and neural structures. Furthermore, in patients with extensive Celiac Plexus and Splanchnic Nerve Block retroperitoneal disease encompassing the celiac plexus, Pancreatic cancer and other upper abdominal malig- spread of injectate may be poor, which may limit the nancies are frequently associated with severe pain and effectiveness of the block. nausea. Opioid analgesics provide good pain control Adverse effects of NCPB or NSB include tem- for most patients in these settings, but may provide porary orthostatic hypotension (caused by decreased inadequate relief to others. Opioids further impede sympathetic vascular tone) and diarrhea (from un- gastrointestinal motility and thereby contribute to opposed parasympathetic innervation to the alimen- nausea or constipation. Afferent pain fibers from up- tary tract). These symptoms generally resolve per abdominal viscera (including the stomach, liver, spontaneously within 1 to 2 days. Other potential gall bladder, pancreas, spleen, kidneys, adrenals, omen- complications include hematoma, infection, local tum, mesentery, small bowel, and proximal colon anesthetic toxicity, pneumothorax, hematuria, and through the splenic flexure) pass through the celiac injury to spinal nerve roots. Paraplegia is a devastat- plexus along the anterolateral surface of the aorta at ing but rare complication after NCPB, with an esti- the T12–L1 vertebral level. The visceral pain fibers mated occurrence of 1 in 700.11 The theoretical then pass through the splanchnic nerves to reach the mechanism of paralysis is injury to, or vasospasm of, spinal cord in the mid-to-lower thoracic levels. Neural the artery of Adamkiewicz, the major segmental ar- transmission of upper abdominal visceral pain can be tery that supplies the anterior two thirds of the distal blocked at either the level of the celiac plexus or spinal cord. NCPB remains a useful tool for pallia- splanchnic nerves. Local anesthetic celiac plexus or tion of abdominal pain in advanced malignancy; it po- splanchnic block may provide temporary pain relief tentially provides 70% to 90% of patients with good and can be used to clarify the extent to which a given to excellent pain relief, with risk for serious adverse patient’s pain may respond to neurolytic blockade. For effects comparable to that of other palliative thera- long-term pain relief, neurolytic celiac plexus block pies (chemotherapy, radiation therapy, or surgical in- (NCPB) or neurolytic splanchnic block (NSB) is per- tervention) often offered to patients with advanced formed with alcohol or phenol. Both blocks are per- malignancy for symptom control. formed under radiologic guidance using either fluoroscopy or computed tomography (CT). Typically, Superior Hypogastric Nerve Block NCPB or NSB provides pain relief for several weeks Superior hypogastric nerve block interrupts pain in- to a few months. These injections can be repeated if nervation of pelvic organs.12 Neurolytic superior hy- pain returns but, because of the short life expectancy pogastric plexus block (NSHB) may provide analgesia of persons with advanced upper abdominal malig- for malignancies involving the cervix, uterus, ovaries, nancy, repeat injection is rarely needed. proximal vagina, testes, prostate, bladder, descending NCPB is most commonly used in pancreatic can- colon, and rectum.13 However, successful NSHB may cer, where its use has been extensively studied.Good not be possible in patients with extensive retroperi- to excellent pain control is achieved in 70% to 90% toneal disease encompassing the plexus, because it of patients undergoing NCPB for pancreatic cancer may prevent appropriate spread of neurolytic solution.

854 Original Article

Eidelman et al.

Furthermore, NSHB is not likely to provide good Table 2 Head and Neck Nerve Blocks pain control if the tumor invades somatic or neural Nerve Block Indication structures. These blocks are performed with radio- Trigeminal Facial cancer pain in the distribution graphic imaging to guide needle placement. Although (gasserian) of the V2 and V3 divisions of the NSHB is generally well tolerated, potential compli- ganglion trigeminal nerve cations include hematoma formation, inadvertent Mandibular nerve Painful malignancy involving the injection into the iliac vessels, ureteral puncture, jaw and anterior two thirds of tongue lumbar or sacral nerve root injury, perforation of an Maxillary nerve Painful tumors of the middle third intervertebral disc, sexual dysfunction, and changes in of the face (maxilla, cheek, nasal cavity, hard palate) bowel or bladder function. Case reports suggest that Glossopharyngeal Localized pain arising from the base NSHB is an effective tool in managing pelvic visceral nerve of tongue, soft palate, or cancer pain, but no controlled clinical trials have been nasopharynx published. Impar Ganglion Block Neurolysis of the impar ganglion has been reported to deafferentation pain, and therefore should only be alleviate cancer pain originating from perineal vis- used in patients with intractable pain and short life cera.13,14 The impar ganglion, a solitary structure that expectancy. marks the termination of the paravertebral sympathetic chain, is located anterior to the sacrococcygeal Nerve Blocks for Thoracic Chest Wall Pain junction. Local tumor invasion, especially from rec- Chemical neurolysis of intercostal nerves may be used tal cancer, may decrease the efficacy of the block by for pain originating from the thoracic wall, including preventing spread of the neurolytic solution. metastatic lesions involving the ribs. One case series Complications include perforation of the rectum and describes 46 cancer patients whose pain was success- periosteal injection. No controlled studies are avail- fully managed with serial phenol intercostal nerve able to determine the efficacy of this injection. blocks,16 but little additional literature documents the efficacy of this technique. Pneumothorax is an infre- quent complication when the correct technique is used.17 Nerve Blocks for Head and Neck Cancer Pain Other potential complications include vascular up- Head and neck cancer pain is often challenging to take of the injected solution and a single report of manage because of the erosive nature of the malig- paraplegia.18 Neurolytic thoracic paravertebral block nancy, dense sensory innervation of the region, and can be used for pain involving a small number of tho- regional anatomic changes from surgical or radiation racic segments.19 The procedure involves injection of therapy. Although most head and neck cancer pain solution in the space immediately lateral to the ver- is controlled with systemic analgesics,15 neural block- tebral body to target the somatic spinal nerves as they ade may be indicated for refractory pain. Carefully emerge from the intervertebral foramina. Because of assessing the location and nature of the pain is im- the complex anatomy of the thoracic region, experts portant to applying the most appropriate nerve block recommend that the technique be performed with ra- (Table 2). Neural destructive lesions in the head and diologic imaging.20 Complications include pneumoth- neck must be managed carefully to avoid increasing orax and inadvertent spread of solution into the patient morbidity through excessive or unintended epidural or spinal space. Because of the potential for sensory or motor deficits. A trial of local anesthetic significant untoward events, neurolytic intercostal block is often used to help determine the extent of and thoracic paravertebral blocks are limited to in- sensory or motor deficit expected from neurolysis. tractable pain in cancer patients with poor prognosis. Neurolytic blockade of the gasserian (trigeminal) Neurolytic blockade of the chest wall is less com- ganglion has been largely replaced by radiofrequency monly used now than in the past because of the avail- lesioning techniques that provide better control of ability of spinal analgesic techniques (see next page), lesion distribution and greater analgesic efficacy. which offer greater efficacy and lower risk for adverse Peripheral neurolysis in the head and neck may be effects. Spinal analgesic therapies have the added associated with delayed development of neuritis or burden (and cost) of an implanted pump/reservoir

Original Article 855

Interventional Therapies for Cancer Pain Management

and spinal catheter, in addition to the ongoing need for pain relief or the burden of undergoing repeated for refill and maintenance of the spinal drug admin- procedures. istration system. Furthermore, an emerging therapy for palliating chronic chest wall pain is pulsed radiofre- Spinal Administration of Analgesics quency of the dorsal root ganglion or intercostal The potential usefulness of spinal analgesics, includ- nerves.21 ing epidural and subarachnoid (intrathecal) administration, has greatly expanded in recent years. Spinal Neuroaxial Neurolysis analgesic therapies for cancer pain are not limited to Spinal neurolytic blocks are used less frequently now opioid alone but often involve custom compounded than in prior decades, likely reflecting a more sophis- solutions, which may include opioid, local anesthetic, ticated use of systemic and other spinal analgesic clonidine, or ziconotide.24 Simple, percutaneous therapies and increased awareness of potential com- epidural, or subarachnoid (intrathecal) catheters are plications of neurolytic techniques. Although no still used to provide pain relief to patients with ad- prospective studies have been reported of spinal neu- vanced disease and very short life expectancy (days rolytic techniques in the management of cancer pain, to weeks), but for long-term use (several weeks to case series indicate a potential role for these proce- years), subarachnoid catheters with implanted infusion dures when other analgesic measures have failed. The pumps are used most often. Examples of settings in subarachnoid injection of a neurolytic agent is a tech- which patients may benefit from spinal analgesics nique that aims to produce a chemical posterior rhi- include brachial plexopathy from axillary recurrence zotomy, thereby interrupting transmission of pain of breast cancer or Pancoast tumor, lumbar or sacral signals from a painful area.22 Subarachnoid neurolytic metastases with radicular pain, or intractable pain blocks are best used for bilateral perineal pain in from intra-abdominal malignancies.23 terminally ill patients with preexisting colostomy The improved analgesic efficacy of spinal opioid and permanent bladder catheter (because of risk for administration compared with systemic administra- postneurolysis incontinence) or in those with well- tion is caused by regional delivery of opioid close to localized, unilateral pain of the chest wall. Published spinal sites of action, with reduction in adverse effects data suggest that subarachnoid neurolysis may pro- (especially sedation) through reduced delivery of opi- vide good results in as many as 60% of patients, with oid to the brain. Sedation and other opioid-related relief lasting for up to 12 months.23 With appropriate adverse effects may be further reduced with spinal ad- patient selection and meticulous technique, compli- ministration of nonopioid analgesics. If the catheter cation rates range from 1% to 14%, which may be tip is close to the spinal dermatomal level of pain, acceptable for certain patients. For example, if a blad- moderate doses of spinal local anesthetic may produce der catheter is already used, bladder incontinence marked analgesia without significant muscle weak- may not be an issue; however, ambulatory, continent ness. Clonidine has significant analgesic effect when patients will almost never accept the possibility of administered spinally, especially when used for neuro- incontinence even if pain relief through alternative pathic pain. Ziconotide, a synthetic analogue of a pep- methods is inferior. tide toxin found in venomous marine cone snails, is a Epidural injection on neurolytic solution is gen- potent spinal analgesic but clinical use has been lim- erally considered for intractable pain only when other ited by significant adverse effects. With the availabil- interventions, such as spinal analgesics, subarachnoid ity of revised dosing guidelines shown to minimize neurolysis, or sympathetic neurolytic blockade, are adverse effects,25 ziconotide will likely be used increas- judged inappropriate. Epidural neurolysis is typically ingly for resistant cancer pain. Although a potent tool accomplished with daily injections of small volumes for pain control, combinations of spinal analgesics of neurolytic solution administered through a tempo- must be used cautiously. It is well recognized that rary percutaneous catheter over a few days, so that the the efficacy of spinal opioid may be limited by seda- extent of neurolysis is titrated to the degree required.22 tion, but nonopioid spinal analgesics also have dose- Although they improve the accuracy of neurolysis, re- limiting adverse effects: local anesthetic may cause peated epidural neurolytic injections over several days numbness, weakness, and hypotension, and clonidine may be unacceptable because of the resultant delay may cause sedation, hypotension, and bradycardia.

856 Original Article

Eidelman et al.

With careful selection and dose titration, combina- needles. PV and PK provide good to excellent pain tions of spinal analgesics may be used effectively, and relief in 80% to 90% of persons with painful compres- often with tolerable adverse effects, for palliation of sion fractures caused by osteoporosis and 50% to otherwise intractable cancer pain. 85% of those with painful compression fractures of Percutaneous epidural catheters are commonly neoplastic origin.31 Neither PV nor PK need delay used for epidural analgesia for acute pain (postopera- treatment of spinal metastases with radiation therapy, tive, traumatic, obstetrical) but also can be used in but instead may provide rapid onset of pain relief, cancer pain management when pain relief is needed which may facilitate the use of appropriate antitumor for short periods (e.g., pathologic fracture pending sur- therapies. gical fixation, severe pain limiting positioning for In PV, the bone cement is injected through the diagnostic imaging or radiation therapy, severe pain needles into the interstices of the vertebral body at the very end of life). Percutaneous subarachnoid marrow space. PK is similar to PV, with the addition catheters have been used in cancer pain management, of steps to include temporary placement of a high- but special attention to sterile technique is needed to pressure balloon into the vertebral body. Inflating the minimize the incidence of bacterial meningitis.26,27 PK balloon develops a cavity within the vertebral body Simple percutaneous spinal catheters are inexpensive that is subsequently filled with bone cement. PK bal- and require little or no specialized equipment for place- loon expansion may partially restore vertebral height, ment in most patients, allowing their use even in ter- but lack of consensus exists regarding the selection of minal cancer patients who may be unable to be PV versus PK for managing painful vertebral compres- transferred to a hospital or procedure center for more sion fractures.32 Comparative studies between PV and invasive therapies.28 Over several weeks to months, PK are currently unavailable. Serious complications re- risks for infection or catheter displacement increase sig- lated to PV or PK are rare but can be devastating.33 The nificantly with percutaneous spinal catheters. For bone cement can extrude outside the vertebral body longer-term use, subarachnoid catheters are tunneled into the spinal canal, causing neural compromise. to implanted infusion pumps. Implanted pumps are Furthermore, venous embolism of cement may occur, refilled every 1 to 2 months using a needle placed per- producing symptomatic pulmonary embolism or even cutaneously through an injection port into the pump cardiovascular collapse. Because the risk for serious reservoir. If the efficacy of spinal analgesia is uncertain adverse effects is low and the likelihood of providing in a specific clinical setting, a temporary trial of spinal good to excellent pain relief is reasonably high, PV analgesics (through a percutaneous catheter) is typi- and PK are good options for managing pain caused cally undertaken before proceeding with spinal pump by vertebral compression fractures. implantation. Guidelines directing the use of spinal When considering PV or PK, radiologic exami- 24 29 analgesics and techniques are available, but the clin- nation of the spine is necessary to evaluate the loca- ical use of spinal analgesics therapies remains largely tion and number of involved vertebrae. Neuraxial empiric. Prospective clinical trials are needed to fur- imaging (magnetic resonance imaging [MRI] or CT ther clarify the role of spinal analgesic therapies in scan) is used to evaluate possible extension of bone managing various cancer pain syndromes. fragments or tumor into the spinal canal. PV and PK are most likely to result in pain improvement when Percutaneous Vertebroplasty and used to treat recent compression fractures, although Kyphoplasty for Painful Compression pain relief can be obtained from fractures older than Fractures 1 year.34 MRI is the most useful imaging modality for Percutaneous vertebroplasty (PV) and kyphoplasty detecting acute vertebral fractures, because vertebral (PK) are related procedures for managing painful marrow edema (indicative of an acute compression vertebral compression fractures caused by osteoporo- fracture) is readily identified. Bone scan is less useful sis, primary neoplasm (e.g., multiple myeloma), or because it will identify increased activity in compressed metastatic tumor.30,31 These image-guided minimally in- vertebrae for up to 2 years after onset of the fracture, vasive procedures stabilize a compressed vertebra by long after the fracture may have spontaneously stabi- introducing polymethylmethacrylate bone cement lized. PV of PK can be performed at more then one ver- into the compressed vertebral body through large-bore tebral level if multiple compression fractures are

Original Article 857

Interventional Therapies for Cancer Pain Management

present. PV and PK are important tools in the multi- future experts will consider genetic profiling to guide modal management of malignancy-related spinal interventional pain treatment planning. pain.32 Percutaneous sacroplasty is a variation of vertebroplasty that involves the injection of bone cement into Conclusions symptomatic sacral insufficiency fractures.35 Sacroplasty Although most cancer pain is controlled with oral has been shown to provide rapid relief and stabiliza- analgesic therapy, interventional pain therapies play tion of primary sacral tumors and metastases.35,36 a valuable role in the multimodal management of can- Furthermore, several percutaneous ablative techniques cer pain. Oncologists and palliative care specialists have been developed for palliation of painful bone must understand the precise indications and appro- metastases. Radiofrequency ablation is a minimally priate timing of pain procedures so that the interven- invasive procedure that destroys cancerous tissue us- tions can be successfully implemented. As further ing local thermal lesions.37 Percutaneous cryoablation advances are made in interventional therapies, their uses freezing to produce destruction of tissue,38 and mi- effectiveness will likely improve and their indications crowave ablation is an emerging technique that uses may expand. Because clinical use of interventional electromagnetic energy to eradicate primary tumor or pain therapies is currently largely empiric, an impor- bony metastases.39 Benefit may be associated with com- tant goal is to improve the quality and volume of clin- bining interventional procedures, such as vertebro- ical research to formulate evidence-based practice plasty and radiofrequency ablation of medial branch paradigms for interventional pain therapies. For- nerves, in managing resistant pain from vertebral com- tunately for persons with inadequately controlled can- pression fractures.40 Further studies are warranted to cer pain, interventional therapies are available that compare the efficacy and safety of each ablative ther- may significantly enhance pain relief. apy and to determine the feasibility of combination techniques. References 1. Ingham JM, Foley KM. Pain and the barriers to its relief at the end of life: a lesson for improving end of life health care. Hosp J 1998; Emerging Areas of Research 13:89–100. Pulsed radiofrequency (PRF) is an emerging interven- 2. Burton AW, Fanciullo GJ, Beasley RD, Fisch MJ. Chronic pain in the tion that uses brief bursts of radiofrequency energy to cancer survivor: a new frontier. Pain Med 2007;8:189–198. nondestructively alter central and peripheral compo- 3. Cancer pain relief and palliative care: report of a WHO expert com- mittee. Technical Report Series 804. Geneva: World Health nents of the nociceptive pathways. PRF is a potential Organization; 1990. substitute for neurodestructive techniques such as 4. Meuser T, Pietruck C, Radbruch L, et al. Symptoms during cancer chemical neurolysis or radiofrequency thermal abla- pain treatment following WHO-guidelines: a longitudinal follow- tion. PRF is appealing because it does not produce up study of symptom prevalence, severity and etiology. Pain 2001; 93:247–257. deafferentation pain or neuritis. PRF is performed by 5. Ventafridda V, Tamburini M, Caraceni A, et al. A validation study percutaneously inserting electrodes in proximity to of the WHO method for cancer pain relief. Cancer 1987;59:850–856. target neural structures, including dorsal root ganglion 6. Eidelman A, Carr DB. Taxonomy of cancer pain. In: de Leon-Casasola or peripheral nerves. The electrical field produced by OA, ed. Cancer Pain: Pharmacologic, Interventional, and Palliative PRF has been shown in vitro to alter synaptic trans- Approaches. Philadelphia: Saunders Elsevier; 2006:3–12. 41–43 7. Staats PS, Hekmat H, Sauter P, et al. The effects of alcohol celiac mission and produce neurobiologic effects. The plexus block, pain, and mood on longevity in patients with unre- clinical literature supporting the efficacy of PRF is sectable pancreatic cancer:a double-blinded, randomized, placebo- growing, and recently a randomized doubled-blinded controlled study. Pain Med 2002;2:28–34. trial supported the efficacy of PRF for palliation of 8. Smith TJ, Staats PS, Deer T, et al. Randomized clinical trial of an 44 implantable drug delivery system compared with comprehensive chronic pain. In the future, PRF could have signifi- medical management for refractory cancer pain: impact on pain, cant usefulness in treating cancer pain. Much atten- drug-related toxicity, and survival. J Clin Oncol 2002;20: tion has been focused recently on pharmacogenetics: 4040–4049. the genetic-mediated, interindividual differences in 9. Wong GY, Schroeder DR, Carns PE, et al. Effect of neurolytic celiac 45 plexus block on pain relief, quality of life, and survival in patients drug response. This individual variability may also with unresectable pancreatic cancer: a randomized controlled trial. extend to interventional therapies. Perhaps in the JAMA 2004;291:1092–1099.

858 Original Article

Eidelman et al.

10. Mao J. Opioid-induced abnormal pain sensitivity: implications in 28. Mercadante S. Intrathecal morphine and bupivacaine in advanced clinical opioid therapy. Pain 2000;100:213–217. cancer pain patients implanted at home. J Pain Symptom Manage 11. Davies DD. Incidence of major complications of neurolytic coeliac 1994;9:201–207. plexus block. J R Soc Med 1993;86:264–266. 29. Burton AW, Rajagopal A, Shah HN, et al. Epidural and intrathecal 12. Plancarte R, de Leon-Casasola OA, El-Helaly M, et al. Neurolytic analgesia is effective in treating refractory cancer pain. Pain Med superior hypogastric plexus block for chronic pelvic pain associated 2004;3:239–247. with cancer. Reg Anesth 1997;22:562–568. 30. Burton AW, Reddy SK, Shah HN, et al. Percutaneous vertebroplasty— 13. de Leon-Casasola OA. Critical evaluation of chemical neurolysis of a technique to treat refractory spinal pain in the setting of advanced the sympathetic axis for cancer pain. Cancer Control 2000;7:142–148. metastatic cancer: a case series. J Pain Symptom Manage 2005;30: 87–95. 14. Ho KY, Nagi PA, Gray L, et al. An alternative approach to ganglion impar neurolysis under computed tomography guidance for recur- 31. Jensen ME, Kallmes DE. Percutaneous vertebroplasty in the treat- rent vulva cancer. Anesthesiology 2006;105:861–862. ment of malignant spine disease. Cancer J 2002;8:194–206. 15. Grond S, Zech D, Lynch J, et al. Validation of World Health 32. Mathis JM, Ortiz AO, Zoarski GH. Vertebroplasty versus kypho- Organization guidelines for pain relief in head and neck cancer. A plasty: a comparison and contrast. AJNR Am J Neuroradiol prospective study. Ann Otol Rhinol Laryngol 1993;102:342–348. 2004;25:840–845. 16. Doyle D. Nerve blocks in advanced cancer. Practitioner 1982;226:539, 33. Nussbaum DA, Gailloud P, Murphy K. A review of complications as- 541–544. sociated with vertebroplasty and kyphoplasty as reported to the Food and Drug Administration medical device related web site. J Vasc 17. Moore DC, Bridenbaugh LD. Intercostal nerve block in 4333 patients: indications, technique, and complications. Anesth Analg Interv Radiol 2004;15:1185–1192. 1962;41:1–11. 34. Brown DB, Gilula LA, Sehgal M, et al. Treatment of chronic symptomatic vertebral compression fractures with percutaneous vertebro- 18. Kowalewski R, Schurch B, Hodler J, et al. Persistent paraplegia after an aqueous 7.5% phenol solution to the anterior motor root plasty. AJR Am J Roentgenol 2004;182:319–322. for intercostal neurolysis: a case report. Arch Phys Med Rehabil 35. Uemura A, Matsusako M, Numaguchi Y, et al. Percutaneous 2002;83:283–285. sacroplasty for hemorrhagic metastases from hepatocellular carci- noma. AJNR Am J Neuroradiol 2005;26:493–495. 19. Antila H, Kirvela O. Neurolytic thoracic paravertebral block in cancer pain. A clinical report. Acta Anaesthesiol Scand 1998; 36. Masala S, Konda D, Massari F, Simonetti G. Sacroplasty and iliac 42:581–585. osteoplasty under combined CT and fluoroscopic guidance. Spine 2006;31:E667–669. 20. Purcell-Jones G, Pither CE, Justins DM. Paravertebral somatic nerve block: a clinical, radiographic, and computed tomographic study in 37. Goetz M, Callstrom M, Charboneau J, et al. Percutaneous image- chronic pain patients. Anesth Analg 1989;68:32–39. guided radiofrequency ablation of painful metastases involving bone: a multicenter study. J Clin Oncol 2004;22:300–306. 21. Cohen SP, Sireci A, Wu CL, et al. Pulsed radiofrequency of the dorsal root ganglia is superior to pharmacotherapy or pulsed radiofre- 38. Robinson D, Yassin M, Nevo Z. Cryotherapy of musculoskeletal quency of the intercostal nerves in the treatment of chronic tumors—from basic science to clinical results. Technol Cancer Res postsurgical thoracic pain. Pain Physician 2006;9:227–235. Treat 2004;3:371–375. 22. Patt RB, Cousins MJ. Techniques for neurolytic neural blockade. In: 39. Fan QY, Ma BA, Qlu XC. Preliminary report on treatment of bone Cousins MJ, Bridenbaugh PO, eds. Neural Blockade in Clinical tumors with microwave-induced hyperthermia. Bioelectromagnetics Anesthesia and Management of Pain, 3rd ed. Philadelphia: 1996;17:218–222. Lippincott-Raven; 1998:1007–1061. 40. Masala S, Roselli M, Massari F, et al. Radiofrequency heat ablation 23. Swarm RA, Karanikolas M, Cousins MJ. Anaesthetic techniques for and vertebroplasty in the treatment of neoplastic vertebral body frac- pain control. In: Doyle D, Hanks G, Cherny NI, et al., eds. Oxford tures. Anticancer Res 2004;24:3129–3133. Textbook of Palliative Medicine, 3rd ed. Oxford: Oxford University 41. Higuchi Y, Nashold BS Jr, Sluijter M, et al. Exposure of the dorsal Press; 2004:378–396. root ganglion in rats to pulsed radiofrequency currents activates dor- 24. Hassenbusch SJ, Portenoy RK, Cousins M, et al. Polyanalgesic sal horn lamina I and II neurons. Neurosurgery 2002;50:850–855. Consensus Conference 2003: an update on the management of pain 42. Van Zundert J, de Louw AJ, Joosten EA, et al. Pulsed and continu- by intraspinal drug delivery—report of an expert panel. J Pain ous radiofrequency current adjacent to the cervical dorsal root gan- Symptom Manage 2004;27:540–563. glion of the rat induces late cellular activity in the dorsal horn. 25. Rauck RL, Wallace MS, Leong MS, et al. A randomized, double- Anesthesiology 2005;102:125–131. blind, placebo-controlled study of intrathecal ziconotide in adults 43. Hamann W, Abou-Sherif S, Thompson S, et al. Pulsed radiofre- with severe chronic pain. J Pain Symptom Manage 2006;31: quency applied to dorsal root ganglia causes a selective increase in 393–406. ATF3 in small neurons. Eur J Pain 2006;10:171–176. 26. Du Pen S. Complications of neuraxial infusion in cancer patients. 44. Van Zundert J, Patijn J, Kessels A. Pulsed radiofrequency adjacent Oncology (Williston Park) 1999;13:45–51. to the cervical dorsal root ganglion in chronic cervical radicular 27. van Dongen RT, Crul BJ, De Bock M. Long-term intrathecal infu- pain: a double blind sham controlled randomized clinical trial. Pain sion of morphine and morphine/bupivacaine mixtures in the treat- 2007;127:173–182. ment of cancer pain: a retrospective analysis of 51 cases. Pain 45. Mogil JS, Sternberg WF, Marek P, et al. The genetics of pain and 1993;55:119–123. pain inhibition. Proc Natl Acad Sci USA 1996;93:3048–3055.