Neuraxial Infusion for Pain Control: When, Why, and What to Do After the Implant

May 1, 1999
Peter S. Staats, MD

Oncology, ONCOLOGY Vol 13 No 5, Volume 13, Issue 5

Neuraxial infusion therapy is an excellent option for selected patients with severe pain. Both epidural and intrathecal systems can be effective for multiple pains and are titratable, nondestructive, and very safe.

ABSTRACT: Neuraxial infusion therapy is an excellent option for selected patients with severe pain. Both epidural and intrathecal systems can be effective for multiple pains and are titratable, nondestructive, and very safe. Intraspinal therapy requires low opioid doses, has no motor, sensory, or sympathetic effects, and may have a lower side-effect liability than systemic therapy. Although most oncologists do not refer patients for intrathecal therapy, a recent study indicated that more oncologists would if they knew more about the therapy and if patients requested it. When permanent systems are used, close follow-up is essential. To obtain the maximum benefit from intraspinal therapy, pain management physicians and oncologists must communicate with each other and work together as partners. [ONCOLOGY 13(Suppl 2):58-62, 1999]


The use of neuraxial infusion for cancer pain first requires appropriate education for the oncologist. Many oncologists are not familiar with intraspinal analgesia and may not be able to contribute to clinical decisions involved in patient selection and timing of therapy. There are no strict indications or guidelines for this technique, and each physician must develop his or her own approach based on knowledge of the therapy and clinical judgment. Physicians who treat cancer patients must understand ongoing patient management after a permanent system is implanted. Oncologists may believe that once a device is implanted, the work is done. Implantable infusion therapy, however—much like medical management—involves multiple medications, multiple dose titrations, and constant monitoring to establish a favorable analgesia-to-side-effect profile.

Barriers to Pain Relief

The acceptance of intraspinal therapy has been impeded by numerous barriers. These include:

  • Lack of knowledge of the therapy.[1]

  • Lack of double-blinded, randomized, controlled trials demonstrating the efficacy of this therapy and its superiority over conventional therapies; the current efficacy data are limited to prospective and retrospective surveys.

  • Financial barriers

  • Fear of invasiveness

  • Barriers to providing pain relief in general.

Eight million people in the United States have cancer,[2] and 50% have pain, as estimated by the World Health Organization. Therefore, 4 million patients have cancer pain.[3] With optimal medical management, 75% to 90% of these patients attain pain relief. Four hundred thousand cancer patients in the United States have cancer pain each year that cannot be controlled with routine systemic therapy under the best of circumstances. True outcomes are actually much worse because of undertreatment.[4,5]

Medical therapies remain the mainstay of chronic cancer pain management. The World Health Organization Ladder for Chronic Cancer Pain Management (Figure 1) is a valuable tool, but was intended for both developing and developed countries and may not be sufficient for developed countries.[6] Between 75% and 90% of patients can have their pain controlled by following this ladder. It also provides a framework for managing patients with intractable pain. But the ladder offers no guidance for the management of side effects or the selection of patients for interventional therapies.

There are different approaches to managing the patient with pain that is poorly responsive to opioids. Oncologists are particularly good at managing side effects and know when palliative chemotherapy and palliative radiotherapy are indicated. They are not as familiar with neurolytic blocks, intraspinal infusion therapies, and neurosurgical options. In contrast, pain management physicians are more familiar with neurolytic blocks, infusion therapies, opioid rotation, and the use of coanalgesics. Clinicians must eliminate barriers between disciplines (Figure 2). The oncologist should understand when neurolytic blocks and infusion therapies are indicated and the pain physician must understand the full spectrum of palliative care with medical therapies, as well as when chemotherapy and radiation therapy are needed.

Role of Intraspinal Therapy

Although medical therapies remain the mainstay of pain control, side effects may limit the benefits of therapy. Side effects include sedation, mental cloudiness, and constipation.[7] It is at these times that we need to consider and position neuraxial infusion therapies in the spectrum of options.

Both the Agency for Health Care Policy and Research and the National Comprehensive Cancer Network have guidelines for interventional therapies such as neurolytic blocks and infusion therapies.[8] If these guidelines were followed, the number of patients who are treated with neurolytic blocks or infusion therapies would probably increase substantially.[9-11] There are also many more options for intraspinal treatment. With intrathecal clonidine, local anesthetics, and hydrophilic and lipophilic opioids, and even new drugs like intrathecal ziconitide, the role of interventional neuraxial infusions should be increasing.

It is possible that a more thoughtful approach to the use of invasive therapies might be based on assessment of the disease process and likelihood of responding to medical and interventional therapies. For example, a patient with pancreatic cancer may respond to a celiac plexus block early in the disease; this procedure has been shown to improve pain relief, mood, and, possibly, life expectancy.[12,13] If a patient with cancer has a new onset of back pain, the epidural metastasis must be considered, and radiation therapy may be recommended early. Likewise, there might be times when clinicians can predict that a patient would do poorly with systemic opioids, and could then intervene with intrathecal infusion approaches early to gain as much benefit as possible. More research is needed to explore this potential.

It is important for pain management specialists to discuss with colleagues in oncology the potential advantages of epidural and intrathecal infusions. Epidural and intrathecal infusions:

  • Can be effective for multiple pains

  • Are titratable

  • Are nondestructive and safe

  • Require low doses of opioids, and can have a lower side-effect liability than the systemic opioids

  • Offer the possibility of adding intraspinal coanalgesics

  • Can be cost-effective.[14]

Furthermore, neuraxial opioids have no motor, sensory, or sympathetic effects. The normal conversions from systemic to intrathecal opioids range anywhere from 200:1 to 600:1. The oral/epidural conversion for morphine is 30:1, and for oral/intrathecal conversion, 300:1. The lower dose needed to produce effects may lead to a better side-effect profile in most patients.[15]

Intrathecal Infusion Studies

The theoretical advantages of intraspinal therapy are complemented by several retrospective studies and case reports that intrathecal morphine is effective in the treatment of cancer pain.[15-18] Although most surveys of intraspinal therapy revealed high success rates, many patients entered into these surveys were not required to develop side effects prior to proceeding to an implanted device. It also is not clear how much work was done to control side effects before the patient received the implanted device, or whether opioid rotation was tried. For this reason, spinal therapy is usually still considered an approach for managing patients with pain that is refractory to an optimally administered systemic opioid regimen.

Need to Increase Awareness

Recently, a study assessing attitudes of oncologists toward interventional cancer pain management was conducted.[1] A series of qualitative (74) and quantitative (230) interviews were carried out in a sample group of 304 cancer health care specialists. Qualitative tests included focus group discussions and quantitative evaluations included computer-aided telephone interviews.

Pain assessment techniques used by cancer care personnel had little emphasis on the more accurate multidimensional assessment techniques. Extended release morphine sulfate preparations were found to be the most commonly prescribed analgesic in this project with lower doses (0 to 300 mg/day) considered to be within their margins of comfort by 73% of those studied. A total of 81% of oncologists reported patient referral to pain specialists to be a part of their current practice (Figure 3). The familiarity with intrathecal therapy was low, with only 46% having referred patients for intrathecal therapy in the previous 12 months. The invasive nature of device placement was deemed a drawback by 42% of physicians. The study indicates that continued education of both patient and physician is essential to ensure that all means of treatment are available to cancer pain patients.

Follow-up Is Essential

Implanting the device is clearly the easy part of intraspinal therapy. Knowing when to implant (indications), when not to implant (have reasonable medical therapies been exhausted), and, mostly, understanding the pharmacologic options are the keys to a successful outcome. There has to be a close communication between the oncologist and the pain management team to achieve optimal outcomes.

As the disease progresses, it would be a mistake to simply titrate the dose. One has to evaluate the patient, the severity of the pain, the location of the pain, and the type of pain, because each of these factors will impact how the clinician proceeds with therapy. Systemic therapies may be added for neuropathic pain; intraspinal opioids might be switched if side effects develop; or on occasion other interventional approaches could be considered.

Morphine, for example, is a very hydrophilic drug. It is a very good analgesic, but is associated with side effects. It spreads throughout the cerebrospinal fluid. Many patients will need to switch opioids and perhaps add to other coanalgesics.

Intrathecal opiates commonly used include morphine, hydromorphone (Dilaudid), sufentanil, fentanyl, and meperidine. The differences in lipid solubility and the relative starting doses are listed in Table 1. The nonopioids include the local anesthetics and clonidine. Other drugs, such as the neuron-specific calcium channel blockers, have been shown to be effective in chronic cancer pain.[19-22]

Options When Intraspinal Morphine Is Failing

The growing sophistication of intraspinal therapy is well-illustrated by exploring the options that exist when intraspinal morphine is failing. As Figure 4 shows, doses are usually increased to about 20 mg a day of morphine, and sometimes higher. If the pain is neuropathic, clonidine or bupivacaine can be added. If it is nociceptive, switching to more lipophilic opioids makes sense.

There are also other options. Intra-cisternal catheters are effective for head and neck pain. The catheter enters the upper thoracic spine and is placed in the cisterna magna for infusions of local anesthetics. In Appelgren et al’s group of 13 patients treated with intracisternal infusion, 11 had significant pain relief, going from a maximal pain intensity of about nine down to about two.[23]

With any pain management option, it is important that the pain specialists work as a team with the oncologists. To provide excellent patient care, pain specialists must provide feedback on patients by letters, copies of notes, and phone calls to oncologists, and by sending patients back to the referring physicians. Pain specialists can let oncologists know about pain management options, while respecting the role of palliative radiation therapy, chemotherapy, and hormonal therapy.


  • The World Health Organization Ladder for Chronic Cancer Pain Management is a very valuable tool for nonpain specialists.

  • Using the WHO ladder, 75% to 90% of patients will have adequate pain control.

  • Interventional strategies can improve pain control, symptom management, and outcomes in many patients with cancer.

  • Neuraxial infusion therapy is an excellent therapy for patients with severe pain.

  • A thorough knowledge of the pharmacology and options is essential for good patient outcomes.

  • Novel analgesics have improved the outcome of therapy.

  • Close communication with oncologists about options is needed.

Questions From the Audience:

Russell K. Portenoy, MD: You say that the management of implanted therapies does not stop after the device is implanted. Who should take responsibility for the patient after the pump is in?

Peter S. Staats, MD: Before the device goes in, the person who implants it needs to have a plan for that patient. In some centers, a neurosurgeon works well with an anesthesiologist, and the neurosurgeon implants the device and the anesthesiologist follows the patient. In some centers, the pain management physician is an anesthesiologist who implants and refills the pumps. Practically speaking, a pain management specialist could implant the device and have an oncologist follow the patient and the device, but it is rare for oncologists to be familiar with the full range of intrathecal pharmacologic options.

Dr. Portenoy: Would you advocate then that a referral for consideration of a pump be viewed by the oncologist as turning over the pain care to the anesthesiologist or turning over the palliative care to the anesthesiologist? There is clearly more needed than just pump refills.

Dr. Staats: Intrathecal therapy should be a benefit to the patient and oncologists. I know that oncologists may be concerned that they will lose their patient if they refer for this therapy. I would support working in a partnership with the oncologist.

Michael H. Levy, MD, PhD: I think that the partnership is really appropriate. We collaborate with cardiologists. The cardiologists will have their medications that the oncologists do not adjust, but if the oncologists see low blood pressure or high heart rate, they could make initial adjustments and then call the cardiologist. I think the consultative collaboration is very critical. The one thing that I have seen that is really problematic is that sometimes an oncologist will refer a patient to us who had a pump put in somewhere and the oncologist does not know what is in the pump, does not know when it needs to be refilled. So, I think the communication needs to go both ways.

Dr. Portenoy: Before we leave the issue of joint management, can you talk about how nurses or other professionals working with physicians fit into this and what specific instructions are given to patients and caregivers in terms of who they contact when they have problems? In other words, how do you make it simple for the people who are actually using the therapy? And what role do the professionals who work with the physicians play?

Dr. Staats: That would be an individual choice and it works differently in different centers. In our center, when patients have problems with their pain control, they will call my nurse. We have a contact person available during the week, 8 AM to 5 PM, who is a nurse that the patient knows. In the evening, there is a fellow who takes beeper calls all night and knows when to call me. So the patients really have access to either a nurse or a physician 24 hours a day, 7 days a week. It is important that the patient have access to someone knowledgeable about this therapy. It is also important to the oncologist to know that if someone gets into trouble with uncontrolled pain, there is no contraindication to adding systemic opioids. I have had implanted patients very close to the end of life and we add intravenous infusions of opioids because we’re not adequately controlling the pain and sedation is not an issue. There is no contraindication to doing that.

Elliot Krames, MD:I have a problem with more than one doctor prescribing pain medications. Intrathecal therapy is a prescription for pain medication and if the oncologist is going to go ahead and also prescribe pain medications on top of that, I think that you are going to get into problems. I believe that once a patient is referred for intrathecal therapy, then the physician who manages the intrathecal therapy should manage systemic opioid therapy as well and should manage the side effects from the opioid therapy. There is a lot for the oncologist to do managing medical therapy. In my practice, when we receive a referral for intrathecal therapy, there is an upfront agreement that we’re going to take over the pain management of that patient.

Dr. Staats: What I suggested was that this needs to be viewed as a partnership. The reality is that I do take much of the medical pain therapies, but I do not try to exclude the oncologist. This is always with the permission of the oncologist. There are times when the patient has a new metastasis and we need to discuss treatment with the oncologist. Is chemotherapy indicated here? Is radiotherapy indicated here? What is the role of steroids? I view it as a partnership.


1. O’Gorman DA, Staats PS, Traffas M: Attitudes to interventional pain management among cancer health care personnel. Abstract presented at the American Pain Society Annual Meeting, November 1998, San Diego, Calif.

2. NCI Fact Book. US Department of Health and Human Services, 1996, p. 36.

3. Cancer Facts and Figures. American Cancer Society, 1997.

4. Stjernsward J, Teoh N: The scope of the cancer pain problem. Adv Pain Res Ther 16:7-12, 1990.

5. Foley KM: Controversies in cancer pain: Medical perspectives. Cancer 6:2257-2265, 1989.

6. Staats PS, Carr D, Turk D: Cancer pain: The World Health Organization Ladder Revisited. Abstract presented at the American Pain Society Annual Meeting, November 1998, San Diego, Calif.

7. Staats PS, Hogan L: Strategies to manage opioid related side effects. Int Med (In press).

8. Jacox A, Carr DB, Payne R, et al: Clinical Practice Guideline Number 9: Management of Cancer Pain. Rockville, MD: US Dept of Health and Human Services, Agency for Heatlh Care Policy and Research; 1994. AHCPR publication 94-0592.

9. Dougherty P, Staats PS: Intrathecal analgesia: From basic science to clinical practice. Anesthesiology (Accepted with revisions).

10. Staats PS, Mitchell VM: Future directions for intrathecal analgesia. Prog Anesthesiology 19:367-382, 1997.

11. Staats PS: Cancer pain: Beyond the ladder. J Back Musculoskel Rehab 10:69-80, 1998.

12. Lillimore KD, et al: Chemical splanchnicectomy in patients with unresectable pancreatic cancer: A prospective randomized trial. Ann Surg 217:447-457, 1993.

13. Staats PS, Hemkat H, Sauter P, et al: The effects of alcohol, negative mood, and postoperative pain on life expectancy in patients with pancreatic cancer. Abstract presented at the American Pain Society Annual Meeting, October 1997.

14. Bedder MD, Burchiel K, Larson SA: Cost analysis of two implantable narcotic delivery systems. J Pain Sympt Manage 6:368-373, 1991.

15. Oakley J, Staats PS. Intraspinal infusion devices, in Raj PP (ed): Practical Management of Pain, 3rd ed. St. Louis, Mosby Year Book, 1999 (In press).

16. Follett KA, Hitchon PW, Piper J, et al: Response of intractable pain to continuous intrathecal morphine. A retrospective study. Pain 49:21-25, 1992.

17. Onofrio BM, Yaksh TL: Long-term pain relief produced by intrathecal infusion in 53 patients. J Neurosurg 72:200-209, 1990.

18. Hassenbusch SJ, Pillay PK, Magdinec M, et al. Constant infusion of morphine for intractable cancer pain using an implanted pump. J Neurosurg 73:405-409, 1990.

19. Sjorbert M, Appelgren L, Einalsen S, et al: Long-term intrathecal morphine and bupivacaine in ‘refactory’ cancer pain. Results from first series of 52 patients. Acta Anaesthesiol Scand 35:30-43, 1991.

20. Krames ES. Intrathecal infusional therapies for intractable pain: Patient management guidelines. J Pain Symptom Manage 8:36-46, 1993.

21. Kowal A, Staats PS: Intractable pain: A new technique for attack in patients with an implanted intrathecal infusion pump. Regional Anesth 22:584, 1997.

22. Staats PS, et al: Chronic intractable neuropathic pain: Marked analgesic efficacy of ziconotide. Abstract presented at the proceedings of the International Neuromodulation Society, September 1998, Lucerne, Switzerland.

23. Appelgren L, Janson M, Nitescu P, et al: Continuous intracisternal and high cervical intrathecal bupivacaine analgesia in refractory head and neck pain. Anesthesiology 84:256-72, 1996.