Access to the PPM Journal and newsletters is FREE for clinicians.
10 Articles in Volume 6, Issue #1
Do Topical Herbal Agents Provide Pain Relief?
Infusion Catheter Epidural
New Report of a High-Dose Morphine Metabolite
Pain Education and Pain Educators
Suspecting and Diagnosing Arachnoiditis
Tennant Blood Study, First Update
The Demise of Multidisciplinary Pain Management Clinics?
The Dimensions of Pain
The Role of Psychology in Pain Management

Infusion Catheter Epidural

Continuous infusion of lumbar epidural corticosteroid—via insertion of an implanted epidural catheter connected to an external infusion pump in an ambulatory patient—effectively reduces inflammation and pain by “bathing” the suspect spinal structure over a three day period.

A lumbar epidural injection of a corticoid has been the treatment of choice for radiculopathy for nearly fifty years. Patients who “fail” conservative care routinely are, in most cases, candidates for lumbar epidurals prior to consideration of surgery. Although approximately 80% of patients on whom the author performs epidurals have had some degree of pain relief, only about 50% have significant sustainable relief, and able to return to a fairly normal functional status. Unfortunately, the other 50% have either no relief, partial relief (i.e. 20% better after two epidurals), or transient relief (i.e. 90% better after three epidurals, but back to baseline by four weeks following the third epidural). After performing approximately 6,000 procedures over the past 14 years, the author has investigated why some patients do so well, and some others don’t.

Assuming all patients have the epidurals performed under flouroscopy with contrast, and that different approaches are attempted in failed patients (i.e. an interlaminar approach is attempted first with no success, and a transforaminal is attempted second, again with no success), what factors could otherwise explain this discrepancy? In the author’s opinion, variabilities in re-absorption of the corticosteroid may explain, in part, some of these clinical differences seen. The epidural space is extremely vascular, and the author has witnessed cases where the contrast dye is completely re-absorbed in as fast as three to four minutes but, at other times, may take significantly longer. One would postulate the same process must likely occur with the corticosteroid as well, particularly since corticosteroids are somewhat water-soluble. That may also explain why one may see some systemic effects of the corticosteroids in some patients but not in others. If this is true, is it possible that the corticosteroid just doesn’t linger long enough around the inflamed tissues to have a long lasting optimal effect?

Physiology of Locally Injected Corticoids

It has long been assumed that corticosteroids produce their pain-relieving effects when injected into joints or the epidural space by suppressing the inflammation, which may be either a cause or result of nerve injury. While this mechanism is at least partly responsible for the effectiveness of corticosteroid injection, it is also possible that some corticosteroids produce an anabolic effect at the cellular level.1 Varying clinical results may be obtained by various forms and analogues of cortisone. It is unknown whether corticoids metabolize while in a container or post-injection, but this is clearly possible because cortisol rapidly metabolizes into a number of compounds, such as androstenedione and dehydroepiandosterone (DHEA), which are clearly anabolic in nature.

To possibly enhance the effects in the past, corticosteroids had been injected in a reasonably insoluble base (i.e., “Depo”) form to keep the corticosteroid at the target area longer than could be achieved with a purely water-soluble corticosteroid. The corresponding enhanced clinical response, with a longer “soaking” in a “target” area, is essentially the same fundamental physiologic mechanism underlying the three-day epidural infusion technique described here. Simply put, exposing the inflamed tissues to a continuous infusion of corticosteroids may work better than just a one time bolus which is rapidly reabsorbed.

A New, Epidural Corticosteroid Infusion Technique

To achieve the advantage of longer term corticosteroid treatment, an indwelling catheter with an external infusion pump has been developed. (Infusion Catheter Epidural ICE Kit, Sgarlato Lab, San Jose, California). Previous to this report, 10 patients have been reported in a clinical trial.2 An additional 50 more patients, meeting the indications listed below, have since undergone the procedure. Results are very promising.

The technique requires an 18G Toughy spinal needle placed 1-3 levels above the affected suspect segment, using a retrograde approach, into the epidural space. The catheter, with stylet, is next advanced through the spinal needle and tunneled 1-3 segments in a caudal direction in the epidural space, securely implanting the catheter within the epidural space. Thus micro-movements of the catheter should not cause significant catheter migration, and should not dislodge the catheter out of the epidural space.

Spinal needle and catheter placement are verified with contrast dye under fluoroscopy. An initial bolus of injectate is initially given. The catheter is next externalized and subsequently tunneled in the subcutaneous tissues to further secure it and help reduce the risk of infection. The catheter is then attached to an external infusion pump. A combination of corticosteroid and normal saline are injected into the infusion pump, and a pre-set flow rate of 1cc per hour over 72 hours is implemented.


The ideal candidate for continuous corticoid epidural infusion is a patient who has had either a transient response (i.e. 80% improvement after one or two epidurals, but returns to baseline status within a few weeks) or a partial response (i.e. 25% better after two lumbar epidurals) to lumbar epidural injections. In this subset of patients, one could hypothesize that they did indeed have a favorable response to the bolus of corticosteroid injected into the epidural space based on their clinical improvement, albeit transient or partial. By now administering a greater dose of corticosteroid over a prolonged period of time (72 hours) to the inflamed tissues via a continuous infusion, one would anticipate a greater and/or more prolonged favorable clinical response. By breaking the pain cycle in these patients for a prolonged period of time, ICE may represent a surgery-sparing procedure in this population.


The infusion catheter epidural procedure is contraindicated in febrile or pregnant patients. Other contraindications include known anaphylic reaction to contrast dye, evidence of systemic or localized infection in the procedural field, a compromised immune system, bleeding diathesis, or concurrent use of anticoagulants.

Illustrative Case Report

Patient History

A 58 year old white male was referred by his orthopedic spine surgeon for consideration of performing the Infusion Catheter Epidural (ICE). He had three previous epidurals, all of which gave him only partial and transient relief of his pain. His pain was self-described as a sharp, achy pain, radiating down both legs, but much worse on the right. He had been symptomatic for at least ten months. His MRI demonstrated multilevel foraminal narrowing, most severe on the left at L3-4 and L4-5, and on the right at L4-5 and L5-S1. He also has multilevel central stenosis and lateral recess stenosis at L3-4. His EMG’s demonstrated bilateral L5 radiculopathies. He had endured extensive physical therapy, NSAID’s, exercise, and had significantly modified his activity but, despite all of this, was still symptomatic and progressively worsening. He had no other significant medical surgical history, and was not taking any medication. He had no known drug allergies, including medical dyes, and had no significant adverse reaction to the previous epidurals. He is a retired merchant sailor, and does not smoke or drink alcohol routinely. Review of systems was unremarkable. Physical exam was significant for weakness in the L5-S1 distribution, more pronounced on the right (for both plantar and dorsi-flexion).


After informed consent was obtained, the patient was brought into the operative suite and placed in the prone position. Prior to beginning the procedure, under aseptic conditions, the Infusion Pump was filled with a solution containing 69cc of sodium chloride and 3cc of 4.0mg/cc of Dexamthesone. (Dexamthesone was used because of its increased solubility in water in order to avoid particulation of the steroid in the pump). The on/off clamp of the pump was then placed in the open position, and proper flow of the pump was observed. The pump was next primed by releasing the green cap from the flow regulator to allow trapped air to escape. Once all of the trapped air was observed to be released, the green cap was secured to the flow regulator. The protective cap was then placed on the pump port to ensure sterility, and the on-off clamp was placed back in the closed position. The infusion pump was then set aside.

Following adequate local anesthesia, a three and a half inch 18G Touhy spinal needle with the bevel faced caudally, was slowly advanced under flouroscopic guidance utilizing a retrograde interlaminar approach. The spinal needle was angled into a cephalad to caudal direction at approximately sixty degrees into the Right L3-4 interspace without difficulty. This atypical positioning of the spinal needle was performed to allow for catheter placement. At this time, there was no withdrawal of blood or CSF with aspiration. Lateral view demonstrated proper placement of the spinal needle. Verification was initially obtained utilizing loss of resistance technique, followed by injection of a non-ionized based radio-opaque contrast medium through the spinal needle and visualized to be present in the epidural space under flouroscopy.

Epidurogram at this time demonstrated adequate flow of the dye in the epidural space, with no evidence of blockage, no intravascular flow of the dye, and no flow in the sub-arachnoid space. Following verification with the dye, 2cc of 40 mg/cc of alcohol-free Kenalog and 3cc of bacteriostatic normal saline were injected through the spinal needle at the L3-4 level due to significant disease at this level. Following this, a 20G radiopaque catheter with stylet was carefully inserted through the spinal needle and slowly advanced and tunneled in a caudal direction in the epidural space down to the Right L5 level, where it was visualized to be securely implanted under flouroscopic guidance. The stylet was then removed. There was no withdrawal of blood or CSF at this time. Contrast dye was again injected through the catheter, which verified continued epidural presence under flouroscopy. Epidurogram again demonstrated no intravascular flow, no sub-arachnoid flow, and no blockage of flow of the dye.

Following this final verification with the dye, the spinal needle was removed while the catheter was securely left in place. The catheter was then looped twice, and adequately taped utilizing a tegaderm patch adhered to the patients back. The catheter was next securely connected to the Infusion Pump by attaching an alligator clip to the catheter, and securing it to the pump. The Infusion Pump was attached to the accompanying patients belt harness. The on-off valve was then placed in the open position to begin flow. The flow regulator was pre-set at a flow rate of 1cc per hour. Discharge instructions were dispensed prior to the patient’s departure from the surgery center. The patient was instructed to follow-up with the author in 72 hours for evaluation and removal of the catheter. He was given a prescription of Ceftin for three days for infection prophylaxis.

Follow Up

At follow-up, 72 hours after insertion, the infusion pump was inspected and found to be intact with all of the medicine in the pump having been discharged. The patient related he was at least 80-90% improved, further stating it was the best he had felt in nearly a year. Subjectively, he had a much more dramatic improvement with ICE than with any of the three previous epidurals. He denied any adverse reaction or complications. He was afebrile, with all vital signs being stable. He was then brought into the operative suite where he was placed in the prone position. The tegaderm patch was gently removed. The patient was instructed to take a few deep breaths. Upon exhaling, the proximal end of the externalized catheter was gently pulled and advanced in a cephalad direction in a slow continuous motion, until it was completely removed from the patient. The catheter was closely inspected at this point and was found to be intact, with no sheared off parts.

That patient’s incision site was also closely inspected. There was no significant sero-sanguinous or purulent discharge noted. The surrounding skin was intact, with no significant erythema, or other abnormalities noted. No sign of infection was present. The wound site was non-tender to palpation. Skin tone was normal, with no increased warmth. Antibacterial ointment was then placed over the wound site, and a band-aid applied. The patient tolerated the procedure well. There were no complications to report. Follow up at three days, two weeks, and ninety days post-procedure revealed the patient continued to be doing extremely well. He maintained a subjective 80-90 percent improvement in his pain, and relates he was able to ambulate without any significant discomfort. His motor strength also significantly improved. He had no adverse reaction or complications to report.


The findings to date are very encouraging. Continuous epidural corticoid infusion over a period of three days represents a new advance in technology and provides the interventionalist with another weapon in his/her arsenal to combat inflammation and pain from spinal disorders in a safe and effective manner. Corticosteroids are time proven and—by delivering them in a larger volume, over a more prolonged period of time, with increased saturation to the inflamed tissues—may offer the patient another opportunity for pain relief and improved functional status without surgery. This technique may be useful to physicians of all persuasions in bringing corticoids and perhaps additional agents to a target site of pain and inflammation over a prolonged period. The benefits of a continuous “bathing” of affected tissues over an extended period of time appear to far exceed that of the heretofore single bolus, short-term corticoid injection.

Last updated on: May 16, 2011
close X