The workshop was designed to pick up on the major themes and unanswered questions raised in the plenary sessions. The audience (self-selected) turned out to be composed of people from a variety of clinical backgrounds and, importantly, representative of a wide variety of settings, from hospice sisters working with limited medical input to physicians working in pain services in tertiary institutions. Despite this difference in clinical context, there was surprising uniformity in agreeing on the problems that we all face and in the questions that people still regard as unanswered. Perhaps most important among the opening statements were the difficulties expressed by the nurses working in the circumstance of limited expert medical cover, limited either in the sense that there was no expert medical cover or by distance=mexpert cover was available, but only at a considerable distance. These people expressed a clear view that guidelines would improve the management of pain when simple measures were no longer adequate.
The session began with an attempt to define the role that invasive interventions played in cancer pain management; this arose from the fact that nerve blocks and neurosurgery had received relatively little attention in the program. The response from the audience suggested that for a substantial proportion there was no possibility of obtaining such interventions for their patients, even had the interventions been regarded as effective.
For those who did have access to neurosurgery and nerve blocks, the pattern emerged of considerable decrease in their use over the past 10 years. This decline appeared to be partly due to lack of availability and partly due to lack of evidence of effectiveness. Some participants reported that celiac plexus block could be of value in abdominal pain, specifically carcinoma of the pancreas. Few members of the audience referred patients for cordotomy, either percutaneous or by open neurosurgical approach. It was suggested that cordotomy could be of value in unilateral pain, but not at the high cervical level. It was therefore unlikely to be helpful in pain associated with a Pancoast's tumor. There was apparently little recourse to neurosurgical pain-relieving procedures other than cordotomy.
Epidural and intrathecal blocks with local anesthetic were used by some members of the audience when pain was difficult to control with conventional drug measures. There was little consensus about the effectiveness of epidural steroids as a therapeutic measure in the management of back pain due to metastasis. Epidural delivery of opioids alone or combinations of local anesthetic and opioids was discussed later.
The lecture by Dr. Portenoy was used as the framework to try to pinpoint areas in drug control of neuropathic pain where further research was needed. Most members of the audience had difficulty in controlling neuropathic pain adequately with opioids alone. Many audience members, most particularly the nurses working with limited expert medical support, expressed a clear need for guidelines to treatment of ``difficult'' pains. Most used the two drug classes, antidepressants and anticonvulsants, together with opioids, starting with antidepressants and using anticonvulsants if the antidepressants failed. It was not clear whether or not there was a consensus about the dose of antidepressant that one would rise to before declaring no response and switching to anticonvulsants. There was apparently little experience of the more selective antidepressants.
Among the anticonvulsants there were differences reported both in the drugs that practitioners used as first- and second-line and in the effectiveness reported. Some audience members were clearly impressed with the usefulness of anticonvulsants; others were skeptical. One of the points in Dr. Portenoy's lecture was the potential of oral formulations of local anesthetic drugs. Among those members of the workshop audience who had experience using these drugs in neuropathic pain, there were mixed reports of their effectiveness. Those who had tried intravenous local anesthetic challenges reported mixed feelings about the ability of the challenge to predict response with the oral formulation. It seemed unclear whether any one anticonvulsant was more or less effective than any other, independent of adverse effects. Professor Foley drew attention to the reports of subcutaneous local anesthetic infusion, but audience members appeared to have little experience of this route. The pharmacological control of neuropathic pain seems therefore to continue to be an obvious focal point for clinical research.
There was an interesting disparity in the views of the audience about the use of steroids for the management of neuropathic pain. On the one hand, there were audience members who used high doses of parenteral steroids to manage incipient or actual spinal cord compression but whose use of steroids appeared to be restricted to that circumstance, and, on the other hand, there were audience members who used oral steroids routinely in the management of neuropathic pain, using relatively small doses (2 to 4 mg of dexamethasone three times daily). Your reporter was struck by the fact that the nursing members of the audience who had limited medical cover appeared to be the strongest proponents of steroid use in this manner, despite protestations by medical audience members that tolerance occurred rapidly. This was denied vehemently. Clearly corticosteroid use is another area for the spotlight of research.
Some surprisingly negative views were elicited about the effectiveness of spinal opioids on their own in cancer pain management. One member of the audience described eloquently what was clearly considerable experience of implanting devices for opioid delivery, and concluded by saying that he did not believe that there was sufficient benefit to justify the intervention.
When the discussion moved to the topic of infusions of local anesthetic and opioid combinations there was a clear split in the audience, because experience of this approach appeared to be limited largely to those working outside the United States. Audience members from Australia and New Zealand led the discussion. They described the use of continuous infusions of low doses of local anesthetic combined with opioid, using simple syringe drivers (not implanted) attached to tunneled percutaneous epidural catheters. At optimal dosage pain was well controlled without sensory or motor block, and this technique was manageable in the patient's home. There was clear enthusiasm for this approach for patients who were not responding to simpler measures. Other audience members quite appropriately pointed out that similar claims had been made for spinal opioids in the past and that spinal opioids alone had not proved to be a panacea. The need for controlled clinical trials was stressed.
For your reporter, this was a useful workshop. It illustrated a number of the difficulties that we face in cancer pain management. The first problem is to ensure that the drugs that can provide pain control for the majority of patients are available, and it is clear that, while great strides have been made, there are still countries and states where patients suffer unnecessarily. The second problem is the management of those pains that do not respond to straightforward measures. Those working without expert medical cover expressed the wish for guidelines. Ideally such guidelines should be evidence-based, and it was clear in the workshop that there are still great gaps in the evidence. One difficulty is the difference in the clinical settings in which we all work. Some see predominantly ambulatory patients, whereas, at the other extreme, some care for terminal patients. This means that didactic guidelines may not be appropriate, because judgments about the balance between effectiveness and adverse effects will differ for patients who are still active members of the community compared to those who are confined to bed. The other difficulty is the lack of data from this patient population. Enthusiasms for novel approaches, such as the combination epidural infusions, should be underpinned by randomized controlled comparisons with the alternatives, even though such comparisons may be very difficult in this patient group. Our interventions should be both effective and appropriate.