Follow Us On
Contact Us
info@ampainsoc.org847.375.4715
American Pain Society
4700 W Lake Ave
Glenview, IL 60025
Mark Sullivan, MD PhD, Department Editor
Alex Cahana, MD DAAPM
Department Editor’s Note: The cases below were presented to the Ethics Special Interest Group at the 2002 American Pain Society Meeting in Baltimore. They illustrate some of the more confusing, unintended consequences of interventional pain management at the end of life. They serve to remind us that pain occurs within the complex homeostatic system of the person.
A high prevalence of unrelieved pain has been reported in medical and surgical wards, as well as oncology clinics, despite decades of effort to provide clinicians with various therapeutic alternatives (American Pain Society Quality of Care Committee, 1995). However, “high technology” pain treatment interventions such as patient-controlled analgesia, epidural and subarachnoid catheters, and implantable intrathecal pumps, are increasingly being used for the minority of cases in which “simpler” measures do not suffice (American Pain Society Task Force on Pain, Symptoms, and End of Life Care, 1997).
Evidence-based reviews of the literature on intrathecal delivery of pain medications and clinical guidelines as well as long-term outcome studies are available (Portenoy & Hassenbusch, 2000). Quality of analgesia on appropriately selected and trailed patients is good, reliable, and reproducible and incidence of side effects or complications appears acceptable (Wallace & Yaksh, 2000). Nonetheless, the use of implantable technology is new enough that data is still insufficient to make authoritative statements about its overall value. Thus the answer to the question, “Is optimal pain relief optimal for patients at an end-of-life context,” remains unclear.
Three patients in our institute, all suffering from advanced malignancies, presented to our service with intractable pain and unacceptable side effects, notably mental clouding (Table 1). All underwent a 7-day epidural trial with fentanyl and bupivacaine, and two of them were implanted with a programmable intrathecal pump by a standard technique described elsewhere (Patt & Hassenbusch, 2001). Mr D. refused a pump and continued his previous regimen.
| |||||||||||||||||||||||||
After receiving their intrathecal pump, both patients had more than 50% pain relief with a marked improvement of their cognitive impairment (Table 2). However, associated with this improvement was a significant increase in depression, anxiety, and diminished well-being (as measured by Beck Depression Inventory, Speilberger Anxiety Trait Inventory, and Edmonton Symptom Assessment Scale). Patients were treated in-hospital at our palliative care center, and were seen by members of our psychology services. Though activities of daily living improved over time, both patients presented conjugal conflicts and entered a family therapy program.
| ||||||||||||||||
The nature of these conflicts included interpersonal disputes between patient and significant other (“Stop asking me just about the pain and the nausea.”), as well as vice versa (“Stop telling me what to do, I am not your slave.”). Matters regarding previous familial stressors such as children, ex-spouses, and in-laws, became predominant issues of discontent. Finally, cancer and death itself emerged as an extreme emotional burden (“You were dead already, what are you doing here again? I can’t take it to lose you again.”).
These two cases exemplify the complexity of unforeseen consequences following our medical interventions. Indications for implantable devices are seemingly straightforward, workup as well as procedures are standardized, and probable outcomes can be conveyed to our patients. If this is the case, why did our patients have this unfavorable evolution? Could we have anticipated these risks? If not, should this influence our therapeutic decisions?
Attempting to analyze this outcome is completely speculative. Lack of euphoric effects with intrathecal opioids (being a dose-dependent phenomenon) may contribute directly to the apparition of depression and anxiety. Similarly, the improved cognitive perception itself may lead patients to realize their end-of-life situation and precipitate a depressive state. Or, the new cognitive state may impose renewed conjugal dynamics already predetermined in the previous “catatonic” state resulting in additional stress originating from neglected or unresolved issues. Finally, perhaps this new “high-tech” device might demand too much of an investment (medical and personal) in a situation where bereavement has been completed and a “resurrection” is not sought.
Whatever the predisposition, pretext, or premorbid situation, this outcome as described is difficult to accept and provokes some ethical considerations regarding our interventions.
We usually regard refractory pain, severe uncontrollable side effects of our oral pharmacotherapy, and comprehension of the therapeutic plan by the patient as prime indications for the use of implantable devices. In the cases presented, it seems that the psychological distress following the procedures is linked to the fact that these patients suffered from uncontrolled pain and severe side effects for many months prior to the intervention. This fact is not of minor importance. Apparently unopposed pain not only imprints a “cellular” scar, but a behavioral and interrelational predisposition regarding attitude toward pain (Chapman & Garvin, 1999).
Thus, if timing is crucial, one must also ask when should we perform these procedures? Is the end of life “too late”? Is there a “too late”? Is a “better-late-than-never” attitude appropriate? These questions stem from an attempt to balance the obligation of non-maleficence, which asserts the obligation not to inflict harm on others (hence the famous primum non nocere maxim) with our obligation of beneficence, to take positive steps to help others. This distinction is important since it contains within it the concept of “proportionality,” by which the cost-benefit of our intervention is weighed. And if this cost-benefit is measured in terms of quality of life outcomes, one wonders if this was achieved in our cases. “one has an obligation to prevent serious evil or harm when one can do so without seriously interfering with one’s life plans or style” (Beauchamp & Childress, 2001).
Finally, what are the solutions ahead? Taking a strict “do no harm” approach to intervention at end of life risks denying patients the benefits of interventional pain therapies. An earlier intervention that prevents prolonged epochs of unopposed pain is, of course, our desire. Timely interventions at the end of life with appropriate psychological support might be the practical and reasonable paradigm.
The human experiences of pain and suffering are highly complex psychological and existential processes and are not merely two terms signifying the same concept. Ethics of pain management and medical science are intimately linked. Cognizance of these issues is only the beginning of a long process of corrective action and improved outcomes. We have presented these cases to stimulate this process.
Department Editor’s Note: These cases raise the paradoxical possibility that good pain control may increase, rather than decrease, the need for psychological support at the end of life. Why could this be? I don’t have the answer, but will leave you with two quotes. Karl Marx stated, “The only antidote to mental suffering is physical pain.” Some psychiatric patients, for example, find they can survive their mental torment only by cutting on themselves. Tolstoy describes this paradox in The Death of Ivan Ilyich: “He slept less and less; they gave him opium and began to inject morphine. But this did not relieve him. The dull pain he experienced in the half sleep condition at first only relieved him as a change, but then it became as bad, or even more agonizing, than the open pain.” Later, Ivan Ilyich says to himself on his deathbed, “And now it is all done and there is only death.”Alex Cahana is an assistant professor responsible for the Interventional Pain Program, Department of Anesthesiology, Pharmacology, Intensive Care, Geneva University Hospital, Geneva, Switzerland.
American Pain Society Quality of Care Committee. (1995). Quality improvement guidelines for the treatment of acute and cancer pain. Journal of the American Medical Association, 274(23), 1874–1880.
American Pain Society Task Force on Pain, Symptoms, and End of Life Care. (1997). Treatment of pain at the end of life. APS Bulletin, 7(1), 1–3.
Beauchamp, T.L., & Childress, J.F. (2001). Principles of biomedical ethics (5th ed.). New York: Oxford University Press.
Chapman, R.C., & Garvin, J. (1999). Suffering: The contributions of persistent pain. Lancet, 353, 2233–2237.
Patt, R.B., & Hassenbusch, S.J. (2001). Implantable technology for pain control. In S.D. Waldman (Ed.), Interventional pain management (2nd ed., pp. 654–670). Philadelphia, PA: WB Saunders.
Portenoy, R.K., & Hassenbusch, S.J. (2000). PolyAnalgesic Consensus Conference 2000. Journal of Pain and Symptom Management, 20, S1–S50.
Wallace, M., & Yaksh, T.L. (2000). Long term spinal analgesic delivery: A review of pre-clinical and clinical literature. Regional Anesthesia and Pain Medicine, 25(2), 117–157.
