APOS Clinical E-Mail Update #12
13 October 2004

In this Update:

How does a psychological intervention affect cancer patients and health behaviors?

Andersen BL, Farrar WB, Golden-Kreutz DM, Glaser R, Emery CF, Crespin TR, Shapiro CL, Carson WE.  Psychological, behavioral and immune changes after a psychological intervention: a clinical trial.  J Clin Oncol 2004;22:3570-3580.

The goal in cancer medicine is to improve disease management, longevity, and quality of longevity in cancer patients.  This study of 227 women surgically treated for regional breast cancer looks at psychological state, health behavior, and biological effects after random assignment to a psychological intervention, a small group treatment with one session per week for 4 months.  The study is set in the context of 8 prior randomized studies of psychological intervention in cancer patients; four of the eight improved survival.  In general, such interventions reduce psychological distress particularly in the most distressed.  In this study patients who received the intervention had significantly reduced anxiety, improved perception of social support, better diet, and reduced smoking.

Besides behavioral and psychological differences, the study found a difference in immune parameters.  The markers of T-cell proliferation in response to both phytohemagglutinin and concanavalin A remained stable or increased for the intervention patients but declined for patients who did not receive the intervention.

In the intervention, two clinical psychologists met in groups of 8 to 12 patients weekly for 1.5 hours for 18 sessions.  Patients who were absent were telephoned by a therapist to provide support and to discuss the session topic.  Session 1-18 focused on understanding stress responses and progressive muscle relaxation training.  Other sessions concerned relaxation training, positive coping, problem solving and the ability to identify a social network and communication skills.  A food intake diary and education, stretching and walking protocol 20 minutes per day, three times per week, were also added.  Patients who smoked were referred to other services and given group support for cessation.  Disease and treatment information, assertive communication skills, monitoring of treatment and follow-up appointments, and goal setting were added to improve adherence.

When the patients were assessed after the intervention, total mood disturbance and anxiety (on the Profile of Mood States) were reduced in those who had the intervention.  Fatigue also improved significantly with the intervention.  This intervention tried to improve the patient’s social adjustment in the real world with the hope that better social adjustment leads to better quality of life and better treatment adherence.  The inclusion of health behaviors as intervention targets adds an important dimension to the psychological intervention literature in cancer patients and may elucidate keys to better survival.

[back to top]

What is the protocol when an error is suspected in an intrathecal pump?

Coyne PJ, Hansen LA, Laird J, Buster P, Smith TJ.  Massive hydromorphone dose delivered subcutaneously instead of intrathecally: Guidelines for prevention and management of opioid, local anestetic and clonidine overdose.  J Pain Symp Manage 2004;28:273-6.

Occasionally in advanced cancer patients drugs are delivered intrathecally by catheter or pump.  Usually the medication is an opioid analgesic, but bupivicaine and clonidine are also used as adjunctive anesthetics.  This paper discusses the complication of a massive hydromorphone dose delivered subcutaneously in a patient with sickle cell anemia.  The needle had become dislodged when the patient moved and mistakenly delivered the drug, a two-month supply, into a subcutaneous pocket.  Within minutes, the patient was sleepy and nauseated.  The problem was recognized, and naloxone was started as ICU admission was considered.  The outcome of this case was good, but it led to a discussion of nursing principles for intrathecal pumps, actions to take when an inaccurate intrathecal pump medication delivery is suspected.  Parallel suggestions are made for a protocol for inaccurate refill of local anesthetic or clonidine.  These may be useful to the varied staff, not as familiar with pump technology, who may be covering for cancer patients.  Besides recognition, alert of trained personnel, monitor of vital signs, and preparation for code, these principles include querying the pump, turning it off, and aspirating potential opioid if possible.

[back to top]

Can patients “give up” or “hold on” in the face of death for an important occasion?

Skala J, Freedland KE.  Death takes a raincheck.  Psychosomatic Med 2004;66:382-386.

These authors review 18 studies since the early 70s that tried to find out if patients can give up or hold on in relation to symbolically important occasions such as holidays or birthdays.  They did not find evidence of a psycho-physiological mechanism that enables people to postpone or hasten their own death.

[back to top]

What are the oncologists’ feelings about a request for physician-assisted suicide?

Meyskens FL.  Fetting JH.  Pulling the trigger.  J Clin Oncol. 2004; 22:3644-3645.

The physician’s countertransference to rare patient requests for euthanasia is the topic of this poem by Dr. Meyskens discussed by Dr. Fetting.  Fetting sets out the common taboo feelings: not looking forward to the patient visit, not wanting to call back the patient, feelings of failure, feeling defensive and angry.  These responses make a distance between patient and doctor just when the patient who makes this request especially seeks attention and help.  He emphasizes that a request for help requires hope, and any help of any kind may break the cycle of helplessness.  The specific help in this setting is expert knowledge on relief of suffering and a larger and more comprehensive care team.  He pleads for the hopeful aspects of bringing hospice and palliative care to the bedside.

[back to top]

What do we understand about CNS metastases in breast cancer?

Lin NU, Bellon JR, Winer EP.  CNS metastases in breast cancer.  J Clin Oncol 2004;22:3608-3617.

These authors review the recognition and management of central nervous system metastases in breast cancer, a more common problem now that treatment for the underlying disease has improved.  They make the point that breast cancer is now the second most common cause of CNS metastases.  The frequency of CNS metastases is estimated at 10-16% in Stage IV cancer patients of the 1960s and 70s, but is present in 30% at autopsy.  Metastases follow a vascular distribution by hematogenous spread, tend to be more commonly supratentorial, to have a predilection for vascular border zones and gray- and white-matter junctions.  Young age is a risk factor as is HER2 overexpression.  Trastuzumab (Herceptin) does not cross the blood-brain barrier, so the incidence of CNS metastases may increase as systemic control improves.

Headache is the presenting symptoms in 24-48% cases, but mental status changes and cognitive disturbances occur in 24-34% of patients.  Leptomeningeal metastases also present with non-localizing headache or pain and with deficits in cranial nerves 3, 4, 6, 7, and 8.  The diagnostic evaluation requires gadolinium-enhanced magnetic resonance imaging, which has an advantage over computerized tomography, particularly in the posterior fossa and brainstem.  The MRI is also estimated to be 76% sensitive for leptomeningeal spread.  The role of flurorodeoxyglucose positron emission tomography remains undefined.  Historically the one-year survival of patients with brain metastases of breast cancer has been 20%.  Outcome favors younger patients with normal Karnofsky performance scores and control of the primary tumor.  Stereotactic radiosurgery has advanced as an alternative to neurosurgery in the last decade.  The authors go on to review treatment options in all modalities.

[back to top]

Does the economic burden of cancer persist even among survivors?

Yabroff KR, Lawrence WF, Clauser S, Davis WW, Brown ML.  Burden of illness in cancer survivors: findings from a population-based national sample.  J National Cancer Institute 2004;96:1322-30.

From the 2000 National Health Interview Survey, an annual multi-stage survey of civilian non-institutionalized people in the US, Yabroff et al. found that 1823 cancer survivors had a greater burden of illness even after 11 years as long-term survivors.  Compared to matched controls, they had lower scores on a summary measure of health, lost more days from work, and were more likely to report their health as fair or poor.  The study describes the intangible burden of illness even in survivors beyond 5 years that can add to mortality costs, lost productivity among caregivers, and direct costs of cancer care to delineate the overall cancer burden.  These results were consistent across tumor types.

However, the survivors of lung and other cancers with short survival reported more burden than survivors of breast, colorectal, prostate, and all other cancers.

The researchers identified 2151 men and women who reported that they had been diagnosed with cancer and used the remaining 30 195 as potential controls.  The controls were matched on age, sex, educational attainment.  Utility was measured by the Helath Activities and Limitations Index (HALex).  Lost productivity due to morbidity was measured by whether the individual had a job in the past year, limitations in work ability due to health problems, limitations in amount or kind of work because of health problems, and number of days lost from work in the past year in those with jobs.

[back to top]

Are psychiatric patients more at risk for cancer?

Carney CP, Woolson RF, Jones L, Noyes R, Doebbeling BN.  Occurrence of cancer among people with mental health claims in an insured population.  Psychosomatic Medicine 2004;66:735-743.

In this retrospective cohort study of administrative claims data in Iowa Blue Cross/Blue Shield between 1989 and 1993, insured people who had claims for mental disorder had a greater risk of certain cancers and developed malignancies at younger ages.  The odds ratio for cancer in women who had mental health claims was 1.023 and for men with mental health claims 1.10.  Patients with a mental disorder had a greater likelihood of a central nervous system tumor or respiratory tumor, and there was a trend toward a greater odds ratio for oral cavity tumors in men.  These data would fit with an association between mental symptoms as the presenting symptoms of a brain tumor.  They would also fit with the increased likelihood of smoking in patients with mental disorder and the role of smoking as a carcinogen.  In the past, it has been thought that schizophrenia might be protective against cancer, and depression has been thought to increase the likelihood of cancer.

The study population of 722,139 included men and women 18 to 64, 95% white, 55% urban.  The mental health cohort was two-thirds women compared to half the controls.  Men were followed an average of 20 months and women 23 months from their first mental health claim.  For those with mental health disorders who later developed a malignancy, mood disorders accounted for 32% in men and 43% in women.  The rate of mood disorders was no different in men with or without cancer.  Among women, mood disorder was slightly more common in those who developed cancer.

The median length of time between psychiatric presentation and first brain tumor claim was 1.5 years, implying that the brain tumor itself may be causing the first psychiatric presentation.  The higher rate of lung and head and neck tumors suggests a relationship to smoking that may not have been apparent in previous studies because of methodological difficulties.

[back to top]

How reliable is the physician report of adverse symptoms by chemotherapy toxicity criteria?

Fromme EK, Eilers KM, Mori M, Hsieh Y-C, Beer TM.  How accurate is clinician reporting of chemotherapy adverse effects? A comparison with patient-reported symptoms from the Quality-of-Life Questionnaire C30.  J Clin Oncol 2004;22:3485-3490.

In a phase II trial of 37 men with metastatic, androgen-indepenent prostate cancer treated with weekly calitriol and docetaxel, oncologists completed an Adverse Event log using the Common Toxicity Criteria of the National Cancer Institute.  The physicians had regular clinical responsibility for the patients as they became symptomatic, and the medical record was checked to make sure that all recorded adverse events were reported.  Patients completed the Quality of Life Questionnaire of the European Organization for the Research and Treatment of Cancer (QLQ 30) before treatment and every 4 weeks. 

This paper reports a comparison of the adverse event symptoms found by the physicians performing the clinical trial and those symptoms that patients reported on the questionnaire.  If the QLQ 30 were considered the standard, physician sensitivity and specificity were only 47% and 68% respectively.  The physicians did not report about half of the symptoms reported in the QLQ as adverse events, and the QLQ did not detect approximately half of the symptom adverse events reported by the physician.  The eight symptoms were fatigue, pain, dyspnea, sleep disturbance, appetite disturbance, nausea, diarrhea, and constipation.  The QLQ30 has not been used to identify adverse events in the past, but this study highlights the variability in adverse event reports and the difference between the reports of patient and physician.

[back to top]

How should clinicians respond when the patient refuses assessment of capacity?

Hurst SA.  When patients refuse assessment of decision-making capacity.  Arch Intern Med. 2004; 164:1757-1760

If the patient will not state why he is refusing beneficial treatment, the assessment of decision-making capacity is difficult.  If the patient is incompetent, then the decision is referred to a proxy who would stand for their prior stated wishes or their best interest in combination.  Refusal to give reasons does not indicate incompetence.  This paper discusses whether there is a duty of conversation and how it would apply in such a case, how requests by the patient to be left alone might be approached, and how the physician chooses risks in uncertainty.  The clinician does his best to obtain the patient’s trust, engages in a dialogue or gets others to talk with the patient.  The gravity of the risk affects whether the patient is subsequently treated as if incompetent.  The rationale is then explained to the patient.  Again, the physician can seek a conversation with the patient.


Donna B. Greenberg, MD, Associate Professor of Psychiatry at Harvard Medical School and Psychiatric Consultant in the Massachusetts General Hospital Cancer Center, Dana Farber Partners Cancer Care