Palliative sedation-still a complex clinical issue!

It is my anecdotal impression that, in reviewing the recent contents of palliative care journals, it is hard not to find an article on palliative sedation [...]

Establishing a multicentre clinical research network: lessons learned

BACKGROUND

Within many health care disciplines, research networks have emerged to connect researchers who are physically separated, to facilitate sharing of expertise and resources, and to exchange valuable skills. A multicentre research network committed to studying difficult cancer pain problems was launched in 2004 as part of a Canadian initiative to increase palliative and end-of-life care research capacity. Funding was received for 5 years to support network activities.

METHODS

Mid-way through the 5-year granting period, an external review panel provided a formal mid-grant evaluation. Concurrently, an internal evaluation of the network by survey of its members was conducted. Based on feedback from both evaluations and on a review of the literature, we identified several components believed to be relevant to the development of a successful clinical cancer research network.

RESULTS

THESE COMMON ELEMENTS OF SUCCESSFUL CLINICAL CANCER RESEARCH NETWORKS WERE IDENTIFIED: shared vision, formal governance policies and terms of reference, infrastructure support, regular and effective communication, an accountability framework, a succession planning strategy to address membership change over time, multiple strategies to engage network members, regular review of goals and timelines, and a balance between structure and creativity.

CONCLUSIONS

In establishing and conducting a multi-year, multicentre clinical cancer research network, network members were led to reflect on the factors that contributed most to the achievement of network goals. Several specific factors were identified that seemed to be highly relevant in promoting success. These observations are presented to foster further discussion on the successful design and operation of research networks.

Global warming in the palliative care research environment: adapting to change

Advocates of palliative care research have often described the cold and difficult environment that has constrained the development of research internationally. The development of palliative care research has been slow over the last few decades and has met with resistance and sometimes hostility to the idea of conducting research in 'vulnerable populations'. The seeds of advocacy for research can be found in palliative care literature from the 1980s and early 1990s. Although we have much to do, we need to recognize that palliative care research development has come a long way. Of particular note is the development of well-funded collaboratives that now exist in Europe, Canada, Australia and the USA. The European Association for Palliative Care and the International Association for Hospice and Palliative Care has recognized the need to develop and promote global research initiatives, with a special focus on developing countries. Time is needed to develop good research evidence and in a more complex healthcare environment takes increasingly more resources to be productive. The increased support (global warming) evident in the increased funding opportunities available to palliative care researchers in a number of countries brings both benefits and challenges. There is evidence that the advocacy of individuals such as Kathleen Foley, Neil MacDonald, Balfour Mount, Vittorio Ventafridda, Robert Twycross and Geoff Hanks is now providing fertile ground and a much friendlier environment for a new generation of interdisciplinary palliative care research. We have achieved many of the goals necessary to avoid failure of the 'palliative care experiment', and need to accept the challenge of our present climate and adapt and take advantage of the change.

Pre-admission escalation rate of daily opioid consumption (PERDOC), and total morphine equivalent daily dose on the first complete day of admission (D1-MEDD) to a tertiary-level palliative care unit (TPCU): Correlates and predictors in patients with advan

9065

Background: Although animal laboratory studies attest to opioid tolerance (OT), it can be difficult in clinical practice to determine whether substantive escalation in opioid dose (average of >5% per day between initial daily dose and maximum daily dose to date) is due to disease progression (DP) or decreased opioid responsiveness, which in turn may be due to OT or other factors. Our study aims were to determine (1) the frequency of PERDOC>5%, (2) the correlates and predictors of PERDOC>5% and D1-MEDD. Methods: We retrospectively examined TPCU patient database records for demographics, physician rating of PERDOC in the Edmonton Staging System (ESS) for cancer pain classification, Edmonton Symptom Assessment System (ESAS) scores, and D1-MEDD. Consecutive 1st admission data on patients surviving >3 days were included in the initial analysis. Using complete data, logistic regression and multiple regression models were created with PERDOC>5% and logn D1-MEDD, respectively, as dependent variables. Results: From 1,351 patients who met the initial descriptive analysis eligibility criteria, 1212 (90%) had an ESS rating for PERDOC. The prevalence of PERDOC>5% was 274/1212 (19.3%). Bivariate analysis (N=969, complete data) showed that PERDOC>5% was positively associated (p<0.05) with younger age, neuropathic pain component (NPC), a pathological level of psychological distress (PLPD), substance abuse, higher D1-MEDD, and higher ESAS pain score (ESAS-P). In the multivariate analysis, NPC (Odds ratio: 2.1, 95% confidence interval: 1.5–2.9), PLPD (1.7, 1.2–2.6), and higher D1-MEDD (2.2, 1.03–4.7) were the strongest independent positive predictors, and ESAS-P (1.01, 1.003–1.02) remained as a weaker predictor. In the D1-MEDD regression model, positive predictors (p<0.05) were younger age, NPC, incident pain, PERDOC>5%, PLPD, ESAS-P and ESAS anxiety scores. Conclusions: Aside from OT and DP, the multiple predictors identified for PERDOC>5% and D1-MEDD underscore the need for a systematic multidimensional assessment of cancer pain that incorporates psychological and physical characteristics.

No significant financial relationships to disclose.

Location of death in Canada. A comparison of 20th-century hospital and nonhospital locations of death and corresponding population trends

This report compares 20th-century Canadian hospital and nonhospital location-of-death trends and corresponding population mortality trends. One of the chief findings is a hospitalization-of-death trend, with deaths in hospital peaking in 1994 at 80.5% of all deaths. The rise in hospitalization was more pronounced in the years prior to the development of a national health care program (1966). Another key finding is a gradual reduction since 1994 in hospital deaths, with this reduction occurring across all sociodemographic variables. This suggests nonhospital care options are needed to support what may be an ongoing shift away from hospitalized death and dying.

Amputation and the prevention of phantom pain

Although it has been proposed that preoperative analgesia with epidural administration of analgesics may prevent long-term phantom pain, published results to date have been contradictory and controversial. In this case report, we describe a 41-year-old man with local recurrence of squamous cell carcinoma of the anus who underwent a hemipelvectomy. Preoperatively he had a significant neuropathic pain syndrome requiring oxycodone 60 mg every 4 hours. An epidural infusion of morphine and bupivacaine was started 24 hours preoperatively and discontinued on the third postoperative day. Over the next 10 days the oxycodone was gradually decreased and eventually discontinued prior to discharge. A review of the literature reveals conflicting reports on the benefit of preoperative epidural pain management in the prevention of postoperative pain syndromes. Conflicting research and conclusions of commentators leaves unanswered questions for clinicians. Nevertheless, we do know that we need to provide the best pain relief for patients both before and after amputation. This may require a combination of the oral, subcutaneous or intravenous, and epidural routes.

Sedation for delirium and other symptoms in terminally ill patients in Edmonton

The use of sedation and the management of delirium and other difficult symptoms in terminally ill patients in Edmonton has been reported previously. The focus of this study was to assess the prevalence in the Edmonton region of difficult symptoms requiring sedation at the end of life. Data were collected for 50 consecutive patients at each of (a) the tertiary palliative care unit, (b) the consulting palliative care program at the Royal Alexandra Hospital (acute care), and (c) three hospice inpatient units in the city. Patients on the tertiary palliative care unit were significantly younger. Assessments confirmed the more problematic physical and psychosocial issues of patients in the tertiary palliative care unit. These patients had more difficult pain syndromes and required significantly higher doses of daily opioids. Approximately 80% of patients in all three settings developed delirium prior to death. Pharmacological management of this problem was needed by 40% in the acute care setting, and by 80% in the tertiary palliative care unit. The patients sedated varied from 4% in the hospice setting to 10% in the tertiary palliative care unit. Of the 150 patients, nine were sedated for delirium, one for dyspnea. The prevalence of delirium and other symptoms requiring sedation in our area is relatively low compared to others reported in the literature. Demographic variability between the three Edmonton settings highlights the need for caution in comparing results of different palliative care groups. It is possible that some variability in the use of sedation internationally is due to cultural differences. The infrequent deliberate use of sedation in Edmonton suggests that improved management has resulted in fewer distressing symptoms at the end of life. This is of benefit to patients and to family members who are with them during this time.

A multicentre international study of sedation for uncontrolled symptoms in terminally ill patients

The issue of symptom management at the end of life and the need to use sedation has become a controversial topic. This debate has been intensified by the suggestion that sedation may correlate with 'slow euthanasia'. The need to have more facts and less anecdote was a motivating factor in this multicentre study. Four palliative care programmes in Israel, South Africa, and Spain agreed to participate. The target population was palliative care patients in an inpatient setting. Information was collected on demographics, major symptom distress, and intent and need to use sedatives in the last week of life. Further data on level of consciousness, adequacy of symptom control, and opioids and psychotropic agents used during the final week of life was recorded. As the final week of life can be difficult to predict, treating physicians were asked to complete the data at the time of death. The data available for analysis included 100 patients each from Israel and Madrid, 94 patients from Durban, and 93 patients from Cape Town. More than 90% of patients required medical management for pain, dyspnoea, delirium and/or nausea in the final week of life. The intent to sedate varied from 15% to 36%, with delirium being the most common problem requiring sedation. There were variations in the need to sedate patients for dyspnoea, and existential and family distress. Midazolam was the most common medication prescribed to achieve sedation. The diversity in symptom distress, intent to sedate and use of sedatives, provides further knowledge in characterizing and describing the use of deliberate pharmacological sedation for problematic symptoms at the end of life. The international nature of the patient population studied enhances our understanding of potential differences in definition of symptom issues, variation of clinical practice, and cultural and psychosocial influences.

Home versus hospice inpatient care: discharge characteristics of palliative care patients in an acute care hospital

This prospective survey was initiated to identify factors that helped and hindered home discharge for 100 consecutive patients who did not require further specialist palliative or acute care. Information was collected on demographics, functional ability (using the Palliative Performance Scale [PPS] and Karnofsky Performance Scale [KPS]), cognitive function at discharge as measured by the Mini-Mental State Examination (MMSE), home support circumstances, and patient and family preference for discharge. 59 patients were discharged home and 41 were transferred to a hospice. Younger patients with younger caregivers were discharged home more often. Patients with better MMSE and better functional ability (PPS and KPS) were also more likely to go home. Patients going home were more likely to be married. Preference for site of discharge was met for 76% of patients and 90% of families. Of the patients going to a hospice, 24% of patients and 7% of families preferred a home discharge. More physical support at home could have facilitated a home discharge for 13 patients. Functionally dependent and cognitively impaired patients were generally unable to return home. To support patients and their families in an environment of their choice, access to increased physical support in the home must be addressed.

Sedation for uncontrolled symptoms in a South African hospice

The need to sedate terminally ill patients for uncontrolled symptoms has been previously documented in a few reports. A retrospective consecutive chart review was undertaken at a hospice in Cape Town, South Africa, to develop an understanding of the local experience and assess the potential for improved patient management. Twenty-three of seventy-six (30%) patients received sedating therapies: twenty patients for delirium, two patients for delirium and dyspnea, and one patient for dyspnea alone. Fourteen patients were sedated with a continuous subcutaneous infusion of midazolam, seven patients with intermittent doses of benzodiazepines, and two patients with chlorpromazine and lorazepam. The mean midazolam dose was 29 mg per day (median 30 mg; range 15-60 mg per day). Patients were sedated on average 2.5 days before death (median 1 day; range 4 hours-12 days). The mean equivalent daily dose of parenteral morphine in the last week of life showed a significantly higher mean for the sedated group, as compared to the nonsedated group. There was minimal investigation of reversible causes for delirium, none of the patients underwent an opioid rotation, and the opioid dose was seldom decreased. None of the patients received parenteral hydration. The prevalence for the use of sedating treatment is consistent with the range of other literature reports. Nevertheless, the wide disparity in the reported prevalence of these problems, and the ethical concerns raised by the relative frequency of this sedative approach, cannot be ignored.

Innovative palliative care in Edmonton

PROBLEM BEING ADDRESSED

Access to palliative care in Edmonton has been hampered by uneven development, poor distribution of services, and more recently, economic restraints. Family physicians' involvement in palliative care has been hindered by the variety of access points, poor coordination, and inadequate reimbursement for time-consuming and difficult patient care situations.

OBJECTIVE OF PROGRAM

To provide high-quality palliative care throughout Edmonton in all settings, with patients able to move easily throughout the components of the program; to lower costs by having fewer palliative care patients die in acute care facilities; and to ensure that family physicians receive support to care for most patients at home or in palliative care units.

MAIN COMPONENTS OF PROGRAM

The program includes a regional office, home care, and consultant teams. A specialized 14-bed palliative care unit provides acute care. Family physicians are the primary caregivers in the 56 palliative continuing care unit beds.

CONCLUSIONS

This program appears to meet most of the need for palliative care in Edmonton. Family physicians, with support from consulting teams, have a central role. Evaluation is ongoing; an important issue is how best to support patients dying at home.

When to treat dehydration in a terminally ill patient?

The need to treat dehydration in terminally ill patients has become a very controversial topic. Numerous reports in the literature illustrate opposing view-points from both clinical and ethical perspectives. Arguments for the maintenance of hydration in terminally ill patients have tended to come from "the traditional medical model". Many health care professionals looking after terminally ill patients have reacted to the generalized use of intravenous fluids in dying patients and the perceived negative effects of this management. Our palliative care group has argued that the viewpoint that dehydration in dying patients is not a cause of symptom distress overlooks commonly reported problems, such as agitated delirium, that can be prevented or reversed by the management of dehydration. This review presents a summary of the traditional arguments, a different perspective on the controversy, biochemical parameters reported in terminally ill cancer patients, recent dehydration research, and the use of hypodermoclysis and rectal hydration. We conclude that the data reported to date are insufficient to allow a final conclusion on the benefit or harm of dehydration in terminally ill patients. Nevertheless, it is worth considering that while some dying patients may not suffer any ill effects from dehydration, there may be others who do manifest symptoms, such as confusion or opioid toxicity, that might be alleviated or prevented by parenteral hydration.

Rapid discontinuation of hypnotics in terminal cancer patients: a prospective study

PURPOSE

To determine the proportion of patients receiving hypnotics upon admission to a palliative care unit, the frequency and intensity of withdrawal symptoms after rapid hypnotic discontinuation, and the effect of discontinuation on insomnia and cognitive failure.

PATIENTS AND METHODS

120 consecutive admitted patients. Rapid hypnotic discontinuation (1-4 days) was attempted. Insomnia (visual analogue 0 = best, 100 = worst for insomnia, restedness during the morning and difficulty falling asleep), cognition (Mini-Mental State Questionnaire), and withdrawal signs were monitored.

RESULTS

Upon admission, 92/120 patients (77%) had been receiving hypnotics for a mean of 11 (standard deviation = 8) weeks. 4/92 patients (4%) refused hypnotic discontinuation. Acute mild withdrawal was observed in 2/88 patients (2%). The intensity of insomnia was not significantly different, while cognition significantly improved after hypnotic discontinuation.

CONCLUSION

A large proportion of terminal cancer patients receives hypnotic drugs chronically. These drugs are probably not useful for the treatment of their insomnia, and rapid discontinuation can be safely achieved in most patients.

Volume of hydration in terminal cancer patients

In this retrospective study we reviewed the volume and modality of hydration of consecutive series of terminal cancer patients in two different settings. In a palliative care unit 203/290 admitted patients received subcutaneous hydration for 12 +/- 8 days at a daily volume of 1015 +/- 135 ml/day. At the cancer center, 30 consecutive similar patients received intravenous hydration for 11.5 +/- 5 days (P > 0.2) but at a daily volume of 2080 +/- 720 ml/day (P < 0.001). None of the palliative care unit patients required discontinuation of hydration because of complications. Hypodermoclysis was administered mainly as a continuous infusion, an overnight infusion, or in one to three 1-h boluses in 62 (31%), 98 (48%) and 43 (21%) patients, respectively. Our findings suggest that, in some settings, patients may be receiving excessive volumes of hydration by less comfortable routes such as the intravenous route. Increased education and research in this area are badly needed.

Integrating medical and surgical treatments in gastrointestinal, genitourinary, and biliary obstruction in patients with cancer

Although local expertise needs to be considered, the following general guidelines for the palliation of malignant biliary obstruction have been proposed: (1) patients in good general condition with small tumors should undergo laparotomy to assess resectability and undergo surgical bypass; (2) patients with advanced disease and poor general condition are suitable for endoscopic stenting to reduce jaundice, pruritus, and risk of cholangitis; (3) patients with poor cognition and limited life expectancy should be managed pharmacologically as appropriate; and (4) research is required to assess optimal management for patients between these extremes; however, clinicians can use the previously described guidelines to make these difficult management decisions.

Decreased opioid doses used on a palliative care unit

We have previously published data on our use of opioids in the last week of life. A change in our pattern of opioid use, i.e. switching opioids more frequently and using high-dose methadone suppositories, appears to have resulted in a decrease in the number of patients requiring high-dose opioids. A retrospective chart review of 100 consecutive patients treated on our palliative care unit during 1992 was completed and compared to the original data from 1990. Results confirmed a decrease in the range of opioids used, as well as a statistically significant decrease in the daily opioid dose in the last week of life. We believe that this difference is most likely due to the use of methadone in patients showing either a poor response to other opioids or a rapid development to tolerance, as well as switching opioids more frequently to take advantage of incomplete cross-tolerance.

The frequency of alcoholism among patients with pain due to terminal cancer

The purpose of this retrospective study was to determine the prevalence of alcoholism among terminally ill cancer patients when assessed by multidisciplinary interviews and by the CAGE Questionnaire. We reviewed the charts of 100 consecutive patients assessed by a multidisciplinary team for the presence of alcoholism during 1989, and 100 consecutive patients assessed by the CAGE Questionnaire during 1992. Alcoholism was diagnosed in 28/100 patients during 1989 (28%) and 18/66 patients during 1992 (27%). Thirty-four patients were unable to complete the CAGE Questionnaire in 1992 because of sedation or cognitive impairment; six of these patients (17%) were found to be alcoholics after multidisciplinary assessment. Only 9/28 (32%) and 8/24 (33%) patients diagnosed as alcoholics during 1989 and 1992, respectively, had been previously diagnosed as alcoholics according to the medical charts. The mean equivalent daily dose of morphine during admission and on Day 2 during 1992 were 153 +/- 193 mg and 183 +/- 198 for alcoholic patients, versus 58 +/- 80 and 70 +/- 79 mg for nonalcoholics (P = 0.06 and 0.03, respectively). The maximal dose of opioid and the pain intensity during admission, however, were not significantly different between alcoholics and nonalcoholics. Our results suggest that alcoholism is highly prevalent and underdiagnosed among symptomatic terminally ill cancer patients. The CAGE Questionnaire should be used for screening for alcoholism in this population. When multidimensional assessment and management of pain is applied, the outcome of alcoholic patients appears to be similar to that of nonalcoholics.

Is this opioid analgesic tolerance?

Tolerance to the analgesic effect of opioids is a poorly understood phenomenon. While generally accepted to be an uncommon problem, it clearly can present major management difficulties in some patients. This case report illustrates different aspects of tolerance, describes a management approach using different opioids, and provides a focus for discussion of some current developments in understanding and managing this problem.

Comparison of two different concentrations of hyaluronidase in patients receiving one-hour infusions of hypodermoclysis

The purpose of this randomized, double-blind study was to compare 300 units of hyaluronidase per one-half liter to 150 units per one-half liter in patients receiving brief infusions for subcutaneous hydration. Twenty-five evaluable patients were randomized to receive a local injection of 300 units of hyaluronidase or 150 units of hyaluronidase immediately before two 1-hr infusions of two-thirds dextrose 5% and one-third normal saline solution (500 cc volume). The following day a crossover took place, and patients received the alternate treatment before each of the two 1-hr infusions. The intensity and swelling as reported by the patient (visual analogue scale 0-100), and the intensity of edema and rash as assessed by the investigator (score 0-4) were not significantly different between groups. The patients' and investigators' final choice was also not significantly different. Patients could not distinguish between bolus and their previous experience with overnight clysis. Our results suggest that brief infusions are well tolerated for subcutaneous hydration of patients with advanced cancer. A concentration of 150 units of hyaluronidase per one-half liter is well tolerated in this population.

Reversible delirium in terminally ill patients

Delirium is reported to be a common problem in terminally ill patients. The poor prognosis given to these patients may result in the failure to recognize the causes that are easily treated and may be reversible. We present four patients in whom a comprehensive assessment revealed a number of reversible causes of delirium, resulting in a treatment approach that significantly improved the patients' cognition and quality of life.

Palliative care in Edmonton

The Palliative Care Program at the Edmonton General Hospital, Edmonton, Canada provides a comprehensive multidisciplinary service to terminally ill patients in the area. The political, clinical, educational, and research developments impacting on this program are described.

The use of hypodermoclysis for rehydration in terminally ill cancer patients

The need to treat dehydration in terminally ill patients to minimize symptom distress remains a controversial issue. Hypodermoclysis (HDC) is a simple technique for rehydration that offers many advantages over the intravenous route. In this prospective open study of 100 consecutive patients who died on a palliative care unit, we recorded our indications for, and use of, HDC. Of the 100 patients, 69 received HDC for an average of 14 +/- 18 days during an average admission of 35 +/- 41 days. The 31 patients who did not receive HDC had an average admission of 22 +/- 24 days, and appeared to have different characteristics than the HDC group. HDC was well tolerated in most patients at an average volume of 1203 +/- 505 mL/day. These results confirm that HDC for dehydration is a safe and effective technique and suggest the need for further research to clarify the role of rehydration in assisting symptom control.

Symptom control in terminally ill patients with malignant bowel obstruction (MBO)

The inadequacy of prolonged conservative management with nasogastric suction and intravenous fluids for terminally ill patients with bowel obstruction has long been recognized. Using previous reports and our experience on the Palliative Care Unit at the Edmonton General Hospital, we have developed a basic approach to bowel obstruction management. In a review of 100 consecutive patients who died on our Palliative Care Unit, 15 required medical management for bowel obstruction. Evaluation of these cases suggests that intensive medical management can provide good symptom control without using intravenous lines and with minimal use of nasogastric tubes.

The use of urinary catheters in terminally ill cancer patients

Recent reports have commented on the need to improve our knowledge and management of urinary problems in terminally ill patients. We conducted a prospective study in 61 consecutive patients admitted to our palliative care unit (PCU), who were assessed for urinary problems, use of urinary catheters, and management of problems associated with the catheters. A total of 74% (45 of 61) of the patients required a catheter before death, with 23 (38%) being admitted to the PCU with a catheter, and 22 patients (36%) requiring a catheter during admission to PCU. We present our findings regarding the duration of catheter use, indications for catheters, type of catheters used, and complications of catheters. The data collected suggest that, although urinary problems and catheter use are common in terminal illness, if strict guidelines are followed there is no demonstrable mortality, and morbidity associated with catheter use is outweighed by patient benefit.

The assessment of pain intensity in patients with cognitive failure: a preliminary report

We reviewed our experience with 14 consecutive patients with cancer pain who developed severe cognitive failure that reverted either spontaneously or after specific treatment. In 3 patients who developed a nonagitated cognitive failure episode (CFE), there was no difference in the pain intensity measured by the patient before and after the episode and that measured by the nurse during the episode. In 11 patients who developed an agitated CFE, pain intensity assessed by a nurse during the CFE was significantly higher than the patient's assessment, both before and after the CFE. Patients who developed agitated CFE received a mean of 5 +/- 2 extra doses of narcotics per day, versus a mean of 2.17 +/- 1.6 doses in the average patient in our unit (P less than 0.01). Upon complete recovery, none of the 14 patients recalled having had any discomfort during the CFE. Problematic conflict between staff and family was detected in 4 of 11 cases of agitated CFE (36%), versus an expected 13 of 260 cases (5%, P less than 0.01). We conclude that (a) patients who recover from a severe CFE have no memory of pain; (b) medical and nursing staff are likely to overestimate the level of pain of patients with agitated CFE; and (c) agitated CFE in patients with cancer pain is a major source of distress for the patients' families and staff.