Surg Clin N Am 85 (2005) 303–313

Palliative Care in the Surgical ICU

Anne Charlotte Mosenthal, MD, FACSDivision Surgical Critical Care, New Jersey Medical School,

University of Medicine & Dentistry of New Jersey-University Hospital,

150 Bergen Street, Mezzanine 233, Newark, NJ 07103, USA

Mr. Jones is in the surgical ICU (SICU) on a respirator. This is his thirdadmission to the SICU for ventilator support. Three weeks ago he hada pancreatoduodenectomy for pancreatic cancer and his initial recovery wasuneventful; however, on postoperative Day 5 he developed pneumonia,rapid atrial fibrillation, and pancreatic ascites. Despite antibiotics, thisprogressed to respiratory failure and he was placed on the ventilator. Heimproved and 1 week later was extubated and transferred to the hospitalfloor, only to relapse within 24 hours with respiratory failure, sepsis, andrenal failure. He was readmitted to the SICU for ventilator support,continuous renal replacement therapy, and antibiotics. He improves over 3weeks time and is now dialysis-free with adequate renal function; however,ventilator weaning is slow with multiple exacerbations of hypoxia and fever.His family states that because of his advanced age and multiple underlyinghealth problems, the patient would not want to be maintained on ‘‘thesemachines’’ for the remainder of his life. The patient had communicated thesewishes to his physicians before surgery. The nurses suggest that ongoing lifesupport is prolonging the patient’s suffering; however, the surgeon notesthat the prognosis for cancer survival is excellent, and hopes that with time,better nutrition, and slow ventilator weaning, the patient’s multiple organfailure will improve and he will survive to be discharged.

This patient’s lingering course exemplifies the dilemmas that areencountered increasingly in the SICU when prognosis is unclear andmultiple organ failure waxes and wanes. Is Mr. Jones likely to survive todischarge with an acceptable quality of life? Or will he die in the ICU whilereceiving mechanical ventilation? If he is likely to die, when should caretransition to a palliative approach, if at all? As critical care becomesincreasingly sophisticated, these sorts of decisions have become part ofstandard practice for physicians in the ICU.

E-mail address: mosentac@umdnj.edu

0039-6109/05/$ - see front matter � 2005 Elsevier Inc. All rights reserved.

doi:10.1016/j.suc.2005.01.001 surgical.theclinics.com

304 MOSENTHAL

The ICU traditionally has focused on prolonging survival and curingdisease with advanced technology; 10% to 20% of patients in the ICU willdie before discharge from the hospital [1]. Among ICU survivors, mortalityremains high at 6 months, particularly among the elderly [2]. Moreover, oneof five American deaths occurs in the hospital during or after an ICU stay [3].The landmark Study to Understand Prognoses and Preferences forOutcomes and Risks of Treatment (SUPPORT) study demonstrated thata large proportion of dying patients in ICUs received unwanted life supportand therapies with inadequate attention to pain and suffering in the last 3days of life [4,5]. This occurs despite the fact that in the last decade, medicalpractice found that most deaths in the ICU are associated with withholdingor withdrawal of life support [6,7] These trends suggest that expertise inpalliative care is an essential part of the armamentarium of the critical carephysician, to ensure pain and symptommanagement in the ICU and to applythe appropriate principles of withholding or withdrawing of life support.

Although it is increasingly clear that the principles of palliative care mustbe integrated into the ICU to ensure high-quality end-of-life care, how bestto do this is not. The particular nature of critically ill surgical patients,critical care surgeons, and the culture of the ICU brings forth specialdifficulties in integrating palliative care into critical care. First, although10% to 20% of patients who are admitted to ICU will die during theirhospitalization, it is difficult to know with certainty which ones will do soand when. Just as it is unclear whether Mr. Jones in the aforementioned casestudy is dying or merely following the natural history of multiple organfailure, it is difficult to predict the actual likelihood of mortality for a specificpatient. Many predictive models that have been developed for the ICU (eg,Acute Physiology and Chronic Health Evaluation, Multiple OrganDysfunction Syndrome, Trauma Injury and Severity Score), although usefulfor stratifying populations, have not proven to be helpful for clinicians whoare faced with making decisions about individual patients. The literaturesuggests that clinicians rarely use prognostic or predictive scores or modelsfor clinical decision making; even when they are informed of the likelihoodof death for a particular patient in the ICU based on such models, they donot make a transition to a palliative approach [5,8]. The culture of the ICUtraditionally views mortality and survival as dichotomous variables [9];often death must be imminent before clinicians will change from a curativetreatment plan to a palliative one. For example, physicians decide towithdraw mechanical ventilation when they judge patients to have poorlong-term cognitive function or less than a 10% likelihood of survival [10].The wide range of practices regarding end-of-life care and withdrawal andwithholding support in ICUs illustrate how prognostic uncertainty andvaried definitions of medical futility affect palliative care in the ICU setting.

Most patients who are admitted to the ICU do not have clear advancedirectives and are unable to participate in medical decisions; this compoundsprognostic uncertainty. Only 5% of patients in the ICU can report their

305PALLIATIVE CARE IN THE SURGICAL ICU

end-of-life preferences, their symptoms, or participate in treatment decisions[6]. Much of the decision-making relies upon family members as surrogates.Often, they do not know their loved ones’ preferences, are bereavedthemselves, and are ill-prepared to participate in serious end-of-lifedecisions. Furthermore, neither the family nor the patient knows theattending surgeon or intensivist well.

Lastly, almost all patients who enter the SICU, and the physicians whocare for them, arrive with goals for curative intent and assumptions that allthat technology has to offer will bring about a positive outcome. Whencoupled with the prognostic uncertainty that was described above, itbecomes difficult to know prospectively for each patient when or if goals ofcare should become palliative. This challenge to palliative care in the ICUsetting can be met in several ways. First, quality care in the ICU mustinclude palliative principles, such as pain and symptom relief and goodcommunication for all patients, not only those who are imminently dying orat the end of life. Data suggest that attention to these areas early in ICU careimproves the overall quality of critical care, in general, and facilitatesspecific decision-making around end-of-life care [11,12]. Second, a large partof palliative care in the ICU requires particular expertise in the transition ingoals of care and the ability to accomplish curative and palliative goals, inparallel for the same patient. This transition, in the setting of acute life-threatening illness, defines the inherent difficulties and differences inpalliative care in the critical care setting. It often must be done acutely, ina compressed time frame, with incomplete information. To accomplish thistransition, physicians must have skill in communication for shared decision-making and identifying goals of care, knowledge of the principles ofwithdrawal and withholding of life support, and familiarity with grief andbereavement support.

The following sections review the four essential areas of palliative care:communication, withholding and withdrawal of organ systems support,pain and nonpain symptom management, and bereavement support as theyapply to the SICU setting.

Communication

Effective communication is the mainstay of quality palliative care in theSICU. The integration of palliative care into the intensive care setting relieson early, effective communication between physicians, nurses, the patient,and family; if done well, this will facilitate later difficult transitions in goalsof care and end–of-life decision-making. Several studies showed thatfamilies of patients in the ICU value communication, particularly withphysicians, as highly as medical care and will rate overall quality of criticalcare based on communication [12,13]. Communication between health careproviders also is important for maintenance of the interdisciplinary teamthat is required for good palliative care.

306 MOSENTHAL

Although communication long has been regarded as a nice, butnonessential, part of the practice of medicine, particularly surgery, therecognition of competencies that are expected of professionalism validatedcommunication as an essential skill of the surgeon. More importantly, thereis increasing evidence in the literature that communication interventions inthe ICU have a significant impact on quality of care, length of stay, andpatient and family satisfaction. Lilly et al [11,14] found that a familymeeting within 72 hours of admission to the ICU decreased the length ofstay. This was particularly effective if discussion specifically addressed thepatient’s goals of care and time-limited therapies, based on the clinicalresponse of the patient. A randomized, prospective trial that used a differentmodel of ethics consultation on nonbeneficial life support found this to bea useful way to resolve conflicts and avoid the prolonged use of futile lifesupport. Length of stay in the ICU and prolonged mechanical ventilationwere decreased after the implementation of ethics consultation [15].

Why these interventions work is not entirely clear, although improvedresolution of conflicts is a likely benefit. Conflict around end-of-life care isfrequent and arises from different understandings of the prognosis, lack ofunderstanding of the patient’s preferences, family grief, and poorcommunication between all parties. Readiness to accept death as a likelyoutcome for a patient or willingness to entertain withdrawal of life supportfollows a different time course for each individual, whether he receives orprovides treatment. Nurses may precede physicians in this process [13].Qualitative data from the SUPPORT study suggest that families mustaccept death of the patient before they can be ready to accept palliativegoals of care [16]. Early communication around admission to the ICU,including support for families, may facilitate end–of-life care and decision-making around withdrawal and withholding of life support [17,18].

Despite the demonstrated importance of communication for palliativecare in the ICU setting, there is little evidence about the qualitative aspectsof communication and how they affect end-of-life care in the ICU. There isa large body of literature on physician–patient communication, but itsapplicability to the critical care setting is unclear. The compressed time-frame of surgical critical illness, uncertain prognosis, incapacity of thepatient, and reliance on bereaved surrogates significantly alter the nature ofthe communication. Even with this realization, little is known about whichinterventions might improve quality of communication. Studies thatexamined the audiotapes of family conferences in the ICU founda correlation between the amount of time that physicians spend listeningand the family’s satisfaction with the communication [19]. In the surgicalsetting, even less has been studied about surgeon’s communications skillsand attitudes around end of life. One study suggests that the critical carenurse plays an important role as mediator and communicator betweensurgeon and family in the SICU [20]. Lessons also can be drawn fromthe SUPPORT study; nurse-facilitated communication intervention with

307PALLIATIVE CARE IN THE SURGICAL ICU

physicians and patients did not change end-of-life care significantly withrespect to do not resuscitate orders or withdrawal of life support. Nursespecialists providing routinely updated information to physicians onprognosis and patient preferences was not enough to change practice [5].Whether this reflected deficiencies in the quality of communication or someother aspect of care remains open to speculation. Despite these inconsistentdata, recommendations exist for appropriate communication while pro-viding end-of-life care in the critical care setting. Curtis [12] suggests thefollowing six steps for a family meeting, for a patient who is in the ICUpatient, based on a modification from earlier principles adapted to thecritical care setting. These include: (1) adequate preparation for thediscussion, including proper setting, selection of appropriate members ofthe team, and timing; (2) assessment of family understanding of thesituation; (3) discussion of prognosis; (4) clarification of possible outcomesand goals of care; (5) listen and provide empathetic support; and (6) providea follow-up plan and recommendations for treatment. Further study isneeded to identify specific elements of communication in the ICU,particularly in the surgical setting that may improve outcome.

Withdrawal and withholding of life support

The option of withdrawal and withholding of life support, based oninformed patient consent, in the ICU is considered to be standard practice;it probably occurs in this setting more than in other health care settings.Prendergast et al [21] found that 48% of patients who die in the ICU havelife support withdrawn or withheld, whereas a larger percentage havecardiopulmonary resuscitation (CPR) withheld [21]. Expertise in thedecision making about withholding and withdrawing treatments and itsclinical application is an essential component of palliative care in the ICU.The decision to withdraw organ systems support must be based on patientpreferences and goals, knowledge of the likely outcome of continuingsystems support versus withdrawing it, and the risks and benefits ofprolonging organ system supportive care. Shared decision-making is the firstgoal in this process, and is based on communication between physicians,nurses and the patient-family. Decision making can be complicated by theterm ‘‘life support,’’ which can set unrealistic expectations about what thetreatment ultimately will deliver. Some individuals are afraid or reluctant towithdraw life support—even when they are fully aware that ongoingtreatment will not achieve survival or quality of life—because stopping ‘‘lifesupport’’ makes them seem or feel as if they are facilitating death orabandoning the patient. Confusion about the ethical and legal basis forwithdrawal and withholding of life support persists, particularly in mattersof withdrawal of mechanical ventilation. This is reflected in the many studiesthat describe the wide variability in physician practice about these matters.When presented with patient vignettes and several choices for withdrawal or

308 MOSENTHAL

withholding of CPR, dialysis, or ventilatory support, physicians chosedifferent strategies [22,23]. Studies showed that physicians prefer towithdraw treatments in a particular order, often withdrawing vasopressoragents first and the ventilator last, if at all [24], or they will withholda therapy, but not withdraw it once instituted [25,26]. Ethical and legalprecedents are clear that withholding and withdrawing therapy are ethicallyequivalent and that the patient’s autonomy and right to refuse therapyprevail, even if withdrawal may lead to death [27].

After a decision is made to withdraw ventilatory support, theinterdisciplinary team and the family must be prepared. Families shouldbe informed of symptoms that can occur following extubation (dyspnea,anxiety, noisy secretions) and reassured that the means to control symptomsrapidly and aggressively will be at hand. It is critical at this point that thepatient, family, and ICU staff understand that the intent of the medication isto control symptoms. Although these medications may induce deep sedationor even contribute to the patient’s demise, the principle of ‘‘double effect’’makes this ethically permissible. Equally important as discussing thesymptoms that may occur following ventilator withdrawal, is cautioningthe family that demise may not occur immediately following ventilatorwithdrawal. Patients who have had mechanical ventilation withdrawn havebeen known to survive hours, days, and even longer. Families often want toknow the time from ventilator withdrawal to death; this is highly variableand dependent on clinical situations. Generally, health care providers arepoor at predicting this, but families should be given a range of times and beprepared for immediate demise if this is a possibility.

It should be determined if the family wishes to be present duringventilator withdrawal or extubation and what provisions should be madefor their support during this time. The patient should be premedicated andsedated before removal of the ventilator and extubation. Opioids are themedication of choice for treatment of dyspnea; benzodiazepines areprescribed for anxiety that inevitably accompanies dyspnea. The ventilatorcan be withdrawn by two basic methods: direct extubation or ‘‘terminalweaning.’’ There is little evidence in the literature to support one techniqueover another. Some investigators describe a combination of both, witha rapid wean to minimal ventilator settings to allow for appropriate controlof symptoms, followed by disconnecting the ventilator, then extubation [28].Selection of a method should be based on the patient’s goals, discomfort,and ability to communicate. If the patient hopes to talk and interact with hisfamily with minimal interference, then extubation is necessary. If the patientis unconscious or airway obstruction threatens the patient’s and the family’scomfort, ventilator weaning without removal of the endotracheal ortracheostomy tube is preferred. A physician needs to be immediatelyavailable with additional opioids and benzodiazepines to assess ongoingsymptoms and treat them accordingly. It is critical that symptoms bedocumented, and assessments of pain or discomfort are noted clearly when

309PALLIATIVE CARE IN THE SURGICAL ICU

opioids are increased. Concerns on the part of nurses and physicians abouthastening death with the administration of opioids for dyspnea have notbeen supported by the literature. Studies that examined the time to deathafter withdrawal of the ventilator found no difference between those whowere medicated for dyspnea and those who were not [29].

Pain and symptom management

Although the traditional imperative of ICU care is to cure, there is anequal obligation to relieve suffering. This is true for all patients, not merelythose who are dying. Therapy to resuscitate and support life need notpreclude attention to relief of suffering, nor should clinicians be deterredfrom treating pain aggressively because of fears of destabilizing vital systemfunction. To the contrary, the literature suggests that the physiologicresponse to unrelieved pain and stress is more deleterious to the critically illpatient from tachycardia and agitation that result in increased myocardialoxygen demand [30,31]. Others studies showed that there are long-termadverse consequences of inadequate pain relief and sedation in criticallyinjured burn patients following prolonged ICU stays. These include a greaterincidence of acute stress disorders and posttraumatic stress syndrome [32].

Multiple studies have documented the deficiencies in pain and symptomrelief in ICUs. In the SUPPORT Study, more than 60% of families of dying,critically ill patients reported that their loved one had multiple severesymptoms (including pain) in the last 3 days of life [5]. Nelson et al [33]prospectively studied critically ill patients who had cancer who could reporttheir own symptoms. A significant proportion of them experienced pain,anxiety, dyspnea, thirst, and hunger. Severe pain and discomfort have beenreported by patients during routine nursing and medical procedures in theICU, such as suctioning and turning [34,35]. Although the literature reportsthe barriers and deficiencies in pain and symptom management, there arefew evidence-based studies on successful pain management strategies forcritically ill patients.

The inability of many patients to self-report their symptoms is a barrierto satisfactory pain management in the ICU. Accurate assessment of pain bynursing staff can be extremely difficult; surveys of patients for retrospectivepain reports are fraught with methodologic difficulties. Proxy symptomreports have been studied and validated in patients who had cancer, but ingeneral, health care providers underestimate pain when compared with thepatient reports; family proxies do not do much better [36,37]. Reliance ontachycardia, dyspnea, and hypertension as proxies for pain in unconsciouspatients is not sufficient because there are too many variables that affectthese vital signs. Several new assessment tools have been developed,specifically for the noncommunicative patient in the ICU. These rely on vitalsigns and on nonverbal cues, such as grimacing, splinting, restlessness, andstiffness, with points awarded for these behaviors during rest, turning, or

310 MOSENTHAL

movement [38,39]. With the advent of assessment tools, all paininterventions should be based on repetitive assessments and reassessmentsfor response to therapy. The use of pain medications, solely as needed, is nolonger standard of care; parameters that are based on numerical scalesshould govern titration of medication. The use of protocols for painmanagement and sedation were shown to decrease ventilator days, decreasethe length of stay in the ICU, and diminish the frequency of a pain scorethat is greater than 2 among patients in the ICU [40,41].

Opioids are the mainstay of pain management in the ICU; continuousintravenous (IV) administration is the preferred approach. Dose titrationprotocols with guidelines for rescue doses should be implemented. Forpatients who are dying imminently or who require aggressive palliative care,non-IV routes—with the exception of intramuscular injection—can beconsidered as a way of minimizing the problems that are associated with IVdelivery. Attention should be paid to decreasing as many routine criticalcare nursing procedures (eg, turning, suctioning, intravenous catheters,blood draws, frequent vital signs) as possible, to minimize the symptomsfrom these seemingly innocuous activities. Increasing data reveal thatsignificant symptoms are related to these procedures [33], particularlyamong the dying.

Bereavement and family support

Excellent palliative care in the ICU focuses on the patient and family asthe unit of care. It is documented increasingly that families of dying,critically ill patients have extensive psychosocial needs and increased rates ofmental and physical illness. More often than not, the patient is incapacitatedcognitively and the family—while in a crisis situation and bereavedthemselves—must be surrogates for decision making. Good communicationis valued by families for emotional support as much as it is valued forinformation and planning. This may prevent later conflicts and facilitateend-of-life decision-making. The positive impact of hospital-based bereave-ment services, pastoral care, or family support personnel is becomingincreasingly apparent on the long-term psychosocial functioning ofsurviving families and on other outcomes, such as organ donation [42,43].They are a vital part of comprehensive palliative care in the ICU setting.Multiple studies showed that the way in which bad news is delivered mayhave life-long effects on the survivors, grief and bereavement process [44,45].For all of these reasons, intensive care surgeons and other health careproviders should be trained in communication skills, specifically forbreaking bad news.

It also is clear that the opportunity to say ‘‘good-bye’’ to loved ones andthe degree of satisfaction of how this was accomplished are importantcorrelates with posttraumatic stress disorder during bereavement. Familiesshould have as much opportunity to be with loved ones at the bedside as

311PALLIATIVE CARE IN THE SURGICAL ICU

possible. The opportunity for a family’s presence at the bedside of thepatient in the ICU is a critical part of successful palliative care. Unrestrictedvisiting hours in ICUs is becoming the norm; this sociocultural change hasgone a long way toward improving the quality of death and dying in ICUs.Studies showed that the presence of family does not affect care adversely,and probably improves it [46]. The presence of family during CPR is morecontroversial, but data increasingly suggest that this is helpful duringbereavement [47]. If families are present, a bereavement counselor, socialworker, pastoral care or other specially-trained personnel should be presentto support the family, explain procedures, and communicate last wishes andpreferences to the health care staff. Despite the high technologyenvironment of the ICU, attention can be paid to the spiritual care anddeath rituals, as appropriate. Families usually need support in after-deathprocedures; many have little knowledge of how to navigate the socialpathways (eg, funeral homes) following death in the hospital. Social workeror spiritual care worker assistance in these matters should be part of theinterdisciplinary care that is provided in the ICU.

Summary

Palliative care for the critically ill has become an increasingly importantcomponent of care in the SICU. As the population ages, medical technologycontinues to offer new treatments that can prolong life, and more and moreAmericans die in the hospital in critical care settings, the appropriatemanagement of the end-of-life must be part of the clinical expertise ofsurgeons and intensivists. Part of this expertise must include the componentsof palliative care (eg, pain and symptom management, psychosocialsupport, communication skills, shared decision-making) and specializedareas of withdrawal and withholding of life support. Integrating palliativecare expertise into the SICU is not straightforward; understanding whenand how to make the transition from curative to palliative care can befraught with uncertainty regarding prognosis and patient preferences.Attention to the principles of good pain management, communication withpatient and family, and discussion of goals of care are not just for patientswho are at the end-of-life, but are appropriate care for all critically illpatients, regardless of prognosis. In this framework, ‘‘intensive care’’encompasses palliative and curative care.

References

[1] KnausWA, Draper EA,Wagner DP, et al. An evaluation of outcome from intensive care in

major medical centers. Ann Intern Med 1986;104:410–8.

[2] Chelluri L, Pinsky MR, Donahoe MP, et al. Longterm outcome of critically ill elderly

patients requiring intensive care. JAMA 1993;269:3119–23.

312 MOSENTHAL

[3] Angus DC, Barnato AE, Linde-Zwirble WT, et al. The use of intensive care at the end of life

in the United States: an epidemiologic study. Crit Care Med 2004;32:638–43.

[4] Lynn J, Teno JM, Phillips RS, et al. Perceptions by family members of the dying experience

of older and seriously ill patients. SUPPORT Investigators. Study to Understand Prognoses

and Preferences for Outcomes and Risks of Treatments. Ann Intern Med 1997;126:97–106.

[5] The SUPPORT Investigators. A controlled trial to improve care for the seriously ill and

hospitalized patients. JAMA 1995;224:1591–8.

[6] Prendergast T, Luce J. Increasing incidence of withholding and withdrawal of life support

from the critically ill. Am J Respir Crit Care Med 1997;155:15–20.

[7] Faber-LangendoenK. Amulti-institutional study of care given to patients dying in hospital:

ethical and practice implications. Arch Intern Med 1996;156:2130–6.

[8] Knaus WA, Rauss A, Alperovitch A, et al. Do objective estimates of chances for survival

influence decision to withhold or withdraw treatment? The French Multicentric Group of

ICU Research. Med Decis Making 1990;10:163–71.

[9] Barnato AE, Angus DC. Value and role of intensive care unit outcome prediction models in

end-of-life decision making. Crit Care Clin 2004;20:345–62.

[10] Cook D, Rocker G, Marshall J, et al. Withdrawal of mechanical ventilation in anticipation

of death in the intensive care unit. N Engl J Med 2003;349:1123–32.

[11] Lilly CM, De Meo DL, Sonna LA, et al. An intensive communication intervention for the

critically ill. Am J Med 2000;109:469–75.

[12] Curtis JR. Communicating about end of life care with patients and families in the intensive

care unit. Crit Care Clin 2004;20:363–80.

[13] Hickey M. What are the needs of families of critically ill patients? A review of the literature

since 1976. Heart Lung 1990;19:401–15.

[14] Lilly CM, Sonna LA, Haley KJ, et al. Intensive communication: four-year follow-up from

a clinical practice study. Crit Care Med 2003;31:S394–9.

[15] SchneidermanLJ,GilmerT, TeetzelHD, et al. Effect of ethics consultations on nonbeneficial

life-sustaining treatments in the intensive care setting: a randomized controlled trial. JAMA

2003;290:1166–72.

[16] Murphy PA, Price DM, StevensM, et al. Under the radar: Ccontributions of the SUPPORT

nurses. Nurs Outlook 2001;49:238–42.

[17] Way J, Back AL, Curtis JR. Withdrawing life support and resolution of conflict with

families. BMJ 2002;325:1342–5.

[18] TildenVP,Tolle SW,Garland JM, et al.Decisions about life-sustaining treatment. Impact of

physicians’ behaviors on the family. Arch Intern Med 1995;155:633–8.

[19] MCDonagh JR, Elliott TB, Engelberg RA, et al. Studying communication about end-of-life-

care during the ICU family conference: development of a framework. J Crit Care 2002;17:

147–60.

[20] Cassell J, Buchman TG, Street S, et al. Surgeons, intensivists and the covenant of care:

administrative models and values affecting care at the end of life. Crit Care Med 2003;31:

1551–7.

[21] Prendergast T, J Claessens MT, Luce JM. A national survey of end-of-life care for critically

ill patients. Am J Respir Crit Care Med 1998;158:1163–7.

[22] Cook DJ, Guyatt GH, Jaeschke R, et al. Determinants in Canadian health care workers of

the decision to withdraw life support from the critically ill. JAMA 1995;273:703–8.

[23] Hinshaw DB, Pawlik T, Mosenthal AC, et al. When do we stop, and how do we do it?

Medical futility and withdrawal of care. J Am Coll Surg 2003;196:621–49.

[24] Solomon MZ, O’Donnell L, Jenning B, et al. Decisions near the end of life: professional

views on life sustaining treatments. Am J Public Health 1993;83(1):14–23.

[25] Asch DA, Christakis NA. Why do some physicians prefer to withdraw some forms of life

support over others? Med Care 1996;34:103–11.

[26] Christakis NA, Asch DA. Biases in how physicians choose to withdraw life support. Lancet

1993;342:642–6.

313PALLIATIVE CARE IN THE SURGICAL ICU

[27] Presidents Commission for the Study of Ethical Problems in Medicine and Biomedical and

Behavioral Research. Deciding to forego life-sustaining treatment: a report on the ethical,

medical and legal issues in treatment decisions. Washington, DC: US Government Printing

Office; 1983.

[28] Rubenfeld GD. Principles and practice of withdrawing life-sustaining treatments. Crit Care

Clin 2004;20:435–52.

[29] Wilson WC, Smedira NG, Fink C, et al. Ordering and administration of sedatives and

analgesics during the withholding and withdrawal of life support from critically ill patients.

JAMA 1992;267:949–53.

[30] Lewis AM, Whipple JK, Michael KA, et al. The effect of analgesic treatment on the

physiological consequences of acute pain. Am J Hosp Pharm 1994;51:1539–54.

[31] Epstein J, Breslow MJ. The stress response of critical illness. Crit Care Clin 1999;15:17–33.

[32] Saxe G, Stoddard F, Courney D, et al. Relationship between acute morphine and the course

of PTSD in children with burns. J Am Acad Child Adolesc Psychiatry 2001;40:915–21.

[33] Nelson J, Meier D, Oei E, et al. Self-reported symptom experience of critically ill cancer

patients receiving intensive care. Crit Care Med 2001;29:277–82.

[34] Puntillo KA. Dimensions of procedural pain and its analgesic management in critically ill

surgical patients. Am J Crit Care 1994;3:116–22.

[35] Turner JS, Briggs SJ, Springhorn HE, et al. Patients recollection of intensive care unit

experience. Crit Care Med 1990;18:966–8.

[36] Choiniere M, Melzack R, Girard N, et al. Comparisons between patients’ and nurses’

assessment of pain and medication efficacy in severe burn injuries. Pain 1990;40:143–52.

[37] Desbiens NA, Mueller-Rizner N. How well do surrogates assess the pain of seriously ill

patients? Crit Care Med 2000;28:1347–52.

[38] Puntillo KA, Miaskowski C, Kehrle K, et al. Relationship between behavioral and

physiological indicators of pain, critical care patients’ self-reports of pain and opioid

administration. Crit Care Med 1997;25:1159–66.

[39] Puntillo KA, Stannard D, Miaskowski C, et al. Use of a pain assessment and intervention

notation (PAIN) tool in critical care nursing practice: nurses evaluations. Heart Lung 2002;

31:303–14.

[40] Kress J, Pohlman AS, O’Connor MF, et al. Daily interruption of sedative infusions in

critically ill patients undergoing mechanical ventilation. New Engl J Med 2000;342:1471–7.

[41] Caswell DR, Williams JP, Vallejo M. Improving pain management in critical care. J Qual

Improvement 1996;22:702–12.

[42] IversonK.Grave words: notifying survivors about sudden unexpected deaths. Tucson (AZ):

Galen Press; 1999.

[43] Oliver RC, Sturtevant J, Scheetz J, et al. Beneficial effects of a hospital bereavement

intervention program after traumatic childhood death. J Trauma 2001;50:440–8.

[44] Linyear AS, Tartaglia A. Family communication coordination: a program to increase organ

donation. J Transplant Coord 1999;9:165–74.

[45] Jurkevich GJ, Pierce B, Pananen L, et al. Giving bad news: the family perspective. J Trauma

2000;48:865–73.

[46] Simpson T, Wilson T, Mucken N, et al. Implementation and evaluation of a liberalized

visiting policy. Am J Crit Care 1996;5:420–6.

[47] Meyers TA, Eickhard DJ, Guzzzetta C, et al. Family presence during intensive procedures

and resuscitation: the experience of familymembers, nurses and physicians. Am JNurs 2000;

100:32–42.