Surg Clin N Am 86 (2006) 1065–1092

Palliating Patients Who Have

Unresectable Colorectal Cancer:
Creating the Right Framework
and Salient Symptom Management
Geoffrey P. Dunn, MD
Department of Surgery and Palliative Care Consultation Service, Hamot Medical Center,
2050 South shore Drive, Erie, PA 16505, USA

Over 56,000 individuals succumb to colorectal cancer each year in the

United States [1], and many more thousands carry psychological, social,
and spiritual burdens associated with all stages of this illness. Despite the
availability of effective therapies for cancer pain, recognition of the impor-
tance of a patient-centered approach, and increasingly effective therapies for
advanced disease, care for individuals who have advanced colorectal cancer
has been acknowledged as a difficult problem in several reviews by colorec-
tal surgeons [2,3]. Numerous reports have documented that medical care of
those who have serious and advanced illness suffers from poor symptom
control, conflicts in decision-making, lack of direction, and erosion of the
patient family’s physical, mental, and economic health [4–8]. The discrep-
ancy between what is considered ideal care and what is often done suggests
a much more profound barrier to successful palliation than simple lack of
knowledge, technology, or interest on the part of surgeons. Nevertheless,
there has been growing recognition by major surgical organizations of the
importance of good palliative care [9–11], not only to patients and their fam-
ilies, but to surgical staff participating in their care. The key elements of
palliative caredeffective communication, diligent symptom control, and
continuity of caredare well within the reach of practitioners and medical
institutions, given a clear vision of the framework and context of palliative
care. This article discusses the framework of palliative care for patients who
have advanced colorectal malignancy, as well as specific interventions for
selected symptoms and problems associated with it.

E-mail address: [email protected]

0039-6109/06/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.suc.2006.05.008 surgical.theclinics.com
1066 DUNN

Palliation has been defined as making less serious, to make less severe
without curing, to reduce the pain or intensity of, to mitigate, and alleviate.
It comes from the Latin palliare, ‘‘to cloak or cover.’’ Unfortunately, the ad-
jective form, palliative, has been used in contrasting ways in surgical litera-
ture, reinforcing the deep-seated ambivalence many surgeons have towards
palliative situations [12]. It has been used to describe: (1) surgery to relieve
symptoms, knowing beforehand that all of the tumor could not be removed;
(2) surgery in which microscopic or gross disease was knowingly or unknow-
ingly left behind; and (3) surgery for recurrent or persistent disease after
primary treatment failure (ie, ‘‘salvage surgery’’). In addition to these dis-
crepancies, the term is usually used as a description of the surgeon’s intent,
not a measurable result from the patient’s perspective. The surgical litera-
ture, similar to the literature on palliative chemotherapy [13], offers scant
guidance about the effectiveness of palliative interventions because of the
lack of prospective, randomized controlled studies and the tendency to mea-
sure only survival and morbidity outcomes. Unlike survival, the only one
who can truly answer if palliation has been achieved is the patient.
In a symposium on surgical palliative care [14], surgical oncologist Blake
Cady succinctly outlined the rules of engagement for major palliative resec-
tions. The principles and considerations he discussed can be just as usefully
applied to all palliative interventions, including nonsurgical therapies. He
prefaced these by noting that one of the major reasons for continued fol-
low-up in clinic and office after a initial curative treatment is for the contin-
uation of emotional and psychological support of the patients. He
discouraged ‘‘technical’’ follow-updthe routine post-treatment procession
of tumor markers, blood studies, radiographs, and scans that contribute
nothing to the long-term success of patients after initial curative multidisci-
plinary measures. While cautioning against overlooking opportunities
for cure (ie, resection of a solitary liver metastasis from a colorectal cancer
with favorable biological characteristics), he noted that the decision to
intervene for cure recognizes that recurrent disease is a pattern- and not
a time-dependent phenomenon. Recognizing this, ongoing searches for
asymptomatic metastases are not useful. If the patient is asymptomatic, pal-
liation is not needed. Pre-emptive palliation assumes suffering is as predict-
ably preventable as disease. This may be true in some cases, but it will
require a methodology of proof that has not yet been developed.
For surgeons seeking perspective when making management decisions for
palliative care, Cady commented, ‘‘It is easier to make day-to-day surgical
decisions in the framework of some overall surgical philosophy. Certainly
if you have a mature surgical philosophy that understands the vicissitudes
of life, you’re better equipped to deal with some of these situations than
a surgeon who thinks that ‘I can cure all problems’’’ [14]. To proceed
with palliative interventions, the surgeon must understand three key ele-
ments: (1) the psychology of the particular patient, (2) the biology of the dis-
ease, and (3) the effectiveness of the particular operation. Each of these
 PALLIATING UNRESECTABLE COLORECTAL CANCER 1067

elements is considered through weighing alternative examples. A summary

of his approach to decision making with some expansion is tabulated in
Table 1.
Cady completed his advice on cautioning not to allow the patient to think
we [surgeons] are magicians. Good guidance for palliation will always in-
clude some type of nonsurgical therapy. The goals of palliation must be
clearly and honestly explained to your patient, your team, and especially
your residents and referring physicians.

Palliative care needs assessment

The immediate goal of patient assessment in end-of-life care is to permit
the relief of suffering, though an additional goal is the identification of sour-
ces of strength that may be helpful for the patient and family. Trauma care
has shown how well surgeons can make assessments of complicated situa-
tions in an emotionally charged atmosphere. Making a palliative assessment
can foster enough trust that the surgeon may be asked by the previously hes-
itant patient for information about prognosis and what symptoms to expect
as a result of progressing disease. A useful palliative assessment must reflect
more than the physical problems, even if they dominate the immediate pic-
ture. Each surgeon has his own comfort level for the degree to which he ex-
plores the nonphysical concerns, but with experience and reinforcement, the

Table 1
Decision making for palliative interventions
Element of decision making Considerations
Psyche of the patient Active versus passive
Mature versus panicked
Realistic versus magical thinking
Physiology of the patient Functional versus debilitated
Actual age versus chronologic age
Preserved nutrition versus hypoalbuminemic
Biology of the disease Single disease versus multiple comorbidities
Slow-growing versus aggressive
Nature of the intervention Evidence-based versus ‘‘N of one’’
(personal anecdotal experience)
Routine versus uncommon
Low versus high morbidity risk (physical, financial, social)
Straight-forward versus difficult procedure
One of many options versus no other option
Profile of the surgeon Empathic healer versus technician
Experienced versus inexperienced
Cautious versus ‘‘cowboy’’
Realistic versus magical thinking or projecting
Collegial versus isolated and dominating
Modified from Cady B, Easson A, Aboulafia A, et al. Part 1: Surgical palliation of advanced
illnessdwhat’s new, what’s helpful. J Am Coll Surg 2005;200(1):115–27.
1068 DUNN

surgeon will experience an increased sense of satisfaction with his increasing

depth of participation in discussion about the other dimensions.
Assessment for palliative care can be seen as a staging procedure for the
dimensions of distress, analogous to the approach used to stage the extent of
disease. Along with the stepwise procedure for giving bad news, it should be
considered part of the surgeon’s cognitive operative repertoire. Following
the general guidelines for communication outlined by Buckman in his land-
mark book, How to Break Bad News [15], a thorough assessment for palli-
ative care can be done in the amount of time it takes to do a secondary
survey using advanced trauma life support (ATLS) protocol.
Staging for palliative care not only provides a reliable patient-centered
platform for further interventions, but also creates a unique bond between
the practitioner making the assessment and the individual assessed, not un-
like that which forms between a surgeon and an individual she has operated
on. Because of the inherent vulnerability of a patient revealing his innermost
fears and yearnings while in a physically precarious condition, this interven-
tion is as invasive as any physical operation, and deserves the same attention
and skill. The procedure of exploring and staging an individual’s perception
of life is based on many small, careful steps and a few big steps, any one of
which could be irreversible.

Delivery models of palliative care

There are two formal delivery models of palliative care in the US medical
system that surgeons should be aware of in their capacities of consultant and
primary physician. The first of these is hospice, which has been recognized
and compensated through Medicare legislation since 1983; the other model
is palliative care support (eg, consult services, clinics, and designated in-hos-
pital units), which can be seen as an application of the hospice model to
a wider spectrum of illness without the restraint of prognostic consider-
ations. Both delivery models share the following features: (1) patient/family
is the unit of care, (2) relief of distress and promotion of quality of life is the
clinical goal, and (3) care is provided using an interdisciplinary approach. A
hospice program provides palliative care for terminally ill patients and their
families. All hospice programs are palliative care programs, but palliative
care is not limited to terminally ill patients. Palliative care consultation in
hospitals and in outpatient settings has became increasingly available since
the later 1990s because of increasing public interest in improved end-of-life
care and the favorable impact on patient care made by hospice programs in
this country since its introduction to this country in the later 1970s. Pallia-
tive care can be given concurrently with all other appropriate medical care.
Palliative care programs have proliferated rapidly in the United States
since 2000 [16], numbering about 1,027 (in 25% of hospitals) in 2003.
Some of these programs are directed by and actively supported by surgeons.
Palliative care programs were more likely to be offered in larger hospitals,
 PALLIATING UNRESECTABLE COLORECTAL CANCER 1069

teaching hospitals, veterans’ hospitals, and hospitals designated as Ameri-

can College of Surgeons’ Commission on Cancer approved cancer hospitals.
Multiple reviews [17–21] have documented improved pain and non-pain
symptom management, improved patient and family satisfaction, reduced
critical care unit and hospital stay, decreased in-hospital deaths, and signif-
icant cost reductions. A high rate of implementation of recommendations
from palliative care teams has been noted [22]. Currently, fewer than 2000
physicians are certified in palliative care in the United Statesdfar fewer
than necessary to staff existing and future palliative care and hospice pro-
grams. Participation by surgeons as medical consultants, volunteers, and re-
ferral sources would greatly help in bridging this gap.
Criteria for palliative care referral include: (1) patient/family request, (2)
‘‘triggered’’ referrals (ie, protocols that initiate a referral of any patient in an
intensive care setting over a defined period of time, or any patient who has
a Karnofsky status of less than 50), or (3) by clinician judgment for a given
patient. Palliative care consultation referral criteria are listed in Box 1. Some
hospitals have designated units for palliative care; others use a consultative
model, often in conjunction with pain management services, with or without
a designated unit.
For any individual whose treatment is palliative and no longer life-pro-
longing, hospice referral from the hospital or outpatient setting should be
considered if there is a physician-certified prognosis of less than 6 months

Box 1. Indications for palliative care consultation

Patient has an illness typified by progressive deterioration and
worsening symptoms, often ending fatally.

Patient has limiting/threatening conditions with declining
functional status, mental or cognitive function.

Suboptimal control of pain or other distressing symptoms

Patient/family would benefit from clarification of goals and
plan of care, or resolution of ethical dilemmas.

Patient/surrogate declines further invasive or curative
procedures, preferring comfort-oriented symptom
management only.

Patients on medical/surgical or critical care units who are
expected to die imminently or shortly following hospital
discharge

Bereavement support of hospital workers, particularly after the
death of a colleague

Courtesy of Robert A. Milch, MD, Buffalo, NY

1070 DUNN

if the disease pursues its usual course. Eligibility requirements for Medicare
hospice benefit are listed in Box 2. Any terminal diagnosis is appropriate as
long as its prognosis of 6 months or less is confirmed by two physicians. A
do not (attempt to) resuscitate order (DNR or DNAR) is not required for
enrollment into a hospice program, nor is it allowed to be under Medicare
regulations. The presence of a DNR order has been mistakenly used as
a marker for preferences about end-of-life care [23]. As a practical matter,
most patients or surrogates who have had appropriate counseling prior to
election of hospice care are prepared to discuss and authorize a DNR order.
Another inaccurate belief about hospice and palliative care is that it em-
phasizes withdrawal of burdensome care and usage of sedatives and opioids.
Theoretically, no treatments are automatically excluded from consideration
by hospices as long as the criterion for a treatment rests on its proven ability
to achieve symptom control, regardless of its impact on the underlying dis-
ease process. Depending on the resources and size of the hospice program
(Medicare reimbursement is only $125 per day per patient under its capi-
tated system [24]), services vary. Hospice benefit team members and services
are listed in Box 3. Benefits may include the options of intravenous (IV) and
total parenteral nutrition (TPN) therapy, chemotherapeutic agents, and
transfusion. Despite the effectiveness of some surgical procedures such as
gastrointestinal (GI) tract stenting, video-assisted thoracoscopy (VATS),
and open fixation of pathologic fractures, they are not usually considered
for patients enrolled in hospice. Possible reasons for this include: (1) lateness
of hospice referrals, with survival of only a few weeks to days, despite the
hospice admission criterion that allows patients who have up to 6 months
life expectancy; (2) cost of the interventions that is prohibitive given the lim-
ited resources of many hospice programs; and (3) hospice professionals’ lack
of familiarity with palliative surgical techniques.
Although a consulting surgeon may not be in a position to make a direct
referral to palliative care or hospice services, the surgeon’s interest and sup-
port for these interventions on behalf of a patient, when indicated, will be
deeply appreciated by the patient and his family. Recommendations made

Box 2. Eligibility for Medicare hospice benefit

Patient is eligible for Medicare Part A.

Two physicians must certify that patient has a condition whose
prognosis is associated with a survival of 6 months or less if
the illness pursues its natural course.

Patient (surrogate if patient not competent) must sign form
electing hospice benefit.

Hospice care must be provided by a Medicare-certified hospice
program.
 PALLIATING UNRESECTABLE COLORECTAL CANCER 1071

Box 3. Medicare hospice benefit team and services

Core team members

Physician

Registered nurse

Social worker

Pastoral (spiritual) or other counselor

Minimum percentage of volunteer hours required
Services

Pain and non-pain symptom management

Assistance of patient and family with the emotional, social, and
spiritual aspects of dying

Provides medications for control of symptoms, medical
supplies, and durable medical equipment.

Education of family on care of the patient

Specialized therapy services (speech, massage, and physical
therapy)

Short-term inpatient care (respite care) when pain or
symptoms are unmanageable at home, or when caregiver
needs respite

Bereavement care for surviving family and friends for 1 year
from the date of the death

for these interventions are quite likely to be heeded if the surgeon’s relation-
ship with them has been positive.

Symptoms management of advanced colorectal cancer

Symptom burden is high with advanced cancer, and an increasing num-
ber of symptoms correlates with higher degrees of psychologic distress and
poorer quality of life [25]. Symptoms are classified as cancer-related, cancer
treatment-related, and non-cancer–related. As a general rule, physical symp-
toms that are immediately distracting are addressed before nonphysical
symptoms, which may ultimately be more significant for the individual if
not mitigated. Intervention is not deferred until definitive diagnosis and
treatment plan is made, but should begin promptly while the evaluation is
initiated, especially when symptoms are severe. Symptoms from comorbid
illness can trump cancer-related symptoms and should be triaged accord-
ingly (ie, severe dyspnea from advanced chronic obstructive pulmonary
disease [COPD] triages to a higher priority of treatment than mild cancer-
related pain).
Most of the symptoms of advanced colorectal cancer are encountered in
non-GI cancers, with the important exceptions of GI tract obstruction and
fistula. Symptom management is the work of palliative care, but not its goal.
1072 DUNN

Although new assessment scales for specific symptoms and quality of life
are continuously being introduced and revised, validated and culturally spe-
cific instruments should be used whenever possible, especially in the setting
of clinical research. Well-known and validated assessment scales include the
Functional Assessment of CancerdGeneral Version (FACT G) [26] and the
Edmonton Symptom Assessment System (ESAS) [27].
The Karnofsky scale, well known to oncologists, assesses physical func-
tion, not degree of distress or success of palliation. Numerous studies
have correlated low Karnofsky scores with shortened survival in advanced
cancer. It has been used, sometimes naı̈vely, as an index of quality of life be-
cause of this association; however, loss of physical function does not neces-
sarily preclude quality of life, and preservation of physical function does not
guarantee it.

Pain
The incidence of pain in advanced colorectal cancer is not known, though
up to 90% of individuals in some categories of advanced cancer report pain
[28], often severe [29], and often undertreated [30,31]. Barriers to effective
pain management have been extensively discussed. Fears shared by all par-
ticipants in the patient encounter have created many of these. Some of these
fears are based on lack of information or exaggeration of known risks,
whereas others are based on the psychology of life-limiting disease. Fear
of addiction, fear of respiratory suppression, and fear of scrutiny by regula-
tory agencies have been consistently identified as barriers. More subtle bar-
riers are rooted in the association of opioids with imminent demise, because
of medical practice in the past (‘‘They got the morphine out when he only
had a few days left.’’) and the reluctance of a patient (or caregiver) to ac-
knowledge progressing disease as the reason for escalation of dose. The dif-
ferential diagnosis for ‘‘refractory’’ cancer pain includes denial or avoidance
of the recognition of progressing disease by either patient or care giver.
Pain is classified into acute or chronic, nociceptive (somatic and visceral)
or neuropathic. Some pain syndromes are a mix of each of these groupings.
The classification has therapeutic implications because of the differing un-
derlying physiologic mechanisms of each class. The potential causes of
pain in patients who have advanced colorectal cancer include pain from
cancer, pain from cancer-related treatments, and pain unrelated to cancer.
Mechanisms of pain include tumor infiltration, including direct infiltration
of nerves; malignant bowel obstruction (MBO); inflammatory changes; che-
motherapy or surgically induced neuropathy; hepatic capsular distention;
and nonmalignant causes.
The clinical expression and the psychology of chronic pain differs from
acute pain. The visible signs of increased sympathetic dischargedgrimacing,
pallor, tachycardia, and high blood pressuredobservable and measurable in
acute pain, are muted or absent in chronic pain, giving the misleading
 PALLIATING UNRESECTABLE COLORECTAL CANCER 1073

impression that ‘‘He does not look like he is in pain.’’ The psychological and
existential impact of acute pain that has evolved to chronic pain can be
summarized as the distress of having an unpleasant sensory-emotional expe-
rience that has outlived its purposefulness. For some unfortunate individ-
uals, chronic pain becomes a metaphor for life itself, which may in some
way be linked to their increased risk of depression and suicide.
Improved cancer pain management is an institutional responsibility as
well as the surgeon’s individual moral and professional responsibility. In
the summary of the American Pain Society Recommendations for Improv-
ing the Quality of Acute and Cancer Pain Management [32], the authors,
based on a systematic review of quality improvement in pain management
from 1994 to 1994, recommend that ‘‘all care settings should formulate
a structured, multilevel systems approach (sensitive to the type of pain
and setting of care) that focuses on five primary areas: (1) prompt recogni-
tion and treatment of pain, (2) involvement of patients in the pain manage-
ment plan, (3) improved treatment patterns, (4) regular reassessment and
adjustment of the pain management plan as needed, and (5) measurement
of processes and outcomes of pain management.’’
Physicians believe that their failure to manage pain effectively is related to
their inability to assess it properly [29]. Twycross [33] proposed a mnemonic,
modified by Ray [34]: PQRSTU (think of an EKG of a hypokalemic pa-
tient!). Table 2 outlines this mnemonic for initial and subsequent pain as-
sessment. Pain assessment should include some acknowledgement of the
three nonphysical types of pain: (1) psychological, (2) social (economic),
and (3) spiritual. The sum of physical and nonphysical pain has been fre-
quently referred to in the palliative care literature as ‘‘total pain’’ [35].
The numerous mechanisms of cancer-related pain provide numerous tar-
gets for pharmacotherapy and other types of intervention for the relief of
pain. Depending on the nature of the symptoms and the tolerances and
wishes of the patient, any single agent or combination of agents can be
used for the management of pain, in addition to achieving other goals of
therapy such as prolonged survival (chemotherapy), improved appetite
(steroids), or control of fever (nonsteroidals). Cytotoxic chemotherapy,
nonsteroidal anti-inflammatory agents (NSAIDs), steroidal agents, antide-
pressants, and opioid analgesics, have been used singly or in combination
for the relief of cancer pain or cancer treatment-related pain.
All of these modalities have the potential for major side effects, including
death, which have to be anticipated in this vulnerable patient population
that frequently has major comorbidities. Therapeutic nihilism, often di-
rected towards the ‘‘terminal’’ patient in the hospital, nursing home, and
hospice setting, should not discourage the surgeon from pursuing uncom-
monly used pain management strategies known to be highly effective for cer-
tain pain syndromes when simpler approaches are inadequate (eg, using
axial analgesia via an epidural catheter instead of high-dose systemic opioids
with debilitating side-effects). By contrast, caution must be exercised when
1074 DUNN

Table 2
PQRSTU mnemonic for assessment of pain
Provoked or What makes pain better? Worse? Effect of position, climate, company,
palliated by prayer, meditation, procedures
Quality of pain Have patient describe quality of pain in own words:

Nocioceptive
Somatic (skin, muscle, bone): described as achy, stabbing, throbbing.
Visceral (bowel, bladder): described as crampy, gnawing, achy, sharp

Neuropathic (nerve damage)
Burning, tingling, sharp, shooting, electric shock-like
Quality of pain can give clue to etiology and guide selection of
analgesic:

Nociceptive pain generally responds to opioids and anti-inflammatory
agents.

Neuropathic pain is less responsive to opioids and may require
addition of an adjuvant agent (tricyclic antidepressant or
anticonvulsant).
Region and referral Guides imaging and regional therapies
Severity Visual analogue scale (VAS): 0–10.
Patient is asked to rank pain: 0¼no pain, 10¼worst pain imaginable

Mild pain¼1–3

Moderate pain¼4–6

Severe pain¼7–10, pain emergency
Severity ranking can be used to guide pace of titration; eg, if pain is
‘‘severe,’’ opioid dose can be increased up to 50%–100% over a
24-hour period.
Temporal Need to know for identifying end-of-dose failure of analgesic
Utilized medications What has worked or not worked and why. Opportunity to clarify
confusion of ‘‘allergy’’ with drug side-effects
Adapted from Ray JB. Pharmacologic management of pain. Surg Onc Clin N Am
2001;10(1):73–4.

prescribing the familiar: NSAIDs such as ibuprofen, available over the

counter, can lead to renal compromise, GI hemorrhage, and platelet
dysfunction.
NSAIDS are effective in the management of cancer pain [36]. They
are included in the first step of the World Health Organization (WHO)
step-ladder for cancer pain control [37], outlined in Box 4. They can be given
concurrently with opioids, but should not be given to patients receiving cor-
ticosteroids or anticoagulants, or to thrombocytopenic patients. NSAIDs
should be avoided in patients who have renal impairment, congestive heart
failure, and liver disease. Extreme caution is also advised in the treatment of
the elderly. The risk of renal impairment from NSAIDs is increased in pa-
tients of advanced age, those with pre-existing renal impairment, heart fail-
ure, hepatic dysfunction, or hypovolemia; those with concomitant therapy
with other nephrotoxic agents (including diuretics); and those with elevated
angiotensin II or catecholamine levels [38]. Toxicity is the most significant
factor limiting long-term use [39]. Major complications are not necessarily
preceded by minor ones.
 PALLIATING UNRESECTABLE COLORECTAL CANCER 1075

Box 4. World Health Organization’s analgesic ladder: ‘‘By mouth,

by the clock, by the ladder, with attention to detail’’
Step 1: mild pain, visual analogue scale (VAS) 1–3

Acetaminophen, aspirin, NSAIDs

May add a nonopioid analgesic or an adjuvant agent. Adjuvant
or coanalgesic agents are drugs that enhance analgesic
efficacy of opioids, treat concurrent symptoms that exacerbate
pain, or provide independent analgesia for specific types of
pain [36] (eg, a tricyclic antidepressant for treatment of
neuropathic pain).
Step 2: moderate pain, VAS 4–6

Acetaminophen or aspirin compounded with hydrocodone,
codeine, oxycodone, dihydrocodeine

NSAIDs as used in Step 1

May add a nonopioid analgesic or an adjuvant agent
Step 3: Severe pain, VAS 7–10, pain emergency

Morphine: ‘‘gold standard’’; most flexibility in dosing forms

Hydromorphone: useful in renal failure patients and for
subcutaneous infusions

Oxycodone: no IV form; growing stigma due to illicit use

Fentanyl: extremely fast-acting, transcutaneous dosing
available

Methadone: cheap and effective; not a first-line agent

Levorphanol (rarely used)

NSAIDs as used in Step 1

May add a nonopioid analgesic or an adjuvant agent

Opioids are the mainstay of steps two and three of the WHO ladder
(Table 3). Initiating them does not preclude ongoing use of nonopioid
agents. Reassurance that use of opioids for the management of pain will
not cause addiction is frequently necessary to encourage compliance. It is
helpful to distinguish physical dependence from addiction to reinforce
this. This also educates patients about the dangers of voluntary or unin-
tended sudden withdrawal of opioids. Physical dependence is neuroadapta-
tion to exogenous opioids, usually occurring after continuous dosing over 7
to 10 days. Sudden termination of exogenous opioids or administration of
opioid-reversing agents (naloxone) precipitates an abstinence syndrome
that is, somewhat misleadingly, the very picture of the popular stereotype
of the heroin addict: bodily aches, abdominal pain, cutis anserina (goose
flesh), sweating, nausea, vomiting, diarrhea, psychosis, hallucinations,
diarrhea, tachycardia, and hypertension. Precipitation of the abstinence
 1076
Table 3
Approximate opioid equivalences for management of moderate to severe pain. This conversion chart is a guideline only. Inter- and intra-individual variability
of response to opioids requires the dose to be individualized and titrated to effect.
Analgesic IM, SC, IV route (mg) PO route (mg) Comments
Morphine 10 30 Immediate and controlled-release formulation available.
Analgesic metabolite, M6G, can accumulate with renal impairment,
producing prolonged analgesia. Accumulation of analgesically
inactive metabolite M3G may cause neurotoxicity (myoclonus,
hyperalgesia, cognitive dysfunction).
Fentanyl 10 mcg IV z 1 mg IV morphine Short half-life, but at steady state, slow elimination from tissues can lead
25 mcg/hr patch q72h z 50 mg to prolonged half-life (up to 12 hours).
oral morphine/24h Opioid-naı̈ve patients should not be started with more than 12 mcg/hr
transdermally.

DUNN
Hydromorphone 1.5 7.5 Useful alternative to morphine, but oral dosing impractical for very high
doses because of tablet burden/dose.
Tolerated by patients with renal impairment.
Methadone [87] Ratios relative to methadone If daily morphine dose before switch is 30–90 mg, estimated dose ratio
depend on the dose of the (EDR) ¼ 4:1 (ie, 4 mg morphine ¼ 1mg methadone); 90–300 mg
previous opioid morphine, EDR ¼ 8:1; above 300 mg morphine, EDR ¼ 12:1.
Methadone has a variable and long half-life. Dosing interval every 8–12
hours.
Oxycodone _ 20 Immediate and controlled-release formulation available.
Used for moderate pain when combined with nonopioid (Step 2, WHO
Analgesic Ladder).
Can be used like oral morphine for severe pain when used as a single
entity. (Step 3, WHO Analgesic Ladder)
Dosing information adapted from Principles of analgesic use in acute pain and cancer pain, 5th edition. American Pain Society; Glenview, Illinois.
2003. p. 1–74.
 PALLIATING UNRESECTABLE COLORECTAL CANCER 1077

syndrome should be anticipated when rapidly withdrawing opioids follow-

ing procedures that significantly reduce pain (anesthetic blockade).
The converse of this situation is the syndrome of opioid overdosage that
can emerge as the result of marked opioid sparing following a pain-reducing
procedure without a reduction of the previous opioid dose.
Both of these scenarios can occur in a patient who has a painful sacral
recurrence of rectal cancer and who has been treated with high doses of sys-
temic opioids, and then undergoes a successful anesthetic block. Close mon-
itoring with frequent and careful evaluation is necessary for a period of 12 to
24 hours following pain-ablating procedures done to patients receiving
chronic opioid therapy.
True addiction has been defined as ‘‘a primary, chronic, neurobiological
disease, with genetic, psychological, and environmental factors influencing
its development and manifestations. It is characterized by behaviors that in-
clude one or more of the following: impaired control over drug use, compul-
sive use, continued use despite harm, and craving’’ [40]. Drug dependence is
physiologic and adaptive; drug addiction is pathologic.
Whenever opioid doses are adjusted, it is important to explain the reason
to patients, especially when death appears imminent. Patients often errone-
ously assume that the reason for dose escalation is tolerance to the drug in-
stead of progressing disease, almost always the real reason. It is critical that
patients near death and their caregivers understand that the purpose of the
opioids is the relief of pain (or dyspnea), not deliberate hastening of death;
however, according to the US Supreme Court’s 1997 opinion in Vaco v Quill
[41], the hastening of death resulting from the use of opioids for palliation of
pain should not be a deterrent to their use.
When initiating opioid therapy, common side effects should be antici-
pated and discussed. Counseling about and prophylaxis for possible opioid
side effects increases compliance. Constipation is a universal opioid side ef-
fect to which tolerance does not develop (the ‘‘other’’ neurologic crisis be-
sides unrelieved pain!). Prophylaxis for constipation should be started at
the commencement of opioid therapy using a stool softener and a stimulant
laxative such as senna. Bulk-forming agents such as psyllium or methylcel-
lulose should be avoided, because of their propensity to form impactions
with decreased fluid intake. Nausea is a frequent side effect of opioids; tol-
erance usually develops within a few days. Opioid-induced nausea can be
pre-empted by use of an antiemetic, a useful maneuver in patients who
have a history of opioid-induced nausea. Stimulants (methylphenidate)
can be used for persistent opioid-induced sedation, another opioid side-
effect to which tolerance normally develops fairly quickly.
The most feared opioid side effect is respiratory depression. Tolerance to
this, fortunately, develops quickly, making it an uncommon problem with
chronic usage. The risk of respiratory depression is greatest in the opioid-na-
ı̈ve, the elderly, patients who have impaired renal function, and those receiv-
ing sedating medications. Because respiratory depression is always preceded
1078 DUNN

by increasing sedation with opioid overdosage, careful monitoring provides

the opportunity to stop the agent until it is metabolized, and then resume it
at a lower dose. This is much less traumatic then full opioid reversal, which
can precipitate opioid withdrawal syndrome and abrupt return of severe
pain. When opioid reversal is necessary, dilute a 0.4 mg ampoule of nalox-
one with 10 ml saline and give 1 ml of this solution every 5 minutes until
partial reversal has occurred [42]. This may need to be repeated.
Because of the potential for dose-limiting side effects and reluctance of
many patients to escalate opioid doses, a strategy using the least amount
of opioid necessary for relief of pain is employed. Nonopioids and other
treatment modalities can be used with opioids for ‘‘opioid sparing’’ to ac-
complish this. An example of this is the concurrent use of systemic opioids
in conjunction with regional therapies such as external beam radiation or
epidural anesthesia.
Although many patients and surgeons are familiar with oxycodone as
a component of frequently prescribed compound medications for postoper-
ative pain, they may not be aware that the acetaminophen content of com-
pound drugs (usually 325 mg) limits their use for cancer pain management.
The maximum 24-hour allowance for acetaminophen in an adult who had
no liver disease or heavy alcohol consumption is 3000 mg. The amount
would be even less in the elderly. In a compound containing 325 mg of acet-
aminophen and 5 mg of oxycodone, the maximum daily amount of oxyco-
done that could be given as an agent compounded with acetaminophen
without incurring acetaminophen toxicity would be less than 50 mgdfar
short of what is often required for the effective relief of cancer pain. In in-
stances where oxycodone is prescribed, using the uncompounded form is
recommended to avoid potential acetaminophen toxicity and to provide
greater flexibility in dosing for breakthrough pain and conversion to long-
acting forms.
Two other medications encountered commonly in surgical practice, dolo-
phine and meperidine, have no role in cancer pain management because of
potential side-effects of their metabolites that can occur in the presence of
renal compromise, particularly in the elderly.
Codeine is of limited usefulness in cancer pain management for several
reasons. Codeine is metabolized to morphine, accounting, in part, for its an-
algesic property, though 7% of Caucasians genetically lack this capacity
[43]. It has unpredictable intramuscular absorption and cannot be given in-
travenously. It is sold compounded, limiting its use to the ceiling dose of
acetaminophen in the compound. Some studies have shown increasing unto-
ward side-effects for doses above that, yet diminishing returns for analgesic
effect [44].
Use of codeine for cancer patients should be restricted to perioperative
pain control for procedure-related pain that will resolve (eg, laparoscopy,
hernia repair; as an induction analgesic agent for mild to moderate cancer
pain; and control of cough and diarrhea).
 PALLIATING UNRESECTABLE COLORECTAL CANCER 1079

The pronounced anti-inflammatory properties of steroids (hydrocortisone,

dexamethasone) have made them useful as adjuvants to management of can-
cer pain complicated by neural compromise, especially if there are other symp-
toms such as anorexia, nausea, and generalized malaise that may also respond
to them. Future palliative care research should look more closely at the rela-
tionship between tumor necrosis factor (TNF) levels and palliation outcomes,
based on the suppressive property steroids are known to have upon TNF pro-
duction and the presence of TNF in many of the cytokine mediated metabolic
pathways generating systemic symptoms of advanced cancer. Activation of
some of the same pathways can also result from systemic cancer treatments.
The use of steroids must be considered in light of several possible signif-
icant side-effects that include immunosuppression, peptic ulceration, hyper-
gylcemia, psychological disturbances, and myopathy. Prophylaxis against
peptic ulceration is provided by concurrent use of a proton pump inhibitor.
Although the large majority (O80%) of patients who have cancer pain
will have their pain well-controlled using the WHO stepladder approach
[45], some patients will require procedural interventions for control of
pain because of failure to respond or presence of unacceptable side-effects
with pharmacotherapy. These interventions run the spectrum from direct
approaches correcting, reducing, or bypassing pathology to approaches di-
rected towards the neural basis of pain. Interventions can be operative or
minimally invasive operative procedures, percutaneous interventions, and
radiation therapy.
The description of specific interventions is beyond the scope of this
article.
Amersi and colleagues have recently written an excellent review [46] of
operative and ‘‘minimally invasive’’ interventions for the palliation of symp-
toms from advanced colorectal malignancy. They point out that studies
have previously been nonrandomized, case-control descriptions or case se-
ries that may have an inherent selection bias. Despite the diverse menu of
operative interventions reviewed, they conclude that the best approach to
palliation is one that is individualized, interdisciplinary, and incorporates
surgical and nonsurgical modalities. Recent evidence has shown that formal
education in palliative care significantly influences surgical judgment [47].
Radiation therapy has been used to mitigate symptoms of pain, bleeding,
obstruction, and fistula in advanced colorectal malignancy. The morbidity
of treatment is related to the anatomic position of the region treated, the ra-
diation field, and dose. A standard treatment regimen for symptomatic re-
current rectal disease, a particularly anguishing problem, consists of 40 Gy
in 20 fractions to 54 Gy in 30 fractions, often with systemic cytotoxic che-
motherapy to enhance radiation response [48]. Morbidity from radiation,
immediate and delayed, include strictures, fistula, bleeding, and ulcers.
Pharmacologic palliation of side effects from palliative radiation therapy
is usually possible. Endoscopic and surgical interventions are reserved for
more extreme complications in patients who have better prognosis.
1080 DUNN

Bonica and Benedetti [49] have proposed two additional steps to the
WHO analgesic stepladder for ‘‘very severe’’ and ‘‘excruciating’’ categories
of pain, which, if present at presentation, allow for immediate bypassing of
steps for control of pain. These additional steps, in addition to potent opioid
analgesics and coanalgesics, include neurolytic blocks, central nervous sys-
tem (CNS) stimulation procedures, and neurosurgery.

Nausea and vomiting

Nausea and vomiting is a common symptom complex in advanced cancer
[50], with a prevalence reported at 40% during the last 6 weeks of life [51]. It
can occur as a symptom with an identifiable etiology (opioids, bowel ob-
struction) or as a chronic syndrome characterized by some authors as a man-
ifestation of autonomic failure [52]. Similar to pain management, knowledge
of the physiology of nausea and vomiting allowing targeting of specific re-
ceptors is helpful in the selection of therapy. Though closely related, nausea
should be seen as a separate entity from vomiting, because each is a distinct
subjective experience and has an anatomically distinct locus in the CNS as-
sociated with it. The central mediation of nausea and vomiting is located in
two centersdthe chemoreceptor trigger zone (CTZ), which senses distress;
and the vomiting center (VC), which responds to the distress. Both of these
areas are extensively integrated with the rest of the central nervous system,
with far-ranging sympathetic and parasympathetic input. In addition, the
metabolic milieu of the body is continuously presented to the exquisitely
sensitive chemosensors of the CTZ through their direct contact with cere-
brospinal fluid. The psychological and practical implications of the two cen-
ters differing roles (nausea center ¼ distress, vomiting center ¼ ‘‘fighting
back’’) are helpful in counseling patients about the management of nausea
and vomiting, particularly in the nonoperative management of MBO [53].
The VOMIT acronym (Vestibular, Obstruction, Motility, Inflammatory,
Toxins) [54] is a somewhat graphic way of classifying the etiologies of nau-
sea and vomiting. For each etiologic class, specific receptors for nausea are
implicated providing the rationale for the use of the blockers specific for the
receptors (Table 4). The identification of activity in the inferior frontal gyrus
of the cortex (association cortex) during the sensation of nausea [55] demon-
strates the importance of cortical input for this perception, and suggests a ra-
tionale for therapies directed towards this region of the central nervous
system (eg, anxiolytics for anticipatory nausea).
Although the management of nausea and vomiting in the setting of ad-
vanced colorectal cancer is heavily weighted towards pharmacotherapy,
nonpharmacologic approaches, pending formal evidence-based confirma-
tion of their efficacy, should be encouraged if the therapy is not known to
induce harm. Chronic nausea is a psychologically debilitating symptom.
Anything that fosters a sense of hope, even openness to trials of complemen-
tary therapies, is welcomed by patients. Openness develops the basis of trust
 PALLIATING UNRESECTABLE COLORECTAL CANCER 1081

Table 4
VOMIT acronym. Nausea and vomiting: etiologies, receptors, drugs
Etiology Receptors Blockers/stimulants
Vestibular Cholinergic, histaminic Scopolamine patch,
autonomic failure, promethazine
hypovolemia
Obstruction Cholinergic, histaminic, Senna products (stimulate
constipation, 5HT3 myenteric plexus)
but not mechanical
obstruction
Motility Cholinergic, Histaminic, Metoclopropamide
Upper intestinal tract 5HT3 (contraindicated in mechanical
dysmotility bowel obstruction)
Inflammatory Cholinergic, Histaminic, Promethazine
Infection, inflammation 5HT3.
(radiation therapy) Neurokinin 1

Toxins Dopamine 2, 5HT3 Prochlorperazine, haloperidol,

Opioids, chemotherapy, odansetron (no established
uremia, digitalis superiority as first-line agent,
expensive)
Adapted from Hallenbeck J. Fast fact and concept #5: Causes of nausea and vomiting
(V.O.M.I.T.). August 2005, 2nd edition. End-of-Life Physician Education Resource Center.
Available at: www.eperc.mcw.edu.

necessary to encourage consideration of more aggressive approaches, in-

cluding operative and chemotherapy interventions.

Bowel obstruction
The management of MBO from advanced cancer is the paradigmatic
challenge of surgical palliative care (Table 5). Surgeons who have opened
the abdominal cavity and have been greeted by straw-colored ascites and
the tactile sensation of diffuse tumor studding of the peritoneum know the
sinking feeling: ‘‘What am I doing here? I wish I could have avoided this.
Now what am I going to do?’’
Colorectal cancer accounts for approximately 10% to 28% of instances
of MBO [56]. MBO is a heterogeneous clinical syndrome whose manage-
ment is influenced by the levels of obstruction, the pattern of disease, clinical
features of the disease related to prognosis, prior treatment, and patient
preferences. Management decisions must also account for the possibility
of a benign etiology for bowel obstruction in the patient who have previ-
ously diagnosed malignancy; the incidence of benign causes of obstruction
in these patients approaches 50% in some series [57].
Plain film radiography remains the initial evaluation of patients who have
suspected bowel obstruction. It can accurately diagnose small bowel ob-
struction up to 60% of the time [58]. The diagnostic findings of bowel ob-
struction, dilated loops of small bowel, multiple air fluid levels on upright
1082 DUNN

Table 5
Pharmacotherapy of malignant bowel obstruction (MBO)
Drug Action Dose
Opioids (morphine, Antiperistaltic effect reduces Titrate to relief of pain.
hydromorphone) cramping, targets pain from Both can be given PO, SC, IV, or
bowel distention or tumor SL (see Table 4 for PO to IV or
invasion. SC conversion ratios).
Antisecretory agents (1) Octreotide minimizes Octreotide: 0.3 mg/day SC. Can
(octreotide, volume of emeses betitrated up to 0.6 mg/day. Can
scopolamine, (2) Minimize gut distention be given as infusion or
glycopyrrolate) by reducing gastrointestinal scheduled BID dosing. Favored
secretions. for rapid reduction of GI
(3) Anticholinergic antisecretory secretions.
agents slow gut peristalsis, Expensive.
making them effective for Scopolamine: 1–2 patches every 3
abdominal cramping. days. Anticholinergic side effects
include dry mouth, urinary
retention, confusion.
Convenient.
Glycopyrrolate: start at 0.1 to
0.2 mg Q6hr SC PRN adjusted
to renal and hepatic function
[86]. Some anticholiergic side-
effects, though no CNS effects,
unlike scopolamine
Central-acting Relief of nausea Haloperidol: 5 to 15 mg/day SC,
antiemetics IV, or PO.
(haloperidol Chlorpromazine: 25 to 100 mg
chlorpromazine, TID PO, PR, or IV. Use if
hydroxyzine) sedation desired.
Avoid above in patients with
Parkinsonism or patients taking
other drugs with potential
Parkinsonian side-effects (ie,
metoclopropamide).
Dystonia can be reversed with
diphenhydramine 1 mg/kg or
benztropine 0.02–0.05 mg/kg/
dose, (maximum ¼ 4 mg IV)
Hydroxyzine: 100–200 mg/day
PO, SC, or IV.
Steroids (1) Central antiemetic effect Dexamethasone: 12–16 mg/day.
(dexamethasone) (2) Reduces inflammatory Discontinue if no relief after 5
component of tumor days. Helpful for concommitant
causing mechanical bowel pain from hepatic distention
obstruction from metastases or symptomatic
(3) Appetite stimulant CNS metastases.
Propulsive agents Increase peristaltic activity. Metoclopramide: 60 mg/day PO,
(metoclopropamide, Contraindicated in complete SC, IV. Also can act as
amidorizoate) obstruction, but may have a centrally active antiemetic in
use in partial MBO. higher doses. Can worsen colic
and may not be used with
anticholinergics.
 PALLIATING UNRESECTABLE COLORECTAL CANCER 1083

views, and absent or minimal intraluminal gas distal to the point of obstruc-
tion, may be muted or misinterpreted in patients who have malignant ob-
struction. Tumor can encase the bowel, preventing distention, can cause
bowel dilatation from paresis due to infiltration of the mesentery and auto-
nomic nerves, and can obstruct at more than one point.
Because of the limitations of plain film radiography, CT imaging has in-
creasingly become the gold standard for diagnosis, especially for suspected
malignant obstruction. It is sensitive, particularly for high-grade obstruc-
tion, and can identify the cause of obstruction in over 95% of cases [59]. Be-
cause of its specificity, it is the preferred initial screening modality for
patients presenting with suspected bowel obstruction who have abdominal
tenderness, leukocytosis, or fever [60].
Adjuncts to the primary imaging modalities include oral small bowel fol-
low-through with barium or water-soluble contrast agent, enteroclysis, and
contrast enema. These can be helpful in identifying low-grade obstructions
and in establishing the diagnosis of carcinomatosis [61]dgenerally a sign of
nonoperability.
Based on accumulating data, Krouse and colleagues [62] recommend the
use of CT in the evaluation of suspected MBO, especially if invasive treat-
ment approaches are being considered, because there are no comparative
studies comparing cost, patient comfort, and accuracy of current radio-
graphic diagnostic modalities.
At presentation of MBO, nasogastric drainage is routinely initiated before
plain radiographs are obtained. This controls nausea and vomiting promptly,
but at the cost of having a nasogastric tube inserted and left in place. Nasogas-
tric drainage should be used only as a temporizing measure for the relief of
obstructive symptoms in patients who have advanced or terminal colorectal
cancer. No one needs to ‘‘have a tube in until they die’’ anymore when satis-
factory control of symptoms is almost always possible with operative interven-
tion, minimally invasive procedures, or pharmacotherapy. Although bowel
obstruction in the setting of advanced malignancy will spontaneously resolve
during a trial of nasogastric suction in some cases, a course of either operative
or nonoperative management should be planned for durable relief of symp-
toms. Consultation with palliative care services, when available, is helpful to
assist in framing the context of future care through skilled communication,
pharmacologic management of obstructive symptoms, and attention to social,
psychologic, and spiritual aspects of progressing life-limiting illness.
Surgeons are familiar with operative approaches for the relief of MBO, as
well as some of the other technically difficult problems encountered in the
setting of intra-abdominal malignancy (ie, radiation enteritis and bowel fis-
tulas); however, because of the unique nature of an individual’s suffering
and the reasons for it, prospective randomized trials of palliative surgical
interventions have not been performed in the past, and will only be conceiv-
able when research methodologies used in the social sciences are employed
in addition to those used in the natural sciences.
1084 DUNN

Mortality and morbidity are sobering when considering operation for

MBO: operative mortality has been reported to be as high as 32% [63]
and morbidity as high as 42% [64], with reobstruction rates up to 50%
[65]. In a review of 823 palliative surgical procedures by Cady and colleagues
[66], a major postoperative complication reduced the probability of symp-
tom relief to 17%, and the combination of a complication-free palliative
procedure and symptom resolution until death (their ideal outcome) oc-
curred in only 16% of patients.
Relative contraindications to surgery for malignant obstruction are many
and include: diffuse intraperitoneal carcinomatosis, palpable multiple intra-
abdominal masses, liver metastases, extra-abdominal metastases, pleural ef-
fusions, multiple sites of partial obstruction or prolonged transit time of
contrast on intestinal radiographs, ascites, cachexia or hypoalbuminemia,
advanced age, poor performance status, recurrence following recent laparot-
omy for malignant obstruction, previous abdominal radiation therapy, and
disease refractory to chemotherapy. Reports of success and patterns of fail-
ure are inconsistent: one study demonstrated that multiple sites of obstruc-
tion by colorectal cancer did not correlate with failure to resume bowel
function or postoperative death [67]. In weighing options for surgery in
the obstructed patient who has known malignancy, it must be remembered
that a certain percentage of these (up to 48%) are due to benign causes [68].
Operation is preceded by the process of informed consent, in which it is
clearly understood by patient, surrogates, and family that the primary goal
of treatment is improvement or preservation of quality of life, without nec-
essarily having any impact on the disease process itself. Nonabandonment
by the surgeon and other members of the care team should be assured; if op-
portunity allows, the subject of advance directives should be broached. Ar-
rangements for possible postoperative axial analgesia (ie, epidural catheter
placement) should be discussed with anesthesiologist consultants.
Spiritual concerns and needs should be assessed for appropriate
ministrations.
The principles of operative management outlined by Krouse [69] include
liberal use of bypass and stomas (preoperative enterostomal therapist con-
sultation advised) in avoiding extensive dissection and performance of gas-
trostomy in cases where no resection or bypass for relief of obstruction can
be performed. Laparoscopic procedures will inevitably become more com-
monly performed, though no large series for this indication have been
reported.
Intestinal stenting is an attractive alternative to surgery that, in addition
to relief of obstruction, can permit bowel preparation for a resection and
primary closure. Covered stents have been used for the control of intestinal
fistulas. Complications are most likely during or immediately after insertion.
These include perforation, bleeding, malpositioning, or migration. Tumor
ingrowth with reobstruction can occur from days to months after insertion.
Laser therapy, usually with Nd:Yag lasers, can be used for recanalization
 PALLIATING UNRESECTABLE COLORECTAL CANCER 1085

through intraluminal malignancy, but not for extrinsically compressing or

completely obstructing lesions. Usually several treatments are required. Re-
lief has been reported in 75% to 80% of patients [70].
Although 20 years have elapsed since Baines and colleagues reported suc-
cessful management of symptoms of inoperable MBO using pharmacother-
apy alone [71], nonoperative approaches other than nasogastric tube
drainage have been slow to find their way into surgical practice. Lack of
this information and knowledge of its satisfactory outcomes have probably
led to many unnecessary laparotomies. Several studies documenting a very
low incidence of strangulation secondary to malignant obstruction [72,73]
can reassure the surgeon that a nonoperative approach to its management
will not hazard intrabdominal septic catastrophe.
The goal of pharmacotherapy for MBO is relief of symptoms (pain, nau-
sea, and fewer than two emeses/day), not resolution of obstruction. In some
cases, however, relief of obstruction may occur if obstruction is partial. Suc-
cessful symptom control for MBO can be anticipated using combination
therapy with an opioid, an anticholinergic (or antisecretory agent such as
octreotide), and an antiemetic without intestinal drainage tubes, surgery,
or intravenous fluids [74]. A prospective randomized study of patients re-
ceiving antisecretory therapy for inoperable MBO [75] demonstrated that
patients receiving octreotide had a more rapid reduction in GI secretions (al-
lowing for earlier removal of a nasogastric tube) than patients receiving an
anticholinergic antisecretory drug (scopolamine butylbromide). All patients
studied were assessed for pain, nausea, dry mouth thirst, abdominal disten-
tion, dyspnea, and drowsiness. A significant observation, given the emo-
tional issue of hydration at end of life, was the lack of difference in the
daily thirst and dry mouth intensity in relation to the amount of parenteral
hydration or the treatment provided. Also observed was that patients receiv-
ing less parenteral hydration presented with significantly more nausea and
drowsiness. Pain control was good in all patients studied and most were
cared for at home [75].
During pharmacotherapy for MBO, patients will tolerate an occasional
emesis as long as nausea is well-controlled. This discrepancy can be used
to advantage if the patient desires to eat or drinkdas long as nausea is ab-
sent many patients are willing to risk or even accept an occasional emesis in
order to try a small amount of food or liquid. As a general rule, the more
distal the obstruction, the more likely it can be managed successfully with
pharmacotherapy alone. For resistant symptoms, high intestinal obstruc-
tion, or burdensome MBO-related medication side effects, placement of
a percutaneous endoscopic gastrostomy (PEG) tube for drainage should
be considered.
Mercadante [76] has reported success in reversing MBO using an aggres-
sive and early multiple agent therapy (dexamethasone, octreotide, metoclo-
propamide, and gastrografin) that combines propulsive and antisecretory
agents. In addition to symptom relief, bowel function probably resumed
1086 DUNN

because treatment was initiated before permanently obstructing impaction

and edema developed.

Ascites
Malignant ascites is a harbinger of very limited prognosis, associated with
a median survival of roughly 2 months. If no further disease-directed therapy
is anticipated, the finding of malignant ascites would satisfy admission crite-
ria for hospice care, following the appropriate course of disclosure and con-
sent of the patient. Malignant ascites is usually caused by diffuse peritoneal
and mesenteric tumor infiltration, accounting for its poor prognosis. Knowl-
edge of this pathogensesis is important in understanding the rationale and
limitations of therapy. Unless liver involvement with tumor has led to portal
hemodynamic changes, diuretics useful in nonmalignant ascites are of limited
or no use for the malignant variety, and they may even complicate manage-
ment of other terminal symptoms by inducing prerenal azotemia, leading to
lethargy and confusion from accumulating drug metabolites. Symptoms
caused by ascites are fullness, bloating, pain, dyspnea, nausea, and body im-
age-related distress.
Drainage is the definitive symptom management approach, except in pa-
tients actively dying, in which case relief of symptoms (pain and dyspnea) is
best managed by opioids. Even if chemotherapy is expected to resolve intra-
abdominal disease, drainage is still necessary for temporary symptom relief.
No particular drainage procedure has emerged as evidence-based best. The
simplest approach is paracentesis, which has the advantages of speed, simplic-
ity, flexibility of treatment setting (it can be done in the home or office), and
low complication rate. Ultrasound has been increasingly used, especially
when loculated ascites is suspected. Drainage of malignant ascites is less likely
to be followed by hemodynamic instability than nonmalignant varieties. Vol-
umes of 5 liters drained at a sitting have been reported as safe [77], though
larger volumes can be justified to relieve severe distress, especially dyspnea.
More durable drainage approaches include placement of a pigtail catheter,
Tenkhoff catheter [78], pleurex catheter [79], or peritoneal port [80]. Peritoneal
venous shunting has been used for malignant ascites, though morbidity re-
lated to infection, congestive heart failure, coagulopathy, venous thrombosis,
and shunt occlusion from protein-rich ascites, as well as the need for hospital-
ization, are sobering downsides to palliative benefit. Nondrainage approaches
for ascites management include systemic or intraperitoneal instillation of che-
motherapeutic agents, or intracavitary injection of sclerosing agents, though
these approaches may take weeks to realize symptomatic benefit.

Cachexia-anorexia
The incidence of cachexia-anorexia syndrome (or cancer cachexia syn-
drome) in patients who have advanced colorectal malignancy is not known,
 PALLIATING UNRESECTABLE COLORECTAL CANCER 1087

though it is most common in solid tumors (pancreas, stomach, lung) except

breast [81]. Both anorexia and cachexia are quite common in patients pre-
senting to palliative care or hospice programs.
Cancer cachexia has been defined as ‘‘a wasting syndrome involving loss
of muscle and fat directly caused by tumor factors, or indirectly caused by
an aberrant host response to tumor presence’’ [82]. Cancer cachexia is char-
acterized by breakdown of skeletal muscle and deranged fat and carbohy-
drate metabolism, despite adequate or forced nutritional intake. Although
the appearance of this syndrome outwardly mimics simple starvation, its
metabolic basis is profoundly different. To suggest to patients ‘‘you need
to eat more’’ in the face of advanced cancer cachexia overlooks this differ-
ence and creates confusion and doubt about medical guidance. Many eti-
ologies for cachexia and anorexia in cancer patients are reversible,
however, and a diligent search for these should not be prevented by a sense
of resignation about advanced illness. Megesterol acetate, dexamethasone,
cannabinoids, and other agents have been used successfully to enhance ap-
petite and sense of well-being, but have not been shown to reverse ca-
chexia-anorexia syndrome. Omega-3 fatty acids found in some fish oil
supplements show some promise in the treatment of this syndrome
[83,84]. See Box 5 for general recommendations for the management of
cancer cahexia syndrome.
The inevitability of appetite and weight loss in patients who have pro-
gressing and incurable colorectal cancer requires surgeons to be familiar
with this syndrome in order to respond helpfully to the equally inevitable
questions from patients and families about what can be done and not
done about it. Because the association between feeding and caring is so cen-
tral to most cultures, the surgeon must walk the delicate line between truth-
fulness about the limitations of feeding or medication for this condition, and
not taking away hope or impugning caregivers’ good intentions. The word
‘‘starvation’’ should be used carefully in discussions about weight loss in ad-
vanced cancer when one considers the different physiologic and moral con-
notations of weight loss in starvation, forced starvation, and cancer-induced
cachexia. To imply or allow it to be implied that neglect or lack of aggres-
siveness is the reason for weight loss when it is the result of an irreversible
physiologic process risks future bereavement complications. Even in the
most far advanced cases of cachexia, surgeons can use their well-earned
credibility on matters of nutrition and gastrointestinal function to comfort
a family, by clarifying the reasons for weight loss and its implications for
care giving. Gently redirecting a family’s attention from feeding to other ex-
pressions of concern is almost always deeply appreciated by a patient no
longer possessing the appetite or strength to consume food. Weight loss
in this context does not mean hunger. McCann and colleagues [85] demon-
strated that hunger and thirst are uncommon complaints in terminally ill
cancer patients. When these symptoms occurred, small amounts of food
or water were needed for symptom relief.
Box 5. Recommendations for management of cancer cachexia
syndrome
1. Regularly assess weight, change in appetite, food intake.
2. Obtain sHb, sAlbumin, and C-reactive protein as a baseline
study.
3. Systematically rule out treatable secondary causes of
wasting.
4. Refer for nutritional counseling from a team with special
interest in wasting disorders.
5. Encourage patient participation in a rehabilitation program
for consideration of the use of specific nutritional and
pharmacologic interventions.
6. Follow-up visits should review antitumor therapy, tumor
volume, symptom control, weight, appetite, and function.
7. Regular review of the full medication profile of patients who
are wasting. These might include drugs that could have
a favorable effect on anorexia-cachexia (cardiac agents such
as the statins, angiotensin-converting enzyme [ACE]
inhibitors) and other agents that may be deleterious (eg,
herbal medications laced with corticosteroids).
8. It is reasonable for patients to increase the amount of
eicosapentaenoic acid (EPA)-rich omega-3 fatty acids in their
diet.
9. Testosterone status should be established in cancer patients
who have the anorexia-cachexia syndrome. If clearly
reduced, physiologic testosterone supplementation should
be considered after discussion with the patient.
10. Patients must be assured of a reasonable intake of amino
acids. Protein-containing foods are indicated, and rich
sources of both essential and nonessential amino acids will
support any anabolic potential.
11. Clinical researchers are encouraged to familiarize themselves
with the work of colleagues in sports medicine, AIDS, and
geriatrics.
12. Patients should be encouraged to keep active or take part in
tailored exercise programs.
13. Surgeon researchers have led the way in studies on acute
stress reactions related to surgery and trauma, and the role
of nutrients in combating these events. These can guide
further anorexia-cachexia trials, and should be expanded to
encompass studies on wasting associated with chronic
disease.

Adapted from MacDonald N, Easson A, Mazurak V, et al. Understanding and

managing cancer cachexia. J Am Coll Surg 2003;197:143–61. 82; with permission.
 PALLIATING UNRESECTABLE COLORECTAL CANCER 1089

The surgeon can mitigate his sense of frustration or failure in this and
other scenarios of terminal illness by willingness to participate in the process
of reframing the problem in a larger and redefined context of meaning. This
is the ultimate test of the primary principles of all surgical palliative care:
nonabandonment and preservation of hope.

Summary
The last phases of colorectal malignant illness may be the most challeng-
ing and saddening for all involved, but they offer opportunities to become
the most rewarding. This transformation of hopelessness to fulfillment re-
quires a willingness by surgeon, patient, and patient’s family to trust one an-
other to realistically set goals of care, stick together, and not let the
treatment of the disease become a surrogate for treating the suffering that
characterizes grave illness.

References
[1] American Cancer Society. Cancer facts and figures 2005. Atlanta (GA): American Cancer
Society; 2005. Available at: www.cancer.org/docroot/CRI/content/CRI_2_4_1x_what_are_
the_key_statistics_for_colon_and_rectum_cancer.asp?rnav-cri. Accessed June 30, 2006.
[2] DeCosse JJ, Cennerazzo WJ. Quality-of-life management of patients with colorectal cancer.
CA Cancer J Clin 1997;47:198–206.
[3] Amersi F, Stamos MJ, Ko CY. Palliative care for colorectal cancer. Surg Oncol Clin N Am
2004;13(3):467–77.
[4] The SUPPORT principal Investigators. A controlled trial to improve care for seriously ill
hospitalized patients: The study to understand prognoses and preferences for outcomes
and risks of treatments (SUPPORT). JAMA 1995;274:1591–8.
[5] Billings JA. Module 1: Gaps in the end-of-life care of cancer patients. Disseminating end-
of-life education to cancer cCenters. The National Cancer Institute, 2001.
[6] Emanuel EJ, Fairlough DL, Slutsman J, et al. Understanding economic and other burdens of
terminal illness: the experience of patients and their caregivers. Ann Intern Med 2000;132:
451–9.
[7] Covinsky KE, Goldman L, Cook EF, et al. The impact of serious illness on patients’ families.
JAMA 1994;272:1839–44.
[8] Morrison RS, Meier D. Palliative care. N Engl J Med 2004;350:2582–90.
[9] American College of Surgeons. Principles of care at end of life. Bull Am Coll Surg 1998;83:
46.
[10] American College of Surgeons. Principles of palliative care. Bull Am Coll Surg 2005;90(8):
34–5.
[11] American Board of Surgery, Inc. American Board of Surgery Qualifying Examination.
Booklet of information. July 2005–June 2006. Available at: www.absurgery.org/xfer/ABS_
BookletOfInfo05.pdf. Accessed January 23, 2006.
[12] Finlayson CA, Eisenberg BL. Palliative pelvic exenteration: patient selection and results.
Oncology 1996;10(4):479–84 [Huntingt].
[13] Colorectal Meta-analysis Collaboration. Palliative chemotherapy for advanced or meta-
static colorectal cancer. Cochrane Database Syst Rev 2000;2:CD001545.
[14] Cady B, Easson A, Aboulafia A, et al. Part 1: Surgical palliation of advanced illnessdwhat’s
new, what’s helpful. J Am Coll Surg 2005;200(1):115–27.
1090 DUNN

[15] Buckman R. How to break bad news. Baltimore (MD): Johns Hopkins Press; 1992. p. 65–97.
[16] Morrison SR, Maroney-Galin C, Kralovec PD, et al. The growth of palliative care programs
in United States hospitals. J Palliat Med 2005;8(6):1127–34.
[17] Morrison S, Meier DE. Palliative care. N Engl J Med 2004;350:2582–90.
[18] Elsayem A, Swint K, Fisch M, et al. Palliative care inpatient service in a comprehensive can-
cer center: clinical and financial outcomes. Clin Oncol 2004;22:2008–14.
[19] Meier DE. Palliative care in hospitals: making the case. New York: Center to Advance Pal-
liative Care; 2002.
[20] Higginson IJ, Finlay IG, Goodwin DM, et al. Do hospital-based palliative teams improve
care for patients or families at the end of life? J Pain Symptom Manage 2002;23:96–106.
[21] Campbell ML, Guzman JA. Impact of a proactive approach to improve end-of-life care in
a medical ICU. Chest 2003;123:266–71.
[22] Manfredi PL, Morrison RS, Morris J, et al. Palliative care consultations: how do they impact
the care of hospitalized patients? J Pain Symtom Manage 2000;20(3):166–73.
[23] Beach MC, Morrison RS. The effect of do-not-resuscitate orders on physician decision-
making. J Am Geriatr Soc 2002;50:2057–61.
[24] National Hospice and Palliative Care organization. Available at: www.nhpco.org/Files/
members/TheCostsofHospiceCare-Millman.pdf. Accessed February 23, 2006.
[25] Portenoy RK, Thaler HT, Kornblith AB, et al. The Memorial Symptom Assessment Scale:
an instrument for the evaluation of symptom prevalence, characteristics, and distress. Eur J
Cancer 1994;30A(9):1326–36.
[26] Cella DF, Tulsky DS, Gray G, et al. The Functional Assessment of Cancer Therapy Scale;
development of and validation of the general measure. J Clin Oncol 1993;11:570–9.
[27] Bruera E, Kuehn N, Miller MJ, et al. The Edmonton Symptom Assessment System. (ESAS):
a simple method for the assessment of palliative care patients. J Palliat Care 1991;7(2):6–9.
[28] Larue F, Colleau SM, Brasseur L, et al. Multicentre study of cancer pain and its treatment
in France. BMJ 1995;310:1034–7.
[29] Von Roenn JH, Cleeland CS, Gonin R, et al. Physicians’ attitudes and practice in cancer pain
management: a survey from the Eastern Cooperative Oncology Group. Ann Intern Med
1993;119:121–6.
[30] Cleeland CS, Gonin R, Hatfield AK, et al. Pain and its treatment in outpatients with meta-
static cancer. N Eng J Med 1994;330:592–6.
[31] Wolfe J, Grier HE, Klar N, et al. Symptoms and suffering at the end of life in children with
cancer. N Eng J Med 2000;342:326–33.
[32] American Pain Society Quality of Care Task Force. American Pain Society recommenda-
tions for improving the quality of acute and cancer pain management. Arch Intern Med
2005;165(14):1574–80.
[33] Twycross RG. Pain and analgesics. Curr Med Res Opin 1978;5:497–505.
[34] Ray JB. Pharmacologic management of pain. Surg Oncol Clin N Am 2001;10(1):71–87.
[35] Saunders S, Sykes N. The management of terminal malignant disease. 3rd edition. London:
Edward Arnold; 1993.
[36] Levy M. Pharmacological treatment of cancer pain. N Engl J Med 1996;335:1124–32.
[37] World Health Organiztion. Cancer pain relief and palliative care. Report of a WHO expert
committee. World health Organization Technical Report Series, 804. Geneva (Switzerland):
World Health Organization; 1990. p. 1–75.
[38] Jacox A, Carr DB, Payne R, et al. Mangement of cancer pain. AHCPR publication No.
94-0592: Clinical practice guideline No. 9. Rockville (MD): U.S. Department of Health and
Human Services, Public Health Service; March 1994. p. 1–257.
[39] Pace V. Use of non-steroidal anti-inflammatory drugs in cancer. Palliat Med 1995;9:273–86.
[40] Definitions related to the use of opioids for the treatment of pain: a consensus document
from the American Academy of Pain Medicine, the American Pain Society, and the Amer-
ican Society of Addiction medicine. 2002. available at: www.painmed.org/productpub/
statements/pdfs/definition.pdf. Accessed April 5, 2006.
 PALLIATING UNRESECTABLE COLORECTAL CANCER 1091

[41] Vaco v Quill, 521 US 793 (1997).

[42] Storey P, Knight C. UNIPAC three: assessment and treatment of pain in the terminally ill.
2nd edition. Larchmont (NY): Mary Ann Liebert Publishers; 2003. p. 1–91.
[43] Sindrup SH, Brosen K. The pharmacogenetics of codeine hypoalgesia. Pharmacogenetics
1995;5(6):335–46.
[44] Lipman AG, Gauthier ME. Pharmacology of opioid drugs: basic principles. In: Portnoy R,
Bruera E, editors. Topics in palliative care, vol. 1. New York: Oxford University Press; 1997.
p. 1–316.
[45] Stjernsward J, Colleau SM, Ventafridda V. The World health Organization Cancer Pain
and Palliative Care Program: past, present, and future. J Pain Symptom Manage 1996;
12:65–72.
[46] Amersi F, Stamos MJ, Ko CY. Palliative care of colorectal cancer. Surg Oncol Clin N Am
2004;13:467–77.
[47] Galante JM, Bowles TL, Khatri VJ, et al. Experience and attitudes of surgeons toward pal-
liation in cancer. Arch Surg 2005;140:873–80.
[48] Howell D. Radiation therapy in the palliation of gastrointestinal malignancies. Gastro-
enterol Clin North Am 2006;35:125–30.
[49] Bonica JJ, Benedetti C. Management of cancer pain. In: Moosa AR, Robson MC, Schimpff
SC, editors. Comprehensive textbook of oncology. Baltimore (MD): Williams &Wilkins;
1986. p. 462.
[50] Coyle N, Adelhardt J, Foley KM, et al. Character of terminal illness in the advanced cancer
patient: pain and other symptoms during the last four weeks of life. J Pain Symptom Manage
1990;5(2):83–93.
[51] Reuben DB, Mor V. Nausea and vomiting in terminal cancer patients. Arch Intern Med
1986;146(10):2021–3.
[52] Ventafridda V, Sbanotto A, Burnhill R. Palliative care. In: Garden OJ, Geraaghty JG, Na-
gorney DM, editors. Liver metastases. biology, diagnosis, and treatment. London: Springer-
Verlag Limited; 1998. p. 1–207.
[53] Mount BM. Bowel obstruction. In: MacDonald N, editor. Palliative medicine. A case based
manual. Toronto: Oxford Medical Publications; 1998. p. 1–318.
[54] Hallenbeck J. Fast fact and concept #005; treatment of nausea and vomiting. Available at:
www.eperc.mcw.edu/FastFact/ff_005.htm. Accessed February 24, 2006.
[55] Miller AD, Rowley HA, Roberts TPL, et al. Human cortical activity during vestibular and
drug-induced detected using M.S.I. In: Highstein SM, Cohen B, Buettner-Ennever JA, edi-
tors. New directions in vestibular research. 1996; Ann NY Acad Ser 781. p. 670–2.
[56] Davis MP, Nouneh D. Modern management of cancer-related intestinal obstruction. Curr
Pain Headache Rep 2001;5:257–64.
[57] Legendre H, Vahhuyse F, Caroli-Bosc FX, et al. Survival and quality of life after palliative
surgery for neoplastic gastrointestinal obstruction. Eur J Surg Oncol 2001;27:364–7.
[58] Maglinte DD, Kelvin FM, O’Connor K, et al. Current status of small bowel radiography.
Abdom Imaging 1996;21:247–57.
[59] Maglinte DD, Reyes BL, Harmon BH, et al. Reliability and role of plain film radiography
and CT in the diagnosis of small-bowel obstruction. Am J Roentgenol 1996;167(6):1451–5.
[60] American College of Surgeons, Division of Education. Surgical Education and Self-Assess-
ment Program (SESAP) No. 12. Vol. 1. 2004. p. 1–626.
[61] Bundrick T, Cho SR, Ammann A, et al. Intraperitoneal carcinomatosis: incidence of its ra-
diographic findings and description of a new sign. Br J Radiol 1983;56:13–6.
[62] Krouse RS, McCahill LE, Easson AM, et al. When the sun can set on an unoperated bowel
obstruction: management of malignant bowel obstruction. J Am Coll Surg 2002;195:117–28.
[63] Averbach AM, Sugarbaker PH. Recurrent intra-abdominal cancer with intestinal obstruc-
tion. Int Surg 1995;80(2):141–6.
[64] Makela J, Kiviniemi H, Laitinen S, et al. Surgical management of intestinal obstruction after
treatment for cancer. Eur J Surg 1991;157:73–7.
1092 DUNN

[65] Miner T, Jaques DP, Paty PB, et al. Symptom control in patients with locally recurrent rectal
cancer. Ann Surg Oncol 2003;10(1):72–9.
[66] Cady B, Miner T, Morgenthaler A. Part 2: Surgical palliation of advanced illnessdwhat’s
new, what’s helpful. J Am Coll Surg 2005;2005(2):281–90.
[67] Lau PWK, Lorentz TG. Results of surgery for malignant bowel obstruction in advanced, un-
resectable, recurrent colorectal cancer. Dis Colon Rectum 1993;36:61–4.
[68] Legendre H, Vah Huyse F, Caroli-Bosc FX, et al. Survival and quality of life after palliative
surgery for neoplastic gastrointestinal obstruction. Eur J Surg Oncol 2001;27:364–7.
[69] Krouse RS. Surgical management of malignant bowel obstruction. Surg Oncol Clin N
Am;13:479–90.
[70] Adler DG, Merwat SN. Endoscopic approaches for palliation of luminal gastrointestinal
obstruction. Gastroenterol Clin North Am 2006;35:65–82.
[71] Baines M, Oliver DJ, Carter RL. Medical management of intestinal obstruction with ad-
vanced malignant disease: a clinical and pathological study. Lancet 1985;II:990–3.
[72] Butler JA, Cameron BL, Morrow M, et al. Small bowel obstruction in patients with a prior
history of cancer. Am J Surg 1991;162:244–9.
[73] Rubin SC. Intestinal obstruction in advanced ovarian cancer: what does the patient want?
Gynecol Oncol 1999;75:311–2.
[74] Ripamonti C, Conno FD, Ventafridda V, et al. Management of bowel obstruction in ad-
vanced and terminal cancer patients. Ann Oncol 1993;4:15–21.
[75] Ripamonti C, Mercadante S, Groff L, et al. Role of octreotide, scopolamine butylbromide,
and hydration in symptom control of patients with inoperable bowel obstruction and naso-
gastric tubes: a prospective randomized trial. J Pain Symptom Manage 2000;19(1):23–34.
[76] Mercadante S, Ferrera P, Villari P, et al. Aggressive pharmacological treatment for reversing
malignant bowel obstruction. J Pain Symptom Manage 2004;28(4):412–6.
[77] Muir JC. Ascites. In: Von Roenn J, Smith TJ, Loprinzi CL, et al, editors. ASCO curriculum:
optimizing cancer care. The importance of symptom management, vol. 1. Dubuque (IA):
Kendall/Hunt; 2001. p. 1–31.
[78] Lomas DA, Wallis PJ, Stockley RA. Palliation of malignant ascites with a Tenkhoff catheter.
Thorax 1989;44:928.
[79] Richard HM, Coldwell DM, Boyd-Kranis RI, et al. Pleurex tunneled catheter in the manage-
ment of malignant ascites. J Vasc Interv Radiol 2001;2:373–5.
[80] Savin MA, Kirsch MJ, Romano WJ, et al. Peritoneal ports for treatment of intractable as-
cites. J Vasc Interv Radiol 2005;16:363–8.
[81] Kern KA, Norton JA. Cancer cachexia. J Parenter Enteral Nutr 1988;12(3):286–98.
[82] MacDonald N, Easson A, Mazurak V, et al. Understanding and managing cancer cachexia.
J Am Coll Surg 2003;197(1):143–61.
[83] Wigmore SJ, Barber MD, Ross JA, et al. Effect of oral eicosapentaenoic acid on weight loss
in patients with pancreatic cancer. Nutr Cancer 2000;36:177–84.
[84] Gogos CA, Ginopoulos P, Salsa B, et al. Dietary omega-3 polyunsaturated fatty acids plus
vitamin E restore immunodeficiency and prolong survival for severely ill patients with gen-
eralized malignancy: a randomized control trial. Cancer 1998;82:395–402.
[85] McCann RM, Hall WJ, Groth-Juncker A. Comfort care for terminally ill patients. The ap-
propriate use of nutrition and hydration. JAMA 1994;272:1263–6.
[86] Davis MP, Furste A. Glycopyrrolate: a useful drug in the palliation of mechanical bowel ob-
struction. J Pain Symptom Manage 1999;18(3):153–4.
[87] Ripamonti C, Bianchi M. The use of methadone for cancer pain. Hematol Oncol Clin N Am
2002;16:543–55.