Introduction
Constipation is a common problem among persons receiving narcotics
for treatment of pain. Although it occurs very commonly, it is often underassessed
and undertreated by both physicians and nurses, leaving patients to grapple
with this problem alone.1-5 Because patients often do not understand
the mechanism that is causing the constipation, their self-treatment may be
less than effective. Narcotic-induced constipation is a problem caused by the
treatment provided by health care professionals and thus is a problem that should
be addressed by them.
Prevalence
Limited studies of prevalence of constipation have been conducted
on healthy persons or on persons with cancer. Prevalence in the general population
is believed to be approximately 2%. Among the elderly in long-term care facilities,
prevalence of constipation ranges between 40% and 50%. Although overall prevalence
among persons with cancer is unknown, some studies investigated patients receiving
hospice care and active curative treatment. Studies of hospitalized patients
receiving cancer treatment ranged from 70% to 100%.2-4 Studies of
patients with advanced cancer who are receiving hospice care show prevalence
of constipation to range between 23% and 84%.2,3,5-9 Walsh6
found constipation to be the most common side effect of morphine among 688 hospice
patients with a prevalence of 48%.
Defining Constipation
A significant problem in assessing constipation has been the varying
definitions used by patients and health care providers. Some patients believe
they are constipated if they do not have a bowel movement every day. Other patients
expect to have only three bowel movements per week, so if they have three, they
believe they are not constipated — and in both cases these patients may (or
may not) be correct. One of the largest studies of bowel function ever conducted
was reported in 1965.10 Of the 1,055 persons included in the study,
99% reported bowel movements in the range of three per day to three per week.
Any number outside of this range was considered by these researchers to be abnormal.
However, with such a wide range of normal, it is easy to see how misleading
it would be to characterize a person as constipated solely on the basis of the
number of bowel movements.
Others have attempted to define constipation by looking at more
than just frequency of bowel movements. Table 1 lists 22 defining characteristics
of constipation resulting from a study of 300 healthy adults by McShane and
McLane in 1985.11 The Constipation Assessment Scale, first published
in 1989, presents eight characteristic symptoms and allows the patient to self-report
on a valid and reliable scale.4 These symptoms are also presented
in Table 1.
| Table
1. — Defining Characteristics of Constipation From Two Prospective Studies |
| McShane
and McLane (1985)11 |
McMillan
and Williams (1989)4 |
| Abdominal distention |
Inability to pass stool |
Abdominal distention/bloating |
| Abdominal growling |
Increased abdominal pressure |
Change in gas passed |
| Abdominal mass |
Indigestion |
Inability to pass stool |
| Abdominal pain |
Oozing liquid stool |
Less frequent bowel movements |
| Blood with stool |
Rectal fullness |
Oozing liquid stool |
| Change in abdominal size |
Rectal mass |
Rectal fullness or pressure |
| Change in flatus |
Rectal pain with stool |
Rectal pain at bowel movement |
| Change in frequency |
Rectal pressure |
Small volume of stool |
| Decreased appetite |
Small volume of stool |
|
| Dry, hard stool |
Straining at stool |
|
| Headache |
Swollen rectal veins |
|
An acceptable definition of constipation for clinical practice
would be the following: a decrease in the frequency of passage of formed
stools and characterized by stools that are hard and difficult to pass.
This definition suggests four characteristics that may be presented
in the acronym DISH: difficult to pass, infrequent compared to normal, smaller
than normal, and hard.
Etiologies
Although narcotics are a common cause of constipation in persons
with cancer, there are many other potential causes as well. Constipation may
be classified into one of three types: primary, secondary, or iatrogenically
induced (Table 2).12 Narcotic-induced constipation falls under the
third category.
| Table 2. —
Etiologies of Constipation |
Primary
|
Secondary |
Iatrogenically Induced |
| Due to: |
Due to: |
Due to drugs such as: |
Lack of fiber, exercise,
fluid, or time and privacy
for defecating |
Parkinson’s disease* |
Antacids |
| Intestinal obstruction |
Anticholinergics |
| Volvulus |
Antidepressants |
| Adhesions |
Antihistamines |
| Stroke |
Barium sulfate |
| Hypercalcemia |
Calcium channel blockers |
| Hypokalemia |
Drugs for Parkinsonism* |
| Diverticulosis |
Ganglionic blockers |
| Spinal cord compression |
Hypotensives |
| Rectocele |
Iron supplements |
| |
Monoamine oxidase inhibitors |
| |
Opiates |
| |
Psychotherapeutic drugs |
| |
Vinca alkaloids** |
|
* Controversy remains about whether constipation is caused by drugs
for Parkinsonism, the disease itself, or both.
** In one study,31 100% of patients receiving vinblastine,
a vinca alkaloid, became constipated and one had paralytic ileus.
|
Primary or simple constipation results from extrinsic factors
that affect bowel function, such as inadequate dietary fiber, inadequate fluid
intake, decreased physical activity, and inadequate time or privacy for defecation.12
Any or all of these factors may lead to decreased bowel motility and increased
transit time for fecal material, which causes fluid to be reabsorbed and stools
to be dry and hard. Patients with advanced cancer are already at risk for many
of these factors and therefore are candidates for problems with constipation
even before narcotics are administered.
Secondary constipation results from pathologic changes.
Examples that may be seen in persons with cancer include partial intestinal
obstruction, spinal cord compression at a level that affects bowel function,
and metabolic effects of hypercalcemia and hypokalemia.12 Of course,
patients with cancer may have other problems as well; concomitant conditions
that may occur in persons with cancer and contribute to constipation are listed
in Table 2.13
Iatrogenically induced constipation is caused by administration
of pharmacologic agents. Although narcotics are chief offenders in the cancer
setting, a host of other drugs also may be implicated.12 Examples
of these are listed in Table 2.
Effects of Opiate Narcotics
Opiates affect bowel function in a variety of ways. First, opiates
delay gastric emptying by producing gastroparesis secondary to spasm in the
antropyloric region. This action appears to stem from the central nervous system
and be dopamine mediated. Opiates also delay stool transit through the small
bowel, thus increasing transit time.14 This effect is greatest in
the jejunum and is related to an increase in nonpropulsive contractions. Colonic
transit time also is increased. The basis for this physiologic response to opiates
was studied in the 1940s, before the days of the modern institutional review
board and protection of human subjects. Researchers administered varying doses
of morphine and hydromorphone to healthy male volunteers and studied the responses
of the ileum and colon. The drugs significantly increased the amount of time
needed for fecal material to move through by interfering with normal tone and
contractility. While segmental motor tone and contractility were increased,
longitudinal propulsive peristalsis was decreased. Churning motions
occurred in some cases but without adequate propulsive activity. The change
in peristalsis began to occur 5 to 25 minutes after the administration of the
opiate, depending on the amount of the dose.15
Opiates also may cause the patient to experience a decreased urge
to defecate. The constipating effect of morphine was confirmed in 1989 by a
group of patients receiving at least 30 mg of morphine per day; evaluation showed
that 100% of these patients were constipated.4 In summary, opiates
decrease peristalsis and increase stool transit time, drying out the stool and
causing constipation. The effect is immediate and is dose related. Treatment
of narcotic-induced constipation should be based on an understanding of this
physiologic response to opiates.
Assessing Constipation
Since managing constipation requires a complete understanding
of the origins of the problem, complete assessment is critical. However, most
patients are attuned to self-management of constipation and may not think to
mention it to a health care provider. Thus, patients at risk must be asked specifically
about their bowel function. If a patient is receiving opiate analgesics, he
or she will have a problem with constipation. However, there may be other complicating
factors that need attention.16
History
On the patient’s history, the essential information to obtain
includes: (1) daily activity level, including participation in regular exercise,
(2) food preferences and dislikes, (3) amount of fiber and fluid intake, (4)
ability to chew and swallow, (5) current medications, including over-the-counter
drugs (Table 2), (6) underlying medical problems (Table 2), and (7) recent changes
in stool or bowel habits, eg, pain with stool, change in pattern or size of
stool, diarrhea alternating with constipation accompanied by abdominal pain
or discomfort, bright red blood or black tarry appearance of stools (ask for
time of last bowel movement). In addition, a change in mental status with confusion
and increased agitation, elevated temperature, incontinence,
and unexplained falls may sometimes be the only presenting symptoms
of constipation in the elderly; therefore, maintain a high index of suspicion.
Physical Examination
A complete physical examination is essential when constipation
is suspected. This examination must include oral, abdominal, and anorectal examinations.16
Oral Examination: Dentition should be checked to
ensure adequacy of chewing, and the oral cavity should be examined for lesions
or tumors that could interfere with chewing, tasting, or swallowing. Not uncommonly,
ill-fitting dentures make chewing so difficult or uncomfortable that the elderly
patient takes the path of least resistance and eats only soft foods that require
no chewing, such as custards, but also contain no fiber and thus contribute
to primary constipation.
Abdominal Examination:
The patient should empty the bladder before the examination begins. Inspection:
Begin by looking for bloating, distention, or bulges. When distention is noted,
consider the possibility of obesity, fluid, flatus, or tumor. If these are ruled
out, then the possibility of feces must be considered. Auscultation:
Listen for hyperactive bowel sounds or absent bowel sounds. If no bowel sounds
are readily heard, listen continuously for five minutes before establishing
the absence of bowel sounds. Absent bowel sounds may indicate paralytic ileus,
whereas increased bowel sounds may indicate impending diarrhea. Percussion:
On percussion, gas in the colon produces tympany. Dullness is heard over intestinal
fluid and feces. Palpation: Light palpation can detect abdominal tenderness
and muscular resistance associated with chronic constipation. When abdominal
pain occurs on coughing or with light palpation and/or when rebound tenderness
is detected, peritoneal inflammation must be suspected. Deep palpation may be
used to detect abdominal masses including a sausage-like mass of fecal material
in the left colon. A feces-filled colon is suggestive of constipation.
Anorectal Examination: A digital rectal examination
may reveal a stool-filled rectum, suggesting constipation. Endoscopic examination
of the anus and rectum can reveal not only an impacted fecal mass, but also
fissures, hemorrhoids, ulceration, or rectal ulcers. Look also for the presence
of blood or for alterations of the vascular pattern that might indicate ischemic
proctitis.
Follow-up Evaluation
A complete history and physical examination conducted by a physician,
nurse practitioner, or physician’s assistant is necessary for the initial diagnosis
of constipation. However, ongoing assessment also is necessary to monitor the
effect of treatment. For this purpose, the Constipation Assessment Scale (CAS)
was developed. It is an easy-to-read, eight-item, self-report tool that is clinically
useful in a variety of settings.2-4,17 The CAS is valid and reliable,
and it takes approximately two minutes to complete.4 The eight items
focus on the symptoms of constipation that are most universal (Table 1). The
patient is asked to rate each symptom as "no problem" (0), "some
problem" (1), or "severe problem" (2). Item scores are added
together for a total that may range from 0 (no constipation) to 16 (worst possible
constipation). Readability is at approximately the sixth-grade level, so most
patients can read and understand the CAS. A series of these CAS assessments
completed over a period of days or weeks provides an ongoing standardized assessment
of the cancer patient’s constipation.
Preventing Constipation
Preventing constipation should be related to the patient’s risk
factors. Assessment of these risk factors should place the patient’s risk into
one of the three categories described earlier: primary, secondary, or iatrogenically
induced constipation.
Primary Constipation
If the patient is at risk for primary constipation because of
eating habits or lack of exercise or other lifestyle factors, it is useful to
attempt lifestyle changes. However, in excessively debilitated or anorectic
cancer patients, increasing fiber and exercise may not be feasible alternatives.
For patients who are well enough to cooperate, prevention strategies for primary
constipation include the following18,19:
· Increased dietary fiber — Fiber holds water in the stool,
making it heavier and faster. It also may increase flatus.
· Increased fluid intake — Water has to be available to be
absorbed. The thirst mechanism may be diminished in elderly patients, and patients
experiencing nausea may be unwilling to drink. However, they must keep up fluid
intake because fiber without fluid can become hard and difficult to pass.
· Adequate exercise — The movement of the body helps the bowels
to be mobile. Inactivity helps the bowels to stay inactive.
· Adequate time and privacy for toileting — Some patients are
just unable to defecate in a public bathroom and will wait until they get home.
Likewise, hospitalized patients and residents in nursing homes who need help
to toilet may find it difficult to defecate. Holding the stool in the large
bowel allows the water to be reabsorbed, which causes the stool to be hard and
difficult to pass. Patients should be taught to plan a time for good bowel hygiene.
Karam and Nies20 reported use of a bowel management
regimen for the elderly that included all of the required elements: fluids,
exercises, fiber, and a regular toileting time. The program was implemented
at three levels — mild, moderate, and aggressive.
The mild
program included a minimum of 1500 mL of fluid per day, abdominal and pelvic
exercises, 1 oz of fiber supplement daily (Table 3), 5 to 15 minutes on the
commode after each meal, and ambulation of at least 50 feet twice per day. These
same elements were included in the moderate and aggressive programs,
but the amount of fiber supplement was increased to 2 oz and 3 oz per day, respectively.
As a result of this protocol, laxative use was markedly decreased and spontaneous
bowel movements significantly increased. This appears to be an effective prevention
protocol in otherwise healthy elderly adults.
Table
3. — Recipe for
Fiber Supplement |
|
Ingredient*
|
Amount
|
|
All-bran cerea
|
2 cups
|
|
Applesauce
|
2 cups
|
| Prune
juice |
1
cup |
|
* Mix well and refrigerate.
|
Secondary Constipation
Secondary constipation may not be
as amenable to prevention as primary constipation. Because it is caused by concurrent
diseases, there may be little available in the way of prevention.21
The prevention protocol recommended for primary prevention may be effective
in some cases, but in others, it may be necessary to treat the underlying disease
or begin a constipation treatment program.
Iatrogenically Induced Constipation
Unfortunately, much of the treatment
given to patients can cause constipation, and polypharmacy is a particular problem
among elderly patients.22 For example, one specialist may prescribe
a calcium channel blocker and another may prescribe an antidepressant. Then
cancer treatments are added, and another specialist may prescribe one of the
vinca alkaloids and an opiate for pain. As drugs are added, the risk of constipation
mounts.
When patients come into a cancer setting,
the nurse may make a list of current medications, but how is that list used?
It is important when taking the drug history to look for drugs that may cause
constipation. If a patient reports taking one or more of these drugs, a careful
assessment for constipation must be conducted. In a few rare cases, it may be
possible to substitute a less constipating drug. However, this will not be possible
in most cases, and ongoing management of constipation will become necessary.
While it may be tempting to withhold opiates to avoid constipation, this leads
to impaired quality of life; a better approach is to give the narcotic analgesic
to manage the pain and then treat the resulting constipation appropriately.
Managing Narcotic-Induced Constipation
The treatment of narcotic-induced constipation should include,
to whatever degree possible, the use of prevention strategies to avoid primary
constipation. However, this alone will not be sufficient for patients receiving
opiate narcotics. Other stra
tegies involve the use of various pharmacologic agents, given
both orally and rectally, and enemas. The choice of strategy depends on the
underlying cause of the constipation and its severity.
Several types of pharmacologic agents are available for treating
constipation, including osmotic laxatives, emollient or lubricant cathartics,
bulk cathartics, and stimulant cathartics.13,16
Osmotic Laxatives
Two commonly used osmotic laxatives are the disaccharides and
the saline cathartics. Disaccharides exert an osmotic effect because they are
not absorbed or metabolized in the small bowel. This increases bulk via fermentation
in the bowel. Unfortunately, the disaccharides may have the side effect of cramps,
abdominal distention, and flatulence, which can be uncomfortable for the patient.
In addition, disaccharides are usually in a liquid form and may not be palatable
to patients. Examples are lactulose and sorbitol.13,16,23 Saline
cathartics employ osmotic forces to pull fluid into the bowel to increase the
weight of the stool and soften it. An example is Milk of Magnesia.13,16
Emollient or Lubricant Cathartics
Emollient or lubricant cathartics soften stools. An example of
this is docusate sodium (marketed as Surfak).13-16 Some older patients
may have developed a habit of taking mineral oil as a lubricant. Because of
its tendency to wash out fat-soluble vitamins, mineral oil may not be the best
choice. If the patient persists in using it, he or she should be instructed
to avoid taking it at or near mealtimes. Glycerin suppositories also are a type
of lubricant.
Bulk Cathartics
Bulk cathartics increase mass and soften
stools. Several over-the-counter bulk cathartics are available. An example is
psyllium,18 marketed as Metamucil.
Stimulant Cathartics
Stimulant cathartics promote intestinal motility, which is the
action that opiates suppress. These are available for both oral and rectal administration.
Examples include the senna-based products such as Senokot and the stimulant
bisacodyl marketed as Dulcolax. These can cause some cramping and may be unacceptable
to patients for that reason.13,16 This effect may be diminished by
spreading the dose out over the day, perhaps giving small doses with each meal
and a slightly larger dose at bedtime. In addition, it should be noted that
the cause of the narcotic-induced constipation, the analgesic, also should decrease
the impact of cramping in this population of patients, thereby making it somewhat
less problematic.
A comparative study of the efficacy of lactulose and senna was
conducted using terminal cancer patients (n=91).24 Both laxatives
were found to be equally effective in treating narcotic-induced constipation,
but senna was recommended because of its lower cost compared with lactulose.
Prostaglandins/Prokinetic Drugs
Prostaglandins work by changing water and electrolyte absorption
in the intestines. It also is believed that the laxative effect may be due to
the motor effect of prostaglandins, but this is not yet clear. Researchers found
that misoprostol, marketed as Cytotec, increased the weight and frequency of
stools and shortened colonic transit time in patients with severe chronic constipation.25
Cisapride, an agent that stimulates the upper gastrointestinal tract, is useful
in patients with spinal cord injury or Parkinson’s disease.21 Colchicine,
used to treat gouty arthritis, is useful in treatment of chronic constipation.26
No studies on the effectiveness of these drugs for narcotic-induced constipation
have been reported.
Naloxone
Orally administered naloxone is an opioid-antagonist that has
the ability to antagonize the gastrointestinal effects of opioids. For this
reason, it has been studied as a treatment for opioid-induced constipation.
In low and slowly escalating doses, naloxone has been shown to have a laxative
effect. However, adverse reactions including withdrawal and return of pain have
been demonstrated at varying dose levels. Thus naloxone is not recommended for
routine use and its use should be avoided in outpatients.27-30
Combination Agents
Some of these pharmacologic agents are available in combinations.
For example, senna, a stimulant, is combined with docusate sodium, a stool softener,
and sold as Senokot-S. This particular combination is recommended for narcotic-induced
constipation
because it helps to overcome the specific problems caused by narcotics.
Because opiates are known to decrease longitudinal propulsive contractions that
help to move stool downward and because they also allow stool to remain in the
colon where it dries out and becomes hard, this combination seems ideal. The
mechanism of action of senna is stimulation of longitudinal peristalsis (the
action that opiates interfere with), and docusate sodium helps to soften stools
that have become dry and hard.16
The effect of opiates on bowel function is so consistent that
some experts recommend administering senna and docusate sodium on a scheduled
basis rather than when necessary.30 An individualized program that
includes careful follow-up evaluation should be developed for each patient.
Enemas
Although both nurses and patients dislike them, enemas sometimes
are unavoidable. If the patient is impacted, an enema may be the only way to
break up hardened fecal matter and wash it out. However, once the impaction
has been removed, enemas should be used sparingly. They tend to wash out the
normal mucus in the colon that provides lubrication for stools. Therefore, enemas
should be considered an extreme measure rather than part of routine treatment.13,16
Conclusions
Constipation is common in patients with cancer because of their
many risk factors, and in a cancer patient receiving opiates, constipation is
almost guaranteed. Unfortunately, this potentially serious problem is often
overlooked and undermanaged. Although constipation may seem like a minor complication
when compared to a life-threatening disease such as cancer, it can become a
major detriment to quality of life if it is not well managed.
References
1. Bruera E, Suarez-Almazor M,
Velasco A, et al. The assessment of constipation in terminal cancer patients
admitted to a palliative care unit: a retrospective review. J Pain Symptom
Manage. 1994;9:515-519.
2. McMillan SC, Tittle M. A descriptive study of the management
of pain and pain-related side effects in a cancer center and a hospice. Hosp
J. 1995;10:89-108.
3. Tittle M, McMillan SC. Pain and pain-related side effects
in an ICU and on a surgical unit: nurses’ management. Am J Crit Care.
1994;3:25-30.
4. McMillan SC, Williams FA. Validity and reliability of the
Constipation Assessment Scale. Cancer Nurs. 1989;12:183-188.
5. Addington-Hall JM, MacDonald LD, Anderson HR, et al. Dying
from cancer: the views of the bereaved family and friends about the experiences
of terminally ill patients. Palliat Med. 1991;5:207-214.
6. Walsh TD. Oral morphine in chronic cancer pain. Pain.
1984;18:1-11.
7. Vainio A, Auvinen A. Prevalence of symptoms among patients
with advanced cancer: an international collaborative study. J Pain Symptom
Manage. 1996;12:3-10.
8. Donnelly S, Walsh D, Rybicki L. The symptoms of advanced
cancer: identification of clinical and research priorities by assessment of
prevalence and severity. J Palliat Care. 1995;11:27-32.
9. Weitzner MA, Moody LN, McMillan SC. Symptom management issues
in hospice care. Am J Hosp Palliat Care. 1997;14:190-195.
10. Connell AM, Hilton C, Irvin EG, et al. Variation in bowel
habits in two population samples. Br Med J. 1965;2:1095-1099.
11. McShane RE, McLane AM. Constipation: consensual and empirical
validation. Nurs Clin North Am. 1985;20:801-808.
12. Cimprich B. Symptom management: constipation. Cancer
Nurs. 1985;8:39-43.
13. Snapes WJ. Therapy in patients with constipation. Home
HealthCare Consult. 1996;3:4-17.
14. Maguire LC, Yon JL, Miller E. Prevention of narcotic-induced
constipation. N Engl J Med. 1981;305:1651.
15. Adler HF, Atkinson AJ, Ivy AC. Effect of morphine and dilaudid
on the ileum and of morphine, dilaudid, and atropine on the colon of man. Arch
Intern Med. 1942; 69:974-985.
16. Fallon M, O’Neill B. ABC of palliative care. Constipation
and diarrhoea. Br Med J. 1997;315:1293-1296.
17. Broussard BS. The constipation assessment scale for pregnancy.
J Obstet Gynecol Neonatal Nurs. 1998;27:297-301.
18. Cheskin LJ, Kamal N, Crowell MD, et al. Mechanisms of constipation
in older persons and effects of fiber compared with placebo. J Am Geriatr
Soc. 1995;43:666-669.
19. Hanlon M. Patients with constipation requiring a bowel
management program. J Wound Ostomy Continence Nurs. 1997;24:325-329.
20. Karam SE, Nies DM. Student/staff collaboration: a pilot
bowel management program. J Gerontol Nurs. 1994;20:32-40.
21. Geders JM, Gaing A, Bauman WA, et al. The effect of cisapride
on segmental colonic transit time in patients with spinal cord injury. Am
J Gastroenterol. 1995;90: 285-289.
22. Corcoran MB. Polypharmacy in the older patient with cancer.
Cancer Control. 1997;4:419-428.
23. Kleessen B, Sykura B, Zunft HJ, et al. Effects of inulin
and lactose on fecal microflora, microbial activity, and bowel habit in elderly
constipated persons. Am J Clin Nutr. 1997;65:1397-1402.
24. Agra Y, Sacristan A, Gonzalez M, et al. Efficacy of senna
versus lactulose in terminal cancer patients treated with opioids. J Pain
Symptom Manage. 1998;15:1-7.
25. Soffer EE, Metcalf A, Launspach J. Misoprostol is effective
treatment for patients with severe chronic constipation. Dig Dis Sci.
1994;39:929-933.
26. Verne GN, Eaker EY, Davis RH, et al. Colchicine is an effective
treatment for patients with chronic constipation: an open-label trial. Dig
Dis Sci. 1997;42:1959-1963.
27. Culpepper-Morgan JA, Inturrisi CE, Portenoy RK, et al.
Treatment of opioid-induced constipation with oral naloxone: a pilot study.
Clin Pharmacol Ther. 1992; 52:90-95.
28. Yuan CS, Foss JF, Osinski J, et al. The safety and efficacy
of oral methylnaltrexone in preventing morphine-induced delay in oral-cecal
transit time. Clin Pharmacol Ther. 1997;61:467-475.
29. Sykes NP. An investigation of the ability of oral naloxone
to correct opioid-related constipation in patients with advanced cancer. Palliat
Med. 1996;10:135-144.
30. Walsh D. Commentary on research in cancer pain and palliative
care: assessment and treatment of opioid-induced constipation. Cancer Pain
Release. 1998;11:6-7.
31. Kavanagh JJ, Copeland LJ, Gershenson DM, et al. Continuous-infusion
vinblastine in refractory carcinoma of the cervix: a phase II trial. Gynecol
Oncol. 1985;21:211-214.