book cover

Palliative Care Perspectives : Chapter 5: Non-Pain Symptom Management : Bowel Obstruction

This material is based on a presentation to the American Academy of Hospice and Palliative Medicine with copresenter Richard Meyer, M.D., June 1999.

Excellent palliation of bowel obstruction requires experience and skill. Bowel obstruction is one of the most difficult processes to palliate, yet recent advances in understanding and new therapeutic approaches allow us to do a much better job than was possible only a few years ago.30 That problems like bowel obstruction exist is a compelling reason for why we need palliative care specialists and for why home hospice is not always enough. I have found it very difficult to get bowel obstruction under control when initiating therapy in the home. However, when I can adjust medications on an inpatient unit, I often get people on a regimen that can be administered in home hospice, as happened in the case of Mr. A mentioned at the beginning of this chapter. For those clinicians who believe that palliative care is nothing more than good intentions and a morphine infusion, learning about bowel obstruction can be an eye-opener. Competency requires an understanding of physiology and pharmacology as well as skills in assessment and communication.

Here, I address only the care of malignant bowel obstruction, as this is the type that most commonly arises in palliative care. To what extent this discussion may be applicable to other types of bowel obstruction is unclear. I know of no papers that address the application of the therapies presented here outside of malignant bowel obstruction.


Bowel obstruction arises most commonly as a complication of ovarian or colon cancer. In one series 42% of patients with ovarian cancer developed obstruction.31 Primary presentation of colon cancer with obstruction is usually treated with surgical resection. The incidence of obstruction in colorectal cancer as a late complication is approximately 10%.32 Other tumors that may give rise to bowel obstruction are gastric, pancreatic, cervical, bladder, endometrial, mesothelial (of peritoneum), carcinoma, and melanoma.

Pathophysiology of Bowel Obstruction

The goal of therapy is to improve quality of life by compensating for the complex and distressing physiologic changes that arise secondary to the obstructive process. This goal is best achieved by normalizing gut function proximal to the obstruction.

The intestinal tract did not evolve to compensate for mechanical bowel obstruction caused by an immobile, fixed lesion such as cancer. The physiologic changes that arise with obstruction would be adaptive to reversible forms of bowel obstruction that may have occurred for our ancient ancestors, but they are maladaptive for patients with cancer. Similarly, kidneys demonstrate a maladaptive response to heart failure. Decreased renal perfusion is sensed as dehydration. Fluid is retained to compensate when, in fact, the patient is drowning. We adjust for this by "overruling" the kidneys, telling them to get rid of salt and water and not hold on to them. The kidneys can be mistaken. What "misunderstanding" arises in bowel obstruction?

Imagine that you have been very hungry. Your tribe finally hunts down a mastodon, and it is time for a feast. You gorge yourself, eating great chunks of meat and causing a temporary obstruction. Your body would respond in the following way.

  1. Mechanoreceptors and chemoreceptors would be stimulated by the distention caused by the large build-up of food proximal to the blockage. These receptors would tell your brain to stop eating.
  2. The intestine proximal to the blockage would begin hypersecreting fluid, trying to flood the system and wash the intestinal contents downstream.
  3. Intestinal motility would increase, further trying to push contents downstream and causing cramping.

With luck, you would live to hunt another day. While this approach works well for ingested mastodons, it works poorly for malignant bowel obstruction.

A delicate balance of fluid absorption and secretion from and into the lumen is normally maintained. Studies have demonstrated that with obstruction the balance is shifted strongly in favor of secretion.33 Increased secretion of fluid results in further intestinal dilatation, cramping, and frank nausea and vomiting. In one small study Huchison measured serotonin levels in women with ovarian cancer with and without bowel obstruction. High levels of urinary 5HIAA, a serotonin metabolite, were found only in obstructed women.34 This pattern was very similar to the pattern of serotonin elevation found in patients treated with cisplatin. In cisplatin therapy enterochromaffin cells in the intestine secrete serotonin, which binds 5HT3 receptors in the gut, thereby stimulating nausea.35 It is therefore quite possible, but not yet proven, that a similar process occurs in bowel obstruction.

A vicious cycle is entered wherein hypersecretion (associated with cramping in the early stage) is followed by dilatation and vomiting, followed by further secretion and vomiting. Dehydration and electrolyte disturbances quickly result, leading to death (and misery) if an intervention is not made (Fig. 5.2).

Graphic showing vicious cycle of bowel obstruction.
Figure 5.2. Pathophysiology of bowel obstruction - a vicious cycle.

Traditional Medical Approaches

Traditional "conservative" management, labeled "drip and suck" therapy by some authors, is an extrapolation from perioperative management for obstruction.36,37 It certainly makes sense if operating on a vomiting patient with a dilated gut to decompress the gut with an NG tube and restore intravascular volume. However, there is no data that supports this approach as a long-term therapy for malignant bowel obstruction. Indeed, multiple studies have shown dismal outcomes with this approach alone.38-41 Theoretically, IV hydration, in addition to restoring intravascular volume, also increases hydrostatic pressure in the villi and therefore could increase secretion into the lumen, contributing to the vicious cycle. Mr. A. was initially treated with a traditional drip and suck approach, which still seems to be the dominant approach used in medical care.

Early Palliative Approaches

Early palliative approaches stressed symptomatic relief. It was assumed that the gut was nonfunctional, and therefore no attempt was made to normalize function. Indeed, symptomatic relief, in effect, put the gut to sleep. Anticholinergic drugs, such as hyoscine, both decreased secretion into the gut and decreased motility, thereby alleviating cramping.32,42 Opioids were also stressed, both to reduce motility and treat pain directly. These approaches are still used when normalization of gut function is impossible, as it often is in very proximal gut obstruction.

Steroids have been used in the hope of relieving obstruction by reducing swelling around obstructing growths, although their efficacy in this regard is debatable. Bowel obstruction is a very dynamic process, frequently reverting from total to partial obstruction and back in as many as 50% of cases. Only one controlled study of the use of steroids in bowel obstruction has been done. It showed no evidence that steroids were helpful in reducing the degree of obstruction. A major problem in this study was the very high rate of spontaneous conversion from total to partial obstruction.43 Steroids may nevertheless be useful in bowel obstruction by decreasing bowel and peritoneal inflammation and by acting as appetite stimulants.

Recent Approaches

Recent approaches have tried to normalize gut function to the extent possible in addition to palliating symptoms directly. In practice, the ability to normalize and use the proximal gut is highly dependent on the level of obstruction. Many cases of malignant obstruction that are appropriate for nonsurgical, palliative approaches have multiple sites of obstruction, most frequently in the jejunum or ileum.44 It is not uncommon to have many feet of potentially functional intestine proximal to the rate-limiting site of obstruction. Very proximal obstructions prohibits normalization. However, very proximal and very distal obstructions may be amenable to stent placement that results in significant palliation by forcing open the gut lumen using an expandable wire mesh stent.45,46


The most important drug in the therapy of bowel obstruction is octreotide. An analogue of the hormone somatostatin, it significantly reduces secretion into the gut. In one study by Mangili, 13 patients with ovarian cancer-related obstruction had NG aspirate volumes measured. Mean drainage decreased from 1687 ml/day to < 50 ml/day.47 Octreotide is generally well tolerated. It appears to have minimal effects on motility. Doses range from 100 mcg BID to 200 mcg TID IV or SC.48 Octreotide can result in significant improvements in nausea and vomiting.49 However, because this appears to be due to decreased secretion of fluid into the gut, it usually takes 24 to 48 hours for this effect to become apparent. A recent randomized controlled trial that compared octreotide treatment to scopolamine butylbromide (an anticholinergic agent unavailable in the United States) demonstrated that octreotide was superior in reducing intestinal secretions.50 A new long-acting depo version of octreotide has been developed. I am unfamiliar with any studies that used it in bowel obstruction. In theory, one injection can last up to a month. However, it is very expensive.

Promotility agents

Promotility agents can be used if cramping is not present and if the intention is to normalize and use the proximal gut. Traditionally, clinicians have believed that promotility agents are contraindicated in bowel obstruction because increased motility could worsen cramping and theoretically result in gut perforation. However, reports of the beneficial effects of promotility drugs are beginning to appear in the literature.36,51 There are, to my reading, no data to substantiate the concern about inducing perforation. I am unaware of a single case report that associates bowel perforation or rupture with promotility drug use. Metoclopramide is the drug of choice for this purpose. Metoclopramide works by binding 5HT4 receptors and releasing acetylcholine, which in turn binds cholinergic receptors and results in increased motility. Understanding this is important, as concomitant use of drugs with anticholinergic effects, such as scopolamine, promethazine, or amitriptyline, may antagonize this action and reduce efficacy. Dosing is usually begun at 10 mg TID AC PO and gradually increased. Care should be used with metoclopramide in the presence of renal failure, as it is renally excreted, and in patients with Parkinson's disease because of dopamine receptor blockade. For large bowel obstruction a combination of metoclopramide with a large bowel stimulant, such as senna, will probably have to suffice until new motility agents are identified. Experts in the field have warned that promotility drugs should not be used in complete bowel obstruction, although the evidence base for this seems weak, as mentioned above.30 In practice it is not always easy to distinguish total from partial obstruction. Frequently, obstruction progresses from partial to total and back again, making the advice not to use promotility drugs in total obstruction easier to give in principle than to follow in practice.

Direct antiemetics

If cramping is present or if the intent is to rest the bowel, as with patients no longer capable of eating or drinking, anticholinergic and antihistaminic antiemetics such as promethazine may be used. (See section on nausea.) Glyco-pyrrolate, a more locally acting anticholinergic drug, can be given orally or parenterally. It can reduce cramping, intestinal secretion, and nausea.

If the goal is to normalize gut function, anticholinergic agents should be avoided, because they both inhibit motility and block the use of metoclopramide. Many experts use haloperidol as a first-line agent in part because it does not affect motility. However, a pharmacologic rationale for this is lacking, as it is unclear that dopamine receptors are significantly involved in the pathophysiology of obstructive nausea. I suspect 5HT3 antagonists, such as ondansetron, may be the agents of choice, based on the limited data presented above suggesting 5HT3-mediated nausea and the fact that they have limited effects on motility.52 They are expensive, however, and the evidence base for this practice has not been established.

Other interventions include:

NG/venting gastrostomies

Although NG tube placement is a bad approach, I believe, for long-term management in most cases, it can be very helpful for initial gut decompression. I often keep NG tubes in place during initiation of octreotide. Venting gastrostomies have been used, much like PEG tubes for feeding, as a long-term alternative to NG tubes for decompression. Although certainly less uncomfortable than NG tubes, there is minimal evidence that they result in less nausea or distention. No studies have compared venting gastrostomies to long-term octreotide therapy. A consensus panel of the European Association of Palliative Care recommended that venting gastrostomies be used only if medications fail to control nausea.30


I have given patients who are not acutely vomiting simethicone in order to encourage burping of stomach and intestinal gas. Although not yet documented as efficacious in the literature, in my experience this is nontoxic and simply makes sense. One cause of intestinal dilatation is intestinal gas (largely nitrogen) that cannot be disposed of "the old-fashioned way" via flatus. Studies have shown that the vast majority of gas seen on an abdominal X-ray is swallowed.33 To the extent dilatation directly contributes to the vicious cycle described above, it makes sense to do everything possible to minimize dilatation, including encouraging burping. The usual dose is 80 mg BID to TID.

Low-fiber diet

Most patients with bowel obstruction who are able to eat should be on a low-fiber diet. This is essential if they are trying to eat with a total obstruction (as is, in fact, sometimes possible). Patients with complete obstruction who do eat often vomit or regurgitate every few days. Although this is not pleasant, it appears to be more pleasant than dying with an NG tube.


Opioids are very effective in dealing with the cramping of bowel obstruction and are usually needed for pain management associated with advanced malignant disease. However, they can have undesirable effects on motility if one is trying to normalize gut function. As a general rule, pain management trumps motility management. That is, if a patient needs an opioid for pain, give it. As previously discussed, the fentanyl patch may have a lesser effect on GI motility than do other agents. It is often preferred, as well, because the oral route is generally unreliable in bowel obstruction.

Psychosocial support

So far, the emphasis has been on physiology. Bowel obstruction also requires psychosocial support. Patients with distal obstruction often become distended, which alters body image and can be distressing. While most patients hate NG tubes, they can also become dependent on them and may resist suggestions to discontinue them. This may be because when they were initially placed they did provide relief. Such patients also probably fear possible tube replacement. Tubes although discouraged as long-term therapy, may also represent medical caring, and thus patients and families may view suggestions to discontinue them as potential abandonment. The rationale for discontinuation of any therapy, as discussed elsewhere, must be carefully explained. As discussed later in the section on cachexia and anorexia, the inability to eat or drink normally causes an intense grief reaction in patients and families. Adjusting the diet, usually to low-fiber liquid-based formulations, may allow nurturing to continue even in the presence of complete bowel obstruction.

What Does Success Look Like in the Treatment of Bowel Obstruction?

In my experience most patients with bowel obstruction can have NG tubes and IVs discontinued. In some very proximal cases I have continued both if medications do not adequately control secretions. The symptoms of pain, bloating, nausea, and vomiting should be amendable to palliation in virtually all cases. In the most rewarding cases I have treated, such as Mr. A., patients with total distal bowel obstruction who probably would have died within days with traditional drip and suck therapy have been able to resume eating, improve their functional status, and be discharged to home, living for weeks to a few months post-obstruction. Although the treatment of bowel obstruction is difficult, it can be very rewarding. Controlled trials have not been performed on all the interventions suggested above. I hope they will be done in the near future in hopes of furthering our understanding and expanding our treatment options.

<<< Previous Next >>> [ Go Up ]

book cover

Palliative Care Perspectives

James L. Hallenbeck, M.D.

Order hardback from
Order paperback from

Copyright 2003 by Oxford University Press, Inc.

The online version of this book is used with permission of the publisher and author on web sites affiliated with the Inter-Institutional Collaborating Network on End-of-life Care (IICN), sponsored by Growth House, Inc.