Malignant Bowel Obstruction in a Patient Who Wants to Go Home
Ms. K is 61 with recurrent high-grade serous ovarian cancer, peritoneal carcinomatosis, and ascites. She has had two lines of chemotherapy and is not a surgical candidate. She arrives with three days of vomiting (about 1.5 L daily), colicky abdominal pain, no flatus or stool, and a tense, distended abdomen. CT shows small bowel obstruction with two transition points, no perforation, and no site amenable to stenting. She is fully alert, has a life expectancy of weeks, and says her goal is to be at home and "not throwing up." Her medications are hydromorphone 2 mg every 4 hours as needed, ondansetron 8 mg every 8 hours, metoclopramide 10 mg every 6 hours as needed for nausea, and dexamethasone 4 mg daily. A nasogastric tube drained 1.2 L on insertion.
1. Which of her four medications would you stop, continue, or change today, and why? What would you add?
Read the worked answer
Stop metoclopramide. Prokinetic agents are ineffective and potentially harmful in complete obstruction - they increase colic and can worsen pain and vomiting.
Stop ondansetron. 5-HT3 antagonists are better suited to chemotherapy- and vagally mediated nausea; in obstruction they add little and cost a subcutaneous site. Simplify.
Convert hydromorphone to a scheduled parenteral (subcutaneous) dose with a breakthrough dose rather than as-needed only. Oral absorption is unreliable with obstruction; colicky pain may also respond to an antispasmodic such as hyoscine butylbromide, which has the added benefit of reducing secretions.
Increase dexamethasone to 6–16 mg daily (8 mg is a common starting point), given subcutaneously or IV. It is antiemetic and anti-inflammatory and can partially resolve obstruction in a minority of patients; reassess in 3–5 days and use the lowest effective dose, watching glucose.
Add octreotide 100–300 mcg subcutaneously every 8 hours, or a continuous 20–60 mcg/hour infusion. It reduces gastrointestinal secretions and vomiting and works even when an NG tube is not in place; allow 24–72 hours for full effect.
Add a second antiemetic with a different mechanism such as haloperidol 0.5–1 mg subcutaneously every 6–8 hours, or olanzapine 2.5–5 mg, which also calms agitation.
Hold stimulant laxatives in complete obstruction. Treat mouth care aggressively; limit IV fluids because they increase secretions and vomiting.
2. She asks whether she can go home. What is your discharge plan, and what conversations must occur first?
Read the worked answer
Yes, home is feasible, but not without infrastructure.
Decide the decompression strategy. An NG tube at home is burdensome and a common reason for readmission. Options: keep the NG while waiting 48–72 hours to see if octreotide and dexamethasone work; place a venting percutaneous gastrostomy for longer-term drainage; or, if vomiting comes under control, remove it entirely.
Set up a continuous subcutaneous infusion delivering octreotide, the opioid, and an antiemetic, with a reliable pump and a plan for who refills it at 2 a.m.
Confirm caregiver capacity honestly: who is at home, can they manage the pump and the tube, do they have transportation, and is there 24-hour hospice coverage with an on-call nurse who can come.
Send anticipatory prescriptions for pain crisis, agitation, secretions, and vomiting, plus a written emergency plan that says whom to call rather than 911 by default.
Conversations before discharge:
- Restate her goals and confirm that comfort at home is the priority.
- Tell her plainly what can go wrong and what you would do about it.
- Name the triggers for returning to an inpatient hospice bed, so the decision is pre-made rather than made in a crisis.
- Include her brother, and ask what he is most worried about.
3. Her brother asks whether stopping a feeding tube and IV fluids means she will "starve to death." How do you answer?
Read the worked answer
Acknowledge the fear first - it comes from love, and it deserves a direct answer rather than a deflection.
Then reframe what is happening. With a complete obstruction, the intestine cannot move contents forward; adding formula or fluid increases vomiting, abdominal distension, secretions, and aspiration risk. In this setting artificial nutrition has not been shown to prolong life or improve comfort, and it often worsens the symptoms she most wants relieved.
Separate hunger from thirst. Hunger is uncommon at end of life and is relieved by small tastes of what she enjoys, if she tolerates them. Thirst is real but is relieved by ice chips, sips, lip balm, and frequent mouth care - not by a liter of intravenous fluid.
Explain the mechanism honestly: she is dying of ovarian cancer, not of hunger. The body's gradual shutdown is part of the disease trajectory, and starvation is not the cause of death.
Address fluids individually rather than as a package: a modest trial of hypodermoclysis may be reasonable if she has severe thirst, hypercalcemia, or opioid toxicity with rising creatinine - but not as a default, and not if it worsens her vomiting.
Close the loop: ask what he is afraid will happen, and what his sister would want him to know. Offer to include him in the next goals-of-care discussion so he is a participant rather than a spectator.