A Husband Who Believes Feeding Her Is the Only Thing Left He Can Do
Her husband is certain that starting IV nutrition is the one thing standing between her and starvation. The disagreement isn't about how much he loves her — it's about whether her body, in the state cachexia has put it in, can actually use what a feeding line would provide.
Eleanor B., 68, and her husband have been married forty-five years, and in the three weeks since her oral intake began dropping in earnest, he has taken to preparing her favorite dishes from scratch each evening regardless of how little she manages to eat, as though the act of cooking itself were something he still had control over. She has advanced ovarian cancer with diffuse peritoneal carcinomatosis, no discrete mechanical obstruction on her most recent CT, and a pattern of early satiety, intermittent nausea, and progressive anorexia consistent with cancer-associated cachexia rather than a blockage a tube or a stent could bypass.
Her oral intake is now estimated at under a quarter of her caloric needs, and her husband has asked directly, more than once, for IV nutrition to be started — convinced, in his words, that she is "being allowed to starve." What he has not yet had explained to him in those terms is that cachexia is not simple starvation: it is a cytokine-driven catabolic process, the same systemic inflammatory signaling behind much of her weight loss, that continues consuming muscle and fat regardless of how many calories are actually provided, which is the physiologic reason parenteral nutrition has not been shown to reverse it or extend survival in patients whose anorexia has this specific cause. ASCO's cancer cachexia guideline (Roeland et al., 2020) recommends against routine tube feeding or parenteral nutrition for precisely the population she belongs to — advanced cancer, cachexia-predominant anorexia, no obstruction to bypass — and she matches every one of those criteria rather than sitting at the edge of them. It carries real risks of its own — catheter infection, fluid overload, and in a patient this catabolic, a genuine risk of refeeding complications if nutrition is reintroduced aggressively. What hasn't yet been tried, separately from that question entirely, is whether her nausea and appetite could still be meaningfully improved through medication alone. That option is narrower than it sounds: the same guideline found no agent worth calling standard of care, and mirtazapine, the drug most often reached for here on reputation, was later tested head-to-head against placebo by Hunter et al. in 2021 and did no better. What remains is a short course of a corticosteroid or a progesterone analog — weak evidence, but the only evidence there is, and nobody has yet offered her either one.
In the family meeting room, explaining why a feeding line isn't a simple yes
I don't think parenteral nutrition is the right recommendation here. Cachexia isn't simple starvation — it's a cytokine-driven catabolic process that keeps consuming muscle and fat regardless of how many calories we provide, which is exactly why artificial nutrition hasn't been shown to reverse it or extend survival in patients whose anorexia has this cause. ASCO's cachexia guideline — Roeland and colleagues, 2020 — recommends against routine parenteral nutrition for this exact population, and she isn't a borderline member of it. It also carries real risks of its own: catheter infection, fluid overload, refeeding complications in someone this catabolic. I want to support oral intake as tolerated and help her husband understand what's actually happening to her body.
I agree with the physiology, and I still don't think an outright refusal is the only honest answer to give her husband tonight. A time-limited trial — a defined number of days, defined markers we'd actually look for, an agreed point where it stops if we don't see them — is a real, recognized middle path in exactly this kind of disagreement. It isn't the same as committing to indefinite TPN, and it gives him something concrete rather than a flat no.
Explaining the physiology once and moving straight to refusal risks him hearing "we've given up," even when that isn't what we mean.
Before we get to whether a feeding line helps, I want to make sure we're not conflating two different questions. "TPN won't fix her cachexia" and "there's nothing left to try for her appetite" aren't the same statement, and right now nobody has actually tried optimizing her nausea or trialing an appetite stimulant. If those don't move her oral intake at all, that tells us something real. If they do, it may change what her husband is actually asking us to solve.
One thing I'd flag before anyone writes the order: mirtazapine is the reflex here, and Hunter et al. tested it against placebo in 2021 for exactly this indication and found no difference. If we're going to try something, it should be a short course of dexamethasone — one of the two agent classes ASCO's guideline actually names, and it works on her nausea at the same time. Let's hold off on the parenteral nutrition conversation for a few days, start scheduled antiemetics alongside it, and bring her husband back into a follow-up conversation once we know whether that ceiling is higher than it looks tonight.
Agreed: hold the parenteral nutrition decision for now, start scheduled antiemetic therapy and a short, time-limited course of dexamethasone for appetite — mirtazapine explicitly not chosen, given a placebo-controlled trial showing no benefit in this indication — and bring her husband back for a follow-up family meeting in several days to reassess her oral intake and revisit the nutrition-support conversation with a clearer picture.
Not agreed: the palliative care specialist felt this risked implying to her husband that a feeding line remained a live option contingent only on today's trial not working, when the underlying physiologic case against it wouldn't meaningfully change either way. The dietitian and oncologist held that giving the symptom-management trial an honest chance first, and being explicit about why, would make whatever conversation followed a more genuine one for him to hear, whichever direction it ultimately went. The disagreement was not resolved that night; the follow-up meeting was scheduled with both perspectives intentionally still in view.