Inhaled Epoprostenol or Nitric Oxide: A Postpartum Lung That Doesn't Read Like the Textbook Case
A single patient, ten days postpartum, still hypoxemic after eighteen hours prone. Two inhaled vasodilators tie on efficacy and differ tenfold in price — and the argument for breaking the tie on safety is routinely made backwards.
C.M. delivered by emergency cesarean ten days ago after severe preeclampsia was diagnosed at 34 weeks, with blood pressures peaking at 178/110 and enough proteinuria that she spent four days on a magnesium sulfate infusion before delivery; her daughter is healthy and already home with her husband, a fact the bedside nurse says she asks about first thing every morning she's been lucid enough to ask. What was expected to be a routine postpartum recovery instead became progressive pulmonary edema over the following week — early diuresis helped for the first two days before her oxygen requirement began climbing again despite it — and by ICU day 2 she met criteria for ARDS. She has now been proned for eighteen hours, and her PaO2/FiO2 ratio remains at 88 on FiO2 0.9 — refractory hypoxemia despite the maneuver that was supposed to be the next step after failing conventional management.
Her echocardiogram shows a mildly elevated estimated pulmonary artery pressure with preserved right ventricular function — not a dramatic finding on its own, but a real one, and consistent with the elevated pulmonary vascular tone preeclampsia is known to leave behind even after delivery. Her preeclampsia labs have otherwise been reassuring — platelets recovered into the normal range by ICU day 3, liver enzymes trending down — which is part of why the persistent hypoxemia now reads as a distinct pulmonary problem rather than a continuation of her systemic preeclamptic process. Those recovered platelets are the detail the room keeps circling back to. Inhaled epoprostenol and inhaled nitric oxide produce roughly equivalent short-term oxygenation improvements in the general refractory-hypoxemia population, with epoprostenol costing a small fraction of nitric oxide's daily price and neither shown to improve mortality — normally reason enough to let cost decide a tie. But epoprostenol is a prostacyclin, and prostacyclin inhibits platelet aggregation; she is ten days from an emergency cesarean and three days from a platelet count that was not yet normal. The tie the cost argument depends on is a tie in the general population, and she is ten days postpartum with a surgical wound and a pulmonary vasculature preeclampsia has already stiffened.
Choosing between two drugs that mostly tie
I'd start inhaled epoprostenol first. Neither drug has shown a mortality benefit in refractory hypoxemia, and the oxygenation response is roughly equivalent between them — no randomized head-to-head trial shows nitric oxide is actually superior. Epoprostenol costs a small fraction of nitric oxide's daily price. When efficacy is genuinely a tie, cost stewardship should be what decides, and there's no reason to spend thousands more a day for a drug with no demonstrated edge.
The cost argument only holds if the two drugs are actually a tie for her specifically, and I don't think they are — though I want to be careful about which direction the safety argument runs, because it's commonly stated backwards. Rebound pulmonary hypertension on withdrawal is a documented problem with nitric oxide, described since Miller's 1995 report and mechanistically tied to endothelin-1 rises during therapy; it is not an epoprostenol phenomenon. That's an argument for weaning iNO deliberately, not for avoiding it. What actually separates the two drugs for her is what epoprostenol is: a prostacyclin, with real platelet-inhibitory activity, in a woman ten days from an emergency cesarean whose platelets were still recovering three days ago. It also drops systemic pressure, and her MAP is 74 without much room.
And the efficacy-parity data you're leaning on comes from general ARDS populations — not from postpartum patients with a fresh surgical wound and preeclampsia-stiffened pulmonary vasculature. That's the gap I'm pointing at, and it isn't a gap cost can close.
On the drug itself I'd start where you would — epoprostenol first, and I'd add that the platelet concern is real but manageable rather than disqualifying: inhaled prostacyclin has far less systemic antiplatelet effect than the intravenous route, and she's ten days out from surgery with a normal count, not two. Serial counts and a look at the wound cover it. Where I'd part company with both of you is treating this as a choice to be settled by argument at all. A fixed proportion of patients don't meaningfully respond to either agent, and neither of your positions survives her being one of them. If it works, we've answered the practical question without needing to resolve the broader debate tonight; if it doesn't, that result should drive the next step more than either of your priors does.
Agreed: start inhaled epoprostenol for a defined four-hour trial with an explicit PaO2/FiO2 response threshold as the continue/stop rule, with echocardiography before and after to check for any change in right ventricular function or estimated pulmonary artery pressure.
Not agreed:
What happens if she fails to respond adequately to epoprostenol. The pulmonologist wants to try inhaled nitric oxide next — more pulmonary-selective, no antiplatelet effect, and the rebound-on-withdrawal risk manageable with a planned wean rather than disqualifying; the critical care physician would rather move directly to the next definitive step — reinforced proning strategy or ECMO evaluation — without cycling through a second unproven inhaled agent and losing time on a drug trial that already failed once in a related mechanism.