A Pulmonary Artery Aneurysm and the Drug the Room Must Not Reach For
A single patient with a newly discovered pulmonary artery aneurysm from Behçet's syndrome. The disagreement isn't over whether to treat aggressively — it's over how fast infliximab can actually work against a structure that could rupture before it does.
Tomer A., 29, drives for a rideshare app most nights, work he picked specifically because the flexible hours let him manage the flare-and-remission rhythm of Behçet's disease he's lived with for five years — recurrent oral and genital ulcers, a bout of uveitis two years ago, all reasonably controlled since on colchicine alone. None of that prepared him for the hemoptysis that started three days ago, small streaks at first, alongside a pleuritic ache he assumed was a pulled muscle until the blood made him reconsider. CT of his chest found the real cause: a 2.8cm aneurysm in a branch of his right pulmonary artery, a rare but genuinely life-threatening manifestation of Behçet's disease that carries a real risk of fatal hemoptysis if it ruptures. There's no active extravasation on today's imaging, no hemodynamic instability, and his coagulation studies are normal — none of which changes what the aneurysm itself represents, or how fast that could change.
Major vascular Behçet's has historically been treated with cyclophosphamide-based immunosuppression, and it still may have a role in select patients — particularly those without pulmonary artery involvement specifically, or with disease refractory to other agents. But international guidance now endorses high-dose glucocorticoids with either cyclophosphamide or a TNF inhibitor as induction for exactly Tomer's situation — major vascular or nervous-system involvement — and the first randomized comparison of the two has since come down on infliximab's side. Saadoun and colleagues (NEJM Evidence, 2024) randomized 52 patients with major vascular or CNS Behçet's to infliximab or cyclophosphamide and found a superior complete response rate at 22 weeks, with fewer adverse events overall. Two things that trial does not establish, and that matter for how much weight to put on it here: it was a phase 2 study, and serious adverse events were numerically slightly higher on infliximab (15% vs 12%), so it is a result about efficacy at 22 weeks rather than a demonstration of long-term safety.
One detail matters enough to state plainly rather than let it go unspoken in a room focused on immunosuppression: nothing about this aneurysm should prompt anticoagulation. Behçet's vascular lesions are inflammatory and structurally friable, not primarily thrombotic the way most vascular emergencies are, and anticoagulating a patient with an unruptured pulmonary artery aneurysm risks converting a serious but currently stable lesion into a catastrophic hemorrhage — the opposite of what a clinician trained on other vascular diseases might reach for by habit.
How fast the drug has to work
Guidance now puts a TNF inhibitor on equal footing with cyclophosphamide for major vascular disease, and Saadoun's randomized comparison pushes me further than that — superior complete response at 22 weeks. His presentation is a clean match: vascular disease, no prior biologic exposure, no contraindication. I'd start high-dose glucocorticoids and infliximab together now, and explicitly not anticoagulation — this aneurysm is inflammatory and friable, not a clot risk, and treating it like one could be catastrophic.
I agree completely on withholding anticoagulation — that's not in question. What I'd push on is timing: infliximab, even at its fastest, takes days to achieve meaningful drug levels and longer than that to produce real anti-inflammatory effect on the vessel wall itself. If this aneurysm is at meaningful near-term rupture risk, I'm not confident a biologic alone acts fast enough, and I'd want interventional radiology involved now to assess whether embolization should happen in parallel rather than as a fallback.
I'm not arguing against infliximab as the right immunosuppressive choice — I'm arguing that 'right drug' and 'fast enough drug' aren't automatically the same question here, and this lesion's size and behavior should be assessed for that specifically, not assumed safe just because today's scan shows no active bleeding.
Both of those tracks can run together rather than in sequence. High-dose glucocorticoids have a faster onset than infliximab and should be started immediately regardless of the biologic timeline, covering the gap while infliximab's own effect builds. In parallel, I'd have interventional radiology evaluate the aneurysm's anatomy specifically for embolization candidacy — that's a mechanical decision, not a pharmacologic one, and it doesn't need to wait on how the immunosuppression conversation resolves. Cyclophosphamide stays reserved for inadequate response to this combination, consistent with its narrower current role in pulmonary-artery-specific disease. The one thing everyone in this room needs to say out loud before anyone else touches his chart: no anticoagulation, regardless of what any other consultant might otherwise reach for on seeing 'aneurysm' and 'vascular.'
Agreed: start high-dose glucocorticoids immediately and infliximab for induction, with interventional radiology consulted in parallel to assess embolization candidacy — and explicit, unanimous agreement that anticoagulation is not to be started under any circumstance related to this lesion.
Not fully agreed: the pulmonologist remains less certain than the rheumatologist that pharmacologic therapy alone is moving fast enough relative to the aneurysm's actual rupture risk, and would revisit urgent embolization if IR's anatomic assessment raises any concern — a threshold the rheumatologist views as appropriately conservative rather than premature.