Bleeding That the Anticoagulant Revealed, Not the Anticoagulant That Caused It
A mechanical heart valve makes stopping anticoagulation dangerous. New gross hematuria at a therapeutic INR makes ignoring it dangerous too. The disagreement is over how much room exists to do both at once.
H.N., 74, retired four years ago after four decades on a factory floor and has since put most of that newfound time into a vegetable garden that, by his daughter's account, has taken over the entire back half of his yard. He has a mechanical mitral valve, placed six years ago, and has been on stable warfarin therapy since with a target INR of 2.5 to 3.5 — the range the 2020 ACC/AHA valve guideline sets for a mechanical mitral prosthesis, centred on 3.0 rather than the 2.5 it gives the aortic position, reflecting the greater thrombogenicity of a valve in mitral position. He presents with three days of painless gross hematuria, no clots, no flank pain, no fever, and no recent trauma or instrumentation.
His INR today is 3.1 — within his target range, not supratherapeutic — which means simple over-anticoagulation does not explain the bleeding on its own. He has no prior episodes of hematuria, no known bladder or kidney disease, and no urinary symptoms otherwise. He is hemodynamically stable, and his hemoglobin is only mildly reduced from his known baseline — a detail that matters because it argues against a large, ongoing bleed and toward a source that, whatever it turns out to be, has not yet caused clinically significant blood loss. He is, understandably, anxious about anything that might affect the valve that has kept him alive and gardening for six years. His only other medical issue is mild osteoarthritis in both knees, managed with acetaminophen rather than an NSAID specifically because of his anticoagulation — a distinction his cardiologist has been careful about for years, which is part of why today's bleeding feels to him like a genuine surprise rather than something he'd been warned to expect.
His valve was placed for rheumatic mitral stenosis, a diagnosis that traces back to a childhood strep infection he only learned the significance of decades later, when valvular damage from it finally became symptomatic in his fifties. That history is part of why he takes his anticoagulation as seriously as he does — he has never missed a dose in six years by his own and his daughter's account — and part of why a bleeding episode with no clear anticoagulation failure to explain it has left him more unsettled than the bleeding's actual severity would otherwise justify.
A therapeutic INR that still needs explaining
His INR is in range, not supratherapeutic — that rules out simple over-anticoagulation as the explanation, which tells us this bleeding is very likely drug-revealed, not drug-caused. I'd proceed with cystoscopy and CT urogram while continuing his anticoagulation exactly as it is. His valve's thrombotic risk from any under-anticoagulation is severe, and the bleeding itself hasn't been dangerous.
I'd still want to reduce his bleeding risk while that workup happens. He's at 3.1, the upper edge of his target window even though it's technically in range — trimming toward the lower half, say 2.5 to 3.0, while we work this up seems like a reasonable, low-cost way to ease active bleeding without abandoning protection against valve thrombosis, which I don't want to risk even briefly.
You're right that some flexibility exists here — a same-range trim toward 2.5 to 3.0 is genuinely low-risk, and I wouldn't argue against it on the numbers alone.
But I'd correct the specific framing you used — "without abandoning protection" and "risk even briefly" both imply this trim carries some real thrombosis exposure to weigh against the bleeding benefit. The ACC/AHA guideline sets that 2.5-to-3.5 band as acceptable in its entirety, and the thrombosis risk behind it is established from true interruption or a genuinely subtherapeutic INR — not from a dose adjustment that keeps him inside his own target range the whole time. What you're proposing isn't a meaningful thrombosis risk at all — it's simply a reasonable way to ease bleeding, and it doesn't deserve the caution you're bringing to it.
Agreed: continue warfarin with the target trimmed to the lower half of his range (2.5–3.0) pending urologic workup — cystoscopy and CT urogram — with no interruption and no reversal agent given.
Not agreed: whether cardiology or urology should hold final authority over the anticoagulation target if the workup reveals a bleeding source requiring an invasive procedure, such as tumor resection, that would need a much more significant, temporary interruption. Left open pending the workup's actual result.