Hydralazine/Isosorbide Dinitrate: A Guideline Written for a Trial That Wasn't About Him
Persistent HFrEF symptoms on fully optimized quadruple GDMT. The next guideline-recommended add-on has a real mortality benefit behind it — proven in a population, and by a pathway, that don't technically describe this patient.
R.O., a 70-year-old white man, spent thirty-eight years as a machinist before retiring, and now spends most days in his garage restoring vintage motorcycles — engines he can still take apart and rebuild from memory, even though the last several months have left him needing to stop and catch his breath halfway through a job that used to take him an afternoon without a second thought. He was diagnosed with nonischemic HFrEF eighteen months ago, EF 28% at diagnosis, and has been titrated over that time to target doses of all four pillars of GDMT: sacubitril/valsartan, a beta-blocker, spironolactone, and an SGLT2 inhibitor. He has tolerated the ARNI without any issue — no symptomatic hypotension, no meaningful renal function change, potassium consistently normal. Despite three months at target doses of all four agents, he remains NYHA class II–III: a 6MWT last month covered less distance than the one six months earlier, and he reports needing to pause twice climbing the stairs to his workshop, something he never used to think about.
Hydralazine combined with isosorbide dinitrate is the next guideline-recommended step for persistent HFrEF symptoms despite optimized therapy, but the evidence behind that recommendation is narrower than the guideline language alone suggests. The trial that established its mortality benefit, A-HeFT, enrolled only self-identified Black patients with NYHA III–IV symptoms, and the current heart-failure guideline's Class I recommendation for the combination is written specifically for that population on top of GDMT. A separate, much weaker Class 2b recommendation exists for patients of any race, but only for those who cannot be given an ACE inhibitor, ARB, or ARNI because of intolerance or renal insufficiency — and R.O. takes his ARNI without any difficulty, with stable renal function. Neither guideline pathway technically describes him. The earlier V-HeFT II trial, conducted in a general population, found enalapril outperformed hydralazine-isosorbide dinitrate on mortality — evidence against reflexively extending A-HeFT's finding across racial lines, even though the biological mechanism proposed to explain the original finding, differences in nitric-oxide bioavailability and renin-angiotensin-system activity, offers no bedside test to confirm or rule out for any individual patient.
Heart failure clinic, symptom-escalation visit
I'd add hydralazine-isosorbide dinitrate today. The proposed mechanism behind A-HeFT's finding — nitric-oxide bioavailability, RAAS activity — is a biological mechanism, not a population restriction in principle. He has real, persistent symptoms on fully optimized therapy, and I don't think a guideline letter-grade should be the reason we leave a symptomatic patient undertreated.
I'm not saying the evidence for him is as strong as it was for A-HeFT's actual population — I'm saying "not proven in this population" is different from "shown not to work."
Neither guideline pathway actually describes him — not the Class I population, and not the Class 2b intolerance-or-renal-insufficiency criterion, since he tolerates his ARNI fine and his kidneys are stable. And 2b is the weakest recommendation the guideline issues short of saying nothing. That's not a technicality; it's the reason two different trials produced two different answers in two different populations. V-HeFT II, in a general population, found enalapril beat this exact combination on mortality. I'd rather confirm there's genuinely nothing else to optimize before reaching for a therapy whose only positive trial doesn't describe him.
I'm not disputing the proposed mechanism is plausible — I'm disputing that plausible is the same as demonstrated in his case.
He doesn't have a clean alternative sitting on the table — his heart rate is already in range where ivabradine wouldn't add anything, and his QRS is too narrow for CRT to be a consideration. Given that, I'd offer a defined trial: start it, set a real reassessment point, and treat the evidence gap honestly rather than resolving it by assumption in either direction.
Agreed: start hydralazine 25mg TID and isosorbide dinitrate 20mg TID, up-titrate as tolerated, continue the existing four-pillar GDMT unchanged, and reassess symptoms and repeat the 6MWT at eight to twelve weeks.
Not agreed, and carried forward explicitly rather than smoothed over:
The Heart Failure Cardiologist's read is that the mechanism generalized as expected; the Clinical Pharmacologist's position is that this remains one anecdote, not evidence the guideline gap should close.
The Clinical Pharmacologist's caution is reinforced, and the group returns to a narrower set of remaining options with less time to spare.
The Clinical Pharmacologist maintained, even after agreeing to the trial, that the underlying evidence gap for patients like R.O. is real and unresolved by today's plan — the bounded trial was accepted as a reasonable way to act under that uncertainty, not as a resolution of it.