ARNI Timing After MI: How Long to Wait on a Still-Rising Creatinine
Three days after an anterior STEMI, sacubitril/valsartan looks like the right next drug for his new low ejection fraction. The disagreement is whether his creatinine — still moving, not yet a settled number — is reliable enough yet to base today's decision on.
D.H., a 61-year-old man, has spent the last several months picking his granddaughter up from preschool twice a week, a routine he took on after his daughter went back to work and one he privately credits with getting him moving again after a stretch of skipped cardiology follow-ups. His hypertension has been a fixture for fifteen years — managed adequately, not tightly, on amlodipine alone, with home readings he admits he stopped logging around the same time his last set of labs, drawn eight months ago, showed a creatinine already a shade above where his primary care doctor wanted it. He arrived by ambulance after forty minutes of crushing substernal chest pressure that didn't resolve with a neighbor's spare nitroglycerin, a field ECG already showing the anterior ST elevations that sent him straight to the cath lab.
Primary PCI opened a proximal LAD occlusion within fifty-five minutes of arrival — one drug-eluting stent, roughly 175 mL of contrast, a case complex enough to run long but not unusually so. The post-procedure echocardiogram found an ejection fraction of 33%, the anteroapical wall no longer moving at all: real, new left ventricular dysfunction, not just an at-risk zone that recovers with reperfusion. That finding is what makes sacubitril/valsartan relevant here at all — the drug proven to reduce heart failure hospitalization and cardiovascular death in chronic HFrEF has since been tested specifically in patients like him, enrolled within days of a first MI with a newly reduced ejection fraction and no prior heart failure history, and by that trial's own eligibility rules he would have qualified twice over: the low ejection fraction on its own, and a baseline eGFR already under 60 as an added risk factor. What complicates the decision is a second finding moving in the same window. His creatinine has climbed on each of the three mornings since the procedure — 1.3 on admission, to 1.6, to 1.8 today — tracking his eGFR down from a baseline of 58 to 40, with no confirmed floor yet, and urine output that has stayed adequate throughout. Contrast-associated kidney injury typically peaks around forty-eight to seventy-two hours post-exposure and turns around over the following week, which would put him close to a turning point — but that describes what usually happens, not what his own creatinine is about to do, and a drug whose renal safety margin was defined in a trial that excluded anyone below an eGFR of 30 is being asked to answer that question today, not next week.
Day 3, before rounds
Start sacubitril/valsartan today. He fits PARADISE-MI on both counts that matter — an ejection fraction of 33% and a baseline eGFR already under 60 as an augmenting risk factor — and he's inside the exact half-day-to-seven-day randomization window the trial actually studied. Today's eGFR is 40. That's above the trial's own exclusion cutoff of 30 and above the label's threshold for a reduced starting dose, so nothing in his current labs argues for anything other than the standard dose.
I'm not dismissing the renal trajectory — a creatinine that hasn't turned yet is real information, not noise. But the headline result everyone quotes, "sacubitril/valsartan didn't beat ramipril after MI," missed its prespecified 15% target while still landing on a 10% relative reduction, and the same investigators' own sensitivity analysis using investigator-reported timing found a hazard ratio of 0.85 with a p-value of 0.01. Whatever benefit exists here looks concentrated in patients captured early. That's exactly what makes me reluctant to let this slide to hospital day six, or worse, to an outpatient visit six weeks out that not everyone reliably makes.
Contrast-associated kidney injury typically peaks forty-eight to seventy-two hours after exposure and turns around over the following week — but "typically" is a description of a population, not a promise about this creatinine, and his has climbed on all three mornings we've checked it, with no confirmed nadir yet. An ARNI's renal effect isn't neutral on a kidney in that state: the angiotensin-receptor component relaxes efferent arteriolar tone, which lowers intraglomerular filtration pressure — genuinely protective in stable chronic kidney disease over years, but a real further hit to GFR when applied on top of an acute injury that hasn't declared its floor. His eGFR hasn't crossed either threshold that matters here — the label's 30, or the trial's own exclusion cutoff of 30 — but it's fallen eighteen points in seventy-two hours with no confirmed floor. That trajectory, not the day-three number in isolation, is what worries me. Those thresholds aren't arbitrary numbers this committee picked; they're the line the trial's own investigators drew for exactly this drug.
The trial's enrollment window was half a day to seven days, not "today or never." Waiting two or three more days for a clearer trend still lands well inside that window — nothing about the evidence itself is lost by waiting. What's actually being traded is a few days' delay against real uncertainty about whether this kidney is still injuring or already turning the corner.
Start today — but at 24/26 mg twice daily. And I want to correct the framing, because it isn't the off-label caution it sounds like: the label recommends that starting dose for anyone previously taking low doses of an ACE inhibitor or ARB, and lisinopril 5 mg is a low dose. The renal trajectory is a second reason to land there, not the only one. The target maintenance dose is 97/103 mg either way, reached by doubling every two to four weeks as tolerated — the starting dose isn't an efficacy decision, it's a first-week safety margin, and choosing the lower one costs almost nothing if the caution turns out to be unneeded while meaningfully limiting exposure if it doesn't.
Two practical points that follow from that. He's still on lisinopril, started at admission before this conversation happened — an ACE inhibitor and an ARNI can't be co-administered, on angioedema risk, so the switch requires the full labeled 36-hour washout after his last lisinopril dose before the first sacubitril/valsartan dose, which actually sets the clock time he can receive it today. And rather than the standard two-to-four-week interval before any escalation decision, recheck creatinine and potassium at 72 hours — right at the tail of the expected injury peak — before deciding whether the normal uptitration schedule proceeds.
Agreed: sacubitril/valsartan started today at the reduced 24/26 mg twice-daily dose, timed to the full 36-hour washout after his last dose of lisinopril. Creatinine and potassium rechecked at 72 hours, before any decision about the standard two-to-four-week uptitration schedule.
Not agreed, and the reason the follow-up plan carries a branch point rather than a single expectation:
Uptitration proceeds on the standard schedule. Nobody revisits this.
Nephrology reads that as reason to hold the current dose longer before escalating. Cardiology reads a modestly higher number on an unchanged low dose as expected post-contrast noise, not evidence the drug is the problem, and would still escalate on schedule absent a clear inflection.
Cardiology and nephrology left the room holding genuinely different readings of what that same 72-hour number would mean — not a disagreement anyone smoothed over, just one neither side needed to resolve today to agree on what to do right now.