Three-Vessel Disease in Diabetes: Bypass or Stents
A single patient whose coronary anatomy is genuinely revascularizable by either approach, which is exactly what turns this into a real decision rather than an anatomic foregone conclusion. The disagreement is about which five-year tradeoff — death and MI risk against an early excess of stroke — she should be the one to choose.
L.F., a 64-year-old woman, still spends her Tuesday afternoons back in the same elementary school where she taught fourth grade for over two decades, now reading picture books aloud to a new generation of eight-year-olds in the volunteer program she joined the year she retired. She has had type 2 diabetes for twelve years, and despite metformin and a basal insulin regimen her hemoglobin A1c has run in the 8s more often than not — her hypertension, by contrast, has been reliably controlled on a single agent for years. Four days ago she came to the emergency department with exertional chest pressure that didn't resolve with rest, and her troponin confirmed an NSTEMI.
Coronary angiography, performed the following morning, shows three-vessel disease — significant lesions in the LAD, the circumflex, and the right coronary artery — in a pattern the interventional and surgical teams both describe as genuinely revascularizable by either approach, which is exactly what makes the choice a real one rather than an anatomic foregone conclusion. The FREEDOM trial's own patient population looked a great deal like her: diabetic, multivessel disease, and a five-year outcome in which bypass surgery reduced death and myocardial infarction compared with drug-eluting-stent PCI, at the cost of a higher stroke rate — concentrated in the period around surgery, though the difference was still measurable at five years. That tradeoff, not the anatomy, is what the team now has to actually weigh against her own preferences and her own realistic surgical risk, which her preserved kidney function and the absence of any prior cardiac or thoracic surgery both put at the lower end of what a patient her age might carry into an operating room.
Weighing the five-year tradeoff
FREEDOM is real and I'm not going to argue it away, but it's also more than a decade old at enrollment and ran largely on earlier-generation drug-eluting stents. Contemporary stent technology and more aggressive post-PCI antiplatelet and lipid management have narrowed that gap in more recent registries. She's told me directly she wants to avoid a sternotomy if there's a reasonable path that doesn't require one, and three-vessel disease without left main involvement is a reasonable path.
The gap narrowing in registries doesn't unseat a randomized trial built specifically to answer this question in patients who look like her. FREEDOM wasn't ambiguous — CABG reduced death and MI in diabetics with multivessel disease, and the mechanism behind that isn't really about stent generation, it's about complete surgical revascularization addressing all three vessels durably in a population whose diabetes itself accelerates disease progression in whatever's left unstented.
Her preference matters and I'm not dismissing it, but a preference against surgery isn't the same as a preference against the outcome surgery is more likely to deliver — those need to be discussed as two separate things, not folded into one.
Whichever procedure she has, the diabetes itself needs a plan that doesn't wait for the cardiac decision to get made first. Her A1c in the 8s is its own driver of the disease we're trying to treat surgically or percutaneously, and an SGLT2 inhibitor belongs in her regimen for its independent cardiovascular benefit, on top of whatever glycemic improvement it adds. The timing is the part to get right: if she goes to surgery, the label requires an SGLT2 inhibitor be held at least three days beforehand because of euglycemic ketoacidosis risk, so starting one this week and operating next week is precisely the sequence to avoid. Write it into the discharge plan to begin once she is postoperative and eating normally. Her insulin regimen needs a perioperative plan well before the morning of the operation, not worked out the night before.
After the tradeoff was framed directly — not "invasive versus less invasive" but "a real reduction in five-year death and MI, against a real increase in perioperative stroke risk" — she chose bypass surgery. CABG was scheduled for the following week; empagliflozin committed to but deliberately deferred to the postoperative period rather than started in the intervening days, per the label's three-day preoperative hold; endocrinology arranged to finalize a perioperative insulin plan before the operative date rather than the night before it.