Stent-to-stent comparison trials have routinely excluded or under-recruited patients with diabetes and multivessel disease — the very group with the worst outcomes. TUXEDO-2 was an investigator-initiated, open-label, multicentre trial in India with a 2x2 factorial design. It randomised 1,800 patients undergoing percutaneous coronary intervention to an ultra-thin biodegradable-polymer sirolimus-eluting stent or a durable-polymer everolimus-eluting stent, and separately to ticagrelor or prasugrel. Mean age was 60.3 years and 28% were women.
At one year, target lesion failure — cardiac death, target vessel myocardial infarction or ischaemia-driven target lesion revascularisation — occurred in 70 patients (7.92%) with the sirolimus stent and 78 (8.75%) with the everolimus stent. The risk difference of -0.83 percentage points, one-sided upper 97.5% confidence bound 3.42%, met the prespecified noninferiority margin of 4.5% (p=0.005 for noninferiority). Cardiac death (3.6% versus 3.4%), target vessel myocardial infarction (6.61% versus 7.54%) and ischaemia-driven target lesion revascularisation (0.8% versus 1.0%) were all similar, as was stent thrombosis (1.0% versus 0.7%).
This matters for Indian practice specifically. The trial recruited Indian patients at Indian centres, with a locally manufactured stent tested against the standard comparator, and the event rates it reports are the real-world numbers for this population — a nearly 8% target lesion failure rate at one year in diabetic multivessel disease, which is a sobering figure whichever stent is used.
Noninferiority is not superiority. What the trial establishes is that stent choice is not the lever here. The lever is medical therapy, glycaemic control and completeness of revascularisation.
- Either stent is a defensible choice in diabetic multivessel disease; let cost and availability decide
- Quote the real number when counselling: roughly 8% target lesion failure at one year in this population, on either device
- Do not read noninferiority as equivalence on outcomes the trial did not power for, including anything beyond one year
- Spend the consultation on secondary prevention rather than on stent selection — that is where the residual risk sits
- Note the 2x2 design also randomised ticagrelor versus prasugrel, so antiplatelet choice was balanced rather than standardised
The statistics, in plain English
A noninferiority trial asks whether the new option is not unacceptably worse, and the answer depends entirely on the margin chosen in advance. Here that margin was 4.5 percentage points, and the upper confidence bound came in at 3.42 — inside the margin, so noninferiority is declared, but the data are still compatible with the sirolimus stent being up to 3.4 percentage points worse. A margin of 4.5 points against a control event rate of 8.75% is generous. The result is real; it is a weaker claim than 'the same'.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for cardiology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free