A note for colleagues.
Why this trial exists
Diabetic patients with multivessel disease are the group in whom stent-platform differences should matter most — diffuse disease, longer lesions, higher restenosis and event rates — yet they are precisely the group that head-to-head drug-eluting stent trials have historically excluded or under-enrolled. TUXEDO-2 was designed to close that gap, and it was conducted entirely in Indian centres.
Design
Investigator-initiated, prospective, open-label, multicentre, 2 × 2 factorial randomised controlled trial across sites in India. A total of 1,800 patients with diabetes and multivessel disease undergoing PCI were randomised 1:1 to an ultra-thin biodegradable-polymer sirolimus-eluting stent (Supraflex Cruz BP-SES) or a durable-polymer everolimus-eluting stent (Xience DP-EES) — and, in the second randomisation, to ticagrelor or prasugrel. All received guideline-directed medical therapy; intravascular imaging and physiology were used where feasible.
Primary endpoint: target lesion failure (TLF) at 1 year — a composite of cardiac death, target-vessel myocardial infarction, or ischaemia-driven target lesion revascularisation. Noninferiority margin 4.5% (1-sided upper 97.5% confidence bound).
Mean age 60.3 years; 28% women; predominantly three-vessel disease (reported as approximately 85%), with a mean of 2.88 stents per patient.
Results at 1 year
- TLF: 7.92% (BP-SES) vs 8.75% (DP-EES); risk difference −0.83 percentage points, 1-sided upper 97.5% bound 3.42% — meeting the prespecified noninferiority margin (P = 0.005).
- Cardiac death: 3.6% vs 3.4%.
- Target-vessel MI: 6.61% vs 7.54%.
- Ischaemia-driven TLR: 0.8% vs 1.0%.
- Non-fatal MI: 4.7% vs 6.4%. Stent thrombosis: 1.0% vs 0.7%.
- Device, lesion and procedural success 99.7% in both arms; procedural complications 1.09% vs 1.00%.
Reading it critically
This is noninferiority, not superiority — the trial does not establish that the ultra-thin BP-SES is better, only that it is not meaningfully worse against a 4.5% margin, which is generous relative to the observed event rates. The design is open-label, and the study was funded by an unrestricted grant from the sirolimus-stent manufacturer (SMT, Surat), with that company's chief medical officer among the authors, though the funder is stated to have had no role in design, analysis or submission. Follow-up is 1 year; late catch-up with biodegradable polymers is a question only longer follow-up can answer. The DAPT randomisation (ticagrelor vs prasugrel) is reported separately and is worth reading alongside.
Two figures stand out for practice: an ischaemia-driven TLR under 1% in diabetic multivessel disease is remarkably low, but a TLF of roughly 8% driven largely by target-vessel MI and cardiac death is a reminder that in this population the residual risk is atherosclerotic and metabolic, not primarily device-related.
Why it matters in central Gujarat
Diabetes prevalence in our catchment is high, and diffuse multivessel disease in relatively young patients is routine in the Anand and Kheda cath lab. TUXEDO-2 is one of the few large randomised datasets generated in Indian patients, at Indian centres, in exactly this phenotype — including a Gujarat site (Bankers Heart Institute, Vadodara). Practically, it supports the position that a contemporary ultra-thin biodegradable-polymer sirolimus-eluting stent is a reasonable platform choice in diabetic multivessel PCI, and it strengthens the case for what actually drives the residual 8%: aggressive lipid lowering, glycaemic control, and imaging-guided optimisation rather than stent brand selection.
Sources
- Kaul U, et al. Ultra-Thin Sirolimus-Eluting Versus Everolimus-Eluting Stents in Diabetic Multivessel Coronary Artery Disease Patients: The TUXEDO-2 Trial. J Am Coll Cardiol. 2026 Jun 24 (online ahead of print). doi:10.1016/j.jacc.2026.05.025 — jacc.org
- PubMed record (abstract and disclosures): pubmed.ncbi.nlm.nih.gov/42383943
- TCT 2025 presentation coverage: Cardiac Interventions Today
- Trial registration: CTRI/2019/11/022088
Summary prepared for colleagues; readers should refer to the primary publication for full methods and data.