Critical review · Apr 2, 2026
ORBITA-CTO: chronic total occlusion PCI versus placebo in stable angina
Khan S, Sajjad U, Fawaz S, Butt H, Simpson R, Ibrahim A, et al. A Randomized, Placebo-Controlled Trial of Chronic Total Occlusion Percutaneous Coronary Intervention in Stable Angina - ORBITA-CTO. J Am Coll Cardiol. 2026 (pre-proof).
doi:10.1016/j.jacc.2026.03.027 · Randomized, double-blind, placebo-controlled trial with Bayesian analysis (United Kingdom; NCT05142215), n = 50
Summary
ORBITA-CTO randomized 50 patients between October 2021 and October 2025. Criteria were single-vessel CTO, confirmed ischemia and viability, J-CTO ≤3 and no relevant bystander disease. Randomization 1:1 to CTO PCI or placebo took place after dual-injection angiography.
All anti-anginals were stopped at randomization. Symptoms were recorded daily on the ORBITA-app for 6 months, with patient-initiated re-titration. The primary endpoint was an angina symptom score, analyzed with a Bayesian ordinal model of daily repeated measures.
To protect blinding, patients wore noise-cancelling headphones under deep sedation. Clocks were removed from rooms, handovers were scripted and both arms stayed overnight. The Bang blinding index was close to zero (−0.08 in the PCI group, 0.04 in placebo).
| Outcome | CTO PCI vs placebo | Pr(benefit) |
|---|---|---|
| Angina score (primary) | OR 4.4 (95% CrI 1.6 to 12.7) | 0.996 |
| Daily angina episodes | OR 4.4 (95% CrI 1.6 to 11.8) | 0.997 |
| Angina-free days over 6 months | +30.6 (95% CrI 11.1 to 50.7) | >0.999 |
| SAQ angina frequency | +10.7 (1.4 to 20.2) | 0.988 |
| SAQ physical limitation | +13.5 (4.5 to 22.3) | >0.999 |
| SAQ quality of life | +18.2 (5.4 to 30.5) | 0.997 |
| SAQ summary score | +13.7 (4.2 to 23.2) | 0.997 |
CCS class improved. Dyspnea and EQ-5D-5L did not differ. Technical success was 96% (24/25), with IVUS in 92%. There were 60 anti-anginal up-titrations during follow-up: 21 in the PCI group and 39 in placebo. Data covered 8631 follow-up days.
Strengths
- Verified blinding, with a Bang index close to zero.
- Daily symptom data over 8631 follow-up days, with a Bayesian model suited to repeated measures.
- Both arms received an active medical strategy, with symptom-driven anti-anginal re-titration.
- High procedural quality: 96% technical success, 92% IVUS, two experienced operators.
- One failed PCI case, averaging 6 angina episodes per day, was kept in the intention-to-treat analysis. This biases against PCI, and the effect held.
Methodological issues
- Small sample. With 50 patients, credible intervals are wide.
- Selected lesions at experienced centers. The result does not extend to multivessel disease, high-complexity CTOs (J-CTO 4 to 5) or centers with success rates below 96%.
- No hard endpoints. The trial was not designed for prognosis.
- No effect on dyspnea or generic quality of life. Dyspnea and EQ-5D-5L did not differ between groups.
Opinion
EuroCTO (Werner et al.) showed symptom improvement with CTO PCI versus medical therapy at 1 year, sustained at 3 years. But it was open-label, with 17.5% crossover from medical therapy to PCI. Without blinding, the placebo contribution to symptom relief was unknown.
ORBITA-CTO measured that contribution. Both arms improved, so the placebo response was large. The PCI effect was immediate and sustained, and the placebo group needed more anti-anginals over time.
For years, CTO PCI skeptics pointed to the lack of blinded evidence. That point now has an answer: CTO PCI reduces angina beyond placebo, in well-selected patients at experienced centers. Randomizing 50 patients took 4 years.
Some called the design an insult, for comparing PCI with “placebo without pills.” PCI was not compared with nothing. Both arms had symptom-driven anti-anginal up-titration throughout follow-up (39 up-titrations with placebo and 21 with PCI). The comparison was PCI versus an active medical strategy. I know of no other design that isolates the effect of PCI. We ask for randomized, blinded evidence for what we do, and this trial delivered it.
I do not perform CTO PCI. My assessment is about the method, not the result.