TL;DR. ORBITA-CTO is the first randomized, double-blind, placebo-controlled trial of CTO PCI for angina relief. In 50 patients with single-vessel CTO and refractory angina despite optimal medical therapy, CTO PCI produced a strongly positive Bayesian signal for symptom improvement: OR 4.38 (95% CrI 1.57–12.69), Pr[Benefit] = 99.6%, with 30.6 more angina-free days over 6 months. The methodology is robust, the effect size is large and clinically meaningful. The main constraint is sample size: this is a signal-generating trial, not a definitive outcomes study, and patient selection was narrow by design.
Background
The clinical rationale for CTO PCI has rested on three pillars: angina relief, reduction of ischemic burden, and potential improvement in left ventricular function. Of these, angina relief is the most consistent benefit reported in registries and observational series, yet the field has lacked a controlled trial designed to separate the specific effect of revascularization from placebo, procedural experience, and natural symptom fluctuation.
The ORBITA program established the placebo-controlled RCT methodology for coronary intervention, first applied to stable single-vessel non-occlusive disease (ORBITA-1, 2017, Lancet) and then extended to fractional flow reserve guidance (ORBITA-2, 2023, NEJM). ORBITA-CTO applies the same framework to the most complex end of the spectrum: a true chronic total occlusion, where the technical challenge is higher, the ischemic burden typically larger, and the expected symptomatic benefit arguably most compelling.
The trial was conducted at Essex Cardiothoracic Centre and Mid and South Essex NHS Trust, United Kingdom, with Sarosh Khan as lead operator and John Davies as principal investigator, with Rasha Al-Lamee as co-investigator. Abbott Vascular provided independent funding with no role in data analysis or reporting. The trial was presented as a Late-Breaking Clinical Trial at ACC.26 in March 2026 and simultaneously published in JACC.
Study design
ORBITA-CTO randomized 50 patients 1:1 to CTO PCI (n = 25) or a sham procedure (n = 25). Patients had at least one CTO confirmed by angiography, CCS class II or higher angina despite optimal medical therapy, and a single-vessel culprit anatomy to maintain procedural blinding. Multivessel disease and prior CABG were exclusion criteria.
The sham procedure employed auditory isolation: patients received general anesthesia or deep sedation with headphones playing music, preventing them from hearing procedural sounds (wire manipulation, balloon inflation, stent deployment) and thereby maintaining blinding fidelity. Both groups had the same skin preparation, femoral or radial access, and a diagnostic angiogram performed. The CTO PCI group then received full revascularization; the sham group had the sheath removed after the diagnostic run. Patients, clinical care teams, and outcome assessors were blinded to allocation throughout the 6-month follow-up period.
The primary endpoint was a validated composite angina symptom score at 6 months, assessed via a structured questionnaire capturing angina frequency, severity, and nitrate use. Analysis used Bayesian ordinal regression, specified a priori, with a non-informative prior. The pre-specified threshold for a positive trial was Pr[Benefit] > 95%. Secondary endpoints included angina-free days, CCS functional class, exercise treadmill test duration, and quality-of-life scores (SAQ, EQ-5D).
Key results
The primary endpoint was strongly positive. CTO PCI produced an OR of 4.38 (95% CrI 1.57–12.69) for improvement in the composite angina symptom score, with Pr[Benefit] = 99.6%, far exceeding the pre-specified threshold. The secondary angina-free days endpoint showed 30.6 additional angina-free days in the CTO PCI group over 6 months (95% CrI 11.1–50.7), with Pr[Benefit] > 99.9%. CCS functional class improved by a mean of 0.70 classes (95% CrI 0.27–1.13), also with Pr[Benefit] > 99.9%. Exercise tolerance and quality-of-life scores were directionally consistent with the primary result.
Blinding fidelity was assessed post-procedure and reported as excellent across both groups, with all three groups (CTO PCI, sham, and research staff) scoring near zero on blinding suspicion metrics, confirming that the auditory isolation protocol effectively masked procedural allocation.
No in-hospital deaths, coronary perforations requiring pericardiocentesis, or strokes occurred. The CTO PCI arm showed the expected low-grade periprocedural event profile (minor troponin elevation in a minority of cases). At 6 months, no significant difference in major adverse events was observed between groups.
Critical appraisal
The methodological rigor of ORBITA-CTO is high. Blinding was maintained through a validated auditory isolation protocol that the group had previously applied in ORBITA-1 and ORBITA-2, and blinding fidelity data are reported transparently. The Bayesian primary analysis is appropriate for a small-sample exploratory trial: it provides a probability statement rather than a binary p-value, and the non-informative prior prevents the authors from inflating the posterior with exogenous optimism.
The effect size is large and clinically interpretable. An OR of 4.38 for the composite symptom score and 30.6 additional angina-free days over 6 months translate directly into patient experience. The consistency across multiple pre-specified secondary endpoints (CCS class, exercise duration, SAQ domains) reinforces the signal.
The key limitation is sample size, and the authors acknowledge it explicitly. With 25 patients per arm, the trial is powered to detect a signal, not to provide definitive evidence of clinical benefit in the broad CTO population. The confidence interval for the OR runs from 1.57 to 12.69 — clinically meaningful at the lower bound, but wide enough to reflect uncertainty that cannot be resolved without a larger study.
The single-center design at a high-volume UK CTO program and the strict single-vessel anatomy requirement also limit generalizability. Post-CABG patients, multivessel disease, and CTOs in the context of impaired LV function — the populations most debated in real practice — were excluded. Whether the magnitude of benefit observed here applies to a more complex, mixed case mix is unknown.
A second point worth examining is the ORBITA lineage. ORBITA-1 produced a null result for FFR-negative single-vessel disease — a finding that generated significant controversy and led to the ORBITA-2 design (which was positive). The CTO context is mechanistically different: the ischemic burden per vessel is generally larger, total occlusion implies complete absence of antegrade perfusion, and the baseline symptom burden in ORBITA-CTO was higher than in ORBITA-1. The positive result here is therefore not paradoxical relative to the prior ORBITA program — it is mechanistically expected — but the contrast with ORBITA-1 reinforces the importance of patient selection.
The Bayesian framework also deserves explicit comment for readers less familiar with the methodology. The pre-specified threshold of Pr[Benefit] > 95% is more conservative than it first appears: it requires the posterior distribution of the treatment effect to assign at least 95% probability to any benefit. A result of 99.6% means that conditional on the data and the prior, the model assigns near-certain probability to CTO PCI being better than sham on the primary endpoint. This is not a p-value. The clinical interpretation should emphasize the CrI, the magnitude of the OR, and the secondary endpoints together, rather than treating 99.6% as equivalent to a frequentist p < 0.05.
Take-home for the cathlab
ORBITA-CTO provides the most rigorous evidence to date that CTO PCI produces meaningful, sustained angina relief beyond placebo in patients with single-vessel CTO and refractory symptoms on optimal medical therapy. The effect size is large enough that it is unlikely to disappear in a larger trial, though the exact magnitude remains uncertain.
For practical decision-making: this trial supports offering CTO PCI to well-selected patients with symptomatic single-vessel CTO unresponsive to medical therapy, particularly when the CTO territory is viable and the operator has the technical proficiency to complete the procedure safely. The caveat is that ORBITA-CTO does not address mortality, LV function recovery, or outcomes in multivessel or post-CABG CTO. These remain the targets of future larger trials.
The auditory isolation methodology is now validated across three randomized coronary intervention trials. Any future blinded CTO trial that does not use this protocol — or an equivalent — should be viewed critically. The ORBITA program has set the methodological standard.
| Domain | Assessment | Notes |
|---|---|---|
| Blinding | Robust | Auditory isolation; blinding fidelity data reported |
| Primary endpoint | Appropriate | Composite symptom score; Bayesian ordinal regression |
| Effect size | Large | OR 4.38 (CrI 1.57–12.69); 30.6 angina-free days |
| Sample size | Limited | n=50; signal-generating, not definitive |
| Generalizability | Narrow | Single-vessel, no CABG, single center, high-volume operator |
| Safety | Acceptable | No major periprocedural events in CTO PCI arm |
| Conflict of interest | Managed | Abbott funding; independent analysis confirmed |