Drug-Coated Balloon Only Strategy After Recanalization of Chronic Total Coronary Occlusions (NoTrace-CTO)

Design:

Prospective, single-center, investigator-initiated, single-arm pilot study conducted at the Heart Institute (InCor), Hospital das Clinicas, University of Sao Paulo Medical School. Consecutive eligible patients are enrolled by convenience sampling from the institutional chronic total occlusion (CTO) recanalization waiting list, following routine care pathways without modification of clinical prioritization or therapeutic decision-making.

Index procedure:

Recanalization follows standard interventional practice under the hybrid CTO algorithm, including antegrade wire escalation, antegrade dissection and re-entry, and retrograde approaches. Baseline lesion complexity is graded with the Multicenter CTO Registry of Japan (J-CTO) and Prospective Global Registry for the Study of Chronic Total Occlusion Intervention (PROGRESS-CTO) scores. Lesion preparation uses sequential semi-compliant and non-compliant balloon predilatation sized approximately 1:1 to the intravascular ultrasound (IVUS) measured reference lumen diameter; cutting or scoring balloons are used at operator discretion where fibrocalcific tissue prevents adequate luminal gain. Drug-coated balloon (DCB) delivery proceeds only after angiographic and IVUS confirmation of adequate lesion preparation, defined as absence of flow-limiting or otherwise significant dissection, and satisfactory conditions for drug delivery.

Device and technique:

The AGENT paclitaxel-coated percutaneous transluminal coronary angioplasty (PTCA) balloon catheter (2 micrograms/mm2 dose density) is sized 1:1 to the IVUS-measured reference vessel diameter, with length selected to cover the full treated segment. Each inflation is maintained for a minimum of 120 seconds at nominal pressure, a pre-specified protocol requirement exceeding the instructions-for-use reference inflation time, adopted to prolong balloon-wall contact in recanalized fibrocalcific segments. Where more than one balloon is required, each inflation independently meets the minimum duration, with at least 2 mm of overlap documented by IVUS. No permanent metallic scaffold is implanted as part of the strategy.

Imaging:

Intravascular ultrasound is acquired with a rotational 60 MHz high-definition system by automated motorized pullback at 0.5 mm/s, extending at least 5 mm beyond each device edge, before and immediately after DCB angioplasty and again at nine months. Serial pullbacks are co-registered using fixed anatomical landmarks. Quantitative coronary angiography (QCA) is performed offline using validated edge-detection software with guiding-catheter calibration and matched angiographic projections between time-points. Analyses are performed by independent readers at the InCor core laboratory blinded to clinical data and time-point; inter- and intra-observer reproducibility are assessed by intraclass correlation coefficient and Bland-Altman limits of agreement, and approximately 10% of paired IVUS studies undergo external over-read by an independent reader unaffiliated with InCor.

Bailout:

If a flow-limiting dissection (Thrombolysis in Myocardial Infarction [TIMI] flow below 3) occurs after DCB dilation and cannot be optimized with repeat inflation, stent implantation is performed. Such cases are classified as technical failure of the DCB-only strategy, count as failures in the DCB-only strategy success proportion, and are retained in the intention-to-treat sensitivity analysis with a pre-specified penalized late lumen loss value rather than being treated as missing.

Antithrombotic therapy:

Dual antiplatelet therapy (DAPT) for 6 months is standard, abbreviated to 3 months in patients at high bleeding risk based on clinical judgment and validated scores.

Follow-up:

Elective angiographic and IVUS reassessment of the target vessel is performed at nine months. Patients declining invasive reassessment may accept coronary computed tomography angiography as a fallback; these patients contribute exploratory descriptive data but not paired QCA or IVUS data to the primary endpoint. Cumulative contrast volume and radiation exposure across both procedures are prospectively tracked.

Analysis populations:

The per-protocol population, comprising enrolled and successfully recanalized lesions that complete the DCB-only strategy without bailout stenting, is the primary analysis population for late lumen loss. The intention-to-treat population is analyzed as a pre-specified, mandatory co-reported sensitivity analysis in which bailout-stented lesions are assigned a penalized late lumen loss value set a priori at 0.50 mm, with rank-based worst-rank imputation additionally reported. DCB-only strategy success is calculated on the full intention-to-treat denominator.

Statistical approach:

Sample size was determined under a one-sample performance-goal framework. The benchmark, derived from a Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) aligned pooled analysis of six contemporary second-generation drug-eluting stent (DES) CTO cohorts under independent core-laboratory QCA using a random-effects model with Hartung-Knapp-Sidik-Jonkman adjustment, is a mean in-stent late lumen loss of 0.20 mm with a pooled within-study standard deviation of approximately 0.50 mm. Under a one-sample t-test with one-sided alpha of 0.025 and 90% power, 30 evaluable patients with paired angiographic follow-up are required; allowing approximately 20% attrition, target enrollment is 38. The primary objective is met if the upper bound of the one-sided 97.5% confidence interval for mean in-balloon late lumen loss lies below 0.50 mm. The full type I error is allocated to this single primary endpoint; all other endpoints are supportive or exploratory.