Native coronary artery disease in post-CABG patients often progresses more aggressively than in patients without prior surgery. Contributing mechanisms include competitive flow from bypass grafts, chronic hypoperfusion leading to distal vessel atrophy, and a high prevalence of diffuse fibrocalcific disease. Native lesions in this population frequently present with chronic total occlusions (CTOs), extensive calcification, long-segment disease, and complex lesion architecture. These features often require prolonged procedure times, extensive lesion preparation, high-pressure inflations, atherectomy, or advanced CTO techniques. Such interventions increase ischemic burden, biomarker release, and periprocedural myocardial infarction (MI) risk. ^,

SVG degeneration follows a triphasic progression distinct from native atherosclerosis ^, : (1) 0 to 1 year—intimal hyperplasia from smooth muscle proliferation; (2) 1 to 10 years—accelerated atherosclerosis with lipid-rich, friable plaque prone to embolization; and (3) >10 years—advanced degenerative remodeling (ectasia or aneurysm), thrombus, and diffuse degeneration. These processes create an inherently embolic substrate during SVG PCI, predisposing to distal embolization, no-reflow, and periprocedural MI. ^,,

Proximal SVG degeneration and embolic-risk stratification can be used to individualize procedural strategy. Intravascular ultrasound (IVUS) features associated with no-reflow and adverse outcomes in SVG PCI include degenerated graft morphology, intraluminal mass or thrombus, and multiple plaque ruptures. In higher-risk graft morphologies, embolic protection and careful lesion preparation become especially important. The PROXIMAL trial demonstrated that proximal embolic protection could reduce the composite of death, MI, emergency CABG, or target-vessel revascularization during SVG PCI compared with no protection.

Native PCI carries a higher early procedural hazard in complex anatomy due to prolonged ischemia and the need for atherectomy, long stent lengths, and CTO techniques, while offering better long-term durability when successful. ^, SVG PCI may have a lower early hazard when graft morphology is favorable (focal disease, limited thrombus, nonaneurysmal or nonectatic), but longer-term outcomes are limited by progressive graft degeneration and late failure. ^, Accordingly, target selection should explicitly weigh near-term procedural risk against longer-term durability and the likelihood of repeat revascularization.

Before the PROCTOR trial, the native-first strategy was supported entirely by observational data. Large national and international registries consistently reported higher rates of mortality, MI, and repeat revascularization following SVG PCI compared with native-vessel PCI. ^,,,

The VA-CART Registry found that SVG PCI independently predicted increased short- and long-term major adverse cardiac events (MACE). Similarly, the Netherlands Heart Registration demonstrated higher mortality and MACE with SVG PCI compared with native PCI. These findings reinforced the long-held preference for targeting the native vessel whenever anatomically feasible.

Multiple cohort studies have corroborated the registry findings. Yang et al. observed significantly higher rates of no-reflow, peri‑procedural MI, and in-hospital MACE following SVG PCI. Broughton et al. reported worse early and long-term outcomes with SVG PCI, while Varghese et al. demonstrated inferior angiographic and clinical outcomes following SVG PCI compared with native PCI. ^,

Together, these studies emphasized the procedural hazards inherent to SVG PCI, particularly in degenerative grafts with friable plaque, thrombus, or aneurysmal changes.

Meta-analytic evidence further solidified the native-first paradigm. Farag et al. analyzed data from over 40,000 post-CABG patients and reported significantly lower rates of MI, MACE, and repeat revascularization with native-vessel PCI compared with SVG PCI. Systematic reviews and large registry/cohort data have shown consistent advantage toward native PCI across diverse patient subsets and lesion profiles (see Table 1 ). ^,,,,,

Comparison between SVG PCI and native PCI is inherently biased in observational datasets because the target vessel is not randomly selected. In practice, SVG PCI is frequently performed when the native vessel is chronically occluded, heavily calcified, diffusely diseased, has poor distal targets, or has previously failed PCI, whereas native PCI is favored when anatomy is straightforward and offers greater long-term durability. ^,,, Similarly, graft PCI may be preferentially selected when the graft is focal, of acceptable caliber, and clinically aligned with the ischemic territory, while severely degenerated grafts may be avoided altogether. These competing selection effects create substantial confounding by indication and anatomy, limiting causal inference from nonrandomized comparisons and underscoring the need for strategy-level randomized data.

PROCTOR is the first randomized strategy trial comparing native-vessel PCI versus SVG PCI in post-CABG patients. ^, The original protocol planned 584 patients with 3-year angiographic follow-up and a MACE primary endpoint, powered (0.80) to test superiority of a native-vessel strategy; enrollment was terminated early after randomization of 220 patients, and available results therefore represent early, underpowered evidence that should be interpreted cautiously. Heart-team adjudication required that both a native coronary target and an SVG target supplying the ischemic territory were technically and clinically suitable, enriching the randomized population for favorable graft anatomy and complex native anatomy. SVG lesions <1 year from CABG were excluded per protocol. The primary published analysis reports 1-year clinical outcomes only, with longer-term follow-up prespecified in the protocol.

PROCTOR randomized patients to either an SVG PCI strategy or a native-vessel PCI strategy when both targets were deemed feasible. In the native-vessel strategy, lesion complexity was substantial, with most targets reported as CTOs (approximately 83%) and retrograde techniques used in about 50% of cases, reflecting a population in which native PCI carried high early procedural hazard. This procedural complexity plausibly contributed to the early excess of PCI-related MI observed in the native arm and supports the concept that the randomized comparison was between carefully selected SVG PCI and technically challenging native PCI.

The primary endpoint was 1-year MACE, defined as all-cause mortality, nonfatal target-territory MI, and clinically indicated target-territory revascularization. At 1 year, MACE occurred in 34% of patients randomized to a native coronary PCI strategy compared with 19% randomized to an SVG PCI strategy (hazard ratio 2.14; p = 0.006). A key driver of this difference was a higher rate of PCI-related (periprocedural) MI in the native strategy arm (13% vs 1%), with more frequent subsequent revascularization in the native arm. There was no statistically significant between-group difference in all-cause mortality at 1 year. Collectively, these findings support a distinct early hazard associated with technically complex native-vessel PCI in the PROCTOR population, while longer-term comparative durability remains unknown given the prespecified longer follow-up still pending.

Several mechanistic considerations likely explain why an SVG PCI strategy was associated with better early outcomes in PROCTOR.

First, native-vessel PCI in PROCTOR frequently involved highly complex anatomy, including CTOs, long lesion length, heavy calcification, and extensive lesion preparation, all of which increase procedure time, ischemic burden, and the risk of biomarker-defined PCI-related MI. In PROCTOR, the native arm was characterized by a high prevalence of CTO PCI and frequent use of retrograde techniques, plausibly accounting for the observed early excess in PCI-related MI.

Second, the trial design was enriched for favorable SVG anatomy. PROCTOR preferentially enrolled patients with SVG lesions that were technically suitable and less likely to exhibit high-risk features such as heavy thrombus, severe ectasia, aneurysmal degeneration, or diffuse friable plaque. This selection likely attenuated the early hazards of SVG PCI that are often seen in unselected, real-world graft interventions.

Third, contemporary SVG PCI has benefited from advances in technique and adjunctive therapy. Modern drug-eluting stents, intravascular imaging–guided optimization, refined antithrombotic strategies, and selective embolic protection can reduce periprocedural complications, particularly for focal SVG lesions with limited thrombus. Pre-PCI IVUS features have also been associated with the no-reflow phenomenon after SVG intervention, supporting an imaging-guided, risk-adapted approach to lesion handling and protection strategies, including proximal embolic protection in selected cases. ^,

Finally, the temporal risk profiles differ. Native PCI, especially CTO PCI, may carry a greater early procedural hazard but offers the potential for greater long-term durability when successful, whereas SVG PCI can appear safer early in carefully selected grafts but remains vulnerable to the biology of progressive graft degeneration over time. Accordingly, PROCTOR’s early findings should be interpreted within this time-dependent framework, and longer-term follow-up will be essential to determine whether early SVG advantages persist.

Although PROCTOR is the first randomized comparison of native coronary versus SVG PCI strategies after CABG, several limitations constrain inference and generalizability.

The follow-up duration is short. The published analysis reports 1-year outcomes, which primarily capture early procedural hazards and early repeat revascularization. This horizon does not fully reflect the longer-term SVG failure curve, which often accelerates beyond the first few years, and therefore, the durability question remains unresolved. ^,

The randomized cohort was highly selected. Eligibility required that both an SVG target and a native target supplying the ischemic territory were technically and clinically suitable, and protocol-based screening excluded higher-risk graft morphologies and lesion features. As a result, the trial does not represent the full spectrum of graft degeneration commonly encountered in clinical practice.

There was a substantial imbalance in procedural complexity. Native PCI frequently involves CTOs, severe calcification, and advanced techniques. This complexity likely contributed to the early excess in PCI-related MI and may not be generalizable to post-CABG patients with simpler native targets.

The study was underpowered by hard endpoints and long-term durability. Enrollment was terminated early relative to the original design, limiting power for mortality and stroke and rendering subgroup analyses exploratory.

Endpoint definitions may influence interpretation. Biomarker-driven definitions of PCI-related MI can disproportionately classify complex native PCI as MI events, which can amplify apparent differences between strategies when procedural complexity differs substantially.

Finally, absolute effect measures were not emphasized (e.g., absolute risk reduction, number needed to treat, number needed to harm), which would help clinicians interpret the magnitude of benefit and tradeoffs in practice, particularly given an approximate 15% absolute difference in 1-year MACE in this selected population.