Although distal left main (LM) disease has been treated with coronary artery bypass grafting for several decades, the advances in stent designs, adjunctive pharmacotherapy, and the evolution of intravascular imaging to guide and optimize percutaneous coronary interventions (PCI), make interventional treatment a viable alternative in cases with low to intermediate syntax score (1). In an attempt to improve immediate and long-term clinical outcomes, multiple stenting techniques have been proposed, and disagreement exists on whether an upfront single- or two-stent strategy is superior (2). In cases of severe disease in both branches, where an upfront 2-stent strategy is planned, guidelines recommend the DK-Crush (DKC) technique based on evidence from trials comparing the outcomes of this strategy to classic Crush (3) and Culotte stenting (4, 5). Although the discussion regarding the relative merits of different strategies remains open, 2-stent techniques involve many technical steps and demand expertise (6). Consequently, any efforts at simplifying these processes, including reducing procedural times, contrast and x-ray exposure while preserving patient outcomes are welcome. In a recently published paper, we reported the outcomes of a randomized trial comparing DKC with a simplified variant involving a shorter stent crush length and only one rewiring. We reported that this variation, which we called reverse-T-and-protrusion (rTAP) and which resembles a previously reported technique named “cone flare DKC”(7), was non-inferior to DKC in terms of side branch (SB) ostial expansion and was associated with shortened procedural times (8). In the present study, we compare the intermediate-term outcomes assessed by optical coherence tomography (OCT) of patients with LM bifurcation disease, randomized to either the DKC technique or rTAP. The purpose of the new proposed technique is to minimize procedural time and complexity without compromising patient outcomes.

The baseline clinical characteristics, procedural characteristics, and procedural outcomes, as well as OCT analysis results, of the 52 enrolled patients were previously reported (8). There were no differences between groups in any of the clinical, procedural, or imaging variables except for ostial side branch expansion, which was significantly larger in the rTAP at the end of the procedure (8).

The differences between rTAP and DKC have been extensively described in a previous paper and are summarized in Figure 1 (8). The major differences include the use of a systematic alternate high-pressure inflation of MB and SB balloons following SB stent implantation and the need for one rewiring in rTAP (instead of two in DKC). In rTAP, this alternate balloon dilation results in the splaying of the proximal SB stent struts along the circumference of the SB ostium. In DKC, the first MB balloon inflation followed by the first rewiring and kissing results in crushing of the proximal SB stent struts on one side.

In our previous publication, the rTAP technique proved to be non-inferior to the gold standard DKC technique in terms of short-term efficacy and safety in the treatment of complex unprotected LM bifurcation lesions. As compared to DKC, rTAP reduced procedural time and improved side branch ostial expansion (8). The current study aims to compare the outcomes, as assessed by OCT, of both techniques at the intermediate term, i.e., 6-months follow-up.

The primary endpoint that we sought to examine is the serial change in the side branch ostial area in the 2 groups. No significant difference was detected in that respect between the 2 study arms (Table 3). Despite that a significantly better SB ostial stent expansion was achieved in rTAP vs. DKC at baseline, both groups had a similar ostial stent area at the end of the procedure (median >5 mm² in both groups) (8). It is also crucial to note that a follow-up duration of 6 months does not cover the typical time-frame for restenosis in drug-eluting stents, and thus it is critical to observe how this plays out on the longer-term. Different authors have previously pointed out that multiple factors can influence such serial change, including features assessed by 3D-OCT: the SB jailing configuration by the MB stent struts, the site of crossing through the MB stent struts into the SB to perform the KBI (11, 12). These aspects warrant examination in further studies with 3D-OCT reconstruction.

From another perspective, stent strut apposition, tissue coverage, and the extent of neointimal growth are key elements to evaluate in the follow-up of different stenting techniques. The higher percentage of malapposed struts shown by follow-up OCT (including in part SB-ostium-jailing struts) in rTAP compared to DKC, albeit not reaching statistical significance, was not observed immediately after implantation. This can signify a higher incidence of late acquired malapposition. However, since the same stent type and similar deployment pressures were used in the two groups, the mechanisms of this phenomenon are unclear, but it might be hypothesized that the repeated high-pressure inflations might have resulted in more extensive vascular trauma. A more plausible explanation though is that the numerically fewer major malappositions detected at baseline in the DKC group (8), aided neointimal bridging of lesser degrees of malapposition (13–15) in this group as opposed to rTAP.

This tendency for a higher percentage of malapposed struts raises concern because of the unsettled relationship with stent thrombosis. However, it is important to note that the percentage of uncovered stent struts was not different between the 2 study groups, with the caveat that it can be difficult to discern neointima from fibrin depositions on malapposed struts by OCT. Interestingly, none of the patients in the 2 study arms had in-stent thrombotic events, or any kind of MI so far. However, as patients emerge from the protective umbrella of DAPT beyond 1 year, and should this malapposition remain unbridged, there is a higher theoretical risk of late and very late stent thrombosis.

On the other hand, the trend for larger neointimal area relative to the stent area in DKC could be explained by the repeated kissing, which creates more arterial wall stress and trauma, as previously shown by computer simulation, which may in turn trigger an excessive healing response (16). Another interesting finding was the significantly smaller luminal areas in the main branch distal to the bifurcation in DKC, associated with a tendency for smaller stent area, and larger neointimal area relative to the stent area, despite no difference in the used stent or post dilation-balloon diameters. This could be accounted for by the higher burden of calcification (as qualitatively judged on angiography—Supplementary Table S1) in the DKC group, precluding the parent artery stent from expanding to the same extent as in the rTAP group.

Although the current study was not powered to detect differences in clinical outcomes, it is worth noting that both groups showed low rates of target vessel failure which is reassuring from a safety standpoint.

At 6-months, OCT showed a comparable change in the SB ostial area (primary endpoint) in rTAP vs. DKC. There were non-statistically significant trends for smaller luminal areas in the confluence polygon and the distal parent vessel, and larger neointimal areas relative to the stent areas, in DKC, as opposed to more malapposed stent struts in rTAP. The clinical impact of such signals is yet to be seen on longer term follow-up, as well as in other studies adequately powered to examine clinical endpoints.