Coronary artery perforation located in a coronary artery bypass graft treated with new-generation single-layer polytetrafluorethylene-covered stent: results from a multicenter Registry

To the Editors

Stents covered with polytetrafluorethylene(PTFE) have been widely investigated in several settings of perforations, and in different types of coronaryartery and bypass graft disease [^1,2]. In cases of uncontrollable bleeding after perforation, implantation of covered stents (CS) represents a potentially lifesaving strategy [^1,2].

Recently, we reported that implantationof a new-generation single-layer PTFE-CS, for the treatment of coronary perforations, showedhigh technical success rates, favorable angiographic and clinical efficacy, and safety profile, especiallywith regard to thrombotic events [³].

Data concerning safety and efficacy in the specifically challenging setting of saphenous-vein-graft (SVG) perforation are scant. Therefore, we report in this analysis clinical and angiographic outcomes of patients with coronary artery perforation located in an SVG treated with a new-generation single-layer PTFE-CS.

We identified a total of four patients who underwent CS implantation using new-generation single-layer PTFE-CSs [BeGraft coronary stent (Bentley InnoMed GmbH, Hechingen, Germany) a Cobalt Chrome (L-605), open-cell platform covered with a single-layer of a microporous PTFE membrane (thickness of 89 ± 25 μm), which is clamped at the proximal and distal stent ends], in an SVG due to perforation during percutaneous coronary intervention (PCI).

Coronary artery perforation is classified according to the Ellis classification. Type I is defined by the development of an extraluminal crater without extravasation, Type II by the development of a pericardial or myocardial blush without contrast jet extravasation, and Type III by the development of an extravasation jet through a frank (>1 mm) perforation or cavity spilling into an anatomic cavity chamber. Cardiac tamponade results from pericardial fluid collection and is diagnosed by echocardiographic and clinical features, which include hypotension and tachycardia. Technical success was defined as the absence of any blood loss (contrast medium extravasation due to vessel perforation or rupture) confirmed at angiography after CS implantation, without acute deterioration of clinical status or the need for urgent surgical conversion.

Clinical endpoints were target lesion revascularization defined as any repeat PCI of the target lesion (the previously treated segment + 5 mm) or repeat bypass surgery of the target vessel, myocardial infarction (MI) defined in accordance with the universal definition of MI, stent thrombosis (ST) defined in accordance with the definition of the academic research consortium, all-cause mortality and cardiovascular mortality, in hospital and at follow-up.

Angiographic endpoints of interest were in-segment binary angiographic restenosis, defined as diameter stenosis of at least 50% in the in-segment area; late lumen loss in the treated area, defined as the difference between minimal lumen diameter postprocedure and minimal lumen diameter at follow-up and percent diameter stenosis in segment.

Baseline patient and lesion characteristics are displayed in Table 1. In brief, patients were of high age (77.1 ± 11.2) with various cardiovascular risk factors and impaired ejection fraction (46.5 ± 2.1) and have undergone several coronary revascularizations including SVG in all and PCI in 50% of cases. This is also reflected in the very high lesion complexity type B2/C in all cases and chronic total occlusion or severe calcification in 50%. The setting of primary index PCI was elective in the majority of cases (75%).

Table 1 - Demographic and clinicalcharacteristics

Demographic and clinical characteristics	Patients (n = 4)
Age (years)	77.1 ± 11.2
Male	2 (50.0)
Hypertension	4 (100.0)
Dyslipidemia	4 (100.0)
Diabetes	2 (50.0)
Current smoking	0 (0.0)
Previous myocardial infarction	1 (25.0)
Previous PCI	2 (50.0)
Previous CABG	4 (100.0)
Clinical presentation
Stable angina	3 (75.0)
NSTE ACS	1 (25.0)
STEMI	0 (0.0)
Ejection fractiona (%)	46.5 ± 2.1
Lesion complexity
Type B2/C	4 (100)
Bifurcation	0 (0.0)
Chronic total occlusion (CTO)	2 (50.0)
Severe calcification (≥Grade 3)	2 (50.0)
CTO and/or severe calcification	4 (100.0)
Coronary vein graft
To RCA	2 (50.0)
To LCx	2 (50.0)
Mechanism of perforation
Balloon predilation	1 (25.0)
Stent implantation	1 (25.0)
Balloon postdilation	2 (50.0)
Perforation grade (Ellis grade)
Type III	4 (100.0)

Data are shown as absolute numbers and percentage (%) or mean ± SD.

CABG, coronary bypass graft; NA, not applicable; NSTE ACS, non-ST-elevation acute coronary syndrome; PCI, percutaneous coronary intervention.

^aData available for 50% of patients.

Concerning periprocedural perforation, the mechanism of perforation was rupture caused by balloon dilation in every case. However, cases differed concerning the timing of perforation in terms of balloon pre- or postdilation or rupture during stent implantation (Table 1). Interestingly, maximum balloon pressure of balloon dilation (non-compliant balloon in every case with rupture during pre- or postdilation) was moderate (13 atm ± 6.7) and quantitative coronary angiography analysis revealed no balloon vessel mismatch (nominal balloon diameter 4.25 ± 1.1 vs. vessel diameter 4.19 ± 0.19; ratio 1:1.01) in three of four cases. However, in one case (Fig. 1), nominal balloon diameter of 6.0 mm was reported during postdilation in a vessel diameter of 4.19 mm (ratio: 1:1.4).

Fig. 1: New-generation single-layer covered stent implantation in a case of saphenous-vein-graft perforation at index-procedure and at follow-up angiography.

Frank perforation (Ellis grade III) was observed in every case, resulting in a potentially life-threatening complication with massive contrast extravasation. Given the history of SVG and pericardiotomy, pericardial effusion with tamponade requiring emergent pericardiocentesis via pig-tail catheter was only observed in one case with very distal perforation of a graft to right coronary artery. Perforation site was successfully sealed by implantation of one PTFE-CS in 75% of cases. One case required implantation of 3 PTFE-CS due to geographical miss in a complex R. posterolateralis sinister to R. descendens posterior dexter jump-graft anatomy. Technical success was achieved in all cases.

Angiographic follow-up at 165.6 ± 103.4 days was performed within routine clinical practice as appointed control in two of four cases with favorable 22.5 ± 12.1% diameter stenosis and late lumen loss 0.12 ± 0.44 mm. However, in one of these cases, repeat elective angiography at more than 24 months after CS implantation revealed complete occlusive restenosis of the target vessel/SVG.

Clinical follow-up was available out to 16.8 ± 12.1 months with no clinically relevant event, including death, MI, any revascularization, or ST reported out to 12 months after CS implantation. Despite favorable angiographic result at 6 months follow-up, a repeated elective angiography at timepoint of more than 24 months after CS implantation revealed complete occlusion of target vessel/SVG in terms of occlusive restenosis in one case.

The introduction of new-generation CSs led to a decrease in the percentage of patients requiring emergency surgery and improved clinical outcomes compared with their early generation counterparts [^3,4]. The latter relies on bare, stainless steel backbones, with a thicker crossing profile, by virtue of the so-called ‘sandwich design’ (with PTFE coverage captured between two stent-strut layers). The single-layer design of current CS enhances the possibility to approach challenging coronary anatomies and increases the availability of different device measures, as well as the deliverability and trackability of the devices. In line, treatment with new-generation CS is reported to be successful in a high proportion of patients with emergent coronary artery perforation [^3,4]. Second, concerns with respect to a limited efficacy and safety of early-generation PTFE-CSs [^1,2], especially with regard to thrombotic events seem to be less pronounced with new-generation CS [^3,4].

In the specific setting of SVG perforation, some further considerations should be taken into account. Emergency surgical repair is often challenging given the patients’ history of previous thoracotomy. In general, coronary stent-based percutaneous interventional strategies in SVG have been reported to have worse outcome compared with interventions in native vessels both in terms of safety and in terms of restenosis [⁵]. These issues may also impact significantly on the clinical performance of CS in this setting. Therefore, although prompt potentially lifesaving intervention is required, preferably with highly deliverable new-generation CS, these devices must also provide acceptable mid-to-long–term results in terms of efficacy and safety. In the current analysis, new-generation CS showed high technical success rates and low clinical event rates out to 1 year. Especially, the absence of thrombotic events at 12 months is reassuring and is in line with previous reports that new-generation single-layer CSs may have overcome this CS-specific limitation [^3,4]. However, given the extremely small sample size of the current analysis, definite conclusions concerning the efficacy of CS in the specific setting of SVG perforations should be drawn with caution and the current results should be regarded as hypothesis-generating.

In the specific setting of SVG perforation in patients undergoing coronary intervention, treatment with new-generation single-layer CSs showed high technical success rates and low clinical event rates out to 12 months.

Acknowledgements

Conflicts of interest

S.K. reports consulting and lecture fees from Astra Zeneca, Bristol Myers Squibb, and Translumina. For the remaining authors, there are no conflicts of interest.