Independent OCT Registry on Very Late Bioresorbable Scaffold Thrombosis

Background: The 3-year clinical outcome data of ABSORB II study failed to reach the two co-primary endpoints of vasomotion and late loss, but also reported a higher risk of definite or probable stent thrombosis throughout 3 years (3% vs 0%, p=0.03) with the Absorb BVS compared with the metallic EES as well as myocardial infarction.(1) In addition, a numerically higher rate of definite very late device thrombosis (2% vs 0%, p=0.19) was observed. In ABSORB Japan study, malapposition and late strut discontinuities were observed at the timepoint of very late scaffold thrombosis in all the 4 cases of definite very late scaffold thrombosis.(2,3) There is mounting evidence from other studies with extended follow-up supporting the notion of an increased risk of very late scaffold thrombosis. While the causes of acute scaffold thrombosis are explained at least in part (i.e. underexpansion, increased strut thickness), the causes underlying thrombotic events occurring beyond one year after scaffold implantation remain unclear and require investigation in an independent manner.

The investigators have previously reported that scaffold discontinuity and restenosis during the resorption process may provide mechanistic explanations for very late scaffold thrombosis.(4) As this report was based on a limited number of cases, the investigators extend the OCT database to comprehensively study the phenomenon very late scaffold thrombosis.

Rationale: The investigators consider the establishment of an independent OCT registry of paramount important as more than 150 000 patients have been implanted with Absorb BVS worldwide. The findings may provide important insights to identify patients at increased risk in addition to considerations regarding long-term antiplatelet therapy.

Objectives: The specific aim is to systematically analyse scaffold thrombosis occurring beyond 1 year after implantation of a Absorb BVS 1.1 investigated by OCT at the timepoint of thrombosis (either prior or after thrombus aspiration).

Primary and secondary endpoint/outcome(s): The primary endpoint is the underlying main reason for stent thrombosis as per quantitative and qualitative analysis and per expert panel consensus. To investigate the underlying mechanisms of very late scaffold thrombosis, the investigators will systematically assess the following parameters:

• Distal and proximal reference area, mm2
• Scaffold area, mm2
• Lumen area, mm2
• Incomplete scaffold apposition area, mm2
• Neointimal area, mm2
• Intracoronary thrombus area, mm2
• Percent area stenosis, %
• Scaffold expansion, %
• Eccentricity index
• Presence of macrophage accumulations, %
• Uncovered struts, %
• Malapposed struts, %
• Neointimal thickness, µm
• Incomplete scaffold apposition distance, µm
• Thrombus area, mm2

Other study variables: Retrospective data about clinical and drugs history of patients, procedural features and any intracoronary imaging information will be collected and analysed. This will include data for any imaging information between index implantation and very late scaffold thrombosis, when available, in order to investigate the underlying mechanisms of very late scaffold thrombosis.

Inclusion Criteria:

1. Any patients who suffered from scaffold thrombosis occurring beyond 1 year after implantation of a Absorb BVS 1.1 and investigated by OCT at the timepoint of thrombosis.
2. Sufficient quality of the OCT recording allowing for CoreLab analysis.

Exclusion Criteria: No exclusion criteria will be applied.

OCT assessment: Commercially available ILUMIEN (St. Jude Medical) or OFDI (Terumo) system are used. The analysis will be conducted at the Core laboratory at Bern University Hospital, Bern, Switzerland using a systematic and validated approach.

The lumen contour is drawn by semi-automated detection using QCU-CMS software, following the endoluminal contour of the neointima with manual correction wherever required. In case of thrombus with low attenuation, the lumen contour can still be drawn. In case of high attenuation, the lumen contour was extrapolated when the lumen contour was visible in at least three quadrants. The scaffold area was measured by joining the middle points of the signal poor core of the abluminal side of the struts if at least one strut was clearly visible in every quadrant. Distal and proximal reference area and scaffold expansion were calculated as previously reported by the MUSIC (Multicenter Ultrasound Stenting in Coronaries) investigators.(5) The neointimal thickness is measured from the abluminal border of the black strut core to the lumen. Uncovered struts are defined in the absence of a homogenous coverage by neointima. If the apposed struts is not covered by neointima but rather by irregular tissue, struts are categorized as "apposed uncovered with superimposed thrombus". If the coverage of apposed struts cannot be assessed owing to the presence of attenuating thrombus precluding the evaluation of neointima, the struts were categorized as "apposed possibly uncovered". Malapposed struts were defined as struts where the abluminal surface was clearly separated from the vessel wall. Incomplete scaffold apposition (ISA) distance was measured from the abluminal border of the strut core to the vessel wall. If malapposed struts are partially or completely surrounded by thrombus, struts were regarded as malapposed with thrombus. The area between the backside of the struts and the vessel wall was measured as ISA area, while the neointimal area is measured as scaffold area - (lumen area - ISA area). Intracoronary thrombus was identified as any irregular mass inside the lumen, with a sharp border and various degrees of attenuation.

Angiography assessment: The analysis will be conducted at the Core laboratory at Bern University Hospital, Bern, Switzerland using a conventional quantitative coronary angiography software (Qangio XA).

Determination of sample size: Up a total number of 30 patients who suffered from very late scaffold thrombosis and underwent OCT will be analysed retrospectively. This sample size is justified by the need to have precise estimates of event rates in terms of narrow 95% confidence intervals.

Planned analysis: Descriptive statistics will be provided. The investigators will assess geographic association between distribution of thrombus within scaffold and abnormal findings assessed by OCT. Statistical analyses will be performed with R (R Foundation for Statistical Computing, Vienna, Austria). Continuous variables will be summarized as mean ± standard deviation or median ± 95% confidence interval, depending on normality of distribution. The investigators will assess underlying reasons for scaffold thrombosis based on the same methodology with previous publication.(6)