Ultra-Short Term Evaluation of Coronary Vessel Wall Changes in Reference Segments Adjacent to Culprit Lesions in ST-Segment Elevation Myocardial Infarction

Background. Culprit lesions of ST-segment elevation myocardial infarction (STEMI) patients are friable, soft, and prone to disruption during primary percutaneous coronary intervention (pPCI). The presence of dissections in reference vessel segments (RVSs), adjacent to stented culprit lesions, and dynamic luminal changes in proximal or distal RVSs have not yet been investigated. We therefore sought to assess the healing patterns of edge dissections and the changes of lumen area at RVSs within 1 week post stent implantation in patients with STEMI. Methods. In the MATRIX trial (ClinicalTrials.gov NCT01433627), optical coherence tomography (OCT) was performed at the end of pPCI and within 1 week during staged PCI. The RVS dissection was defined as: type 1 = flap; type 2 = cavity; type 3 = double barrel; and type 4 = fissure. We compared separately the fate of residual dissection and luminal area/dimension by OCT in the target vessel between pPCI and staged PCI, including 1-year clinical outcomes. Results. Out of 151 patients, 46 patients had dissections in 50 RVSs and did not experience worse clinical outcome. Dissections were 44% type 1, 28% type 2, 12% type 3, and 16% type 4. Overall, 18% of the dissections healed. The mean lumen area of the RVS enlarged in 82 patients (59%) from pPCI to staged PCI. Compared with the proximal RVS, there was a significant increase in the lumen diameter at the distal RVS (0.06 ± 0.25 mm vs -0.01 ± 0.21 mm; P=.01). Conclusion. Dissections occur frequently after pPCI. One-fifth of them heal within 1 week and do not seem to negatively impact clinical outcomes. Distal RVS lumen area increased compared with proximal RVS, likely reflecting a different vasoconstriction pattern over time.

J INVASIVE CARDIOL 2021;33(12):E923-E930. Epub 2021 November 18.

Key words: dissection, optical coherence tomography, reference vessel segment, ST-segment elevation myocardial infarction, vasoconstriction

The goal of primary percutaneous coronary intervention (PCI) with stent implantation is coronary flow restoration and vessel wall scaffolding at the culprit sites, in order to avoid vessel recoil and reocclusion. Since most of culprit lesions are friable, soft, and easy dilatable, one of the most frequent complication is proximal or distal stent edge (reference vessel segment) dissection.¹ The presence of dissection, either at the proximal and/or distal edge segments of the vessel, is associated with reinfarction and stent thrombosis.^2,3 Intravascular ultrasound identifies approximately 5%-10% of edge dissections.^3,4 However, optical coherence tomography (OCT) studies have revealed that vessel injury occurs much more commonly, in up to 28%.⁵ Given the high incidence of vessel edge dissections following PCI and the possible link to future adverse events, a deep understanding of the natural healing course of such a phenomenon is clinically relevant.^6,7 In addition, the reference vessel segment dimensions, proximal and distal to the culprit lesion, experience dynamic changes in lumen area after stent implantation; this may be due to various physiological and hemodynamic changes taking place after restoration of flow.

We sought to assess the healing patterns of edge dissections and the dynamic changes of lumen area within 1 week after stent implantation in patients with ST-segment elevation myocardial infarction (STEMI) in the setting of the OCT substudy of the large MATRIX (Minimizing Adverse Hemorrhagic Events by Trans-radial Access Site and angioX) program (ClinicalTrials.gov Identifier: NCT01433627). The clinical perspective of the study is shown in Table 1.

Study population. The MATRIX study was a randomized, multicenter, prospective, open-label, factorial trial that compared the effectiveness and safety of radial or femoral access sites and unfractionated heparin with provisional glycoprotein IIb/IIIa inhibitor (GPI) or bivalirudin in patients with acute coronary syndrome undergoing invasive management. Patients were followed up with an on-site outpatient clinical visit at 30 days and 1 year. Adverse cardiovascular events were independently adjudicated by a blinded clinical events committee.^8,9 The MATRIX OCT substudy details have been previously described.¹⁰ A total of 151 patients were included in this study between September 2013 and November 2015, across 16 sites in Italy, the Netherlands, and Spain, among STEMI patients with multivessel disease who were scheduled to undergo staged PCI within index hospitalization. All included patients underwent OCT assessment of treated vessel with an adequate acquisition length to address the stented, distal, and proximal length of the vessel. The procedure technique was left to the discretion of the operator and all patients underwent OCT assessment of the infarcted artery at the end of the primary PCI as well as at the time of staged revascularization. The study flow chart is shown in Figure 1.

OCT image acquisition and analysis. OCT images was similarly undertaken during primary and staged PCI. We performed calibration of the images and defined the region of interest by measuring 5 mm distal and proximal to the implanted stent edge in the target vessel. The analysis was done at every 1mm.

Edge dissections were classified as follows (Figure 2):¹¹ (1) flap: a flap structure lifted from the vessel wall, which otherwise displayed an interrupted luminal contour; (2) cavity: dissection-like cavity shape due to fracture of vessel wall structure; (3) double-barrel dissection: dissection with a false lumen separated from the true lumen by a cap; or (4) fissure: a visible split delineating a flap-like structure not lifted from the vessel wall, which otherwise displayed an uninterrupted luminal contour.

We also measured the dissection length and depth in a cross-sectional frame. We assessed the plaque composition at the edges and scored them by considering the most predominant tissue type.

The following plaque types were defined:¹² (1) fibrous plaque: a structure with high backscattering and a relatively homogenous OCT signal; (2) lipid plaque: a lesion with an OCT-delineated fibrous cap and a necrotic core (signal-poor region within an atherosclerotic plaque, with poorly delineated borders, a fast OCT signal drop-off, and little or no OCT signal backscattering); or (3) calcified plaque: appears as a signal-poor or heterogeneous region with a sharply delineated border.

For the assessment of luminal changes, we calculated average (distal and proximal of each 5 mm) reference vessel area.

The images were analyzed at MedStar Cardiovascular Research Network Core Lab in Washington DC. The readers were blinded to the clinical presentation and laboratory results. Image analysis was performed with QCU-CMS offline software (LKEB, Leiden University).

Statistical analysis. Continuous variables are expressed as mean ± standard deviation, and categorical variables as percentages. Luminal dimension changes between primary and staged PCI were compared with 2-sided paired t-test. Statistical significance was defined as a P-value <.05. Statistical analysis was performed with SPSS, version 22 (SPSS).

Among a total of 151 patients included in the study, 12 patients were excluded due to poor image quality at the reference vessel segments (Figure 1). The clinical features of the patient population are summarized in Table 2. The mean age was 63.6 ± 11.1 years and 123 patients (81.4%) were men. The left anterior descending coronary artery was involved in 43.8% of the subjects and drug-eluting stents were implanted in the majority of cases (92.8%) (Table 3).

Incidence and healing of coronary vessel wall dissections. A total of 50 edge dissections were found in 46 patients (33%). Four cases had dissections in both the proximal and distal reference vessel segments. In 76% of cases, the plaque morphology at dissection site was lipid plaque (Table 4). The distribution of dissection type was as follows: 22 type 1 (44%), 14 type 2 (28%), 6 type 3 (12%), and 8 type 4 (16%). A total of 9 dissections (18%) healed within 1 week, of which 5 (23%), 0 (0%), 1 (17%), and 3 (37.5%) were type 1, 2, 3, or 4, respectively (Figure 3 and Figure 4). There was no difference in dissection length (923 ± 411 µm vs 1179 ± 862 µm; P=.39) or the dissection depth (359 ± 180 µm vs 545 ± 299 µm; P=.08) between healed vs non-healed dissections, respectively.

The composite rate of all-cause mortality, myocardial infarction, and stroke at 1 year was 6.1% (n = 8) and was similarly distributed between patients with and without dissection (Table 5).

Dynamic coronary luminal changes at reference segments (adjacent to culprit sites). The majority of reference vessel segments (in 82 patients; 59%) increased in lumen dimension within 1 week after primary PCI. At baseline, the mean lumen area was 8.47 ± 2.72 mm² at the proximal edge and 6.03 ± 0.21 mm² at the distal edge. There were significant differences in the mean diameter and area changes in the distal vs proximal reference vessel segments (0.06 ± 0.25 mm vs -0.01 ± 0.21 mm, respectively [P=.01] and 0.28 ± 1.06 mm² vs 0.01 ± 1.13 mm², respectively [P=.02]) between primary and staged PCI (Table 6). Baseline mean reference vessel segment area was smaller in the reference vessel segment enlargement group than in the non-reference vessel segment enlargement group in primary PCI cases (6.89 ± 2.33 mm² vs 7.66 ± 2.10 mm², respectively; P=.049).

The percentage of reference vessel segment enlargement trended higher among patients who received prolonged high-dose bivalirudin compared with heparin, short-term bivalirudin, and prolonged standard-dose bivalirudin (73.4% vs 42.8% vs 58.9% vs 52.9%, respectively; P=.08). Patients with or without reference vessel segment enlargement did not differ in terms of 1-year clinical outcomes (Table 5).

In this report, focusing on the reference segments adjacent to the culprit lesion in patients suffering from STEMI and treated with PCI, we found that PCIs are inevitably associated with vascular injury, leading to intimal tears and subsequent edge dissections.^13,14 After using a specific and systematic approach in classifying the type and degree of dissection, as previously defined,¹¹ we found the following: (1) dissection flap (type 1) was most commonly observed; (2) 18% of the OCT-detected edge dissections healed uneventfully within 1 week; and (3) the distal (but not the proximal) coronary lumen diameter increased within a week.

Incidence and evolution of edge dissections. Edge dissections during coronary stenting using angiography have been reported in 1.7%-6.4% of cases,^15,16 while the incidence rates have ranged from 5%-23% using intravascular ultrasound.^4,17 In a study by Chamie et al,¹⁸ a total of 395 stent edges in 230 patients (42 patients with STEMI) were analyzed using OCT and the overall incidence of OCT-detected edge dissections was 37.8%, of which only 16% were detected using angiography. Similarly, Radu et al¹¹ reported 22 dissections in 57 patients using OCT, of which only 2 (9%) were discovered using angiography (STEMI was an exclusion criteria).These differences are related to the resolution of the imaging techniques, as a frame by frame analysis of OCT-based imaging proved to be superior to contrast-based imaging. In the present study, it was demonstrated that 33% had a dissection at 50 stent edges, which was in line with other studies.^11,18 We found that 44% of the vessel injuries were type 1 dissection flaps, with 18% of the total dissections healing at 1 week follow-up, and the majority classified as type 4 fissures (38%). Similarly, Radu et al¹¹ showed that 96% of flap dissections healed at 1-year follow-up. Kume et al¹⁹ observed that in 36 stable angina patients with 12 edge dissections, all but 1 patient was fully healed at follow-up (median, 188 days). It was concluded that the longer the circumferential extension of the dissection, the greater impediment it posed to vascular healing. Due to the low prevalence, we can only hypothesize that the longitudinal extension/depth of the dissection takes precedence over the maximum length in the healing process, which we can observe in our report as type 4 dissections healed more successfully compared with type 1-3 dissections, which had deeper extensions. In this report, 18% healed within 1 week. This ultra-short follow-up has never been reported for the investigation of early healing of edge dissection in a STEMI population. The absence of clinical events, despite a significant proportion of edge dissections with incomplete healing status at 1 week follow-up, shows that these phenomena are most likely “benign” in the majority of cases and that they resolve at long-term follow-up.

Lumen size dynamic changes at reference segments. During OCT-guided PCI, the size of the stent is selected based on the pre-PCI mean of the proximal and distal lumen diameters, but it is possible that due to flow restoration, physiology, and hemodynamic changes within the target coronary artery the vessel lumen size in the reference vessel segment might change in size. The placement of a stent itself causes changes in the pressure of the reference vessel segment, resulting in vasodilation or vasoconstriction of the vessel. Our study consisted of STEMI patients with obstructive stenotic lesions within a target vessel, which subsequently led to vessel shrinkage through elastic recoil due to lack of blood flow and acute drop in pressure beyond the target lesions. A previous study that correlated local pressure-flow characteristics of stenotic lesions in a stable setting (STEMI was excluded) demonstrated shrinkage of the distal vessel segment beyond the target area of the stenosis as well as enlargement of the lumen proximal to the target lesion at baseline before PCI.²⁰ After intervention, it was discovered that the mean lumen diameter of the distal reference segment increased 0.05 mm (from 2.57 ± 0.6 mm to 2.62 ± 0.64 mm), whereas the proximal reference segment decreased after stenting. In this report, we found a similar diameter change in the distal and proximal reference vessel segments within 1 week after primary PCI, which means an improvement of vasoconstriction in STEMI. Whether there is a further increase in lumen dimensions after 1 week remains to be investigated.

The MATRIX trial was conducted comparing standard-dose heparin, standard-dose bivalirudin, and high-dose bivalirudin. It was observed that patients treated with high-dose bivalirudin had a numerically higher proportion of enlargement in the reference vessel segment. We hypothesize that this may be due to the antithrombin effect of high-dose bivalirudin, wherein the vasoconstrictive effects of thrombin are mitigated and hence result in vasodilation.

Study limitations. The MATRIX-OCT study investigated the difference of OCT findings between primary and staged PCI in a sizable, but still small number of patients (n = 151) for assessing the impact of these observations on clinical outcomes.

We used OCT to assess the distal and proximal reference segments, adjacent to the culprit lesion, within 1 week after primary PCI in STEMI patients, and found that approximately 1 in 3 patients had residual dissections after stenting, which did not apparently impact on outcomes despite the fact that only 1 in 4 healed shortly after intervention. The lumen area increased overall in the vast majority of patients, especially in the distal reference vessel segment compared with the proximal reference vessel segment, likely reflecting a different vasoconstriction pattern over time.

• Left Ventricular Free-Wall Rupture After Late-Admission STEMI 
• Vascular Complications of Percutaneous Coronary Intervention Via Distal Radial Artery Approach in Patients With Acute Myocardial Infarction With and Without ST-Segment Elevation
• Safety of Same-Day Discharge After Percutaneous Coronary Intervention in Selected Patients With Non-ST Elevation Acute Coronary Syndrome
• Frailty Among Patients With Acute ST-Elevation Myocardial Infarction in the United States: The Impact of the Primary Percutaneous Coronary Intervention on In-Hospital Outcomes