Access to side branches with a sharply angulated origin: usefulness of a specific wire for chronic occlusions

INTRODUCTION

The optimal strategy and timing of complete revascularization in patients with ST-segment elevation myocardial infarction (STEMI) and multivessel coronary artery disease remains unclear, and current recommendations are controversial.1 According to 2023 European Society of Cardiology (ESC) guidelines, complete revascularization, based solely on angiographic severity, is recommended in “stable” STEMI patients.2 Conversely, the 2023 Asia-Pacific Expert Consensus Document suggested a treatment strategy of non-culprit lesions (NCL) based on angiographic severity and invasive physiological assessment with fractional flow reserve (FFR) or non-hyperemic pressure ratios for patients with STEMI.3

FFR and non-hyperemic pressure ratios may be inaccurate in acute coronary syndrome (ACS), as hyperemic flow may be reduced due to microcirculatory dysfunction, while the resting flow may be higher due to neurohumoral compensatory mechanisms.4

Angiography-derived indices are additional physiological tools. They need ≥ 1 angiographic projections plus frame count analysis and/or aortic pressure that may also be different in the acute setting.

Furthermore, drugs such as hypolipidemic agents may promote plaque regression, potentially impacting the results of physiological assessment after a few months into therapy.5

Our main objective was to evaluate the changes in invasive physiological measurements of NCL between the acute and staged phases of ACS.

Secondly, we aimed to evaluate the effects of different therapies on physiological measurements.

METHODS

Eligibility criteria

We included studies that evaluated the physiology of NCL during acute and staged interventions for ACS. Studies conducted on assessments following percutaneous coronary interventions of non-culprit vessels, or with patients with chronic coronary syndrome were excluded.

Case reports, conference abstracts, commentaries, editorials, and reviews were excluded as well. An initial protocol was registered in PROSPERO with registration No. CRD42024574683.

Search strategy, and study selection

We conducted the search across ClinicalTrials.gov, Embase (via Ovid), Google Scholar, PubMed, and Web of Science from inception through 26 April 2024 (initial search). We used the “Review articles” filter in Google Scholar and the “Topic” field in Web of Science. No language restrictions were applied.

Duplicates were removed using Deduplicator (SR-Accelerator) software. Title/abstract and full text screening was conducted independently by 2 authors using Rayyan software.

Back in July, 2 authors conducted a backward and forward citation analysis of the included articles using Citationchaser software.

The search strings were repeated in 6 December 2024 (in Embase, sources with invalid date limits were excluded). Simultaneously, we looked into any online conference news on imaging modalities and physiological measurements.6 Additionally, we looked into the “Slide Library” section using the “2024” filter on another web page.7

Finally, we manually reviewed the references of the articles included after the initial search.

All discrepancies were resolved by consensus.

Selection process was recorded in sufficient detail to complete a Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) flow diagram.8

Data extraction

The following data were extracted from each article: a) study characteristics; b) population characteristics; c) type of physiological index(es); d) follow-up duration; e) primary endpoint.

The primary endpoint was the variation between acute and staged indices regarding statistical significance, mean difference (MD), and disagreement on revascularization decision.

One author extracted the data, and another one checked it independently. We contacted the authors of eligible studies when clarifications were needed.

Risk of bias assessment

Risk of bias was assessed using the Joanna Briggs Institute (JBI) critical appraisal tools,9-11 as appropriate.

Two authors independently assessed the risk of bias for each study. We used red for high, yellow for moderate, and green for low risk of bias based on positive answers being ≤ 49%, 50%-69%, or ≥ 70%.

Data synthesis

We conducted a descriptive synthesis of the evidence. Results from data extraction were shown in separate tables based on risk of bias, or bubble charts. Some data were rounded to the nearest integer (age, diameter of stenosis of NCL, and follow-up) or 2 decimal places (MD).

Unless otherwise specified, P values < .05 were considered statistically significant. When MDs were unreported, they were estimated by calculating the difference between staged and acute mean values. When required, a formula for estimating the means was applied.12

In bubble charts, the size of the bubbles represents the number of patients or lesions if the former was not reported. Acute−/staged+ disagreement indicates an acute value above the threshold, with the staged value below the revascularization cut-off. Acute+/staged− disagreement represents the opposite.

RESULTS

Characteristics of the articles, participants, and indices

Results of the search and selection processes are shown in figure 1. Extracted data are shown in table 1 and table 2.

Figure 1. PRISMA flow diagram. PRISMA, preferred reporting items for systematic reviews and meta-analyses.

Table 1. Extracted data of studies with low risk of bias

A total of 20 articles were included13-32 (1 article in the form of a conference presentation).19 Publication years went from 2010 through 2024. The total number of reported patients was 4379.

In every publication, the patients are predominantly men and non-diabetic. The main clinical presentation was STEMI, except for 3 studies.19,29,31

The following methods were assessed: a) angiography-derived: Murray law-based quantitative flow ratio (muQFR), quantitative flow ratio (QFR), vessel FFR (vFFR); b) hyperemic (FFR); and c) non-hyperemic indices: instantaneous wave-free ratio (iFR), resting distal-to-aortic coronary pressure ratio (P_d/P_a). When reported, the FFR was obtained using adenosine.

Reported patients for each index are as follows: 2340 (muQFR), 1187 (QFR), 710 (FFR), 243 (iFR), 92 (vFFR), and 73 (resting P_d/P_a).

Risk of bias

The studies mainly used an observational (cohort) design. Cohort studies on angiography-derived methods were retrospective, except for 1 article on QFR.28 Those on FFR and non-hyperemic indices were prospective, except for 2 substudies.22,26

QFR was also evaluated by 1 quasi-experimental study27 and 1 randomized controlled trial.13

Finally, the FFR was assessed by 2 randomized controlled trials, in samples with predominance of non-ST-segment elevation myocardial infarction (NSTEMI).19,31,33

Results are shown in table 1 of the supplementary data, table 2 of the supplementary data, and table 3 of the supplementary data. There were no studies with high risk of bias.

Table 2. Data drawn from studies with moderate risk of bias

Primary endpoint

Statistical significance

There were no articles on relevant changes in muQFR,16,30 resting P_d/P_a,25 and vFFR17 at the follow-up.

A significant difference in FFR, iFR, and QFR was found in 3, 2, and 1 article(s),14,22-26 respectively. In 1 study, the difference in QFR was non-significant, with a significance threshold of .001.28

These variations were seen in cohorts of STEMI patients.14,22-26 Studies including non-ST-segment elevation acute coronary syndrome (NSTEACS) did not show any relevant differences regarding the QFR15 or the FFR.19,21,29,31

A total of 4 articles20,22,25,26 evaluated > 1 method. The iFR and FFR were both stable in the study by Musto et al.,20 while the iFR was more stable than the FFR in a different article.25 The QFR was compared to both the FFR26 and the iFR.22 Unlike these indices, the QFR did not show any significant changes in staged phases.22,26

Mean differences

The most valued indices showed varying results. muQFR had MD values close to 0 in both studies.16,30

QFR variations were observed at both lower22,26,28 and higher values.14,15,18 Conversely, the FFR and the iFR varied towards smaller and greater values, respectively.22-26,29,31 Their MDs ranged from − 0.02 to + 0.06 (QFR), − 0.03 to 0.00 (FFR), and 0.00 to + 0.03 (iFR).14,19-21,24,25,28 MD values of 0.01 were observed more often.

In STEMI patients, the MDs of the FFR, the iFR, and the QFR were close to 0 only in studies with mean follow-ups of < 1 week.20,32 In studies including NSTEACS, the FFR MDs were close to 0 after longer mean follow-ups (> 1 month).19,21 Furthermore, Ntalianis et al. showed a greater stability of FFR in patients with NSTEMI (MD, 0.00) vs those with STEMI (MD, − 0.02).21

Disagreement

Disagreement in the indication for revascularization is shown in figure 2. MDs of 0.01 resulted in variable disagreements: 5%-18%.15,18,23,25

Figure 2. Disagreement between acute and staged values in the indication for PCI. B, Barauskas; C, Cortés; E, Erbay; FFR, fractional flow reserve; H, Huang; iFR, instantaneous wave-free ratio; K, Kirigaya; L, Li; muQFR, Murray law-based QFR; N, Ntalianis; PCI, percutaneous coronary intervention; QFR, quantitative flow ratio; SE, Sejr-Hansen; SH, Shukla; SP, Spitaleri; T, Thim; V, van der Hoeven; vFFR, vessel fractional flow reserve.

Unlike the QFR, the FFR and the iFR consistently showed a higher frequency of one type of disagreement: acute−/staged+ for FFR,21,23,25 and acute+/staged− for iFR.24,25

Secondary endpoint

A total of 4 studies compared the effects of different drugs on the physiological parameters.13,19,27,31

Ticagrelor (which can increase the levels of adenosine) was compared to clopidogrel and no significant differences were found in the FFR of non-culprit vessels after 6 months of treatment.31

Another 3 studies compared a proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitor (eg, alirocumab or evolocumab) plus high-intensity statin (HIST) (eg, rosuvastatin 20 mg/day) vs statin-only therapy.13,19,27

In a nonrandomized study, the QFR values were significantly higher in the evolucumab group at 12 months.27 However, in 2 randomized studies, no significant differences were observed across the 2 treatment groups in the QFR at 12 months or in the FFR at 3 months, respectively.13,19

DISCUSSION

The main findings of this systematic review are these: firstly, in STEMI patients, the muQFR, resting P_d/P_a, and vFFR indices remained relatively stable in retrospective and/or small studies. The FFR, iFR, and QFR showed variability between acute and staged phases. Secondly, the FFR did not change significantly in prospective cohorts or randomized controlled trials including NSTEACS. Thirdly, the QFR was more stable than both the FFR and the iFR in direct comparisons, although only the FFR and the iFR exhibited consistent directions of change. Fourthly, PCSK9 inhibitors added to HIST did not influence physiological measurements compared with HIST in randomized controlled trials.

The muQFR demonstrated stability in a large sample of patients. This index is based on a single angiographic view, unlike other angiography-based methods that require 2 angiographic projections. This characteristic might reduce observer variability and enhance reliability. Future prospective and comparative studies are needed to confirm the validity of this method.

Although low variations for FFR, iFR, and QFR were observed in cohorts of STEMI patients,20,32 these studies were limited by short-term follow-ups. Thim et al. found a non-significant change in the iFR with 5-day follow-ups, whereas there were significant changes with ≥ 5 day follow-ups.24 Therefore, physiological disarrangements initiated at the acute moment of STEMI might still exist if a staged procedure is conducted close to the index event.24,25

Angiographic, hemodynamic, and microcirculatory variables may alter acute physiologic assessment and account for the higher reliability of the FFR in NSTEACS vs STEMI.

In patients with microvascular dysfunction, epicardial blood flow cannot increase sufficiently during maximal hyperemia, thus causing a reduced pressure gradient across the stenotic lesion,29 and higher FFR values.

In STEMI patients, microcirculatory indices (coronary flow reserve and index of microcirculatory resistance) were significantly worse during the acute phase, along with a higher FFR.25 Conversely, studies including NSTEACS did not show any significant differences in the coronary flow reserve and/or index of microcirculatory resistance at the follow-up.21,29,31

Furthermore, STEMI patients showed greater acute angiographic severity, along with lower QFR or iFR values,14,22 which may be attributed to vasoconstriction typically occurring during the acute phase.

Consequently, the FFR seems more reliable in NSTEACS vs STEMI due to reduced acute microcirculatory impairment and/or vasoconstriction.

Literature trials support the use of the FFR in NCL of NSTEMI during the acute phase (eg, within the index hospitalization).34,35 In contrast, acute FFR-guided complete revascularization did not show any significant benefits in terms of death or myocardial infarction in STEMI patients.36-39

The higher stability of QFR when directly compared to the FFR or the iFR was limited to a small number of patients in post-hoc substudies.22,26 A MD of 0.01 sometimes led to non-trivial disagreement on revascularization decision,25 likely due to baseline values being near the cut-off. Therefore, it is essential to have an index which remains stable or demonstrates consistent changes, such as the FFR and the iFR. Similarly, these indices demonstrated a greater frequency of a specific type of disagreement (methodological variations–wire positioning–may explain the less frequent cases of disagreement).24

Therefore, the FFR and the iFR could be considered in the acute STEMI as an alternative to delayed assessments,25 considering that the FFR tends to decrease and the iFR tends to increase. The FFR could guide the revascularization of positive lesions (FFR ≤ 0.80).25 In patients with a FFR > 0.80, acute iFR assessment can be used to delay the revascularization of negative NCL (iFR > 0.89).24 In the remaining cases (iFR ≤ 0.89), some authors suggested a staged reevaluation.24 At least 5 days after the index procedure should go by. This was the minimum time needed to observe the initial resolution of acute physiological disturbances.24

Finally, when plaques are correctly identified as functionally negative, they may still be vulnerable and associated with adverse events. NCL exhibiting thin-cap fibroatheromas as defined by optical coherence tomography, and having a muQFR ≤ 0.80, showed the highest event rate,40 which suggests that imaging can offer additional prognostic information.

PCSK9 inhibitors have shown minimal impact on coronary physiology, despite greatly reducing low-density lipoprotein-cholesterol (LDL-C) levels. A large treatment effect on HIST only,19 minor flow limitation at baseline, and microvascular compensation may account for this finding.13

However, combining alirocumab with HIST resulted in a greater increase in cap thickness of fibroatheromas vs statin monotherapy as assessed by optical coherence tomography.41 Moreover, lower LDL-C levels after an ACS are associated with the occurrence of fewer cardiovascular events.2 Therefore, PCSK9 inhibitor treatment is recommended in patients who do not reach their LDL-C target despite maximum tolerated statin and ezetimibe therapy.2

Limitations

The wide variety of indices to assess coronary physiology has led to a lack of evidence on some of them; similarly, few studies made direct comparisons among such indices.

Our evaluations are mainly based on observational studies with a very different follow-ups.

Angiography-based methods frequently exhibited bias due to their retrospective analysis. Some patients were excluded because of the poor quality of angiographies or anatomic issues, such as ostial lesion or severe vascular tortuosity. Some angiographies were not obtained optimally according to the specific acquisition guide.

CONCLUSIONS

The assessment of functional indices for NCL during the initial procedure for STEMI is not absolutely reliable. This evidence is due to potential variability of the FFR, the iFR, and the QFR outside the acute phase. Although variation was not significant for muQFR, resting Pd/Pa, and vFFR, retrospective and/or limited data limit the generalizability of these findings.

Both the FFR and the iFR showed consistent directions of change. Therefore, during an acute STEMI, the FFR can guide the revascularization of positive NCL, while the iFR can help defer revascularization of negative NCL. A negative FFR with a positive iFR should be reevaluated.

The FFR shows robust data supporting its use in NLC of NSTEMI during the acute phase, meaning that it is a more reliable index for initial ACS procedures.

DATA AVAILABILITY

Search string for Google Scholar: “acute coronary syndrome”|”myocardial infarction” “fractional flow reserve”|FFR| “hyperemic ind”|”resting ind”|iFR|”instantaneous wave-free ratio”| “angiography-based ind”|”angiography-derived ind”|QFR|”quantitative flow ratio”|OFR staged|repeated|later. The remaining search strings are available upon request.

FUNDING

None declared.

ETHICAL CONSIDERATIONS

Ethical committee and patient’s informed consent: not applicable. We followed the SAGER guidelines with respect to possible sex/gender bias.

STATEMENT ON THE USE OF ARTIFICIAL INTELLIGENCE

Microsoft Copilot was used to help edit the English version of the text.

AUTHORS’ CONTRIBUTIONS

F. Vergni designed the work. F. Vergni, S. Buscarini, L. Ciurlanti, and F.L. Gurgoglione contributed to data acquisition (screening, and/or extraction). F. Vergni, and L. Ciurlanti conducted the critical appraisal. F. Vergni, and S. Buscarini contributed to data interpretation. F. Vergni, and F.L. Gurgoglione drafted, edited and reviewed the work. F. Vergni, S. Buscarini, L. Ciurlanti, F.L. Gurgoglione, F. Pellone, and M. Luzi approved the final version for publication.

CONFLICTS OF INTEREST

None declared.

SUPPLEMENTARY DATA

REFERENCES

INTRODUCTION

Coronary artery disease is the leading cause of mortality worldwide.1 Despite the safety involved in deferring invasive treatment in patients with stable coronary artery disease without functionally significant lesions,2 a percentage of patients experience cardiovascular events at the long-term follow-up.3 It has been reported that cardiovascular events not only depend on the degree of coronary obstruction assessed by intracoronary physiology4-5 but also on the global atherosclerotic burden and its vulnerability assessed by intracoronary imaging modalities.6-8

The new era of coronary physiology is based on predicting fractional flow reserve by reconstructing the coronary tree using angiography and computational fluid dynamics.9-10 Estimating quantitative flow ratio (QFR) is the most validated method of the ones currently available.

QFR—which predicts fractional flow reserve^10-11—has proven to be a better tool than angiography alone to guide the need for lesion revascularization12 and shown long-term prognostic value13. Furthermore, it provides quantitative information out of the 3D reconstruction of the coronary tree, including minimum diameter and area, reference diameters, luminal volume, and atherosclerotic plaque volume in the studied vessel. However, the prognostic value of this quantitative analysis has not been sufficiently studied.

The main aim of this study was to determine the prognostic value of global plaque volume (GPV) in patients with stable coronary artery disease without functionally significant lesions at a 5-year follow-up.

METHODS

We conducted a retrospective observational study on a cohort of patients from 6 tertiary referral centers.

Study population

Patients who underwent coronary angiography from January through December 2015 for suspected stable coronary artery disease were included. Each participant center retrospectively enrolled all patients who underwent coronary angiography for suspected stable coronary artery disease and met the inclusion criteria. Patients with chronic total coronary occlusions, prior coronary artery bypass graft surgery, or inadequate angiographic quality for analysis were excluded. Additionally, patients whose angiographic analysis revealed a positive QFR study (< 0.80) in any coronary territory were excluded. The principal investigator conducted a retrospective follow-up at each center within the next 5 years following the index procedure. Baseline and procedural characteristics, and events at the follow-up were collected by local investigators. The study fully complied the good clinical practice principles and regulations set forth in the Declaration of Helsinki for research with human subjects. The study protocol was approved by the ethics committee of the reference hospital (Hospital Clínico Universitario de Valladolid) and the institutional review boards, including informed consent obtained from participants or, alternatively, approval for retrospective data analysis under ethical committee supervision.

Angiographic analysis

A blinded angiographic analysis of diagnostic coronary angiograms was performed by trained analysts at a centralized imaging unit (Icicorelab, Valladolid) using specialized software (QAngio XA 3D QFR, Medis Medical Imaging System, The Netherlands). A 3D reconstruction of the 3 major coronary vessels was performed using 2 different projections with > 25° of separation. For the right and left circumflex coronary arteries, the proximal marker was manually placed at the vessel ostium, while for the left anterior descending coronary artery, it was placed at the left main coronary artery ostium. The distal marker was placed at the end of the coronary artery. Plaque volume was estimated by calculating the difference between the theoretical reference vessel volume in the absence of atherosclerotic disease and the estimated vessel volume in angiography using QFR software via quantitative analysis. Reference diameters, minimum diameter, and minimum area were obtained for each vessel. Considering contrast flow through the coronary tree, QFR was calculated according to FAVOR II standards for the physiological significance of coronary lesions. Patients with functionally significant disease (QFR < 0.80) were excluded.

Statistical analysis

Categorical variables are expressed as totals and percentages, and continuous ones as means and standard deviations. GPV was estimated as the sum of plaque volume across 3 coronary territories.

The primary endpoint—major adverse cardiovascular events (MACE)—was a composite of cardiac death, acute myocardial infarction, or all-cause unplanned hospital admission.

An optimal GPV cutoff as a predictor of MACE was determined using the receiver operating characteristic (ROC) curve as the value with the maximum Youden index. Multivariate logistic regression models were used to calculate the odds ratio and 95% confidence interval as independent predictors for MACE. Variables with P < .20 in the univariate analysis were included in the multivariate model as covariates.

Event-free survival was compared using Kaplan-Meier and Mantel-Haenszel analyses. All probability values were two-tailed, and P < .05 was considered statistically significant. Statistical analysis was performed using Stata (16.1, StataCorp, College Station, United States).

RESULTS

Descriptive population analysis

A total of 803 patients were evaluated for inclusion in the registry, 122 of whom (15.2%) were excluded due to chronic occlusions in ≥ 1 coronary territory, 17 (2.12%) due to previous surgical myocardial revascularization, and 159 (19.2%) due to inadequate angiographic analysis in, at least, 1 coronary territory. Among the remaining patients, 228 (45.1%) had significant coronary artery disease (QFR < 0.80) in, at least, 1 coronary territory, which left a final cohort of 277 patients. Patient flowchart is shown in figure 1.

Figure 1. Flowchart of the patient selection process for inclusion in the study. CABG, coronary artery bypass graft; CTO, chronic total coronary occlusion; QFR, quantitative flow ratio.

The mean age of the population was 65.8 years (most were hypertensive [74.4%] men [66.1%]). Table 1 illustrates the baseline characteristics of the population. The median follow-up was 69 months, during which time 5 patients were lost to follow-up.

Table 1. Baseline characteristics of the included population

Variable	n/mean	Proportion/SD
Female Sex	94	33.9%
Hypertension	206	74.3%
Diabetes mellitus	106	38.2%
Dyslipidemia	188	67.9%
Smoking	121	43.7%
Chronic kidney disease	21	7.6%
Peripheral arterial disease	14	5.1%
Previous ischemic heart disease	105	37.9%
Age (years)	65.8	12.2
Weight (kg)	78.0	15.0
Height (cm)	156.2	36.8
Left ventricular ejection fraction (%)	57.4	9.3
SD, standard deviation.

Angiographic analysis

Mean plaque volume in the study population was 170.5 mm³ (± 16.5); mean QFR was 0.95. Table 2 illustrates the overall means from the angiographic analysis according to the coronary territory studied. Plaque volume was independently analyzed for each coronary territory and was significantly higher in the right (243 mm³) vs the left anterior descending (161.4 mm³) and left circumflex coronary arteries (172.9 mm³). Data on this analysis by coronary territories are shown in table 1 and figure 1 of the supplementary data.

Table 2. Characteristics of the angiographic analysis performed in the 3 coronary territories using quantitative flow ratio

Variable	Mean	SD	95%CI
QFR	0.95	0.37	0.95-0.96
Length	76.99	13.21	75.22-78.77
Proximal diameter	3.18	0.47	3.11-3.24
Distal diameter	1.99	0.34	1.95-2.04
Reference diameter	2.69	0.42	2.58-2.70
Minimum lumen diameter	1.76	0.34	1.72-1.81
Percent diameter stenosis	33.81	6.44	32.95-34.68
Stenosis area (%)	38.72	9.59	37.43-40.01
Minimum lumen area	3.53	1.30	3.35-3.70
Lumen volume	295.5	242.25	262.83-328.12
Plaque volume	170.54	240.24	138.17-202.91
SD, standard deviation; 95%CI, 95% confidence interval; QFR, quantitative flow ratio.

Prognostic value of global plaque volume

The primary event (MACE) occurred in 116 patients, which amounts to 42.7% of the cohort at the follow-up. Among these patients, 11% died, 2.6% suffered an acute myocardial infarction, and 38.1% required unplanned hospitalization. Patients who developed MACE had a significantly higher GPV (231.6 vs 111.8 mm³; P < .001), as well as those with a higher mortality rate (255.2 mm³ vs 154.3 mm³; P = .04) or unplanned hospitalizations (235.0 mm³ vs 125.4 mm³; P < .001). However, there were no significant differences in patients who experienced acute myocardial infarction (235.1 mm³ vs 169.3 mm³; P = .51).

The optimal GPV cutoff to predict events was set at 44 mm³ based on ROC curve analysis (sensitivity, 64%; specificity, 65.8%; LR⁺, 1.9; LR^–, 0.6).

Table 3 illustrates the study of the main determinants of the primary event. Variables with a significance level of P < .10 were included in the multivariate analysis. In the final model, age and GPV were independent predictors. A GPV > 44 mm³ was associated with a 2.8-fold higher risk of events at the follow-up (figure 2).

Table 3. Uni- and multivariate analysis of determinants of the main event

Determinants of the main event	Univariate analysis		Multivariate analysis
	OR	95%CI	OR	95%CI
Sex, female	1.09	0.66-1.81
Age*	1.03	1.01-1.10	1.03	1.00-1.07
Hypertension*	2.26	1.26-4.07	1.70	0.82-3.53
Diabetes mellitus	1.18	0.72-1.93
Dyslipidemia	1.04	0.62-1.73
Smoking	1.01	0.72-1.42
Chronic kidney disease	1.00	0.41-2.46
Peripheral arterial disease	1.37	0.47-4.01
Previous ischemic heart disease*	1.52	0.93-2.50	1.46	0.80-2.68
LVEF	0.98	0.96-1.01
GPV (> 44 mm3)*	1.93	1.17-3.18	2.80	1.51-5.21
Reference vessel diameter*	2.20	1.12-4.35	1.62	0.75-3.50
* P values < .10 were included in the multivariate analysis. 95%CI, 95% confidence interval; GPV, global plaque volume; LVEF, left ventricular ejection fraction; OR, odds ratio.

Figure 2. Kaplan-Meier curve showing the patients’ event-free survival based on their global plaque volume.

DISCUSSION

The main finding of this study is that GPV quantification emerged as an independent prognostic factor in patients without functionally significant coronary artery disease, which demonstrated that those with a higher GPV experienced more events at the follow-up. The optimal GPV cutoff for event prediction was set at 44 mm³. This study emphasizes the importance of anatomically characterizing coronary arteries without significant lesions.

Despite the absence of significant coronary artery obstructions, some patients still experience events during follow-up.14 In patients with a negative QFR functional study, it has been reported that the 5-year rate of events—cardiac death, target vessel myocardial infarction—is 11.6%,3 similar to our findings, where mortality rate was 11% and acute myocardial infarction occurred in 2.6% of patients. Determining the difference between the actual vessel diameter and the estimated diameter obtained through 3D reconstruction from QFR-based angiography has been used in other studies.15 This estimation—previously derived from coronary computed tomography16-17—has demonstrated the prognostic significance of plaque volume differences between normal and non-obstructive coronary arteries. These differences have also been confirmed using invasive imaging modalities such as intravascular ultrasound.18 Although angiography-derived percent luminal stenosis shows poor concordance with myocardial ischemia,19 a greater degree of coronary stenosis (percent diameter stenosis > 50%) is associated with a higher event rate at the 2-year follow-up in patients without functionally significant coronary lesions.20 The present study takes a step further into the minimally invasive characterization of atherosclerotic burden using easy-to-implement 3D coronary tree reconstruction technology as an independent prognostic factor in patients without functionally significant coronary lesions. In this regard, this study is consistent with recent studies which demonstrated that subclinical atherosclerosis burden—measured by vascular ultrasound for carotid plaque quantification and computed tomography for coronary calcium scoring—in asymptomatic individuals is independently associated with all-cause mortality.21

Based on these findings, GPV measurement enables the identification of patients who, despite having no significant coronary lesions, are at risk of developing events within the next 5 years, allowing for intensified treatment and cardiovascular risk factor control. However, this study has limitations, including its retrospective design for patient inclusion and recruitment, the use of indirect methods—such as QFR—to estimate plaque volume, and the inability of this method to describe plaque characteristics, or potential lipid plaque vulnerability. Of note, the estimated plaque volume in each coronary artery was not specifically correlated with events in that territory but rather with overall adverse cardiovascular events. Therefore, further studies are needed to confirm or refute this hypothesis.

CONCLUSIONS

Plaque volume, calculated by 3D coronary tree reconstruction, is an independent predictor of events in patients with suspected stable ischemic heart disease without significant coronary artery disease. The optimal GPV cutoff for event prediction is 44 mm³.

FUNDING

C. Cortés received funding through the Río Hortega contract CM22/00168 and Miguel Servet CP24/00128 from Instituto de Salud Carlos III (Madrid, Spain).

ETHICAL CONSIDERATIONS

The present study was conducted in full compliance with clinical practice guidelines set forth in the Declaration of Helsinki for clinical research and was approved by the ethics committees of the reference hospital (Hospital Clínico Universitario de Valladolid) and other participant centers. Possible sex- and gender-related biases were also considered.

DECLARATION ON THE USE OF ARTIFICIAL INTELLIGENCE

No artificial intelligence was used in the writing of this text.

AUTHORS’ CONTRIBUTIONS

C. Cortés and J. Ruiz-Ruiz participated in study design, data analysis, manuscript drafting, and critical review. C. Fernández and M. García participated in data collection and result analysis. F. Rivero and R. López-Palop assisted in data collection. S. Blasco and A. Freites contributed to statistical analysis. L. Scorpiglione and M. Rosario Ortas Nadal collaborated in data interpretation. O. Jiménez participated in manuscript preparation and initial review. J.A. San Román Calvar and I.J. Amat-Santos conducted the final review and approved the version for publication.

CONFLICTS OF INTEREST

None declared.

REFERENCES

INTRODUCTION

Percutaneous coronary intervention (PCI) is the first-line therapy in patients with ST-segment-elevation myocardial infarction (STEMI).1 The STEMI-related culprit vessel is usually considered as one of the most important prognostic factors in STEMI patients.2,3 This assumption comes from previous studies –conducted in the pre-reperfusion or thrombolysis era– which showed that left anterior descending artery (LAD)-related STEMIs were associated with worse clinical outcomes compared with right coronary (RCA) and left circumflex artery (LCX)-related lesions.4-9

However, in the contemporary era of primary PCI there are limited data about the prognostic impact of LAD as the STEMI-related culprit vessel especially in a very long follow-up.10,11

Therefore, the aim of this study was to investigate the impact of the LAD as the STEMI-related culprit vessel on very long-term clinical outcomes in STEMI patients undergoing primary PCI enrolled in the EXAMINATION-EXTEND study (10-year follow-up of the EXAMINATION trial).

METHODS

Study design and patients

The EXAMINATION trial (NCT00828087) was an all-comer, multicenter, prospective, 1:1 randomized, 2-arm, single-blind, controlled trial conducted at 12 centers across 3 countries to assess the superiority of EES (Xience V) vs BMS (Multilink Vision, Abbott Vascular) in STEMI patients regarding the primary endpoint of all-cause mortality, any myocardial infarction, and any revascularization at 1 year. The study had broad inclusion criteria and few exclusion criteria to ensure an all-comer STEMI population representative of the routine clinical practice. The study outcomes have been reported up to the year 5.12,13 After that, it was reinitiated as the EXAMINATION- EXTEND study to evaluate patient- and device-oriented composite endpoints at 10 years. The latter is registered at ClinicalTrials.gov (NCT04462315) as an investigator-driven extension of follow-up of the EXAMINATION trial. An independent study monitor (ADKNOMA, Barcelona, Spain) verified the adequacy of the extended follow-up and events reported. All events were adjudicated and classified by an independent event adjudication committee blinded to the therapy groups (Barcicore Lab, Barcelona, Spain). The 10-year primary endpoint results of the EXAMINATION-EXTEND study have been previously published.14 For the aim of this study, baseline, procedural characteristics and outcomes were stratified according to the STEMI-related culprit vessel (LAD vs others). All centers participating in the EXAMINATION trial received the approval of their Medical Ethics Committee, and all enrolled patients who had already signed their written informed consent forms. Medical ethics committee approval for EXAMINATION- EXTEND was granted at the institutions of the principal investigators (Hospital Clínic and Hospital Bellvitge, Barcelona, Spain), and the requirement to obtain informed consent to gather information on 10- year events was waived. The study complied with the Declaration of Helsinki.

Study endpoints

The primary endpoint of this study was the patient-oriented composite endpoint of all-cause mortality, any myocardial infarction, or any revascularization at 10 years. Secondary endpoints were each individual components of the primary endpoint, device-oriented composite endpoint (cardiac death, target-vessel myocardial infarction, target lesion revascularization), its individual components and stent thrombosis. Detailed descriptions of the study endpoints and definitions have been published previously.15

Statistical analysis

Continuous variables are expressed as median (interquartile range; IQR), and categorical variables as absolute and relative frequencies (percentages).

Baseline clinical, angiographic, and procedural characteristics were compared between the groups stratified by the STEMI-related artery (LAD vs other vessels) using the Wilcoxon rank sum test, the chi-square, or Fisher’s exact test, where appropriate.

Time-to-event curves for POCE and all-cause death were plotted using the one minus the Kaplan-Meier estimate and the cumulative incidence function for other outcomes. The incidence of events at the follow-up was compared between groups using log-rank or Grey’s test. Landmark analyses were also performed, setting landmark points at 1 and 5 years.

The association between LAD as a STEMI-related culprit vessel and events was analyzed in univariable and multivariable cause-specific Cox regression models. Covariates were added to the multivariable model in 2 blocks. The first model included all clinically relevant baseline characteristics variables with P < .1 in the between-groups comparison (LAD vs other vessels), i.e., sex, smoking status, peripheral vascular disease, previous PCI, previous CABG, previous MI, and Killip class. The second model (expanded adjustment) included both the baseline characteristics and the left ventricular ejection fraction (LVEF) at discharge.

Two-tailed P-value < .05 was considered statistically significant. All statistical analyses were performed using R (R Core Team (2022). R: a language for statistical computing. R Foundation for Statistical Computing, Austria) with the following packages: survival, tidycmprsk, jskm, and gtsummary.

RESULTS

Patient characteristics

In 631 (42%) out of the 1498 STEMI patients included in the EXAMINATION EXTEND trial, the LAD was the culprit vessel (LAD-STEMI group), whereas in 867 patients (58%) it was not (non- LAD-STEMI group). Patients’ inclusion flowchart is shown in figure 1.

Figure 1. Study flowchart. A total of 1498 patients were initially recruited. At 10 years, clinical follow-up was obtained in 95.2% of the patients. LAD, left anterior descending artery; STEMI, ST-elevation myocardial infarction.

LAD-STEMI group had a higher incidence of active smokers, advanced Killip class and more depressed LVEF vs the non-LAD-STEMI group, which, however, exhibited a higher incidence of peripheral vascular disease, previous MI and previous PCI (table 1). Also, although non-statistically significant, the frequency of late comers and bailout PCI was numerically higher in the LAD-STEMI group.

Table 1. Baseline clinical characteristics

Clinical characteristics	Overall (N = 1498)a	LAD-STEMI (N = 631)a	Non-LAD-STEMI (N = 867)a	Pb
Age (years)	61 [51-71]	61 [51-71]	61 [51-70]	.778
Sex (male)	1,244 (83%)	512 (81%)	732 (84%)	.094
Current smoker	415 (28%)	197 (31%)	218 (25%)	.009
Hyperlipidemia	655 (44%)	268 (43%)	387 (45%)	.419
Hypertension	725 (48%)	307 (49%)	418 (48%)	.843
Peripheral vascular disease	55 (3.7%)	14 (2.2%)	41 (4.7%)	.011
Previous stroke	31 (2.1%)	14 (2.2%)	17 (2.0%)	.726
Previous myocardial infarction	80 (5.3%)	23 (3.7%)	57 (6.6%)	.013
Previous PCI	61 (4.1%)	18 (2.9%)	43 (5.0%)	.042
Previous CABG	10 (0.7%)	1 (0.2%)	9 (1.0%)	.052
Clinical presentation				.126
PCI (< 12 h)	1,268 (85%)	520 (82%)	748 (86%)
Bailout PCI	98 (6.5%)	51 (8.1%)	47 (5.4%)
PCI after successful thrombolysis	34 (2.3%)	14 (2.2%)	20 (2.3%)
Late comer (> 12 h and < 48 h)	97 (6.5%)	46 (7.3%)	51 (5.9%)
Killip class				< .001
I	1,337 (90%)	525 (83%)	812 (94%)
II	115 (7.7%)	76 (12%)	39 (4.5%)
III	23 (1.5%)	20 (3.2%)	3 (0.3%)
IV	18 (1.2%)	8 (1.3%)	10 (1.2%)
In-hospital LVEF (%)	52 (45, 58)	46 [40-55]	55 [50-60]	< .001
Symptom onset to first medical contact time (hours)	1.38 (0.70, 3.00)	1.27 [0.67-3.00]	1.47 [0.75-3.00]	.353
CABG, coronary artery bypass graft; LVEF, left ventricular ejection fraction; PCI, percutaneous coronary intervention. a>Median [interquartile range] or frequency (%). bWilcoxon rank sum test; Pearson’s chi-squared test; Fisher’s exact test.

Regarding procedural data, LAD-STEMI group received smaller stent diameter (3.12 mm vs 3.26 mm; P = .001) and had a lower incidence of ST-segment resolution than the non-LAD-STEMI group (73%, vs 50%; P = .001) (table 2). The use of GP IIb/IIIa inhibitors was numerically lower in the LAD-STEMI group, although the differences between groups were not statistically significant. Of note, almost half of the patients (46%) with LAD-STEMI had the lesion in the proximal LAD compared with 44% of them who had it in the mid/distal LAD.

Table 2. Angiographic and procedural characteristics

Procedural characteristics	Overall (N = 1498)a	LAD-related STEMI (N = 631)a	Non-LAD-related STEMI (N = 867)a	Pb
STEMI-related culprit vessel				N/A
LAD	631 (42)	631 (100)	0 (0)
LMCA	3 (0.2)	0 (0)	3 (0.3)
RCA	650 (43)	0 (0)	650 (75)
LCx	207 (14)	0 (0)	207 (24)
SVG	7 (0.5)	0 (0)	7 (0.8)
Multivessel disease	188 (13)	72 (11)	116 (13)	.256
Total ischemic time (hours)	3.9 [2.7-6.8]	4.0 [2.7-7.3]	3.9 [2.7-6.3]	.366
Manual thrombectomy	976 (65)	405 (64)	571 (66)	.502
Glycoprotein IIb/IIIa inhibitor	785 (52)	312 (49)	473 (55)	.051
Direct stenting	885 (60)	390 (63)	495 (59)	.113
Stent type				.312
DES	751 (50)	326 (52)	425 (49)
BMS	747 (50)	305 (48)	442 (51)
No. of stents	1.39 (0.65)	1.37 (0.63)	1.40 (0.66)	.428
Total stent length (mm)	23 (18-35)	23 (18-33)	23 (18-35)	.154
Maximal stent diameter (mm)	3.20 (0.45)	3.12 (0.40)	3.26 (0.47)	< .001
Post-dilatation	221 (15)	97 (15)	124 (14)	.564
TIMI flow after PCI				.607
0	26 (1.7)	9 (1.4)	17 (2.0)
1	12 (0.8)	5 (0.8)	7 (0.8)
2	59 (4.0)	29 (4.6)	30 (3.5)
3	1396 (94)	584 (93)	812 (94)
ST-segment resolution > 70%	852 (63)	285 (50)	567 (73)	< .001
BMS, bare metal stent; CABG, coronary artery bypass graft. DES, drug-eluting stent; LAD, left anterior descending coronary artery, LCx, left circumflex artery; LMCA, left main coronary artery; PCI, percutaneous coronary intervention; RCA, right coronary artery; STEMI: ST-segment elevation myocardial infarction; SVG, saphenous venous graft; TIMI, thrombolysis in myocardial infarction. aMedian [interquartile range], mean (standard deviation) or frequency (%). bFisher’s exact test; Pearson’s chi-squared test; Wilcoxon rank sum test.

Ten-year outcomes

At the 10-year follow-up, POCE did not differ between LAD-STEMI and non-LAD-STEMI group (adjusted HR, 0.95; 95%CI, 0.79-1.13; P = .56) (figure 2). Moreover, no differences were found in terms of each individual component of POCE (all-cause mortality, MI, any revascularization) (figure 3) and other secondary endpoints (figure 1 of the supplementary data). Furthermore, when the expanded adjustment was performed and LVEF was included in the multivariable analysis, there were no inter-group differences between (table 3).

Figure 2. Central illustration. Outcomes of patients with ST-segment elevation myocardial infarction according to the culprit vessel at the 10-year follow-up. LAD, left anterior descending coronary artery; STEMI: ST-segment elevation myocardial infarction; POCE: patient-oriented composite endpoint.

Figure 3. Time-to-event curves for the patient-oriented composite endpoint (A), all-cause mortality (B), myocardial infarction (C), and any revascularization (D) in patients stratified according to the culprit vessel. LAD, left anterior descending coronary artery; MI, myocardial infarction; STEMI, ST-segment elevation myocardial infarction; POCE, patient-oriented composite endpoint.

Table 3. Ten-year outcomes

10-year outcomes	LAD-related STEMI (N = 631)	Non-LAD-related STEMI (N = 867)	Unadjusted HR (95%CI)	P	Adjusted HR (95%CI)	Pa	Expanded adjusted HR (95%CI)	Pb
Patient-oriented composite endpointc	220 (34.9)	307 (35.4)	0.99 (0.83-1.17)	.87	0.95 (0.79-1.13)	.56	0.98 (0.78-1.23)	.86
All-cause mortalityd	131 (21.6)	179 (21.2)	1.02 (0.82-1.28)	.84	0.93 (0.74-1.18)	.56	0.81 (0.59-1.09)	.17
Any myocardial infarctione	33 (5.5)	53 (6.3)	0.86 (0.56-1.33)	.50	0.93 (0.60-1.45)	.76	1.14 (0.67-1.93)	.61
Any revascularization	108 (17.4)	161 (18.8)	0.93 (0.73-1.18)	.55	0.96 (0.75-1.22)	.72	1.12 (0.83-1.52)	.45
Device-oriented composite endpointf	94 (14.3)	132 (14.2)	0.98 (0.75-1.28)	.88	0.91 (0.70-1.20)	.50	0.95 (0.67-1.35)	.77
Cardiac death	72 (9.8)	95 (10.0)	1.06 (0.78- 1.44)	.71	0.89 (0.65- 1.23)	.49	0.71 (0.47-1.09)	.12
Target vessel myocardial infarction	16 (2.6)	36 (4.2)	0.62 (0.34-1.11)	.10	0.69 (0.38-1.25)	.22	0.87 (0.43-1.77)	.71
Target lesion revascularization	44 (7.0)	63 (7.3)	0.97 (0.66-1.43)	.89	1.01 (0.68-1.49)	.96	1.20 (0.76-1.93)	.43
Definite/probable stent thrombosisg	17 (2.7)	28 (3.3)	0.84 (0.46-1.54)	.57	0.83 (0.45-1.55)	.57	0.80 (0.38-1.73)	.58
95%CI, 95% confidence interval; HR, hazard ratio; LAD, left anterior descending artery, STEMI: ST-elevation myocardial infarction. Data are expressed as no. (%). aCause-specific Cox regression model adjusted for sex, smoking status, peripheral vascular disease, previous percutaneous coronary intervention, previous coronary artery bypass graft, previous myocardial infarction, and Killip class. bCause-specific Cox regression expanded model, adjusted for baseline comorbidities and left ventricular ejection fraction at discharge. cComposite endpoint of all-cause death, any recurrent myocardial infarction, and any revascularization. dDeath was adjudicated according to the Academic Research Consortium definition. eMyocardial infarction was adjudicated according to the World Health Organization extended definition. fComposite endpoint of cardiac death, target vessel myocardial infarction, target lesion revascularization, and stent thrombosis. gStent thrombosis was defined according to the Academic Research Consortium definition.

Landmark analyses

POCE landmark analysis showed no differences between the 2 groups across different time points. (figure 4A). Looking specifically at the various POCE individual components, the LAD-STEMI group exhibited a higher rate of all-cause mortality within the first year vs the non-LAD-STEMI group (p = 0.041), but this difference disappeared thereafter (figure 4B). Between years 0 and 1, there was also a trend toward a lower rate of myocardial infarction in the LAD-STEMI group vs the non-LAD-STEMI group (p = 0.081), which disappeared after year 1 (figure 4C). No differences were ever found regarding any revascularization (figure 4D) or other secondary endpoints between the 2 groups (figure 2 of the supplementary data).

Figure 4. Landmark analysis for the patient-oriented composite endpoint (A), all-cause mortality (B), myocardial infarction (C), and any revascularization (D) in patients stratified according to the culprit vessel. LAD, left anterior descending coronary artery; MI, myocardial infarction; STEMI, ST-segment elevation myocardial infarction; POCE, patient-oriented composite endpoint.

DISCUSSION

The main findings of this study can be summarized as follows: a) STEMI patients with LAD as the culprit vessel have a different baseline clinical profile vs STEMI patients with other culprit vessels; b) in the contemporary era of primary PCI, LAD as the STEMI-related culprit vessel did not bring worse very long-term outcomes compared with other coronary vessels; c) nevertheless, between years 0 and 1 the LAD-STEMI group exhibited a higher all-cause mortality rate, which disappeared thereafter compared with non-LAD-STEMI group.

Cardiology community knows (as reflected by the ESC guidelines on the management of acute coronary syndromes) that STEMI with LAD involvement as culprit vessel is a clinical marker of high risk of further events.1 LAD-related STEMI represents, approximately, 40% up to 50% of all STEMIs,12,16 and its worse prognosis has been related to the large myocardium covered by the LAD flow compared with the myocardium supplied by other coronary vessels. Of note, those studies were performed in the pre-reperfusion4-7 and early thrombolysis/PCI era,8,9 when PCIs were still not widely available. In the PCI era, there are very few studies (with short or mid-term follow-ups ranging from 1 to 3 years) reporting that LAD-STEMI is associated with an increased risk of stroke, heart failure, all-cause mortality10,17 and cardiovascular death11 after the PCI.

In our analysis, conducted in a cohort where the PCI was extensively performed, LAD as the STEMI culprit vessel did not appear to confer a worse prognosis to patients at the 1- or even 10-year follow-up. Of interest, LAD-STEMI patients exhibited the classical clinical features related to LAD, such as advanced Killip class at the time of presentation, lower ST-segment resolution and lower LVEF, which is similar to previous studies.8-11,17 All these unfavorable clinical characteristics are indeed related to the large amount of myocardium damaged in a LAD-STEMI with subsequent heart failure and ventricular arrhythmias.17-19 Nevertheless, this did not translate into a worse, very long-term clinical outcome. Significantly, even after accounting for variations in LVEF (which we addressed separately in our model due to its perceived role in the outcome cascade) the results showed no differences. This observation stands in contrast to earlier evidence, where the higher mortality rate in this cohort had been partially attributed to the subsequent decline in LVEF after STEMI.9,10

Several explanations may be claimed to understand our main finding. It may be hypothesized that worse outcome related to anterior STEMI may have been overcome by the introduction of the PCI with quick myocardial reperfusion. Pharmacological treatment has been also improved from thrombolysis to the PCI era, not only in terms of antiplatelet agents, but also in terms of secondary prevention (high intensity statins and angiotensin converting enzyme inhibitors/angiotensin receptor blockers or angiotensin receptor/neprilysin inhibitors for left ventricular dysfunction).20-23 Furthermore, in our study, the LAD-STEMI group had a higher proportion of active smokers. Smoking cessation remains the most critical preventive measure for coronary artery disease. The relationship between smoking and cardiovascular outcomes has been a matter of discussion, as some studies have suggested improved cardiovascular outcomes, even in the long term, among smokers who experienced STEMI.24 However, many of these studies were observational registries conducted in the pre-PCI era. Recent evidence indicates that smoking is associated with more post-PCI long-term adverse outcomes.25 Therefore, the so-called “smoker’s paradox” might be better explained by factors such as younger age and a lower prevalence of other risk factors among smokers. Indeed, in our study, while the LAD-STEMI group had a higher proportion of smokers, they had a lower prevalence of other risk factors, such as peripheral vascular disease and a history of prior PCI or MI.

Last, but not least, in landmark analysis we found that between years 0 and 1, all-cause mortality was more common in the LAD-STEMI group. Notably, in this period, there was a numerically higher number of cardiac deaths (although not statistically significant, P = .12), a similar finding to other existing evidence that found a higher relatively short-term mortality in the LAD-STEMI group within the first 30 days. In these studies, the elevated short-term mortality was associated with acute sequelae, such as heart failure and was also speculated to be connected to other lethal complications, such as ventricular arrhythmias, cardiogenic shock or mechanical complications.10,11 In our cohort, we found a trend towards a higher rate of reinfarction in the non-LAD-STEMI group (P = .081) that was largely unrelated to TLR, TVMI, or stent thrombosis. This observation contrasts with previous literature that reported a more common occurrence of reinfarction at the follow-up in patients with the SVG as the culprit vessel26 as well as the LAD,8 but not in LCx or the RCA.9-11

Our 10-year follow-up revealed similar clinical event rates between LAD-STEMI and non-LAD-STEMI group, indicating absence of long- term divergence. Previous studies showed a favorable post-acute phase prognosis for LAD-STEMI patients,10,11 which is consistent with our findings. In fact, non-cardiac factors seem to impact long-term mortality more than infarct location does.19 Thus, patients with STEMI should receive uniform management focused on secondary prevention strategies, regardless of the culprit vessel. Unfortunately, insufficient long-term data collection limits deeper insights into these outcomes (such as the presence of heart failure, optimal medical therapy, or other comorbidities).

Limitations

This study presents several limitations. First, this is a non-prespecified post-hoc analysis of the EXAMINATION-EXTEND study and therefore its conclusions must be considered only hypothesis generating. The association between infarction and outcomes may be driven by confounders which have not been recorded in the study. Then, several clinical and procedural characteristics were not available for the analysis, such as specified in-hospital or follow-up clinical data, like optimal medical treatment or compliance to medication at the follow-up.

CONCLUSIONS

In a contemporary cohort of STEMI patients, there were no differences in POCE between LAD as the STEMI-related culprit vessel and other vessels at the 10-year follow-up. However, within the first year after STEMI, all-cause death was more common in the LAD-STEMI group. Our results should be considered as hypothesis-generating. Further studies are needed to specifically assess the relationship between infarction location and outcomes in a contemporary setting where interventional and medical treatments are optimized.

FUNDING

The EXAMINATION-EXTEND study was funded by an unrestricted grant of Abbott Vascular to the Spanish Society of Cardiology (promoter). P. Vidal Calés has been supported by a research grant provided by Hospital Clínic at Barcelona, Spain.

ETHICAL CONSIDERATIONS

The study fully complied with the Declaration of Helsinki and was approved by our Institutional Review Committee. All patients signed a written informed consent form before being included in this study. The clinical ethics committee gave its approval for the analysis of the data collected. In this work, SAGER guidelines regarding sex and gender bias have been followed.

STATEMENT ON THE USE OF ARTIFICIAL INTELLIGENCE

No artificial intelligence tools were used during the preparation of this work.

AUTHORS’ CONTRIBUTIONS

The authors declare they meet the full criteria and requirements for authorship and have reviewed and agree with the content of the article. P. Vidal Calés, K. Bujak, R. Rinaldi, A. Salazar Rodríguez, S. Brugaletta and M. Sabaté contributed to conceptualization, design, data analysis and drafting of the manuscript. L. Ortega-Paz, J. Gómez-Lara, V. Jiménez-Diaz, M. Jiménez, P. Jiménez-Quevedo, R. Diletti, P. Bordes, G. Campo, A. Silvestro, J. Maristany, X. Flores, A. De Miguel-Castro, A. Íñiguez, A. Ielasi, M. Tespili, M. Lenzen, N. Gonzalo, M. Tebaldi, S. Biscaglia, R. Romaguera, J.A. Gómez-Hospital and P. W. Serruys reviewed and edited the manuscript.

CONFLICTS OF INTEREST

M. Sabaté declares he has received consulting fees from Abbott Vascular and iVascular outside the submitted work. R. Romaguera is associate editor of REC: Interventional Cardiology. The journal’s editorial procedure to ensure impartial handling of the manuscript has been followed. The rest of the authors declared no conflicts of interest whatsoever.

REFERENCES

INTRODUCTION

Currently, the number of percutaneous coronary interventions (PCI) involving moderate-to-severe calcified plaques is increasing due to a progressively aging population and extending procedural indications into more comorbid patients. The presence of such calcification is extremely relevant as it is strongly associated with worse outcomes, specially by means of stent underexpansion, a potent predictor of stent thrombosis or in-stent restenosis.1-3 Moreover, calcified plaques can make advancing the devices difficult and trigger stent deformation and entrapment, coronary artery dissection, or perforation.1,4-6. Currently, there is a growing interest in the assessment of plaque morphology and its modification prior to stent implantation, which has led to the development of multiple tools such as rotational atherectomy, lithotripsy, orbitational atherectomy, cutting balloons (CB) and scoring balloons (SB).7-11 The latter are easy to use and aim to create a controlled fracture of calcium deposits and plaque dilatation to facilitate stenting.12-14 However, despite their theoretical simplicity, these devices are bulky and stiff, making it difficult to cross the lesion at the first attempt, navigate the vessel, and recross the lesion once inflated. Therefore, there is a need for a more trackable and better-profiled SB to improve the uptake of these devices to treat calcified coronary artery disease.

The newly designed Naviscore SB (iVascular, Spain) seeks to address these drawbacks. Its structure is based on 125-µm thick nitinol laser cut filaments arranged in an axial pattern placed over a semi-compliant high-pressure balloon with a nominal pressure of 8 atm, a rated burst pressure of 20 atm, and a mean burst pressure of 26 atm (figure 1). A nylon compensation tube in the shaft helps to re-wrap during balloon deflation. The mechanical properties of nitinol tend to regain its original shape once the balloon has been deflated. The nylon compensation tube elongates once the balloon has been inflated and due to its elastic properties, it regains its original length when deflated (video 1 of the supplementary data). The 2 mechanisms produce a powerful re-wrapping of the entire system when the balloon has been deflated, regaining its original crossing profile, which allows for easy lesion recross and further dilatations as many times as required. Axial distribution of scoring elements provides a high push against calcified lesions. Nitinol elastic properties provide a better navigability through tortuous calcific vessels compared with rigid scoring elements, such as stainless steel. The durable hydrophilic coating of the Hydrax Plus catheter (iVascular, Spain) significantly reduces its coefficient of friction to 0.04 by increasing slip and navigability. Also, its axial design enables a far larger contact area with the vessel wall compared with other devices with spiral configuration of nitinol filaments such as the AngioSculpt catheter (Philips Healthcare, The Netherlands) (figure 2). In vitro testing (iVascular, Spain) was conducted to measure the crossing profile of different SBs using a non-contact laser meter where the profile is calculated through the shadow that has been created. This allows us to measure the profile without exerting any pressure on the device.15 The Naviscore crossing profile is 5% lower than Angiosculpt, and 31% lower than Wolverine (Boston Scientific, United States). This catheter is available in a wide range of measures from 1.5 mm up to 3.5 mm in diameter and from 6.0 mm up to 15 mm in length, all of them compatible with a 6-Fr guiding catheter.

Figure 1. Structure of the Naviscore SB. MBP, mean burst pressure; RBP, rated burst pressure.

Figure 2. In vitro model assessment of the AngioSculpt scoring surface (upper image) vs the Naviscore (lower image). The Naviscore scoring surface is 6 times larger than that of the AngioSculpt.

The present study aims to demonstrate the safety and efficacy profile associated with crossing and treating calcified coronary lesions with the Naviscore SB.

METHODS

The Naviscore first-in-man study is a multicentric and prospective registry that evaluated the device safety and efficacy profile in the treatment of calcified lesions in 85 patients from 10 centers (9 in Spain and 1 in Portugal), all from the Euro 4C Group, founded in 2018 and focused on the cardiac care of calcified and complex patients. All operators involved in this study were experts in the treatment of calcified coronary lesions and familiar with most tools designed to treat such lesions.

Inclusion criteria were the presence of de novo moderate-to-severe calcified lesions by angiographic criteria in the native coronary tree of patients with chronic coronary syndrome scheduled for a PCI due to symptom persistence despite optimal medical therapy and/or evidence of inducible ischemia. The only exclusion criterion was the presence of the patient’s hemodynamic compromise.

The study was designed to assess the safety and efficacy profile of Naviscore in terms of delivery success when used as the first device to dilate the lesion, plaque modification capabilities, and complications. Consequently, operators were asked to use the Naviscore in all cases as the first device to cross and dilate the lesion. However, in cases of failed lesion crossing, dilatation with a small balloon was recommended with subsequent re-use of the same Naviscore catheter.

Operators involved in the study had little prior experience with the Naviscore in, at least, 3 cases and were asked to include, at least, 5 patients in the study. The operators assessed the performance of the catheter in each procedure in terms of pushability, navigability, crossing, deflation time, re-wrap, recrossing capabilities and ease of retrieval, and made a subjective comparison with their routinely used SB or CB.

The baseline clinical characteristics were recorded prior to the procedure and angiographical and optical coherence tomography (OCT) images were analyzed separately by 2 different operators. Coronary angiography was performed using, at least, 2 orthogonal projections to show stenosis as it is commonly used in the routine clinical practice. The view with the most severe stenosis was selected for the quantitative analysis of the lesion before and after the PCI. Lesion calcification was angiographically categorized as none/mild, moderate (radiopacities were only noted during the cardiac cycle movement prior to contrast injection) or severe (radiopacities noted without cardiac movement prior to contrast injection involving both sides of the arterial lumen).16 Lesions were categorized as A, B1, B2 and C based on the modified ACC/AHA Task Force classification, which is in turn, based on the morphology and potential complexity of the PCI.17 Procedural success was defined as an angiographically residual percent diameter stenosis < 30% after stent implantation, absence of major complications and final Thrombolysis in Myocardial Infarction (TIMI) grade-3 flow.18 OCT analysis was performed as recommended in the routine clinical practice: lesion and proximal and distal references within 5 mm were used to estimate diameters and areas. Calcium cracks were defined as fissures involving a calcified region.19,20

Statistical analysis

Data are expressed as mean ± standard deviation for continuous variables with a normal distribution, median and interquartile range [IQR] for continuous variables with a non-Gaussian distribution, and counts and percentages for categorical data.

Statistical analyses were performed using the Stata software version 16.1 (College Station, TX, United States).

Ethical considerations

Informed consent was obtained from all the patients and the study was approved by the Research Ethics Committee. The authors declare that procedures were followed according to the regulations established by the Clinical Research and Ethics Committee and the Declaration of Helsinki of the World Medical Association.

RESULTS

From November 2021 through February 2022, a total of 85 patients—80% males—with a mean age of 71 ± 11 years were included in the present study. One center included a total of 21 patients and the remaining 9, between 5 and 10 patients each. Baseline patient and lesion characteristics are shown in table 1. Regarding comorbidities, the prevalence of diabetes mellitus, dyslipidemia, hypertension, and chronic kidney disease was 44%, 70%, 75%, and 18% respectively. Prior revascularization was present in 43% of the patients (PCI in 38% and coronary artery bypass graft in 16%). The left anterior descending coronary artery was the most common location of target lesions (41%), followed by the right coronary artery (28%), left circumflex artery (16%) and left main coronary artery (15%). Most lesions (87%) were categorized as type B2/C, 66% were severely calcified and 34% had moderate calcification by angiographic assessment. Chronic total occlusion was reported in 10% of treated lesions. Reference vessel diameter was 3.0 ± 0.5 mm; mean lesion length, 20.3 ± 9.4 mm; and diameter stenosis, 81.4 ± 12%.

Table 1. Baseline clinical and angiographic characteristics

Clinical and angiographic characteristics (n = 85)	n (%)
Age, years	71 ± 11
Male	68 (80%)
Diabetes	37 (44%)
Dyslipidemia	59 (70%)
Hypertension	67 (75%)
Chronic kidney disease	15 (18%)
Current/former smokers	53 (62%)
Prior PCI/CABG	37 (43%)
Type B2/C lesions	74 (87%)
Severe calcification	56 (66%)
Moderate calcification	29 (34%)
Basal percent diameter stenosis	81 ± 12%
Chronic total occlusion	8 (10%)
Lesion location: Left main coronary artery	13 (15%)
LAD	35 (41%)
RCA	24 (28%)
LCx	13 (16%)
CABG, coronary artery bypass graft; LAD, left anterior descending coronary artery; LCx, left circumflex artery; PCI, percutaneous coronary intervention; RCA, right coronary artery.

The Naviscore catheter diameters used to dilate the lesions were 2.0 mm (21%), 2.5 mm (38%), 3.0 mm (31%), and 3.5 mm (10%). Mean number of device inflations was 2.7 ± 1.5 times.

The Naviscore crossing performance of is shown in figure 3. Despite the strong recommendation to use Naviscore as the first device, some operators decided to use Rotablator or small balloons first in 10 patients due to severely narrowed and/or calcified vessels. In all those cases, the Naviscore successfully crossed and dilated the lesion after the first attempt. In the 75 patients in whom the Naviscore was used as the first device, the lesions were crossed and treated successfully in 57 (76%) of them. In the remaining 18 (24%) patients, the Naviscore crossed the lesion after pre-dilatation with a small balloon in 16 (89%) patients. Only 2 patients had non-crossable lesions.

Figure 3. Crossing performance of the Naviscore.

PCI results and in-hospital outcomes are shown in table 2. Procedural success was achieved in 94% of cases. The mean lesion percent diameter stenosis decreased from 81.4 ± 12% at baseline to 33.3 ± 8.5% after Naviscore dilatation, with a residual percent diameter stenosis of 7.5 ± 2.6% after stent implantation. There were no in-hospital major adverse cardiovascular events or any cases of perioperative perforation or device entrapment. Coronary dissections occurred in 13% of the cases (all of them type A or B) and resolved after stent implantation.

Table 2. Angiographic and in-hospital results

Angiographic and in-hospital clinical results (n = 85)	n (%)
Procedural success: residual percent diameter stenosis < 30% after stenting, absence of major complications and TIMI grade-3 flow	80 (94%)
Percent diameter stenosis pre-Naviscore	81 ± 12%
Percent diameter stenosis post-Naviscore	33 ± 8.5%
Percent diameter stenosis post-stenting	7.5 ± 2.6%
MACE (in-hospital)	0%
Death, MI, emergency CABG	0%
Perforation	0%
Limiting flow dissection	0%
Type A or B dissection	11 (13%)
Device entrapment	0%
CABG, coronary artery bypass graft; MACE, major adverse cardiovascular events; MI, myocardial infarction; TIMI, Thrombolysis in Myocardial Infarction.

Ten procedures were OCT-guided. Pre-dilatation analysis could only be performed in 5 lesions; the OCT catheter could not cross the remaining lesions. Four of those had Fujino’s scores8 of 4 and in 2 of them the nodules protruded into the lumen. After dilatation, all lesions exhibited dissections that covered the intima and the media. Fractures were seen in all calcified plaques, which were deeper and wider in non-nodular calcified regions. Enlargement of lumen area after treatment with the Naviscore and correct stent apposition and expansion was observed in all imaging-guided cases (figure 4).

Figure 4. Clinical examples of 2 different lesions treated with the Naviscore. Angiography and baseline OCT (A) after dilatation with the Naviscore catheter (B) and post-stent implantation (C). On the left side, panel A shows a severely stenotic fibrocalcific plaque on the left anterior descending coronary artery that Naviscore (B) modifies creating calcium fractures (*) and dissection (arrow) resulting in stent implantation with good apposition and expansion (C). The right side shows a severely calcified plaque on the right coronary artery (A) with an arc of calcium of 180º at its proximal edge (lateral OCT picture) and 360º at its distal edge (central OCT picture) that the Naviscore modifies (B) creating calcium fractures (*) and dissection (arrow) resulting in stent implantation with good apposition and expansion (C). OCT, optical coherence tomography.

Table 3 shows the subjective performance of the Naviscore as evaluated by the operators of the present study. The Naviscore performance including push, navigability, crossing, deflation time, re-wrap, recrossing capabilities and ease of retrieval was deemed superior to the Wolverine, NSE Alpha (Nipro Co. Ltd., Japan), AngioSculpt, and Scoreflex (OrbusNeich, China).

Table 3. Subjective performance of the Naviscore compared with traditional and scoring or cutting balloons of participant centers. Push, capacity to cross the lesion, deflation time, rewrap and recrossing capabilities were the most valued characteristics of the Naviscore

Parameter	Push	Navigability	Crossing	Friction	Device visibility	Deflation time	Rewrap	Recrossing capability	Ease of retrieving	Global evaluation
Better	60%	54%	63%	52%	34%	65%	65%	69%	53%	78%
Equal	38%	46%	35%	46%	65%	34%	34%	29%	46%	21%
Worse	2%	0%	2%	2%	1%	1%	1%	2%	1%	1%

DISCUSSION

Findings of this first-in-man registry with the new SB Naviscore in moderately to severely calcified coronary lesions performed in CHIP (complex and high-risk intervention in indicated patients) by highly experienced operators on this field can be summarized as follows: a) the Naviscore was able to cross the lesions as the first device in 3 out of 4 patients in such difficult scenario; b) this device proved to be effective to treat complex coronary lesions, with procedural success rates of 94%; c) the Naviscore was safe as no major dissection, perforation, or device entrapment were observed and d) the performance of the Naviscore SB was better compared with other commercially available SB and CB as subjectively assessed by the experienced operators in this study.

Calcified coronary lesions account for up to 30% of lesions scheduled for PCI and are associated with worse clinical outcomes.1 Furthermore, these lesions are probably the most challenging ones for PCI operators. Thus, it is of paramount importance to develop specific devices for this scenario.21-23 Although several plaque-modification techniques have appeared in recent years, there are not very many head-to-head comparisons, thus complicating the choice between them. In contrast, several treatment combinations and algorithms have been published.24-27 Ablation techniques, such as rotational or orbital atherectomy, are especially indicated in uncrossable or undilatable lesions with balloon catheters. However, since there are more potential complications and a steeper learning curve associated with these therapies, developing new tools with a better crossing profile would be very positive in this scenario. The good crossability of the Naviscore SB showed in the present study is probably related to its unique nitinol structure in an axial configuration. CB such as the Wolverine or the NSE Alpha have a similar axial configuration of their cutting elements. However, the crossing profile of the Naviscore is 31% smaller than the CB. Although a comparative study on the crossing capabilities of those devices is not available, such a different profile favors the superior crossing capabilities of the Navis­core device. In fact, the operators of the present registry highlighted the ability to cross and recross lesions as one of the best features of the device compared with their usual CB or SB. Compared with the AngioSculpt—a SB that shares a nitinol structure with the Naviscore— the helical configuration of nitinol filaments in front of the axial alignment of the Naviscore nitinol filaments can make a difference. Axial alignment adds push to the device through the lesion, while the helical nitinol configuration can deform the structure under friction, thus reducing its navigability and, in some cases, cause device entrapment.28 Furthermore, as shown in figure 2, helical distribution of nitinol significantly reduces the nitinol scoring surface in front of an axial distribution.

The efficacy of the Naviscore balloon has proven to be good in the present study, with a procedural success rate of 94%. Furthermore, quantitative angiographic analysis showed a reduction of basal stenosis from 81% to 31% after Naviscore dilatation and to 7.5 ± 2.6% after stent implantation. Finally, the OCT evaluation confirmed the presence of extensive calcium fractures caused by the scoring filaments (figure 4). As the balloon gradually inflates, the radial forces concentrate along the surface of the nitinol scoring elements, resulting in a more controlled balloon expansion, increasing the force of the nitinol frame filaments to break down the calcified plaques.29 In vitro experiments comparing a simple SB (Scoreflex) with a conventional balloon catheter to dilate concentric tubes of calcium revealed that the inflation pressure required to break down the calcium tubes was consistently lower with SB. Finite element analysis revealed that the first main stress applied to the calcified plaque was, at least, 3-fold higher when inflating the balloon catheter with scoring elements.30 Naviscore has the largest scoring surface in the SB current market, being 6 times more extensive than the AngioSculpt (figure 2). Pressure concentration of the scoring elements is the mechanism of the increased ability of SB to dilate calcified lesions and facilitate stent expansion. Residual stenosis after stent placement was 7.5 ± 2.6% in our study.

Finally, the Naviscore proved to be safe in the present study, which could be justified by the mechanism of action of the device that uses nitinol filaments as the anchor to avoid balloon slippage, and allows balloon controlled expansion, minimizing the risk of barotrauma, coronary dissection, and perforation. Using OCT imaging, SB broke down the calcified lesion without the undesirable dissection of noncalcified segments, thus allowing successful stent implantation with adequate expansion.29,30

Limitations

One limitation of the study is its own design as a registry and therefore, the absence of a randomized comparator. Instead, expert operators in the treatment of calcified coronary lesions were asked to subjectively compare the device at test with their commonly used SB or CB in terms of push, cross/recross, rewrap, navigability and time of deflation. The subjective nature of this assessment, while providing valuable information, could be a limitation.

Another limitation is the sample size of the study, especially the size of the population involved in the OCT imaging analysis. Unfortunately, in our setting, the use of this technique to analyze calcified lesions, although on the rise, is still far from what would be recommended. However, and despite the limitations in terms of number, the cases analyzed with intracoronary imaging homogeneously show us the effect of the device under study—calcium fractures, dissection and increase in luminal area—as well as the optimal stent expansion.

The small sample of patients in this study does not allow for a disaggregated analysis by sex to draw any valid results.

CONCLUSIONS

The Naviscore SB is a step ahead in this field with an innovative design using a nitinol frame with axial distribution of filaments placed over a high-pressure balloon to improve current SB or CB designs. This provides a strong pushing capability and flexibility to cross the most difficult calcified lesions in 3 out of 4 patients and easily navigate through tortuous anatomy. Superior scoring surface provides strong plaque modification capabilities by facilitating calcium fractures and controlled dissections, and ultimately, optimal stent expansion. Uniform and controlled balloon expansion and the anchor effect provided by the nitinol frame minimizes the risk of uncontrolled dissections and distal embolization, thus providing an outstanding safety profile, confirmed in this study by the absence of major adverse cardiovascular events, device entrapment or flow- limiting dissections. Therefore, the Naviscore can be considered as the front-line SB, either alone or in combination with atheroablative techniques in the treatment of moderate-to-severe calcified lesions.

FUNDING

This study was partly funded by iVascular, Barcelona, Spain, who provided the devices to perform the study.

ETHICAL CONSIDERATIONS

Informed consent was obtained from all patients and the study was approved by the Research Ethics Committee. The authors declare that the procedures were followed in full compliance with the regulations set forth by the Clinical Research and Ethics Committee and Declaration of Helsinki of the World Medical Association. In accordance with the regulations of the SAGER guidelines, the small sample of patients in this study does not allow for a disaggregated analysis by sex to draw any valid results.

STATEMENT ON THE USE OF ARTIFICIAL INTELLIGENCE

No artificial intelligence has been used in the development of this paper.

AUTHORS’ CONTRIBUTIONS

A. Serra Peñaranda designed the protocol, database and study outline, participated in data collection, coordinated data analysis and interpretation, and drafted the article. E. Fernández Peregrina participated in data collection, data analysis and interpretation and drafted the article. M. Jiménez Kockar participated in data collection, data analysis and interpretation, and performed the statistical analysis. B. García del Blanco, S. Romani, J. Martín-Moreiras, E. Pinar Bermúdez, A. Rodrigues, S. Ojeda, N. Gonzalo López, A. Regueiro and A. Serrador Frutos participated in data collection and critically revised the manuscript. All authors gave their final approval to the last version for publication.

CONFLICTS OF INTEREST

S. Ojeda is an associate editor of REC: Interventional Cardiology. The journal’s editorial procedure to ensure impartial handling of the manuscript has been followed. A. Serra Peñaranda and Ander Regueiro received consulting fees from iVascular, Barcelona, Spain. The remaining authors declared no conflicts of interest whatsoever.

REFERENCES