Outcomes of Unprotected Left Main Percutaneous Coronary Intervention: A Single-Centre Experience.

INTRODUCTION: When the baseline anatomical syntax score-I (SxSI) is more than or equal to 33, percutaneous coronary intervention (PCI) of unprotected left main (UPLM) is discouraged and considered as high-risk of adverse cardiac events. We designed this study to compare the outcomes of UPLM-PCI between the low to intermediate-syntax score (SxSIlow/int.) group (defined as SxSI <33) and the high-syntax score (SxSIhigh) group (defined as SxSI more than or equal 33) with a subanalysis to explore the predictive role of intravascular images (IVI).
METHODS: The study is a retrospective observational analysis in a prospective cohort. The prospectively gathered data of consecutive patients, who were enrolled from January 2018 to December 2018, were analyzed at 1-year regarding the primary outcome of major adverse cardiovascular and cerebrovascular events (MACCE). An independent committee calculated the SxSI and reviewed all the events.
RESULTS: Baseline data of 2144 patients were considered for analysis. 1245 underwent PCI and 32 (2.57%) had PCI of UPLM; of these, 24 (75%) were in SxSIlow/int. group and 8 (25%) in SxSIhigh group. The mean SxSI was 20.3 ± 6.5 and 37.1 ± 4.9 for SxSIlow/int. and SxSIhigh groups, respectively. Compared with patients in the SxSIlow/int., patients in SxSIhigh group had a greater syntax score-II (<0.0001), which reflects the expected 4-year mortality after PCI by combining the SxSI and clinical variables.Interestingly, despite the fact that the proportion of subjects who sustained MACCE was almost three times more among the SxSIhigh compared to their SxSIlow/int. counterparts [1 (12.5%) vs. 1 (4.2%), P = 0.444], this difference was not statistically significant.It was noted that a higher proportion of patients in whom intravascular imaging (IVI) was not used to guide PCI sustained MACCE compared with that in whom IVI was utilized (2 [50%] vs. [0%], P = 0.012). There is a strong, negative, statistically significant association between the use of IVI and the occurrence of MAACE (Phi=-0.681), which reflects that the use of imaging is associated with significantly lower probability of developing MACCE.
CONCLUSION: As demonstrated in this real-world cohort, PCI of UPLM provides excellent outcome at one year regardless of the initial anatomical syntax score. Furthermore, it appears that IVI utilization during the index PCI procedure of UPLM is a potent and independent predictor of MACCE. Copyright:

INTRODUCTION

Approximately 4%–8% of coronary angiograms display significant unprotected left main (UPLM) coronary artery disease.[12] This diagnosis is of great clinical importance as severe narrowing of the left main coronary artery (LMCA) jeopardizes a large area of myocardium and increases the risk of major adverse cardiac events.[34] Indeed, depending on the anatomical dominance, around 70% to 100% of the left ventricular myocardium is supplied by the LMCA.[25] Hence, hemodynamically significant disease of the LM frequently precipitates lethal complications.[5]

Intervention for UPLM has been studied in several randomized controlled trials (RCTs) that have compared percutaneous revascularization to the gold-standard surgical revascularization.[678] The international guidelines are based mainly on the SYNTAX (Synergy Between Percutaneous Coronary Intervention with TAXUS and Cardiac Surgery) trial[9] with support from other less potent randomized trials that were considered underpowered to give a firm statement on the optimal revascularization approach including the LE MANS (Left Main coronary Artery Stenting) and PRECOMBAT (Bypass Surgery Versus Angioplasty Using Sirolimus-Eluting Stent in Patients With Left Main Coronary Artery Disease) trials.[1011] The lack of a solid answer triggered the conduct of two large RCTs analyzing the outcomes of coronary artery bypass graft (CABG) versus PCI for UPLM treatment, namely, EXCEL (Evaluation of Xience versus Coronary Artery Bypass Surgery for Effectiveness of Left Main Revascularization) and NOBLE (Nordic-Baltic-British Left Main Revascularization Study) trials, both of which have been recently published but not yet reflected in the guidelines.[1213]

The debate regarding best revascularization strategy for UPLM is still an open-ended discussion. In the past, CABG was recommended as the first strategy for UPLM disease.[13] Nevertheless, with notable advances in PCI techniques and stent technologies assisted with cutting-edge intravascular imaging (IVI) and powerful novel antiplatelet agents, PCI for UPLM has emerged as an acceptable non-inferior alternative treatment strategy for certain candidates as there seems to be no significant variation in terms of major adverse cardiac and cerebrovascular event (MACCE) incidence between PCI with deployment of drug-eluting stents (DES) and CABG.[1415]

Stenting of UPLM is currently endorsed by the latest European Society of Cardiology (ESC) guidelines as an acceptable revascularization approach, specifically when the lesion is ostial, mid shaft, and/or the anatomical SYNTAX score-I (SxSI) is < 33 (SxSIlow/int.).[16] PCI in SxSI of more than or equal 33 (SxSIhigh) identifies a subgroup of patients with a significantly high risk of adverse cardiac events independent of age, gender, clinical presentation, left ventricular systolic function, syntax score-II, and degree of revascularization. Hence, the ESC guideline discouraged PCI of UPLM for SxSIhigh group (Class III A).[1617]

We designed this retrospective observational study to compare one-year LM-PCI outcomes between SxSIlow/int. and SxSIhigh groups among patients treated according to the ESC guidelines at a state-of-the-art cardiac center with a sub-analysis to explore the predictive role of intravascular imaging (IVI) utilization in term of MACCE.

MATERIALS AND METHODS

The study is a retrospective observational analysis in a prospective cohort (32 of patients enrolled in “Bahrain Left Main registry” from January 2018 to December 2018). The study took place at Mohammed Bin Khalifa Cardiac Centre (MKCC), Riffa, Kingdom of Bahrain.

All patients who underwent percutaneous intervention for UPLM received rigorous consultation prior to PCI regarding the best revascularization strategy (in particular, the SxSIhigh group refusing CABG) by a dedicated heart-team at our institution consisting of invasive and non-invasive cardiologists, cardiac surgeons and cardiac anesthesiologists unless LM-stenting was deemed an emergency life-saving procedure. The local Ethics committee approved the study, and all patients gave written informed consent in compliance with the Declaration of Helsinki.

Exclusion criteria were patients' refusal of informed consent to participate in the registry or lack of capacity to consent or high probability of non-adherence to follow-up requirements.

Left main intervention

The access for the procedure whether radial or femoral and the PCI technique including plaque-debulking strategy and temporary mechanical circulatory support system were left for the discretion of the operator. All patients received antiplatelet therapy as per the ESC guidelines with close follow-up (at least twice) at outpatient clinic during the first year post intervention. The use of intravascular imaging to guide the procedure was highly encouraged (if applicable) based on the case scenario.

Intravascular images

Intravascular ultrasound (IVUS) was recommended over Optical Coherence Tomography (OCT), in view of its superiority in imaging the left main (LM) coronary artery, whereas OCT may result in suboptimal images when performed in large-caliber LM, ostial lesion or when renal function is impaired and minimum contrast usage is vital. After defining plaque burden, distribution and composition, IVUS clarified true lumen diameters to aid stent sizing and ensure better determination of proximal and distal healthy landing zones. Stent malapposition/underexpansion and angiographically unnoticed edge-dissection are all detectable by IVUS with a view to preventing stent failure and major future complications thus, a post LM-stenting IVUS-pullback is mandatory when it is safe to do so.

Clinical outcomes

All-cause death was defined as all cause cardiovascular and non-cardiovascular death, in addition to unwitnessed death and death of unknown cause post left main intervention. Stroke was defined as an acute episode of focal or global neurological dysfunction caused by brain, spinal cord, or retinal vascular injury as a result of hemorrhage or infarction.

Cerebrovascular events comprised stroke and transient ischemic attack (TIA). Target vessel revascularization (TVR) included any repeat left main coronary revascularization (either percutaneous or surgical). Myocardial infarction was defined based on the 2018 universal definition including peri-procedural myocardial infarction (MI) in patients with unstable angina (UA).

The primary endpoint of our study was MACCE defined as the composite of cardiac death, TVR, stroke, or non-fatal MI at one-year follow-up.

Statistical analysis

The data were analyzed using IBM Corp. Released 2019. (IBM SPSS Statistics for Windows, Version 26.0. Armonk, NY: IBM Corp) where the categorical variables (dichotomous and ordinal) were summarized calculating the categories' frequencies and percentages. Continuous variables were summarized by calculating the mean and the standard deviation (mean ±SD). The significance of the difference between any categories were examined by the Fisher's exact test armed with Monte Carlo simulation of 1000 samples to obtain a more precise estimate of the P value. To estimate the strength of the relationship (e.g., use of intravascular images and occurrence of MACCE), two directional measures of association were used: Phi and the Uncertainty coefficient.

For exploring the relationship between an independent categorical and a dependent continuous variable (e.g., syntax-1 score and serum Creatinine level), a Monte Carlo simulated Mann–Whitney test was used, considering the test assumptions and the sample size. The significance of any relationship or a difference was established using Alpha 0.05 as a cut-off point.

RESULTS

Baseline characteristics of the entire study population

Of the 1245 who underwent PCI in 2018 at our catheterization laboratories, 32 (2.57%) had PCI of UPLM; of these, 24 (75%) were in SxSIlow/int. group and 8 (25%) in SxSIhigh group. The major characteristics of both groups are shown in Table 1. Of note, SxSI score was 20.3 ± 6.5 and 37.1 ± 4.9 for SxSIlow/int. and SxSIhigh, respectively and as expected was statistically significant (P < 0.0001). Compared with patients in the SxSI low/int. group, patients in SxSIhigh group had a higher baseline creatinine (P = 0.028), higher incidence of hypertension (P = 0.030), as well as higher syntax score-II (SxSII) (P < 0.0001). The latter reflects that they had greater burden of comorbidities in addition to more complex coronary anatomy with a greater risk of 4-year mortality after PCI. SxSII was calculated by combining the SxSI score and clinical variables using the online calculation tool ().

Table 1

Baseline characteristics of the entire study population

Variables	Syntax-1 score (<33) (n=24)	Syntax-1 score (>33) (n=8)	P
Male gender, n (%)	17 (70.8)	8 (100)	0.103
Age (years), mean (±SD)	62.8 (±13.6)	53.6 (±12.8)	0.073
Smoking, n (%)	9 (75)	3 (25)	1.000
Hypertension, n (%)	18 (75)	2 (25)	0.030
Diabetes mellitus, n (%)	15 (62.5)	3 (37.5)	0.252
Dyslipidemia, n (%)	20 (83.3)	4 (50)	0.152
Malignancy, n (%)	1 (4.2)	0	1.000
COPD, n (%)	1 (4.2)	0	1.000
Stroke or TIA, n (%)	4 (16.7)	0	0.550
PVD, n (%)	0	0	1.000
Creatinine, mean (±SD)	80.6 (±24.3)	100.2 (±21.8)	0.028
eGFR, mean (±SD)	57.9 (±5.2)	58.1 (±3.4)	0.642
WBC, mean (±SD)	7.4 (±2.3)	13.3 (±8.9)	0.142
Hemoglobin, mean (±SD)	11.7 (±2)	12.6 (±1.5)	0.240
Total cholesterol, mean (±SD)	4.3 (±1.5)	4.1 (±1.6)	0.460
LDL, mean (±SD)	2.4 (±1)	2.7 (±1.5)	0.751
Triglyceride, mean (±SD)	1.8 (±1.2)	1.3 (±0.4)	0.277
ACS, n (%)	3 (12.5)	2 (25)	0.578
Syntax-1 score, mean (±SD)	20.3 (±6.5)	37.5 (±4.9)	<0.0001
Syntax-2 score, mean (±SD)	16.9 (±12.9)	34.1 (±6.5)	<0.0001
Ejection fraction, mean (±SD)	46.6 (±13.4)	36.8 (±14.3)	0.104
Number of stents, n (%)	0.96 (±0.3)	1 (±0)	0.909
IVI, n (%)	22 (91.7)	6 (75)	0.254
Cutting balloon, n (%)	4 (16.7)	0	0.335
Rotablation, n (%)	3 (12.5)	0	0.555
IABP, n (%)	2 (8.3)	1 (12.5)	1.000
Impella, n (%)	0	1 (12.5)	0.250
Inotropes, n (%)	1 (4.2)	2 (25)	0.147
Ventilator, n (%)	0	2 (25)	0.056
CPR, n (%)	0	1 (12.5)	0.250
Antithrombotic therapy, n (%)
Clopidogrel	12 (50)	3 (37.5)	0.648
Ticagrelor	11 (45.8)	5 (62.5)
OAC	1 (4.2)	0

ACS: Acute coronary syndrome, COPD: Chronic obstructive airway disease, CPR: Cardiopulmonary resuscitation, eGFR: Estimated glomerular filtration rate, IABP: Intra-aortic balloon pump, LDL: Low-density lipoprotein, OAC: Oral anticoagulant, PVD: Peripheral vascular disease, TIA: Transient ischemic attack, SD: Standard deviation, WBC: White blood cell, IVI: Intravascular images

There was a small and statistically non-significant variation between the two groups in terms of other baseline clinical characteristics, laboratory investigations, clinical presentation, lesion and procedural details, and antiplatelet therapy postintervention. There was a trend toward more ventilator use in SxSIhigh group but this was also statistically non-significant.

One-year outcome

The analysis of one-year all cause death, TVR, stroke or TIA, and non-fatal MI showed no significant differences between the two groups [Table 2]. Interestingly, despite the fact that the proportion of subjects who sustained MACCE was almost three times higher in SxSIhigh group when compared to their SxSIlow/int. counterparts (1 [12.5%] vs. 1 [4.2%], P = 0.444), this difference was not statistically significant [Figure 1].

Table 2

One-year outcomes for low to intermediatesyntax score versus high-syntax score groups

Outcome	Syntax-1 score (<33), n (%)	Syntax-1 score (>33), n (%)	P
MACCE	1 (4.2)	1 (12.5)	0.444
Death	0	0	-
TVR	1 (4.2)	0	1.000
Stroke or TIA	0	0	-
Nonfatal MI	0	1 (12.5)	0.250

MACCE: Major adverse cardiovascular and cerebrovascular event, MI: Myocardial infarction, TIA: Transient ischemic attack, TVR: Target vessel revascularization

Figure 1

Bar chart compares the proportions of patients who had major adverse cardiovascular and cerebrovascular events across the two patients' categories: SxSIlow/int. and SxSIhigh. There is no statistically significant difference (1 [4.2%] vs. 1 [12.5%], P = 0.444). This finding suggests that the occurrence of major adverse cardiovascular and cerebrovascular events is independent from the patients' SxSI score

Impact of using intravascular imaging (IVI) on one-year MACCE:

The overall use of IVI (IVUS or OCT) during left main intervention in our registry was exceptionally high (87.5%). A significantly higher proportion of patients in whom IVI was not used during the procedures sustained MACCE compared with those in which IVI was used (2 [50%] vs. [0%], P = 0.012). There is a strong, negative, statistically significant association between the use imaging and the development of MAACE (Phi=−0.681), which suggests that the use of imaging is associated [Figure 2] with significantly lower probability of developing MACCE.

Figure 2

Bar chart compares the proportions of patients who had major adverse cardiovascular and cerebrovascular events across the two patients' categories: whom coronary intravascular images were used versus those intravascular images not used to guide left main intervention (0 [0%] vs. 2 [50%], P = 0.012), which suggests that the use of images is associated with significantly low probability of developing major adverse cardiovascular and cerebrovascular events

It was noted that the positive predictive value of IVI for MACCE was significantly increased by 62.9% which reflects the importance of IVI in the context of LM coronary artery intervention [Table 3].

Table 3

The impact of intravascular images on 1-year major adverse cardiovascular and cerebrovascular events

Variable Use of IVI	Categories, n (%)	MACCE, n (%)	Phi	Uncertainty coefficient	P
Yes	28 (87.5)	0	−0.681	0.629	0.012
No	4 (12.5)	2 (50)

Statistically significant difference with Fisher’s exact test (χ2 =14.9, df=1, P=0.012) and Monte Carlo simulation of 1000 samples at Alpha 0.05. MACCE: Major adverse cardiovascular and cerebrovascular events, IVI: Intravascular images

DISCUSSION

PCI when performed in patients with UPLM and high complexity lesions (SYNTAX scores ≥33) is a controversial practice. The 2018 ESC guideline downgraded percutaneous revascularization in this category to Class-III A.[1517] In our study, there was no significant difference in terms of hard end points comparing SxSIhigh with SxSIlow/int. groups. The study proved that, with current phenomenal advances in the field of interventional cardiology, the anatomical complexity of coronary arteries is not solely the major determinant of clinical outcomes. Indeed, we had excellent outcomes with very rare adverse cardiac events thanks to state-of-the-art technologies involving stent engineering, potent antiplatelet agents, and cutting-edge intracoronary imaging including IVUS, OCT, and near-infrared spectroscopy-IVUS.

Our results were compatible with previous studies such as the MAIN-COMPARE (Revascularization for Unprotected Left Main Coronary Artery Stenosis: Comparison of Percutaneous Coronary Angioplasty versus Surgical Revascularization) trial, which showed that the three-year rate of all-cause death was less in IVUS-guided PCI compared with angiography-guided stenting (4.7% versus 16%; P < 0.05).[18] Qian et al. also showed that IVUS-guidance was associated with significantly lower incidence of the composite of all-cause death and MI (hazard ratio [HR]: 0.65; 95% confidence interval: 0.50–0.84; P = 0.00).[19] Gao et al. reached the same conclusion after propensity score matching, with lower MACCE in imaging guided PCI, mainly driven by a decrease in cardiac mortality and TVR.[20] In an analysis of four Spanish registries, De la Torre et al. found better three-year outcomes with IVUS-guided revascularization with a statistically significant reduction in mortality, TVR and MI.[21]

There have been several explanations for the benefits of IVUS-guidance. First, IVUS-guided PCI provided a more precise delineation of lesion severity and length. The Arterial Revascularization Trial (ART) showed that total stent length was an independent predictor of stent thrombosis at three years (HR 1.14, P = 0.0037).[22] Second, left main disease may not have recognizable reference segments; this greatly influences the evaluation of lesion severity and stenting strategy including stent size and length selection.[23] Third, angiographic MEDINA classification of left main bifurcation lesions is often deceptive; IVUS assessment provides a more accurate assessment of LMCA disease extension into the proximal left anterior descending artery (LAD) and/or left circumflex artery (LCX).[24]

Oviedo et al. found that IVUS analysis of the LAD/diagonal and left main bifurcations revealed that most lesions had diffuse plaques extending from the main vessel into the main branch at a much higher rate than detected angiographically.[2526] Thus, IVUS is useful in selecting the most favorable PCI technique and in getting the best acute post-procedure results; this then translates into better long-term outcomes. In our study, the concept of IVI-guided PCI was implemented for the majority of patients no single case of adverse events was reported in IVI group at one-year follow-up regardless of the anatomical complexity of epicardial coronary lesions.

While there are no large randomized clinical trials to assess whether IVI-guided UPLM-PCI is associated with better long-term clinical outcomes, convincing data from large cohorts suggest a long-term mortality benefit in patients undergoing IVUS-guided compared with angiography-guided UPLM-PCI.[27282930]

Our study had several limitations, including the use of single center data; MKCC (Mohammed bin Khalifa Cardiac Centre) is the only cardiac center with full-service catheterization laboratories in the Kingdom of Bahrain. Secondly, it was left to the operator's discretion whether to use IVUS or rely on angiography alone. Therefore, despite rigorous statistical adjustment, unmeasured confounders may have influenced the outcomes. The current study is limited also by being a retrospective analysis, although from a prospectively acquired data set. However, all patients received drug eluting stents and were treated according to the current guidelines, and thus, there is less risk of erroneous conclusions from allocation bias. Moreover, in view of the sample size, it was not feasible to divide this population into acute coronary syndrome (ACS) versus non-ACS LM-PCI as would be a reflection of daily practice and may influence the outcome, as utilization of imaging in the acute setting may not always be possible.

CONCLUSION

LM disease represents one of the most challenging situations for interventional cardiologists. However, as demonstrated in this real-world cohort, with tremendous advances in coronary stent engineering and guideline-based management backed up with IVI technology; percutaneous revascularization of UPLM provides for an excellent outcome regardless of the initial SxSI score. Furthermore, it appears that IVI utilization during intervention for UPLM is a potent and independent predictor of MACCE.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.