Trends in Clinical Practice and Outcomes After Percutaneous Coronary Intervention of Unprotected Left Main Coronary Artery.

Background The use of percutaneous coronary intervention (PCI) to treat unprotected left main coronary artery disease has expanded rapidly in the past decade. We aimed to describe nationwide trends in clinical practice and outcomes after PCI for left main coronary artery disease. Methods and Results Patients (n=4085) enrolled in the SCAAR (Swedish Coronary Angiography and Angioplasty Registry) as undergoing PCI for left main coronary artery disease from 2005 to 2017 were included. A count regression model was used to analyze time-related differences in procedural characteristics. The 3-year major adverse cardiovascular and cerebrovascular event rate defined as death, myocardial infarction, stroke, and repeat revascularization was calculated with the Kaplan-Meier estimator and Cox proportional hazard model. The number of annual PCI procedures grew from 121 in 2005 to 589 in 2017 (389%). The increase was greater for men (479%) and individuals with diabetes (500%). Periprocedural complications occurred in 7.9%, decreasing from 10% to 6% during the study period. A major adverse cardiovascular and cerebrovascular event occurred in 35.7% of patients, falling from 45.6% to 23.9% (hazard ratio, 0.56; 95% CI, 0.41-0.78; P=0.001). Radial artery access rose from 21.5% to 74.2% and intracoronary diagnostic procedures from 14.0% to 53.3%. Use of bare-metal stents and first-generation drug-eluting stents fell from 19.0% and 71.9%, respectively, to 0, with use of new-generation drug-eluting stents increasing to 95.2%. Conclusions Recent changes in clinical practice relating to PCI for left main coronary artery disease are characterized by a 4-fold rise in procedures conducted, increased use of evidence-based adjunctive treatment strategies, intracoronary diagnostics, newer stents, and more favorable outcomes.

chronic coronary syndrome

Everolimus‐Eluting Stents or Bypass Surgery for Left Main Coronary Artery Disease

Kaplan‐Meier

Acute and Late Outcomes of Unprotected Left Main Stenting in Comparison With Surgical Revascularization

Major adverse cardiovascular and cerebrovascular event

Percutaneous Coronary Angioplasty Versus Coronary Artery Bypass Grafting in Treatment of Unprotected Left Main Stenosis: A Prospective, Randomised, Open‐Label, Non‐inferiority Trial

non–ST‐segment–elevation acute coronary syndrome

percutaneous coronary intervention of left main coronary artery disease

Randomized Trial of Stents Versus Bypass Surgery for Left Main Coronary Artery Disease

Swedish Coronary Angiography and Angioplasty Registry

Swedish Web‐System for Enhancement and Development of Evidence‐Based Care in Heart Disease Evaluated According to Recommended Therapies

ST‐segment–elevation acute coronary syndrome

Percutaneous Coronary Intervention Versus Coronary‐Artery Bypass Grafting for Severe Coronary Artery Disease

Clinical Perspective

What Is New

In an all‐comer nationwide population, the annual number of percutaneous coronary intervention procedures for unprotected left main coronary artery disease increased by ≈400%, while rates of periprocedural complications and major adverse cardiovascular and cerebrovascular events decreased by ≈40%.

What Are the Clinical Implications?

These findings support the current guideline recommendations endorsing percutaneous coronary intervention as a treatment option in patients with unprotected left main coronary artery disease.

Treatment of unprotected left main coronary artery disease with percutaneous coronary intervention (PCI‐LMCA) has increased rapidly during the past decade, owing to results of randomized trials showing comparable results of PCI and coronary artery bypass grafting (CABG). , , , , , In addition, improvements have been made in the field of coronary intervention. Stents have been refined by a gradual reduction in strut thickness and superior biocompatibility of the drug‐carrying polymer, resulting in a decreased incidence of in‐stent restenosis and stent thrombosis. Newer bifurcation techniques and the use of intravascular ultrasound as an adjunct diagnostic tool for stent sizing and detection of periprocedural complications are increasingly employed, , , , as is antithrombotic therapy with modern P2Y12 inhibitors and improved adherence to secondary prophylactic medications. Recent guidelines endorse PCI‐LMCA as a treatment option for patients with low anatomic complexity (Class Ia), while maintaining a class IIb and III recommendation for complex anatomies. ,

We aimed to quantify and describe time‐related changes in clinical practice and outcomes associated with PCI‐LMCA in a real‐world all‐comer patient population over a 12‐year period using the SCAAR (Swedish Coronary Angiography and Angioplasty Registry).

Methods

Data Sources

The SCAAR registry is part of the nationwide SWEDEHEART (Swedish Web‐System for Enhancement and Development of Evidence‐Based Care in Heart Disease Evaluated According to Recommended Therapies) registry, a national registry of the Swedish health authorities, receiving no commercial funding. The registry records all coronary angiographies and interventions in Sweden and describes each procedure with up to 250 variables. A unique personal identification number allows for longitudinal follow‐up of individuals undergoing a repeat angiography at any hospital in Sweden. Data were linked to the National Population Registry and National Patient Registry by the epidemiologic center of the Swedish National Board of Health and Welfare using the personal identification number to obtain censorship dates and death status for each individual. The authors had full access to the data, and the corresponding author takes responsibility for the analyses performed. The data set is legally restricted because of Swedish patient privacy and secrecy laws and the Uppsala University and Uppsala Clinical Research Center legal department. Data are available upon reasonable request to the Data Protection Officer at Uppsala County Council at landstinget@lul.se.

Study Design

The study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines and was approved by the Regional Ethical Review Board in Lund. The SCAAR registry as part of the SWEDEHEART registry is an anonymized quality registry, and patients are informed about their participation and their right to decline participation. Therefore, no informed consent is legally required for patient inclusion. The primary objective was to describe temporal trends of PCI‐LMCA in a nationwide all‐comer population with respect to angiographic characteristics; periprocedural treatment; use of PCI techniques such as radial versus femoral arterial access; and complete versus incomplete revascularization, intracoronary diagnostic procedures, periprocedural complications, and long‐term outcome. All patients with chronic coronary syndrome (CCS) and acute coronary syndrome (ACS) who underwent unprotected PCI‐LMCA from 2005 through 2017 were included. Exclusion criteria and flowchart are presented in Figure S1.

Outcomes

The primary outcome was a major adverse cardiovascular and cerebrovascular event (MACCE) within 3 years, defined as death from any cause, first occurrence myocardial infarction regardless of culprit lesion location, stroke, or repeat revascularization (target lesion revascularization with PCI or new CABG). Secondary outcomes were the independent components of the primary outcome along with in‐stent restenosis as identified on a subsequent coronary angiography as stenosis in a previously inserted stent. Vital status was obtained from the National Population Registry. Stroke information was obtained from the National Patient Registry and defined by the International Classification of Diseases Tenth Revision (ICD‐10) codes I60, I61, I62, I63, or I64. Target lesion revascularization was defined as repeat PCI in left main coronary artery disease and was assumed in patients undergoing CABG after PCI‐LMCA, with date of surgery obtained from the National Patient Registry. Data of target lesion revascularization treated with PCI, in‐stent restenosis, and new myocardial infarction were obtained from SCAAR. Myocardial infarction was defined according to the fourth universal definition and verified by coronary angiography. Records of periprocedural complications were obtained from SCAAR and defined as the composite of all complications including hypotension requiring vasoactive drugs, serious arrhythmia, neurological complications, perforation, cardiac tamponade, complications leading to emergency CABG, death in the catheterization lab, procedure‐related death, or any complication documented by the PCI operator occurring at the catheterization lab. All outcomes were ascertained up to April 1, 2018.

Statistical Analysis

A Poisson count regression model was used to assess temporal trends in PCI‐LMCA using calendar year as categorical variable and annual number of patients as outcome variable in the entire population as well as in selected subgroups. The subgroups of interest were sex, diabetes, age (<75 years versus ≥75 years), stable CCS, non–ST‐segment–elevation acute coronary syndrome (NSTE‐ACS), and ST‐segment–elevation acute coronary syndrome (STE‐ACS). Temporal trends in outcome were analyzed by estimating event rates with the Kaplan‐Meier (KM) estimator for each year of admission during the study period and hazard ratios with 95% CIs were estimated with Cox proportional hazard models using year of admission as a categorical, independent variable. Analyses were conducted on complete case data. The proportion of missing values is presented in Table S1. For descriptive purposes, patient characteristics were assessed in 2 time periods, before and after 2013. A 2‐sided P<0.05 was considered significant. All analyses were performed using STATA MP version 16.1 for Macintosh (StataCorp, College Station, TX).

Results

Patient Characteristics and Temporal Trends

A total of 4085 patients with PCI‐LMCA were included in the study. The median age of the study population was 74 years (interquartile range, 66–82), and 1165 (28.5%) patients were women (Table). A total of 948 (23.2%) patients presented with CCS; 2266 (55.5%) with NSTE‐ACS, and 871 (21.3%) with STE‐ACS (Table). A total of 323 patients (10.4%) presented with cardiogenic shock, which decreased from 19.7% before 2013 to 6.2% in 2013 to 2017. The number of patients with PCI‐LMCA grew from 121 in 2005 to 589 in 2017, a 386% increase (Figure 1A and Table S2). The increase was greater in men, from 76 to 440 (479%) compared with 45 to 149 (231%) in women (Figure 1A and Table S2); and in patients with diabetes (500%) compared with those without (379%). The increase was less pronounced in patients presenting with STE‐ACS (197%) compared with CCS (485%) and NSTE‐ACS (447%). No difference in number of procedures in age groups was found (Figure 1B and Table S2). A total of 2217 (54.3%) patients were discussed at a multidisciplinary heart team meeting, among whom 42.8% were declined from CABG before PCI (Table and Figure 1C). In remaining patients who were not declined from CABG, PCI was deemed the preferred option of revascularization. The proportion of patients declined from CABG decreased from 63.2% in 2005 to 31.5% in 2017, whereas the proportion of patients not declined from CABG but in whom PCI is preferred increased from 36.8% to 68.5%. Isolated left main coronary artery lesions were observed in 33% of patients. PCI‐LMCA, together with PCI of either proximal left anterior descending artery or circumflex artery, or both, was performed in 48% of patients (Figure 2).

Table  

Patient Characteristics

Baseline table	Total	Year 2012 or earlier	Year 2013 or later
	4085 (100.0%)	1584 (38.8%)	2501 (61.2%)
Variable
Age, y, median (IQR)	74.0 (66.0–82.0)	75.0 (66.0–82.0)	74.0 (67.0–81.0)
Body mass index	25.9 (23.8–28.7)	25.7 (23.7–28.4)	26.1 (23.9–29.0)
Male, n (%)	2920 (71.5)	1111 (70.1)	1809 (72.3)
Female, n (%)	1165 (28.5)	473 (29.9)	692 (27.7)
Smoking status, n (%)
Never smoked	1695 (46.5)	658 (48.3)	1037 (45.4)
Previous smoker	1445 (39.6)	498 (36.6)	947 (41.5)
Current smoker	505 (13.9)	205 (15.1)	300 (13.1)
Medical history, n (%)
Diabetes	938 (23.3)	344 (22.3)	594 (23.9)
Hypertension	2847 (71.9)	959 (63.8)	1888 (76.9)
Hyperlipidemia	2481 (62.9)	882 (58.9)	1599 (65.3)
History of myocardial infarction, n (%)	1346 (34.1)	517 (34.6)	829 (33.8)
History of PCI	1166 (28.6)	337 (21.3)	829 (33.1)
Stroke	452 (11.1)	190 (12.0)	262 (10.5)
Chronic heart failure	484 (11.8)	209 (13.2)	275 (11.0)
Renal failure	231 (5.7)	73 (4.6)	158 (6.3)
In‐hospital characteristics, median (IQR)
Creatinine, μmol/L	88.0 (74.0–108.0)	90.0 (75.0–111.0)	87.0 (73.0–107.0)
Estimated glomerular filtration rate–MDRD4, (mL/min per 1.73 m2	73.3 (56.0–90.0)	71.5 (54.8–87.0)	74.5 (56.9–90.9)
Killip class, n (%)
Killip I	2458 (79.1)	646 (67.3)	1812 (84.3)
Killip II	222 (7.1)	79 (8.2)	143 (6.7)
Killip III	106 (3.4)	46 (4.8)	60 (2.8)
Killip IV	323 (10.4)	189 (19.7)	134 (6.2)
Indication for angiography, n (%)
Chronic coronary syndrome	948 (23.2)	306 (19.3)	642 (25.7)
Non–ST‐segment–elevation ACS	2266 (55.5)	841 (53.1)	1425 (57.0)
ST‐segment–elevation ACS	871 (21.3)	437 (27.6)	434 (17.4)
Office/duty hours—angiography, n (%)
Planned—office hours	1254 (31.8)	475 (31.1)	779 (32.3)
Acute—office hours	483 (12.2)	272 (17.8)	211 (8.7)
Acute—duty hours	736 (18.7)	325 (21.3)	411 (17.0)
Subacute—office hours	1264 (32.0)	417 (27.3)	847 (35.1)
Subacute—duty hours	207 (5.2)	40 (2.6)	167 (6.9)
Vascular approach, n (%)
Femoral artery	1500 (36.7)	834 (52.7)	666 (26.6)
Radial artery	2392 (58.6)	685 (43.3)	1707 (68.3)
Combined/other	191 (4.7)	63 (4.0)	128 (5.1)
Treatment before angiography, n (%)
Clopidogrel/Ticlopidin	1966 (48.2)	1204 (76.2)	762 (30.5)
Prasugrel	49 (1.2)	33 (2.1)	16 (0.6)
Ticagrelor	1350 (33.0)	88 (5.6)	1262 (50.5)
Aspirin	3803 (93.3)	1448 (91.7)	2355 (94.3)
Heparin	3327 (81.4)	1137 (71.8)	2190 (87.6)
Bivalirudin	803 (19.9)	435 (28.2)	368 (14.7)
Glycoprotein IIB/IIIA, within 24 h	448 (11.0)	305 (19.3)	143 (5.7)
Discussed on multidisciplinary heart team	2217 (54.3)	913 (57.6)	1304 (52.1)
Not declined CABG but PCI preferred*	1269 (57.2)	455 (49.8)	814 (62.4)
Declined CABG*	948 (42.8)	458 (50.2)	490 (37.6)
Ad hoc PCI (not discussed)	1471 (36.0)	561 (38.1)	910 (61.9)
Stent diameter, mm	4.0 (3.5–4.0)	3.5 (3.5–4.0)	4.0 (3.5–4.5)
Stent length, mm	16.0 (12.0–23.0)	16.0 (12.0–20.0)	18.0 (14.0–24.0)
Stent pressure inflation, kPa	20.0 (18.0–20.0)	20.0 (18.0–21.0)	20.0 (18.0–20.0)
Number of stents implanted, n (%)
0	242 (5.9)	126 (8.0)	116 (4.6)
1	1462 (35.8)	626 (39.5)	836 (33.4)
2	1194 (29.2)	464 (29.3)	730 (29.2)
3	636 (15.6)	199 (12.6)	437 (17.5)
≥4	551 (13.5)	169 (10.7)	382 (15.3)

ACS indicates acute coronary syndrome; CABG, coronary artery bypass graft; IQR, interquartile range; MDRD4, Modification of Diet in Renal Disease 4; and PCI, percutaneous coronary intervention.

Proportion of patients discussed on multidisciplinary heart team.

Figure 1

Temporal trends in demographics, clinical presentation, and treatment in patients with LMCA treated with PCI.

Temporal trends in PCI‐treated unprotected left main coronary artery disease by (A) patient characteristics; (B) clinical presentation; (C) multidisciplinary heart team decision; (D) periprocedural treatment; (E) PCI techniques and anatomic/physiological diagnostic procedures; and (F) stent details. All panels but panel C show absolute number of patients per year. Figure 1C shows the proportion of patients declined from CABG and those that were not declined but in whom PCI was preferred by number of patients discussed at a multidisciplinary heart team meeting, whereas the proportion of patients undergoing ad hoc PCI is presented as a proportion of all PCI‐LMCA. BMS indicates bare‐metal stent; CCS, chronic coronary syndrome; DES, drug‐eluting stent; DM, diabetes mellitus; FFR/iFR, fractional flow reserve/instant wave‐free ratio; GPIIB/IIA, glycoprotein IIB/IIA; IVUS, intravascular ultrasonography; NSTE‐ACS, non–ST‐segment–elevation acute coronary syndrome; OCT, optical coherence tomography; PCI‐LMCA, percutaneous coronary intervention for unprotected left main coronary artery disease; and STE‐ACS, ST‐segment–elevation acute coronary syndrome.

Figure 2

Anatomic pattern of LMCA lesions treated with PCI.

(A) Proportion of anatomic locations of coronary artery lesions of the studied cohort and (B) their classifications according to American College of Cardiology/American Heart Association. Cx indicates circumflex artery; LAD, left anterior descending artery; LM, left main; LMCA indicates left main coronary artery disease; and PCI, percutaneous coronary intervention.

Periprocedural, Diagnostic, and Therapeutic Procedures

During the study period, treatment with the potent P2Y12 inhibitors ticagrelor/prasugrel became more common and use of clopidogrel declined (Figure 1D). Periprocedural heparin use increased, while provision of bivalirudin and glycoprotein IIB/IIIA inhibitor remained relatively consistent. The proportion of radial access grew from 21.5% in 2005 to 74.2% in 2017 (Figure 1E). Intracoronary diagnostic procedures (intravascular ultrasound, optical coherence tomography, or instant wave‐free ratio/fractional flow reserve) increased from 14.0% in 2005 to 53.3% in 2017. The use of intravascular ultrasound increased from 9.1% to 29.4%; optical coherence tomography from 1.3% when first introduced in 2010 to 18.7%, and instant wave‐free ratio/FFR from 5.0% to 12.8%. Bare‐metal stents and first‐generation drug‐eluting stents, used in 19.0% and 71.9% of cases in 2005, respectively, were phased out by 2017, replaced with new‐generation drug‐eluting stents (95.2%) (Figure 1F).

Periprocedural Complications and Outcome

Periprocedural complications occurred in 322 (7.9%) patients (Figure 3). The most common complication was death in the catheterization lab, which occurred in 117 (2.9%) patients, with an additional 0.7% procedure‐related deaths. The rate of periprocedural complications increased from 10% in 2005 to 16% in 2008 before falling steadily to 6% in 2017 (Figure 3). The 3‐year KM event rate for MACCE was 35.7% (1339); death 28.2% (1058); target lesion revascularization with PCI 4.0% (131); new CABG 2.5% (75); stroke 2.2% (66); myocardial infarction 3.5% (110); and in‐stent restenosis 1.5% (47) (Figure 4A and Table S3). The MACCE rate was higher in women, 39.3% (429) compared with 34.2% (910) in males (log‐rank P<0.001) (Figure 4B and Table S3). Patients aged ≥75 years had a significantly higher MACCE event rate of 42.6% (789) compared with 28.8% (550) for <75 (log‐rank P<0.001) (Figure 4C and Table S3). Similarly, 44.9% (386) of patients with diabetes experienced a MACCE compared with 32.1% (908) of those without diabetes (log‐rank P<0.001) (Figure 4D and Table S3). The MACCE rate for patients presenting with STE‐ACS was 57.9% (487) compared with 34.7% (708) for NSTE‐ACS and 17.4% (144) for patients with CCS (log‐rank P<0.001) (Figure 4E and Table S3). The KM event rate for patients treated with a bare‐metal stent was 56.1% (281), 36.9% (153) for old drug‐eluting stents, and 30.2% (775) for new‐generation drug‐eluting stents (log‐rank P<0.001) (Figure 4F). The KM event rates for 3‐year MACCE remained stationary from 2005 to 2010 and fell thereafter from 40.5% (95) in 2010 to 23.9% (114) in 2017 (Figure 5 and Table S4). The overall reduction in MACCE from 2005 to 2017 was 44% (hazard ratio, 0.56; 95% CI, 0.41–0.78; P=0.001). Decline in MACCE was observed in all subgroups but was not significant in women (35.8%–32.5%), age ≥75 years (42.8%–29.3%), diabetes (52.9%–29.8%), NSTE‐ACS (37.1%–22.6%), or STE‐ACS (66.7%–50.2%) (Table S4). Remaining results of the primary outcome are presented in Table S3 and Table S4.

Figure 3

Periprocedural complications in PCI‐LMCA.

A, Frequency of periprocedural complications. B, Temporal trends in periprocedural complications as 2‐year running average. CABG indicates coronary artery bypass grafting; PCI‐LMCA, percutaneous coronary intervention for unprotected left main coronary artery disease.

Figure 4

Kaplan‐Meier failure estimates of primary end point of PCI‐LMCA.

A, Cumulative incidence and Kaplan‐Meier event rates of the primary outcome of MACCE within 3 years defined as the first occurrence of all‐cause death, repeat revascularization (target lesion revascularization or CABG), stroke, or new myocardial infarction. B through F, Cumulative incidence of MACCE according to sex, age group, diabetes status, clinical presentation, and stent type. BMS indicates bare‐metal stent; CABG, coronary artery bypass grafting; CCS, chronic coronary syndrome; DES, drug‐eluting stent; MACCE, major adverse cardiovascular and cerebrovascular event; NSTE‐ACS, non–ST‐segment–elevation acute coronary syndrome; PCI‐LMCA, percutaneous coronary intervention for unprotected left main coronary artery disease; and STE‐ACS, ST‐segment–elevation acute coronary syndrome.

Figure 5

Temporal trends in long‐term outcome of PCI‐LMCA.

A, Three‐year risk of primary and secondary outcomes as 2‐year running average of the Kaplan‐Meier estimates. B, Three‐year risk of MACCE over time as 2‐year running average of the Kaplan‐Meier estimates. C, Three‐year Kaplan‐Meier event rates of the primary outcome together with hazard ratio and 95% CI by year. CCS indicates chronic coronary syndrome; DM, diabetes mellitus; MACCE, major adverse cardiovascular and cerebrovascular event; NSTE‐ACS, non–ST‐segment–elevation acute coronary syndrome; STE‐ACS, ST‐segment–elevation acute coronary syndrome; and TLR, target lesion revascularization.

The KM death rate decreased from 36.4% (44) in 2005 to 19.5% (98) in 2017 and was higher in women at 32.2% versus 26.5% in men (Table S3). While the mortality rate decreased significantly for men, from 39.5% in 2005 to 16.8% in 2017, the decline was less pronounced in women, 31.1% to 27.5%. Remaining secondary outcome measures, subgroup analyses, and temporal trends are presented in Table S3 and Table S4. Figure 6 illustrates landmark analysis at 30 days showing that nearly 45% of all MACCE occurred within 30 days of the procedure (570/1339), corresponding to a KM event rate of 14.0%. Figure 7 illustrates outcome by American Heart Association stenosis classification, with MACCE event rates ranging from 28.9% to 44.3%. Lesions classified as C or C bifurcations were associated with worse outcome.

Figure 6

Landmark analysis of MACCE at 30 days after PCI‐LMCA.

Cumulative incidence and Kaplan‐Meier event rates of the primary outcome major adverse cardiovascular and cerebrovascular event (MACCE) within and after 30 days. PCI‐LMCA indicates percutaneous coronary intervention for unprotected left main coronary artery disease.

Figure 7

Outcome according to American Heart Association (AHA) lesion complexity.

Lesions classified as C or C bifurcation according to the American Heart Association lesion classification were associated with highest incidence of MACCE within 3 years. MACCE indicates major adverse cardiovascular and cerebrovascular event.

Discussion

In this longitudinal nationwide population‐based study, we quantified changes in clinical practice and outcomes in patients with unprotected left main coronary artery disease treated with PCI from 2005 through 2017. The principal finding was a 4‐fold increase in PCI‐LMCA procedures conducted. This increase was greater in men and in patients with diabetes and was accompanied by a nearly 40% decrease in periprocedural complications and 3‐year MACCE risk.

Results of Previous Studies

The use of PCI‐LMCA has increased significantly since the introduction of new stents and the publication of the randomized clinical trials PRECOMBAT (Randomized Trial of Stents Versus Bypass Surgery for Left Main Coronary Artery Disease), NOBLE (Percutaneous Coronary Angioplasty Versus Coronary Artery Bypass Grafting in Treatment of Unprotected Left Main Stenosis: A Prospective, Randomised, Open‐Label, Non‐inferiority Trial), EXCEL (Everolimus‐Eluting Stents or Bypass Surgery for Left Main Coronary Artery Disease), SYNTAX (Percutaneous Coronary Intervention Versus Coronary‐Artery Bypass Grafting for Severe Coronary Artery Disease), and LE MANS (Acute and Late Outcomes of Unprotected Left Main Stenting in Comparison With Surgical Revascularization). The 5‐year MACCE rates found in these trials ranged from 17.5% to 36.9%, , , , and the 10‐year rate observed in the PRECOMBAT and LE MANS trials ranged from 29.8% to 52.2%. ,

In a real‐world setting, we observed a 3‐year MACCE event rate of 35.7%, which is higher than that reported in the available randomized studies. This difference can be attributed to several factors; for example, current guidelines recommend PCI as an alternative treatment in patients with low/intermediate lesion complexity. Although the SYNTAX score is not captured in the registry, only 33% of patients in this study exhibited an isolated left main coronary artery lesion, and nearly 50% presented with a bifurcation lesion or multivessel disease. As we included all consecutively treated patients in Sweden, advanced age and high prevalence of risk factors such as diabetes, hypertension, previous infarct, previous stroke, and previous PCI were common. It is therefore reasonable to assume that a proportion of patients with intermediate and high lesion complexity, likely rejected for surgery because of comorbidity burden, but in whom PCI was considered reasonable, were included. This is supported by the high proportion of patients with contraindications to surgery who were declined from CABG by the multidisciplinary heart team. In addition, a high proportion of patients presented with STE‐ACS (21.3%) and cardiogenic shock (10.4%), conditions that are associated with markedly higher MACCE rates, likely contributing to the higher rate of adverse outcomes, as patients presenting with these conditions had a MACCE rate approaching 60%.

Observational real‐world data of long‐term outcome after PCI‐LMCA are scarce, with most being limited in sample size or follow‐up time. Available studies report a 5‐year MACE rate of 34.4% (n=383, age 72.3±9.7 years) and 38.6% (n=421, age 68.4±11.5 years). A large study (n=11 264) reported a 1‐year death rate of 11.5%. Lee et al observed a 3‐year MACCE rate of 16.0% in 1658 individuals treated with new‐generation stents. To the best of our knowledge, the present study is the largest to quantify long‐term outcomes after PCI‐LMCA and the only conducted in a nationwide population‐based setting.

Temporal Trends

Over the course of the study, we observed more PCI‐LMCAs conducted, along with a concomitant improvement in outcomes. The more favorable outcomes are multifactorial and relate to progress in several areas associated with an increase in novel evidence‐based treatment strategies. The shift to radial artery access has reduced bleeding rates, improving short‐term outcome. The more potent P2Y12 blockers prasugrel and ticagrelor as antiplatelet therapy may contribute to the reduced frequency of new myocardial infarction and, together with improved stent design and delivery systems with thinner stent struts, biocompatible polymers, and more effective drug delivery systems, may explain the low frequency of angiographically verified restenosis and repeat revascularizations. , , The use of intracoronary diagnostic procedures is associated with more accurate stent sizing/apposition and, consequently, better outcomes and reduced incidence of stent thrombosis. Advances in PCI technique (not investigated in this study), , along with increased skill of PCI operators as a consequence of the expansion in number of procedures performed, could contribute to the reduction in periprocedural complications. Finally, it cannot be ruled out that more favorable outcomes can result from improved risk stratification and selection of patients referred for PCI. Patients treated at the end of the study period tended to have a greater number of comorbidities such as diabetes, hypertension, history of PCI, and renal failure but were generally younger and presented less often with cardiogenic shock, both factors predictive of a better outcome.

Knowledge Gaps

The lack of advancement in outcomes in women and patients of advanced age is of particular concern. Whether this is attributable to more complex lesions, poor risk stratification, or a shorter remaining lifespan warrants further investigation. We observed a significant increase of PCI‐LMCA conducted in individuals with diabetes, a subgroup of patients who, in general, benefit more from CABG than from PCI. In this group, the observation of a 50% higher MACCE is worrisome (Table S4, Figure 4 and Figure 5). Whether this may reflect reduced adherence to guideline recommendations or serious comorbidities putting these patients at high or prohibitively high risk in surgery needs to be investigated. Finally, procedures in patients presenting with NSTE‐ACS and CCS increased 4‐ to 5‐fold. Only those presenting with CCS showed a convincing trend of improved outcomes during the study. The reason for the static situation in those presenting with ACS warrants investigation. It is possible that patients surviving the procedure succumb to hemodynamic instability (all patients with cardiogenic shock presented with ACS) or, high frailty attributable to more advanced age (median, 76.0 years versus 70.5 years), increasing risk of major complications such as bleeding.

Limitations

We acknowledge some important limitations. The SCAAR does not record SYNTAX scores. Instead, the description of coronary lesions relies on a simple Composite Autonomic Severity Score diagram that records the degree of stenosis and lesion complexity. Lesions were reported as requiring treatment with PCI, bifurcations, and engagement of proximal segments of the left anterior descending artery and circumflex artery. The inability to accurately capture the SYNTAX score in our study, degree of calcification, and Medina classification renders our registry study difficult to directly compare to randomized controlled trials and registry studies using these definitions. In addition, the lack of a SYNTAX score makes it difficult to assess exact reason why PCI was performed as opposed to CABG. Finally, our scope was to investigate temporal trends in patients with PCI‐LMCA; hence, patients revascularized with CABG were not included in the analysis. Studies assessing temporal trends in revascularization with CABG could provide further insight into a possible shift in revascularization strategy with respect to patients with left main coronary artery treated with PCI.

Conclusions

The years 2005 through 2017 saw a 4‐fold expansion in PCI‐LMCA procedures conducted with an increase in implementation of evidence‐based treatment strategies including use of newer stents, recently developed anatomic and physiological diagnostic procedures, and advanced adjunctive pharmacological treatment, accompanied by a concomitant decline in periprocedural complications and improved long‐term outcome.

Sources of Funding

This work was supported by the Bundy Academy, the Märta Winkler foundation, and the Anna‐Lisa and Sven‐Eric Lundgren foundation for medical research. The sponsors were not involved in the study design; collection, analysis, or interpretation of data; writing of the manuscript; approving the manuscript; or the decision to submit the manuscript for publication.

Disclosures

Dr Angerås declares he has received speaker fee from Abbot (Boston, MA) and a research grant from Abbott. The remaining authors declare no support for the submitted work from any organization, no financial relationship in the past 3 years with any organization that might have an interest in the submitted work, and no other relationships or activities that could appear to have influenced the submitted work.

Tables S1–S5

Figure S1

Click here for additional data file.