Abstract
Background:
In patients with severe coronary artery disease (CAD) requiring coronary revascularization, the prevalence of surgical ineligibility and its clinical effect on long-term outcomes remain unclear.
Methods and Results:
Among 15,939 patients with first coronary revascularization in the CREDO-Kyoto percutaneous coronary intervention (PCI)/coronary artery bypass grafting (CABG) registry cohort-2, we identified 3,982 patients with triple-vessel or left main disease (PCI: n=2,188, and CABG: n=1,794). Surgical ineligibility as documented in hospital charts was present in 142 (6.5%) of 2,188 PCI-patients, which was mainly related to comorbidities and advanced age. The cumulative 5-year incidence of the primary outcome measure (all-cause death/myocardial infarction/stroke) was much higher in PCI-patients with surgical ineligibility than in PCI-patients without surgical ineligibility and in CABG-patients (52.5%, 27.6%, and 24.0%, respectively, log-rank P<0.001). After adjusting for confounders, the excess risk of PCI-patients with surgical ineligibility relative to CABG-patients was substantial (hazard ratio [HR] 1.97, 95% CI 1.51–2.58, P<0.001), while the excess risk of PCI-patients without surgical ineligibility relative to CABG-patients was modest, but remained significant (HR 1.37, 95% CI 1.19–1.59, P<0.001).
Conclusions:
Among patients with severe CAD, PCI-patients with surgical ineligibility had worse long-term outcomes as compared with those without surgical ineligibility and CABG-patients.
The 5-year results of the SYNergy between percutaneous coronary intervention with TAXus and cardiac surgery (SYNTAX) randomized trial, which compared coronary artery bypass grafting (CABG) and percutaneous coronary intervention (PCI) for the treatment of severe coronary artery disease (CAD) (triple-vessel disease [TVD] or left main CAD [LMCAD]), demonstrated significantly better cardiovascular outcomes after CABG than after PCI even with the usage of drug-eluting stents (DES), especially in patients with high levels of coronary anatomic complexity.1
Other recent randomized trials also reported favorable outcomes with CABG relative to PCI in patients with TVD or LMCAD.2–4
Accordingly, current clinical guidelines recommend CABG as the standard of care for this population;5–7
however, those randomized trials enrolled only selected patients who were suitable for both treatment options of CABG and PCI. There is a scarcity of studies assessing the prevalence of patients ineligible for CABG and their long-term outcomes after PCI, although the number of patients who have to receive PCI because of surgical ineligibility might be increasing in the current aging society.8–10
Therefore, we sought to evaluate the prevalence of surgical ineligibility and its clinical effect on long-term outcomes in patients with TVD or LMCAD in real-world clinical practice.
Methods
Study Population
The Coronary Revascularization Demonstrating Outcome Study in Kyoto (CREDO-Kyoto) PCI/CABG registry cohort-2 is a physician-initiated, non-company-sponsored, multicenter registry enrolling consecutive patients undergoing their first coronary revascularization at 26 centers in Japan between January 2005 and December 2007.11
The relevant ethics committees in all participating centers (Supplementary Appendix 1) approved the study protocol. Because of the retrospective enrollment, the requirement for written informed consent from the patients was waived, however, we excluded patients who refused participation in the study when contacted for follow-up. This strategy is concordant with the guidelines of the Japanese Ministry of Health, Labor and Welfare.
Among 15,939 patients enrolled in the registry, we identified 3,982 patients with TVD or LMCAD (PCI: n=2,188, CABG: n=1,794) (Figure 1), excluding those who refused to participate (n=99), with concomitant non-coronary surgery (n=609), acute myocardial infarction (MI) presentation (n=4,892), and single- or double-vessel disease (n=6,357).
Definitions and Outcome Measures
Clinical, angiographic, and procedural data were collected from hospital charts or hospital databases according to the prespecified definitions by experienced clinical research coordinators from an independent clinical research organization (Research Institute for Production Development, Kyoto, Japan:
Supplementary Appendix 2). Patients were regarded as ineligible for surgery when the terms such as “contraindicated for surgery” or “too high risk for surgery” were documented in hospital charts. We did not take formal consultation with surgeons into account for adjudication on surgical ineligibility. Each patient with surgical ineligibility had at least 1 reason, and those reasons for surgical ineligibility were also evaluated. Patients without clear documentation of surgical ineligibility were regarded as eligible for surgery. The primary outcome measure of this study was a composite of all-cause death, MI, and stroke (death/MI/stroke) at 5 years. The secondary outcome measures included the individual components of the primary outcome measure as well as cardiac death, sudden cardiac death, non-cardiac death, major bleeding, any coronary revascularization, and hospitalization for heart failure (HF). Death was regarded as cardiac in origin unless obvious non-cardiac cause could be identified. Any death during the index hospitalization for coronary revascularization was regarded as cardiac death. MI was defined according to the definition in the Arterial Revascularization Therapy Study.12
Within 1 week of the index procedure, only Q-wave MI was adjudicated as MI. Stroke was defined as ischemic or hemorrhagic stroke occurring either during the index hospitalization or requiring hospitalization with symptoms lasting >24 h. Hospitalization for HF was defined as hospitalization for worsening HF requiring intravenous drug therapy. Bleeding was defined according to the Global Utilization of Streptokinase and Tissue plasminogen activator for Occluded coronary arteries (GUSTO) classification.13
GUSTO moderate or severe bleeding was adjudicated as major bleeding. Any coronary revascularization was defined as either PCI or CABG for any reason. Scheduled staged coronary revascularization procedures performed within 3 months of the initial procedure were not regarded as follow-up events but included in the index procedure. Follow-up data were obtained from hospital charts or by contacting patients, their relatives and/or referring physicians. Events such as death, MI, and stroke were adjudicated by the clinical event committee (Supplementary Appendix 3). The information on surgical ineligibility was collected from hospital charts at 5-year follow-up between January 2012 and June 2012.
Statistical Analysis
Categorical variables are presented as number and percentage and were compared by chi-square test across the 3 groups of patients comprising CABG-patients, PCI-patients with and without surgical ineligibility. Continuous variables are expressed as mean±standard deviation or median with interquartile range (IQR). Continuous variables were compared by ANOVA or Kruskal-Wallis test based on their distributions across the 3 groups. Cumulative incidence was estimated by the Kaplan-Meier method, and the differences were assessed with the log-rank test. We used Cox proportional hazard models to estimate the hazard ratios (HRs) and their 95% confidence intervals (CIs) of PCI-patients with and without surgical ineligibility relative to CABG-patients for the primary and secondary outcome measures, adjusting for the 30 clinically relevant factors listed in
Table 1. We assigned the dummy variables for PCI-patients with and without surgical ineligibility. The continuous variables were dichotomized by clinically meaningful reference values or median values. Proportional hazard assumptions for the risk-adjusting variables were assessed on the plots of log (time) vs. log [-log (survival)] stratified by the variable. The assumptions were verified to be acceptable for all the variables. Statistical analyses were performed with JMP 12.0 software (SAS Institute Inc., Cary, NC, USA). All statistical analyses were 2-tailed. P<0.05 was considered statistically significant.
Table 1.
Baseline Characteristics of Study Patients
| |
CABG
(n=1,794) |
PCI (n=2,188) |
P value |
Without surgical
ineligibility (n=2,046) |
With surgical
ineligibility (n=142) |
| Patients’ characteristics |
| Age (years) |
68.5±9.0 |
69.6±10.0 |
75.9±9.7 |
<0.001 |
| ≥75 years* |
513 (28.6) |
707 (34.6) |
86 (60.6) |
<0.001 |
| Men* |
1,334 (74.4) |
1,467 (71.7) |
87 (61.3) |
0.002 |
| BMI <25.0 kg/m2* |
1,276 (71.1) |
1,356 (66.3) |
112 (78.9) |
<0.001 |
| ACS |
167 (9.3) |
215 (10.5) |
17 (12.0) |
0.34 |
| Hypertension* |
1,512 (84.3) |
1,779 (87.0) |
126 (88.7) |
0.04 |
| Diabetes mellitus* |
933 (52.0) |
1,004 (49.1) |
61 (43.0) |
0.04 |
| On insulin therapy |
309 (17.2) |
262 (12.8) |
23 (16.2) |
<0.001 |
| Current smoker* |
436 (24.3) |
517 (25.3) |
23 (16.2) |
0.04 |
| HF* |
387 (21.6) |
408 (19.9) |
44 (31.0) |
0.009 |
| EF ≤40% |
217 (12.9) |
213 (12.0) |
19 (15.3) |
0.49 |
| MR grade ≥3/4* |
53 (3.0) |
119 (5.8) |
14 (9.9) |
<0.001 |
| Prior MI* |
396 (22.1) |
385 (18.8) |
30 (21.1) |
0.04 |
| Prior stroke (symptomatic)* |
248 (13.8) |
315 (15.4) |
31 (21.8) |
0.03 |
| Peripheral vascular disease* |
227 (12.7) |
241 (11.8) |
15 (10.6) |
0.60 |
| eGFR <30 mL/min/1.73 m2, without hemodialysis* |
139 (7.8) |
110 (5.4) |
12 (8.5) |
0.008 |
| Hemodialysis* |
118 (6.6) |
108 (5.3) |
14 (9.9) |
0.046 |
| Anemia (Hb <11.0 g/dL)* |
346 (19.3) |
316 (15.4) |
39 (27.5) |
<0.001 |
| Thrombocytopenia (platelets <100×109/L)* |
41 (2.3) |
28 (1.4) |
5 (3.5) |
0.04 |
| COPD* |
42 (2.3) |
61 (3.0) |
11 (7.8) |
0.006 |
| Liver cirrhosis* |
53 (3.0) |
64 (3.1) |
7 (4.9) |
0.48 |
| Malignancy* |
187 (10.4) |
220 (10.8) |
30 (21.1) |
0.002 |
| Lesion characteristics |
| TVD only |
1,154 (64.3) |
1,731 (84.6) |
93 (65.5) |
<0.001 |
| LMCAD, any |
640 (35.7) |
315 (15.4) |
49 (34.5) |
<0.001 |
| Left main only |
57 (8.9) |
30 (9.5) |
1 (2.0) |
<0.001 |
| Left main+single vessel |
108 (16.9) |
82 (26.0) |
7 (14.3) |
| Left main+double vessel |
182 (28.4) |
112 (35.6) |
19 (38.8) |
| Left main+triple vessel |
293 (45.8) |
91 (28.9) |
22 (44.9) |
| SYNTAX score |
29 (22–38) |
23 (17–30) |
27 (20–35) |
<0.001 |
| Low <23 |
411 (25.8) |
950 (47.3) |
45 (32.4) |
<0.001 |
| Intermediate 23–32 |
564 (35.4) |
721 (35.9) |
47 (33.8) |
| High ≥33 |
617 (38.8) |
337 (16.8) |
47 (33.8) |
| No. of target lesions or anastomoses |
3 (3–4) |
2 (1–3) |
2 (1–3) |
<0.001 |
| Target of proximal LAD* |
1,569 (87.5) |
1,264 (61.8) |
82 (57.8) |
<0.001 |
| Target of CTO* |
759 (42.3) |
436 (21.3) |
24 (16.9) |
<0.001 |
| Stent use |
|
1,947 (95.2) |
134 (94.4) |
0.68 |
| DES use |
|
1,497 (73.2) |
105 (73.9) |
0.84 |
| Total stents (n) |
|
2 (2–4) |
3 (2–4) |
0.36 |
| Total stent length (mm) |
|
53 (31–82) |
54 (31–87) |
0.57 |
| Internal thoracic artery use |
1,760 (98.1) |
|
|
|
| Off-pump |
1,139 (63.5) |
|
|
|
| Medications at hospital discharge |
| Thienopyridines |
181 (10.1) |
2,021 (98.8) |
140 (98.6) |
<0.001 |
| Aspirin |
1,767 (98.5) |
2,015 (98.5) |
139 (97.9) |
0.86 |
| Cilostazol* |
136 (7.6) |
219 (10.7) |
11 (7.8) |
0.003 |
| Statins* |
548 (30.6) |
1,058 (51.7) |
68 (47.9) |
<0.001 |
| β-blocker* |
474 (26.4) |
615 (30.1) |
51 (35.9) |
0.007 |
| ACEI/ARB* |
557 (31.1) |
1,134 (55.4) |
83 (58.5) |
<0.001 |
| Nitrates* |
621 (34.6) |
915 (44.7) |
59 (41.6) |
<0.001 |
| Calcium-channel blocker* |
911 (50.8) |
1,058 (51.7) |
66 (46.5) |
0.45 |
| Nicorandil* |
737 (41.1) |
527 (25.8) |
46 (32.4) |
<0.001 |
| Warfarin* |
673 (37.5) |
166 (8.1) |
14 (9.9) |
<0.001 |
| Proton pump inhibitor* |
733 (40.9) |
448 (21.9) |
47 (33.1) |
<0.001 |
| Histamine type-2 receptor blocker* |
605 (33.7) |
470 (23.0) |
32 (22.5) |
<0.001 |
Data are missing for EF in 398 patients, and for SYNTAX score in 243 patients. Categorical variables are presented as number (%), and continuous variables are presented as mean±standard deviation or median and 25–75th percentiles. Categorical variables were compared by chi-square test, and continuous variables were compared by ANOVA or Kruskal-Wallis test based on their distributions, across 3 groups of patients comprising CABG-patients, PCI-patients with and without surgical ineligibility. *Risk adjusting variables selected for Cox proportional hazard models. ACEI, angiotensin-converting enzyme inhibitor; ACS, acute coronary syndrome; ARB, angiotensin receptor blocker; BMI, body mass index; CABG, coronary artery bypass grafting; COPD, chronic obstructive pulmonary disease; CTO, chronic total occlusion; DES, drug-eluting stent; EF, ejection fraction; eGFR, estimated glomerular filtration rate; Hb, hemoglobin; HF, heart failure; LAD, left anterior descending artery; LMCAD, left main coronary artery disease; MI, myocardial infarction; MR, mitral regurgitation; PCI, percutaneous coronary intervention; SYNTAX, SYNergy between percutaneous coronary intervention with TAXus and cardiac surgery; TVD, triple-vessel disease.
Results
Study Population and Baseline Characteristics
Among the 3,982 patients with TVD or LMCAD, surgical ineligibility was determined in 142 (6.5%) of 2,188 patients treated with PCI. Therefore, the current study population consisted of 3 groups of patients: CABG-patients (n=1,794), PCI-patients without surgical ineligibility (n=2,046), and PCI-patients with surgical ineligibility (n=142) (Figure 1). PCI-patients with surgical ineligibility were significantly older, and more often had HF, anemia, chronic obstructive pulmonary disease, and malignancy than CABG-patients and PCI-patients without surgical ineligibility. Regarding coronary anatomic complexity, CABG-patients had greater numbers of target lesions, higher prevalence of involvement of proximal left anterior descending coronary artery and chronic total occlusion than PCI-patients with and without surgical ineligibility. In terms of PCI procedural characteristics, the rate of stent usage, and DES in particular, as well as total numbers of stents and total stent length were not significantly different between PCI-patients with and without surgical ineligibility. In terms of medications at hospital discharge, evidence-based medicines such as thienopyridines, statins, β-blockers, and angiotensin-converting enzyme inhibitor/angiotensin receptor blockers were more often prescribed for PCI-patients with and without surgical ineligibility than for CABG-patients (Table 1).
Reasons for Surgical Ineligibility
Reasons for surgical ineligibility were categorized as follows: comorbidities (67.6%), advanced age (37.3%), poor quality of distal vessels (8.5%), no graft material for anastomosis (1.4%), and other reasons (4.9%) (Figure 2A). Comorbidities included cerebrovascular disease (35.0%), malignancy (19.4%), lung disease (16.5%), immobile disease (14.6%), renal insufficiency (9.7%), congestive HF (6.8%), and peripheral artery disease (5.8%) in decreasing order (Figure 2B).
Long-Term Clinical Outcomes
Median follow-up duration was 5.1 (IQR: 4.2–5.9) years, and complete 1-, 3-, and 5-year clinical follow-up information was obtained in 97.8%, 95.9%, and 71.7% of patients, respectively. The cumulative 5-year incidence of the primary outcome measure (death/MI/stroke) was much higher in PCI-patients with surgical ineligibility than in both PCI-patients without surgical ineligibility and CABG-patients (52.5%, 27.6%, and 24.0%, respectively, log-rank P<0.001) (Figure 3). After adjusting confounders, the excess risk of PCI-patients with surgical ineligibility relative to CABG-patients for the primary outcome measure was substantial (HR 1.97, 95% CI 1.51–2.58, P<0.001) (Table 2), and was consistent both in the TVD (Supplementary Table 3, Supplementary Figure 1) and LMCAD subgroups (Supplementary Table 4, Supplementary Figure 2). The excess risk of PCI-patients without surgical ineligibility relative to CABG-patients for the primary outcome measure was modest, but remained significant (HR 1.37, 95% CI 1.19–1.59, P<0.001) (Table 2). The results for all-cause death, cardiac death, and sudden cardiac death followed the same trend as those for the primary outcome measure. PCI-patients with surgical ineligibility had significantly higher adjusted risk for non-cardiac death than CABG-patients, although it was not significantly different between PCI-patients without surgical ineligibility and CABG-patients. Regardless of the presence or absence of surgical ineligibility, PCI-patients as compared with CABG-patients were associated with significantly higher adjusted risk for MI, any coronary revascularization, and hospitalization for HF, neutral risk for stroke, and lower risk for major bleeding (Tables 2,3). PCI-patients as compared with CABG-patients were associated with significantly higher crude risk for major bleeding beyond 30 days after the procedure, although it was no longer significant after adjusting confounders (Table 3).
Table 2.
Primary Outcome and Its Individual Components
| Endpoint |
Total
patients (n) |
No. of patients with
event (cumulative
5-year incidence, %) |
Crude HR
(95% CI) |
P value |
Adjusted HR
(95% CI) |
P value |
| Death/MI/stroke |
| CABG |
1,794 |
462 (24.0) |
Ref. |
|
Ref. |
|
| PCI without surgical ineligibility |
2,046 |
618 (27.6) |
1.21 (1.07–1.36) |
0.002 |
1.37 (1.19–1.59) |
<0.001 |
| PCI with surgical ineligibility |
142 |
74 (52.5) |
2.62 (2.03–3.32) |
<0.001 |
1.97 (1.51–2.58) |
<0.001 |
| All-cause death |
| CABG |
1,794 |
341 (17.7) |
Ref. |
|
Ref. |
|
| PCI without surgical ineligibility |
2,046 |
437 (19.6) |
1.14 (0.99–1.31) |
0.07 |
1.31 (1.10–1.56) |
0.002 |
| PCI with surgical ineligibility |
142 |
65 (46.7) |
3.14 (2.40–4.09) |
<0.001 |
2.16 (1.61–2.89) |
<0.001 |
| Cardiac death |
| CABG |
1,794 |
141 (7.7) |
Ref. |
|
Ref. |
|
| PCI without surgical ineligibility |
2,046 |
197 (9.1) |
1.24 (1.00–1.54) |
0.049 |
1.41 (1.08–1.83) |
0.01 |
| PCI with surgical ineligibility |
142 |
29 (22.0) |
3.28 (2.16–4.82) |
<0.001 |
2.05 (1.31–3.19) |
0.001 |
| Sudden cardiac death |
| CABG |
1,794 |
41 (2.0) |
Ref. |
|
Ref. |
|
| PCI without surgical ineligibility |
2,046 |
68 (3.4) |
1.48 (1.01–2.19) |
0.046 |
1.70 (1.07–2.71) |
0.025 |
| PCI with surgical ineligibility |
142 |
12 (8.5) |
4.82 (2.42–8.89) |
<0.001 |
3.51 (1.70–7.23) |
<0.001 |
| Non-cardiac death |
| CABG |
1,794 |
200 (10.8) |
Ref. |
|
Ref. |
|
| PCI without surgical ineligibility |
2,046 |
240 (11.6) |
1.07 (0.89–1.29) |
0.48 |
1.23 (0.98–1.55) |
0.07 |
| PCI with surgical ineligibility |
142 |
36 (31.7) |
3.03 (2.13–4.32) |
<0.001 |
2.24 (1.52–3.31) |
<0.001 |
| MI |
| CABG |
1,794 |
54 (3.1) |
Ref. |
|
Ref. |
|
| PCI without surgical ineligibility |
2,046 |
138 (6.6) |
2.30 (1.68–3.15) |
<0.001 |
2.55 (1.74–3.71) |
<0.001 |
| PCI with surgical ineligibility |
142 |
12 (9.9) |
3.41 (1.83–6.39) |
<0.001 |
3.36 (1.72–6.56) |
<0.001 |
| Stroke |
| CABG |
1,794 |
158 (8.5) |
Ref. |
|
Ref. |
|
| PCI without surgical ineligibility |
2,046 |
168 (7.9) |
0.94 (0.76–1.17) |
0.57 |
1.00 (0.76–1.31) |
0.98 |
| PCI with surgical ineligibility |
142 |
12 (10.5) |
1.19 (0.63–2.05) |
0.57 |
0.93 (0.50–1.73) |
0.83 |
Numbers of patients with event were counted until the end of follow-up. HRs with 95% CIs of PCI-patients with and without surgical ineligibility relative to CABG-patients for the clinical endpoints were estimated throughout the entire follow-up period by Cox proportional hazard models. Cumulative 5-year incidence was estimated by the Kaplan-Meier method. CI, confidence interval; HR, hazard ratio. Other abbreviations as in Table 1.
Table 3.
Other Secondary Outcomes
| Endpoint |
Total
patients (n) |
No. of patients with
event (cumulative
5-year incidence, %) |
Crude HR
(95% CI) |
P value |
Adjusted HR
(95% CI) |
P value |
| Major bleeding |
| CABG |
1,794 |
1,014 (56.2) |
Ref. |
|
Ref. |
|
| PCI without surgical ineligibility |
2,046 |
284 (13.8) |
0.19 (0.17–0.22) |
<0.001 |
0.18 (0.15–0.21) |
<0.001 |
| PCI with surgical ineligibility |
142 |
28 (21.4) |
0.30 (0.20–0.42) |
<0.001 |
0.20 (0.13–0.29) |
<0.001 |
| Major bleeding beyond 30 days after procedure |
| CABG |
1,794 |
58 (6.3) |
Ref. |
|
Ref. |
|
| PCI without surgical ineligibility |
2,046 |
225 (11.2) |
1.78 (1.35–2.40) |
<0.001 |
1.14 (0.81–1.60) |
0.45 |
| PCI with surgical ineligibility |
142 |
16 (14.1) |
2.38 (1.32–4.04) |
0.005 |
1.04 (0.57–1.90) |
0.91 |
| Any coronary revascularization |
| CABG |
1,794 |
250 (14.3) |
Ref. |
|
Ref. |
|
| PCI without surgical ineligibility |
2,046 |
916 (47.0) |
4.01 (3.49–4.62) |
<0.001 |
4.30 (3.64–5.08) |
<0.001 |
| PCI with surgical ineligibility |
142 |
58 (48.6) |
4.42 (3.29–5.84) |
<0.001 |
5.26 (3.87–7.14) |
<0.001 |
| Ischemia-driven coronary revascularization |
| CABG |
1,794 |
106 (5.9) |
Ref. |
|
Ref. |
|
| PCI without surgical ineligibility |
2,046 |
454 (23.2) |
4.18 (3.39–5.17) |
<0.001 |
3.89 (3.03–5.00) |
<0.001 |
| PCI with surgical ineligibility |
142 |
23 (20.5) |
3.72 (2.37–5.84) |
<0.001 |
3.71 (2.31–5.96) |
<0.001 |
| Hospitalization for HF |
| CABG |
1,794 |
158 (9.0) |
Ref. |
|
Ref. |
|
| PCI without surgical ineligibility |
2,046 |
234 (11.7) |
1.33 (1.09–1.63) |
0.006 |
1.48 (1.15–1.90) |
0.002 |
| PCI with surgical ineligibility |
142 |
24 (20.1) |
2.44 (1.55–3.68) |
<0.001 |
1.87 (1.18–2.95) |
0.008 |
Numbers of patients with event were counted until the end of follow-up. HRs with 95% CIs of PCI-patients with and without surgical ineligibility relative to CABG-patients for the clinical endpoints were estimated throughout the entire follow-up period by Cox proportional hazard models. Cumulative 5-year incidence was estimated by the Kaplan-Meier method. Abbreviations as in Tables 1,2.
Discussion
The main findings of the present study were as follows: (1) cumulative 5-year incidence of death/MI/stroke in PCI-patients with surgical ineligibility was much higher than that in both PCI-patients without surgical ineligibility and CABG-patients; and (2) even after adjusting for confounders, the magnitude of excess risk for death/MI/stroke in PCI-patients with surgical ineligibility relative to CABG-patients was substantially larger as compared with that in PCI-patients without surgical ineligibility.
Randomized comparison is the scientifically solid approach to comparing CABG and PCI in patients with TVD or LMCAD. However, randomized controlled trials comparing CABG and PCI generally exclude patients with advanced age and/or severe comorbidities, who were often considered ineligible for surgery.8
The current analysis using a large-scale registry of patients undergoing first coronary revascularization demonstrated that PCI-patients with surgical ineligibility had extremely poor outcomes as compared with CABG-patients and PCI-patients without surgical ineligibility. Previous studies involving patients who had undergone PCI for multivessel disease or LMCAD also reported that surgically ineligible patients had significantly worse cardiovascular outcomes than those without surgical ineligibility, and furthermore, that surgical ineligibility was independently associated with increased mortality even after adjusting for various clinical risk factors listed in standard risk scores.9,10
Assessment of surgical risk scores such as the European System for Cardiac Operative Risk Evaluation (EuroSCORE) and the Society of Thoracic Surgeons (STS) score is helpful in predicting short-term operative mortality to some extent, but might be insufficient for determining whether patients are surgically ineligible or not, because of the lack of important variables such as frailty as a component of the risk score.14,15
We previously reported better long-term outcomes after CABG than after PCI in patients with TVD or LMCAD from the same PCI/CABG registry.16,17
However, the present study suggested that the extremely poor outcomes in PCI-patients with surgical ineligibility might have a substantial influence on the overall outcomes after PCI relative to CABG. The proportion of patients with surgical ineligibility in the present study (6.5%) was almost consistent with the SYNTAX trial in which 6.4% of patients were enrolled in the PCI registry.8
However, several studies have reported much higher proportions of patients with surgical ineligibility (54.5% in patients with LMCAD, and 21.5% in patients with multivessel disease or LMCAD),9,10
indicating that the prevalence of surgical ineligibility in the present study was underestimated because of being a retrospective analysis based on actual documentation in hospital charts. Therefore, prospective exclusion of patients with surgical ineligibility might be important to minimize the influence of selection bias when we compare CABG and PCI in future observational studies.
Study Limitations
There are several important limitations. First and most importantly, surgical ineligibility was identified retrospectively based on documentation in hospital charts. Underestimation of surgical ineligibility should be taken into account. Furthermore, it was unknown whether or not surgical ineligibility was appropriately evaluated by a heart team including both cardiac surgeons and interventional cardiologists. The assessment of surgical ineligibility without a heart team might lead to misclassification of patients as surgically ineligible, which could have some influence on the clinical outcomes of PCI-patients with and without surgical ineligibility. Second, selection bias and residual confounding are inevitable even in a comparison between PCI-patients without surgical ineligibility and CABG-patients. Third, surgical risk scores such as the EuroSCORE and the STS score were not available in this study. Fourth, the proportion of patients with complete or incomplete revascularization could not be evaluated in this study because there were no data on hemodynamic significance and ischemic burden of residual stenoses. Finally, the prescription rates of evidence-based medicines such as statins and β-blockers in the present study were lower as compared with the current standard. Considering the likelihood of less complete revascularization in PCI-patients with surgical ineligibility, their long-term outcomes might be improved by more stringent medical therapy, although the optimal treatment strategy for this population is still under debate, and it is unknown which is the best in patients with surgical ineligibility: complete revascularization, culprit-only revascularization, or medical therapy only.
Conclusions
In patients with severe CAD, PCI-patients with surgical ineligibility had worse long-term outcomes as compared with both PCI-patients without surgical ineligibility and CABG-patients.
Acknowledgments
We appreciate the support and collaboration of the co-investigators participating in the CREDO-Kyoto PCI/CABG registry cohort-2. We are indebted to the outstanding effort of the clinical research coordinators for data collection.
Sources of Funding
This study was supported by the Pharmaceuticals and Medical Devices Agency (PMDA) in Japan.
Disclosures
None of the authors has any conflict of interest to disclose.
Supplementary Files
Please find supplementary file(s);
http://dx.doi.org/10.1253/circj.CJ-19-0440
References
- 1.
Mohr FW, Morice MC, Kappetein AP, Feldman TE, Ståhle E, Colombo A, et al. Coronary artery bypass graft surgery versus percutaneous coronary intervention in patients with three-vessel disease and left main coronary disease: 5-year follow-up of the randomised, clinical SYNTAX trial. Lancet 2013; 381: 629–638.
- 2.
Farkouh ME, Domanski M, Sleeper LA, Siami FS, Dangas G, Mack M, et al. Strategies for multivessel revascularization in patients with diabetes. N Engl J Med 2012; 367: 2375–2384.
- 3.
Park SJ, Ahn JM, Kim YH, Park DW, Yun SC, Lee JY, et al. Trial of everolimus-eluting stents or bypass surgery for coronary disease. N Engl J Med 2015; 372: 1204–1212.
- 4.
Mäkikallio T, Holm NR, Lindsay M, Spence MS, Erglis A, Menown IBA, et al. Percutaneous coronary angioplasty versus coronary artery bypass grafting in treatment of unprotected left main stenosis (NOBLE): A prospective, randomised, open-label, non-inferiority trial. Lancet 2016; 388: 2743–2752.
- 5.
Fihn SD, Gardin JM, Abrams J, Berra K, Blankenship JC, Dallas AP, et al. 2012 ACCF/AHA/ACP/AATS/PCNA/SCAI/STS Guideline for the diagnosis and management of patients with stable ischemic heart disease: A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, and the American College of Physicians, American Association for Thoracic Surgery, Preventive Cardiovascular Nurses Association, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons. J Am Coll Cardiol 2012; 60: e44–e164.
- 6.
Fihn SD, Blankenship JC, Alexander KP, Bittl JA, Byrne JG, Fletcher BJ, et al. 2014 ACC/AHA/AATS/PCNA/SCAI/STS focused update of the guideline for the diagnosis and management of patients with stable ischemic heart disease: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines, and the American Association for Thoracic Surgery, Preventive Cardiovascular Nurses Association, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons. Circulation 2014; 130: 1749–1767.
- 7.
Neumann FJ, Sousa-Uva M, Ahlsson A, Alfonso F, Banning AP, Benedetto U, et al. 2018 ESC/EACTS Guidelines on myocardial revascularization. Eur Heart J 2019; 40: 87–165.
- 8.
Head SJ, Holmes DR Jr, Mack MJ, Serruys PW, Mohr FW, Morice MC, et al. Risk profile and 3-year outcomes from the SYNTAX percutaneous coronary intervention and coronary artery bypass grafting nested registries. JACC Cardiovasc Interv 2012; 5: 618–625.
- 9.
McNulty EJ, Ng W, Spertus JA, Zaroff JG, Yeh RW, Ren XM, et al. Surgical candidacy and selection biases in nonemergent left main stenting: Implications for observational studies. JACC Cardiovasc Interv 2011; 4: 1020–1027.
- 10.
Waldo SW, Secemsky EA, O’Brien C, Kennedy KF, Pomerantsev E, Sundt TM 3rd, et al. Surgical ineligibility and mortality among patients with unprotected left main or multivessel coronary artery disease undergoing percutaneous coronary intervention. Circulation 2014; 130: 2295–2301.
- 11.
Kimura T, Morimoto T, Furukawa Y, Nakagawa Y, Kadota K, Iwabuchi M, et al. Long-term safety and efficacy of sirolimus-eluting stents versus bare-metal stents in real world clinical practice in Japan. Cardiovasc Interv Ther 2011; 26: 234–245.
- 12.
Serruys PW, Unger F, Sousa JE, Jatene A, Bonnier HJ, Schonberger JP, et al. Comparison of coronary-artery bypass surgery and stenting for the treatment of multivessel disease. N Engl J Med 2001; 344: 1117–1124.
- 13.
GUSTO investigators. An international randomized trial comparing four thrombolytic strategies for acute myocardial infarction. N Engl J Med 1993; 329: 673–682.
- 14.
Nashef SA, Roques F, Sharples LD, Nilsson J, Smith C, Goldstone AR, et al. EuroSCORE II. Eur J Cardiothorac Surg 2012; 41: 734–744; discussion 744–745.
- 15.
Shahian DM, O’Brien SM, Filardo G, Ferraris VA, Haan CK, Rich JB, et al. The Society of Thoracic Surgeons 2008 cardiac surgery risk models. Part 1: Coronary artery bypass grafting surgery. Ann Thorac Surg 2009; 88: S2–S22.
- 16.
Shiomi H, Morimoto T, Furukawa Y, Nakagawa Y, Tazaki J, Sakata R, et al. Comparison of five-year outcome of percutaneous coronary intervention with coronary artery bypass grafting in triple-vessel coronary artery disease (from the Coronary Revascularization Demonstrating Outcome Study in Kyoto PCI/CABG Registry Cohort-2). Am J Cardiol 2015; 116: 59–65.
- 17.
Shiomi H, Morimoto T, Furukawa Y, Nakagawa Y, Sakata R, Okabayashi H, et al. Comparison of percutaneous coronary intervention with coronary artery bypass grafting in unprotected left main coronary artery disease: 5-year outcome from CREDO-Kyoto PCI/CABG Registry Cohort-2. Circ J 2015; 79: 1282–1289.
