Data on the outcome of young patients after coronary artery bypass grafting (CABG) and percutaneous coronary intervention (PCI) are scarce. Data on 2,209 consecutive patients aged ≤50 years who underwent CABG or PCI were retrospectively collected from 15 European institutions. PCI and CABG had similar 30-day mortality rates (0.8% vs 1.4%, p = 0.27), late survival (at 5 years, 97.8% vs 94.9%, p = 0.082), and freedom from stroke (at 5 years, 98.0% and 98.0%, p = 0.731). PCI was associated with significantly lower freedom from major adverse cardiac and cerebrovascular events (at 5 years, 73.9% vs 85.0%, p <0.0001), repeat revascularization (at 5 years, 77.6% vs 92.5%, p <0.0001), and myocardial infarction (at 5 years, 89.9% vs 96.6%, p <0.0001) compared with CABG. These findings were confirmed in propensity score–adjusted and matched analyses. Freedom from major adverse cardiac and cerebrovascular events after PCI was particularly low in diabetics (at 5 years, 58.0% vs 75.9%, p <0.0001) and in patients with multivessel disease (at 5 years, 63.6% vs 85.1%, p <0.0001). PCI in patients with ST elevation myocardial infarction was associated with significantly better 5-year survival (97.5% vs 88.8%, p = 0.001), which was driven by its lower 30-day mortality rate (1.5% vs 6.0%, p = 0.017). In conclusion, patients aged ≤50 years have an excellent immediate outcome after either PCI or CABG with similar long-term survival when used according to the current clinical practice. PCI was associated with significantly lower freedom from myocardial infarction and repeat revascularization.
Coronary artery bypass grafting (CABG) seems to be more durable compared with percutaneous coronary intervention (PCI), particularly when multiple arterial grafts are used. However, no formal comparative analysis of these 2 treatment methods has been previously performed in patients with premature coronary artery disease. In this multicenter study, we sought to compare the indications, clinical characteristics, and immediate and late outcomes after PCI versus CABG in patients aged ≤50 years in contemporary clinical practice.
Methods
The Coronary aRtery diseAse in younG adultS study is a retrospective study performed by collecting data from 15 European centers of cardiac surgery and cardiology. This trial was designed by the principal investigator (FB) and study chair (KEJA), and the study protocol was approved by the institutional review board at each participating center. This study was registered at ClinicalTrials.gov (no. NCT01838746 ). The principal investigator (FB) had unrestricted access to the data after the database was locked, prepared all drafts of the manuscript, and made the decision to submit the manuscript for publication. Eligible study participants were patients aged 18 to 50 years who received a diagnosis of stable angina, unstable angina, silent ischemia, non–ST elevation myocardial infarction or ST elevation myocardial infarction (STEMI), and underwent PCI or CABG. Data on preoperative and procedural variables and on the immediate outcome was retrieved from patients’ records. Late events were recorded by contacting patients and their relatives and/or their general practitioner and by checking patients’ record for any event of interest. In countries with centralized referral pathway, only patients residing in the referral area were included in the study, and data on their late outcome were retrieved from patients’ records. The primary end point of this study was a composite of major adverse cardiac and cerebrovascular events (MACCE), including death from any cause, myocardial infarction, stroke, and repeat coronary revascularization during the follow-up. Secondary end points included the individual components of the primary end point: death from any cause, myocardial infarction, stroke, and repeat coronary revascularization. Furthermore, we evaluated need for de novo dialysis and reoperation for bleeding as secondary immediate adverse events.
Statistical analysis was performed using SPSS statistical software (version 20; IBM SPSS Inc., Chicago, Illinois). Continuous data are reported as mean and SD. Nominal variables are reported as counts and percentages. No attempt to replace missing data was made. Baseline clinical characteristics, procedural data, and immediate postoperative outcome for the 2 study groups were compared using Mann-Whitney test for continuous variables and the chi-square test or Fisher’s exact test for categorical variables, as appropriate. Logistic regression was used for multivariate analysis of immediate postoperative events. The Kaplan-Meier and Cox methods were used to estimate the late outcome of the 2 study groups and to evaluate the impact of baseline variables on late adverse events. We used a propensity score approach to account for baseline differences at admission between treatment groups. The propensity score was estimated by logistic regression including all baseline variables. The area under the receiver operating characteristic curve was used to represent the discriminatory ability of the regression model. Propensity score was used for risk adjustment as a covariate in multivariate analyses assessing all predefined outcome end points. Furthermore, one-to-one propensity score matching was performed using a caliper of 0.02 of the SD of propensity score, that is, 0.006. Sensitivity analyses of the outcome in subgroups of interest were performed as well. A p <0.05 was considered statistically significant.
Results
From January 2002 to December 2012, a total of 2,209 consecutive patients underwent either CABG (592 patients) or PCI (1,617 patients). The characteristics of these patients significantly differed in a number of baseline, operative, and postoperative variables ( Tables 1 and 2 ). The estimated propensity score for CABG versus PCI had an area under the receiver operating characteristic curve of 0.909 (95% confidence interval 0.895 to 0.923, Hosmer-Lemeshow test: p = 0.802), and the independent predictors were single-vessel disease, STEMI, dyslipidemia, left main stenosis, previous CABG, and previous myocardial infarction. The immediate outcome was good in both groups with rather low rates of mortality and morbidity with PCI yielding a significantly lower risk of stroke, need for dialysis, and major bleeding compared with CABG ( Table 3 ). PCI was associated with a trend toward a better survival at 5 years (97.8% vs 94.9%, p = 0.082, Figure 1 ) but significantly worse freedom from MACCE ( Figure 2 ), myocardial infarction, and repeat revascularization ( Figure 3 ) compared with CABG ( Table 3 ).
| Clinical Variables | PCI (n = 1617) | CABG (n = 592) | p-Value |
|---|---|---|---|
| Age (years) | 45.0 ± 4.5 | 45.9 ± 3.9 | <0.0001 |
| Women | 239 (14.8%) | 74 (12.5%) | 0.18 |
| Family history of coronary artery disease | 803 (50.7%) | 306 (52.7%) | 0.44 |
| Dyslipidemia | 573 (35.6%) | 399 (68.0%) | <0.0001 |
| Treatment for hypertension | 583 (36.1%) | 338 (57.5%) | <0.0001 |
| Smoker | 1142 (70.7%) | 370 (62.7%) | <0.0001 |
| Diabetes | 207 (12.8%) | 141 (23.9%) | <0.0001 |
| Extracardiac arteriopathy | 30 (1.9%) | 63 (10.7%) | <0.0001 |
| Estimated glomerular filtration rate (mL/min/1.73 m 2 ) | 99 ± 24 | 95 ± 42 | <0.0001 |
| Estimated glomerular filtration rate <60 mL/min/1.73 m 2 | 51 (3.2%) | 40 (6.9%) | <0.0001 |
| Congestive heart failure | 8 (0.5%) | 47 (8.0%) | <0.0001 |
| Prior transient ischemic attack or stroke | 24 (1.5%) | 10 (1.7%) | 0.73 |
| Prior myocardial infarction | 117 (7.3%) | 213 (36.3%) | <0.0001 |
| Prior percutaneous coronary intervention | 87 (5.4%) | 88 (15.0%) | <0.0001 |
| Prior coronary artery bypass grafting | 11 (0.7%) | 2 (0.3%) | 0.53 |
| Indication | <0.0001 | ||
| Stable angina pectoris | 278 (17.2%) | 231 (39.5%) | |
| Unstable angina pectoris | 161 (10.0%) | 165 (28.2%) | |
| Non–ST-elevation myocardial infarction | 437 (27.1%) | 106 (18.1%) | |
| ST-elevation myocardial infarction | 737 (45.7%) | 83 (14.2%) | |
| Urgency | <0.0001 | ||
| Elective | 430 (26.7%) | 272 (46.0%) | |
| Urgent | 798 (49.5%) | 284 (48.1%) | |
| Emergency | 383 (23.8%) | 35 (5.9%) | |
| Left main stenosis | 19 (1.2%) | 136 (23.4%) | <0.0001 |
| Left ventricular ejection fraction ∗ | <0.0001 | ||
| >50% | 729 (60.2%) | 402 (69.0%) | |
| 30–50% | 470 (38.8%) | 162 (27.8%) | |
| <30% | 12 (1.0%) | 19 (3.3%) | |
| No. narrowed coronary arteries | 1.4 ± 0.7 | 2.6 ± 0.6 | <0.0001 |
| 1 | 1079 (66.8%) | 48 (8.1%) | |
| 2 | 386 (23.9%) | 152 (25.7%) | |
| 3 | 150 (9.3%) | 392 (66.2%) |
| Operative Variables | PCI (n = 1617) | CABG (n = 592) | p-Value |
|---|---|---|---|
| Coronary artery bypass grafting | |||
| At least one internal mammary a. graft | — | 576 (97.3%) | — |
| Both internal mammary a. grafts | — | 87 (14.7%) | — |
| Radial artery graft | — | 93 (15.8%) | — |
| At least two arterial grafts | — | 163 (27.5%) | — |
| Three arterial grafts | — | 16 (2.6%) | — |
| No. of distal anastomoses | — | 3.0 ± 1.1 | — |
| Revascularized coronary artery | |||
| Left anterior descending artery | — | 576 (97.3%) | — |
| Circumflex artery | — | 467 (78.9%) | — |
| Right coronary artery | — | 428 (72.3%) | — |
| Percutaneous coronary intervention | |||
| At least one drug eluting stent | 818 (50.9%) | — | — |
| Bare metal stent | 750 (46.7%) | — | — |
| Balloon angioplasty | 38 (2.4%) | — | — |
| No. of treated coronary arteries | 1.2 ± 0.4 | — | — |
| No. of treated lesions | 1.3 ± 0.6 | — | — |
| PCI of left main stenosis | 14 (0.9%) | — | — |
| Revascularized coronary artery | |||
| Left anterior descending artery | 919 (56.8%) | — | — |
| Circumflex artery | 351 (21.7%) | — | — |
| Right coronary artery | 620 (38.3%) | — | — |
| Medication at discharge | |||
| Aspirin | 1589 (98.2%) | 542 (91.6%) | <0.0001 |
| Ticlopidine | 25 (1.5%) | 3 (0.5%) | 0.06 |
| Clopidogrel | 1505 (93.0%) | 96 (16.2%) | <0.0001 |
| Ticagrelor | 12 (0.7%) | 0 | 0.04 |
| Prasugrel | 46 (2.8%) | 4 (0.7%) | 0.006 |
| Warfarin | 65 (4.0%) | 55 (9.3%) | <0.0001 |
| Statin | 1545 (95.5%) | 398 (67.2%) | <0.0001 |
| Angiotensin converting enzyme-inhibitor | 998 (61.7%) | 234 (39.5%) | <0.0001 |
| Beta-blockers | 1429 (88.3%) | 447 (75.5%) | <0.0001 |
| Diuretics | 153 (9.5%) | 282 (47.6%) | <0.0001 |
| Calcium channel blockers | 117 (7.2%) | 66 (11.1%) | 0.01 |
| Outcome End-points | PCI (n = 1617) | CABG (n = 592) | Univariate Analysis p-Value | Propensity Score Adjusted Analysis OR/RR, 95% CI |
|---|---|---|---|---|
| Immediate outcome | ||||
| 30-Day mortality | 13 (0.8%) | 8 (1.4%) | 0.27 | 0.283, 0.055–1.462 |
| 30-Day repeat revascularization | 35 (2.2%) | 7 (1.2%) | 0.13 | 2.917, 0.883–9.638 |
| Stroke | 0 (0%) | 4 (0.7%) | 0.005 | 0.000 |
| De novo dialysis | 2 (0.1%) | 7 (1.2%) | 0.002 | 0.191, 0.022–1.679 |
| Late outcome | ||||
| Survival | 0.08 | 0.694, 0.284–1.699 | ||
| 1-Year | 98.9% | 98.3% | ||
| 3-Year | 98.1% | 96.3% | ||
| 5-Year | 97.8% | 94.9% | ||
| Freedom from repeat revascularization | <0.0001 | 9.176, 5.865–14.356 | ||
| 1-Year | 89.2% | 96.3% | ||
| 3-Year | 83.0% | 95.1% | ||
| 5-Year | 77.6% | 92.5% | ||
| Freedom from stroke | 0.73 | 2.383, 0.684–8.303 | ||
| 1-Year | 99.6% | 99.3% | ||
| 3-Year | 98.5% | 99.3% | ||
| 5-Year | 98.0% | 98.0% | ||
| Freedom from myocardial infarction | <0.0001 | 10.627, 4.925–22.834 | ||
| 1-Year | 97.0% | 98.9% | ||
| 3-Year | 92.9% | 98.4% | ||
| 5-Year | 89.9% | 96.6% | ||
| Freedom from MACCE | <0.0001 | 4.648, 3.203–6.746 | ||
| 1-Year | 87.5% | 93.8% | ||
| 3-Year | 79.2% | 90.1% | ||
| 5-Year | 73.9% | 85.0% |
Propensity score–adjusted multivariate analysis confirmed the findings of univariate analysis ( Table 3 ). One-to-one propensity score matching provided 235 pairs. The prevalence of depressed left ventricular function, left main stenosis, STEMI, and one-vessel disease was similar in patients who underwent PCI and CABG ( Table 4 ). The immediate outcome, late survival, and freedom from stroke were similar between the study groups ( Table 5 ). However, PCI was associated with significantly lower freedom from myocardial infarction (85.8% vs 99.5%, p <0.0001), repeat revascularization (at 5 years, 77.6% vs 92.5%, p <0.0001), and MACCE (58.7% vs 90.2%, p <0.0001; Table 5 ).
| Clinical Variables | PCI (n = 235) | CABG (n = 235) | p-Value |
|---|---|---|---|
| Age (years) | 45.8 ± 4.1 | 46.1 ± 3.8 | 0.55 |
| Female gender | 25 (10.6%) | 25 (10.6%) | 1.00 |
| Family history of coronary artery disease | 133 (56.6%) | 128 (54.5%) | 0.71 |
| Dyslipidemia | 147 (62.6%) | 134 (57.0%) | 0.22 |
| Treatment for hypertension | 126 (53.6%) | 120 (51.1%) | 0.58 |
| Smoking habit | 156 (66.4%) | 155 (66.0%) | 0.92 |
| Diabetes | 44 (18.7%) | 43 (18.3%) | 0.91 |
| Extracardiac arteriopathy | 13 (5.5%) | 11 (4.7%) | 0.68 |
| Estimated glomerular filtration rate (mL/min/1.73 m 2 ) | 95 ± 24 | 98 ± 21 | 0.58 |
| Estimated glomerular filtration rate <60 mL/min/1.73 m 2 | 15 (6.4%) | 6 (2.6%) | 0.04 |
| Congestive heart failure | 3 (1.3%) | 8 (3.4%) | 0.22 |
| Prior transient ischemic attack or stroke | 8 (3.4%) | 4 (1.7%) | 0.38 |
| Prior myocardial infarction | 45 (19.1%) | 45 (19.1%) | 1.00 |
| Prior percutaneous coronary intervention | 30 (12.8%) | 24 (10.2%) | 0.39 |
| Prior coronary artery bypass grafting | 3 (1.3%) | 2 (0.9%) | 0.65 |
| Indication | <0.0001 | ||
| Stable angina pectoris | 70 (29.8%) | 97 (41.3%) | |
| Unstable angina pectoris | 29 (12.3%) | 49 (20.9%) | |
| Non–ST-elevation myocardial infarction | 90 (38.3%) | 42 (17.9%) | |
| ST-elevation myocardial infarction | 46 (19.6%) | 47 (20.0%) | 1.000 |
| Urgency | 0.70 | ||
| Elective | 116 (49.4%) | 107 (45.5%) | |
| Urgent | 106 (45.1%) | 113 (48.1%) | |
| Emergency | 13 (5.5%) | 15 (6.4%) | |
| Left main stenosis | 15 (6.4%) | 14 (6.0%) | |
| Left ventricular ejection fraction | 0.75 | ||
| >50% | 163 (69.4%) | 167 (71.1%) | |
| 30–50% | 67 (28.5%) | 65 (27.7%) | |
| <30% | 5 (2.1%) | 3 (1.3%) | 0.85 |
| No. diseased vessels | <0.0001 | ||
| 1 | 43 (18.3%) | 39 (16.6%) | 0.63 |
| 2 | 135 (57.4%) | 44 (18.7%) | |
| 3 | 57 (24.3%) | 152 (64.7%) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
