Risk Prediction After Coronary Artery Bypass Grafting Combined With Coronary Endarterectomy

Coronary endarterectomy (CE) combined with coronary artery bypass grafting (CABG) is used for complete revascularization of diffusely diseased coronary arteries. Nevertheless, studies reported an increased risk after this procedure. Therefore, risk prediction in these patients is essential. Patients who underwent CABG + CE during September 2008 and July 2022 at our center were retrospectively recruited. A total of 32 characteristics were analyzed. The least absolute shrinkage and selection operator regression were used for the feature selection, and multivariable Cox regression was applied to develop a nomogram for risk prediction. The primary outcome was the major adverse cardiovascular and cerebrovascular events (MACCE), a composite of all-cause death, nonfatal myocardial infarction, repeat revascularization, and stroke. A total of 570 patients with 601 CE targets, including left anterior descending (41.4%), right coronary artery (43.9%), left circumflex artery (6.8%), and diagonal branches/intermedius ramidus (8.0%), were enrolled. The mean age was 61.0 ± 8.9 years, and 77.7% were men. A total of 4 features were identified as the predictors of MACCE, including age ≥65 years (hazard ratio [HR] 2.12, 95% confidence interval [CI] 1.38 to 3.25, p <0.001), left main disease (HR 2.56, 95% CI 1.46 to 4.49, p = 0.001), mitral regurgitation (≥mild, HR 1.91, 95% CI 1.01 to 3.65, p = 0.049), and left anterior descending endarterectomy (HR 1.69, 95% CI 1.09 to 2.62, p = 0.018), and a nomogram for the 1- and 3-year MACCE prediction was developed. The model showed relatively good discrimination (C-index 0.68), calibration, and clinical usefulness. In conclusion, the nomogram provides estimation of the 1- and 3-year MACCE risk after CABG + CE. Coronary endarterectomy (CE) combined with coronary artery bypass grafting (CABG) is used for complete revascularization of diffusely diseased coronary arteries. Nevertheless, studies reported an increased risk after this procedure. Therefore, risk prediction in these patients is essential. Patients who underwent CABG + CE during September 2008 and July 2022 at our center were retrospectively recruited. A total of 32 characteristics were analyzed. The least absolute shrinkage and selection operator regression were used for the feature selection, and multivariable Cox regression was applied to develop a nomogram for risk prediction. The primary outcome was the major adverse cardiovascular and cerebrovascular events (MACCE), a composite of all-cause death, nonfatal myocardial infarction, repeat revascularization, and stroke. A total of 570 patients with 601 CE targets, including left anterior descending (41.4%), right coronary artery (43.9%), left circumflex artery (6.8%), and diagonal branches/intermedius ramidus (8.0%), were enrolled. The mean age was 61.0 ± 8.9 years, and 77.7% were men. A total of 4 features were identified as the predictors of MACCE, including age ≥65 years (hazard ratio [HR] 2.12, 95% confidence interval [CI] 1.38 to 3.25, p <0.001), left main disease (HR 2.56, 95% CI 1.46 to 4.49, p = 0.001), mitral regurgitation (≥mild, HR 1.91, 95% CI 1.01 to 3.65, p = 0.049), and left anterior descending endarterectomy (HR 1.69, 95% CI 1.09 to 2.62, p = 0.018), and a nomogram for the 1- and 3-year MACCE prediction was developed. The model showed relatively good discrimination (C-index 0.68), calibration, and clinical usefulness. In conclusion, the nomogram provides estimation of the 1- and 3-year MACCE risk after CABG + CE. Patients with diffuse coronary artery disease not amenable to revascularization through conventional approaches (e.g., percutaneous coronary intervention or coronary artery bypass grafting [CABG]) are at a huge risk of early and long-term mortality.1Lozano I Capin E de la Hera JM Llosa JC Carro A López-Palop R Diffuse coronary artery disease not amenable to revascularization: long-term prognosis.Rev Esp Cardiol (Engl Ed). 2015; 68: 631-633Crossref PubMed Scopus (21) Google Scholar In these patients, coronary endarterectomy (CE) combined with CABG might be the only choice for achieving complete revascularization and relieving symptoms.2Chang Y Shih CT Lai ST Early results of the advanced coronary endarterectomy combined with CABG in the treatment of coronary artery occlusive disease.Zhonghua Yi Xue Za Zhi (Taipei). 1994; 54: 156-159PubMed Google Scholar The major concerns for the clinical application of CE are its high risk of postoperative myocardial infarction and mortality and low patency rate.3Ferraris VA Harrah JD Moritz DM Striz M Striz D Ferraris SP Long-term angiographic results of coronary endarterectomy.Ann Thorac Surg. 2000; 69: 1737-1743Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar, 4Atik FA Dallan LA de Oliveira SA Lisboa LA Platania F Cabral RH Jatene AD Myocardial revascularization with coronary endarterectomy. Stratification of risk factors for early mortality.Arq Bras Cardiol. 2000; 75: 269-280Crossref PubMed Scopus (20) Google Scholar, 5Song Y Xu F Du J Zhang J Feng W Coronary endarterectomy with coronary artery bypass graft decreases graft patency compared with isolated coronary artery bypass graft: a meta-analysis.Interact Cardiovasc Thorac Surg. 2017; 25: 30-36Crossref PubMed Scopus (25) Google Scholar With the improvement in the surgical techniques and the perioperative antithrombotic therapy, studies reported improved clinical outcomes after CABG + CE.6Tiemuerniyazi X Yan H Song Y Nan Y Xu F Feng W Mid-term outcomes of coronary endarterectomy combined with coronary artery bypass grafting.Interact Cardiovasc Thorac Surg. 2021; 32: 188-195Crossref PubMed Scopus (3) Google Scholar, 7Chow SC Ho JY Kwok MW Fujikawa T Lim K Wan S Wong RH Coronary endarterectomy in coronary artery disease: factors affecting graft patency and survival.Asian Cardiovasc Thorac Ann. 2022; 30: 147-155Crossref PubMed Scopus (1) Google Scholar, 8Eris C Engin M Sunbul SA As AK Erdolu B Early postoperative results of on-pump coronary endarterectomy: is gender a risk factor?.Heart Surg Forum. 2021; 24: E662-E669Crossref PubMed Scopus (3) Google Scholar However, researchers also noticed that CE is still associated with an increased risk of repeat revascularization9Janiec M Ragnarsson S Nozohoor S Long-term outcome after coronary endarterectomy adjunct to coronary artery bypass grafting.Interact Cardiovasc Thorac Surg. 2019; 29: 22-27Crossref PubMed Scopus (10) Google Scholar and higher short-term mortality.10Wang C Chen J Gu C Li J Analysis of survival after coronary endarterectomy combined with coronary artery bypass grafting compared with isolated coronary artery bypass grafting: a meta-analysis.Interact Cardiovasc Thorac Surg. 2019; 29: 393-401Crossref PubMed Scopus (11) Google Scholar Recent meta-analyses also demonstrate a higher risk of mortality and cerebrovascular events in patients who underwent CABG + CE than those receiving isolated CABG.11Wang C Chen J Gu C Qiao R Li J Impact of risk factors and surgical techniques in coronary endarterectomy: a network meta-analysis.Interact Cardiovasc Thorac Surg. 2019; 29: 355-364Crossref PubMed Scopus (8) Google Scholar,12Wang J Gu C Gao M Yu W Yu Y Adjunct coronary endarterectomy increases cerebrovascular accident after coronary artery bypass grafting: a meta-analysis of 17,600 patients.Int J Cardiol. 2015; 182: 79-81Abstract Full Text Full Text PDF PubMed Google Scholar Therefore, further risk estimation after CABG + CE is essential. In this study, we aimed to summarize the clinical outcomes of patients who underwent CABG + CE at our center during the last 14 years, explore the potential risk factors for the clinical outcomes, and develop a risk prediction model for adverse events. In this single-center retrospective cohort study, adult (aged >18 years) patients who underwent CABG + CE during September 2008 to July 2022 were retrospectively recruited. Patients with any of the following conditions were excluded: (1) patients who underwent concomitant valvular (aortic or mitral valve) surgery, (2) patients who received concomitant left ventricular reconstruction for ventricular aneurysm, or (3) patients who were complicated with hypertrophic cardiomyopathy or received concomitant myectomy for left ventricular or midventricular obstruction. The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Fuwai Hospital (approval ID: 2022-1849) and individual consent for this retrospective analysis was waived. All the patients underwent a CE procedure, followed by CABG, with or without cardiopulmonary bypass. CE was performed in those with diffusely diseased (severely stenotic, occluded, or calcified) coronary artery, with or without involving the major side branches or severe long-segment in-stent restenosis, where complete revascularization could not be achieved through neither by percutaneous coronary intervention nor by isolated CABG. Whether to perform CE was decided intraoperatively according to the condition of the coronary arteries. Because of the perpendicular distribution of the perforators, the open CE technique was preferred in the left anterior descending (LAD) artery to achieve a complete removal of the plaque, whereas the closed CE technique was preferred on the other coronary territories. The operative techniques for CE were described in detail previously.6Tiemuerniyazi X Yan H Song Y Nan Y Xu F Feng W Mid-term outcomes of coronary endarterectomy combined with coronary artery bypass grafting.Interact Cardiovasc Thorac Surg. 2021; 32: 188-195Crossref PubMed Scopus (3) Google Scholar Simply, a small incision was made on the coronary artery to remove the atherosclerotic plaque during a closed procedure, whereas a long-segment incision was applied in the open technique. The surgical procedure for CE was completed by fully experienced cardiac surgeons who could apply CE whenever feasible. The perioperative data were extracted from the electronic hospital records, whereas the follow-up was completed through outpatient visits and phone call interviews. For several patients who could not be reached by either of the methods, a home visit was made. The primary outcome was the occurrence of major adverse cardiovascular and cerebrovascular events (MACCE), which was defined as a composite of all-cause death, nonfatal myocardial infarction, repeat revascularization, and stroke (both hemorrhagic and ischemic stroke). Age was categorized using a cutoff of 65 years. Preoperative chronic kidney failure was defined as a medical history of renal failure, dialysis, or a preoperative creatinine level of >133 µmol/L. Postoperative myocardial infarction was diagnosed if all of the following criteria were met: (1) within 48 hours postoperatively; (2) elevation of high-sensitivity cardiac troponin I more than 5 times; and (3) any of the following criteria: new-onset pathologic Q wave or left bundle branch block, new-onset ventricular motion abnormality, or findings of graft occlusion in the coronary computed tomography. Operative death was defined as death within 1 month after the surgery. Continuous variables were expressed as means with SDs if they followed a normal distribution; otherwise, they were reported as medians with twenty-fifth and seventy-fifth quartiles. Categorical variables were presented as numbers with percentages. The least absolute shrinkage and selection operator (LASSO) regression was used for the feature selection. A multivariable Cox regression model was used to develop the final risk prediction model, followed by the generation of a nomogram. The performance of the model was tested using discrimination ability (C-index) and calibration plot. To further assess the discriminative ability of the model, the patients were divided into low-risk (risk score less than median) and high-risk (equal or more than median) groups, using the individualized risk score calculated using the nomogram. Then, the MACCE-free survival of the 2 groups were calculated using the Kaplan–Meier method and compared using the log-rank test. In addition, a decision curve analysis was performed to evaluate the clinical usefulness of the primary model. All the tests were 2-sided, and a p <0.05 was considered statistically significant. Statistical analyses were performed using R 4.2.0 (Core Team, Vienna, Austria). A total of 570 patients were enrolled, with a mean age of 61.0 ± 8.9 years. Most of the patients were men (77.7%). The majority of the patients’ conditions were complicated with hypertension (70.2%), dyslipidemia (74.6%), and triple-vessel disease (64.2%), whereas 44.6% of them had diabetes mellitus and 35.3% had a history of myocardial infarction. A total of 7 patients (1.2%) underwent emergent surgery, and 35.4% underwent an on-pump procedure. A total of 601 coronary arteries were endarterectomized, including LAD (41.4%), right coronary artery (43.9%), left circumflex artery (6.8%), and diagonal branches/intermedius ramidus (8.0%; Table 1, Figure 1).Table 1Patient characteristicsVariablesOverall cohort (N=570)Age (years), mean ± SD61.0 ± 8.9 Age ≥ 65 years, no (%)212 (37.2)Sex (male), no (%)443 (77.7)Body mass index (kg/m2), median [Q1, Q3]25.7 [23.8, 27.6] Body mass index ≥ 25kg/m2, no (%)339 (59.5)Smoking, no (%)324 (56.8)Hypertension, no (%)400 (70.2)Dyslipidemia, no (%)425 (74.6)Diabetes mellitus, no (%)254 (44.6)Chronic lung disease, no (%)5 (0.9)Peripheral artery disease, no (%)49 (8.6)Stroke, no (%)78 (13.7)Atrial fibrillation, no (%)13 (2.3)Myocardial infarction, no (%)201 (35.3)Percutaneous coronary intervention, no (%)95 (16.7)NYHA III or IV, no (%)208 (36.5)Congestive heart failure, no (%)59 (10.4)Chronic kidney failure, no (%)15 (2.6)Left main disease, no (%)49 (8.6)Triple vessel disease, no (%)366 (64.2)Ejection fraction (%), mean ± SD59.1 ± 8.5LVEDD (mm), mean ± SD50.0 ± 5.8Mitral regurgitation (mild or more), no (%)64 (11.2)Emergent surgery, no (%)7 (1.2)On-pump, no (%)202 (35.4)Number of grafts, mean ± SD3.4 ± 0.9Coronary endarterectomy Left anterior descending, no (%)249 (43.7) Right coronary artery, no (%)264 (46.3) Left circumflex artery, no (%)41 (7.2) Diagonal branches or intermedius ramidus, no (%)48 (8.4)Postoperative warfarin, no (%)119 (20.9)LVEDD = left ventricular end-diastolic diameter; NYHA = New York Heart Association; SD = standard deviation. Open table in a new tab LVEDD = left ventricular end-diastolic diameter; NYHA = New York Heart Association; SD = standard deviation. The median postoperative hospital stay was 7.0 (6.0 to 9.0) days. A total of 7 patients (1.2%) experienced operative death, 19 (3.3%) underwent thoracotomy for bleeding after the surgery, and 160 (28.1%) received transfusion. The occurrence of perioperative myocardial infarction, new-onset stroke, and acute kidney injury were 6.0%, 0.9%, and 18.1%, respectively, whereas the occurrence of new-onset atrial fibrillation was 18.9% (Table 2). Of the 7 patients who suffered from death, 1 was at critical state preoperatively and received emergent surgery. This patient received extracorporeal membrane oxygenation assist and continuous renal replacement therapy for renal failure postoperatively but died of multiple organ dysfunction. Other causes for operative death included: severe pulmonary infection or renal failure, followed by multiple organ dysfunction (n = 2), intraoperative ventricular fibrillation unresponsive to defibrillation (n = 1), pulmonary embolism (n = 1), and postoperative cardiac tamponade (underwent re-exploration for bleeding) or ventricular fibrillation (irresponsive to defibrillation, underwent re-exploration for bleeding), followed by multiple organ dysfunction (n = 2).Table 2Postoperative and follow-up outcomesVariablesOverall cohort (N=570)PostoperativeTransfusion, no (%)160 (28.1)Thoracotomy for bleeding, no (%)19 (3.3)Intra-aortic balloon pump, no (%)18 (3.2)ICU-stay (hours), median [Q1, Q3]48.0 [22.3, 96.0]Intubation time (hours), median [Q1, Q3]16.0 [13.0, 20.0]New-onset stroke, no (%)5 (0.9)New-onset atrial fibrillation, no (%)108 (18.9)Acute kidney injury103 (18.1) Stage 189 (15.6) Stage 26 (1.1) Stage 38 (1.4)Postoperative myocardial infarction, no (%)34 (6.0)Operative death, no (%)7 (1.2)Ejection fraction (%), mean ± SD57.9 ± 7.4LVEDD (mm), mean ± SD47.1 ± 5.5Mitral regurgitation (mild or more), no (%)6 (1.1)Postoperative hospital-stay (days), median [Q1, Q3]7.0 [6.0, 9.0]Follow-upMedian follow-up time (months), median [Q1, Q3]23.1 [12.8, 55.8]All-cause death, no (%)21 (3.7)MACCE, no (%)84 (14.7)LVEDD = left ventricular end-diastolic diameter; MACCE = major adverse cardiovascular and cerebrovascular events; SD = standard deviation. Open table in a new tab LVEDD = left ventricular end-diastolic diameter; MACCE = major adverse cardiovascular and cerebrovascular events; SD = standard deviation. The features included in the LASSO regression were as follows: (1) preoperative features, including age, gender, body mass index, smoking, history of hypertension, diabetes mellitus, chronic lung disease, peripheral artery disease, stroke, atrial fibrillation, myocardial infarction, previous percutaneous coronary intervention, New York Heart Association class III or IV, congestive heart failure, chronic kidney failure, left main disease, triple-vessel disease, ejection fraction, left ventricular end-diastolic diameter, and mild or more mitral regurgitation; (2) operative characteristics, including emergent surgery, use of cardiopulmonary bypass (on-pump or not), number of coronary grafts, LAD endarterectomy or not, and the experience of surgeon (completion of >20 CE or not); and (3) postoperative variables: use of warfarin in addition to the antiplatelet therapy. The LASSO regression resulted in the selection of 4 variables, including age, left main disease, mild or more mitral regurgitation, and LAD endarterectomy (Figure 2). A multivariable Cox regression model was developed using the 4 variables. The results showed that age ≥65 years (hazard ratio [HR] 2.12, 95% confidence interval [CI] 1.38 to 3.25, p <0.001), left main disease (HR 2.56, 95% CI 1.46 to 4.49, p = 0.001), mild or more mitral regurgitation (HR 1.91, 95% CI 1.01 to 3.65, p = 0.049), and LAD endarterectomy (HR 1.69, 95% CI 1.09 to 2.62, p = 0.018) were the strong predictors of MACCE (Supplementary Table 1). A nomogram including these variables was developed (Figure 2). The model showed relatively good discrimination (C-index 0.68). In addition, the calibration plot for the prediction of 1- and 3-year MACCE risk suggested a consistent performance of the model (Figure 3). Patients were divided into low-risk (score < 6) and high-risk (score ≥ 6) groups according to the individualized score calculated by the nomogram. The survival curve showed a significant increase regarding the occurrence of MACCE in the high-risk group (HR 2.43, 95% CI 1.54 to 3.83, p <0.001) compared with the low-risk group, indicating that the discriminative ability of the model was good (Figure 3). In addition, the decision curve analysis showed that adding the experience of the surgeon (model 2) did not improve the clinical usefulness of the original model (model 1; Figure 3). In this retrospective cohort study, we reported the midterm outcome after CABG + CE for diffuse coronary artery disease and analyzed the potential risk factors for the adverse events. We noticed that advanced age (≥65 years), left main disease, mitral regurgitation (≥mild), and LAD endarterectomy were the strong predictors of midterm MACCE. Furthermore, we developed a risk prediction model for the estimation of the 1- and 3-year MACCE risk. To the best of our knowledge, this was the first study aiming to establish a risk prediction model after CABG + CE. With the rapid improvement of percutaneous coronary intervention strategies and surgical revascularization techniques, the hazard of coronary artery disease has decreased. Currently, the clinical guidelines emphasize the importance of complete coronary revascularization to improve the prognosis of the patients.13Lawton JS, Tamis-Holland JE, Bangalore S, Bates ER, Beckie TM, Bischoff JM, Bittl JA, Cohen MG, DiMaio JM, Don CW, Fremes SE, Gaudino MF, Goldberger ZD, Grant MC, Jaswal JB, Kurlansky PA, Mehran R, Metkus TS Jr, Nnacheta LC, Rao SV, Sellke FW, Sharma G, Yong CM, Zwischenberger BA. 2021 ACC/AHA/SCAI guideline for coronary artery revascularization: a report of the American College of Cardiology/American Heart Association Joint Committee on clinical practice guidelines [published correction appears in Circulation 2022;145:e772]. Circulation 2022;145:e18–e114.Google Scholar Nonetheless, cardiac surgeons are facing more and more diffusely diseased coronary arteries, where complete coronary revascularization cannot be achieved through conventional intervention procedures, leaving the patients at a high risk. According to the previous studies, 5-year mortality in diffuse coronary artery disease not amenable to revascularization can be up to more than 60%, and the risk increases with age.1Lozano I Capin E de la Hera JM Llosa JC Carro A López-Palop R Diffuse coronary artery disease not amenable to revascularization: long-term prognosis.Rev Esp Cardiol (Engl Ed). 2015; 68: 631-633Crossref PubMed Scopus (21) Google Scholar In these patients, CABG + CE is the final option for complete revascularization. In this study, we enrolled 570 patients with diffuse coronary artery disease. All the patients underwent CE on at least 1 coronary artery, followed by CABG, and complete revascularization was achieved in all of the patients. Debates exist regarding the application of CABG + CE. The results from the early studies indicate high postoperative myocardial infarction and mortality rate after CABG + CE.14Livesay JJ Cooley DA Hallman GL Reul GJ Ott DA Duncan JM Frazier OH Early and late results of coronary endarterectomy. Analysis of 3,369 patients.J Thorac Cardiovasc Surg. 1986; 92: 649-660Abstract Full Text PDF PubMed Google Scholar Furthermore, the graft patency is also 1 of the concerns. During the last 2 decades, with the improvement of the surgical techniques and perioperative antithrombotic management, studies reported improved outcomes after CABG + CE,15Djalilian AR Shumway SJ Adjunctive coronary endarterectomy: improved safety in modern cardiac surgery.Ann Thorac Surg. 1995; 60: 1749-1754Abstract Full Text PDF PubMed Scopus (57) Google Scholar and researchers also pointed out that cardiac surgeons should realize the importance of CE in patients with diffuse coronary artery disease.16Tiruvoipati R Loubani M Peek G Coronary endarterectomy in the current era.Curr Opin Cardiol. 2005; 20: 517-520Crossref PubMed Scopus (17) Google Scholar Ranjan et al17Ranjan R Adhikary D Mandal S Seedher A Adhikary AB Outcome of coronary endarterectomy with coronary artery bypass grafting in patients with diffuse coronary artery disease in Bangladesh: a retrospective cohort study.JRSM Cardiovasc Dis. 2017; 62048004017732658PubMed Google Scholar reported in a retrospective study of 1,000 patients who underwent CABG + CE that the postoperative myocardial infarction rate is as low as 3.1% and the operative death is only 1.9%, and the long-term outcome is also promising. However, in the latest study including the largest population from the Society of Thoracic Surgeons (STS) Adult Cardiac Surgery Database, researchers observed that operative mortality (3.2%) and postoperative myocardial infarction (6.8%) after CABG + CE is nearly twice as the isolated CABG.18Kelly JJ Han JJ Desai ND Iyengar A Acker AM Grau-Sepulveda M Zwischenberger BA Jawitz OK Hargrove WC Szeto WY Williams ML Coronary endarterectomy: analysis of the Society of Thoracic Surgeons adult cardiac surgery database.Ann Thorac Surg. 2022; 114: 667-674Abstract Full Text Full Text PDF PubMed Scopus (6) Google Scholar The pooled results from meta-analyses also suggest compromised outcomes after CABG + CE,10Wang C Chen J Gu C Li J Analysis of survival after coronary endarterectomy combined with coronary artery bypass grafting compared with isolated coronary artery bypass grafting: a meta-analysis.Interact Cardiovasc Thorac Surg. 2019; 29: 393-401Crossref PubMed Scopus (11) Google Scholar,11Wang C Chen J Gu C Qiao R Li J Impact of risk factors and surgical techniques in coronary endarterectomy: a network meta-analysis.Interact Cardiovasc Thorac Surg. 2019; 29: 355-364Crossref PubMed Scopus (8) Google Scholar especially short-term outcomes.19Zhang W, Wu H. Outcomes of coronary artery bypass grafting with coronary endarterectomy: a systematic review and meta-analysis of recent studies [published online December 18, 2022]. Perfusion doi:10.1177/02676591221147418.Google Scholar Adding the fact that CE is technically demanding, many cardiac surgeons still hold a cautious attitude toward this procedure. According to the STS Adult Cardiac Surgery Database,18Kelly JJ Han JJ Desai ND Iyengar A Acker AM Grau-Sepulveda M Zwischenberger BA Jawitz OK Hargrove WC Szeto WY Williams ML Coronary endarterectomy: analysis of the Society of Thoracic Surgeons adult cardiac surgery database.Ann Thorac Surg. 2022; 114: 667-674Abstract Full Text Full Text PDF PubMed Scopus (6) Google Scholar only 2.9% of the patients who underwent CABG received a concomitant CE procedure. In this study, we observed that the operative mortality after CABG + CE was only 1.2%, whereas the occurrence of postoperative myocardial infarction was as high as 6.0%, which was in line with the study from the STS Adult Cardiac Surgery Database.18Kelly JJ Han JJ Desai ND Iyengar A Acker AM Grau-Sepulveda M Zwischenberger BA Jawitz OK Hargrove WC Szeto WY Williams ML Coronary endarterectomy: analysis of the Society of Thoracic Surgeons adult cardiac surgery database.Ann Thorac Surg. 2022; 114: 667-674Abstract Full Text Full Text PDF PubMed Scopus (6) Google Scholar In addition, the midterm survival outcome was also acceptable, which was also consistent with the previous studies.17Ranjan R Adhikary D Mandal S Seedher A Adhikary AB Outcome of coronary endarterectomy with coronary artery bypass grafting in patients with diffuse coronary artery disease in Bangladesh: a retrospective cohort study.JRSM Cardiovasc Dis. 2017; 62048004017732658PubMed Google Scholar Therefore, although the rate of postoperative myocardial infarction is higher, CABG + CE can be achieved with acceptable early and midterm survival outcomes. As mentioned previously, although it might be associated with increased risk, CABG + CE is the only option for patients with diffuse coronary artery disease that cannot achieve complete coronary revascularization. Therefore, risk prediction is essential. Livesay et al14Livesay JJ Cooley DA Hallman GL Reul GJ Ott DA Duncan JM Frazier OH Early and late results of coronary endarterectomy. Analysis of 3,369 patients.J Thorac Cardiovasc Surg. 1986; 92: 649-660Abstract Full Text PDF PubMed Google Scholar noticed in their study that age, reduced ejection fraction, and LAD endarterectomy were associated with increased perioperative risk. Researchers also pointed out that the result of CE is highly dependent of the surgeon's technique and the completeness of the endarterectomy.20Wallsh E Weinstein GS Coronary endarterectomy as an adjunct to coronary artery bypass: a perspective.Int J Cardiol. 1984; 5: 767-772Abstract Full Text PDF PubMed Scopus (1) Google Scholar The endarterectomy target (LAD or non-LAD endarterectomy), on-pump or off-pump surgery, and open or closed technique are the other factors that attracted the researchers’ attention.11Wang C Chen J Gu C Qiao R Li J Impact of risk factors and surgical techniques in coronary endarterectomy: a network meta-analysis.Interact Cardiovasc Thorac Surg. 2019; 29: 355-364Crossref PubMed Scopus (8) Google Scholar,21Zhu P Ye X Chen A Liu J Wang Z Zhou M Zhao Q Does the site of coronary endarterectomy have an impact on the clinical outcomes and graft patency?.Interact Cardiovasc Thorac Surg. 2019; 29: 402-408Crossref PubMed Scopus (8) Google Scholar In this study, we enrolled preoperative and operative features to investigate the potential risk factors for the adverse events after CABG + CE. The surgeon's experience was also 1 of the factors analyzed. It is worth mentioning that open CE was preferred in the revascularization of the LAD, considering the anatomic characteristics of the perforator branches, whereas the closed technique was preferred in non-LAD territories. We noticed that advanced age (≥65 years), existence of left main disease, mitral regurgitation (even mild level), and LAD endarterectomy were the strong predictors of MACCE after CABG + CE. We also developed a nomogram to predict the individualized risk of adverse events, including the 4 factors. The nomogram showed good discrimination ability and calibration. Unfortunately, however, because the limitation in the sample size and single-center study design, an external validation was not completed. More studies are needed. This study has several limitations. First, this is a retrospective cohort study from a single center. Therefore, the bias caused by the study design cannot be avoided. In addition, because the application of the CE procedure is less frequent, the sample size, especially the number of primary end points, is limited, which may have compromised the power of the tests. Furthermore, external validation was also not performed. Third, the angiographic results were not available for most of the patients. Therefore, risk prediction for graft patency was not developed. In conclusion, the nomogram incorporating age, left main disease, mitral regurgitation, and LAD endarterectomy provides the estimation of the 1- and 3-year MACCE risk after CABG + CE. The authors have no conflicts of interest to declare. Dr. Feng takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. Download .docx (.02 MB) Help with docx files