Propensity-matched analysis of a side-clamp versus an anastomosis assist device in cases of isolated coronary artery bypass grafting

Propensity-matched analysis of a side-clamp versus an anastomosis assist device in cases of... Abstract OBJECTIVES The use of an anastomosis assist device during coronary artery bypass grafting (CABG) is considered less invasive for the ascending aorta than the use of a side-biting clamp (Side-clamp) and to possibly be associated with a lower incidence of postoperative stroke. However, this benefit has not yet been clearly demonstrated. This study was to evaluate whether the use of an anastomosis assist device will minimize the postoperative stroke and other complications in patients undergoing off-pump CABG in comparison with the use of the Side-clamp. METHODS Patients undergoing isolated off-pump CABG were retrospectively reviewed using the Japan Adult Cardiovascular Surgery Database (2013–2016). We performed a one-to-one matched analysis based on the estimated propensity scores of those who underwent off-pump CABG with an anastomosis assist device (Device group, n = 14 213) or a side-biting clamp (Side-clamp group, n = 7374) and obtained 2 cohorts (n = 7348 each). We compared the early outcomes and the details of postoperative complications using the Pearson’s χ2 test. P-values of <0.05 were considered to indicate statistical significance. RESULTS No significant differences were observed in the rates of 30-day mortality (Side-clamp versus Device: 0.8% vs 0.8%, P = 0.93) or stroke (1.4% vs 1.4%, P = 0.46). Transient ischaemic attack/reversible ischaemic neurological deficit/delirium occurred more frequently in the Side-clamp group (1.3% vs 0.9%, P = 0.020), whereas new-onset atrial fibrillation (11.0% vs 12.8%, P < 0.001) and prolonged ventilation (2.0% vs 2.9%, P < 0.001) occurred more frequently in the Device group. There was no difference in the length of intensive care unit stay. CONCLUSIONS The use of an anastomosis assist device partially provided better results with regard to the transient neurological complications; however, no overall benefit was observed in this study. Anastomosis assist device, Off-pump coronary artery bypass grafting, Stroke, Side-biting clamp INTRODUCTION The manipulation of the ascending aorta during coronary artery bypass grafting (CABG) may result in various postoperative complications, such as stroke. The postoperative impact of a variety of aortic manipulations during CABG, for example the use of a side-biting clamp (Side-clamp), the use of an anastomosis assist device, the aorta no-touch technique and the use of cardiopulmonary bypass with arterial cannulation to the ascending aorta, has been evaluated in previous studies. Although cerebrovascular events have frequently among major postoperative complications, the discussion has not reached a conclusion [1–3], and the efficacy of the use of an anastomosis assist device in comparison to a side-biting clamp has not been adequately evaluated to date. This study evaluated the incidence of perioperative complications with stroke after off-pump CABG (OPCAB) as the primary end point. This was achieved by utilizing the nationwide Japan Adult Cardiovascular Surgery Database (JCVSD). A propensity-matched analysis was performed to compare the short-term outcomes of OPCAB procedures performed using the Side-clamp with those performed using an anastomosis assist device, in order to justify the use of an anastomosis assist device in the treatment of our patients. METHODS Japan Adult Cardiovascular Surgery Database JCVSD is a nationwide registry of patients undergoing cardiovascular surgery in Japan. Data collection was conducted using the National Clinical Database and was linked to cardiovascular surgery certification. Data from 590 cardiac surgery units located throughout Japan (close to 100% of the total number of such units) were recorded in the JCVSD adult section registry between January 2013 and December 2016. The methods of data collection and the data content of the JCVSD adult section have been previously described [4, 5]. The data registration project was approved by the institutional review board of the Japanese Surgical Society. The JCVSD variables and the relevant definitions (available online at http://jacvsd.umin.jp) were identical—for the most part—to those of the Society of Thoracic Surgeons (STS) National Adult Cardiac Database (available online at http://sts.org), with some modifications. Most of the data components collected in the database are required data components (i.e. the system forbids registration unless data are entered). As this study utilized only these main data components, almost no records were missing data. A limited number of cases [184 out of 56 102 isolated CABGs (0.3%)] that lacked information on the patients’ age, sex or 30-day postoperative status, which we believe arose from a system error, were deleted from the database. Study population Data on the patients who underwent isolated OPCAB using either the Side-clamp or an anastomosis assist device (i.e. HEARTSTRING, ENCLOSE II or PAS-PORT) in the years 2013–2016 and who were registered in the JCVSD were extracted from the database. The patients who underwent conversion to open-heart surgery as well as those without ascending aortic manipulation were excluded from this study (Fig. 1). Figure 1: View largeDownload slide The algorithm of patient enrolment. The number of isolated CABG cases from 2013 to 2016 was initially 56 102; a total of 21 587 cases were extracted after applying the exclusion criteria. CABG: coronary artery bypass grafting; JCVSD: Japan Adult Cardiovascular Surgery Database; OPCAB: off-pump CABG. Figure 1: View largeDownload slide The algorithm of patient enrolment. The number of isolated CABG cases from 2013 to 2016 was initially 56 102; a total of 21 587 cases were extracted after applying the exclusion criteria. CABG: coronary artery bypass grafting; JCVSD: Japan Adult Cardiovascular Surgery Database; OPCAB: off-pump CABG. Study outcome As the primary outcome, we assessed the incidence of postoperative stroke within 30 days after the surgery or during hospitalization. Stroke was defined in the registry as ‘cerebrovascular events with paralytic symptoms relevant to central nervous system persisting for more than 72 h during hospitalization’. As secondary outcomes, we assessed the rates of all-cause postoperative 30-day death, all-cause surgical death (death within 30 days or during the same hospitalization), transient neurological complications [transient ischaemic attack, reversible ischaemic neurological deficit or delirium], renal failure with a serum creatinine level of >2.0 mg/dl plus twice the level of preoperative serum creatinine, newly initiated dialysis, new-onset atrial fibrillation (AF) requiring any intervention, prolonged ventilation lasting >72 h, deep sternal infection (with infection penetrating the mediastinal cavity requiring debridement/drainage, with a positive culture, or for which antibiotics were administered), or reoperation for bleeding or graft occlusion. Complications that occurred within 30 days of surgery or during hospitalization were recorded in the database. We also assessed the operative time. Patient and procedural characteristics We extracted the information on patient characteristics and procedural characteristics which we considered to be associated with the outcomes, as well as those potentially affecting the surgeon’s choice between an anastomosis assist device or the Side-clamp, from the database. These included age, sex, body mass index, smoking status, comorbidities (diabetes mellitus, renal dysfunction, chronic dialysis, hyperlipidaemia, hypertension, carotid artery disease, chronic occlusive pulmonary disease, thoracic aortic disease, peripheral artery disease, congestive heart failure, history of cerebrovascular accident, past percutaneous coronary intervention and past CABG), preoperative conditions (presentation: acute myocardial infarction, angina, asymptomatic ischaemia, old myocardial infarction, other, presence of cardiac shock, New York Heart Association classification, left ventricular function and presence of mitral insufficiency), preoperative medication use [digitalis, beta blockers, acetylsalicylic acid (ASA), non-ASA antiplatelets, warfarin, direct oral anticoagulants, other anticoagulants and steroids] and procedural characteristics (urgency of surgery, whether or not a preoperative catheter examination was conducted, number of stenotic vessels, presence of left main trunk stenosis, use of the bilateral internal thoracic arteries and number of anastomoses). The patients were given ASA within 5 days before surgery. Statistical analysis We tabulated the patient characteristics by the method of aortic manipulation (i.e. Side-clamp or an anastomosis assist device), the number and percentage for categorical variables, and the mean and standard deviation for age. The characteristics were compared between the 2 groups using standardized mean differences, which were calculated using the SAS macro published by Yang and Dalton [6]. We estimated the propensity score (PS) for the use of an anastomosis assist device against the use of the Side-clamp for each patient, by fitting a logistic regression model with all of the above-mentioned patient and procedural characteristics. Then, using the SAS macro provided by Coca-Perraillon [7], we conducted greedy nearest neighbour matching between the 2 groups, with a caliper of 0.2 SD of the logit (PS). We checked the balance of the patient and procedural characteristics of the matched cohort using the above-described method. We compared the incidence of the study end points between the matched Side-clamp and Device patients using the McNemar’s test and Wilcoxon Signed-rank test for paired samples for binary outcomes and continuous outcomes, respectively. All of the statistical tests were 2-tailed, and P-values of <0.05 were considered to indicate statistical significance. All analyses were conducted using SAS 9.4 (SAS Institute Inc., Cary, NC, USA). RESULTS A total of 33 407 patients who underwent OPCAB during the study period in the JCVSD were identified. After excluding those who did not meet the inclusion criteria, data on 7374 patients and 14 213 patients who underwent OPCAB using the Side-clamp or an anastomosis assist device, respectively, were extracted from the database. Patient procedural characteristics In general, the patient and procedural characteristics of the 2 groups were similar, with a few exceptions (Table 1). In the Device group, the age of the patients was slightly older [mean age (standard deviation); Side-clamp 69.1 (9.8) years; Device 70.2 (9.4)], and the frequency of patients with a history of smoking was higher (Side-clamp 48.6%; Device 53.4%). The incidence of carotid artery disease in the patients in the Device group (9.6%) was higher than that in the patients in the Side-clamp group (7.1%). The frequency of procedures to treat old myocardial infarction was slightly higher in the Side-clamp group. The use of ASA and non-ASA antiplatelet therapy was higher among patients in the Side-clamp group, whereas anticoagulant use was higher among the patients in the Device group. A higher percentage of the patients in the Side-clamp group underwent a larger number (>3) of anastomoses. All other characteristics were similar between the 2 groups with a standardized difference of <0.07. Table 1: Preoperative characteristics of the patients Characteristics  Before matching     After matching     Side-clamp (n = 7374)  Device (n = 14 213)  SD  Side-clamp (n = 7348)  Device (n = 7348)  SD  Demographics   Age (years), mean (std)  69.1 (9.8)  70.2 (9.4)  −0.11  69.1 (9.8)  69.1 (9.8)  0.00   Male, n (%)  5681 (77)  11 154 (78.5)  −0.03  5663 (77.1)  5663 (77.1)  0.00   BMI ≥30 (kg/m2), n (%)  388 (5.3)  668 (4.7)  0.03  385 (5.2)  382 (5.2)  0.00   Smoking, n (%)  3584 (48.6)  7586 (53.4)  −0.10  3580 (48.7)  3561 (48)  0.01   Recent smoking (<1 month), n (%)  1184 (16.1)  2310 (16.3)  −0.01  1180 (16.1)  1188 (16.2)  0.00  Comorbidities, n (%)   Diabetes mellitus  4010 (54.4)  7690 (54.1)  0.01  3994 (54.4)  4008 (54.6)  0.00   Renal dysfunction  2044 (27.7)  3965 (27.9)  0.00  2036 (27.7)  2033 (27.7)  0.00   Dialysis  731 (9.9)  1545 (10.9)  −0.03  731 (10.0)  733 (10.0)  0.00   Hyperlipidaemia  4877 (66.1)  9070 (63.8)  0.05  4855 (66.1)  4853 (65.1)  0.00   Hypertension  5759 (78.1)  11 240 (79.1)  −0.02  5742 (78.1)  5739 (78.1)  0.00   Cerebrovascular accident  855 (11.6)  1871 (13.2)  −0.05  854 (11.6)  888 (12.1)  −0.01   Carotid artery disease  527 (7.1)  1368 (9.6)  −0.09  527 (7.2)  554 (7.5)  −0.01   COPD  1086 (14.7)  2144 (15.1)  −0.01  1084 (14.8)  1081 (14.7)  0.00   Thoracic aortic disease  89 (1.2)  279 (2)  −0.06  89 (1.2)  77 (1.1)  0.02   Peripheral artery disease  1082 (14.7)  2703 (19)  −0.12  1082 (14.7)  1092 (14.9)  0.00   Congestive heart failure  2045 (27.7)  3329 (23.4)  0.10  2023 (27.5)  2003 (27.3)  0.01   Past PCI  1907 (25.9)  3587 (25.2)  0.01  1898 (25.8)  1910 (26.0)  0.00   Past CABG  30 (0.4)  83 (0.6)  0.00  0  0  0.00  Presentation, n (%)      0.08      0.02   Primary diagnosis                Acute myocardial infarction  1519 (10.7)  801 (10.9)    793 (10.8)  784 (10.7)      Angina  11 084 (78.0)  5721 (77.6)    5709 (77.7)  5689 (77.4)      Asymptomatic ischaemia  535 (3.8)  241 (3.3)    239 (3.3)  252 (3.4)      Old myocardial infarction  590 (4.2)  261 (3.5)    261 (3.6)  283 (3.9)      Other  485 (3.4)  350 (4.8)    346 (4.7)  340 (4.6)     Cardiac shock  206 (2.8)  346 (2.4)  0.02  202 (2.8)  220 (3.0)  −0.01   NYHA classification III–IV  1267 (17.2)  2298 (16.2)  0.03  1259 (17.1)  1257 (17.1)  0.00   Left ventricle function      0.01      0.01    Good  3735 (50.7)  7270 (51.2)    3725 (50.7)  3687 (50.2)      Medium  3284 (44.5)  6293 (44.3)    3269 (44.5)  3313 (45.1)      Bad  355 (4.8)  650 (4.6)    354 (4.8)  348 (4.7)     Mitral insufficiency II/III/IV  1236 (16.8)  2364 (16.6)  0.00  1232 (16.8)  1207 (16.4)  0.01   Preoperative catheter exam      0.04      0.02    Performed: LMT/3-vessel disease  2447 (33.2)  4706 (33.1)    2439 (33.2)  2463 (33.5)      Performed: LMT/1–2 vessel disease  519 (7.0)  1119 (7.9)    581 (7.1)  515 (7.0)      Performed: no LMT/3-vessel disease  3341 (45.3)  6271 (44.1)    3328 (45.3)  3315 (45.1)      Performed: no LMT/1–2 vessel disease  889 (12.1)  1727 (12.2)    885 (12.0)  861 (11.7)      Not performed  178 (2.4)  390 (2.7)    178 (2.4)  194 (2.6)    Procedural characteristics, n (%)   Bilateral ITA use  2648 (35.9)  4919 (34.6)  0.03  2642 (36.0)  2663 (36.2)  −0.01   Number of anastomosis      0.08      0.01    <2  1633 (22.1)  3570 (25.1)    1630 (22.2)  1630 (22.2)      3  2911 (39.5)  5548 (39)    2895 (39.4)  2880 (39.2)      4 or 5  2602 (35.3)  4710 (33.1)    2595 (35.3)  2614 (35.6)      6 or more  228 (3.1)  385 (2.7)    228 (3.1)  224 (3.1)     Urgency      0.05      0.01    Elective  6162 (83.6)  12 095 (85.1)    6145 (83.6)  6143 (83.6)      Urgent  720 (9.8)  1348 (9.5)    717 (9.8)  705 (9.6)      Emergent  492 (6.7)  770 (5.4)    486 (6.6)  500 (6.8)    Preoperative medication use, n (%)   Digitalis  28 (0.4)  72 (0.5)  −0.02  28 (0.4)  28 (0.4)  0.00   Beta blocker  2629 (35.7)  4947 (34.8)  0.02  2620 (35.7)  2603 (35.4)  0.00   ASA  3469 (47)  5295 (37.3)  0.20  3444 (46.9)  3478 (47.3)  −0.01   Non-ASA antiplatelets  777 (10.5)  1177 (8.3)  0.08  763 (10.4)  791 (10.8)  −0.01   Warfarin  2101 (28.5)  4027 (28.3)  0.00  2097 (28.5)  2077 (28.3)  0.01   Warfarin replacements  42 (0.6)  108 (0.8)  −0.02  42 (0.6)  37 (0.5)  0.01   Anticoags  771 (10.5)  2122 (14.9)  −0.13  771 (10.5)  768 (10.5)  0.00   Steroids  81 (1.1)  210 (1.5)  −0.03  80 (1.1)  88 (1.2)  −0.01  Characteristics  Before matching     After matching     Side-clamp (n = 7374)  Device (n = 14 213)  SD  Side-clamp (n = 7348)  Device (n = 7348)  SD  Demographics   Age (years), mean (std)  69.1 (9.8)  70.2 (9.4)  −0.11  69.1 (9.8)  69.1 (9.8)  0.00   Male, n (%)  5681 (77)  11 154 (78.5)  −0.03  5663 (77.1)  5663 (77.1)  0.00   BMI ≥30 (kg/m2), n (%)  388 (5.3)  668 (4.7)  0.03  385 (5.2)  382 (5.2)  0.00   Smoking, n (%)  3584 (48.6)  7586 (53.4)  −0.10  3580 (48.7)  3561 (48)  0.01   Recent smoking (<1 month), n (%)  1184 (16.1)  2310 (16.3)  −0.01  1180 (16.1)  1188 (16.2)  0.00  Comorbidities, n (%)   Diabetes mellitus  4010 (54.4)  7690 (54.1)  0.01  3994 (54.4)  4008 (54.6)  0.00   Renal dysfunction  2044 (27.7)  3965 (27.9)  0.00  2036 (27.7)  2033 (27.7)  0.00   Dialysis  731 (9.9)  1545 (10.9)  −0.03  731 (10.0)  733 (10.0)  0.00   Hyperlipidaemia  4877 (66.1)  9070 (63.8)  0.05  4855 (66.1)  4853 (65.1)  0.00   Hypertension  5759 (78.1)  11 240 (79.1)  −0.02  5742 (78.1)  5739 (78.1)  0.00   Cerebrovascular accident  855 (11.6)  1871 (13.2)  −0.05  854 (11.6)  888 (12.1)  −0.01   Carotid artery disease  527 (7.1)  1368 (9.6)  −0.09  527 (7.2)  554 (7.5)  −0.01   COPD  1086 (14.7)  2144 (15.1)  −0.01  1084 (14.8)  1081 (14.7)  0.00   Thoracic aortic disease  89 (1.2)  279 (2)  −0.06  89 (1.2)  77 (1.1)  0.02   Peripheral artery disease  1082 (14.7)  2703 (19)  −0.12  1082 (14.7)  1092 (14.9)  0.00   Congestive heart failure  2045 (27.7)  3329 (23.4)  0.10  2023 (27.5)  2003 (27.3)  0.01   Past PCI  1907 (25.9)  3587 (25.2)  0.01  1898 (25.8)  1910 (26.0)  0.00   Past CABG  30 (0.4)  83 (0.6)  0.00  0  0  0.00  Presentation, n (%)      0.08      0.02   Primary diagnosis                Acute myocardial infarction  1519 (10.7)  801 (10.9)    793 (10.8)  784 (10.7)      Angina  11 084 (78.0)  5721 (77.6)    5709 (77.7)  5689 (77.4)      Asymptomatic ischaemia  535 (3.8)  241 (3.3)    239 (3.3)  252 (3.4)      Old myocardial infarction  590 (4.2)  261 (3.5)    261 (3.6)  283 (3.9)      Other  485 (3.4)  350 (4.8)    346 (4.7)  340 (4.6)     Cardiac shock  206 (2.8)  346 (2.4)  0.02  202 (2.8)  220 (3.0)  −0.01   NYHA classification III–IV  1267 (17.2)  2298 (16.2)  0.03  1259 (17.1)  1257 (17.1)  0.00   Left ventricle function      0.01      0.01    Good  3735 (50.7)  7270 (51.2)    3725 (50.7)  3687 (50.2)      Medium  3284 (44.5)  6293 (44.3)    3269 (44.5)  3313 (45.1)      Bad  355 (4.8)  650 (4.6)    354 (4.8)  348 (4.7)     Mitral insufficiency II/III/IV  1236 (16.8)  2364 (16.6)  0.00  1232 (16.8)  1207 (16.4)  0.01   Preoperative catheter exam      0.04      0.02    Performed: LMT/3-vessel disease  2447 (33.2)  4706 (33.1)    2439 (33.2)  2463 (33.5)      Performed: LMT/1–2 vessel disease  519 (7.0)  1119 (7.9)    581 (7.1)  515 (7.0)      Performed: no LMT/3-vessel disease  3341 (45.3)  6271 (44.1)    3328 (45.3)  3315 (45.1)      Performed: no LMT/1–2 vessel disease  889 (12.1)  1727 (12.2)    885 (12.0)  861 (11.7)      Not performed  178 (2.4)  390 (2.7)    178 (2.4)  194 (2.6)    Procedural characteristics, n (%)   Bilateral ITA use  2648 (35.9)  4919 (34.6)  0.03  2642 (36.0)  2663 (36.2)  −0.01   Number of anastomosis      0.08      0.01    <2  1633 (22.1)  3570 (25.1)    1630 (22.2)  1630 (22.2)      3  2911 (39.5)  5548 (39)    2895 (39.4)  2880 (39.2)      4 or 5  2602 (35.3)  4710 (33.1)    2595 (35.3)  2614 (35.6)      6 or more  228 (3.1)  385 (2.7)    228 (3.1)  224 (3.1)     Urgency      0.05      0.01    Elective  6162 (83.6)  12 095 (85.1)    6145 (83.6)  6143 (83.6)      Urgent  720 (9.8)  1348 (9.5)    717 (9.8)  705 (9.6)      Emergent  492 (6.7)  770 (5.4)    486 (6.6)  500 (6.8)    Preoperative medication use, n (%)   Digitalis  28 (0.4)  72 (0.5)  −0.02  28 (0.4)  28 (0.4)  0.00   Beta blocker  2629 (35.7)  4947 (34.8)  0.02  2620 (35.7)  2603 (35.4)  0.00   ASA  3469 (47)  5295 (37.3)  0.20  3444 (46.9)  3478 (47.3)  −0.01   Non-ASA antiplatelets  777 (10.5)  1177 (8.3)  0.08  763 (10.4)  791 (10.8)  −0.01   Warfarin  2101 (28.5)  4027 (28.3)  0.00  2097 (28.5)  2077 (28.3)  0.01   Warfarin replacements  42 (0.6)  108 (0.8)  −0.02  42 (0.6)  37 (0.5)  0.01   Anticoags  771 (10.5)  2122 (14.9)  −0.13  771 (10.5)  768 (10.5)  0.00   Steroids  81 (1.1)  210 (1.5)  −0.03  80 (1.1)  88 (1.2)  −0.01  ASA: acetylsalicylic acid; BMI: body mass index; CABG: coronary artery bypass graft; COPD: chronic obstructive pulmonary disease; ITA: internal thoracic artery; LMT: left main trunk; NYHA: New York Heart Association; PCI: percutaneous transcatheter intervention; SD: standardized difference; std: standard deviation. Table 1: Preoperative characteristics of the patients Characteristics  Before matching     After matching     Side-clamp (n = 7374)  Device (n = 14 213)  SD  Side-clamp (n = 7348)  Device (n = 7348)  SD  Demographics   Age (years), mean (std)  69.1 (9.8)  70.2 (9.4)  −0.11  69.1 (9.8)  69.1 (9.8)  0.00   Male, n (%)  5681 (77)  11 154 (78.5)  −0.03  5663 (77.1)  5663 (77.1)  0.00   BMI ≥30 (kg/m2), n (%)  388 (5.3)  668 (4.7)  0.03  385 (5.2)  382 (5.2)  0.00   Smoking, n (%)  3584 (48.6)  7586 (53.4)  −0.10  3580 (48.7)  3561 (48)  0.01   Recent smoking (<1 month), n (%)  1184 (16.1)  2310 (16.3)  −0.01  1180 (16.1)  1188 (16.2)  0.00  Comorbidities, n (%)   Diabetes mellitus  4010 (54.4)  7690 (54.1)  0.01  3994 (54.4)  4008 (54.6)  0.00   Renal dysfunction  2044 (27.7)  3965 (27.9)  0.00  2036 (27.7)  2033 (27.7)  0.00   Dialysis  731 (9.9)  1545 (10.9)  −0.03  731 (10.0)  733 (10.0)  0.00   Hyperlipidaemia  4877 (66.1)  9070 (63.8)  0.05  4855 (66.1)  4853 (65.1)  0.00   Hypertension  5759 (78.1)  11 240 (79.1)  −0.02  5742 (78.1)  5739 (78.1)  0.00   Cerebrovascular accident  855 (11.6)  1871 (13.2)  −0.05  854 (11.6)  888 (12.1)  −0.01   Carotid artery disease  527 (7.1)  1368 (9.6)  −0.09  527 (7.2)  554 (7.5)  −0.01   COPD  1086 (14.7)  2144 (15.1)  −0.01  1084 (14.8)  1081 (14.7)  0.00   Thoracic aortic disease  89 (1.2)  279 (2)  −0.06  89 (1.2)  77 (1.1)  0.02   Peripheral artery disease  1082 (14.7)  2703 (19)  −0.12  1082 (14.7)  1092 (14.9)  0.00   Congestive heart failure  2045 (27.7)  3329 (23.4)  0.10  2023 (27.5)  2003 (27.3)  0.01   Past PCI  1907 (25.9)  3587 (25.2)  0.01  1898 (25.8)  1910 (26.0)  0.00   Past CABG  30 (0.4)  83 (0.6)  0.00  0  0  0.00  Presentation, n (%)      0.08      0.02   Primary diagnosis                Acute myocardial infarction  1519 (10.7)  801 (10.9)    793 (10.8)  784 (10.7)      Angina  11 084 (78.0)  5721 (77.6)    5709 (77.7)  5689 (77.4)      Asymptomatic ischaemia  535 (3.8)  241 (3.3)    239 (3.3)  252 (3.4)      Old myocardial infarction  590 (4.2)  261 (3.5)    261 (3.6)  283 (3.9)      Other  485 (3.4)  350 (4.8)    346 (4.7)  340 (4.6)     Cardiac shock  206 (2.8)  346 (2.4)  0.02  202 (2.8)  220 (3.0)  −0.01   NYHA classification III–IV  1267 (17.2)  2298 (16.2)  0.03  1259 (17.1)  1257 (17.1)  0.00   Left ventricle function      0.01      0.01    Good  3735 (50.7)  7270 (51.2)    3725 (50.7)  3687 (50.2)      Medium  3284 (44.5)  6293 (44.3)    3269 (44.5)  3313 (45.1)      Bad  355 (4.8)  650 (4.6)    354 (4.8)  348 (4.7)     Mitral insufficiency II/III/IV  1236 (16.8)  2364 (16.6)  0.00  1232 (16.8)  1207 (16.4)  0.01   Preoperative catheter exam      0.04      0.02    Performed: LMT/3-vessel disease  2447 (33.2)  4706 (33.1)    2439 (33.2)  2463 (33.5)      Performed: LMT/1–2 vessel disease  519 (7.0)  1119 (7.9)    581 (7.1)  515 (7.0)      Performed: no LMT/3-vessel disease  3341 (45.3)  6271 (44.1)    3328 (45.3)  3315 (45.1)      Performed: no LMT/1–2 vessel disease  889 (12.1)  1727 (12.2)    885 (12.0)  861 (11.7)      Not performed  178 (2.4)  390 (2.7)    178 (2.4)  194 (2.6)    Procedural characteristics, n (%)   Bilateral ITA use  2648 (35.9)  4919 (34.6)  0.03  2642 (36.0)  2663 (36.2)  −0.01   Number of anastomosis      0.08      0.01    <2  1633 (22.1)  3570 (25.1)    1630 (22.2)  1630 (22.2)      3  2911 (39.5)  5548 (39)    2895 (39.4)  2880 (39.2)      4 or 5  2602 (35.3)  4710 (33.1)    2595 (35.3)  2614 (35.6)      6 or more  228 (3.1)  385 (2.7)    228 (3.1)  224 (3.1)     Urgency      0.05      0.01    Elective  6162 (83.6)  12 095 (85.1)    6145 (83.6)  6143 (83.6)      Urgent  720 (9.8)  1348 (9.5)    717 (9.8)  705 (9.6)      Emergent  492 (6.7)  770 (5.4)    486 (6.6)  500 (6.8)    Preoperative medication use, n (%)   Digitalis  28 (0.4)  72 (0.5)  −0.02  28 (0.4)  28 (0.4)  0.00   Beta blocker  2629 (35.7)  4947 (34.8)  0.02  2620 (35.7)  2603 (35.4)  0.00   ASA  3469 (47)  5295 (37.3)  0.20  3444 (46.9)  3478 (47.3)  −0.01   Non-ASA antiplatelets  777 (10.5)  1177 (8.3)  0.08  763 (10.4)  791 (10.8)  −0.01   Warfarin  2101 (28.5)  4027 (28.3)  0.00  2097 (28.5)  2077 (28.3)  0.01   Warfarin replacements  42 (0.6)  108 (0.8)  −0.02  42 (0.6)  37 (0.5)  0.01   Anticoags  771 (10.5)  2122 (14.9)  −0.13  771 (10.5)  768 (10.5)  0.00   Steroids  81 (1.1)  210 (1.5)  −0.03  80 (1.1)  88 (1.2)  −0.01  Characteristics  Before matching     After matching     Side-clamp (n = 7374)  Device (n = 14 213)  SD  Side-clamp (n = 7348)  Device (n = 7348)  SD  Demographics   Age (years), mean (std)  69.1 (9.8)  70.2 (9.4)  −0.11  69.1 (9.8)  69.1 (9.8)  0.00   Male, n (%)  5681 (77)  11 154 (78.5)  −0.03  5663 (77.1)  5663 (77.1)  0.00   BMI ≥30 (kg/m2), n (%)  388 (5.3)  668 (4.7)  0.03  385 (5.2)  382 (5.2)  0.00   Smoking, n (%)  3584 (48.6)  7586 (53.4)  −0.10  3580 (48.7)  3561 (48)  0.01   Recent smoking (<1 month), n (%)  1184 (16.1)  2310 (16.3)  −0.01  1180 (16.1)  1188 (16.2)  0.00  Comorbidities, n (%)   Diabetes mellitus  4010 (54.4)  7690 (54.1)  0.01  3994 (54.4)  4008 (54.6)  0.00   Renal dysfunction  2044 (27.7)  3965 (27.9)  0.00  2036 (27.7)  2033 (27.7)  0.00   Dialysis  731 (9.9)  1545 (10.9)  −0.03  731 (10.0)  733 (10.0)  0.00   Hyperlipidaemia  4877 (66.1)  9070 (63.8)  0.05  4855 (66.1)  4853 (65.1)  0.00   Hypertension  5759 (78.1)  11 240 (79.1)  −0.02  5742 (78.1)  5739 (78.1)  0.00   Cerebrovascular accident  855 (11.6)  1871 (13.2)  −0.05  854 (11.6)  888 (12.1)  −0.01   Carotid artery disease  527 (7.1)  1368 (9.6)  −0.09  527 (7.2)  554 (7.5)  −0.01   COPD  1086 (14.7)  2144 (15.1)  −0.01  1084 (14.8)  1081 (14.7)  0.00   Thoracic aortic disease  89 (1.2)  279 (2)  −0.06  89 (1.2)  77 (1.1)  0.02   Peripheral artery disease  1082 (14.7)  2703 (19)  −0.12  1082 (14.7)  1092 (14.9)  0.00   Congestive heart failure  2045 (27.7)  3329 (23.4)  0.10  2023 (27.5)  2003 (27.3)  0.01   Past PCI  1907 (25.9)  3587 (25.2)  0.01  1898 (25.8)  1910 (26.0)  0.00   Past CABG  30 (0.4)  83 (0.6)  0.00  0  0  0.00  Presentation, n (%)      0.08      0.02   Primary diagnosis                Acute myocardial infarction  1519 (10.7)  801 (10.9)    793 (10.8)  784 (10.7)      Angina  11 084 (78.0)  5721 (77.6)    5709 (77.7)  5689 (77.4)      Asymptomatic ischaemia  535 (3.8)  241 (3.3)    239 (3.3)  252 (3.4)      Old myocardial infarction  590 (4.2)  261 (3.5)    261 (3.6)  283 (3.9)      Other  485 (3.4)  350 (4.8)    346 (4.7)  340 (4.6)     Cardiac shock  206 (2.8)  346 (2.4)  0.02  202 (2.8)  220 (3.0)  −0.01   NYHA classification III–IV  1267 (17.2)  2298 (16.2)  0.03  1259 (17.1)  1257 (17.1)  0.00   Left ventricle function      0.01      0.01    Good  3735 (50.7)  7270 (51.2)    3725 (50.7)  3687 (50.2)      Medium  3284 (44.5)  6293 (44.3)    3269 (44.5)  3313 (45.1)      Bad  355 (4.8)  650 (4.6)    354 (4.8)  348 (4.7)     Mitral insufficiency II/III/IV  1236 (16.8)  2364 (16.6)  0.00  1232 (16.8)  1207 (16.4)  0.01   Preoperative catheter exam      0.04      0.02    Performed: LMT/3-vessel disease  2447 (33.2)  4706 (33.1)    2439 (33.2)  2463 (33.5)      Performed: LMT/1–2 vessel disease  519 (7.0)  1119 (7.9)    581 (7.1)  515 (7.0)      Performed: no LMT/3-vessel disease  3341 (45.3)  6271 (44.1)    3328 (45.3)  3315 (45.1)      Performed: no LMT/1–2 vessel disease  889 (12.1)  1727 (12.2)    885 (12.0)  861 (11.7)      Not performed  178 (2.4)  390 (2.7)    178 (2.4)  194 (2.6)    Procedural characteristics, n (%)   Bilateral ITA use  2648 (35.9)  4919 (34.6)  0.03  2642 (36.0)  2663 (36.2)  −0.01   Number of anastomosis      0.08      0.01    <2  1633 (22.1)  3570 (25.1)    1630 (22.2)  1630 (22.2)      3  2911 (39.5)  5548 (39)    2895 (39.4)  2880 (39.2)      4 or 5  2602 (35.3)  4710 (33.1)    2595 (35.3)  2614 (35.6)      6 or more  228 (3.1)  385 (2.7)    228 (3.1)  224 (3.1)     Urgency      0.05      0.01    Elective  6162 (83.6)  12 095 (85.1)    6145 (83.6)  6143 (83.6)      Urgent  720 (9.8)  1348 (9.5)    717 (9.8)  705 (9.6)      Emergent  492 (6.7)  770 (5.4)    486 (6.6)  500 (6.8)    Preoperative medication use, n (%)   Digitalis  28 (0.4)  72 (0.5)  −0.02  28 (0.4)  28 (0.4)  0.00   Beta blocker  2629 (35.7)  4947 (34.8)  0.02  2620 (35.7)  2603 (35.4)  0.00   ASA  3469 (47)  5295 (37.3)  0.20  3444 (46.9)  3478 (47.3)  −0.01   Non-ASA antiplatelets  777 (10.5)  1177 (8.3)  0.08  763 (10.4)  791 (10.8)  −0.01   Warfarin  2101 (28.5)  4027 (28.3)  0.00  2097 (28.5)  2077 (28.3)  0.01   Warfarin replacements  42 (0.6)  108 (0.8)  −0.02  42 (0.6)  37 (0.5)  0.01   Anticoags  771 (10.5)  2122 (14.9)  −0.13  771 (10.5)  768 (10.5)  0.00   Steroids  81 (1.1)  210 (1.5)  −0.03  80 (1.1)  88 (1.2)  −0.01  ASA: acetylsalicylic acid; BMI: body mass index; CABG: coronary artery bypass graft; COPD: chronic obstructive pulmonary disease; ITA: internal thoracic artery; LMT: left main trunk; NYHA: New York Heart Association; PCI: percutaneous transcatheter intervention; SD: standardized difference; std: standard deviation. The propensity score distribution and balance of characteristics in the matched cohort The c-statistic of the PS model was 0.602. As expected, there was a large amount of overlap in the distribution of the PSs between the 2 groups before matching (Fig. 2). We matched 7348 Side-clamp and 7348 Device patients based on their PSs. The distribution of the PS values of the 2 groups were almost completely identical in the matched cohort (Fig. 2). The balance of the patient and procedural characteristics between the 2 groups was further improved, with a standardized mean difference of <0.03 for all variables. Figure 2: View largeDownload slide The propensity score distribution of the Side-clamp (blue solid line) and the Device (red solid line) groups (A) before and (B) after matching are demonstrated, which shows that the 2 cohorts were very well matched. Figure 2: View largeDownload slide The propensity score distribution of the Side-clamp (blue solid line) and the Device (red solid line) groups (A) before and (B) after matching are demonstrated, which shows that the 2 cohorts were very well matched. Short-term clinical results The short-term postoperative outcomes in the matched cohort are presented in Table 2. The incidence of the primary outcome of stroke was similar in the 2 groups, with 99 (1.4%) patients in the Side-clamp group and 89 (1.2%) patients in the Device group (P = 0.46). The rates of 30-day and operative mortality were also similar in the 2 groups (operative mortality: Side-clamp 1.4%; Device 1.5%; P = 0.78). The incidence of major complications [operative death, stroke, newly initiated dialysis, prolonged ventilation (>24 h) due to pulmonary complications, deep sternal wound infection and reoperation for bleeding] did not differ to a statistically significant extent (Side-clamp 7.0%; Device 7.5%; P = 0.23). With regard to the postoperative morbidities, significant differences were observed in the incidence of transient ischaemic attack/reversible ischaemic neurological deficit/delirium [Side-clamp n = 93 (1.3%); Device n = 64 (0.9%); P = 0.020], new-onset AF [Side-clamp n = 807 (11.0%); Device n = 943 (12.8%); P <0.001] and prolonged ventilation [Side-clamp n = 147 (2%); Device n = 212 (2.9%); P < 0.001). The overall length of intensive care unit (ICU) stay was similar. With regard to the procedural results, the patients in the Device required homologous blood transfusion more frequently than those in the Side-clamp group (Side-clamp 54.6%; Device 60.0%; P < 0.001), and the average operation time was significantly shorter in the Side-clamp group [mean (standard deviation); Side-clamp 304.5 (99.5) min; Device 319.5 (97.9) min]. The days in the ICU were very similar between the 2 groups but were found to be marginally shorter for the Side-clamp group (median days in ICU: 3 days for Side-clamp, 3 days for Device, P-value 0.036). Table 2: Comparison of the early postoperative results between ‘Side-clamp’ and ‘Device’ Outcomes  Side-clamp (n = 7348)  Device (n = 7348)  P-value*  30-Day death, n (%)  61 (0.8)  62 (0.8)  0.93  Operative death, n (%)  106 (1.4)  110 (1.5)  0.78  Major complicationsa, n (%)  514 (7.0)  552 (7.5)  0.23  Stroke, n (%)  99 (1.4)  89 (1.2)  0.46  TIA/RIND/delirium, n (%)  93 (1.3)  64 (0.9)  0.020  Renal failure, n (%)  174 (2.4)  196 (2.7)  0.25  Newly initiated dialysis, n (%)  123 (1.7)  123 (1.7)  1.00  New-onset AF, n (%)  807 (11.0)  943 (12.8)  <0.001  Prolonged ventilationb, n (%)  147 (2.0)  212 (2.9)  <0.001  Postoperative ventilator use (h), median (25th–75th percentile)  8 (4–16.5)  8 (4–17)  <0.001  Deep sternal wound infection, n (%)  107 (1.5)  106 (1.4)  0.95  Reoperation for bleeding, n (%)  83 (1.1)  79 (1.1)  0.75  Reoperation for graft occlusion, n (%)  95 (1.3)  95 (1.3)  1.00  Days in ICU, median (25th–75th percentile)  3 (2–4)  3 (2–4)  0.12  Homologous transfusion use, n (%)  4008 (54.6)  4406 (60.0)  <0.001  Operation time (min), median (25th–75th percentile)  297 (241–357)  309 (255–370)  <0.001  Outcomes  Side-clamp (n = 7348)  Device (n = 7348)  P-value*  30-Day death, n (%)  61 (0.8)  62 (0.8)  0.93  Operative death, n (%)  106 (1.4)  110 (1.5)  0.78  Major complicationsa, n (%)  514 (7.0)  552 (7.5)  0.23  Stroke, n (%)  99 (1.4)  89 (1.2)  0.46  TIA/RIND/delirium, n (%)  93 (1.3)  64 (0.9)  0.020  Renal failure, n (%)  174 (2.4)  196 (2.7)  0.25  Newly initiated dialysis, n (%)  123 (1.7)  123 (1.7)  1.00  New-onset AF, n (%)  807 (11.0)  943 (12.8)  <0.001  Prolonged ventilationb, n (%)  147 (2.0)  212 (2.9)  <0.001  Postoperative ventilator use (h), median (25th–75th percentile)  8 (4–16.5)  8 (4–17)  <0.001  Deep sternal wound infection, n (%)  107 (1.5)  106 (1.4)  0.95  Reoperation for bleeding, n (%)  83 (1.1)  79 (1.1)  0.75  Reoperation for graft occlusion, n (%)  95 (1.3)  95 (1.3)  1.00  Days in ICU, median (25th–75th percentile)  3 (2–4)  3 (2–4)  0.12  Homologous transfusion use, n (%)  4008 (54.6)  4406 (60.0)  <0.001  Operation time (min), median (25th–75th percentile)  297 (241–357)  309 (255–370)  <0.001  a Major complications include operative death, stroke, newly initiated dialysis, prolonged ventilation (>24 h) due to pulmonary complications, deep sternal wound infection and reoperation for bleeds. b Ventilation >24 h due to pulmonary complications. * P-values from Pearson’s χ2 test or Wilcoxon rank-sum test. AF: atrial fibrillation; ICU: intensive care unit; RIND: reversible ischaemic neurological deficit; TIA: transient ischaemic attack. Table 2: Comparison of the early postoperative results between ‘Side-clamp’ and ‘Device’ Outcomes  Side-clamp (n = 7348)  Device (n = 7348)  P-value*  30-Day death, n (%)  61 (0.8)  62 (0.8)  0.93  Operative death, n (%)  106 (1.4)  110 (1.5)  0.78  Major complicationsa, n (%)  514 (7.0)  552 (7.5)  0.23  Stroke, n (%)  99 (1.4)  89 (1.2)  0.46  TIA/RIND/delirium, n (%)  93 (1.3)  64 (0.9)  0.020  Renal failure, n (%)  174 (2.4)  196 (2.7)  0.25  Newly initiated dialysis, n (%)  123 (1.7)  123 (1.7)  1.00  New-onset AF, n (%)  807 (11.0)  943 (12.8)  <0.001  Prolonged ventilationb, n (%)  147 (2.0)  212 (2.9)  <0.001  Postoperative ventilator use (h), median (25th–75th percentile)  8 (4–16.5)  8 (4–17)  <0.001  Deep sternal wound infection, n (%)  107 (1.5)  106 (1.4)  0.95  Reoperation for bleeding, n (%)  83 (1.1)  79 (1.1)  0.75  Reoperation for graft occlusion, n (%)  95 (1.3)  95 (1.3)  1.00  Days in ICU, median (25th–75th percentile)  3 (2–4)  3 (2–4)  0.12  Homologous transfusion use, n (%)  4008 (54.6)  4406 (60.0)  <0.001  Operation time (min), median (25th–75th percentile)  297 (241–357)  309 (255–370)  <0.001  Outcomes  Side-clamp (n = 7348)  Device (n = 7348)  P-value*  30-Day death, n (%)  61 (0.8)  62 (0.8)  0.93  Operative death, n (%)  106 (1.4)  110 (1.5)  0.78  Major complicationsa, n (%)  514 (7.0)  552 (7.5)  0.23  Stroke, n (%)  99 (1.4)  89 (1.2)  0.46  TIA/RIND/delirium, n (%)  93 (1.3)  64 (0.9)  0.020  Renal failure, n (%)  174 (2.4)  196 (2.7)  0.25  Newly initiated dialysis, n (%)  123 (1.7)  123 (1.7)  1.00  New-onset AF, n (%)  807 (11.0)  943 (12.8)  <0.001  Prolonged ventilationb, n (%)  147 (2.0)  212 (2.9)  <0.001  Postoperative ventilator use (h), median (25th–75th percentile)  8 (4–16.5)  8 (4–17)  <0.001  Deep sternal wound infection, n (%)  107 (1.5)  106 (1.4)  0.95  Reoperation for bleeding, n (%)  83 (1.1)  79 (1.1)  0.75  Reoperation for graft occlusion, n (%)  95 (1.3)  95 (1.3)  1.00  Days in ICU, median (25th–75th percentile)  3 (2–4)  3 (2–4)  0.12  Homologous transfusion use, n (%)  4008 (54.6)  4406 (60.0)  <0.001  Operation time (min), median (25th–75th percentile)  297 (241–357)  309 (255–370)  <0.001  a Major complications include operative death, stroke, newly initiated dialysis, prolonged ventilation (>24 h) due to pulmonary complications, deep sternal wound infection and reoperation for bleeds. b Ventilation >24 h due to pulmonary complications. * P-values from Pearson’s χ2 test or Wilcoxon rank-sum test. AF: atrial fibrillation; ICU: intensive care unit; RIND: reversible ischaemic neurological deficit; TIA: transient ischaemic attack. DISCUSSION During CABG, the manipulation of the ascending aorta, for which various degrees of invasive manipulation have been conceived, is often thought to increase the risk of postoperative cerebrovascular events. These include arterial cannulation for the establishment of cardiopulmonary bypass, the use of cross clamping, the use of Side-clamps and the use of anastomosis assist devices. The risks associated with the above-mentioned manipulations have been investigated in a series of studies [1–3, 8, 9]. In addition, the efficacy of the aorta no-touch technique has also been discussed previously [10–12]. In this study, we focused on the benefit of using an anastomosis assist device in reducing postoperative cerebrovascular events, especially stroke; however, at the time of writing, a consensus has not been reached with regard to the efficacy of anastomosis assist devices in this respect [1, 12–18]. The current study question originally arose from a previous report [13], in which we hypothesized that the frequent use of anastomosis assist devices might be effective for reducing postoperative cerebrovascular complications. As ∼60% of CABG procedures are performed in an off-pump manner, and given the fact that anastomosis assist devices such as the HEARTSTRING, ENCLOSE II and PAS-PORT systems are frequently used in Japan, a rigorous comparison focusing on the differences between the impact of Side-clamp and anastomosis assist device use in the ascending aorta has become possible. To the best of our knowledge, this is the largest clinical study to investigate the effect of anastomosis assist devices in OPCAB cases. A significant majority of the cases were included in the analysis (7348 of the 7374 Side-clamp cases and 7348 of the 14 213 Device cases after stratification), supporting the overall reliability of the results that were obtained. The overall background factors were well balanced, because the estimated rates of mortality and morbidity were similar between the 2 groups. This study sought to quantify the protective effect of anastomosis assist device use, as an emerging technique (second in efficacy to the aorta no-touch method) in combatting postoperative cerebrovascular events to improve postoperative complications; several favourable clinical outcomes have been published in the past [1, 2, 12–16, 19–26]. The benefit of the aorta no-touch technique is an accepted issue today; thus, to make the comparison as simple and straightforward as possible, we only compared the Side-clamp and Device groups. Our expectations were partially met in that the use of an anastomosis assist device provides a benefit in terms of minimizing the incidence of perioperative transient neurological complications with regard to the transient ischaemic attack/reversible ischaemic neurological deficit/delirium; on the other hand, the stroke rate was similar between the 2 groups. This result—with regard to the stroke rate—was compatible with the results of a network meta-analysis reported by Zhao et al [12]. In their report, the use of partial clamping and the use of a HEARTSTRING device were associated with a reduced rate of stroke in comparison with on-pump CABG; however, the comparison between the partial clamp and anastomosis assist device use revealed no significant difference in the stroke rate. There was no clear description regarding the degree of coronary artery disease in the target cohort, and this factor may not have been fully assessed elsewhere in the past. Our database did not record the frequency of the application of clamping manoeuvres in each case, which might have interfered with the outcome regarding the cerebrovascular event rate. Our results, which showed a higher incidence of new-onset AF and prolonged ventilation in the Device group, also need to be discussed. With regard to the AF, we matched the preoperative characteristics in detail, including ASA and beta-blocker use; thus, it can be stated that the background characteristics of the patient had minimal influence on this result. The notable findings were that the blood transfusion rate was higher and the operation time was longer in the Device group. Bleeding during the procedure might be a serious issue with the use of anastomosis assist devices regarding technical achievement in handling, and this issue has resulted in the higher requirement of homologous blood in the Device group. Although excessive bleeding did not seem to result in the higher requirement of re-exploration, it could be one factor that resulted in the higher incidence of AF in the Device group. In 2014, Alameddine et al. [27] reported that perioperative blood transfusion may be associated with excess AF following CABG and that this risk was associated with the amount of transfusion. Another report from the Cleveland Clinic demonstrated that perioperative red blood cell transfusion is associated with an increased risk of postoperative morbidity after isolated CABG [28]. They did not clearly comment on AF but they noted that red cell transfusion was a significant risk factor for prolonged ventilation (>24 h), and this was compatible with the results of our study. Limitations This study was associated with some limitations. Although this study was performed using a national database to provide clinical information on a very high-volume cohort, we could not always obtain data on very specific issues with regard to the use of specific anastomosis assist devices or how they were applied to the aorta, or the degree of aortic disease. Our database is periodically reviewed and updated with more appropriate questionnaires. The newest version of the database (2017∼) contains information on the status of the aorta, and a further update of this study may provide more relevant results to better address our clinical questions. Another limitation was the fact that this study primarily focused on the detailed short-term outcomes and did not investigate the long-term results because the data were extracted from clinical results accumulated from 590 cardiac surgical units located throughout Japan. Thus far, the submission of 100% of the data to the JCVSD under third-party surveillance has only been achieved over the short term; however, it is anticipated that this national database will include long-term follow-up data in the future and thereby allow for the analysis of the long-term outcomes. CONCLUSIONS In conclusion, when the perioperative characteristics, short-term surgical results and risk profiles for OPCAB, were analysed. The use of an anastomosis assist device provided better outcomes than Side-clamp with respect to minimizing postoperative transient neurological complications but at the same time increased the risk of blood transfusion, new-onset AF and prolonged ventilation, and the overall benefit of using an anastomosis assist device was not demonstrated. Further stratification of the patients, such as by the severity of coronary artery disease, may yield additional information on the patients in whom device use is expected to be beneficial. In the future, it might be also necessary to investigate the economic benefit of device use by balancing the cost of the anastomosis assist device with the extra cost of postoperative complications. Further investigation of the means to reduce the rate of postoperative complications following OPCAB will require the appropriate selection of relevant procedures based on multiple aspects. ACKNOWLEDGEMENTS The authors thank Shinichi Takamoto for his valuable contribution to the organization of the data set and this article. Conflict of interest: none declared. REFERENCES 1 Biancari F, Yli-Pyky S. Meta-analysis on the use of the Heartstring anastomotic device to prevent stroke in patients undergoing off-pump coronary artery bypass grafting. Eur J Cardiothorac Surg  2011; 40: 1236– 40. Google Scholar PubMed  2 Thourani VH, Razavi SA, Nguyen TC, Kilgo PD, Puskas JD, Guyton RA et al.   Incidence of postoperative stroke using the Heartstring device in 1,380 coronary artery bypass graft patients with mild to severe atherosclerosis of the ascending aorta. Ann Thorac Surg  2014; 97: 2066– 72. Google Scholar CrossRef Search ADS PubMed  3 Arrigoni SC, Mecozzi G, Grandjean JG, Hillege JL, Kappetein AP, Mariani MA. Off-pump no-touch technique: 3-year results compared with the SYNTAX trial. Interact CardioVasc Thorac Surg  2015; 20: 601– 4. Google Scholar CrossRef Search ADS PubMed  4 Motomura N, Miyata H, Tsukihara H, Okada M, Takamoto S; Japan Cardiovascular Surgery Database Organization. First report on 30-day and operative mortality in risk model of isolated coronary artery bypass grafting in Japan. Ann Thorac Surg  2008; 86: 1866– 72. Google Scholar CrossRef Search ADS PubMed  5 Saito A, Motomura N, Miyata H, Takamoto S, Kyo S, Ono M. Age-specific risk stratification in 13488 isolated coronary artery bypass grafting procedures. Interact CardioVasc Thorac Surg  2011; 12: 575– 80. Google Scholar CrossRef Search ADS PubMed  6 Yang D, Dalton JE. A unified approach to measuring the effect size between two groups using SAS®. SAS Global Forum;  2012, 1– 6. Paper 335. 7 Coca-Perraillon M. Local and global optimal propensity score matching. SAS Global Forum ; 2007, 1– 9. Paper 185. 8 Lamy A, Tong W, Devereaux PJ, Gao P, Gafni A, Singh K et al.   The cost implications of off-pump versus on-pump coronary artery bypass graft surgery at one year. Ann Thorac Surg  2014; 98: 1620– 5. Google Scholar CrossRef Search ADS PubMed  9 Trehan N, Mishra M, Sharma OP, Mishra A, Kasliwal RR. Further reduction in stroke after off-pump coronary artery bypass grafting: a 10-year experience. Ann Thorac Surg  2001; 72: S1026– 32. Google Scholar CrossRef Search ADS PubMed  10 Halbersma WB, Arrigoni SC, Mecozzi G, Grandjean JG, Kappetein AP, van der Palen J et al.   Four-year outcome of OPCAB no-touch with total arterial Y-graft: making the best treatment a daily practice. Ann Thorac Surg  2009; 88: 796– 801. Google Scholar CrossRef Search ADS PubMed  11 Davierwala PM, Leontyev S, Misfeld M, Rastan A, Holzhey D, Lehmann S et al.   No-touch aorta off-pump coronary bypass operation: arteriovenous composite grafts may be used as a last resort. Ann Thorac Surg  2013; 95: 846– 52. Google Scholar CrossRef Search ADS PubMed  12 Zhao DF, Edelman JJ, Seco M, Bannon PG, Wilson MK, Byrom MJ et al.   Coronary artery bypass grafting with and without manipulation of the ascending aorta: a network meta-analysis. J Am Coll Cardiol  2017; 69: 924– 36. Google Scholar CrossRef Search ADS PubMed  13 Dohmen G, Hatam N, Goetzenich A, Mahnken A, Autschbach R, Spillner J. PAS-Port(R) clampless proximal anastomotic device for coronary bypass surgery in porcelain aorta. Eur J Cardiothorac Surg  2011; 39: 49– 52. Google Scholar CrossRef Search ADS PubMed  14 Aranki SF, Shekar PS, Ehsan A, Byrne-Taft M, Couper GS. Evaluation of the enclose proximal anastomosis device in coronary artery bypass grafting. Ann Thorac Surg  2005; 80: 1091– 5. Google Scholar CrossRef Search ADS PubMed  15 Akpinar B, Guden M, Sagbas E, Sanisoglu I, Ergenoglu MU, Turkoglu C. Clinical experience with the Novare Enclose II manual proximal anastomotic device during off-pump coronary artery surgery. Eur J Cardiothorac Surg  2005; 27: 1070– 3. Google Scholar CrossRef Search ADS PubMed  16 Medalion B, Meirson D, Hauptman E, Sasson L, Schachner A. Initial experience with the Heartstring proximal anastomotic system. J Thorac Cardiovasc Surg  2004; 128: 273– 7. Google Scholar CrossRef Search ADS PubMed  17 Biancari F, Lahtinen J, Ojala R, Ahvenjarvi L, Jartti A, Mosorin M et al.   Spyder aortic connector system in off-pump coronary artery bypass surgery. Ann Thorac Surg  2007; 84: 254– 7. Google Scholar CrossRef Search ADS PubMed  18 Saito A, Miyata H, Motomura N, Ono M, Takamoto S; Japan Cardiovascular Surgery Database Organization. Propensity-matched analysis of bilateral internal mammary artery vs single internal mammary artery in 7702 cases of isolated coronary artery bypass grafting. Eur J Cardiothorac Surg  2013; 44: 711– 17. Google Scholar CrossRef Search ADS PubMed  19 Patel NC, Hemli JM. Anastomotic devices in coronary artery surgery: it is about the anastomosis? Multimed Man Cardiothorac Surg  2013; 2013: mmt019. Google Scholar CrossRef Search ADS PubMed  20 Kawasaki M, Fujii T, Hara M, Sasaki Y, Katayanagi T, Okuma S et al.   Morphological evaluation of proximal anastomosis by PAS-Port((R)) system in patients with long-term patent grafts. Ann Thorac Cardiovasc Surg  2015; 21: 172– 7. Google Scholar CrossRef Search ADS PubMed  21 Ayad RF, Bhella PS, Dockery WD, Schussler JM. Patency of vein graft anastomoses facilitated with the hexalon device. Ann Thorac Surg  2011; 91: 894– 8. Google Scholar CrossRef Search ADS PubMed  22 Wiklund L, Setina M, Tsang K, Cusimano R, Yau T. A multicenter prospective randomized trial of a second-generation anastomotic device in coronary artery bypass surgery. J Thorac Cardiovasc Surg  2010; 139: 741– 7. Google Scholar CrossRef Search ADS PubMed  23 Kikuchi K, Tambara K, Yamamoto T, Yamasaki M, Hirose H, Amano A. The use of Enclose((R))II anastomosis assist device for the proximal coronary branch anastomosis to vascular graft. Ann Vasc Dis  2010; 3: 84– 6. Google Scholar CrossRef Search ADS PubMed  24 Puskas JD, Halkos ME, Balkhy H, Caskey M, Connolly M, Crouch J et al.   Evaluation of the PAS-Port Proximal Anastomosis System in coronary artery bypass surgery (the EPIC trial). J Thorac Cardiovasc Surg  2009; 138: 125– 32. Google Scholar CrossRef Search ADS PubMed  25 Manabe S, Fukui T, Miyajima K, Watanabe Y, Matsuyama S, Shimokawa T et al.   Impact of proximal anastomosis procedures on stroke in off-pump coronary artery bypass grafting. J Card Surg  2009; 24: 644– 9. Google Scholar CrossRef Search ADS PubMed  26 Athanasiou T, Asrafian H, Krasopoulos G, Purkayastha S, Malinowski V, Al-Ruzzeh S et al.   Clampless arterial coronary artery bypass grafting with the use of magnetic coupling devices. J Card Surg  2006; 21: 69– 74. Google Scholar CrossRef Search ADS PubMed  27 Alameddine AK, Visintainer P, Alimov VK, Rousou JA. Blood transfusion and the risk of atrial fibrillation after cardiac surgery. J Card Surg  2014; 29: 593– 9. Google Scholar CrossRef Search ADS PubMed  28 Koch CG, Li L, Duncan AI, Mihaljevic T, Cosgrove DM, Loop FD et al.   Morbidity and mortality risk associated with red blood cell and blood-component transfusion in isolated coronary artery bypass grafting. Crit Care Med  2006; 34: 1608– 16. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png European Journal of Cardio-Thoracic Surgery Oxford University Press

Propensity-matched analysis of a side-clamp versus an anastomosis assist device in cases of isolated coronary artery bypass grafting

Loading next page...
 
/lp/ou_press/propensity-matched-analysis-of-a-side-clamp-versus-an-anastomosis-Q2gRYD8Iyg
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
ISSN
1010-7940
eISSN
1873-734X
D.O.I.
10.1093/ejcts/ezy177
Publisher site
See Article on Publisher Site

Abstract

Abstract OBJECTIVES The use of an anastomosis assist device during coronary artery bypass grafting (CABG) is considered less invasive for the ascending aorta than the use of a side-biting clamp (Side-clamp) and to possibly be associated with a lower incidence of postoperative stroke. However, this benefit has not yet been clearly demonstrated. This study was to evaluate whether the use of an anastomosis assist device will minimize the postoperative stroke and other complications in patients undergoing off-pump CABG in comparison with the use of the Side-clamp. METHODS Patients undergoing isolated off-pump CABG were retrospectively reviewed using the Japan Adult Cardiovascular Surgery Database (2013–2016). We performed a one-to-one matched analysis based on the estimated propensity scores of those who underwent off-pump CABG with an anastomosis assist device (Device group, n = 14 213) or a side-biting clamp (Side-clamp group, n = 7374) and obtained 2 cohorts (n = 7348 each). We compared the early outcomes and the details of postoperative complications using the Pearson’s χ2 test. P-values of <0.05 were considered to indicate statistical significance. RESULTS No significant differences were observed in the rates of 30-day mortality (Side-clamp versus Device: 0.8% vs 0.8%, P = 0.93) or stroke (1.4% vs 1.4%, P = 0.46). Transient ischaemic attack/reversible ischaemic neurological deficit/delirium occurred more frequently in the Side-clamp group (1.3% vs 0.9%, P = 0.020), whereas new-onset atrial fibrillation (11.0% vs 12.8%, P < 0.001) and prolonged ventilation (2.0% vs 2.9%, P < 0.001) occurred more frequently in the Device group. There was no difference in the length of intensive care unit stay. CONCLUSIONS The use of an anastomosis assist device partially provided better results with regard to the transient neurological complications; however, no overall benefit was observed in this study. Anastomosis assist device, Off-pump coronary artery bypass grafting, Stroke, Side-biting clamp INTRODUCTION The manipulation of the ascending aorta during coronary artery bypass grafting (CABG) may result in various postoperative complications, such as stroke. The postoperative impact of a variety of aortic manipulations during CABG, for example the use of a side-biting clamp (Side-clamp), the use of an anastomosis assist device, the aorta no-touch technique and the use of cardiopulmonary bypass with arterial cannulation to the ascending aorta, has been evaluated in previous studies. Although cerebrovascular events have frequently among major postoperative complications, the discussion has not reached a conclusion [1–3], and the efficacy of the use of an anastomosis assist device in comparison to a side-biting clamp has not been adequately evaluated to date. This study evaluated the incidence of perioperative complications with stroke after off-pump CABG (OPCAB) as the primary end point. This was achieved by utilizing the nationwide Japan Adult Cardiovascular Surgery Database (JCVSD). A propensity-matched analysis was performed to compare the short-term outcomes of OPCAB procedures performed using the Side-clamp with those performed using an anastomosis assist device, in order to justify the use of an anastomosis assist device in the treatment of our patients. METHODS Japan Adult Cardiovascular Surgery Database JCVSD is a nationwide registry of patients undergoing cardiovascular surgery in Japan. Data collection was conducted using the National Clinical Database and was linked to cardiovascular surgery certification. Data from 590 cardiac surgery units located throughout Japan (close to 100% of the total number of such units) were recorded in the JCVSD adult section registry between January 2013 and December 2016. The methods of data collection and the data content of the JCVSD adult section have been previously described [4, 5]. The data registration project was approved by the institutional review board of the Japanese Surgical Society. The JCVSD variables and the relevant definitions (available online at http://jacvsd.umin.jp) were identical—for the most part—to those of the Society of Thoracic Surgeons (STS) National Adult Cardiac Database (available online at http://sts.org), with some modifications. Most of the data components collected in the database are required data components (i.e. the system forbids registration unless data are entered). As this study utilized only these main data components, almost no records were missing data. A limited number of cases [184 out of 56 102 isolated CABGs (0.3%)] that lacked information on the patients’ age, sex or 30-day postoperative status, which we believe arose from a system error, were deleted from the database. Study population Data on the patients who underwent isolated OPCAB using either the Side-clamp or an anastomosis assist device (i.e. HEARTSTRING, ENCLOSE II or PAS-PORT) in the years 2013–2016 and who were registered in the JCVSD were extracted from the database. The patients who underwent conversion to open-heart surgery as well as those without ascending aortic manipulation were excluded from this study (Fig. 1). Figure 1: View largeDownload slide The algorithm of patient enrolment. The number of isolated CABG cases from 2013 to 2016 was initially 56 102; a total of 21 587 cases were extracted after applying the exclusion criteria. CABG: coronary artery bypass grafting; JCVSD: Japan Adult Cardiovascular Surgery Database; OPCAB: off-pump CABG. Figure 1: View largeDownload slide The algorithm of patient enrolment. The number of isolated CABG cases from 2013 to 2016 was initially 56 102; a total of 21 587 cases were extracted after applying the exclusion criteria. CABG: coronary artery bypass grafting; JCVSD: Japan Adult Cardiovascular Surgery Database; OPCAB: off-pump CABG. Study outcome As the primary outcome, we assessed the incidence of postoperative stroke within 30 days after the surgery or during hospitalization. Stroke was defined in the registry as ‘cerebrovascular events with paralytic symptoms relevant to central nervous system persisting for more than 72 h during hospitalization’. As secondary outcomes, we assessed the rates of all-cause postoperative 30-day death, all-cause surgical death (death within 30 days or during the same hospitalization), transient neurological complications [transient ischaemic attack, reversible ischaemic neurological deficit or delirium], renal failure with a serum creatinine level of >2.0 mg/dl plus twice the level of preoperative serum creatinine, newly initiated dialysis, new-onset atrial fibrillation (AF) requiring any intervention, prolonged ventilation lasting >72 h, deep sternal infection (with infection penetrating the mediastinal cavity requiring debridement/drainage, with a positive culture, or for which antibiotics were administered), or reoperation for bleeding or graft occlusion. Complications that occurred within 30 days of surgery or during hospitalization were recorded in the database. We also assessed the operative time. Patient and procedural characteristics We extracted the information on patient characteristics and procedural characteristics which we considered to be associated with the outcomes, as well as those potentially affecting the surgeon’s choice between an anastomosis assist device or the Side-clamp, from the database. These included age, sex, body mass index, smoking status, comorbidities (diabetes mellitus, renal dysfunction, chronic dialysis, hyperlipidaemia, hypertension, carotid artery disease, chronic occlusive pulmonary disease, thoracic aortic disease, peripheral artery disease, congestive heart failure, history of cerebrovascular accident, past percutaneous coronary intervention and past CABG), preoperative conditions (presentation: acute myocardial infarction, angina, asymptomatic ischaemia, old myocardial infarction, other, presence of cardiac shock, New York Heart Association classification, left ventricular function and presence of mitral insufficiency), preoperative medication use [digitalis, beta blockers, acetylsalicylic acid (ASA), non-ASA antiplatelets, warfarin, direct oral anticoagulants, other anticoagulants and steroids] and procedural characteristics (urgency of surgery, whether or not a preoperative catheter examination was conducted, number of stenotic vessels, presence of left main trunk stenosis, use of the bilateral internal thoracic arteries and number of anastomoses). The patients were given ASA within 5 days before surgery. Statistical analysis We tabulated the patient characteristics by the method of aortic manipulation (i.e. Side-clamp or an anastomosis assist device), the number and percentage for categorical variables, and the mean and standard deviation for age. The characteristics were compared between the 2 groups using standardized mean differences, which were calculated using the SAS macro published by Yang and Dalton [6]. We estimated the propensity score (PS) for the use of an anastomosis assist device against the use of the Side-clamp for each patient, by fitting a logistic regression model with all of the above-mentioned patient and procedural characteristics. Then, using the SAS macro provided by Coca-Perraillon [7], we conducted greedy nearest neighbour matching between the 2 groups, with a caliper of 0.2 SD of the logit (PS). We checked the balance of the patient and procedural characteristics of the matched cohort using the above-described method. We compared the incidence of the study end points between the matched Side-clamp and Device patients using the McNemar’s test and Wilcoxon Signed-rank test for paired samples for binary outcomes and continuous outcomes, respectively. All of the statistical tests were 2-tailed, and P-values of <0.05 were considered to indicate statistical significance. All analyses were conducted using SAS 9.4 (SAS Institute Inc., Cary, NC, USA). RESULTS A total of 33 407 patients who underwent OPCAB during the study period in the JCVSD were identified. After excluding those who did not meet the inclusion criteria, data on 7374 patients and 14 213 patients who underwent OPCAB using the Side-clamp or an anastomosis assist device, respectively, were extracted from the database. Patient procedural characteristics In general, the patient and procedural characteristics of the 2 groups were similar, with a few exceptions (Table 1). In the Device group, the age of the patients was slightly older [mean age (standard deviation); Side-clamp 69.1 (9.8) years; Device 70.2 (9.4)], and the frequency of patients with a history of smoking was higher (Side-clamp 48.6%; Device 53.4%). The incidence of carotid artery disease in the patients in the Device group (9.6%) was higher than that in the patients in the Side-clamp group (7.1%). The frequency of procedures to treat old myocardial infarction was slightly higher in the Side-clamp group. The use of ASA and non-ASA antiplatelet therapy was higher among patients in the Side-clamp group, whereas anticoagulant use was higher among the patients in the Device group. A higher percentage of the patients in the Side-clamp group underwent a larger number (>3) of anastomoses. All other characteristics were similar between the 2 groups with a standardized difference of <0.07. Table 1: Preoperative characteristics of the patients Characteristics  Before matching     After matching     Side-clamp (n = 7374)  Device (n = 14 213)  SD  Side-clamp (n = 7348)  Device (n = 7348)  SD  Demographics   Age (years), mean (std)  69.1 (9.8)  70.2 (9.4)  −0.11  69.1 (9.8)  69.1 (9.8)  0.00   Male, n (%)  5681 (77)  11 154 (78.5)  −0.03  5663 (77.1)  5663 (77.1)  0.00   BMI ≥30 (kg/m2), n (%)  388 (5.3)  668 (4.7)  0.03  385 (5.2)  382 (5.2)  0.00   Smoking, n (%)  3584 (48.6)  7586 (53.4)  −0.10  3580 (48.7)  3561 (48)  0.01   Recent smoking (<1 month), n (%)  1184 (16.1)  2310 (16.3)  −0.01  1180 (16.1)  1188 (16.2)  0.00  Comorbidities, n (%)   Diabetes mellitus  4010 (54.4)  7690 (54.1)  0.01  3994 (54.4)  4008 (54.6)  0.00   Renal dysfunction  2044 (27.7)  3965 (27.9)  0.00  2036 (27.7)  2033 (27.7)  0.00   Dialysis  731 (9.9)  1545 (10.9)  −0.03  731 (10.0)  733 (10.0)  0.00   Hyperlipidaemia  4877 (66.1)  9070 (63.8)  0.05  4855 (66.1)  4853 (65.1)  0.00   Hypertension  5759 (78.1)  11 240 (79.1)  −0.02  5742 (78.1)  5739 (78.1)  0.00   Cerebrovascular accident  855 (11.6)  1871 (13.2)  −0.05  854 (11.6)  888 (12.1)  −0.01   Carotid artery disease  527 (7.1)  1368 (9.6)  −0.09  527 (7.2)  554 (7.5)  −0.01   COPD  1086 (14.7)  2144 (15.1)  −0.01  1084 (14.8)  1081 (14.7)  0.00   Thoracic aortic disease  89 (1.2)  279 (2)  −0.06  89 (1.2)  77 (1.1)  0.02   Peripheral artery disease  1082 (14.7)  2703 (19)  −0.12  1082 (14.7)  1092 (14.9)  0.00   Congestive heart failure  2045 (27.7)  3329 (23.4)  0.10  2023 (27.5)  2003 (27.3)  0.01   Past PCI  1907 (25.9)  3587 (25.2)  0.01  1898 (25.8)  1910 (26.0)  0.00   Past CABG  30 (0.4)  83 (0.6)  0.00  0  0  0.00  Presentation, n (%)      0.08      0.02   Primary diagnosis                Acute myocardial infarction  1519 (10.7)  801 (10.9)    793 (10.8)  784 (10.7)      Angina  11 084 (78.0)  5721 (77.6)    5709 (77.7)  5689 (77.4)      Asymptomatic ischaemia  535 (3.8)  241 (3.3)    239 (3.3)  252 (3.4)      Old myocardial infarction  590 (4.2)  261 (3.5)    261 (3.6)  283 (3.9)      Other  485 (3.4)  350 (4.8)    346 (4.7)  340 (4.6)     Cardiac shock  206 (2.8)  346 (2.4)  0.02  202 (2.8)  220 (3.0)  −0.01   NYHA classification III–IV  1267 (17.2)  2298 (16.2)  0.03  1259 (17.1)  1257 (17.1)  0.00   Left ventricle function      0.01      0.01    Good  3735 (50.7)  7270 (51.2)    3725 (50.7)  3687 (50.2)      Medium  3284 (44.5)  6293 (44.3)    3269 (44.5)  3313 (45.1)      Bad  355 (4.8)  650 (4.6)    354 (4.8)  348 (4.7)     Mitral insufficiency II/III/IV  1236 (16.8)  2364 (16.6)  0.00  1232 (16.8)  1207 (16.4)  0.01   Preoperative catheter exam      0.04      0.02    Performed: LMT/3-vessel disease  2447 (33.2)  4706 (33.1)    2439 (33.2)  2463 (33.5)      Performed: LMT/1–2 vessel disease  519 (7.0)  1119 (7.9)    581 (7.1)  515 (7.0)      Performed: no LMT/3-vessel disease  3341 (45.3)  6271 (44.1)    3328 (45.3)  3315 (45.1)      Performed: no LMT/1–2 vessel disease  889 (12.1)  1727 (12.2)    885 (12.0)  861 (11.7)      Not performed  178 (2.4)  390 (2.7)    178 (2.4)  194 (2.6)    Procedural characteristics, n (%)   Bilateral ITA use  2648 (35.9)  4919 (34.6)  0.03  2642 (36.0)  2663 (36.2)  −0.01   Number of anastomosis      0.08      0.01    <2  1633 (22.1)  3570 (25.1)    1630 (22.2)  1630 (22.2)      3  2911 (39.5)  5548 (39)    2895 (39.4)  2880 (39.2)      4 or 5  2602 (35.3)  4710 (33.1)    2595 (35.3)  2614 (35.6)      6 or more  228 (3.1)  385 (2.7)    228 (3.1)  224 (3.1)     Urgency      0.05      0.01    Elective  6162 (83.6)  12 095 (85.1)    6145 (83.6)  6143 (83.6)      Urgent  720 (9.8)  1348 (9.5)    717 (9.8)  705 (9.6)      Emergent  492 (6.7)  770 (5.4)    486 (6.6)  500 (6.8)    Preoperative medication use, n (%)   Digitalis  28 (0.4)  72 (0.5)  −0.02  28 (0.4)  28 (0.4)  0.00   Beta blocker  2629 (35.7)  4947 (34.8)  0.02  2620 (35.7)  2603 (35.4)  0.00   ASA  3469 (47)  5295 (37.3)  0.20  3444 (46.9)  3478 (47.3)  −0.01   Non-ASA antiplatelets  777 (10.5)  1177 (8.3)  0.08  763 (10.4)  791 (10.8)  −0.01   Warfarin  2101 (28.5)  4027 (28.3)  0.00  2097 (28.5)  2077 (28.3)  0.01   Warfarin replacements  42 (0.6)  108 (0.8)  −0.02  42 (0.6)  37 (0.5)  0.01   Anticoags  771 (10.5)  2122 (14.9)  −0.13  771 (10.5)  768 (10.5)  0.00   Steroids  81 (1.1)  210 (1.5)  −0.03  80 (1.1)  88 (1.2)  −0.01  Characteristics  Before matching     After matching     Side-clamp (n = 7374)  Device (n = 14 213)  SD  Side-clamp (n = 7348)  Device (n = 7348)  SD  Demographics   Age (years), mean (std)  69.1 (9.8)  70.2 (9.4)  −0.11  69.1 (9.8)  69.1 (9.8)  0.00   Male, n (%)  5681 (77)  11 154 (78.5)  −0.03  5663 (77.1)  5663 (77.1)  0.00   BMI ≥30 (kg/m2), n (%)  388 (5.3)  668 (4.7)  0.03  385 (5.2)  382 (5.2)  0.00   Smoking, n (%)  3584 (48.6)  7586 (53.4)  −0.10  3580 (48.7)  3561 (48)  0.01   Recent smoking (<1 month), n (%)  1184 (16.1)  2310 (16.3)  −0.01  1180 (16.1)  1188 (16.2)  0.00  Comorbidities, n (%)   Diabetes mellitus  4010 (54.4)  7690 (54.1)  0.01  3994 (54.4)  4008 (54.6)  0.00   Renal dysfunction  2044 (27.7)  3965 (27.9)  0.00  2036 (27.7)  2033 (27.7)  0.00   Dialysis  731 (9.9)  1545 (10.9)  −0.03  731 (10.0)  733 (10.0)  0.00   Hyperlipidaemia  4877 (66.1)  9070 (63.8)  0.05  4855 (66.1)  4853 (65.1)  0.00   Hypertension  5759 (78.1)  11 240 (79.1)  −0.02  5742 (78.1)  5739 (78.1)  0.00   Cerebrovascular accident  855 (11.6)  1871 (13.2)  −0.05  854 (11.6)  888 (12.1)  −0.01   Carotid artery disease  527 (7.1)  1368 (9.6)  −0.09  527 (7.2)  554 (7.5)  −0.01   COPD  1086 (14.7)  2144 (15.1)  −0.01  1084 (14.8)  1081 (14.7)  0.00   Thoracic aortic disease  89 (1.2)  279 (2)  −0.06  89 (1.2)  77 (1.1)  0.02   Peripheral artery disease  1082 (14.7)  2703 (19)  −0.12  1082 (14.7)  1092 (14.9)  0.00   Congestive heart failure  2045 (27.7)  3329 (23.4)  0.10  2023 (27.5)  2003 (27.3)  0.01   Past PCI  1907 (25.9)  3587 (25.2)  0.01  1898 (25.8)  1910 (26.0)  0.00   Past CABG  30 (0.4)  83 (0.6)  0.00  0  0  0.00  Presentation, n (%)      0.08      0.02   Primary diagnosis                Acute myocardial infarction  1519 (10.7)  801 (10.9)    793 (10.8)  784 (10.7)      Angina  11 084 (78.0)  5721 (77.6)    5709 (77.7)  5689 (77.4)      Asymptomatic ischaemia  535 (3.8)  241 (3.3)    239 (3.3)  252 (3.4)      Old myocardial infarction  590 (4.2)  261 (3.5)    261 (3.6)  283 (3.9)      Other  485 (3.4)  350 (4.8)    346 (4.7)  340 (4.6)     Cardiac shock  206 (2.8)  346 (2.4)  0.02  202 (2.8)  220 (3.0)  −0.01   NYHA classification III–IV  1267 (17.2)  2298 (16.2)  0.03  1259 (17.1)  1257 (17.1)  0.00   Left ventricle function      0.01      0.01    Good  3735 (50.7)  7270 (51.2)    3725 (50.7)  3687 (50.2)      Medium  3284 (44.5)  6293 (44.3)    3269 (44.5)  3313 (45.1)      Bad  355 (4.8)  650 (4.6)    354 (4.8)  348 (4.7)     Mitral insufficiency II/III/IV  1236 (16.8)  2364 (16.6)  0.00  1232 (16.8)  1207 (16.4)  0.01   Preoperative catheter exam      0.04      0.02    Performed: LMT/3-vessel disease  2447 (33.2)  4706 (33.1)    2439 (33.2)  2463 (33.5)      Performed: LMT/1–2 vessel disease  519 (7.0)  1119 (7.9)    581 (7.1)  515 (7.0)      Performed: no LMT/3-vessel disease  3341 (45.3)  6271 (44.1)    3328 (45.3)  3315 (45.1)      Performed: no LMT/1–2 vessel disease  889 (12.1)  1727 (12.2)    885 (12.0)  861 (11.7)      Not performed  178 (2.4)  390 (2.7)    178 (2.4)  194 (2.6)    Procedural characteristics, n (%)   Bilateral ITA use  2648 (35.9)  4919 (34.6)  0.03  2642 (36.0)  2663 (36.2)  −0.01   Number of anastomosis      0.08      0.01    <2  1633 (22.1)  3570 (25.1)    1630 (22.2)  1630 (22.2)      3  2911 (39.5)  5548 (39)    2895 (39.4)  2880 (39.2)      4 or 5  2602 (35.3)  4710 (33.1)    2595 (35.3)  2614 (35.6)      6 or more  228 (3.1)  385 (2.7)    228 (3.1)  224 (3.1)     Urgency      0.05      0.01    Elective  6162 (83.6)  12 095 (85.1)    6145 (83.6)  6143 (83.6)      Urgent  720 (9.8)  1348 (9.5)    717 (9.8)  705 (9.6)      Emergent  492 (6.7)  770 (5.4)    486 (6.6)  500 (6.8)    Preoperative medication use, n (%)   Digitalis  28 (0.4)  72 (0.5)  −0.02  28 (0.4)  28 (0.4)  0.00   Beta blocker  2629 (35.7)  4947 (34.8)  0.02  2620 (35.7)  2603 (35.4)  0.00   ASA  3469 (47)  5295 (37.3)  0.20  3444 (46.9)  3478 (47.3)  −0.01   Non-ASA antiplatelets  777 (10.5)  1177 (8.3)  0.08  763 (10.4)  791 (10.8)  −0.01   Warfarin  2101 (28.5)  4027 (28.3)  0.00  2097 (28.5)  2077 (28.3)  0.01   Warfarin replacements  42 (0.6)  108 (0.8)  −0.02  42 (0.6)  37 (0.5)  0.01   Anticoags  771 (10.5)  2122 (14.9)  −0.13  771 (10.5)  768 (10.5)  0.00   Steroids  81 (1.1)  210 (1.5)  −0.03  80 (1.1)  88 (1.2)  −0.01  ASA: acetylsalicylic acid; BMI: body mass index; CABG: coronary artery bypass graft; COPD: chronic obstructive pulmonary disease; ITA: internal thoracic artery; LMT: left main trunk; NYHA: New York Heart Association; PCI: percutaneous transcatheter intervention; SD: standardized difference; std: standard deviation. Table 1: Preoperative characteristics of the patients Characteristics  Before matching     After matching     Side-clamp (n = 7374)  Device (n = 14 213)  SD  Side-clamp (n = 7348)  Device (n = 7348)  SD  Demographics   Age (years), mean (std)  69.1 (9.8)  70.2 (9.4)  −0.11  69.1 (9.8)  69.1 (9.8)  0.00   Male, n (%)  5681 (77)  11 154 (78.5)  −0.03  5663 (77.1)  5663 (77.1)  0.00   BMI ≥30 (kg/m2), n (%)  388 (5.3)  668 (4.7)  0.03  385 (5.2)  382 (5.2)  0.00   Smoking, n (%)  3584 (48.6)  7586 (53.4)  −0.10  3580 (48.7)  3561 (48)  0.01   Recent smoking (<1 month), n (%)  1184 (16.1)  2310 (16.3)  −0.01  1180 (16.1)  1188 (16.2)  0.00  Comorbidities, n (%)   Diabetes mellitus  4010 (54.4)  7690 (54.1)  0.01  3994 (54.4)  4008 (54.6)  0.00   Renal dysfunction  2044 (27.7)  3965 (27.9)  0.00  2036 (27.7)  2033 (27.7)  0.00   Dialysis  731 (9.9)  1545 (10.9)  −0.03  731 (10.0)  733 (10.0)  0.00   Hyperlipidaemia  4877 (66.1)  9070 (63.8)  0.05  4855 (66.1)  4853 (65.1)  0.00   Hypertension  5759 (78.1)  11 240 (79.1)  −0.02  5742 (78.1)  5739 (78.1)  0.00   Cerebrovascular accident  855 (11.6)  1871 (13.2)  −0.05  854 (11.6)  888 (12.1)  −0.01   Carotid artery disease  527 (7.1)  1368 (9.6)  −0.09  527 (7.2)  554 (7.5)  −0.01   COPD  1086 (14.7)  2144 (15.1)  −0.01  1084 (14.8)  1081 (14.7)  0.00   Thoracic aortic disease  89 (1.2)  279 (2)  −0.06  89 (1.2)  77 (1.1)  0.02   Peripheral artery disease  1082 (14.7)  2703 (19)  −0.12  1082 (14.7)  1092 (14.9)  0.00   Congestive heart failure  2045 (27.7)  3329 (23.4)  0.10  2023 (27.5)  2003 (27.3)  0.01   Past PCI  1907 (25.9)  3587 (25.2)  0.01  1898 (25.8)  1910 (26.0)  0.00   Past CABG  30 (0.4)  83 (0.6)  0.00  0  0  0.00  Presentation, n (%)      0.08      0.02   Primary diagnosis                Acute myocardial infarction  1519 (10.7)  801 (10.9)    793 (10.8)  784 (10.7)      Angina  11 084 (78.0)  5721 (77.6)    5709 (77.7)  5689 (77.4)      Asymptomatic ischaemia  535 (3.8)  241 (3.3)    239 (3.3)  252 (3.4)      Old myocardial infarction  590 (4.2)  261 (3.5)    261 (3.6)  283 (3.9)      Other  485 (3.4)  350 (4.8)    346 (4.7)  340 (4.6)     Cardiac shock  206 (2.8)  346 (2.4)  0.02  202 (2.8)  220 (3.0)  −0.01   NYHA classification III–IV  1267 (17.2)  2298 (16.2)  0.03  1259 (17.1)  1257 (17.1)  0.00   Left ventricle function      0.01      0.01    Good  3735 (50.7)  7270 (51.2)    3725 (50.7)  3687 (50.2)      Medium  3284 (44.5)  6293 (44.3)    3269 (44.5)  3313 (45.1)      Bad  355 (4.8)  650 (4.6)    354 (4.8)  348 (4.7)     Mitral insufficiency II/III/IV  1236 (16.8)  2364 (16.6)  0.00  1232 (16.8)  1207 (16.4)  0.01   Preoperative catheter exam      0.04      0.02    Performed: LMT/3-vessel disease  2447 (33.2)  4706 (33.1)    2439 (33.2)  2463 (33.5)      Performed: LMT/1–2 vessel disease  519 (7.0)  1119 (7.9)    581 (7.1)  515 (7.0)      Performed: no LMT/3-vessel disease  3341 (45.3)  6271 (44.1)    3328 (45.3)  3315 (45.1)      Performed: no LMT/1–2 vessel disease  889 (12.1)  1727 (12.2)    885 (12.0)  861 (11.7)      Not performed  178 (2.4)  390 (2.7)    178 (2.4)  194 (2.6)    Procedural characteristics, n (%)   Bilateral ITA use  2648 (35.9)  4919 (34.6)  0.03  2642 (36.0)  2663 (36.2)  −0.01   Number of anastomosis      0.08      0.01    <2  1633 (22.1)  3570 (25.1)    1630 (22.2)  1630 (22.2)      3  2911 (39.5)  5548 (39)    2895 (39.4)  2880 (39.2)      4 or 5  2602 (35.3)  4710 (33.1)    2595 (35.3)  2614 (35.6)      6 or more  228 (3.1)  385 (2.7)    228 (3.1)  224 (3.1)     Urgency      0.05      0.01    Elective  6162 (83.6)  12 095 (85.1)    6145 (83.6)  6143 (83.6)      Urgent  720 (9.8)  1348 (9.5)    717 (9.8)  705 (9.6)      Emergent  492 (6.7)  770 (5.4)    486 (6.6)  500 (6.8)    Preoperative medication use, n (%)   Digitalis  28 (0.4)  72 (0.5)  −0.02  28 (0.4)  28 (0.4)  0.00   Beta blocker  2629 (35.7)  4947 (34.8)  0.02  2620 (35.7)  2603 (35.4)  0.00   ASA  3469 (47)  5295 (37.3)  0.20  3444 (46.9)  3478 (47.3)  −0.01   Non-ASA antiplatelets  777 (10.5)  1177 (8.3)  0.08  763 (10.4)  791 (10.8)  −0.01   Warfarin  2101 (28.5)  4027 (28.3)  0.00  2097 (28.5)  2077 (28.3)  0.01   Warfarin replacements  42 (0.6)  108 (0.8)  −0.02  42 (0.6)  37 (0.5)  0.01   Anticoags  771 (10.5)  2122 (14.9)  −0.13  771 (10.5)  768 (10.5)  0.00   Steroids  81 (1.1)  210 (1.5)  −0.03  80 (1.1)  88 (1.2)  −0.01  Characteristics  Before matching     After matching     Side-clamp (n = 7374)  Device (n = 14 213)  SD  Side-clamp (n = 7348)  Device (n = 7348)  SD  Demographics   Age (years), mean (std)  69.1 (9.8)  70.2 (9.4)  −0.11  69.1 (9.8)  69.1 (9.8)  0.00   Male, n (%)  5681 (77)  11 154 (78.5)  −0.03  5663 (77.1)  5663 (77.1)  0.00   BMI ≥30 (kg/m2), n (%)  388 (5.3)  668 (4.7)  0.03  385 (5.2)  382 (5.2)  0.00   Smoking, n (%)  3584 (48.6)  7586 (53.4)  −0.10  3580 (48.7)  3561 (48)  0.01   Recent smoking (<1 month), n (%)  1184 (16.1)  2310 (16.3)  −0.01  1180 (16.1)  1188 (16.2)  0.00  Comorbidities, n (%)   Diabetes mellitus  4010 (54.4)  7690 (54.1)  0.01  3994 (54.4)  4008 (54.6)  0.00   Renal dysfunction  2044 (27.7)  3965 (27.9)  0.00  2036 (27.7)  2033 (27.7)  0.00   Dialysis  731 (9.9)  1545 (10.9)  −0.03  731 (10.0)  733 (10.0)  0.00   Hyperlipidaemia  4877 (66.1)  9070 (63.8)  0.05  4855 (66.1)  4853 (65.1)  0.00   Hypertension  5759 (78.1)  11 240 (79.1)  −0.02  5742 (78.1)  5739 (78.1)  0.00   Cerebrovascular accident  855 (11.6)  1871 (13.2)  −0.05  854 (11.6)  888 (12.1)  −0.01   Carotid artery disease  527 (7.1)  1368 (9.6)  −0.09  527 (7.2)  554 (7.5)  −0.01   COPD  1086 (14.7)  2144 (15.1)  −0.01  1084 (14.8)  1081 (14.7)  0.00   Thoracic aortic disease  89 (1.2)  279 (2)  −0.06  89 (1.2)  77 (1.1)  0.02   Peripheral artery disease  1082 (14.7)  2703 (19)  −0.12  1082 (14.7)  1092 (14.9)  0.00   Congestive heart failure  2045 (27.7)  3329 (23.4)  0.10  2023 (27.5)  2003 (27.3)  0.01   Past PCI  1907 (25.9)  3587 (25.2)  0.01  1898 (25.8)  1910 (26.0)  0.00   Past CABG  30 (0.4)  83 (0.6)  0.00  0  0  0.00  Presentation, n (%)      0.08      0.02   Primary diagnosis                Acute myocardial infarction  1519 (10.7)  801 (10.9)    793 (10.8)  784 (10.7)      Angina  11 084 (78.0)  5721 (77.6)    5709 (77.7)  5689 (77.4)      Asymptomatic ischaemia  535 (3.8)  241 (3.3)    239 (3.3)  252 (3.4)      Old myocardial infarction  590 (4.2)  261 (3.5)    261 (3.6)  283 (3.9)      Other  485 (3.4)  350 (4.8)    346 (4.7)  340 (4.6)     Cardiac shock  206 (2.8)  346 (2.4)  0.02  202 (2.8)  220 (3.0)  −0.01   NYHA classification III–IV  1267 (17.2)  2298 (16.2)  0.03  1259 (17.1)  1257 (17.1)  0.00   Left ventricle function      0.01      0.01    Good  3735 (50.7)  7270 (51.2)    3725 (50.7)  3687 (50.2)      Medium  3284 (44.5)  6293 (44.3)    3269 (44.5)  3313 (45.1)      Bad  355 (4.8)  650 (4.6)    354 (4.8)  348 (4.7)     Mitral insufficiency II/III/IV  1236 (16.8)  2364 (16.6)  0.00  1232 (16.8)  1207 (16.4)  0.01   Preoperative catheter exam      0.04      0.02    Performed: LMT/3-vessel disease  2447 (33.2)  4706 (33.1)    2439 (33.2)  2463 (33.5)      Performed: LMT/1–2 vessel disease  519 (7.0)  1119 (7.9)    581 (7.1)  515 (7.0)      Performed: no LMT/3-vessel disease  3341 (45.3)  6271 (44.1)    3328 (45.3)  3315 (45.1)      Performed: no LMT/1–2 vessel disease  889 (12.1)  1727 (12.2)    885 (12.0)  861 (11.7)      Not performed  178 (2.4)  390 (2.7)    178 (2.4)  194 (2.6)    Procedural characteristics, n (%)   Bilateral ITA use  2648 (35.9)  4919 (34.6)  0.03  2642 (36.0)  2663 (36.2)  −0.01   Number of anastomosis      0.08      0.01    <2  1633 (22.1)  3570 (25.1)    1630 (22.2)  1630 (22.2)      3  2911 (39.5)  5548 (39)    2895 (39.4)  2880 (39.2)      4 or 5  2602 (35.3)  4710 (33.1)    2595 (35.3)  2614 (35.6)      6 or more  228 (3.1)  385 (2.7)    228 (3.1)  224 (3.1)     Urgency      0.05      0.01    Elective  6162 (83.6)  12 095 (85.1)    6145 (83.6)  6143 (83.6)      Urgent  720 (9.8)  1348 (9.5)    717 (9.8)  705 (9.6)      Emergent  492 (6.7)  770 (5.4)    486 (6.6)  500 (6.8)    Preoperative medication use, n (%)   Digitalis  28 (0.4)  72 (0.5)  −0.02  28 (0.4)  28 (0.4)  0.00   Beta blocker  2629 (35.7)  4947 (34.8)  0.02  2620 (35.7)  2603 (35.4)  0.00   ASA  3469 (47)  5295 (37.3)  0.20  3444 (46.9)  3478 (47.3)  −0.01   Non-ASA antiplatelets  777 (10.5)  1177 (8.3)  0.08  763 (10.4)  791 (10.8)  −0.01   Warfarin  2101 (28.5)  4027 (28.3)  0.00  2097 (28.5)  2077 (28.3)  0.01   Warfarin replacements  42 (0.6)  108 (0.8)  −0.02  42 (0.6)  37 (0.5)  0.01   Anticoags  771 (10.5)  2122 (14.9)  −0.13  771 (10.5)  768 (10.5)  0.00   Steroids  81 (1.1)  210 (1.5)  −0.03  80 (1.1)  88 (1.2)  −0.01  ASA: acetylsalicylic acid; BMI: body mass index; CABG: coronary artery bypass graft; COPD: chronic obstructive pulmonary disease; ITA: internal thoracic artery; LMT: left main trunk; NYHA: New York Heart Association; PCI: percutaneous transcatheter intervention; SD: standardized difference; std: standard deviation. The propensity score distribution and balance of characteristics in the matched cohort The c-statistic of the PS model was 0.602. As expected, there was a large amount of overlap in the distribution of the PSs between the 2 groups before matching (Fig. 2). We matched 7348 Side-clamp and 7348 Device patients based on their PSs. The distribution of the PS values of the 2 groups were almost completely identical in the matched cohort (Fig. 2). The balance of the patient and procedural characteristics between the 2 groups was further improved, with a standardized mean difference of <0.03 for all variables. Figure 2: View largeDownload slide The propensity score distribution of the Side-clamp (blue solid line) and the Device (red solid line) groups (A) before and (B) after matching are demonstrated, which shows that the 2 cohorts were very well matched. Figure 2: View largeDownload slide The propensity score distribution of the Side-clamp (blue solid line) and the Device (red solid line) groups (A) before and (B) after matching are demonstrated, which shows that the 2 cohorts were very well matched. Short-term clinical results The short-term postoperative outcomes in the matched cohort are presented in Table 2. The incidence of the primary outcome of stroke was similar in the 2 groups, with 99 (1.4%) patients in the Side-clamp group and 89 (1.2%) patients in the Device group (P = 0.46). The rates of 30-day and operative mortality were also similar in the 2 groups (operative mortality: Side-clamp 1.4%; Device 1.5%; P = 0.78). The incidence of major complications [operative death, stroke, newly initiated dialysis, prolonged ventilation (>24 h) due to pulmonary complications, deep sternal wound infection and reoperation for bleeding] did not differ to a statistically significant extent (Side-clamp 7.0%; Device 7.5%; P = 0.23). With regard to the postoperative morbidities, significant differences were observed in the incidence of transient ischaemic attack/reversible ischaemic neurological deficit/delirium [Side-clamp n = 93 (1.3%); Device n = 64 (0.9%); P = 0.020], new-onset AF [Side-clamp n = 807 (11.0%); Device n = 943 (12.8%); P <0.001] and prolonged ventilation [Side-clamp n = 147 (2%); Device n = 212 (2.9%); P < 0.001). The overall length of intensive care unit (ICU) stay was similar. With regard to the procedural results, the patients in the Device required homologous blood transfusion more frequently than those in the Side-clamp group (Side-clamp 54.6%; Device 60.0%; P < 0.001), and the average operation time was significantly shorter in the Side-clamp group [mean (standard deviation); Side-clamp 304.5 (99.5) min; Device 319.5 (97.9) min]. The days in the ICU were very similar between the 2 groups but were found to be marginally shorter for the Side-clamp group (median days in ICU: 3 days for Side-clamp, 3 days for Device, P-value 0.036). Table 2: Comparison of the early postoperative results between ‘Side-clamp’ and ‘Device’ Outcomes  Side-clamp (n = 7348)  Device (n = 7348)  P-value*  30-Day death, n (%)  61 (0.8)  62 (0.8)  0.93  Operative death, n (%)  106 (1.4)  110 (1.5)  0.78  Major complicationsa, n (%)  514 (7.0)  552 (7.5)  0.23  Stroke, n (%)  99 (1.4)  89 (1.2)  0.46  TIA/RIND/delirium, n (%)  93 (1.3)  64 (0.9)  0.020  Renal failure, n (%)  174 (2.4)  196 (2.7)  0.25  Newly initiated dialysis, n (%)  123 (1.7)  123 (1.7)  1.00  New-onset AF, n (%)  807 (11.0)  943 (12.8)  <0.001  Prolonged ventilationb, n (%)  147 (2.0)  212 (2.9)  <0.001  Postoperative ventilator use (h), median (25th–75th percentile)  8 (4–16.5)  8 (4–17)  <0.001  Deep sternal wound infection, n (%)  107 (1.5)  106 (1.4)  0.95  Reoperation for bleeding, n (%)  83 (1.1)  79 (1.1)  0.75  Reoperation for graft occlusion, n (%)  95 (1.3)  95 (1.3)  1.00  Days in ICU, median (25th–75th percentile)  3 (2–4)  3 (2–4)  0.12  Homologous transfusion use, n (%)  4008 (54.6)  4406 (60.0)  <0.001  Operation time (min), median (25th–75th percentile)  297 (241–357)  309 (255–370)  <0.001  Outcomes  Side-clamp (n = 7348)  Device (n = 7348)  P-value*  30-Day death, n (%)  61 (0.8)  62 (0.8)  0.93  Operative death, n (%)  106 (1.4)  110 (1.5)  0.78  Major complicationsa, n (%)  514 (7.0)  552 (7.5)  0.23  Stroke, n (%)  99 (1.4)  89 (1.2)  0.46  TIA/RIND/delirium, n (%)  93 (1.3)  64 (0.9)  0.020  Renal failure, n (%)  174 (2.4)  196 (2.7)  0.25  Newly initiated dialysis, n (%)  123 (1.7)  123 (1.7)  1.00  New-onset AF, n (%)  807 (11.0)  943 (12.8)  <0.001  Prolonged ventilationb, n (%)  147 (2.0)  212 (2.9)  <0.001  Postoperative ventilator use (h), median (25th–75th percentile)  8 (4–16.5)  8 (4–17)  <0.001  Deep sternal wound infection, n (%)  107 (1.5)  106 (1.4)  0.95  Reoperation for bleeding, n (%)  83 (1.1)  79 (1.1)  0.75  Reoperation for graft occlusion, n (%)  95 (1.3)  95 (1.3)  1.00  Days in ICU, median (25th–75th percentile)  3 (2–4)  3 (2–4)  0.12  Homologous transfusion use, n (%)  4008 (54.6)  4406 (60.0)  <0.001  Operation time (min), median (25th–75th percentile)  297 (241–357)  309 (255–370)  <0.001  a Major complications include operative death, stroke, newly initiated dialysis, prolonged ventilation (>24 h) due to pulmonary complications, deep sternal wound infection and reoperation for bleeds. b Ventilation >24 h due to pulmonary complications. * P-values from Pearson’s χ2 test or Wilcoxon rank-sum test. AF: atrial fibrillation; ICU: intensive care unit; RIND: reversible ischaemic neurological deficit; TIA: transient ischaemic attack. Table 2: Comparison of the early postoperative results between ‘Side-clamp’ and ‘Device’ Outcomes  Side-clamp (n = 7348)  Device (n = 7348)  P-value*  30-Day death, n (%)  61 (0.8)  62 (0.8)  0.93  Operative death, n (%)  106 (1.4)  110 (1.5)  0.78  Major complicationsa, n (%)  514 (7.0)  552 (7.5)  0.23  Stroke, n (%)  99 (1.4)  89 (1.2)  0.46  TIA/RIND/delirium, n (%)  93 (1.3)  64 (0.9)  0.020  Renal failure, n (%)  174 (2.4)  196 (2.7)  0.25  Newly initiated dialysis, n (%)  123 (1.7)  123 (1.7)  1.00  New-onset AF, n (%)  807 (11.0)  943 (12.8)  <0.001  Prolonged ventilationb, n (%)  147 (2.0)  212 (2.9)  <0.001  Postoperative ventilator use (h), median (25th–75th percentile)  8 (4–16.5)  8 (4–17)  <0.001  Deep sternal wound infection, n (%)  107 (1.5)  106 (1.4)  0.95  Reoperation for bleeding, n (%)  83 (1.1)  79 (1.1)  0.75  Reoperation for graft occlusion, n (%)  95 (1.3)  95 (1.3)  1.00  Days in ICU, median (25th–75th percentile)  3 (2–4)  3 (2–4)  0.12  Homologous transfusion use, n (%)  4008 (54.6)  4406 (60.0)  <0.001  Operation time (min), median (25th–75th percentile)  297 (241–357)  309 (255–370)  <0.001  Outcomes  Side-clamp (n = 7348)  Device (n = 7348)  P-value*  30-Day death, n (%)  61 (0.8)  62 (0.8)  0.93  Operative death, n (%)  106 (1.4)  110 (1.5)  0.78  Major complicationsa, n (%)  514 (7.0)  552 (7.5)  0.23  Stroke, n (%)  99 (1.4)  89 (1.2)  0.46  TIA/RIND/delirium, n (%)  93 (1.3)  64 (0.9)  0.020  Renal failure, n (%)  174 (2.4)  196 (2.7)  0.25  Newly initiated dialysis, n (%)  123 (1.7)  123 (1.7)  1.00  New-onset AF, n (%)  807 (11.0)  943 (12.8)  <0.001  Prolonged ventilationb, n (%)  147 (2.0)  212 (2.9)  <0.001  Postoperative ventilator use (h), median (25th–75th percentile)  8 (4–16.5)  8 (4–17)  <0.001  Deep sternal wound infection, n (%)  107 (1.5)  106 (1.4)  0.95  Reoperation for bleeding, n (%)  83 (1.1)  79 (1.1)  0.75  Reoperation for graft occlusion, n (%)  95 (1.3)  95 (1.3)  1.00  Days in ICU, median (25th–75th percentile)  3 (2–4)  3 (2–4)  0.12  Homologous transfusion use, n (%)  4008 (54.6)  4406 (60.0)  <0.001  Operation time (min), median (25th–75th percentile)  297 (241–357)  309 (255–370)  <0.001  a Major complications include operative death, stroke, newly initiated dialysis, prolonged ventilation (>24 h) due to pulmonary complications, deep sternal wound infection and reoperation for bleeds. b Ventilation >24 h due to pulmonary complications. * P-values from Pearson’s χ2 test or Wilcoxon rank-sum test. AF: atrial fibrillation; ICU: intensive care unit; RIND: reversible ischaemic neurological deficit; TIA: transient ischaemic attack. DISCUSSION During CABG, the manipulation of the ascending aorta, for which various degrees of invasive manipulation have been conceived, is often thought to increase the risk of postoperative cerebrovascular events. These include arterial cannulation for the establishment of cardiopulmonary bypass, the use of cross clamping, the use of Side-clamps and the use of anastomosis assist devices. The risks associated with the above-mentioned manipulations have been investigated in a series of studies [1–3, 8, 9]. In addition, the efficacy of the aorta no-touch technique has also been discussed previously [10–12]. In this study, we focused on the benefit of using an anastomosis assist device in reducing postoperative cerebrovascular events, especially stroke; however, at the time of writing, a consensus has not been reached with regard to the efficacy of anastomosis assist devices in this respect [1, 12–18]. The current study question originally arose from a previous report [13], in which we hypothesized that the frequent use of anastomosis assist devices might be effective for reducing postoperative cerebrovascular complications. As ∼60% of CABG procedures are performed in an off-pump manner, and given the fact that anastomosis assist devices such as the HEARTSTRING, ENCLOSE II and PAS-PORT systems are frequently used in Japan, a rigorous comparison focusing on the differences between the impact of Side-clamp and anastomosis assist device use in the ascending aorta has become possible. To the best of our knowledge, this is the largest clinical study to investigate the effect of anastomosis assist devices in OPCAB cases. A significant majority of the cases were included in the analysis (7348 of the 7374 Side-clamp cases and 7348 of the 14 213 Device cases after stratification), supporting the overall reliability of the results that were obtained. The overall background factors were well balanced, because the estimated rates of mortality and morbidity were similar between the 2 groups. This study sought to quantify the protective effect of anastomosis assist device use, as an emerging technique (second in efficacy to the aorta no-touch method) in combatting postoperative cerebrovascular events to improve postoperative complications; several favourable clinical outcomes have been published in the past [1, 2, 12–16, 19–26]. The benefit of the aorta no-touch technique is an accepted issue today; thus, to make the comparison as simple and straightforward as possible, we only compared the Side-clamp and Device groups. Our expectations were partially met in that the use of an anastomosis assist device provides a benefit in terms of minimizing the incidence of perioperative transient neurological complications with regard to the transient ischaemic attack/reversible ischaemic neurological deficit/delirium; on the other hand, the stroke rate was similar between the 2 groups. This result—with regard to the stroke rate—was compatible with the results of a network meta-analysis reported by Zhao et al [12]. In their report, the use of partial clamping and the use of a HEARTSTRING device were associated with a reduced rate of stroke in comparison with on-pump CABG; however, the comparison between the partial clamp and anastomosis assist device use revealed no significant difference in the stroke rate. There was no clear description regarding the degree of coronary artery disease in the target cohort, and this factor may not have been fully assessed elsewhere in the past. Our database did not record the frequency of the application of clamping manoeuvres in each case, which might have interfered with the outcome regarding the cerebrovascular event rate. Our results, which showed a higher incidence of new-onset AF and prolonged ventilation in the Device group, also need to be discussed. With regard to the AF, we matched the preoperative characteristics in detail, including ASA and beta-blocker use; thus, it can be stated that the background characteristics of the patient had minimal influence on this result. The notable findings were that the blood transfusion rate was higher and the operation time was longer in the Device group. Bleeding during the procedure might be a serious issue with the use of anastomosis assist devices regarding technical achievement in handling, and this issue has resulted in the higher requirement of homologous blood in the Device group. Although excessive bleeding did not seem to result in the higher requirement of re-exploration, it could be one factor that resulted in the higher incidence of AF in the Device group. In 2014, Alameddine et al. [27] reported that perioperative blood transfusion may be associated with excess AF following CABG and that this risk was associated with the amount of transfusion. Another report from the Cleveland Clinic demonstrated that perioperative red blood cell transfusion is associated with an increased risk of postoperative morbidity after isolated CABG [28]. They did not clearly comment on AF but they noted that red cell transfusion was a significant risk factor for prolonged ventilation (>24 h), and this was compatible with the results of our study. Limitations This study was associated with some limitations. Although this study was performed using a national database to provide clinical information on a very high-volume cohort, we could not always obtain data on very specific issues with regard to the use of specific anastomosis assist devices or how they were applied to the aorta, or the degree of aortic disease. Our database is periodically reviewed and updated with more appropriate questionnaires. The newest version of the database (2017∼) contains information on the status of the aorta, and a further update of this study may provide more relevant results to better address our clinical questions. Another limitation was the fact that this study primarily focused on the detailed short-term outcomes and did not investigate the long-term results because the data were extracted from clinical results accumulated from 590 cardiac surgical units located throughout Japan. Thus far, the submission of 100% of the data to the JCVSD under third-party surveillance has only been achieved over the short term; however, it is anticipated that this national database will include long-term follow-up data in the future and thereby allow for the analysis of the long-term outcomes. CONCLUSIONS In conclusion, when the perioperative characteristics, short-term surgical results and risk profiles for OPCAB, were analysed. The use of an anastomosis assist device provided better outcomes than Side-clamp with respect to minimizing postoperative transient neurological complications but at the same time increased the risk of blood transfusion, new-onset AF and prolonged ventilation, and the overall benefit of using an anastomosis assist device was not demonstrated. Further stratification of the patients, such as by the severity of coronary artery disease, may yield additional information on the patients in whom device use is expected to be beneficial. In the future, it might be also necessary to investigate the economic benefit of device use by balancing the cost of the anastomosis assist device with the extra cost of postoperative complications. Further investigation of the means to reduce the rate of postoperative complications following OPCAB will require the appropriate selection of relevant procedures based on multiple aspects. ACKNOWLEDGEMENTS The authors thank Shinichi Takamoto for his valuable contribution to the organization of the data set and this article. Conflict of interest: none declared. REFERENCES 1 Biancari F, Yli-Pyky S. Meta-analysis on the use of the Heartstring anastomotic device to prevent stroke in patients undergoing off-pump coronary artery bypass grafting. Eur J Cardiothorac Surg  2011; 40: 1236– 40. Google Scholar PubMed  2 Thourani VH, Razavi SA, Nguyen TC, Kilgo PD, Puskas JD, Guyton RA et al.   Incidence of postoperative stroke using the Heartstring device in 1,380 coronary artery bypass graft patients with mild to severe atherosclerosis of the ascending aorta. Ann Thorac Surg  2014; 97: 2066– 72. Google Scholar CrossRef Search ADS PubMed  3 Arrigoni SC, Mecozzi G, Grandjean JG, Hillege JL, Kappetein AP, Mariani MA. Off-pump no-touch technique: 3-year results compared with the SYNTAX trial. Interact CardioVasc Thorac Surg  2015; 20: 601– 4. Google Scholar CrossRef Search ADS PubMed  4 Motomura N, Miyata H, Tsukihara H, Okada M, Takamoto S; Japan Cardiovascular Surgery Database Organization. First report on 30-day and operative mortality in risk model of isolated coronary artery bypass grafting in Japan. Ann Thorac Surg  2008; 86: 1866– 72. Google Scholar CrossRef Search ADS PubMed  5 Saito A, Motomura N, Miyata H, Takamoto S, Kyo S, Ono M. Age-specific risk stratification in 13488 isolated coronary artery bypass grafting procedures. Interact CardioVasc Thorac Surg  2011; 12: 575– 80. Google Scholar CrossRef Search ADS PubMed  6 Yang D, Dalton JE. A unified approach to measuring the effect size between two groups using SAS®. SAS Global Forum;  2012, 1– 6. Paper 335. 7 Coca-Perraillon M. Local and global optimal propensity score matching. SAS Global Forum ; 2007, 1– 9. Paper 185. 8 Lamy A, Tong W, Devereaux PJ, Gao P, Gafni A, Singh K et al.   The cost implications of off-pump versus on-pump coronary artery bypass graft surgery at one year. Ann Thorac Surg  2014; 98: 1620– 5. Google Scholar CrossRef Search ADS PubMed  9 Trehan N, Mishra M, Sharma OP, Mishra A, Kasliwal RR. Further reduction in stroke after off-pump coronary artery bypass grafting: a 10-year experience. Ann Thorac Surg  2001; 72: S1026– 32. Google Scholar CrossRef Search ADS PubMed  10 Halbersma WB, Arrigoni SC, Mecozzi G, Grandjean JG, Kappetein AP, van der Palen J et al.   Four-year outcome of OPCAB no-touch with total arterial Y-graft: making the best treatment a daily practice. Ann Thorac Surg  2009; 88: 796– 801. Google Scholar CrossRef Search ADS PubMed  11 Davierwala PM, Leontyev S, Misfeld M, Rastan A, Holzhey D, Lehmann S et al.   No-touch aorta off-pump coronary bypass operation: arteriovenous composite grafts may be used as a last resort. Ann Thorac Surg  2013; 95: 846– 52. Google Scholar CrossRef Search ADS PubMed  12 Zhao DF, Edelman JJ, Seco M, Bannon PG, Wilson MK, Byrom MJ et al.   Coronary artery bypass grafting with and without manipulation of the ascending aorta: a network meta-analysis. J Am Coll Cardiol  2017; 69: 924– 36. Google Scholar CrossRef Search ADS PubMed  13 Dohmen G, Hatam N, Goetzenich A, Mahnken A, Autschbach R, Spillner J. PAS-Port(R) clampless proximal anastomotic device for coronary bypass surgery in porcelain aorta. Eur J Cardiothorac Surg  2011; 39: 49– 52. Google Scholar CrossRef Search ADS PubMed  14 Aranki SF, Shekar PS, Ehsan A, Byrne-Taft M, Couper GS. Evaluation of the enclose proximal anastomosis device in coronary artery bypass grafting. Ann Thorac Surg  2005; 80: 1091– 5. Google Scholar CrossRef Search ADS PubMed  15 Akpinar B, Guden M, Sagbas E, Sanisoglu I, Ergenoglu MU, Turkoglu C. Clinical experience with the Novare Enclose II manual proximal anastomotic device during off-pump coronary artery surgery. Eur J Cardiothorac Surg  2005; 27: 1070– 3. Google Scholar CrossRef Search ADS PubMed  16 Medalion B, Meirson D, Hauptman E, Sasson L, Schachner A. Initial experience with the Heartstring proximal anastomotic system. J Thorac Cardiovasc Surg  2004; 128: 273– 7. Google Scholar CrossRef Search ADS PubMed  17 Biancari F, Lahtinen J, Ojala R, Ahvenjarvi L, Jartti A, Mosorin M et al.   Spyder aortic connector system in off-pump coronary artery bypass surgery. Ann Thorac Surg  2007; 84: 254– 7. Google Scholar CrossRef Search ADS PubMed  18 Saito A, Miyata H, Motomura N, Ono M, Takamoto S; Japan Cardiovascular Surgery Database Organization. Propensity-matched analysis of bilateral internal mammary artery vs single internal mammary artery in 7702 cases of isolated coronary artery bypass grafting. Eur J Cardiothorac Surg  2013; 44: 711– 17. Google Scholar CrossRef Search ADS PubMed  19 Patel NC, Hemli JM. Anastomotic devices in coronary artery surgery: it is about the anastomosis? Multimed Man Cardiothorac Surg  2013; 2013: mmt019. Google Scholar CrossRef Search ADS PubMed  20 Kawasaki M, Fujii T, Hara M, Sasaki Y, Katayanagi T, Okuma S et al.   Morphological evaluation of proximal anastomosis by PAS-Port((R)) system in patients with long-term patent grafts. Ann Thorac Cardiovasc Surg  2015; 21: 172– 7. Google Scholar CrossRef Search ADS PubMed  21 Ayad RF, Bhella PS, Dockery WD, Schussler JM. Patency of vein graft anastomoses facilitated with the hexalon device. Ann Thorac Surg  2011; 91: 894– 8. Google Scholar CrossRef Search ADS PubMed  22 Wiklund L, Setina M, Tsang K, Cusimano R, Yau T. A multicenter prospective randomized trial of a second-generation anastomotic device in coronary artery bypass surgery. J Thorac Cardiovasc Surg  2010; 139: 741– 7. Google Scholar CrossRef Search ADS PubMed  23 Kikuchi K, Tambara K, Yamamoto T, Yamasaki M, Hirose H, Amano A. The use of Enclose((R))II anastomosis assist device for the proximal coronary branch anastomosis to vascular graft. Ann Vasc Dis  2010; 3: 84– 6. Google Scholar CrossRef Search ADS PubMed  24 Puskas JD, Halkos ME, Balkhy H, Caskey M, Connolly M, Crouch J et al.   Evaluation of the PAS-Port Proximal Anastomosis System in coronary artery bypass surgery (the EPIC trial). J Thorac Cardiovasc Surg  2009; 138: 125– 32. Google Scholar CrossRef Search ADS PubMed  25 Manabe S, Fukui T, Miyajima K, Watanabe Y, Matsuyama S, Shimokawa T et al.   Impact of proximal anastomosis procedures on stroke in off-pump coronary artery bypass grafting. J Card Surg  2009; 24: 644– 9. Google Scholar CrossRef Search ADS PubMed  26 Athanasiou T, Asrafian H, Krasopoulos G, Purkayastha S, Malinowski V, Al-Ruzzeh S et al.   Clampless arterial coronary artery bypass grafting with the use of magnetic coupling devices. J Card Surg  2006; 21: 69– 74. Google Scholar CrossRef Search ADS PubMed  27 Alameddine AK, Visintainer P, Alimov VK, Rousou JA. Blood transfusion and the risk of atrial fibrillation after cardiac surgery. J Card Surg  2014; 29: 593– 9. Google Scholar CrossRef Search ADS PubMed  28 Koch CG, Li L, Duncan AI, Mihaljevic T, Cosgrove DM, Loop FD et al.   Morbidity and mortality risk associated with red blood cell and blood-component transfusion in isolated coronary artery bypass grafting. Crit Care Med  2006; 34: 1608– 16. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

European Journal of Cardio-Thoracic SurgeryOxford University Press

Published: May 2, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off