Ten-year real-life effectiveness of coronary artery bypass using radial artery or great saphenous vein grafts in a single centre Chinese hospital

Transcription

1 Interactive CardioVascular and Thoracic Surgery 25 (2017) doi: /icvts/ivx174 Advance Access publication 16 June 2017 ORIGINAL ARTICLE Cite this article as: Zhu Y, Chen A, Wang Z, Liu J, Cai J, Zhou M et al. Ten-year real-life effectiveness of coronary artery bypass using radial artery or great saphenous vein grafts in a single centre Chinese hospital. Interact CardioVasc Thorac Surg 2017;25: Ten-year real-life effectiveness of coronary artery bypass using radial artery or great saphenous vein grafts in a single centre Chinese hospital Yunpeng Zhu,AnqingChen, Zhe Wang, Jun Liu, Junfeng Cai, Mi Zhou and Qiang Zhao* Department of Cardiac Surgery, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China * Corresponding author. Department of Cardiac Surgery, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai , China. Tel: ; fax: ; Zq11607@rjh.com.cn (Q. Zhao). Received 28 September 2016; received in revised form 20 February 2017; accepted 21 March 2017 Abstract OBJECTIVES: Long-term effectiveness of coronary artery bypass grafting using radial artery (RA) or great saphenous vein (SVG) grafts as a second conduit was compared. METHODS: Patients received simple elective off-pump coronary artery bypass involving both the left internal thoracic artery (LITA) and the left anterior descending artery between January 1999 and December 2005 at Ruijin Hospital, Shanghai Jiaotong University School of Medicine, China. RA graft patients (n = 147 LITA + RA and n = 61 LITA + RA + SVG) were matched with SVG graft patients (n = 208 LITA + SVG). Mean follow-up was 86.5 months. RESULTS: Baseline characteristics were comparable before and after surgery. Intraoperative hospital mortality was not significantly different. In all, 378 (90.9%) patients completed postoperative follow-up (197 in the RA and 181 in SVG). Overall survival was significantly better in the RA group (Log-rank, P = 0.017) with 88% 10-year survival in the RA group and 81% in the SVG group. All-cause mortality was significantly lower in the RA group (hazard ratio 0.42, 95% confidence interval , P = 0.020). Major adverse cardiovascular event-free survival was significantly better in the RA group than in the SVG group (Log-rank, P = 0.019). No significant difference in the length of postoperative angina relief was found. CONCLUSIONS: Using the RA as the secondary graft for coronary artery bypass grafting improved long-term postoperative survival and reduced the incidence of postoperative major adverse cardiovascular events. Keywords: Coronary artery bypass grafting Internal thoracic artery Radial artery Saphenous vein Patient outcome assessment INTRODUCTION Coronary artery bypass grafting (CABG) has long been established as an efficient and safe procedure for the treatment of coronary heart disease [1]. Using grafts from the left internal thoracic artery (LITA) to the left anterior descending artery (LAD) improves the effectiveness of CABG [2] and the LITA has been regarded as the gold standard for CABG [3]. In many cases, more than 1 artery is affected by atherosclerosis and the clinical choice has often been to use the great saphenous vein (SVG) as a second conduit [4]. In fact CABG with LITA to LAD and SVG to other target vessels has now become a standard method for the treatment of multi-vessel lesions in Chinese/East Asian patients, and its effectiveness is encouraging [5]. The success of LITA has also aroused interest in the use of other arteries and has promoted the advancement of full arterial CABG. Grafts from the radial artery (RA) have been widely These authors contributed equally to this work. investigated for their high long-term patency, being easy to harvest, high feasibility for various target vessels due to the length and lumen diameter, low incidence of severe complications [6], and the clinical results are satisfying [7, 8]. Studies suggest that using RA grafts could improve prognosis, especially by increasing the long-term survival and decreasing the long-term incidence of cardiovascular events. Thus, it is fairly well established that RA grafts are preferable to SVG grafts as a second conduit in CABG [9, 10]. However, a recent review concluded that the quality of evidence for the use of RA over SVG grafts was poor and it could only be weakly recommended [11]. Unfortunately, only a few medical centres in China perform CABG using RA grafts, and only very few related studies, which have mainly focused on the short-term effectiveness after operation, have been reported to encourage the use of RA grafts [12 14]. No randomized clinical trial or large sample sized casecontrol study has been performed in China to investigate whether using RA grafts to perform full arterial grafting ADULT CARDIAC VC The Author Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.

2 560 Y. Zhu et al. / Interactive CardioVascular and Thoracic Surgery (LITA + RA) or the standard method plus RA (LITA + RA + SVG) could further improve the effectiveness of CABG. Here we performed a retrospective, case-matched, observational study to compare the long-term effectiveness of using RA grafts for CABG with conventional SVG CABG in patients treated in our clinical centre. MATERIAL AND METHODS Patients Patients who received simple elective CABG between January 1999 and December 2005 in the clinical centre of Ruijin Hospital, Shanghai Jiaotong University School of Medicine who met the following criteria were included in this retrospective study: (i) aged between 18 and 80 years; (ii) indications for CABG [15] and (iii) underwent LITA-LAD CABG, and either RA or SVG grafts to other target vessels. Patients with 1 or more of the following features were excluded: (i) the CABG protocol did not include LITA-LAD; (ii) patients underwent other cardiac surgery at the same time; (iii) patients were treated with on-pump CABG or (iv) patients had emergency surgery. The study was approved by the ethics committee of the hospital and all the included patients signed informed consent. Study design From 1389 patients considered for inclusion, 230 patients had undergone CABG using grafts from the RA according to the best judgement of the surgeon. For each of these 230 patients, 1 matched control patient with a SVG graft was selected. The matching was performed so that the matched patients were the same in terms of age within 3 years, gender, Canadian Cardiovascular Society (CCS) classification of preoperative angina [16], New York Heart Association grading of heart function, and number of coronary arteries involved (±1). Finally, 416 patients were included in the study, 208 were in the RA group (including 147 received LITA + RA and 61 received LITA + RA + SVG) and 208 were in the SVG group (all received LITA + SVG). In the RA group, the RA was preferentially used for the revascularization of the target vessels with more severe stenosis in order to achieve a better long-term patency rate, but when the vessel length of RA graft was exhausted, then the SVG was used as a supplement. So, for patients in the LITA + RA + SVG group the RA was deployed to the artery with the tightest stenosis irrespective of its size or importance. The SVG target vessels had to have more than 50% stenosis (estimated and judged by surgeons). All cases had sequential grafting. Clinical data collection and examination Intraoperative data including patient demographics and clinical evaluations including type and number of bypass grafts, preoperative myocardial infarction (MI), history of percutaneous coronary interventions/cabg treatment were obtained from medical records. Enrolment forms used in a Chinese registry study of surgical treatment for coronary heart diseases were used for the data collection. Preoperative angina was classified according to criteria issued by CCS [16]. Symptoms of hyperlipidaemia included hypercholesterolaemia and hypertriglyceridaemia and the diagnostic criteria were total cholesterol >5.2 mmol/l or total triglycerides >1.7 mmol/l. If the level of low-density lipoprotein cholesterol was >3.64 mmol/l, this was also considered as positive result. Preoperative blood examinations were also routinely performed to evaluate hyperlipidaemia. History of cerebrovascular accident was diagnosed when the patient had been in a coma for 24 continuous hours or more, or had experienced central nervous system disorders (including limb weakness, conscious disturbance, aphasia and visual field loss) for 72 continuous hours or more. Simple transient ischaemic attack was not recorded as cerebrovascular accident. Peripheral vascular diseases indicated mainly arterial diseases including carotid artery lesions, RA lesions, femoral artery lesions and abdominal aortic lesions. Preoperative colour vascular ultrasound examinations were routinely performed to evaluate disorders of the bilateral carotid arteries and femoral arteries, and stenosis >70% was considered positive. The patients were followed-up by clinic visit, once every 6 months, and 378 patients completed postoperative follow-up (including 197 patients in the RA group and 181 patients in the SVG group; the overall follow-up rate was 90.9%). The mean follow-up time of the patients was 82.5 months (ranging from 62 to 139 months). Follow-up forms used in a Chinese registry study of surgical treatment for coronary heart diseases were used for the data collection. The outcome evaluated from follow-up was the occurrence of major adverse cardiovascular events (MACE). Controlled postoperative hypertension was defined as postoperative blood pressure <140/90 mmhg. Highest systolic blood pressure >_140/90 mmhg (self-reported by patients) or diagnosis of hypertension without receiving regular medication or monitoring were considered uncontrolled. Controlled postoperative diabetes was defined as fasting blood-glucose <7 mmol/l. Random postoperative fasting blood-glucose >_7 mmol/l (selfreported by patients) or patients diagnosed with diabetes who did not receive regular medication or monitoring were considered uncontrolled. Controlled postoperative hyperlipidaemia was defined as low-density lipoprotein <2.6 mmol/l and total triglycerides <1.7 mmol/l. Random postoperative low-density lipoprotein >_2.6 mmol/l or total triglycerides >_1.7 mmol/l (patients self-report) or patients diagnosed with hyperlipidaemia who did not receive regular medication or monitoring were considered uncontrolled. Regular postoperative physical exercise was defined as an accumulated exercise time >_3.5 h/week; body weight control was defined as postoperative body weightpreoperative body weight <_3 kg (using body weight measured in the past 6 months). Surgical processes The coronary anastomosis procedure was performed by the same surgeon in each patient. The graft vessel harvest was undertaken by another surgeon of the same surgical group. Surgeons performed flow measurements during surgery and completed vascular anastomosis according to unified standards. Each surgical procedure was undertaken by the same surgeon. In order to harvest the RA graft, the Allen test was routinely performed before the operation, and the non-preferred arm was selected. The non-touch isolation technique was used [17]. After administration of heparin, the proximal and distal end of the RA was ligated, and then RA graft was harvested. The length of the RA graft harvested did not extend beyond the radial recurrent artery and rasceta. The RA graft was wrapped with wet gauze with

3 Y. Zhu et al. / Interactive CardioVascular and Thoracic Surgery 561 Table 1: Characteristics papaverine until use. The RA graft was sequentially anastomosed to the left circumflex artery, lateral posterior branch or posterior descending branch of the right coronary artery. Continuous suturing of the RA graft to the coronary artery was performed using 7-0 prolene thread. The proximal end was then routinely anastomosed to LITA using T-Graft anastomosis. Local administration of papaverine was immediately provided by spraying it onto the RA graft in the operating room, to prevent the RA from spasm and then the patient was transferred from the operating room to intensive care unit, during which time intravenous injection of nitrate and diltiazem (1 ug/kg/min) was performed. Oral intake of nitrates for 1 year and diltiazem for 6 months after operation was also performed. Outcomes Preoperative patient clinical data RA group SVG group (n = 208) (n = 208) n (%) n (%) P-value Age (years) 55.9 ± ± Age >70 years 32 (15.4) 37 (17.8) Female 25 (12.0) 25 (12.0) 1.00 Angina (CCS) 3.1 ± ± CCS Grade 4 58 (27.9) 54 (26.0) NYHA grade 3.2 ± ± LVEF (%) 63.0 ± ± Previous MI 101 (48.6) 85 (40.9) Previous PCI 40 (19.2) 33 (15.9) COPD 14 (6.7%) 16 (7.7) Creatinine >2.5 mg/dl 6 (2.9) 5 (2.4) Cerebrovascular accident 23 (11.1) 21 (10.1) Hypertension 157 (75.5) 170 (81.7) Diabetes 71 (34.1) 78 (37.5) Hyperlipidaemia 114 (54.8) 125 (60.1) Peripheral vascular disorders 3 (1.4) 12 (5.7) Smoking 118 (56.7) 121 (58.1) Number of involved coronary 3.56 ± ± arteries Left main coronary artery lesions 55 (26.4) 47 (22.6) Triple vessel lesions 176 (84.6) 183 (88.0) Number of grafts 3.28 ± ± All patients received ASA 100 mg or clopidogrel 75 mg for antiplatelet treatment, so this was the same in both groups. NYHA: New York Heart Association; LVEF: left ventricular ejection fraction; COPD: chronic obstructive pulmonary disease; PCI: percutaneous coronary interventions; MI: myocardial infarction; CCS: Canadian Cardiovascular Society; RA: radial artery; SVG: great saphenous vein; ASA: acetylsalicylic acid. The primary outcomes of the study included overall survival, postoperative MACE-free rate and postoperative angina remission. All-cause mortality was used for the comparison of overall survival. MACE included cardiac death, postoperative MI and target vessel revascularization. Target vessel revascularization was defined as lesions of the grafted vessel or distal coronary artery that require secondary percutaneous coronary interventions or CABG of the target vessel; coronary artery lesions beyond the target vessels that required revascularization were not considered as MACE. Postoperative angina was diagnosed according to the following index: (i) induction factor: activity, satiation or excitement; (ii) position of pain: substernal, precordial, interscapular or subxiphoid; (iii) feature of pain: constriction, congestion, tightness, burning pain or dull. Postoperative pains with features similar to the features before operation were diagnosed as defined angina, while stinging pain, electric shock-like pain or pain caused by cough or inhalation could be excluded; (iv) pain duration: about several minutes to tens of minutes. Pains with durations of only several seconds or longer than several hours or days but with no changes in pain severity or pain could be excluded; (v) radiating pain: radiating pain with the radiation area covered neck, jaw, left shoulder and arm, and upper abdomen could be diagnosed as defined angina and (vi) remission: pain remission could be found after rest or oral intake of nitrates. Statistical analysis Categorical data were described as frequencies and percentages, and continuous data were described as means and standard divisions. SPSS 16.0 software (SPSS Inc., Chicago, IL, USA) was used for the statistical analysis. Chi-square test was used to compare the categorical data between the 2 groups as well as the data before and after operation, while t-test was used to compare the continuous data with normal distribution. Kaplan Meier survival curves and log-rank test were used to compare the overall survival, MACE-free time and time to angina remission. Cox proportional-hazards regression model was used to calculate hazard ratios of the results with significance. P < 0.05 was considered statistically significant. RESULTS Comparison of preoperative baseline data Table 1 shows the baseline data of the patients in the 2 groups before operation. No significant difference in age, sex, preoperative angina classification (CCS), New York Heart Association grade, number of involved coronary arteries, previous MI and percutaneous coronary interventions treatment was found between the 2 groups. Distribution of other factors that could influence the outcomes of the patients (including hypertension, diabetes, hyperlipidaemia and previous cerebrovascular accidents) were also not significantly different between the 2 groups. In contrast, more patients in the SVG group had a history of peripheral vascular disorders (P = 0.018). This is likely to be because the RA group underwent an Allen test before the procedure. It is well known that the presence of peripheral vascular disease adversely affects the outcomes of patients undergoing coronary revascularization. The mean number of grafts was 3.28 ± 0.88 and 3.32 ± 0.70 for the patients in the RA and SVG groups, respectively, and the difference was not significant (P > 0.05). Comparison of postoperative clinical data Table 2 shows postoperative data of the patients in the 2 groups. Postoperative cerebrovascular accidents, rates of hypertension, diabetes, hyperlipidaemia control, rate of postoperative smoking and postoperative weight control were not significantly different between the 2 groups. In contrast, more patients in the RA group were found to have undertaken regular physical activity than in the SVG group (86.8% vs 70.2%, P = 0.008). ADULT CARDIAC

4 562 Y. Zhu et al. / Interactive CardioVascular and Thoracic Surgery Table 2: Postoperative patient clinical data Characteristics RA group (n = 197) SVG group (n = 181) P-value Hypertension control 152 (77.3%) 138 (76.2%) Diabetes control 135 (68.7%) 118 (65.2%) Hyperlipidaemia control 133 (67.2%) 126 (69.6%) Postoperative smoking 29 (14.9%) 25 (13.8%) Physical exercise 161 (81.7%) 127 (70.2%) Body weight control 147 (74.6%) 144 (79.6%) RA: radial artery; SVG: great saphenous vein. Comparison of follow-up data Two and 4 patients in the RA and SVG groups, respectively, died during the intraoperative period (including hospital stay and within 30 days after operation), the difference was not statistically important (0.8% vs 1.7%, P > 0.05). During long-term follow-up, 8 and 17 patients from RA and SVG groups died, respectively, and the difference was statistically significant (4.1% vs 9.4%, P = 0.037). Figure 1 shows the Kaplan Meier overall survival curve of the 2 groups, overall survival was significantly better in the RA group than in the SVG group (log-rank test, P = 0.017). The 1-, 3-, 5-, 8- and 10-year accumulative survival was 99%, 98%, 97%, 94% and 88%, respectively, for the RA group, and 97%, 96%, 90%, 87% and 81%, respectively, for the SVG group. Cox proportional-hazards regression showed that the all-cause mortality was significantly lower in the RA group (hazard ratio 0.42, 95% confidence interval , P =0.020). The causes of death during follow-up are shown in Table 3. Figure 2 shows the Kaplan Meier MACE-free time curve of the 2 groups, MACE-free time was better in the RA group than in the SVG group (log-rank test, P = 0.019). The 1-, 3-, 5-, 8- and 10-year accumulative MACE-free time was 99%, 97%, 95%, 88% and 74%, respectively, in the RA group, and 94%, 90%, 88%, 78% and 68%, respectively, in the SVG group. Cox proportional-hazards regression showed that the MACE rate was significantly lower in the RA group (hazard ratio 0.50, 95% confidence interval , P = 0.022). Figure 3 shows the angina remission time curve in the 2 groups, and the difference was not significantly different (log-rank test, P = 0.078). The 1-, 3-, 5-, 8- and 10-year accumulative angina remission rate was 97%, 95%, 90%, 74% and 65%, respectively, in the RA group, and 93%, 87%, 86%, 62% and 57%, respectively, in the SVG group, and the differences were not statistically significant. DISCUSSION This study compared the long-term effectiveness of RA grafts with SVG grafts for CABG alongside LITA. We found that using the RA as the secondary graft for CABG improved long-term postoperative survival and reduced the incidence of postoperative MACE. This provides evidence that RA grafting has a role to play in the treatment of coronary artery disease in China. The improved long-term overall survival, lower all-cause mortality, improved MACE-free time in the RA group all suggest that, in agreement with many previous studies, RA grafts may be considered before SVG grafts alongside LITA. Two important advances, namely the non-touch technique and combined drug therapy for arterial spasm, have substantially Figure 1: Kaplan Meier overall survival curve of the 2 groups. Overall survival was significantly better in the radial artery (RA) group than in the great saphenous vein (SVG) group (log-rank test, P = 0.017). Table 3: Death cause Causes of death during follow-up RA (n =8) Myocardial infarction 2 6 Heart failure 1 2 Cerebral haemorrhage 0 1 Pericardial effusion 0 1 Renal failure 0 1 Tumour 2 2 Others 1 1 Unknown 2 1 RA: radial artery; SVG: great saphenous vein. SVG (n = 15) improved RA use [17]. Early arterial spasm results from mechanical stimulation induced by surgical processes, exogenous active substance that mediates vasoconstriction, and local release of nerve vascular constricting factors, neurotransmitters and hormones. Gentle surgical processes substantially reduce the risk of RA spasm. Most anti-arterial spasm drugs are phosphodiesterase receptor antagonists, including papaverine, calcium channel blockers, nitrates and phenoxybenzamine (a potential a-adrenergic receptor antagonist) [18 21]. Here, all RA grafts were harvested using non-touch technique. Papaverine, nitrates and calcium channel blockers were used in combination to inhibit arterial spasm, and diltiazem was orally administrated until 6 months after the operation. There are cases where VSG grafts might be preferred, in cases such as in the elderly, whose arterial grafts might often be diseased [22]. In this study, only patients <65 years, in whom the collateral circulation of ulnar artery was good (confirmed by Allens test), and stenosis of the target vessel was >70%, were selected. As age could substantially affect the long-term survival of patients, the patients in the present study were 1:1 matched by age and several other factors including preoperative angina grade (CCS), New York Heart

5 Y. Zhu et al. / Interactive CardioVascular and Thoracic Surgery 563 Figure 2: Kaplan Meier major adverse cardiovascular events (MACE)-free time curve of the 2 groups. MACE-free time was better in the radial artery (RA) group than in the great saphenous vein (SVG) group (log-rank test, P = 0.019). Figure 3: Kaplan Meier angina remission time curve of the 2 groups. There was no significant difference in angina remission time in the radial artery (RA) and great saphenous vein (SVG) groups (log-rank test, P = 0.078). Association grade, and number of involved coronary arteries. Risk factors of coronary heart diseases including hypertension, diabetes and hyperlipidaemia were comparable in the 2 groups. Peripheral vascular disorders are also a risk factor for RA occlusion [23]. Here, an Allen test found that fewer patients in the RA group had peripheral vascular disorders at baseline. Patients with insufficient vascularized vessels had significantly higher operative mortality, recurrence rate of angina, and 5-year MI incidence [24]. No significant difference in the bypass number was found between the RA and SVG groups (3.28 ± 0.88 vs 3.32 ± 0.70). The treatment strategies for the patients were decided according to preoperative coronary angiogram results to ensure revascularization could be completed in 1 stage. For the RA group with multi-vessel lesions, if the length of RA graft was not enough to complete full revascularization, grafts from SVG were used. As hybrid technology advances, arterial bypassing and stents could be better combined to resolve the limitations of insufficient RA graft length. The postoperative baseline clinical data were similar between the 2 groups except for postoperative physical exercises. We supposed that the improvement of the motion function in the RA group could result in better return to work and society. The EAST study [25] found that 47% of the patients that had been treated with CABG could engage in moderate or intense activities 3 years after operation, 38.5% of patients older than 61 years returned to the work and 79% of patients less than 61 years returned to work 1 year after operation. Control of hypertension, diabetes, hyperlipidaemia and body weight, as well as postoperative smoking could also affect the progression of coronary arterial diseases or disorders of the bypass grafts could also affect the postoperative clinical effectiveness. In this study, these factors were comparable between the RA and SVG groups. Perioperative mortality was similar between the 2 groups, suggesting that CABG using RA grafts is a safe procedure. However, the long-term mortality was significantly lower in the RA group than in the SVG group, suggesting that using RA grafts could effectively improve the long-term survival of the patients. An early-stage survival database of patients that received CABG using RA graft has already been established. In a perspective study [26], 3-year overall survival of the patients was 97.2% and 3-year MACE-free rate was 89.6% (MACE was defined differently to the present study). Very few studies have reported postoperative angina remission rate, but Acar et al. [8] found the 5-year angina remission rate was 88.7%, which is in accordance with this study. Several retrospective studies with larger sample sizes have validated the encouraging long-term clinical outcomes of CABG using RA grafts. A 14-year follow-up data from 4271 patients (2711 patients received LITA + RA + SVG and 1560 patients received LITA + SVG treatment) was carried out. The 1-, 5- and 10-year accumulative survival was 98.3%, 93.9% and 83.1% in the RA group, and 97.2%, 88.7% and 74.3% in the SVG group, respectively. This demonstrated better long-term outcomes in the RA group [10]. After the patients were stratified by several factors including age, sex and diabetes, the results further demonstrated the better long-term outcomes in the RA group in each subgroup. The long-term outcomes between patients that received LITA + RA (full arterial bypassing) and conventional LITA + SVG strategies were compared in a casecontrol study [9]. After 4743 patients (612 in the RA group) were carefully matched with age and other demographic features, the 6-year survival was significantly better in the RA group; the significance was more pronounced in the diabetes-, age- and female-subgroups. Recently, the 12-year outcomes of these patients were reported [27], which further demonstrated the significantly better long-term survival in the RA group; interestingly, the significance was only found in the 3-vessel lesion subgroup but not in the 2-vessel lesion subgroup. The grouping strategy of the present study was slightly different from the studies above. However, the 139-month (11.5 years) follow-up data also showed that using RA graft could significantly improve the long-term survival after CABG treatment, and the improvement tended to be more pronounced with time. The findings of this study were in accordance with the other retrospective studies. ADULT CARDIAC

6 564 Y. Zhu et al. / Interactive CardioVascular and Thoracic Surgery Preliminary results of several ongoing randomized perspective clinical studies including RSVP [28], RAPCO [29] and RAPS [30] showed that CABG with RA grafts could result in satisfactory early- and mid-term clinical outcomes and patency. We believe that their further results could validate the long-term outcomes of CABG using RA graft as the secondary resource of bypass graft. Limitations The major limitation of this study is the retrospective design. Although the matching strategy resulted in good comparability in baseline characteristics between the 2 groups, the significant difference between the groups in terms of peripheral vascular disease may have biased the results. In addition, the selection of the graft used was made by the surgeon performing the procedure without specific guidelines, therefore it is likely that there is a selection bias in the results. Compared with well-designed randomized clinical trials, the design of the present study is an unavoidable limitation. The sample size of the present study was also limited. In addition, obtaining imaging data during the follow-up period is still a challenge. Many other studies have used coronary angiogram or computed tomography scanning for patients with recurrent angina. We lost 9% of patients at follow-up, this means that the long-term outcomes of the study may be biased. CONCLUSION This study provides more evidence for beneficial outcomes of patients treated with LITA and RA grafts compared to the more standard LITA and SVG treatment. The RA group showed improved long-term postoperative survival and a reduced incidence of postoperative MACE. Conflict of interest: none declared. REFERENCES [1] Barner HB. Operative treatment of coronary atherosclerosis. Ann Thorac Surg 2008;85: [2] Loop FD, Lytle BW, Cosgrove DM, Stewart RW, Goormastic M, Williams GW et al. Influence of the internal-mammary-artery graft on 10-year survival and other cardiac events. N Engl J Med 1986;314:1 6. [3] Taggart DP, D Amico R, Altman DG. Effect of arterial revascularisation on survival: a systematic review of studies comparing bilateral and single internal mammary arteries. Lancet 2001;358: [4] Legare JF, Hassan A, Buth KJ, Sullivan JA. The effect of total arterial grafting on medium-term outcomes following coronary artery bypass grafting. J Cardiothorac Surg 2007;2:44. [5] Li J, Liu Y, Zheng J, Bai T, Liu Y, Wang X et al. The patency of sequential and individual vein coronary bypass grafts: a systematic review. Ann Thorac Surg 2011;92: [6] Al-Sabti HA, Al Kindi A, Al-Rasadi K, Banerjee Y, Al-Hashmi K, Al-Hinai A. Saphenous vein graft vs. radial artery graft searching for the best second coronary artery bypass graft. J Saudi Heart Assoc 2013;25: [7] Hayward PA, Hare DL, Gordon I, Buxton BF. Effect of radial artery or saphenous vein conduit for the second graft on 6-year clinical outcome after coronary artery bypass grafting. Results of a randomised trial. Eur J Cardiothorac Surg 2008;34: [8] Acar C, Ramsheyi A, Pagny JY, Jebara V, Barrier P, Fabiani JN et al. The radial artery for coronary artery bypass grafting: clinical and angiographic results at five years. J Thorac Cardiovasc Surg 1998;116: [9] Zacharias A, Habib RH, Schwann TA, Riordan CJ, Durham SJ, Shah A. Improved survival with radial artery versus vein conduits in coronary bypass surgery with left internal thoracic artery to left anterior descending artery grafting. Circulation 2004;109: [10] Tranbaugh RF, Dimitrova KR, Friedmann P, Geller CM, Harris LJ, Stelzer P et al. Radial artery conduits improve long-term survival after coronary artery bypass grafting. Ann Thorac Surg 2010;90: [11] Zhang H, Wang ZW, Wu HB, Hu XP, Zhou Z, Xu P. Radial artery graft vs. saphenous vein graft for coronary artery bypass surgery: which conduit offers better efficacy? Herz 2014;39: [12] Cheng Z, Zhao Z, Quan X, Wang S, Zhao J, Sun J et al. Twelve years experience and clinical results of using the radial artery for coronary revascularization. Chin Med J (Engl) 2014;127: [13] Tsai FC, Yeh TF, Jing Lin P. Use of graft flow measurement and computerized tomography angiography to evaluate patency of endoscopically harvested radial artery as sequential graft in coronary artery bypass surgery. J Cardiovasc Surg (Torino) 2014;55: [14] Chen X, Chen X, Xie D, Meng F, Shi K, Xu M. Radial artery as conduit is safe and effective in coronary bypass surgery in the elderly: singlecentral results from 326 patients. Circ J 2009;73: [15] Hillis LD, Smith PK, Anderson JL, Bittl JA, Bridges CR, Byrne JG et al ACCF/AHA Guideline for Coronary Artery Bypass Graft Surgery. A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Developed in collaboration with the American Association for Thoracic Surgery, Society of Cardiovascular Anesthesiologists, and Society of Thoracic Surgeons. J Am Coll Cardiol 2011;58:e [16] Campeau L. Letter: Grading of angina pectoris. Circulation 1976;54: [17] Verma S, Szmitko PE, Weisel RD, Bonneau D, Latter D, Errett L et al. Should radial arteries be used routinely for coronary artery bypass grafting? Circulation 2004;110:e40 6. [18] Wei W, Floten HS, He GW. Interaction between vasodilators and vasopressin in internal mammary artery and clinical significance. Ann Thorac Surg 2002;73: [19] Wei W, Chen ZW, Yang Q, Jin H, Furnary A, Yao XQ et al. Vasorelaxation induced by vascular endothelial growth factor in the human internal mammary artery and radial artery. Vascul Pharmacol 2007;46: [20] He GW, Fan L, Furnary A, Yang Q. A new antispastic solution for arterial grafting: nicardipine and nitroglycerin cocktail in preparation of internal thoracic and radial arteries for coronary surgery. J Thorac Cardiovasc Surg 2008;136:673 80, 680.e1 2. [21] Gaudino M, Glieca F, Luciani N, Alessandrini F, Possati G. Clinical and angiographic effects of chronic calcium channel blocker therapy continued beyond first postoperative year in patients with radial artery grafts: results of a prospective randomized investigation. Circulation 2001;104:I64 7. [22] Song SW, Sul SY, Lee HJ, Yoo KJ. Comparison of the radial artery and saphenous vein as composite grafts in off-pump coronary artery bypass grafting in elderly patients: a randomized controlled trial. Korean Circ J 2012;42: [23] Desai ND, Naylor CD, Kiss A, Cohen EA, Feder-Elituv R, Miwa S et al. Impact of patient and target-vessel characteristics on arterial and venous bypass graft patency: insight from a randomized trial. Circulation 2007;115: [24] King SB 3rd, Kosinski AS, Guyton RA, Lembo NJ, Weintraub WS. Eightyear mortality in the Emory Angioplasty versus Surgery Trial (EAST). J Am Coll Cardiol 2000;35: [25] Weintraub WS, Mauldin PD, Becker E, Kosinski AS, King SB 3rd. A comparison of the costs of and quality of life after coronary angioplasty or coronary surgery for multivessel coronary artery disease. Results from the Emory Angioplasty Versus Surgery Trial (EAST). Circulation 1995;92: [26] Amano A, Hirose H, Takahashi A, Nagano N. Coronary artery bypass grafting using the radial artery: midterm results in a Japanese institute. Ann Thorac Surg 2001;72: [27] Zacharias A, Schwann TA, Riordan CJ, Durham SJ, Shah AS, Engoren M et al. Late outcomes after radial artery versus saphenous vein grafting during reoperative coronary artery bypass surgery. J Thorac Cardiovasc Surg 2010;139: e4. [28] Collins P, Webb CM, Chong CF, Moat NE. Radial artery versus saphenous vein patency randomized trial: five-year angiographic follow-up. Circulation 2008;117: [29] Hayward PA, Buxton BF. Mid-term results of the radial artery patency and clinical outcomes randomized trial. Ann Cardiothorac Surg 2013;2: [30] Deb S, Cohen EA, Singh SK, Une D, Laupacis A, Fremes SE. Radial artery and saphenous vein patency more than 5 years after coronary artery bypass surgery: results from RAPS (Radial Artery Patency Study). J Am Coll Cardiol 2012;60:28 35.