No-touch Harvesting in Off-pump Coronary Artery Bypass Grafting With Sequential Saphenous Vein Graft: A Single-center Prospective Randomized Controlled Trial

Background: Coronary artery bypass grafting (CABG) materials comprise the combination of the left internal mammary artery and saphenous vein. The patency rate of vein grafts is considered not ideal; more studies support that the no-touch (NT) procedure can improve the patency rate of vein grafts. However, it is not clear that the NT technology is used in the sequential saphenous vein grafting during off-pump coronary artery bypass surgery. This study explored whether the NT technique is safety and ecacy compared to the conventional manner in the off-pump coronary artery bypass surgery using sequential vein graft. Methods: This was a prospective single-center randomized controlled clinical trial. A total of 200 patients undergoing off-pump coronary artery bypass grafting in the sequential saphenous graft were randomly assigned to two groups: The no-touch(NT) and the conventional (CON) groups. Perioperative and postoperative data were collected prospectively during the hospital stay. The occlusion of sequential grafts was measured by cardiac computed tomography angiography (CCTA) 3-months post CABG. Leg wound complications were followed up. The mean diameter of sequential grafts for the rst 100 patients was measured using CCTA, 3-months after the operation. Results: The primary endpoint was that there was no difference in occlusion of sequential venous grafts between the two groups (NT: 9/180 (4.4%), CON: 5/194 (1.5%), p=0.22). Similarly, there were no differences in composite clinical events (NT: 2/91 (2.2%), CON: 1/96 (1.0%), p=0.96). There was no difference in the leg wound complications between the two groups (NT: 8/91 (8.8%), CON: 4/96 (4.2%), p=0.20). However, there was a signicant difference in the average diameter of sequential grafts between the two groups (NT: (2.98±0.42), CON: (3.26±0.51), p=0.005). Conclusions: The early clinical results suggest that the NT technique is safety and ecacy compared to the conventional technique in sequential grafting in off-pump coronary artery bypass surgery. The sequential grafts early expansion in the NT technique is not as pronounced as the conventional technique, which may have a long-term protective effect on the grafts.


Introduction
Coronary artery disease remains one of the major serious diseases that threaten human health. Coronary artery bypass surgery is an effective way to handle the multi-vessel disease, complex lesions, and severe left main coronary heart disease [1][2][3]. The graft materials are mainly the combination of the left internal mammary artery and saphenous vein (SV) [4]. Early graft failure after CABG was present in up to 12% of grafts [5]. The internal mammary artery is the gold graft [6]. However, the low patency rate of the saphenous vein remains a big challenge [7]. One month after the operation, the main factor affecting the venous patency rate was the formation of acute thrombus, whereas hyperplasia of venous intima was the main factor 1-12 months after the surgery. The long-term patency rate after the operation was closely linked to venous atherosclerosis [8]. These factors signi cantly affected the early and late patency rate of the saphenous vein and further affected the safety and life quality of the patients [9]. Therefore, it is essential to determine how to improve the patency rate of the venous graft.
In 1996, the Swedish expert Souza's team initiated the NT technology, which retained part of the tissue surrounding the vein; the vein did not dilate manually after harvesting it. This reduced the damage to the vein's intimal, decreasing vascular smooth muscle cell activation, and protected against distensioninduced damage [10,11]. It also preserved intact vasa vasorum; this might present an excellent protective effect [12]. The clinical trials yielded encouraging results; the NT technique was safe and feasible. After 16 years of follow-up, the patency rate of venous grafts in the NT group was found to be comparable to that of the left internal mammary artery [13][14][15].
The NT technology is recommended as class IIa in the 2018ESC/EACTS guide [16]. Previous studies on NT were mostly performed under cardiopulmonary bypass, and most of the vein grafts were individual coronary artery bypass grafts. However, there were few studies on off-pump bypass surgery with sequential venous grafts. Increasing clinical studies show that sequential graft is as safe and effective as a single graft [17][18][19]. It was not clear NT technology was used in the sequential saphenous vein grafting during off-pump coronary artery bypass surgery. The actual effect of NT technology in off-pump bypass surgery with sequential venous grafts needs to be further veri ed through further clinical trials. This study aimed to test the reliability of the NT technology in off-pump CABG in sequential venous grafts. It compared and analyzed the clinical data of the two groups pre and post-operation, as well as the incidence of graft occlusion and clinical events three months after the operation. Our hypothesis is that NT technique is safety and e cacy compared to the conventional technique in the sequential grafts during off-pump coronary artery bypass surgery.

Study design
This was a single-center, randomized controlled study comparing the difference between NT and the traditional vein harvested in the off-pump bypass surgery with sequential venous graft. Two hundred patients with at least three branches of coronary artery diseases, needing the left internal mammary artery and a sequential vein graft only were recruited. The enrollment of participants was determined by selecting random envelopes. The ow-chart and study design schedule are presented in Fig. 1.

Participants
This study recruited 200 patients with coronary heart disease who were to undergo off-pump coronary artery bypass grafting in cardiac surgery center China, from December 2018 to April 2020.
The inclusion criteria were as follows: (a)Aged 18-80 years; (b) at least three-vessel coronary artery disease; (c) voluntarily joined the study and signed the informed consent. The exclusion criteria were: (a) Simultaneous operations (such as heart valve or lung or abdominal surgery); (b) emergent surgery; (c) ejection fraction ≤ 35%; (d) complicated with the interventricular septal perforation and ventricular aneurysm; (e) redo CABG; (f) internal diameter of great saphenous vein ≤ 0.20 cm, or varicose great saphenous vein, or venous tortuosity; (g) complicated with the severe malignant tumor or other serious systemic diseases; (h) severe renal insu ciency(creatinine > 200 µmol/L); (i) dual antiplatelet taboo (j) severe peripheral vascular disease; (k) allergic to the radiocontrast agent; (l)participating in other clinical trials at the same time.

Randomization
Participants will be randomly assigned (on a 1:1 ratio) to the NT and the CON groups with a random permuted block length of 4 patients per block, to ensure that trial groups at each block were balanced.
Randomization was based on the computer-generated random digits table. All the random numbers were put in a sealed opaque envelope, sealed with a stapler, and locked in the le cabinet. Patients who satis ed all inclusion criteria and did not meet any of the exclusion criteria would open the cabinet to take out the envelope and draw lots. The speci c scienti c research secretary kept and facilitated the random drawing of lots. The study patients were blinded.

Surgical Techniques
All patients were examined using bilateral great saphenous vein ultrasonography and marked before the operation. The patients received off-pump coronary artery bypass surgery with only one sequential venous graft. If hemodynamic instability or ventricular brillation occurred during off-pump bypass surgery, and the drug couldn't maintain stability, cardiopulmonary bypass would be established and surgery perform on-pump. Then the patient was excluded from the study. Leg wounds were sutured in three consecutive layers, and the leg wound was pressurized with an elastic bandage. The bandage was removed on the second-day post-operation and elastic socks worn for three months. If the trunk of the vein was very slender, or varicose or tortuous, it would be converted using the traditional technique, and the patient was excluded from this study.

Nt Group
The leg incision was cut longitudinally along the ultrasound mapping line made before the operation, and the trunk of the vein was exposed. When the trunk of the vein was dissociated, about 2 mm of the surrounding tissue was retained on both the left and right sides. The vein was not dilated after harvesting. It was marked with a signal pen to make preparations for distinguishing the course and direction of the anastomosed vein. After removal, the vein was stored in a mixture containing heparinized saline and papaverine hydrochloride. The vein was anastomosed proximally with the ascending aorta using a 6 − 0 polypropylene. The vein would be fully pre-dilated by aortic pressure and then examined for leakage. The sequential vein graft anastomoses were constructed end-to-side in the left and right crown systems using 7 − 0 or 8 − 0 polypropylene. The part of the vein graft with a longer distance between the two anastomoses was sutured to the heart's surface to prevent the vein from being twisted and kinked.

Control Group
The leg skin was cut longitudinally along the preoperative ultrasound marking line to expose the trunk and separate the visible branches. When the trunk of the vein dissociated, the surrounding tissue was not retained. The ankle vein was xed using a small adaptor that connected to a 10 ml syringe. The vein was dilated using a syringe lled with heparinized saline, checked for leakage, and placed in a mixture of heparinized saline and papaverine. Other operational processes were similar to the NT group.

Medicine
All patients were given anticoagulant therapy with low molecular weight heparin after admission, and the oral anticoagulants discontinued. Patients with underlying diseases (such as hypertension, diabetes, and hyperlipidemia) continued to take medicines until the day before the operation. Aspirin and statins were resumed 24 hours after the operation.

Follow Up
All patients underwent cardiac computed tomography angiography (CCTA) three months after the operation. The clinical events, recovery of the leg, and sternal wound complications were followed up.

Primary Outcome Measure
The primary clinical end event was the occlusion rate of venous grafts three months after the operation. This was detected using cardiac computed tomography angiography (CCTA). Evaluation of graft failure: The number of failures was calculated by distal anastomosis. The graft and anastomotic failure were evaluated according to the FitzGibbon classi cation system [20]. FitzGibbon-A refered to a wide range of unobstructed grafts or less than 50% narrow grafts; FitzGibbon-B was a limited ow graft with a narrowing higher than 50%. FitzGibbon-O refered to an occlusive graft without blood ow. In this study, FitzGibbon-A/B was used as patency and FitzGibbon-O as graft failure. The diseased graft was also regarded as a lesion if the lesion was located at the proximal /distal anastomosis site or the graft trunk.
In this study, all patients in the two groups accepted sequential grafts, which had never been before. In order to know the effectiveness of the sequential grafts more early, CCTA detection was carried out in advance three months after operation. This was different from the primary outcome measure initially planned for one year after the surgery.
2. Secondary Outcome Measure 1. The composite clinical events: Major adverse cardiac and cerebrovascular events (MACCEs), including death from any cause, myocardial infarction, stroke, and repeat revascularization.

The leg wound complications.
Leg complications were de ned as the wound not closing after 1 months, festering wound, or necrotic tissues present in the incision.

Postoperative mortality
The respiratory complications, myocardial infarction, stroke, acute renal failure, and atrial brillation in the post-operation were analyzed.

The average diameter of sequential grafts at three months
The saphenous vein expansion might have an effect on the long-term patency of the sequential grafts.
Therefore, we used CCTA to compare the average diameter of the sequential graft of the two groups at 3 months after the surgery. There was no such plan at the beginning of the study in the protocol. The cardiac computed tomography angiography (CCTA) was used to measure the diameter of the sequential grafts in the proximal, middle, and distal segments, and then the average diameter of the grafts was calculated. The early expansion of sequential grafts was achieved by measuring the rst 100 patients' mean diameter of sequential grafts, 3months after the operation.

Sample size calculation
Using the difference test of the comparison of the two groups of rates, α = 0.05, the degree of control is 90%. The results of meta-analytic evidence indicate that the one-year patency rate of vein grafts with the traditional method is 85%, and that of the no-touch way is expected to 95%. According to the formula, the sample size of each group is 188 cases, and then increased by 15% in case of loss of follow-up, and the sample size of each group is 217 cases. Every patient has a sequential venous graft, with at least two anastomoses. According to the anastomotic ratio of 1 to 2, and one anastomotic occlusion is an occurrence of the event, so the number of cases is about 200. SPSS 22.0 for Mac (IBM SPSS Statistics) was used for statistical analyses. Continuous variables were reported as the mean +/− standard deviation or median (interquartile range) (IQR). Categorical variables were reported as the absolute frequency and as a percentage. The Student t-test was applied for continuous data with equal or unequal variances. The Mann-Whitney U test was applied for continuous data that were not normally distributed. Pearson's X 2 and Fisher's exact tests were used for categorical data. Statistical signi cance was accepted at p<0.05.

Patients
A total of 673 patients were recruited from December 2018 to April 2020 in Beijing Anzhen Hospital China. Finally, 200 patients were enrolled in the trial. A total of 100 patients were randomly assigned to the NT group and 100 patients to the CON group. A total of 187 patients (91 patients in the NT group and 96 patients in the CON group) received a follow-up survey. Thirteen patients (NT: 9, CON: 4) did not undergo the surgical technique assigned for non-adherence to the study design requirements. The ow of patients through the trial up to three months of follow-up is shown (Fig. 1). The groups were well matched in age, sex, body-mass index, smoking, left ventricular ejection fraction, previous medical history, and so on (Table 1). Intraoperative and postoperative data are shown in Table 2 and Table 3.

Treatment
Six patients in the NT group were converted to cardiopulmonary bypass because of hemodynamic instability. Two patients did not perform sequential venous graft owing to the absence of a suitable anastomosis site during the operation. A no-touch vein could not be harvest in one patient due to a slender vein. Therefore, these nine patients were excluded from the NT group. Likewise, three patients in the CON group were converted to cardiopulmonary bypass because of hemodynamic instability, and the sequential venous graft was not performed in one patient. These also were excluded. There were 85 (93.4%) internal mammary arteries used in the NT group and 93(96.9%) in the CON group.

Primary outcomes
The total number of follow-up sequential vein anastomoses in the two groups was: NT (180/205(87.8%)),and CON (194/221(87.8%)) (Table4). The occlusion of sequential graft anastomoses at three months was not signi cantly different between the NT and the CON group: (NT: 9/180(5.0%), CON: 5/194(2.6%), p = 0.22) (Fig. 2). In the NT group, seven sequential grafts occluded without blood ow from the anastomosis of the obtuse marginal branch to the posterior descending branch. One sequential graft occluded without blood ow starting from the beginning of the vein graft. In the CON group, one sequential graft occluded from the anastomosis of the obtuse marginal branch to the posterior descending branch. Two sequential grafts occluded from anastomosis of the diagonal and posterior left ventricular branch to the posterior descending branch. One sequential graft occluded without blood ow from the beginning of the vein graft.

Secondary Outcomes
There was no difference in the composite endpoint between the two groups: NT (2/91(2.2%)) and CON (1/96(1.0%)) (p = 0.96). However, one patient in the NT group died fteen days after discharge due to respiratory failure; the patient was 76 years old, with a history of asthma. One patient in the NT group underwent PCI (percutaneous coronary intervention) before follow-up because of occlusion of the right graft and frequent angina pectoris. Three stents were implanted. In the CON group, one patient with cerebral infarction was hospitalized one month after discharge.

Postoperative Mortality
There was a higher blood transfusion in group B than in group A: NT (10/91 (11.0%)), and CON (19/96(19.8%)) (p = 0.10) ( Table 4), but it was not statistically signi cant. There was no signi cant difference in postoperative bleeding (NT: 400 (250-600), CON: 400 (250,700), p = 0.60), and hemoglobin at the postoperative rst day (NT: 103.58 ± 16.71, CON: 102.71 ± 17.00, p = 0.73) between the two groups. There was no difference in the duration of surgery, intubation time, intensive care unit (ICU) stay time, and hospital stay between the two groups. There was also no difference observed in respiratory complications, myocardial infarction, stroke, acute renal failure, and atrial brillation in post-operation.

The Average Diameter Of Sequential Grafts
Among the rst 100 patients, 2 in the NT group and 1 in the CON group failed the follow-up CTA examination. The diameters of sequential grafts measuring 48 and 49 were determined in the NT group and the CON group, respectively. The diameters of the proximal, middle, and distal segments of the sequential grafts were measured respectively. Since 4 cases of distal segment were occluded in the NT group, 44 cases of the distal diameter were measured. The CON group had a similar situation, so 48 cases were measured in the distal segment in the end. The detailed results were shown in Table 5.

Discussion
This study aimed to explore whether the NT technique in patients undergoing off-pump coronary artery bypass grafting with the sequential saphenous vein grafting was safe and effective. To clarify the effect of NT technique in sequential grafts, the clinical data of the two groups were compared and analyzed during and post-operatively, as well as the incidence of graft occlusion and clinical events 3-months postoperation. The results showed that the NT technique is safety and e cacy compared to the conventional vein grafting technique in the sequential grafts during off-pump coronary artery bypass surgery.
Studies in related elds demonstrated that the NT technology is safe and feasible, however, the leg wound complications did not present the same as in the NT group. Likewise, our study suggested that the leg wound complications in the NT group were slightly more than those in the CON group. In previous studies, a drainage tube was routinely placed after an operation on the leg wound [21]. However, no such drainage tube was used in our study, which may explain the high number of wound complications in the NT group.
In this study, there were more incidences of venous graft occlusion in the NT group; although the difference, between the two groups, was not signi cant (Fig. 2), attention should be paid to it. Studies show that there is no signi cant difference in the patency rate of the venous grafts between the no-touch and the conventional vein grafts at the early stages, and there was a trend of lower occlusion rate in the no-touch group. However, the results of this study were slightly different probably due to the following factor: Most of the occlusion of the grafts occurred near the posterior descending branch in the NT group whereas this study used a sequential graft, which could reach the posterior descending branch from the farthest point of the aorta. The graft was relatively long, and the graft was not manually dilated. The length of the graft may have been too long and the distal posterior descending branch couldn't be fully expanded by the aortic blood pressure. Moreover, the vein of the Asian population differs from that of the Swedish; the diameter of the vein is smaller, and it was bound by perivascular tissue, which made the distal end of the vein grafting di cult to expand.
Most of the previous studies on NT technology were retrospective, whereas prospective and randomized controlled studies were mainly from the Swedish Souza team. Most studies were performed under cardiopulmonary bypass, or the saphenous vein was individually graft. This study performed NT technology all in off-pump CABG with the sequential grafts. In on-pump surgery, the empty and accid heart hampered the correct estimation of graft con gurations in a full-beating heart. The use of off-pump CABG could prevent these challenges; off-pump procedures appeared to be associated with a reduced risk of early morbidities, such as a stroke, and fewer transfusion [22][23][24]. There was only one saphenous vein graft in the present study, and they were all sequential grafts, which per graft had at least two anastomoses. There was no difference between the sequential graft and single graft in MACCEs events in multiple previous studies. One advantage of the sequential graft is using a minimal length of SVG. There is also less tendency for the graft to kink because of the perivascular tissue in the NT group. However, failure of a single sequential conduit jeopardizes all of the anastomoses along that graft segment.
As in the Souza trial, ultrasound was used to locate the saphenous vein in all patients before operation.
The size of the great saphenous vein in Asian people is different from that in European people such as the Swedish; it is often thin in the Asian population. The pre-experiment experience showed that when the ultrasound reported the vein as thin, it would be challenging to succeed in harvesting a no-touch vein because of its neness, even breaking the trunk of the vein. Preoperative ultrasound can reduce unnecessary incisions and save operation time.
One of the characteristics of the NT technology was that the vein was not manually expanded after harvesting. The expansion of the vein graft depended entirely on the blood pressure of the aorta. In this case, the degree of postoperative expansion of the graft obtained using NT may be weaker than that of the conventional group. Therefore, CCTA was used to compare the average diameter of the sequential graft of the two groups 3 months after surgery. The results indicate that the average diameter of veins in the NT group is signi cantly less than that in the conventional group (Fig. 3). This suggested that the NT technology reduced the damage caused by vein expansion while ensuring the safety and effectiveness of the surgery. This milder vein expansion may have a protective effect on the long-term patency of the sequential grafts.
There were limitations in this study: The study was a single-center research; The patients in the group were mainly from the north of China, which may present a bias in the results. A multicenter, large sample, prospective randomized controlled study may be carried out in the future, which can provide more de nitive evidence for the effect of NT technology; the primary endpoint event was the occlusion rate of the venous graft three months after the operation. The Souza team's study conducted 72 cases 8.5 years after the surgery and 27 cases16 years in each group. The longer the follow-up time, the more signi cant difference appeared between the two groups. Therefore, this study will continue to follow up the patients for one, ve, ten years, or even longer; nally, samples sizes were determined for the variables at the cardiac computed tomography follow-up one year after the surgery and not for the end points that were the focus of this study. More extensive studies with long-term follow-up with clinical endpoints are warranted.

Conclusions
The early clinical results suggest that the NT technique is safety and e cacy compared to the conventional vein grafting technique in the sequential saphenous vein grafting during off-pump coronary artery bypass surgery. The sequential grafts early expansion in the NT technique is not as distinct as the conventional technique, which may have a long-term protective effect on the grafts. These ndings add to the accumulating evidence that no-touch is an alternative manner in sequential saphenous vein grafting during off-pump coronary artery bypass surgery. Large-scale prospective randomized controlled studies currently underway may be able to more fully prove the clinical effects of NT technology in the future.  Comparison of average diameter of sequential grafts

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. CONSORT2010Checklist.doc Protocol.docx