The impact of perivascular tissue preservation on 5-year patency of saphenous vein composite grafts

Abstract OBJECTIVES This retrospective study was conducted to evaluate the impact of saphenous vein (SV) harvesting with versus without perivascular tissue on the 5-year angiographic patency in coronary artery bypass grafting. METHODS Among the 944 patients who received coronary artery bypass grafting between 2010 and 2015, 579 patients who received off-pump coronary artery bypass grafting using 1 SV as a Y-composite graft based on the in situ left internal thoracic artery were enrolled. SV harvesting was performed using no-touch technique without perivascular tissue (the NoPVT group) in 342 patients and with perivascular tissue (the PVT group) in 237 patients. Follow-up duration was 84.0 months (interquartile range 66.5–105.4). Propensity score matching was performed, and long-term clinical outcomes and angiographic patency were compared. RESULTS The average number of distal anastomoses per patient was comparable between the groups, although more SV grafts were anastomosed to left anterior descending territory in the PVT group than in the NoPVT group. Overall survival and cumulative incidence of cardiac death were comparable between the groups, whereas cumulative incidence of target vessel revascularization (1.3% vs 4.3% at 5 year, P = 0.009) and that of major adverse cardiac events (7.3% vs 9.9% at 5 year, P = 0.035) were lower in the PVT group than in the NoPVT group. One-year and 5-year angiographic patency rates of the SV grafts were higher in the PVT group than in the NoPVT group [97.0% vs 91.7% (P = 0.004) and 96.3% vs 89.9% (P = 0.007), respectively]. CONCLUSIONS SV grafts harvested using no-touch technique with perivascular tissue further improved the 5-year patency of SV composite grafts compared with those without perivascular tissue.


INTRODUCTION
The internal thoracic artery is considered the 1st graft of choice for coronary artery bypass grafting (CABG) because of its excellent long-term patency and survival benefit [1,2].Regarding the 2nd graft of choice, the use of multiple arterial grafting is still limited, and this strategy is performed in only 3.9-34.2% of CABG patients in North America and Europe [3].In contrast, the saphenous vein (SV) graft, which was introduced more than 50 years ago [4], remains the most commonly used conduit despite the limitations of high failure rates in the short and long term [5][6][7].
Based on the no-touch (NT) SV harvesting technique, in which the SV is harvested and treated with minimal manipulation and manual intraluminal distension of the SV is avoided [8], a novel NT-SV grafting strategy that harvests the NT-SV with its surrounding perivascular tissue (PVT) was 1st introduced in 1996 [9], and previous studies have reported superior short-and longterm graft patency of novel NT-SV grafts compared with that of conventional SV grafts [10][11][12].In addition to the benefits of preventing traumatic manipulation, avoiding manual luminal distension and preserving the vasa vasorum, previous studies have suggested that preserving the PVT might have additional benefits for the NT-SV [13,14].
Therefore, the present retrospective observational singlecentre study was conducted to investigate the impact of PVT preservation during SV harvest on the 5-year angiographic patency of SV grafts.

Patient selection
The study protocol was reviewed by the Institutional Review Board and approved as a minimal risk retrospective study (approval number H-2201-116-1293), and individual consent was waived.Of 994 patients who received CABG between January 2010 and December 2015 at our institution, primary isolated off-pump CABG was performed in 880 patients.Among these patients, 579 patients who received off-pump CABG using the SV as a Y-composite graft based on in situ left internal thoracic artery (LITA) were enrolled in the present study.The SV was harvested using the NT technique with PVT (the PVT group) and without PVT (the NoPVT group) in 237 and 342 patients, respectively (Fig. 1).

Operative techniques and revascularization strategy
The operative procedures and surgical strategies of CABG have been illustrated in previous studies [15,16].During the study period, anaortic off-pump CABG using Y-composite grafting based on the in situ LITA was the preferred revascularization strategy for CABG at our institution to minimize cerebrovascular complications [17].The SV has been used as the preferred 2nd conduit of choice to construct Y-composite grafts based on our studies showing non-inferiority of the SV compared to arterial conduits [16,18].The NT-SV was harvested without PVT until September 2013.Since October 2013, the NT-SV was harvested with a substantial margin of PVT on both sides of the SV and with thin layers of adherent connective tissues anteriorly and posteriorly (Fig. 2).
The NT-SV was harvested preferentially from a lower leg to alleviate the possible size mismatch with native coronary arteries or ITAs, although it resulted in numbness around the lower leg incision in most of the patients.Immediately after the harvest, the reversed SV was anastomosed to the posterior aspect of the mid-LITA in an end-to-side fashion to construct a Y-composite graft.After the construction of the Y-composite graft, the left anterior descending coronary artery territory was revascularized 1st.Then, the left circumflex coronary artery territory was revascularized, followed by right coronary artery territory.A sequential anastomotic technique was also used as needed.

Evaluation of clinical outcomes
Operative mortality was defined as any mortality within 30 days after surgery or during the same hospitalization.After discharge, all patients visited the outpatient clinic regularly at 3-to 6month intervals.If the patients had not visited the outpatient clinic as scheduled, their survival and conditions were checked by telephone.For the patients in the loss of follow-ups, their survival data were obtained from the National Health Insurance Service and the Statistics Korea, a central organization for Cardiac death was defined as any death related to cardiac events, including sudden death during follow-up.Reintervention included both percutaneous coronary intervention and coronary reoperation.Reintervention was further divided into target vessel revascularization (TVR), which was defined as coronary interventions of target vessels that were bypassed during the index operation, and non-TVR, which was defined as coronary interventions for a new lesion outside the target vessels.Major adverse cardiac events included cardiac death, acute myocardial infarction and coronary reintervention.

Angiographic evaluation of graft patency
Early, 1-year and 5-year follow-up coronary angiograms were performed based on the patient's consent, regardless of angina symptoms.Patients who died, refused angiographic evaluation or had renal function impairment but were not dependent on renal replacement therapy were excluded from angiographic follow-up.Early postoperative [1 day (IQR 1-1)] graft evaluation was performed in 99.1% (574 out of 579) of the study patients with conventional angiography.One-year [12.5 months (IQR 12.1-13.3)]graft evaluation was performed in 491 patients (84.8%) using conventional angiography and multidetector computed tomography (MDCT) in 274 and 217 patients, respectively.Five-year [60.5 months (IQR 59.9-62.1)]graft evaluation was performed in 386 patients (66.7%) with conventional angiography and MDCT in 85 and 301 patients, respectively (Supplementary Material, Table S1).

Statistical analysis
Statistical analysis was performed using IBM SPSS version 25.0 (IBM Corp., Armonk, NY, USA) and SAS version 9.4 (SAS Institute Inc., Cary, NC, USA).Continuous variables are presented as the  mean ± standard deviation or median with interquartile range, and categorical variables are presented as the number and proportion.Propensity score-matched analysis was performed to adjust the differences in preoperative characteristics.Eighteen variables in baseline characteristics were selected for the propensity score-matched analysis including age, sex, smoking, body mass index >25.0kg/m 2 , hypertension, diabetes mellitus, dyslipidaemia, history of stroke, chronic kidney disease, chronic obstructive pulmonary disease, atrial fibrillation, peripheral vascular disease, previous percutaneous coronary intervention, left ventricular ejection fraction <0.35, preoperative diagnosis, 3vessel disease, left main disease and emergency operation.After calculating the propensity scores, 232 pairs were matched using a nearest neighbourhood within a calliper width of 0.1 in propensity scores and with a ratio of 1:1.The balance of covariates between the groups was evaluated with the standardized mean difference.A standardized mean difference of <0.100 was considered as a negligible difference between the groups.
In the unmatched population, the 2 groups were compared using the chi-square test or Fisher's exact test for categorical variables and Student's t-test or Wilcoxon rank sum test for continuous variables.In the matched population, McNemar tests were performed for categorical variables, and Wilcoxon signed rank sum tests or paired t tests were performed for continuous variables.Comparison of the graft patency rates between the 2 groups was performed using logistic regression with generalized estimating equations to account for the clustering of multiple measurements per subject.
For longitudinal data analyses, overall survival was estimated using the Kaplan-Meier method and compared with the log-rank test.Cardiac death, TVR, reintervention and major adverse cardiac events were analysed with cumulative incidence curves.Noncardiac death was considered a competing event for cardiac death, and all-cause death was considered a competing event for the other outcomes.Comparison of cumulative incidence curves between the 2 groups was performed using a subdistribution hazards model [19].A P value of <0.050 was considered to indicate statistical significance.

Patient characteristics
The mean age was 67.0 ± 9.7 years, and 24.2% (140 out of 579) of the patients were female.There were no differences in demographic data or preoperative risk factors between the 2 groups, except for a higher proportion of patients with a history of percutaneous coronary intervention in the PVT group than in the NoPVT group (19.8% vs 13.5%, P ¼ 0.040).After propensity score matching, the differences in patient characteristics between the groups significantly reduced (Table 1).

Operative data
The number of distal anastomoses per patient ranged from 2 to 6, and was comparable between the groups.The number of distal anastomoses per LITA was smaller in the PVT group than in the NoPVT group, whereas that per SV was greater in the PVT group than in the NoPVT group.These findings regarding the numbers of distal anastomoses remained consistent after propensity score matching (Table 2).

Early and long-term clinical outcomes
Operative mortality was 0.9% (5 out of 579 patients) without an inter-group difference.There were no significant differences in the occurrence rates of postoperative complications between the 2 groups both in the entire population and in the matched population, except for the higher incidence of SV harvest site complications in the PVT group (Table 3).

CORONARY DISEASE
cumulative incidence of TVR was lower in the PVT group than in the NoPVT group (1.3% vs 3.2%, P ¼ 0.027) (Supplementary Material, Fig. S1).In the matched population, 5-year overall survival rates and cumulative incidences of cardiac death consistently demonstrated no significant differences between the groups.However, the cumulative incidence of TVR (1.3% vs 4.3% at 5 year, P ¼ 0.009) and that of major adverse cardiac events (7.3% vs 9.9% at 5 year, P ¼ 0.035) were significantly lower in the PVT group than in the NoPVT group.The cumulative incidence of reintervention was also lower in the PVT group with a marginal significance (2.2% vs 5.2% at 5 year, P ¼ 0.057) (Fig. 3).

Angiographic graft patency
Early, 1-year and 5-year postoperative overall graft patency rates in the PVT versus NoPVT groups were 99.2% vs 97.9%, 97.6% vs 93.3% and 96.5% vs 92.2% in the matched cohort, respectively.All the obstructions occurred in distal anastomosis or distal graft, and no obstruction occurred at the level of Y-graft.There were no significant differences in graft patency rates of LITA between the 2 groups [early, 1 year and 5 years ¼ 99.6% vs 99.6% (P ¼ 0.941), 99.0% vs 97.0% (P ¼ 0.154) and 97.0% vs 97.2% (P ¼ 0.959), respectively], whereas patency rates of SV were significantly higher in the PVT group than in the NoPVT group at early, 1-year and 5-year angiograms (99.0%vs 97.1%, 97.0% vs 91.7% and 96.3% vs 89.9%, respectively) (Table 4).When the graft patency rates of the SV were analysed based on the coronary artery territory, they were highest when anastomosed to the left anterior descending coronary artery territory and lowest when anastomosed to the right coronary artery territory in both groups and in all time periods before and after matching.The patency rates of the SV to the right coronary artery territory showed significant differences between the 2 groups at 1 and 5 years [94.7% (161 of 170 distal anastomoses) vs 83.4% (141 of 169 distal anastomoses), P ¼ 0.003 and 93.3% (126 of 135 distal anastomoses) vs 82.6% (100 of 121 distal anastomoses), P ¼ 0.013, respectively], whereas there were no significant differences in the patency rates of the SV to the left anterior descending coronary artery territory in all time periods (Table 5).

DISCUSSION
The present study demonstrated that preservation of PVT in NT-SV harvesting might be beneficial in terms of 5-year angiographic patency and cumulative incidence of TVR after CABG.The concept of preserving the wall characteristics of SV grafts by the NT technique, which was inspired by the poor patency rates of SV grafts after CABG, was introduced in the early period of CABG, but it was not popularized [8].A novel NT-SV grafting strategy was then introduced in 1996 [9].Following the harvest methods of NT-SV as originally described, the novel strategy included preservation of the surrounding PVT.Although recent guidelines recommend NT-SV as Class IIa when an open technique is used [20], data regarding the benefits of NT-SV with PVT in CABG are limited because most studies have been reported from Souza and his colleagues, who introduced this technique [10-12, 21, 22].
In addition to the benefit of the NT technique that prevents the SV wall structures, particularly endothelial integrity, from mechanical damage during surgery, preserving the PVT in the novel NT technique has several theoretical advantages.These included (i) saving the vasa vasorum that prevents the SV from transmural ischaemic damage and (ii) buffering against pulsatile stress of the arterial flow that reduces neointimal and medial thickening [23][24][25][26].With this theoretical background, a previous randomized controlled trial enrolled 52 patients in each NT-SV and control group.That study showed that aortocoronary NT-SV with PVT resulted in favourable patency rates of 95.4%, 90% and 83% at 1.5, 8.5 and 16 years after CABG, respectively, although 16-year patency was evaluated in only 27 patients with 75 distal anastomoses [10,11,21].Another randomized trial that

CORONARY DISEASE
enrolled 250 patients from multiple institutions demonstrated that the composite rate of severe stenosis or occlusion at 12 months after CABG was 7.8% (8 of 102 grafts) in the NT-SV group [27].
In the present study, the NT-SV grafts were all used as composite grafts based on in situ LITA.Study results showed that 1year and 5-year patency of the NT-SV grafts without PVT were 92.3% and 91.5%, respectively, which might be lower than the former randomized controlled trial but comparable to the latter randomized controlled trial.The patency rates of the NT-SV in the PVT group were significantly higher than those in the NoPVT group and were comparable with those in previous studies using NT-SV with PVT.
In addition to the theoretical advantages described above, there has been growing evidence regarding the impact of the PVT on the NT-SV.Previous studies have suggested that adipose tissue in the PVT interacts with vascular smooth muscle cells and modulates vascular tone as well as externally supports vascular structures [28,29].Another study revealed that it is a major source of nitric oxide, which plays crucial roles in suppressing atherosclerosis [14].In that study, the authors analysed residual SVs obtained during CABG and demonstrated that the PVT expressed high levels of enzymes related to nitric oxide synthesis, such as argininosuccinate synthase 1 and endothelial nitric oxide synthase.
The present study confirmed that the PVT of the NT-SV is beneficial for patency of the NT-SV grafts up to 5 years after surgery.These improved patencies might result in favourable clinical outcomes in the PVT group, although those outcomes were not significantly different from those of the NoPVT group, except for the cumulative incidence of TVR.

Study limitations
There are several limitations to the present study that must be recognized.First, this study was a single institutional study and was not performed in a prospective randomized manner, although all consecutive patients who met the inclusion criteria were enrolled and propensity score matching was performed.Second, 1-year and 5-year graft evaluations were not performed in all the study patients, and graft patency rates might be overestimated.Third, there was a discrepancy in the modality, conventional angiogram versus MDCT, to evaluate the graft patency.Although the diagnostic efficacy of MDCT has been well demonstrated as excellent [30], this discrepancy could affect the results.Fourth, patients were recruited to this study at different time points with no overlap.Although propensity score matching was performed to overcome retrospective nature of the present study, other changes to surgical practice may have influenced the study results.Fifth, the effect of early technical failure to the long-term patency should also be considered in the interpretation of the results because the difference in graft patency existed at early angiographic evaluation.

CONCLUSIONS
Preserving the PVT during NT-SV harvest might improve graft patency rates of the SV up to 5 years after CABG.
statistics under the Ministry of Strategy and Finance.The clinical follow-up was closed on 31 December 2021.Follow-up data were complete in 100% and 99.5% (576 out of 579 patients) for survival and the other clinical outcomes, respectively, with a median follow-up duration of 84.0 months [interquartile range (IQR) 66.5-105.4].

Figure 3 :
Figure 3: Comparison of (A) overall survival, (B) cumulative incidence of cardiac death, (C) cumulative incidence of target vessel revascularization (TVR), (D) cumulative incidence of reintervention and (E) cumulative incidence of major adverse cardiac events (MACEs) between the groups in the propensity score-matched population.MACE: major adverse cardiac event; NoPVT: without PVT; PVT: perivascular tissue; TVR: target vessel revascularization.

Table 1 :
Preoperative characteristics and risk factors for the study patients Continuous variables are presented as the average ± standard deviation for normally distributed variables, and median with interquartile range for non-normally distributed variables.PVT: perivascular tissue; SMD: standardized mean difference.

Table 2 :
Comparison of the number of distal anastomoses between the groups LAD: left anterior descending; LCx: left circumflex; LITA: left internal thoracic artery; PVT: perivascular tissue; RCA: right coronary artery; SV: saphenous vein.

Table 3 :
Comparison of early clinical outcomes between the groups

Table 4 :
Comparison of early, 1-year and 5-year angiographic patency rates of all anastomoses between the groups

Table 5 :
Comparison of early, 1-year and 5-year angiographic patency rates of the anastomoses with saphenous vein grafts between the groups LAD: left anterior descending; LCX: left circumflex; PVT: perivascular tissue; RCA: right coronary artery; SV: saphenous vein.