No-touch versus conventional vein harvesting in coronary bypass surgery: A meta-analysis of randomized controlled trials

Abstract

Objectives

The study evaluated whether no-touch saphenous vein harvesting (NTH) improves graft patency compared with conventional harvesting (CH) and whether this translates into clinical benefit in coronary artery bypass grafting.

Methods

A systematic literature review was performed during May 2025 to retrieve randomized controlled trials comparing NTH and CH in coronary artery bypass grafting. Primary outcomes included graft occlusion per graft and per patient, and graft failure per patient. Secondary outcomes included all-cause death, myocardial infarction, repeat revascularization, recurrent angina, and leg wound complications. Data were extracted and pooled analyses were performed using a random-effects model.

Results

Ten randomized controlled trials with 4251 patients (NTH: 2143, CH: 2108) and 4848 grafts (NTH: 2415, CH: 2433) were included. Weighted mean follow-up was 39.0 months (range, 0.2-192 months). NTH was associated with a lower risk of graft occlusion per graft (risk ratio [RR], 0.70; 95% CI, 0.61-0.81), per patient (RR, 0.71; 95% CI, 0.60-0.84), and graft failure per patient (RR, 0.74; 95% CI, 0.64-0.87). Secondary outcomes showed no significant differences in all-cause death (hazard ratio [HR], 0.87; 95% CI, 0.43-1.79), myocardial infarction (HR, 0.79; 95% CI, 0.34-1.87), repeat revascularization (HR, 0.80; 95% CI, 0.32-1.95), and recurrent angina (HR, 0.59; 95% CI, 0.11-3.33), except for leg wound complications, which were significantly higher with NTH (RR, 1.78; 95% CI, 1.30-2.43).

Conclusions

NTH significantly improves graft patency compared with CH but confers no midterm advantage in major clinical outcomes, with increased leg wound complications. Large-scale studies with extended follow-up are needed to define its clinical influence.

Graphical Abstract

Central Picture Legends: Meta-analysis of RCTs comparing no-touch and conventional saphenous vein harvesting.

Central Message.

No-touch vein harvesting improves graft patency compared with conventional methods but does not confer midterm clinical benefit and increases leg wound complications.

Perspective.

The optimal technique for harvesting the saphenous vein during CABG remains a topic of debate. Although the no-touch harvesting technique has demonstrated better patency rates compared with the conventional method, it does not provide midterm benefits and is associated with a higher incidence of leg wound complications. These findings may inform surgical decisions during CABG procedures.

Coronary artery disease is the leading cause of death in the United States,^1,2 and surgical revascularization plays a critical role in its management. Coronary artery bypass grafting (CABG) is the standard treatment for complex lesions, with long-term graft patency being a key determinant of success. Total arterial graft revascularization has gained evidence supporting superior patency and survival compared with venous grafting.^3,4 However, saphenous vein grafts (SVGs) remain the most frequently employed secondary grafts in up to 80% to 90% of cases.^5,6 The major concern associated with SVGs is suboptimal long-term durability, with approximately 25% of SVGs occluding within the first year and 50% to 60% patency at 10 years.6, 7, 8

The underlying mechanisms of SVG failure were theorized to stem from endothelial injury and reduced nitric oxide formation due to skeletonization. To mitigate such shortcomings, the no-touch harvesting (NTH) technique, first described by Souza in 1996,⁹ was developed. The NTH technique preserves the adventitia and surrounding tissue without skeletonization of the grafts, resulting in reduced mechanical trauma, maintained endogenous nitric oxide release, and attenuated inflammatory response.¹⁰ Furthermore, NTH preserves the vasa vasorum, potentially maintaining vascular wall perfusion and limiting intimal hyperplasia.¹¹

Several randomized controlled trials (RCTs) have demonstrated improved graft patency with NTH compared with conventional harvesting (CH).12, 13, 14 The 2018 European Society of Cardiology/European Association for Cardio-Thoracic Surgery guidelines recommend NTH for open harvesting as a Class IIa indication.¹⁵ The American College of Cardiology and the American Heart Association have also endorsed NTH as part of best practices for the use of bypass conduits in CABG, particularly in patients at low risk for wound complications.¹⁶ However, its adoption has remained limited due to concerns regarding increased leg wound complications and uncertainty about its effect on clinical outcomes, including all-cause death and repeat revascularization. A prior Bayesian meta-analysis of RCTs reconfirmed the NTH's patency benefit but found no difference in clinical outcomes due to lack of statistical power.¹⁷ Moreover, key clinical outcomes such as revascularization, recurrent angina, and myocardial infarction were not assessed, prompting the need for an updated analysis incorporating these end points.

In contrast to prior meta-analyses,¹⁷ our study incorporates the latest randomized data from the PATENCY¹⁸ and SWEDEGRAFT¹⁹ trials and expands the end point spectrum to include clinical outcomes such as angina recurrence, myocardial infarction, and repeat revascularization.

Methods

This systematic review and meta-analysis were conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and registered with PROSPERO (CRD420251106230). Given the nature of our study, institutional review board or informed written consent for publication were not required.

Literature Search

A comprehensive literature search was independently performed by 2 authors (T.Y. and T.S.) using the Medline, Embase, Cochrane Central databases to identify RCTs comparing NTH versus CH SVG techniques in patients with coronary artery disease undergoing CABG. The search included all relevant studies published up to May 2025. References of included articles and relevant reviews were also screened manually. The detailed search strategies are demonstrated in Tables E1, E2, and E3. No language restrictions were applied. The study selection process according to Preferred Reporting Items for Systematic Reviews and Meta-Analysis 2020 guidelines are summarized in Figure 1.

Figure 1.

Preferred Reporting Items for Systematic reviews and Meta-Analyses 2020 flow diagram illustrating the selection process of randomized controlled trials.

Eligibility Criteria

Studies were eligible if they met the following inclusion criteria: RCT design, comparison of NTH and CH techniques for harvesting SVGs, and reporting outcomes on graft patency or clinical events. Studies were excluded if they lacked relevant outcome data or were available only in abstract form.

Risk of Bias in Individual Studies

Data extraction and risk of bias assessment were performed independently by 2 authors using the Cochrane Risk of Bias 2 tool.²⁰ Disagreements were resolved through consensus.

Summary Measures

The primary outcomes were graft occlusion per graft, graft occlusion per patient, and graft failure, defined as ≥50% stenosis or complete occlusion of the SVG. The secondary outcomes included all-cause death, myocardial infarction (MI), repeat revascularization, angina recurrence, and leg wound complications. Outcome definitions for each study were summarized in Table E4, showing notable variability in outcomes such as graft failure and leg wound complications.

For graft occlusion, graft failure, and leg wound complications, risk ratios (RRs) with corresponding 95% CIs were extracted. If RRs were not available, RRs were calculated using the event numbers and number of patients. For clinical outcomes during follow-up period, including all- cause death, MI, repeat revascularization, and angina recurrence, hazard ratios (HRs) with corresponding 95% CIs were extracted. When HRs were unavailable, RRs were derived from event counts and used as approximations.

Statistical Analysis

All statistical analyses were conducted using R software version 4.4.2 (R Foundation for Statistical Computing) with the meta package (version 8.1.0). Dichotomous outcomes were analyzed using a random-effects model. Pooled estimates were reported as RR or HR with corresponding 95% CIs. The Mantel-Haenszel method was applied for outcomes reported as RR, and the generic inverse variance method was applied for those reported as HR. Between-study heterogeneity was assessed using the I² statistic, τ², and Cochran's Q test. Heterogeneity was considered substantial when I² >50% or P < .05 for the Q statistic. Forest plots were constructed to visualize pooled effect sizes on a logarithmic scale.

Secondary Analyses

In addition to relative measures, absolute event rates for each outcome were collected and summarized. Funnel plot asymmetry was examined visually; Egger's linear regression test was not used due to the small number of studies. Leave-1-out sensitivity analyses were performed to assess the influence of individual studies on pooled estimates. A meta-regression analysis was conducted to evaluate the relationship between baseline characteristics of patients, including age, sex, or diabetes mellitus (DM) and outcomes of graft occlusion per patient and all-cause death.

Results

A total of 10 RCTs involving 4251 patients (NTH group; n = 2143, CH group; n = 2108) were included in this meta-analysis.^12,18,19,21, 22, 23, 24, 25, 26, 27 These patients received a total of 4848 SVGs, with 2415 in the NTH group and 2433 in the CH group. The weighted mean follow-up duration was 39.0 months, ranging from 0.2 to 192 months across the studies. The baseline patient characteristics of the included RCTs are summarized in Table 1. Mean age of patients ranged from 58.0 to 74.5 years.^12,27

Table 1.

Baseline characteristics of patients in Included randomized controlled trials comparing no-touch and conventional vein harvesting

First author	Year	Enrollment period	Mean follow-up duration (mo)	Patients (NTH/CH)	Age (y)	Sex male (%)	BMI mean	Smoking (%)	DM (%)	HT (%)	HL (%)	PAD (%)	Previous MI (%)	Previous stroke (%)	Previous PCI (%)	LVEF (mean/mean %)
Tian18	2025	2017-2019	36	1337/1318	61.0/60.9	78.7/78.1	25.7/25.5	53.8/55.5	36.4/35.2	64.6/61.7	67.2/68.8	7.7/6.6	21.0/20.5	3.5/2.7	14.5/15.3	60.8/60.3
Thelin19	2025	2018-2020	53	454/446	66.9/67.1	88.1/87.9	28.0/28.0	N/A	28.4/25.8	N/A	N/A	4.6/3.6	N/A	N/A	20.3/19.3	N/A
Angelini21	2021	2009-2013	12	51/45	67.3/64.5	90.2/91.1	27.6/28.1	70.6/62.2	17.6/15.6	80.4/84.4	88.2/97.8	0.0/0.0	N/A	N/A	N/A	N/A
Hou22	2021	2018-2019	3	50/50	61.0/59.8	92.0/94.0	31.0/30.0	52.0/56.0	36.0/40.0	58.0/60.0	24.0/22.0	28.0/28.0	46.0/44.0	8.0/10.0	12.0/12.0	60.0/60.0
Jiang23	2020	2017-2017	0.21	30/29	64.5/64.2	70.0/69.0	N/A	16.7/17.2	26.7/31.0	53.3/44.8	46.7/37.9	N/A	N/A	10.0/24.1	N/A	59.5/60.0
Deb24	2019	2011-2013	12.7	127/123	65.5/64.0	83.5/91.9	28.2/29.0	62.2/66.7	34.6/34.1	75.6/83.7	N/A	2.4/4.1	39.4/39.8	7.1/6.5	14.2/21.1	53.2/52.1
Pettersen25	2017	2013-2014	6	49/51	63.4/65.0	91.8/82.4	27.6/26.5	51.0/35.3	2.0/3.9	N/A	N/A	N/A	28.6/41.2	N/A	N/A	56.0/57.0
Samano12	2015	1993-1997	192	45/46	77.6/71.4	55.6/50.0	N/A	0.0/8.7	22.2/17.4	33.3/39.1	57.8/54.3	N/A	N/A	N/A	N/A	N/A
Johansson26	2009	1993-1997	102	45/46	66.8/66.7	95.6/82.6	N/A	0.0/15.2	40.0/23.9	60.0/34.8	82.2/76.1	N/A	57.8/58.7	N/A	N/A	N/A
Souza27	2002	1993-1997	16.5	45/46	58.0/58.0	84.4/87.0	N/A	0.0/19.6	N/A	N/A	64.4/63.0	N/A	55.6/63.0	N/A	N/A	67.0/63.0

Values are presented as mean (%) or mean ± SD, stratified by study group. NTH, No-touch saphenous vein harvesting; CH, conventional harvesting; BMI, body mass index; DM, diabetes mellitus; HT, hypertension; HL, hyperlipidemia; PAD, peripheral artery disease; MI, myocardial infarction; PCI, percutaneous coronary intervention; LVEF, left ventricular ejection fraction; N/A, not reported in the original study.

Primary Analyses

Graft occlusion occurred in 10% of patients in the NTH group and 14% in the CH group. The NTH group was associated with a significantly lower risk of graft occlusion (per graft: RR, 0.70; 95% CI, 0.61-0.81; P < .0001; I² = 0.0%, per patient: RR, 0.71; 95% CI, 0.60-0.84; P < .0001; I² = 0.0%) (Figure 2, A and B). Additionally, the NTH group demonstrated reduced rates of graft failure per patient (RR, 0.74; 95% CI, 0.64-0.87; P = .0002; I² = 0.0%) (Figure 2, C).

Figure 2.

Forest plots comparing no-touch saphenous vein harvesting versus conventional harvesting techniques in coronary artery bypass grafting. A, Graft occlusion per graft. B, Graft occlusion per patient. C, Graft failure per patient. RR, Risk ratio.

For secondary clinical outcomes, we found no difference in all-cause death (HR, 0.87; 95% CI, 0.43-1.79; P = .604; I² = 20.5%) (Figure 3, A). Similarly, no statistically significant differences were observed for MI (HR, 0.79; 95% CI, 0.34-1.87; P = .453; I² = 54.7%) (Figure 3, B), repeat revascularization (HR, 0.80; 95% CI, 0.32-1.95; P = .458; I² = 47.6%) (Figure 3, C), and recurrent angina (HR, 0.59; 95% CI, 0.11-3.33; P = .106; I² = 61.1%) (Figure 3, D). In contrast, the incidence of leg wound complications was more frequent in the NTH group (14% vs 7%), with relative measures favoring the CH group (HR, 1.78; 95% CI, 1.30-2.43; P = .0003; I² = 47.6%) (Figure 3, E). The absolute event rates for each clinical outcome are summarized in Table E5.

Figure 3.

Forest plots comparing no-touch saphenous vein harvesting versus conventional harvesting techniques in coronary artery bypass grafting for clinical outcomes. A, All-cause death. B, Myocardial infarction. C, Repeat revascularization. D, Angina recurrence. E, Leg wound complications. HR, Hazard ratio; RR, risk ratio.

Secondary Analyses

Significant heterogeneity was present in MI (I² = 56.8%) and angina recurrence (I² = 61.1%); other outcomes did not report significant heterogeneity. Leave-1-out sensitivity analyses were consistent with the primary analyses, confirming the benefits of NTH on graft patency (Figure E1). The funnel plot is presented in Figure E2, and no publication bias was present on visual assessment.

Figure E1.

Leave-1-out sensitivity analyses for each clinical outcome comparing no-touch saphenous vein harvesting (NTH) versus conventional harvesting (CH) techniques in coronary artery bypass grafting (CABG). A, Graft occlusion per graft. B, Graft occlusion per patient. C, Graft failure. D, All-cause death. E, Myocardial infarction (MI). F, Repeat revascularization. G, Angina recurrence. H, Leg wound complications. Each plot shows the pooled risk ratio (RR) or hazard ratio (HR) after sequentially omitting 1 study at a time.

Figure E2.

Funnel plots assessing publication bias in studies comparing no-touch saphenous vein harvesting (NTH) versus conventional harvesting (CH) techniques in coronary artery bypass grafting (CABG). A, Graft occlusion per graft. B, Graft occlusion per patient. C, Graft failure. D, All-cause death. E, Myocardial infarction (MI). F, Repeat revascularization. G, Angina recurrence. H, Leg wound complications. The plots display SE versus log risk or hazard ratio. Symmetry suggests low risk of publication bias.

Meta-regression analyses for graft occlusion per patient and all-cause death revealed no statistically significant associations with age, sex, or DM (Table 2). Risk of bias assessments using the Cochrane Risk of Bias 2 tool are summarized in Figure E3, indicating high risk in 3 studies^12,22,23 and some concerns in 4 studies.^21,24,26,27 A sensitivity analysis excluding high-risk-of-bias studies (n = 3)^12,22,23 yielded similar pooled estimates (graft occlusion per graft RR, 0.63; 95% CI, 0.52-0.78; P < .0001; I² = 0.0, graft occlusion per patient RR, 0.72; 95% CI, 0.60-0.86; P = .0002; I² = 0.0%), confirming the robustness of the results (Figure E4). Other secondary outcomes did not include any high-risk studies.

Table 2.

Univariable meta-regression analyses for graft occlusion per patient and all-cause death

Variable	Graft occlusion per patient				All-cause death
	LogHR	SE	HR	95% CI	LogHR	SE	HR	95% CI
Age	0.003	0.022	1.003	0.961-1.047	−0.103	0.083	0.902	0.767-1.061
Male sex	0.758	0.979	2.135	0.313-14.557	−6.311	4.121	0.002	0.000-5.851
DM	−0.447	1.678	0.64	0.024-17.149	−3.1	5.09	0.045	0.000-967.972

HR, Hazard ratio; DM, diabetes mellitus.

Figure E3.

Risk of bias assessment using the Cochrane Risk of Bias 2 tool for Included randomized controlled trials.

Discussion

This updated meta-analysis of 10 RCTs demonstrates that NTH technique is associated with favorable graft patency compared with the CH technique (Figure 4). Consistent with prior reports, this patency advantage was observed in both graft-level and patient-level analyses. However, this advantage did not translate to measurable clinical advantages, including mortality, repeat revascularization, MI, and recurrent anginal episodes. Our study also demonstrated higher leg wound complication rate in the NTH technique, as previously reported.¹⁷ Clinicians should interpret these findings with caution, recognizing that improved angiographic patency has not yet translated into reductions in death or ischemic events within current follow-up durations.

Figure 4.

Graphical summary of the meta-analysis. Ten randomized controlled trials (RCTs) were analyzed, showing superior graft patency in the no-touch group. However, this advantage did not translate into improved clinical outcomes and was associated with higher rates of leg wound complications. RR, Risk ratio; HR, hazard ratio; MI, myocardial infarction.

In the present analysis, NTH was associated with significantly lower rates of graft occlusion compared with CH, with consistent benefits observed across studies. This finding aligns with previously proposed mechanisms that support the better patency of NTH. The technique preserves the perivascular tissue and avoids direct manipulation of the vein, thereby minimizing endothelial injury and preserving the vein's antithrombotic properties. Previous studies have shown that NTH exhibits greater endothelial nitric oxide synthase expression, which helps maintain vasodilation and inhibit platelet aggregation.¹⁰ NTH also preserves an intact vasa vasorum, which may improve vessel wall perfusion and reduce ischemia-induced smooth muscle proliferation and intimal hyperplasia.¹¹ These biological mechanisms likely contribute to the lower occlusion rates observed in our meta-analysis. Additionally, procedural nuances such as the use of syringe dilation during vein preparation may influence endothelial integrity and subsequent graft patency.²⁸ However, these technical details were inconsistently reported among the included trials, precluding subgroup analysis based on this variable.

Despite superior patency, NTH did not significantly reduce clinical end points such as all-cause death, MI, or repeat revascularization in the included trials. This aligns with prior systematic reviews and registry studies suggesting that enhanced angiographic outcomes do not always translate to better clinical end points, particularly within the relatively short follow-up periods available in most trials.¹⁷ It is plausible that the benefits of improved patency may become more clinically evident over longer durations, as suggested by long-term observational data showing sustained patency and reduced event rates with NTH at 8 to 16 years post-CABG.^12,29

The increased incidence of leg wound complications associated with NTH harvesting remains a critical limitation of the technique. Multiple studies have reported that the wider tissue margin involved in NTH results in higher rates of infection, delayed healing, and patient discomfort.^21,24 Although some trials used standardized definitions and follow-up protocols for leg wound assessment, considerable variation remains, highlighting the need for unified criteria in future studies. Innovations such as minimally invasive or endoscopic NTH harvesting may help mitigate this complication, although data supporting such approaches remain limited.^30,31 Patient selection remains crucial. Individuals with obesity, DM, or peripheral vascular disease may have elevated risk of wound morbidity, limiting the generalizability of open NTH harvesting. Future studies should evaluate minimally invasive or endoscopic NTH approaches as potential solutions to reduce leg wound complications.³² Additionally, outcome definitions differed across trials, especially for wound-related complications, which may partly explain the heterogeneity observed. Nevertheless, the overall direction and magnitude of effect remained consistent.

Our findings complement those of recent meta-analyses, including 1 by Deng and colleagues¹⁷ that incorporated both frequentist and Bayesian methods and similarly concluded that NTH improves patency but not survival, and increases wound complications. Compared with their work, our meta-analysis included additional trial and updated results from the PATENCY trial and the SWEDEGRAFT trial, enabling a broader assessment of outcomes beyond angiographic end points.^18,19 Moreover, we incorporated a wider range of clinical outcomes, to better characterize the trade-offs of NTH in routine surgical practice. This updated analysis therefore provides the most comprehensive synthesis of randomized evidence to date, extending beyond angiographic patency to encompass clinically meaningful outcomes.

Although the SWEDEGRAFT trial, with 3.5 to 4.4 years of follow-up, demonstrated comparable composite clinical outcomes,¹⁹ the PATENCY trial, at 3 years, showed favorable graft occlusion, MI, and repeat revascularizations rates with the no-touch technique.¹⁸ This discrepancy may reflect the underpowered design of SWEDEGRAFT,¹⁹ which assumed a 50% vein graft failure rate based on the 2-decade-old The PRoject of Ex-vivo Vein graft ENgineering via Transfection IV (PREVENT-IV) trial.³³ These findings further underscore the value of pooled analyses to increase statistical power. Moreover, the divergent results highlight the need for larger, longer-term trials with consideration for factors, including patient selection, conduit configuration, use of multiple arterial grafts, and follow-up duration to better determine drivers of graft outcomes.

Long-term randomized follow-up is essential to determine whether the patency benefit of NTH ultimately translates into survival advantage. Future trials should assess endoscopic NTH³² techniques that may preserve graft integrity while minimizing wound complications.

Limitations

This study has several limitations. First, given the nature of a meta-analysis, our study is inherent to the limitations associated with the included RCTs. Second, the median follow-up period of the included studies was 39.0 months. Although the overall follow-up in this study is longer than the prior meta-analyses, it may still be insufficient to detect differences in clinical outcomes.¹⁷ Furthermore, variations in surgical technique, leg wound assessment, and perioperative care across studies may have caused bias, despite overall low statistical heterogeneity. Additionally, some HRs were approximated from RRs derived from event data. Although this approach is acceptable according to the Cochrane Handbook for Systematic Reviews of Interventions,³⁴ it may introduce bias due to the lack of time-to-event information. Furthermore, the limited number of trials included in this meta-analysis renders the meta-regression analyses exploratory in nature. Subgroup or meta-regression analyses based on operative strategy were not feasible due to inconsistent reporting of details, including on-pump versus off-pump techniques and target vessel locations. Future studies are warranted to investigate these critical characteristics. It is worth noting that some of the included trials were conducted more than 2 decades ago. However, the patient sample from these earlier studies was relatively small, with the majority of patients from contemporary trials. Future studies investigating the influence of advances in surgical technique and perioperative management are warranted. Finally, most of the included trials were underpowered to evaluate hard clinical end points and many were single-center studies, which limits generalizability.

Conclusions

NTH significantly improves graft patency compared with CH but shows no significant advantage in major clinical outcomes and is associated with increased leg wound complications. Large-scale trials with long-term follow-up are warranted to better define the clinical utility of the NTH technique.

Conflict of Interest Statement

Dr Fukuhara is a consultant for Terumo Aortic, Medtronic, Edwards Lifesciences, and Artivion. Dr Kaneko is a speaker for Medtronic and Edwards Lifesciences and a consultant for Baylis Medical and 4C Medical. Dr Takayama is a consultant for Artivion and Edwards Lifesciences. All other authors reported no conflicts of interest.

The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest.

Appendix E1

Figure E4.

Sensitivity analysis excluding high-risk-of-bias studies. A, Graft occlusion per graft. B, Graft occlusion per patient. RR, Risk ratio.

Table E1.

Search strategy for Medline

Search No.	Query	Results
1	(“Coronary Artery Bypass” [Mesh] OR “coronary artery bypass grafting” OR CABG OR “coronary artery bypass surgery”) AND (“Saphenous Vein” [Mesh] OR “saphenous vein graft” OR “vein harvesting”)	5505
2	(“No-touch” OR “no touch” OR “non-touch” OR “non touch” OR atraumatic OR “minimally traumatic” OR “pedicled” OR “pedicle”)	46,485
3	(“Conventional” OR “conventional harvesting” OR “traditional harvesting”)	695,067
4	(randomized controlled trial [pt] OR controlled clinical trial [pt] OR randomized [tiab] OR placebo [tiab] OR “clinical trial” [tiab])	1,317,259
5	(animals [mh] NOT humans [mh])	5,353,154
6	#1 AND #2 AND #3 AND #4 NOT #5	31

Table E2.

Search strategy for Embase

Search no.	Query	Results
1	(‘coronary artery bypass’/exp OR ‘coronary artery bypass grafting’:ti, ab OR CABG:ti, ab OR ‘coronary artery bypass surgery’:ti,ab)	125,241
2	(‘saphenous vein’/exp OR ‘saphenous vein graft’:ti, ab OR ‘vein harvesting’:ti,ab)	22,979
3	(‘no touch’:ti, ab OR ‘no-touch’:ti, ab OR ‘non-touch’:ti, ab OR ‘non touch’:ti, ab OR atraumatic:ti, ab OR ‘minimally traumatic’:ti, ab OR pedicled:ti, ab OR pedicle:ti,ab)	57,815
4	(‘conventional’:ti, ab OR ‘conventional harvesting’:ti, ab OR ‘traditional harvesting’:ti,ab)	900,682
5	(‘randomized controlled trial’/exp OR ‘controlled clinical trial’/exp OR randomized:ti, ab OR placebo:ti, ab OR ‘clinical trial’:ti,ab)	2,158,313
6	([animals]/lim NOT [humans]/lim)	6,722,641
7	#1 AND #2 AND #3 AND #4 AND #5 NOT #6	37

Table E3.

Search strategy for Cochrane Central

Search no.	Query	Results
1	(coronary artery bypass OR CABG OR coronary artery bypass grafting OR coronary artery bypass surgery)	15,639
2	(coronary artery bypass OR CABG OR coronary artery bypass grafting OR coronary artery bypass surgery)	2079
3	(“no-touch” OR “no touch” OR “non-touch” OR “non touch” OR atraumatic OR “minimally traumatic” OR pedicled OR pedicle)	2793
4	(conventional OR “conventional harvesting” OR “traditional harvesting”)	88,230
5	#1 AND #2 AND #3 AND #4	44

Table E4.

Definition of each outcome

First author	Year	Graft occlusion	Graft failure	Myocardial infarction	Repeat revascularization	Angina recurrence	Leg complications
Tian18	2025	Conduit did not fill with contrast or if a string sign was observed in any segment on CTA at 36 mo	Graft occlusion or graft stenosis >50% but not occluded on CTA at 36 mo	Elevation of cardiac troponin values with at least 1 value above the 99th percentile upper reference limit, plus evidence of myocardial necrosis over 36-mo follow-up	Coronary revascularization includes coronary artery grafting or angiography, as well as other percutaneous coronary interventions designed to treat coronary artery lesions over 36-mo follow-up	Ischemic chest pain or equivalent, and no evidence of MI, and no clear alternative explanation over 36-mo follow-up	Wound requiring surgical intervention over 3 mo
Thelin19	2025	Occlusion on CTA at 42 mo	At least 1 vein graft occluded or stenosed >50% on CTA at 42 mo	Not clearly specified, events over 53-mo follow-up	Not clearly specified, events over 53-mo follow-up	Inquiries were made regarding physical limitations, recurrent angina, and health status using the Seattle Angina Questionnaire-7 over 53-mo follow-up	Postoperative leg wound complications, contact with health care, wound dressing by a nurse, wound assessment by physician, surgical intervention of leg wound, antibiotics treatment due to leg wound complications at 3 mo
Angelini21	2021	Occlusion on IVUS or angiography at 12 mo	Not documented	Not documented	Not documented	Angina-related quality of life and symptoms were evaluated using the Seattle Angina Questionnaire over 12 mo	Persistent pain at 3 mo
Hou22	2021	FitzGibbon-O∗ on CTA at 3 mo	Not documented	Not documented	Not documented	Not documented	Not documented
Jiang23	2020	Occlusion on CTA at 0.21 mo	Not documented	Not documented	Not documented	Not documented	Postoperative infection, follow-up duration not clearly specified
Deb24	2019	Complete (100%) occlusion on CTA at 12.7 mo	50%-99% stenosis or complete occlusion at 12.7 mo	Not clearly specified, events over 12.7-mo follow-up	Not clearly specified, events over 12.7-mo follow-up	Not documented	Cumulative incidence, including infection, necrosis, dehiscence, drainage, and fluid collection at 1 y
Pettersen25	2017	Occlusion of angiography, including OCT	Not documented	Not documented	Not clearly specified, events over 6-mo follow-up	Not documented	Infection treated surgically or with antibiotics or as a wound with positive microbial culture at 6 wk
Samano12	2015	Graft that was not opacified by contrast on CTA at 192 mo	Not documented	Not documented	Not documented	Not documented	Not documented
Johansson26	2009	Not documented (angiography)	Not documented	Not clearly specified, events over 102-mo follow-up	PCI or re-CABG over 102-mo follow-up	CCS class I-IV events over 102-mo follow-up	Not documented
Souza27	2002	Occlusion on angiography at 18 mo	Not documented	Not documented	Not documented	Not documented	Cellulitis and infection, follow-up duration not clearly specified

CTA, Computed tomography angiography; MI, myocardial infarction; IVUS, intravascular ultrasound; OCT, optical coherence tomography; PCI, percutaneous coronary intervention; CABG, coronary artery bypass grafting; CCS, Canadian Cardiovascular Society.

^∗An occlusive graft without blood flow.

Table E5.

Absolute event rates

Outcome	NTH	CH	Pooled relative effect∗	Contributing data
Graft occlusion per graft	190/2462 (8)	292/2482 (12)	RR, 0.70 (0.61-0.81)	Tian 202518; Angelini 202121; Hou 202122; Deb 201924; Pettersen 201725; Samano 201512
Graft occlusion per patient	199/2013 (10)	277/1983 (14)	RR, 0.71 (0.60-0.84)	Tian 202518; Thelin 202519; Hou 202122; Deb 201924; Samano 201512
Graft failure	239/1918 (12)	318/1887 (17)	RR, 0.74 (0.64-0.87)	Tian 202518; Thelin 202519; Deb 201924
All-cause death	64/2012 (3)	66/1984 (3)	HR, 0.87 (0.43-1.79)	Tian 202518; Thelin 202519; Deb 201924; Pettersen 201725; Johansson 200926
MI	57/1963 (3)	72/1933 (4)	HR, 0.80 (0.45-1.41)	Tian 202518; Thelin 202519; Deb 201924; Johansson 200926
Repeat revascularization	67/2012 (3)	73/1984 (4)	HR, 0.80 (0.32-1.95)	Tian 202518; Thelin 202519; Deb 201924; Pettersen 201725; Johansson 200926
Angina recurrence	95/1433 (7)	137/1409 (10)	HR, 0.59 (0.11-3.33)	Tian 202518; Angelini 202121; Johansson 200926
Leg complications	292/2063 (14)	147/2029 (7)	RR, 1.78 (1.30-2.43)	Tian 202518; Thelin 202519; Angelini 202121; Deb 201924; Pettersen 201725; Souza 200227

Values are presented as n/N (%) unless otherwise noted. NTH, No-touch saphenous vein harvesting; CH, conventional harvesting; RR, risk ratio; HR, hazard ratio; MI, myocardial infarction.

^∗Pooled relative effect is presented as HR (95% CI) or RR (95% CI).