Share this article on:

Endoscopic Vascular Harvest in Coronary Artery Bypass Grafting Surgery: A Meta-Analysis of Randomized Trials and Controlled Trials

Cheng, Davy MD*; Allen, Keith MD†; Cohn, William MD‡; Connolly, Mark MD§; Edgerton, James MD¶; Falk, Volkmar MD‖; Martin, Janet Pharm D*; Ohtsuka, Toshiya MD#; Vitali, Richard PA§

Innovations: Technology & Techniques in Cardiothoracic & Vascular Surgery: Winter 2005 - Volume 1 - Issue 2 - pp 61-74
doi: 10.1097/01.gim.0000196316.48694.41
Original Articles

Objective: This meta-analysis sought to determine whether endoscopic vascular graft harvesting (EVH) improves clinical and resource outcomes compared with conventional open graft harvesting (OVH) in adults undergoing coronary artery bypass surgery.

Methods: A comprehensive search was undertaken to identify all randomized and nonrandomized trials of EVH versus OVH up to April 2005. The primary outcome was wound complications. Secondary outcomes included any other clinical morbidity and resource utilization. Odds ratios (OR), weighted mean differences (WMD), or standardized mean differences (SMD) and their 95% confidence intervals (95% CI) were analyzed.

Results: Thirty-six trials of 9,632 patients undergoing saphenous vein harvest met the inclusion criteria (13 randomized; 23 nonrandomized). Risk of wound complications was significantly reduced by EVH compared with OVH (OR 0.31, 95% CI 0.23-0.41). Similarly, the risk of wound infections was significantly reduced (OR 0.23, 95% CI 0.20–0.53; P < 0.0001). Need for surgical wound intervention was also significantly reduced (OR 0.16, 95% CI 0.08–0.29). The incidence of pain, neuralgia, and patient satisfaction was improved with EVH compared with OVH. Postoperative myocardial infarction, stroke, reintervention for ischemia or angina recurrence, and mortality were not significantly different. Operative time was significantly increased (WMD 15.26 minutes; 95% CI 0.01, 30.51), hospital length of stay was reduced (WMD –0.85 days; 95% CI –1.55, –0.15), and readmissions were reduced (OR 0.53, 95% CI 0.29–0.98). Costs were insufficiently reported to allow for aggregate analysis.

Conclusions: Endoscopic vascular graft harvesting of the saphenous vein reduces wound complications and improves patient satisfaction and resource utilization. Further research is required to determine the incremental cost-effectiveness of EVH versus OVH.

From the *Department of Anesthesia & Perioperative Medicine, London Health Sciences Centre, The University of Western Ontario, London, ON, Canada; †The Heart Center of Indiana, Division of Cardiothoracic Surgery, Indianapolis, IN; ‡Texas Heart Institute, Houston, TX; §St. Michael's Medical Center, Newark, NJ; ¶Cardiopulmonary Research Science and Technology Institute, Dallas, TX; ‖Klinik fur Herzchirurgie, Herzzentrum Leipzig, Germany; #Department of Cardiothoracic Surgery, Tokyo Metropolitan Fuchu General Hospital, Tokyo, Japan.

Support for the meta-analysis and a consensus conference was provided by the International Society for Minimally Invasive Cardiothoracic Surgery (ISMICS), which has received unrestricted educational grants from industries that produce surgical technologies.

Presented at the ISMICS 8th Annual Meeting, New York, NY, June 1–4, 2005.

Address correspondence and reprint requests to Dr. Davy Cheng, LHSC-University Hospital, 339 Windermere Road, London, Ontario, Canada N6A 5A5; e-mail:

Coronary artery bypass grafting (CABG) surgery is still one of the most common surgeries performed worldwide, with over 500,000 surgeries performed annually. Saphenous vein (SV) bypass grafting remains predominant in coronary bypass surgery, despite increasing interest in arterial conduits. Saphenous vein harvested using a longitudinal open technique results in an incidence of wound complications (infection, cellulitis, drainage, dehiscence, delayed healing, lymphangitis, sepsis, and limb amputation) ranging from 2% to 25%, creating an important clinical and economic burden.1–4 In addition, open SV harvest is associated with postoperative pain, delayed ambulation, neuropathy, and long-term pain and scarring that may contribute to reduced patient satisfaction. After CABG, some patients report that the discomfort from the leg incision used to harvest the saphenous vein is greater than that from the sternotomy.3 Furthermore, statistics suggest that patients referred for CABG are increasingly older and sicker, with increased baseline risk of complications such as wound infections and death.

In an attempt to reduce the risk of these complications, and to improve patient satisfaction, there has been increasing interest in endoscopic SV harvest. A number of clinical trials have reported short- and long-term clinical and resource outcomes with endoscopic harvest compared with conventional technique. Each of these trials individually is underpowered to provide adequate estimates of potential clinical and economic advantages compared with conventional graft harvesting. However, each of these trials singularly has had insufficient sample size to rule out clinically important differences in selected clinical and economic outcomes. Combining these trials through meta-analysis would increase the power to detect important differences between endoscopic and conventional conduit harvest.

Presently, no comprehensive and methodologically rigorous meta-analysis of randomized trials comparing endoscopic vein and artery harvest with open harvest for CABG exists. Two related meta-analyses of minimally invasive harvesting focused only on vein harvest and combined nonendoscopic harvest with endoscopic technique.5,6 Additionally, a number of key clinical and economic outcomes other than wound infection were not addressed, and more recent trials have not been included in these earlier meta-analyses. Also notable, these earlier meta-analyses used liberal definitions for wound complications. Therefore, a current and comprehensive meta-analysis is needed to determine the benefits of endoscopic graft harvesting to patients (improved outcomes), practitioners (improved quality of care), and providers (improved cost-effectiveness).

Back to Top | Article Outline


This systematic review with meta-analysis sought to determine whether endoscopic vascular graft harvesting improves clinical and resource outcomes compared with conventional graft harvesting in adults undergoing coronary artery bypass surgery.

Back to Top | Article Outline


This meta-analysis of randomized and nonrandomized controlled trials was performed in accordance with state-of-the-art methodologic recommendations (ie, as per QUOROM Consensus and Cochrane Collaboration recommendations)7,8 and according to a protocol that prespecified outcomes, search strategies, inclusion criteria, and statistical analyses. Both randomized and nonrandomized comparative trials were sought, with the intention of analyzing the trials in aggregate, and separately, in order to determine whether nonrandomized data differed from the randomized data.

Back to Top | Article Outline

Definition of Clinical Endpoints

The primary endpoint was defined as the incidence of wound infection at the harvest site. Secondary endpoints were defined as wound complications, need for antibiotics, death, stroke, postoperative myocardial infarction, vein repairs, quality of vascular integrity, angiographic stenosis, angina recurrence or need for repeat revascularization (CABG or percutaneous coronary intervention), reexploration for bleeding, blood loss, pain, sensory disturbances, patient satisfaction, time to harvest, closure time, duration of surgery, time to ambulation, intensive care unit (ICU) length of stay, total hospital length of stay, physician and emergency room visits, homecare visits, readmissions for wound complications, and costs.

Back to Top | Article Outline

Literature Search

A comprehensive literature search of Medline, Embase, Cochrane Central, Current Contents, and Science Citation Index using keywords and variants of “minimally invasive,” “endoscopic,” “vein OR artery,” and “coronary artery bypass,” was performed from the earliest available date to April 2005. The most recent 12 months of relevant surgical and anesthesia journals were hand-searched, and databases of conference abstracts were reviewed electronically. Experts were contacted to solicit unpublished studies.

Back to Top | Article Outline

Inclusion Criteria

Eligible trials had to be randomized or nonrandomized controlled trials in adults undergoing coronary artery bypass surgery, whether on-pump or off-pump, and whether arterial or venous grafts were performed. Published and unpublished trials were included, in any language. Trials of combined surgical procedures were excluded, and studies that evaluated robotic surgery or nonendoscopic minimally invasive harvest techniques were excluded.

Back to Top | Article Outline

Data Extraction and Definitions

Two authors independently extracted the following data points: baseline demographics including number of patients, inclusion/exclusion criteria for patient entry to study, age, sex, diabetes, obesity, peripheral vascular disease, number of grafts performed, site of harvest, type of wound infections or complications, off-pump or on-pump CABG, surgical time, ventilation time, length of stay, costs, and all other relevant clinical outcomes. Discrepancies were resolved by consensus.

Wound infections were defined as per the author of the study, and included cellulitis when “wound infection” was not separately reported. Wound complications were defined as per the study authors’ definitions, and were recorded under this category only when reported by the study authors as “wound complications.” In general, “wound complications” usually comprised the composite of a number of subtypes of wound problems, including infection, cellulitis, abscess, drainage, seroma, lymphocele, dehiscence, necrosis, debridement, surgical wound intervention, and sometimes hematoma or inflammation. Importantly, subtypes of wound complications reported individually within the trials were not simply summed in order to create a composite for “wound complications,” because this would risk double-counting (ie, patients with one wound complication are more likely to present with additional complication subtypes, and would be represented more than once under the definition of wound complications if the subcomplications were summed). Surgical intervention for wound included debridement, amputations, fasciotomies, and tissue flaps. Patient dissatisfaction was defined as one minus the proportion of satisfied patients when “satisfaction” or “patient preference” was reported in the trials. Major coronary adverse events (MACE) was defined as per the study authors’ definitions, and usually included all-cause mortality, need for reintervention due to ischemia, myocardial infarction, stroke, and angina recurrence. Myocardial infarction was defined as per the study authors’ definitions. Quality of vascular integrity was defined by endothelial damage scores (typically 1 to 5, with 5 indicating maximum damage or no integrity), medial continuity scores, and purported markers of endothelial injury (ie, interleukin [IL]-1, IL-2, IL-10, and intracellular adhesion molecule [ICAM]) from examination of vein segments.

Back to Top | Article Outline

Data Analysis

Odds ratios and their 95% confidence intervals (OR, 95% CI) were calculated for discrete data. For continuous data, weighted mean differences (WMD, 95% CI) were calculated, except for endothelial scores and patient satisfaction scores, where the standardized mean difference (SMD, 95% CI) was used when appropriate. For discrete data with significant results, the number needed to treat (NNT, 95% CI) was calculated.9,10 Heterogeneity was explored through the Q-statistic, and by calculating the I2 statistic.11,12 Summary odds ratios and weighted mean differences were calculated using the fixed effects model when statistical heterogeneity was not found (ie, Q-test P value > 0.10 and I2 < 50%). The random effects model was used when statistical heterogeneity was found (ie, Q-test P value < 0.10 or I2 > 50%). Statistical significance for overall effect was defined as P < 0.05 or a confidence that excluded the value 1.00 for odds ratios and 0.00 for weighted or standardized mean differences. Sensitivity analysis was planned a priori for redo surgeries, diabetic patients, elderly patients (< 70 years), and publication status. Subgroup analysis was planned for randomized and nonrandomized studies. Publication bias was explored though visual inspection of funnel plots.

Back to Top | Article Outline


Study Identification

In searching for randomized trials, over 500 studies were screened, and 20 were identified as potentially relevant and retrieved for systematic review. Of these, 13 randomized trials published in 15 papers met the inclusion criteria, for a total of 1,319 randomized patients (Figure 1, QUOROM Flowchart).2,4,13–25

In searching for nonrandomized controlled trials, over 800 studies were screened, and 27 were identified as potentially relevant and retrieved for review. Of these, 23 nonrandomized trials published in 27 papers met the inclusion criteria, for a total of 8,313 nonrandomized patients (Figure 2, QUOROM Flowchart).3,26–51 Therefore, in total, 36 randomized and nonrandomized trials were included in this meta-analysis, including a total of 9,632 patients. All included studies were published. Two studies were published in languages other than English.36,51

Table 1 and 2 outline the characteristics of the included randomized and nonrandomized trials, respectively. All trials examined endoscopic harvest of the saphenous vein, and no trials of arterial endoscopic harvest were found. Whereas the comparator group in most trials used conventional open SV harvest, in some trials a modified bridge approach was used. Although many trials were conducted within the United States, other countries including Europe, Canada, and Asia were also represented. Most studies were conducted during 1996 to 2001 (range: 1995 to 2002).

A total of 4.3% of patients were reported to require conversion from EVH to OVH. For these patients, outcomes were not generally reported separately, and in some cases these patients were excluded by the authors of the original trials or were not analyzed by intention-to-treat.

Table 3 outlines the aggregate baseline characteristics of the patients included in the randomized and nonrandomized studies. Average age at baseline was 64.6 years; approximately 27% were female, 30% were diabetic patients, 20% were obese patients, and over 17% had peripheral vascular disease. In randomized studies, baseline characteristics did not differ significantly between EVH and OVH groups. However, in nonrandomized studies, the EVH group had significantly fewer females (24.7% vs. 28.0%, P = 0.04), and was more likely to be obese (22.3% vs. 18%, P = 0.02). This baseline imbalance in key prognostic characteristics is concerning, and may be evidence of selection bias, wherein patients who are male or obese were preferentially selected to be treated endoscopically due to the perception that they would benefit most from EVH compared with OVH. The mean length of vein harvested and the mean number of grafts performed did not differ significantly between groups (Table 3).

Clear evidence of publication bias was not found after visual inspection of funnel plots; however, inadequate power for some clinical outcomes precluded adequate analysis of bias. Heterogeneity between trials was found for a number of outcomes (Table 4), necessitating that the random effects model be used to incorporate the uncertainty between trials.

Back to Top | Article Outline

Clinical Outcomes

The forest plots given in Figure 3 display the results, individually and in aggregate, for all studies that reported on each relevant clinical outcome. When there was statistical heterogeneity, the difference between randomized and nonrandomized results was scrutinized to determine whether study design (and, presumably, selection bias in the nonrandomized trials could explain the heterogeneity). Sensitivity analysis for redo surgeries, diabetic patients, elderly patients, and publication status age was not possible due to inadequate data.

Heterogeneity was detected for the outcomes of wound complications, edema, hematoma, harvest vein repairs, pain scores in hospital, patient dissatisfaction, harvest time, operative time, closure time, time to ambulation, and hospital length of stay. When only the randomized data are analyzed for these heterogeneous data, the heterogeneity is generally reduced (or eliminated), suggesting that the nonrandomized data accounts for the heterogeneity. Therefore, for these clinical outcomes, the results of the randomized data should be given higher priority in clinical decision-making than the results of the nonrandomized data, due to the risk of bias in the latter.

Back to Top | Article Outline

Wound Complications

The risk of wound complication was significantly reduced by 69% with EVH compared with OVH (OR 0.31, 95% CI 0.23–0.41; P < 0.0001) with an estimated NNT of 11 (95% CI, 10 to 13). Similarly, the risk of wound infections were significantly reduced by 67% with EVH compared with OVH (OR 0.23, 95% CI 0.20–0.53; P < 0.0001), with an estimated NNT of 19 (95% CI, 15 to 25). The need for surgical intervention for wound infection was also significantly reduced (OR 0.16, 95% CI 0.08–0.29; P < 0.0001), with a NNT of 45 (35 to 60). The risk for most of the individual subcomponents of the definition of wound complications, including edema, seroma/lymphocele, necrosis, wound drainage, dehiscence, and need for antibiotics was significantly reduced (odds reductions ranging from 56% to 94%), with corresponding NNT estimates ranging from 4 to 45 (Table 4). However, hematomas and abscess formation were not significantly different between EVH and OVH groups.

Back to Top | Article Outline

Quality of Harvested Conduit

Although the mean number of venous repair stitches per person was increased by a mean of 1.11 stitches per patient with EVH compared with OVH (WMD 1.11, 95% CI 0.52–1.69, P = 0.0002), the odds of requiring harvest vein repair during the harvest procedure was not significantly different (OR 2.39, 95% CI 0.95–6.05, P = 0.07). Endothelial damage scores were not significantly different (SMD –0.09, 95% CI –0.27, 0.09, P = 0.3), and medial continuity score was not significantly different (WMD 0.00, 95% CI –0.22, 0.22, P = 1.0). Biochemical markers (IL-1, IL-2, IL-10, and ICAM) were reported in one nonrandomized trial only and were not significantly different between EVH and OVH (Table 4).

Major coronary adverse events (MACE) such as postoperative myocardial infarction, stroke, rethoracotomy for bleeding, reintervention for ischemia or angina recurrence, and all-cause mortality were not significantly different for EVH versus OVH at 30 days (Table 4) or up to 5 years.13 Clinical outcomes at 30 days were as follows: all-cause mortality (OR 0.71, 95% CI 0.34–1.48, P = 0.4), postoperative myocardial infarction (OR 1.02, 95% CI 0.58–1.78, P = 0.9), angina recurrence or reintervention for ischemia (OR 1.06, 95% CI 0.38–2.96, P = 0.9), and rethoracotomy for bleeding (OR 0.91, 95% CI 0.09–8.89, P = 0.9). One randomized trial reported survival free of MACE (death, myocardial infarction, heart failure, or angina recurrence) at up to 5 years, and found no significant difference (75% vs. 74%, P = 0.85).13

Angiographic outcomes were reported in only a few trials, generally at up to 3 to 6 months of follow-up, and were not significantly different (Fig. 3). Similarly, no difference was found for graft occlusion at 3 to 6 months (OR 1.30, 95% CI 0.79–2.15, P = 0.3) and stenosis greater than 50% after 3 months (OR 0.79, 95% CI 0.44–1.45, P = 0.5).

Back to Top | Article Outline

Pain and Mobility

The incidence of moderate to severe pain was reduced with EVH compared with OVH in the postoperative period by 74% (OR 0.26, 95% CI 0.12–0.55, P < 0.0001) and by 83% at up to 6 months of follow-up (OR 0.17, 95% CI 0.05–0.60, P = 0.006). The incidence of mobility disturbance at postoperative day 2 to 4 was reduced by 69% (OR 0.31, 95% CI 0.15–0.65, P = 0.002). Similarly, the incidence of neuralgia was reduced by 75% at 4 to 6 weeks, and by 78% at 3 to 6 months of follow-up (Table 4). A significant difference in neuralgia was no longer found at 12 months (1 trial). Visual analog score (VAS) for pain on postoperative day 0 or 1 was significantly reduced by more than 2 points with EVH versus OVH (WMD –2.18, 95% CI –3.56 to –0.79 points), and by less than 1 point at 4 to 6 weeks of follow-up (WMD –0.35, 95% CI –0.58, –0.12 points). In-hospital morphine equivalents use was significantly reduced by 0.14 mg/kg (WMD –0.14, 95% CI –0.20, –0.08 mg/kg). Duration of analgesia use was significantly reduced by a weighted mean of nearly 14 days (WMD –13.8 days, 95% CI, –21.22 to –6.38 days). The number of patients with restricted mobility on postoperative day 2 to 4 was significantly reduced (OR 0.31, 95% CI 0.15–0.65), and at day 30 (OR 0.06, 95% CI 0.008–0.49). Mean time to ambulation was not significantly different (Table 4).

Back to Top | Article Outline

Patient Satisfaction

Measures of patient satisfaction were infrequently reported, and therefore the resulting estimates are based on very few trials. Patient scores for subjective assessment of mobility were significantly improved for EVH compared with OVH at discharge (WMD 2.40 points, 95% CI 2.05 to 2.75). Patients’ satisfaction with pain experience during hospitalization was significantly improved (WMD 1.88 points, 95% CI 1.49–2.28), but was no longer significant at 6 weeks postdischarge (WMD 0.24 points, 95% CI –0.09 to 0.57). Patient satisfaction scores for cosmetic result was significantly higher with EVH compared with OVH at discharge (WMD 2.33 points, 95% CI 1.9–2.75) and at 6 weeks (WMD 0.93 points, 95% CI 0.58–1.27). The incidence of patient dissatisfaction with EVH overall surgical results compared with OVH was not significantly reduced (OR 0.005, 95% CI 0.00–14.6), but incidence of cosmetic dissatisfaction was significantly reduced with EVH compared with OVH (OR 0.24, 95% CI 0.08–0.71, P = 0.009).

Back to Top | Article Outline

Quality of Graft Harvested

Mean number of vein repairs was significantly increased in EVH compared with OVH (WMD 1.11, 95% CI 0.52–1.69, P = 0.0002). Clinical outcomes that may ultimately be an expression of the quality of graft include myocardial infarction, rethoracotomy for bleeding, angina recurrence or reintervention for ischemia, and death. Aggregate analysis and subanalysis of randomized and nonrandomized data showed no difference between EVH and OVH for myocardial infarction, rethoracotomy, angina recurrence or reintervention, and death over the short term (ie, at 4–6 weeks) (Table 4, Fig. 3). One trial reported clinical outcomes at 5 years, and showed no difference in 5-year event-free survival (freedom from myocardial infarction, angina recurrence, congestive heart failure, or death).13

Back to Top | Article Outline

Histology Studies

When vein segments were analyzed from small subsets of patients were observed by microscopy, they showed preservation of vascular integrity when defined by no difference in immunoperoxidase stains (vimentin, Factor VIII, and CD31) and no evidence of trauma of the saphenous vein wall shown on hematoxylin and eosin stains.33 Similarly, no difference in evidence of mechanical or thermal damage was found under stained light microscopy examination.32 Measurement of IL-1, IL-2, and IL-10 did not differ significantly in samples taken during EVH and OVH technique. Although ICAM levels were reduced, the clinical significance remains undefined.27–30 Aggregate analysis of studies that reported a 5-point endothelial damage score showed no significant difference between estimates of endothelial damage between EVH and OVH groups (SMD –0.09, –0.27 to 0.09).

Back to Top | Article Outline

Resource Utilization

The mean time required to harvest the graft was increased (WMD 11.23 minutes, 95% CI 7.12–15.34, P < 0.0001), mean closure time was significantly reduced (WMD –23.12 minutes; 95% CI –30.57, –15.66; P < 0.0001), and total operative time was significantly increased (WMD 15.26 minutes, 95% CI 0.01–30.51, P < 0.05) for EVH compared to OVH. The rate of vascular harvest was not significantly different with EVH and OVH (WMD –0.10 cm/min; 95% CI –0.34, 0.14; P = 0.4). Total blood loss was not reduced with EVH (–9.57 mL; 95% CI –56.55, 37.41; P = 0.7), and incidence of blood product transfusion was not reported.

Time to ambulation was not significantly different with EVH and OVH (–0.38 days; 95% CI –1.17, 0.41; P = 0.4). Total hospital length of stay was reduced significantly (WMD –0.85 days; 95% CI –1.55, –0.15; P = 0.02), but ICU length of stay was not reduced (WMD 0.07; 95% CI –0.20, 0.34; P = 0.6) with EVH.

Aggregate analysis of physician visits and emergency room visits showed significant reduction (OR 0.26, 95% CI 0.12–0.57, P = 0.001), and need for nursing or home care services was also significantly reduced (OR 0.17, 95% CI 0.08–0.38, P < 0.0001) with EVH compared with OVH. In addition, hospital readmissions for wound complications were significantly reduced (OR 0.53, 95% CI 0.29–0.98, P = 0.04) with EVH.

Although two studies that reported costs from the payer perspective in the United States suggested cost savings ranging from US $68 to $1,500 savings per patient undergoing EVH instead of OVH (Table 5), these cost estimates should be interpreted with caution due to variations in definitions of applicable costs (direct and indirect, in hospital and follow-up costs). A formal assessment of the incremental cost-effectiveness ratio would be necessary before conclusions about the relative cost-effectiveness of the two techniques could be made.

Back to Top | Article Outline

Study Quality and Heterogeneity

Subanalysis of randomized trials showed similar results as for the combined randomized and nonrandomized data (see Fig. 3). However, significant heterogeneity between trials was detected for some of the categories of wound complications. In most cases, for outcomes showing statistically significant heterogeneity, the excess heterogeneity was attributable to the nonrandomized cohort of studies. This emphasizes the need to rely first on highest levels of evidence to inform decision making.

Back to Top | Article Outline


This meta-analysis suggests that EVH results in improved clinical outcomes and may have favorable implications for resource utilization.

Back to Top | Article Outline

Clinical Outcome

Endoscopic harvest of saphenous vein results in reduced risk of wound complications compared with OVH. Specific subtypes of wound complications that were reduced included wound infections, antibiotics given, edema, seroma/lymphocele, necrosis, wound drainage, dehiscence, and surgical wound interventions. For each of these endpoints, the reduction in risk was not only statistically significant, but also likely to be considered clinically significant. Assuming that these results are applicable to patients outside of the clinical setting, these results predict that for every 1,000 patients (of similar characteristics to those included in this meta-analysis) who undergo EVH instead of OVH, there would be on average 91 fewer patients with at least 1 wound complication, 250 fewer experiencing edema, 71 fewer with wound drainage, 62 fewer with dehiscence, 53 fewer with wound infections, 48 fewer with necrosis, 40 fewer requiring antibiotics, and 31 fewer with seroma/lymphocele. In addition, average pain scores over the short and long term would decrease, incidence of neuralgia would decrease, and patient satisfaction would be improved if EVH were used rather than OVH.

Back to Top | Article Outline

Resource Outcomes

The results of this meta-analysis also suggest the potential for favorable impact on resource utilization, such that for every 1,000 patients treated with EVH instead of OVH, there would be approximately 850 fewer hospital days of stay, 100 fewer patients requiring physician or ER visits, 33 fewer patients requiring nurse or homecare visits, 22 fewer patients requiring surgical intervention for wound complications, and 7 fewer hospital admissions, and over 13,000 fewer days of analgesic use (mostly outpatient use). However, the mean harvest time and total operative time would be increased by over 10 minutes per patient, and overall operation time by about 15 minutes per patient.

Judgments about whether the costs of EVH are reasonable and feasible depend on the definition of cost-effectiveness and on the perspective taken (ie, the different perspectives of patients, physicians, hospitals, health care systems, society would result in differing definitions of relevant costs and outcomes). Because formal economic analyses have not been conducted for studies of EVH and OVH, there are limited data to base recommendations on the resource implications of the two techniques.

Back to Top | Article Outline

Strength and Limitations

The results of this meta-analysis should be interpreted in light of its strengths and limitations. Although this meta-analysis represents a comprehensive analysis of currently available randomized and nonrandomized evidence, it is possible that new evidence will emerge to better inform the longer-term outcomes, and this newer evidence will need to be incorporated over time. In addition, it is also important to note that the results of this analysis are technique- and skill-dependent, and the results described herein are no guarantee for outcomes produced by individual surgeons or physician assistants who may have less technical competence or experience than those represented within the trials. It is clear, from examination of Tables 1 and 2 that the included trials give mixed representation of surgeons (single or multiple) and physician's assistants who conducted the vein harvest, some of whom were very early in their experience, and others of whom had significant experience before conducting the trial. Because a number of the trials described their experience as early in their learning curve, it is likely safe to suggest that the results of this meta-analysis provide a conservative estimate of the potential benefit that would have been possible had only experienced operators been included in the trials.

A significant limitation of the results of this meta-analysis is that some of the prognostically significant characteristics of the patients differed significantly at baseline. These baseline differences were often attributable to the nonrandomized trials. This is of particular concern since it suggests that younger, male, obese patients were preferentially selected to undergo EVH rather than OVH, which could potentially result in bias. Fortunately, we have subanalyzed the randomized trials separately from the nonrandomized trials so that the results of the methodologically rigorous trials can be considered in isolation of the selection biases of the nonrandomized trials. When the randomized trials are considered separately from the nonrandomized trials, the conclusions remain unchanged for the majority of outcomes (Fig. 3).

Although endothelial damage scores and markers or endothelial injury were considered to inform the question of quality of conduit harvested, no guidelines and no agreement exists in the literature regarding the validity of these indices of quality of conduit harvested. In particular, there are no guidelines for appropriate preparation and valid measurement of markers of injury in vein segments, no consensus is available regarding which biochemical markers represent a valid marker of injury, and no empirical information is available to inform whether these markers (and at what threshold) represent clinically meaningful associations with clinical outcomes such as graft patency and MACE. Similar limitations exist for other indicators of endothelial injury, such as visual inspection of endothelial intimal integrity through electron microscopy. Therefore, these preliminary indicators of graft quality should be interpreted with caution. Rather, greater emphasis should be placed on the more clinically relevant outcomes of myocardial infarction, need for reintervention, heart failure, and survival over the longer term.

None of the clinically relevant cardiovascular outcomes or overall survival differed significantly for EVH compared with OVH; however, the confidence intervals remain wide since the power to rule out significant differences was relatively low, even with aggregate analysis of all trials through meta-analysis. Inadequate power also hindered the precision of outcome estimates for graft patency, since few studies reported angiographic outcomes, and of those that did, a significant proportion of patients did not undergo angiography during longer-term follow-up. Therefore, the angiographic results should be interpreted conservatively at this time, until further studies with adequate follow-up become available. Because measures of patient satisfaction were infrequently reported, the resulting should be interpreted with caution, and further research should be encouraged in this area.

Back to Top | Article Outline

A Consensus Conference and Statement on EVH Versus OVH

A consensus conference was facilitated by the International Society for Minimally Invasive Cardiac Surgery (ISMICS) to clarify, for clinical practitioners and health care planners or administrators, the suggested role of endoscopic vascular harvest relative to open vascular harvest in adults undergoing CABG surgery. The primary objective of the consensus conference was to provide the best available evidence on the question: “Does endoscopic vascular graft harvesting (EVH) compared to conventional open vascular harvesting (OVH) in adults undergoing coronary artery bypass surgery improve clinical and resource outcomes?”52 Furthermore, economic analysis from the perspective of the institution and the health care system were examined to determine whether the clinical benefits of EVH would be worthy of the required resource shifts. The ISMICS Consensus Conference suggested that “EVH is recommended to reduce wound related complications, improve patient satisfaction, and decrease postoperative pain, hospital LOS, and outpatient wound management resources when compared to OVH (Class I; Level A).” Based on quality of conduit harvested, either endoscopic or open vein harvest technique may be used (Class IIa; Level B). Based on major adverse cardiac events and angiographic patency at 6 months, either endoscopic or open vein harvest technique may be used (Class IIa; Level A). It also recommended that given these statements, EVH should be considered standard of care for patients who require saphenous vein grafts for coronary revascularization (Class I; Level B).

However, it should be cautioned that the very important issue of operator technical competency must be further explored and adequate training standards and technical competence should be considered before widespread implementation of EVH as a standard of care. In addition, future research should address long-term angiographic patency, MACE, and cost-effectiveness in patients undergoing saphenous vein harvest and in patients undergoing endoarterial harvest.

Back to Top | Article Outline


1. Goldsborough MA, Miller MH, Gibson J, et al. Prevalence of leg wound complications after coronary artery bypass grafting: Determination of risk factors. Am J Crit Care. 1999;8:153.
2. Allen KB, Griffith GL, Heimansohn DA, et al. Endoscopic versus traditional saphenous vein harvesting: A prospective, randomized trial. Ann Thorac Surg. 1998;66:26–32.
3. Bitondo JM, Daggett WM, Torchiana DF, et al. Endoscopic versus open saphenous vein harvest: A comparison of postoperative wound complications. Ann Thorac Surg. 2002;73:523–528.
4. Bonde P, Graham AN, MacGowan SW. Endoscopic vein harvest: Advantages and limitations. Ann Thorac Surg. 2004;77:2076–2082.
5. Athanasiou T, Aziz O, Skapinakis P, et al. Leg wound infection after coronary artery bypass grafting: A meta-analysis comparing minimally invasive versus conventional vein harvesting. Ann Thorac Surg. 2003;76:2141–2146.
6. Athanasiou T, Aziz O, Al-Ruzzeh S, et al. Are wound healing disturbances and length of hospital stay reduced with minimally invasive vein harvest? A meta-analysis. Eur J Cardiothorac Surg. 2004;26:1015–1026.
7. Alderson P, Green S, Higgins J. Cochrane Reviewers’ Handbook 4.2.1, (updated December 2003). Chichester, UK, John Wiley & Sons, Ltd, 2004.
8. Moher D, Cook DJ, Eastwood S, et al. Improving the quality of reports of meta-analyses of randomized controlled trials. The QUOROM statement. Quality of Reporting of Meta-analyses. Lancet. 1999;354:1896–900.
9. Altman DG. Confidence intervals for the number needed to treat. BMJ. 1998;317:1309–12.
10. Cates CJ. Simpson's paradox and calculation of number needed to treat from meta-analysis. BMC Med Res Methodol. 2002;2:1.
11. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ 2003;327:557–560.
12. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med. 2002;21:1539–1558.
13. Allen KB, Heimansohn DA, Robison RJ, et al. Influence of endoscopic versus traditional saphenectomy on event-free survival: Five-year follow-up of a prospective randomized trial. Heart Surg Forum. 2003;6:E143–E145.
14. Bonde P, Graham AN MacGowan SW. Endoscopic vein harvest: Early results of a prospective trial with open vein harvest. Heart Surg Forum. 2002;5:S378–S391.
15. Carpino PA, Khabbaz KR, Bojar RM, et al. Clinical benefits of endoscopic vein harvesting in patients with risk factors for saphenectomy wound infections undergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg. 2000;119:69–76.
16. Cisowski M, Wites M, Gerber W, et al. A Minimally invasive saphenous vein harvesting for coronary artery bypass grafting– comparison of three less invasive methods. Med Sci Monit. 2000;6:735–739.
17. Fabricius AM, Diegeler A, Doll N, et al. Minimally invasive saphenous vein harvesting techniques: Morphology and postoperative outcome. Ann Thorac Surg. 2000;70:473–8.
18. Folliguet TA, Le Bret E, Moneta A, et al. Endoscopic saphenous vein harvesting versus ’open’ technique. A prospective study. Eur J Cardiothorac Surg. 1998;13:662–666.
19. Hayward TZI, Hey LA, Newman LL, et al. Endoscopic versus open saphenous vein harvest: The effect on postoperative outcomes. Ann Thorac Surg. 1999;68:2107–2011.
20. Kiaii B, Moon BC, Massel D, et al. A prospective randomized trial of endoscopic versus conventional harvesting of the saphenous vein in coronary artery bypass surgery. J Thorac Cardiovasc Surg. 2002;123:204–212.
21. Perrault LP, Jeanmart H, Bilodeaul L, et al. Early quantitative coronary angiography of saphenous vein grafts for coronary artery bypass grafting harvested by means of open versus endoscopic saphenectomy: A prospective randomized trial. J Thorac Cardiovasc Surg. 2004;127:1402–1407.
22. Puskas JD, Wright CE, Miller PK, et al. A randomized trial of endoscopic versus open saphenous vein harvest in coronary bypass surgery. Ann Thorac Surg. 1999;68:1509–1512.
23. Lamm P, Juchem G, Milz S, et al. Continuous graft perfusion: optimizing the quality of saphenous vein grafts. Heart Surg Forum. 2002;5:S355–S361.
24. Schurr UP, Lachat ML, Reuthebuch O, et al. Endoscopic saphenous vein harvesting for CABG–A randomized prospective trial. Thorac Cardiovasc Surg. 2002;50:160–163.
25. Yun KL, Wu Y, Aharonian V, et al. Randomized trial of endoscopic versus open vein harvest for coronary artery bypass grafting: Six-month patency rates. J Thorac Cardiovasc Surg. 2005;129:496–503.
26. Allen KB, Heimansohn DA, Robison RJ, et al. Risk factors for leg wound complications following endoscopic versus traditional saphenous vein harvesting. Heart Surg Forum. 2000;3:325–330.
27. Alrawi S, Balaya F, Raju R, et al. Nitric oxide expression by interleukin-10 in the endoscopic and open methods of vein harvesting in coronary artery bypass surgery. JSLS. 2002;6:5–9.
28. Alrawi SJ, Samee M, Ramanathan R, et al. Intercellular and vascular cell adhesion molecule levels in endoscopic and open saphenous vein harvesting for coronary artery bypass surgery. Heart Surg Forum. 2000;3:241–245.
29. Alrawi SJ, Ramanathan R, Satheesan R, et al. Interleukin-1, Interleukin-2, and Interleukin-10 expression in different techniques of saphenous vein harvesting. Heart Surg Forum. 2001;42–47.
30. Alrawi SJ, Balaya F, Raju R, et al. A comparative study of endothelial cell injury during open and endoscopic saphenectomy: An electron microscopic evaluation. Heart Surg Forum. 2005;4:120–127.
31. Brandt CP, Greene GC, Pollard TR, et al. Review of efforts to decrease costly leg wound complications in the medicare population following coronary revascularization. Heart Surg Forum. 2003;6:258–263.
32. Coppoolse R, Rees W, Krech R, et al. Routine minimal invasive vein harvesting reduces postoperative morbidity in cardiac bypass procedures. Clinical report of 1400 patients. Eur J Cardiothorac Surg. 1999;16:S61–S66.
33. Crouch JD, O’Hair, DP, Keuler JP, et al. Open versus endoscopic saphenous vein harvesting: Wound complications and vein quality. Ann Thorac Surg. 1999;68:1513–1516.
34. Davis Z, Jacobs HK, Zhang M, et al. Endoscopic vein harvest for coronary artery bypass grafting: Technique and outcomes. J Thorac Cardiovasc Surg. 1998;116:228–235.
35. Felisky CD, Paull DL, Hille ME, et al. Endoscopic greater saphenous vein harvesting reduces the morbidity of coronary artery bypass surgery. Am J Surg. 2002;183:576–579.
36. Folliguet T, Sverac-Bastide R, Le Bret E, et al. Prelevement du greffon saphene intern pour pontage aorto-coronaire par methode video-chirurgie. [Harvesting of an internal saphenous graft for aorto-coronary bypass using video-surgery.] Arch Mal Coeur Vaiss, 1999;92:859–66.
37. Galbraith GF, Pica-Furey W. A retrospective comparative study of open and endoscopic saphenous vein harvesting. J Endovasc Ther. 2000;7:460–468.
38. Griffith GL, Allen KB, Waller BF, et al. Endoscopic and traditional saphenous vein harvest: A histological comparison. Ann Thorac Surg. 2000;69:520–523.
39. Horvath KD, Gray D, Benton L, et al. Operative outcomes of minimally invasive saphenous vein harvest. Am J Surg. 1998;175:391–395.
40. Isgro F, Weisse U, Voss B, et al. Minimally invasive saphenous vein harvesting: Is there an improvement of the results with the endoscopic approach? Eur J Cardiothorac Surg. 1999 16(Suppl 2):S58–S60.
41. Kan CD, Luo CY, Yang YJ. Endoscopic saphenous vein harvest decreases leg wound complication in coronary artery bypass grafting patients. J Card Surg. 1999;14:157–162.
42. Li CY, Lai ST, Yu TJ, et al. Endoscopic vein harvest for coronary artery bypass surgery. Zonghua Yi Xue Za Zhi (Taipei). 1998;61:276–280.
43. Li JY, Wang SS, Lin FY, et al. Video-assisted endoscopic saphenous vein harvesting for coronary artery bypass grafting. J Formos Med Assoc. 1998;97:819–825.
44. Lutz CW, Hillmann R, Lutter G, et al. Endoscopic vs. conventional vein harvesting: First results with a new, non-disposable system. Thorac Cardiovasc Surg. 2001;49:321–327.
45. Marty B, von Segesser, LK, Tozzi P, et al. Benefits of endoscopic vein harvesting. World J Surg. 2000;24:1104–1108.
46. Mayer DM, Rogers TE, Jessen ME, et al. Histologic evidence of the safety of endoscopic saphenous vein graft preparation. Ann Thorac Surg. 2000;70:487–491.
47. Morris RJ, Butler MT, Samuels LE. Minimally invasive saphenous vein harvesting. Ann Thorac Surg. 1998;66:1026–1028.
48. Pagni S, Ulfe EA, Montgomery WD, et al. Clinical experience with the video-assisted saphenectomy procedure for coronary bypass operations. Ann Thorac Surg. 1998;66:1626–1631.
49. Patel AN, Hebeler RF, Hamman BL, et al. Prospective analysis of endoscopic vein harvesting. Am J Surg. 2001;182:716–719.
50. Rodrigus IE, Stockman B, Amsel BJ, et al. Should we use video-assisted endoscopic vein harvesting as a standard technique? Heart Surg Forum. 2000;4:53–55.
51. Terrini A, Graffigna A, Martinelli L. Riduzione dell’invasività in corso di rivascolarizzazione chirurgica: il prelievo della vena grande safena con technica videoendoscopica. Ital Heart J Suppl. 2000;1:674–678.
52. Allen K, Cheng D, Cohn W, et al. Endoscopic vascular harvest in coronary artery bypass grafting surgery: A consensus statement of the International Society of Minimally Invasive Cardiothoracic Surgery (ISMICS) 2005. Innovations. 2005;1:51–60.

meta-analysis; endoscopic vascular harvest; coronary artery bypass surgery

© 2005 Lippincott Williams & Wilkins, Inc.