|Home | About | Journals | Submit | Contact Us | Français|
Open saphenous vein harvesting can be associated with wound complications, incision pain, prolonged convalescence, and poor cosmetic results. Endoscopic vein harvesting has been widely used for prevention of these problems. We compared outcomes of open and endoscopic vein harvesting for coronary artery bypass grafting at the Texas Heart Institute.
We retrospectively analyzed data from 1,573 consecutive coronary artery bypass procedures performed at our institution during a 20-month period. Each procedure included saphenectomy by endoscopic vein harvesting (n=588) performed by physician assistants, or by traditional open vein harvesting (n=985) performed by physicians or physician assistants. The primary outcome variable was the incidence of postoperative leg infections.
Both groups were similar in terms of preoperative risk factors. After surgery, leg wound infections were significantly less frequent in the endoscopic vein harvesting group (3/588, 0.5%) than in the open vein harvesting group (27/985, 2.7%; P <0.002). The most common organism involved in leg infections was Staphylococcus (20/30, 66%): S. aureus was present in 14 of 30 infections (47%). Open vein harvesting was the only significant independent risk factor for leg infection.
We conclude that endoscopic vein harvesting reduces leg wound infections, is safe and reliable, and should be the standard of care when venous conduits are required for coronary artery bypass grafting and vascular procedures. Although the transition from open to endoscopic vein harvesting can be challenging in institutions, it can be successful if operators receive adequate training in endoscopic technique and are supported by surgeons and staff.
Open saphenous vein harvesting (OVH) is often associated with wound complications,1 incision pain, prolonged convalescence, and poor cosmetic results. Endoscopic vein harvesting (EVH), a less invasive alternative to OVH, has been shown in several studies to be associated with a 64% to 83% lower incidence of wound complications.2–8 This reduction in risk is particularly great for diabetic and obese patients, who have a high baseline risk of wound infections.4,9 In addition, multiple clinical studies have shown that EVH reduces wound-related postoperative care time,10 postoperative pain,6 time to ambulation, length of hospital stay, and readmission rate5 without compromising vein graft patency.10–14
Herein, we compare the outcomes of EVH and OVH procedures at our institution and discuss the difficulties encountered when our EVH program began.
Patient Selection. We retrospectively analyzed data collected from 1,573 consecutive patients who had undergone coronary artery bypass grafting (either alone or in combination with other procedures) with use of saphenous vein grafts (SVGs) from January 2004 through August 2005. Saphenectomy was performed with either EVH (n=588) or OVH (n=985). In accordance with World Health Organization standards, patients were classified as obese if the body mass index exceeded 30 kg/m2.
Leg Wound Complications. Leg wound complications were classified either as leg infections or as noninfective wound healing disturbance. Postoperative leg infections, defined according to the national Centers for Disease Control and Prevention standards, had to have at least one of the following features: purulent drainage, isolated organisms, at least 1 sign or symptom of infection (such as pain, tenderness, swelling, redness, or heat), physician diagnosis, spontaneous dehiscence, or abscess. In addition, the infection had to occur within 30 days of surgery. Leg wounds were evaluated by physicians, intensive care unit nurses, clinic nurses, and independent infectious disease service nurses. The diagnosis of infection was made by a physician. Wounds were monitored for infection during home health nursing visits. No data were available on noninfective wound healing disturbances, so our study was limited to leg infections.
Surgical Techniques. For both EVH and OVH, patients were prepared with povidone-iodine solution; this included a circumferential preparation of the legs. Patients were placed in a supine position, with both legs externally rotated and elevated on rolled sheets that were placed under the thighs to avoid pressure on the peroneal nerve.
Each EVH procedure was performed by 1 of 3 physician assistants (PAs) using the Clearglide® Endoscopic Vessel Harvesting System (CardioVations, Ethicon, Inc., a Johnson & Johnson company; Somerville, NJ). The system consists of an endoscopic dissector, endoscopic retractor, pigtail loop dissector, and bipolar cauterizing device. Before the study period began, each PA had completed an industry-sponsored training event that included observation and cadaveric surgical experience.
In each EVH procedure, a 2- to 3-cm longitudinal incision was made medially above or below the knee. If the entire SV was needed, 2 separate incisions were made, 1 above and 1 below the knee. After initial identification of the SV, a plane of dissection along the anterior surface of the vein was established using the endoscopic dissector (CardioVations). Harvesting was initially directed proximally and then distally. Branches were divided using bipolar cautery. The vein was divided distally using clips, silk ties, or bipolar cautery, and a small stab incision was made over the chosen proximal division site. The vein was pulled through the stab incision and its proximal patency was maintained, while the vein was cannulated and prepared with heparinized blood or saline, and branches were ligated using titanium clips or 3–0 silk ties. Bacitracin antibiotic solution was used for tunnel irrigation in patients who had diabetes. The incisions were closed with absorbable subcutaneous and subcuticular sutures. Closed-system drainage and elastic bandage wrappings were used in patients who were expected to have extended cardiopulmonary bypass times or who had excessive tunnel bleeding.
Open vein harvesting was performed either by a surgeon or by a PA. Longitudinal incisions were made over the course of the SV, starting at the groin. The length of the incision depended on the amount of vein required. Most open harvests were performed using the bridge technique (with skip incisions). Once exposed, side branches were clipped or ligated with 3–0 silk ties. Irrigation with antibiotic solution was used in diabetic patients. The incisions were closed in layers, with absorbable subcutaneous and subcuticular sutures. Closed-system drainage and elastic bandage wrapping were used in some patients.
Statistical Analysis. Univariate analyses were performed to compare the OVH and EVH groups in terms of risk factors for postoperative wound infection. Continuous variables were analyzed with 2-sample t-tests, and discrete variables were analyzed with 2 tests. Multiple logistic regression analysis was used to look for independent risk factors for wound infection; the variables included vein harvesting method, age, sex, diabetes mellitus, hypertension, renal disease, obesity, peripheral vascular disease, New York Heart Association functional class, and tobacco use. Test results were considered significant when P <0.05.
The OVH and EVH groups were similar in terms of preoperative risk factors, which are presented in Table I. More distal anastomoses were performed in the EVH group (3.2 ± 1.0) than in the OVH group (2.9 ± 1.1; P <0.001), because some surgeons chose to use OVH when only a very short segment of SV was required (to minimize equipment costs and to enable the limited number of PAs to participate in concurrent procedures that required longer SVGs). The mean hospital length of stay was similar in the EVH group (12.6 ± 13.0 days) and the OVH group (12.9 ± 12.3 days).
Fewer postoperative leg wound infections occurred in the EVH group (3/588, 0.5%) than in the OVH group (27/985, 2.74%; P <0.002). Among leg infections, the most common organism isolated from culture was Staphylococcus (20/30, 66%). In particular, S. aureus was isolated in almost half of those cultured (14/30, 47%). In a multiple logistic regression, OVH was the only significant independent risk factor for leg infection (P <0.001).
Despite the proven benefits of EVH, many surgical and heart centers continue to harvest SV segments using open techniques. This resistance to change has been attributed to various concerns about EVH, including increased cost, potential vein-quality problems, increased technical difficulty, and longer harvesting times. We performed this study in order to compare EVH with OVH and to review the overall experience at our institution.
Reduction of Postoperative Infection. After SVG harvesting, common leg wound complications include leg wound infections and noninfective wound healing disturbances such as dehiscence, lymphedema, ecchymosis, seroma, and hematoma. Several studies have shown that female sex, obesity, diabetes, and peripheral vascular disease are risk factors for these leg wound complications.2,3,14 Allen and colleagues2,14 found that EVH modified this risk factor profile, in that sex and peripheral vascular disease did not affect leg wound complication rates in EVH patients. However, obese and diabetic patients were still at greater risk of developing leg wound complications after saphenectomy with EVH.
In our study population, the use of EVH in lieu of OVH substantially reduced the incidence of leg wound infections. In our analysis, OVH was the only independent predictor of postoperative leg wound infections; obesity, diabetes, and the other risk factors that we examined had no effect on infection rates. The results of this study suggest that EVH should be the standard vein harvesting method for cardiac and vascular procedures requiring the use of a greater SVG.
Transition from OVH to EVH. Endoscopic vein harvesting has been finding greater acceptance within the specialty of cardiovascular surgery for years. At the Texas Heart Institute (THI), EVH has been used since the 1990s15; however, it was not routinely accepted as the standard of care for patients requiring a greater SVG until 2003. During the transition, many challenges were encountered and overcome, including justifying increased equipment and personnel costs, changing surgeon and support staff preconceptions about EVH technology and results, and giving the staff sufficient time to gain proficiency in performing and supporting the procedure.
The transition from traditional OVH to EVH can be frustrating for everyone involved. Initially, the change may dramatically increase surgical times, as well as the amount and cost of surgical supplies used in each procedure. The transition was especially challenging at our institution, because the support staff was accustomed to performing procedures very efficiently, and the traditionally rapid vein harvesting times could not be matched by the PAs before they became fully proficient at EVH. Furthermore, the complexity of the EVH technology did not seem consonant with our founder's motto, “Modify, simplify, and apply.” However, as the staff became more comfortable in supporting EVH and found the outcomes to be favorable, attitudes toward this more complex, initially more time-consuming technology became more positive, and we began performing EVH more frequently (Fig. 1, with data extended to include all of 2005).
Vein Harvesting Time. Saphenectomy can take longer with EVH than with OVH. The 3 PAs involved in this study have estimated that their average harvesting times were 10 to 40 minutes, depending on the length of vein harvested. Harvesting times decreased dramatically as the PAs gained proficiency. Vein harvesting time depends on the individual PA's experience and the characteristics of both the patient and the vein. Although harvesting times may be longer with EVH, the associated incisions are smaller and closure times are much shorter, resulting in similar overall procedure times.
Vein Quality. While deciding whether to continue the EVH program at THI, the surgeons considered the effects that using EVH could have on various aspects of the coronary artery bypass procedure, including the quality of the harvested conduits. If EVH were to yield vein grafts of lower quality than those harvested with OVH, then maintaining the EVH program would not be worth the effort. During the initial learning period, SVGs harvested using EVH did require more repairs. However, as the PAs became more comfortable and consistent in performing EVH, vein quality improved dramatically; consequently, more surgeons were willing to permit EVH to be performed on their patients. Currently, most THI surgeons are of the opinion that EVH and OVH produce vein grafts of similar quality and that EVH is of great benefit to patients. Clinical studies have shown that EVH and OVH produce conduits with similar 6-month and 5-year patency rates.12,16
Physician Assistants as Resources for EVH. Because THI is a training institution, surgical fellows and residents spend 6 months to 2 years training within the institute. Although these clinicians are certainly capable of learning to perform EVH, their transience would make maintenance of an EVH-trained workforce difficult. For this reason, PAs, whose positions are permanent, were added to the THI surgical staff in September 2003. This added stability made the lengthy training in EVH worth the effort. Initially, patience was required of the surgeons to allow the PAs to gain proficiency with the new technique. However, once the skills were in place, rapid completion of procedures and efficient use of operating room time resumed.
Of the 3 PAs who performed EVH during the study period, none had previous experience with EVH, although all had been trained in its use. One PA had been a cardiac surgeon in a foreign country for many years and had harvested radial arteries with an endoscope. Another had many years of general surgery experience but none with EVH. The 3rd PA was a recent graduate with no cardiac surgery or EVH experience. Various educational and support strategies, including presurgical practice sessions with inanimate models and cadaver legs, proctored initial sessions, and immediate performance evaluation, were found to be helpful to PAs during their 1st few cases. According to our experience, the operators became proficient at EVH after performing 15 to 35 vein harvests.
Study Limitations. This study has some limitations. No data were available on noninfective wound healing disturbances such as dehiscence, lymphedema, ecchymosis, seroma, and hematoma; therefore, we could not include these in the present study, either as outcome variables or as potential predictors of hospital length of stay. Similarly, data pertaining to harvesting and wound closure times were not collected. These are disadvantages associated with retrospective study. However, the 3 PAs whose work was included in the study reported a dramatic decrease in harvesting times over the period studied. Despite the limitations of the study, the substantial difference in the rates of leg infection, as well as anecdotal reports of greater postoperative mobility and satisfaction by the patients, encourages further use of the endoscopic procedure.
Endoscopic vein harvesting is now the standard of care for most patients undergoing cardiovascular operations that require SVGs at THI. The adjustment period—including the training required for PAs to perform EVH and the effort to gain the acceptance of the surgical and support staff—was a challenge. However, the substantial benefits of EVH for our patients—fewer wound infections, improved cosmesis, and earlier ambulation—have made the transition well worth the effort.
Because of these benefits, we believe that EVH should be the standard of care in all cardiac or vascular procedures that require a venous conduit for bypass grafting. Our successful transition to the consistent use of EVH techniques should be reproducible in most institutions.
Stephen N. Palmer, PhD, ELS, provided editorial support.
Address for reprints: David A. Ott, MD, Texas Heart Institute, MC 3-258, P.O. Box 20345, Houston, TX 77225-0345. E-mail: ude.cmt.iht.traeh@ttoad