After coronary artery bypass grafting (CABG), many

Transcription

1 Endoscopic Versus Open Saphenous Vein Harvest: A Comparison of Postoperative Wound Complications Jerene M. Bitondo, PAC, Willard M. Daggett, MD, David F. Torchiana, MD, Cary W. Akins, MD, Alan D. Hilgenberg, MD, Gus J. Vlahakes, MD, Joren C. Madsen, MD, Thomas E. MacGillivray, MD, and Arvind K. Agnihotri, MD Division of Cardiac Surgery, Massachusetts General Hospital, Boston, Massachusetts Background. Wound complications associated with long incisions used to harvest the greater saphenous vein are well documented. Recent reports suggest that techniques of endoscopic vein harvest may result in decreased wound complications. A prospective, nonrandomized study was developed to compare outcomes of open versus endoscopic vein harvest procedures. Methods. There were 106 patients in the open vein harvest group, and 154 patients in the endoscopic vein harvest group. Patient characteristics and demographics were similar in both groups. Wound complications identified were dehiscence, drainage for greater than 2 weeks postoperatively, cellulitis, hematoma, and seroma/ lymphocele. Results. Wound complications were significantly less in the endoscopic vein harvest group (9 of 133, 6.8%) versus the open vein harvest group (26 of 92, 28.3%), p less than By multivariable analysis with logistic regression, the open vein harvest technique was the only risk factor for postoperative leg wound complication (relative risk 4.0). Conclusions. Endoscopic vein harvest offered improved patient outcomes in terms of wound healing compared with the open vein harvest technique. (Ann Thorac Surg 2002;73:523 8) 2002 by The Society of Thoracic Surgeons After coronary artery bypass grafting (CABG), many patients have anecdotally reported experiencing more discomfort from the leg incision used to harvest the saphenous vein than from the sternotomy. Numerous reports [1 3] have documented complications associated with the long, continuous incision traditionally used to harvest the greater saphenous vein (GSV). With the advent of endoscopic techniques [4 6] for harvesting the GSV, a decrease in incisional pain and complications such as skin necrosis, hematomas, and infections has been reported [7, 8]. To test this hypothesis, we designed and carried out a prospective, nonrandomized study of patients undergoing CABG to compare the outcomes of the traditional open technique versus the endoscopic technique for harvesting the GSV. Accepted for publication Sept 21, Address reprint requests to Dr Torchiana, Division of Cardiac Surgery, Massachusetts General Hospital, Bulfinch 119, 55 Fruit St, Boston, MA 02114; dtorchiana@partners.org. Material and Methods Patients From June 1998 through October 1998, two groups of patients were prospectively identified. Both groups consisted of patients undergoing CABG for whom a segment of GSV was selected as a conduit for bypass grafts. These groups included patients that may have had additional procedures along with a CABG as well as those undergoing reoperative procedures (Table 1). Emergent cases were excluded from both groups. The first group consisted of 106 patients in which the GSV was removed using the open vein harvesting (OVH) technique. The second group consisted of 154 patients in which the GSV was removed using the endoscopic vein harvesting (EVH) technique. Patients were placed in each group by either specific surgeon preference or availability of a physician assistant trained in EVH. Operative Procedure All patients in both groups were positioned, prepared, and draped in the standard fashion. The hips were externally rotated, and the lower extremities were flexed at the knee and secured at the ankle. Supports were placed under the thighs and calves. Perioperative care, including prophylactic antibiotic treatment, was the same for both groups of patients. In the OVH group, the vein harvest procedure was performed by a second-year or third-year general surgery resident supervised by a staff surgeon. An initial incision was made in the groin and continued until the length of vein needed was obtained. If the required length of vein involved going into the lower leg, the skin 2002 by The Society of Thoracic Surgeons /02/$22.00 Published by Elsevier Science Inc PII S (01)

2 524 BITONDO ET AL Ann Thorac Surg ENDOSCOPIC VERSUS OPEN VEIN HARVEST 2002;73:523 8 Table 1. Operative Procedure (n 106) (n 154) Primary CABG Reoperative CABG 10 8 CABG plus other procedures a a Four patients in the OVH group and 1 in the EVH were reoperative/ combined procedures. CABG coronary artery bypass grafting; EVH endoscopic vein harvest; OVH open vein harvest. incision was interrupted at the bend of the knee, and a second incision was begun at the level of the tibial plateau. After the vein was exposed to the desired length, the tributary branches on the patient side were occluded with metallic clips and the branch was divided. The two ends of the vein were then ligated and divided. The vein was then prepared with suture ligatures. Metal clips were applied based on surgeon preference. After hemostasis was obtained, the incision was irrigated with an antibiotic solution and closed immediately. The closure consisted of interrupted, absorbable sutures for the subcutaneous layer and a running 4-0 absorbable suture for the subcuticular layer. A closed suction drain was inserted in occasional patients if deemed necessary based on patient body habitus or tendency to bleed. Staples were added on selected patients based on surgeon preference or patient body habitus. At the conclusion of the procedure, a dry dressing was applied the length of the incision. The leg was wrapped with an elastic bandage for 2 days. For the EVH patients, the vein was harvested using the VasoView system (Guidant Cardiac and Vascular Surgery, Menlo Park, CA). This procedure was performed by one of four people at various levels of endoscopic training. The group consisted of three physician assistants and one cardiac surgeon. For this procedure, a 2-cm to 2.5-cm transverse incision was made just above the knee, and the vein was identified (Fig 1). With the aid of carbon dioxide (CO 2 ) insufflation, a dissection cannula was used to isolate the vein and its surrounding branches. Once isolation of the vein was achieved, a second endoscopic instrument was used through which a scissors connected to a bipolar cautery was inserted to cauterize and cut the tributary branches. A 1-cm to 1.5-cm incision was made in the groin to gain control of the proximal vein for ligation and division. If additional length of vein was required from the lower leg, the instruments were inserted into the initial incision, and the vein was followed distally using the same technique. An additional 1-cm incision was made in the lower leg to expose the end of the vein for ligation and division. The vein was removed in one continuous segment. The vein was then prepared with suture ligatures. Metal clips were applied based on surgeon preference After hemostasis was confirmed, the tunnel created by this technique was irrigated with antibiotic solution. The incisions were closed immediately after the vein was removed. Closure consisted of one or two interrupted, absorbable stitches and a running, 4-0 absorbable suture for the subcuticular layer. A drain was inserted in occasional patients if needed based on body habitus or tendency toward bleeding. At the end of the operation, a dry dressing was applied to the incisions. The leg was wrapped with an elastic bandage for 2 days. Follow-up Patient follow-up was obtained 6 to 8 weeks after surgery by a physician, physician assistant, or a registered nurse. Follow-up was done by direct examination, telephone conversation, or written letter requesting completion of a form that addressed wound healing, level of pain, and wound complications. Late follow-up was obtained in 92 patients in the OVH group (87%) and 133 in the EVH group (86%). For patients who could not be directly contacted, follow-up information was sought from treating physicians, which accounted for 12% of the total with follow-up in the OVH group and 17.3% in the EVH group. The remaining 13% to 14% of patients were not available for follow-up due to residence out of the state or country, early mortality, or loss of contact. Statistical Analysis Analysis was performed using the intention-to-treat principle. Continuous data were presented as the mean value SD, and qualitative data as a percentage. Unpaired Student s t tests and Mann-Whitney U test compared group means. Nonparametric comparisons were performed using 2 and Fisher s exact tests. A p value of 0.05 was considered significant. Multivariable logistic statistical analysis was performed using a stepwise methodology with StatView (SAS Institute Inc, Cary, NC). Confidence intervals are given as 1 SD. Fig 1. Photograph of leg incision identifying the vein. Left leg with the initial 2.5-cm transverse incision proximal to the bend of the knee. The greater saphenous vein is identified and isolated with a vessel loop. Results The patient characteristics for the entire series of patients, shown in Table 2, did not differ significantly between the two groups. Obesity was more than twice as

3 Ann Thorac Surg BITONDO ET AL 2002;73:523 8 ENDOSCOPIC VERSUS OPEN VEIN HARVEST 525 Table 2. Patient Characteristics Table 4. Complications (n 106) (n 154) (n 92) (n 133) No. % No. % p Value No. % No. % p Value Gender Male Age (years) Diabetes Obesity a PVD a Obesity BMI (body mass index) 32 in males and 34 in females. Dehiscence Drainage a Cellulitis Hematoma Infected lymphocele Seroma Total a Persistent drainage lasting greater than 2 weeks postoperatively. EVH endoscopic vein harvest; PVD peripheral vascular disease. OVH open vein harvest; EVH endoscopic vein harvest; OVH open vein harvest. common in the EVH group, but the difference failed to attain statistical significance. Table 3 documents the operative procedure for the two groups. Harvest time began when the initial incision was made and ended when the vein was removed from the leg. Harvest times and length of vein were similar in both groups. Mean closure time was minutes in the OVH group versus minutes in the endoscopic group (p ). Follow-up information was obtained for 92 patients in the OVH group and 133 patients in the EVH group, and no significant differences in the patient characteristics between the two groups of patients for whom follow-up was available were noted. There were 26 of 92 (28.3%) patients in the OVH group and 9 of 133 (6.8%) patients in the EVH group who required follow-up evaluation specifically for suspected complications associated with their leg incision (p ) (Table 4). Not all complaints required interventions; however, all sought additional medical advice from a member of our service or from their primary care provider. Dehiscence, defined as wound edge separation that required a minimum of wet to dry dressing changes, occurred in 7 patients (7.6%) in the OVH group. Five patients were treated with dressing changes and home health nursing visits; 1 required oral antibiotics. One patient was readmitted to the hospital and treated with Table 3. Vein Harvesting Procedure (n 106) (n 154) p Value Harvest time a min min 0.1 Closing time a min min Length of vein b 42.1 cm cm a Mean times and standard deviations. Harvest time is the actual time of vein harvest. Preparation of the vein, including ligation of branches, was performed in both groups after vein removal. b Mean length and standard deviations. EVH endoscopic vein harvest; OVH open vein harvest. dressing changes and oral antibiotics. Another patient required prolongation of hospital stay for dressing changes and intravenous antibiotics before undergoing a surgical revision of bilateral thigh incisions. In the EVH group, 1 person developed fat necrosis of a leg incision that required dressing changes and home health nursing follow-up. In the OVH group, 10 patients (10.9%) were seen for follow-up with persistent drainage from their leg incisions greater than 2 weeks postoperatively. One person in the EVH group also had persistent drainage. Two patients in the OVH group and 1 in the EVH group required dressing changes and oral antibiotics for cellulitis. One patient in the OVH group was referred for follow-up by his home health nurse because of a hematoma but required no intervention. Seromas comprised the majority of the EVH postoperative complications. Six patients were evaluated for a fluid collection at the incision just above the knee. Three of the 6 were treated with aspiration of the fluid and application of a compression dressing for 24 hours. No intervention was required in the other 3 patients, and all wounds healed per primum. In the OVH group, 4 of the 5 patients who developed a seroma had the fluid aspirated and a compression dressing applied for 24 hours. One patient in the OVH group required readmission for an infected lymphocele, which was treated by incision and drainage, dressing changes, and intravenous antibiotics. None of the patients in the EVH group required readmission. Nine patients in the EVH group were converted to an open approach. These patients remained in the EVH group due to the intention-to-treat principle as described above under statistical analysis. Eight of these patients were captured in follow-up, and none developed complications associated with their leg incisions. However, further analysis of these patients did not change statistical outcomes based on inclusion in the EVH group or change to the OVH group. A multivariable analysis with logistic regression was performed. Utilizing the patient variables shown in Table 5, open vein harvest was the only risk factor found to be

4 526 BITONDO ET AL Ann Thorac Surg ENDOSCOPIC VERSUS OPEN VEIN HARVEST 2002;73:523 8 Table 5. Patient Variables Potential Risk Factor Total Number of Patients Wound Complications No. % No. % RR CI p Value Harvest technique Open Endoscopic Smoker Age 71 years Female Renal failure PVD Diabetes HTN IABP Obesity Reoperative procedure Additional procedure CI confidence interval; HTN hypertension; IABP intraaortic balloon pump; obesity BMI (body mass index) 32 in males and 34 in females; PVD peripheral vascular disease; Renal failure required preoperative dialysis; RR relative risk. a significant predictor of postoperative leg wound complications (p ). A relative risk of 4.0 (CI 1.97 to 8.20) was attributed to the open vein harvest technique. The study was reviewed and approved by the Human Studies Committee of the Massachusetts General Hospital. Comment Differences in wound complication rates between the traditional open technique and the endoscopic technique for saphenous vein harvest during CABG have been reported by other authors [7, 8]. In this study, we observed a significant improvement in the outcomes of the leg incisions in the EVH group of patients. With the comorbid factors that afflict many patients with coronary artery disease, such as obesity, diabetes, and peripheral vascular disease, an endoscopic approach to vein harvesting is beneficial compared with the more traumatic open technique. With EVH, there is minimal disruption of the skin along with arguably less subcutaneous trauma, both of which aid improved wound healing (Fig 2). Patient perception of the technique is an important consideration when choosing this technique. The incision is all that the patient sees of the vein harvesting procedure once they begin their recuperation. Small incisions along with the decreased pain [9, 10] are often an im- Fig 2. Comparison photographs of leg incisions. (A) Incision made to harvest the greater saphenous vein from the groin to the proximal calf using the traditional open harvest technique. (B) Incision made to harvest the greater saphenous vein from the groin to midcalf using the endoscopic technique. Each photograph was taken on postoperative day 5.

5 Ann Thorac Surg BITONDO ET AL 2002;73:523 8 ENDOSCOPIC VERSUS OPEN VEIN HARVEST 527 provement over the expectation that many patients have before undergoing their procedure. Patients who have undergone a reoperative CABG frequently report an improvement in the healing of the leg incision through which their vein was harvested endoscopically compared with their first operation, in which the vein was harvested by the open technique. Even those who stated that they had no difficulty in the healing of their leg incision with their first CABG report that the endoscopic approach resulted in less pain and improved appearance of their wound. Endoscopic vein harvest techniques continue to evolve and improve, such that currently only one or two 5-mm stab incisions are used to obtain both the proximal and distal vein. A single sterile adhesive strip is all that is often required to approximate these incisions. Cost may be an important consideration when choosing an endoscopic approach to harvesting the GSV. Each endoscopic procedure has a definite added cost due to the expense of the disposable equipment required along with an initial investment in the nondisposable equipment (monitor, camera, light source, and CO 2 insufflator). However, overall savings based on improved wound healing and, therefore, fewer additional treatments may counterbalance the added cost of the equipment. Additional treatments for wound complications may range from dressing supplies and outpatient services to readmission and surgical revisions of the affected areas. We were unable to perform a valid actual cost comparison of the two procedures in our study because the patients were not randomized and, therefore, differences other than the leg procedure could impact the costs; and care was administered through other providers such as home health nurses and primary care clinicians where accurate costs could not be obtained. These factors address only the tangible costs. Quality of life for patients who may experience permanent scars in addition to delay in return to employment or an active lifestyle are not easily measured. Another consideration is the time it takes to learn and master this technique. Several different EVH systems are available [4 6, 11], each of which offers a unique approach. We have observed that it takes several cases to gain comfort and ease with the use of the equipment. An additional 10 to 20 cases may be necessary to feel confident with the procedure. The time it takes to become fully adept at this technique varies and is operator dependent. Of the 4 people harvesting the vein in the EVH group, 2 had performed five or fewer cases before the beginning of this study. Others have emphasized that as one becomes more experienced with the endoscopic procedure, the harvest time decreases [4, 6]. Increasing experience with EVH also yields a decreased conversion rate (those cases that were selected for EVH but in which the endoscopic procedure could not be completed, requiring conversion to open harvest). The EVH group contained 9 such patients. The majority of conversions occurred during an individual operator s first 25 cases. After this number of cases, the occurrence of conversion decreased dramatically. The two main reasons for conversion are inability to locate the vein and vein size. Certain landmarks are used to determine the typical course of the GSV. One is usually able to locate the GSV approximately 3 cm medial to the medial femoral epicondyle. The GSV usually lies in the groove of the sartorius muscle and the gracilis muscle. Once the landmarks are identified, a transverse incision is made just above the knee. On selected patients, typically the very obese, these landmarks may not be clear, and the use of vein mapping by preoperative ultrasound has been helpful. Allen and Shaar [12] have shown that intraoperative ultrasound can be used successfully to determine the location of the GSV in order to make the incision at the correct location. Doppler ultrasound may also be used to assist in documenting the path of the GSV along the leg if the operator creates increased flow during external compression of the vein while the probe is held over the vein proximally. The ease with which the GSV is located is an important part of the procedure. If there is much difficulty at this part of the procedure, there will likely be a greater area dissected and chance of increasing the incision size. Both of these factors may affect the rest of the case by impairing the seal needed to maintain proper CO 2 insufflation to continue with the dissection and harvest of the vein. It was felt that if a greater area of tissue was dissected to locate the GSV, there was an increased incidence of seromas. Dissection that is kept to a minimum by more experienced operators has led to a decrease in the frequency of seromas. The second most common reason for conversion is vein size. If the vein that is located appears to be of inadequate size, either too small or too large, a decision may be made to look for a more suitably sized vein in the other leg or to convert to an open technique beginning at the groin or ankle. The GSV in the lower leg may be preferred in some patients due to vein size or patient size. The lower leg vein may be more uniform in caliber and more comparable in size with the coronary arteries in some patients. In these cases, we begin at the knee and dissect the lower leg vein first. The EVH system we use offers sustained CO 2 insufflation, which keeps the working space open with a low gas flow (1 L/min) and low pressure (10 to 12 mm Hg). This technique allows freedom of both hands to perform the dissection, isolation of the vein and its branches, and the cauterization and dividing of the branches. Vitale and associates [13] have recently reported safety in the use of CO 2 during endoscopic vein harvesting; we use the same system but with a flow of CO 2 of only 1 L/min. We have not seen CO 2 embolization with this technique, as was suggested by Chavanon and associates [14]. In occasional cases, we have noted hypercapnea with pco 2 as high as 62, which was corrected by increasing patient ventilation. When we began in November 1997, we attempted to select patients who were thought to be ideal for using this system. Over the following 2 years, we have attempted to use EVH for all CABG patients who require a segment of GSV. Figure 3 demonstrates how our proportion of EVH cases has risen over the years.

6 528 BITONDO ET AL Ann Thorac Surg ENDOSCOPIC VERSUS OPEN VEIN HARVEST 2002;73:523 8 We thank Chi-Ming Chow, MDCM, MSc, and Yuchiao Chang, PhD, for their statistical assistance, Karen Lynch, BSN, for her assistance with creating the database, and Debra J. Skoniecki, MS, ANP, for her assistance with patient follow-up. We also wish to mention S. Kenneson, PA-C, and S. Wasserman, PA-C, for their technical assistance with endoscopic vein harvesting. Jerene M. Bitondo, PA-C, and Willard M. Daggett, MD, have become consultants for Guidant Corporation after the completion of this study. Neither the institution nor the individuals received financial support for the study. Health care professionals responsible for the evaluation of the incisions received no financial benefit. Fig 3. Percentage of cases in which the greater saphenous vein was harvested endoscopically. Individual quarters are shown on x-axis. Percentages of total coronary artery bypass graft (CABG) cases are shown on the y-axis. (EVH endoscopic vein harvest.) One aspect of this procedure that remains to be evaluated is the potential effect of EVH on long-term vein graft patency. Although this procedure is still in its relative infancy, there is no evidence that early graft closure has resulted. The mean number of repair stitches used during the time of this study was 1.7 for the OVH group and 1.8 for the EVH group. Several studies have evaluated the quality of the veins harvested endoscopically [8, 15, 16]. They reported no histologic differences between segments of vein harvested endoscopically versus open. In addition, we have not observed any change in the clinical course of patients followed after CABG with EVH. Patient and physician acceptance has led to increased request and use of this approach. The advantages of this technique are most evident in the morbidly obese and diabetic population, where a traditional long, continuous incision carries a greater risk of morbidity. Limitations of this study include lack of randomization, different cardiac surgical procedures in the two groups, which may influence general recovery, incomplete follow-up, and reliance on patient, not physician, evaluation of the incision. In conclusion, we have shown that the advantages of EVH as an alternative method to the open technique of saphenous vein harvesting are substantial. The inferences that we have drawn from this study are: (a) endoscopic compared with open vein harvesting offered improved patient outcomes in terms of wound healing; and (b) by logistic regression analysis, the only significant predictor of impaired wound healing was open vein harvesting. References 1. DeLaria GA, Hunter JA, Goldin MD, Serry C, Javid H, Najafi H. Leg wound complications associated with coronary revascularization. J Thorac Cardiovasc Surg 1981;81: Utley JR, Thomason ME, Wallace EJ, et al. Preoperative correlates of impaired wound healing after saphenous vein excision. J Thorac Cardiovasc Surg 1989;98: Wipke-Tevis DD, Stotts NA, Skov P, et al. Frequency, manifestations, and correlates of impaired healing of saphenous vein harvest incisions. Heart, Lung 1996;25: Crouch J, Keuler J, Kleinman K, et al. Endoscopic saphenous vein harvesting for coronary artery bypass grafting. Proceedings of the 6th World Congress of Endoscopic Surgery, Rome, Italy, June 1998: Kan CD, Luo CY, Yang YJ. Endoscopic saphenous vein harvest decreases leg wound complications in coronary artery bypass grafting. J Card Surg 1999;14: Carrizio GJ, Livesay JJ, Luy L. Endoscopic harvesting of the greater saphenous vein for aortocoronary bypass grafting. Tex Heart Inst J 1999;26: Allen KB, Griffith GL, Heimansohn DA, et al. Endoscopic versus traditional saphenous vein harvesting: a prospective randomized trial. Ann Thorac Surg 1998;66: Crouch JD, O Hair DP, Keuler JP, et al. Open verses endoscopic saphenous vein harvesting wound complications and vein quality. Ann Thorac Surg 1999;68: 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: Morris RJ, Butler MT, Samuels LE. Minimally invasive saphenous vein harvesting. Ann Thorac Surg 1998;66: Tevaearai HT, Mueller XM, von Segesser LK. Minimally invasive harvest of the saphenous vein for coronary artery bypass grafting. Ann Thorac Surg 1997;63:S Allen KB, Shaar CJ. Facile location of the saphenous vein during endoscopic vessel harvesting. Ann Thorac Surg 2000; 69: Vitali RM, Reddy RC, Molinaro PJ, et al. Hemodynamic effects of carbon dioxide insufflation during endoscopic vein harvesting. Ann Thorac Surg 2000;70: Chavanon O, Tremblay C, Delay D, et al. Carbon dioxide embolism during endoscopic saphenectomy for coronary artery bypass surgery. J Thorac Cardiovasc Surg 1999;118: Griffith GL, Allen KB, Waller BF, et al. Endoscopic and traditional saphenous vein harvest: a histologic comparison. Ann Thorac Surg 2000;69: Meyer DM, Rogers TE, Jessen ME, et al. Histologic evidence of the safety of endoscopic saphenous vein graft preparation. Ann Thorac Surg 2000;70: