Comparison of Bypass with Endoscopically Harvested Internal Saphenous Vein versus Bypass with Surgically Harvested Internal Saphenous Vein for Lower Limb Arterial Disease

Clinical Research Comparison of Bypass with Endoscopically Harvested Internal Saphenous Vein versus Bypass with Surgically Harvested Internal Saphenous Vein for Lower Limb Arterial Disease  Javier Peinado Cebri an, Angel Flores Herrero, Christian Leonel Salgado Lopetegui, Marı´a Pilar Lamarca Mendoza, Ricardo Montoya Ching, Santiago Estebanez Seco, Jose Ignacio Leal Lorenzo, Jose Gil Sales, and Antonio Orgaz Perez-Grueso, Toledo, Spain

Background: Patients with lower limb arterial disease have a high risk for complications related with surgical wounds. The endoscopic extraction of the great saphenous vein (GSV) is a less invasive alternative to the conventional surgical extraction. Methods: A clinical and ultrasonographic follow-up was carried out on the lower limb bypass with GSV performed in our institution between years 2007 and 2012. Patients were selected for open or endoscopic harvesting depending on the surgeon assigned (endoscopic or open surgeon). Follow-up was performed at 1, 3, 6, and 12 months after surgery and annually thereafter. All the GSV endoscopic harvestings (GSVEH) were performed by the same surgeon. Data for primary, assisted, and secondary patency and amputation-free survival were analyzed. Anatomopathalogic analysis were performed on pares of samples of the same vein dissected surgically and endoscopically from the same patient. Results: Sixty bypass surgery has been performed on 60 patients (54 men and 6 women), 30 with GSVEH (50%), and 30 with GSV open harvesting (GSVOH). All patients were intervened for critical limb ischemia (Rutherford cathegory 4, 5, and 6). Significant differences were found between both groups for suture dehiscence (GSVEH 0%, GSVOH 20%, P ¼ 0.01) and infection (GSVEH 3%, GSVOH 30% P, 0.006). No significant differences were found between both groups regarding to primary patency, assisted primary patency, or amputation-free survival. An anatomopathologic comparison of segments of veins extracted surgically and endoscopically of the same patients did not show any significant differences. Conclusions: Although no statistically significant differences were found between GSVOH and GSVEH bypass for lower limb revascularization, there is a trend toward poorer patency rates for the endoscopic technique. GSVEH lowers the risks for infection and dehiscence of surgical wounds.

Conflict of Interest: None. Funding: None. Servicio de Angiologı´a y Cirugı´a Vascular, Hospital Virgen de la Salud Toledo, Toledo, Spain. Correspondence to: Javier Peinado Cebrian, MD, Servicio de Angiologı´a y Cirugı´a Vascular, Hospital Virgen de la Salud, Barber 30, Toledo 45003, Spain; E-mail: [email protected]

Ann Vasc Surg 2015; -: 1–10 http://dx.doi.org/10.1016/j.avsg.2015.05.007 Ó 2015 Elsevier Inc. All rights reserved. Manuscript received: July 14, 2014; manuscript accepted: May 5, 2015; published online: ---.

1

2 Peinado Cebri an et al.

INTRODUCTION The lower limb revascularization using the bypass technique with autologous great saphenous vein (GSV) was originally described by Kunlin in 1948.1 Since this first experience, the method has shown high patency rates on a long and midterm basis, and it has become the first choice for revascularization in a wide group of patients with lower limb arterial pathology.2 However, there is a high morbidity rate associated with the long incisions required for the extraction of the GSV. In an attempt to reduce this morbidity, some minimally invasive techniques have been developed. One example is the incorporation of the GSV endoscopic harvesting (GSVEH), initially described for coronary bypass surgery, where it has proved to decrease the rate of surgical wound complications and postoperative pain.3e5 However, there is a wide discrepancy in the literature regarding the comparability in terms of patency of the coronary bypass performed with endoscopically extracted GSV, against those with the open-harvested GSV (GSVOH).6,7 The application of the GSVEH in the lower limb revascularization vascular surgery has also proved to reduce the morbidity related to surgical wounds and to reduce the costs of surgery compared with those of the conventional dissection.8 However, there is a lack of evidence regarding the results of the endoscopic technique for lower limb revascularization in critical ischemia. We have performed a nonrandomized prospective study comparing the results of this technique against open GSV harvesting in terms of woundrelated complication rates (infection, dehiscence of suture, hemorrhage, hematoma). We have compared too patency rates and amputation-free survival between both techniques, as well as postoperative hospital stay. An anatomopathologic evaluation of the veins harvested either endoscopically and surgically has been performed to detect differences in tissue damage between the GSV harvesting procedures.

PATIENTS AND METHODS Patients All patients undergoing surgical lower limb revascularization with contralateral GSV bypass for critical lower limb ischemia (Rutherford stages 4, 5, and 6), intervened in our center between July 2007 and January 2011 where included for the analysis. Given that in case of available ipsilateral GSV the elective technique in our institution is the in situ configuration, only contralateral GSV grafts

Annals of Vascular Surgery

were considered for enrollment. The indication for lower limb revascularization surgery with contralateral GSV graft bypass was established in all patients based on the current clinical guidelines (Trans-AtlanticInter-Society Consensus II).2 The adequacy of the contralateral GSV to be used as a graft was established by preoperative ultrasonographic study. The ultrasound criteria for the exclusion of GSV were: diameter less than 3 mm in standing position, venous incompetence and/or signs of thrombophlebitis. All the GSVEH bypass were performed by the same senior surgeon, who was the only operator during endoscopic vein harvesting and who performed exclusively GSVEH during the study period. The patients who underwent GSVOH bypass were intervened by the remaining 3 senior surgeons from our service. All the open harvesting of GSV was performed by senior surgeons. Of 87 patients undergoing bypass revascularization with contralateral GSV in our center during the recruitment period, 60 were included for analysis. The remaining patients were excluded because of their negative to participate in the study. After signing an informed consent, a sequential sampling was conducted, assigning the first 30 patients to endoscopic technique, and the subsequent 30 patients to GSVOH. The study protocol was reviewed and approved by the investigation and ethical committee from the institution. Operative Method Open extraction of the GSV. After ultrasonographic location of the GSV and its branches, surgical dissection of the GSV was performed at the groin to control the saphenofemoral junction. Skipeskin incisions were performed, when needed, to ligate the collateral branches. Long skin incisions were avoided when possible, trying not to make longcontinuous incisions. All the veins harvested with open technique were inserted as a bypass graft in an inverted disposition. Tunneling was performed anatomically and using a tunneler in all cases. Endoscopic extraction of GSV. The Maquet Vasoview Hemopro endoscopic Vessel Harvesting System (Maquet Cardiovascular, LLC, Wayne, NJ) was used for endoscopic harvesting of the GSV in all cases. After ultrasonographic identification of the GSV in its immediately suprapopliteal portion, dissection proceeded through a 2-cm long skin incision. Through this incision a blunt dissection endoscopic device was introduced in a cephalad direction, followed by the inflation of the sealing balloon and the low pressure injection of CO2 in the tissue surrounding

Vol.

-,

No.

-, -

2015

Endoscopic versus open harvested saphenous vein for bypass in CLI 3

GSV, which facilitates the dissection and provides enough working space. After the complete blunt dissection of the GSV in its suprapopliteal portion, dissection of the infrapopliteal segment proceeded through the same incision. The dissector was then extracted and bipolar cauterization of all the collateral branches of the infrapopliteal and crural segments was performed. Once the required segment of GSV was free to be extracted, the transection of the proximal end and ligation of the remaining stump proceeded through an endoscopically guided 5-mm length inguinal incision. The distal end was dissected and transected surgically, through a 2-cm long incision. The extracted conduit was anastomosed to the receiving and donor vessels in inverted configuration previous anatomic tunneling using a surgical tunneler. At the end of the procedure, a drainage is left (Redon 6e8F) throughout the whole length of the dissection tunnel of the extracted GSV. Independent of the GSV harvesting method, the donor leg was bandaged with cotton and elastic bandage applying gentle pressure.

(defined as clinical signs of infection plus positive bacteriological cultures), presence of suture dehiscence of the GSV harvesting wound, presence of new trophic lesions, curation of previous trophic lesions, absolute systolic arterial pressure in distal arteries, mayor amputations suffered during surveillance, and need for bypass salvage interventions during surveillance. Given that there were not any wound with an underlying graft, we did not use any vascular surgery specific wound infection classification, but a general one, separating wound infections by severity as follows9:

Perioperative Anticoagulation and/or Antiaggregation Treatment

The ultrasound studies were performed in all cases by one of our vascular surgeons in our vascular laboratory homologated by the ‘‘Board for Noninvasive Vascular Diagnostic from The Spanish Society of Vascular Surgery.’’ In all cases, the Toshiba SSA770A Aplio 80 DU imaging system was used (Toshiba Medical Imaging, Tokio, Japan). The 7.5-MHz lineal probe model PLT-704AT was used in all explorations. All surveillance explorations were performed with the patient in a supine position. The insonation Doppler angle was 60 in all cases except on very rare cases in which this could not be reached because of complex anatomy. End diastolic velocity (EDV) and peak systolic velocity (PSV) in proximal and distal anastomosis and the central segment of the conduit were obtained. The whole conduit was explored to identify stenotic lesions, which were confirmed with PSV and EDV measurements at the site of maximum stenosis, and/or PSV at maximum stenosis/PSV proximal-to-stenosis ratio. All bypass with images of stenosis that showed PSV ratio >2.5 were classified as failing graft, and included on surgical waiting list for a bypass rescue intervention. Patients who underwent bypass salvage surgery were revised after 1, 6, and 12 months from the intervention, and yearly thereafter.

All patient under acetylsalicylic acid treatment continued medication preoperatively and postoperatively. Patients under clopidogrel were advised to retire treatment 7 days before surgery and reinitiated medication immediately after intervention. Patients under chronic anticoagulation for cardiac disease replaced oral anticoagulation for intravenous or low molecular weight heparin, to normalize International Normalized Ratio before the intervention. During the first 24 postoperative hours heparin was continued and oral anticoagulation reinitiated thereafter. In case of significant bleeding all anticoagulation therapies are discontinued in our institution until hemorrhage is controlled. All patients received intraoperative 100 UI/kg of intravenous heparin, after GSV was harvested and target arteries were dissected. We do not use routine revertion of intraoperative anticoagulation at the end of the procedure unless hemorrhagic complication or difficulties for hemostasis occurs. Follow-up Intervened patients underwent a clinical revision, physical examination and Dupplex ultrasound exploration the day before to hospital discharge, as well as 1, 6, and 12 months after the intervention, and annually thereafter. The parameters collected in the physical examination included: palpation of pulses, the presence of surgical wound infection

1. Incisional surgical site infection (SSI): limited to the incisional site. A. Superficial: affecting only the skin and subcutaneous tissue. B. Deep: affecting muscle and/or fascia. 2. Organ and/or space SSI: involving any part of the anatomy, other than the incision, opened or manipulated during the intervention.

Anatomopathologic Analysis When using endoscopic harvesting, it is necessary in some cases to dissect the distal portion of the vein

4 Peinado Cebri an et al.

surgically, because of the difficulties of working in a thin subcutaneous tissue and with a vein that many times becomes extra fascial in the infrapopliteal location. In all these cases, a segment of the surgically exposed and one of the endoscopically harvested GSV were reserved for anatomopathologic evaluation. The segments were submitted in buffered formalin, and embebbed in paraffin for microscope examination. An 1-mm cross sectional samples of each segment were stained with hemathoxilin and eosin. An immunohistochemical analysis of an endothelium-specific glycoprotein, CD34 was performed as well, to identify differences in the endothelial integrity between samples (Fig. 4). An anathomopathologist, blinded for the harvesting method, recorded the frequency of abnormalities in samples. These frequencies were compared between groups using Fisher’s exact test. Statistical Analysis Continuous variables were expressed as mean + standard deviation. The normality of the distributions was checked by the KolmogoroveSmirnov test. The categorical data were expressed as percentage and absolute frequencies, and were compared by using the exact Fisher’s test. The comparison of continuous variables between cohorts was carried out with the Student’s t test for independent samples. Patency was analyzed by means of survival test using KaplaneMeier tables. The log-rank test (Mentel Cox) was used to compare the survival between cohorts. The impact of the GSV processing method in the patency was analyzed by means of the Cox correlation test. Statistical significance has been defined as a bilateral P < 0.05. The statistical analysis was carried out using the 15.0 version of the SPSS computer software for Windows (SPSS Inc., Chicago, IL).

RESULTS Patient Demographics The prospective data registered from 60 patients who underwent lower limb revascularization surgery by means of GSV bypass performed in our center between March 2007 and November 2012 have been analyzed. A total of 54 patients were male. The average age was 68 ± 12 years. None of the patients chosen for GSVEH were reconverted to open technique. No significant statistical differences were found in the distribution of cardiovascular risk factors, but for hyperlipidemia (GSVEH 19 [63%] vs. GSVOH 11 [37%]; P, 0.02). There were

Annals of Vascular Surgery

no differences between cohorts in the distribution of chronic medical treatment (antihypertensives, anticoagulation, simple antiaggregation, double antiaggregation, and statins; Table I). GSV Diameter No statistically significant differences were found between both groups in the diameter of the GSV used (GSVEH 3.55 ± 0.55 mm, GSVOH 3.83 ± 0.59 mm, P ¼ 0.07; Table II). Indications and Type of Pathology No statistically significant differences were found between cohorts in the type of pathology for which surgery was indicated (Table II). All patients were intervened for chronic lower limb ischemia, of which 17 were in Rutherford stage 4, 39 in stage 5, and 4 in stage 6. None of the patients underwent surgery for intermittent claudication. No differences were found regarding the distal anastomosis territory: 38 bypass were performed to third popliteal portion (21 GSVOH vs. 17 GSVEH; P ¼ 0.2), 15 to tibioperoneal trunk (6 GSVOH vs. 9 GSVEH; P ¼ 0.28), and 7 bypass to anterior tibial artery (3 GSVOH vs. 4 GSVEH; P ¼ 0.5; Table II). Operation and Hospitalization Times The average total operating time was 254.63 + 66.75 min for those patients with GSVEH and 191.60 + 71.86 min for those patients operated by the open method (P < 0.01). Postoperative hospital stay was significantly shorter for those patients operated on by the endoscopic method (3.4 ± 1 days GSVEH vs. 4.6 ± 1.4 days GSVOH; P < 0.05). Systemic Perioperative Complications and Perioperative Mortality None of the patients showed signs of acute perioperative renal impairment or acute perioperative myocardial infarct. One patient in the GSVOH group died during hospitalization because of congestive heart failure. Two patients of the GSVEH group died during the perioperative period because of ischemic and hemorrhagic stroke, respectively (P ¼ 0.6). Surgical Wound-Related Complications Significant statistical differences were found regarding the mean global GSV wound-related complications (GSVOH, 16 vs. GSVEH, 4; P ¼ 0.017). These differences were mainly because of the

Vol.

-,

No.

-, -

2015

Endoscopic versus open harvested saphenous vein for bypass in CLI 5

Fig. 1. Primary patency. KaplaneMeier tables comparing GSVEH versus GEVOH.

number of suture dehiscences (defined as absence of apposition of the cutaneous edges with a clear continuity solution) in the GSV extraction wound (GSVOH, 6 vs. GSVEH, 0; P ¼ 0.01) and infection of the GSV harvesting wound (GSVOH, 9 vs. GSVEH, 1; P ¼ 0.06: Table III). All the wound infections detected were classified as superficial SSI. All of them received antibiotic treatment, without need for surgical intervention in any of the cases. Empiric intravenous antibiotic therapy was initiated in all patients with clinical infection until a microbiological culture was available to deescalate treatment. Reinterventions, Patency, and Limb Salvage Six patients required intervention in a short postoperative period (<30 days) through early occlusion (GSVEH, 3 vs. GSVOH, 3; P ¼ 1). All the bypass that showed early occlusion were in the infrapopliteal group. Two of the bypass early occluded in the GSVOH group, were rescued by surgical thrombectomy. The only case that was unable to be rescued required mayor limb amputation. In the group of patients with GSVEH, 1 of the conduits early occluded was rescued. Of those unable to be

rescued, 1 received major amputation of the intervened limb. The remaining patient died before amputation due to hemorrhagic stroke. Seven bypass required rescue surgery during follow-up (GSVOH, 3 vs. GSVEH, 4; P ¼ 0.69). In all of the cases, stenosis in the distal anastomosis was found during surveillance, in which simple angioplasty with paclitaxel eluting balloon was performed. With a mean follow-up of 31 months, the global primary patency rate has been 61 ± 8.3% at 31 months. The primary patency at 31 months of the GSVEH bypass has been 55 ± 11.7%. The primary patency of the GSVOH bypass at 31 months has been 70 ± 9.8%. No significant differences have been found regarding primary patency between the 2 groups (log rank, P ¼ 0.64; Fig. 1). The Cox regression study has not shown a statistically significant influence of the GSV processing method over the primary patency (P ¼ 0.77). The overall secondary patency at 31 months has been of 75 ± 7.3%. For the GSVEH bypass, the secondary has been 74 ± 12.3% at 31 months, and 77 ± 9.3% for the GSVOH bypass. No significant differences have been found between the 2 groups (log rank: P ¼ 0.68; Fig. 2). The Cox regression study has

6 Peinado Cebri an et al.

Annals of Vascular Surgery

Fig. 2. Secondary patency. KaplaneMeier tables comparing GSVEH versus GEVOH.

not shown a statistically significant influence of the GSV processing method over the secondary patency rate (P ¼ 0.91). The global amputation-free survival at 31 months has been 88.3 ± 5.6%. No statistically significant differences were found regarding the amputation-free survival between the 2 groups (82.6 + 9.3% GSVEH vs. 94.7 + 5.1% GSVOH, log rank; P ¼ 0.41: Fig. 3). The Cox regression study did not show a statistically significant influence of the GSV dissection method on the amputation-free survival rate (P ¼ 0.43). Of the 10 patients who underwent amputation during the follow-up (7 GSVEH vs. 3 GSVOH; P ¼ 0.3) in 3 cases the bypass remained patent at the time of amputation (GSVOH, 2 vs. GSVEH, 1) and required amputation because of complication of the ischemic lesions at the foot.

fragments of either the surgically, and the endoscopically dissected vein were sent for anatomopathologic analysis. These segments were analyzed by 1 anatomopathologist blinded for the GSV dissection method, to detect differences in the frequency of endothelial and vessel wall integrity abnormalities between both techniques. No differences were found in the frequency of alterations in the intima, media, or adventitial layers of the veins, with no evidence of significant damage in any of them. The immunohistochemical analysis of the endothelial-specific glycoprotein CD34 showed intense staining of the endothelial layer in the veins dissected by both techniques. In summary, there was no way to differentiate what kind of dissection was used for each vein segment in any of the pairs of samples.

Anatomopathologic Study

DISCUSSION

In the 10 patients of the GSVEH group who needed surgical dissection of the infrapopliteal GSV because of technical difficulties (extra fascial disposition, multiple bifurcations, thin subcutaneous tissue),

The GSVEH to be used as a conduit for bypass, has been developed over the past few years to diminish the high rate of complications associated with the surgical wounds needed for surgical harvesting.

Vol.

-,

No.

-, -

2015

Endoscopic versus open harvested saphenous vein for bypass in CLI 7

Fig. 3. Amputation-free survival. KaplaneMeier tables comparing GSVEH versus GEVOH.

Fig. 4. Immunohistochemical staining (CD34) of a pair a samples from the same patients of GSVOH versus GSVEH.

There is a wide heterogeneity in the results described in the literature for this method, both for patency rates and with reference to the reduction in the rate of complications compared with the open harvesting method. Moreover, most of studies published have been developed for coronary revascularization, not being the results extrapolable to the lower limb revascularization territory for several reasons: The need for longer conduits in the lower limb implies greater incisions, increasing the risk for complications of surgical wounds. Furthermore, the clinical evolution of such wounds in patients

with ischemic pathology of lower limb is not comparable to that developed in patients without peripheral arterial pathology. The first reference to the use of this method for revascularization of inferior limbs was published in 1996 by Lumsden et al.10 Wartman et al.11 have recently presented their results comparing the endoscopic method with the open method, having found differences solely in the amount of postoperative narcotics used, being these significantly lower in patients with GSVEH bypass. Previously, in 2004, Gazoni et al.,12 in an also retrospective analysis, found

8 Peinado Cebri an et al.

Annals of Vascular Surgery

Table I. Patients baseline characteristics and comorbidities GSVEH (n ¼ 30)

Variable

Age, years Sex Male Female Diabetes Hypertension Hyperlipidemia Former smoker Current smoker Alcohol Renal impairment Statins Antiaggregation Simple Double Anticoagulation Cilostazol Antihypertensive Mean body mass index

Table III. Surgery-related complications within 30 days

GSVOH (n ¼ 30)

65 ± 13

70 ± 11

27 3 19 22 19 7 6 1 3 20

(90) (10) (63) (73) (63) (23) (21) (3) (10) (67)

27 3 18 18 11 10 11 4 4 10

(90) (10) (60) (60) (37) (33) (37) (13) (13) (33)

8 5 2 1 20 28

(27) (17) (7) (3) (67)

11 4 4 3 15 28

(37) (13) (13) (10) (50)

P

0.12 1

0.79 0.35 0.02 0.39 0.1 0.08 0.92 0.23 0.43 0.72 0.19 0.35 0.34 1

Bold value indicates statistically significant results (P < 0.05). % in brackets.

Table II. Surgery-related variables Variable

GSV diameter (mm) Postoperative hospital stay (days) Operative time (min) CLI Target vessels Anterior tibial artery Third popliteal portion Tibioperoneal trunk

GSVEH (n ¼ 30)

GSVOH (n ¼ 30)

P

33.55 ± 0.55

3.82 ± 0.59

0.07

2.57 ± 1

2.63 ± 0.48

0.001

255 ± 67

192 ± 72

0.001

30

30

d 0.56

4

3

17

21

9

6

Bold values indicate statistically significant results (P < 0.05). CLI, critical limb ischemia.

significant differences in favor of the GSVEH in terms of complication rates, patency, and limb salvage. One of the main criticism against the endoscopic method has been the supposed increase of endothelial damage suffered by the vein during dissection, postulating this as the main cause of patency decrease in the series of conduits with negative

Variable a

Early bypass occlusion GSV wound hematomab GSV wound bleedingc GSV wound infection GSV wound dehiscence Mean global GSV wound-related complications

GSVEH (n ¼ 30)

GSVOH (n ¼ 30)

3 3 0 1 0 13%

3 1 0 9 6 56%

P

d 0.31 d <0.01 <0.01 <0.05

Bold values indicate statistically significant results (P < 0.05). a Early bypass occlusion: occlusion within 30 postoperative days. b Wound hematoma: hematoma at the GSV harvesting bed. c Wound bleeding: bleeding at the GSV wound that required reintervention or blood transfusion.

results for GSVEH.10,12 Nevertheless, several histologic studies have tried to find differences in the quality of endoscopically harvested vein against that harvested by the open method3,13,14 without conclusive results, except that published by Rousou et al.15 in 2009, who found structural and functional alterations, without correlations, however, with lower patency of the conduit. The absence, in our study, of significant differences between the endothelium of pairs of fragments of vein harvested by the open method and the endoscopical method in the same patient, support the qualitative equivalence, in histologic terms, of both types of veins. The consistency in the endoscopic harvesting method assured by the fact that all the veins in the GSVEH group have been processed by the same surgeon, together with the use of a most recent dissection harvesting system (Maquet Vasoview Hemopro endoscopic Vessel Harvesting System; Maquet Cardiovascular, LLC, Wayne NJ), with bipolar cauterization of collateral branches, have contributed to minimize vein trauma during harvesting. Although surgery time has been significantly greater in those patients intervened with the endoscopic method, this has been comparable to that used by other groups in similar studies 11. We must mention here that the learning curve considered for this technique is a minimum of 30 cases. Although the endoscopic surgeon in this series has an experience of more than 100 endoscopic abdominal and thoracic procedures, all the GSV endoscopic dissections analyzed in this study were performed during his learning curve, which could have influence the procedure time and other results. Statistically significant differences have been found on the overall complications rate related to

Vol.

-,

No.

-, -

2015

surgical wounds, secondary to a significant increase in the rate of infection and dehiscence of the surgical wound. This finding has had a direct influence on the shorter hospital postoperative stay found in the GSVEH group, mainly secondary to the awaiting for microbiological culture results and the need for intravenous antibiotics administration. Although an economical impact study has not been carried out in this work, the reduction in hospitalization time could explain the lower cost of this method against the open method found by Garcı´a Altes et al.8 in 2011. We have not found any significant statistical difference between both groups in terms of primary, assisted primary nor secondary patency. It is important to point out that 100% of the interventions were carried out in a limb salvage intention (Rutherford 4, 5, and 6), which implies a rate of critical limb ischemia greater than that gathered in similar studies.10e12,16 The rate of limb salvation has been comparable to that registered in the literature, with an overall amputation-free survival of 88%, without any significant statistical difference between the 2 groups. However, although not statistically significant, possibly because of lack of power of the study, we have been observed a trend toward greater patency rates for GSVOH bypass.

CRITICISMS Although the distribution of the variables has been homogeneous, maybe because of the sequential sampling used, the nonrandomized assignment of the subjects and the comparison of one ‘‘endoscopic’’ surgeon with 3 ‘‘open harvesting’’ surgeons is a possible source of selection bias that we must mention here. The relatively small patient population limits the value of our study, presenting a lack of power to detect statistically significant differences. This could explain the absence of significance regarding patency rates differences, despite the fact that a clear trend toward poorer patencies has been observed in the endoscopic cohort. Unfortunately, no data were collected regarding pain and time to ambulation. Such important variables in terms of quality of life could have provided significant data in favor of endoscopic harvesting. They should be considered for future research.

CONCLUSIONS The surgical wounds performed for the GSVEH present a significantly lower rate of infection and suture

Endoscopic versus open harvested saphenous vein for bypass in CLI 9

dehiscence, which could contribute to a lesser hospitalization time observed in this group of patients. Although not statistically significant, a greater primary patency rate has been observed for open harvesting method over endoscopic method. No statistically significant difference has been observed in terms of amputation-free survival between cohorts. The histologic quality of the conduit is not affected by the harvesting method, as long as it is carried out with the right materials and method, as suggested in the histologic analysis by the absence of differences between the pairs of samples obtained by different methods, on veins obtained from the same patient. Nevertheless, it is still deemed necessary to publish new randomized clinical trials to base these decisions and indications on greater scientific evidence.

REFERENCES 1. Kunlin J. [Long vein transplantation in treatment of ischemia caused by arteritis]. Rev Chir 1951;70:206e35. 2. Norgren L, Hiatt WR, Dormandy JA, et al. Inter-Society Consensus for the Management of Peripheral Arterial Disease (TASC II). J Vasc Surg 2007;45(Suppl S):S5e67. 3. 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;123:204e12. 4. Felisky CD, Paull DL, Hill ME, et al. Endoscopic greater saphenous vein harvesting reduces the morbidity of coronary artery bypass surgery. Am J Surg 2002;183:576e9. 5. 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;129:496e503. 6. Lopes RD, Hafley GE, Allen KB, et al. Endoscopic versus open vein-graft harvesting in coronary-artery bypass surgery. N Engl J Med 2009;361:235e44. 7. Dacey LJ, Braxton JH Jr, Kramer RS, et al. Long-term outcomes of endoscopic vein harvesting after coronary artery bypass grafting. Circulation 2011;123:147e53. 8. Garcia-Altes A, Peiro S. A systematic review of costeffectiveness evidence of endoscopic saphenous vein harvesting: is it efficient? Eur J Vasc Endovasc Surg 2011;41:831e6. 9. Horan TC, Gaynes RP, Martone WJ, et al. CDC definitions of nosocomial surgical site infections, 1992: a modification of CDC definitions of surgical wound infections. Infect Control Hosp Epidemiol 1992;13:606e8. 10. Lumsden AB, Eaves FF 3rd, Ofenloch JC, et al. Subcutaneous, video-assisted saphenous vein harvest: report of the first 30 cases. Cardiovasc Surg 1996;4:771e6. 11. Wartman SM, Woo K, Herscu G, et al. Endoscopic vein harvest for infrainguinal arterial bypass. J Vasc Surg 2013;57:1489e94. 12. Gazoni LM, Carty R, Skinner J, et al. Endoscopic versus open saphenous vein harvest for femoral to below the knee arterial bypass using saphenous vein graft. J Vasc Surg 2006;44:282e7. discussion 7e8. 13. Meyer DM, Rogers TE, Jessen ME, et al. Histologic evidence of the safety of endoscopic saphenous vein graft preparation. Ann Thorac Surg 2000;70:487e91.

10 Peinado Cebri an et al.

14. Bonde P, Graham A, MacGowan S. Endoscopic vein harvest: early results of a prospective trial with open vein harvest. The Heart Surg Forum 2002;5(Suppl 4):S378e91. 15. Rousou LJ, Taylor KB, Lu XG, et al. Saphenous vein conduits harvested by endoscopic technique exhibit

Annals of Vascular Surgery

structural and functional damage. Ann Thorac Surg 2009;87:62e70. 16. Julliard W, Katzen J, Nabozny M, et al. Long-term results of endoscopic versus open saphenous vein harvest for lower extremity bypass. Ann Vasc Surg 2011;25:101e7.