---

Distal revascularization of lower extremities: a 13-year experience
(Portuguese PDF version)

Antonio Vieira de Mello¹, Cristina M. Torres dos Santos², Bruno Cals de Oliveira³, Renata Vieira de Mello³

1. Director of the Angiology and Vascular Surgery Clinic of Centro Médico Barrashopping - Rio de Janeiro.
2. Vice-Director of the Angiology and Vascular Surgery Clinic of Centro Médico Barrashopping - Rio de Janeiro.
3. Assistant physicians of the Angiology and Vascular Surgery Clinic of Centro Médico Barrashopping - Rio de Janeiro.

Correspondence:
Dr. Antonio Vieira de Mello
Av. das Américas, 4666/Gr. 324
Centro Médico Barrashopping
CEP 22649-900 - Rio de Janeiro - RJ
Tel.: +55 21 2430.9333/Fax: +55 21 2430.9332
E-mail: vieirademello@uol.com.br

---

ABSTRACT

Objectives: To present and discuss the experience, the results and the lessons learnt from the use of in situ saphenous vein graft technique during 13 years.

Methods: Between December 1986 and December 1999, 212 distal revascularizations with in situ saphenous vein graft were performed in order to correct atherosclerosis obliterans in the lower extremities of 186 patients. Forty-two revascularizations were excluded from the study: associated surgical procedures in the aortoiliac segment to improve the inflow, partial grafts ( in situ plus reversed saphenous vein grafts) and reoperations. One hundred and seven "pure" revascularizations with in situ saphenous vein were assessed. The patients were followed during 13 years.

Results: The immediate mortality rate was 6% and the late mortality rate was 70%. There was 31% of non-vascular complications and 39% of vascular complications. The study registered 21 cases (12%) of failing grafts that were submitted to a second surgery in late postoperative period. There was 8% of immediate thrombosis and 19.4% of late thrombosis. The analysis of patency (Kaplan-Meier curves) showed cumulative percentages of primary patency of 72%, 55% and 32% in the first, third and sixth year, respectively; and cumulative percentages of secondary patency of 82%, 77%, 67% and 48% in the first, third, sixth and ninth year, respectively. The difference between the cumulative percentages of primary and secondary patencies was statistically significant (P < 0.05).

Conclusions: Distal revascularization of the lower extremities with in situ saphenous vein offers satisfactory results of patency, when it is performed according to the criteria of surgical recommendation, arteriographic study and surgical technique with endothelial protection. The careful postoperative follow-up makes the recovery of failing grafts possible and improves the cumulative indexes of patency.

Key words: atherosclerosis; revascularization; saphenous vein
Palavras-chave: aterosclerose, revascularização, veia safena.

J Vasc Br 2002;1(3):181-92

---

INTRODUCTION

Revascularization of distal arteries of the lower extremities for the salvage of limbs with endstage atherosclerotic occlusive disease can still be considered one of the greatest challenges to reconstructive vascular surgery. The success of such revascularizations highly depends on the surgical technique, and its functional life is related to variable factors, such as functional status, diameter and length of the venous graft, blood supply and flow. Moreover, the low flow rates observed in the distal canals exacerbate the process of platelet reactivity and contribute to increasing the functional boundaries of biological grafts or artificial prostheses. Therefore, the autologous vein is the most recommended conduit for such revascularizations.¹,²

Several studies have shown that the autologous vein has a higher patency if compared to artificial grafts in infrainguinal revascularizations, even when regarding above-knee procedures.³,⁴,⁵ However, there is still a lack of in-depth prospective and randomized studies that allow for a definite decision regarding the best technique employing the autologous vein in distal revascularizations: reversed or in situ saphenous vein? Theoretically, the in situ saphenous vein offers obvious advantages in relation to the veins extracted from their beds: minimal injury caused by handling, less endothelial exposure to warm ischemia in nonphysiological solutions, hydrostatic dilations with controlled pressures, preservation of the vasa vasorum and of adventitious innervation, milder endothelial lesion assessed throughthe improved balance ratio prostacyclin/thromboxane A2, better machting between anastomoses and better hemodynamic performance since the graft becomes progressively thinner, which increases the flow rate.^6-12 However, the studies that have been published by the most acknowledged experts of the above mentioned techniques, Robert P. Leather, in situ saphenous vein,¹³ and Lloyd Taylor, reversed saphenous vein,¹⁴ presented similar results. In either case, evidence leads to the conclusion that the most important aspect for obtaining positive and long-lasting results with autologous vein revascularizations is the endothelial protection of the venous graft and of the donor and receptor arteries.⁹,^15-17

In this study we present our experience during a strict and detailed 13-year follow-up with the use of the in situ saphenous vein in distal revascularizations of the lower limbs.

PATIENTS AND METHODS

Between December 1986 and December 1999, we performed 212 below-knee (or tibial, or distal) revascularizations with in situ saphenous vein in 186 patients. Among all the procedures, five were performed on the same patient (bilateral revascularizations, at different stages) and 21 procedures were reoperations with the purpose of recovering failing grafts. The correction of lesions in the aortoiliac territory by means of tubular prosthesis or endovascular procedures in order to improve blood supply was necessary in six cases. In 15 cases, in situ saphenous vein graft received additional inverted segments from other autogenous veins, due to lesions caused by the valvulotome or due to defects of the vein itself (fibrosis). Such grafts (in situ saphenous vein plus reversed saphenous vein) are called primary partial grafts since they were composed in the first surgery of limb revascularization (Figure 1). When the additional segment of the reversed autologous vein occurs during reoperation, the graft is called secondary partial graft.

Figure 1 - Injured distal segment of in situ greater saphenous vein replaced in the same primary surgery with a reversed lesser saphenous vein segment (arrow), re-establishing the flow to the proximal anterior tibial artery. Example of primary partial in situ saphenous vein graft (in situ + reversed).

After subtracting 21 cases of reoperation, six simultaneous surgical procedures in the aortoiliac territory and 15 cases of primary partial grafts from the 212 procedures of distal revascularization, there were 170 "pure" procedures of distal revascularization with in situ saphenous vein, which will be analyzed and discussed in this study.

Data collected and the results of this study were organized and analyzed as Kaplan-Meier curves (life table), according to the guidelines of the ad hoc committee of the International Society of Cardiovascular Surgery and the North-American Chapter of Vascular Surgery, headed by Dr. Robert Rutherford, published in 1986 and reviewed and republished in 1997.¹⁸,¹⁹ The significance indices of the differences between the results were obtained through the chi-squared test and Wilcoxin Signed Rank Test.²⁰

Regarding etiology, among 186 patients submitted to surgery, 184 presented atherosclerosis obliterans (99%) and, two cases (1%) presented the etiology of popliteal entrapment syndrome. Male patients represented 56.4% (105 patients) of the sample and 43.6% were female patients (81 patients). The subjects' age ranged from 39 to 89 years, with a mean of 64 years.

The surgical treatment was recommended with the aim of limb salvage in 95% of the 170 procedures performed. Among those cases, 143 (84%) were cases of necrosis and 18 (11%) consisted of untreatable rest pain. Only nine patients (5%) were submitted to surgery in order to solve selected problems of seriously incapacitating claudication for their lifestyle.

The main comorbidities associated with atherosclerotic disease were diabetes, arterial hypertension, smoking habit, heart diseases, chronic obstructive pulmonary disease (COPD), previous stroke and chronic renal insufficiency (Table 1).

Table 1 - Comorbidities in 186 patients

Disease	n. of patients	%
Diabetes	144	61
Hypertension	112	60
Smoking	102	59
Coronary disease/myocardiopathy	98	53
Chronic obstructive pulmonary disease	28	15
Previous stroke	17	9
Chronic renal insufficiency	2	1

Before the surgery, all patients were submitted to complete arteriographic studies, from the infrarenal abdominal aorta to the plantar arches. Most tests were performed by our team, using a customized technique, by means of percutaneous puncture of the abdominal aorta, with dos Santos needle, and of the common femoral artery, with the use of Cournand needle, due to the conditions available in our service at that time (Hospital Geral de Jacarepaguá - SUS-RJ). Since 1996, all arteriographic exams have been performed with the use of a catheter at accredited services of vascular radiology.

The surgeries were performed by means of continuous incision and complete exposure of the internal saphenous vein. Its route was previously drawn on the skin and infiltrated with papaverine solution in order to avoid spasmodic injury to the endothelium, according to the technique recommended by LoGerfo et al..¹⁶ The closure of the continuous incisions was always performed by means of mechanical suture of the skin, which was removed after three weeks, usually at outpatient services. Delicate surgical equipment were used to avoid traumatic clamping of the vessels, such as Mills and Rothon dissection clamps and Fogarty clamps, with silicone handles, Yasargyl's clips and Castroviejo needle holders and microsurgical scissors. For the ligation of venous collaterals we always use Ligaclips with the purpose of avoiding stenosis due to ligation by means of threads.

Devalvulations were performed with Mills-Leather valvulotomes, antegrade and retrograde models, always making sure that the vein was being dilated with colloidal solution (Dextran 70) and with controlled pressure of at most 300 mmHg, before introducing the valvulotomes in order to reduce contact with the venous endothelium.

We usually use efficient illumination with coaxial fiber-optic Frontolux and magnification with personalized 3x magnifying loupe. Another mandatory routine procedure is to check the quality of the surgery through a Doppler ultrasound and completion arteriography after the last anastomosis, by means of proximal puncture of the venous graft with a Jelco 22 needle.

The donor arteries (where the proximal anastomoses were performed) and the receptor arteries (which received the distal anastomoses) are shown in Table 2.

Table 2 - Donor and receptor arteries in 170 procedures

Donor artery	n. of cases	%
Common femoral	27	16.0
Superficial femoral	83	49.0
Deep femoral	40	23.5
Proximal popliteal	2	1.0
Distal popliteal	18	10.5
Receptor artery	n. of cases	%
Fibular	77	45
Posterior tibial	45	26
Anterior tibial	23	14
Dorsalis pedis	17	10
Tibiofibular trunk	5	3
Lateral plantar	3	2

Associated surgical procedures were performed and they are presented in Table 3. It is important to emphasize that more than half of the cases submitted to surgery (64.7%) needed minor and major debridements of the feet, which were always performed during the same surgery, after the revascularizations were accomplished.

Table 3 - Associated surgeries in 70 cases

Procedure	n. of cases	%
Intraluminal dilation	7	4.0
Endarterectomy of receptor artery	18	10.5
Debridement/minor amputation	74	43.5
Amputation of forefoot	36	21.0
Total	135	79.0

The patients submitted to surgery were followed as outpatients through visits every three months, during the first year after the surgery, and twice a year, one year after the surgery; even though, 25 (15%) patients were lost to follow-up. The shortest follow-up period lasted for three months and the longest follow-up lasted for 156 months (13 years), with a mean of six years. The in situ saphenous grafts are easy to monitor during the postoperative period due to their superficial localization, which allows for pulse palpation and auscultation with ultrasound Doppler along the whole subcutaneous route. Up to 1996, we had only a portable ultrasound Doppler in order to measure the pressure rates and a bidirectional ultrasound Doppler, with graphic recording, to perform the assessment of outpatients after surgery. We know that the pressure rates with the Doppler ultrasound are only able to detect critical stenosis and the color duplex scan is more sensitive to diagnosing early stenosis of the affected grafts and arteries,²¹ but the duplex scan became available only after 1996.

Our current protocol of postoperative follow-up of patients submitted to revascularization of the lower limbs consists of permanent use of a platelet antiaggregator (100 mg of aspirin/day); outpatient reviews every three months during the first year after the surgery and every six months thereafter; symptomatology assessment, physical examination, accurate palpation of peripheral pulses and of the venous graft, and assessment of trophic lesions (whenever they are present); arterial pressure measurements and calculation of the ankle/brachial index with ultrasound Doppler; and assessment, through color duplex scan of the limb submitted to surgery, with measurement of the speed spectrum collected at the donor and receptor arteries and at seven different points in the route of the graft. In addition, the following criteria of the ad hoc committee, mentioned above,¹⁸,¹⁹ are used to diagnose failing grafts:

1- recurrence of symptoms;

2- reduction in the amplitude or disappearance of previously palpable pulses in the native arteries or in the graft (always in subcutaneous position);

3- reduction of the ankle/brachial pressure index with ultrasound Doppler of 0.15 or higher, regarding the best rate observed in the immediate postoperative period;

4- reduction of the graft diameter above 75% in the assessment with color duplex scan - velocity rate (VR) above 3.5 and peak systolic velocity (PSV) above 300 cm/sec. All patients presenting such a condition are submitted to arteriographic study and surgical review in order to correct the failing graft. We adopted the definition by Frank Veith et al.²² for failing graft: grafts that present severe hemodynamic alteration before total occlusion. To be classified as failing, the graft must be pervious in most part of its extension, either with severe subocclusive stenosis, or despite the occlusion in proximal or distal segment, being clearly patent in its remaining extension. When the failing graft, without occlusion, is reoperated and recovered, it is included in the actuarial analysis (life table) as assisted primary patency (recovery before the occlusion). When reoperation is performed in order to recover the partially occluded graft, it is included in the statistics as secondary patency (recovery after partial occlusion). When the venous graft thromboses completely (along its whole extension), it is considered a failed, thrombosed or occluded graft, without chances for recovery and should be excluded from the life-table analysis. Revascularization, in these cases, might only occur through another surgery, a redo surgery.¹⁸,¹⁹,²²

RESULTS

Early mortality, during the first 30 days, affected 10 patients (6%) and late mortality, after 30 days, occurred in 119 cases (70%). These data are coherent with the fact that patients that need distal bypasses represent a group of subjects often affected by systemic wasting of the atherosclerotic disease. Such patients belong to older age groups, present less favorable arterial anatomies and, usually, have severe associated comorbidities (Table 1), which explains the high rates of early and late mortality, whose causes are listed in Tables 4 and 5.

Table 4 - Causes of immediate death in 170 cases

Cause	n. of cases	%
Multiple organ failure	3	2.0
Acute myocardial infarction	2	1.0
Pulmonary embolism	2	1.0
Congestive heart failure	1	0.5
Graft rupture ("explosion")	1	0.5
Anesthetic accident in debridement	1	0.5
Total	10	6.0

Table 5 - Causes of late death in 170 cases

Cause	n. of cases	%
Acute myocardial infarction	48	28.0
Unknown causes	30	17.5
Cancer	18	10.5
Stroke	15	9.0
Respiratory insulfficiency	3	1.7
Complicated diabetes	3	1.7
Digestive hemorrhage	1	0.5
Ruptured cerebral aneursmy	1	0.5
Total	119	70.0

There were 53 (31%) nonvascular complications (they did not occur in grafts or in anastomoses). Most of them (40 cases or 23.5%) were caused by the type of continuous incision employed, which was mandatory for the safe handling of Mills-Leather valvulotome. However, in only four cases (2.3%) of extensive necroses of the skin such incisions were directly responsible for the occlusion of the respective grafts (Table 6).

Table 6 - Nonvascular and vascular complications in 170 cases

Nonvascular complications	n. of cases	%
Localized skin necrosis	16	9.0
Amputation with patent graft	10	6.0
Lymphocele	7	4.0
Cellulitis	6	3.5
Subcutaneous hematoma	6	3.5
Extensive skin necrosis	3	2.0
Localized subcutaneous infection	3	2.0
Extensive subcutaneous infection	1	0.5
Extensive necrotizing erysipelas(fatal)	1	0.5
Total	53	31.0
Vascular complications
Thrombosis	47	28.0
Failing grafts	21	12.0
Residual valve	4	2.0
Varicose vein rupture	2	1.0
Rupture due to infection	1	0.5
Severe arteriovenous fistula	1	0.5
Total	67	39.0

There were 67 (39%) vascular complications, which were located in the grafts or in the anastomoses (Table 6). There were three cases (1.5%) of graft rupture, with severe hemorrhage, all of them during the early postoperative period. One of them was caused by infection; another one was due to varicose vein rupture, and the other one due to venous blow out. Only the case of varicose vein rupture was recovered by means of raphes, in three episodes of hemorrhage due to varicose vein rupture, on the 9th, 15th and 21st days after surgery. In the 4th month after the surgery, the graft became totally occluded, and the patient underwent below-knee amputation. The infection case was solved through ligation of the exposed and infected segment of the in situ vein and through the inclusion of a shunt with reversed vein bypass. The case of blow out of a proximal segment of the in situ graft led to hemorrhagic shock and death of the patient, even though the patient received quick assistance, still at the ICU, on the first day after the surgery. The internal saphenous vein used presented normal appearance, and the histological exam of the ruptured segment, just below the proximal anastomosis, at the superficial femoral artery, did not present anomalies. It seems that the rupture occurred only because the vein did not tolerate the patient's arterial hypertension, which led to its explosion at a weak point. The 83-year-old patient could not resist the devastating hemorrhage. There was only one case (0.5%) of severe arteriovenous fistula, which caused massive and growing edema of the revascularized lower limb. The fistula was ligated in the fourth month after the surgery, with local anesthesia and at the outpatient service, since it was easily located by means of palpation of the typical fremitus and due to the subcutaneous position of the in situ saphenous vein; the edema has completely resolved. There was an incidence of four cases (2.3%) of residual venous valves, and three of them produced unrecoverable graft thrombosis, in the early postoperative period on the 7th, 12th and 21st day after the surgery, respectively. Only one case could be recovered through reoperation, by means of resection of the residual valve and repair of the autologous vein, in the 4th month after the surgery (Figure 2).

Figure 2 - Postoperative arteriography of in situ failing saphenous bypass graft due to thrombus formation in residual valve (not fully lysed). The patient presented recurrent symptoms and purple toe syndrome caused by microembolism.

The incidence of failing grafts was of 21 cases (12%), all of them were reoperated in late postoperative period, as shown in Table 7. Immediate thrombosis occurred in 14 cases (8%) and 33 (19.4%) cases presented late thrombosis, totaling 47 cases (27.6%). Among the cases of late thrombosis, 16 (48.4%) took place during the first year after the surgery.

Table 7 - Operated and recovered failing grafts in170 cases

Type of failure	n. of cases	%
Occlusion of the distal mobilized segment	10	6.0
Occlusion of donor artery	3	1.7
Intimal hyperplasia of proximal anastomosis	2	1.0
Stenosis of receptor artery	2	1.0
Intimal hyperplasia of distal anastomosis	1	0.5
Stenosis of the medial segment of the graft	1	0.5
Anastomotic aneurysm (distal anastomosis)	1	0.5
Aneurysm of the common femoral artery	1	0.5
Total	21	12.0

The recovery of limbs with failing grafts was possible in 80% of the cases, with a percentage of major amputation of 20%. The recovery of limbs with thrombosed grafts was possible only in 45% of the cases, with a rate of major amputation of 55%. (Table 8). Note that when there was total thrombosis of the grafts, reoperations were performed as an attempt to save the limbs with acute occlusions, by implementing new grafts, that is by redoing the surgery. We were not able to recover totally thrombosed grafts.

Table 8 - Percentage of recovery of failing and thrombosed grafts

	Failure (n = 21)	Thrombosis (n = 47)
Recovery	80%	45%
Amputation	20%	55%

Through the analysis of the data mentioned above, which were obtained from strict outpatient follow-up of the patients submitted to below-knee revascularizations, we were able to design actuarial analysis with Kaplan-Meier curves (life tables). In the analyses, the cumulative percentage of primary patencies were compared to the cumulative percentage of secondary patencies, which were the result of the inclusion in the statistics of reoperated failing grafts that were recovered before their total occlusion (secondary patency). Such actuarial analyses showed cumulative percentage of primary patency of 72% in the first year, 55% in the third year and 32% after the sixth year, when the standard deviation exceeds 10% and the statistical indices are not reliable, certainly due to the significant 15% loss to patient follow-up (Figure 3). The cumulative percentages of secondary patency were 82% in the first year, 77% in the third year, 67% in the sixth year and 48% in the ninth year, up to this moment, with a standard deviation of 9.1%. After the ninth year, such standard deviation is higher than 10% and the statistical data of secondary patency are not reliable, due to the 15% loss to follow-up and to the small amount of patients still at risk (Figure 4). The difference between the cumulative percentages of primary and secondary patencies was statistically significant (P < 0.05).

Figure 3 - Rate of primary cumulative patency in tibial shunts with in situ saphenous graft (n = 170), in a 13-year period (December 1986 to December 1999). The numbers above the curve stand for grafts at risk and those below the curve represent the cumulative rate of patency in respective years. The dotted line represents an SD greater 10% and unreliable statistical data.

SD = standard deviation.

Figure 4 - Rate of secondary cumulative patency (assisted with reoperations) in tibial shunts with in situ saphenous grafts (n = 170), in a 13-year period (December 1986 to December 1999). The numbers above the curve stand for grafts at risk and those below the curve represent the cumulative rate of patency in respective years. The dotted line represents an SD greater 10% and unreliable statistical data.

SD = standard deviation

DISCUSSION

Regarding the Brazilian context, a reasonable experience with the use of both reversed and in situ techniques has been acquired. Our initial preference, up to December 1986, was the use of reversed saphenous vein as the first choice in infrainguinal revascularizations. In a seven-year period (from 1979 to 1993), our rate of primary patency in 131 infrainguinal revascularizations with reversed saphenous vein was 40%.²³ After the publications by the group from Albany, United States, in 1979,²⁴ the in situ saphenous vein technique was shown to the world with excellent results, which were reproduced in several medical services.^25-28 From December 1986, we started using this technique as the first choice for lower limb revascularizations. We still use the reversed autologous vein, but only if the in situ technique cannot be used. During the same time period of seven years (from 1986 to 1993), our rate of primary cumulative patency in 146 patients by using the in situ saphenous vein for infrainguinal revascularizations was 59% and the secondary patency was 72%.² The adoption of the in situ saphenous vein technique for infrainguinal revascularizations was remarkable, since, after December 1986, we started using extremely delicate techniques and materials. Consequently, we learnt how to carefully protect venous and arterial endothelia, by avoiding touching or damaging them with the instruments, and we employed this special in situ surgical technique in the reversed saphenous vein: vein removal by means of continuous incision (less traumatic), use of papaverine in order to avoid detrimental spasms to the endothelium, ligation of collateral vessels with clips, clamping with extremely delicate material and technique that avoids touching the endothelium, clamps with silicone jaws and subcutaneous venous graft placement. By using such careful procedures to protect the vessels, we believe that the results can really be similar in both reversed or in situ techniques, according to the studies by Taylor, an enthusiastic defender of the reversed saphenous vein technique.¹⁴ Although the difference in our initial results has been favorable to the use of in situ saphenous vein, we do not consider our comparison valid, since the cases of reversed saphenous vein submitted to surgery before December 1986 did not benefit from the technical advances and extreme careful endothelial protection adopted by our group after the implementation of the in situ saphenous vein technique. Another important difference - totally practical and without documented proof - that we observed between both techniques is that the in situ technique seems to offer more opportunities to save the failing grafts. There were cases of patients that presented occluded in situ grafts in their distal segments, with patent proximal segments and who kept their pulse during weeks, thanks to the collaterals (arteriovenous fistulas), which we never observed with reversed veins. Therefore, we believe that due to the greater technical ease of carrying out surgeries, still today the in situ saphenous vein technique is our procedure of choice for any kind of infrainguinal revascularization. We avoid using grafts of nonautologous veins in lower limb revascularizations, even using veins of the upper limbs whenever necessary. Such option can be explained by the great superiority of the results obtained through autologous veins, which has been presented in several modern studies, comparing them to artificial and biological grafts (PTFE and umbilical vein), even in above-knee revascularizations.³,⁴,⁵ We believe that the idea of sparing the autologous vein with the purpose of using it after the artificial prosthesis occludes in the first surgery is not valid. We should base our decision on the strictness of formal indications of lower limb revascularizations and use the best technique and the best graft in the first procedure. By doing that we offer the best treatment to the patient, avoiding great suffering and inconvenience caused by reoperations of early occluded artificial prostheses whose results will always be worse than the primary surgeries performed with autologous veins.²⁹ With a perfect diagnosis, a good technique and a strict follow-up, there are great possibilities of achieving excellent rates of secondary patency and satisfactory quality of life for patients with several associated risk factors that are responsible for their short survival. Next, we will present the lessons we learnt during more than two decades from dealing with autologous veins, especially regarding in situ saphenous vein.

All patients submitted to surgery were assessed through arteriographic studies, including aortoiliac territory and the lower limb considered for revascularization, with complete visualization of the plantar arches. We only recommended arteriography if surgery was intended and we never recommended it only as a diagnostic procedure; we attempted to perform the arteriography only in the limb to be revascularized. During approximately 20 years, up to 1996, the arteriographic studies had been performed by the members of group, using our own technique³⁰ of percutaneous puncture of the infra-renal aorta with dos Santos needle, and percutaneous puncture of the common femoral artery with Cournand needle. Such technique, which can be employed even in patients without palpable femoral pulses, offers, in most cases, arteriographies with excellent resolutions of the distal arteries of lower limbs, sometimes better than the arteriographies performed with a catheter. Even today it can be used in certain services with scarce material resources and those that do not have sophisticated equipment or means to use the catheter. The arteriographic study must be complete and conclusive, allowing for the preoperative planning of the surgery, as well as the prognosis, in order to perfectly detect a favorable or unfavorable arterial anatomy. We will hardly miss a distal artery liable to revascularization, even in a critically ischemic limb, when the arteriographic study is performed by means of an appropriate technique. We do not accept, up to this moment, the performance of surgical revascularization of lower limbs without arteriography.In the rare cases in which we were not able to contrast the distal arteries, by means of ultrasound Doppler, we located the sound of a possible recipient artery and completed the arteriographic study intraoperatively, after dissection and direct puncture of the target artery.

As shown in Table 2, the profunda femoris was the second more often used as donor to the proximal anastomosis. It is an excellent choice, since it is not usually affected by atherosclerosis and it can be reached through the internal saphenous vein without complicated procedures. It usually remains patent even after common femoral artery occlusion and it allows the easy reoperation of failing grafts. Whenever the superficial femoral artery is occluded, our second choice is the profunda femoris artery, since, in order to reach the common femoral artery with the internal saphenous vein, we have to cut and remove part of the anterior wall of the common femoral vein. In a survey that included 1,000 cases, Leather et al.¹³ have shown that there is no difference in the endurance of the patency of anastomosed grafts in the donor common, superficial or deep femoral arteries as well as in the receptor below-knee arteries (anterior and posterior tibial, fibular). The fibular artery was most commonly used as receptor (Table 2), because, among the below-knee arterial trunks, it is usually the artery that is least affected by atherosclerosis. We used the medial route, through the internal surface of the leg, to access this artery. When the anterior tibial artery is the receptor, we pass the in situ venous graft over the tibia subcutaneously.

Mills-Leather valvulotome³¹ (antegrade and retrograde models) has been used to present day. This tool requires total exposure of the venous graft by continuous incision of the skin, which has received some criticism due to to the posssibility of healing complications. The problem is that Mills-Leather valvulotome needs to be used under direct view, in order to prevent injury to the venous graft and this requires a continuous skin incision. We consider the use of this valvulotome relatively safe, and the severe complications with continuous incisions only occurred due to extensive necrosis of the skin in four grafts (2,3%), which occluded at the beginning of our study. We learnt that the route of the internal saphenous vein has to be drawn on the skin before the surgery, so that the incision is made exactly over the vein trajectory, thus avoiding skin detachments. After such precaution was taken, we did not have any further healing complications. The closure of the skin with mechanical suture substantially simplifies and shortens surgical time, and for that reason, we now use it in abdominal surgeries as well. The removal of clips should be made by an appropriate remover, in at least three weeks, when healing will have already occurred and local fibrosis will soothe the pain.

The endothelial protection of the venous graft and of the arteries involved in surgery is crucial and directly influences the results in the short and long term.¹⁷ Such protection includes the use of delicate and adequate material, so as to avoid touching the endothelium, excellent visual acuity, with Frontolux illumination and use of magnifying glasses, as previously mentioned, as well as meticulous care with the application of the valvulotome, which should never be carried out without previous dilation of the vein, with controlled pressure (maximum of 300 mmHg), in order for the valvulotome to travel through the vein and come in contact with the endothelium as minimally as possible .

Injury to the graft should be avoided during surgery. We conducted a statistical survey between December 1986 and August 1993, which was presented at the Pan-American Congress of Vascular Surgery in Rio de Janeiro, in 1993, in which we compared the evolution of 203 "pure" in situ saphenous grafts (grafts free of any repair or extensions) with 17 primary partial grafts (which received segments from reversed veins during primary surgery) and with 16 secondary partial grafts (which received segments from autogenous veins during reoperations). During that time, immediate thrombosis was present in 53% of the primary grafts; while in secondary grafts, immediate and late thromboses occurred in 6% and 25% of the cases, respectively, with a total rate of 31%; in pure grafts, the total rate of thrombosis in the same time period amounted to 21%, of which 8% corresponded to immediate thrombosis and 13% represented late thrombosis (Table 9). This shows that surgeons have to take extra care in order to avoid injury to the grafts, which would oblige them to use extensions with reversed autogenous veins in primary surgeries. The patency of such primary partial grafts is significantly lower than that of pure grafts. Secondary partial grafts, that is, those that received venous segments during reoperations, in late postoperative periods, have a better patency than primary partial grafts. This may result from the fact that thrombo reactivity is limited exclusively to the new venous segment received during reoperation, as the patent in situ venous graft should be totally adapted to the organism.

Table 9 - Incidence of thrombosis in pure and partial in situ grafts

	Pure in situ (n = 203)	Primary partial (n = 17)	Secondary partial (n = 16)
Immediate	8%	53%	6%
Late	13%	-	25%
Total	21%	53%	31%

We have verified that 48.4% of thrombosed grafts in our patients occurred within the first year after the surgery, which suggests that outcome is steadfastly associated with the surgical procedure, especially at the learning curve stage, showing that improved results are heavily dependent on the surgeons and their techniques. Table 7 shows that the major cause of failing grafts was the occlusion of the internal saphenous segment used for distal anastomosis. Shah et al.³² have observed that when the length of the distal segment exceeds 10 cm, the incidence of thrombosis in the referred segment remarkably increases. Therefore, we should attempt to mobilize the venous segment as minimally as possible, with rigorous endothelial protection, at the time of distal anastomosis.

The use of an in situ varicose saphenous vein in our experiment was disastrous. However, in a personal communication, Dhiraj Shah, current head of the Surgical Department of the Albany Medical Center, United States, which is known for having the largest experience with in situ saphenous revascularizations on a worldwide basis, has affirmed that moderately varicose veins are good to be used, since they do not require devalvulation.

The quality of the surgery should be checked in all cases, with no exception, by means of Doppler ultrasound intraoperatively, and completion arteriography after the last anastomosis. Doppler ultrasound is used to assess the quality of the flow and to locate arteriovenous fistulas; arteriography is used to confirm perfect anastomosis, presence of residual fistulas or valves and, especially, technical failures that cannot be detected by Doppler ultrasound and which, if not corrected immediately, could cause early thrombosis of the grafts and surgical failure (Figures 5 and 6). Thanks to these resources, the incidence of residual valves and arteriovenous fistulas were quite low in our patients (2.3% and 0.5%, respectively).

Figure 5 - Round lesion caused by traumatic clamping of the fibular artery, receptor of the distal anastomosis (arrow). Had arteriography not been performed, this lesion would have gone unnoticed, since, at the beginning it neither causes reduction of pulse rates nor alters the sound of intraoperative Doppler scan.

Figure 6 - Complementary angiography showing injury to the graft caused by the valvulotome (arrow). This lesion might not be detected only with Doppler ultrasound or by palpation, since it does not reduce pulse rates and does not alter the sound until thrombosis occurs, if left untreated.

Table 3 shows that 64.5 % of the cases submitted to surgery required surgical debridement and partial amputations of the forefoot, which is acceptable, since 80% of the surgical indications included patients with necrosis. These debridements are always performed in the same surgery, after revascularization and dressings of the incisions, which are isolated from the necrotic areas. We have never had infectious contamination or complications, since we always attempt to perform broad debridements and, whenever possible, definitive ones, by totally resecting all necrosed or nonviable tissues and bones, in order to avoid later debridements, which are troublesome and cause a lot of suffering to the patients and their families.

A rigorous postoperative follow-up is imperative, since it allows the surgeon to assess the technique used and to improve results, also helping to detect failing grafts before occlusion. It is common knowledge that reoperations of failing grafts are much better than surgeries aimed at recovering totally thrombosed grafts (Table 8).

In conclusion, distal revascularization of lower limbs with critical ischemia by means of in situ saphenous grafts offers satisfactory patency if carried out according to stringent indication criteria, with careful endothelial protection and regular postoperative follow-up. A rigorous and regular follow-up of revascularized patients allows the recovery of failing grafts and significant improvement of patency.

REFERENCES

1. Chang B, Leather RP, Kaufman J, Kupinski AM, Leopold PW, Shah DM. Hemodynamic characteristics of failing infrainguinal in situ vein bypass. J Vasc Surg 1990;12(5):596-9.

2. Leather RP, Powers SR, Karmody AM. A reappraisal of the in situ saphenous vein arterial bypass. Surgery 1979;86:453-60.

3. Veterans Administration Cooperative Group 141. Comparative evaluation of prosthetic, reversed and in situ vein bypass grafts in distal popliteal and tibial-peroneal revascularization. Arch Surg 1988:123:434-8.

4. Cranley FF, Hajner CD. Revascularization of the femoropopliteal arteries using saphenous vein, polytetrafluorethylene and umbilical vein grafts. Arch Surg 1982;117:1543-50.

5. Eickhoff JH, Broome A, Ericsson BF, et al. Four years results of a prospective randomized clinical trial comparing PTFE and human umbilical vein for bellow-knee femoropopliteal bypass surgery. J Vasc Surg 1987;6 506-11.

6. Buchbinder D, Sigh JK, Karmody AM, Leather RP. Comparison of patency rate and structural change of in-situ and reverse vein arterial bypass. J Surg Res 1981;30:213-16.

7. Leather RP, Corson JD, Naraynsingh V, et al. Technique and results of in situ bypass. Cont Surg 1984;24:31-6.

8. Abbott W, Wieland S, Anstone WG. Structural changes during preparation of autogenous venous grafts. Surgery 1974;76:1031-6.

9. Bush HL, McCabe ME, Nesbeth DC. Functional injury of vein graft endothelium. Role of hyperthermia and distension. Arch Surg 1984; 119: 770-779.

10. Bandyk DF. Postoperative surveillance of femorodistal grafts: the application of Echo-Doppler (Duplex) ultrasonic scanning. In: Bergan JJ, Yao JST, editors. Reoperative Arterial Surgery. Orlando: Grune & Stratton Inc.; 1986. p. 59-79.

11. Leather RP, Corson JD, Karmody AM. Instrumental evolution of the valve incision method of in situ saphenous vein bypass. J Vasc Surg 1984;1:113-8.

12. Leather RP, Karmody AM. In situ saphenous vein arterial bypass for the treatment of limb ischemia. New York: Year Book Medical Publishers, Inc.; 1986. p. 175.

13. Leather RP, Shah DM, Shang B, Kaufman J. Resurrection of the in situ saphenous vein bypass 1,000 cases later. Ann Surg 1988;208:435-42.

14. Taylor LM, Edwards JM, Porter JM. Present status of reversed vein bypass grafts: five years results of a modern series. J Vasc Surg 1990;11:193-206.

15. Bonchek LI. Prevention of endothelial damage during preparation of saphenous veins for bypass grafting. J Thorac Cardiovasc Surg 1980;79:911-15.

16. LoGerfo FW, Quist WC, Crawshaw HW. An improved technique for endothelial morphology in vein grafts. Surgery 1981;90:1015-19.

17. Bunt TJ. Iatrogenic Vascular Injury. 1st ed. New York: Futura Publishing Co. Inc.; 1990.

18. Rutherford RP, Flanigan DP, Gupta SK, et al. Suggested standards for reports dealing with lower extremity ischemia. J Vasc Surg 1986;4:80-97.

19. Rutherford RB, Baker JD, Ernst CE, et al. Recommended standards for reports dealing with lower extremity ischemia: revised version. J Vasc Surg 1997;26:517-38.

20. Knapp RJ. Basic Statistics. 1st ed. New York: Wiley Medical Publication; 1978.

21. Bandik DF, Schmitt DD, Seabrook GR, et al. Monitoring functional patency of in situ saphenous vein bypass: the impact of a surveillance protocol and elective revision. J Vasc Surg 1989;9:286-96.

22. Veith FJ, Weiser RK, Gupta KS, et al. Diagnosis and management of failing lower extremity arterial reconstructions. J Cardiovasc Surg 1984;25:381-4.

23. Santos CMT. Durabilidade das revascularizações infra-guinais com safena in situ. Recuperação de enxertos falhando. Comparação de perviedades primária e secundária. Rev Ang Cir Vasc 1995;4:46-58.

24. Leather RP, Powers SR, Karmody AM. A reappraisal of the in situ saphenous vein arterial bypass: its use in limb salvage. Surgery 1979;86:453-61.

25. Fogle MA, Withmore AD, Couch NP, Mannick JA. A comparison of in situ and reversed saphenous vein grafts for infrainguinal reconstruction. J Vasc Surg 1987;5:46-52.

26. Carney WI, Balko A, Barrett M. In situ saphenous vein femoropopliteal and infrapopliteal bypass: a two years experience. Arch Surg 1985;120:812-16.

27. Bushbinder D, Rollins DL, Samrow CE, et al. In situ tibial reconstruction: state-of-the-art or passing fancy. Ann Surg 1988;207:184-8.

28. Vieira de Mello A, Santos CMT, Silva JLCN, et al. In situ saphenous vein arterial bypass for infrainguinal revascularization: initial experience using an open technique. Angiology 1991;10:126-32.

29. Henhe PK, Proctor MS, Zajkowiski PJ, et al. Tissue loss, early primary graft occlusion, female gender and a prohibitive failure rate of secondary infrainguinal arterial reconstruction. J Vasc Surg 2002;35:902-9.

30. Vieira de Mello A, Santos CMT, Silva JLCN, Portilho MA. Visualização arteriográfica pré operatória dos arcos plantares. Radiol Bras 1984;17:84-8.

31. Leather RP, Corson JD, Karmody AM. Instrumental evolution of the valve incision method of in situ saphenous vein bypass. J Vasc Surg 1984;1:113-23.

32. Shah DM, Darling III RC, Shang BB, Fitzgerald K, Paty PSK, Leather RP. Long term results of in situ saphenous vein bypass analysis of 2058 cases. Ann Surg 1995;222:438-48.