SciELO - Scientific Electronic Library Online

vol.9 número1Prevalência da estenose aterosclerótica do tronco celíaco e da artéria mesentérica superior na arteriopatia oclusiva dos membros inferioresTunelizador maleável moldável universal na cirurgia de revascularização do membro inferior isquêmico índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados




Links relacionados


Jornal Vascular Brasileiro

versão impressa ISSN 1677-5449

J. vasc. bras. vol.9 no.1 Porto Alegre  2010  Epub 23-Abr-2010 



Temporal venous arterialization of the diabetic foot



Francisco Lengua, Andrés La Madrid, Carlos Acosta, Jorge Vargas

Department of Thoracic and Cardiovascular Surgery, Hospital de Policía, Lima, Perú





Background: The first idea of surgeons (1902) to avoid amputations due to ischemia was to deviate the arterial flow to the venous system using an arteriovenous fistula between adjacent vessels; however, the results were unreliable. Since then, sympathectomies, endarterectomies and bypasses have been created, and more recently, other medical, surgical, and endovascular advances have been used. However, amputations continue to be performed worldwide mainly in diabetic patients. The arterialization of the foot veins, based on the old idea of inverted blood flow, is a new possibility for these patients who, otherwise, could lose their limbs.
Objective: To demonstrate that arterialization of the foot veins in diabetic patients with neuroischemic lesions, usually infected (diabetic foot), is an effective and long-lasting method, even though the bypass only works temporally.
Patients and method: From January 2000 to February 2009, 59 patients with diabetic foot were threated by means of arterialization of the foot veins. An early death was not included in the analysis. Of the 58 remaining patients, 44 were male and 14 were female. Their mean age was 71 years old: (53-91 years). Fifty-four of them were classified as being Fontaine IV and four were IIIB. The arterialization was conducted using an inverted venous graft proximally anastomized to an artery presenting good flow (external iliac, femoral or popliteal arteries) distally in the internal marginal vein of the foot, eliminating the valves of the foot arch veins.
Results: Of the 58 patients who underwent arterialization, 12 failed due to early thrombosis, having amputated limbs, and 46 succeeded (79%): six at short-term, 12 at medium-term, and 28 at long-term. In the groups who were considered to have a successful outcome (two), there were 38 late bypass thrombosis (95%) and four relapses of the ischemic symptoms. There was no case of heart overload or varices.
Conclusion: The arterialization of the diabetic foot is possible, effective and long-lasting probably thanks to the induction of a neo-arteriogenesis and neo-angiogenesis that maintain the benefits even after the bypass has occluded (temporal function).

Key words: Limb salvation, diabetic foot, temporal arterialization, limb amputation.




In the early 20th century, the first idea of surgeons (San Martín, 1902, Spain, and Jaboulay, 1902, France) to avoid amputations due to ischemia was to deviate the arterial blood flow to the venous system using a fistula between adjacent vessels. Subsequently, this technique has been used by different surgeons from different countries, but it has been almost completely abandoned because of its uncertain results. In diabetics, it was used for the first time by Halsted and Vaughan, 1912 (USA); however, they did not achieve good results.

In 1948, Kunlin (France),1 after having done the first bypass surgery successfully, started a new era in the treatment of ischemia.

The improvement in the suture material and angiographic techniques, and the use of microsurgery have allowed the implantation of bypasses in small arteries at the ankle and foot level;2 sometimes even with bone resection.3 However, for a significant number of patients, all conventional therapeutic resources, both medical and surgical, are still unavailable.4 These patients are condemned to a limb amputation.

Following the ideas of Leger,5 which states that "our angiographic findings, in contrast, proved over and over again that there may be upstream blood flow in the venous system" and with the help of Kunlin's bypass, we returned to this old idea, but we changed its method of use: arterialization of foot veins (AFV) using a bypass (Figure 1). Therefore, this technique was first used in 1974 in three diabetic patients with promising results.6 However, the technique was not well received and, only 20 years later, it started to be used because of publications about its successfully use by other vascular surgery teams.7-10



Currently, there is a tendency to group diabetic patients with neuroischaemic and infected trophic disorders in their feet into the "diabetic foot" category (tissue necrosis, ulcerations, and gangrenes), thus unifying one of the most feared complications of diabetes mellitus. Because of their severity, these injuries lead to limb amputation from 10 to 20 times more often in diabetic patients than in non-diabetic ones.11


Material and methods

Between January 2000 and February 2009, 59 patients with diabetic foot were treated with AVF. One of the patients died on the 18th postoperative day and, therefore, this case was not included in the study. Of the 58 remaining patients, 44 were men and 14 were women. Their mean age was 71 years old (range: 53-91), with 50 patients having type 2 or non-insulin-dependent diabetes and eight patients having type 1 diabetes (World Health Organization - WHO). Fifty-four of these patients were classified as Fontaine stage IV and four were stage IIIB. Thirty-five patients were hypertensive, 15 had cardiac rhythm disorders, 12 had coronary artery disease, eight had renal failure (of these, two were on hemodialysis), six suffered from Parkinson's disease at an initial stage, five had sequelae of myocardial infarction, four had diabetic retinitis, and three had prostate cancer.

The arteriography comprising the whole limb showed a severe impairment of the leg and foot arteries. In two patients, surgical exploration was done before arterialization: dorsalis pedis and posterior tibial artery. The distal venous system was clinically evaluated and by means of Doppler scan.

In terms of graft, we preferred the internal saphenous vein removed from one or both limbs, as appropriate.

Graft was done in only one venous segment in 32 cases, with 27 cases of inverted flow and five cases of non-inverted flow, but with prior destruction of the valves. It was composed of two vein segments in 18 cases. In eight cases there were mixed grafts [polytetrafluoroethylene (PTFE) plus reversed vein always distally placed], with six sequential grafts (PTFE plus vein and one using vein donated by a family member) and two cases of mixed direct grafts. In 25 cases, the donor artery was the popliteal, while in 27 cases, it was the superficial femoral artery, in three cases we used the common femoral artery, and in other three cases the external iliac artery.

The destruction of the foot valves was performed using a set of five flexible valvulotomes, four of them were 24-cm long, equipped with short olives with tapered tip implanted at an angle of about 10 degrees for destruction of the valves in the veins of the foot dorsum, and one of the valvulotomes was a small 6-cm long device with an olive implanted without angulation for destruction of the ostial valves of the collateral veins discharging into the phlebotomy. The rupture of the valves was achieved by means of a thud, turning the olive into the vessel lumen to prevent tearing the vein wall (Figure 2).



Surgical technique

We describe the implementation of a bypass between two segments (12-14).

The anesthesia method of choice was the use of an epidural catheter. The necrotic lesion was isolated with a sterile, waterproof material. It protects the heel (from prolonged compression) using a surgical glove filled with sterile water placed under the calcaneal tendon (soleus/twin tendon that inserts into the calcaneus).

The intervention was conducted by two teams: the first team removed the internal saphenous veins from both thighs to obtain the required length for the vein graft, prepared and implemented one of the veins in the donor artery (superficial femoral or other). The same team performed an oblique longitudinal incision in the lower portion of the inner side of the knee to tunnel the proximal portion of the bypass path. The second team performed the dissection of the marginal vein of the foot. If the path was visible in this vein, a curved incision was made with upper opening in the pre-malleolar fossa about 10 mm below the path of this vein to prevent the incision from "falling" on the same vein, compromising the anastomosis (skin necrosis). Traction threads were placed on the edge of the surgical wound comprising skin, subcutaneous cell tissue, and superficial dorsal fascia. During this exposure of internal marginal vein (as well as during the closure of the surgical wound), the use of dissection tweezers and separators is not allowed to prevent skin necrosis.

If the path of the internal marginal vein is not visible, with a small incision in front of the internal malleolus it is possible to isolate a segment of the pre-malleolar internal saphenous vein and to pass it around a wire whose traction will show the path of the internal marginal vein.

The path of the bypass is tunneled in the leg in whose bed a plastic tube (No. 24) was placed through which the graft will pass easily without rotation (the metal tunneler should not be used because it is rigid and does not adapt to the shape of the tibia, producing too much detachment of the skin with risk of consequent necrosis).

When the graft is in place, the tube is removed and its distal end is cut to prevent anastomosis in "pockets." A longitudinal phlebotomy as deemed necessary (in relation to the position of collaterals of the exposed segment of the internal marginal vein, whose ostial valves must be destroyed) is performed slightly to the outer edge of the internal marginal vein (in relation to the axis of the body) so that the anastomosis is a little "ejected." This successful phlebotomy ruptures the accessible valves of the foot veins with the material designed for that purpose (the arterial stents only rebend the valves against the wall) (Figure 3). A suture with prolene 7/0 was performed with magnifying lenses, except for the distal angle of the anastomosis, which is done with separated stitches. The two venous segments underwent termino-terminal anastomosis in the incision made in the knee. The clamping of the internal marginal vein and its collaterals was carried out by traction of a thin thread. When there is risk of compression of the anastomosis in the closure of the surgical wound, its edges can be left semi-sutured (up to 6 mm); however, the whole vessel exposed should be always covered with skin by means of one or two isolated stitches.



In the immediate postoperative period, patients received low-molecular-weight heparin for a week and then continued receiving warfarin.



Evaluation criteria for successful cases. Successful cases were those in which an amputation with loss of support (heel) could be avoided for at least 1 year, with permeable bypass for at least 1 month, with suppression of pain, healing of the necrotic lesions (with or without partial amputation of the foot), and maintenance of these benefits despite the occlusion of the bypass. Furthermore, following the traditional classification, we considered three types of successful cases: short term (between 1 month and 1 year), medium term (between 1 and 5 years), and long term (5 years or longer).

Of the 58 patients, 12 were failures and underwent amputation in spite of the attempts to solve obstruction, nine supracondyles and three infracondyles. The other 46 were successful cases (79%), with six short-term cases, 12 medium-term, and 28 long-term cases. According to the criteria of successful cases mentioned above, only the medium- and long-term successful cases were assessed.

Of the 40 medium- and long-term successful cases, 38 suffered thrombosis (95%), seven patients died (two in the group of medium term and five in the long-term group) due to intercurrent diseases, and three patients were lost for follow-up after a mean period of 2 years. All patients were followed up for a mean period of 4 years and 4 months. The mean bypass duration was 8 months. The healing period of necrotic lesions varied depending on the extent, site, and infection degree of the necroses; however, this healing was obtained between a mean period of 3 and 4 months and in a quarter of the cases it was obtained with the help of skin autografts. Two patients still have a permeable bypass, one after 1 year and 10 months and the other after 6 years and 10 months.



In the successful cases with late thrombosis (after 1 month), there were four recurrences of ischemia. One patient died of sepsis without being reoperated; the other three cases were successfully treated with removal of obstructions and angioplasty of the distal anastomosis. We also found four partial thrombosis in those cases with long phlebotomy (5 cm), of which two were treated by percutaneous dilatation and the other two, whose stenosis included the distal portion of the anastomosis, were kept under observation (Figure 4), having kept their limb until the moment this article was prepared: 3 years.



In addition, there were 16 skin necroses of the surgical wound of the foot without compromising the anastomosis, five cases of repeated postoperative progression of necroses of the foot with permeable bypass treated with surgical cleaning, three cases needed a transmetatarsal amputation, three cases of painful necrotic syndromes, and two cases of painful hyperesthesias triggered at the slightest touch on the skin, all of them treated with lumbar sympathetic blocks. There were three false paths without consequences. There were not cases of cardiac overload or varicose veins of the operated limb.



According to Leriche, who says that "medicine has nothing to gain with the contempt of clinical fact in favor of annotations in figures,"15, our results are provided in a simple and classic manner, based on clinical observation, Doppler ultrasound and angiographic controls. Moreover, the parameters adopted in the evaluation of the results of classic bypasses are not fully applicable to foot arterialization. In fact, if thrombosis of the arterialization bypasses occurs between the first and sixth month after implantation, amputation is delayed or prevented even over the medium term. It is possible that the arterialization bypass has promoted the induction of neo-collaterals (arteriogenesis: Figures 5, 6) and neo-arterioles (arteriogenesis) 16, 17 responsible for the improvement of the ischemic lesions of the feet, even after occlusion bypass graft. In conventional bypasses, preservation of the limb is closely linked to the bypass duration.18-21




Toursarkissiam,22 after analyzing 80 diabetic patients who received distal bypasses, had an overall rate of amputations of over 50%, with a mean time of bypass permeability of 13 months and a 14-month follow-up after the occlusion of the bypass. In our series, we saved 48 limbs, with a mean bypass permeability of 8 months and a mean follow-up of 4 years and 4 months. There were not any cases of major amputation. Moreover, of these 80 distal bypasses, only 35 (44%) were implanted in the arteries below the malleoli (31 in the pedal arteries and four in the plantar arteries). In our view, these bypasses are the only ones that could be compared with our foot arterializations in a randomized study. On the other hand, in the classic bypasses, the majority of the stenosis occurred late and mostly in the proximal anastomosis,23 while, in the AFV, most stenosis occurred in the distal anastomosis and had an earlier onset. By contrast, in the classic bypass revascularization, suppression of ischemic symptoms was fast, since it occurred by means of physiological method, while, in the AFV, the same recovery was slow, sometimes even with progression of the necroses despite the bypass permeability, which is explained by a revascularization done, at least for some time, using a non-physiological method.

Mar and Ramos24 (Tampico, Mexico), personal communication, using a technique similar to ours, obtained a 75% success rate in 21 arterializations (in press).

There are other methods to arterialize foot veins, as proposed by Busato et al. 25,26 (Ponta Grossa, Brazil), who, in a study in press entitled "The use of long saphenous vein 'in situ' to the arterialization of the venous arch of the foot," mentioned a series of 16 arterialized cases with the variant in question, with good results in most cases.

The successful results of the AFV in diabetic patients, which in the past accounted for 64%,12 with phlebotomy as needed and a more effective destruction of the valves, is currently 79%. These results were achieved in patients who could not receive a classic distal bypass.

We are aware of the fact that our study lacks evidence of our findings; evidence that are unfortunately not available to us because of the lack of resources; however, the results of our study are the product of a long clinical observation and radiological controls.



The arterialization of the veins in the diabetic foot being effective and durable even if the bypass only works temporarily is an additional alternative in patients for whom conventional medical and surgical treatments are not available.



1. Kunlin J. Le traitement de la ischémie artéritique par la greffe veineuse longue. Rev Chir. 1951;206-35.         [ Links ]

2. Pomposelli FB, Jepsen SJ, Gibbon GW, Campbell DR, Logerfo FW. Efficacy of the dorsal pedal bypass for limb salvage in diabetic patients: Short term observations; J Vasc Surg. 1990;111:745-52.         [ Links ]

3. Ascer E, Veith F, Gupta SK. Bypasses to plantar arteries and other tibial branches. An extended approach to limb salvage. J Vasc Surg. 1988;8:434-41.         [ Links ]

4. Campbell WB, Verfaille P, Ridler BM, Thomson JF. Non operative treatment of advanced limb ischaemia: the division for palliative care. Eur J Vasc Endovasc Surg. 2000;19:246-49.         [ Links ]

5. Leger L. Essais de revscularisation d'un membre par anastomose artério-veineuse termoino-terminal. Press Med. 1949;57:1198-200.         [ Links ]

6. Lengua F. Technique d'artérialisation du réseau veineux du pied. Press Med. 1975;4:1039-42        [ Links ]

7. Pokrovskii AV, Chupin DV, Khorovets AG. Arterialization of the foot venous system in the treatment of the critical limb ischaemia and distal bed occlusion. Ang Vasc Surg. 1996;4:73-93.         [ Links ]

8. Taylor RS, Belli A, Jacob S. Distal venous arterialisation for salvage of critically ischaemic inoperable limbs. Lancet. 1999;354:1962-65.         [ Links ]

9. Engelke C, Morgan RA, Quarmby JW, Taylor RS. Distal venous arterialization for lower limb salvage: Angiographic appearance and interventional procedures. Radiographies. 2001;21:1239-50.         [ Links ]

10. Rowe VL, Hood DB, Liphan J, Terramini T, Torres G, Katz S. Initial experience with dorsal venous arterialization for limb salvage. Ann Vasc Surg. 2002;16:187-92.         [ Links ]

11. Herrison C, Simon L. Le pied diabétique. Paris, Milan, Barcelona: Masson Ed; 1993. p. 5.         [ Links ]

12. Lengua F. Le pontage d'artérialisation veineuse distale peut-il être bénéfique au pied diabétique avec nécrose? Chirurgie. 1994-1995;120:1346-52.         [ Links ]

13. Lengua F, Cavero C, Barriga H, Zuñe A. L'artérialisation veineuse périphérique temporaire peut-elle éviter des amputations pour ischémie? Médecine et Armées. 2004;32:103-108.         [ Links ]

14. Lengua Almora F. Artérialisation du pied pour ischémie. Dernière chance avant l'amputation chez le diabétique. Lima: Delvi; 2007.         [ Links ]

15. Leriche R. Bases de la chirurgie physiologique. Mass Ed; 1955. Préface.         [ Links ]

16. Wahlberg E. Angiogenesis and arteriogenesis in limb ischaemia. J Vasc Surg. 2003;38:198-203.         [ Links ]

17. Mousa AY. Angiogenesis in limb ischemia. In: Whittemore AD, editor. Advances in Vascular Surgery. 2004. p. 122.         [ Links ]

18. Veith FJ, Weiser RK, Gupta SK, et al. Diagnosis in management of failing lower extremity arterial reconstruction prior to graft occlusion. J Cardiovasc Surg. 1984;25:381-4.         [ Links ]

19. Isaksson L, Lundgren F. Vein bypass surgery to the foot in patients with diabetes in critical ischaemia. Br J Surg. 1994;81:517-20.         [ Links ]

20. Pomposelli F, Maracaccio EJ, Gibbon GW, Campbell DR, Logerfo FW. Dorsalis pedis bypass durable limb salvage for foot ischaemia in patients with diabetes mellitus. J Vasc Surg. 1995; 2:375-84.         [ Links ]

21. Panayiotopoulos YP, Reidy JF, Taylor PR. The concept of knee salvage: why does a failed femoral/pedal arterial bypass not affect the amputation level? Eur J Vasc Endovasc Surg. 1997;13:447-85.         [ Links ]

22. Toursarkissiam B, Shireman PK, Schoolfield J, Blumoff RI. Evolution après Occlusion d'un pontage Distal chez le Diabétique. Ann Chir Vasc. 2003;17:670-5.         [ Links ]

23. Darling RC, Roddy SP, Chang BB, et al. Long-term results of revised infrainguinal arterial reconstruction. J Vasc Surg. 2002;35:773-8.         [ Links ]

24. Mar CHR, Ramos G et al. Salvamento del pie diabético con la arterialización de las venas del pie. Cir Gral. 2007;29S1:135.         [ Links ]

25. Busato CR, Utrabo CAL, Housome JK, Gomes RZ. Arterialização do arco venoso do pé para tratamento da isquemia crítica sem leito distal. Cir Vasc Angiol. 1999;15:117-21.         [ Links ]

26. Busato CR, Lima Utrabo CA, Zanetti Gomes R, et al. Arterialization of the venous arch of the foot for the treatment of thromboangeiitis obliterans. J Vasc Bras. 2008;7:267-71.         [ Links ]



Servicio de Cirugía torácica y C-V, Hospital de Policía
Av. Brasil s/n - Lima, Perú

Manuscript received Apr 6 2009, accepted for publication Jan 13 2010.



No conflicts of interest declared concerning the publication of this article.

Creative Commons License Todo o conteúdo deste periódico, exceto onde está identificado, está licenciado sob uma Licença Creative Commons