- Email: [email protected]
Distal in situ saphenous vein grafts for limb salvage
Distal In Situ Saphenous Vein Grafts for Limb Salvage increased Operative Blood Flow and Postoperative Patency
Harry L. Bush, Jr., MD, Boston, Massachusetts Christopher A. Corey, MD, Boston, Massachusetts Donald C. Nabseth, MD, Boston, Massachusetts
Thrombosis of vein grafts in the early postoperative period remains the major obstacle to graft patency and limb salvage in distal reconstruction of the leg. Surgeons have ascribed these early failures of reversed vein grafts to technical problems, such as inadequate inflow or outflow, inadequate vein size, poor quality of the vein wall, anastomotic or clamping techniques, hypercoagulability, or vessel mismatch at the anastomotic sites. However, increasing attention is being focused on the role of the endothelium in preventing graft thrombosis. Experimental evidence has demonstrated that the in situ vein graft technique preserves the morphologic integrity of the endothelium plus the metabolic capacity to produce prostacyclin, a potent inhibitor of platelet aggregation on a graft wall. These functional and morphologic benefits are distinctly abnormal in the reversed vein graft [I $1. Leather et al [3] have recently revitalized the in situ technique for arterial reconstruction of the leg in patients with critical ischemia. Their clinical reports document the dramatic clinical results with in situ vein grafts, especially in the infrapopliteal position [4]. The increased utilization of saphenous veins and the increased patency of these grafts allow more patients with advanced ischemia to have successful revascularization. The present series was initiated to corroborate the experience of Leather et al [3,4]. This report includes our entire experience with the in situ technique for revascularization of the distal popliteal and infrapopliteal arteries for limb salvage. From the Departments of Surgery, Boston Veterans Administration Medical Center and Tufts University School of Medicine, Boston, Massachusetts. Requests for reprints should be addressed to Harry L. Bush, Jr., MD. Dapartment of Surgery, Veterans Administration Medical Center, 150 South Huntington Avenue, Boston, Massachusetts 02130. Presented at the 63rd Annual Meeting of the New England Surgical Society, Bretton Woods, New Hampshire, Octobar 15-17, 1962.
542
Material
and Methods
Twenty-eight patients had a total of 30 bypass grafts to the distal popliteal or tibial arteries since December 1979. All patients were men and the mean age was 62.5 years (range 39 to 81 years). Associated cardiovascular risk factors included tobacco abuse in 26 patients, hypertension in 13, diabetes mellitus in 10, and angina or an old myocardial infarction in 8. All patients had critical ischemia with the leading indication for revascularization being ischemic ulceration in 10 limbs, cellulitis in 3 limbs, digital gangrene in 2 limbs, and pain at rest in the remaining 15 limbs. Angiography was performed in all patients and revealed adequate runoff to the ankle in 21 of the 30 limbs. In nine case additional angiograms were required early in the operative procedure to define the patency of distal blood vessels before reconstruction. Runoff was defined as the number of patent blood vessels connecting the anastomotic site with the pedal arch. The mean runoff was 1.2 f 0.1 vessels. The runoff was three vessels in 3 limbs, two vessels in 4 limbs, one vessel in 19 limbs, and no vessels in 4 limbs. At surgery, the common femoral artery and its branches, along with the saphenofemoral junction, were exposed through a vertical incision. The first 6 cm of the saphenous vein were dissected, the saphenous vein was divided at the saphenofemoral junction, and the femoral vein was closed with a running 5-O monofilament vascular suture. The first two to three valves were incised using scissors inserted through the orifice of the saphenous vein, the saphenous bulb was anastomosed to the common femoral or superficial femoral artery, and all clamps removed. The proximal anastomosis was performed to the common femoral artery in 6 limbs, to the junction of the common and superficial femoral arteries in 11 limbs, and to the proximal superficial femoral artery in 5 limbs. In addition, two anastomoses originated from the common and profunda femoral artery patch grafts and six from the hoods of prosthetic inflow grafts. The saphenous vein was exposed for its entire length, leaving the overlying superficial fascia intact. Competent valves in the arterialized vein could easily be detected and were rendered incompetent in one of three ways. The first nine operations were performed using Hall’s The American Journal of Surgery
Distal In Situ Vein Grafts
[5] technique for a transverse venotomy, direct valve excision, and closure with interrupted 7-O monofilament vascular sutures. Later, a scissor with a shaft that was 1 mm in diameter and 10 cm in length was custom-made. The size allowed introduction of the scissor through a side branch with direct incision of each valve cusp. In the past year, a valvulotome was inserted through the distal end of the saphenous vein and was used to disrupt the valve cusps [3,6]. To protect the vasa vasorum, the superficial fascia was incised only to ligate side branches of the vein. The quality of the saphenous veins varied greatly. Sixteen of the 30 saphenous veins were distinctly abnormal which probably would have precluded their use as a reversed vein graft. Eleven of these 16 veins included a segment of the graft with a diameter of less than 3.5 mm before surgical manipulation. These 11 patients included 4 with bifurcation of the saphenous vein in the thigh with two branches running parallel to each other and subsequently rejoining (usually near the knee) to form a common trunk. The other five saphenous veins had segments below the knee that had thick walls and diminished diameters. Pathologic examination confirmed a chronic inflammatory process. Two of the five patients required distal replacement with a short reversed vein graft. The other three patients’ saphenous veins were judged to be marginally adequate for use as a bypass conduit. The distal 6 to 7 cm of the saphenous vein was dissected free from the subcutaneous tissue and brought gently down to the recipient artery. The recipient artery was controlled using elastic Silastic@ loops. Anastomoses were performed with a running 6-O monofilament vascular suture in the popliteal artery and an interrupted 7-O monofilament suture in the tibia1 arteries. The distal anastomoses included 13 grafts to the distal (below the knee) popliteal artery, 3 to the tibioperoneal trunk, 8 to the posterior tibia1 artery, 4 to the anterior tibial artery, and 2 to the peroneal artery. Composite grafts were used in three bypass procedures. One patient had a 6 cm interposition prosthetic graft which ran from an aortofemoral limb to the proximal saphenous vein. Two patientshad overt phlebitic changes in the distal portion of the saphenous vein that was needed for bypass, therefore, the distal ends of their grafts were replaced with a reversed cephalic vein (8 cm in length) and a reversed saphenous vein (6 cm in length). After reconstruction, operative arteriograms were performed to detect retained valve cusps, arteriovenous fistulas, and other more standard technical problems (Figure 1). In addition, systolic Doppler pressures at the ankle was compared with the brachial systolic pressures to derive the ankle systolic pressure index (ASPI). Electromagnetic flowmeter methodology was used to measure mean rate of blood flow in the basal state and after vasodilation with papaverine (30 mg) injected into the common femoral artery. With the flow probe positioned in the groin, occlusion of the distal end of the vein graft caused the blood flow in the graft to fall to zero, thus confirming that all arteriovenous fistulas had been ligated. During the operative procedure, the patients were systemically heparinized initially and then were maintained with hourly bolus injections of heparin. In addition, infusion of low molecular weight dextran was started intraoperatively and maintained for 3 days at 25 ml/h. Qn the third postoperative day the patients began receiving aspirin, 300 mg/day orally. Patients were followed in the special vascular clinic by staff surgeons and 100 percent follow-up was obtained. Volume 145, April 1993
Flgure 1. Operatlve angiogram after artertallzatton of the sapheIK)us vetn. Two common pruMem5 am noted:a smal atterlove~~~ ftstula that f/Its the deep venous system and a paritalty competent retalned valve.
Results Patency of the vein graft was defined as a palpable pulse within the graft or a systemic Doppler pressure measured with the inflation cuff positioned over the distal graft. All 30 grafts were patent 30 days postoperatively. Five of the patients with patent grafts had reexploration for retained competent valves or graft thrombosis in the early postoperative period. Longitudinal follow-up revealed that all grafts remained patent except for one which failed 22 months after surgery (Figure 2). This failure was marked by the appearance of intermittent claudication and a decrease in the ASP1 to 0.63. Rest pain and ischemic ulceration did not develop before this patient’s death from a myocardial infarction 5 months after graft failure. Life table analysis showed a cumulative graft patency of 100 percent until 24 months postoperatively when the patency rate decreased to 82 percent for the 30 grafts. All 30 extremities had been salvaged at their last follow-up examination or at the time of death. One patient required digital amputations and one required a transmetatarsal amputation for existing gangrenous lesions present preoperatively. Three patients died between 6 and 12 weeks after surgery. Two of the patients died at home, one from a myocardial infarction and 1 from an acute pulmonary embolus, and the third died in the hospital from sepsis related to an ischemic hip disarticulation 543
0
L 1
12
6 TIME
(months)
wound on the contralateral side. A fourth patient died 27 months postoperatively from a myocardial infarction. After angiography confirmed the absence of arteriovenous fistulas or technical defects, electromagnetic flowmeter measurements in 20 grafts revealed a basal blood flow of 161 f 31 ml/min (Figure 3). Vasodilation of the distal arterial tree with 30 mg of papaverine resulted in a maximum increase in the flow rate to 267 f 48 ml/min. The lack of functional arteriovenous fistulas was confirmed by a decrease in graft flow to zero during occlusion of the distal end of the graft. The difference in graft flow was not significantly different when .distal popliteal grafts were compared with infrapopliteal grafts, nor when patients with diabetes mellitus were compared with nondiabetic patients. The mean value for the preoperative ASP1 was 0.27 f 0.13 in the 30 limbs (Figure 4). This rose to 0.85 f 0.21 within 48 hours after reconstruction. The ASP1 at last follow-up in all limbs was 0.89 f 0.27 (n = 26). The postoperative ASP1 was an invaluable monitor when the postoperative values were inappropriately low for the known runoff. Residual competent valve cusps in two of three patients and a late stenotic lesion in the graft of another patient were detected before thrombosis of the graft occurred. Early detection of the lesions allowed successful graft revision. Early in the series, two patients had immediate postoperative thrombosis of their grafts due to extrinsic compression. Both grafts were anastomosed to the distal popliteal artery and were allowed to lie directly behind the medial condyle of the tibia as the graft coursed from its subcutaneous position into the popliteal fossa. The patients had unusually bulky gastrocnemius muscles, the belly of that muscle
544
18 I
241
Ffgure 2. Ltfe table analysts of 30 In dtu grafts whkh depkts the patency rate for bekw-knee artertal reconstructlon.
compressed the graft against the boney prominence when the leg was extended. Both patients had successful thrombectomies of their grafts and relief of extrinsic pressure. No further technical abnormality was found on arteriograms obtained after thrombectomy and the patients were discharged with patent grafts. Three grafts were found to have residual competent valve leaflets, despite “normal” operative arteriograms immediately after reconstruction. One patient had palpable pedal pulses and systemic ankle Doppler pressures for 48 hours before his graft acutely thrombosed. After successful thrombectomy, the operative arteriogram showed a partially competent valve at the mid-thigh level. After excision, the patient was discharged with a patent graft and palpable pedal pulses. Two patients had patent grafts but inappropriately low ASP1 after reconstruction. Repeat postoperative angiograms showed residual valve cusps to be competent in one patient at the groin level and in another patient at the infrapopliteal level. Reexploration in both patients provided successful division of the competent valve cusps and the patients were discharged with patent grafts. Retained valves were not encountered when direct valve excision or incision of the valve cusp with scissors was used. All three retained valves occurred when the valve disruption was attempted using the Leather valvulotome. One patient returned 10 weeks postoperatively with a patent graft and fluctuating cyanosis of the toes. His ASP1 had been 0.94 at discharge and on follow-up visits varied irregularly from 0.60 to 0.90. Angiography showed a focal stenotic lesion 3 cm proximal to the popliteal anastomosis with a small mural thrombosis distal to the stenosis. At reexploration the external part of the vein appeared normal;
The American doumal of Surgery
Distal In Situ Vein Grafts
I’
3001%~ 1 .cI-
N-20
1SD
Nr30
x E 1 g 200t
g ii! 8 s m
5
f
0.6 , -
a f
loo-
t iz 0
L-
BASAL
-
VASODILATED
F&urn 3. Bloodflow thmughgrafl after arterkvemus flsMas wets l&tatd and technkal problems were ellminated. Mean basalgraft flow k 161 f 31 mum/n. Vasodilatkn with papaverlne (30 mg) into the common femoral artery results in a flow of 267 f 48 m/Lmin.
however, a tight stenotic, circumferential lesion was found, which was possibly related to a valve on the clamp site. This segment was resected and a 4 cm reversed saphenous vein interposition graft was inserted. The patient was discharged with systemic Doppler pressures and palpable pedal pulses. Comments The importance of the saphenous vein in arterial reconstruction below the knee is well established. For the past 15 years, use of a reversed segment of the saphenous vein has been the established technique for restoring flow below the knee after atherosclerotic occlusion of the superficial femoral artery and its branches. However, the reversed saphenous vein bypass to the infrapopliteal artery has been associated with a graft thrombosis rate of 28 to 38 percent at 1 month postoperatively and a 44 to 47 percent rate at 1 year postoperatively [7-101. The failure rates were lower at the distal popliteal artery level but were still 5 to 18 percent at 1 month and 17 to 34 percent at 1 year postoperatively [8,11,12]. After the first year, the vein grafts showed steady attrition (4 to 5 percent graft failures per year). This early failure rate of distal vein grafts has not decreased significantly despite advances in techniques for excising and preserving the vein grafts, improvements in sutures and vascular clamping techniques, and postoperative manipulation of thrombogenesis.
Volume 145, April 1993
0.5,-
r
PREOP
48HOURS POSTOP
LATE FOLLOWUP
Figure 4. Ankle systolic pressure Index (ASPI) measured preoperatively (0.27 f 0.73), 48 hours postoperative/y (0.83 f 0.27), and at last follow-up (0.89 f 0.27).
Revival of the in situ vein graft technique by Leather et al [3,4] provided the first significant decrease in early graft thrombosis. In addition to several theoretical advantages, two practical features were demonstrated. First, more saphenous veins could be utilized than could be with the traditional reversed technique. Second, the patency rates of the in situ grafts were far superior at both the distal popliteal and infrapopliteal artery levels. Since graft patency is the best insurance for limb salvage, the lack of an adequate saphenous vein for revascularization is a major loss to the patient. We observed that 16 of the 30 veins utilized would not have met most vascular surgeon’s criteria for an “adequate” saphenous vein. This mirrors the observation that 25 to 40 percent of patients who present with critical ischemia due to occlusion of the superficial artery do not have an adequate saphenous vein for use as a reversed femoropopliteal artery bypass graft [4,6,13]. The excellent patency and limb salvage rates in our series appear to confirm the clinical experience of Leather et al [3,4]. Although the follow-up period was relatively short in our study, the vein grafts remained patent throughout the period of highest risk (that is, the first year). The goal of reconstructive surgery should be to restore the circulatory capacity of the leg to normal. Among other features it should provide a conduit capable of supplying normal blood flow. Brismar et al [14] reported on the blood flow in the superficial femoral artery in patients without arterial insuffi-
545
Bush et al
Fwn?~ 5. Anglograms penbmwd 8 months afief cfeafhm of In situ femqosferior tibial bypass grafi. Lefl, fhs match between fhs pmximal saphenous vein and the femoral artery in the groin. Right, the vessel match between the distal saphenous vein and the dlstelposterlor t/b/al artery.
ciency while they were under anesthesia. The mean flow was 160 f 30 ml/min and increased to a maximum of 200 ml/min after volume loading with 900 ml of whole blood. Blood flow in reversed vein grafts has not achieved this level [15-211. However, our data indicated a mean flow of 161 f 31 ml/min, with a maximum increase to 267 f 48 ml/min after vasodilation with papaverine. Therefore, the in situ vein graft technique seems to restore a normal volume rate of blood flow to the leg. The reason is not clear from this study. Certainly the lack of technical errors, the more anastomotic matching (Figure 5), the improved endothelial preservation, and the capacity to produce prostacyclin are all possible causes. Whether patency is related to this improved graft flow as a causeand-effect relation is not known. However, the restoration of normal blood flow to the foot is a distinct advantage of the in situ technique over the reversed vein graft technique. In conclusion, the in situ vein graft bypass is an effective technique for distal reconstruction. However, it is not a panacea for a suboptimal surgical technique. The operation is tedious and marked by numerous changes for technical failure. The grafts must be monitored closely after operation. Aggressive diagnosis and intervention for residual defects has been required to avoid failure for purely technical reasons. However, when the in situ technique has been successfully applied, the results in terms of graft patency and limb salvage have been gratifying.
546
Summary Early failure remains a major obstacle to successful distal bypass surgery using vein grafts for limb salvage. Thirty distal bypass graft procedures were performed for limb salvage using the in situ technique. Grafts were anastomosed to the distal popliteal artery in 13 patients and to the infrapopliteal artery in 17 patients. Sixteen patients had inadequate saphenous veins for reversed vein grafts. The mean blood flow measured through these grafts (n = 20) was 164 f 22 ml/min and increased to 278 f 31 ml/min after administration of 30 mg of papaverine. All grafts were patent at the time of hospital discharge and patients were followed for 1 to 28 months. Life table analysis of the 30 procedures shows a patency of 100 percent at 18 months follow-up. One graft subsequently failed at 22 months. Long-term limb salvage was achieved in 100 percent of the patients in this series. The excellent blood flow through these grafts suggests that the in situ vein graft technique may be more favorable for arterial reconstruction than the reversed vein graft technique. Our preliminary data confirm the observations of Leather et al [3,4], that the rates of vein utilization and graft patency are higher with the in situ technique.
References 1. Buchbinder D, Singh JK, Karmody AM, Leather RP. Shah M. Comparison of patency rate and structural changes of in situ
The American Journal of Surgery
Distal
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12. 13.
14.
15.
18.
17. 18.
19.
20.
21.
and reversed vein arterial bypass. J Surg Res 1981;30: 213-22. Bush HL Jr, Hong SI, Deykin D, Nabseth DC. The effect of surgical trauma on prostacyclin production by vein grafts. Surg Forum 1982;33:483-5. Leather RP, Powers SR, Karmody AM. A reappraisal of the “in situ” saphenous vein arterial bypass: its use in limb salvage. Surgery 1979;86:453-60. Leather RP, Shah DM, Karmody AM. lnfrapopliteal arterial bypass for limb salvage: increased patency and utilization of the saphenous vein used “in situ.” Surgery 1981;90: 1000-7. Hall KV, Rostad H. In situ vein bypass in the treatment of femoropopliteal atherosclerotic disease: a ten year study. Am J Surg 1978;136:158-61. Mills NL, Oschner JL. Valvulotomy of valves in the saphenous vein graft before coronary artery bypass. J Thorac Cardiovast Surg 1978;71:878-79. Kacoyanis GP, Whittemore AD, Couch NP, Mannick JA. Femorotibial and femoroperoneal bypass vein grafts. Arch Surg 1981;116:1529-33. Szilagyi DE, Hageman JH, Smith RF, Elliott JP, Brown F, Dietz P. Autogenous vein grafting in femoropopliteal atherosclerosis: the limits of its effectiveness. Surgery 1979;86: 836-49. Reichle FA, Mattinson MW, Rankin KP. lnfrapopliteal arterial reconstructions in the severely ischemic lower extremity. Ann Surg 1980;191:59-65. lmparato AM, Kim GE, Madayag M, Haveson SP. The results of tibia1 artery reconstruction procedures. Surg Gynecol Obstet 1974;138:33-8. LoGerfo FW, Corson JD. Mannick JA. Improved results with femoropopliteal vein grafts for limb salvage. Arch Surg 1977;112:567-70. DeWeese JA, Rob CG. Autogenous venous grafts ten years later. Surgery 1977;82:775-84. O’Donnell JA, Brener BJ, Brief DK, Alpert J, Parsonnet V. Realistic expectations for patients having lower extremity bypass surgery for limb salvage. Arch Surg 1977;112: 1356-63. Brismar B, Cronestrand R, Forfeldt L, Tuhlin-Dannfelt A. Leg blood flow and central circulation at various blood volumes: a perioperative study of nine patients with varicose veins. Clin Sci Mol Med 1977;53:349-54. Barner HB. Judd DR, Kaiser GC, Willman VL, Hanlon CR. Blood flow in femoropopliteal bypass vein grafts. Arch Surg 1968;96:619-27. Bliss BP. Pressure, flow and peripheral resistance measurements during surgery for femoropopliteal occlusion. Stand J Clin Lab Invest 1973;31(suppl 128):179-83. Bernhard VM. lntraoperative monitoring of femorotibial bypass grafts. Surg Clin North Am 1974;54:77-84. Dedichen H. Prognostic significance of intraoperative flow and pressure measurements in reconstructive vascular surgery. Stand J Clin Lab Invest 1973:31(suooI 128):189-92. Mundt ED, Darling RC, Moran JM. Guantitative correlation of distal arterial outflow and patency of femoropopliteal reversed saphenous vein grafts with lntraoperative flow and pressure measurements. Surgery 1969;65:197-206. Sonnenfeld T, Cronestrand R. Factors determining outcome of reversed saphenous vein femoropopliteal bypass grafts. Br J Surg 1980;67:842-48. Terry HJ, Allan JS, Taylor GW. The relationship between blood flow and failure of femoropopliteal reconstructive arterial surgery. Br J Surg 1972;59:549-51.
Discussion William Abbott (Boston, MA): The in situ or nonreversed saphenous vein technique has four theoretic advantages: The first is geometric. With the nonreversed in situ method, there is a large-to-small taper from the proximal
Volume 145, April 1983
In Situ Vein Grafts
to the distal end, and the size or diameter match at the popliteal or tibia1 anastomosis is far more conducive to hemodynamic streamlining. The second theoretic advantage is that a longer length of vein will more frequently be available. Thus, the need to resort to alternative conduits, synthetics, or other materials is reduced. This factor has been borne out in the Albany experience [4] where in more than 90 percent of the researchers’ primary femoropopliteal reconstructions they were able to use vein. This is a much higher incidence than what has been reported from most other medical centers, my own included. The third and fourth theoretic advantages have to do with the fact that the vein may be better preserved. The in situ, nondissected vein may be less damaged and may more adequately retain endothelial cells. Since the vein is not dissected, its wall is not devascularized and therefore, there may be better retention of the mechanical properties of the vein; that is, its compliance or ability to transmit pulsatile energy may be better. Dr. Richard Cambria and I have demonstrated in an experimental model that the integrity of endothelial cells is better served by the less traumatic method of vein preparation associated with the in situ technique. Based on these advantages, Dr. David Brewster and I have been using the in situ method at the Massachusetts General Hospital over the past year and have performed about 20 operations using it. Our experience has been extremely favorable; the theoretic advantages seem to be real, and the technique is especially valuable for distal popliteal, tibial, and peroneal artery revascularizations. In fact, in my opinion it is the‘method of choice for these operations at this time. I caution, however, that it is technically demanding, and it has some pitfalls that are different from those experienced by the surgeon who is used to preparing veins by the dissection and reversal method. Nonetheless, I think that with the passage of time, the superiority of this technique will become evident and it will gain widespread acceptance.
John D. Carson (Albany, NY): The authors are to be commended on their excellent early results in this small series of in situ vein bypasses. Their initial high patency rate confirms the data in my own small personal experience with 12 in situ vein bypasses to the infrapopliteal vessels and 9 in situ vein bypasses to the popliteal artery performed since July 1981. Drs. Leather, Karmody, and myself now have a consecutive series of 349 in situ vein bypasses that we have followed from March 1976 to the present time. Two hundred one of these bypasses are to infrapopliteal vessels. The early high patency rate achieved by Dr. Bush and his colleagues is quite different from that achieved in most reported series of reversed vein bypass grafts where there has been a sharp decrease in the patency rate in the first 12 months. Their confirmation of the Albany data showing initially high patency rates, coupled with a vein utilization rate of 93 percent, should encourage the continued use of the in situ technique for limb salvage. Over the past 18 months we have been using a valve cutter to incise the valves in the vein from the groin to the knee level. The remainder of the valves are incised with either a valvulotome or valve scissors. The use of the cutter makes the in situ operation a little less tedious because the only valves that need to be visualized directly are those cut with the scissors or the valvulotome. In the total series to date a missed valve has been found in six patients in the immediate perioperative period, which required subse-
547
Bush et al
quent revision with excision of the offending valve. In these six patients the bypass grafts have been salvaged. In addition, we detected nine stenosis which developed in the in situ bypass grafts within 1 year of implantation. Close monitoring in the perioperative and early follow-up periods is needed to detect these problems. The 75 percent cummulative patency rate in the Albany series at 48 months should encourage the authors and others to pursue this exciting technique.
Charles Bucknam (Hartford, CT): It is difficult to ignore the success of those who have been using this technique. The principles espoused by the authors and the group in Albany are logical, and the procedure is fairly easy to learn. In the past year, four surgeons in Hartford with experience at the Hartford Hospital, Newington Veterans Administration Hospital, and the University of Connecticut Health Center have operated on 35 patients and carried out 37 graft procedures. Limb salvage was the indication in 22 patients and disabling claudication in 15. Most commonly, distal anastomoses were to the popliteal artery, with five to the anterior tibial, four to the posterior tibial, and two to the peroneal arteries. The patency rate was about 95 percent. There was one early occlusion which could not be repaired and one late occlusion. There were also three early complications which were corrected and there was essentially no early mortality. I have two questions for the authors. Is Dr. Bush comfortable with the angulation of the graft which sometimes is necessary to bring the in situ graft into the popliteal artery? This often is approximated a 90 degree angle which is greater than our usual angle when we are bringing the reversed vein through the anatomic position between the two heads of the gastrocnemius muscles. Also, do you recommend this technique when the vein appears to be of adequate diameter in a patient in whom it would seem to be quite easy to carry out the conventional reversed saphenous vein technique? John Mannick (Boston, MA): I am convinced that the authors have achieved fine results with this new technique for doing distal bypasses. The thing I object to is one of their explanations for its success, namely, increased rate
548
of blood flow. Flow characteristics may be improved, and that might be important in a distal anastomosis, but volume of flow should be equal in both directions for any given vein. The critical determinant of flow is the outflow bed. The size of the vein itself does not contribute much to the flow rate until the internal diameter decreases to about 3 mm.
Harry L. Bush, Jr. (closing): I would like to start in reverse with Dr. Mannick’s question and observation. Our observation is that there appears to be a surprisingly high rate of blood flow in the absence of arteriovenous fistula and in the presence of what is admittedly fairly poor runoff to the foot. There may be some theoretical reasons why the graft flow is better in the in situ rather than the reversed vein graft. It may be related to the vein diameter or to the characteristics of the vein wall. A reversed vein graft has poor endothelial coverage, and secondary platelet aggregation may release a significant amount of thromboxane into the distal arteriolar circulation. If this is true, thromboxane (a potent vasoconstrictor of the arteriolar system) may lead to the diminished intraoperative flow noted in reversed vein grafts. The in situ vein graft has both a better preserved endothelial monolayer and a higher rate of blood flow through the graft. But, I would agree that this increased flow may not be the reason why the graft remains patent. There were three patients in our series who had retained cusps evident postoperatively. Each of the three was reexplored with division of the cusp, and each had a patent graft at the time of discharge. At present, we are limiting this procedure to what I think are traditional indications for distal reconstruction, that is, people who have evidence of critical limb &hernia in terms of rest pain, ischemic ulcers, or tissue loss. Dr. Bucknam’s question about the angle of the popliteal artery is an important one. Increasingly, we have tried to bring the vein in at a more gradual angle, which may bring the vein into the tibioperoneal trunk and not into the main popliteal artery. At the present time, if a patient has a large-caliber saphenous vein, we would offer him an in situ technique as opposed to a reverse technique. It is tedious and it has many technical pitfalls, but it can be applied effectively to patients with critical &hernia.
The American Journal of Surgery
Harry L. Bush, Jr., MD, Boston, Massachusetts Christopher A. Corey, MD, Boston, Massachusetts Donald C. Nabseth, MD, Boston, Massachusetts
Thrombosis of vein grafts in the early postoperative period remains the major obstacle to graft patency and limb salvage in distal reconstruction of the leg. Surgeons have ascribed these early failures of reversed vein grafts to technical problems, such as inadequate inflow or outflow, inadequate vein size, poor quality of the vein wall, anastomotic or clamping techniques, hypercoagulability, or vessel mismatch at the anastomotic sites. However, increasing attention is being focused on the role of the endothelium in preventing graft thrombosis. Experimental evidence has demonstrated that the in situ vein graft technique preserves the morphologic integrity of the endothelium plus the metabolic capacity to produce prostacyclin, a potent inhibitor of platelet aggregation on a graft wall. These functional and morphologic benefits are distinctly abnormal in the reversed vein graft [I $1. Leather et al [3] have recently revitalized the in situ technique for arterial reconstruction of the leg in patients with critical ischemia. Their clinical reports document the dramatic clinical results with in situ vein grafts, especially in the infrapopliteal position [4]. The increased utilization of saphenous veins and the increased patency of these grafts allow more patients with advanced ischemia to have successful revascularization. The present series was initiated to corroborate the experience of Leather et al [3,4]. This report includes our entire experience with the in situ technique for revascularization of the distal popliteal and infrapopliteal arteries for limb salvage. From the Departments of Surgery, Boston Veterans Administration Medical Center and Tufts University School of Medicine, Boston, Massachusetts. Requests for reprints should be addressed to Harry L. Bush, Jr., MD. Dapartment of Surgery, Veterans Administration Medical Center, 150 South Huntington Avenue, Boston, Massachusetts 02130. Presented at the 63rd Annual Meeting of the New England Surgical Society, Bretton Woods, New Hampshire, Octobar 15-17, 1962.
542
Material
and Methods
Twenty-eight patients had a total of 30 bypass grafts to the distal popliteal or tibial arteries since December 1979. All patients were men and the mean age was 62.5 years (range 39 to 81 years). Associated cardiovascular risk factors included tobacco abuse in 26 patients, hypertension in 13, diabetes mellitus in 10, and angina or an old myocardial infarction in 8. All patients had critical ischemia with the leading indication for revascularization being ischemic ulceration in 10 limbs, cellulitis in 3 limbs, digital gangrene in 2 limbs, and pain at rest in the remaining 15 limbs. Angiography was performed in all patients and revealed adequate runoff to the ankle in 21 of the 30 limbs. In nine case additional angiograms were required early in the operative procedure to define the patency of distal blood vessels before reconstruction. Runoff was defined as the number of patent blood vessels connecting the anastomotic site with the pedal arch. The mean runoff was 1.2 f 0.1 vessels. The runoff was three vessels in 3 limbs, two vessels in 4 limbs, one vessel in 19 limbs, and no vessels in 4 limbs. At surgery, the common femoral artery and its branches, along with the saphenofemoral junction, were exposed through a vertical incision. The first 6 cm of the saphenous vein were dissected, the saphenous vein was divided at the saphenofemoral junction, and the femoral vein was closed with a running 5-O monofilament vascular suture. The first two to three valves were incised using scissors inserted through the orifice of the saphenous vein, the saphenous bulb was anastomosed to the common femoral or superficial femoral artery, and all clamps removed. The proximal anastomosis was performed to the common femoral artery in 6 limbs, to the junction of the common and superficial femoral arteries in 11 limbs, and to the proximal superficial femoral artery in 5 limbs. In addition, two anastomoses originated from the common and profunda femoral artery patch grafts and six from the hoods of prosthetic inflow grafts. The saphenous vein was exposed for its entire length, leaving the overlying superficial fascia intact. Competent valves in the arterialized vein could easily be detected and were rendered incompetent in one of three ways. The first nine operations were performed using Hall’s The American Journal of Surgery
Distal In Situ Vein Grafts
[5] technique for a transverse venotomy, direct valve excision, and closure with interrupted 7-O monofilament vascular sutures. Later, a scissor with a shaft that was 1 mm in diameter and 10 cm in length was custom-made. The size allowed introduction of the scissor through a side branch with direct incision of each valve cusp. In the past year, a valvulotome was inserted through the distal end of the saphenous vein and was used to disrupt the valve cusps [3,6]. To protect the vasa vasorum, the superficial fascia was incised only to ligate side branches of the vein. The quality of the saphenous veins varied greatly. Sixteen of the 30 saphenous veins were distinctly abnormal which probably would have precluded their use as a reversed vein graft. Eleven of these 16 veins included a segment of the graft with a diameter of less than 3.5 mm before surgical manipulation. These 11 patients included 4 with bifurcation of the saphenous vein in the thigh with two branches running parallel to each other and subsequently rejoining (usually near the knee) to form a common trunk. The other five saphenous veins had segments below the knee that had thick walls and diminished diameters. Pathologic examination confirmed a chronic inflammatory process. Two of the five patients required distal replacement with a short reversed vein graft. The other three patients’ saphenous veins were judged to be marginally adequate for use as a bypass conduit. The distal 6 to 7 cm of the saphenous vein was dissected free from the subcutaneous tissue and brought gently down to the recipient artery. The recipient artery was controlled using elastic Silastic@ loops. Anastomoses were performed with a running 6-O monofilament vascular suture in the popliteal artery and an interrupted 7-O monofilament suture in the tibia1 arteries. The distal anastomoses included 13 grafts to the distal (below the knee) popliteal artery, 3 to the tibioperoneal trunk, 8 to the posterior tibia1 artery, 4 to the anterior tibial artery, and 2 to the peroneal artery. Composite grafts were used in three bypass procedures. One patient had a 6 cm interposition prosthetic graft which ran from an aortofemoral limb to the proximal saphenous vein. Two patientshad overt phlebitic changes in the distal portion of the saphenous vein that was needed for bypass, therefore, the distal ends of their grafts were replaced with a reversed cephalic vein (8 cm in length) and a reversed saphenous vein (6 cm in length). After reconstruction, operative arteriograms were performed to detect retained valve cusps, arteriovenous fistulas, and other more standard technical problems (Figure 1). In addition, systolic Doppler pressures at the ankle was compared with the brachial systolic pressures to derive the ankle systolic pressure index (ASPI). Electromagnetic flowmeter methodology was used to measure mean rate of blood flow in the basal state and after vasodilation with papaverine (30 mg) injected into the common femoral artery. With the flow probe positioned in the groin, occlusion of the distal end of the vein graft caused the blood flow in the graft to fall to zero, thus confirming that all arteriovenous fistulas had been ligated. During the operative procedure, the patients were systemically heparinized initially and then were maintained with hourly bolus injections of heparin. In addition, infusion of low molecular weight dextran was started intraoperatively and maintained for 3 days at 25 ml/h. Qn the third postoperative day the patients began receiving aspirin, 300 mg/day orally. Patients were followed in the special vascular clinic by staff surgeons and 100 percent follow-up was obtained. Volume 145, April 1993
Flgure 1. Operatlve angiogram after artertallzatton of the sapheIK)us vetn. Two common pruMem5 am noted:a smal atterlove~~~ ftstula that f/Its the deep venous system and a paritalty competent retalned valve.
Results Patency of the vein graft was defined as a palpable pulse within the graft or a systemic Doppler pressure measured with the inflation cuff positioned over the distal graft. All 30 grafts were patent 30 days postoperatively. Five of the patients with patent grafts had reexploration for retained competent valves or graft thrombosis in the early postoperative period. Longitudinal follow-up revealed that all grafts remained patent except for one which failed 22 months after surgery (Figure 2). This failure was marked by the appearance of intermittent claudication and a decrease in the ASP1 to 0.63. Rest pain and ischemic ulceration did not develop before this patient’s death from a myocardial infarction 5 months after graft failure. Life table analysis showed a cumulative graft patency of 100 percent until 24 months postoperatively when the patency rate decreased to 82 percent for the 30 grafts. All 30 extremities had been salvaged at their last follow-up examination or at the time of death. One patient required digital amputations and one required a transmetatarsal amputation for existing gangrenous lesions present preoperatively. Three patients died between 6 and 12 weeks after surgery. Two of the patients died at home, one from a myocardial infarction and 1 from an acute pulmonary embolus, and the third died in the hospital from sepsis related to an ischemic hip disarticulation 543
0
L 1
12
6 TIME
(months)
wound on the contralateral side. A fourth patient died 27 months postoperatively from a myocardial infarction. After angiography confirmed the absence of arteriovenous fistulas or technical defects, electromagnetic flowmeter measurements in 20 grafts revealed a basal blood flow of 161 f 31 ml/min (Figure 3). Vasodilation of the distal arterial tree with 30 mg of papaverine resulted in a maximum increase in the flow rate to 267 f 48 ml/min. The lack of functional arteriovenous fistulas was confirmed by a decrease in graft flow to zero during occlusion of the distal end of the graft. The difference in graft flow was not significantly different when .distal popliteal grafts were compared with infrapopliteal grafts, nor when patients with diabetes mellitus were compared with nondiabetic patients. The mean value for the preoperative ASP1 was 0.27 f 0.13 in the 30 limbs (Figure 4). This rose to 0.85 f 0.21 within 48 hours after reconstruction. The ASP1 at last follow-up in all limbs was 0.89 f 0.27 (n = 26). The postoperative ASP1 was an invaluable monitor when the postoperative values were inappropriately low for the known runoff. Residual competent valve cusps in two of three patients and a late stenotic lesion in the graft of another patient were detected before thrombosis of the graft occurred. Early detection of the lesions allowed successful graft revision. Early in the series, two patients had immediate postoperative thrombosis of their grafts due to extrinsic compression. Both grafts were anastomosed to the distal popliteal artery and were allowed to lie directly behind the medial condyle of the tibia as the graft coursed from its subcutaneous position into the popliteal fossa. The patients had unusually bulky gastrocnemius muscles, the belly of that muscle
544
18 I
241
Ffgure 2. Ltfe table analysts of 30 In dtu grafts whkh depkts the patency rate for bekw-knee artertal reconstructlon.
compressed the graft against the boney prominence when the leg was extended. Both patients had successful thrombectomies of their grafts and relief of extrinsic pressure. No further technical abnormality was found on arteriograms obtained after thrombectomy and the patients were discharged with patent grafts. Three grafts were found to have residual competent valve leaflets, despite “normal” operative arteriograms immediately after reconstruction. One patient had palpable pedal pulses and systemic ankle Doppler pressures for 48 hours before his graft acutely thrombosed. After successful thrombectomy, the operative arteriogram showed a partially competent valve at the mid-thigh level. After excision, the patient was discharged with a patent graft and palpable pedal pulses. Two patients had patent grafts but inappropriately low ASP1 after reconstruction. Repeat postoperative angiograms showed residual valve cusps to be competent in one patient at the groin level and in another patient at the infrapopliteal level. Reexploration in both patients provided successful division of the competent valve cusps and the patients were discharged with patent grafts. Retained valves were not encountered when direct valve excision or incision of the valve cusp with scissors was used. All three retained valves occurred when the valve disruption was attempted using the Leather valvulotome. One patient returned 10 weeks postoperatively with a patent graft and fluctuating cyanosis of the toes. His ASP1 had been 0.94 at discharge and on follow-up visits varied irregularly from 0.60 to 0.90. Angiography showed a focal stenotic lesion 3 cm proximal to the popliteal anastomosis with a small mural thrombosis distal to the stenosis. At reexploration the external part of the vein appeared normal;
The American doumal of Surgery
Distal In Situ Vein Grafts
I’
3001%~ 1 .cI-
N-20
1SD
Nr30
x E 1 g 200t
g ii! 8 s m
5
f
0.6 , -
a f
loo-
t iz 0
L-
BASAL
-
VASODILATED
F&urn 3. Bloodflow thmughgrafl after arterkvemus flsMas wets l&tatd and technkal problems were ellminated. Mean basalgraft flow k 161 f 31 mum/n. Vasodilatkn with papaverlne (30 mg) into the common femoral artery results in a flow of 267 f 48 m/Lmin.
however, a tight stenotic, circumferential lesion was found, which was possibly related to a valve on the clamp site. This segment was resected and a 4 cm reversed saphenous vein interposition graft was inserted. The patient was discharged with systemic Doppler pressures and palpable pedal pulses. Comments The importance of the saphenous vein in arterial reconstruction below the knee is well established. For the past 15 years, use of a reversed segment of the saphenous vein has been the established technique for restoring flow below the knee after atherosclerotic occlusion of the superficial femoral artery and its branches. However, the reversed saphenous vein bypass to the infrapopliteal artery has been associated with a graft thrombosis rate of 28 to 38 percent at 1 month postoperatively and a 44 to 47 percent rate at 1 year postoperatively [7-101. The failure rates were lower at the distal popliteal artery level but were still 5 to 18 percent at 1 month and 17 to 34 percent at 1 year postoperatively [8,11,12]. After the first year, the vein grafts showed steady attrition (4 to 5 percent graft failures per year). This early failure rate of distal vein grafts has not decreased significantly despite advances in techniques for excising and preserving the vein grafts, improvements in sutures and vascular clamping techniques, and postoperative manipulation of thrombogenesis.
Volume 145, April 1993
0.5,-
r
PREOP
48HOURS POSTOP
LATE FOLLOWUP
Figure 4. Ankle systolic pressure Index (ASPI) measured preoperatively (0.27 f 0.73), 48 hours postoperative/y (0.83 f 0.27), and at last follow-up (0.89 f 0.27).
Revival of the in situ vein graft technique by Leather et al [3,4] provided the first significant decrease in early graft thrombosis. In addition to several theoretical advantages, two practical features were demonstrated. First, more saphenous veins could be utilized than could be with the traditional reversed technique. Second, the patency rates of the in situ grafts were far superior at both the distal popliteal and infrapopliteal artery levels. Since graft patency is the best insurance for limb salvage, the lack of an adequate saphenous vein for revascularization is a major loss to the patient. We observed that 16 of the 30 veins utilized would not have met most vascular surgeon’s criteria for an “adequate” saphenous vein. This mirrors the observation that 25 to 40 percent of patients who present with critical ischemia due to occlusion of the superficial artery do not have an adequate saphenous vein for use as a reversed femoropopliteal artery bypass graft [4,6,13]. The excellent patency and limb salvage rates in our series appear to confirm the clinical experience of Leather et al [3,4]. Although the follow-up period was relatively short in our study, the vein grafts remained patent throughout the period of highest risk (that is, the first year). The goal of reconstructive surgery should be to restore the circulatory capacity of the leg to normal. Among other features it should provide a conduit capable of supplying normal blood flow. Brismar et al [14] reported on the blood flow in the superficial femoral artery in patients without arterial insuffi-
545
Bush et al
Fwn?~ 5. Anglograms penbmwd 8 months afief cfeafhm of In situ femqosferior tibial bypass grafi. Lefl, fhs match between fhs pmximal saphenous vein and the femoral artery in the groin. Right, the vessel match between the distal saphenous vein and the dlstelposterlor t/b/al artery.
ciency while they were under anesthesia. The mean flow was 160 f 30 ml/min and increased to a maximum of 200 ml/min after volume loading with 900 ml of whole blood. Blood flow in reversed vein grafts has not achieved this level [15-211. However, our data indicated a mean flow of 161 f 31 ml/min, with a maximum increase to 267 f 48 ml/min after vasodilation with papaverine. Therefore, the in situ vein graft technique seems to restore a normal volume rate of blood flow to the leg. The reason is not clear from this study. Certainly the lack of technical errors, the more anastomotic matching (Figure 5), the improved endothelial preservation, and the capacity to produce prostacyclin are all possible causes. Whether patency is related to this improved graft flow as a causeand-effect relation is not known. However, the restoration of normal blood flow to the foot is a distinct advantage of the in situ technique over the reversed vein graft technique. In conclusion, the in situ vein graft bypass is an effective technique for distal reconstruction. However, it is not a panacea for a suboptimal surgical technique. The operation is tedious and marked by numerous changes for technical failure. The grafts must be monitored closely after operation. Aggressive diagnosis and intervention for residual defects has been required to avoid failure for purely technical reasons. However, when the in situ technique has been successfully applied, the results in terms of graft patency and limb salvage have been gratifying.
546
Summary Early failure remains a major obstacle to successful distal bypass surgery using vein grafts for limb salvage. Thirty distal bypass graft procedures were performed for limb salvage using the in situ technique. Grafts were anastomosed to the distal popliteal artery in 13 patients and to the infrapopliteal artery in 17 patients. Sixteen patients had inadequate saphenous veins for reversed vein grafts. The mean blood flow measured through these grafts (n = 20) was 164 f 22 ml/min and increased to 278 f 31 ml/min after administration of 30 mg of papaverine. All grafts were patent at the time of hospital discharge and patients were followed for 1 to 28 months. Life table analysis of the 30 procedures shows a patency of 100 percent at 18 months follow-up. One graft subsequently failed at 22 months. Long-term limb salvage was achieved in 100 percent of the patients in this series. The excellent blood flow through these grafts suggests that the in situ vein graft technique may be more favorable for arterial reconstruction than the reversed vein graft technique. Our preliminary data confirm the observations of Leather et al [3,4], that the rates of vein utilization and graft patency are higher with the in situ technique.
References 1. Buchbinder D, Singh JK, Karmody AM, Leather RP. Shah M. Comparison of patency rate and structural changes of in situ
The American Journal of Surgery
Distal
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12. 13.
14.
15.
18.
17. 18.
19.
20.
21.
and reversed vein arterial bypass. J Surg Res 1981;30: 213-22. Bush HL Jr, Hong SI, Deykin D, Nabseth DC. The effect of surgical trauma on prostacyclin production by vein grafts. Surg Forum 1982;33:483-5. Leather RP, Powers SR, Karmody AM. A reappraisal of the “in situ” saphenous vein arterial bypass: its use in limb salvage. Surgery 1979;86:453-60. Leather RP, Shah DM, Karmody AM. lnfrapopliteal arterial bypass for limb salvage: increased patency and utilization of the saphenous vein used “in situ.” Surgery 1981;90: 1000-7. Hall KV, Rostad H. In situ vein bypass in the treatment of femoropopliteal atherosclerotic disease: a ten year study. Am J Surg 1978;136:158-61. Mills NL, Oschner JL. Valvulotomy of valves in the saphenous vein graft before coronary artery bypass. J Thorac Cardiovast Surg 1978;71:878-79. Kacoyanis GP, Whittemore AD, Couch NP, Mannick JA. Femorotibial and femoroperoneal bypass vein grafts. Arch Surg 1981;116:1529-33. Szilagyi DE, Hageman JH, Smith RF, Elliott JP, Brown F, Dietz P. Autogenous vein grafting in femoropopliteal atherosclerosis: the limits of its effectiveness. Surgery 1979;86: 836-49. Reichle FA, Mattinson MW, Rankin KP. lnfrapopliteal arterial reconstructions in the severely ischemic lower extremity. Ann Surg 1980;191:59-65. lmparato AM, Kim GE, Madayag M, Haveson SP. The results of tibia1 artery reconstruction procedures. Surg Gynecol Obstet 1974;138:33-8. LoGerfo FW, Corson JD. Mannick JA. Improved results with femoropopliteal vein grafts for limb salvage. Arch Surg 1977;112:567-70. DeWeese JA, Rob CG. Autogenous venous grafts ten years later. Surgery 1977;82:775-84. O’Donnell JA, Brener BJ, Brief DK, Alpert J, Parsonnet V. Realistic expectations for patients having lower extremity bypass surgery for limb salvage. Arch Surg 1977;112: 1356-63. Brismar B, Cronestrand R, Forfeldt L, Tuhlin-Dannfelt A. Leg blood flow and central circulation at various blood volumes: a perioperative study of nine patients with varicose veins. Clin Sci Mol Med 1977;53:349-54. Barner HB. Judd DR, Kaiser GC, Willman VL, Hanlon CR. Blood flow in femoropopliteal bypass vein grafts. Arch Surg 1968;96:619-27. Bliss BP. Pressure, flow and peripheral resistance measurements during surgery for femoropopliteal occlusion. Stand J Clin Lab Invest 1973;31(suppl 128):179-83. Bernhard VM. lntraoperative monitoring of femorotibial bypass grafts. Surg Clin North Am 1974;54:77-84. Dedichen H. Prognostic significance of intraoperative flow and pressure measurements in reconstructive vascular surgery. Stand J Clin Lab Invest 1973:31(suooI 128):189-92. Mundt ED, Darling RC, Moran JM. Guantitative correlation of distal arterial outflow and patency of femoropopliteal reversed saphenous vein grafts with lntraoperative flow and pressure measurements. Surgery 1969;65:197-206. Sonnenfeld T, Cronestrand R. Factors determining outcome of reversed saphenous vein femoropopliteal bypass grafts. Br J Surg 1980;67:842-48. Terry HJ, Allan JS, Taylor GW. The relationship between blood flow and failure of femoropopliteal reconstructive arterial surgery. Br J Surg 1972;59:549-51.
Discussion William Abbott (Boston, MA): The in situ or nonreversed saphenous vein technique has four theoretic advantages: The first is geometric. With the nonreversed in situ method, there is a large-to-small taper from the proximal
Volume 145, April 1983
In Situ Vein Grafts
to the distal end, and the size or diameter match at the popliteal or tibia1 anastomosis is far more conducive to hemodynamic streamlining. The second theoretic advantage is that a longer length of vein will more frequently be available. Thus, the need to resort to alternative conduits, synthetics, or other materials is reduced. This factor has been borne out in the Albany experience [4] where in more than 90 percent of the researchers’ primary femoropopliteal reconstructions they were able to use vein. This is a much higher incidence than what has been reported from most other medical centers, my own included. The third and fourth theoretic advantages have to do with the fact that the vein may be better preserved. The in situ, nondissected vein may be less damaged and may more adequately retain endothelial cells. Since the vein is not dissected, its wall is not devascularized and therefore, there may be better retention of the mechanical properties of the vein; that is, its compliance or ability to transmit pulsatile energy may be better. Dr. Richard Cambria and I have demonstrated in an experimental model that the integrity of endothelial cells is better served by the less traumatic method of vein preparation associated with the in situ technique. Based on these advantages, Dr. David Brewster and I have been using the in situ method at the Massachusetts General Hospital over the past year and have performed about 20 operations using it. Our experience has been extremely favorable; the theoretic advantages seem to be real, and the technique is especially valuable for distal popliteal, tibial, and peroneal artery revascularizations. In fact, in my opinion it is the‘method of choice for these operations at this time. I caution, however, that it is technically demanding, and it has some pitfalls that are different from those experienced by the surgeon who is used to preparing veins by the dissection and reversal method. Nonetheless, I think that with the passage of time, the superiority of this technique will become evident and it will gain widespread acceptance.
John D. Carson (Albany, NY): The authors are to be commended on their excellent early results in this small series of in situ vein bypasses. Their initial high patency rate confirms the data in my own small personal experience with 12 in situ vein bypasses to the infrapopliteal vessels and 9 in situ vein bypasses to the popliteal artery performed since July 1981. Drs. Leather, Karmody, and myself now have a consecutive series of 349 in situ vein bypasses that we have followed from March 1976 to the present time. Two hundred one of these bypasses are to infrapopliteal vessels. The early high patency rate achieved by Dr. Bush and his colleagues is quite different from that achieved in most reported series of reversed vein bypass grafts where there has been a sharp decrease in the patency rate in the first 12 months. Their confirmation of the Albany data showing initially high patency rates, coupled with a vein utilization rate of 93 percent, should encourage the continued use of the in situ technique for limb salvage. Over the past 18 months we have been using a valve cutter to incise the valves in the vein from the groin to the knee level. The remainder of the valves are incised with either a valvulotome or valve scissors. The use of the cutter makes the in situ operation a little less tedious because the only valves that need to be visualized directly are those cut with the scissors or the valvulotome. In the total series to date a missed valve has been found in six patients in the immediate perioperative period, which required subse-
547
Bush et al
quent revision with excision of the offending valve. In these six patients the bypass grafts have been salvaged. In addition, we detected nine stenosis which developed in the in situ bypass grafts within 1 year of implantation. Close monitoring in the perioperative and early follow-up periods is needed to detect these problems. The 75 percent cummulative patency rate in the Albany series at 48 months should encourage the authors and others to pursue this exciting technique.
Charles Bucknam (Hartford, CT): It is difficult to ignore the success of those who have been using this technique. The principles espoused by the authors and the group in Albany are logical, and the procedure is fairly easy to learn. In the past year, four surgeons in Hartford with experience at the Hartford Hospital, Newington Veterans Administration Hospital, and the University of Connecticut Health Center have operated on 35 patients and carried out 37 graft procedures. Limb salvage was the indication in 22 patients and disabling claudication in 15. Most commonly, distal anastomoses were to the popliteal artery, with five to the anterior tibial, four to the posterior tibial, and two to the peroneal arteries. The patency rate was about 95 percent. There was one early occlusion which could not be repaired and one late occlusion. There were also three early complications which were corrected and there was essentially no early mortality. I have two questions for the authors. Is Dr. Bush comfortable with the angulation of the graft which sometimes is necessary to bring the in situ graft into the popliteal artery? This often is approximated a 90 degree angle which is greater than our usual angle when we are bringing the reversed vein through the anatomic position between the two heads of the gastrocnemius muscles. Also, do you recommend this technique when the vein appears to be of adequate diameter in a patient in whom it would seem to be quite easy to carry out the conventional reversed saphenous vein technique? John Mannick (Boston, MA): I am convinced that the authors have achieved fine results with this new technique for doing distal bypasses. The thing I object to is one of their explanations for its success, namely, increased rate
548
of blood flow. Flow characteristics may be improved, and that might be important in a distal anastomosis, but volume of flow should be equal in both directions for any given vein. The critical determinant of flow is the outflow bed. The size of the vein itself does not contribute much to the flow rate until the internal diameter decreases to about 3 mm.
Harry L. Bush, Jr. (closing): I would like to start in reverse with Dr. Mannick’s question and observation. Our observation is that there appears to be a surprisingly high rate of blood flow in the absence of arteriovenous fistula and in the presence of what is admittedly fairly poor runoff to the foot. There may be some theoretical reasons why the graft flow is better in the in situ rather than the reversed vein graft. It may be related to the vein diameter or to the characteristics of the vein wall. A reversed vein graft has poor endothelial coverage, and secondary platelet aggregation may release a significant amount of thromboxane into the distal arteriolar circulation. If this is true, thromboxane (a potent vasoconstrictor of the arteriolar system) may lead to the diminished intraoperative flow noted in reversed vein grafts. The in situ vein graft has both a better preserved endothelial monolayer and a higher rate of blood flow through the graft. But, I would agree that this increased flow may not be the reason why the graft remains patent. There were three patients in our series who had retained cusps evident postoperatively. Each of the three was reexplored with division of the cusp, and each had a patent graft at the time of discharge. At present, we are limiting this procedure to what I think are traditional indications for distal reconstruction, that is, people who have evidence of critical limb &hernia in terms of rest pain, ischemic ulcers, or tissue loss. Dr. Bucknam’s question about the angle of the popliteal artery is an important one. Increasingly, we have tried to bring the vein in at a more gradual angle, which may bring the vein into the tibioperoneal trunk and not into the main popliteal artery. At the present time, if a patient has a large-caliber saphenous vein, we would offer him an in situ technique as opposed to a reverse technique. It is tedious and it has many technical pitfalls, but it can be applied effectively to patients with critical &hernia.
The American Journal of Surgery