- Email: [email protected]
In Situ Bypass to the Dorsalis Pedis and Tibial Arteries at the Ankle
In Situ Bypass to the Dorsalis Pedis and Tibia1 Arteries at the Ankle Arnold W. Levine, MD, Robert C. Davis, MD, Robert 0. Gingery, MD, Denise D. Anderegg, PAC, Oakland, California
Over a three-year period (1984-1987), we performed 98 in situ saphenous vein bypasses to the tibial arteries in 89 patients. In 68 of these grafts, the distal anastomoses were above ankle level. All but one of these grafts were performed for limb salvage indications. Seven (10%) of these grafts were done in patients on chronic renal dialysis. In 30 grafts, the distal anastomoses were to the ankle level. Indications for surgery were risk of limb loss in all of these patients, with tissue loss in 29 (97%). Six (20%) were done in patients on chronic renal dialysis. Operative mortality was 3% in both groups. Late mortality was 13% in the above-ankle group, and 27% in the ankle level group. Secondary patency for the above-ankle group was 97%, 85%, 8l%, and 81% at 30 days, one year, two years, and three years. Primary patency was 91%,67%, and 58% at 30 days, one year, and two years, after which the standard error is greater than 10%. Secondary patency for the grafts at the ankle level was 100% and 82% at 30 days and one and one-half years. Primary patency rates were 93% and 68% at 30 days and one year. In situ bypass grafts at the ankle level had patency rates equivalent to grafts with distal anastomoses above the ankle. Patients with distal bypasses usually presented with tissue loss and had a higher late mortality rate. Careful follow-up and operative intervention when changes in graft velocities or indices are recognized, markedly improves the durability of the in situ saphenous vein bypass. KEY WORDS: Arteries, tibial; arteries, dorsalis pedis; saphenous vein bypass.
Severe tibial occlusive disease, with or without concomitant femoral-popliteal occlusive disease, presents a unique challenge to the vascular surgeon. Often, the only arteries suitable for bypass are at the ankle level or below. Runoff is poor and a long, nonthrombogenic conduit is required to attain limb salvage. The in situ saphenous vein has theoretical advantages that make it an attractive conduit in From General Vascular Surgery Medical Group, Inc., Oakland, California. Presented at the Annual Meeting of the Peripheral Vascular Surgery Society, June 12, 1988, Chicago, Illinois. Reprint requests: Arnold W . Levine, MD, 2940 Webster Street, Oakland, California 94609. 205
these difficult bypass situations. Vein-artery size match at anastomotic sites and the ability to utilize vein segments down to two millimeters in internal diameter simplifies the technical demands of the operation. The controlled distention of the vein, inherent in the technique and absence of endothelial ischemia may play a role in resistance to thrombosis. A three-year experience of in situ vein bypasses to the tibial arteries, done in a private practice setting, is reviewed with special attention to grafts to the dorsalis pedis and tibial arteries at ankle level. Primary and secondary patency rates and patient characteristics are compared in grafts above the ankle and grafts to the ankle level.
206
IN SITW BYPASS TO T H E DORSALIS PEDlS
ANNALSOF VASCULAR SURGERY
was used for identification of missed arteriovenous fistula, incompletely lysed valve leaflets, and anasNinety-eight tibial bypasses were performed in 89 tomotic technical error prior to completion arteripatients. This includes our first through our ninety- ography. After the perioperative period, patients were foleighth consecutive in situ tibial bypass. Absence of a suitable outflow vessel on standard aortofemoral lowed at three-month intervals. Ankle-brachial arteriography did not exclude the patient unless an Doppler indices were obtained at each office visit. operative arteriogram or vessel exploration con- Graft velocity measurements [2] were obtained firmed the arteriogram findings. Patients without a when graft length went beyond the top of the blood relatively nonsclerotic in situ saphenous vein at pressure cuff at the ankle, when the distal anastoleast 2 mm in diameter were excluded from this mosis was to the peroneal artery or to isolated study and usually had grafts placed with contralat- segments of tibial arteries, or if a minor reduction in era1 saphenous or cephalic veins. Doppler index (< .15) was recorded and its signifiThe indications for surgery in the 68 limbs with cance was uncertain. Grafts with ankle Doppler anastomoses above the ankle (69% of all grafts) in indices that decrease 0.15 or more, or with graft 60 patients, were ischemic rest pain in 25 (37%), velocities less than 45 cm/sec, were evaluated by ischemic ulceration in 22 (32%). and gangrene in 21 duplex ultrasound. Grafts with flow defects to ac(31%). Forty-two (62%) grafts were done in limbs of count for the changes in these parameters were diabetic patients, 53 (78%) in patients who had a repaired surgically, usually without preoperative positive smoking history, and seven (10%) in pa- arteriography. Graft patency rates were determined tients who were on chronic renal dialysis. by life-table method [3]. Indications for surgery in the 30 limbs with distal anastomoses to the ankle level were ischemic rest pain in one (3%), ischemic ulceration in 10 (33%) RESULTS and gangrene in 19 (64%). Therefore, 97% of these grafts were performed for tissue loss. Twenty-four Three deaths (3%) occurred within 30 days of (80%) of these grafts were done in limbs of diabetic surgery. One of these patients had a bypass to the patients, 16 (53%) in patients who had a cigarette ankle level (3%). Two of the deaths were related to smoking history, and six (20%) in patients who were cardiac complications and the cause of the third is on chronic renal dialysis. unknown. During the late follow-up period, 16 In the grafts with distal anastomoses above the (16%) patients died. Seven (23%) were patients with ankle level, 67 (98%) had the proximal anastomoses ankle level grafts and nine (13%) of these were in the femoral area, either the common femoral, the patients with distal anastomoses above ankle level. Operative complications included three myocarproximal deep femoral, the proximal superficial femoral arteries or to the distal end of an inflow dial infarctions and two cardiac arrythmias. Three bypass to the femoral artery. One (1%) originated patients with chronic renal failure prior to surgery from the distal popliteal artery. The distal anasto- had worsening of their renal function in the postopmoses were to the posterior tibial in 32 (47%) grafts, erative period and required temporary dialysis. One anterior tibial artery in six (9%), peroneal artery in patient developed transient cerebral ischemia and was treated with carotid endarterectomy, and an29 (43%), and tibial peroneal trunk in one (1%). In the 30 grafts with bypasses to the ankle level, other patient developed acute cholecystitis and conthe proximal anastomosis was in the femoral region gestive heart failure. She underwent cholecystecin 25 (83%) grafts. tomy. Another patient developed pneumonia in the Five (17%) of these bypasses originated from the postoperative period that was successfully treated distal popliteal artery. The distal anastomoses were with antibiotics. Wound-edge necrosis occurred after I7 operaeither to the dorsalis pedis artery in 18 (60%) grafts, to the posterior tibial artery within 5 cm at the ankle tions (17%). Fourteen of these were noted in diajoint space in five (30%) grafts, or the distal 5 cm of betic patients, five of whom were dialysis patients. the anterior tibial artery in three (10%). Ankle level grafts accounted for eight (26%) of The operative technique has been previously de- these complications, and grafts above the ankle scribed [I]. Modification of the technique for this only accounted for nine (13%). Two patients develseries includes the use of the forward cutting val- oped hematomas that required drainage. vulotome for valve lysis in addition to the Karmody One wound infection resulted in limb loss. A vein valve scissors and retrograde valvulotome. diabetic patient developed hemorrhage from his Vessel occlusion for the construction of the distal autogenous profundaplasty which was done at the anastomosis was accomplished with Jacobsen bull- time of his bypass. He was found to have a candida dog clamps, vessel loops, or by pneumatic tourni- wound infection and he refused amphotericin therquet at the discretion of the operating surgeon. apy. He underwent debridement of his profunda Continuous-wave Doppler without spectral analysis femoris artery. With a patent femoral to midposteMATERIAL AND METHODS
VOLUME3
I N SITU BYPASS TO THE DORSALIS PEDIS
NO 3 - 1989
TABLE I.-Primary
207
and secondary life-table patency data of 98 tibial in situ saphenous vein bypasses
Intervals Primary patency 0-30 days 30 days-6 mos 6 mos-1 year 1-1 Ih years 1 l1-2 years 2-2% years 2%3 years Secondary patency 0-30 days 30 days-6 mos 6 mos-1 year 1-1 M years 2 ' 11 1 years 2-2112 years 3'112 years
No. grafts at risk
No. grafts failing
Death
Withdrawn'
98 83 45 23 13 9 3
8 17 1 3 1 1 0
3 6 3 1 1 0 0
98 89 57 31 19 11
2 6 3 1 1 0
5
0
3 6 4 1 2 0 0
Interval failure rate (%)
Interval patency rate
4 15 18 6 2 5 3
8.4 23.4 2.9 15.4 8.7 15.4
4 19 20 10 5 6 4
("4
Cumulative patency rate (?/.)
Standard patency rate (%)
0
91.6 76.6 97.1 84.6 91.3 84.6 100
91.6 70.2 68.0 57.0 52.0 44.0 44.0
2.9 5.1 5.3 7.1 8.0 10.0 10.0
2.1 7.8 6.7 3.9 6.5 0 0
97.9 92.2 93.3 96.1 93.5 100 100
97.9 90.3 84.2 80.9 75.6 75.6 75.6
1.5 3.3 4.6 5.5 7.2 7.2 7.2
'Reasons for withdrawal of grafts were duration and loss of follow-up
rior tibial bypass, he went on to necrose his proxiLate revisions (> 30 days) were required to mal thigh muscles and required a hip disarticulation maintain graft patency or salvage occluded grafts five weeks after his tibial bypass. This is considered when the patients reported soon after their occlua failed graft in the life tables. sions in 20 grafts. Six of these were ankle level Three grafts required revision for disappointing grafts. Twenty revisions in 16 patent grafts conhemodynamic results in the early postoperative sisted of vein patch angioplasty (lo), vein segment period (30 days). One patient had replacement of a replacement (21, jump graft to distal artery ( 2 ) , vein small vein segment with reversed vein and another patch angioplasty and ligation of arteriovenous fishad ligation of arteriovenous fistula. The third pa- tula ( l ) , ligation of arteriovenous fistula alone (3), tient had exploration of the distal anastomosis and iliofemoral bypass ( I ) , and femorofemoral bypass dilatation of the dorsalis pedis artery distal to the (1). These inflow procedures were done to improve anastomosis when spasm was identified. Four grafts the flow of in situ grafts, not to treat limb-threatoccluded in the first 24 hours and were salvaged by ening ischemia. Four occluded grafts were salvaged thrombectomy alone in one patient and thrombec- (maintained patency for at least 30 days) with tomy and vein patch angioplasty or replacement of thrombectomy plus vein angioplasty ( 2 ) , replacesmall vein segments in three patients. The patient ment of vein segment ( 2 ) , or jump graft to a distal who only required thrombectomy had a preopera- artery (1). tive platelet count of greater than one million. She Primary and secondary patency rates are preunderwent chemotherapy for her myeloprolifera- sented in Tables I. I1 and 111. Table I reports all tive disease and maintained graft patency. grafts, Table I1 grafts to ankle level and Table I11
TABLE 11.-Primary
Intervals Primary patency 0-30 days 1-6 mos 6 mos-1 year Secondary patency 0-30 days
1-6 mos 6 mos-1 year 1-1 112 years
and secondary life-table patency data of 30 tibial in situ saphenous vein bypasses to ankle level and below
No. grafts at risk
No. grafts failing
Death
30 25 13
2 6 0
1 1 1
2 5 6
7.0 27.3 0
93.0 72.7 100
93.0 67.6 67.6
4.8 9.5 9.5
30 26 15 6
0 2 1 0
1 1
3 8 6 2
0 9.3 9.1 0
100 90.7 90.9 100
100 90.7 82.5 82.5
0 6.3 9.7 9.7
2
1
'Reason for withdrawal of grafts were duration and loss of follow-up
Interval failure rate Withdrawn* (Yo)
Interval patency rate
(W
Cumulative patency rate (%)
Standard patency rate (%)
208
ANNALSOF VASCULAR SURGERY
IN SITU BYPASS TO THE DORSALIS PEDIS
TABLE Ill.-Primary
Intervals Primary patency 0-30 days 1-6 rnos 6 rnos-1 year 1-1 VZ years 1l/L-2 years Secondary patency 0-30 days 1-6 rnos 6 rnos-1 year 1-1 1/2 years 1%2 years 2-2% years 2%3 years
and secondary life-table patency data of 68 above ankle level tibial in situ saphenous vein bypasses No. grafts at risk
No. grafts failina
Interval failure rate Withdrawn'
(%I
Interval patency rate 1%)
Cumulative patency rate (%I
Standard patency rate (%)
Death
68 59 32 17 10
6 12 1 2 0
2 5 2 0 1
1 10 12 5 1
9.0 23.3 4.0 13.8
91.o 76.7 96.0 86.2 100
91.o 69.8 67.0 57.8 57.8
3.5 6.0 6.4 8.2 8.2
68 63 43 25 16 11 5
2 4 2 1 0 0 0
2 5 2 0 1 1 0
1 11 14 8 4 5 5
3.0 7.3 5.7 4.8 0 0
97.0 92.7 94.3 95.2 100 100 100
97.0 89.9 84.8 80.8 80.8 80.8 80.8
2.1 3.9 5.1 6.3 6.3 6.3 6.3
0
0
'Reasons for withdrawal of grafts were duration and loss of follow-up
reports grafts above ankle level. Figure 1 compares the above ankle level grafts to the ankle level bypasses. Two of the grafts listed as failures in the 30 day to six month interval were patent grafts that
100%.
0 Craftsaboveankle 1 0
+ Grafts above ankle
20
were sacrificed during amputation. One of these amputations was due to more extensive foot necrosis than was appreciated at the time of bypass, and the other was caused by complications resulting from a candida wound infection as discussed previously. All successful grafts resulted in limb salvage, at least to the transmetatarsal level, except one patient who had a Symes amputation after her open transmetatarsal amputation failed.
W Craftstoankle 10 0 Grafts to ankle 20
DISCUSSION
90%-
80%-
70%-
60%-
50%-
1
I
I
lmo.
I
1
6mo.
llr.
I
I
I
1Hyr.
Ar.
2Hyr.
3yr.
Fig. 1. Comparison of patenc rates of grafts to ankle level and below and grafts above ankle level. (Numbers listed are patent grafts at end of interval.)
The patency data in this series demonstrates equivalent results when comparing grafts to the dorsalis pedis and tibial arteries at ankle level to grafts above the ankle. This supports an aggressive approach to limb salvage in the patient with either no distal vessels suitable for bypass on the preoperative arteriogram or vessels that are occluded to ankle level. It also suggests that bypass to the dorsalis pedis or tibial arteries at ankle level may be preferable to a compromised tibial outflow, such as an isolated segment tibial artery or a peroneal artery with poor collateralization to the foot, if adequate vein is available. The marked improvement in secondary over primary patency rates emphasizes the importance of careful and frequent follow-up to improve the durability of these operations. We agree with the conclusions of Bandyk and associates [4] that a protocol of graft surveillance, early revision of hemodynamically abnormal but patent grafts, and an aggressive approach to the recently thrombosed graft is rewarded by excellent secondary patency. Wound edge necrosis and infection complicated a substantial number (17%) of these operations and
VOLUME 3 NO 3 - 1989
IN SITU BYPASS TO THE DORSALIS PEDIS
occurred more often in patients with ankle level grafts. This may represent the poor healing potential of chronically ischemic skin in the diabetic patient and the dialysis patient (N = 13) who had a 38% incidence of this complication. Skin flaps made by placing the incision away from the vein may have contributed. In the latter part of this series, we have used preoperative duplex ultrasound vein mapping in an effort to reduce this problem. Although operative mortality was equal in the two groups of patients, later mortality was higher in the patients who required ankle level grafts. They presented with more advanced ischemia (tissue loss in 97%), and there was a higher incidence of diabetes mellitus and chronic renal failure requiring dialysis in this group. This may indicate that this group selects patients who are at a more advanced stage of complications from long-term diabetes mellitus. Buchbinder and colleagues [ 5 ] demonstrated excellent results with the in situ saphenous vein as a conduit for “femoral-foot bypass” defined as “to, or below, malleolar level.” In their series, the in situ vein performed better than reversed vein and polytetrafluoroethylene. They also noted a 10% lower limb salvage rate than patency rate. We had two patients in our series that had major amputations in limbs with patent grafts (considered failed grafts in the life tables), although they were not in the population with ankle level grafts. An additional patient who was diabetic and on dialysis had a graft to the dorsalis pedis artery that remained patent for six weeks. In that time period, she demonstrated no signs of healing her transmetatarsal amputation or her distal surgical wounds. Her graft occluded after the decision was made to amputate, but before it was done. This graft is listed as a failed graft in the 30 day to six month interval. All of the other
209
patients with ankle level grafts healed their ulcers or minor amputations. CONCLUSION In situ saphenous vein grafts to the dorsalis pedis and tibial arteries at the ankle demonstrate patency rates equal to grafts above the ankle level, and have provided limb salvage to a substantial number of patients with severe tibial occlusive disease. More of these patients with ankle level outflow have diabetes mellitus or chronic renal failure requiring dialysis. Wound healing is a greater problem in these patients. They also present more often with tissue loss and suffer a greater late mortality. Graft surveillance and early revision improves the durability of these tibial and ankle level grafts. The in situ saphenous vein is our conduit of choice for the difficult bypass situations to vessels at the ankle level or below. REFERENCES LEVINE AW, BANDYK DF, BONIER PH, TOWNE JB. Lessons learned in adopting the in situ saphenous vein bypass. J Vusc Surg 1985;2: 145-153. BANDYK DF, C A T 0 RF, TOWNE JB. A low flow velocity predicts failure of femoropopliteal and femorotibial bypass grafts. Surgery 1985;98:799-809. CUTLER SJ, EDERER F. Maximum utilization of the life table method in analyzing survival. J Chronic Dis 1958; 81699-7 I 2. BANDYK DF, KAEBNICK HW, STEWARTGW, TOWNE JB. Durability of the in situ saphenous vein arterial bypass: a comparison of primary and secondary patency. J Vusc Surg 1987 5 :256-268. BUCHBINDER D. PASCH AR, ROLLINS DL, DILLON BC, DOUGLAS DJ, SCHULER J J , FLANIGAN DP. Results of arterial reconstruction of the foot. Arch Surg 1986:121:673-677.
Over a three-year period (1984-1987), we performed 98 in situ saphenous vein bypasses to the tibial arteries in 89 patients. In 68 of these grafts, the distal anastomoses were above ankle level. All but one of these grafts were performed for limb salvage indications. Seven (10%) of these grafts were done in patients on chronic renal dialysis. In 30 grafts, the distal anastomoses were to the ankle level. Indications for surgery were risk of limb loss in all of these patients, with tissue loss in 29 (97%). Six (20%) were done in patients on chronic renal dialysis. Operative mortality was 3% in both groups. Late mortality was 13% in the above-ankle group, and 27% in the ankle level group. Secondary patency for the above-ankle group was 97%, 85%, 8l%, and 81% at 30 days, one year, two years, and three years. Primary patency was 91%,67%, and 58% at 30 days, one year, and two years, after which the standard error is greater than 10%. Secondary patency for the grafts at the ankle level was 100% and 82% at 30 days and one and one-half years. Primary patency rates were 93% and 68% at 30 days and one year. In situ bypass grafts at the ankle level had patency rates equivalent to grafts with distal anastomoses above the ankle. Patients with distal bypasses usually presented with tissue loss and had a higher late mortality rate. Careful follow-up and operative intervention when changes in graft velocities or indices are recognized, markedly improves the durability of the in situ saphenous vein bypass. KEY WORDS: Arteries, tibial; arteries, dorsalis pedis; saphenous vein bypass.
Severe tibial occlusive disease, with or without concomitant femoral-popliteal occlusive disease, presents a unique challenge to the vascular surgeon. Often, the only arteries suitable for bypass are at the ankle level or below. Runoff is poor and a long, nonthrombogenic conduit is required to attain limb salvage. The in situ saphenous vein has theoretical advantages that make it an attractive conduit in From General Vascular Surgery Medical Group, Inc., Oakland, California. Presented at the Annual Meeting of the Peripheral Vascular Surgery Society, June 12, 1988, Chicago, Illinois. Reprint requests: Arnold W . Levine, MD, 2940 Webster Street, Oakland, California 94609. 205
these difficult bypass situations. Vein-artery size match at anastomotic sites and the ability to utilize vein segments down to two millimeters in internal diameter simplifies the technical demands of the operation. The controlled distention of the vein, inherent in the technique and absence of endothelial ischemia may play a role in resistance to thrombosis. A three-year experience of in situ vein bypasses to the tibial arteries, done in a private practice setting, is reviewed with special attention to grafts to the dorsalis pedis and tibial arteries at ankle level. Primary and secondary patency rates and patient characteristics are compared in grafts above the ankle and grafts to the ankle level.
206
IN SITW BYPASS TO T H E DORSALIS PEDlS
ANNALSOF VASCULAR SURGERY
was used for identification of missed arteriovenous fistula, incompletely lysed valve leaflets, and anasNinety-eight tibial bypasses were performed in 89 tomotic technical error prior to completion arteripatients. This includes our first through our ninety- ography. After the perioperative period, patients were foleighth consecutive in situ tibial bypass. Absence of a suitable outflow vessel on standard aortofemoral lowed at three-month intervals. Ankle-brachial arteriography did not exclude the patient unless an Doppler indices were obtained at each office visit. operative arteriogram or vessel exploration con- Graft velocity measurements [2] were obtained firmed the arteriogram findings. Patients without a when graft length went beyond the top of the blood relatively nonsclerotic in situ saphenous vein at pressure cuff at the ankle, when the distal anastoleast 2 mm in diameter were excluded from this mosis was to the peroneal artery or to isolated study and usually had grafts placed with contralat- segments of tibial arteries, or if a minor reduction in era1 saphenous or cephalic veins. Doppler index (< .15) was recorded and its signifiThe indications for surgery in the 68 limbs with cance was uncertain. Grafts with ankle Doppler anastomoses above the ankle (69% of all grafts) in indices that decrease 0.15 or more, or with graft 60 patients, were ischemic rest pain in 25 (37%), velocities less than 45 cm/sec, were evaluated by ischemic ulceration in 22 (32%). and gangrene in 21 duplex ultrasound. Grafts with flow defects to ac(31%). Forty-two (62%) grafts were done in limbs of count for the changes in these parameters were diabetic patients, 53 (78%) in patients who had a repaired surgically, usually without preoperative positive smoking history, and seven (10%) in pa- arteriography. Graft patency rates were determined tients who were on chronic renal dialysis. by life-table method [3]. Indications for surgery in the 30 limbs with distal anastomoses to the ankle level were ischemic rest pain in one (3%), ischemic ulceration in 10 (33%) RESULTS and gangrene in 19 (64%). Therefore, 97% of these grafts were performed for tissue loss. Twenty-four Three deaths (3%) occurred within 30 days of (80%) of these grafts were done in limbs of diabetic surgery. One of these patients had a bypass to the patients, 16 (53%) in patients who had a cigarette ankle level (3%). Two of the deaths were related to smoking history, and six (20%) in patients who were cardiac complications and the cause of the third is on chronic renal dialysis. unknown. During the late follow-up period, 16 In the grafts with distal anastomoses above the (16%) patients died. Seven (23%) were patients with ankle level, 67 (98%) had the proximal anastomoses ankle level grafts and nine (13%) of these were in the femoral area, either the common femoral, the patients with distal anastomoses above ankle level. Operative complications included three myocarproximal deep femoral, the proximal superficial femoral arteries or to the distal end of an inflow dial infarctions and two cardiac arrythmias. Three bypass to the femoral artery. One (1%) originated patients with chronic renal failure prior to surgery from the distal popliteal artery. The distal anasto- had worsening of their renal function in the postopmoses were to the posterior tibial in 32 (47%) grafts, erative period and required temporary dialysis. One anterior tibial artery in six (9%), peroneal artery in patient developed transient cerebral ischemia and was treated with carotid endarterectomy, and an29 (43%), and tibial peroneal trunk in one (1%). In the 30 grafts with bypasses to the ankle level, other patient developed acute cholecystitis and conthe proximal anastomosis was in the femoral region gestive heart failure. She underwent cholecystecin 25 (83%) grafts. tomy. Another patient developed pneumonia in the Five (17%) of these bypasses originated from the postoperative period that was successfully treated distal popliteal artery. The distal anastomoses were with antibiotics. Wound-edge necrosis occurred after I7 operaeither to the dorsalis pedis artery in 18 (60%) grafts, to the posterior tibial artery within 5 cm at the ankle tions (17%). Fourteen of these were noted in diajoint space in five (30%) grafts, or the distal 5 cm of betic patients, five of whom were dialysis patients. the anterior tibial artery in three (10%). Ankle level grafts accounted for eight (26%) of The operative technique has been previously de- these complications, and grafts above the ankle scribed [I]. Modification of the technique for this only accounted for nine (13%). Two patients develseries includes the use of the forward cutting val- oped hematomas that required drainage. vulotome for valve lysis in addition to the Karmody One wound infection resulted in limb loss. A vein valve scissors and retrograde valvulotome. diabetic patient developed hemorrhage from his Vessel occlusion for the construction of the distal autogenous profundaplasty which was done at the anastomosis was accomplished with Jacobsen bull- time of his bypass. He was found to have a candida dog clamps, vessel loops, or by pneumatic tourni- wound infection and he refused amphotericin therquet at the discretion of the operating surgeon. apy. He underwent debridement of his profunda Continuous-wave Doppler without spectral analysis femoris artery. With a patent femoral to midposteMATERIAL AND METHODS
VOLUME3
I N SITU BYPASS TO THE DORSALIS PEDIS
NO 3 - 1989
TABLE I.-Primary
207
and secondary life-table patency data of 98 tibial in situ saphenous vein bypasses
Intervals Primary patency 0-30 days 30 days-6 mos 6 mos-1 year 1-1 Ih years 1 l1-2 years 2-2% years 2%3 years Secondary patency 0-30 days 30 days-6 mos 6 mos-1 year 1-1 M years 2 ' 11 1 years 2-2112 years 3'112 years
No. grafts at risk
No. grafts failing
Death
Withdrawn'
98 83 45 23 13 9 3
8 17 1 3 1 1 0
3 6 3 1 1 0 0
98 89 57 31 19 11
2 6 3 1 1 0
5
0
3 6 4 1 2 0 0
Interval failure rate (%)
Interval patency rate
4 15 18 6 2 5 3
8.4 23.4 2.9 15.4 8.7 15.4
4 19 20 10 5 6 4
("4
Cumulative patency rate (?/.)
Standard patency rate (%)
0
91.6 76.6 97.1 84.6 91.3 84.6 100
91.6 70.2 68.0 57.0 52.0 44.0 44.0
2.9 5.1 5.3 7.1 8.0 10.0 10.0
2.1 7.8 6.7 3.9 6.5 0 0
97.9 92.2 93.3 96.1 93.5 100 100
97.9 90.3 84.2 80.9 75.6 75.6 75.6
1.5 3.3 4.6 5.5 7.2 7.2 7.2
'Reasons for withdrawal of grafts were duration and loss of follow-up
rior tibial bypass, he went on to necrose his proxiLate revisions (> 30 days) were required to mal thigh muscles and required a hip disarticulation maintain graft patency or salvage occluded grafts five weeks after his tibial bypass. This is considered when the patients reported soon after their occlua failed graft in the life tables. sions in 20 grafts. Six of these were ankle level Three grafts required revision for disappointing grafts. Twenty revisions in 16 patent grafts conhemodynamic results in the early postoperative sisted of vein patch angioplasty (lo), vein segment period (30 days). One patient had replacement of a replacement (21, jump graft to distal artery ( 2 ) , vein small vein segment with reversed vein and another patch angioplasty and ligation of arteriovenous fishad ligation of arteriovenous fistula. The third pa- tula ( l ) , ligation of arteriovenous fistula alone (3), tient had exploration of the distal anastomosis and iliofemoral bypass ( I ) , and femorofemoral bypass dilatation of the dorsalis pedis artery distal to the (1). These inflow procedures were done to improve anastomosis when spasm was identified. Four grafts the flow of in situ grafts, not to treat limb-threatoccluded in the first 24 hours and were salvaged by ening ischemia. Four occluded grafts were salvaged thrombectomy alone in one patient and thrombec- (maintained patency for at least 30 days) with tomy and vein patch angioplasty or replacement of thrombectomy plus vein angioplasty ( 2 ) , replacesmall vein segments in three patients. The patient ment of vein segment ( 2 ) , or jump graft to a distal who only required thrombectomy had a preopera- artery (1). tive platelet count of greater than one million. She Primary and secondary patency rates are preunderwent chemotherapy for her myeloprolifera- sented in Tables I. I1 and 111. Table I reports all tive disease and maintained graft patency. grafts, Table I1 grafts to ankle level and Table I11
TABLE 11.-Primary
Intervals Primary patency 0-30 days 1-6 mos 6 mos-1 year Secondary patency 0-30 days
1-6 mos 6 mos-1 year 1-1 112 years
and secondary life-table patency data of 30 tibial in situ saphenous vein bypasses to ankle level and below
No. grafts at risk
No. grafts failing
Death
30 25 13
2 6 0
1 1 1
2 5 6
7.0 27.3 0
93.0 72.7 100
93.0 67.6 67.6
4.8 9.5 9.5
30 26 15 6
0 2 1 0
1 1
3 8 6 2
0 9.3 9.1 0
100 90.7 90.9 100
100 90.7 82.5 82.5
0 6.3 9.7 9.7
2
1
'Reason for withdrawal of grafts were duration and loss of follow-up
Interval failure rate Withdrawn* (Yo)
Interval patency rate
(W
Cumulative patency rate (%)
Standard patency rate (%)
208
ANNALSOF VASCULAR SURGERY
IN SITU BYPASS TO THE DORSALIS PEDIS
TABLE Ill.-Primary
Intervals Primary patency 0-30 days 1-6 rnos 6 rnos-1 year 1-1 VZ years 1l/L-2 years Secondary patency 0-30 days 1-6 rnos 6 rnos-1 year 1-1 1/2 years 1%2 years 2-2% years 2%3 years
and secondary life-table patency data of 68 above ankle level tibial in situ saphenous vein bypasses No. grafts at risk
No. grafts failina
Interval failure rate Withdrawn'
(%I
Interval patency rate 1%)
Cumulative patency rate (%I
Standard patency rate (%)
Death
68 59 32 17 10
6 12 1 2 0
2 5 2 0 1
1 10 12 5 1
9.0 23.3 4.0 13.8
91.o 76.7 96.0 86.2 100
91.o 69.8 67.0 57.8 57.8
3.5 6.0 6.4 8.2 8.2
68 63 43 25 16 11 5
2 4 2 1 0 0 0
2 5 2 0 1 1 0
1 11 14 8 4 5 5
3.0 7.3 5.7 4.8 0 0
97.0 92.7 94.3 95.2 100 100 100
97.0 89.9 84.8 80.8 80.8 80.8 80.8
2.1 3.9 5.1 6.3 6.3 6.3 6.3
0
0
'Reasons for withdrawal of grafts were duration and loss of follow-up
reports grafts above ankle level. Figure 1 compares the above ankle level grafts to the ankle level bypasses. Two of the grafts listed as failures in the 30 day to six month interval were patent grafts that
100%.
0 Craftsaboveankle 1 0
+ Grafts above ankle
20
were sacrificed during amputation. One of these amputations was due to more extensive foot necrosis than was appreciated at the time of bypass, and the other was caused by complications resulting from a candida wound infection as discussed previously. All successful grafts resulted in limb salvage, at least to the transmetatarsal level, except one patient who had a Symes amputation after her open transmetatarsal amputation failed.
W Craftstoankle 10 0 Grafts to ankle 20
DISCUSSION
90%-
80%-
70%-
60%-
50%-
1
I
I
lmo.
I
1
6mo.
llr.
I
I
I
1Hyr.
Ar.
2Hyr.
3yr.
Fig. 1. Comparison of patenc rates of grafts to ankle level and below and grafts above ankle level. (Numbers listed are patent grafts at end of interval.)
The patency data in this series demonstrates equivalent results when comparing grafts to the dorsalis pedis and tibial arteries at ankle level to grafts above the ankle. This supports an aggressive approach to limb salvage in the patient with either no distal vessels suitable for bypass on the preoperative arteriogram or vessels that are occluded to ankle level. It also suggests that bypass to the dorsalis pedis or tibial arteries at ankle level may be preferable to a compromised tibial outflow, such as an isolated segment tibial artery or a peroneal artery with poor collateralization to the foot, if adequate vein is available. The marked improvement in secondary over primary patency rates emphasizes the importance of careful and frequent follow-up to improve the durability of these operations. We agree with the conclusions of Bandyk and associates [4] that a protocol of graft surveillance, early revision of hemodynamically abnormal but patent grafts, and an aggressive approach to the recently thrombosed graft is rewarded by excellent secondary patency. Wound edge necrosis and infection complicated a substantial number (17%) of these operations and
VOLUME 3 NO 3 - 1989
IN SITU BYPASS TO THE DORSALIS PEDIS
occurred more often in patients with ankle level grafts. This may represent the poor healing potential of chronically ischemic skin in the diabetic patient and the dialysis patient (N = 13) who had a 38% incidence of this complication. Skin flaps made by placing the incision away from the vein may have contributed. In the latter part of this series, we have used preoperative duplex ultrasound vein mapping in an effort to reduce this problem. Although operative mortality was equal in the two groups of patients, later mortality was higher in the patients who required ankle level grafts. They presented with more advanced ischemia (tissue loss in 97%), and there was a higher incidence of diabetes mellitus and chronic renal failure requiring dialysis in this group. This may indicate that this group selects patients who are at a more advanced stage of complications from long-term diabetes mellitus. Buchbinder and colleagues [ 5 ] demonstrated excellent results with the in situ saphenous vein as a conduit for “femoral-foot bypass” defined as “to, or below, malleolar level.” In their series, the in situ vein performed better than reversed vein and polytetrafluoroethylene. They also noted a 10% lower limb salvage rate than patency rate. We had two patients in our series that had major amputations in limbs with patent grafts (considered failed grafts in the life tables), although they were not in the population with ankle level grafts. An additional patient who was diabetic and on dialysis had a graft to the dorsalis pedis artery that remained patent for six weeks. In that time period, she demonstrated no signs of healing her transmetatarsal amputation or her distal surgical wounds. Her graft occluded after the decision was made to amputate, but before it was done. This graft is listed as a failed graft in the 30 day to six month interval. All of the other
209
patients with ankle level grafts healed their ulcers or minor amputations. CONCLUSION In situ saphenous vein grafts to the dorsalis pedis and tibial arteries at the ankle demonstrate patency rates equal to grafts above the ankle level, and have provided limb salvage to a substantial number of patients with severe tibial occlusive disease. More of these patients with ankle level outflow have diabetes mellitus or chronic renal failure requiring dialysis. Wound healing is a greater problem in these patients. They also present more often with tissue loss and suffer a greater late mortality. Graft surveillance and early revision improves the durability of these tibial and ankle level grafts. The in situ saphenous vein is our conduit of choice for the difficult bypass situations to vessels at the ankle level or below. REFERENCES LEVINE AW, BANDYK DF, BONIER PH, TOWNE JB. Lessons learned in adopting the in situ saphenous vein bypass. J Vusc Surg 1985;2: 145-153. BANDYK DF, C A T 0 RF, TOWNE JB. A low flow velocity predicts failure of femoropopliteal and femorotibial bypass grafts. Surgery 1985;98:799-809. CUTLER SJ, EDERER F. Maximum utilization of the life table method in analyzing survival. J Chronic Dis 1958; 81699-7 I 2. BANDYK DF, KAEBNICK HW, STEWARTGW, TOWNE JB. Durability of the in situ saphenous vein arterial bypass: a comparison of primary and secondary patency. J Vusc Surg 1987 5 :256-268. BUCHBINDER D. PASCH AR, ROLLINS DL, DILLON BC, DOUGLAS DJ, SCHULER J J , FLANIGAN DP. Results of arterial reconstruction of the foot. Arch Surg 1986:121:673-677.