- Email: [email protected]
Factors determining the outcome of crural and pedal revascularisation for critical limb ischaemia
Eur J VascSurg 7, 82-86 (1993)
Factors Determining the Outcome of Crural and Pedal Revascularisation for Critical Limb Ischaemia Jan H. M. Tordoir, Jeroen P. L. van der Plas, Michael J. H. M. Jacobs and Peter J. E. H. M. Kitslaar Department of Surgery, Academic Hospital Maastricht, Maastricht, The Netherlands The influences of clinical factors, site of distal anastomosis, type of graft and angiographic run-off, on graft patency and limb salvage following 141 femorocrural and pedal bypasses in 121 patients were investigated retrospectively. The grafts consisted of 111 femorocrural and 30 pedal bypasses; 49% of the patients had diabetes mellitus. Venous grafts were implanted in 116 limbs, using either in situ vein (65), reversed vein (38) or composite vein (13) graft. Twenty-five prosthetic grafts (14 PTFE and 11 umbilical veins) were inserted. After 1, 2, 3 and 4 years of follow-up, the primary cumulative patency rates for all grafts were respectively 67, 61, 55 and 55%, and the secondary patency rates were 75, 70, 64 and 64%. The site of distal anastomosis had no influence on graft patency rate; neither was there any significant effect of clinical risk factors and run-off on graft patency. Prosthetic grafts showed significantly lower patencies compared to venous grafts and appeared to be the only independent prognostic risk factor for graft failure (multivariate analysis; p = 0.03). Overall limb salvage rate was 84% at 3 years. There were four amputations with patent grafts. The limb salvage rates for in situ vein, reversed/composite vein and prosthetic grafts were 89, 79 and 66% at 3 years, respectively. Various bypass grafts to the cruraI and pedal arteries are successful and durable. The use of prosthetic grafts results in significantly lower patency rates, but appears to be effectivefor limb salvage. Key Words: Femorocrural bypass; Limb ischaemia; Patency rates.
Introduction Femorocrural and pedal bypass operations are performed increasingly for the treatment of critical limb ischaemia. The results of these distal vascular reconstructions are usually acceptable with patency rates of 80 and 60% after 1 and 3 years of follow-up, respectively. 1,2 The success of vascular reconstruction to the crural vessels is, however, still limited by several factors. Patients with previously failed vascular procedures may lack autogenous vein, necessitating implantation of prosthetic grafts. Graft occlusion is usually the result of the thrombogenicity of the artificial material. Diabetic patients may suffer from distal atherosclerosis and extensive infection in the foot, which makes distal bypass grafting extremely difficult and which may result in amputation, despite a patent bypass. To determine the influences of clinical and local factors, we performed a retrospective study of 141 consecutive bypasses to the crural and pedal arteries,
performed for limb salvage over a 5-year period. Clinical risk factors, bypass type and material, site of distal anastomosis and the quality of crural and runoff vessels were analysed and the influences of these variables on graft patencies and limb salvage rates were established.
Patients and Methods
During a 5-year period from January 1986 to July 1991, 121 patients underwent 141 bypasses to the crural and pedal arteries, because of critical limb ischaemia. Single grafts were performed in 101 patients; 10 patients had two bypasses in the same leg and another 10 patients had bypass surgery for ischaemia of both legs. The indication for reconstruction was rest pain in 47% and non-healing foot lesions in 53%. The patient group consisted of 70 men and 51 women with a mean age of 72 years (range 3599). Previous vascular operations had been performed in 42 patients. Clinical risk factors analysed included the presence of hypertension, cardiac disPlease address all correspondence to: Jan H. M. Tordoir, Department of Surgery, AcademicHospital, Postbox5800, 6202AZ Maas- ease, cerebral ischaemia, diabetes mellitus and smoking (Table 1). tricht, The Netherlands.
0950-821X/93/010082+05 $08.00/0© 1993Grune & StrattonLtd.
Factors Determining the Outcome of Femorocrural Bypasses
Table 1. Risk factors in patients (n = 121) with femorodistal bypasses
n
%
Cardiac disease
61
Cerebral disease
21
50 17
Diabetes mellitus
56
46
Hypertension
37
31
Smoking
61
50
Table 2. Proximal and distal anastomosis in 141 femorodistal bypasses
n
%
Proximal Femoral artery
127
9o
Popliteal artery
14
10
Distal Anterior tibial artery
40
28
Posterior tibial artery
42
30
Peroneal artery
29
21
Pedal/plantar artery
30
21
Saphenous vein grafts were used in 116 cases and 25 prosthetic (14 PTFE and 11 umbilical vein) grafts were implanted either because of absence of the saphenous vein or because of the presence of an unsuitable or small calibre vein. External ring-supported polytetrafluoroethylene (PTFE; Gore-tex), 6 mm in diameter and h u m a n umbilical vein (HUV; Biograft; Meadox inc.) comprised the prosthetic grafts. Reversed saphenous veins were used in 38 cases, in situ vein grafts in 65 cases and composite venous grafts in 13. Proximal and distal anastomotic sites are listed in Table 2. Thirty distal anastomoses were performed to the dorsal pedal or plantar arteries. Additional arteriovenous fistulas at the distal anastomosis to increase graft flow were created in 10 PTFE prostheses. A deep anatomic route for graft insertion was performed in 45 cases; 96 grafts were implanted in a subcutaneous tunnel. Intraoperative continuous wave (CW) Doppler measurements and completion angiography were performed after each reconstruction. The status of the distal run-off was assessed in each patient by means of the pre- and intraoperative angiograms on the basis of the number and patency of the crural vessels and the presence or absence of the pedal arch, as proposed by the Ad Hoc Corn-
83
mittee on reporting standards for lower extremity ischaemia. 3 Run-off was judged to be good when three crural vessels were patent and to be poor when three crural vessels were absent or partially occluded. Also, a run-off score of the bypass was calculated as good (open outflow vessels distal to the anastomosis) or poor (totally occluded outflow vessels). Haemodynamic improvement after femorodistal bypass operation was measured non-invasively by comparing pre- and postoperative ankle and toe blood pressures and ankle-brachial pressure indices. Postoperative measurements were made 7-10 days after operation. Clinical and Doppler examinations were performed at 3 monthly intervals to assess graft patency. Follow-up ranged from 3-62 months (mean 18 months). Statistical analysis of the different variables was performed with the non-parametric Mann-Whitney test. Stepwise multiple regression analysis was used to determine independent risk factors. Differences in cumulative patency rates among groups were assessed by comparison of life tables with the LeeDesu test.
Results
The incidence of postoperative complications and early graft failure is shown in Table 3. There were 41 graft occlusions; 14 grafts were patent after revision, 27 grafts failed resulting in 19 amputations. Six patients died of cardiac or cerebrovascular disease within I month of operation. Table 3. Early (<30 days) postoperative complications and graft failure
n
%
24
17
Cardiac failure
9
6
Graft occlusion
41
29
6
5
Wound complications
Death
Late graft failure occurred in 12 grafts; three grafts were successfully revised because of a significant stenosis (>70% diameter reduction). Primary and secondary cumulative patencies (calculated by the life table method) for all grafts were 67 and 75%, respectively, after 1 year, and 55 and 64%, respectively, after 3 years of follow-up (Fig. 1). Secondary patency rates for the different graft types are displayed in Fig. 2. At 1 year, secondary patency was Eur J VascSurgVol 7, January1993
84
J.H.M.
Tordoir
tOOIT II ] .....
I00
6
- .....
86
L. . . . . . . . . .
18
I ,
I.............................
.........
I
15 4O
2O
0
12
24 Follow-up (months)
36
48
Fig. 1. Primary ( - - - ) a n d secondary (--) cumulative patency rates for all grafts. I00
~
8O
i
14
~, 60
12
t . . . . . . . . . .
8
.....
,
8
; i. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
~
4o
6
I
~.)
2O i
0
12
i
24 Follow-up (months)
.i._
36
48
et al.
0.03). The cumulative secondary patency rates of pedal bypasses were similar to the secondary patency rates of bypasses to the crural vessels at 1, 2 and 3 years of follow-up (Fig. 3). Clinical variables, including sex, age, smoking habits, diabetes mellitus, cardiac disease and ischaemic stadium did not significantly affect graft patency. Neither was the angiographically determined outflow and run-off score of importance for graft failure. Bypasses with poor run-off vessels performed as well as bypasses with good run-off. Also, the site of the distal anastomosis had no influence on primary and secondary patency rates. Multivariate analysis showed that the use of prosthetic graft material for bypass surgery was the only single independent variable correlating with graft failure (R2 value 0.3669; p = 0.025). Six amputations were needed after a mean follow-up period of 7 months because of graft failure (2), extensive foot infection with a patent graft (2) and infection in the prosthetic graft making explantation necessary (2). Overall, a total of 25 major amputations were needed in 141 limbs. Diabetic patients had a significant higher incidence of amputations compared to non-diabetic patients (17/68 versus 8/73; p = 0.05). Overall limb salvage rates after crural and pedal revascularisation were 87 and 82%, respectively, after 3 years of follow-up (Fig. 4). No statistically signifi-
Fig. 2. Secondary patency rates of in situ vein (--), reversed vein ( - - ) and prosthetic ( . . . . ) grafts. Lee-Desu p = 0.03.
IO0
121
I00 ~
8O
g 6o
~
80 o~
lu6 117
76 78
12
561 . . . . . . . . . 50
_30 i ! ..............................
19
60 l
~54 7[ I L. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 1 ......
17
J
~
2O
40
m 20
0
0
12
24 Follow-up (months)
:36
48
Fig. 3. Secondary patency rates of crural (--) versus pedal (- -) bypasses. Lee-Desu p = 0.91.
80% for in situ grafts, 78% for reversed/composite vein grafts and 57% for prosthetic grafts. At 3 years, these rates were 80, 60 and 42%, respectively. The differences in 3 years patencies between the vein grafts (in situ plus reversed/composite vein grafts) and prosthetic grafts were statistically significant (p = Eur J Vasc Surg Vol 7, January 1993
12
24 Follow-up (months)
36
48
Fig. 4. Limb salvage (--) and survival ( - - - ) rates for femorocrural a n d pedal bypass operations.
cant differences in limb salvage rates were found between in situ vein, reversed/composite vein and prosthetic grafts (Fig. 5). Grafts with the distal anastomosis to the peroneal artery had a higher amputation rate (p = 0.05). There were a total of 36 late deaths after a follow-up of 4 years, resulting in a survival rate of 68% (Fig. 4). In Table 4, the results of the non-invasive
Factors Determining the Outcome of Femorocrural Bypasses
85
follow-up, 66% of the grafts in diabetic patients were patent versus 62% in non-diabetics. However, ......... _30 ........ '_L .......... 8O diabetic patients had a significantly higher ampu'25 2 41 2e 17 i0 tation rate of 25%, compared to 11% in non-diabetic 2 18 II 9 4 o 60 patients. Occlusion of foot arteries (one patient), extensive infection (two patients) and prosthetic graft o 40 infection with the need for graft explanation despite an open bypass (two patients), contributed to ampud tation in diabetics. 20 Poor run-off vessels and occlusion of the pedal arch have been reported as a cause of graft failure. In 0 12 24 36 48 our patient group, we could not find a significant Follow-up (months) correlation between the status of the distal outflow Fig. 5. Limb salvage rates after in situ vein (--), reversed vein vessels and good or poor run-off on graft patency and (---) and prosthetic graft (.... ) bypass operations. Lee-Desup = limb salvage. Collateral circulation might have 0.06. provided adequate distal run-off and low resistance to the bypass but we did not perform outflow resistTable 4. Pre- and postoperative ankle-brachial indices, ankle and toe pressures ance measurements, as others have done. 5 The site of distal anastomosis proved to have no Preoperative Postoperative influence on graft patency. Bypasses to the peroneal Mean ankle-brachialindex 0.41 0.74 artery appeared to have a higher amputation rate, however, using multivariate analysis this observation Mean ankle pressure (mmHg) 60 108 was proved not to be of statistical significance. RelaMean toe pressure (mmHg) 15 70 tively more prosthetic grafts were anastomosed to the peroneal artery (31%) than to tibial arteries (18%) and this may be the reason for the higher amputation measurements are outlined. The mean ankle-brachial rate. The patency rates of bypasses to crural vessels indices and toe pressures rose from respectively 0.41 were lower compared with those to pedal arteries and 15 m m H g preoperatively to respectively 0.74 and (respectively, 64 and 69% after 3 years of follow-up). 70 m m H g postoperatively (patients with incompres- However, in the femorocrural bypass group, 24 prossible ankle arteries excluded; n = 24). No statistical thetic grafts were inserted (22%), whereas the in situ significant differences in preoperative ankle-brachial vein technique was used almost exclusively in the indices and toe pressures between patients with and femoropedal bypass group; only one prosthesis (3%) without failed grafts or limb salvage and limb loss being implanted in this latter group (p = 0.001). were found. Excluding the prosthetic bypass grafts, similar patency rates of 71 and 69%, respectively, were calculated. The superiority of the autogenous saphenous Discussion vein in the reversed or in situ position as the conduit of choice for infrapopliteal vascular reconstruction is Distal bypasses to crural and pedal arteries can now widely acknowledged. Nevertheless, there is an be performed successfully for the treatment of increasing number of patients requiring distal revaspatients with critical limb ischaemia. However, there cularisation without a satisfactory saphenous vein or remain certain patients in whom these distal bypass an already consumed saphenous vein due to preoperations fail, usually resulting in limb amputation. vious vascular reconstruction. In these patients, Diabetic patients may show distal occlusive disease, upper extremity and lesser saphenous veins can be calcified arteries and extended tissue necrosis, which used with acceptable results. 6"7 makes distal anastomosis and salvage of the foot exThe use of prosthetic material in the infrapoplitremely difficult.4 Diabetes mellitus occurred in 49% teal position has been debated in the literature. Howof our patient population. The results showed that ever, reasonable results using PTFE prostheses to the the presence of diabetes had no effect on the develop- crural level have been published recently. 8"9 Adment of postoperative complications nor on mor- ditional venous cuffs and distal arteriovenous (AV) tality. Cumulative graft patencies were similar in fistulas are reported to improve the patency of PTFE diabetic and non-diabetic patients. After 3 years of bypass grafts. 1° During the last 2 years of this study I O0
i! 5
_
I
L
I
Eur J VascSurg Vol 7, January1993
86
J . H . M . Tordoir et al.
we routinely created AV fistulas at the distal anastomosis of prosthetic grafts to increase graft flow. In these patients we have found a significant improvement in microcirculatory parameters after vascular reconstruction. 11 Secondary graft patencies at 3 years follow-up of the in situ vein, reversed vein and prosthetic graft were significantly different with lower patencies for the PTFE and umbilical vein grafts. Multivariate statistical analysis showed that the use of prosthetic material for femorodistal bypass operations was the only independent risk factor for graft failure. Surprisingly, this outcome did not correlate with a poor limb salvage. Limb salvage rates for prosthetic grafts were lower than those for vein grafts, but these differences were not statistically significant and 66% of limbs could be saved using prosthetic grafts. Our results indicate that femorocrural or pedal bypass operations for critical limb ischaemia can be performed with fair to good results. Clinical factors like diabetes mellitus, poor distal run-off and site of distal anastomosis had no adverse effect on the functioning and patency of the graft. In diabetics, significantly more amputations were required because of persistent foot infection or ischaemia. Implantation of prosthetic grafts resulted in lower graft patency and limb salvage rates, but in the presence of an unsuitable or absent vein the use of a prosthesis is preferable to primary amputaion.
References 1 ANDROSG, HARRISRW, SALLES-CUNHASX, DULAWALB, OBLATH RW, APYAN RL. Bypass grafts to the ankle and foot. J Vasc Surg 1988; 7: 785-794. 2 VEITH FJ, GUPTA SK, WENGERTER KR, et al. Changing arteriosclerotic disease patterns and management strategies in lower limb threatening ischaemia. Ann Surg 1990; 212: 402-414. 3 RUTHERFORDRB, FLANIGANP, GUPTA SK, JOHNSTON KW, KARMODY A, et al. Suggested standards for reports dealing with lower extremity ischaemia. J Vasc Surg 1986; 4: 80-94. 4 DIETZEK AM, GUPTA SK, KRAM HB, WENGERTER KR, VE~THFJ. Limb loss with patent infra-inguinal bypasses. Eur J Vasc Surg 1990; 4: 413-417. 5 ASCERE, VEITHFJ, WHITE-FLORESSA, MORIN L, GUPTASK. lntraoperative outflow resistance as a predictor of late patency of femoropopltteal and infrapopliteal arterial bypasses. J Vasc Surg 1987; 5: 820-827. 6 BALSHIJD, CANTELMONL, MENZOIANJO, LoGERFO FW. The use of arm veins for infrainguinal bypass in end-stage peripheral vascular disease. Arch Surg 1989; 124: 1078-1081. 7 FEINBERGRL, WINTERRP, WHEELERJR, GREGORYRT, SNYDERSO, GAYLE RG, PARENT FN, ADCOCK GD. The use of composite grafts in femorocrural bypasses performed for limb salvage: A review of 108 consecutive cases and comparison with 57 in situ saphenous vein bypasses. J Vasc Surg 1990; 12: 257-263. 8 WOLFE JHN, TYRRELL MR. Justifying arterial reconstruction to crural vessels--even with a prosthetic graft. Br J Surg 1991; 78: 897-899. 9 TYRRELLMR, WOLFEJHN. New prosthetic venous collar anastomotic technique--combining the best of other procedures. Br J Surg 1991; 78; 1016-1017. 10 JACOBS MJHM, GREGORIC ID, REUL GJ. Prosthetic graft placement and creation of a distal arteriovenous fistula for secondary vascular reconstruction in patients with severe limb ischaemia. J Vasc Surg 1992; 15: 612-618. 11 JACOBS MJHM, REUL GJ, UBBINK DT, TORDOIR JHM, KITSLAAR PJEHM. Creation of a distal arteriovenous fistula improves microcirculatory haemodynamics of prosthetic graft bypass in secondary limb salvage procedures. J Vasc Surg, in press.
Accepted 1 September 1992
Eur J Vasc Surg Vol 7, January 1993
Factors Determining the Outcome of Crural and Pedal Revascularisation for Critical Limb Ischaemia Jan H. M. Tordoir, Jeroen P. L. van der Plas, Michael J. H. M. Jacobs and Peter J. E. H. M. Kitslaar Department of Surgery, Academic Hospital Maastricht, Maastricht, The Netherlands The influences of clinical factors, site of distal anastomosis, type of graft and angiographic run-off, on graft patency and limb salvage following 141 femorocrural and pedal bypasses in 121 patients were investigated retrospectively. The grafts consisted of 111 femorocrural and 30 pedal bypasses; 49% of the patients had diabetes mellitus. Venous grafts were implanted in 116 limbs, using either in situ vein (65), reversed vein (38) or composite vein (13) graft. Twenty-five prosthetic grafts (14 PTFE and 11 umbilical veins) were inserted. After 1, 2, 3 and 4 years of follow-up, the primary cumulative patency rates for all grafts were respectively 67, 61, 55 and 55%, and the secondary patency rates were 75, 70, 64 and 64%. The site of distal anastomosis had no influence on graft patency rate; neither was there any significant effect of clinical risk factors and run-off on graft patency. Prosthetic grafts showed significantly lower patencies compared to venous grafts and appeared to be the only independent prognostic risk factor for graft failure (multivariate analysis; p = 0.03). Overall limb salvage rate was 84% at 3 years. There were four amputations with patent grafts. The limb salvage rates for in situ vein, reversed/composite vein and prosthetic grafts were 89, 79 and 66% at 3 years, respectively. Various bypass grafts to the cruraI and pedal arteries are successful and durable. The use of prosthetic grafts results in significantly lower patency rates, but appears to be effectivefor limb salvage. Key Words: Femorocrural bypass; Limb ischaemia; Patency rates.
Introduction Femorocrural and pedal bypass operations are performed increasingly for the treatment of critical limb ischaemia. The results of these distal vascular reconstructions are usually acceptable with patency rates of 80 and 60% after 1 and 3 years of follow-up, respectively. 1,2 The success of vascular reconstruction to the crural vessels is, however, still limited by several factors. Patients with previously failed vascular procedures may lack autogenous vein, necessitating implantation of prosthetic grafts. Graft occlusion is usually the result of the thrombogenicity of the artificial material. Diabetic patients may suffer from distal atherosclerosis and extensive infection in the foot, which makes distal bypass grafting extremely difficult and which may result in amputation, despite a patent bypass. To determine the influences of clinical and local factors, we performed a retrospective study of 141 consecutive bypasses to the crural and pedal arteries,
performed for limb salvage over a 5-year period. Clinical risk factors, bypass type and material, site of distal anastomosis and the quality of crural and runoff vessels were analysed and the influences of these variables on graft patencies and limb salvage rates were established.
Patients and Methods
During a 5-year period from January 1986 to July 1991, 121 patients underwent 141 bypasses to the crural and pedal arteries, because of critical limb ischaemia. Single grafts were performed in 101 patients; 10 patients had two bypasses in the same leg and another 10 patients had bypass surgery for ischaemia of both legs. The indication for reconstruction was rest pain in 47% and non-healing foot lesions in 53%. The patient group consisted of 70 men and 51 women with a mean age of 72 years (range 3599). Previous vascular operations had been performed in 42 patients. Clinical risk factors analysed included the presence of hypertension, cardiac disPlease address all correspondence to: Jan H. M. Tordoir, Department of Surgery, AcademicHospital, Postbox5800, 6202AZ Maas- ease, cerebral ischaemia, diabetes mellitus and smoking (Table 1). tricht, The Netherlands.
0950-821X/93/010082+05 $08.00/0© 1993Grune & StrattonLtd.
Factors Determining the Outcome of Femorocrural Bypasses
Table 1. Risk factors in patients (n = 121) with femorodistal bypasses
n
%
Cardiac disease
61
Cerebral disease
21
50 17
Diabetes mellitus
56
46
Hypertension
37
31
Smoking
61
50
Table 2. Proximal and distal anastomosis in 141 femorodistal bypasses
n
%
Proximal Femoral artery
127
9o
Popliteal artery
14
10
Distal Anterior tibial artery
40
28
Posterior tibial artery
42
30
Peroneal artery
29
21
Pedal/plantar artery
30
21
Saphenous vein grafts were used in 116 cases and 25 prosthetic (14 PTFE and 11 umbilical vein) grafts were implanted either because of absence of the saphenous vein or because of the presence of an unsuitable or small calibre vein. External ring-supported polytetrafluoroethylene (PTFE; Gore-tex), 6 mm in diameter and h u m a n umbilical vein (HUV; Biograft; Meadox inc.) comprised the prosthetic grafts. Reversed saphenous veins were used in 38 cases, in situ vein grafts in 65 cases and composite venous grafts in 13. Proximal and distal anastomotic sites are listed in Table 2. Thirty distal anastomoses were performed to the dorsal pedal or plantar arteries. Additional arteriovenous fistulas at the distal anastomosis to increase graft flow were created in 10 PTFE prostheses. A deep anatomic route for graft insertion was performed in 45 cases; 96 grafts were implanted in a subcutaneous tunnel. Intraoperative continuous wave (CW) Doppler measurements and completion angiography were performed after each reconstruction. The status of the distal run-off was assessed in each patient by means of the pre- and intraoperative angiograms on the basis of the number and patency of the crural vessels and the presence or absence of the pedal arch, as proposed by the Ad Hoc Corn-
83
mittee on reporting standards for lower extremity ischaemia. 3 Run-off was judged to be good when three crural vessels were patent and to be poor when three crural vessels were absent or partially occluded. Also, a run-off score of the bypass was calculated as good (open outflow vessels distal to the anastomosis) or poor (totally occluded outflow vessels). Haemodynamic improvement after femorodistal bypass operation was measured non-invasively by comparing pre- and postoperative ankle and toe blood pressures and ankle-brachial pressure indices. Postoperative measurements were made 7-10 days after operation. Clinical and Doppler examinations were performed at 3 monthly intervals to assess graft patency. Follow-up ranged from 3-62 months (mean 18 months). Statistical analysis of the different variables was performed with the non-parametric Mann-Whitney test. Stepwise multiple regression analysis was used to determine independent risk factors. Differences in cumulative patency rates among groups were assessed by comparison of life tables with the LeeDesu test.
Results
The incidence of postoperative complications and early graft failure is shown in Table 3. There were 41 graft occlusions; 14 grafts were patent after revision, 27 grafts failed resulting in 19 amputations. Six patients died of cardiac or cerebrovascular disease within I month of operation. Table 3. Early (<30 days) postoperative complications and graft failure
n
%
24
17
Cardiac failure
9
6
Graft occlusion
41
29
6
5
Wound complications
Death
Late graft failure occurred in 12 grafts; three grafts were successfully revised because of a significant stenosis (>70% diameter reduction). Primary and secondary cumulative patencies (calculated by the life table method) for all grafts were 67 and 75%, respectively, after 1 year, and 55 and 64%, respectively, after 3 years of follow-up (Fig. 1). Secondary patency rates for the different graft types are displayed in Fig. 2. At 1 year, secondary patency was Eur J VascSurgVol 7, January1993
84
J.H.M.
Tordoir
tOOIT II ] .....
I00
6
- .....
86
L. . . . . . . . . .
18
I ,
I.............................
.........
I
15 4O
2O
0
12
24 Follow-up (months)
36
48
Fig. 1. Primary ( - - - ) a n d secondary (--) cumulative patency rates for all grafts. I00
~
8O
i
14
~, 60
12
t . . . . . . . . . .
8
.....
,
8
; i. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
~
4o
6
I
~.)
2O i
0
12
i
24 Follow-up (months)
.i._
36
48
et al.
0.03). The cumulative secondary patency rates of pedal bypasses were similar to the secondary patency rates of bypasses to the crural vessels at 1, 2 and 3 years of follow-up (Fig. 3). Clinical variables, including sex, age, smoking habits, diabetes mellitus, cardiac disease and ischaemic stadium did not significantly affect graft patency. Neither was the angiographically determined outflow and run-off score of importance for graft failure. Bypasses with poor run-off vessels performed as well as bypasses with good run-off. Also, the site of the distal anastomosis had no influence on primary and secondary patency rates. Multivariate analysis showed that the use of prosthetic graft material for bypass surgery was the only single independent variable correlating with graft failure (R2 value 0.3669; p = 0.025). Six amputations were needed after a mean follow-up period of 7 months because of graft failure (2), extensive foot infection with a patent graft (2) and infection in the prosthetic graft making explantation necessary (2). Overall, a total of 25 major amputations were needed in 141 limbs. Diabetic patients had a significant higher incidence of amputations compared to non-diabetic patients (17/68 versus 8/73; p = 0.05). Overall limb salvage rates after crural and pedal revascularisation were 87 and 82%, respectively, after 3 years of follow-up (Fig. 4). No statistically signifi-
Fig. 2. Secondary patency rates of in situ vein (--), reversed vein ( - - ) and prosthetic ( . . . . ) grafts. Lee-Desu p = 0.03.
IO0
121
I00 ~
8O
g 6o
~
80 o~
lu6 117
76 78
12
561 . . . . . . . . . 50
_30 i ! ..............................
19
60 l
~54 7[ I L. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 1 ......
17
J
~
2O
40
m 20
0
0
12
24 Follow-up (months)
:36
48
Fig. 3. Secondary patency rates of crural (--) versus pedal (- -) bypasses. Lee-Desu p = 0.91.
80% for in situ grafts, 78% for reversed/composite vein grafts and 57% for prosthetic grafts. At 3 years, these rates were 80, 60 and 42%, respectively. The differences in 3 years patencies between the vein grafts (in situ plus reversed/composite vein grafts) and prosthetic grafts were statistically significant (p = Eur J Vasc Surg Vol 7, January 1993
12
24 Follow-up (months)
36
48
Fig. 4. Limb salvage (--) and survival ( - - - ) rates for femorocrural a n d pedal bypass operations.
cant differences in limb salvage rates were found between in situ vein, reversed/composite vein and prosthetic grafts (Fig. 5). Grafts with the distal anastomosis to the peroneal artery had a higher amputation rate (p = 0.05). There were a total of 36 late deaths after a follow-up of 4 years, resulting in a survival rate of 68% (Fig. 4). In Table 4, the results of the non-invasive
Factors Determining the Outcome of Femorocrural Bypasses
85
follow-up, 66% of the grafts in diabetic patients were patent versus 62% in non-diabetics. However, ......... _30 ........ '_L .......... 8O diabetic patients had a significantly higher ampu'25 2 41 2e 17 i0 tation rate of 25%, compared to 11% in non-diabetic 2 18 II 9 4 o 60 patients. Occlusion of foot arteries (one patient), extensive infection (two patients) and prosthetic graft o 40 infection with the need for graft explanation despite an open bypass (two patients), contributed to ampud tation in diabetics. 20 Poor run-off vessels and occlusion of the pedal arch have been reported as a cause of graft failure. In 0 12 24 36 48 our patient group, we could not find a significant Follow-up (months) correlation between the status of the distal outflow Fig. 5. Limb salvage rates after in situ vein (--), reversed vein vessels and good or poor run-off on graft patency and (---) and prosthetic graft (.... ) bypass operations. Lee-Desup = limb salvage. Collateral circulation might have 0.06. provided adequate distal run-off and low resistance to the bypass but we did not perform outflow resistTable 4. Pre- and postoperative ankle-brachial indices, ankle and toe pressures ance measurements, as others have done. 5 The site of distal anastomosis proved to have no Preoperative Postoperative influence on graft patency. Bypasses to the peroneal Mean ankle-brachialindex 0.41 0.74 artery appeared to have a higher amputation rate, however, using multivariate analysis this observation Mean ankle pressure (mmHg) 60 108 was proved not to be of statistical significance. RelaMean toe pressure (mmHg) 15 70 tively more prosthetic grafts were anastomosed to the peroneal artery (31%) than to tibial arteries (18%) and this may be the reason for the higher amputation measurements are outlined. The mean ankle-brachial rate. The patency rates of bypasses to crural vessels indices and toe pressures rose from respectively 0.41 were lower compared with those to pedal arteries and 15 m m H g preoperatively to respectively 0.74 and (respectively, 64 and 69% after 3 years of follow-up). 70 m m H g postoperatively (patients with incompres- However, in the femorocrural bypass group, 24 prossible ankle arteries excluded; n = 24). No statistical thetic grafts were inserted (22%), whereas the in situ significant differences in preoperative ankle-brachial vein technique was used almost exclusively in the indices and toe pressures between patients with and femoropedal bypass group; only one prosthesis (3%) without failed grafts or limb salvage and limb loss being implanted in this latter group (p = 0.001). were found. Excluding the prosthetic bypass grafts, similar patency rates of 71 and 69%, respectively, were calculated. The superiority of the autogenous saphenous Discussion vein in the reversed or in situ position as the conduit of choice for infrapopliteal vascular reconstruction is Distal bypasses to crural and pedal arteries can now widely acknowledged. Nevertheless, there is an be performed successfully for the treatment of increasing number of patients requiring distal revaspatients with critical limb ischaemia. However, there cularisation without a satisfactory saphenous vein or remain certain patients in whom these distal bypass an already consumed saphenous vein due to preoperations fail, usually resulting in limb amputation. vious vascular reconstruction. In these patients, Diabetic patients may show distal occlusive disease, upper extremity and lesser saphenous veins can be calcified arteries and extended tissue necrosis, which used with acceptable results. 6"7 makes distal anastomosis and salvage of the foot exThe use of prosthetic material in the infrapoplitremely difficult.4 Diabetes mellitus occurred in 49% teal position has been debated in the literature. Howof our patient population. The results showed that ever, reasonable results using PTFE prostheses to the the presence of diabetes had no effect on the develop- crural level have been published recently. 8"9 Adment of postoperative complications nor on mor- ditional venous cuffs and distal arteriovenous (AV) tality. Cumulative graft patencies were similar in fistulas are reported to improve the patency of PTFE diabetic and non-diabetic patients. After 3 years of bypass grafts. 1° During the last 2 years of this study I O0
i! 5
_
I
L
I
Eur J VascSurg Vol 7, January1993
86
J . H . M . Tordoir et al.
we routinely created AV fistulas at the distal anastomosis of prosthetic grafts to increase graft flow. In these patients we have found a significant improvement in microcirculatory parameters after vascular reconstruction. 11 Secondary graft patencies at 3 years follow-up of the in situ vein, reversed vein and prosthetic graft were significantly different with lower patencies for the PTFE and umbilical vein grafts. Multivariate statistical analysis showed that the use of prosthetic material for femorodistal bypass operations was the only independent risk factor for graft failure. Surprisingly, this outcome did not correlate with a poor limb salvage. Limb salvage rates for prosthetic grafts were lower than those for vein grafts, but these differences were not statistically significant and 66% of limbs could be saved using prosthetic grafts. Our results indicate that femorocrural or pedal bypass operations for critical limb ischaemia can be performed with fair to good results. Clinical factors like diabetes mellitus, poor distal run-off and site of distal anastomosis had no adverse effect on the functioning and patency of the graft. In diabetics, significantly more amputations were required because of persistent foot infection or ischaemia. Implantation of prosthetic grafts resulted in lower graft patency and limb salvage rates, but in the presence of an unsuitable or absent vein the use of a prosthesis is preferable to primary amputaion.
References 1 ANDROSG, HARRISRW, SALLES-CUNHASX, DULAWALB, OBLATH RW, APYAN RL. Bypass grafts to the ankle and foot. J Vasc Surg 1988; 7: 785-794. 2 VEITH FJ, GUPTA SK, WENGERTER KR, et al. Changing arteriosclerotic disease patterns and management strategies in lower limb threatening ischaemia. Ann Surg 1990; 212: 402-414. 3 RUTHERFORDRB, FLANIGANP, GUPTA SK, JOHNSTON KW, KARMODY A, et al. Suggested standards for reports dealing with lower extremity ischaemia. J Vasc Surg 1986; 4: 80-94. 4 DIETZEK AM, GUPTA SK, KRAM HB, WENGERTER KR, VE~THFJ. Limb loss with patent infra-inguinal bypasses. Eur J Vasc Surg 1990; 4: 413-417. 5 ASCERE, VEITHFJ, WHITE-FLORESSA, MORIN L, GUPTASK. lntraoperative outflow resistance as a predictor of late patency of femoropopltteal and infrapopliteal arterial bypasses. J Vasc Surg 1987; 5: 820-827. 6 BALSHIJD, CANTELMONL, MENZOIANJO, LoGERFO FW. The use of arm veins for infrainguinal bypass in end-stage peripheral vascular disease. Arch Surg 1989; 124: 1078-1081. 7 FEINBERGRL, WINTERRP, WHEELERJR, GREGORYRT, SNYDERSO, GAYLE RG, PARENT FN, ADCOCK GD. The use of composite grafts in femorocrural bypasses performed for limb salvage: A review of 108 consecutive cases and comparison with 57 in situ saphenous vein bypasses. J Vasc Surg 1990; 12: 257-263. 8 WOLFE JHN, TYRRELL MR. Justifying arterial reconstruction to crural vessels--even with a prosthetic graft. Br J Surg 1991; 78: 897-899. 9 TYRRELLMR, WOLFEJHN. New prosthetic venous collar anastomotic technique--combining the best of other procedures. Br J Surg 1991; 78; 1016-1017. 10 JACOBS MJHM, GREGORIC ID, REUL GJ. Prosthetic graft placement and creation of a distal arteriovenous fistula for secondary vascular reconstruction in patients with severe limb ischaemia. J Vasc Surg 1992; 15: 612-618. 11 JACOBS MJHM, REUL GJ, UBBINK DT, TORDOIR JHM, KITSLAAR PJEHM. Creation of a distal arteriovenous fistula improves microcirculatory haemodynamics of prosthetic graft bypass in secondary limb salvage procedures. J Vasc Surg, in press.
Accepted 1 September 1992
Eur J Vasc Surg Vol 7, January 1993