The profunda femoris: A durable outflow vessel in aortofemoral surgery E d m o n d J. Prendiville, FRCSI, Paul E. Burke, MD, FRCSI, Mary Paula Colgan, MD, Bee L. Wee, MB, Dermot J. Moore, MD, FRCSI, and D. Gregor Shanik, MD, FRCSI, FACS, Dublin, Ireland Aorta-common femoral artery bypass is the standard operation for relief of aortoiliac occlusive disease. When extensive superficial femoral artery disease coexists, the profunda femoris, even in its distal portion, may be used as the outflow vessel. To test this assumption we compared cumulative patency, limb salvage, and the need for distal bypass of 134 aorta-profunda femoris and 151 aorta-common femoral artery bypasses performed consecutively for aortoiliac occlusive disease over a 12-year period. We also analyzed results of proximal (n = I03) and distal (n = 31) aortoprofunda bypasses. Angiographic and noninvasive studies showed greater disease in limbs undergoing aorta-profunda femoris bypass. However, no difference was observed in cumulative patency (91% ± 6% vs 96% -+ 3%) or limb salvage (90% ± 6% vs 94% ± 3%) at 5 years. Seventeen distal bypasses in the group undergoing profimda femoris bypass and 20 distal bypasses in the group undergoing common femoral artery bypass were required to maintain limb salvage. Proximal and distal aorta-profunda femoris bypasses showed no difference in cumulative patency (91% -+ 9% vs 95% -+ 6%) or limb salvage (94% in each group) at 3 years. Standard aorta-common femoral artery and aorta-profunda femoris bypass provide cumulative patency and limb salvage exceeding 90% at 5 years; concomitant or subsequent distal bypass was required in 12% or limbs undergoing aorta-profunda femoris bypasses. Both proximal and distal profunda femoris arteries provide a durable outflow tract when aortoiliac and femoropopliteal occlusive disease are combined. (J VAsc SURG 1992;16:23-9.)
Aorta-common femoral artery (A-CFA) bypass is the standard operation for relief of aortoiliac occlusive disease and has excellent results, with patency rates greater than 85% at 5 years) However, significant CFA outflow obstruction frequently coexists with aortoiliac disease, and may necessitate distal bypass surgery to maintain graft patency and limb salvage. The profunda femoris (PFA), through its many branches, forms extensive communication with the popliteal artery at the knee and the tibial vessels in the calf. In the presence of superficial femoral artery (SFA) occlusion, the PFA provides the major collateral circulation through which blood reaches the distal limb. Although atherosclerosis does not affect the PFA as severely as the SFA, Beales et al.2 reviewing angiograms from ischemic limbs, observed atheroma in the ostium of the PFA in 39% of cases. Martin et al. s noted that the atheroma affecting the From the Department of Vascular Surgery, St, James'sHospital, Dublin, Ireland. Reprint requests: ProfessorD.G. Shanik, Departmentof Vascular Surgery, St. James's Hospital, Dublin 8. Ireland. 24/1/34a45
PFA was confined to the ostium or proximal segment in 74% of cases but affected the mid and distal segments in 26%. In the presence of coexisting aortoiliac occlusive disease and CFA outflow obstruction, use of the PFA as the outflow vessel in aortofemoral bypass will optimize the collateral circulation, and may lessen the need for distal bypass surgery. To test this hypothesis we have reviewed the results of all aortofemoral bypass procedures performed for aortoiliac occlusive disease in our unit and compared cumulative patency, limb salvage, and the need for distal bypass surgery in A-CFA and A-PFA procedures. When atheroma cxists beyond the ostium of the PFA it may be necessary to perform a more distal PFA bypass. We also analyzed the results ofaortofemoral bypass to the proximal, mid, and distal segments of the PFA. PATIENTS A N D M E T H O D S A review of all aortofemoral bypass procedures performed in St. James's Hospital over a 12-year period, between January 1978 and December 1989, identified a total of 285 grafts. O f these, 145 grafts 23
24 Prendiville et al.
were inserted for aortoiliac occlusive disease. Bypass procedures for aneurysmal disease were excluded from this study. Two hundred eighty-five limbs were revascularized. In 151 limbs the CFA was used as the outflow vessel, and in 134 limbs the PFA was used. O f A-PFA bypasses 103 were to the proximal and 31 to the mid or distal segments of the PFA. All records were reviewed by two of the authors (E.J.P. and P.E.B.) and compiled according to the standards suggested by the Ad Hoc Committee on Reporting Standards for Reports Dealing with Lower Limb Ischemia. 4 The patient's age and sex as well as a history of cigarette smoking, diabetes, hypertension, and coronary artery disease were noted. All patients underwent preoperative noninvasive laboratory investigations to determine segmental lower limb pressures, ankle/brachial index. Preoperative angiograms were performed in all patients, and angiograms were reviewed for the presence of atherosclerotic occlusive disease affecting the SFA and/or PFA. Disease of these vessels was graded into complete occlusion, 50% to 99% stenosis, less than 50% stenosis, or no disease. The indications for surgery were disabling clandication (grade I, chronic limb ischemia) and critical limb ischemia. Critical limb ischemia was further defined as ischemic rest pain (grade II, chronic limb ischemia) and acute tissue loss as a result ofischemia (grade III, chronic limb ischemia). All procedures were performed with Dacron bifurcation grafts. The aortic anastomosis was usually performed end to end with continuous 2.0 polypropylene sutures. The distal anastomosis was performed end to side with continuous 5.0 or 6.0 polypropylene. When the distal anastomosis was to the common femoral artery even in the presence of a femoral endarterectomy or an extended tongue onto the PFA this was considered an A-CFA. Systemic anticoagulation (heparin) was administered before damping the aorta and not reversed. Prophylactic antibiotics were used routinely. After completion of the anastomoses, graft patency and runoffwas assessed with use of a hand-held Doppler probe. Patient follow-up Noninvasive studies were performed after operation to obtain objective evidence of improved limb perfusion. After discharge, patients were followed-up at 6 weeks, 3 months, 6 months, i year, and annually thereafter. Noninvasive studies were performed at each visit. For the purpose of this study all patients not seen during the previous 6 months were reviewed
Journal of VASCULAR SURGERY
in a special clinic, and clinical examination and noninvasive studies were repeated. Graft occlusion was suspected if there was deterioration in the clinical status of the limb (i.e., a return or worsening of claudication or critical ischemia) and was confirmed by clinical examination and noninvasive studies plus or minus angiography. Two percent of patients were lost to follow-up annually. Graft patency Primary graft patency was defined as uninterrupted patency of the graft from the time of operation. Secondary patency was defined as restoration of flow in an occluded graft by thrombectomy, thrombolysis, patch angioplasty, and/or extension of the graft to a site distal to the cause of obstruction. Extension of a graft because of distal atherosclerosis was not considered as graft failure, if the graft was patent at the time of the second operation. Major limb amputation in the presence of a patent graft was not considered a graft failure. However, removal of a patent graft because of infection was considered a failure of the graft. Statistics Results of cumulative patency and limb salvage were analyzed by actuarial methods and presented in the form of life tables. Comparison of life-table estimates was performed using the log rank test for significance. Comparisons of demographic features, indications for surgery, noninvasive studies, and angiographic findings between the groups were made with use of the Student t test on samples of equal variance. Differences were considered to have reached statistical significance at the 95% level @ = 0.05). RESULTS Patient characteristics Age and sex distribution was similar between the two groups. No significant difference was observed in the incidence of hypertension, ischemic heart disease, diabetes, or in the prevalence of smokers between both groups (Table I). Subjective evaluation of the indications for surgery showed no difference in the severity of disease between limbs undergoing A-CFA bypass as Compared with A-PFA bypass, with similar incidence of claudication (35% vs 25%), rest pain (48% vs 57%), and tissue loss (17% vs 18%), between the two groups. However, objective evaluation of circulation showed evidence of more severe disease in those limbs revascularised by the A-PFA route. The mean preoperative ankle/brachial index
Volume 16 Number 1 July 1992
Use of PFA as ouq~w vessel 25
Table 11. Preoperative
Table I. Patient characteristics No. of patients Mean age (yr) Hypertension Ischemic heart disease Smokers Respiratory disease Diabetes
Total
A-CFA
A-PFA
143 63.5 47 (33%) 23 (16%) 138 (97%) 99 (69%) 16 (11%)
76 62.5 25 (33%) 12 (16%) 73 (96%) 48 (63%) 10 (13%)
67 64.4 22 (33%) 11 (16%) 65 (97%) 51 (76%) 6 (9%)
was significantly lower in the A-PFA group (0.42) as compared with the A-CFA group (0.70, p < 0.0002). Review of the preoperative angiograms (Table II) showed severe SFA disease ( > 5 0 % stenosis or complete occlusion) in 97% of limbs revascularized by A-PFA as opposed to 58% in the A-CFA group. In a similar manner a greater incidence of severe disease was seen in the PFA in limbs undergoing A-PFA bypass (59%) as compared with A-CFA bypass (21%). The overall 30-day mortality rate was 3.0%. Graft patency Mean follow-up for A-PFA was 32 months, ranging from i to 139 months and for A-CFA was 39 months, ranging from 1 to 142 months. Life tables for primary and secondary graft patency, comparing the two groups, are displayed on Figs. 1 and 2. Primary graft patency for A-PFA bypass was 94% at 3 years and 89% at 5 years, and for the A-CFA bypass was 95% and 92%, respectively. Differences between A-PFA and A-CFA were not statistically significant. Secondary patency for A-PFA was 96% at 3 years and 91% at 5 years, and for A-CFA were 99% and 96%, respectively. These differences again were not statistically significant. Ten grafts occluded in the A-PFA group. Two were reopened by thrombectomy alone and two by thrombectomy and distal extension of the graft to the popliteal artery. No attempt to reopen the graft was made in the remaining six. Seven grafts occluded in the A-CFA group. Two were thrombectomized alone. In two, the graft was revised with extension of the distal anastomosis to the PFA and in one a distal extension to the tibial vessels with in situ vein was performed. Two grafts were not reopened. It was not possible to pinpoint the cause of graft occlusion. Distal bypass A total of 37 (13%) distal bypass procedures to the popliteal or tibial vessels were required to maintain limb salvage or graft patency. Sixteen
angiographic findings SFA disease No. of limbs Occlusion 50% to 99% stenosis < 50% stenosis Normal PFA disease No. of limbs Occlusion
Total
A-CFA
A-PFA
209 126 (60%) 34 (17%)
110 43 (39%) 21 (19%)
99 83 (84%) 13 (13%)
23 (11%) 26 (12%)
21 (19%) 25 (23%)
2 (2%) 1 (1%)
194 42 (22%)
96 12 (12%)
98 30 (31%)
50% to 99%
36 (18%)
9 (9%)
27 (28%)
stenosis < 50% stenosis Normal
40 (21%) 76 (39%)
18 (20%) 57 (59%)
22 (22%) 19 (19%)
(12%) were performed in the A-PFA group and 21 (14%) in the A-CFA group. Nineteen grafts occluded during follow-up resulting in 17 amputations, eight above-knee, and nine below-knee amputations. Most above-knee amputations resulted from loss of PFA flow. Five distal bypasses were performed concomitantly in both groups, all in patients with critical ischemia. Four were to the above-knee popliteal, three to the below-knee popliteal, and three to the tibial vessels. Five grafts occluded (three in the A-PFA group and two in the A-CFA group) between i and 8 months. Of the occluded grafts two resulted in above-knee amputations (both A-CFA group) and two below-knee amputations (both A-PFA group). Eleven limbs in the A-PFA group underwent subsequent distal revascularization: two after thrombectomy of an occluded graft and nine for progression of distal disease. Three were to the above-knee and four to the below-knee popliteal and four were to tibia/vessels. Four occluded between 1 day and 5 months; one below-knee and three tibial grafts. Three resulted in below-knee and one in above-knee amputations. Seven grafts were patent at between 4 and 50 months. Sixteen limbs in the A-CFA group had subsequent distal revascularization: one after thrombectomy of an occluded graft and 15 for progression of distal disease. Four were to the above-knee popliteal, six to the below-knee popliteal, and six to the tibial vessels. Ten grafts failed between 1 week and 91 months of which nine resulted in major limb amputations, five above-knee and four below-knee. Six remain patent between 3 weeks and 47 months.
Journa} c VASCULAt
26
Prendiville et al.
SURGER~
v
"~O 0 1~36
t23
112
88
63
49
5
1
2
k
2
T
99
94
74
48
2'4
3'6
38
r A-CFA i A-~FA
O_ 0 C 0._ © >
"4
60~ 40 2O
E C}
0
0
1'2
£8
6'0
Time (months) Fig. 1. Life-table curves of A-CFA (n = 151) and A-PFA (n patency rates (-+ SEM, p = NS, log rank test).
137
100 F
124
113
T 9
101
89 r
96
63
49
>
-
39
r A-CFA A-PFA
~- 801 o_ I ®
134) comparing primary
50
cO
~c_
=
36
22
4'8
6'0
j a@-
t j
20-'
Z
0
1'2
2'4 3'6 Time (months)
'
Fig. 2. Life-table curves ofA-CFA (n = 151) and A-PFA (n = 134) comparing secondary patency rates ( - SEM, p = NS, log rank test).
Limb salvage Limb salvage at 5 years was 90% in A-PFA and 94% in A-CFA group (Fig. 3); differences were not significant. Nine major limb amputations were performed in the A-PFA group, six above-knee and three below-knee. Three patients had patent A-PFA grafts at the time of amputation; however, two had distal bypass grafts that were occluded at amputation. One aortobiprofunda graft was removed because of infection, and the limbs were revascularized by an axillobifemoral graft. The latter graft subsequently failed, resulting in bilateral above-knee amputations. Four amputations occurred within the first 6 months
after operation (two within the first month) and one after 5 years. Ten major limb amputations were performed in the A-CFA group, seven above and three below-knee. Four grafts were open at amputation, but all had occluded distal grafts. Four amputations were performed within the first 6 months, and three were carried out later than 5 years after operation. Proximal versus mid/distal PFA One hundred thirty-four limbs were revascularized by the A-PFA route. Of these the distal anastomosis was to the proximal PFA in 103 limbs
Volume 16 Number 1 July 1992
Use of PFA as outflow vessel 27
137 14
123
1001~2°~ 50
112 95
89
67
51
T
T
T
3g I A-CFA
75
50
[ 36
i A_pFA 21
48
60
50
,--4 40
.~1 20 4J ,--t 0
12
2~ 56 Time (months)
Fig. 3. Life-table curve ofA-CFA (n = 151) and A-PFA (n = 134) comparing limb salvage rates ( -+-SEM, p -- NS, log rank test). and to the mid or distal PFA in 31 limbs. Because of the smaller numbers in the latter group it was not meaningful to make comparisons beyond 3 years. Three-year primary patency for A-prox PFA was 94% and for A-mid/distal PFA was 91.5%. Secondary patency was 96% and 95%, respectively (Fig. 4). Limb salvage at 3 years was 94% in both groups (Fig. 5). To maintain limb salvage 12 (13%) distal bypasses were performed in the A-Prox PFA group and 4 (11.5%) in the A-mid/distal PFA group. None of the above differences reached statistical significance. DISCUSSION
A review of our results of aortofemoral bypass using the PFA as outflow has revealed that (1) cumulative patency and limb salvage at 5 years is similar to results for aorta-common femoral bypass, (2) despite the high incidence of severe occlusive disease of the SFA, few distal bypass procedures were required to maintain limb salvage, (3) results of bypass to the mid or distal PFA were similar, at 3 years, to results for proximal PFA bypass, and (4) in the overall series of 19 major limb amputations, 16 had had concomitant (5 of 10) or subsequent (11 of 27) distal infrainguinal bypasses. Coexisting atherosclerotic occlusive disease of the aortoiliac and femoropopliteal vessels may result in graft failure if either is bypassed alone, because of inadequate inflow or outflow. However, long prosthetic grafts, from aorta to popliteal or distal vessels, result in poor long-term patency and limb salvage, s,6 The PFA and its branches run in parallel with the SFA
and form extensive communications with the popliteal and distal lower limb vessels. HemodynamicaUy, blood flow through the PFA is equal to that through the CFA, 7 when both the superficial and PFA are patent, and is twice that through a femoropopliteal vein graft, s Therefore, flow in a graft anastomosed directly to the PFA, beyond an area of stenosis or occlusion, should equal flow in a graft to the CFA when the SFA and PFA are patent. Our results show no significant difference in aortofemoral graft function whether the distal anastomosis is to the PFA or CFA. Use of the PFA to revascularize the lower limb has remained controversial since first described by Morris et al.9 in 1961. Most reports ofPFA revascularization describe the use of profundaplasty alone or in combination with an inflow procedure to the CFA. Profundaplasty alone, although showing good shortterm results, ~°-12has not shown any long-term benefit at 3 and 5 years in limbs operated on for limbthreatening ischemia and has led many surgeons to regard the PFA as a poor outflow vessel, is to be used only in patients who are poor risk for long vascular reconstructions. 14 Sequential A-CFA bypass and extended profundaplasty have improved limb salvage in the critically ischemic limb. is Few reports exist in the literature of bypass directly to the PFA. Youmans et al.16 described a small series of 15 limbs where bypass was performed to the PFA up to 3 inches from its origin. Fourteen patients had significant symptomatic improvement. Ouriel et al.,17 in a larger series of 53 inflow procedures, 3 cm or more from the PFA origin, reported 96% cumulative patcncy and 86%
Journal of VASCULAR SURGERY
28 Prendiville et al.
96
83
78
65
41 T A-PR~3X P~A
23
L 18
l 18
!
~5
i A-MIO/OIST 9 ~FA
I
°° 8o]1 CZ >,
i
60q
i
G O-
_...s
(D
o!
1'2 2'4 Time (months)
0
5'6
Fig. 4. Life-table curves ofA-prox PFA (n = 103) and A-mid/dist PFA (n = 34) comparing secondary patency rates ( + SEM, p = NS, log rank test).
m lOO
23
18
97
83
1.4
18
13
77
62
9 A-~IO/DISr
ad
A-PROX PFA
80 60
"~
40
> .M
2o
4-1
,-4
0 rj
0
24
I'2 Time
3'6
(months)
Fig. 5. Life-table curves ofA-prox PFA (n = 103) and A-mid/dist PFA (n = 34) comparing limb salvage rates (-+ SEM, p = NS, log rank test). limb salvage at 4 years. Our results compare favorably to these for both overall A-PFA and aorta-mid/distal PFA bypass. Concomitant aortofemoral and femoropopliteal bypass has been advocated by some authors as the surgery of choice for coexisting aortoiliac and femoropopliteal occlusive disease, xs However, long-term results are poor, with disappointing long-term patency in the infrainguinal grafts. Benson et al.6 in a review of 12 patients with combined aortofemoral and femoropopliteal grafts, observed satisfactory relief of symptoms in only four patients. The femoropopliteal graft had occluded within 18 months in nine cases, with subsequent failure of the
aortofemoral graft in six. In only one case did the aortofemoral graft occlude first. Heyden et al.s reported 39% cumulative patency and 35% limb salvage at 5 years in 66 limbs with combined procedures, compared with 77% and 70%, respectively for aortofemoral bypass with profundaplasty. Our results show 91% cumulative patency and 90% limb salvage at 5 years after A-PFA bypass. Several factors may be responsible for these improved results, including better patient selection for surgery, improved surgical technique, and a more aggressive approach to reintervention. Sixteen of the 19 major limb amputations in our series had had concomitant or subsequent infrainguinal bypass surgery. M-
Volume 16 Number 1 July 1992
though this probably represents more severe distal disease in these limbs, one may speculate that graft failure may have been aggravated by increased platelet activation by the greater graft length resulting in increased intimal hyperplasia at the distal anastomosis and progression of atheroma in the distal vessels. 19,20 Atheroma of the PFA most commonly affects the ostium and proximal vessel, 3 and even in the presence of a completely occluded PFA, which is not visualized on angiography, it is our experience that exploration of the PFA will usually reveal a distal vessel that is soft and free ofatheroma. Although the number of bypass procedures in our series to the mid and distal segments of the PFA were small, the results at 3 years for graft patency and limb salvage were similar to those for proximal PFA bypass. Our results are similar to those of Ouriel et al.17 for bypass to the distal PFA. CONCLUSION
The PFA is a crucial vessel for maintenance of blood flow to the distal lower limb in the presence of severe occlusive disease of the SFA. When aortoiliac and femoropopliteal atherosclerotic disease coexist, aortofemoral bypass using the PFA as the outflow vessel results in excellent long-term graft patency and limb salvage. Where the PFA is occluded or atheroma exists beyond the proximal segment of the vessel, exploration of the mid or distal segments will usually reveal a soft, disease-free segment, which is a durable outflow vessel with results similar to proximal PFA bypass. Despite the high incidence of severe atherosclerosis affecting the SFA few distal bypass procedures were required to maintain limb salvage whether proximal, mid, or distal segments of the PFA were used as the outflow vessel. We recommend the use of the PFA as an outflow vessel in aortofemoral bypass surgery when aortoiliac and femoropopliteal atherosclerotic occlusive disease coexist, rather than concomitant A-CFA and femoropopliteal/distal bypass grafts. REFERENCES 1. Crawford ES, Bomberger RA, Glaeser DH, Salen SA, Russell WL. Aorto-iliac occlusive disease: factors influencing survival and function following reconstructive operation over a twenty-five year period. Surgery 1981;90:1055-67. 2. Beales JSM, Adcock FA, Frawley JS, et al. The radiological assessment of disease of the profunda femoris artery. Br J Radiol 1971;44:854-9.
Use of PFA as ouq~ow vessel 29
3. Martin P, Frawley JE, Barahas AP, Rosengarten DS. On the surgery of atherosclerosis of the profunda femoris artery. Surgery 1972;71:182-9. 4. Rutherford RB, Flanigan DP, Gupta SK, et al. Suggested standards for reports dealing with lower extremity ischemia. J VASCSURG 1986;4:80-94. 5. Heyden B, Vollmar J, Vass EU. Principles for operations for combined aortoiliac and femoropopliteal occlusive lesions. Surg Gynecol Obstet 1980; 151: 519-24. 6. Benson JR, Whelan TJ, Cohen A, Spencer FC. Combined aorto-iliac and femoral popliteal occlusive disease: limitations of total aortofemoropopliteal bypass. Ann Surg 1966;163: 121-30. 7. Bernard VM, Ray LI, Militello JP. The role of angioplasty of the profunda femoris artery in revascularization of the ischemic limb. Surg Gynecol Obstet 1976;142:840-4. 8. Martin P, Jamieson C. The rational for and measurement after profundapalsty. Surg Clin North Am 1974;54:95-109. 9. Morris GC, Edwards W, Cooley DA, Crawford ES, DeBakey ME. Surgical importance of the profunda femoris artery. Arch Surg 1961;82:52-7. 10. Cotton LT, Roberts VT. Extended deep femoral angioplasty: an alternative to femoropopliteal bypass. Br J Surg 1975;62: 340-3. 11. David TE, Drezner AD. Extended profundaplasty for limb salvage. Surgery 1978;84:758-63. 12. Leather RP, Shah DM, Karmody AM. The use of extended profundaplasty in limb salvage. Am J Surg 1978;136: 359-62. 13. Ward AS, Morris-Jones W. Long term results of profimdaplasty in femoropopliteal arterial occlusion. Br J Surg 1977;64:365-7. 14. Miksic K, Novak B. Profunda femoris revasoalarization in limb salvage. J Cardiovasc Surg 1986;27:544-51. 15. Pearce WH, Kempczinski RF. Extended autogenous profundaplasty and aortofemoral grafting: an alternative to synchronous distal bypass. J VASESURG 1984;1:455-8. 16. Youmans CR, Hopkins JW, Derrick JR. Transinguinal distal profunda femoris revascularization. Am J Surg 1969;118: 909-14. 17. Ouriel K, DeWesse JA, Ricotta JJ, Green RM. Revascularization of the distal profunda femoris artery in the reconstructive treatment of aortoiliac occlusive disease. J VAsc SURG 1987;6:217-20. 18. Dardik H, Ibrahim IM, Jarrah M, Sussman D, Dardik I. Synchronous aortofemoral and iliofemoral bypass with revascularization of the lower extremity. Surg Gynecol Obstet 1979;149:676-80. 19. LoGerfo FW, Quist WC, Nowak MD, Crawshaw HM, Haudenschild CC. Downstream anastomotic hyperplasia: a mechanism of failure in Dacron arterial grafts. Ann Surg 1983; 197:479-83. 20. O'Donnell TF, Mackey W, McCuUough JL, et al. Correlation of operative findings with angiographic and noninvasive haemodynamic factors associated with failure of polytetrafluoroethylene grafts. J VASCSUV,G 1984;1:136-48. Submitted Mar. 4, 1991; accepted Oct. 11, 1991.
Aorta-common femoral artery (A-CFA) bypass is the standard operation for relief of aortoiliac occlusive disease and has excellent results, with patency rates greater than 85% at 5 years) However, significant CFA outflow obstruction frequently coexists with aortoiliac disease, and may necessitate distal bypass surgery to maintain graft patency and limb salvage. The profunda femoris (PFA), through its many branches, forms extensive communication with the popliteal artery at the knee and the tibial vessels in the calf. In the presence of superficial femoral artery (SFA) occlusion, the PFA provides the major collateral circulation through which blood reaches the distal limb. Although atherosclerosis does not affect the PFA as severely as the SFA, Beales et al.2 reviewing angiograms from ischemic limbs, observed atheroma in the ostium of the PFA in 39% of cases. Martin et al. s noted that the atheroma affecting the From the Department of Vascular Surgery, St, James'sHospital, Dublin, Ireland. Reprint requests: ProfessorD.G. Shanik, Departmentof Vascular Surgery, St. James's Hospital, Dublin 8. Ireland. 24/1/34a45
PFA was confined to the ostium or proximal segment in 74% of cases but affected the mid and distal segments in 26%. In the presence of coexisting aortoiliac occlusive disease and CFA outflow obstruction, use of the PFA as the outflow vessel in aortofemoral bypass will optimize the collateral circulation, and may lessen the need for distal bypass surgery. To test this hypothesis we have reviewed the results of all aortofemoral bypass procedures performed for aortoiliac occlusive disease in our unit and compared cumulative patency, limb salvage, and the need for distal bypass surgery in A-CFA and A-PFA procedures. When atheroma cxists beyond the ostium of the PFA it may be necessary to perform a more distal PFA bypass. We also analyzed the results ofaortofemoral bypass to the proximal, mid, and distal segments of the PFA. PATIENTS A N D M E T H O D S A review of all aortofemoral bypass procedures performed in St. James's Hospital over a 12-year period, between January 1978 and December 1989, identified a total of 285 grafts. O f these, 145 grafts 23
24 Prendiville et al.
were inserted for aortoiliac occlusive disease. Bypass procedures for aneurysmal disease were excluded from this study. Two hundred eighty-five limbs were revascularized. In 151 limbs the CFA was used as the outflow vessel, and in 134 limbs the PFA was used. O f A-PFA bypasses 103 were to the proximal and 31 to the mid or distal segments of the PFA. All records were reviewed by two of the authors (E.J.P. and P.E.B.) and compiled according to the standards suggested by the Ad Hoc Committee on Reporting Standards for Reports Dealing with Lower Limb Ischemia. 4 The patient's age and sex as well as a history of cigarette smoking, diabetes, hypertension, and coronary artery disease were noted. All patients underwent preoperative noninvasive laboratory investigations to determine segmental lower limb pressures, ankle/brachial index. Preoperative angiograms were performed in all patients, and angiograms were reviewed for the presence of atherosclerotic occlusive disease affecting the SFA and/or PFA. Disease of these vessels was graded into complete occlusion, 50% to 99% stenosis, less than 50% stenosis, or no disease. The indications for surgery were disabling clandication (grade I, chronic limb ischemia) and critical limb ischemia. Critical limb ischemia was further defined as ischemic rest pain (grade II, chronic limb ischemia) and acute tissue loss as a result ofischemia (grade III, chronic limb ischemia). All procedures were performed with Dacron bifurcation grafts. The aortic anastomosis was usually performed end to end with continuous 2.0 polypropylene sutures. The distal anastomosis was performed end to side with continuous 5.0 or 6.0 polypropylene. When the distal anastomosis was to the common femoral artery even in the presence of a femoral endarterectomy or an extended tongue onto the PFA this was considered an A-CFA. Systemic anticoagulation (heparin) was administered before damping the aorta and not reversed. Prophylactic antibiotics were used routinely. After completion of the anastomoses, graft patency and runoffwas assessed with use of a hand-held Doppler probe. Patient follow-up Noninvasive studies were performed after operation to obtain objective evidence of improved limb perfusion. After discharge, patients were followed-up at 6 weeks, 3 months, 6 months, i year, and annually thereafter. Noninvasive studies were performed at each visit. For the purpose of this study all patients not seen during the previous 6 months were reviewed
Journal of VASCULAR SURGERY
in a special clinic, and clinical examination and noninvasive studies were repeated. Graft occlusion was suspected if there was deterioration in the clinical status of the limb (i.e., a return or worsening of claudication or critical ischemia) and was confirmed by clinical examination and noninvasive studies plus or minus angiography. Two percent of patients were lost to follow-up annually. Graft patency Primary graft patency was defined as uninterrupted patency of the graft from the time of operation. Secondary patency was defined as restoration of flow in an occluded graft by thrombectomy, thrombolysis, patch angioplasty, and/or extension of the graft to a site distal to the cause of obstruction. Extension of a graft because of distal atherosclerosis was not considered as graft failure, if the graft was patent at the time of the second operation. Major limb amputation in the presence of a patent graft was not considered a graft failure. However, removal of a patent graft because of infection was considered a failure of the graft. Statistics Results of cumulative patency and limb salvage were analyzed by actuarial methods and presented in the form of life tables. Comparison of life-table estimates was performed using the log rank test for significance. Comparisons of demographic features, indications for surgery, noninvasive studies, and angiographic findings between the groups were made with use of the Student t test on samples of equal variance. Differences were considered to have reached statistical significance at the 95% level @ = 0.05). RESULTS Patient characteristics Age and sex distribution was similar between the two groups. No significant difference was observed in the incidence of hypertension, ischemic heart disease, diabetes, or in the prevalence of smokers between both groups (Table I). Subjective evaluation of the indications for surgery showed no difference in the severity of disease between limbs undergoing A-CFA bypass as Compared with A-PFA bypass, with similar incidence of claudication (35% vs 25%), rest pain (48% vs 57%), and tissue loss (17% vs 18%), between the two groups. However, objective evaluation of circulation showed evidence of more severe disease in those limbs revascularised by the A-PFA route. The mean preoperative ankle/brachial index
Volume 16 Number 1 July 1992
Use of PFA as ouq~w vessel 25
Table 11. Preoperative
Table I. Patient characteristics No. of patients Mean age (yr) Hypertension Ischemic heart disease Smokers Respiratory disease Diabetes
Total
A-CFA
A-PFA
143 63.5 47 (33%) 23 (16%) 138 (97%) 99 (69%) 16 (11%)
76 62.5 25 (33%) 12 (16%) 73 (96%) 48 (63%) 10 (13%)
67 64.4 22 (33%) 11 (16%) 65 (97%) 51 (76%) 6 (9%)
was significantly lower in the A-PFA group (0.42) as compared with the A-CFA group (0.70, p < 0.0002). Review of the preoperative angiograms (Table II) showed severe SFA disease ( > 5 0 % stenosis or complete occlusion) in 97% of limbs revascularized by A-PFA as opposed to 58% in the A-CFA group. In a similar manner a greater incidence of severe disease was seen in the PFA in limbs undergoing A-PFA bypass (59%) as compared with A-CFA bypass (21%). The overall 30-day mortality rate was 3.0%. Graft patency Mean follow-up for A-PFA was 32 months, ranging from i to 139 months and for A-CFA was 39 months, ranging from 1 to 142 months. Life tables for primary and secondary graft patency, comparing the two groups, are displayed on Figs. 1 and 2. Primary graft patency for A-PFA bypass was 94% at 3 years and 89% at 5 years, and for the A-CFA bypass was 95% and 92%, respectively. Differences between A-PFA and A-CFA were not statistically significant. Secondary patency for A-PFA was 96% at 3 years and 91% at 5 years, and for A-CFA were 99% and 96%, respectively. These differences again were not statistically significant. Ten grafts occluded in the A-PFA group. Two were reopened by thrombectomy alone and two by thrombectomy and distal extension of the graft to the popliteal artery. No attempt to reopen the graft was made in the remaining six. Seven grafts occluded in the A-CFA group. Two were thrombectomized alone. In two, the graft was revised with extension of the distal anastomosis to the PFA and in one a distal extension to the tibial vessels with in situ vein was performed. Two grafts were not reopened. It was not possible to pinpoint the cause of graft occlusion. Distal bypass A total of 37 (13%) distal bypass procedures to the popliteal or tibial vessels were required to maintain limb salvage or graft patency. Sixteen
angiographic findings SFA disease No. of limbs Occlusion 50% to 99% stenosis < 50% stenosis Normal PFA disease No. of limbs Occlusion
Total
A-CFA
A-PFA
209 126 (60%) 34 (17%)
110 43 (39%) 21 (19%)
99 83 (84%) 13 (13%)
23 (11%) 26 (12%)
21 (19%) 25 (23%)
2 (2%) 1 (1%)
194 42 (22%)
96 12 (12%)
98 30 (31%)
50% to 99%
36 (18%)
9 (9%)
27 (28%)
stenosis < 50% stenosis Normal
40 (21%) 76 (39%)
18 (20%) 57 (59%)
22 (22%) 19 (19%)
(12%) were performed in the A-PFA group and 21 (14%) in the A-CFA group. Nineteen grafts occluded during follow-up resulting in 17 amputations, eight above-knee, and nine below-knee amputations. Most above-knee amputations resulted from loss of PFA flow. Five distal bypasses were performed concomitantly in both groups, all in patients with critical ischemia. Four were to the above-knee popliteal, three to the below-knee popliteal, and three to the tibial vessels. Five grafts occluded (three in the A-PFA group and two in the A-CFA group) between i and 8 months. Of the occluded grafts two resulted in above-knee amputations (both A-CFA group) and two below-knee amputations (both A-PFA group). Eleven limbs in the A-PFA group underwent subsequent distal revascularization: two after thrombectomy of an occluded graft and nine for progression of distal disease. Three were to the above-knee and four to the below-knee popliteal and four were to tibia/vessels. Four occluded between 1 day and 5 months; one below-knee and three tibial grafts. Three resulted in below-knee and one in above-knee amputations. Seven grafts were patent at between 4 and 50 months. Sixteen limbs in the A-CFA group had subsequent distal revascularization: one after thrombectomy of an occluded graft and 15 for progression of distal disease. Four were to the above-knee popliteal, six to the below-knee popliteal, and six to the tibial vessels. Ten grafts failed between 1 week and 91 months of which nine resulted in major limb amputations, five above-knee and four below-knee. Six remain patent between 3 weeks and 47 months.
Journa} c VASCULAt
26
Prendiville et al.
SURGER~
v
"~O 0 1~36
t23
112
88
63
49
5
1
2
k
2
T
99
94
74
48
2'4
3'6
38
r A-CFA i A-~FA
O_ 0 C 0._ © >
"4
60~ 40 2O
E C}
0
0
1'2
£8
6'0
Time (months) Fig. 1. Life-table curves of A-CFA (n = 151) and A-PFA (n patency rates (-+ SEM, p = NS, log rank test).
137
100 F
124
113
T 9
101
89 r
96
63
49
>
-
39
r A-CFA A-PFA
~- 801 o_ I ®
134) comparing primary
50
cO
~c_
=
36
22
4'8
6'0
j a@-
t j
20-'
Z
0
1'2
2'4 3'6 Time (months)
'
Fig. 2. Life-table curves ofA-CFA (n = 151) and A-PFA (n = 134) comparing secondary patency rates ( - SEM, p = NS, log rank test).
Limb salvage Limb salvage at 5 years was 90% in A-PFA and 94% in A-CFA group (Fig. 3); differences were not significant. Nine major limb amputations were performed in the A-PFA group, six above-knee and three below-knee. Three patients had patent A-PFA grafts at the time of amputation; however, two had distal bypass grafts that were occluded at amputation. One aortobiprofunda graft was removed because of infection, and the limbs were revascularized by an axillobifemoral graft. The latter graft subsequently failed, resulting in bilateral above-knee amputations. Four amputations occurred within the first 6 months
after operation (two within the first month) and one after 5 years. Ten major limb amputations were performed in the A-CFA group, seven above and three below-knee. Four grafts were open at amputation, but all had occluded distal grafts. Four amputations were performed within the first 6 months, and three were carried out later than 5 years after operation. Proximal versus mid/distal PFA One hundred thirty-four limbs were revascularized by the A-PFA route. Of these the distal anastomosis was to the proximal PFA in 103 limbs
Volume 16 Number 1 July 1992
Use of PFA as outflow vessel 27
137 14
123
1001~2°~ 50
112 95
89
67
51
T
T
T
3g I A-CFA
75
50
[ 36
i A_pFA 21
48
60
50
,--4 40
.~1 20 4J ,--t 0
12
2~ 56 Time (months)
Fig. 3. Life-table curve ofA-CFA (n = 151) and A-PFA (n = 134) comparing limb salvage rates ( -+-SEM, p -- NS, log rank test). and to the mid or distal PFA in 31 limbs. Because of the smaller numbers in the latter group it was not meaningful to make comparisons beyond 3 years. Three-year primary patency for A-prox PFA was 94% and for A-mid/distal PFA was 91.5%. Secondary patency was 96% and 95%, respectively (Fig. 4). Limb salvage at 3 years was 94% in both groups (Fig. 5). To maintain limb salvage 12 (13%) distal bypasses were performed in the A-Prox PFA group and 4 (11.5%) in the A-mid/distal PFA group. None of the above differences reached statistical significance. DISCUSSION
A review of our results of aortofemoral bypass using the PFA as outflow has revealed that (1) cumulative patency and limb salvage at 5 years is similar to results for aorta-common femoral bypass, (2) despite the high incidence of severe occlusive disease of the SFA, few distal bypass procedures were required to maintain limb salvage, (3) results of bypass to the mid or distal PFA were similar, at 3 years, to results for proximal PFA bypass, and (4) in the overall series of 19 major limb amputations, 16 had had concomitant (5 of 10) or subsequent (11 of 27) distal infrainguinal bypasses. Coexisting atherosclerotic occlusive disease of the aortoiliac and femoropopliteal vessels may result in graft failure if either is bypassed alone, because of inadequate inflow or outflow. However, long prosthetic grafts, from aorta to popliteal or distal vessels, result in poor long-term patency and limb salvage, s,6 The PFA and its branches run in parallel with the SFA
and form extensive communications with the popliteal and distal lower limb vessels. HemodynamicaUy, blood flow through the PFA is equal to that through the CFA, 7 when both the superficial and PFA are patent, and is twice that through a femoropopliteal vein graft, s Therefore, flow in a graft anastomosed directly to the PFA, beyond an area of stenosis or occlusion, should equal flow in a graft to the CFA when the SFA and PFA are patent. Our results show no significant difference in aortofemoral graft function whether the distal anastomosis is to the PFA or CFA. Use of the PFA to revascularize the lower limb has remained controversial since first described by Morris et al.9 in 1961. Most reports ofPFA revascularization describe the use of profundaplasty alone or in combination with an inflow procedure to the CFA. Profundaplasty alone, although showing good shortterm results, ~°-12has not shown any long-term benefit at 3 and 5 years in limbs operated on for limbthreatening ischemia and has led many surgeons to regard the PFA as a poor outflow vessel, is to be used only in patients who are poor risk for long vascular reconstructions. 14 Sequential A-CFA bypass and extended profundaplasty have improved limb salvage in the critically ischemic limb. is Few reports exist in the literature of bypass directly to the PFA. Youmans et al.16 described a small series of 15 limbs where bypass was performed to the PFA up to 3 inches from its origin. Fourteen patients had significant symptomatic improvement. Ouriel et al.,17 in a larger series of 53 inflow procedures, 3 cm or more from the PFA origin, reported 96% cumulative patcncy and 86%
Journal of VASCULAR SURGERY
28 Prendiville et al.
96
83
78
65
41 T A-PR~3X P~A
23
L 18
l 18
!
~5
i A-MIO/OIST 9 ~FA
I
°° 8o]1 CZ >,
i
60q
i
G O-
_...s
(D
o!
1'2 2'4 Time (months)
0
5'6
Fig. 4. Life-table curves ofA-prox PFA (n = 103) and A-mid/dist PFA (n = 34) comparing secondary patency rates ( + SEM, p = NS, log rank test).
m lOO
23
18
97
83
1.4
18
13
77
62
9 A-~IO/DISr
ad
A-PROX PFA
80 60
"~
40
> .M
2o
4-1
,-4
0 rj
0
24
I'2 Time
3'6
(months)
Fig. 5. Life-table curves ofA-prox PFA (n = 103) and A-mid/dist PFA (n = 34) comparing limb salvage rates (-+ SEM, p = NS, log rank test). limb salvage at 4 years. Our results compare favorably to these for both overall A-PFA and aorta-mid/distal PFA bypass. Concomitant aortofemoral and femoropopliteal bypass has been advocated by some authors as the surgery of choice for coexisting aortoiliac and femoropopliteal occlusive disease, xs However, long-term results are poor, with disappointing long-term patency in the infrainguinal grafts. Benson et al.6 in a review of 12 patients with combined aortofemoral and femoropopliteal grafts, observed satisfactory relief of symptoms in only four patients. The femoropopliteal graft had occluded within 18 months in nine cases, with subsequent failure of the
aortofemoral graft in six. In only one case did the aortofemoral graft occlude first. Heyden et al.s reported 39% cumulative patency and 35% limb salvage at 5 years in 66 limbs with combined procedures, compared with 77% and 70%, respectively for aortofemoral bypass with profundaplasty. Our results show 91% cumulative patency and 90% limb salvage at 5 years after A-PFA bypass. Several factors may be responsible for these improved results, including better patient selection for surgery, improved surgical technique, and a more aggressive approach to reintervention. Sixteen of the 19 major limb amputations in our series had had concomitant or subsequent infrainguinal bypass surgery. M-
Volume 16 Number 1 July 1992
though this probably represents more severe distal disease in these limbs, one may speculate that graft failure may have been aggravated by increased platelet activation by the greater graft length resulting in increased intimal hyperplasia at the distal anastomosis and progression of atheroma in the distal vessels. 19,20 Atheroma of the PFA most commonly affects the ostium and proximal vessel, 3 and even in the presence of a completely occluded PFA, which is not visualized on angiography, it is our experience that exploration of the PFA will usually reveal a distal vessel that is soft and free ofatheroma. Although the number of bypass procedures in our series to the mid and distal segments of the PFA were small, the results at 3 years for graft patency and limb salvage were similar to those for proximal PFA bypass. Our results are similar to those of Ouriel et al.17 for bypass to the distal PFA. CONCLUSION
The PFA is a crucial vessel for maintenance of blood flow to the distal lower limb in the presence of severe occlusive disease of the SFA. When aortoiliac and femoropopliteal atherosclerotic disease coexist, aortofemoral bypass using the PFA as the outflow vessel results in excellent long-term graft patency and limb salvage. Where the PFA is occluded or atheroma exists beyond the proximal segment of the vessel, exploration of the mid or distal segments will usually reveal a soft, disease-free segment, which is a durable outflow vessel with results similar to proximal PFA bypass. Despite the high incidence of severe atherosclerosis affecting the SFA few distal bypass procedures were required to maintain limb salvage whether proximal, mid, or distal segments of the PFA were used as the outflow vessel. We recommend the use of the PFA as an outflow vessel in aortofemoral bypass surgery when aortoiliac and femoropopliteal atherosclerotic occlusive disease coexist, rather than concomitant A-CFA and femoropopliteal/distal bypass grafts. REFERENCES 1. Crawford ES, Bomberger RA, Glaeser DH, Salen SA, Russell WL. Aorto-iliac occlusive disease: factors influencing survival and function following reconstructive operation over a twenty-five year period. Surgery 1981;90:1055-67. 2. Beales JSM, Adcock FA, Frawley JS, et al. The radiological assessment of disease of the profunda femoris artery. Br J Radiol 1971;44:854-9.
Use of PFA as ouq~ow vessel 29
3. Martin P, Frawley JE, Barahas AP, Rosengarten DS. On the surgery of atherosclerosis of the profunda femoris artery. Surgery 1972;71:182-9. 4. Rutherford RB, Flanigan DP, Gupta SK, et al. Suggested standards for reports dealing with lower extremity ischemia. J VASCSURG 1986;4:80-94. 5. Heyden B, Vollmar J, Vass EU. Principles for operations for combined aortoiliac and femoropopliteal occlusive lesions. Surg Gynecol Obstet 1980; 151: 519-24. 6. Benson JR, Whelan TJ, Cohen A, Spencer FC. Combined aorto-iliac and femoral popliteal occlusive disease: limitations of total aortofemoropopliteal bypass. Ann Surg 1966;163: 121-30. 7. Bernard VM, Ray LI, Militello JP. The role of angioplasty of the profunda femoris artery in revascularization of the ischemic limb. Surg Gynecol Obstet 1976;142:840-4. 8. Martin P, Jamieson C. The rational for and measurement after profundapalsty. Surg Clin North Am 1974;54:95-109. 9. Morris GC, Edwards W, Cooley DA, Crawford ES, DeBakey ME. Surgical importance of the profunda femoris artery. Arch Surg 1961;82:52-7. 10. Cotton LT, Roberts VT. Extended deep femoral angioplasty: an alternative to femoropopliteal bypass. Br J Surg 1975;62: 340-3. 11. David TE, Drezner AD. Extended profundaplasty for limb salvage. Surgery 1978;84:758-63. 12. Leather RP, Shah DM, Karmody AM. The use of extended profundaplasty in limb salvage. Am J Surg 1978;136: 359-62. 13. Ward AS, Morris-Jones W. Long term results of profimdaplasty in femoropopliteal arterial occlusion. Br J Surg 1977;64:365-7. 14. Miksic K, Novak B. Profunda femoris revasoalarization in limb salvage. J Cardiovasc Surg 1986;27:544-51. 15. Pearce WH, Kempczinski RF. Extended autogenous profundaplasty and aortofemoral grafting: an alternative to synchronous distal bypass. J VASESURG 1984;1:455-8. 16. Youmans CR, Hopkins JW, Derrick JR. Transinguinal distal profunda femoris revascularization. Am J Surg 1969;118: 909-14. 17. Ouriel K, DeWesse JA, Ricotta JJ, Green RM. Revascularization of the distal profunda femoris artery in the reconstructive treatment of aortoiliac occlusive disease. J VAsc SURG 1987;6:217-20. 18. Dardik H, Ibrahim IM, Jarrah M, Sussman D, Dardik I. Synchronous aortofemoral and iliofemoral bypass with revascularization of the lower extremity. Surg Gynecol Obstet 1979;149:676-80. 19. LoGerfo FW, Quist WC, Nowak MD, Crawshaw HM, Haudenschild CC. Downstream anastomotic hyperplasia: a mechanism of failure in Dacron arterial grafts. Ann Surg 1983; 197:479-83. 20. O'Donnell TF, Mackey W, McCuUough JL, et al. Correlation of operative findings with angiographic and noninvasive haemodynamic factors associated with failure of polytetrafluoroethylene grafts. J VASCSUV,G 1984;1:136-48. Submitted Mar. 4, 1991; accepted Oct. 11, 1991.
