Autogenous vein grafts for femorofemoral revascularization in contaminated or infected fields Albert G. Hakaim, MD,* Norman R. Hertzer, MD, Patrick J. O'Hara, MD, Leonard P. Krajewski, MD, and Edwin G. Beven, MD, Cleveland) Ohio

Purpose: The purpose of this study was to determine the indications and the long-term results for femorofemoral revascularization with autogenous vein grafts (AVG) rather than conventional synthetic materials. Methods: A consecutive series of 13 men and 12 women (mean age 64 years) receiving femorofemoral AVG was collected during a retrospective review of our experience from 1979 to 1992. Sixteen of these 25 patients required revascularization for acute ischemia in the presence of potential groin contamination (group 1), and the remaining nine had frank infections involving previous synthetic grafts (group 2). New grafts were constructed with the greater saphenous vein in 23 patients (92%) and with the cephalic vein in two. Results: One patient in each group (8%) died after operation with patent grafts. Preoperative and postoperative ankle{brachial indexes (ABI) were available for 12 patients in group 1 and for seven patients in group 2. There was significant improvement in the ABI among 11 patients with ischemia in group 1 (mean 0.33 ± 0.13; P = 0.0001), whereas no change in the ABI occurred in conjunction with the replacement of patent infected grafts among six patients in group 2 (mean 0.04 ± 0.17;p = NS). One patient from each group sustained a reduction in postoperative ABI for the recipient limb despite a patent AVG. Cumulative 3-year survival and primary patency estimates for all 23 operative survivors were 63% and 75%, respectively. Late results seemed to be comparable in group 1 and group 2, but each contained too few patients for adequate analysis. Conclusions: Despite their infrequent use, AVG seem to represent a reasonable alternative to synthetic bypass for femorofemoral revascularization in patients having either contaminated wounds or established graft infections. (J VASe SURG 1994;19:912-5.)

Synthetic materials undoubtedly are used in the vast majority of patients who require elective femo­rofemoral bypass, and only a few published reports during the past 20 years have described the occasional use of autogenous vein grafts (AVG) for femoro­femoral revascularization. In some representative early series, the choice of A VG for elective procedures may merely have reflected an ipitial reluctance to construct synthetic grafts under circumstances in which the greater saphenous vein (GSV) might have

superior results.1,2 Once abundant data became available to indicate that the late patency rates of fabric or polytetrafluoroethy1ene (PTFE) grafts are acceptable in the femorofemoral position, A VG largely were abandoned to conserve the saphenous system for other purposes (e.g., coronary or tibial bypass) to which it is uniquely suited.

Unlike the elective setting, however, AVG fre­quendy are indispensible for revascularization within a contaminated surgical field. Mannick and Maini3

and Ehrenfeld et al.4 have demonstrated the versatil­ity of femorofemoral bypass with GSV in a small number of patients having either groin sepsis or an established infection involving one limb of a syn­thetic aortobifemoral graft. Despite the fact that these and other similar indications for femorofemoral A VG have been relatively unusual in the past, they seem likely to be encountered more commonly in the future because of the potential risk for simultaneous arterial injury and groin contamination associated

From the Department of Vascular Surgery, The Cleveland Clinic Foundation, Cleveland.

Reprint requests: Norman R. Hertzer, MD, Department of Vascular Surgery, The Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, OH 44195.

*Current address: Department of Surgery, University Hospital, 88 E. Newton St., Boston, MA 02118.

Copyright © 1994 by The Society for Vascular Surgery and International Society for Cardiovascular Surgery, North Ameri­can Chapter.

0741-5214/94/$3.00 + 0 24/1/51994

912

JOURNAL OF VASCULAR SURGERY Volwne 19, Nwnber 5

with femoral catheterization in candidates for percu­taneous coronary angioplasty or peripheral endovas­cular therapy. Because the durability of femorofemo­ral A VG remains uncertain, the purpose of this report is to review their outcome at our center.

MATERIAL AND METHODS

Patient information. From 1979 through 1992, a total of 25 femorofemoral A VG were constructed in 13 men and 12 women having a mean age of 64 years (range 43 to 84 years). In all of these patients, A VG were used in preference to synthetic grafts because of either a perceived risk for postoperative wound infections (group 1, n = 16) or the presence of established infections already involving previous synthetic grafts in one or both groins (group 2, n = 9).

Group 1. Six (37%) of the 16 patients in this group had recently undergone a prolonged period of femoral artery catheterization for thrombolytic therapy or transluminal balloon angioplasty lasting a minimum of several hours. Another six patients required urgent femorofemoral revascularization soon after incidental iliac arterial occlusion or ligation in conjunction with transabdominal surgical proce­dures, all of which involved bowel resections or enterotomies. The remaining four patients (25%) were undergoing treatment for malignant neoplasms, systemic lupus erythematosus, or unrelated illnesses requiring corticosteroid management. In each case, potential contamination of at least one of the groin incisions or the subcutaneous suprapubic tunnel connecting them seemed sufficiently likely to warrant autogenous femorofemoral reconstruction.

Group 2. The nine patients in this group were admitted with clinical evidence of infections involv­ing previous synthetic femorofemoral (n = 4) or axillofemoral (n = 2) grafts, or one limb of previous aortobifemoral grafts (n = 3). Seven of these grafts were Dacron, and two were PTFE. Despite convinc­ing local indications of graft infections in all nine patients, the responsible organisms (Staphylococcus species) could be documented in only four.

Surgical management. Reversed GSV was used for femorofemoral revascularization in a total of 23 patients, but it was necessary to construct cephalic vein grafts in the remaining two. The common femoral artery or a segment of the superficial or deep femoral artery immediately adjacent to it was used as the inflow source for all procedures, and the outflow anastomosis was performed with either the contralat­eral common femoral artery or a contiguous segment of deep femoral artery. Each patient received broad

Hakaim et at. 913

spectrum antibiotic treatment during the periopera­tive period, which was modified appropriately when­ever specific organisms were identified after surgical debridement.

Statistical analysis. The follow-up interval for this series extends to 78 months (mean 22 months, median 13 months). Data concerning patient survival were collected during a telephone canvass, but conclusions regarding graft patency were strictly limited to the objective results of the physical examination and noninvasive vascular testing per­formed at the time of the most recent outpatient visit.

Paired comparisons were evaluated with Stu­dent's t test. Standard cumulative methods were used to estimate survival and primary patency rates, and standard errors were calculated according to the Greenwood formula. The statistical significance of apparent differences between group 1 and group 2 were further explored with the log rank test and Fisher's (two-tailed) exact test.

RESULTS

There was one postoperative death in each of the two study groups, for an overall early mortality rate of8%. In group 1, a 74-year'-0Id man who had been admitted with an acute abdominal catastrophe un­derwent urgent celiotomy after preoperative angio­grams had demonstrated occlusion of the infrarenal segment of the aorta. The sigmoid colon was found to be necrotic and was resected in conjunction with a Hartmann procedure. Transfemoral thrombectomy reestablished adequate inflow to the right lower extremity, and femorofemoral revascularization of the contralateral limb was performed with GSV. This patient died within 24 hours because of systemic sepsis and respiratory failure. In group 2, a 54-year­old woman required preliminary construction of an axillofemoral PTFE graft, together with autogenous femorofemoral bypass (GSV) as a staged procedure before excision of an infected aortic bifurcation graft. She died with cardiorespiratory failure 10 days after her second operation.

Postoperative graft patency was determined by clinical assessment and anklefbrachial indexes (ABI) in the 23 operative survivors. Preoperative ABI were unavailable for three patients in group 1; two of these patients each required unanticipated femorofemoral AVG immediately after one common iliac artery had been resected in continuity with pelvic neoplasms, and the third patient already had undergone an above-knee amputation distal to the outflow anasto­mosis ofhis femorofemoral bypass. Preoperative ABI for the remaining 12 patients in group 1 ranged from

914 Hakaim et ai. JOURNAL OF VASCULAR SURGERY

May 1994

Table I. Summary of preoperative and postoperative ABI in two patients whose ABI declined in the recipient limb after autogenous femorofemoral bypass

Donor limb Recipient limb

Preoperative ABI PostoperatWe ABI Preoperative ABI Postoperative ABI

Group 1 Group 2

0.23* 1.00

0.75* 1.00

0.75 1.00

0.58 0.56

*Procedure performed in conjunction with synthetic axillofemoral bypass on donor side.

Table II. Cumulative survival and patency rates for the 23 operative survivors in this series

Group 1

No. %

Survival rate 1 year 15 100 2 years 8 100 3 years 4 43

Primary patency rate 1 year 15 92 2 years 8 76 3 years 3 76

0.16 to 0.67, and there was significant overall improvement in the postoperative ABI in this subset (mean 0.33 ± 0.13; P = 0.0001). With the excep­tion of a single patient who required emergency surgical treatment for an infected femoral false aneurysm, preoperative ABI were obtained for all members of group 2 (range 0.29 to 1.0). Because the infected grafts in most of these patients still were patent before their excision, the femorofemoral A VG used to replace them were not associated with significant improvement in postoperative ABI (mean 0.04 ± 0.17; P = NS). One patient in each group sustained a reduction in postoperative ABI for the recipient limb despite a patent AVG (Table I). This finding suggests that the diameter of the GSV may sometimes be marginal for femorofemoral bypass, yet neither of these patients has required amputation.

Five (22%) of the 23 operative survivors died during the late follow-up period, including two patients in group 1 (13%) and three in group 2 (37%). None of these deaths were related to com­plications involving femorofemoral A VG, and no new or recurrent groin infections occurred in either group of patients. Two AVG in group 1 (13%) and two others in group 2 (25%) eventually occluded, for an overall crude failure rate of 17%. Two of these four grafts were salvaged by thrombectomy and appro­priate revision of inflow arteries. No late amputations were necessary.

Group 2 Total

No. % No. %

8 100 23 100 5 78 12 90 3 78 7 63

8 71 23 84 4 71 12 75 3 71 6 75

Conceding their statistical unreliability, cumula­tive data concerning survival and graft patency are presented in Table II. Excluding both operative deaths, the 2- and 3-year survival estimates were 90% and 63%, respectively. The overall primary graft pa­tency estimate at each of these intervals was 75%. No significant differences were identified between group 1 and group 2 with respect to any of these compari­sons, but relatively few patients still were eligible for consideration during the third postoperative year. The limited follow-up period in group 1 probably is related to the fact that many of these patients either had undergone contemporary endovascular therapy in the recent past or were undergoing concurrent treatment for malignant disease at the time their femorofemoral AVG were constructed. Nevertheless, it may be encouraging to note that only one graft has failed subsequent to the first year of observation.

DISCUSSION

Reports by McCaughan and Kahn5 and Vetto6

were among the first to describe successful synthetic femorofemoral bypass, and in a later review of the hemodynamic effects of femorofemoral grafts, Pers­son et al.7 eventually suggested that the diameter of the GSV often may be suboptimal ( < 6 mm) for this particular approach to lower extremity revasculariza­tion. In fact, several subsequent series comprising a total of 369 femorofemoral grafts included only 9

JOURNAL OF VASCULAR SURGERY Volume 19, Number 5

(2.4%) constructed with GSV.8-11 We currently prefer PTFE for elective femorofemoral bypass, and approximately ,30 such grafts have been constructed annually by the members of our department since we began to record these data prospectively in 1989. Although our results in patients having potential groin contamination or frank: infections do not imply that femorofemoral A VG should supplant the use of synthetic grafts under other circumstances, the 3-year cumulative patency estimate of 75% in our experi­ence probably is consistent with the figure of 57% recently reported for fabric and PTFE grafts by Peder et al. 12 Only two of our patients required femoro­femoral bypass with cephalic vein grafts, one of which failed in the early postoperative period. No conclu­sions are possible on the basis of such limited data, but we previously have found that the 3-year patency rate « 50%) for femoropopliteal or femorotibial bypass with cephalic vein grafts did not meet the standard usually expected for GSV.13

According to several reports, synthetic graft infections occur in approximately 1 % to 6% of patients, most of which involve grafts w~th at least one anastomosis in the groin. 14-16 Although conven­tional strategy in such a situation often is to replace the infected material with a new synthetic graft routed through a clean surgical field, isolated reports of relatively few patients have indicated that in situ revascularization with A VG represents an inflow alternative to axillo-superficial femoral or axil­lopopliteal bypass in selected patients with potential groin contamination or frank: sepsis.4 ,17 Because of the absence of recurrent infections and the satisfac­tory intermediate-term patency rate for femorofemo­ral A VG in our series, we believe this approach is cautiously justified under the indications we have adopted. Assuming continued growth in the field of trans femoral endovascular therapy, these indications may become far more common in the future than they ever have been in the past.

Ming Tan, PhD, performed the statistical analysis for this study, and Joseph Bonafede, MD, participated in data retrieval.

Hakaim et al. 915

REFERENCES

1. Trimble IR, Stonesifer GL, Wilgis EFS, Montague Ae. Criteria for femoro-femoral bypass from clinical and hemo­dynamic studies. Ann Surg 1972;175:985-98.

2. DeLaurentis DA, Sala LE, Russell E, McCombs PRo A twelve year experience with axillofemoral and femorofe­moral bypass operations. Surg Gynecol Obstet 1978;147: 881-8.

3. Mannick JA, Maini BS. Femorofemoral grafting: indications and late results. Am J Surg 1978;136:190-4.

4. Ehrenfeld WK., Wilbur BG, Olcott CN, Stoney RJ. Autog­enous tissue reconsttuction in the management of infected prosthetic grafts. Surgery 1979;85:82-92.

5. McCaughan JJ, Kahn SF. Cross-over graft for unilateral occlusive disease of the iliofemoral atteries. Ann Surg 1960; 26:26-8.

6. Vetto RM. The treattnent of unilateral iliac artery obsttuction with a transabdominal, subcutaneous, femorofemoral graft. Surg Gynecol Obstet 1961;52:342-5.

7. Persson AV, Dyer VE, West LS. Femoral-to-femoral bypass graft. Surg Clin North Am 1980;60:537-44.

8. Mosley JG, Marston A. Long term results of 66 femoral-to­femoral by-pass grafts: a 9-year follow-up. Br J Surg 1983; 70:631-4.

9. Kalman PG, Hosang M, Johnson KW, Walker PM. The current role for femoro-femoral bypass. J VASC SURG 1987; 6:71-6.

10. Farber MA, Hollier LH, Eubanks R, Ochsner JL, Bowen Je. Femorofemoral bypass: a profile of graft failure. South Med J 1990;83:1437-43.

11. Fahal AH, McDonald AM, Marston A. Femorofemoral bypass in unilateral iliac artery occlusion. Br J Surg 1989;76: 22-5.

12. Perler BA, Burdick JF, Williams GM. Femoro-femoral or ilio-femoral bypass for unilateral inflow reconsttuction. Am J Surg 1991;161:426-30.

13. Sesto ME, Sullivan TM, Hertzer NR, Krajewski LP, O'Hara PI, Beven EG. Cephalic vein grafts for lower extremity revascularization. J VASe SURG 1992;15:543-9.

14. Fry WI, Lindenauer SM. Infection complicating the use of plastic atterial implants. Arch Surg 1967;94:600-5.

15. Goldstone I, Moore WS. Infection in vascular prosthesis. Am J Surg 1974;128:225-32.

16. Grafe WR, Watson RC, Dineen P. Infections of vascular anastomoses. Vasc Surg 1971;5:204-10.

17. Lorentzen IE, Nielsen OM. Aortobifemoral bypass with autogenous saphenous vein in treattnent of paninfected aortic bifurcation graft. J VASC SURG 1986;3:666-8.

Submitted July 12, 1993; accepted Oct. 1, 1993.