Polytetrafluoroethylene interposition grafts for carotid reconstruction Michael J. Sise, M D , Michael E. Ivy, M D , R o s e Malanche, CVT, and Kim R. Ranbarger, M D , San Diego, Calif. Polytetrafluoroethylene interposition grafts were used for carotid reconstruction in 26 operations performed in 23 patients during a 7-year period. There were 10 men and 13 women with a mean age of 66 years. The indication for interposition graft placement was recurrent stenosis with inability to perform endarterectomy in nine operations, a severe kink or technical problem during endarterectomy in nine, aneurysmal changes of the artery in six, and thrombosis of the endarterectomy site segment with stroke in two. There were no perioperative deaths and no further perioperative neurologic complications. Follow-up with performance of yearly duplex scans was obtained in all patients, and the mean duration of follow-up was 36 months. One patient (4%) died of end-stage pulmonary disease during the study interval. Occlusion of the polytetrafluorocthylene graft occurred at 12 months in one patient (4%). Hemodynamically significant restenosis developed in two patients (8%), and two patients (8%) had mild restenosis. Evidence of a second restenosis developed in three of the nine patients who underwent polytetrafluoroethylene interposition graft placement for carotid restenosis. On the basis of our study results, we conclude that polytetrafluoroethylene can be used effectively for carotid reconstruction when an interposition graft is required. However, recurrent stenosis occurs. Patients who undergo polytetrafluoroethylene interposition graft placement for carotid restenosis appear to be at high risk for a second restenosis. (J VAs¢ SURG 1992;16:601-8.)
Carotid endarterectomy remains a technically demanding procedure with little room for error. The endarterectomy site is at high risk for the development of platelet aggregation and thromboembolic sequelae. Thrombotic occlusion and platelet emboli from this source are major causes o f perioperative stroke, xa Arterial closure after endarterectomy requires careful attention to the preservation of normal arterial contour to avoid turbulent flow. Endarterectomy for recurrent stenosis, carotid aneurysm, inadvertent deep dissection, and coils or kinks are the most common causes o f difficulty in maintaining adequate arterial contour. We have used polytetrafluoroethylene (PTFE) interposition grafts for caFrom the Department of Surgery (VascularSurgeryDivision) and Clinical Investigation Department, Naval Hospital, San Diego, Calif. Work performed and supported under the United States Navy Clinical Investigation Program, study number 84-16-2021. The opinions or assertions expressed herein are those of the authors and are not to be construed as officialor as reflectingthe views of the Navy Department or the Naval Services at large. Presented at the SeventhAnnual Meeting of the Western Vascular Society, Mani, Hawaii, Jan. 11-15, 1992. Reprint requests: CAPT Michael Ji Sise, MC, USN, c/o Clinical Investigation Department, Naval Hospital San Diego, San Diego, CA 92134-5000.
24/6/39750
rotid reconstruction when restoring the carotid bifurcation contour by simple arterial closure or patch angioplasty was not possible. We studied our patients with PTFE interposition grafts to determine the efficacy of this approach and the long-term clinical outcome. METHODS
Beginning in 1984 we adopted PTFE interposition graft placement during carotid endarterectomy whenever primary closure or patch angioplasty would not result in sufficient restoration of the normal contour of the carotid bifurcation after endarterectomy. We also chose this procedure for repair o f carotid aneurysm and for immediate reoperation for carotid thrombosis after endarterectomy. This approach was based on the ease with which PTFE grafts are used in this setting and the ability to reliably choose an appropriate-sized interposition graft to restore the bifurcation contour. The technique of PTFE interposition grafting was directed toward using the shortest possible graft, fashioning a smooth transition from the common carotid artery through the bifurcation, and establishing a distal anastomosis to normal internal carotid artery. The portion of internal carotid artery that was diseased or 601
602 Sise et al.
PTFE INTERPOSITION
Fig. 1. Configuration of PTFE interposition graft used in reconstruction of carotid bifurcation. previously endarterectomized was resected and the common carotid artery trimmed to fashion a beveled site for proximal anastomosis while the origin of the external carotid artery was preserved (Fig. 1). However, in two operations, the extent of disease precluded the preservation of the distal common carotid artery and an interposition with end-to-side reimplantation of the external carotid artery was used. Six millimeter thin-walled PTFE grafts were used in all but two operations, in which tapered grafts with a distal diameter of 4.5 mm were used to obtain a proper size match with the internal carotid artery. We routinely used intraluminal shunts. The PTFE interposition graft was fashioned to the appropriate size and placed over the shunt. The anastomosis of the internal carotid artery to the PTFE graft was performed first, followed by the proximal anastomosis. All operations were performed by attending vascular surgeons or by surgical trainees with an attending vascular surgeon as teaching assistant. Intraoperative
Journal of VASCULAR SURGERY
completion angiograms were routinely performed (Fig. 2). Careful postoperative neurologic examinations were performed and recorded. Duplex scans (Advanced Technology Laboratories, Bothell, Wash.) were obtained within the first month after operation. Subsequent follow-up included repeat duplex scanning and Vascular Outpatient Clinic visits at 3 and 6 months and annually thereafter. Arch and carotid digital arteriography was performed whenever hemodynamically significant recurrent stenosis or contralateral stenosis was detected on duplex scan. Antiplatelet therapy with aspirin or aspirin in combination with dipyridamole was routinely administered perioperatively and was continued postoperatively. Data collection was performed through the mechanism of the Naval Hospital Vascular Registry. Data abstracted included history and physical examination, indication for operation, indication for PTFE interposition graft placement, perioperative course, and the results of subsequent follow-up evaluations. Preoperative and intraoperative completion angiograms were reviewed as were subsequent studies for recurrent ipsilateral or contralateral stenoses. Data were analyzed by the Kaplan-Meier method and chi-square analysis. 4 RESULTS During the 7-year study interval, 401 carotid operations for carotid stenosis or aneurysm were performed at Naval Hospital San Diego. PTFE interposition grafts were used in 26 (6%) operations performed in 23 patients. There were 10 men and 13 women with a mean age of 66 years. Indications for endarterectomy included asymptomatic stenosis in 12 patients (46%), transient ischemic attack in four (15%), amaurosis fugax in eight (31%), stroke with good recovery in one (4%), and carotid bifurcation aneurysm in one (4%). Risk factors included hypertension in 83% of the patients, tobacco use in 96%, hyperlipidemia in 39%, and insulin-dependent diabetes in 4%. The indications for PTFE graft placement are summarized in Table I. In the nine patients with recurrent carotid stenosis, the PTFE grafts were used when endarterectomy could not be performed or when restenosis occurred in the presence of a prior patch angioplasty. In four operations (16%), a focal aneurysm with a blisterlike appearance was encountered and endarterectomy resulted in a thinned-out eccentric arterial contour that did not appear reparable with patch angioplasty. A kink, either above the endarterectomy site or in the
Volume 16 Number 4 October 1992
PTFE interpositionfor carotid endarterectoray 603
Table L Indication for PTFE interposition graft placement n (%)
Recurrent carotid artery, stenosis, endarterectomy not feasible True aneurysm of the carotid bifurcation Focal aneurysm of the carotid artery Significant kink of the internal carotid artery Technical error during endarterectomy Thrombosed endarterectomized segment with stroke
9 (36) 2 4 5 4 2
Total
26
(8) (16) (20) (12) (8)
region of arterial closure, led to PTFE interposition graft placement when it appeared to be hemodynamically significant on the completion angiogram. Technical errors leading to PTFE interposition grafting included endarterectomy into an inappropriately deep plane in two patients and a significant arterial tear in the internal carotid artery in one patient. PTFE interposition grafting was used in two patients in whom thrombosis of the endarterectomized segment developed. In one patient, thrombosis was found on the completion angiogram. Rethrombosis was noted on a second angiogram in this patient after a Dacron patch (Meadox Medicals, Inc., Oakland, N.J.) had been placed. Placement of a PTFE interposition graft was performed and the artery remained patent without recurrent thrombus. This patient awoke with a mild motor weakness and a patent artery on noninvasive evaluation. The second patient awoke without a neurologic deficit after carotid endarterectomy and patch angioplasty but hemiparesis developed within 2 hours. On immediate return to the operating room the endarterectomy site was found to contain extensive platelet thrombus. A PTFE interposition graft was placed and the patient awoke with the same neurologic findings. One patient who underwent PTFE interposition grafting for recurrent carotid stenosis awoke with a permanent ipsilateral true vocal cord paralysis, and two patients required treatment of reperfusion headaches in the early postoperative period. Follow-up was obtained in all 23 patients and ranged from 13 to 87 months with a mean of 36 months. One patient (4%) died of end-stage pulmonary disease 13 months after operation. Asymptomatic occlusion of the PTFE graft occurred at 12 months in one patient (4%). In two patients (8%) hemodynamically significant restenosis developed, and two patients (8%) had mild restenosis at a mean of 40 months after operation (Table II). Symptoms occurred in only one patient (patient 1) with a
Fig. 2. Intraoperative completion angiogram after PTFE graft placement for carotid bifurcation reconstruction in patient with hemodynamicaUy significant carotid restenosis. contralateral carotid artery occlusion and were orthostatic in nature. This patient underwent graft replacement with a second PTFE interposition graft. The second graft remained patent until death as a result of progressive pulmonary disease 13 months later. There were no other neurologic events in the surviving 22 patients during the follow-up period. The restenosis and occlusion-free survival rates of the study group calculated by the Kaplan-Meier method are detailed in Table III. The 1-year and 3-year marks appeared to represent significant intervals in the development of restenosis in this group. No single risk factor was remarkable in this small group of patients with restenosis. Only one patient continued to smoke and two were hyperlipidemic. All of the patients in this group were compliant with antiplatelet therapy. The rate of restenosis after placement of a PTFE interposition graft for recurrent carotid stenosis compared with that in those who had not had a prior carotid operation was higher but not statistically significant (33% vs 12%).
604
Journal of VASCULAR SURGERY
Sise et al.
Table H. Patients with recurrent carotid stenosis after PTFE interposition graft placement Patient
Age (yr)/Sex
Indication for PTFE graft
History
Onset restenosis,symptoms
Outcome
HemodynamicaUy significant restenosis or occlusion 1 71/Male TIA, restenosis, contralat- Restenosis, dense fibrosis 39 months, dizziness eral occlusion
2 : ~"3
58/Male
&symptomatic restenosis, hyperlipidemia 67/Female Amaurosis fugax, continued smoking, diabetes
Mild to moderate stenosis 4 66/Female Asymptomatic restenosis 5
63/Female Amaurosis fugax, hyperlipidemia
At reoperation for fibrous hyperplasia, PTFE replaced, died 13 months later of pulmonary disease Restenosis, dense fibrosis 12 months, asymptomatic Occluded, asymptomatic
Intraoperative thrombosis
87 months, asymptomatic Antiplatelet therapy, remains asymptomatic
Restenosis, dense fibrosis 39 months, asymptomatic 60% stenosis, remains stable, asymptomatic Redundant artery with 13 months, asymptomatic 40% stenosis, remains kink stable, asymptomatic
T/A, Transient ischemic attacks.
Table HI. Recurrent stenosis after PTFE interposition graft placement (Kaplan-Meier method)
Interval (too)
No. of patiems at risk at start~endinterval
No. of recurrences in interval
Hemodynamically significant stenosis or 0-12 26/26 13-39 25/14 All stenoses or occlusions 0-12 26/26 13-15 25/22 16-39 21/14
Interval recurrence (%)
Recurrence-free survival at end of interval (%)
Cummulative recurrence-freemrvival at end of interval (%)
3.8 7.1
96.2 92.9
96.2 89.3
3.8 4.5 14.3
96.2 95.5 85.7
96.2 91.9 78.8
occlusion 1 1 1 1 2
Digital or magnetic resonance arteriography was obtained in the three patients with significant restenosis. Occlusion was found in one patient and diffuse narrowing was found in the other two patients (Fig. 3). An arteriogram, which confirmed the duplex scan findings, was also obtained in one patient with mild stenosis. Two patients with normal duplex scans of the area of the PTFE graft underwent digital arteriography for suspected contralateral stenosis and were found to have patent grafts with a normal bifurcation contour (Fig. 4). DISCUSSION The efficacy of carotid endarterectomy in stroke prevention is largely dependent on the minimization of Perioperative complications. Thrombotic occlusion and emboli from the endarterectomy site remain major causes of neurologic complications and must be prevented to sustain good results from this operation. We adopted the use of PTFE interposition
graft placement when primary closure or patch angioplasty did not adequately restore the carotid bifurcation contour in an effort to minimize the chances of turbulent flow and thrombus formation. Our selection of PTFE graft material for carotid reconstruction in this setting was based not only on the ease with which it is used but also on a review of the other options available to restore carotid bifurcation contour. A variety of techniques from resection and primary anastomosis to complicated interpositions have been described, s-° A review of the available literature failed to reveal significant early or late advantages to any one of these techniques. The simplest approach appeared to be resection of the diseased internal carotid artery and PTFE interposition graft placement. Although we found this approach to be effective in the perioperative period, we undertook this study to determine whether or not the long-term results were acceptable. The initial goal of safely reconstructing the
Volume 16 Number 4 October 1992
PTFE interpositionfor carotid endarterectomy 605
Fig. 3. Diffuse internal carotid artery stenosis 87 months after PTFE interposition graft placement. Duplex scan had remained normal throughout follow-up before this time.
Fig. 4. Follow-up arteriogram 18 months after PTFE interposition grafting. Note carotid bifurcation with preservation of contour.
carotid bifurcation was met with the use of PTFE interposition grafts. Intraoperative completion angiography documented the restoration of the normal carotid bifurcation contour with this technique. There were no perioperative complications specifically related to the use of the PTFE graft, and the initial postoperative duplex scan documented a patent carotid bifurcation in all patients. Long-term follow-up, however, revealed that this reconstruction was not free from the risk of restenosis. Although follow-up revealed an overall restenosis rate of 20%, hemodynamically significant recurrent stenosis or occlusion occurred in 12% of the patients. There were no thromboembolic complications encountered in any of the patients. The only neurologic symptoms appeared to be hemodynamic in nature and occurred in a patient with bilateral occlusive disease. The overall survival rate was 96% at a mean of 3 years. The expected rate of restenosis after carotid endarterectomy is difficult to quantify. Reported
rates vary from 1% to 33% and may be related to the duration of follow-up and the persistence of the investigators in obtaining imaging studies, x°qa Our findings of a 12% significant stenosis or occlusion rate and a survival and stenosis-free rate of 89% at 39 months appear to be in keeping with results of series based on careful duplex scanning over similar time intervals. 1416 Further difficulty is encountered with quantification of the recurrent stenosis rate after carotid reconstruction for restenosis. 17-19We found a 33% recurrence rate in this group, and the only neurologic symptoms in our series occurred in one of the patients in this group. Graft stenosis and occlusion with PTFE grafts have been extensively investigated. Our findings with PTFE grafts in the reconstruction of the carotid bifurcation appear to be somewhat more favorable than the results found in reconstructions in other areas. 2°,21 The reasons for this finding cannot be investigated in a small series such as ours. However, the short nature of these grafts and the high flow rates
606
Journal of VASCULAR SURGERY
Sise et al.
at the carotid bifurcation m a y be the key factors that allow patency rates with P T F E grafts comparable to those in other m e t h o d s o f carotid reconstruction. P T F E grafts appear to be effective for carotid reconstruction w h e n an interposition graft is required. T h e carotid bifurcation c o n t o u r is easily restored and the risk o f perioperative t h r o m b o e m bolic complications is apparently low. A l t h o u g h early patency is g o o d , recurrent stenosis occurs. The rate o f restenosis does n o t appear significantly different than rates observed with other forms o f carotid reconstruction. Patients w h o u n d e r g o P T F E interposition graft placement for recurrent carotid stenosis appear to be at high risk for a second restenosis.
10,
11, 12. 13. 14. 15.
REFERENCES
1. ~ CD, Hertzer NR. Postoperative stroke and late neurologic complications after carotid endarterectomy. Arch Surg 1981;116:1561-8. 2. Steed DL, Peitzman AB, Grundy BL. Causes of stroke in endarterectomy. Surgery 1982;92:634-41. 3. Dooner J, Kuechler P. Salvageafter postoperative thrombosis of the carotid artery. Am J Surg 1990;159:525-6. 4. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Star Assoc 1958;53:457-81. 5. Comfier JM, Cormier F, Laurian C, et al. Polytetrafluoroethylene bypassfor revascularizationof the atherosderotic carotid artery: late results. Ann Vasc Surg 1987;1:564-71. 6. Clagetr GP, Patterson CB, Fisher DF, et al. Vein patch versus primary closure for carotid endarterectomy. J VASC SURG 1989;9:213-23. 7. Rosenthal D, Archie JP, Garcia-Rinaldi R, et al. Carotid patch angioplasty: immediate and long-term results. J VASCSURG 1990;12:326-33. 8. Vanmaele R, Van Schil P, De Maesneer M. Closure of the internal carotid after endarterectomy: the advantages of patch angioplasty without its disadvantages. Ann Vasc Surg 1990; 4:81-4. 9. Collins PS, Orrecchia P, Gomez E. A technique for correction
16. 17. 18. 19. 20.
21.
of carotid kinks and coils following endarterectomy. Ann Vasc Surg 1991;5:116-20. Reilly LM, Okuhn SP, Rapp JH, et al. Recurrent carotid stenosis: a consequence of local or systemic factors? The influence of unrepaired defects. J VASCSURG 1990;11:44860. Green RM, McNamara JA, Ouriel K, et al. The clinical course of residual carotid arterial disease. J VASC SURG 1991;13: 112-20. Callow AD, Mackey WC. Long-term follow-up of surgically managed carotid bifurcation stenosis: justification for an aggressive approach. Ann Surg 1989;210:308-16. Cook JM, Thompson BW, Barnes RW. Is routine duplex examination after carotid endarterectomy justified? J VASC SURG 1990;12:334-40. DeGroote RD, Lynch TG, Jamil Z, et al. Carotid restenosis: long-term noninvasive follow-up after carotid endarterectomy. Stroke 1987;18:1031-6. Sawchuk AP, Flanigan DP, Machi J, et al. The fate of unrepaired minor technical defects detected by intraoperative ultrasonography during carotid endarterectomy. J VASCSURG 1989;9:671-6. Healy DA, Zierler RE, Nicholls SC, et al. Long-term follow-up and clinical outcome of carotid restenosis. J VASC SURG 1989;10:662-9. Kazmers A, Zierler RE, Huang TW, et al. Reoperative carotid surgery. Am J Surg 1988;156:346-52. Edwards WH, Edwards WH, Mulherin JL, et al. Recurrent carotid artery stenosis. Ann Surg 1989;209:662-8. Nitzberg RS, Mackey WC, Prendville E, et al. Long-term follow-up of patients operated on for recurrent carotid stenosis. J VASCSUP,G 1991;13:121-7. Veith FJ, Gupta SK, Ascer E, et al. Six-year prospective multi-center randomized comparison of autologous saphenous vein and expanded polytetrafluoroethylene grafts in infrainguinal arterial reconstructions. J VASCSrORG1986;3: 104-14. Rutherford RB, Jones DN, Bergentz SE, et al. Factors affecting the patency of infrainguinal bypass. J VASCSUItG 1988;8:236-46.
Submitted Feb. 13, 1992; accepted May 25, 1992.
DISCUSSION Dr. Victor M. Bernhard (Tucson, Ariz.). Dr. Sise and his colleagues have provided us with an opportunity to appraise a moderately large series of PTFE replacements of the carotid artery at and distal to the bifurcation. In addition to the report of immediate results, they have rigorously followed up their patients over the long term with duplex scans, which I think is essential. In this group of 26 carotid artery replacements, the operative mortality rate was zero and a perioperative neurologic deficit occurred as a result of the interposition graft procedure in only one patient. Late occlusion occurred in one patient. In two patients severe recurrent stenoses
developed, and in one of these cases the stenosis was symptomatic and the patient required a second PTFE graft. Two other patients had mild to moderate asymptomatic restenoses. These are excellent results, and the experience of this series and from a larger series reported by Cormier support the authors' contention that PTFE may be a reliable conduit for repair of carotid artery defects when routine endarterectomy with or without patching is neither safe nor feasible. About 4 or 5 years ago Dr. Cormier and his group reported on 62 grafts in roughly the same position. The
Volume 16 Number 4 October 1992
operative procedure was not exactly the same. The grafts were actually longer, usually coming from the common carotid, and were inserted because there was extensive common carotid disease, and some of them came from other sources such as the subclavian or the arch. Nevertheless, a PTFE graft was placed in the carotid artery position in 62 repairs in 1789 patients, roughly about half the frequency that was reported by Dr. Sise. They also had a zero mortality rate. The incidences of transient deficits, permanent deficits, early asymptomatic occlusions, and so on were very low, and the 4-year patency rate by careful observation was 96%. There are several questions I would like to address to the authors. When discussing technique, the authors note that the use of PTFE replacement grafts permits the surgeon to pick the size graft that will most effectively fit the distal vessel, and in two o f their patients they used a 4.5 mm taper. Were either of these grafts associated with subsequent stenoses, and how long were these patients followed up so that we will really know what the narrower graft may do? The need for graft interposition seems to be relatively high in the authors' overall experience with 401 carotid artery repairs, especially for the correction ofrestenosis and for kinks and coils. In these two subcategories, what was the percentage of patients in whom the more conventional techniques such as patch angioplasty or arterial shortening were abandoned in favor of PTFE interposition grafting? In most reports and in our own experience with recurrent stenosis, the need for carotid replacement rather than endarterectomy or patch angioplasty, or both of these procedures, ranges between 3% and 7%. The pathologic basis of recurrent stenosis in the one patient with resection of the interposition graft was not described, and we should know what was found. In the other patient with a severe stenosis in whom an arteriogram was presented, did the duplex scan suggest layering thrombus, which might explain the appearance of diffuse narrowing in what appeared to be the entire length of the PTFE graft that was inserted? Restenosis occurred in five of the interposition grafts or 19% in this series o f which three appeared in the subset of nine carotid repairs for restenosis of a prior endarterectomy. There is a paucity of data as the authors noted regarding the likelihood of re-restenosis under those circumstances. Nevertheless, this may have been a target group, although the numbers in the authors' series were too small for statistical significance. Will they continue to use or recommend PTFE interposition grafts as the preferred alternative in this subgroup rather than be more persistent in their efforts to accomplish patch angioplasty or to use a vein graft when an interposition graft is required? Despite the fine results reported, this is a retrospective study without comparison with other techniques. Admittedly, the acquisition of sufficient data to make these comparisons may be difficult or impossible. However, a broader experience and a longer follow-up in the manner prescribed by the authors will be required before we
PTFE interpositionfor carotid endartcrectoray 607
abandon conventional methods in favor of PTFE interposition grafting. In the meantime, this report supports the use o f PTFE as a possibly safe and reliable replacement for the carotid bulb and proximal internal carotid artery when no other options are available. Dr. Michael I. Sise. There were no problems with the two 4.5 mm interposition grafts. They have been followed up for 2 and 3 years now, respectively. The patients in whom restenosis developed all received 6 mm thin-walled PTFE grafts. The patient in whom occlusion occurred had undergone an operation for restenosis once with a patch angioplasty and then a second operation for restenosis in which we proceeded with an interposition graft. In those patients who underwent operation for restenosis during the 7-year period, roughly half of the arteries were closed after repeat endarterectomy with patch angioplasty. We routinely patch in all carotid endarterectomies with PTFE. We stopped resecting for kinks and coils in favor of placing the interposition graft. We found that resecting, sewing the distal resection line of internal carotid artery to the proximal internal carotid artery or to the bifurcation, and then placing a patch was more timeconsuming, more difficult to align, and technically less satisfactory than the PTFE interposition. If we were going to go so far as to resect and patch we would prefer to do an interposition graft. The patient in whom the recurrent stenosis developed and who underwent a second PTFE interposition graft procedure was the only patient for whom we have data on the pathologic basis of the restenosis. We found fibrous hyperplasia, typical of recurrent carotid stenosis. These grafts are very short, and there is not an area of graft of any significant length between the proximal anastomosis and • the distal anastomosis. We could not do a graded analysis of the pathologic changes along the graft. In the patient who had no symptoms and who had the diffuse stenosis we have done a number of imaging studies and we have not seen layered thrombus. It appears to be noncalcified plaque. Will we continue to use PTFE interposition grafts? I want to reiterate that these were arterial closures that were difficult. We do not set out to use PTFE grafts. We set out to do an endarterectomy and do a patch angioplasty if we think it is feasible and if we think we will get a good restoration of contour. We believe there is a window of vulnerability when we perform this operation from the time we clamp the carotid artery until the time we unclamp and through the first couple of hours. That is the time when we can make the difference with our perioperative stroke rate. We believe that PTFE grafting, in certain patients in whom arterial closure is difficult, offers an advantage in restoring contour in a fashion that prevents thromboembolic complications. So in those settings we will continue to use PTFE interposition grafts, but we will also continue to aggres-
608
Sise et al.
sively follow up our patients. I want to reiterate that this procedure is a good way to get out of trouble. This is not a primarily preferred procedure. Dr. Ronald J. Stoney (San Francisco, Calif.). My experience with prosthetic carotid artery replacement has been in secondary restenoses in which I originally used knitted Dacron conduits and have more recently used PTFE grafts. Fortunately there have not been any subsequent recurrences with the PTFE and only one with the knitted Dacron grafts. However, there are probably only half a dozen in each group. I think your idea that this technique should be considered in the patient with myointimal hyperplasia causing restenosis is a good one. Such lesions may involve a very difficuk technical challenge in patching, or if you have problems you may attempt to cover yourself with the use of a conduit. Dr. Ehrenfeld has been interested in generous patching of recurrent carotid stenoses and argues that a larger postreconstruction lumen might be beneficial. Because you cannot prevent re-recurrence, a larger lumen ~ accommodate restenosis and still provide a residual lumen that will be acceptable. It is tempting to think that in removal of the arterial segment in which stenosis has recurred any replacement with a graft might be preferable. A few times I have put a PTFE conduit in for first-time recurrence, and they have worked very well. My question is this: in the occasional patient with a primary myointimal recurrence, will you use a tubular graft rather than a patch? Dr. Sise. The decision to use an interposition graft is strictly related to the ability to get what we would consider an acceptable contour after endarterectomy. We try to endarterectomize the recurrent carotid stenosis, as I am sure most of you do. If we cannot do it and if it looks like there is a smooth fibrous area, we consider primary patching, but those patients are few and far between. There are important issues related to recurrent carotid artery stenosis. Dr. Strandness and his group presented an important paper 4 years ago in Kauai, Hawaii, to this society in which they documented a significant carotid restenosis rate after carotid endarterectomy. However, these lesions were usually asymptomatic. There are little data from long-term follow-up with
Journal of VASCULAR SURGERY
imaging studies concerning the results after a second operation for carotid artery stenosis. It is important for those of us with registries with large groups of patients that we take a look at our patients to closely follow up those who have undergone operation for carotid artery restenosis. We believe this group is at an extremely high risk for recurrent disease. Dr. Eugene F. Bernstein (La Jolla, Calif.). It is my sense that in Europe vascular surgeons do vein interposition reconstructions for the carotid artery more frequently than we do. A number of friends I have there say something like, "Oh, if we have a little problem where the lesion is too long or where there is a kink or there is something else, we just put in a piece of saphenous vein, and we probably do it one in five operations." I wonder if you would comment about venous interposition versus PTFE grafting when an interposition is required and also distinguish between the situation in which you have planned to perform interposition grafting versus the situation in which you have not planned to do it but are forced to do it for technical reasons during the operation. Dr. Sise. A lot of agonizing went into what to do with these patients. In the early 1980s we had three patients with extremely difficult recurrent stenoses in whom we placed vein interpositions. All had early complete occlusions. We went to the literature and found a paucity of data on what happens to saphenous vein interpositions. In the few studies with follow-up, there was a significant rate of recurrent stenosis or total occlusion. We did not wish to add a saphenous vein harvest, which may or may not result in an acceptable size match to the carotid bifurcation, when we knew that there was not good documentation that vein was superior to PTFE. The PTFE graft is easy to use in this position, and we therefore chose to use it when an interposition was required. I do not believe that anyone in this society does not gain more and more respect for carotid artery reconstruction independent of any external critical analysis such as that which our nonsurgical colleagues are placing on us. Certainly the more operations we do, the more we develop the "play-off game" mentality that every carotid artery operation must be successful.
Carotid endarterectomy remains a technically demanding procedure with little room for error. The endarterectomy site is at high risk for the development of platelet aggregation and thromboembolic sequelae. Thrombotic occlusion and platelet emboli from this source are major causes o f perioperative stroke, xa Arterial closure after endarterectomy requires careful attention to the preservation of normal arterial contour to avoid turbulent flow. Endarterectomy for recurrent stenosis, carotid aneurysm, inadvertent deep dissection, and coils or kinks are the most common causes o f difficulty in maintaining adequate arterial contour. We have used polytetrafluoroethylene (PTFE) interposition grafts for caFrom the Department of Surgery (VascularSurgeryDivision) and Clinical Investigation Department, Naval Hospital, San Diego, Calif. Work performed and supported under the United States Navy Clinical Investigation Program, study number 84-16-2021. The opinions or assertions expressed herein are those of the authors and are not to be construed as officialor as reflectingthe views of the Navy Department or the Naval Services at large. Presented at the SeventhAnnual Meeting of the Western Vascular Society, Mani, Hawaii, Jan. 11-15, 1992. Reprint requests: CAPT Michael Ji Sise, MC, USN, c/o Clinical Investigation Department, Naval Hospital San Diego, San Diego, CA 92134-5000.
24/6/39750
rotid reconstruction when restoring the carotid bifurcation contour by simple arterial closure or patch angioplasty was not possible. We studied our patients with PTFE interposition grafts to determine the efficacy of this approach and the long-term clinical outcome. METHODS
Beginning in 1984 we adopted PTFE interposition graft placement during carotid endarterectomy whenever primary closure or patch angioplasty would not result in sufficient restoration of the normal contour of the carotid bifurcation after endarterectomy. We also chose this procedure for repair o f carotid aneurysm and for immediate reoperation for carotid thrombosis after endarterectomy. This approach was based on the ease with which PTFE grafts are used in this setting and the ability to reliably choose an appropriate-sized interposition graft to restore the bifurcation contour. The technique of PTFE interposition grafting was directed toward using the shortest possible graft, fashioning a smooth transition from the common carotid artery through the bifurcation, and establishing a distal anastomosis to normal internal carotid artery. The portion of internal carotid artery that was diseased or 601
602 Sise et al.
PTFE INTERPOSITION
Fig. 1. Configuration of PTFE interposition graft used in reconstruction of carotid bifurcation. previously endarterectomized was resected and the common carotid artery trimmed to fashion a beveled site for proximal anastomosis while the origin of the external carotid artery was preserved (Fig. 1). However, in two operations, the extent of disease precluded the preservation of the distal common carotid artery and an interposition with end-to-side reimplantation of the external carotid artery was used. Six millimeter thin-walled PTFE grafts were used in all but two operations, in which tapered grafts with a distal diameter of 4.5 mm were used to obtain a proper size match with the internal carotid artery. We routinely used intraluminal shunts. The PTFE interposition graft was fashioned to the appropriate size and placed over the shunt. The anastomosis of the internal carotid artery to the PTFE graft was performed first, followed by the proximal anastomosis. All operations were performed by attending vascular surgeons or by surgical trainees with an attending vascular surgeon as teaching assistant. Intraoperative
Journal of VASCULAR SURGERY
completion angiograms were routinely performed (Fig. 2). Careful postoperative neurologic examinations were performed and recorded. Duplex scans (Advanced Technology Laboratories, Bothell, Wash.) were obtained within the first month after operation. Subsequent follow-up included repeat duplex scanning and Vascular Outpatient Clinic visits at 3 and 6 months and annually thereafter. Arch and carotid digital arteriography was performed whenever hemodynamically significant recurrent stenosis or contralateral stenosis was detected on duplex scan. Antiplatelet therapy with aspirin or aspirin in combination with dipyridamole was routinely administered perioperatively and was continued postoperatively. Data collection was performed through the mechanism of the Naval Hospital Vascular Registry. Data abstracted included history and physical examination, indication for operation, indication for PTFE interposition graft placement, perioperative course, and the results of subsequent follow-up evaluations. Preoperative and intraoperative completion angiograms were reviewed as were subsequent studies for recurrent ipsilateral or contralateral stenoses. Data were analyzed by the Kaplan-Meier method and chi-square analysis. 4 RESULTS During the 7-year study interval, 401 carotid operations for carotid stenosis or aneurysm were performed at Naval Hospital San Diego. PTFE interposition grafts were used in 26 (6%) operations performed in 23 patients. There were 10 men and 13 women with a mean age of 66 years. Indications for endarterectomy included asymptomatic stenosis in 12 patients (46%), transient ischemic attack in four (15%), amaurosis fugax in eight (31%), stroke with good recovery in one (4%), and carotid bifurcation aneurysm in one (4%). Risk factors included hypertension in 83% of the patients, tobacco use in 96%, hyperlipidemia in 39%, and insulin-dependent diabetes in 4%. The indications for PTFE graft placement are summarized in Table I. In the nine patients with recurrent carotid stenosis, the PTFE grafts were used when endarterectomy could not be performed or when restenosis occurred in the presence of a prior patch angioplasty. In four operations (16%), a focal aneurysm with a blisterlike appearance was encountered and endarterectomy resulted in a thinned-out eccentric arterial contour that did not appear reparable with patch angioplasty. A kink, either above the endarterectomy site or in the
Volume 16 Number 4 October 1992
PTFE interpositionfor carotid endarterectoray 603
Table L Indication for PTFE interposition graft placement n (%)
Recurrent carotid artery, stenosis, endarterectomy not feasible True aneurysm of the carotid bifurcation Focal aneurysm of the carotid artery Significant kink of the internal carotid artery Technical error during endarterectomy Thrombosed endarterectomized segment with stroke
9 (36) 2 4 5 4 2
Total
26
(8) (16) (20) (12) (8)
region of arterial closure, led to PTFE interposition graft placement when it appeared to be hemodynamically significant on the completion angiogram. Technical errors leading to PTFE interposition grafting included endarterectomy into an inappropriately deep plane in two patients and a significant arterial tear in the internal carotid artery in one patient. PTFE interposition grafting was used in two patients in whom thrombosis of the endarterectomized segment developed. In one patient, thrombosis was found on the completion angiogram. Rethrombosis was noted on a second angiogram in this patient after a Dacron patch (Meadox Medicals, Inc., Oakland, N.J.) had been placed. Placement of a PTFE interposition graft was performed and the artery remained patent without recurrent thrombus. This patient awoke with a mild motor weakness and a patent artery on noninvasive evaluation. The second patient awoke without a neurologic deficit after carotid endarterectomy and patch angioplasty but hemiparesis developed within 2 hours. On immediate return to the operating room the endarterectomy site was found to contain extensive platelet thrombus. A PTFE interposition graft was placed and the patient awoke with the same neurologic findings. One patient who underwent PTFE interposition grafting for recurrent carotid stenosis awoke with a permanent ipsilateral true vocal cord paralysis, and two patients required treatment of reperfusion headaches in the early postoperative period. Follow-up was obtained in all 23 patients and ranged from 13 to 87 months with a mean of 36 months. One patient (4%) died of end-stage pulmonary disease 13 months after operation. Asymptomatic occlusion of the PTFE graft occurred at 12 months in one patient (4%). In two patients (8%) hemodynamically significant restenosis developed, and two patients (8%) had mild restenosis at a mean of 40 months after operation (Table II). Symptoms occurred in only one patient (patient 1) with a
Fig. 2. Intraoperative completion angiogram after PTFE graft placement for carotid bifurcation reconstruction in patient with hemodynamicaUy significant carotid restenosis. contralateral carotid artery occlusion and were orthostatic in nature. This patient underwent graft replacement with a second PTFE interposition graft. The second graft remained patent until death as a result of progressive pulmonary disease 13 months later. There were no other neurologic events in the surviving 22 patients during the follow-up period. The restenosis and occlusion-free survival rates of the study group calculated by the Kaplan-Meier method are detailed in Table III. The 1-year and 3-year marks appeared to represent significant intervals in the development of restenosis in this group. No single risk factor was remarkable in this small group of patients with restenosis. Only one patient continued to smoke and two were hyperlipidemic. All of the patients in this group were compliant with antiplatelet therapy. The rate of restenosis after placement of a PTFE interposition graft for recurrent carotid stenosis compared with that in those who had not had a prior carotid operation was higher but not statistically significant (33% vs 12%).
604
Journal of VASCULAR SURGERY
Sise et al.
Table H. Patients with recurrent carotid stenosis after PTFE interposition graft placement Patient
Age (yr)/Sex
Indication for PTFE graft
History
Onset restenosis,symptoms
Outcome
HemodynamicaUy significant restenosis or occlusion 1 71/Male TIA, restenosis, contralat- Restenosis, dense fibrosis 39 months, dizziness eral occlusion
2 : ~"3
58/Male
&symptomatic restenosis, hyperlipidemia 67/Female Amaurosis fugax, continued smoking, diabetes
Mild to moderate stenosis 4 66/Female Asymptomatic restenosis 5
63/Female Amaurosis fugax, hyperlipidemia
At reoperation for fibrous hyperplasia, PTFE replaced, died 13 months later of pulmonary disease Restenosis, dense fibrosis 12 months, asymptomatic Occluded, asymptomatic
Intraoperative thrombosis
87 months, asymptomatic Antiplatelet therapy, remains asymptomatic
Restenosis, dense fibrosis 39 months, asymptomatic 60% stenosis, remains stable, asymptomatic Redundant artery with 13 months, asymptomatic 40% stenosis, remains kink stable, asymptomatic
T/A, Transient ischemic attacks.
Table HI. Recurrent stenosis after PTFE interposition graft placement (Kaplan-Meier method)
Interval (too)
No. of patiems at risk at start~endinterval
No. of recurrences in interval
Hemodynamically significant stenosis or 0-12 26/26 13-39 25/14 All stenoses or occlusions 0-12 26/26 13-15 25/22 16-39 21/14
Interval recurrence (%)
Recurrence-free survival at end of interval (%)
Cummulative recurrence-freemrvival at end of interval (%)
3.8 7.1
96.2 92.9
96.2 89.3
3.8 4.5 14.3
96.2 95.5 85.7
96.2 91.9 78.8
occlusion 1 1 1 1 2
Digital or magnetic resonance arteriography was obtained in the three patients with significant restenosis. Occlusion was found in one patient and diffuse narrowing was found in the other two patients (Fig. 3). An arteriogram, which confirmed the duplex scan findings, was also obtained in one patient with mild stenosis. Two patients with normal duplex scans of the area of the PTFE graft underwent digital arteriography for suspected contralateral stenosis and were found to have patent grafts with a normal bifurcation contour (Fig. 4). DISCUSSION The efficacy of carotid endarterectomy in stroke prevention is largely dependent on the minimization of Perioperative complications. Thrombotic occlusion and emboli from the endarterectomy site remain major causes of neurologic complications and must be prevented to sustain good results from this operation. We adopted the use of PTFE interposition
graft placement when primary closure or patch angioplasty did not adequately restore the carotid bifurcation contour in an effort to minimize the chances of turbulent flow and thrombus formation. Our selection of PTFE graft material for carotid reconstruction in this setting was based not only on the ease with which it is used but also on a review of the other options available to restore carotid bifurcation contour. A variety of techniques from resection and primary anastomosis to complicated interpositions have been described, s-° A review of the available literature failed to reveal significant early or late advantages to any one of these techniques. The simplest approach appeared to be resection of the diseased internal carotid artery and PTFE interposition graft placement. Although we found this approach to be effective in the perioperative period, we undertook this study to determine whether or not the long-term results were acceptable. The initial goal of safely reconstructing the
Volume 16 Number 4 October 1992
PTFE interpositionfor carotid endarterectomy 605
Fig. 3. Diffuse internal carotid artery stenosis 87 months after PTFE interposition graft placement. Duplex scan had remained normal throughout follow-up before this time.
Fig. 4. Follow-up arteriogram 18 months after PTFE interposition grafting. Note carotid bifurcation with preservation of contour.
carotid bifurcation was met with the use of PTFE interposition grafts. Intraoperative completion angiography documented the restoration of the normal carotid bifurcation contour with this technique. There were no perioperative complications specifically related to the use of the PTFE graft, and the initial postoperative duplex scan documented a patent carotid bifurcation in all patients. Long-term follow-up, however, revealed that this reconstruction was not free from the risk of restenosis. Although follow-up revealed an overall restenosis rate of 20%, hemodynamically significant recurrent stenosis or occlusion occurred in 12% of the patients. There were no thromboembolic complications encountered in any of the patients. The only neurologic symptoms appeared to be hemodynamic in nature and occurred in a patient with bilateral occlusive disease. The overall survival rate was 96% at a mean of 3 years. The expected rate of restenosis after carotid endarterectomy is difficult to quantify. Reported
rates vary from 1% to 33% and may be related to the duration of follow-up and the persistence of the investigators in obtaining imaging studies, x°qa Our findings of a 12% significant stenosis or occlusion rate and a survival and stenosis-free rate of 89% at 39 months appear to be in keeping with results of series based on careful duplex scanning over similar time intervals. 1416 Further difficulty is encountered with quantification of the recurrent stenosis rate after carotid reconstruction for restenosis. 17-19We found a 33% recurrence rate in this group, and the only neurologic symptoms in our series occurred in one of the patients in this group. Graft stenosis and occlusion with PTFE grafts have been extensively investigated. Our findings with PTFE grafts in the reconstruction of the carotid bifurcation appear to be somewhat more favorable than the results found in reconstructions in other areas. 2°,21 The reasons for this finding cannot be investigated in a small series such as ours. However, the short nature of these grafts and the high flow rates
606
Journal of VASCULAR SURGERY
Sise et al.
at the carotid bifurcation m a y be the key factors that allow patency rates with P T F E grafts comparable to those in other m e t h o d s o f carotid reconstruction. P T F E grafts appear to be effective for carotid reconstruction w h e n an interposition graft is required. T h e carotid bifurcation c o n t o u r is easily restored and the risk o f perioperative t h r o m b o e m bolic complications is apparently low. A l t h o u g h early patency is g o o d , recurrent stenosis occurs. The rate o f restenosis does n o t appear significantly different than rates observed with other forms o f carotid reconstruction. Patients w h o u n d e r g o P T F E interposition graft placement for recurrent carotid stenosis appear to be at high risk for a second restenosis.
10,
11, 12. 13. 14. 15.
REFERENCES
1. ~ CD, Hertzer NR. Postoperative stroke and late neurologic complications after carotid endarterectomy. Arch Surg 1981;116:1561-8. 2. Steed DL, Peitzman AB, Grundy BL. Causes of stroke in endarterectomy. Surgery 1982;92:634-41. 3. Dooner J, Kuechler P. Salvageafter postoperative thrombosis of the carotid artery. Am J Surg 1990;159:525-6. 4. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Star Assoc 1958;53:457-81. 5. Comfier JM, Cormier F, Laurian C, et al. Polytetrafluoroethylene bypassfor revascularizationof the atherosderotic carotid artery: late results. Ann Vasc Surg 1987;1:564-71. 6. Clagetr GP, Patterson CB, Fisher DF, et al. Vein patch versus primary closure for carotid endarterectomy. J VASC SURG 1989;9:213-23. 7. Rosenthal D, Archie JP, Garcia-Rinaldi R, et al. Carotid patch angioplasty: immediate and long-term results. J VASCSURG 1990;12:326-33. 8. Vanmaele R, Van Schil P, De Maesneer M. Closure of the internal carotid after endarterectomy: the advantages of patch angioplasty without its disadvantages. Ann Vasc Surg 1990; 4:81-4. 9. Collins PS, Orrecchia P, Gomez E. A technique for correction
16. 17. 18. 19. 20.
21.
of carotid kinks and coils following endarterectomy. Ann Vasc Surg 1991;5:116-20. Reilly LM, Okuhn SP, Rapp JH, et al. Recurrent carotid stenosis: a consequence of local or systemic factors? The influence of unrepaired defects. J VASCSURG 1990;11:44860. Green RM, McNamara JA, Ouriel K, et al. The clinical course of residual carotid arterial disease. J VASC SURG 1991;13: 112-20. Callow AD, Mackey WC. Long-term follow-up of surgically managed carotid bifurcation stenosis: justification for an aggressive approach. Ann Surg 1989;210:308-16. Cook JM, Thompson BW, Barnes RW. Is routine duplex examination after carotid endarterectomy justified? J VASC SURG 1990;12:334-40. DeGroote RD, Lynch TG, Jamil Z, et al. Carotid restenosis: long-term noninvasive follow-up after carotid endarterectomy. Stroke 1987;18:1031-6. Sawchuk AP, Flanigan DP, Machi J, et al. The fate of unrepaired minor technical defects detected by intraoperative ultrasonography during carotid endarterectomy. J VASCSURG 1989;9:671-6. Healy DA, Zierler RE, Nicholls SC, et al. Long-term follow-up and clinical outcome of carotid restenosis. J VASC SURG 1989;10:662-9. Kazmers A, Zierler RE, Huang TW, et al. Reoperative carotid surgery. Am J Surg 1988;156:346-52. Edwards WH, Edwards WH, Mulherin JL, et al. Recurrent carotid artery stenosis. Ann Surg 1989;209:662-8. Nitzberg RS, Mackey WC, Prendville E, et al. Long-term follow-up of patients operated on for recurrent carotid stenosis. J VASCSUP,G 1991;13:121-7. Veith FJ, Gupta SK, Ascer E, et al. Six-year prospective multi-center randomized comparison of autologous saphenous vein and expanded polytetrafluoroethylene grafts in infrainguinal arterial reconstructions. J VASCSrORG1986;3: 104-14. Rutherford RB, Jones DN, Bergentz SE, et al. Factors affecting the patency of infrainguinal bypass. J VASCSUItG 1988;8:236-46.
Submitted Feb. 13, 1992; accepted May 25, 1992.
DISCUSSION Dr. Victor M. Bernhard (Tucson, Ariz.). Dr. Sise and his colleagues have provided us with an opportunity to appraise a moderately large series of PTFE replacements of the carotid artery at and distal to the bifurcation. In addition to the report of immediate results, they have rigorously followed up their patients over the long term with duplex scans, which I think is essential. In this group of 26 carotid artery replacements, the operative mortality rate was zero and a perioperative neurologic deficit occurred as a result of the interposition graft procedure in only one patient. Late occlusion occurred in one patient. In two patients severe recurrent stenoses
developed, and in one of these cases the stenosis was symptomatic and the patient required a second PTFE graft. Two other patients had mild to moderate asymptomatic restenoses. These are excellent results, and the experience of this series and from a larger series reported by Cormier support the authors' contention that PTFE may be a reliable conduit for repair of carotid artery defects when routine endarterectomy with or without patching is neither safe nor feasible. About 4 or 5 years ago Dr. Cormier and his group reported on 62 grafts in roughly the same position. The
Volume 16 Number 4 October 1992
operative procedure was not exactly the same. The grafts were actually longer, usually coming from the common carotid, and were inserted because there was extensive common carotid disease, and some of them came from other sources such as the subclavian or the arch. Nevertheless, a PTFE graft was placed in the carotid artery position in 62 repairs in 1789 patients, roughly about half the frequency that was reported by Dr. Sise. They also had a zero mortality rate. The incidences of transient deficits, permanent deficits, early asymptomatic occlusions, and so on were very low, and the 4-year patency rate by careful observation was 96%. There are several questions I would like to address to the authors. When discussing technique, the authors note that the use of PTFE replacement grafts permits the surgeon to pick the size graft that will most effectively fit the distal vessel, and in two o f their patients they used a 4.5 mm taper. Were either of these grafts associated with subsequent stenoses, and how long were these patients followed up so that we will really know what the narrower graft may do? The need for graft interposition seems to be relatively high in the authors' overall experience with 401 carotid artery repairs, especially for the correction ofrestenosis and for kinks and coils. In these two subcategories, what was the percentage of patients in whom the more conventional techniques such as patch angioplasty or arterial shortening were abandoned in favor of PTFE interposition grafting? In most reports and in our own experience with recurrent stenosis, the need for carotid replacement rather than endarterectomy or patch angioplasty, or both of these procedures, ranges between 3% and 7%. The pathologic basis of recurrent stenosis in the one patient with resection of the interposition graft was not described, and we should know what was found. In the other patient with a severe stenosis in whom an arteriogram was presented, did the duplex scan suggest layering thrombus, which might explain the appearance of diffuse narrowing in what appeared to be the entire length of the PTFE graft that was inserted? Restenosis occurred in five of the interposition grafts or 19% in this series o f which three appeared in the subset of nine carotid repairs for restenosis of a prior endarterectomy. There is a paucity of data as the authors noted regarding the likelihood of re-restenosis under those circumstances. Nevertheless, this may have been a target group, although the numbers in the authors' series were too small for statistical significance. Will they continue to use or recommend PTFE interposition grafts as the preferred alternative in this subgroup rather than be more persistent in their efforts to accomplish patch angioplasty or to use a vein graft when an interposition graft is required? Despite the fine results reported, this is a retrospective study without comparison with other techniques. Admittedly, the acquisition of sufficient data to make these comparisons may be difficult or impossible. However, a broader experience and a longer follow-up in the manner prescribed by the authors will be required before we
PTFE interpositionfor carotid endartcrectoray 607
abandon conventional methods in favor of PTFE interposition grafting. In the meantime, this report supports the use o f PTFE as a possibly safe and reliable replacement for the carotid bulb and proximal internal carotid artery when no other options are available. Dr. Michael I. Sise. There were no problems with the two 4.5 mm interposition grafts. They have been followed up for 2 and 3 years now, respectively. The patients in whom restenosis developed all received 6 mm thin-walled PTFE grafts. The patient in whom occlusion occurred had undergone an operation for restenosis once with a patch angioplasty and then a second operation for restenosis in which we proceeded with an interposition graft. In those patients who underwent operation for restenosis during the 7-year period, roughly half of the arteries were closed after repeat endarterectomy with patch angioplasty. We routinely patch in all carotid endarterectomies with PTFE. We stopped resecting for kinks and coils in favor of placing the interposition graft. We found that resecting, sewing the distal resection line of internal carotid artery to the proximal internal carotid artery or to the bifurcation, and then placing a patch was more timeconsuming, more difficult to align, and technically less satisfactory than the PTFE interposition. If we were going to go so far as to resect and patch we would prefer to do an interposition graft. The patient in whom the recurrent stenosis developed and who underwent a second PTFE interposition graft procedure was the only patient for whom we have data on the pathologic basis of the restenosis. We found fibrous hyperplasia, typical of recurrent carotid stenosis. These grafts are very short, and there is not an area of graft of any significant length between the proximal anastomosis and • the distal anastomosis. We could not do a graded analysis of the pathologic changes along the graft. In the patient who had no symptoms and who had the diffuse stenosis we have done a number of imaging studies and we have not seen layered thrombus. It appears to be noncalcified plaque. Will we continue to use PTFE interposition grafts? I want to reiterate that these were arterial closures that were difficult. We do not set out to use PTFE grafts. We set out to do an endarterectomy and do a patch angioplasty if we think it is feasible and if we think we will get a good restoration of contour. We believe there is a window of vulnerability when we perform this operation from the time we clamp the carotid artery until the time we unclamp and through the first couple of hours. That is the time when we can make the difference with our perioperative stroke rate. We believe that PTFE grafting, in certain patients in whom arterial closure is difficult, offers an advantage in restoring contour in a fashion that prevents thromboembolic complications. So in those settings we will continue to use PTFE interposition grafts, but we will also continue to aggres-
608
Sise et al.
sively follow up our patients. I want to reiterate that this procedure is a good way to get out of trouble. This is not a primarily preferred procedure. Dr. Ronald J. Stoney (San Francisco, Calif.). My experience with prosthetic carotid artery replacement has been in secondary restenoses in which I originally used knitted Dacron conduits and have more recently used PTFE grafts. Fortunately there have not been any subsequent recurrences with the PTFE and only one with the knitted Dacron grafts. However, there are probably only half a dozen in each group. I think your idea that this technique should be considered in the patient with myointimal hyperplasia causing restenosis is a good one. Such lesions may involve a very difficuk technical challenge in patching, or if you have problems you may attempt to cover yourself with the use of a conduit. Dr. Ehrenfeld has been interested in generous patching of recurrent carotid stenoses and argues that a larger postreconstruction lumen might be beneficial. Because you cannot prevent re-recurrence, a larger lumen ~ accommodate restenosis and still provide a residual lumen that will be acceptable. It is tempting to think that in removal of the arterial segment in which stenosis has recurred any replacement with a graft might be preferable. A few times I have put a PTFE conduit in for first-time recurrence, and they have worked very well. My question is this: in the occasional patient with a primary myointimal recurrence, will you use a tubular graft rather than a patch? Dr. Sise. The decision to use an interposition graft is strictly related to the ability to get what we would consider an acceptable contour after endarterectomy. We try to endarterectomize the recurrent carotid stenosis, as I am sure most of you do. If we cannot do it and if it looks like there is a smooth fibrous area, we consider primary patching, but those patients are few and far between. There are important issues related to recurrent carotid artery stenosis. Dr. Strandness and his group presented an important paper 4 years ago in Kauai, Hawaii, to this society in which they documented a significant carotid restenosis rate after carotid endarterectomy. However, these lesions were usually asymptomatic. There are little data from long-term follow-up with
Journal of VASCULAR SURGERY
imaging studies concerning the results after a second operation for carotid artery stenosis. It is important for those of us with registries with large groups of patients that we take a look at our patients to closely follow up those who have undergone operation for carotid artery restenosis. We believe this group is at an extremely high risk for recurrent disease. Dr. Eugene F. Bernstein (La Jolla, Calif.). It is my sense that in Europe vascular surgeons do vein interposition reconstructions for the carotid artery more frequently than we do. A number of friends I have there say something like, "Oh, if we have a little problem where the lesion is too long or where there is a kink or there is something else, we just put in a piece of saphenous vein, and we probably do it one in five operations." I wonder if you would comment about venous interposition versus PTFE grafting when an interposition is required and also distinguish between the situation in which you have planned to perform interposition grafting versus the situation in which you have not planned to do it but are forced to do it for technical reasons during the operation. Dr. Sise. A lot of agonizing went into what to do with these patients. In the early 1980s we had three patients with extremely difficult recurrent stenoses in whom we placed vein interpositions. All had early complete occlusions. We went to the literature and found a paucity of data on what happens to saphenous vein interpositions. In the few studies with follow-up, there was a significant rate of recurrent stenosis or total occlusion. We did not wish to add a saphenous vein harvest, which may or may not result in an acceptable size match to the carotid bifurcation, when we knew that there was not good documentation that vein was superior to PTFE. The PTFE graft is easy to use in this position, and we therefore chose to use it when an interposition was required. I do not believe that anyone in this society does not gain more and more respect for carotid artery reconstruction independent of any external critical analysis such as that which our nonsurgical colleagues are placing on us. Certainly the more operations we do, the more we develop the "play-off game" mentality that every carotid artery operation must be successful.
