Gore-Tex Grafts for Replacement of the Superior Vena Cava

Transcription

1 Gore-Tex Grafts for Replacement of the Superior Vena Cava William H. Heydorn, COL, MC, Russ Zajtchuk, LTC, MC, John Miller, CPT, VC, and George F. Schuchmann, LTC, MC, all USA ABSTRACT In an effort to evaluate polytebafluoroethylene (Gore-Tex) as a replacement for large veins, tubular grafts of the material were substituted for the precava in 33 mongrel dogs weighing 15 to 30 kg. Thirteen dogs had grafts 10 cm long by 12 mm wide with a 15- to 30-p pore size; 15 dogs had grafts of the same length and width from 90-/1. pore size material; and 5 dogs had grafts 5 cm long by 12 mm wide, of 90-p pore size Gore-Tex. The eff ect of Gore-Tex tubes with and without spiral support and the differences in anatomical positions of the grafts were evaluated in terms of patency. Dogs were killed if a venogram showed occlusion of the graft. Twelve dogs survived 90 days; the longest survivor was killed at eleven months. At postmortem examination, the graft was patent but extremely narrow at the atrial end. Modification in the fabrication of Gore-Tex may eliminate factors contributing to graft failure; experiments longer than 90 days are necessary to evaluate Gore-Tex as a large vein replacement material. The 90-p pore size material used in this experiment was unsuitable as a canine venous substitute. Tubular grafts of Gore-Tex* (polytetrafluoroethylene) have been used in experimental animals [4, 13, 151 and in man [12, 151 for vein replacement. It has been suggested that this material is superior to other synthetic vein replacements [151. We evaluated Gore-Tex as a replacement for the superior vena cava (precava of the dog), and in this paper we present our findings and assessment of the material. From the Thoracic and Cardiovascular Surgery Service and the Clinical Investigation Service, Fitzsimons Army Medical Center, Denver, CO. The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of the Army or the Department of Defense. Accepted for publication Nov 18, Address reprint requests to Technical Publications Editor, Letterman Army Medical Center, Presidio of San Francisco, CA *W. L. Gore and Associates, Flagstaff, AZ. Material and Methods Thirty-three mongrel dogs weighing 20 to 30 kg were used in this study.t They were anesthetized with intravenously administered thiamylal sodium and ventilated with an endotracheal tube and a Bird respirator using halothane. The precava and right heart were exposed through either the third or fourth intercostal space. The precava was dissected free from the junction of the brachiocephalic veins to its intrapericardial portion. The azygos vein was ligated and divided in all experiments. Heparin (5,000 units) was given just prior to application of occluding clamps. Each anastomosis was done with continuous 6-0 Prolene. In Groups 1 and 2 Gore-Tex of 15- to 30-p pore size was used, whereas in Groups 3 to 6 the tubes were of 90-p pore size. In Groups 1 to 5 the tubes were 10 cm long and 12 mm wide. In Group 6 the tubes were 5 cm long. The precava was either ligated or excised. At the end of the procedure a chest tube was left in the thorax for approximately one hour postoperatively. The tube was connected to negative suction. Dogs were given cephaloridine during the procedure and for 3 days thereafter. Venograms were performed on animals (1) if clinically indicated, (2) after 90 days, andlor (3) before electively killing the animal. Gross and microscopical postmortem pathological examinations were performed on all dogs at the time of death. The six groups were used to evaluate different varieties of Gore-Tex tubes and the effects of their anatomic position on patency. GROUP 1. Eight dogs had a Gore-Tex tube inserted between the right brachiocephalic vein and the lateral aspect of the right atrium. An tin conducting the research described in this report, we adhered to the Guide for Laboratory Animal Facilities and Care, as promulgated by the Committee on the Guide for Laboratory Animal Resources, National Academy of Sciences-National Research Council. 539

2 540 The Annals of Thoracic Surgery Vol 23 No 6 June 1977 ellipse of right atrial wall was excised as a site for the caudad anastomosis. GROUP 2. Five dogs had insertion of a Gore- Tex tube from the lateral aspect of the precava (end-to-side at the level of the brachiocephalic veins) into the lateral aspect of the right atrial wall. A shunt between the right internal thoracic artery and the preca.va was placed just above the cranial anastomosis (Fig 1). GROUP 3. Five dogs had a Gore-Tex tube inserted between the base of the resected right atrial appendage and the precava at the level of the brachiocephalic veins. This experimental model is the saml-3 as the one described by Leshnower and associates [8], except that we used a slightly shorter graft. GROUP 4. In 5 dogs a Gore-Tex tube that had been made noncollapsible by a spiral support about the graft (Fig; 2) was inserted between the transected precava and the open atrial appendage. GROUP 5. Five dogs had a Gore-Tex tube Fig2. Gore-Tex tube with spiralsupport. Fig 1. Gore-Tex betzoem the precava and right atrial wall (white arrows) zcdh proximal internal thoracic arterushunt (b1ackar:row). The caua has not vet been &gated. with spiral support inserted between the transected precava and the right lateral atrial wall. GROUP 6. In 5 (dogs the precava was excised from just below the junction of the brachiocephalic veins to just above its entry into the right atrium. A tubular graft of Gore-Tex was then interposed. Results Twelve of the 33 dogs lived for 90 days in apparent good health. Some demonstrated edema of the head and forelegs before 90 days and either died spontaneously or were killed after a venogram showed occlusion of the graft. The results are summarized i.n the Table. Comment These experiments were stimulated by a patient who had occlusion of the superior vena cava secondary to histoplasmosis. A composite vein graft and proximal arteriovenous fistula was used to bypass {the obstruction, but the graft closed after being patent for one year. This patient and his original operation have been reported by Miller and Sullivan (Patient 8) [lll.

3 541 Heydorn et al: Gore-Tex Grafts for SVC Replacement Results of Precava Bypass with Gore-Tex Tubes in 33 Dogs No. of Grafts No. Graft Pore No. of Early Patent at of Length Size Distal Closures Group Dogs (cm) (PI Anastomosis (range in days) 90 Days Death to 30 Atrial wall 6 (13-45) to 30 Atrial wall 4 (19-49) Atrial 4 (17-34) 1 0 appendage Sa 90 Atrial 4 (21-39) 1 1 appendage Sa 90 Atrial wall 1 (9) Precava 2 (41-78) 3 1 "S indicates that tube had spiral support. A satisfactory prosthetic replacement for large veins in man has not been available. Reviews by Haimovici [7] and Scherck [13] and their colleagues describe numerous experiments that were performed in an attempt to find a dependable venous graft. Unfortunately, inconsistent or poor success rates in experimental models have caused surgeons to be reluctant to use prosthetic material in the venous system. It has been suggested that Gore-Tex, an expanded, highly porous form of Teflon (polytetrafluoroethylene), is superior to other synthetic vein replacement materials [15]. It has been used as an arterial substitute [2, 3, 91, for arteriovenous shunts in patients on dialysis [16], or for construction of aortopulmonary shunts [5] as well as for venous replacement. Fujuwara and associates [41 reported patency at one month in 3 of 10 dogs with 3-cm lengths of Gore-Tex interposed in the precava (2 other dogs with patent grafts died in less than a month). Smith and associates [14] used Gore-Tex to replace 2- to 4-cm segments of the postcava of dogs; they reported an 83% patency rate. Four of these animals were followed for six months and the others for less time. When comparing the results of experiments with large vein replacement (superior vena cava, or precava of dogs in particular), there are several factors that must be considered. These include the length of the graft, use of arteriovenous shunts or azygos vein ligation to enhance flow, type of prosthetic material, site of anas- tomosis (anastomosis into the atrial wall is less likely to be successful than venous interposition), and duration of follow-up. Unless these variables are kept in mind, patency rates and comparative experiments have little value. Twelve of 33 dogs survived for 90 days with patent grafts. Subsequent follow-up has shown that this period is not adequate to predict longterm patency. In Groups 1 and 2, 3 of 12 dogs survived longer than 90 days and were electively killed at 188, 168, and 149 days. Two dogs had minimal narrowing at the atrial suture line, and the third had an apparently perfect result (Fig 3). The graft failures appeared to be caused by thrombosis associated with fibrosis and stricture at the atrial suture line. Stenotic fibrosis at a cardiac suture line has been a problem for other investigators [l, 61 and may prevent a high level of success in the attachment of any graft to the atrium. The addition of an internal thoracic artery fistula 1 cm above the cephalad suture line as suggested by Miller and colleagues [lo] did not improve results in Group 2. Microscopical examination of these 15- to 30-p pore size grafts revealed mild inflammatory reaction with lymphocytes and plasma cells outside the grafts. There was some ingrowth of fibroblasts into the tubes but not far enough to attach to the pseudointimal surface. Soyer and associates [151 have suggested that a larger pore size would permit more rapid ingrowth of capillaries and fibroblasts through the graft, providing attachment and anchorage of the

4 542 The Annals of Thoracic Surgery Vol 23 No 6 June 1977 Fig3. Graft in a Group 1 dog is widelypatentsix months after bypass was inserted. Suture lines are indicated by black arrows and the ligated precava by a white arrow. pseudointima. In order to evaluate this concept, we used 90-p fibril length Gore-Tex in Groups 3 through 6. The experimental model of Leshnower and associates [8] was used in Groups 3 and 4. They had obtained 100 /~ 90-day patency by using 12-cm straight Troyfelt grafts,* whereas other forms of this Dacron material gave somewhat less satisfactory results. The Gore-Tex used in Groups 3 through 6 did not perform as well as the plain felt grafts that were used in their experiments. One dog in Group 3 survived 90 days and then died on day 116 with an occluded graft. The addition of an external spiral support in Group 4 and a change in the distal anastomotic site in Group 5 led to some improvement in 90-day patency rates but did not result in long- *Troyfelt grafts (Ethicon Inc, Somerville, NJ 08876) are no longer available for experimental or clinical use (personal communication from R. L. Kronenthal, Ph.D., Director of Research, Sutures, Ethicon, Inc, January 27, 1976). term patency. A graft in Group 4 was patent at 110 days (Fig 4) and almost occluded at the atrial anastomosis when the animal was killed on day 321. Three of the other long-term survivors in Groups 3 through 6 were followed from 204 to 339 days and had either occluded or very narrow grafts when they were killed. All grafts would have closed if enough time had elapsed. Examination of the grafts showed a fibrotic process at the atrial anastomosis. These 90-p tubes did show complete fibrous ingrowth of cells with neovascularization as early as 14 days (Fig 5). In addition to the fibrous process at the cardiac anastomosis, there was transmural fibrous hyperplasia in areas of the graft away from any suture line. Group 6 eliminated a cardiac suture line; this group provides a comparison with experiments done by others [4,14] in which shorter segments of less porous Gore-Tex were used to replace the precava of dogs. Three of the grafts in Group 6 were patent at 90 days; however, 2 were markedly narrowed when the dogs were killed (98 and 100 days) and the third was closed at nine months. The attempt to increase patency by increasing pore size had not been successful.

5 543 Heydorn et al: Gore-Tex Grafts for SVC Replacement Fig4. Venogram demonstratingpatency of Gore-Tex tube in a Group 4 dog at 110 days. Arrows denote suture lines marked at operation zuith metal clips. The patent but narroz~~ed Iunien is marked ziiith a double arrow. Fig5. High-power microscopical picture of 90-p Gore-Texshowing cellgrowth through full thickness of graft. Black arrow indicates inside (neointima). (HDE; original magnification X100.) The rapid attachment and revascularization of the neointima was associated with extensive external fibrosis, transmural fibrosis, and eventual graft closure (Fig 6). Campbell and associates [2, 31, who worked with expanded polytetrafluoroethylene as an arterial substitute, noted a relationship between patency and pore size. They had a high incidence of early thrombosis with large-pore material. Although early thrombosis was not a frequent problem in this study, it would appear that the longer fibril length material used is not suitable as a canine venous substitute. A new high-hoop-strength graft has been designed to inhibit the degree of fibrosis while still allowing transmural organization. This material is currently being evaluated as a venous substitute. Conclusions From our work with Gore-Tex (polytetrafluoroethylene) in dogs, we have reached the following conclusions: Fig6. Gore-Tex tube (Group 6) nearly occluded by transmural fibrosis (white arrow). The suture lines are not involved (black arrows). 1. Gore-Tex can be used to replace the precava of the dog. The fibril length of the material used has critical importance in determining long-term patency. The 30-p Gore-Tex is preferable to 90-p fibril size Gore-Tex, which is associated with a process of transmural fibrosis.

6 544 The Annals of Thoracic Surgery Vol 23 No 6 June 1977 The 90-p Gore-Tex used in this study is unsuitable as a canine venous substitute. 2. The addition of a proximal arteriovenous shunt did not improve the patency rate when 10-cm lengths of 30-p Gore-Tex were interposed between the precava and the right atrium. 3. Graft failures may occur many months after insertion; 90-day experiments are not sufficient to determine long-term patency. 4. Grafts sewn into the right atrium (atrial wall or atrial appendage) will be affected by fibrosis at the atrial suture line. Perhaps this is related to muscularity of the atrial wall. A way to prevent graft failure caused by this type of fibrosis has not been discovered. 5. The addition of a spiral support to the more porous and collapsible graft material may help maintain patency, but no evidence for this suggestion can be derived from these experiments. 6. Modification in the fabrication of Gore-Tex may eliminate factors contributing to graft failure; further study of this material should be undertaken. Acknowledgments The material for the grafts (Gore-Tex) was provided by W. L. Gore and Associates, Flagstaff, AZ. Members of the staff at Gore and Associates also provided assistance in interpretation of the pathological specimens obtained at postmortem examination. MAJ James Seab, Jr., MC, USA, of the Anatomical Pathology Service at Fitzsimons Army Medical Center, Denver, CO, provided assistance in interpreting the pathological specimens. References Ashburn FS, Sewell WH, Huggins CE: Experimental replacement of the superior vena cava with homologous arteries, and report of a case with malignant obstruction replaced with a heterologous artery. J Thorac Surg 31:618, 1956 Campbell CD, Goldfarb D, Detton DD, et al: Expanded polytetrafluoroethylene as a small artery substitute. Trans Am SOC Artif Intern Organs 20236, Campbell CD, Goldfarb D, Roe R: A small arterial substitute: expanded microporous polytetrafluoroethylene. Patency versus porosity. Ann Surg 182: 138, Fuiuwara Y, Cohn LH, Adams D, et al: Use of Gore-Tex grafts for replacement of the superior and inferior venae cavae. J Thorac Cardiovasc Surg 67:774, Gazzaniga AB, Elliott MP, Sperling DR, et al: Microporous expanded polytetrafluoroethylene arterial prosthesis for construction of aortopulmonary shunts: experimental and clinical results. Ann Thorac Surg 21:322, Gerbode F, Yee J, Rundle FF: Experimental anastomoses of vessels to the heart: possible application to superior vena caval obstruction. Surgery 25:556, Haimovici H, Hoffert PW, Zinicola N, et al: An experimental and. clinical evaluation of grafts in the venous sy,stem. Surg Gynecol Obstet 131:1173, Leshnower AC, Shumacker HB, Shariatzadeh AN: Experimental studies with synthetic venous bypass grafts, including a new experimental model. Am J Surg 128:392, Matsumoto H, Hasegawa T, Fuse K, et al: A new vascular prosthesis for small caliber artery. Surgery 74:519, Miller RE, Corneil NJ, Sullivan FJ: Replacement of superior vena cava with autogenous tissue. Ann Thorac Surg 15:474, Miller RE, Sullivan FJ: Superior vena caval obstruction secondary to fibrosing mediastinitis. Ann Thorac Surg 15:483, Norton L, Eisemm B: Replacement of portal vein during pancreatectomy for carcinoma. Surgery 77:280, Scherck JP, Kerstein MD, Stansel HC: The current status of vena caval replacement. Surgery 76:209, Smith DE, Hamrnon J, Anane-Sefah J, et al: Segmental venous replacement. J Thorac Cardiovasc Surg 69:589, Soyer T, Lempinen M, Cooper P, et al: A new venous prosthesis. Surgery 72:864, Volder JGR, Kirlkham RL, Kolff WJ: A-V shunts created in new ways. Trans Am SOC Artif Intern Organs 193, 1973