Vascular Anastomoses in Growing Vessels: The Use of Absorbable Sutures

Transcription

1 Vascular Anastomoses in Growing Vessels: The Use of Absorbable Sutures John L. Myers, M.D., John A. Waldhausen, M.D., Walter E. Pae, Jr., M.D., Arthur l3. Abt, M.D., G. Allen Prophet, B.S., and William S. Pierce, M.D. ABSTRACT Primary end-to-end infrarenal aortic anastomoses were performed in 36 piglets using two synthetic absorbable suture materials: polydioxanone and coated polyglactin. Animals were killed at 1, 4, and 11 weeks and 6 months following operation. Each aorta was removed, burst-tested to 300 mm Hg, radiographed, and examined histologically. All anastomoses were patent, and no burst-test failures occurred. Stenosis occurred in 14 of 17 animals at 1 and 4 weeks, respectively. One of 5 animals exhibited stenosis at 11 weeks, and none of the 14 animals had stenosis 6 months postoperatively. Histological examination revealed fibrosis replacing areas of disrupted elastics at 6 months in both suture groups. This study suggests that absorbable suture material, in particular polydioxanone, because of its excellent handling characteristics and prolonged tensile strength retention, will be useful for the repair of vascular and cardiac anomalies where growth of the suture line is required. Growth of a primary vascular anastomosis is influenced by the technique of anastomosis as well as the suture material utilized. However, stenosis still remains a major problem in the growing patient. With early operative repair of coarctation of the aorta by end-to-end anastomosis, for example, the incidence of recoarctation in infants under 6 months of age has been reported to be as high as 62% [l]. This study was designed to observe the growth of vascular anastomoses performed with two From the Departments of Surgery and Pathology, College of Medicine, The Pennsylvania State University, The Milton S. Hershey Medical Center, Hershey, PA. Presented at the Eighteenth Annual Meeting of The Society of Thoracic Surgeons, Jan 11-13, 1982, New Orleans, LA. Address reprint requests to Dr. Myers, Department of Surgery, College of Medicine, The Pennsylvania State University, The Milton S. Hershey Medical Center, 500 University Dr, Hershey, PA synthetic absorbable polyester suture materials: monofilament polydioxanone [2] and braided multifilament coated polyglactin [3]. Materials and Methods Thirty-six piglets (body weight 12.7 f 3.3 kg, standard error of the mean [SEMI) were anesthetized utilizing a mixture of halothane and oxygen. Following endotracheal intubation, a left retroperitoneal flank incision was performed; the infrarenal abdominal aorta was exposed and its external diameter measured (Fig 1). Each animal was anticoagulated with heparin (100 Ulkg), and end-to-end anastomosis of the infrarenal abdominal aorta was performed using a continuous suture technique. The site of anastomosis was then marked by a nonabsorbable suture in order that precise identification of the anastomosis could be made at future postmortem examination. All anastomoses were performed with size 6-0 suture material with the aid of 2.5 power magnification loupes. The animals were returned to cages until their skin wounds were healed, and then placed in freerange pens until the time of killing. Animals were killed at 1 week, 4 weeks, 11 weeks, and 6 months following operation. The abdominal aortas were carefully excised, and all branches were ligated. Each vessel was then filled with a water-soluble roentgenographic contrast material at a mean intraluminal pressure of 100 mm Hg, radiographs were taken, and diameters were calculated 1 cm proximal to the anastomosis, at the anastomosis, and 1 cm distal to the anastomotic site. Following this, each aorta was burst-tested at a mean intraluminal pressure of 300 mm Hg (Fig 2). The aorta was opened longitudinally and photographs of the gross specimen were taken. The specimens were fixed in formalin, and representative sections were embedded in paraffin and stained with hematoxylin and eosin and by The Society of Thoracic Surgeons

2 530 The Annals of Thoracic Surgery Vol34 No 5 November 1982 Fig 1. Retroperitoneal approach for anastomosis of the infrarenal abdominal aorta. Inset shows position of the left flank incision. Mercury manometer Anastomosis \ Abdominal aorta \\ n Fig 2. Device for burst-testing the abdominal aorta. A polyvinyl tube is inserted into each end of the vessel and secured with a heavy ligature. One tube is occluded, and the other tube is connected to a saline-filled syringe and mercury manometer. Fig 3. Representative aortic anastomoses performed b with polydioxanone (PDS) and coated polyglactin (CPG) at the time of postmortem examinations. (A) PDS at 6 months. (B) PDS at 11 weeks. (C) PDS at 4 weeks. (D) CPG at 6 months. (E) CPG at 11 weeks. (F) CPG at 4 weeks. The arrow marks the anastomosis.

3 531 Myers et al: Vascular Anastomoses in Growing Vessels

4 532 The Annals of Thoracic Surgery Vol 34 No 5 November 1982 with elastic van Gieson before being examined by routine light microscopy. Fig 4. Radiograph of a representative abdominal aorta filled with roentgenographic contrast material at a pressure of 100 mm Hg. The anastomosis is marked by the arrow. Results Six months following operation, the body weight had increased to a mean of 1,016% of the operative weight (from 12.7 f 3.3 to f 27.8 kg SEM), and the aortic diameter had increased by 155% (from 6.2 f 0.8 to 15.8 f 1.8 mm SEM). There were no postoperative complications or deaths. All anastomoses were patent, and no vessel failed burst-testing at 300 mm Hg pressure. No intraluminal suture, with or without adherent thrombus, was seen in any vessel. Photographs of representative anastomoses in each group examined 4 weeks, 11 weeks, and 6 months postoperatively are shown in Figure 3. These anastomoses examined 6 months postoperatively showed smooth intima at the anastomotic site and were difficult to distinguish from adjacent normal vessel (see Fig 3A, D). Figure 4 is a representative radio- Table 1. Results of 6-0 Polydioxanone Running Suture Technique At Operation At Killing Aortic Average Relative Areaa Time in Areab Anastomosis Change" Pig No. Weight (kg) (mm') Days Weight (kg) (mm') (mm') ( O% ameasurement taken at anastornotic site. "Average Area = (area 1 an proximal + area 1 cm distal to anastomosis) + 2. 'Minus denotes relative decrease and plus denotes relative increase in aortic cross-sectional area.

5 533 Myers et al: Vascular Anastomoses in Growing Vessels Table 2. Results of 6-0 Coated Polyglactin Running Suture Technique At Operation At Killing Aortic Average Relative Areaa Time in Areab Anastomosis ChangeC Pig No. Weight (kg) (mm') Days Weight (kg) (mm') (mm') ( O h ) ameasurement taken at anastomotic site. "Average Area = (area 1 an proximal + area 1 cm distal to anastomosis) t 2. CMinus denotes relative decrease and plus denotes relative increase in aortic cross-sectional area graph of an aorta removed 6 months postoperatively. The relative change in anastomotic area for the polydioxanone and coated polyglactin suture groups is shown in Tables 1 and 2, respectively. In piglets killed at 1 week and 4 weeks postoperatively, stenosis was present in 6 of 8 animals and 8 of 9 animals, respectively. Between 4 weeks and 11 weeks postoperatively there was a significant (p < ) decrease in the incidence of stenosis. Only 1 of 5 animals exhibited stenosis at 11 weeks postoperatively, and none of the 14 animals had stenosis 6 months postoperatively (see Tables 1, 2). Histological examination of the anastomoses revealed areas of medial necrosis and calcification at 1 and 4 weeks. A fibroblastic and histiocytic reaction of varying intensity was present at 1 week, and a well-developed foreign body reaction was noted at 4 weeks. At 11 weeks, the granulomatous foreign body reaction was more localized to areas of degenerating suture material and the foci of necrotic media were no longer present. These areas were replaced by collagen deposits, but still exhibited modest fibroblastic proliferation. At 6 months, suture material could no longer be identified in either specimen and the inflammatory reaction had subsided. Evidence of extensive elastica disruption was apparent, and these areas showed replacement by dense fibrous tissue (Fig 5). There were no discernible differences in the pathological alterations between the two groups. Comment The concept of using absorbable suture materials for vascular anastomoses in growing vessels was first advocated in the early 1950s. Several investigators used catgut suture material in vascular anastomoses with good results in experimental animals [4-81 and clinically in three patients [8]. More recently, polyglycolic acid, a

6 534 The Annals of Thoracic Surgery Vol34 No 5 November 1982 Fig 5. (A) The aorta excised at 4 weeks shows an area of necrosis (large arrow) and a fragment of polydioxanone suture (small arrow) surrounded by inflammatory cells. (H&E with polarized light; X75.) (B) An aorta excised at 11 weeks exhibits scattered fragments of polyglactin suture (arrows) and minimal inflammation. (H&E with polarized light; ~75.) (C) The elastic tissue is absent at the right, and a fragment of polydioxanone suture is noted (arrow) in an aorta excised at 11 weeks. (Elastic van Gieson; X75.) (D) At 6 months, the aorta is composed of dense fibrous tissue. Small fragments of calcified elastic tissue (arrows) are noted. (H&E; X75.) The intimal surface is at the top in all photographs. synthetic absorbable suture, has yielded good results in our laboratory [91 and others [lo-121, and has been used successfully in clinical vascular operations in Taiwan [13]. Despite these experimental studies demonstrating good results in vascular anastomoses performed with catgut and polyglycolic acid suture materials, absorbable sutures have yet to be accepted in clinical vascular or cardiac surgery because of concern that the suture material would be degraded before sufficient anastomotic healing occurred. Two new synthetic absorbable sutures are now available for clinical use: polydioxanone, a monofilament suture material, and coated polyglactin, a multifilament braided suture

7 535 Myers et al: Vascular Anastomoses in Growing Vessels m t 01 1 I I I 1 I I L TIME IN VlVO (WEEKS) Fig 6. The in vivo tensile strength of 6-0 polydioxanone (PDS) and coated polyglactin (CPG) suture materials as a function of time. The dashed line represents the tensile strength of size 6-0 polypropylene (PP). (Data supplied through the courtesy of Ethicon, Inc., Somerville, NJ.) material. The tensile strengths of these two suture materials are shown in Figure 6. Coated polyglactin has lost all tensile strength by 3% weeks; however, polydioxanone still has nearly 50% of its tensile strength at 4 weeks and is not completely absorbed for 6 months. The monofilament polydioxanone had handling characteristics similar to polypropylene and passed through tissues easily with minimal drag, whereas coated polyglactin had moderate tissue drag. Piglets were chosen as the experimental animal for our study because they have been shown to double or triple their aortic diameter from 6 weeks to 6 months of age, which is comparable to the increase observed in the growing child from birth to maturity [14]. Additionally, the histology of the pig aorta closely resembles that of the human aorta [15]. Previous studies from our laboratory using the growing piglet demonstrated some of the complications that may result when nonabsorbable suture (polypropylene) is used for vascular anastomoses where growth is required, namely, stenosis and intraluminal suture with adherent thrombus ~91. In this study, no anastomotic leaks or disruptions were observed in either suture group, and both suture materials eliminated any long-term restraining bands or intraluminal PP suture material. Stenosis seen early was a result of a pursestring effect of the suture at the time of operation; however, as the absorbable suture material was degraded, there was no restraining band, thus allowing the anastomosis to grow with the normal vessel. No stenosis approached hemodynamic importance. There was a significant decrease in the number of animals exhibiting stenosis between 4 and 11 weeks, which coincides with the time in which the tensile strength of polydioxanone suture material had diminished to less than 45% and that of coated polyglactin to zero. Aneurysmal dilatation was not seen in any anastomosis. Our study supports the work of previous investigators and obviates the concern about whether absorbable suture material is safe for for primary vascular anastomoses. Absorbable monofilament polydioxanone would be useful in a variety of procedures to repair vascular and cardiac anomalies in infants and children, such as coarctation of the aorta, interrupted aortic arch, total anomalous pulmonary venous return, Blalock-Taussig anastomosis, and the Senning procedure for transposition of the great vessels. In summary, vascular anastomoses in growing vessels performed with the absorbable suture materials polydioxanone and coated polyglactin resulted in no stenosis at 6 months, a smooth intimal surface at the anastomotic site, no intraluminal suture material, and a strong vessel. There were no burst-test failures at 300 mm Hg. This study suggests that absorbable suture material would be useful for the repair of a variety of vascular and cardiac anomalies in which growth of the suture line or vascular anastomosis is required. Because of its excellent handling characteristics and prolonged tensile strength retention, monofilament polydioxanone is the absorbable suture material of choice for anastomoses in growing vessels. Suture materials for this study were kindly provided by Dr. Richard Kronenthal of Ethicon, Inc., Somerville, NJ. References 1. Tawes RL Jr, Aberdeen E, Waterston DJ, Bonham-Carter RE: Coarctation of the aorta in

8 536 The Annals of Thoracic Surgery Vol34 No 5 November 1982 infants and children. A review of 333 operative cases, including 179 infants. Circulation 39: Suppl 1:173, Ray JA, Doddi N, Regula D, et al: Polydioxanone (PDS), a novel monofilament synthetic absorbable suture. Surg Gynecol Obstet 153:497, Aston SJ, Rees TD: Vicryl sutures. Aesth Plast Surg 1:289, Sauvage LR,' Harkins HN: Growth of vascular anastomoses: an experimental study of the influence of suture type and suture method with a note on certain mechanical factors involved. Bull Johns Hopkins Hosp 91:276, Tawes RL Jr, Aberdeen E, Berry CL: The growth of an aortic anastomosis: an experimental study in piglets. Surgery 64:1122, Sako Y, Chisholm TC, Merindino KA, Varco RL: An experimental evaluation of certain methods of suturing the thoracic aorta. Ann Surg 130:363, Johnson J, Kirby CK: The relationship of the method of suture to the growth of end-to-end arterial anastomoses. Surgery 27:17, Detering RA Jr, Coleman CC Jr, Kee J, Humphreys GH: An experimental evaluation of catgut as a vascular suture material and a report on its clinical use. J Thorac Surg 23:303, Pae WE Jr, Waldhausen JA, Prophet GA, Pierce WS: Primary vascular anastomosis in growing pigs: a comparison of polypropylene and polyglycolic acid sutures. J Thorac Cardiovasc Surg 81:921, Dardik H, Dardik I, Katz AR, et al: A new absorbable synthetic suture in growing and adult primary vascular anastomoses: morphologic study. Surgery 68:1112, Watts DR, Carr SH, Hohf RP: Poly(glyco1ic acid) sutures in canine vascular anastomoses. J Biomed Mater Res 10:867, Alvarez-Corder0 R, Reyes EI, Alcazar MG, Hernandez-Jauregui P: Evaluation of polyglycolic acid sutures in vascular surgery. J Surg Res 15:35, Kuo TP, Huang TJ, Howng SL: The clinical use of synthetic absorbable suture in vascular surgery. Taiwan I Hsueh Hui Tsa Chih 73:45, Moss AJ, Adams FH, O'Loughlin BJ, Dixon WJ: The growth of the normal aorta and of the anastomotic site in infants following surgical resection of coarctation of the aorta. Circulation 19:338, Detweiler DK: Swine in comparative cardiovascular research. In Bustad LK, McClellen RO (eds): Swine in Biomedical Research. Richland, WA, Battelle Memorial Institute, Pacific Northwest Laboratories, 1966, p 301 Discussion DR. JACK J. CURTIS (Columbia, MO): In 1972 I had the memorable experience of working in the laboratory with Dr. Jerry Wada, studying arterial anastornotic growth with absorbable suture. This was at a time when reports of recoarctation of the aorta were plentiful and shunting procedures for cyanotic heart disease were common. We too chose the pig model'for studying arterial anastomotic growth, realizing that the aortic diameter may increase 40O0/o with maturity. We also used a synthetic absorbable suture and continuous suture technique, and were very excited with our observation of unimpeded growth of the arterial anastomosis. Of course, Alexis Carrel had made this observation several decades before, using catgut suture. Indeed, just as reported today, we had difficulty discerning the area of anastomosis. We suggested to a major suture manufacturing company that if they would coat the absorbable suture to reduce arterial drag, they might have a good product for pediatric arterial surgery. They didn't think that the market was available, and I have not had the intestinal fortitude to use the suture in patients despite our investigational observations. We are aware, of course, that absorbable suture has been used successfully in human beings for arterial anastomoses. I would ask the authors if they are now planning to apply their investigational obseivations to their clinical practice. DR. FRITZ DEROM (Ghent, Belgium): In a similar but still uncompleted investigation with polydioxanone, our studies confirm that a running suture does not result in stenosis after 6 months in the growing pig. According to our findings, however, this is only part of the story. We used 12 growing pigs, Qf which 6 with Prolene (controls) and 6 with polydioxanone were killed at 6 months. In all 6 pigs with polydioxanone, we found marked dilatation of the vessel at the anastomotic point. The increase in diameter varied between 30 and 120%. Although there were some stenoses in the control pigs, we did not observe dilatations in them. In these experiments we tried to mimic clinical conditions, and therefore all anastomoses were performed after excision of a portion of the aorta and under a certain amount of tension. We postulate that this tension, which is similar to what we have to deal with in clinical cases, was responsible for the aneurysmal dilatations. DR. MYERS: I thank the discussants for their interesting comments. In answer to Dr. Curtis's question, we do plan to proceed with clinical trials using polydioxanone suture in a variety of cardiac and vascular anomalies, to include atriotomy closure, the Senning procedure for transposition of the great vessels, Blalock-Taussig anastomosis, and for repair of total anomalous pulmonary venous return, coarctation of the aorta, and interrupted aortic arch. As noted in our paper, several investigators have documented in the literature that the aortic growth in

9 537 Myers et al: Vascular Anastomoses in Growing Vessels piglets occurring from approximately 4 weeks to 6 months of age is equivalent to the growth seen in children from a yery young age to maturity. Additionally, the histology of the pig aorta is very similar to that of human beings. For these two reasons, we chose the piglet for our studies. In answer to Dr. Derom s question, the aortogram shown in Figure 3 is representative of all our anastomoses examined at 11 weeks and 6 months. We did not see any dilatation of anastomoses at 6 months. We did not excise a segment of aorta as you have done, and perhaps undue tension on your anastomoses resulted in the aneurysm formation you have shown. In the past, silk, which is ultimately an absorbable suture, has been used for vessel-to-vessel anastomoses without aneurysm formation. Furthermore, as we have shown in previous studies in growing vessels using polypropylene, the suture material straightens as the circumference of the anastomosis increases; with further growth, the polypropylene suture may break or pull through the vessel wall, forming a bowstring of suture material across the vessel lumen. At this point, that portion of the aortic wall no longer has any suture material holding it together and is maintained only by the natural healing process. Erratum The Annals of Thoracic Surgery regrets that an error appeared in a recent article by Dr. Mike1 D. Smith and colleagues (Surgical Correction of Congenital Tricuspid Stenosis, Ann Thorac Surg 34:329, 1982). On page 330, the frames for Figure 1 have been transposed; thus, 1A should be lb, and vice versa.