Initial evaluation of absorbable polydioxanone suture for peripheral vascular surgery

Initial evaluation of absorbable polydioxanone suture for peripheral vascular surgery

Initial evaluation of absorbable polydioxanone suture for peripheral vascular surgery Steven W. Merrell, MD, and Peter F. Lawrence, MD, Salt Lake City...

777KB Sizes 13 Downloads 58 Views

Initial evaluation of absorbable polydioxanone suture for peripheral vascular surgery Steven W. Merrell, MD, and Peter F. Lawrence, MD, Salt Lake City, Utah The long-term integrity of an autogenous vascular anastomosis is primarily dependent on the strength of tissue healing; therefore permanent mechanical support of an autogenous anastomosis with sutures is unnecessary. In this study we evaluated monofilament absorbable polydioxanone as an alternative to polypropylene for suturing autogenous vascular tissue during adult peripheral vascular operations. We used polydioxanone suture for 21 vascular procedures in 20 patients. We evaluated suture handling characteristics during operation and then followed patients with clinical assessments and serial duplex scans to monitor for pseudoaneurysms, anastomotic narrowing, and vessel patency. Indications for surgery included limb salvage (67%), dialysis access (23%), traumatic arteriovenous fistula and claudication (5% each). The operative procedures included infrainguinal bypass (57%), arteriovenous fistula formation (24%), thromboembolectomy (14%), and arteriovenous fistula repair (5%). Polydioxanone sutures were placed in 39 separate test sites (35 arterial, 4 venous). No deaths occurred during operation. Polydioxanone suture was found to have handling characteristics similar to polypropylene. During mean patient follow-up of 7.2 - 0.6 months, we found no cases of anastomotic narrowing or pseudoaneurysms. Actuarial test site patency at 1, 3, and 6 months was 97%, 97%, and 86%, respectively. Polydioxanone suture has handling properties that are

acceptable for use in vascular applications, and it provides adequate mechanical support for sutured vessels to heal. A randomized trial comparing polydioxanone with polypropylene suture will be necessary to determine whether the lack of permanent foreign material in vascular anastomoses can improve long-term patency. (J VASe SvxG 1991;14:452-9.)

Intimal hyperplasia is widely recognized as a significant factor in the late failure of vascular reconstructive procedures. A variety of pathogenetic mechanisms have been proposed, including operative trauma, mural ischemia, compliance mismatch, anastomotic flow disturbances, and platelet deposition? Steen et al. 2 showed that permanent nonabsorbable suture materials incite foreign body reactions in vascular tissue, which may be associated with the formation of intimal hyperplasia and anastomotic narrowing. Since primary healing occurs when an anastomosis is created with autogenous tissues, permanent suture support is unnecessary? These considerations prompted the development of laserFrom the Department of Surgery, the University of Utah School of Medicine and Veterans Administration Medical Center, Salt Lake City. Presented at the Sixth Annual Meeting of the Western Vascular Society, Palm Springs, Calif., Jan. 13-16, I991. Reprint requests: Peter F. Lawrence,MD, Department of Surgery, Universityof Utah School of Medicine, 50 N. Medical Dr., Salt Lake City, UT 84132. 24/6/31838 452

assisted vascular anastomoses (LAVA), in which the amount of permanent suture is minimized and the foreign body reaction in the anastomosis is reduced. 4 The use of biodegradable suture is an attractive alternative to LAVA, since a surgeon may use standard operative techniques and still completely eliminate the chronic effects of a permanent foreign body in a vascular anastomosis. Polydioxanone (Ethicon, Inc., Somerville, N.J.) is a monofilament suture that dissolves by nonenzymatic hydrolysis and absorbs more slowly than multifilament absorbable sutures? Polydioxanone has been studied extensively in animal models of vascular surgery, 2"6u and has been used clinically in pediatric cardiovascular surgery for several years. 12-2o However, the published experience with polydioxanone use in adult cardiovascular surgery has been limited to a few small European series and the use of polydioxanone support sutures for LAVA in angloaccess procedures. 212s This report represents the first domestic series in which polydioxanone suture has been used for adult peripheral vascular surgery.

Volume 14 Number 4 October 1991

METHODS From March through December 1990, patients undergoing autogenous vascular reconstructive procedures of the extremities, lateral repair of arteriotomies or venotomies, and native vessel angioaccess procedures were recruited for this study. We excluded patients who required prosthetic graft material for their vascular procedures, those with severe malnutrition, patients receiving corticosteroids or chemotherapeutic agents, and women of childbearing age. All patients with symptomatic occlusive disease were tested in our noninvasive vascular laboratory with segmental limb pressures, ankle/arm indexes, and pulse volume recordings. Angiography was performed routinely in patients with occlusive peripheral vascular disease, and the decision to proceed with surgery was based on the combined history, physical examination, noninvasive laboratory studies, and angiography. This study was approved by the Institutional Review Board (IRB) of both participating hospitals before enrolling patients. Informed consent was obtained from all patients by use of IRB-approved consent forms and procedures. Data collection and analysis Postoperative data collection included clinical assessment of vessel patency and noninvasive vascular lab studies (segmental limb pressures, anlde/brachial indexes, pulse volume recordings, and serial duplex examinations) in the first, third, and sixth months after operation. Annual follow-up was scheduled thereafter. Duplex scans were also used to evaluate the suture test sites for anastomotic size and pseudoaneurysm formation. Each suture test site was evaluated independently for the purpose of data analysis, and cumulative patency was calculated with use of life-table methods. Data are reported as mean + standard error of the mean, except where noted otherwise. RESULTS The study population included 20 patients who underwent a total of 21 operative procedures (Table I). Polydioxanone sutures were placed in 39 separate test sites, including 35 arterial and 4 venous locations. We used 6-0 suture most frequently (64%), followed by 5-0 (26%), 4-0 and 7-0 sutures (5% each). Three of the four femoropopliteal bypass procedures were performed to below-knee segments. Through a mean follow-up period of 7.2 --- 0.6 months (range, 0.2 to 9.4 months), 95% of all study patients had at least one postoperative duplex scan of their test sites, and images were obtained on 80% at

Evaluation of absorbable suture

453

Table I. Summary of surgical indications and operative procedures Surgical indications Limb salvage Claudication Traumatic arteriovenous fistula Angioaccess for hemodialysis Total Operative procedures Femorotibial bypass Femoropopliteal bypass Thromboernbolectomy Repair of traumatic arteriovenous fistula Native arteriovenous fistula Total

14 (67%) 1 (5%) 1 (5%) 5 (23%) 21 (100%) 8 (38%) 4 (19%) 3 (14%) 1 (5%) 5 (24%) 21 (100%)

least 6 months after operation. All test sites were patent at the time of discharge from the hospital, but three patients sustained late vessel occlusions. Eightyfive percent of all suture test sites were patent at the time of last follow-up, and no pseudoaneurysms were identified. The cmnulative patency of all test sites was 97%, 97%, and 86% at 1, 3, and 6 months, respectively (Fig. 1). One artetiovenous fistula failed because of inadequate venous outflow. At the time of thrombosis, a duplex scan showed a widely patent anastomosis with venous occlusion several centimeters distal to the test site. Another patient underwent thrombectomy for acute arterial ischemia through four separate arteriotomies in the superficial femoral, popliteal, and tibioperoneal mink arteries, and was known to have poor distal runoff. He sustained recurrent thrombosis 8 months after operation, at which time he underwent a successful distal bypass procedure. The third patient underwent a popliteal-posterior tibial bypass with poor distal nmoff vessels. His graft thrombosed 3 weeks later. Duplex scanning showed persistent blood flow through a patent proximal anastomosis, with occlusion in the midportion of the graft. One additional patient developed an arterial "steal" after construction of a proximal arteriovenous fistula. He required elective ligation of the fistula, but the anastomosis was widely patent at the time of his last duplex scan. No other major complications have been observed in this series, and no patients died. The handling characteristics of polydioxanone suture were evaluated by one of the authors in every case (Table II). The major criticism of the suture was the tendency to coil and drag when the suture dried, which was easily corrected by wetting the suture. Other characteristics of a suture that are important to surgeons (strength, knot security, and visibility)

454

Journal of VASCULAR SURGERY

AIerrell and Lawrence

03

1.0

36 I

33

26

03 o3

0.8

~ tO

0.6

° ~

0 Q-

0.4

(1~ "= m

0.2

E o

I

I

I

I

I

I

I

I

I

I

1

2

3

4

5

6

7

8

9

10

0

Months Fig. 1. Life-table analysis of the cumulative proportion of patent test sites during the postoperative period. The numbers shown on the graph represent the absolute number of test sites that are being followed in each interval.

Table II. Intraoperative assessment of suture handling characteristics Memory (stiffness) Tissue drag Fraying Visibility Knot security Strength Overall handling

None~minimal

Moderate

Severe

68% 78% 97%

27% 22% 3%

5% 0% 0%

where uniformly rated "good," and were judged to be comparable to polypropylene.

DISCUSSION The clinical use of absorbable sutures in vascular procedures was reported as early as 1952, 26"27 but previously available absorbable sutures were suboptimal for this purpose. Plain and chronic catgut sutures lose most of their tensile strength in the first 1 to 3 weeks after placement and cause severe inflammatory reaction in tissue. Polyglycolic acid (Dexon, Davis + Geck, American Cyanamid Co., Danbury, Corm.) and polyglactin 910 (Vicryl, Ethicon, Inc.) also absorb rapidly, with near complete loss of tensile strength in the first 4 weeks. 28 Consequently, the duration of mechanical support provided by these sutures may be inadequate to allow sufficient healing of vascular tissue. In addition, the braided

Good

Fair

Poor

100% 100% 100% 89%

0% 0% 0% 11%

0% 0% 0% 0%

configuration of polyglycolic acid and polyglactin 910 may cause undesirable trauma to the vessel wall during placement and increase suture affinity for bacteria. 29 Finally, a strong bias against the use of absorbable suture has been present since these sutures were implicated as a cause of pseudoaneurysms when used with prosthetic graft material. 3° Polydioxanone is an extruded polymer of paradioxanone. ~ Since the polymer is much more pliable than polyglycolic acid or polyglactin 910, it is practical for use as a monofilament suture. In this study we used a newer type of polydioxanone suture (PDS-II, Ethicon, Inc.), which is even less stiffthan the original version. The initial strength of polydioxanone is 1.1 to 1.6 times greater than polypropylene, and it absorbs more slowly than polyglycolic acid and polyglactin 910 via nonenzymatic hydrolysis. 7'1s'28 Polydioxanone retains 58% of its original strength at

Volume 14 Number 4 October 1991

4 weeks and 40% of its strength at 6 weeks, which is still approximately 66% of the strength of polypropylene. 28 Fragments of suture material may be visible microscopically for several months, but absorption is usually complete by 180 days. s Data regarding the healing characteristics of vascular anastomoses suggest that the tensile strength rapidly accumulates in the!first 3 weeks, similar to other sutured tissues. 3134 Therefore the extended absorption kinetics of polydioxanone appeared to provide an adequate safety margin for autogenous vascular reconstructive procedures. 3 Since its introduction, polydioxanone suture has been tested in several vascular animal models, 6-n and it is now used routinely in many pediatric cardiovascular procedures. ~22° The adult experience with polydioxanone in cardiovascular procedures has been limited to a few foreign S t u d i e s . 21-23 To date, no anastomotic aneurysms have been reported after polydioxanone suture use in the laboratory or in clinical cases. The current study was initiated to evaluate polydioxanone as an alternative to LAVA, which we were investigating in the laboratory. 85 It was our impression that LAVA was hampered by technical difficulties in medium-sized vessels, particularly when hooded end-to-side anastomoses were attempted. In addition, the LAVA technique requires specialized training and expensive equipment. The use of absorbable suture is an attractive alternative, since standard surgical techniques can be used to obtain the goal of a completely sutureless anastomosis after suture dissolution. We designed the current protocol to include vascular sites in extremities only, to facilitate clinical assessment and noninvasive duplex scan monitoring of suture test sites. No suture-related complications were identified in follow-up ranging up to 9.4 months. Data regarding suture dissolution indicate that the residual tensile strength ofpolydioxanone suture is negligible beyond 1 to 3 m o n t h s Y '~8 Since the strength of healing vascular tissues progressively increases over time, the maximum risk for development of an anastomotic aneurysm would be in the interval immediately after the suture's loss of tensile strength (i.e., 3 to 6 months). Eighty percent of our patients were monitored for at least 6 months, and no pseudoaneurysms were detected. However, we still consider the results of this study to be preliminary, and we plan ongoing follow-up of all patients indefinitely, as we do routinely for graft surveillance. A detailed analysis of the three cases of vessel occlusion indicated that the suture was probably not related to the vessel thrombosis. All of these patients

Evaluation of absorbable suture

455

were known to have poor distal runoff before the original procedure, and failure of the grafts was not unexpected. In addition, two of the three patients underwent immediate duplex scanning at the time of graft closure, which showed that the thrombosis originated at a site distal to the anastomosis. Both scans showed persistent blood flow through the anastomotic site, with no evidence of adjacent intraluminal thrombus or anastomotic stenosis. Anastomotic intimal hyperplasia is a recognized cause of late vein graft failure, and there is evidence to suggest that the mere presence of suture material may contribute to the formation of intimal hyperplasia. 2'4"36-38 Several authors have shown that polypropylene sutures incite a chronic foreign body reaction, which leads to medial thickening and is associated with intimal hyperplasia. 2'4'3638 In both experimental and clinical studies, White et al.24 have shown disordered healing of collagen and elastin fibers in vessel walls adjacent to the polypropylene aligning sutures used for LAVA. When polypropylene sutures have been compared with dissolving polydioxanone sutures, less tissue response has been identified microscopically. 2'8 It is unclear whether the presence of permanent suture material in vascular tissue impairs healing or causes intimal hyperplasia to a degree that is clinically relevant. However, even small amounts of intimal hyperplasia may have a significant detrimental impact on the patency of small caliber anastomoses, such as pedal bypasses, which are being done with increasing frequency. A randomized, prospective trial comparing the use of polydioxanone suture with polypropylene suture will be necessary to determine whether the use of absorbable suture will improve the long-term patency of small caliber arterial bypass operations or venous reconstructive procedures. Use of absorbable suture may also be well suited for vascular procedures performed in contaminated sites, to eliminate permanent foreign material that may perpetuate the infection locally. One study showed that polydioxanone suture may reduce the risk of hemorrhage and pseudoaneurysm formation in the presence of bacterial contamination, 38 and another study showed that polydioxanone has the lowest affinity for bacteria when compared to other commercially available Sutures. 29 The kinetics of polydioxanone suture dissolution have not been well defined in the presence of active infection, but the above studies may justify additional investigation of the role of polydioxanone suture for the in situ management of vascular trauma and graft infections. In summary, we found that polydioxanone suture

456

Journal of VASCULAR SURGERY

Merrell and Lawrence

performed well in vascular reconstructive operations from a technical standpoint. Through the follow-up period, no cases of suture-related complications have been identified. Our data, along with the published experimental and clinical experience of others, indicate that polydioxanone is safe for use in autogenous vascular reconstructions. In contrast, it is clear that the anastomotic strength of prosthetic vascular grafts is permanently dependent on suture support? ° Therefore polydioxanone suture should not be used in any anastomosis involving prosthetic graft material. Potential applications of polydioxanone suture for vascular surgery include autogenous arterial bypass procedures, endarterectomy closures, tacking sunares of endarterectomy end points, in situ management of graft infections, and vascular trauma.

13.

14.

15.

16.

17.

18.

The authors thank Joyce Brown, RN, and Dana Kobey, RN, for their assistance with data collection, John J. Sullivan, PhD, for statistical consultation, and Beckie Bos and Angela M. Richey for manuscript preparation. REFERENCES 1. Chervu A. Myo-intimal hyperplasia. Semin Vasc Surg 1990; 3:21-8. 2. Steen S, Andersson G, Lowenhielm P, et al. Comparison between absorbable and nonabsorbable, monofilament sutures for end-to-end arterial anastomoses in growing pigs. Surgery 1984;95:202-7. 3. Koss G, Constantinos P, Long F, Kusaba A, Perlman M, Matsumoto T. Absorbable suture materials for vascular anastomoses. Tensile strength and axial pressure studies using polyglycolic acid sutures. Am Surg 1981;47:541-7. 4. White RA, Kopchok G, Donayre C, et al. Argon laser-welded arteriovenous anastomoses. J VASC SURG 1987;6:447-53. 5. Ray JA, Doddi N, Regula D, Williams JA, Melveger A. Polydioxanone (PDS), A novel monnfilament synthetic absorbable suture. Surg Gynecol Obstet 1981;153:497-507. 6. Myers JL, Pae WE, Waldhausen JA, Pierce WS. Vascular anastomoses in growing vessels: comparison of absorbable polydioxanone and nonabsorbable polypropylene monofilament suture materials. Surg Forum 1981;32:339-41. 7. Myers JL, Waldhausen JA, Pae WE, et al. Vascular anastomoses in growing vessels. The use of absorbable sutures. Ann Thorac Surg 1982;34:529-37. 8. Schmitz-Rixen T. Animal experiments in vascular surgery with a new single strand absorbable suture material. Langenbecks Arch Chir 1984;362:53-60. 9. Sato A, Kubo Y, Sasajima T, et al. Role of suture materials in small vascular anastomosis. Comparison of absorbable and nonabsorbable suture materials [Abstract]. Artif Organs 1985;9:322. 10. Verschuere, Francois AK, DeRoose J, et al. Polydioxanone suture material in growing vascular anastomoses. Experimental study. J Thorac Cardiovasc Surg 1985;90:765-70. 11. Friberg LG, Mellgren GW, Erikson BO, Bjorkerud S. Subclavian flap angioplasty with absorbable suture polydioxanone (PDS). Scand J Thorac Cardiovasc Surg 1987;21: 9-14. 12. Castaneda AR, Norwood WI, Jonas RA, et al. Transposition

19.

20.

21.

22.

23. 24.

25.

26.

27.

28.

29.

30. 31.

of the great arteries and intact ventricular septum: anatomical repair in the neonate. Ann Thorac Surg 1984;38:438-43. Campbell DB, Waldhausen JA, Fierce WS, et al. Should elective repair of coarctation of the aorta be done in infancy? J Thorac Cardiovasc Surg 1984;88:929-38. Oh DA, Eren EE, Huhta JC, Gutgesell HP. Surgical treatment for the type II and HI truncus: complete division of the mmcal root with primary repair using absorbable suture. Ann Thorac Surg 1985;40:201-4. Campbell DB, Pae WE, Waldhausen JA. Coarctation of the aorta: current surgical management. World J Surg 1985;9: 543-9. Myers JL, Campbell DB, Waldhausen JA. The use of absorbable monofilament polydioxanone suture in pediatric cardiovascular operations. J Thorac Cardiovasc Surg 1986; 92:771-5. Ziemer G, Jonas RA, Ferry SB, et al. Surgery for coarctation of the aorta in the neonate. Circulation 1986;74(suppl I):1-25-31. Mavroudis C. Anatomical repair of transposition of the great arteries with intact ventricular sepmm in the neonate: guidelines to avoid complications. Ann Thorac Surg 1987; 43:495-501. Mellgren G, Friberg LG, Eriksson BO, et al. Neonatal surgery for coarctation of the aorta. The Gothenburg experience. Scand J Thorac Cardiovasc Surg 1987;21:193-7. Sato A, Kubo Y, Sasajima M, et al. Small diameter vascular anastomoses. The use of monofilament absorbable suture materials polydioxanone suture (PDS) [Abstract]. Artif Organs 1988;12:461. Tuchmann A, Dinstl K. Polydioxanone in vascular surgery. First experience with an absorbable suture in femoro-popliteal bypass. J Cardiovasc Surg 1984;25:225-9. Harjola PT, Ala-Kulju K, Heikkinen L. Polydioxanone in cardiovascular surgery. Thorac Cardiovasc Surg 1984;32: 100-1. Ala-Kul)u K, Verkkaia K, Ketonen F, et ai. Polydioxanone in coronary vascular surgery. J Cardiovasc Surg 1989;30:754-6. White RA, White GH, Fuiitani RM, et al. Initial evaluation of argon laser-assisted vascular anastomoses. J VASE SURa 1989;9:542-7. White RA, Kopchok GE, Vlasak J, ct al. Experimental and early clinical evaluation of vascular anastomoses with argon laser fusion and the use of absorbable guy sutures: a preliminary report. ] VASE SuRa 1990;12:401-8. Deterling RA, Coleman CC, Kee J, Humphreys GH. An experimental evaluation of catgut as a vascular suture material and a report on its clinical use. J Thorac Surg 1952;23:30325. Ku T, Huang T, Howng S. The clinical use of synthetic absorbable suture in vascular surgery. J Forsmosan Med Assoc 1974;73:45-52. Chiu I, Hung C, Chao S, Huang S, How S. Growth of the aortic anastomosis in pigs. Comparison of continuous absorbable suture with nonabsorbable suture. J Thorac Cardiovasc Surg 1988;95:112-8. Chu C, Williams DF. Effect of physical configuration and chemical structure of suture materials on bacterial adhesion. A possible link to wound infection. Am J Surg 1984;147:197204. Moore WS, Hall AD. Late suture failure in the pathogenesis of anastomotic false aneurysms. Ann Surg 1970;172:1064-8. Lowenburg RI, Shumacker HB. Experimental studies in vascular repair. 11. Strength of arteries repaired by end-to-end

Volume 14 Number 4 October 1991

32. 33. 34.

35. 36.

suture, with some notes on growth of anastomosis in young animals. Arch Surg 1949;59:74-83. Gaylis H, Corvese WI', Linton RP, Shaw RS. The rate of healing of arterial autografts. Surgery 1959;45:41-58. Watts DR, Carr SH, Hohf RP. Poly(glycolic acid) sutures in canine vascular anastomoses. J Biomed Mater Res 1976;10: 867-77. Lee S, Fung YC, Matsuda M, Xue H, Schneider D, Han K. The development of mechanical strength of surgically anastomosed arteries sutured with Dexon. J Biomechanics 1985; 18:81-9. Lawrence PF, Li K, Merrell SW, Goodman G. A comparison of absorbable suture and argon laser welding for lateral repair of arteries. J VAsc SURG 1991;14:183-9. Mallon WJ, Joyner WH, Seaber AV, Urbaniak MD. Vascular

Evaluation of absorbable suture

457

response to absorbable sutures in immature arteries. Surg Forum 1984;35:527-9. 37. Vincent HF, Hatem JM, Upshur J, Sade RM. Atrial wound healing with polyglycolic acid and polypropylcne sutures. J Thorac Cardiovasc Surg 1983;86:150-3. 38. Chad C. Wang S. How S. Growth of the aortic anastomosis in pigs: comparison of continuous absorbable suture with nonabsorbable suture. J Thorac Cardiovasc Surg 1988;95: 112-8. 39. Torsello GB, Sandmann W. Lenz W, Rosin H. Experimental studies with absorbable and nonabsorbable sutures in infected canine arterial anastomoses. J VASCSURG 1986;3:135-9. Submitted Feb. 2, 1991; accepted June 17, 1991.

DISCUSSION

Glenn L. Kelly (Englewood, Colo.). Drs. Merrell and Lawrence are to be commended on another original study and for bringing the clinical testing of polydioxanone in adults to America. To my knowledge this is only the second study ofpolydioxanone sutures in adult peripheral vascular surgery, and the first such investigation in the United States. Several features have been previously shown about the way that polydioxanone performs when it is used to create an autogenous vascular anastomosis. First, the tensile strength and longevity of the suture material are adequate to prevent vascular disruption or formation of a false aneurysm. In none of the previous series of clean anastomoses has adequate healing been a problem. This is well confirmed by Drs. Merrell and Lawrence who followed 39 test sites for up to 9.4 months with no evidence of suture-line breakdown. However, purposely excluded from their study were any patients with malnutrition, patients receiving corticosteroids or chemotherapeutic agents, or women of childbearing age. It is understandable that this "first" study might exclude such higher risk patients, but I ask the authors whether they feel that polydioxanone is up to the task when normal healing is impaired. Second, the handling characteristics of polydioxanone, although slightly different from polypropylene, have previously been shown to be acceptable. Overall handling is rated by the authors as good 89% of the time and fair only 11%. Only stiffness and tissue drag were occasionally a nuisance. On the two occasions when I have recently used polydioxanone vascular suture, I have noted the same qualities. Third, in contrast to common opinion, the performance of polydioxanone in contaminated wounds is superior to that ofpolypropylene. Fortunately, none of the wounds in this study became infected. However, work by Torsello and associates in Dfisseldorf showed that in 80 canine arterial anastomoses contaminated with StaphyIococ-

cus aureus, the incidence of vascular hemorrhage was three times greater with polypropylene than with polydioxanone. Furthermore, false aneurysms were found in five wounds, only when polypropylene was used. This may in part be related to previously known reduction of bacterial adherence to the polydioxanone suture material. The fact that when hemorrhage did occur it was an early complication, suggests factors other than chronic localized tissue response to the suture material. Fourth, several former studies indicate that use of polydioxanone suture for anastomosis in growing children or animals can avoid the pursestring effect, usually seen when nonabsorbable suture is used. This is even true when a running suture technique is used. There is one overriding theoretic advantage of a dissolvable vascular suture. This is the possibility that such a suture can significantly improve patency by reducing intimal hyperplasia at the anastomosis. Merrell and Lawrence have just mentioned several of the mechanisms postulated to cause intimal hyperplasia. One of these factors is suture-induced foreign body reaction. This can best be determined by microscopic examination of harvested suture anastomoses. Unfortunately there are no histologic specimens to confirm the efficacy of polydioxanone in the authors' study. Previous studies by others, however, suggest that polydioxanone does invoke at least a moderate tissue reaction, and that fragments of the suture material sometimes linger beyond 6 months. Finally, the patency rate for this series is excellent; however, only 14 of their anastomoses really involve small-caliber vessels. This small number of cases along with a relatively short mean follow-up time of 7.4 months needs further investigation. I am pleased to note that they suggest the need for a randomized, prospective trial comparing the use of polydioxanone and polypropylene, looking at long-term patency. Will they be undertaking such a study? Are the authors aware of any experience with the use of

458

Afferrell and Lawrence

polydioxanone for even more demanding microvascular anastomoses ? I think this study is an imaginative extension of the authors' previous work with laser-assisted anastomosis, and I look forward to seeing fiarther data to support their tenable thesis. Dr. Stevcn W. Merrell. Dr. Kelly, thank you for your constructive comments and thoughtful questions. Your first question pertained to the exclusions of certain categories of patients. I should stress that the use of polydioxanone is not currently approved by the FDA for adult vascular surgery. Polydioxanone may only be used in an IRB study or by obtaining an investigative device exception (IDE) number. Because of the preliminary nature of the experience with polydioxanonc, we feel that studies should continue to exclude patients who arc at risk for poor healing, at least until more data are available regarding long-term results. I should also clarify that the sutures used in this study were the new type PDS II, which is much more pliable than the original type of polydioxanone suture that was introduced in the early 1980s. The handling of PDS II is clearly better suited for use in vascular surgery. We are aware of the literature that shows that the risk of pseudoaneurysms is higher when permanent suture is used in contaminated sites. These findings may be related to the permanent presence of the suture and the inability to clear infection from tissue as long as foreign material is present. In addition, another study showed that polydioxanone suture has the lowest affinity for bacteria when tested against gram-positive and gram-negative bacteria when compared to all the other commercially available suture materials, both absorbable and nonabsorbable. These data suggest that polydioxanone suture may be uniquely well suited for use in infected sites, and may account for the diminished risk of pseudoaneurysms as observed in the study you mentioned. As far as the lack of histological information in our study, we have eneotmtered predictable difficulty obtaining consent from patients with open grafts to harvest their anastomotic sites. We will try to obtain tissue for histology if we reoperate on a site that has been previously operated on or ifa patient requires amputation. The preponderance of literature shows that the amount of inflammation associated with polydioxanone suture is less then that seen with polypropylene. These findings are likely related to the fact that suture dissolution occurs via nonezyzymatic hydrolysis. We agree that long-term follow-up will be necessary to see whether the lack of permanent suture will predispose to late pseudoaneurysm formation. These are preliminary data, but we wanted to gain experience with the use of the material to help us justify a randomized clinical trial, and we do plan ongoing investigations in that regard. Finally, we agree with your comments that indicate that polydioxanone may have a role in microsurgical anasto-

lournal of VASCULAR SURGERY

moses. If the foreign body reaction related to permanent sutures proves to be clinically relevant, the effect on patency will be greatest in small caliber vessels. Dr. Victor M. Bernhard (Tucson, Ariz.). I have a comment about the use of sutures in infected sites. I agree with all that you have said, however, what happens in blowouts, and I think Dr. Kelly alluded to this, depends on how much the infection invades the wall and whether this has been effectively erradicated by debridement before arterial closure. The suture material may have nothing whatsoever to do with healing. We have done biopsies on the residual artery wall for culture after debridement to control this factor in terms of the duration of antibiotic therapy. I suggest that this may be useful for monitoring experimental studies designed to evaluate the efficacy of various suture materials in infected vascular sites. Dr. David B. Roos (Denver, Colo.). I commend the authors on this study. In evaluating your material you talked about the suture. I did not see one other important factor mentioned in the parameters of product evaluation, and that is the vehicle carrying the suture, namely the needle. We all struggled through years of frustration when the manufacturers of vascular sutures were trying to produce a needle that was satisfactory, one that was not too soft so it would not bend, and not too large that it would leave a needle hole larger than the suture causing a lot of bleeding at the suture line. The only solution I have seen to this latter problem is the Gore-Tex (W. L. Gore & Assoc., Elkton, Md.) suture where the needle is actually smaller than the suture. So my question is did you have any problems with the needles because you were not using sutures specifically designed and manufactured for vascular work since they are not yet approved for this purpose, and would you suggest suture manufacturers try to develop a specific absorbable vascular suture and needle for this use that might be easier to get FDA approval? Dr. Merrell. We did have difficulty obtaining sutures initially, but were able to procure them from affiliated pediatric hospitals, where polydioxanone suture use is routinely used by the pediatric cardiovascular surgeons. The sutures were available with RB-1, RB-2, and C-1 needles, so we were able to use the same sizes and types of needles that we normally use with polypropylene. I do not have any specific recommendations regarding suture needle design. The specific handling characteristics that we chose to use in evaluation of sutures were based on our previous experience in working with suture companies in the process of product development. The handling properties are standard parameters that the industry uses to evaluate sutures. Dr. Wiley F. Barker (Los Angeles, Calif.). As the historian for the WVS I want to take you back 100 years and perhaps offer a humorous note. Approximately 100 years ago, Heidenhein achieved the first successful or at least reported repair of an arterial

Volume 14 Number 4 October 1991

injury. And as was common in those days, the senior German professors never injured an artery themselves; it was his assistant who clumsily tore the axillary artery during a mastectomy. And as nearly as I can tell, he successfially repaired it. It was a size three, not 3-0, but size three catgut suture. Shortly after that, Jusinowsky and Doerfler had gone to the laboratory and accomplished successfifl arterial repairs in dogs using mllliner's silk. Heidenhein went to the laboratory himself, and using size three catgut suture reported that you could not do it in dogs and that he did not believe that all this nonsense that Doerfler and Jusinowsky reported with silk was worth following up at all. I think that our objective results are much better than they were 100 years ago. Dr. Merrell. In our literature review we found several reports on the use of catgut for autogenous vascular surgery extending back into the 1950s. In addition, a Taiwanese group of surgeons performed portasystemic shunt procedures and vascular trauma cases with polyglycolic acid suture (Dexon). As a result, significant precedent exists for use of absorbable sutures in vascular tissue, but the technique has just been overlooked through the years. It is noteworthy that no suture-related failures have been reported in the literature, even with sutures that lose their tensile strength much more rapidly than polydioxanone. Dr. Albert D. Hall (Greenbrae, Calif.). Will you be influenced in your selection of suture material by results of this study when you close aortic stumps after removal of infected grafts? In 1962 we first closed the aortic stump with a continuous chromic catgut followed by axillofemoral bypass. The aortic stump healed well. At that time we

Evaluation of absorbable suture

459

were concerned that braided Dacron suture may produce a potential nidus of infection. Dr. Merrell. As I said before, polydioxanone suture is not approved by the FDA, and at the present time we would like to accumulate more experience before we use polydioxanone in infected sites. In addition, I feel that available data are inadequate to predict the absorption dynamics of polydioxanone suture in the presence of infection. It is unknown whether loss of tensile strength occurs more rapidly in the presence of infection. The suture may well have a role in the closure of vessels after infected grafts are removed, but we currently use polypropylene in this setting and will continue to do so until more information is available. We agree that adequate debridement of the infected vessel wall is probably of greater importance than the choice of suture material. Dr. Ronald Stoney (San Francisco, Calif.). My comments relate to sutures in vascular repairs. Jack Wylie was interested in some compromised arterial healing problems one of which was infection. We used chronic catgut and followed those dogs during my year in the lab. We never had one blowout, and a moderate number were done in an infected setting. We also did several clinical cases at the femoral level for infected false aneurysms where autogenous patches and chromic catgut were used. The follow-up in some of those early cases showed no dehiscenses. We should remember there was extensive debridment of both the arterial wall and the adjacent host tissue. This leaves a surgically clean area, and that might have allowed a number of repairs to heal. Other sutures were used in that era as well as chromic catgut that were permanent, but they still provided the same secure healing of these anastomotic repairs.