argon laser—welded arteriovenous anastomoses

Argon laser-welded arteriovenous anastomoses Rodney A. White , M.D., George Kopchok, B.S., Carlos Donayre , M.D., Geoffrey White, M.D., Richard Lyons, M.D. , R o y Fujitani, M.D., Stanley R. Klein, M.D., and Jouni Uit to, M.D., Ph.D., Torrance, Calif

This study compared the healing of laser-welded and sutured canine femoral arteriovenous anastomoses. Arteriovenous fistulas 2 can in length were created bilaterally in the femoral vessels of 10 dogs and were studied at 1 (n = 2), 2 (n = 2), 4 (n = 3), and 8 (n = 3) weeks. In each animal, one anastomosis (control) was closed with running 6-0 polypropylene sutures, and the contralateral anastomosis (experimental) was sealed with an argon laser (0.5 watt, 4 minutes of exposure, 1830 J/cmV1 cm length of anastomosis). At removal all experimental anastomoses were patent without hematomas, aneurysms, or luminal narrowing. Histologic examination at 4 weeks revealed that laser-welded anastomoses had less inflammatory response and almost normal collagen and elastin reorien- ration. At 8 weeks sutured anastomoses had significant intimal hyperplasia whereas laser repairs had normal luminal architecture. Tensile strength and collagen production, measured by the synthesis of hydroxyproline and the steady-state levels of type I and type HI procollagen messenger ribonucleic acids, at the anastomoses and in adjacent vein and artery specimens were similar in sutured and laser-welded repairs at 2, 4, and 8 weeks. We conclude that argon laser welding of anastomoses is an acceptable alternative to suture techniques, with the advantage of improved healing without foreign body response and possible diminished intimal hyperplasia at the anastomotic line. ( J VAsc SURG 1987;6: 447-53.)

Most of the published work on the use of lasers to weld vascular anastomoses deals with microvessels. Welding of microvessels with CO2, Nd:YAG (1.06 p-m), and argon lasers has been shown to be faster, with comparable tensile strength, and no foreign body reaction in comparison to sutured repairs, ls Frazier et al.4 demonstrated that CO2 laser- welded anastomoses of femoral arteries in growing miniswine increase normally in diameter whereas sutured control arteries have restricted growth. Con- troversy exists concerning a 5% to 10% incidence o f aneurysms that are reported to occur during the first month after COs laser repair? White et al.6 noted that vessel fusion was successful and no aneurysms occurred with the CO2 laser at 120 to 130 mW. Epstein and Cooley 7 reported that the optimal C02 laser parameters for small vessel welding are

From the Departments of Surgery (Drs. R. White, Donayre, G. White, Fujitani, Klein, and Mr. Kopchok) and Medicine (Drs. Lyons and Uitto), Harbor-UCLA Medical Center.

Presented at the Second Annual Meeting of the Western Vascular Society, Tucson, Ariz., Jan. 22-25, 1987.

Supported in part by grants HL-32622, AM-28450, and GM-28833 from the U.S. Public Health Service, National In- stitutes of Health.

Reprint requests: Rodney A. White, M.D., Harbor-UCLA Med- ical Center, 1000 West Carson St.~ Torrance, CA 90509.

80 to 120 mW, with the use of 80 to 160 pulses of 0.1 second duration, spot size of 0.2 ram, and 2 to 3 J/mm 2 energy fluence.

Preliminary experiments in our laboratory have shown that CO2, Nd:YAG (1.06 p-m), and argon lasers can all be used to seal 6 to 8 mm canine femoral and jugular veins, al° In these studies we observed that CO2 laser energy at 1 to 2 watt power (400 to 1000 J/cm 2 fluence) did not produce seals that could withstand larger diameter arterial pressure. 11 Nd: YAG (1.06 p,m) laser welds in large arteries were initially successful, but most failed within 20 to 40 minutes. In contrast, the argon laser sealed 2 cm long arteriotomies that healed within 4 to 6 weeks.12 Ash- worth et al.~3 have recently reported healing at 4-week follow-up of end-to-end laser-assisted vascular anastomoses of 4 to 5 mm diameter canine carotid arteries with the use of a milliwatt CO2 laser (150 to 175 mW).

On the basis of our preliminary success with the argon laser to seal arteriotomies in vessels of 5 to 8 mm internal diameter, we have directed our in- vestigations to performing arteriovenous anastomoses. The objective of this study was to compare the healing of sutured and argon laser-welded canine femoral arteriovenous fistulas.

447

448 White et al.

Journal of VASCULAR

SURGERY

Fig. 1. Technique of laser welding of vein-artery anastomoses. Sutures are placed at the apexes of the incisions and at the middle of the posterior wall (A); tension on the suture at the middle of the posterior wall opposes the edges of the repair for welding (B); suture is placed in the middle of the anterior wall and apposes the edges for welding (C).

Fig. 2. Gross appearance of argon laser-welded, canine arteriovenous fistula at 8 weeks. S = traction sutures; 1 = 1 cm length of laser fusion.

METHODS

Bilateral side-to-side arteriovenous fistulas 2 cm in length were created in dogs by isolating the femoral artery and vein in the upper thigh and anasto- mosing the walls o f adjacent arteriotomy and venot- omy incisions. Healing o f the repairs was studied at 1, 2, 4, and 8 weeks with one anastomosis (control)

closed with running 6-0 polypropylene sutures, and the contralateral anastomosis (experimental) welded with an argon laser (Laser Sonics model 555 A, Tri- medyne, Inc., Santa Ana, Calif.). Laser welds were fashioned with the use o f 0.5 watt, via a 0.3 mm fiberoptic delivery system held at i cm from the target spot. The spot size was 0.066 cm 2, with 7.6 W/cm 2

Volume 6 Number 5 November I987 Laser-welded arteriovenous anastomoses 449

Fig. 3. Verhoeff-van Gieson stain of arteriovenous fistulas at 8 weeks: suture (A), argon laser- sealed (B), and argon laser-sealed at the site of a traction suture (C). In each illustration the artery is on the top and the vein is on the bottom. Note that sutured areas in both the suture control (A) and lasered specimen at the site of a traction suture (C) were associated with a marked intimal response. (Arrows mark line of artery-vein fusions; IE = internal elastic laminae; I H = intimal hyperplasia; S = suture holes). (Original magnifications x 40.)

4 5 0 White et al.

Journal of VASCULAR

SURGERY

Table I. Synthesis of radioactive hydroxyproline in sutured and argon laser-welded arteriovenous fistulas

Healing study intervals (wk)

1 2 4 8

Suture Laser Suture Laser Suture Laser Suture Laser

Anastomosis 9.3 2.7 6.6 24.4 9.1 7.1 1.5 20.1 46.0 12.4 1.8 28.7 10.5 2.5 2.9 6.8

5.2 * 9.8 7.6 Vein control 8.8 8.3 8.9 5.7 10.0 25.3 1.3 1.5

15.1 28.8 8.1 11.9 8.7 10.1 5.3 10.3 6.7 3.8 2.4 10.1

Artery control 4.1 26.3 3.4 9.9 2.8 6.4 1.8 1.6 9.5 12.6 7.5 4.1 1.3 .6 1.4 5.6

2.1 12.2 7.1 2.6

NOTE: Values expressed as disintegrations per minute ( × lO t) per milligram of wet tissue. *Specimen lost in processing.

power density, and 1830 J/cm 2 fluence. Total exposure was 240 sec/1 cm length of anastomosis by means of 5-second pulses separated by 0.2 second intervals. Laser-welded arteriovenous fistulas were approximated with a single 6-0 polypropylene suture at the apexes of the incisions and by traction sutures posteriorly and anteriorly at the midportion of the back and front wall of the repairs (Fig. 1). The traction sutures were used to appose the edges of the vessels during laser fusion. Thus, laser welding of the 10 fistulas was accomplished by sealing 40, 1 cm segments, (ie, four segments per anastomosis). Seven of the 40 laser-welded segments required one or two additional interrupted sutures to close small holes that did not fuse adequately. During the laser fusion, the vessels were continuously cooled by drips of saline solution at room temperature to prevent thermal damage. 14

Control and experimental arteriovenous fistulas were removed at 1 (n = 2), 2 (n = 2), 4 (n = 3), and 8 (n = 3) weeks. Histologic examination in- cluded hematoxylin-eosin, trichrome, and Verhoeff- van Gieson stains.

Vascular tissue samples were also removed for in vitro measurements of the synthesis of radioactive hydroxyproline as an index of the rate of collagen production) s The samples were incubated with 30 ~Ci of triated proline for 6 hours at 37 ° C in a Dulbecco's minimum essential medium supple- mented with 20% of dialyzed fetal calf serum and 50 ~g/ml ascorbic acid. After incubation, tritiated hydroxyproline was assayed by a specific radiochemical method, xs The values of radioactive hydroxyproline were expressed as disintegrations per minute per milligram of wet weight of tissue. Samples for tritiated hydroxyproline analysis were taken from the arterio-

venous anastomosis site to quantitate healing, and sections were also removed 1 cm away from the repair in adjacent normal artery and vein, for comparison.

In further studies, the steady-state levels of type I and type III procollagen messenger ribonucleic acids (mRNAs) were determined in tissue samples by molecular hybridizations with specific comple- mentary deoxyribonucleic acid (cDNA) probes. 16'17 For this purpose, total RNA was isolated by cesium- chlorine density gradient-centrifugation and varying amounts were dotted on nitrocellulose filters. The filters were hybridized with the cDNA probes labeled radioactive by nick translation with s2p-labeled nu- cleotides. The mRNA-[S2P]cDNA hybrids were vi- sualized by autoradiography and quantitated by scan- ning densitometryat 700 nm. The ratios of type I and type III procollagen were calculated on the basis of densitometric units and corrected for specific ac- tivity and the length of the probes, and the com- position of the procoUagen.

Tensile strength and extensibility of specimens were quantitated by measuring the load at breaking of specimens from the arteriovenous site and on pieces of normal vein and artery by means of an instrument that traces a load-extension curve, xs A strain of 3 mm/min was applied to 2 mm wide strips of control and experimental specimens and to pieces of normal canine femoral artery.

All animals used in the study received humane care in compliance with the "Principles of Laboratory Animal Care" formulated by the National Society of Medical Research and the "Guide for the Care and Use of Laboratory Animals" prepared by the Na- tional Academy of Sciences and published by the National Institutes of Health (NIH Publication No. 80-23, revised 1978).

Volume 6 Number 5 November 1987 Laser-welded arteriovenous anastorm)ses 4 5 1

Table II. Relative abundance of type I and type III procollagen messenger RNAs in sutured and argon laser-sealed arteriovenous fistulas

Messenger RNA level (U/mg)

Tissue Wk cd (I) od (III) Ratio of .t~e 1/111

Suture anastomosis

Laser anastomosis

1 538 1264 0.58 448 None measurable - -

2 178 663 0 2 4 325 1246 0.23

4 229 213 1.47 31 9 3.20

1345 1381 1.02 8 87 157 0.58

164 245 0.67 1016 117 906

1 550 1278 0.59 360 None measurable - -

2 313 816 0.35 127 80 1.43

4 172 570 0,41 63 13 4,50

1143 1269 0.94 8 137 86 1.66

259 185 1.46 260 364 0.75

RESULTS

Laser-sealed anastomoses required approximately 20 minutes for each repair, whereas sutured anastomoses were formed in approximately 15 minutes. When removed at 1, 2, 4, and 8 weeks, all control and experimental arteriovenous fistulas were patent without hematomas, aneurysms, or luminal dilata- tion. Fig. 2 shows the appearance of a laser-welded fistula at 8 weeks.

Histologic examination of sutured wounds showed granulomatous reaction around the sutures with areas of excessive collagen accumulation and disorientation of the elastic fibers. Laser-welded specimens revealed minimal inflammatory response, almost normal collagen content, and a small gap in the elastic fiber continuity that decreased with time after repair. At 8 weeks, the three sets of anastomoses demonstrated intimal hyperplasia in the sutured repairs and no abnormal finding in the laser-welded specimens, except for intimal thickening at the site of the traction sutures (Fig. 3).

The synthesis of radioactive hydroxyproline, a measure of collagen production, was lower in laser- welded repairs at 1 week, but at 2, 4, and 8 weeks both control and experimental specimens were comparable (Table I). There was also no consistent dif- ference in the tritiated hydroxyproline synthesis in adjacent normal veins and arteries compared with specimens from the arteriovenous site.

Levels of mRNA corresponding to type I and type III procollagen at the site of sutured and laser-

sealed aoastomoses and the ratios of type I and type III procollagen mRNAs as an index of the expression of the corresponding genes are shown in Table II. There was significant variabili W among the values. No trends or differences were apparent be- tween sutured and lasered repairs, nor did the values from the sites of anastomosis differ from the values on specimens of control artery or vein.

The tensile strength of both sutured and laser- welded arteriovenous fistulas increased with time and were approximately equal to normal artery at 4 and 8 weeks (Fig. 4). The extensibility of repairs was also comparable at each interval (Table III).

DISCUSSION

Vascular tissue welding by lasers occurs at energy levels much lower than those required to coagulate or cut. Tissue fusion is performed by directing the beam at the opposed edges of the vessels. The tissue is approximated by stay sutures or nonreflective in- struments. Laser energy is passed back and forth over the anastomotic site until fusion is achieved. Appar- ent to the trained eye are vessel sealing, nonunion caused by inadequate energy delivery, and tissue co- agulation or vaporization from excessive exposure. Fiberoptic laser transmission and hand-eye coordi- nation are adequate to perform repairs in vessels greater than 2 to 3 mm in diameter. Magnification and precise mechanical control of the energy are used for microanastomoses. The laser power (measured in watts), and the amount of energy and time required

452 White et al.

Journal of VASCULAR

SURGERY

Fig. 4. Tensile strength of arteriovenous fistulas closed by suture ( ) or laser ( ..... ) welding at 1, 2, 4, and 8 weeks. Graphs depict the load (kg) vs. extension (mm) of the repairs. The top of the curve on the y-axis corresponds to the load at breaking. Control tensile strength of both normal vein and artery are represented by the respective labeled lines ( . . . . . . . ).

Table I lL Extensibility of repairs

Sacrifice interval (wk)

Specimen 1 2 4 8

Suture 55, 50 --*, 97 75, 58, 60 55, 55, 55 Laser 77, 82 97, 84 82, 55, 37 50, 52, 73

NOTE: Extensibility measured as percentage ( [AL/initial length] × 100). Extensibility of control normal artery was 72%; that of control normal vein was 48%. *Infected specimen.

(energy fluence or power density) vary for the type of laser and for the size of the vessels. Although laser repairs can be fashioned in time intervals comparable to or slightly longer than those required for suture repairs, the optimal wavelengths and laser parameters for different types of seals are not yet established.

In this study we found that argon laser-welded arteriovenous fistulas heal comparably to sutured repairs in early examination of up to 8 weeks. Laser repairs have minimal inflammatory reaction and more normal-appearing collagen and elastin fiber ori- entation compared with sutured wounds. The observation that sutured repairs at 8 weeks had significant intimal hyperplasia, whereas laser-sealed anastomoses had no abnormal findings except for hyperplasia at the sites where the traction sutures remained implicates the sutures in the development of intimal lesions and suggests that argon laser- sealed arteriovenous anastomosis delays or inhibits intimal hyperplastic response.

Quigley et al.19 found that intimal hyperplasia was less in CO2 laser-assisted end-to-end microvascular anastomoses of rat femoral arteries compared with sutured controls at 2 weeks, but equal amounts of hyperplasia occurred when compared with sutured

anastomoses at 6 weeks. These authors postulated that the intimal response at 2 weeks was inhibited by medial injury caused by the CO2 laser and that by 6 weeks the vessels had overcome this inhibition. Recent reports of a 5% to 10% incidence of aneurysms in the first month s and intimal thickening at 1 year a° in COa laser-sealed microvessels have associated these phenomena with a break in the elastic lamina at the site of the COa laser fusion. The break in the elastic lamina of the CO2 repairs is likely to be due to tissue necrosis produced by the 80 ° to 120 ° C temperatures generated during the fusion. 2~ Recently, Ashworth et al.13 have reported healing with no aneurysms at 4-week follow-up of end-to- end laser-assisted vascular anastomoses of canine carotid arteries by means o fa miUiwatt CO2 laser (150 to 175 mW and 2400 to 3550 J/cm 2 energy fluence). They attributed the absence of aneurysms in large vessel seals made with the CO2 laser compared with the high incidence of aneurysms reported in microvessels, to minimal vessel wall thermal damage seen in the large artery repairs.

Recent studies in our laboratory have demonstrated that argon laser welds formed while contin- uous saline irrigation is performed limits the maximal

Volume 6 Number 5 Novcmber 1987 Laser-welded arteriovenous anastomoses 453

temperature at the site of fusion to 44.2 ° + 1.6 ° C (mean _ standard deviation) and results in welds that heal rapidly with minimal or no gap in the elastic lamina. 14,22 This technique of low-temperature argon welding prevents thermal tissue necrosis at the site of vessel fusion and may explain both the lack of aneurysms and the decreased intimal hyperplasia observed in the laser welds in this study. Obviously, the implications of this observation in longer term repairs or on other types of vein-artery anastomoses, such as lower extremity bypasses, await further investi- gation.

The tensile strength of both sutured and laser- welded specimens was essentially equivalent from 2 to 8 weeks. Although laser-welded and sutured wounds had low tensile strengths at 1 week, there was no evidence of wound dehiscence or aneurysms. At 3 and 4 weeks, the tensile strength of both control and experimental repairs was comparable and ap- proached those of normal artery. The lower collagen synthesis observed in laser-welded vessels at 1 week correlates with the lower tensile strength observed at the same time point. This finding may reflect increased collagen synthesis in sutured wounds related to foreign body response to the suture material, or to an inhibitory effect on collagen synthesis caused by the laser radiation. 23

On the basis o f our previous observations and the results of this study, we conclude that argon laser welding of arteriovenous fistulas of 4 to 8 mm internal diameter veins and arteries have several potential advantages over conventional suture techniques. The benefits of argon laser fusion of vascular tissue include the absence of the foreign body response related to sutures and rapid wound healing without aneurysms or excess tissue proliferation. Further work is needed to establish the potential benefits of laser welding of vein-artery anastomoses and to de- lineate the optimal laser parameters and wavelength required to produce vascular tissue fusion. The response of both normal and diseased human vessels is also unknown.

REFERENCES

1. Dew DK, Serbent R, Hart WS, Boynton GC, Byren JD, Evans JG. Laser assisted microsurgical vessel anastomosis techniques: the use of argon and COs lasers (Abstract). Lasers Surg Med 1983;3:135.

2. Serure A, Withers EH, Thomsen S, Morris J. Comparison of carbon dioxide laser-assisted microvascular anastomosis and conventional microvascular sutured anastomosis. Surg Forum 1984;34:634-6.

3. Jain KK. Sutureless microvascular extra-intracranial anastomoses with Nd:YAG laser (Abstract). Lasers Surg Med 1984;3:311-2.

4. Frazier OH, Painvin GA, Morris JM, Thomsen S, Neblett CR. Laser-assisted microvascular anastomoses: angiographic and anatomopathologic studies on growing microvascular anastomoses: preliminary report. Surgery 1985;97:585-90.

5. McCarthy WJ, Hartz RS~ Yao JST, Sottiurai VS, Kwaan HC, Michaelis LL. Vascular anastomoses with laser energy. J VASe SURG 1986;3:32-41.

6. White JV, Dalsing MC, Yao JST, Stewart GJ, Comerota AJ. Tissue fusion effects of the CO2 laser. Surg Forum 1985; 36:455-7.

7. Epstein M, Cooley BC. Electron microscopic study of laser dosimetry for microvascular tissue welding (Abstract). Lasers Surg Med 1986;6:202.

8. White RA, Abergel RP, Klein SR, Kopchok G, Dwyer RM, Uitto J. Laser welding of venotomies. Arch Surg 1986; 121:905-7.

9. White RA, Abergel RP, Lyons R, et al. Laser welding: an alternative method of venous repair. J Surg Res 1986;41: 260-3.

10. White RA, Kopchok G, Donayre C, et al. Large vessel sealing with the argon laser. Lasers Surg Med 1987;7:229-35.

11. White RA, Abergel RP, Lyons R, et al. Biological effects of laser welding on vascular healing. Lasers Surg Med 1986; 6:137-41.

12. White RA, Kopchok G, Donayre C, et al. Comparison of laser welded and sutured arteriotomies. Arch Surg 1986; 121:1133-5.

13. Ashworth EM, Dalsing M, Olson J, Baughman S, Reilly K, Glover J. Laser assisted vascular anastomoses of larger arteries. Lasers Surg Med (In press.)

14. Kopchok G, Grundfest WS, White RA~ et al. Argon laser vascular welding: the thermal component. Proc Soc Photo- optical Instr Eng 1986;712:260-3.

15. Juva K, Prockop DJ. Modified procedure for the assav of 3H and ~C labeled hydro~proline. Anal Biochem 1966;15: 77-86.

16. Thomas PS. Hybridization of denatured RNA and small DNA fragments transferred to nitrocellulose. Proc Natl Acad Sci USA 1980;77:5201-6.

17. Uitto J, Perejda AJ, Abergel RP, Chu M, Ramirez F. Altered steady-state ratio of type I/III procollagen mRNAs correlates with selectively increased type I procoUagen biosynthesis in cultured keloid fibroblasts. Proc Nail Acad Sci USA 1985; 82:5935-9.

18. Coulson WF, Carnes WH. Cardiovascular studies of copper- deficient swine. II. Mechanical properties of the aorta. Lab Invest 1962;11:1316-21.

19. Quigley MR, Bailes JE, Kwaan HC, Cerullo LJ, Block S. Comparison of myointimal hyperplasia in laser-assisted and suture anastomosed arteries. J VASC SURG 1986;4:217-9.

20. Quigley MR, Bailes JE, Kwaan HC, Heiferman K, Cerullo LJ. Microvascular laser-assisted anastomosis: results at one year (Abstract). Lasers Surg Med 1986;6:179.

21. Badeau AF, Lee CE, Morris JR, Thompson S, Malk EG, Welch AJ. Temperature response during microvascular anastomosis using milliwatt CO2 laser (Abstract). Lasers Surg Med 1986;6:179.

22. White RA, Kopchok G, Peng SK, et al. Laser vascular welding--How does it work? Ann Vasc Surg 1987; 1:461-4.

23. Abergel RP, Meeker CA, Lam TS, Dwyer RM, Lesavoy MA, Uitto J. Control of connective tissue metabolism by lasers: recent developments and future prospects. J Am Acad Der- matol 1984; 11:1142-50.

argon laser—welded arteriovenous anastomoses

Documents