| | Effect of twist on flow and patency of vein grafts☆☆☆Presented at the Twelfth Annual Meeting of the Midwestern Vascular Surgical Society, Rochester, Minn., Sept. 23–24, 1988. Abstract This article examines the effect of twist on flow through reversed vein segments in vitro and its effect on graft patency in vivo. Excised canine superficial femoral veins were perfused in vitro with normal saline solution or canine blood. Perfusion was carried out at five pressures and against three outflow resistances. Increasing increments of twist were applied to the outflow end of the vein. Flow was measured at each level of twist. With both saline solution and blood, flow was unaltered until twist reached 140 to 180 degrees. Flow then decreased sharply, stopping completely at 175 to 200 degrees of twist. In vivo experiments were then performed in 13 dogs. Reversed superficial femoral veins were used as end-to-end grafts to bypass the iliac arteries. Each graft was deliberately twisted 0, 45, 90, 135, or 200 degrees. All grafts were harvested 6 months after surgery. Eighteen of 20 grafts twisted 135 degrees or less remained patent. However, all five grafts twisted 200 degrees were thrombosed within 4 hours of surgery (p < 0.05). These data suggest that in patients a slight amount of graft twist probably does not reduce flow; however, more than 135 degrees of twist will greatly reduce flow, leading to early graft thrombosis. (J Vasc Surg 1989;9:651–5.)
Many factors are important for a successful outcome when reversed vein grafts are used to bypass narrowed or occluded arteries. One technical consideration is that the surgeon place vein grafts without twist. If a graft is twisted, flow may be altered possibly leading to early graft thrombosis. Although several authors warn against the twisting of grafts 1, 2 there are no published data to demonstrate the amount of twist required to alter flow or cause thrombosis. This study was undertaken (1) to examine the effect of twist on the flow of saline solution and blood through reversed vein segments in vitro and (2) to test the in vitro findings in an animal model in vivo.
Methods and material  In vitro experiments Femoral veins were obtained from fully anesthetized dogs used for other unrelated acute experiments. The dogs' hindlimbs were positioned in full extension and both femoral veins were exposed. Before excision, venous tributaries were ligated and divided, and two 6-0 polypropylene sutures were placed in the vein wall adventitia to designate the legnth of vein to be removed. The distance between these sutures was measured with the vein in situ. The veins were excised, flushed with heparinized saline solution, and cannulated at each end. These cannulas were mounted on supports in vitro to permit the vessels to be studied as horizontal cylinders (Fig. 1).
The wmounted veins were perfused with saline solution at room temperature at 150 mm Hg allowing the vessels to assume their natural, untwisted position. The vessels were lengthened to in situ length as determined by the adventitial sutures. Twist was applied to the distal end of the vein. A hairline attached to the distal cannula was used to set the desired twist against a fixed protractor. The perfusate was contained in a reservoir the height of which determined perfusion pressure. Vein segments were perfused at five pressures: 25, 50, 75, 100, and 150 mm Hg. Outflow resistance was varied with 19-, 18-, and 15-gauge hypodermic needles attached to the outflow cannula; these three resistances (high, medium, and low) were used to simulate one-, two-, and three-vessel outflow in patients. Flow was determined by measuring the volume of perfusate collected in 1 minute from the vein. Flows were determined at each perfusion pressure and each outflow resistance with twist being applied in both clockwise and counterclockwise directions. Measurements were begun with the vein in its neutral position (zero degrees of twist) and the applied twist was increased in a stepwise fashion. Flow was interrupted proximally after each flow measurement to permit the vein to decompress before the amount of twist was changed. The order in which perfusion pressures and outflow resistances were studied was randomized. Six veins were perfused with normal saline solution at room temperature. Six other veins were perfused with heparinized canine blood. The blood for the latter experiment was obtained from dogs given 10,000 units of heparin systemically. The hematocrit of the blood was measured and was adjusted to 42% by adding normal saline solution as needed. At the completion of the in vitro flow studies, radiographs were taken to provide a visual record of the twisted veins. The vessels were perfused with a dilute suspension of barium sulfate at 100 mm Hg and were again subjected to successive degrees of twist. Radiographs were taken with a Ritter model D (Ritter Co., Ashville, N.C.) dental x-ray machine with a pair of dental films placed end to end immediately beneath the vessel. In vivo experiments Thirteen mongrel dogs were preanesthetized with 25 mg/kg thiamylal sodium (Surital) and maintained under anesthesia with 0.8% halothane and 30% nitrous oxide in oxygen. The animals were positioned with the hindlimbs extended. Under sterile conditions the iliac arteries and femoral veins were exposed bilaterally. After the proximal and distal veins were ligated with 3-0 polypropylene sutures, 5 to 6 cm segments of femoral vein were excised bilaterally. These were rinsed gently with saline solution. The common iliac arteries were clamped proximally and distally, and a short segment of the artery was removed. The excised veins were anastomosed in an end-to-end fashion to serve as reversed interposition grafts. Interrupted 6-0 polypropylene sutures were used for both proximal and distal anastomoses. After the proximal anastomosis was completed a single stitch was placed at the distal end of graft to indicate the 12 o'clock position of the vein with no applied twist. A twist of 0 (no twist), 45, 90, 135, or 200 degrees was imposed with the distal stitch used as an indicator. The distal end of the vein was held in this position, and the distal anastomosis was completed with interrupted 6-0 Prolene sutures. After surgery the animals were housed in separate cages and given water ad lib and lab chow daily. Care of the dogs throughout these studies complied with the “Principles of Laboratory Animal Care” and the “Guide for the Care and Use of Laboratory Animals” (NIH publication No. 80-23, revised 1978). After 6 months the animals were reanesthetized and killed with an overdose of anesthetic while fully anesthetized. The vein grafts were excised and examined for patency. Results In vitro studies Saline perfusion The length of the six veins used was 5.40 ± 1.2 cm (mean ± SD), with a mean diameter of 6.6 ± 0.8 mm at 25 mm Hg, a pressure approximating venous pressure. Table I summarizes the mean saline flow through six veins without twist.
Data are shown for each pressure and outflow resistance. After these baseline flow values were obtained twist was applied. In the unpressurized veins the applied twist distributed itself evenly along the length of the vessel. This remained true when the vein was pressurized but was not greatly twisted. As the amount of applied twist was increased it no longer was distributed uniformly but instead collected as a kink near the proximal (inflow) catheter. Further increases in twist resulted in a sharp reduction of flow with visible migration of the kink from its proximal position to a distal location on the vein. This migration was the result of the driving force of the intraluminal fluid column on the kink and was associated with cessation of flow. Migration of the kink is demonstrated in the radiographs shown in Fig. 2.
Flow data for the six veins perfused with saline solution at each perfusion pressure and outflow resistance are summarized in Fig. 3.
Data are presented as percent of baseline flow levels in the untwisted vessels. The data points designate the mean degrees of twist associated with 25%, 50%, 75%, and 100% reduction in flow. For clarity the overlapping standard errors have been deleted. Fig. 3 shows that flow was not reduced until at least 140 degrees of twist was imposed. With slightly greater degrees of twist, flow dropped sharply. Total cessation of flow occurred at 175 degrees of twist with high-perfusion pressure (150 mm Hg) or 206 degrees with low-perfusion pressure (25 mm Hg). Blood perfusion Vein specimens perfused with heparinized blood had a mean length of 4.47 ± 0.93 cm with a mean diameter of 6.5 ± 0.6 mm at 25 mm Hg. Table II presents baseline blood flows through six veins without twist.
Because of the viscosity of blood, flows were somewhat lower than values obtained with saline solution. Observations made during these experiments confirmed the distribution of twist along the vein segments. The twist was distributed uniformly in the unpressurized vein. It formed a kink at the inflow cannula when first pressurized and flow was begun. With greater twist the kink migrated to the outflow cannula and flow stopped. Fig. 4 summarizes mean blood flow plotted against twist for the six veins perfused with blood.
The data are presented as in Fig. 3. The twist required to reduce flow with blood was remarkably similar to that observed for saline solution. Flow remained at baseline levels until 142 degrees of twist was imposed with high flow and high pressure (150 mm Hg) or 183 degrees of twist with low flow and low pressure (25 mm Hg). Blood flow stopped at 180 to 200 degrees of twist. Thus with both saline solution and blood, flow was not reduced until at least 140 degrees of twist was imposed. At that point flow decreased precipitously; with just a little more twist flow stopped. In vivo studies Table III summarizes the patency rates for 25 vein grafts 6 months after implantation as reversed interposition grafts.
These data show that one nontwisted control graft (zero degrees of twist) and one graft twisted 45 degrees occluded at 6 months. None of the graft twisted 90 or 135 degrees occluded. However, all of the grafts twisted 200 degrees occluded within 4 hours of surgery. These data were statistically significant as tested by χ 2 ( p < 0.05) and are consistent with the in vitro findings described above. Discussion The present experiments show that flow through twisted vein grafts is maintained until at least 140 degrees of twist is imposed. At that point the twist collects as a local kink, is driven by the flowing stream, and flow stops. Because the kink develops in one area and does not distribute itself evenly over the length of the vein, the present findings are applicable to vein grafts of all lengths. However, these results probably do not apply to overstretched veins or rigid synthetic grafts. Such vessels have little tissue slack to accommodate twist. One may predict that twisted, overstretched, and rigid synthetic grafts will have reduced flow at lower degrees of twist than observed here. Veins perfused under high pressure required slightly less twist to reduce flow than did those perfused under low pressure. This resulted from the distensibility characteristics of the vein wall. 3, 4 Veins perfused under low pressure are somewhat slack and therefore can accommodate the local, increased wall deformation that accompanies twisting. However, veins perfused under high pressure are maximally stretched and, when twisted, have less tissue slack to accommodate the twist. Critical levels of twist appear to be independent of flow or outflow resistances. These observations were true for both saline solution and heparinized blood. For simplicity of experimental technique the in vitro experiments used steady perfusion and did not examine the effect of pulsatile pressures. However, it is likely that pulsatile perfusion is subject to the same general findings found in the steady-pressure experiments. This is supported by the agreement of results obtained with steady—pressure in vitro experiments and with pulsatile pressures in the 6-month in vivo dog experiments. These data have clinical importance. They suggest that small degrees of twist probably do not reduce flow through vein grafts in patients in the early post-operative period. This should assure surgeons when it is discovered that vein grafts have been inadvertently subjected to a small amount of twist. With larger degrees of twist, however, flow patterns and shear stress may be altered as blood enters and exits the kinked region of a twisted graft. This may lead to turbulent flow, facilitating the development of intimal hyperplasia and graft thrombosis.
Discussion Dr. Sachinder Hans (Warren, Mich.). It seems to me that the length of the graft employed in this study is too short to make these observations valid in a clinical setting. Would there be differences between a reversed vein bypass in comparison to an in situ graft? Do you recommend that the side branches be cut and not just ligated in an in situ bypass so that if a small kink should occur the graft may adjust to it because of its longer length? Dr. Endean (closing). In answer to your first question, the veins that we used were short, 3 to 6 cm in length. However, I believe that our results do apply to the clinical setting because when a critical amount of twist occurs it collects as a discrete, focal kink. Therefore our observations apply to vein grafts of any length. I might add that longitudinal tension applied to the vein also may be important because we found that veins that were not stretched to their full length required somewhat more twist to arrest flow. In response to your second question the in situ technique whereby the vein branches are left intact may offer an advantage. The segment of vein that has not been dissected should resist twisting. Of course there still is potential for a kink to form in the region of vein that has been mobilized for the distal anastomosis. References 1.
1
Bernhard VM.
Bypass to the popliteal and infrapopliteal arteries.
In: 2nd ed.
Rutherford RB editors.
Vascular surgery. Philadelphia: WB Saunders; 1984;p. 607–619. 2.
2
Lalka SG, Bernhard VM.
Noninfectious complications in vascular surgery.
In:
Moore WS editors.
Vascular surgery: a comprehensive review. Orlando: Grune & Stratton; 1986;p. 959–997. 3.
3
Dobrin PB, Canfield TR, Moran J, Sullivan H, Pifarre R.
Coronary artery bypass: physiological basis for differences in flow with internal mammary artery and saphenous vein grafts.
J Thorac Cardiovasc Surg. 1977;74:445–454. MEDLINE 4.
4
Dobrin PB, Littooy FN, Golan J, Blakeman B, Fareed J.
Mechanical and histological changes in canine vein grafts.
J Surg Res. 1988;44:259–265. MEDLINE |
CrossRef
Maywood and Hines, Ill From the Department of Surgery, Loyola University Medical Center, Maywood, and Hines Veteran's Administration Hospital, Hines ☆ Reprint requests: Eric D. Endean, MD, Division of General Surgery, University of Kentucky Medical Center, 800 Rose St., Lexington, KY 40536. ☆☆ J Vasc Surg 1989;9:651–5 PII: S0741-5214(89)70035-5 © 1989 Society for Vascular Surgery and International Society for Cardiovascular Surgery, North American Chapter. Published by Elsevier Inc. All rights reserved. | |
|
FULL TEXT
