| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1995;110:427-435
© 1995 Mosby, Inc.
SURGERY FOR CONGENITAL HEART DISEASE |
Minneapolis, Minn.
From the Department of Surgery, University of Minnesota, Minneapolis, Minn.
Received for publication Sept. 15, 1994. Accepted for publication Dec. 30, 1994. Address for reprints: J. Ernesto Molina, MD, University of Minnesota, Box 182, 420 Delaware St., SE, Minneapolis, MN 55455.
Abstract
Our goal was to identify the most appropriate material for right ventricle-pulmonary artery conduits in growing animals. We used 100 lambs that were 3 to 4 weeks old (mean weight 11.7 kg). Follow-up was up to 24 months. Group I received plain tubular conduits: (1) Dacron knitted fabric, (2) collagen-coated knitted fabric, (3) Milliknit and Microknit material, (4) woven Dacron fabric, (5) three-dimensional Dacron fabric (crossweave 500 and 800), or (6) polytetrafluoroethylene. Group II received either a (1) woven Dacron fabric conduit with a built-in tissue valve or (2) polytetrafluoroethylene graft with a built-in St. Jude Medical valve. We did angiograms and catheterizations every 3 to 6 months and killed the lambs at 6, 12, 18, or 24 months. Tubular Dacron fabric woven or knitted grafts, regardless of matrix, pore size, thickness, or coating, caused formation of a thick acellular pseudointima buildup, which led to progressive obstruction starting as early as 3 months. Polytetrafluoroethylene grafts in groups I and II showed the formation of thin inner and outer capsules (0.5 mm) and none developed obstruction despite wall calcification. Conduits of woven Dacron fabric with a built-in tissue valve degenerated rapidly, leading to calcification thrombosis and obstruction within 3 months; no lamb survived 12 months. Polytetrafluoroethylene conduits with a St. Jude Medical valve in lambs receiving anticoagulants remained free of obstruction and continued to function well. It appears that synthetic conduits of polytetrafluoroethylene perform well in either of the situations here tested and may be the best choice at present. (J THORACCARDIOVASCSURG1995;110:427-35)
The quest continues for a better right ventricle-pulmonary artery conduit to correct complex congenital cardiac anomalies. The long-term results with synthetic crimped Dacron materials were disappointing because of the high rate of pseudointimal formation and obstruction
1-6 and the consequent need for reoperation.
The introduction of aortic and pulmonary cryopreserved homografts was thought to be a significant advance. Not only did they facilitate surgical implantation, but also their long-term durability was thought to be better. 7,8 However, results of experimental 9-11 and clinical 12-15 studies of their long-term function have been disappointing. Implantation of cryopreserved homografts in growing lambs showed that these conduits are actually dead tissue, 11 which is able to stretch longitudinally and transversely but which invariably leads to calcification thrombosis or aneurysm formation.
The growing mammal tends to calcify dead tissue at a faster rate than the fully grown adult. This tendency has also been shown in children with cryopreserved homograft conduits 12-16 and those with biologic valve prostheses. 17-21 We designed this study to lessen the uncertainty over which type of prosthetic material is the most suitable for right ventricle-pulmonary artery conduits. We wanted to possibly defer from future consideration some types of material that do not perform satisfactorily and thus narrow the possibilities to just a few types.
We tested various grafts constructed with different techniques and used materials with a wide range of porosity, pliability, and permeability and with and without built-in valves. We used young growing lambs and observed them up into adulthood.
MATERIALS AND METHODS
Design.
We implanted conduits in 100 lambs that were 3 to 4 weeks old (mean weight 11.7 kg). All animals received humane treatment in accordance with the "Principles of Laboratory Animal Care" formulated by the National Society for Medical Research and the "Guide for the Care and Use of Laboratory Animals" prepared by the Institute of Laboratory Animal Resources and published by the National Institutes of Health (NIH Publication No. 86-23, revised 1985).
We divided the lambs into two categories. In 80 lambs (group I), we excised the native pulmonary trunks to the bifurcation level, leaving the native pulmonary valve in place. Then the trunk was replaced with a synthetic straight tubular graft. In group II, comprising 20 animals, the native pulmonary valve was excised as well and continuity was reestablished with use of a composite pulmonary valved conduit. There were a total of 10 groups with 10 animals per group. After implantation (see Surgical technique section), the lambs underwent angiography and cardiac catheterization with hemodynamic measurements at 3, 6, 12, 18, and 24 months. Electively, two lambs in each group were killed at 6, 12, and 18 months; the remaining survivors were killed at 24 months. The lambs that died spontaneously in the postoperative period (within 1 month) underwent autopsies and new lambs were substituted to attain long-term follow-up of the same number of implanted conduits.
We used 10 animals to test each type of conduit. We compared the autopsy findings and measurements with the results of the angiographic studies. Throughout the observation period up to the time of autopsy we obtained these measurements: (1) diameter of the conduit, (2) diameter of the distal and proximal pulmonary artery, (3) level of obstruction, if any, (4) thickness of the inner and outer capsule of the conduit, (5) presence of calcifications or thrombosis, or both, and (6) hypertrophy of the right ventricle at the following levels: interventricular septum, anterior wall, lateral wall, and infundibulum.
Conduit materials.
We tested eight different materials as tubular conduits connecting the right ventricle to the pulmonary artery. Various designs of Dacron fabric included the woven type with a porosity of 50 to 150 ml. Because of its low porosity, this graft usually does not need to be preclotted. We used it as a base of comparison to confirm experimentally what happens clinically when it is implanted in the pulmonary circulation. 22,23
We also tested knitted Dacron fabric with a porosity of 2735 ml. Although it needs preclotting, it has been preferred in repairs of abdominal aortic aneurysms and in aortoiliac bypass operations. Because of its high porosity, it promotes better anchoring strength of the pseudointima to the prosthetic material.
We tested knitted material with a finer texture, more pliability, and a thinner wall manufactured by Golaski Laboratories* in two models: the so-called Milliknit (with 40-needle construction) version and the Microknit (with 50-needle construction) version, both with a porosity of 2870 ml. A tubular graft of coated Dacron fabric that has been made impervious to blood or fluid has been available in several forms. The coating is collagen, albumin, or gel. Foreign to the recipient body, this protein material is supposedly absorbed and replaced by fibrous and endothelial tissue that is well anchored to the matrix of the graft and has a smooth inner surface.
Another graft that seemed to be advantageous is the so-called Dacron fabric interlock woven type (Ochsner
) with porosities of 200 or 500 ml. These grafts are pliable and easily handled. They form a transition between the pure woven and knitted types. Classified as having a three-dimensional weave pattern, they have been popular in vascular surgery.
Polytetrafluoroethylene (PTFE) material,
widely used in vascular surgery, has great appeal: it is impervious to blood or fluid and has a high patency rate, even in grafts of small diameter. 24-26 Although the pores of these grafts can vary from 20 to 90 µm, the most common type used in vascular operations has pores of 35 µm. This is the material that we used for our PTFE conduits. All valveless conduits were of the same diameter (16 mm) and of the same length.
In our valved composite conduits group, we used a size 19 bileaflet St. Jude Medical valve prosthesis§ sewn into a PTFE graft (18 mm in diameter). The valve was positioned in the midportion of the conduit, which facilitates tailoring as desired. We also used a second type of conduit made of Dacron material with a built-in valve known as the Ionescu-Shiley valve|| made of bovine pericardium. This type of device has been used since the late 1970s although the isolated pericardial tissue valve manufactured by Shiley, Inc., has not performed well in the past when implanted in children. 21,27
Surgical technique.
The operation is essentially the same as described in our previous report on the use of cryopreserved allografts. 11 In brief, it involves a left thoracotomy with cannulation of the descending thoracic aorta for arterial perfusion with a 3.8 mm metal-tipped cannula and of the right atrial appendage with a single 26F venous cannula for venous drainage.
The body temperature of lamb, supported by cardiopulmonary bypass, is allowed to drift to 34 ° to 35 ° C. The heart is kept beating. The pulmonary trunk is excised from the bifucation to either the outflow tract of the right ventricle with excision of the valve (group II) or to just above the level of the native valve (group I). The chosen conduit is implanted to reestablish continuity (Figs. 1 and 2). To prevent air from being suctioned into the venous cannula via the ventriculotomy, we exert digital compression over the anterior wall of the right ventricle while we complete the proximal anastomosis. The distal anastomosis at the bifurcation of the pulmonary artery is done first. Then the proximal end is sutured circumferentially to the pulmonary outflow tract of the right ventricle. Both distal and proximal anastomoses are done with a running monofilament suture. The air is vented while we complete the proximal anastomosis.
|
|
Angiographic study.
We obtained a right ventriculogram in all the lambs with pressure measurements in the right ventricle and in the conduit every 3 to 6 months. For lambs with valved conduits we took the measurements proximal and, whenever feasible, distal to the valve. To reliably visualize and measure the conduit diameter at the same levels every time, we obtained angiograms in the same lateral projection. While the lamb was alive, we measured conduit diameter on the film by using as a scale the known diameter of the catheter used for the injection.
Cardiac catheterization data.
To determine whether the conduit was becoming obstructed, we took the following measurements at the time of cardiac catheterization: right ventricular peak systolic pressure, main pulmonary artery systolic and diastolic pressures, pulmonary artery mean pressure, and independent right and left mean and systolic/diastolic pressures of the two branches of the pulmonary artery.
In the nonvalved conduits we measured pressure gradients in the right ventricle and the distal pulmonary artery. In the lambs with valved conduits, whenever feasible, we took pressures between the proximal conduit and the conduit beyond the prosthetic valve. We did not attempt to measure the degree of hypertrophy taking place in the right ventricle during the observation period. Instead, we evaluated hypertrophy at death and recorded it as a pathologic finding.
Factors considered in hemodynamic interpretation.
Right ventricular pressures and pressure gradients across the conduit measured during follow-up were not compared until the entire study was completed. The main reason for this was that the prosthetic conduit material is nonstretchable and noncompliant; therefore the lamb's growth alone would create a pressure gradient between the right ventricle and the distal pulmonary arteries, even if the graft did not develop pseudointimal thickening with obstruction of its lumen.
Up to the time we undertook this study, we only had information pertaining to the evolution of cryopreserved allograft conduits implanted in lambs. 11 These previous studies showed that although the allografts were acellular, became aneurysmal, and developed thrombosis and calcification, they did not become obstructed. Therefore, in comparing the evolution of a synthetic graft under similar conditions, we thought it was useful to compare the obstruction rate in the synthetic grafts versus that in the allograft implants previously reported. 11
PTFE, in the end, was the only type of graft that did not develop significant inner capsules or narrowing of the conduit lumen. Therefore we compared any pressure gradient measured during follow-up with the same pressure gradient in the PTFE group. This allowed us to predict the significance of the thick inner capsule or pseudointimal formation in the other grafts.
Preclotting.
Except for the PTFE conduits (valved and nonvalved) and the collagen-coated grafts, all conduits needed to be preclotted. However, some of the intervascular grafts with a porosity less than 500 ml and some of the Golaski grafts of very fine stitching (Milliknit and Microknit 50-needle construction) were purposely not preclotted. In our long-term follow-up, we compared these two ways of handling the grafts. The preclotted grafts with the highest porosity were knitted types with an index of 1200 ml.
Anticoagulation regimen.
For the lambs undergoing implantation of a mechanical prosthetic valved conduit (St. Jude Medical), anticoagulation with warfarin was required for the entire observation period. We monitored proper levels of anticoagulation with daily prothrombin time studies until the level of anticoagulation was considered to be adequate (15 to 20 seconds; 2.5 to 3 international normalized ratio). After 1 week of close observation, the lamb was released to a farm and the prothrombin times were monitored once a month for the 24-month observation period.
The lambs that underwent implantation of a Dacron fabric conduit with a built-in pericardial tissue valve did not receive anticoagulation therapy.
Histopathologic analysis.
We noted the location of areas with thicker pseudointimal formation, the presence of thrombosis, and the development of strictures at the proximal or distal anastomoses. Our histologic analysis involved microscopic sections from the proximal, middle, and distal portions of the conduit. We identified cellularity, calcium deposits, and the nature of the pseudointimal formation.
RESULTS
Growth.
As the lambs grew from 3 to 4 weeks old to 2 years old, the right ventricular pressure also increased steadily; the size of the conduit (16 mm in diameter) became proportionately smaller for a larger cardiac output. Baseline weight at the time of implant was 11.7 ± 0.9 kg. At 1 month the weight was 19.3 kg; at 3 months, 32 kg; at 6 months, 40.2 kg; at 12 months, 61 kg; at 18 months, 71 kg; and at 24 months, 74 kg (Fig. 3).
|
The right ventricular pressure in the lambs with composite valved conduits was as follows. In the woven graft with the built-in pericardial valve prosthesis it was 34 mm Hg at 1 month, 80 mm Hg at 3 months, 90 mm Hg at 6 months, and 119 mm Hg at 12 months. None of the lambs survived beyond that time. In fact, 8 out of 10 died before 9 months. In the lambs with PTFE conduits and St. Jude Medical valves, the right ventricular pressure was 38 mm Hg at 3 months, 40 mm Hg at 6 months, 64 mm Hg at 12 months, 65 mm Hg at 18 months, and 70 mm Hg at 24 months, and all these lambs survived to the end of the 24-month observation period. Pressure gradients in this group were no different from those in the group with PTFE tube grafts without prosthetic valves.
We found no significant difference in right ventricular pressure between Milliknit material, woven, or Ochsner 500 or 800 interlock woven grafts (with or without preclotting) and the collagen-coated intervascular grafts.
Pressure gradients.
The pressure gradients (Fig. 4) obtained during cardiac catheterization increased progressively during the observation period. In the valveless PTFE conduit group, the gradients were as follows: at 3 months, 15 mm; at 6 months, 16 mm; at 12 months, 20 mm; at 18 months, 22 mm; and at 24 months, 24 mm. Dacron fabric grafts showed significant increases in gradients after the ninth month of observation that ranged from 45 mm in the woven graft to almost 60 mm in the knitted. Woven grafts with pericardial valves became obstructed much sooner than all the other Dacron fabric nonvalved grafts. None of these lambs survived beyond 12 months. Only two lambs completed 11 months of observation. Most of the deaths occurred between 6 and 9 months.
|
The diameters of the Dacron interlock woven Ochsner 500 and 800 grafts were 162 mm at 3 months and only 13.0 mm at 24 months. The group of lambs that received the Ochsner interlock grafts with a porosity of 200 ml (no preclotting) had more significant obstruction. The diameter at 3 months was 15.1 mm, and at 18 months it had decreased to 9.9 mm. The diameters of the woven Dacron Milliknit material grafts, and of the collagen-coated knitted grafts were 16.2 mm at 3 months and 12 mm at 6 months. In the collagen-coated grafts, the obstruction seemed to be even more severe.
Valved conduits.
The two types of composite valved conduits compared as follows. The PTFE/St. Jude Medical prosthesis conduit, which was 18.0 mm in diameter at implantation proximal and distal to the valve, was still 18.0 mm at 3 months and at 24 months. However, the woven grafts with the built-in Ionescu-Shiley valve prosthesis became obstructed rapidly: at 1 month their diameter was 14 mm; at 6 months it was 11.1 mm with severe distortion of the valvular prosthesis. At 12 months, the diameter was 9.1 mm. None of these lambs survived beyond 12 months; they all died suddenly while under observation (Fig. 5). We found no difference between the collagen-coated conduit and the preclotted Dacron fabric grafts with a tighter matrix.
|
Table I).
|
Table I). The distribution of calcification was often patchy in islets along the length and the circumference of the graft, but without any sign of thrombosis or protrusion into the lumen. This phenomenon was also observed in the PTFE/St. Jude Medical valved conduits even with equal distribution of calcium proximal and distal to the prosthetic valve although the valve remained completely clean and functional. The conduit itself was smooth and shiny with no ulcerations.
Histopathologic findings.
Microscopic analysis of the conduit showed some differences among the various types. None of the conduit materials had any evidence of a live neointimal formation in the lumen of the graft. The capsule or pseudointima on the inner surface was completely acellular, formed of collagenous amorphous material that was of variable thicknesses even within the same conduit, depending on the level examined.
The inner capsule of the PTFE material (valved or nonvalved), on the other hand, had a thin acellular glistening surface, less than 0.1 mm thick (Table II). Calcification was extensive and spotty throughout the wall, but in none of the specimens did calcification protrude into the lumen or show signs of thrombosis. The outer capsule was also thin, measuring 0.5 mm thick at the most.
|
The knitted grafts, with or without collagen coating, also showed severe and marked thickening of the inner capsule leading to severe obstruction of the conduit. The thickness was 2 to 3 mm, and thickening was mostly confined to the inferior and posterior surfaces of the circumference, where the crimping was tighter. We found no difference in the amount of inner capsule formation between grafts that required preclotting and the impervious collagen-coated grafts.
In the valved Dacron fabric conduits with the built-in tissue valve, the degree of obstruction and calcification was severe. The obstruction was significantly worse because of rigidity and calcification of the valve cusps, in addition to the thick inner capsule of the conduit proximal to the valve.
Microscopic analysis of the anastomotic suture lines showed no abnormalities. No evidence of any fibromuscular hyperplasia was seen at the junction between conduit and pulmonary artery or between the right ventricular outflow tract and the conduit. The suture lines were all clear of thrombosis and obstruction.
DISCUSSION
The most significant problem with conduits implanted in the pulmonary circulation (that is, the right ventricle to the pulmonary arteries) is obstruction by layers of pseudoneointima in the lumen. If the obstruction is significant enough, the patient eventually needs another operation to replace the graft. Synthetic grafts have an appeal because they are always available in any size needed, can be tailored at the desired length, and are in limitless supply. Woven Dacron fabric grafts were the first synthetic conduits used for right ventricle-pulmonary artery connection. The undesirable evolution of these grafts in causing obstruction and distal embolization is documented. 1,3 In our study, we had to include them to demonstrate that the same problems found in human beings also occur in an animal model.
When valves were integrated in these conduits, first porcine valves 28 and later pericardial valves were used. However, obstruction and calcification of the valve became an additional problem. 21
The knitted Dacron fabric grafts preferred for peripheral arterial operations are thicker and hard to sew; their porosity is high and preclotting is needed. The theory was that the new pseudointima would be cellular and better anchored. Some knitted grafts, like the Golaski type, are thinner and have smaller pores; they have performed better in the arterial circulation than other Dacron fabric prostheses. 29-31 Therefore we believed it necessary to test this material in both its versions, the Milliknit and Microknit materials. Our experimental model demonstrated no difference in the degree of pseudointimal formation in these grafts versus that in woven or regular knitted grafts.
Sealing materials have also been incorporated, namely collagen, albumin, or protein gels. 32-34 However, because previous reports showed collagen sealing to work the best, 33 we decided not to test albumin or gels.
Grafts that have performed well for substitution of the thoracic or abdominal aorta (such as the Hemashield graft*) have performed poorly in the pulmonary circulation. 34
PTFE has significant appeal and is widely used in arterial reconstructive procedures. The obvious advantage is its long-term patency rate, even in grafts with a smaller diameter. 24-26 In addition, it is impervious to blood or fluid. It is somewhat rubbery and stiff to handle, but the healing properties are highly desirable. We previously reported that the use of Gore-Tex brand PTFE conduits between the right ventricle and the pulmonary artery worked well. The only difficulties were technical: in grafts with a large diameter, it is difficult to adapt the straight PTFE conduit to the contours of the heart without causing kinking. Nevertheless, several technical modifications can help avoid these problems, although not totally. 35,36 In our experiments, the prosthetic device performed extremely well as far as lack of obstruction of the conduit itself and durability.
We deliberately varied the position of the St. Jude Medical valves within the conduit. In some lambs, the St. Jude Medical valve was placed more proximally toward the outflow tract of the right ventricle. In other lambs, the valve was left in the center of the conduit(Fig. 7), and still in others, the valve was positioned distally almost into the pulmonary artery itself, leaving a longer portion of the conduit in continuity with the right ventricle(Fig. 8). Position of the valve in the conduit did not alter the good results.
The calcification that occurred in the PTFE material has been observed by us in patients who received right ventricle-pulmonary artery conduits of this material. The cause of this occurrence needs to be investigated in future studies, because it has not been observed when the material is implanted in the systemic arterial circulation.
In conclusion, if a synthetic conduit is needed to connect the right ventricle to the pulmonary artery, PTFE material seems to be preferable. It performed well used either as a simple tubular graft with the native pulmonary valve left in place or as a composite conduit with a built-in mechanical St. Jude Medical prosthesis.
Footnotes
*Golaski Laboratories, Inc., Philadelphia,Pa. 
Intervascular, Inc., Freeport, Tex
Gore-Tex is a registered trademark of W. L. Gore & Associates, Inc., Elkton, Md.
§St. Jude Medical, Inc., St. Paul, Minn.
||Shiley, Inc., Irvine, Calif.
*Meadox Medicals, Inc., Oakland, N.J.
References
This article has been cited by other articles:
|
|
 |
|
  W.-H. Kim, S. K. Min, C. H. Choi, J. R. Lee, Y. J. Kim, E.-J. Bae, and C. I. Noh Follow-Up of Shelhigh Porcine Pulmonic Valve Conduits Ann. Thorac. Surg., December 1, 2007; 84(6): 2047 - 2050. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. T. Mahle, A. R. Cuadrado, and V. K. H. Tam Early experience with a modified norwood procedure using right ventricle to pulmonary artery conduit Ann. Thorac. Surg., October 1, 2003; 76(4): 1084 - 1088. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Pearl, D. S. Cooper, K. E. Bove, and P. B. Manning Early failure of the shelhigh pulmonary valve conduit in infants Ann. Thorac. Surg., August 1, 2002; 74(2): 542 - 549. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Aeba, T. Katogi, and S. Kawada Hemashield implantation in young patients with congenital cardiovascular lesions Ann. Thorac. Surg., June 1, 2000; 69(6): 1920 - 1924. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
