Modification of the descending thoracic aortic anastomotic site using biodegradable felt: Study in a canine model with or without basic fibroblast growth factor
Article Outline
- Abstract
- Abstract
- Background
- Materials and methods
- Results
- Discussion
- Conclusion
- Author contributions
- References
- Copyright
Objectives
We investigated the outcomes of reinforcing anastomotic sites using (1) nonbiodegradable polytetrafluoroethylene (PTFE) felt, (2) biodegradable polyglycolic acid (PGA) felt, and (3) PGA felt with basic fibroblast growth factor (bFGF) in a canine descending thoracic aortic replacement model.
Methods
Thirty-seven beagles underwent descending thoracic aorta replacement using a prosthetic graft with one of the above-mentioned reinforcements or no reinforcement for controls. Histologic evaluations were carried out 1 month and 3 months after surgery. The biomechanical strength of the anastomosis was assessed along the longitudinal axis of the aortic segments using a tensile tester. Local compliance at the anastomotic site was also evaluated in the circumferential direction.
Results
The media was significantly thinner in the PTFE group than in the control group (65.8% ± 5.1% vs 95.0% ± 9.3% of normal thickness; P < .05). Relative to the control group, the adventitial layer was significantly thinner in the PTFE group (42.3% ± 8.2% of control; P < .05) but significantly thicker in the PGA and the PGA + bFGF groups (117.2% ± 11.3% and 134.1% ± 14.2% of control, respectively; P < .05). There were more vessels in the adventitial layer in the PGA + bFGF group than in the control, PTFE, and PGA groups (29.2 ± 2.1/mm2 vs 13.8 ± 0.8, 5.4 ± 0.7, 17.0 ± 1.3/mm2, respectively; P < .01). There were no significant differences between the four groups in the failure force at anastomotic sites. Local compliance at the anastomotic site was higher in the PGA group than that in the PTFE group (11.6 ± 1.6 10−6 m2/N vs 5.6 ± 1.9 10−6 m2/N; P < .05).
Conclusion
Reinforcement of the experimental aortic wall with PTFE felt resulted in thinning of the media and adventitia and fewer vessels at the anastomotic site. These histologic changes were not observed when biodegradable felt was used. The bFGF failed to augment the modification of the aortic wall with the exception of increased adventitial vessel number. Biomechanical strength of the anastomosis along the longitudinal axis was comparable in all four groups; however, local vascular compliance was better in the biodegradable PGA felt group.
Clinical Relevance
This investigation was conducted to extend our previous investigation on a biodegradable felt strip into more practical form before we proceed in a clinical application of the new material. We hypothesized that sustaining compression of the aorta by the nonbiodegradable felt strip may cause structural derangement and local ischemia on the aortic wall, which may lead to occurrence of late postoperative false aneurysm after aortic surgery. We attempted to find a clue for preventing adverse effects of reinforcement with a conventional felt strip. We have found that biodegradable felt prevented thinning of both the media and adventitia and increased adventitial vessels with increased vascular compliance at the aortic anastomotic sites.
Background
Suture line disruption is a late complication of aortic surgery that can cause substantial mortality and morbidity, even years after the primary operation.1 Even if the disruption is asymptomatic, surgical management of this complication cannot be delayed. Although the overall incidence of pseudoaneurysms originating from the suture line between the graft and the aorta is reportedly less than 0.5%,2 the incidence of pseudoaneurysm appears to be higher after surgical repair of acute aortic dissection, occurring in 2% to 3% of cases.3, 4, 5 However, limited information is available on the incidence of pseudoaneurysm associated with particular surgical techniques in this context. Little was known about the long-term integrity of Teflon felt-supported suture lines before the report published by Strauch et al.1
In thoracic aortic replacement with a prosthetic vascular graft, nonbiodegradable polytetrafluoroethylene (PTFE) felt is widely used not only for suture line hemostasis but also for reinforcement at the anastomotic site,6 since a number of pseudoaneurysms have been observed in patients in whom a felt strip was not used to reinforce the suture line, as Strauch et al1 described. However, even though aortic suture line disruption was reportedly infrequent with the use of PTFE felt, the incidence was, in fact, 2.3%.1 Furthermore, the long-term changes in the pathohistologic and biomechanical properties of these anastomotic sites are poorly understood. In a previous study, we found that when the aorta was simply wrapped (with no resection of the aorta) in nonbiodegradable PTFE felt, the adventitia in the wrapped segment of the aortic wall had reduced thickness and fewer vessels relative to controls.7 However, we found that these histologic changes did not occur when either biodegradable polyglycolic acid (PGA) felt or PGA felt incorporating basic fibroblast growth factor (bFGF) were used instead of PTFE felt. These histologic changes were reflected in the finding that the aortic wall that was wrapped in PGA felt or PGA plus bFGF was stronger than that wrapped in PTFE felt. In our previous study, we focused on the effects of simply wrapping the aorta with different materials and eliminated confounding factors such as anastomotic technique. To extend our investigation into a more practical form, in the present study, we used these biodegradable felts to reinforce surgical anastomotic sites in a canine descending thoracic aortic replacement model.
Our fundamental hypotheses are that reinforcement with biodegradable felt prevents tissue derangement occurring at the aortic anastomotic site and yields more durable aortic wall compared to nonbiodegradable PTFE felt reinforcement and that an addition of bFGF using local drug delivery system can augment the modification of the aortic wall.
Materials and methods
Preparation of the gelatin-coated PGA felt strips
Gelatin hydrogels were prepared according to the method of Tabata et al.8 Briefly, 400 μL of a 25% (w/w) aqueous solution of glutaraldehyde (GA; Wako Pure Chemical Industries, Osaka, Japan) was mixed with 200 mL of a 5% (w/w) aqueous solution of gelatin (type I collagen, with an isoelectric point of 5.0, extracted from bovine bone using the alkaline method; Nitta Gelatin, Osaka, Japan). The GA-gelatin mixture was then heated to 40°C. The PGA felt strips (10 mm × 100 mm; Gunze Co, Kyoto, Japan) were exposed to corona discharge for 1 minute to enhance gelatin binding and then soaked in the GA-gelatin solution. Gelatin cross-linking was allowed to proceed for 12 hours at 4°C. To block the residual aldehyde groups of GA, the gelatin-coated PGA felt strips were then immersed in a 50-mM aqueous solution of glycine (Wako Pure Chemical Industries) for 1 hour at 37°C. The PGA felt strips were rinsed three times with double-distilled water (DDW) at 37°C, freeze-dried, and then sterilized with ethylene oxide gas.
Preparation of the gelatin-coated PGA felt strips containing bFGF
Commercial aqueous recombinant human bFGF solution (10 mg/mL, with an isoelectric point of 9.6; Kaken Pharmaceutical, Tokyo, Japan) was diluted with DDW to adjust the concentration to 50 μg/mL.7 Two milliliters of the diluted bFGF solution (containing 100 μg of bFGF) was then dropped onto each freeze-dried gelatin hydrogel-coated PGA felt strip in order to incorporate the bFGF into the gelatin hydrogel. Tabata et al9 has reported slow release of bFGF from gelatin hydrogel. The release rate depends on the gelatin concentration. With 5% of gelatin we have used, it is known for bFGF to be released for approximately a 1-month period. We selected the bFGF dose based on our previous animal experiments using beagles.7 The felt strips were allowed to stand for 1 hour at 25°C to allow the bFGF solution to completely soak into the gelatin hydrogel.
Operative procedure
The experiments were carried out on 32 beagles weighing between 9 and 11.5 kg. Thiopental (30 mg/kg) was used for induction, and general anesthesia was maintained by inhalation of 0.5% sevoflurane. The descending thoracic aorta was exposed through a left lateral thoracotomy incision.
The descending thoracic aorta was replaced with an 8-mm-diameter 5-cm-long prosthetic vascular graft (Gelweave; Vascutek Ltd, Scotland, United Kingdom) using a 5-0 polypropylene over-and-over running suture (Prolene; Ethicon, Inc, A Johnson & Johnson Company, Somerville, NJ). The anastomotic sites were reinforced circumferentially with one of three different materials: PTFE felt (porosity of approximately 1700 mL/min/cm2 H2O; Kouno Co, Chiba, Japan; PTFE group), PGA felt (PGA group), or PGA felt plus bFGF (PGA + bFGF group). Both the proximal and distal anastomoses were reinforced with the same material. The width of the felt strip was 10 mm in all cases (Fig 1, A). No reinforcing material was used in the control group. As for features of the PGA felt, the PGA bears a resemblance to PTFE felt. It was hard to begin to use because it had been freeze-dried. But once it is moisturized either with saline or bFGF solution, it becomes soft and flexible. It is easy to handle. Needle hole bleeding through it was not observed. Blood loss and aortic cross-clamp time were not different between the groups.
Fig 1.
Schematic diagrams illustrating the experimental procedures. A, The anastomotic site of the aorta. The aorta was anastomosed with an 8-mm prosthetic vascular graft using a 5-0 polypropylene over-and-over running suture. The anastomotic site was reinforced with a 10-mm-wide felt strip. B, Cross-section of the aorta. The cross-sectional area of the media (Am) and the outer (Do) and inner (Di) circumference of the media were measured. Using these values, the mean thickness of the media in each section was calculated. To calculate the adventitia-media ratio, the cross-sectional area of the adventitia (Aa) was measured. C, The tensile testing apparatus used for the longitudinal extension of the aortic specimens. Am, Cross-sectional area of the media; Aa, Cross-sectional area of the adventitia; Do, outer circumference of the media; Di, inner circumference of the media.
The dogs were killed at 1 month (n = 3 for each of the PTFE, PGA, and PGA + bFGF groups, n = 0 for the control group) and 3 months (n = 6 for each of the PTFE, PGA, and PGA + bFGF groups, n = 5 for the control group) of follow-up, and the descending thoracic aorta was resected. We chose a 3-month time point to assess histology and anastomotic strength, since the biodegradable PGA felt has been proven to be completely resorbed by 3 months, based on our previous study.7 To illustrate the process of biodegradation of the PGA, a 1-month time point was also chosen as another evaluation point. The proximal anastomotic sites were assessed histologically, and the distal anastomotic sites were subjected to tensile testing as described below. To evaluate the thickness of the media as relative value to that of normal aorta, a section of intact, untreated aorta was taken from 2 cm proximal to the proximal anastomotic site.
Dogs were treated in accordance with the Declaration of Helsinki and the “Guiding Principles in the Care and Use of Animals.” The experimental protocol and procedural protocol for animal care were also approved by the Animal Care Committee of the Graduate School of Medicine, Tohoku University.
Histologic analyses
Aortic specimens from the proximal anastomotic sites were cut and fixed in a 4% formalin solution and then embedded in paraffin. The transmural sections were stained with elastica-Masson (EM) stain and hematoxylin-eosin (HE) to allow microscopic determination of the thickness of the media and the adventitia in each section. The anastomotic sites were also stained with antifactor VIII antibody (Cedarlane, Ontario, Canada) to detect vascular endothelial cells within the aortic wall. In the media and the adventitia, the number of vessels per mm2 was counted in eight randomly chosen fields per slide by two pathologists blinded to the test. The average number of vessels per slide was used to assess the vessel density.
The cross-sectional area of the media (Am) and outer circumference (Do) and inner circumference (Di) of the media were measured, and mean thickness of the media at the anastomotic site (Ta) was calculated based on the following equation.10 Ta = 2Am/(Di + Do) (Fig 1, B). The same measurements were performed for a section of intact untreated aorta, and the mean medial thickness of the untreated section was used to assess the relative thickness of the media at the anastomotic site. To analyze the histologic changes in the adventitia, the cross-sectional area of the adventitia (Aa) was measured, and the adventitia-media ratio was calculated as Aa/Am. These histologic measurements were performed using Image-Pro Plus (v 4.0; Media Cybernetics, Silver Spring, Md) for Windows (Microsoft, Redmond, Wash).
To examine the structural components of the media, the collagen-smooth muscle ratio was evaluated. The respective areas of slides represented by collagen fibers and smooth muscle cells were quantitatively determined by image analysis using IP Lab Spectrum software (v 2.3; Signal Analytics Corporation, Fairfax, Va).11
Tensile testing: failure force and failure stress
The aortic specimens from the distal anastomotic sites were placed between the jaws of a tensile tester (Autograph AGS-50D type III; Shimadzu Co, Kyoto, Japan) (Fig 1, C). The tensile tests were carried out along the longitudinal axis of the arterial segments. No preload was applied to the specimens, and each specimen was tested at a constant elongation rate of 3 mm/min until failure. The cross-sectional area of the specimen at the failure site was measured microscopically. The maximum tensile force was designated the failure force, and the failure stress was determined by dividing the failure force by the measured cross-sectional area.
Study to measure aortic compliance at the anastomotic site
Based on the results of the above described experiments on the four groups (PTFE, PGA, PGA + bFGF, and control; the results are described in the following Results section), we elected to further investigate a difference in aortic biomechanical property between the PTFE and PGA felt groups specifically. To evaluate local compliance at the aortic anastomotic site, we conducted another set of experiments. Five beagles underwent the same operative procedures as described above. In this series, the proximal and distal anastomotic sites were reinforced circumferentially with PGA and PTFE felt, respectively. The dogs were killed at 3 months of follow-up, and the descending thoracic aorta was resected. The proximal and distal anastomotic sites were subjected to compliance testing as described below.
Vascular compliance can be assessed in the three-dimensional directions. Evaluation of aortic compliance in the circumferential direction is the most significant and reliable method to characterize the aortic compliance that accommodate dynamic component during cardiac systolic and diastolic phase. Stress-strain relation was, hence, examined in the circumferential direction. A rectangular aortic strip 10 mm by 20 mm in size was resected from the anastomotic site where either PGA or PTFE was placed externally. Aortic wall thickness (together with PTFE felt where PTFE was used) was first measured using a micrometer 2050F (Mitsutoyo, Kawasaki, Japan) to calculate stress on the aortic wall afterward. Two parallel lines were marked horizontally on the intimal side using Crystal violet. The aortic specimen was suspended on a tensile tester (Autograph AGS-50D type III; Shimadzu Co, Kyoto, Japan; Fig 1, C), and both ends of the specimen were grasped firmly by the jaws of the tester. Each specimen was tested at a constant elongation rate of 3 mm/min within a physiologic range of load (<50 g). Stress-strain curve was recorded. Modulus of elasticity and subsequently compliance was calculated within the load range corresponding to blood pressure of 90 to 140 mm Hg.
Statistical analysis
Statistical analysis was performed using Excel for Mac (Microsoft Co, Redmond, Wash) with the add-in software StatMate III (ATMS Co, Ltd, Tokyo, Japan). All data, except from the vascular compliance test, were analyzed by Kruskal-Wallis nonparametric test. If the Kruskal-Wallis test finding was significant, then a post hoc Steel-Dwass multiple comparisons test was used. Data on the compliance test were analyzed by Mann-Whitney test. Experimental results are expressed as mean ± SEM. A difference with a P value of less than .05 was considered significant.
Results
Survival and macroscopic findings
All 37 beagles survived until they were killed. The PGA felt had not dissolved in the animals killed at 1 month after surgery but had completely dissolved in the animals killed at 3 months. Fibrous encapsulation of the aorta at the anastomosis was observed in the animals killed at 3 months.
Microscopic findings
The EM-stained transmural sections of the anastomotic site are shown in Fig 2, A-C, and Fig 3, A-D. At 1 month after surgery, in the PGA and PGA + bFGF groups, residual PGA fibers remained to some extent, and infiltrating inflammatory cells were observed. At 3 months, the PGA fibers were no longer present and had been replaced with connective tissue.
Fig 2.
Features of the aortic wall at the anastomotic site 1 month after anastomosis. A-C, Histologic sections of the aortic wall at the anastomotic site (original magnification ×40, elastica-Masson staining; scale bar, 200 μm). The sites were reinforced with PTFE felt (A), PGA felt (B), or PGA felt with 100 μg bFGF (C). D, Thickness of the media at the anastomotic site (expressed as a percentage of the thickness of the intact descending thoracic aorta). E, Collagen-smooth muscle ratio in the media at the anastomotic site. Values in (D) and (E) are given as mean ± SEM. n.s., Not significant.
Fig 3.
Features of the aortic wall at the anastomotic site 3 months after anastomosis. A-D, Histologic sections of the aortic wall at the anastomotic site (original magnification: ×40, elastica-Masson staining; scale bar, 200 μm). In the controls, anastomotic sites were not reinforced (A). In the experimental groups, the sites were reinforced with PTFE felt (B), PGA felt (C), or PGA felt with 100 μg bFGF (D). E, Thickness of the media at the anastomotic site (expressed as a percentage of the thickness of the intact descending thoracic aorta). F, Collagen-smooth muscle ratio in the media at the anastomotic site. G, Adventitia-media ratio (AMR) at the anastomotic site (calculated as AMR = Aa/Am, where Aa is the cross-sectional area of the adventitia, and Am is the cross-sectional area of the media). Values in (E, F, and G) are given as mean ± SEM. n.s., Not significant.
Media thickness
At 1 month after surgery, there was no statistically significant difference in the thickness of the media among the PTFE, PGA, or PGA + bFGF groups (Fig 2, D). At 3 months, the media in the PTFE group was significantly thinner than that in the control group (65.8% ± 5.1% vs 95.0% ± 9.3% of normal thickness; P < .05). There was no statistically significant difference between the control group and either of the PGA or PGA + bFGF groups (Fig 3, E).
Collagen-smooth muscle ratio in the media
The collagen-smooth muscle ratios in the media of the PTFE, PGA, and PGA + bFGF groups at 1 month and 3 months after surgery were essentially the same (Fig 2, E, and 3, F).
Adventitia thickness and adventitia-media ratio
At 1 month after surgery, PGA fibers still existed around the aorta (Fig 2, B and C), so measurement of the adventitial layer could not be carried out. At 3 months, relative to the control group, the adventitial layer was significantly thinner in the PTFE group (42.3% ± 8.2% of control; P < .05) but significantly thicker in the PGA and the PGA + bFGF groups (117.2% ± 11.3% and 134.1% ± 14.2% of control, respectively; P < .05). When the adventitia-media ratios were calculated, the adventitia-media ratio of the PTFE group was significantly lower than the ratios of the PGA, PGA + bFGF, and control groups (Fig 3, G).
Vessel density in the adventitia
Representative photomicrographs of the adventitia are shown in Fig 4, A-D. At 3 months, significantly fewer vessels were seen in the adventitia in the PTFE group relative to the other groups. The number of vessels in the PTFE group was almost one-third that in the control group. In contrast, there were markedly more vessels in the PGA + bFGF group than in the control group (Fig 4, E).
Fig 4.
Vascularization of the aortic wall at the anastomotic site 3 months after anastomosis. A-D, Aortic wall sections from the anastomotic site stained using anti-factor VIII antibody. Distinct vessels can be seen in the adventitia of the aortic wall. Control (no felt) (A), PTFE felt (B), PGA felt (C), PGA felt with 100 μg bFGF (D). In the PGA + bFGF group, a marked angiogenic effect was observed relative to the other groups (original magnification ×100; scale bar, 100 μm). E, Vessel density in the adventitia at the anastomotic site (expressed as number of vessels per mm2). In the PTFE group, the number of adventitial vessels was less than half the number in the control. In the PGA + bFGF group, significantly more vessels were seen in the adventitia. Values are given as mean ± SEM.
Failure force and failure stress
The failure force and failure stress were assessed at 1 month (Fig 5, A and B) and 3 months (Fig 5, C and D) after anastomosis. There were no significant differences between the groups at either of the two time points.
Fig 5.
Tensile testing data for the anastomotic sites. The failure force and failure stress were assessed at 1 month (A, B) and 3 months (C, D) after anastomosis. No significant differences were seen between any groups at either of the time points. Values are given as mean ± SEM. n.s., Not significant.
Vascular compliance
The vascular compliance in the θ direction (Fig 6, A) was evaluated 3 months after anastomosis using two different reinforcing techniques (Fig 6, B). Local compliance at the anastomotic site was higher in the PGA group than that in the PTFE group (11.6 ± 1.6 10−6 m2/N vs 5.6 ± 1.9 10−6 m2/N; P < .05).
Fig 6.
Evaluation of vascular compliance. A, Vascular compliance is assessed in the three-dimensional directions (θ, γ, z). During cardiac systolic and diastolic phase, the vessel wall is displaced to accommodate dynamic component. Magnitude of displacement in the z direction is relatively small. Displacement in the γ direction is reflected on that in the θ direction. B, The vascular compliance was evaluated 3 months after anastomosis using two different reinforcing techniques (PTFE vs PGA). Local compliance at the anastomotic site was significantly higher in the PGA group than the PTFE group. Values are given as mean ± SEM.
Discussion
In a previous study, we found that there are differences in the structural changes that occur in aortic wall wrapped in three different materials: PTFE, PGA, and PGA with bFGF.7 In the present study, we examined the pathohistologic and biomechanical changes that occur in surgically anastomosed aortic wall reinforced with felt strips made from the above-mentioned materials. We found that reinforcement of the anastomosis with PTFE was associated with thinning of the media and adventitia and a small number of vessels in the adventitia. Furthermore, we also found that these histologic changes did not occur when PGA felt with or without bFGF was used. These experiments show that the vascular modification that occurs in the presence of biodegradable PGA felt is reproducible, even with surgical manipulation (in this case, sutured anastomosis).
The wrapping method has been widely used to secure hemostasis and to reinforce the anastomotic site during thoracic aortic surgery.1, 6 However, once the tissue has healed completely at the anastomotic site, reinforcement is thought to be no longer necessary; thus, the reinforcing material (usually a PTFE felt strip) is needed for only a short period of time. In agreement with our findings, Kinefuchi et al12 also found that use of nonbiodegradable felt resulted in a thin aortic wall, presumably because of long-standing compression and local ischemia of the aortic wall. In the present study, by using immunohistochemical methods, we found that PTFE felt also caused a decrease in the number of adventitial vessels at the anastomotic site.
In the present study, changes in medial thickness were also examined. In the PTFE group, the medial thickness declined from 75.0% ± 7.6% to 65.8% ± 5.1% of normal thickness (ie, thickness of the intact aorta) between 1 and 3 months following aortic replacement. Over the same period of time, medial thickness in the PGA and PGA + bFGF groups was essentially unchanged. Since PGA is hydrolyzed constantly under in vivo conditions and completely dissolves within 3-4 months of implantation,13, 14 it is likely that compression of the native aorta by the reinforcement (in association with intraluminal pressure) would decrease over time. During the course of this study, we macroscopically and microscopically confirmed that the PGA felt had completely disappeared by 3 months after surgery. We infer from this observation that the thickness of the aortic wall was maintained because of the biodegradability of the PGA felt.
With regard to the histologic changes in the adventitia, it was substantially thickened in the PGA and PGA + bFGF groups (Fig 3, C and D), with the area between the media and the suturing holes being replaced with regenerated connective tissue, whereas in the control group, the suture holes existed right adjacent to the media, although the adventitial tissue was preserved between the stitches. In the PTFE group, the adventitia was structurally deranged. These results are comparable to those obtained in our previous study of aortic wrapping. In the present study, there were rather marked differences in the numbers of adventitial vessels in the different groups. At 3 months, the number of adventitial vessels in the PTFE group dropped down to one-third of that in the control group, but the number in the PGA groups was much larger than that in the control group. The vessel density in the PGA + bFGF group doubled during the 3-month follow-up period compared to that in the control. As was also seen in our previous simple wrapping model, the induction of accelerated angiogenesis by bFGF was notable and reproducible in the present anastomotic model. The bFGF is a powerful angiogenic growth factor; however, information on the biologic effects of bFGF in its free form is limited because its in vivo half-life is very short.15 Tabata and Ikada16 reported that bFGF release can be controlled effectively by incorporation into gelatin hydrogels, and the usefulness of bFGF-containing gelatin hydrogels has been proven in various experimental settings.17, 18 Our model is, therefore, another successful example of a gelatin hydrogel-mediated local drug delivery system. Nevertheless, bFGF failed to augment the modification of the aortic wall with the exception of increased adventitial vessel number. Some clinical studies have reported that bFGF exerts an angiogenic function under the pathologic circumstances such as tissue ischemia (eg, limb ischemia, myocardial ischemia).19, 20 Our results, as seen in Fig 4, demonstrate that adventitial vessel density was essentially equal between the PGA felt and control groups in contrast to decreased vessel density as in PTFE felt group, which implies that PGA felt reinforcement, per se, can prevent tissue remodeling brought by PTFE felt. Tissue regeneration activity and elimination of local compression provided by biodegradable PGA might have predominant impact on the aortic anastomotic site. Under these circumstances, bFGF might not exert its potential role evidently in this model.
Although the bFGF-induced enhanced angiogenesis was not directly associated with increased adventitial thickness (there was no difference in adventitial thickness between the PGA and PGA + bFGF groups), the impact of neovascularization may still be important. In fact, we have recently demonstrated that enhanced perigraft angiogenesis can prevent prosthetic graft infection in a rat model.21 In this regard, addition of bFGF may help to prevent pseudoaneurysm formation with improved resistance to local infection. Further study is warranted to elucidate the impact of bFGF-induced angiogenesis at the aortic anastomotic site.
In an attempt to characterize the different anastomotic techniques, we evaluated the biomechanical properties of the aortic wall at the anastomotic site. In our previous study of a simple wrapping model (without surgical anastomosis), in the tensile test of the aortic wall, the PGA-wrapped wall, with or without bFGF, had a twofold greater failure stress value than the PTFE-wrapped wall, when tested in a longitudinal direction. In contrast, in the present study, there were no statistical differences in the results of the tensile test performed at either of the two time points. Failure force obtained here corresponded to 350 to 800 mm Hg, which is beyond physiologic blood pressure range. This means the biodegradable PGA felt yielded as durable anastomotic strength as PTFE did. One can argue that once the reinforcing material is dissolved, the anastomotic site becomes no longer robust. It was not the case in our experiment. There were no incidences of suture line loosening or pseudoaneurysm formation either. To the best of our knowledge, our experiment proved for the first time that suture line does not loosen even if biodegradable material is used as a reinforcement at the aortic anastomosis. In order to better characterize the biomechanical property of the aorta at the anastomotic site, we conducted another set of experiments creating 3-month models where we performed a tensile test in the circumferential direction to evaluate compliance of the aorta. Local compliance at the anastomotic site in the PGA group was twofold compared with that in the PTFE group. The PGA felt reinforcement yielded pliable aortic segment at the anastomotic site, whereas PTFE felt generated rigid segment. In general, compliance mismatch has been known to produce anastomotic shear stress with subsequent damage to the arterial wall. The structural impairment permits tearing out of sutures from the vessel wall. The process is facilitated by prior loss of structural integrity of the vessel wall.22, 23 In this regard, PGA felt may play a beneficial role to reduce compliance mismatch between the intact host artery and the aortic segment that is involved in the anastomosis.
Conclusion
Reinforcement of anastomotic sites with nonbiodegradable PTFE felt resulted in thinning of both the media and adventitia in the aortic wall. These changes were associated with a reduced number of vessels in the adventitial layer. In contrast, when biodegradable PGA felt was used, these histologic changes were not observed. The bFGF failed to augment the modification of the aortic wall with the exception of increased adventitial vessel number. There was no difference in failure force in a longitudinal direction relative to the PTFE group. PGA felt, however, provided more compliant aortic segment at the anastomotic site. Biodegradable PGA felt can prevent the medial thinning and reduction in adventitial vessel density and produce compliant aortic segment compared to PTFE felt when used at the anastomotic site.
Study limitations
Since our study was conducted using a large animal model, the follow-up period was limited to 3 months. It is, therefore, still unknown whether the histologic alterations we observed persist over the long term. Thus, further investigation of the long-term effects of PGA felt with or without bFGF on aortic wall remodeling is required. In the present study, we performed a longitudinal tensile test, but endoluminal pressure tests and compliance tests in the radial direction might also be required to gain a better understanding of the biomechanical properties of the anastomotic site.
This work was supported in part by a research grant from the Uehara Memorial Foundation. The authors wish to express their appreciation to Katsuhiko Oda, Ichiro Yoshioka, and Satoshi Kawatsu for their technical assistance during animal experiments, Hideto Ido, Tsubasa Matsui, Masaki Oi, Takatomo Ushiyama, and Junichi Yamazaki for their superb technical support in performing the tensile tests, Mika Watanabe and Kazumasa Ishida for pathohistologic assessment, and Masako Komatsu for her advice on the statistical analysis.
Author contributions
References
- . Long-term integrity of teflon felt-supported suture lines in aortic surgery. Ann Thorac Surg. 2005;79:796–800
- . Mediastinal false aneurysm after thoracic aortic surgery. Ann Thorac Surg. 2000;70:547–552
- Aortic dissection (Prevalence, cause, and results of late reoperations). J Thorac Cardiovasc Surg. 1994;108:199–205discussion 205-6
- . Surgical strategy for repair of large pseudoaneurysms of the aortic isthmus. Eur J Cardiothorac Surg. 2001;20:1240–1242
- . Hypothermic circulatory arrest in operations on the thoracic aorta (Determinants of operative mortality and neurologic outcome). J Thorac Cardiovasc Surg. 1994;107:788–797discussion 797-9
- . Chronic thoracic and thoracoabdominal aortic disease. In: Lampert RH, Geisel EB editor. Cardiac Surgery. 3rd ed.. Philadelphia: Churchill Livingstone; 2003;p. 1851–1900
- The wrapping method using biodegradable felt strips has a preventive effect on the thinning of the aortic wall: experimental study in the canine aorta. J Vasc Surg. 2006;43:349–356
- . Biodegradation of hydrogel carrier incorporating fibroblast growth factor. Tissue Eng. 1999;5:127–138
- . Vascularization effect of basic fibroblast growth factor released from gelatin hydrogels with different biodegradabilities. Biomaterials. 1999;20:2169–2175
- . A morphological and morphometric study of the intrahepatic portal veins in patients with biliary atresia and portal hypertension. Jpn J Pediatr Surg. 1988;20:133–140
- . Intramural distribution of elastic moduli in thoracic aortas and its relationship to histology: comparison between porcine and bovine thoracic aortas. JSME Intl J series C. 1999;42:568–573
- The influence of vascular wall on wrapping method in the canine aorta. Jpn J Cardiovasc Surg. 1995;24:71–79
- . Biodegradable polyesters for medical and ecological applications. Macromol Rapid Commun. 2000;21:117–132
- . Biodegradable polymers. In: Tsuruta T, Hayashi T, Ishihara K, Kataoka K, Kimura Y editor. Biomedical Applications of Polymeric Materials. Boca Raton: CRC Press, Inc; 1993;p. 163–189
- Effects of chronic systemic administration of basic fibroblast growth factor on collateral development in the canine heart. Circulation. 1995;91:145–153
- . Protein release from gelatin matrices. Adv Drug Deliv Rev. 1998;31:287–301
- . Tissue regeneration based on tissue engineering technology. Congenit Anom (Kyoto). 2004;44:111–124
- . Controlled release of basic fibroblast growth factor promotes healing of the pancreaticojejunal anastomosis: a novel approach toward zero pancreatic fistula. Surgery. 2007;142:734–740
- . Intramyocardial sustained delivery of basic fibroblast growth factor improves angiogenesis and ventricular function in a rat infarct model. Heart Vessels. 2003;18:93–99
- Novel method to enhance sternal healing after harvesting bilateral internal thoracic arteries with use of basic fibroblast growth factor. Circulation. 2000;102(19 Suppl 3):III307–III311
- Enhanced perigraft angiogenesis prevents prosthetic graft infection. Ann Thorac Surg. 2008;86:1278–1284
- . Is compliance mismatch the major cause of anastomotic arterial aneurysm? (Analysis of 42 cases). J Cardiovasc Surg (Torino). 1985;26:147–150
- . Pathogenesis of anastomotic aneurysms. Surgery. 1981;90:509–515
Competition of interest: none.
Drs Hidenori Fujiwara and Yoshikatsu Saiki contributed equally to this manuscript.
The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a competition of interest.
PII: S0741-5214(09)01797-2
doi:10.1016/j.jvs.2009.08.078
© 2010 Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.
