Clinical and Investigative Medicine

Arterial wall mechanical characteristics after treatment in collagenase: an in vitro aneurysm model

Ladan Dadgar, PhD
Yves Marois, MSc
Xiaoyan Deng, PhD
Robert Guidoin, PhD

Clin Invest Med 1997; 20 (1): 25-34.

[résumé]

From the Department of Surgery, Université Laval, and Quebec Biomaterials Institute Inc., Hôpital Saint-François d'Assise, Quebec, Que.

(Original manuscript submitted Oct. 30, 1995; received in revised form Sept. 26, 1996; accepted Oct. 15, 1996)

Reprint requests to: Dr. Robert Guidoin, Laboratoire de chirurgie expérimentale, Local 1701, Pavillon de services, Université Laval, Laval QC G1K 7P4; fax 418 656-7512

Contents

Abstract

 Objective: To investigate the mechanical characteristics of canine aortas treated with buffered collagenase as a first step in developing an animal model of aortic aneurysm for the validation of stent-grafts.

Design: In vitro study of canine aortas.

Interventions: Canine thoraco-abdominal arteries were divided into the descending thoracic aorta, the suprarenal artery and the infrarenal artery; these segments were incubated separately in a buffered collagenase solution for 1 to 6 hours. Some segments were left untreated as controls.

Outcome measures: Mean arterial wall thickness, measured with Vernier callipers and computerized histomorphometric methods, and longitudinal tensile strength of control and treated vessel segments.

Results: The arterial wall thickness decreased with incubation time. After 1 hour of incubation the reduction was approximately 15% for the descending thoracic aorta, 16% for the suprarenal artery and 18% for the infrarenal artery. After 6 hours the total reduction in wall thickness was 32%, 41% and 44% respectively for the 3 segments. The tensile strength of the treated arterial segments also decreased with the incubation period. Initially, the infrarenal segment displayed the greatest strength; however, this was reversed as the period of incubation increased. The inelastic limit of the descending thoracic aorta control segment was reached at 100% elongation, whereas that of the suprarenal artery was reached at 80% and that of the infrarenal artery was reached at 60% elongation. All of the arterial segments became weaker as the period of incubation in buffered collagenase increased.

Conclusion: This in vitro incubation technique successfully altered the structure of the collagen fibre network within the arterial wall. This method may be an option in developing an aneurysm model to test stent-grafts in vivo.

[Table of contents]

Résumé

Objectif : Étudier les caractéristiques mécaniques d'aortes de chien traitées avec de la collagénase tamponnée comme première étape vers la mise au point d'un modèle animal d'anévrisme de l'aorte pour la validation d'endoprothèses.

Conception : Étude in vitro d'aortes de chien.

Interventions : On a divisé des artères thoraco-abdominales de chien en aorte thoracique descendante, artère suprarénale et artère infrarénale. On a fait incuber ces segments séparément dans une solution tamponnée de collagénase pendant 1 à 6 heures. On a gardé des segments non traités comme segments témoins.

Mesures des résultats : Épaisseur moyenne des parois de l'artère mesurée au moyen d'une jauge micrométrique et de méthodes histomorphométriques informatisées, et résistance à la traction longitudinale de segments témoins et de segments de vaisseaux traités.

Résultats : L'épaisseur des parois artérielles a diminué avec la durée de l'incubation. Après 1 heure d'incubation, la réduction a atteint environ 15 % dans le cas de l'aorte thoracique descendante, 16 % dans celui de l'artère suprarénale et 18 % dans celui de l'artère infrarénale. Après 6 heures, la réduction totale de l'épaisseur de la paroi artérielle des 3 segments a atteint 32 %, 41 % et 44 % respectivement. La résistance à la tension des segments d'artère traités a aussi diminué en fonction de la période d'incubation. Au début, le segment infrarénal était le plus résistant, mais le phénomène s'est inversé avec l'allongement de la période d'incubation. Le segment témoin d'aorte thoracique descendante a atteint sa limite d'inélasticité à une élongation de 100 %, contre 80 % dans le cas de l'artère suprarénale et 60 % dans celui de l'artère infrarénale. Tous les segments d'artère se sont affaiblis avec l'allongement de la période d'incubation dans la collagénase tamponnée.

Conclusion : Cette technique d'incubation in vitro a modifié avec succès la structure du réseau de fibres de collagène de la paroi artérielle. Cette méthode peut être une façon d'élaborer un modèle d'anévrisme pour tester des endoprothèses in vivo.

[Table of contents]

Introduction

The implantation of endovascular stented grafts has led the way to aggressive treatment of aneurysms with fewer risks and costs than classic vascular surgery.[1­3] Nevertheless, many questions about the short- and long-term effects of these new devices remain unanswered. To validate transluminally placed endovascular stented grafts, experimental aortic aneurysm models have been developed. The first model involved a surgical patch approach,[4,5] in which an aneurysm was created by replacing an aortic segment with an oversized synthetic patch or with autogenous tissue.[5] The second model was the elastase-induced experimental aneurysm.[6] This model was more elegant and closer to reality, yet difficult to reproduce in large animals.[7]

Endothelial and vascular smooth muscle cells have been shown to contribute very little to the total strength of the arterial wall.[8,9] Rather, the strength of an artery is attributable mainly to its collagen fibres as well as to its elastin. Canine models have shown that aortic tensile strength increases with the artery's distance from the heart.[10] Furthermore, the longitudinal force required to rupture the abdominal aorta is higher than that required to rupture the thoracic aorta.[11] Based on the observation that the abdominal aorta has a higher collagen content than the thoracic aorta, Levene[12] showed that the strength of the artery increases with its collagen-fibre content.

Under normal conditions, there is a balance between synthesis and degradation of collagen and proteoglycans in a healthy artery under repair after an injury. In diseased arteries such as aneurysms this balance may be disturbed as a result of a loss of collagen and tensile strength.[13] Canine arteries[14] may be ruptured and aortic dilatation[15,16] may be obtained by collagenase infusion and by increased aortic collagenase concentration in the aortic wall. To validate the effectiveness of endovascular stented grafts in excluding aneurysms from blood circulation, an animal model of aortic aneurysm by collagenase infusion, with consistent and reproducible results, must be developed. This in vitro study was the first step in a research program to measure the mechanical characteristics of canine aortas treated with collagenase.

[Table of contents]

Material and methods

Preparation of the arterial segments

The entire thoraco-abdominal artery, up to the trifurcation, was carefully removed from 7 adult mongrel dogs, each weighing approximately 18 to 22 kg. The dogs had fasted for 24 hours before surgery. They were then anesthetized with 32 mg/kg of intravenously administered sodium pentobarbital. An anterior midline abdominal incision was made to expose the aorta. The artery was then dissected by carefully ligating the side branches. To better preserve the dimensional and mechanical integrity of the vessel, we took measures to avoid applying any excessive longitudinal force to the vessel during the procedure. Upon removal, the artery was rinsed in a 0.9% saline solution to eliminate any remaining blood elements. Any loose connective tissue covering the vessels was gently trimmed. Each artery was then divided into 3 segments: the descending thoracic aorta (DTA), the suprarenal aorta (SRA) and the infrarenal aorta (IRA) (Fig. 1). Each segment was 6 to 8 cm long, depending on the weight of the dog. The internal diameter of each segment was measured at both ends with a cone-shaped diameter-measuring device. The mean internal diameter of each segment was then calculated.

Preparation of the buffered collagenase solution

A fresh buffered collagenase solution was prepared for all experiments.17 The solution consisted of a 0.05 mol/L Tris-hydrochloric acid buffered solution at pH 7.5 supplemented with calcium chloride (0.36 mmol/L). The solution was then warmed to 37°C and collagenase type 1A (Sigma, St. Louis, Mo.) was added to a final concentration of 500 U/mL. Also added to the solution was 0.02% w/v of sodium azide to prevent bacterial growth.

In vitro incubation study

All of the arterial segments, except the control samples, were placed separately in closed containers containing 50 to 70 mL of the buffered collagenase solution at 37°C, without stirring, for periods of incubation ranging from 1 to 6 hours. Care was taken to ensure that all of the segments were free from folds and completely immersed in the solution. One sample from each arterial segment was selected for each period of incubation. After each incubation period, the arterial segments were rinsed 3 times in distilled water and once in 0.9% saline solution before being mechanically tested.

Histological study

Segments of 0.5 cm from both ends of each vascular section were cut and fixed in a 10% formalin solution, then embedded in paraffin. Sections 5 µm thick were then cut and stained with Masson's Trichrome, hematoxylin­eosin and Weigert for microscopic examination to determine the extent of the enzymatic digestion of the collagenous matrix and to detect any changes in the elastin content of the arterial wall in each segment. The same histological examination was carried out on untreated arterial segments, and the results were used as controls.

Arterial wall thickness

The mean thickness of the arterial wall was measured with a Vernier calliper and confirmed by histomorphometric measurement.

Vernier calliper method: Tissue thickness was measured at 10 locations along the length of each segment, before and after each incubation period, with a Vernier calliper accurate to 0.01 mm. The mean wall thickness was then calculated for each segment.

Histomorphometry: Histomorphometric studies were performed to evaluate the extent of collagen digestion in the media (the tissue layer between the inner elastic lamina and the external lamina) as well as in the adventitia (the peripheral tissue outside the external lamina of the artery). The thickness of the medial and adventitial regions of 2 representative sections of each segment was measured with a precise computerized histomorphometric system, the DADS 560 (Wild Leitz, Wetzlar, Germany). Representative sections of the control aortas located in the thoracic, suprarenal and infrarenal regions were also evaluated. Ten to 15 measurements were taken at 500-µm intervals along each section. Measurements were taken at a magnification of 120 times with an eyepiece mounted perpendicular to the luminal surface.

Tensile testing

Arterial samples 5 to 7 cm long were cut from each remaining segment for longitudinal tensile testing. Each sample was placed between the pneumatic jaws of an electrically driven tensiometer (Instron model 1130, Instron Canada, Ltd., Laval, Que.). The initial length of each sample placed between the jaws was 4.0 cm. The tensile tests were carried out along the longitudinal axis of the arterial segments. The proximal end of the arterial segment with the larger diameter was placed in the upper jaw of the tensiometer. Slippage and tearing of the specimens were minimized by using one serrated jaw opposite a smooth one lined with sandpaper. To reduce the hysteresis effect, each sample was preconditioned 3 times in a cyclic load fashion at a constant elongation rate of 5 cm/min before testing. Load elongation curves were recorded with a synchronized chart recorder at a chart speed of 5 cm/min. The corresponding stresses experienced by the test samples were calculated by normalizing the loads with respect to the initial cross-sectional areas of the arterial wall tested. The calculated stresses were plotted against elongations or strains.

Statistical analysis

Paired 2-tailed Student's t-tests were used to compare the data obtained. A difference with a p value of less than 0.05 was considered significant.

[Table of contents]

Results

Mean aortic internal diameter

The mean internal diameters of the control arterial segments and standard deviations (SDs) are reported in Table 1 The narrowing of 17% between the DTA and the IRA indicated that the internal diameter of the thoraco-abdominal artery decreased as the distance from the heart increased.

Histologic results

The control aorta in the thoracic region showed a thick media with an elaborate elastic network and a very thin peripheral layer of adventitia. The media was thinner and the adventitia was thicker in the suprarenal region than in the thoracic region. In the infrarenal region, the adventitia was thick and granulomatous, characterized by loose collagen with islets of smooth muscle cells delimited by external elastic fibres. The media was thinner than in the suprarenal region, displaying a less condensed elastic network (Fig. 2).

After a 1-hour incubation in buffered collagenase, all of the segments exhibited significant modifications of the luminal surface. Internally, the endothelial cells and the elastic lamina had been digested. Externally, some areas of the adventitia were digested. After 2 hours, the wall thickness had been reduced and the collagen content of the media had begun to alter in all 3 segments (Fig. 3). The adventitia in the descending thoracic segment had been completely digested after 3 hours of incubation, whereas that in the suprarenal region had been digested after 4 hours and that in the infrarenal region had been digested after 5 hours. During the 6-hour incubation period, the media of each segment gradually became thinner, and several areas had been digested by the enzyme, resulting in a more fragile wall (Fig. 4).

Arterial wall thickness

Table 2 shows the thinning of the arterial wall segments after various periods of incubation in the buffered collagenase solution. The wall thickness of all 3 aortic segments had changed significantly (p < 0.05) after each period of incubation. After a 1-hour incubation, the mean wall thickness of the DTA had decreased by 15% on average, gradually dropping by 32% after 6 hours of incubation. Similarly, the thickness of the SRA segment had decreased by 16% after the first hour of incubation. However, after 6 hours, the mean thickness of the SRA segments had decreased by 41%, and the mean thickness of the infrarenal wall had also decreased by approximately 44%. After the first hour of incubation, the mean reduction in wall thickness for the 3 aortic segments was approximately 17%, indicating that the digestion of arterial wall substances had already begun during this early stage of incubation.

Histomorphometric studies

Table 3 and (Fig. 5) show the mean thickness of the media of the DTA, SRA and IRA after varying incubation periods in buffered collagenase. The media of the DTA was significantly reduced as early as 1 hour after incubation, and reduction of the media continued as the incubation period increased. Between 3 and 6 hours, the thickness of the DTA media gradually diminished by 52%, from 2000 µm to 967 µm. Similarly, the media of the SRA showed a gradual loss of approximately 46% (from 1245 µm to 671 µm) during the same incubation period. However, the thickness of the IRA media remained virtually unchanged for up to 5 hours, and decreased by 32% only after 6 hours.

The extent of digestion of the adventitia is shown in (Fig. 6). In contrast to the results obtained for the media, the thickness of the adventitia in the IRA region had decreased by 75% after 1 hour, and the adventitia had been completely digested after 6 hours of incubation. The same events were observed in the SRA adventitia. The adventitia of the DTA segment, however, was virtually eliminated within 3 hours of incubation.

Tensile testing

The stress-strain results for all segments and periods of incubation are shown in Figs. 7 Figs. 8 Figs. 9These figures show the relationship between the calculated stress (force per unit of area) and the strain (% elongation) obtained from the load-elongation curves recorded by an Instron chart recorder. In general, the stretching behaviour of the tested aortic tissues are expressed in terms of 3 distinct phases. In the initial elastic stretching phase, the tissue elongates smoothly and linearly, with increasing stress, until the elastic limit is reached. The slope of the stress-strain curve is the stiffness of the tissue. Beyond this elastic limit, the curve changes its slope and a more rapid increase in stress is required to produce the same rate of elongation. This portion of the curve is called the inelastic phase, and the elongation slope of the curve is the inelastic stiffness. The final phase of the stretching is the breaking point, at which stress drops to zero.

Fig. 7 presents the stress-strain relationship for the DTA segments after various periods of incubation. A distinctive difference in mechanical properties was observed between the control DTA and the treated tissue segments. The elastic limit of the control sample was recorded at 0.2 MPa, corresponding to an elongation of approximately 100%. Beyond this elastic limit, the slope of the stress-strain curve increased dramatically, indicating a significant increase in tissue stiffness. However, as the period of incubation in collagenase increased, the inelastic phase of the curve disappeared. The stiffness of the treated tissue remained almost constant and was much lower than that of the control sample. As the period of incubation increased, the tissue stiffness decreased. After 5 hours of incubation, the stress-strain curves merged, suggesting that the mechanical properties may have been completely altered in the first 5 hours of incubation.

The stress-strain relationship for the SRA segments is shown in Fig. 8. The elastic limit of the control sample was 0.25 MPa, at an elongation of 80%. After a 1-hour incubation, the elongation had increased to 100% for the same elastic limit. After 4 hours the inelastic phase had disappeared, and after 5 hours the mechanical properties of the wall were completely altered.

Fig. 9 shows the stress-strain curves of the IRA segments. The collagenase began altering the mechanical properties of the tissue as early as 2 hours after starting incubation. After the first hour of treatment, the shape of the curve showed no apparent difference from that of the control sample. The 3 stretching phases were distinctive, and the elastic limit was almost the same as that of the control specimen, which was 0.17 MPa at a strain of 60%. However, after a 2-hour incubation, the collagenase began to affect the mechanical properties of the arterial segments. The stiffness of the tissue dropped after 3 hours, and the elastic limit disappeared thereafter. After 5 hours, the stress-strain curves merged.

Fig. 10 shows the rupture tensile strength (rupture force/cross-section area of the test vessel) as a function of incubation time. In the first 3 hours of incubation, the rupture tensile strength was greater in the IRA segment than in the DTA and SRA segments. However, between 4 and 6 hours of incubation, no differences in the value of the rupture tensile strength among the 3 segments were observed.

[Table of contents]

Discussion

The recent development of endovascular stented grafts has revolutionized vascular surgery. Implanted transluminally through the arterial femoral route, these new devices can restore blood circulation in stenotic arteries and exclude aortic aneurysms. However, at this time there is little information on the short- and long-term effects of these endoprostheses after implantation. To investigate more closely the biostability and biofunctionality of endovascular stented grafts for the management and treatment of aneurysms, there is an urgent need to develop a consistent animal model that reproduces weakened aneurysmal arteries like those observed in humans. The surgical-patch technique is still far removed from clinical situations, and the elastase-treatment approach is not ideal because it cannot be reproduced in large animals.[7]

In this study, we successfully altered the strength and integrity of the canine arterial wall by modifying its collagen content. Three segments of the thoracoabdominal arterial tree -- the DTA, the SRA and the IRA -- were incubated in buffered collagenase for periods ranging from 1 to 6 hours. The incubation of the arterial segments in the solution led to a loss in wall thickness (as revealed by the thickness measurements and histomorphometric studies) due to the digestion of the collagen fibres. The loss in thickness was greater in the IRA segments, mainly because the collagen content was higher in the adventitia. The adventitia of the IRA segments was approximately 75% digested after only 1 hour of incubation. Histological results also confirmed that there had been considerable digestion of the adventitia within the first hour of collagenase activity. In contrast to the SRA and IRA segments, the complex elastic media of the DTA segments was more susceptible to enzymatic incubation because of a much greater collagen content. Digestion was observed mainly on the luminal surface by removing the endothelial lining, intima and, in some cases, part of the media.

The stress-strain results indicated that, as the incubation period increased, the wall strength decreased. As collagen is the main load-carrying element in the artery[18] and has a very high Young's modulus (in the order of 109 dyns/cm2), the loss of strength must be linked directly to the digestion of the collagen fibres. Tensile testing results also revealed that the strength was originally higher in the IRA control sample than in the SRA and DTA samples for a given strain value. However, after 3 hours of collagenase incubation, all 3 aortic segments decreased in strength to the point that they ruptured at almost the same deformation position.

Natural tissues, such as arteries, do not fit the definition of an elastic body because their stress-strain relationship shows nonlinear elasticity.[19] Yield points observed in stress-strain studies are believed to be caused by the helical format of collagen fibres and the different layers of collagen in the arterial wall. The yield points are reached when some of the collagen fibres are broken, tranferring the load to the remaining fibres. In our study, yield points were observed in the control samples and in all 3 arterial segments incubated for up to 2 hours, confirming that the level of digestion increased with the period of incubation.

In contrast to stenosis formation, the development of aneurysms predominantly involves the medial layer of the arterial wall.6 Activated macrophages within the aortic media may be responsible for elastase secretion and elastic tissue destruction, leading to dilatation of the wall.[6] Studies by Anidjar and associates6 and by others[14,20,21] have demonstrated that the destruction of the elastin network and elastolytic activity within the media is the necessary initial stage of aneurysmal disorder. However, when perfusing the aorta with other proteases such as collagenase, they observed aneurysmal dilatation with fibrin deposits in the lumen, as well as multiple microscopic aneurysms. Nevertheless, the rupture of the aneurysmal wall was never reported in these studies, and the incubation of the arterial segments already perfused in elastase was shown not to promote elastolysis in the media or aneurysmal dilatation. Their models showed that aneurysms in vivo could be produced by macrophage activation within the media and could be responsible for both the secretion of the elastase and elastolytic activity.

On the other hand, aortic collagenase activity has been shown to be elevated.[15,22] Exposure of the arterial wall to collagenase results in the digestion of the collagen fibre, causing weakness and rupture of the wall.[14] Busuttil, Abou-Zamzam and Machleder[15] reported that abdominal aortic aneurysms possess enzymatic activity that leads to collagen degradation and that this phenomenon is prevalent in all types of aneurysms. The amount and distribution of collagen in the aneurysmal area defines the tensile strength of the aneurysmal wall. Therefore, any change in the amount of collagen has a direct effect on the resistance of the aneurysmal wall to intramural stress caused by blood pressure. Menashi and collaborators[22] supported this finding by confirming that collagenase activity increases in a ruptured aneurysmal aorta. From these studies, it therefore appears that an imbalance between synthesis and degradation of the arterial wall collagen may be responsible for the development of aneurysms and that collagenase is as important to aneurysmal rupture as elastase is to aneurysmal dilatation.[14] This study confirms this hypothesis experimentally, finding that collagenase may indeed significantly reduce the strength of the arterial wall and may cause it to fail, leading to dilatation and rupture.

Conclusion

Arterial wall strength depends on its collagen fibre content. Any disruption of this fibre network causes structural and mechanical alteration of the artery. We successfully developed an in vitro incubation technique to alter the structure of the collagen fibre network within the arterial wall. This method may be useful in developing an experimental aneurysm model to test stent-grafts in vivo.

Acknowledgements

This work was supported by the Medical Research Council of Canada (grant no. MT.7879) and the Quebec Biomaterials Institute Inc. The authors would like to thank Dr. Patrick Stroman for his valuable insight as well as Louisette Martin, Jacques Bastien, Andreas Eberhart and Josée Boulet for their technical assistance.

[Table of contents]

References

  1. Chuter TA, Green RM, Ouriel K, Fiore WM, DeWeese JE. Transfemoral endovascular aortic graft placement. J Vasc Surg 1993;18:185-97.
  2. Moore WS, Vescera CL. Repair of abdominal aortic aneurysm by transfemoral endovascular graft replacement. Ann Surg 1994;220:331-41.
  3. Marin ML, Veith FJ, Cynamon J, Sanchez LA, Lyon RT, Levine BA, et al. Initial experience with transluminally placed endovascular grafts for the treatment of complex vascular lesions. Ann Surg 1995;222:449-69.
  4. Balko A, Piasecki GJ, Shah MD, Carney WI, Hopkins RW, Jackson BT. Transfemoral placement of intraluminal polyurethane prosthesis for aneurysm. J Surg Res 1986;40:305-9.
  5. Criado E, Marston WA, Woosley JT, Ligush J, Chuter TA, Baird C, et al. An aortic aneurysm model for the evaluation of endovascular exclusion prostheses. J Vasc Surg 1995;22:306-15.
  6. Anidjar S, Salzmann JL, Gentric D, Lagneau P, Camilleri JP, Michel JB. Elastase-induced experimental aneurysm in rats. Circulation 1990;82:973-81.
  7. Marinov GR, Marois Y, Pâris E, Roby P, Formichi M, Douville Y, et al. Can the infusion of elastase in the abdominal aorta of the Yucatan miniature swine consistently produce experimental aneurysms? J Invest Surg In press.
  8. MacDonald A. Blood flow in arteries. Baltimore: Williams and Wilkins, 1960:146-76.
  9. Burton AC. Relation of structure to function of tissue of the wall of the blood vessels. Physiol Rev 1954;34:619-42.
  10. Harkness MLR, Harkness RD, McDonald DA. The collagen and elastin content of the arterial wall in the dog. Proc Roy Soc Biol 1957;146:541-51.
  11. Hiertonn TE, Jordon P. Tensile strength of canine aortic segments. Angiology 1956;7:21-6.
  12. Levene CI. Collagen as a tensile component in the developing thick aorta. Br J Exp Pathol 1961;42:89-94.
  13. King TA. Hemodynamics of arterial blood flow. In: Giordano JM, Trout HH III, De Palma RG, editors. Basic science of vascular surgery. Mount Kisco (NY): Futura Publishing, 1988:109-24.
  14. Dobrin P, Baker W, Gley W. Elastolytic and collagenolyic studies of arteries. Arch Surg 1984;119:405-9.
  15. Busuttil R, Abou-Zamzam A, Machleder H. Collagenase activity of the human aorta. Arch Surg 1980;155:1373-8.
  16. Zarins CK, Runyon-Hass A, Zatina MA, Lu CT U, Glagov S. Increased collagenase activity in early aneurysmal dilatation. J Vasc Surg 1986;3:238-48.
  17. Zhang Z, Guidoin R, King M, How T, Marois Y, Laroche G. Removing fresh tissue from explanted polyurethane prosthesis: Which approach facilitates physico-chemical analysis? Biomaterials 1995;16:369-80.
  18. Canfield TR, Dobrin PB. Static elastic properties of blood vessels. In: Skalak R, Chien S, editors. Handbook of bio-engineering. New York: McGraw Hill, 1987:16.1-16.28.
  19. Fung YC. Bioviscoelastic solids. In: Fung YC, editor. Biomechanics: mechanical properties of living tissue. 2nd ed. New York: Springer-Verlag, 1990:242-320.
  20. Sumner DS, Hokanson DE, Strandness DE Jr. Stress-strain characteristics and collagen-elastin content of abdominal aortic aneurysms. Surg Gynecol Obstet 1970;130:459-66.
  21. Campa JS, Greenhalgh RM, Powell JT. Elastin degradation in abdominal aortic aneurysms. Atherosclerosis 1987;65:13-21.
  22. Menashi S, Campa JS, Greenhalgh RM, Powell JT. Collagen in abdominal aortic aneurysms: type, content and degradation. J Vasc Surg 1987;6:578-82.
| CIM: February 1997 / MCE : février 1997 |
CMA Webspinners / >