Elemental composition, morphology and mechanical properties of calcified deposits obtained from abdominal aortic aneurysms

Acta Biomaterialia 2 (2006) 515–520 www.actamat-journals.com

Elemental composition, morphology and mechanical properties of calcified deposits obtained from abdominal aortic aneurysms Steven P. Marra a

a,b,*

, Charles P. Daghlian c, Mark F. Fillinger b, Francis E. Kennedy

a

Thayer School of Engineering, 8000 Cummings Hall, Dartmouth College, Hanover, NH 03755-8000, United States b Section of Vascular Surgery, Dartmouth-Hitchcock Medical Center, Lebanon, NH, United States c Electron Microscope Facility, Dartmouth Medical School, Hanover, NH, United States Received 10 December 2005; received in revised form 19 April 2006; accepted 11 May 2006

Abstract Calcified deposits exist in almost all abdominal aortic aneurysms (AAAs). The significant difference in stiffness between these hard deposits and the compliant arterial wall may result in local stress concentrations and increase the risk of aneurysm rupture. Calcium deposits may also complicate AAA repair by hindering the attachment of a graft or stent-graft to the arterial wall or cause vessel wall injury at the site of balloon dilation or vascular clamp placement. Knowledge of the composition and properties of calcified deposits helps in understanding the risks associated with their presence. This work presents results of elemental composition, microscopic morphology, and mechanical property measurements of human calcified deposits obtained from within AAAs. The elemental analyses indicate the deposits are composed primarily of calcium phosphate with other assorted constituents. Microscopy investigations show a variety of microstructures within the deposits. The mechanical property measurements indicate an average elastic modulus in the range of cortical bone and an average hardness similar to nickel and iron. Ó 2006 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Calcium deposit; Mechanical properties; Elemental analysis; Electron microscopy; Abdominal aortic aneurysm

1. Introduction Calcium deposits are hard aggregates of calcium-rich particles that form in blood vessels. The biochemical mechanisms of vascular calcium deposit formation are complex and not completely known [1,2]. A histological study of excised aortas by Stary [3] indicated that calcium granules form intracellularly and are released into the extracellular matrix through cell death. Stary observed such granules, approximately 5–10 lm in size, in the cytoplasm of smooth muscle cells. The released calcium granules aggregate in the arteries and grow into larger structures. In an earlier work, Bobryshev et al. [4] used electron microscopy to study * Corresponding author. Address: Thayer School of Engineering, 8000 Cummings Hall, Dartmouth College, Hanover, NH 03755-8000, United States. Tel.: +1 603 650 3597; fax: +1 603 650 4928. E-mail address: [email protected] (S.P. Marra).

aortic calcification. Their observations indicate that aortic calcification requires some pre-existing extracellular structural base, such as unesterified cholesterol. Proudfoot and Shanahan [2] also note that there is good evidence from gene studies to suggest that vascular calcification is a regulated process similar to the calcification in bone. Regardless of the mechanisms of calcification formation, the presence of vascular calcium deposits poses serious health risks. One major concern is that a deposit will detach from the arterial wall and lodge in a smaller artery or arteriole further downstream [5,6]. A detached calcification originating in a carotid artery may relocate to a cerebral artery and lead to transient ischemic attacks or stroke. Detached deposits from coronary arteries are known to cause heart attacks. Detachment in other areas of the body may result in local cell and/or tissue death. This work focuses on calcifications located within abdominal aortic aneurysms (AAAs). People above 50 years

1742-7061/$ - see front matter Ó 2006 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.actbio.2006.05.003

516

S.P. Marra et al. / Acta Biomaterialia 2 (2006) 515–520

of age are at the greatest risk of developing an AAA. Aortic calcification is also more common in older people. In a study of aortic computed tomography (CT) scans from 257 patients, male and female, and with no known aortic diseases, Dixon et al. [7] reported that calcification was identified in some patients as early as their fourth decade. Of patients in their sixth decade or older, 77% had some degree of calcification. Aortic calcification is even more prevalent in patients with AAAs. Pillari et al. [8] studied CT scans of AAAs from 55 patients, male and female, ages 60–86, and observed some degree of calcification in each. Previous research by Fillinger et al. [9,10] has demonstrated that aortic aneurysm wall stress is strongly associated with aneurysm rupture risk in abdominal aortic aneurysms. Calcified deposits within an abdominal aortic aneurysm may increase the stress in the aneurysmal wall, and hence increase the risk of aneurysm rupture. The considerable difference in stiffness between the compliant arterial tissue and a hard deposit results in the latter acting as a stress raiser; the deposit limits the deformation of the wall tissue on which it is adhered while causing an increase in the strain of the neighboring tissue. This effect has been demonstrated in AAA finite element studies by Inzoli et al. [11] and Marra et al. [12]. Calcium deposits may also complicate AAA repair by hindering the attachment of a graft or stent-graft to the arterial wall, or cause vessel wall injury at the site of balloon dilation or vascular clamp placement [13]. Knowledge of the composition and properties of calcified deposits is beneficial to understanding the risks associated with their presence. This work presents results of elemental composition, microscopic morphology, and mechanical property measurements of human calcified deposits obtained from within AAAs. Chemical composition studies of aortic calcium deposits have been performed previously, although only on specimens from non-aneurysmal aortas. Schmid et al. [14] analyzed deposits using various procedures including atomic absorption spectroscopy, alkalimetric titration, thermal conductivity measurements, and X-ray diffraction. They observed that both large and small deposits had essentially the same composition, consisting of 71 wt.% apatite,1 9 wt.% carbonate, and 15 wt.% protein. Tomazic [1] also conducted various chemical analyses of aortic deposits and detected primarily calcium phosphates with assorted inorganic constituents including sodium, magnesium, fluoride, and carbonate. X-ray diffraction patterns of the deposits indicated a crystalline structure characteristic of apatite. Tomazic also reported that the amount of organic matter in native (non-deproteinized) deposits was 20– 30 wt.%. 1

Wopenka and Pasteris [15] caution that the word apatite refers to a kind of calcium phosphate with a specific crystal structure and should not be used as an identifier unless such a structure is confirmed. Wopenka and Pasteris also argue that some biological phosphates are mistakenly labeled as apatites.

The microscopic structures of non-aneurysmal aortic calcified deposits have been previously studied by Schmid et al. [14] using scanning electron microscopy (SEM). They observed five different structures in their specimens: (1) smooth-surfaced spheres, (2) spheres of spindle-like, radially arranged particles, (3) networks and bundles of fibers, (4) irregularly-shaped particles with ‘‘fuzzy’’ surfaces, and (5) flat, smooth plates. Similar observations were made by Tomazic in a later SEM study of aortic calcifications. No known information has been published regarding the mechanical properties of vascular calcified deposits. The mechanical properties of mineral apatites have been reported by Grenoble et al. [16], who used ultrasonic experimental procedures to characterize these materials. The lack of published information on the mechanical properties of calcified deposits may be due to their relatively small size. The properties presented in this work were measured using a nanoindentation system specifically designed for testing very small specimens. Nanoindentation measurements have been reported previously for bone by Rho and Pharr [17]. While the elastic modulus values presented by Rho and Pharr do agree well with bulk bone measurements, it should be noted that properties obtained from nanoindentation testing relate to the microstructural scale of the testing specimen, and may deviate from bulk properties. 2. Materials Twelve calcified deposit specimens, from 12 different patients, were studied. The patients ranged in age from 53 to 85 and included both genders. All of the specimens studied were excised from patients during planned or emergency open aneurysm repair surgeries, and from arterial wall tissue that would otherwise have been discarded. No specimens were removed from patients specifically for this work. All calcified specimens were removed manually from the intimal side of the excised aneurysm wall. Most specimens were plate-like, but irregular in shape, and varied in size. Specimen thicknesses ranged from about 0.7 mm to about 2 mm (prior to preparation for testing), and the largest specimen dimensions were approximately 8 mm. 3. Methods and results Approval to use patient tissues for research purposes was obtained before starting this study from the institutional review board (Committee for the Protection of Human Subjects, 8/20/2001) and all subjects provided informed consent. 3.1. Elemental analysis Elemental analyses were performed on four calcified deposit specimens. The specimens were dried in air for several days and then ground flat using a series of increasingly

S.P. Marra et al. / Acta Biomaterialia 2 (2006) 515–520

Fig. 1. Elemental analysis results from a single scan of one calcium deposit specimen. These results indicate large amounts of calcium (Ca) and phosphorus (P) and smaller amounts of oxygen (O), carbon (C), magnesium (Mg), sodium (Na), sulfur (S) and chloride (Cl). Elemental scans of the other specimens gave similar results.

finer grit silicon carbide sanding papers. This grinding process removed any excess tissues from the surface of the deposits and exposed the interior constitution. The specimens were finally polished to a smooth finish using 3 lm alumina powder. Specimens were coated with a thin (a few nm thick) layer of carbon in a Fullam MKII carbon fiber coater (E.F. Fullam, Inc., Latham, NY) to prevent surface charging, and to provide a low atomic number background. Samples were examined in an XL-30 field emission-gun (FEG) environmental scanning electron microscope (ESEM) at 15 KV in high vacuum mode (FEI Co., Hillsboro, OR). X-ray microanalysis was carried out with an EDAX Si (Li) detector (10 mm2 crystal) and EDAX Genesis spectral analysis software (EDAX Inc., Mahwah, NJ). Several measurements were taken for each specimen. Each measurement was of an area with an effective diameter of 2–3 lm. Exact quantitative elemental analysis was not possible due to the inherent surface roughness of the specimens. However the measurements do provide a general indication as to the relative amounts of the compositional elements. All four specimens showed very similar elemental compositions. Fig. 1 presents the results of one specimen scan, but is representative of all the scans. Large amounts of calcium and phosphorus are present, and oxygen and carbon are also abundant (although it is unclear how much of the detected carbon is from the deposit and how much is from the applied coating). Small amounts of sodium, magnesium and sulfur were also detected in all of the specimens. Two of the specimens contained additional traces of chlorine and potassium.

3.2. Microscopy Morphological investigations of the microstructures of the calcium deposits were conducted using the XL-30 ESEM FEG at 15 KV in both secondary and backscattered

517

Fig. 2. Scanning electron microscopy images of a calcium deposit at various locations and magnifications.

electron modes. The specimens were fractured before mounting in the microscope in order to expose the internal structure of the deposits. Specimens were also coated with 60:40 Au:Pd in a Technics Hummer V (Anatech, Springfield, VA) sputter coater to provide conductivity. Micrographs of the fractured deposit surfaces are presented in Fig. 2. As in the works by Schmid et al. and Tomazic, various different structures are evident. Conglomerations of spherical particles and smooth platelike structures approximately 5 lm thick can be seen in Fig. 2(A). These latter structures appear as both single and ‘‘stacked’’ plates and are in the same general orientation as the parent deposit. Smooth spherical particles approximately 2–5 lm in diameter are shown in Fig. 2(B). Fine fibrils resembling spider-webs are also present. Plate-like structures, smaller than those in Fig. 2(A), are shown in Fig. 2(C). This figure also reveals spherical and irregularly-shaped particles with rough surfaces. A closer examination of the rectangular area shown in Fig. 2(C) shows them to be composed of nanometer-sized finger-like projections (Fig. 2(D)). A backscattered electron image of the flat surface of a calcium deposit specimen is shown in Fig. 3. In backscattered electron imaging mode, a flat sample shows atomic number contrast due to the monotonically increasing backscattered electron yield with increasing average atomic number of the sample. In Fig. 3, the areas rich in calcium and phosphorus are brightest and the darker areas are either organic components or partially mineralized areas. The black portions of the image represent either voids or embedding medium. Surface cracks are also evident, which may have formed during the formation of the calcification, or during the specimen preparation process. It is clear from the figure that the calcium deposit composition is not uniform throughout the specimen. Nanoindentations were made in homogeneous areas away from cracks or other defects to the extent possible.

518

S.P. Marra et al. / Acta Biomaterialia 2 (2006) 515–520

The elastic moduli of the specimens were easily computed from these values, the measured effective elastic modulus values, and Eq. (2) Especimen ¼

Fig. 3. Backscattered SEM image of the surface of a calcium deposit. Areas rich in calcium phosphate are brightest and the darker areas are either organic components or partially mineralized areas. The black portions of the image represent either voids or embedding medium.

3.3. Mechanical properties measurements Nanoindentation measurements were obtained using a Hysitron Ubi1 Nanomechanical Test Instrument (Hysitron Inc., Minneapolis, MN). Measurement tests are performed by driving a small indenter (Berkovich diamond tip) into a specimen to a maximum depth of several hundred nanometers, holding it there briefly, and then removing it. The indentation force and depth of penetration are recorded during the loading, hold, and unloading phases. The hardness and elastic (Young’s) modulus of the specimen can be determined from this data [17]. The specimen thickness is roughly five orders of magnitude larger than the indentation depth, and thus the specimen may be assumed to represent a semi-infinite solid. The hardness, H, is defined as: H¼

P max ; A

ð1Þ

where Pmax is the maximum indentation force and A is the resultant projected contact area of the indenter tip at that load. This area is a function of the penetration depth and the indenter tip geometry. The area function, described by Rho and Pharr [17], was calibrated using a fused quartz specimen provided by the nanoindenter manufacturer. The elastic modulus, E, is determined from the 20–95% portion of the unloading force–displacement data. From this data the Hysitron software computes an effective modulus, Eeffective, which includes contributions from both the specimen and the indenter 1 Eeffective

¼

1  m2specimen 1  m2indenter þ ; Especimen Eindenter

Eeffective ð1140 GPaÞð1  0:272 Þ : ð1140 GPaÞ  Eeffective ð1  0:072 Þ

Eight calcium deposit specimens were tested using the nanoindenter. Four of the specimens were dehydrated with ethanol and propylene oxide, and then embedded in Epon 812 epoxy prior to testing. These four were sectioned and mounted such that the direction of indentation was perpendicular to the deposit thickness (see Fig. 4). The other four specimens were dried in air for several days and mounted without embedding (‘‘free’’). The direction of indentation for these specimens was parallel to the deposit thickness. All of the specimens were ground flat using a series of increasingly finer grit silicon carbide sanding papers and then polished to a smooth finish using 3 lm alumina powder. A thin (<50 lm) layer of cyanoacrylate glue was used to attach each specimen to a steel mounting plate that could then be held in place by the testing apparatus. Surface roughness scans were taken of each specimen with the Ubi-1 testing instrument prior to indentation using the indenter tip as a probe for contact atomic force microscopy imaging. Relatively flat areas were chosen for testing, and care was taken to target indents away from discontinuities such as cracks and pores, and away from the specimen boundaries. The area of each indentation was on the order of 3 lm2. Specimen surfaces were re-scanned after testing to confirm the location of the indents. The locations of all of the indents were found to be satisfactory, and thus no indentation results were discarded. The testing protocol for each specimen consisted of a steady loading phase to a maximum indentation force of 1500 lN at a rate of 60 lN/s (quasi-static loading). This was followed by a holding period at 1500 lN for 5 s, followed by an unloading phase back to zero force at a rate of 60 lN/s. At least four separate indentations were made in each specimen. Fig. 5 presents a typical indentation force vs. depth of penetration plot for the calcium deposit specimens. The average elastic modulus and the average hardness of each specimen are presented in Table 1. Despite the heter-

Indenter Embedded Specimen

ð2Þ

where m is the Poisson’s ratio. The elastic modulus and Poisson’s ratio for the indenter used in this work were 1140 GPa and 0.07, respectively. A value of 0.27 was assumed for the Poisson’s ratio of the specimens as this was the value measured by Grenoble et al. for the Poisson’s ratio of mineral hydroxyapaptite at atmospheric pressure.

ð3Þ

Steel Plate

A

Free Specimen

B

Fig. 4. Preparation of calcium deposit specimens for nanoindentation testing. (A) Embedded specimens: deposit is embedded in epoxy and then sectioned. The specimen is mounted in the tester so that the direction of indentation is perpendicular to the thickness direction. (B) Free specimens: deposit is mounted flat so that the indentation occurs in the thickness direction.

S.P. Marra et al. / Acta Biomaterialia 2 (2006) 515–520

519

4. Discussion 4.1. Elemental analysis

Fig. 5. Typical indentation force vs. indenter penetration depth for a calcified deposit specimen. The elastic modulus is determined from the 20– 95% portion of the unloading curve.

Table 1 Average elastic modulus and hardness values measured for each specimen tested using the nanoindenter Mounting

Specimens

No. of indents

Elastic modulus GPa (SD)

Hardness MPa (SD)

Embedded

1 2 3 4

4 4 4 4

21.9 13.7 26.8 35.1

(8.2) (5.1) (8.6) (9.5)

930 415 443 1093

(647) (140) (103) (379)

Free

5 6 7 8

6 5 5 4

31.5 15.8 20.0 16.6

(13.0) (3.2) (4.7) (9.5)

856 722 745 478

(486) (143) (241) (306)

ogeneity of the specimens, all of the elastic modulus values are in the low tens of GPa, and all of the hardness values are in the hundreds of MPa. The average elastic modulus and hardness values for all of the embedded and all of the free specimens are shown in Table 2, along with the average values for all the specimens. Both mounting groups had similar average values of elastic modulus (24.4 GPa embedded, 21.0 GPa free) and hardness (720 MPa embedded, 700 MPa free).

The results of the elemental analyses of the calcium deposits are in agreement with those of Schmid et al. and Tomazic. The large amounts of calcium and phosphorus are not surprising as those are component elements of calcium phosphate. It is not certain whether actual apatite is present since its crystalline structure was not confirmed. The presence of oxygen and carbon was also expected as oxygen is also an element of phosphate, and carbon was used to coat the specimens. Further, both of these elements are constituents of carbonate, cholesterol and other organic compounds. Magnesium is another element commonly found in calcified tissues in both organic and inorganic compounds [18]. Sodium, chloride and potassium are of course essential elements for life and are necessary for many physiological processes. The only element detected in this work that was not reported by either Schmid et al. or Tomazic is sulfur. Sulfur is found in four amino acids and in many proteins, so its detection is not altogether surprising. However, it is interesting that the presence of sulfur has not been reported previously. 4.2. Microscopy The calcium deposit microstructures observed in the SEM images are similar to those reported by Schmid et al. and Tomazic. The orientation of the plates shown in Fig. 2(A) may indicate a preferred crystal growth direction in alignment with the direction of blood flow. The spherical particles and conglomerations of spheres in Fig. 2(A) and (B) were also noted by Schmid et al. and Tomazic. Tomazic identified these particles as hydroxyapatite with some carbonate present. Peters and Epple [19] were able to grow similar crystals of hydroxyapaptite in vitro and remarked that the spherical shape is apparently induced by an interaction of the growing calcium phosphate crystals with cholesterol. The fine fibrils shown in Fig. 2(B) may be some kind of protein or other organic matter that became attached to the aggregated particles. The ‘‘rough’’ surfaces identified in Fig. 2(C) and (D) are possibly the same kind of ‘‘fuzzy’’ surfaces observed by Schmid et al. and/or ‘‘spongy’’ surfaces observed by Tomazic. The origin of the finger-like projections that make up these surfaces is not known at this time. 4.3. Mechanical properties

Table 2 Average elastic modulus and hardness values of the embedded and free specimens, and of all specimens Mounting

Embedded Free All

Average values Elastic modulus GPa (SD)

Hardness MPa (SD)

24.4 (7.9) 21.0 (7.6) 22.7 (7.7)

720 (317) 700 (294) 710 (306)

It was expected that the heterogeneous microstructures of the calcium deposits observed in the SEM images might cause considerable variations in the mechanical properties measured using the nanoindenter. However, all of the measured elastic moduli and hardness values were of similar magnitude for all the tested specimens. The range in the standard deviations for the measurements, from relatively small (e.g., specimen 6) to somewhat considerable (e.g.,

520

S.P. Marra et al. / Acta Biomaterialia 2 (2006) 515–520

specimen 8) is not altogether surprising considering the variation of constituent distribution shown in Fig. 3. Differences in the averaged measured properties between the embedded specimens (indented perpendicular to the thickness direction) and the free specimens (indented parallel to the thickness direction) are minor (24.4 GPa vs. 21.0 GPa for elastic modulus, 720 MPa vs. 700 MPa for hardness), indicating mechanical isotropy. The total average elastic modulus of 22.7 GPa is three orders of magnitude stiffer than AAA wall tissue [20]. This modulus value is in the range of cortical bone (22.4–25.7 GPa [17]), but five times less than that of pure hydroxyapatite (114 GPa [16]). This supports the elemental analysis and microstructural evidence of the deposits being composite structures of calcium phosphate and other constituents. The sensitivity of the assumed Poisson’s ratio on the total average elastic modulus is small. A decrease in the assumed Poisson’s ratio from 0.27 to 0.1 results in a total average elastic modulus of 24.5 GPa; an increase to 0.4 results in a total average elastic modulus of 20.5 GPa. The total average hardness value of 710 MPa is similar to that of certain metals (e.g., nickel and iron). However, most pure ceramics are significantly harder. The aggregate structure, and possibly porosity, of the deposits is a reasonable explanation as to why the measured hardness values are not higher.

[2] [3]

[4]

[5] [6]

[7]

[8]

[9]

[10]

[11]

[12]

5. Conclusions [13]

The elemental analyses and microscopy images indicate that the calcium deposits studied in this work, which were obtained from within AAAs, are very similar to those studied by Schmid et al. and Tomazic, which were taken from non-aneurysmal aortas. This is the first known work to report mechanical property measurements for these deposits. The elastic moduli measurements reveal the calcium deposits to be three orders of magnitude stiffer than the surrounding AAA wall tissue. This significant mismatch in stiffness supports the hypothesis that the deposits act as AAA wall stress raisers and strengthens the argument for their inclusion in AAA stress analyses. The measured hardness values also explain the known difficulty of trying to compress a calcified vessel with a vascular clamp or suture through regions of aortic wall containing such deposits. References [1] Tomazic BB. Characterization of mineral phases in cardiovascular calcification. In: Brown PW, Constantz B, editors. Hydroxyapatite

[14]

[15] [16]

[17]

[18]

[19]

[20]

and related materials. New York, NY: CRC Press; 1994. p. 93–115. Proudfoot D, Shanahan CM. Biology of calcification in vascular cells: Intima versus media. Herz 2001;26:245–51. Stary HC. Natural history of calcium deposits in atherosclerosis progression and regression. Z Kardiologie 2000;89(Suppl. 2):II/ 28–=0?>II/35. Bobryshev YV, Lord RSA, Warren BA. Calcified deposit formation in intimal thickenings of the human aorta. Atherosclerosis 1995;118:9–21. Shah PK. Mechanisms of plaque vulnerability and rupture. J Am Coll Cardiol 2003;41:15S–22S. Zohdi TI, Holzapfel GA, Berger SA. A phenomenological model for atherosclerotic plaque growth and rupture. J Theor Biol 2004;227:437–43. Dixon AK, Lawrence JP, Mitchell JRA. Age-related changes in the abdominal aorta shown by computed tomography. Clin Radiol 1984;35:33–7. Pillari G, Chang JB, Zito J, Cohen JR, Gersten K, Rizzo A, et al. Computed tomography of abdominal aortic aneurysm. Arch Surg 1988;123:727–32. Fillinger MF, Raghavan ML, Marra SP, Cronenwett JL, Kennedy FE. In vivo analysis of mechanical wall stress and abdominal aortic aneurysm rupture risk. J Vasc Surg 2002;36:589–97. Fillinger MF, Marra SP, Raghavan ML, Kennedy FE. Prediction of rupture risk in abdominal aortic aneurysms during observation: Wall stress versus diameter. J Vasc Surg 2003;37:724–32. Inzoli F, Boschetti F, Zappa M, Longo T, Funero R. Biomechanical factors in abdominal aortic aneurysm rupture. Eur J Vasc Surg 1993;7:667–74. Marra SP, Fillinger MF, Chen DT, Dwyer JM, Kennedy FE. The inclusion of calcified plaques in patient-specific finite element models of abdominal aortic aneurysm. J Biomech Eng, submitted for publication. Chaufour X, White GH, Hambly BD, Yu W, May J, Harris JP, et al. Evaluation of the risks of using an oversized balloon catheter in the human infrarenal abdominal aorta. Eur J Vasc Endovasc Surg 1998;16:142–7. Schmid K, McSharry WO, Pameijer CH, Binette JP. Chemical and physicochemical studies on the mineral deposits of the human atherosclerotic aorta. Atherosclerosis 1980;37:199–210. Wopenka B, Pasteris JD. A mineralogical perspective on the apatite in bone. Mater Sci Eng C 2005;25:131–43. Grenoble DE, Katz JL, Dunn KL, Gilmore RS, Murty KL. The elastic properties of hard tissues and apatites. J Biomed Mater Res 1972;6:221–33. Rho J-Y, Pharr GM. Nanoindentation testing of bone. In: An YH, Draughn RA, editors. Mechanical testing of bone and the boneimplant interface. New York, NY: CRC Press; 2000. p. 257–69. LeGeros RZ. Biological and synthetic apatites. In: Brown PW, Constantz B, editors. Hydroxyapatite and related materials. New York, NY: CRC Press; 1994. p. 3–28. Peters F, Epple M. Simulating arterial wall calcification in vitro: Biomimetic crystallization of calcium phosphates under controlled conditions. Z Kardiologie 2001;90(Suppl. 3):III/81–III/85. Raghavan ML, Vorp DA. Toward a biomechanical tool to evaluate rupture potential of abdominal aortic aneurysm: identification of a finite strain constitutive model and evaluation of its applicability. J Biomech 2000;33:475–82.