Low-temperature particulate calciumphosphates for bone regeneration

M. V. F. AraujoV. C. MendesP. ChattopadhyayJ. E. Davies

Authors’ affiliations:M. V. F. Araujo, V. C. Mendes, P. Chattopadhyay,J. E. Davies, Faculty of Dentistry, University ofToronto, 124 Edward Street, Toronto, ON, CanadaM5G 1G6Institute of Biomaterials and BiomedicalEngineering, University of Toronto, 164 CollegeStreet, Toronto, ON, Canada M5S 3G9.

Correspondence to:J. E. DaviesInstitute of Biomaterials and BiomedicalEngineeringUniversity of Toronto164 College StreetToronto, ONCanada M5S 3G9Tel.: (416) 978 1471Fax: (416) 946 5639e-mail: davies@ecf.utoronto.ca

Key words: anorganic bone, bioactive glass, bone substitutes, calcium phosphates, demi-

neralized bone matrix

Abstract

Background: Low-temperature synthesized calcium phosphates are produced by mixing

calcium phosphate powders in an aqueous solution resulting in a precipitated phase. These

compounds can be formulated in several forms (e.g. injectable cements and implantable

blocks), and are commonly used as bone substitutes and drug delivery systems for the

treatment of bone defects. As bone substitutes, calcium phosphates in general offer the

advantages of being biocompatible and osteoconductive.

Aims: The present work employed a machine-based process to derive a reproducible

preparation method for low-temperature calcium phosphate particulate (LTCP). The in vivo

outcomes of LTCP were compared with those of three commercially available bone

substitutes by histomorphometric measurements of bone formation and material

degradation in a rat femur implantation model.

Materials & Methods: Specifically, LTCP, anorganic bovine bone (AB), bioactive glass (BG),

and demineralized bone matrix (DBM) were implanted in defects created in the distal

aspect of rat femora. Reparative bone and particulate volumes of these biomaterials were

evaluated post-operatively using micro-computed tomography and histological analyses at

3, 6, 12, and 16 weeks.

Results & Discussion: Results showed that, despite invoking bone formation, AB, BG, and

DBM were found un-resorbed in situ at 16 weeks. Conversely, LTCP showed an early increase

in bone formation as well as clear evidence of complete degradation and reparative

bone remodelling, resulting in the total reconstitution of the marrow cavity and marrow

tissue.

Conclusion: LTCP promoted increased early bone formation, associated with an improved

degradation rate, compared with the other three bone-substitute biomaterials tested.

Effective bone grafts should provide ade-

quate support during repair without delay-

ing the remodelling process. Thus, one of

the great challenges in the bone biomater-

ials field is to develop bone substitutes that

favor repair, but are resorbed and substi-

tuted by natural bone during tissue turn-

over (Schwartz et al. 2008).

Low-temperature synthesized calcium

phosphates (LTCaP) consist of biomater-

ials obtained from the mixing of equimolar

amounts of calcium phosphate powders in

aqueous solutions at or below room tem-

perature. In the preparation process of these

materials, the grinding of the powders is

done using a machine, whereas the mixing

of the solid with the liquid phase is done by

hand (Brown & Chow 1986; Fukase et al.

1990; Bohner et al. 2005; Ambard & Mue-

ninghoff 2006). These biomaterials are

Date:Accepted 26 September 2009

To cite this article:Araujo MVF, Mendes VC, Chattopadhyay P, Davies JE.Low-temperature particulate calcium phosphates forbone regeneration.Clin. Oral Impl. Res. 21, 2010; 632–641.doi: 10.1111/j.1600-0501.2009.01864.x

632 c� 2010 John Wiley & Sons A/S

biocompatible and osteoconductive, and

are often used as bone substitutes in differ-

ent forms, such as in situ setting injectable

cements (Miyamoto et al. 1997, 1999;

Ooms et al. 2003), solid or porous litho-

morphs (Takagi & Chow 2001; Zhang

et al. 2006), and as set cement granules

(Tamimi et al. 2006, 2007, 2008).

However, the diverse reagent chemistry

described for the preparation of calcium

phosphates, allied with the various implan-

tation models, have resulted in great varia-

bility of the reported in vivo outcomes.

Little calcium phosphate cement resorp-

tion and bone formation were reported at

18 months post-operative in a cat sinus

obliteration model (Friedman et al. 1991);

however, in rat tibial defects, such cements

showed increased bone formation but no

osteoclastic resorption at 8 (Miyamoto

et al. 1997) and 24 weeks (Ooms et al.

2003) after implantation. Conversely, the

replacement of in situ setting calcium

phosphates by bone was reported in cat

parietal defects at 6 months (Costantino

et al. 1992) and in dog facial defects at 9

months (Shindo et al. 1993).

Among the formulations reported in the

literature, particulate CaP biomaterials are

especially attractive as they have an in-

creased surface area, are osteoconductive,

biocompatible (Ginebra et al. 2006), and

can fill any defect size or shape (Gautier et

al. 2001; Tas 2008). The implantation of

three size ranges (10–20; 80–100; 200–

400mm) of ceramic particulate biphasic

calcium phosphate [hydroxyapatite

(HA)þ b-tricalcium phosphate (b-TCP)]

showed that 10–20 and 200–400 mm parti-

cles presented significant bone formation

compared with controls. However, despite

invoking an increased inflammatory reac-

tion, 10–20mm particles promoted both

increased bone formation and particulate

resorption (Malard et al. 1999). An in-

creased bone mineral density, granule re-

sorption, and bone formation were found

after calvaria implantation of brushite at 4

weeks post-operative compared with

BioOsss

(Tamimi et al. 2006) and empty

defects (Tamimi et al. 2007). Similarly, the

implantation of monetite granules (0.2–

1 mm in size) in rabbit’s calvaria showed

signs of particulate resorption and an in-

creased mineral content compared with

empty defects at 4 and 8 weeks post-

operative (Tamimi et al. 2008).

One of the most important characteris-

tics of LTCaP is their resorbability in vivo,

which occurs by the combination of two

processes: (1) dissolution in the in vivo

fluidic environment and (2) cell-mediated

resorption, mainly by osteoclasts. The

in vivo dissolution of these biomaterials is

strictly related to their composition and

particle size (LeGeros & LeGeros 2003),

while their resorption is dictated by the

gradual dissolution of the apatitic phase in

the acidic environment created by protons

secreted through osteoclastic membrane-

bound vacuolar proton pumps (V-ATPases)

(Farina & Gagliardi 1999).

In fact, it has long been known that the

characteristics of calcium phosphates are

related to their preparation methods.

Brown & Chow (1986) developed the first

formulation of LTCaP and, since then,

different formulations have been used for

the production of effective grafting materi-

als (Ambard & Mueninghoff 2006). For

instance, the pore sizes of conventional

calcium phosphate blocks range from a

nanoscale to 10 mm with limited intercon-

nectivity. Therefore, cellular invasion

through the scaffold is constrained and

cell–implant interactions are restricted to

their surface (Apelt et al. 2004). However,

Chow (2000) incorporated macropores into

the matrix of calcium phosphate cements

by the addition of large polyssacharides.

After the cements hardened, they were

soaked in an aqueous solution to dissolve

the polyssacharides, which left large, inter-

connected voids within the material struc-

ture (Chow 2000). Alternative approaches

have been proposed by del Real et al. (2003)

using gas flux and more recently by Tas

(2008) using spherical, porous, CaP gran-

ules produced with NaCl as the porogen.

While the above-mentioned authors

have used different strategies to increase

the biomaterial surface area, the present

work uses a low-temperature calcium

phosphate particulate (LTCP), produced

through a novel self-setting cement pro-

cess, as an alternative bone substitute to

harness the known advantages of increase

in surface area (Araujo 2008), increased

bone formation (Ginebra et al. 2006), and

biodegradation rate (Tamimi et al. 2006,

2007, 2008). Thus, we compared the in

vivo outcomes of LTCP with those of

anorganic bovine bone (AB), bioactive glass

(BG), and demineralized bone matrix

(DBM) in a rat model. We show that

LTCP invokes improved osteogenesis, but

also allows its resorption and substitution

during bone remodelling.

Material and methods

Preparation of LTCP

Calcium phosphate cement was obtained

using the same chemistry proposed by

Brown & Chow (1986). An equimolar

mixture of two calcium phosphate powders

with a phosphate solution was produced,

with a powder to liquid ratio fixed at 2 mg/

ml. The powder reagents were dicalcium

phosphate anhydrous (CaHPO4 – DCPA),

tetracalcium phosphate (Ca4(PO4)2O –

TTCP), and the liquid phase was a 0.2 M

solution of Na2HPO4 and NaH2PO4 in a

4 : 1 ratio (vol/vol).

In order to obtain more reactive cements,

commercial ultrapure DCPA (Sigma, St

Louis, MO, USA) and TTCP with purity

495% (Clarkson Chromatography Pro-

ducts Inc., South Williamsport, PA, USA)

were wet-ground with 100% ethanol at a

vibrational frequency of 30 rps for 25 and

20 min, respectively, using a mixer mill

machine (MM301, Retsch, Burlington,

ON, Canada). The resultant TTCP and

DCPA pastes were allowed to dry for 24 h

in a fume hood and subsequently stored in

a desiccator at 601C. The final particle size

ranges were 0.6–20mm for DCPA and 2–

35 mm for TTCP.

Next, 0.068 g of DCPA and 0.183 g of

TTCP were mixed with 125ml of the

neutral phosphate solution using a 45 ml

zirconium oxide jar with a 20 mm zirco-

nium oxide ball, for 30 s at a vibrational

frequency of 20 rps. The final paste-like

cement was allowed to set for 24 h in the

incubator at 371C, with 5% CO2 and

100% relative humidity. It was subse-

quently ground manually into particles,

using a pestle and mortar, and sieved into

a size range of 90–355 mm. The final LTCP

was stored in glass vials, labelled and

g-sterilized at 2.5 Mrads 60Co.

Other tested biomaterials

Commercial brands of particulate AB

(BioOsss

, Geistlich Biomaterials, Wolhu-

sen, CH, Switzerland) with a particle size

range of 200–1000 mm, BG particles (Bio-

Grans

, Biomet-3i, Palm Beach Gardens,

Araujo et al �Low-temperature particulate calcium phosphates

c� 2010 John Wiley & Sons A/S 633 | Clin. Oral Impl. Res. 21, 2010 / 632–641

FL, USA) with sizes ranging from 300 to

350mm, and granulated DBM provided in a

lipid carrier (Allogenixt, Biomet Micro-

fixation, Jacksonville, FL, USA) were also

tested in this study.

Scanning electron microscopy (SEM)

Qualitative surface analysis was conducted

for AB, BG, and DBM using a Hitachi

S3400 model SEM (Oakville, ON, Canada)

with a resolution of 5 nm and an accelerat-

ing voltage of 5 kV, without any sputter

coating. On the other hand, a quantitative

analysis of LTCP was conducted because it

represents a new compound that needs

complete characterization. For this pur-

pose, a Hitachi S-5200 model of field-

emission scanning electron microscope

with a resolution of approximately 1.8 nm

and an accelerating voltage of 5 kV was

used. The particulate material was car-

bon-coated and affixed to an aluminum

stub using conductive carbon paint (Struc-

ture Probe Inc., West Chester, PA, USA).

Surface area, porosity, and pore size

Gas adsorption (Autosorbs

-1, Quanta-

chrome Instruments, Boynton Beach, FL,

USA) was used to analyze the surface area

of LTCP, AB, and BG through the BET

(Brunauer–Emmett–Teller) method (Bru-

nauer et al. 1938) while pore volumes

were calculated using the BJH (Barret–

Joynet–Halenda) method (Barret et al. 1951).

Specifically, adsorption analysis was con-

ducted by adding successive volumes of

nitrogen gas to the samples cell (at

� 1961C), followed by the measurement

of the equilibrium pressure (P0). Next,

desorption of the samples was obtained

through successively lowering the pressure

of the cell (P) while gas removal was

quantified. Finally, qualitative and quanti-

tative data are obtained by the resulting

isotherms (Gregg & Sing 1982).

Powder X-ray diffraction (PXRD)

Cement reaction kinetics were analyzed

using PXRD. Hand-made CaP cement

ground into particles was also prepared to

serve as control group. Briefly, these ce-

ments were produced using 0.068 g of

DCPA and 0.183 g of TTCP, which were

placed in a sterilized glass dish and mixed

with a glass stirring-rod for 30 s. Next,

125mL of the neutral phosphate solution

was added to the powders and the phases

were mixed thoroughly for 3 min to a

paste-like consistency. The final cement

was allowed to set for 24 h in the incubator

at 371C, with 5% CO2 and 100% relative

humidity. It was subsequently ground into

particles, using a pestle and mortar, and

sieved into a size range of 90–355 mm. Two

grams of hand-made CaP particles and

LTCP were stored in 1.5 ml eppendorf

tubes, which were frozen in liquid nitrogen

(�801C) at 1, 3, 5, and 10 days, and

weekly from 1 to 8 weeks (n¼ 3). The

freezing of the samples aimed at the arrest-

ing of the cement reaction at specific time

points.

The samples were run using a normal

front packing technique on a Bruker AXS

D8 Discover Microdiffraction system (Ma-

dison, WI, USA) with CuKa point-focus

X-ray source operating at 40 kV/40mA.

The system was equipped with a curved

primary graphite monochromator and two-

dimensional (2D) proportional detector.

Data were collected on two frames at

1200 s exposure, which cover a range of

20–581 and the 2D diffraction images were

then integrated with a step size of 0.0051 2yand converted to standard I vs. 2y diffraction

patterns. The phase identification was per-

formed by Diffrac Plust data processing

software Evat v. 8.0 and Search/Matcht

routine. This qualitative phase composition

was established by comparison with the

reference PXRD patterns stored in the Uni-

versity of Toronto Chemistry Department

PDF-2 Database (2001) via Search-Matcht

software. The profile fitting applications

such as Rietveld refinement, quantitative,

and microstructure analyses were performed

using Topast v. 2.1 software.

Surgical procedure and sample harvesting

The surgical protocol was approved by the

Local Animal Care Committee of the

Faculty of Dentistry at the University of

Toronto. Forty male Wistar rats (200–

250 g, Charles River Laboratories, Senne-

ville, QC, Canada) were anesthetized by

inhalation of isofluorane in nitrous oxide

and oxygen (900 ml total flow rate – 4% for

induction and 2% as maintenance). Bupre-

norphine (0.01–0.05 mg/kg) was adminis-

tered subcutaneously as analgesic both

before and after the surgery. To ensure

the same volume of test biomaterial was

implanted at each surgical site, custom-

made stainless-steel syringes with end

tips matching the diameter of the defects

were used to deliver the implants into the

defects. Following placement, a Teflon

membrane was used to cover the bony

defects to prevent the washing out of the

particles by blood from the marrow cavity.

In this manner, 10 mm3 of LTCP, AB, BG,

or DBM (n¼5 samples/group/time-point)

were randomly implanted bilaterally in

single bone defects created in the distal

femoral metaphyses using a 2.3 mm round

dental bur under constant saline irrigation.

After implantation, the muscle layer was

repositioned and sutured (4-0 Polysorbt,

Norwalk, CT, Syneture, USA), and the

skin was closed with 9 mm wound clips

(Becton Dickinson, MD, USA).

During the post-operative period, the

animals were monitored on a daily basis

in the animal care facility of the Faculty of

Dentistry of the University of Toronto.

Five animals were euthanized by cervical

dislocation after CO2 exposure at each of

the 3, 6, 12, and 16 weeks post-operation.

The femora were harvested, trimmed, and

fixed in 10% neutral formalin for at least

24 h.

Micro-computed tomography (MicroCT)

MicroCT scans of each femur were con-

ducted with a microtomography system

(MicroCT40, Scanco Medical, Basserdorf,

Switzerland). All trimmed femoral samples

were stacked (in groups of six) in a cylind-

rical holder (8.5 mm inner diameter,

43 mm height) and scanned at 70 kV and

114 mA. The specimens were scanned in

high-resolution mode.

The final 2D images were composed of

500–700 axial-cut slices, each one with

6 mm in thickness. After scanning and

reconstruction, a standardized region of

interest (ROI) comprising the defect with

the implant and surrounding reparative

bone within the marrow tissue was drawn

at different depths (approximately every 50

slices) of the 2D dataset, so that the final

drawings could be morphed and render a

3D ROI. The quantification of both bone

and particulate material volumes was

possible through their segmentation in dif-

ferent threshold values, which were deter-

mined by the gray-level distribution. In

addition, the values obtained were divided

by an averaged defect size (10 mm3), so that

the measurements were consistently re-

lated to the volume of the initial defect.

Araujo et al �Low-temperature particulate calcium phosphates

634 | Clin. Oral Impl. Res. 21, 2010 / 632–641 c� 2010 John Wiley & Sons A/S

Because the DBM putty was not visible

in the MicroCT images, the histomorpho-

metric evaluation of this group was done

using histology slides. Two slides per sam-

ple per interval were selected and the putty

area was measured using the Image J 1.37v

software (National Institute of Health,

Bethesda, CT, USA) (Abramoff et al.

2004). However, bone volume measure-

ments for this group were conducted using

the MicroCT thresholds as described

above.

Histological analysis

The most representative samples for each

interval observed in the MicroCT analysis

were chosen for histological processing.

The samples were initially dehydrated at

room temperature in increasingly concen-

trated ethanol solutions (70, 95, 100 v/v).

Then, they were embedded in polymethyl-

methacrylate (Osteo-Bed, Polysciences,

Warrington, PA, USA). Thin sections of

the specimens (30–40mm) were produced

by the Exakt cutting and grinding system

(Exakt Technologies Inc., Oklahoma City,

OK, USA) and stained by toluidine blue.

Images in different magnifications were

retrieved using a Leitz Aristoplan micro-

scope, Q imaging Micropublisher 5.0 RTV

digital camera, and Openlab 4.0.4 software

(Improvision Ltd, Coventry, West Mid-

lands, UK).

Statistical analyses

Statistical analyses were performed using

the computer software Statistical Analysis

System (SAS Institute Inc., Cary, NC,

USA). Linear generalized estimating equa-

tions (GEE) were used to examine the

effects of treatment, time, and treatment

over time on bone and particulate volume.

GEE were used rather than traditional ana-

lysis of variance due to the correlated

nature of the data (treatments were applied

bilaterally on each rat) (Liang & Zeger

1986). A Bonferroni correction was used

when conducting post hoc analyses of sta-

tistically significant main effects. For each

model, residuals were analyzed for normal-

ity, and when found to be non-normal,

sensitivity analyses were conducted re-

moving outliers to evaluate the impact on

findings.

Differences among the HA conversion of

hand-made CaP particles and LTCP were

compared using ANOVA analysis of var-

iance followed by Tukey adjustment for

multiple comparisons. P-values o0.05

were considered significant.

Results

Qualitative SEM and histological analyses

Figure 1 shows SEM micrographs of LTCP,

AB, BG, and DBM as received. A qualita-

tive analysis of LTCP revealed more com-

plex features than those of AB and BG.

A topographical comparison between

DBM putty and LTCP was not feasible,

because the former is presented as demi-

neralized bone particles embedded in a lipid

carrier and therefore displays a smooth

surface.

Figure 2 shows histological sections of

LTCP, AB, BG, and DBM at 3 and 16

weeks. A qualitative histological evalua-

tion depicted reparative bone in close con-

tact to all samples at 3 weeks. This was

identified as woven bone, as could be ex-

pected in the rat, due to the large, randomly

arranged, osteocyte lacunae. Small marrow

spaces were observed in the peri-implant

bone, which were populated by neutro-

phils, round mononuclear polarized cells,

and large multinucleated cells suggestive of

osteoclasts. These multinucleated cells

were more often seen around LTCP than

AB, BG, and DBM. Defects filled with

LTCP showed complete remodelling of

the reparative bone resulting in restitution

of the marrow tissue at 16 weeks. Repaired

marrow consisted of colonies of small cells

dispersed among round (fat) cells. There

was no evidence of inflammatory or osteo-

clast-like cells.

Using visual inspection, a comparison

between histological and SEM data showed

the maintenance of the shape and size of

the particulate AB and BAG until the 16-

week mark, demonstrating little sign of

long-term resorption (Figs 1 and 2). Histo-

logical sections of DBM displayed separated

masses of demineralized bone particles sur-

rounded by a thin layer of bony tissue, with

little infiltration of cells (Fig. 2).

Reaction kinetics analysis of hand-madeCaP particulate and LTCP

X-ray diffraction analysis of hand-made

CaP particulate and LTCP confirmed HA

as the major final component of the ce-

ment reaction (approximately 51% and

67%, respectively). In LTCP, the residuals

of DCPA and TTCP were greater than the

Fig. 1. Scanning electron micrographs (SEM) of ‘as received’ (a) low-temperature calcium phosphate particulate

(LTCP), (b) anorganic bovine bone (AB), (c) bioactive glass (BG), and (d) demineralized bone matrix (DBM). The

particulates have sizes ranging from 90 to 355 mm (LTCP), 300 to 350 mm (BG), and 200 to 1000mm (AB). The

demineralized bone particles of DBM are delivered in a lipid carrier, and are thus not visible as individual

particles by SEM.

Araujo et al �Low-temperature particulate calcium phosphates

c� 2010 John Wiley & Sons A/S 635 | Clin. Oral Impl. Res. 21, 2010 / 632–641

second (approximately 4%) b-TCP end

product (Table 1). Thus, LTCP shows an

increased and more consistent HA conver-

sion than hand-made samples, which may

be a result of the mechanized activation

employed in the preparation process of

LTCP (Fig. 3).

Surface area, porosity, and pore size

Table 2 shows the surface area and pore

volume analyses for LTCP, AB, and BG.

LTCP presented higher surface area

(77.2 m2/g) than AB (61.14 m2/g) and BG

(0.231 m2/g). Similarly, an increased

microporosity was found for LTCP (14.61

cm3/g) compared with AB (6.71 cm3/g) and

BG (0.036 cm3/g). On the other hand, AB

showed higher pore volumes (0.36 cm3/g

for pores diameter smaller than 768 A at

maximum pressure¼ 0.974 atm.) com-

pared to LTCP (0.11 cm3/g for pores dia-

meter smaller than 772.8 A at maximum

pressure¼ 0.974 atm.) and BG (0.000413

cm3/g for pores diameter smaller than

848 A at maximum pressure¼ 0.977 atm.).

LTCP increased surface area could be ex-

plained by its small particle size range

allied with high pore interconnectivity gen-

erated by the interlocking of HA crystals

during cement setting (Ginebra et al.

2006). The decreased surface area of BG

beads can be attributed to its solid compo-

sition with absent internal porosity while

the surface area and pore volume values

found for AB are a result of its bovine HA

architecture.

Reparative osteogenesis and particulatedegradation

The post-operative period was uneventful

for all animals, except for two animals that

had fractured femora (AB 16 weeks and

DBM 3 weeks). However, the defects of

contralateral legs were still included in this

study (DBM 16 weeks and AB 3 weeks).

MicroCT slices of the bone and particulate

volumetric measurements for LTCP, AB,

BG, and DBM at 3 and 16 weeks are shown

in Fig. 4. LTCP and BG showed increased

values of reparative bone volume at 3

weeks post-operative, which were statisti-

cally significant compared with AB

(P¼ 0.0005 for LTCP and P¼ 0.0014 for

Table 1. Powder X-ray diffraction of hand-made CaP particulate and low-temperaturecalcium phosphate particulate (LTCP)

Hand-made CaP particulate LTCP

HA DCPA TTCP b-TCP HA DCPA TTCP b-TCP

1d 46.6 32.5 12.7 8.3 79.7 8.1 8.5 3.73d 44.6 30.6 10.1 14.7 81.7 8.2 6.3 3.95d 45.4 34.4 7 13.3 75.8 8.2 15.1 0.91w 67.7 20.4 10.4 1.6 73.1 8.5 17 1.410d 55.4 27 4.1 13.5 78.6 4.7 13.3 3.42w 57.6 23.8 3.9 14.7 72.9 10.1 13.5 4.23w 62.3 21.7 3.3 12.7 71.1 10.4 14.1 4.44w 63.7 23.9 3.9 8.5 73.7 13.8 9 3.45w 58.7 25.4 3.7 12.2 67.7 12.2 15.3 4.86w 55.7 27.6 5.2 11.4 80.6 8.1 9 2.37w 57.6 24 4.8 12.3 84.7 3.6 9.1 2.78w 50.6 31.3 5.5 12.6 67.2 16 12.7 4.1

LTCP reaction occurred faster and resulted in increased HA conversion with decreased amounts of

residual DCPA and b-TCP.

DCPA, dicalcium phosphate anhydrous; TTCP, tetracalcium phosphate; HA, hydroxyapatite; b-TCP,

b-tricalcium phosphate; d, days; w, weeks.

Fig. 2. Toluidine blue-stained resin-embedded sections of low-temperature calcium phosphate particulate (LTCP), anorganic bovine bone (AB), bioactive glass (BG), and

demineralized bone matrix (DBM) at 3 (a, b, c, and d) and 16 weeks (e, f, g, and h). (a) At 3 weeks bone formed an intimate contact with the particulate material of LTCP.

However, the latter was no longer visible, at the same magnification, at 16 weeks due to its almost complete degradation (not shown). However, at higher magnification

(e) it was still possible to see some evidence of residual particles enveloped in cortical bone (see also Fig. 4e). Conversely, reparative trabecular bone was very evident at

both 3 and 16 weeks for the AB, BG, and DBM samples. The lack of obvious differences in the trabecular bone volume at 3 and 16 weeks is evidence of the lack of

degradation of each of these materials in this site. Field widths (a, b, c, d, f, g, h)¼ 810mm; (e)¼142.3 mm.

Araujo et al �Low-temperature particulate calcium phosphates

636 | Clin. Oral Impl. Res. 21, 2010 / 632–641 c� 2010 John Wiley & Sons A/S

BG). Higher bone volumes were found for

BG compared with DBM at 6 weeks

(P¼ 0.0002) and AB at 12 weeks

(P¼ 0.0006). LTCP displayed lower bone

volume compared with all other bio-

materials (Po0.0001) at 12 weeks post-

operative. Moreover, the bone surrounding

LTCP had almost completely remodelled

with restoration of the marrow compart-

ment at 16 weeks post-operative, whereas

the other materials still presented increased

bone volumes within the implantation site

(Po0.0001) (Fig. 5).

Particulate volumes of AB and BG were

statistically higher than those of LTCP at

all intervals. Similar volumes of AB and

BG were observed at 3 (P¼ 0.8051) and 6

weeks (P¼ 0.1646) post-operative. How-

ever, increased amounts of AB and BG

compared with LTCP were found at 3

and 6 weeks (Po0.0001). A slow particle

degradation was observed for AB and BG at

12 weeks post-operative (P¼ 0.0115),

whereas LTCP had almost completely re-

sorbed (Po0.0001). Similarly, increased in

situ volumes of AB and BG were seen at 16

weeks post-operative, while LTCP had

completely remodelled. Little sign of re-

sorption was found for AB particles until

the 12th week mark, time at which it

presented a decrease of 8.6% in volume.

From 12 to 16 weeks post-operative, AB

implants were unchanged (resorption of

8.6% in volume from 3 to 16 weeks post-

operative). BG beads showed a decrease of

23.5% in volume from 3 to 6 weeks post-

operative, degradation which decreased to

11.5% from 6 to 12 weeks post-operative.

Similarly to AB, the volume of BG was

kept the same from 12 to 16 weeks post-

operative (volume degradation of 32.4%

from 3 to 16 weeks post-operative). LTCP

presented a decrease of particulate volume

of 33.3%, 75%, and 100% from 3 to 6, 6 to

12, and 12 to 16 weeks, respectively (Fig.

6). Results of area measurements for DBM

showed increased putty values at the final

interval studied and a decreased degrada-

tion rate from 3 to 16 weeks (Fig. 7). This

material was kept un-resorbed from 3 to 6

weeks, but resorbed 28% from 6 to 12

weeks and 16.7% from 12 to 16 weeks

(resorption of 37.5% in area from 3 to 16

weeks post-operative).

Discussion

The concept of producing granulated

LTCaP for bone regeneration has been

explored by Tamimi et al. (2006, 2007,

2008). However, the histomorphometric

analysis for these studies was obtained

from six randomly chosen histological

slices per sample and evaluated the total

mineral density of the implantation site at

4 weeks (Tamimi et al. 2006, 2007) and 4

and 8 weeks post-operative (Tamimi et al.

2008). Traditional tissue evaluation using

histological 2D sections has been broadly

applied in the bone biomaterial field due to

its ability to generate high-resolution

images, allowing for the analysis at the

cell level. However, quantitative histomor-

phometric analysis of histological sections

may result in inaccurate values of the 3D

material measured (Muller et al. 1998). On

the other hand, MicroCT evaluation does

not render the imaging of cellular composi-

tion, but represents a non-invasive techni-

que, which enables a precise 3D analysis of

bone architecture and healing. In the dental

field, MicroCT has been used mainly for

the characterization of bone architecture,

the quantification of bone–implant con-

tact, and the analysis of dynamic condi-

tions, such as bone formation and

resorption rates (Rebaudi et al. 2004).

Therefore, the correlated histological (qua-

litative) and MicroCT (quantitative) eva-

luations of this work aimed at a thorough

analysis of LTCP, AB, BG, and DBM

in vivo effects.

It should be emphasized that complete

healing of the rat femora model used

herein, where there is initially no trabecu-

lar bone, requires repair of the cortex,

regeneration of the marrow cavity and

marrow tissue, and an absence of reparative

trabecular bone. Thus, while we could

observe that AB, BG, and DBM invoked

bone formation, the presence of high vo-

lumes of these materials within the im-

plantation site at 16 weeks, together with

surrounding reparative trabecular bone, is

evidence of their poor in vivo resorbability.

Moreover, SEM and histological analyses

showed only small structural changes in

AB, BG, and DBM throughout the study.

Fig. 3. Temporal HA conversion of hand-made CaP particulate and LTCP (n¼ 3). The mechanical activation

proposed for low-temperature calcium phosphate particulate (LTCP) may have rendered this material more

reactive because it presented a faster and increased HA conversion than hand-made CaP particulate samples.

Table 2. Surface area and pore volume analyses for low-temperature calcium phosphateparticulate (LTCP), anorganic bovine bone (AB), and bioactive glass (BG)

Surface area (m2/g) Microporosity (cm3/g) Pore volume (cm3/g)

LTCP 77.22 14.61 0.11AB 61.14 6.71 0.36BG 0.231 0.036 0.000413

The BJH method allowed for the measurement of pores diameter smaller than 768 A at 0.974 atm.

maximum pressure for AB, pores diameter smaller than 773 A at 0.974 atm. maximum pressure for

LTCP, and pores diameter smaller than 848 A at 0.977 atm. maximum pressure for BG. LTCP presented

higher surface area and microporosity than AB and BG. AB showed higher pore volumes.

BJH, Barret–Joynet–Halenda.

Araujo et al �Low-temperature particulate calcium phosphates

c� 2010 John Wiley & Sons A/S 637 | Clin. Oral Impl. Res. 21, 2010 / 632–641

On the contrary, complete healing was

achieved in the case of LTCP where, by

16 weeks, the cortex, marrow cavity, and

marrow tissue were all completely regen-

erated, without any residual artificial ma-

terial remaining. Specifically, LTCP was

able to maintain its osteoconductivity even

after significant resorption without imped-

ing tissue turnover. In addition, this im-

proved resorption profile of LTCP may be

the result of the novel preparation methods

used to produce LTCaP, resulting in in-

creased surface area and microporosity (as

showed by BET and BJH analyses). These

characteristics, allied with the smaller par-

ticulate size, offered an increased surface

area for both bone deposition and osteo-

clastic activity.

The slow degradation rate of AB has been

attributed to its composition (bovine HA),

particle size (250–1000 mm) (Yildirim et al.

2000; Araujo et al. 2001), and to possible

structural changes caused by its heating

(approximately 4001C) preparation process

(Tadic & Epple 2004), whereas the degrada-

tion rate of BG depends on the hydrolysis

of the glass network, which disrupts the

granules and favors phagocytosis of its

silica-gel core (Lai 2000; Cordioli et al.

2001; Huygh et al. 2002). DBM purported

osteoinductivity derives from its growth

factor content [bone morphogenetic pro-

teins (BMPs)]. However, there is a great

variability reported concerning DBM

which may be a result of differences in

the total amount of BMPs remaining fol-

lowing processing and sterilization of the

material, carrier (Peterson et al. 2004),

particle size, porosity, and shape (Louis-

Ugbo et al. 2004). This variability is also

partly related to the bone donor. Thus,

Fig. 5. Bone volumes for low-temperature calcium phosphate particulate (LTCP), anorganic bovine bone (AB),

bioactive glass (BG), and demineralized bone matrix (DBM). AB, BG, DBM displayed higher bone volume

values than seen in the normal medullary cavity, with little sign of reparative bone remodelling. LTCP showed

similar reparative bone formation compared with the other biomaterials but an increased remodelling profile by

16 weeks.

Fig. 4. Micro-computed tomography images for bone and particulate volumes of low-temperature calcium phosphate particulate (LTCP), anorganic bovine bone (AB),

bioactive glass (BG), and demineralized bone matrix (DBM) at 3 (a, b, c, d) and 16 weeks (e, f, g, h). (a) At 3 weeks, LTCP was surrounded by trabecular bone within the

marrow cavity and cortical wound site. (e) At 16 weeks both the LTCP, and the surrounding trabecular bone, were completely remodelled; although some particulate

could still be seen enveloped in the healed cortical bone (arrows). The presence of both AB and BG particles, and their enveloping thin trabecular bone, is evident at both 3

and 16 weeks (b & c; f & g, respectively). Similarly, reparative trabecular bone formation is also evident at both 3 and 16 weeks for DBM samples (d and h). Although the

demineralized particulate was not radio-opaque, the presence of trabecular bone at 16 weeks in these samples suggests that residual DBM still exists in the marrow cavity.

Interestingly, while LTCP and DBM samples showed complete healing of the cortical bone (e and h), the margins of the cortical defects for AB and BG samples are still

visible at 16 weeks.

Araujo et al �Low-temperature particulate calcium phosphates

638 | Clin. Oral Impl. Res. 21, 2010 / 632–641 c� 2010 John Wiley & Sons A/S

characteristics such as age, health condi-

tions, drug therapy, and genetic alterations

could be related to donor–host site incom-

patibility (Haas et al. 2002), resulting in

the osteoinductivity of DBM being incon-

sistent (Schwartz et al. 1998).

During early peri-implant healing, osteo-

conductive materials stimulate the recruit-

ment and migration of osteogenic cells,

which deposit bone matrix onto their sur-

faces (Davies 1998; Davies & Hosseini

2000). Ideal bone biomaterials should sti-

mulate increased bone formation during

early endosseous healing but, over time,

should be remodelled by the joint activity

of osteoclasts and osteoblasts (Kohri et al.

1993). In this context, the long-term pre-

sence of bone substitutes in the implanta-

tion site may affect new bone growth and

interfere in the characteristics of the newly

forming tissue (Tamimi et al. 2006), possi-

bly by restricting the formation of new

bone through the granulation tissue (Stav-

ropoulos et al. 2004). Moreover, the total

reconstruction of the natural bone struc-

ture in an endosseous defect site permits

maximal functional tissue remodelling,

which may otherwise be compromised by

the presence of residual biomaterials

(Ooms et al. 2003).

Implantable LTCaPs are osteoconduc-

tive and show a faster resorption than

high-temperature synthesized ceramics

(Ohura et al. 1996). However, information

on their biological performance (i.e. disso-

lution, resorption times, and bone forma-

tion) varies due to inconsistencies in the

reagent chemistry and models used

(Bohner et al. 2005). These materials have

been the targets of several studies. A vast

majority of these reports have focused on

the in vitro evaluation of self-setting ce-

ments, cement blocks, and granules aim-

ing at their mechanical properties (Fukase

et al. 1990; Liu et al. 1997; Ambard &

Mueninghoff 2006; Tas 2008), drug release

capabilities (Bohner et al. 2000), as well as

the effects of powdered cement on cellular

function (Pioletti et al. 2000). On the other

hand, in vivo analyses have been conducted

on CaP blocks (Hamanishi et al. 1996),

injectable cements (Miyamoto et al. 1997,

1999; Ooms et al. 2003), and set cement

particles (Tamimi et al. 2006, 2007, 2008).

Several studies have reported changes in

the initial chemistry of LTCaP aiming at the

modulation of characteristics such as parti-

cle size, surface area, porosity, and pore

sizes. Varying (i) the powder to liquid ratio,

(ii) the particle sizes of the reagents, (iii) the

proportions of the starting powders, (iv)

the composition of the reactants, and (v)

the aqueous solution are all possible. Adding

nucleating agents (Bohner et al. 2005; Gi-

nebra et al. 2006), soluble crystals (Chow

2000; Tas 2008), or gas bubbles (del Real

et al. 2003) to the setting reaction are addi-

tional approaches. As part of the premise of

this work, it was anticipated that machine-

prepared LTCaPs would yield high bone

volumes but, with time, be remodelled and

allow for the complete regeneration of the

marrow tissue. In the current study, we used

a novel, machine-based preparation method

for the production of particulate LTCaP

because they exhibited: (i) a less variable,

and increased, HA conversion and (ii) a

more stable, and decreased, in vitro dissolu-

tion at pH 7.4 than their hand-made coun-

terparts (Araujo 2008).

Conclusion

The LTCP used invoked an increased early

bone formation and improved degradation

rate compared with three currently avail-

able commercial bone substitute materials.

Therefore, LTCP represents an easy and

reproducible means of producing effective

implantable calcium phosphates for bone

regeneration.

Acknowledgements: The authors

would like to acknowledge the Canada

Foundation for Innovation (CFI) and

the Ontario Research and Development

Fig. 6. Particle volumes for low-temperature calcium phosphate particulate (LTCP), anorganic bovine bone

(AB), and bioactive glass (BG). At all intervals, significantly smaller volumes were found for LTCP compared

with AB and BG. AB and BG showed little sign of degradation while LTCP was completely degraded by 16

weeks.

Fig. 7. Demineralized bone matrix area measurements. Increased values and a decreased degradation rate of the

material were found. As demineralized bone matrix (DBM) is not radio-opaque, these area measurements were

derived from histological sections, as explained in the text.

Araujo et al �Low-temperature particulate calcium phosphates

c� 2010 John Wiley & Sons A/S 639 | Clin. Oral Impl. Res. 21, 2010 / 632–641

Challenge Fund (ORDCF) for financial

support, together with an unrestricted

grant-in-aid of research from Biomet 3i,

who also kindly provided the

commercially available biomaterials

used in our experiments. M.V.F.A. was a

recipient of a University of Toronto

Scholarship. We are also grateful for

the technical assistance of Susan Carter

(animal care), Limin Guan (materials

processing), and Charles Victor

(statistical analysis).

References

Abramoff, M.D., Magelhaes, P.J. & Ram, S.J. (2004)

Image Processing with ImageJ. Biophotonics In-

ternational 11: 36–42.

Ambard, A.J. & Mueninghoff, L. (2006) Calcium

phosphate cement: review of mechanical and

biological properties. Journal of Prosthodontics

15: 321–328.

Apelt, D., Theiss, F., El-Warrak, A.O., Zlinszky, K.,

Bettschart-Wolfisberger, R., Bohner, M., Matter,

S., Auer, J.A. & von Rechenberg, B. (2004) In vivo

behavior of three different injectable hydraulic

calcium phosphate cements. Biomaterials 25:

1439–1451.

Araujo, M.G., Carmagnola, D., Berglundh, T., Thi-

lander, B. & Lindhe, J. (2001) Orthodontic move-

ment in bone defects augmented with BioOsss

.

An experimental study in dogs. Journal of Clin-

ical Periodontology 28: 73–80.

Araujo, M.V.F. (2008) Calcium phosphate cements

loaded with Pantoprazole as novel bone substitutes.

MSc Thesis, University of Toronto, Toronto.

Barret, E.P., Joyner, L.G. & Halenda, P.P. (1951)

The determination of pore volume and area dis-

tributions in porous substances. I. Computations

from nitrogen isotherms. Journal of the American

Chemical Society 73: 373–380.

Bohner, M., Gbureck, U. & Barralet, J.E. (2005)

Technological issues for the development of more

efficient calcium phosphate bone cements: a cri-

tical assessment. Biomaterials 26: 6423–6429.

Bohner, M., Lemaitre, J., Merkle, H.P. & Gander, B.

(2000) Control of gentamicin release from a calcium

phosphate cement by admixed poly(acrylic acid).

Journal of Pharmaceutical Sciences 89: 1262–1270.

Brown, W.E. & Chow, L.C. (1986) A new calcium

phosphate, water-setting cement. In: Brown, P.W.,

ed. Cements Research Progress, 352–379. Wester-

ville, OH: American Ceramic Society.

Brunauer, S., Emmett, P.H. & Teller, E. (1938)

Adsorption of gases in multimolecular layers.

Journal of the American Chemical Society 60:

309–319.

Chow, I.C. (2000) Calcium phosphate cements:

chemistry, properties, and applications. Materials

Research Society Symposium Proceedings 599:

27–37.

Cordioli, G., Mazzocco, C., Schepers, E., Brugnolo,

E. & Majzoub, Z. (2001) Maxillary sinus floor

augmentation using bioactive glass granules and

autogenous bone with simultaneous implant pla-

cement. Clinical and histological findings. Clin-

ical Oral Implants Research 12: 270–278.

Costantino, P.D., Friedman, C.D., Jones, K., Chow,

L.C. & Sisson, G.A. (1992) Experimental hydro-

xyapatite cement cranioplasty. Plastic and Recon-

structive Surgery 90: 174–185.

Davies, J.E. (1998) Mechanisms of endosseous in-

tegration. International Journal of Prosthodontics

11: 391–401.

Davies, J.E. & Hosseini, M.M. (2000) Histody-

namics of endosseous wound healing. In: Davies,

J.E., ed Bone Engineering, 1–13. Toronto: em2.

del Real, R.P., Ooms, E., Wolke, J.G., Vallet-Regi,

M. & Jansen, J.A. (2003) In vivo bone response to

porous calcium phosphate cement. Journal of

Biomedical Materials Research Part A 65: 30–36.

Farina, C. & Gagliardi, S. (1999) Selective inhibitors

of the osteoclast vacuolar proton ATPase as novel

bone antiresorptive agents. Drug Discovery Today

4: 163–172.

Friedman, C.D., Costantino, P.D., Jones, K., Chow,

L.C., Pelzer, H.J. & Sisson, G.A (1991) Hydro-

xyapatite cement II: obliteration and reconstruc-

tion of the cat frontal sinus. Archives of

Otolaryngology: Head and Neck Surgery 117:

385–389.

Fukase, Y., Eanes, E.D., Takagi, S., Chow, L.C. &

Brown, W.E. (1990) Setting reactions and

compressive strengths of calcium phosphate

cements. Journal of Dental Research 69:

1852–1856.

Gautier, H., Daculsi, G. & Merle, C. (2001) Asso-

ciation of vancomycin and calcium phosphate by

dynamic compaction: in vitro characterization

and microbiological activity. Biomaterials 22:

2481–2487.

Ginebra, M.P., Traykova, T. & Planell, J.A. (2006)

Calcium phosphate cements as bone drug delivery

systems: a review. Journal of Controlled Release

113: 102–110.

Gregg, S.J. & Sing, K.S.W. (1982) Adsorption, Sur-

face Area, and Porosity. 2nd edition. London:

Academic Press.

Haas, R., Haidvogl, D., Donath, K. & Watzek, G.

(2002) Freeze-dried homogeneous and heteroge-

neous bone for sinus augmentation in sheep. Part

I: histological findings. Clinical Oral Implants

Research 13: 396–404.

Hamanishi, C., Kitamoto, K., Ohura, K., Tanaka, S.

& Doi, Y. (1996) Self-setting, bioactive, and

biodegradable TTCP–DCPD apatite cement. Jour-

nal of Biomedical Materials Research 32:

383–389.

Huygh, A., Schepers, E.J.G., Barbier, L. & Duch-

eyne, P. (2002) Microchemical transformation of

bioactive glass particles of narrow size range, a

0–24 months study. Journal of Materials Science:

Materials in Medicine 13: 315–320.

Kohri, M., Miki, K., Waite, D.E., Nakajima, H. &

Okabe, T. (1993) In vitro stability of biphasic

calcium phosphate ceramics. Biomaterials 14:

299–304.

Lai, W. (2000) Bioactive glass degradation products:

distribution and pathway removal. Ph.D. Disser-

tation, University of Pennsylvania, Philadelphia.

LeGeros, R.Z. & LeGeros, J.P. (2003) Calcium

phosphate bioceramics: past, present and future.

Key Engineering Materials 240–242: 2–10.

Liang, K.Y. & Zeger, S.L. (1986) Longitudinal data

analysis using generalized linear models. Biome-

trika 73: 13–22.

Liu, C., Shen, W., Gu, Y. & Hu, L. (1997) Mechan-

ism of the hardening process for a hydroxyapatite

cement. Journal of Biomedical Materials Re-

search 35: 75–80.

Louis-Ugbo, J., Murakami, H., Kim, H.S., Mina-

mide, A. & Boden, S.D. (2004) Evidence of

osteoinduction by Grafton demineralized bone

matrix in nonhuman primate spinal fusion. Spine

29: 360–366.

Malard, O., Bouler, J.M., Guicheux, J., Heymann,

D., Pilet, P., Coquard, C. & Daculsi, G. (1999)

Influence of biphasic calcium phosphate granulo-

metry on bone ingrowth, ceramic resorption, and

inflammatory reactions: preliminary in vitro and

in vivo study. Journal of Biomedical Materials

Research 46: 103–111.

Miyamoto, Y., Ishikawa, K., Takechi, M., Toh, T.,

Yoshida, Y., Nagayama, M., Kon, M. & Asaoka,

K. (1997) Tissue response to fast-setting calcium

phosphate cement in bone. Journal of Biomedical

Materials Research 37: 457–464.

Miyamoto, Y., Ishikawa, K., Takechi, M., Toh, T.,

Yuasa, T., Nagayama, M. & Suzuki, K. (1999)

Histological and compositional evaluations of

three types of calcium phosphate cements when

implanted in subcutaneous tissue immediately

after mixing. Journal of Biomedical Materials

Research 48: 36–42.

Muller, R., Van Campenhout, H., Van Damme, B.,

Van der Perre, G., Dequeker, J., Hildebrand, T. &

Ruegsegger, P. (1998) Morphometric analysis of

human bone biopsies: a quantitative structural

comparison of histological sections and micro-

computed tomography. Bone 23: 59–66.

Ohura, K., Bohner, M., Hardouin, P., Lemaitre, J.,

Pasquier, G. & Flautre, B. (1996) Resorption of,

and bone formation from, new beta-tricalcium

phosphate–monocalcium phosphate cements: an

in vivo study. Journal of Biomedical Materials

Research 30: 193–200.

Ooms, E.M., Wolke, J.G.C., van de Heuvel, M.T.,

Jeschke, B. & Jansen, J.A. (2003) Histological

evaluation of the bone response to calcium phos-

phate cement implanted in cortical bone. Bioma-

terials 24: 989–1000.

Peterson, B., Whang, P.G., Iglesias, R., Wang, J.C. &

Lieberman, J.R. (2004) Osteoinductivity of com-

mercially available demineralized bone matrix.

Preparations in a spine fusion model. Journal of

Bone and Joint Surgery (American Volume) 86-A:

2243–2250.

Pioletti, D.P., Takei, H., Lin, T., Van Landuyt, P.,

Ma, Q.J., Kwon, S.Y. & Sung, K.L. (2000)

The effects of calcium phosphate cement particles

on osteoblast functions. Biomaterials 21:

1103–1114.

Araujo et al �Low-temperature particulate calcium phosphates

640 | Clin. Oral Impl. Res. 21, 2010 / 632–641 c� 2010 John Wiley & Sons A/S

Rebaudi, A., Koller, B., Laib, A. & Trisi, P. (2004)

Microcomputed topographic analysis of the peri-

implant bone. International Journal of Perio-

dontics and Restorative Dentistry 24: 316–325.

Schwartz, Z., Doukarsky-Marx, T., Nasatzky, E.,

Goultschin, J., Ranly, D.M., Greenspan, D.C.,

Sela, J. & Boyan, B.D. (2008) Bone graft substi-

tutes and marrow restoration. Clinical Oral Im-

plants Research 19: 1233–1245.

Schwartz, Z., Somers, A., Mellonig, J.T., Carnes,

D.L.Jr., Dean, D.D., Cochran, D.L. & Boyan,

B.D. (1998) Ability of commercial demineralized

freeze-dried bone allograft to induce new bone

formation is dependent on donor age but not

gender. Journal of Periodontology 69: 470–478.

Shindo, M.L., Costantino, P.D., Friedman, C.D. &

Chow, L.C. (1993) Facial skeletal augmentation

using hydroxyapatite cement. Archives of Otolar-

yngology: Head and Neck Surgery 119: 185–190.

Stavropoulos, A., Kostopoulos, L., Nyengaard, J.R.

& Karting, T. (2004) Fate of bone formed by

guided tissue regeneration with or without graft-

ing of BioOsss

or BioGrans

. An experimental

study in the rat. Journal of Clinical Perio-

dontology 31: 30–39.

Tadic, D. & Epple, M. (2004) A thorough physico-

chemical characterization of 14 calcium phos-

phate-based bone substitution materials in

comparison to natural bone. Biomaterials 25:

987–994.

Takagi, S. & Chow, L.C. (2001) Formation of

macropores in calcium phosphate cement im-

plants. Journal of Materials Science: Materials

in Medicine 12: 135–139.

Tamimi, F., Torres, J., Kathan, C., Baca, R., Clem-

ente, C., Blanco, L. & Lopez-Cabarcos, E. (2008)

Bone regeneration in rabbit calvaria with novel

monetite granules. Journal of Biomedical Materi-

als Research 87A: 980–985.

Tamimi, F.M., Torres, J., Tresguerres, I., Blanco, L.J.

& Lopez-Cabarcos, E. (2007) Vertical bone aug-

mentation with granulated brushite cement set in

glycolic acid. Journal of Biomedical Materials

Research 81A: 93–102.

Tamimi, F.M., Torres, J., Tresguerres, I., Clemente, C.,

Lopez-Cabarcos, E. & Blanco, L.J. (2006) Bone

augmentation in rabbit calvariae: comparative study

between BioOsss

and a novel b-TCP/DCPD granu-

late. Journal of Clinical Periodontology 33:

922–928.

Tas, A.C. (2008) Preparation of porous apatite gran-

ules from calcium phosphate cement. Journal of

Materials Science: Materials in Medicine 19:

2231–2239.

Yildirim, M., Spiekermann, H., Biesterfeld, S. &

Edelhoff, D. (2000) Maxillary sinus augmentation

using xenogenic bone substitute material

BioOsss

in combination with venous blood. A

histologic and histomorphometric study in hu-

mans. Clinical Oral Implants Research 11:

217–229.

Zhang, Y., Xu, H.H., Takagi, S. & Chow, L.C.

(2006) In-situ hardening hydroxyapatite-based

scaffold for bone repair. Journal of Materials

Science: Materials in Medicine 17: 437–445.

Araujo et al �Low-temperature particulate calcium phosphates

c� 2010 John Wiley & Sons A/S 641 | Clin. Oral Impl. Res. 21, 2010 / 632–641