j o u r n a l o f d e n t i s t r y 3 5 ( 2 0 0 7 ) 8 4 – 9 2

Comparative evaluation of the peri-implant bone tissuemineral density around unloaded titanium dental implants

Tonino Traini a, Marco Degidi a,b, Giovanna Iezzi a, Luciano Artese a, Adriano Piattelli a,*aDental School, University of Chieti-Pescara, Via F. Sciucchi 63, 66100 Chieti, Italyb Private Practice, Bologna, Italy

a r t i c l e i n f o

Article history:

Received 11 August 2005

Received in revised form

8 May 2006

Accepted 15 May 2006

Keywords:

Backscattered electron imaging

Bone density

Light microscopy

Mineral content

Mineral density

Peri-implant bone

Dental implants

Osseointegration

Acid fuchsine

Oral implants

Bone

a b s t r a c t

Objective: The mechanical properties of bone are greatly influenced by the percentages of

organic and mineral constituents. Nevertheless, the information about the mineral content

on a microscopic scale in peri-implant bone is scarce. The aim of this work was to analyze

the bone mineral density of peri-implant bone under different techniques.

Design: Five unloaded titanium dental implants with a micro-structured surface (three XiVE

plus and two Frialit 2, DENTSPLY-Friadent, Mannheim, Germany) were retrieved from the

mandible of five patients after a 6-month period. scanning electron microscopy with

backscattered electron signal (BSE), light microscopy (LM) with a double staining technique,

fluorescence microscopy and confocal laser microscopy were used for measuring micro-

scopic mineral content variations in peri-implant bone. Histomorphometry and image

intensity (grey level) were evaluated using a software package for image analysis.

Results: The low mineral density index (LMDI) for LM was of 29.2 � 3.1 (mean � S.D.), while

the high mineral density index (HMDI) was of 88.2 � 3.6 (mean � S.D.). The one-way ANOVA

analysis showed a significant difference (P < 0.001) among the groups. The pairwise Holm–

Sidak test identified the differences among HMDI indexes for both LM and SEM values and

also for cross-evaluation of the LMDI and HMDI values. The comparison between LMDI

indexes for both SEM and LM did not show any significance. The florescence microscopy

analysis showed clearly the difference between old (high mineralized) and new (low

mineralized) bone tissue near the implant surface. Under confocal laser microscopy the

same sections showed the area of bone modelling closest to implant surface.

Conclusion: In this study it was found that bone around unloaded implants showed a low

mineral density index under all the investigation methods used. It was also found that the

conventional LM technique with the double staining method was able to intensely stain the

bone area with a low mineral content.

# 2006 Elsevier Ltd. All rights reserved.

avai lable at www.sc iencedi rec t .com

journal homepage: www. int l .e lsev ierhea l th .com/ journals / jden

1. Introduction

The treatment of patients with prostheses supported by

endosseous implants has become a frequent restorative

option in clinical practice.1,2 The long term survival/success

* Corresponding author. Tel.: +39 0871 3554083; fax: +39 0871 3554076E-mail address: apiattelli@unich.it (A. Piattelli).

0300-5712/$ – see front matter # 2006 Elsevier Ltd. All rights reservedoi:10.1016/j.jdent.2006.05.002

of osseointegrated implants is a relevant matter for the

clinicians. The characteristics of the implant surface is

particularly important in the early phase of peri-implant bone

healing, and the bone tissue microstructure is mainly related

to the remodelling processes which take place around dental

.

d.

j o u r n a l o f d e n t i s t r y 3 5 ( 2 0 0 7 ) 8 4 – 9 2 85

implants.3–7 Rougher implant surfaces have been reported to

have higher bone-to-implant contact percentages than

machined surfaces.8–10

To achieve osseointegration after the implant placement,

the rougher surfaces seem to improve the de novo bone

formation due to the early surface adhesion of non-collage-

nous proteins like osteopontin and bone sialoprotein.11,12

Later on, calcium phosphate nucleation at the calcium binding

sites on these proteins continues the process of osseointegra-

tion, which is completed by crystal growth and collagen

production with mineralization.13,14 A histological study in

rabbits showed that the micro-fractures, which occur due to

the bone condensing technique, improve the peri-implant

bone formation in the first weeks by the release of cytokines.15

Moreover, the ability of bone to function effectively under

loading conditions depends upon two factors: the properties

and the spatial arrangement of the bone.

It is well known that the mechanical properties of the bone

tissue depend both on the mineral and the matrix constitu-

ents; the mineral phase strongly influences the stiffness and

breaking strain of bone: highly mineralized bone matrix is

stiffer and breaks earlier.16–18 In the cortical bones important

differences will occur when the ash content differences reach

approximately 3–4%.19–21 An increase of mineral (ash) content

was also correlated with enhanced fatigue life.22,23 Reilly and

Burstein24–26 showed that if bone is progressively decalcified,

the initial slope of the stiffness continuously decreases, while

the plastic deformation remains the same. They proposed that

the mineral content controlled the initial stiffness, while the

organic matrix determined the slope of the post-yield region.

Mineral content is a critical parameter affecting macroscopic

mechanical properties, and was found to have a significant

effect on both the nanoscale mechanisms of deformation and

nanomechanical properties.27

It is well established that small adjustments in mineral

content can significantly affect the mechanical properties of

bone,28–32 but the microscopic variation of mineral contents in

bone is difficult to measure. Microradiography (MR) was

developed to evaluate the mineral content at the microscopic

level in thin sections.33 However, the resolution of MR is

limited.34 The backscattered electron (BSE) evaluation has

important advantages in terms of volume resolution.35,36 The

BSE has been successfully applied to study plastic embedded

bone tissue.37–40 SEM is, however, a very costly and time-

consuming technique and so we thought useful to see if it was

possible to evaluate the mineral density of bone using only a

conventional staining under light microscopy.

This study was performed to evaluate the mineral content

of the peri-implant bone, in unloaded dental implants

retrieved from man, under light microscopy and scanning

electron microscopy.

2. Materials and methods

2.1. Specimens preparation

In this study five dental implants were used (three XiVE plus

and two Frialit 2, FRIADENT1, Dentsply, Friadent, Mannheim,

Germany); all the implants had been inserted in the mandible

of five patients (three females and two males). The ages ranged

from 30 to 68 years (median 46 years). All the implants had

been left unloaded for 6 months before retrieval. Written

informed consent to participate in the study, which was

approved by the local Ethics Committee, was obtained for all

patients. All the implants were retrieved with a 5 mm trephine

bur and fixed with 10% buffered formalin for 1 day and then

washed in sodium phosphate at pH 7.2. They were then

dehydrated in graded alcohols and embedded in LR white resin

(London Resin, Berkshire, UK). Undecalcified cut and ground

sections were prepared by using the Precise automated system

1 (Assing, Roma, Italy). Ten sections (two central sections from

each sample) were collected and ground to a final thickness of

30 � 5 mm (mean � S.D.) using a graded series (240–1200) of

silicone carbide grit papers under running water (Exakt

Apparatebau GmbH, Norderstedt, Germany) and were used

for the present investigation.

2.2. Back scattering electrons (BSE) analysis

Five sections were used for scanning electron microscopy

(SEM) analysis under BSE mode. The thin sections were

polished with 0.5 mm alumina to an optical finish and lightly

sputter-coated with gold in an Emitech K 550 (Emitech Ltd.,

Ashford, Kent, UK). The specimens were placed on the storage

of a SEM (LEO 435 Vp, LEO Electron Microscopy Ltd., Cam-

bridge, UK) equipped with tetra solid-state BSE detector. SEM

operating conditions included: 30-kV accelerating voltage;

15 mm working distance; 1.2 nA probe current. The images

(M � N 1024 � 768 grid of pixel) were captured in the BSE mode

with nine scans using a frame average technique. SEM

operating conditions were stored in computer memory and

restored prior to each image capturing. A total of seven images

were taken for each implant section at 117�magnification and

they were used for a map reconstruction using an image

managing software (Adobe Photoshop CS ver. 8.0.1).

2.3. Calibration of the gray levels

The BSE signal (gray scale) was calibrated using the ‘‘atomic

number (Z) contrast’’ of reference materials. Using this

method two different reference materials of known Z were

evaluated under defined conditions. The BSE-images were

calibrated using both carbon (Z = 6) (gray levels 25 � 1) and

aluminium (Z = 13) (gray levels 255 � 1) as reported by

Roschger et al.41 Under the same brightness and contrast

conditions, the BSE gray level was calibrated for bone mineral

density values as follows: the BSE gray levels ranging from 6 to

110 were assumed as standard for the bone with low mineral

density, while the BSE gray levels ranging from 111 to 230 were

used as standard for the bone with high mineral density.

2.4. Analysis of the SEM digital images

All images were calibrated using the software Image J 1.32j

(Wayne Rasband National Institute of Health, USA) applying

the Pythagorean Theorem for distance calibration which

reports the number of pixels between two selected points:

the scale bar was overimposed on each image by SEM. The

linear remapping of the pixel values was used to calibrate the

j o u r n a l o f d e n t i s t r y 3 5 ( 2 0 0 7 ) 8 4 – 9 286

intensity of images. From the inspection of the image

histograms and the correlation between the histology and

levels of mineral density with the ranges of the gray values, we

divided the image histograms into the following gray level

sectors:

� 0

–5 the position of the mean value for LR white resin within

bone (marrow spaces) and osteocytes lacunae;

� 6

–92 bone tissue with a low mineral density;

� 9

4–200 bone tissue with a high mineral density;

� 2

01–255 mean value for titanium.

For each field measured, the following data were recorded:

� T

he total area (mm2) in the field that might have contained

bone.

� T

he area (mm2) in the gray level range, defined by low or high

mineral density.

� T

he mean of the gray values of the area defined by low or

high mineral density.

2.5. Special procedure

The mineralization index was evaluated for each image using

different grey levels forming the region of interest (ROI). Low

mineral density index (LMDI) for SEM images was evaluated

with the following equation:

LMDISEM ¼X92

i¼6

Ai � X GLAt

where Ai was the bone area of interest with grey levels com-

prised among 6–92, X GL was the mean of the grey level range

considered in the area of interest, whileAt was the total area of

the image occupied by bone tissue without considering marr-

row spaces and osteocyte lacunae.

High mineral density index (HMDI) for SEM images was

evaluated using the following equation:

HMDISEM ¼X200

i¼94

Ai � X GLAt

where Ai was the bone area of interest with grey levels com-

prised among 94–200, X GL was the mean of the grey level

range considered in the area of interest, while At was the total

area of the image occupied by bone tissue without considering

marrow spaces and osteocytes lacunae.

2.6. LM analysis

Five sections mounted on glass microscope slides were double

stained with toluidine blueand acid fuchsin solutionsaccording

to the following staining method: 10 min in distilled water;

5 min in toluidine blue solution at room temperature; washing

under running water for 30 s; drying with paper; 3 min in acid

fuchsin solution at room temperature; rinsing under running

water for 30 s; drying with paper and mounting in oil.

The nuclei appeared blue, mineralized bone pale red or

unstained, osteoid and low mineralized bone red-purple. The

samples were investigated using an Axiolab microscope (Carl

Zeiss, Jena, Germany). The microscope was connected to a

digital camera (FinePix S2 pro, Fuji Photo Co. Ltd., Minato-ku,

Tokyo, Japan) interfaced to a monitor and PC (Intel Pentium IV

HT, Intel1, Santa Clara, CA, USA). This optical system was

associated with a software package with image capturing

capabilities (Image-Pro Plus 4.5, Media Cybernetics Inc.,

Immagini & Computer Snc, Milano, Italy). The digitized images

were stored in format JPEG with N �M = 3024 � 2016 grid of

pixels for a 24 bit. A total of seven digital images were taken for

each section at 110�magnifications and they were used for a

map reconstruction using an image managing software

(Adobe Photoshop CS ver. 8.0.1). The images obtained were

converted in gray scale at eight bit where each cell referred as a

pixel in the grid had assigned a value between 0 and 255

(0 = black; 255 = white).

2.7. Analysis of the LM digital images

All images were calibrated as previously described for SEM

digital images. Since there was a different appearance of the

structures under transmitted light and BSE by the inspection

of the image histograms, we divided the image histograms

into the following gray level sectors:

� 0

–35 mean value for titanium (at transmitted light it

appeared black);

� 3

6–110 bone tissue with a low mineral density;

� 1

12–232 bone tissue with a high mineral density;

� 2

33–255 mean value for LR white resin within bone (marrow

spaces) and osteocyte lacunae.

For each field measured, the following data were recorded:

� T

he total area (mm2) in the field that might have contained

bone.

� T

he area (mm2) in the gray level range, defined by low or high

mineral density.

� T

he mean of the gray values of the area defined by low or

high mineral density.

2.8. Special procedure

The mineralization indices were calculated as reported for

SEM images with the following equations:

LMDILM ¼X110

i¼36

Ai � X GLAt

HMDILM ¼X232

i¼112

Ai � X GLAt

where the bone areas of interest Ai were calculated for grey

levels ranging from respectively 36 to 110 for LMDI and 112 to

232 for HMDI.

2.9. Fluorescence and laser scanning microscopy analyses

A stained section was randomly chosen for further investiga-

tions with fluorescence microscopy (Nikon Eclipse E600, Nikon

Corporation, Chijoda-Ku, Tokio, Japan) equipped with a TRITC

j o u r n a l o f d e n t i s t r y 3 5 ( 2 0 0 7 ) 8 4 – 9 2 87

Table 1 – Summary of the data collected

Bone tissue extension Gray levels

At SEM (mm2) At LM (mm2) XGLAt SEM (mean � S.D.) XGLAt LM (mean � S.D.)

Total Area occupied by bone

114.65 � 105 262.80 � 105 86.4 � 33.5 126.6 � 43.1

Bone tissue extension Gray levels

Ai SEM (mm2) Ai LM (mm2) XGLAi SEM (mean � S.D.) XGLAi LM (mean � S.D.)

Area of interest for LMDI

60.56 � 105 93.64 � 105 60.1 � 21.7 81.8 � 17.3

Bone tissue extension Gray levels

Ai SEM (mm2) Ai LM (mm2) XGLAi SEM (mean � S.D.) XGLAi LM (mean � S.D.)

Area of interest for HMDI

57.35 � 105 163.32 � 105 115.7 � 15.2 147.4 � 28.0

filter and with confocal laser microscopy (LSM 510 META, Carl

Zeiss, Jena, Germany).

2.10. Statistical analysis

Statistical analysis of the results was performed using the

package Sigma Stat 3.0 (SPSS Inc., Ekrath, Germany) consider-

ing the mean variation of the mineral indices for BSE and LM

investigations divided for low and high mineral density

values. One-way ANOVA was used to assess the differences

among the treatment groups since the data passed both the

normality test (P = 0.115) and the equal variance test

(P = 0.911). The Holm–Sidak method was used for multi-

comparison procedures to identify the differences. A P-value

of under 0.05 was considered statistically significant.

3. Results

The mineralization indices were calculated according to the

reported equations using the data summarized in Table 1. The

Fig. 1 – Digitally map reconstruction of SEM BSE images at 117�a specific gray level (of 2–256 gray-level step) representing a ce

white arrows indicate the bone areas with low mineral density

mineralized. The osteocyte’s lacunae appear as black spots ins

indicate implant. Asterisk (*) indicates marrow spaces.

images detected by BSE signal (Fig. 1) showed a difference of

the gray level’s means (�S.D.) of 55.6 � 6.5 for low mineral

density versus high mineral density. The mean area of the

images related to bone tissue was of 114.65 � 105 mm2 with a

range of gray level from 6 to 200. The LMDI for SEM was

33.8 � 3.8 (mean � S.D.), while the HMDI was 57.7 � 5.6

(mean � S.D.). The images of conventional staining technique

(Fig. 2), after gray scale conversion (Fig. 3), demonstrate a

difference of the gray level’s means (�S.D.) of 65.6 � 10.7 for

low mineral density versus high mineral density. The area of

the images related to bone tissue was 262.80 � 105 mm2 with a

range of gray level from 40 to 248. The LMDI for LM was

29.2 � 3.1 (mean � S.D.), while the HMDI was 88.2 � 3.6

(mean � S.D.). The one-way ANOVA analysis summarized in

Table 2 showed a significant difference (P < 0.001) among the

mean values of the different groups. The pairwise multi-

comparison test (Table 3) with the Holm–Sidak method

identified the differences among HMDI indices for both LM

and SEM values and also for cross-evaluation between LMDI

and HMDI. The comparison between LMDI indices for both

SEM and LM did not show any significance. The fluorescence

of the bone around unloaded dental implant. Each pixel has

rtain mineral content of the corresponding bone area. The

while, the black arrows indicate bone matrix sites highly

ide the mineralized bone matrix. Double asterisks (**)

j o u r n a l o f d e n t i s t r y 3 5 ( 2 0 0 7 ) 8 4 – 9 288

Fig. 2 – Digitally map reconstruction of LM images at 120�of the bone around unloaded dental implant. Each pixel

has a specific gray level (of 2–256 gray-level step) for each

RGB colour. The section was stained with toluidine blue

and fuchsine acid. The white arrows indicate areas

intensely red stained belonging to low mineral density

bone while, the black arrows referring to unstained or pale

stained areas belonging to highly mineralized bone.

Asterisk (*) indicates marrow spaces. Double asterisks (**)

indicate implant. (For interpretation of the references to

color in this figure legend, the reader is referred to the web

version of the article.)

Fig. 3 – Digitally LM image at 400� of the bone around

unloaded dental implant. The colour image originally

stained with toluidine blue and fuchsine acid, after

converting in gray scale appeared made by each pixel of a

specific gray level (of 2–256 gray-level step). The white

arrows indicate dark-gray areas belonging to bone sites

with both high affinity to stains and low mineral density

while, the black arrows indicate light-gray areas belonging

to highly mineralized bone matrix with low affinity to

stains. Double asterisks (**) indicate titanium. Asterisk (*)

indicates marrow spaces.

microscopy analysis (Fig. 4) showed clearly the difference

between old (high mineralized) and new (low mineralized)

bone tissue near the implant surface. Under confocal laser

scanning microscopy the same sections showed the area of

bone remodelling closest to the implant surface (Fig. 5).

4. Discussion

The optimal mineralization density value for bone strength

has not yet been determined28,29,42–49; nevertheless, the peri-

implant bone density is very important since it is directly

related to the strength and the elastic modulus of the bone as

reported by Misch et al.50 Increasing mineralization density

increases the ability of bone to absorb impact energy,

although this relationship is not linear. Furthermore, a high

mineralized bone becomes increasing liable to fracture,

because microfractures can more readily propagate through

highly mineralized bone.51,52 When a load is applied to softer

bone there is a much higher strain compared to a denser bone,

and, as a result, there is more bone remodelling according to

the Frost Mechanostat Theory.53 Manz,54 in a prospective

human study, observed a relation between density of bone

and marginal bone loss next to the implant surface. The bone

remodelling process results in a non-uniform mineral

distribution in cortical and cancellous bone. Because remo-

delling takes place at the surface of trabeculae, the surface

layer generally has a lower mineral content than the

interstitial bone. There are many clinical reasons for

evaluating the quality of the peri-implant bone, and certainly

the most important is related to improving the dental

implants long-term prognosis. The clinical performance of

the osseointegrated dental implants is mainly related to the

ability of bone tissue to remain attached to the implant

surfaces placed in function. When the bone is loaded by

means of a screwed implant, stress is applied at the bone

implant interface and strain increases inside the bone

structure. The level of the stress and its concentration is

related to the implant shape and bone macroarchitecture

while the level of strain inside the bone tissue is related to the

microstructure of the tissue such as collagen fiber orientation

and mineral density.55–58 Bone remodelling is a physiological

process that adapts the bone tissue to local mechanical

needs; when the relationship between loads and bone

physiological mechanisms will be clearer we probably will

have very long lasting dental implants in function without

bone loss. Considering the implants evaluated in the present

study, we can note that unloaded implants after 6 months are

well integrated, but the mineral density, with both systems of

investigation, appeared mainly low in the bone tissue

interfacing the implant surfaces. High and low mineral

density areas inside the bone tissue are physiological and

are normally related to the different remodelling rates.

Certainly the knowledge about mineral density and load

conditions could be helpful from a clinical point of view in

order to establish the best clinical procedure in implant-

prosthetic dentistry, especially referring to the time sequence

and modality of load application. Not many data related to

peri-implant bone microstructure, such as collagen fiber

orientation and mineral density, are available in dental

j o u r n a l o f d e n t i s t r y 3 5 ( 2 0 0 7 ) 8 4 – 9 2 89

Table 2 – One-way ANOVA analysis

Group name N Missing Mean S.D. S.E.M.

BSE-LMDI 5 0 33.892 3.822 1.709

BSE-HMDI 5 0 57.748 5.652 2.528

LM-LMDI 5 0 29.264 3.115 1.393

LM-HMDI 5 0 88.256 3.642 1.629

Source of variation d.f. SS MS F P

Between groups 3 10960.130 3653.377 210.219 <0.001

Residual 16 278.063 17.379

Total 19 11238.193

The differences in the mean values among the treatment groups are greater than would be expected by chance; there is a statistically

significant difference (P = <0.001). Power of performed test with alpha = 0.050:1.000; normality test: passed (P = 0.115); equal variance test:

passed (P = 0.911).

Table 3 – Pair wise multiple comparison procedures (Holm–Sidak method): overall significance level = 0.05

Comparison Differences of means t Unadjusted P Critical level Significant

LM-HMDI vs. LM-LMDI 58.992 22.374 1.686E�013 0.009 Yes

LM-HMDI vs. BSE-LMDI 54.364 20.619 5.977E�013 0.010 Yes

LM-HMDI vs. BSE-HMDI 30.508 11.571 0.00000000348 0.013 Yes

BSE-HMDI vs. LM-LMDI 28.484 10.803 0.00000000928 0.017 Yes

BSE-HMDI vs. BSE-LMDI 23.856 9.048 0.000000108 0.025 Yes

BSE-LMDI vs. LM-LMDI 4.628 1.755 0.0983 0.050 No

literature probably due to the very costly and time-consum-

ing techniques usually necessary to obtain these results.55–58

The present results demonstrate the ability for the reported

double staining technique to provide very useful information

under LM of the level of mineral density simply by evaluating

the affinity to acid fuchsin. In fact, the comparison of LMDI

between the systems of investigation was not statistically

different and demonstrated very close values. The presence of

significant differences between LMDI and HMDI values within

each investigation system substantiates the different stain

Fig. 4 – (A) Fluorescent micrographs at 200� of the bone around

stained with toluidine blue and fuchsine acid were observed und

indicate implant. Asterisk (*) indicates marrow spaces. The blac

low mineral density index while, the white arrows indicate the

Magnification at 400� of the bone around unloaded dental impla

mineral density index and an intense stain affinity. Double ast

spaces.

affinity for acid fuchsin to bone with different level of

mineralization and also the ability of backscattered electrons

to detect information related to mineral density (Table 4).

Finally, the comparison for HMDI between the systems of

investigation resulted in obviously statistically differences

due to the different bone area extension 114.65 � 105 mm2 for

SEM versus 262.80 � 105 mm2 for LM. This aspect was not

secondary but it emphasizes the importance of the present

results since most of the bone area related to LMDI was

adjacent to the implant surface. Moreover, the time spent for

unloaded dental implant. The undemineralized specimens

er a laser light wavelength of 351 nm. Double asterisks (**)

k arrows indicate the bone areas intensely stained with a

bone matrix pale stained and highly mineralized. (B)

nt. The black arrows indicate newly formed bone with low

erisks (**) indicate implant. Asterisk (*) indicates marrow

j o u r n a l o f d e n t i s t r y 3 5 ( 2 0 0 7 ) 8 4 – 9 290

Fig. 5 – (A) Confocal laser micrograph at 400� of the bone around unloaded dental implant. The black arrows indicate the

bone areas with low mineral density while, white arrows indicate bone matrix sites highly mineralized. Double asterisks

(**) indicate implant. Asterisk (*) indicates marrow spaces in red. (B) Magnification at 450� of the rectangle reported in (A)

after adjusting the signals with filters. The implant appears in red (**). The marrow spaces appear in yellow-green (*). The

low mineralized bone appears in light red-orange (black arrows). The highly mineralized bone appears in dark red-orange

(white arrows). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of

the article.)

specimen preparation does not change, while the quality of

the information is greatly enhanced. Both high and low bone

mineralized areas were readily identified using this method.

Nevertheless, it is necessary to underline that the results

obtained were strongly dependent on the following controlled

variables: concentration of the fixative solution and pH, type of

resin used for specimen embedding, curing method, concen-

tration of stain solutions and time/temperature of action.

In conclusion, within the limitations of the present

study, we found that the bone at the interface of unloaded

dental implants after 6 months showed a low mineral

density index by all the investigation methods used. The

Table 4 – Mineral index plotted vs. investigation system

reported technique was able to stain with a significant

difference of intensity the bone areas with different contents

of mineralization.

Acknowledgments

This work was partially supported by the National Research

Council (CNR), Rome, Italy, by the Ministry of Education,

University, Research (MIUR), Rome, Italy, and by AROD

(Research Association for Dentistry and Dermatology),

Chieti, Italy.

r e f e r e n c e s

1. Meraw SJ, Eckert SE, Yacyshyn CE, Wollan PC. Retrospectivereview of grafting techniques utilized in conjunction withendosseous implant placement. The International Journal ofOral & Maxillofacial Implants 1999;14:744–7.

2. Wiens JL, Forte RA, Wiens JP. The use of distractionosteogenesis to treat hemifacial microsomia: a clinicalreport. Journal of Prosthetic Dentistry 2003;89:11–4.

3. Schwartz Z, Nasazky E, Boyan BD. Surface microtopographyregulates osteointegration: the role of implant surfacemicrotopography in osteointegration. The Alpha Omegan2005;98:9–19.

4. Botticelli D, Berglundh T, Persson LG, Lindhe J. Boneregeneration at implants with turned or roughsurfaces in self-contained defects. An experimentalstudy in the dog. Journal of Clinical Periodontology 2005;32:448–55.

j o u r n a l o f d e n t i s t r y 3 5 ( 2 0 0 7 ) 8 4 – 9 2 91

5. Davies JE. Understanding peri-implant endosseous healing.Journal of Dental Education 2003;67:932–49.

6. Ogawa T, Nishimura I. Different bone integration profiles ofturned and acid-etched implants associated withmodulated expression of extracellular matrix genes. TheInternational Journal of Oral & Maxillofacial Implants2003;18:200–10.

7. Hayakawa T, Yoshinari M, Nemoto K, Wolke JG, Jansen JA.Effect of surface roughness and calcium phosphate coatingon the implant/bone response. Clinical Oral Implants Research2000;11:296–304.

8. Novaes Jr AB, Papalexiou V, Muglia V, Taba Jr M. Influence ofinterimplant distance on gingival papilla formation andbone resorption: clinical-radiographic study in dogs.The International Journal of Oral & Maxillofacial Implants2006;21:45–51.

9. Schwartz Z, Nasazky E, Boyan BD. Surface microtopographyregulates osteointegration: the role of implant surfacemicrotopography in osteointegration. The Alpha Omegan2005;98:9–19.

10. Shalabi MM, Wolke JG, Jansen JA. The effects of implantsurface roughness and surgical technique on implantfixation in an in vitro model. Clinical Oral Implants Research2006;17:172–8.

11. Davies JE. Mechanisms of endosseous integration.International Journal of Prosthodontics 1998;11:391–401.

12. Viguet-Carrin S, Garnero P, Delmas PD. The role of collagenin bone strength. Osteoporosis International 2006;17:319–36.

13. Hohling HJ, Arnold S, Barckhaus RH, Plate U, Wiesmann HP.Structural relationship between the primary crystalformations and the matrix macromolecules in differenthard tissues. Discussion of a general principle. ConnectiveTissue Research 1995;33:171–8.

14. Hohling HJ, Arnold S, Plate U, Stratmann U, Wiesmann HP.Analysis of a general principle of crystal nucleation,formation in the different hard tissues. Advances in DentalResearch 1997;11:462–6.

15. Nkenke E, Kloss F, Wiltfang J, Schultze-Mosgau S, Radespiel-Troger M, Loos K, et al. Histomorphometric and fluorescencemicroscopic analysis of bone remodelling after installationof implants using an osteotome technique. Clinical OralImplants Research 2002;13:595–602.

16. Carter DR, Hayes WC. Bone compressive strength: theinfluence of density and strain rate. Science 1976;194:1174–6.

17. Currey JD. The design of mineralised hard tissues for theirmechanical functions. The Journal of Experimental Biology1999;202:3285–94.

18. Goldman HM, Thomas CD, Clement JG, Bromage TG.Relationships among microstructural properties of bone atthe human midshaft femur. Journal of Anatomy2005;206:127–39.

19. Currey JD. The mechanical consequences of variation in themineral content of bone. Journal of Biomechanics 1969;2:1–11.

20. Currey JD. The relationship between the stiffness andthe mineral content of bone. Journal of Biomechanics1969;2:477–80.

21. Currey JD. Effects of differences in mineralization on themechanical properties of bone. Philosophical Transactions ofthe Royal Socisty of London Series B Biological Sciences1984;304:509–18.

22. Loveridge N, Power J, Reeve J, Boyde A. Bone mineralizationdensity and femoral neck fragility. Bone 2004;35:929–41.

23. Skedros JG, Bloebaum RD, Bachus KN, Boyce TM, ConstantzB. Influence of mineral content and composition ongraylevels in backscattered electron images of bone. Journalof Biomedical Materials Research 1993;27:57–64.

24. Reilly DT, Burstein AH, Frankel VH. The elastic modulus forbone. Journal of Biomechanics 1974;7:271–5.

25. Reilly DT, Burstein AH. Review article. The mechanicalproperties of cortical bone. The Journal of Bone and JointSurgery American Volume 1974;56:1001–22.

26. Reilly DT, Burstein AH. The elastic and ultimate propertiesof compact bone tissue. Journal of Biomechanics 1975;8:393–405.

27. Tai K, Qi HJ, Ortiz C. Effect of mineral content on thenanoindentation properties and nanoscale deformationmechanisms of bovine tibial cortical bone. Journal ofMaterials Science Materials in Medicine 2005;16:947–59.

28. Currey JD. The effect of porosity and mineral content on theYoung’s modulus of elasticity of compact bone. Journal ofBiomechanics 1988;21:131–9.

29. Currey JD. Strain rate and mineral content in fracturemodels of bone. Journal of Orthopaedic Research 1988;6:32–8.

30. Currey JD, Brear K, Zioupos P. The effects of ageing andchanges in mineral content in degrading the toughness ofhuman femora. Journal of Biomechanics 1996;29:257–60.

31. Currey JD. The design of mineralised hard tissues for theirmechanical functions. The Journal of Experimental Biology1999;202:3285–94.

32. Currey JD, Zioupos P. The effect of porous microstructure onthe anisotropy of bone-like tissue: a counterexample. Journalof Biomechanics 2001;34:707–10.

33. Martin RB, Papamichos T, Dannucci GA. Linear calibrationof radiographic mineral density using video-digitizingmethods. Calcified Tissue International 1990;47:82–91.

34. Bloebaum RD, Skedros JG, Vajda EG, Bachus KN, ConstantzBR. Determining mineral content variations inbone using backscattered electron imaging. Bone1997;20:485–90.

35. Bloebaum RD, Bachus KN, Boyce TM. Backscattered electronimaging: the role in calcified tissue and implant analysis.Journal of Biomaterials Applications 1990;5:56–85.

36. Skedros JG, Bloebaum RD, Bachus KN, Boyce TM, ConstantzB. Influence of mineral content and composition ongraylevels in backscattered electron images of bone. Journalof Biomedical Materials Research 1993;27:57–64.

37. Bloebaum RD, Bachus KN, Boyce TM. Backscattered electronimaging: the role in calcified tissue and implant analysis.Journal of Biomaterials Applications 1990;5:56–85.

38. Bloebaum RD, Skedros JG, Vajda EG, Bachus KN, ConstantzBR. Determining mineral content variations in bone usingbackscattered electron imaging. Bone 1997;20:485–90.

39. Roschger P, Fratzl P, Klaushofer K, Rodan G. Mineralizationof cancellous bone after alendronate and sodium fluoridetreatment: a quantitative backscattered electron imagingstudy on minipig ribs. Bone 1997;20:393–7.

40. Skedros JG, Bloebaum RD, Bachus KN, Boyce TM, ConstantzB. Influence of mineral content and composition ongraylevels in backscattered electron images of bone. Journalof Biomedical Materials Research 1993;27:57–64.

41. Roschger P, Fratzl P, Klaushofer K, Rodan G. Mineralizationof cancellous bone after alendronate and sodium fluoridetreatment: a quantitative backscattered electron imagingstudy on minipig ribs. Bone 1997;20:393–7.

42. Currey JD. The mechanical consequences of variation in themineral content of bone. Journal of Biomechanics 1969;2:1–11.

43. Currey JD. The relationship between the stiffness andthe mineral content of bone. Journal of Biomechanics1969;2:477–80.

44. Currey JD. Effects of differences in mineralization on themechanical properties of bone. Philosophical Transactions ofthe Royal Society of London Series B Biological Sciences1984;304:509–18.

45. Currey JD, Brear K, Zioupos P. The effects of ageing andchanges in mineral content in degrading the toughness ofhuman femora. Journal of Biomechanics 1996;29:257–60.

j o u r n a l o f d e n t i s t r y 3 5 ( 2 0 0 7 ) 8 4 – 9 292

46. Currey JD. The design of mineralised hard tissues for theirmechanical functions. The Journal of Experimental Biology1999;202:3285–94.

47. Currey JD, Zioupos P, Davies P, Casino A. Mechanicalproperties of nacre and highly mineralized bone. ProceedingsBiological Sciences 2001;268:107–11.

48. Skedros JG, Bloebaum RD, Bachus KN, Boyce TM, ConstantzB. Influence of mineral content and composition ongraylevels in backscattered electron images of bone. Journalof Biomedical Materials Research 1993;27:57–64.

49. Viguet-Carrin S, Garnero P, Delmas PD. The role of collagenin bone strength. Osteoporosis International 2006;17:319–36.

50. Misch CE, Qu Z, Bidez MW. Mechanical properties oftrabecular bone in the human mandible: implications fordental implant treatment planning and surgical placement.Journal of Oral and Maxillofacial Surgery 1999;57:700–6.

51. Loveridge N, Power J, Reeve J, Boyde A. Bone mineralizationdensity and femoral neck fragility. Bone 2004;35:929–41.

52. Skedros JG, Bloebaum RD, Bachus KN, Boyce TM, ConstantzB. Influence of mineral content and composition ongraylevels in backscattered electron images of bone. Journalof Biomedical Materials Research 1993;27:57–64.

53. Frost HM. Bone ‘‘mass’’ and the ‘‘mechanostat’’: a proposal.The Anatomical Record 1987;219:1–9.

54. Manz MC. Radiographic assessment of peri-implant verticalbone loss: DICRG Interim Report No. 9. Journal of Oral andMaxillofacial Surgery 1997;55:62–71.

55. Traini T, Degidi M, Caputi S, Strocchi R, Di Iorio D, Piattelli A.Collagen fiber orientation in human peri-implant bonearound immediately loaded and unloaded titanium dentalimplants. Journal of Periodontology 2005;76:83–9.

56. Traini T, Degidi M, Strocchi R, Caputi S, Piattelli A. Collagenfiber orientation near dental implants in human bone: dotheir organization reflect differences in loading? Journal ofBiomedical Materials Research Part B Applied Biomaterials2005;74:538–46.

57. Traini T, De Paoli S, Caputi S, Iezzi G, Piattelli A. Collagenfiber orientation near a fractured dental implant after a 5-year loading period: case report. Implant Dentistry2006;15:70–6.

58. Neugebauer J, Traini T, Thams U, Piattelli A, Zoller JE. Peri-implant bone organization under immediate loading state.Circularly polarized light analyses: a minipig study. Journalof Periodontology 2006;77:152–60.