Bone Regeneration Related to Calcium Phosphate-Coated Implants in Osteoporotic Animal Models: A Meta-Analysis

Abstract

Background:

Osteoporosis is a frequent human metabolic bone disorder. Prospectively, global ageing of populations will lead to a major increase of subjects being diagnosed with osteoporosis and in need for dental rehabilitation. However, as local osteoporosis of the jaws affects bone quantity and quality of edentulous regions, osseointegration of dental implants might be hampered. Consequently, calcium phosphate ceramic-coated implants have been suggested to compensate for low bone quantity/density and for impaired bone healing in osteoporosis. Nonetheless, up to now no meta-analytical assessment of the relevant preclinical literature to quantify such a possible positive effect has been undertaken.

Materials and Methods:

PubMed search, limited to animal models, to identify a possible positive effect of calcium phosphate-coated implants on bone regeneration, was carried out. Further, the reference lists of related review articles and publications selected for inclusion in this review were systematically screened. The primary outcome variables were bone-to-implant contact percentage as assessed histomorphometrically and mechanical stability testing.

Results:

The electronic search in the database of the National Library of Medicine resulted in the identification of 2704 titles. These titles were initially screened by the two independent reviewers for possible inclusion, resulting in further consideration of 51 publications. Screening the abstracts led to 22 full-text articles. From these articles, 16 reports were excluded. Finally, six of these original research reports could be selected for evaluation. Additionally, eight publications were identified by manual search. Thus, a total of 14 articles were included for analysis.

Conclusions:

It was concluded that (1) in osteoporotic animal models calcium phosphate ceramic-coated implants are associated with improved bone-to-implant healing as compared to noncoated implants. Moreover, (2) essentially due to quality characteristics of the analyzed original research articles a negative impact of osteoporosis on bone-to-implant healing could not be confirmed. Besides, (3) the established positive bone-to-implant healing effect of calcium phosphate ceramic coatings does not differ between osteoporotic and nonosteoporotic, healthy animal models.

Introduction

Osteoporosis is the most frequent human metabolic bone disorder¹ and due to its important global prevalence considered as a serious public health concern. It is estimated that over 200 million people worldwide suffer from this medical condition.^2,3 Further, global ageing of populations will lead to a major increase of subjects being diagnosed with osteoporosis.⁴ In principle, two fundamental pathogenetic mechanisms, that is, excessive bone resorption resulting in decreased bone mass and micro architectural deterioration of the skeleton and inadequate bone formation response during remodeling, are essential for this heterogeneous bone disorder.⁵ Considering the occurrence of osteoporosis in the maxillofacial area, it is well documented that the pattern of the physiological gender- and age-related bone mineral content and bone mineral density (BMD) changes in the jaws are similar to those in other sites of the skeleton. Likewise, in most cases, systemic osteoporosis may be associated with a severe decrease in bone mineral content of the jaw bones.⁵ In particular, systemic osteoporosis may affect bone quantity and quality of edentulous regions that might be considered for dental implant treatment.^6,7 In consequence, primarily increased trabecular bone width and bone mass and reduced cortical bone width⁶ might challenge dental implant therapy. In addition, inferior osseointegration in bone with established osteopenia/osteoporosis as compared to normal bone^8,9 may amplify that challenge. However, it should be emphasized that the precise underlying molecular mechanisms for impaired bone-to-implant healing are not completely understood. Nonetheless, it might be related to bone healing typified by impaired neovascularization, osteogenesis, and remodeling and impaired mechanical properties.¹⁰ Moreover, it has been found that estrogen receptor expression was dysregulated in bone cells, and assumed that this might contribute to the impaired bone healing in estrogen deficiency-related osteoporosis. Consequently, osteoporosis of the jaw bones might challenge dental implant therapy by means of bone quantity/density and especially in postmenopausal, estrogen deficiency-related cases, with conceivably delayed and/or impaired bone regeneration.¹⁰

In principle, for more challenging situations such as low bone density, improved implant stability, and accelerated bone healing have been shown for certain surface modifications of dental titanium implants.^11–14 Surface modifications of dental titanium implants are in general accomplished by roughening or by altering the chemical composition. Various methods have been developed to create roughened surfaces, for example, titanium plasma spraying, grit-blasting, acid etching, and anodization. Coating of dental titanium implants with calcium phosphate (CaP) ceramic is the most frequently used method for changing the chemical surface composition.¹⁵ It is well known that following implantation, the release of CaP into the peri-implant region increases the saturation of body fluids and results in the precipitation of a biological apatite onto the surface of the implant.^16,17 This layer of biological apatite might contain endogenous proteins and serve as a matrix for osteogenic cell attachment and growth.¹⁸

Cellular interactions with the apatite layer and its contained proteins onto an implant surface are mediated by integrins. These specific cell membrane molecules transduce signals from the surface proteins matrix to the cell and vice versa. Then, the signaling pathways through integrins can regulate bone forming cell activity to a certain extent.^16,17 This desired bone-stimulating behavior of CaP coatings at implant surface is of prime importance when compared to dental implants without CaP coatings.¹⁹ In addition, as the biological fixation of titanium implants to bone tissue is faster with a CaP coating than without,^19,20 it seems rational to assume that the bone regeneration process around the implant is enhanced by the formation of the aforementioned biological apatite layer. Thus, it is not astonishing that especially for osteoporotic bone CaP ceramic-coated implants have been suggested to compensate for low bone quantity/density and for delayed and/or impaired bone regeneration.^21–25 However, up to now no meta-analytical assessment of the relevant literature to quantify such a possible positive effect has been undertaken.

Hence, the aims of this study were (1) to systematically review the literature on the subject of bone healing and CaP ceramic-coated implants in osteoporotic animal models and (2) to investigate bone-to-implant contact (BIC) data and data retrieved from mechanical stability tests by means of a meta-analytical approach.

It was hypothesized that in osteoporotic animal models (1) CaP ceramic-coated implants are associated with improved bone healing as compared to noncoated implants, that (2) CaP ceramic-coated implants overcome a possible negative impact of osteoporosis on bone healing, and (3) that the effect of CaP ceramic coatings is greater in osteoporotic animal models as compared to nonosteoporotic, healthy animal models.

Materials and Methods

Outcome variables

The primary outcome variables were BIC percentage as assessed by histological analysis and biomechanical stability testing.

Inclusion/exclusion criteria

In general, only animal studies presenting BIC percentage or data regarding mechanical testing (push-out test or torque-out test) for CaP-coated implants versus noncoated, control implants in osteoporotic animals versus healthy, control animals were included.

The following detailed inclusion criteria were used:

1. Research article presenting in vivo animal data;

2. Data should be acquired in osteoporotic animals (test) and in healthy, nonosteoporotic animals (control);

3. Data should be presented for calcium phosphate ceramic-coated implants (test) and noncoated (control) implants;

4. Shape or configuration of used implants, that is, threaded (screw-shaped) or nonthreaded (cylinder-shaped), should be indicated;

5. Implantation site should be clearly stated;

6. Healing period should be clearly mentioned;

7. The osteoporotic animal model used should be described conspicuously, for example:(i) Basic characteristics of the animal model should be mentioned (e.g.: species and gender);(ii) Initiation method of osteoporosis should be stated;(iii) Time point of induction (i.e., before −, parallel to −, or after implant installation) should be mentioned;(iv) Method to confirm osteoporotic status of the implantation site should be stated;

8. At least four osteoporotic and control animals should have been included;

9. BIC percentage and/or data retrieved from biomechanical testing (push-out test or torque-out test) have to be presented. Further, torque-out testing data should be presented together with histomorphometrical data.

Studies that did not meet all above mentioned inclusion criteria were excluded.

Search strategy

An electronic search in the database of the National Library of Medicine (www.ncbi.nlm.nih.gov) up to November 7, 2011, was carried out. Only publications in English were considered and the search was narrowed to animals only. The following search strategy was applied: (((((“dental implantation” [MeSH Terms] OR “dental implants” [MeSH Terms]) OR “orthopedics” [MeSH Terms]) OR “implants, experimental” [MeSH Terms]) OR “blade implantation” [MeSH Terms]) OR “implants” [All Fields]) AND ((((surface[All Fields] AND modification[All Fields]) OR “surface coating” [All Fields]) OR “surface treatment”[All Fields]) OR (“surface properties”[MeSH Terms] OR “coated materials, biocompatible” [MeSH Terms])). Further, the reference lists of related review articles and publications selected for inclusion in this review were systematically screened.

Study selection

Two independent reviewers (Hamdan Alghamdi [H.A.] and Rüdiger Junker [R.J.]) initially screened the publication titles and abstracts as identified by the electronic and manual search for possible inclusion. Full texts of all articles that were considered eligible for inclusion by one or both of the reviewers were obtained for further assessment against the stated inclusion criteria (Fig. 1). Both reviewers used a data extraction form to extract the data independently. Any disagreements between the reviewers regarding inclusion of a certain publication or data extraction were resolved by discussion.

FIG. 1.

Articles selection process.

Quantitative data synthesis

As the central aim of the present study was to quantify a possible positive effect of CaP ceramic coatings on the subject of implant osseointegration in osteoporotic bone, a meta-analysis for the effect size of CaP ceramic coating with regard to (1) BIC percentage [%BIC], (2) mechanical push-out testing [N], and (3) mechanical torque-out testing [Ncm] in osteoporotic animal models (test) and in healthy, nonosteoporotic models (control) was performed. For (1) BIC percentage [%BIC], (2) mechanical push-out testing [N], and (3) mechanical torque-out testing [Ncm], Student's t-test was applied to calculate the difference of the weighted means of CaP ceramic-coated implants (test) and noncoated implants (control) within and between osteoporotic animal models (test) and healthy, nonosteoporotic models (control). Next to differences of the weighted means (MD), 95% Confidence Intervals (CI), and statistical significance difference (p<0.05) were obtained. In addition, for meta-analysis, inverse variance weighting was used for pooling. Analyses were performed with the statistical software package R, version 2.10.1 (www.R-project.org).²⁶ In case of heterogeneity, the DerSimonian-Laird estimate was used.²⁷

Results

Study selection

The electronic search in the database of the National Library of Medicine resulted in the identification of 2704 titles. As already mentioned, these titles were initially screened by the two independent reviewers for possible inclusion, resulting in further consideration of 51 publications. Screening the abstracts led to 22 full-text articles, which are detailed in Table 1. From these articles, 16 reports were excluded for reasons mentioned in Table 1. Finally, six of these original research reports could be selected for evaluation. Additionally, eight publications were identified by manual search. Thus, at last a total of 14 articles were included for analysis (Fig. 1). Regarding data extraction and interpretation, any disagreement between the reviewers was resolved by discussion. Included studies are summarized in Table 2.

Table 1.

Full-Text Articles and Finally Included Articles for Evaluation

Article	Included	Excluded/reasons of exclusion
Aldini et al.²¹	x
Aldini et al.³⁰	x
Borsari et al.²⁹	x
Fini et al.³⁹	x
Frenkel et al.⁴⁰	x
Gao et al.⁵³		No BIC% or mechanical test reported for noncoated implants
Gao et al.⁵⁴		No BIC% or mechanical test reported for noncoated implants
Hara et al.²²	x
Hasegawa et al.⁵⁵		Number of animals <3; No healthy animal group included
Hayashi et al.³⁷	x
Hayashi et al.³⁸	x
Hayashi et al.²³	x
Hayashi et al.²⁸	x
Jung et al.⁴⁸	x
Kurth et al.⁵⁶		No BIC% or mechanical test reported for noncoated implants
Keller et al.⁵⁷		No BIC% or mechanical test reported for noncoated implants
Li et al.⁵⁸		No BIC% or mechanical test reported for noncoated implants
Li et al.⁵⁹		No BIC% or mechanical test reported for noncoated implants
Li et al.⁶⁰		No BIC% or mechanical test reported for CaP-coated implants
Motohashi et al.⁶¹		No BIC% or mechanical test reported for noncoated implants
Nakamura et al.²⁴	x
Ohkawa et al.⁶²		No BIC% or mechanical test reported for noncoated implants
Ozawa et al.⁶³		No BIC% or mechanical test reported for CaP-coated implants
Pan et al.⁶⁴		No BIC% or mechanical test reported for noncoated implants
Peter et al.⁶⁵		No BIC% or mechanical test reported for noncoated implants
Rocca et al.²⁵	x
Stadelmann et al.⁶⁶		No BIC% or mechanical test reported for noncoated implants
Tokugawa et al.⁶⁷		No BIC% or mechanical test reported for CaP-coated implants
Tresguerres et al.⁶⁸		No BIC% or mechanical test reported for CaP-coated implants
Vidigal et al.⁴⁷	x

x indicates selected article.

BIC, bone-to-implant contact; CaP, calcium phosphate.

Table 2.

Study Characteristics of the Included Studies

	Osteoporotic animal model			Implantation
Study	Species	Model	Detection method	Site	Period	Implant design	Outcome variable
Aldini et al. ³⁰	20 female rats	OVX, 4 months	BMD	Femur	4 weeks	Cylinder	BIC
					8 weeks	Cylinder	BIC
Nakamura et al.²⁴	64 female rats	OVX, 7 months	BMD	Femur	4 weeks	Cylinder	Push-out
Hayashi et al.³⁸	84 female rats	OVX, 6 months	BMD	Femur	4 weeks	Cylinder	Push-out
Hayashi et al.³⁷	36 female rats	OVX, 6 months	BMD	Femur	4 weeks	Cylinder	Push-out
Hara et al.²²	60 female rats	OVX, 6 months	BMD	Femur	4 weeks	Cylinder	Push-out
Fini et al.³⁹	24 female rats	OVX, 4 months	BMD	Femur	8 weeks	Cylinder	BIC
Hayashi et al.²³	135 female rats	OVX, 4 months	BMD	Tibia	4 weeks	Cylinder	BIC
					8 weeks	Cylinder	BIC
					12 weeks	Cylinder	BIC
					24 weeks	Cylinder	BIC
Hayashi et al.²⁸	88 female rats	sciatic NX, 4 months	SOFTEX-C-SM	Tibia	2 weeks	Cylinder	BIC
					24 weeks	Cylinder	BIC
Vidigal et al.⁴⁷	20 female rabbits	OVX, 4 months	BMD	Tibia	12 weeks	Cylinder	BICPush-out
Jung et al.⁴⁸	36 female rabbits	OVX, 2 months	BMD	Tibia	12 weeks	Screws	BICTorque-out
Rocca et al.²⁵	12 female sheep	OVX, 24 months	Iliac biopsy	Femur	12 weeks	Screws	BICTorque-out
				Tibia	12 weeks	Screws	BICTorque-out
Borsari et al.²⁹	5 female sheep	OVX, 24 months	Iliac biopsy	Tibia	12 weeks	Cylinder	BICPush-out^a
Aldini et al.²¹	8 female sheep	OVX, 24 months	BMD	Vertebra	16 weeks	Screws	BICTorque-out
Frenkel et al.⁴⁰	16 female dogs	OVX, 2 months	BMD	Femur	24 weeks	Chamber	Pull-out

Excluded from meta-analysis; the presented data were not transferrable to the measuring unit for push-out testing [N].

OVX, bilateral ovariectomy; NX, unilateral limb neurectomy; BMD, bone mineral density.

Study characteristics

In 13 out of the included 14 research articles, female animals were submitted to bilateral ovariectomy-induced, estrogen-deficiency osteoporosis (OVX), whereas in one study²⁸ unilateral limb sciatic neurectomy (sciatic NX) was used. Osteoporosis was diagnosed in 11 out of the 14 reports by BMD measurements, in one study by radiographic analysis of bone morphology (SOFTEX-C-SM),²⁸ and in two publications^25,29 by histomorphometrical analysis of iliac bone biopsies. In all 14 reports CaP ceramic-coated implants (test) and noncoated implants (control) were used in osteoporotic animal models (test) and in nonosteoporotic, healthy animal models (control). In principle, from four articles data of more than one experiment could be retrieved. This was possible because in one publication,²⁵ different anatomical sites of implantation (i.e., femur or tibia within a sheep-model) were analyzed separately, and because in the three otherarticles,^23,28,30 different healing periods after implantation were independently studied. As a result, 20 discrete trials were identified (Table 2).

CaP ceramic coating characteristics

Coating characteristics as provided in the included research articles are presented in Table 3.

Table 3.

Characteristics of the Calcium Phosphate Ceramic Coating

Publication	Characteristics of the CaP ceramic coating
Aldini et al.³⁰	Coating technique:
	Enameling technique with slurry (consisting of a dispersing polymer, water, and powders of the bioactive glass in suitable concentration) brushed on the surfaces of the ceramic rod;
	After coating, the samples were heat treated at 1280°C;
	Chemical composition:
	SiO2 43.8%, Ca3(PO4)2 24.2%, CaO 18.4%, Na2O 4.5%, K2O 0.2%, MgO 2.8%, CaF2 5% Ta2O5 1%, La2O3 0.1%
	Physical parameters:
	Thickness: 140 μm
Nakamura et al.²⁴	Coating technique:
	Sand-blasted with Al2O3
	Plasma-sprayed
	Chemical composition:
	Hydroxyapatite
	Crystallinity: 55%
	Physical parameters:
	Thickness: 20 μm
	Roughness (Ra): 4.3±0.6 μm
Hayashi et al.³⁸	Coating technique:
	Plasma-sprayed
	Chemical composition:
	Hydroxyapatite
	Crystallinity: 55%
	Ca/P ratio: 1.66
	Physical parameters:
	Thickness: 20 μm
	Roughness (Ra): 4.3±0.6 μm
Hayashi et al.³⁷	Coating technique:
	Radio frequency-thermal plasma spraying
	Chemical composition:
	Hydroxyapatite
	Crystallinity: 81%
	Physical parameters:
	Thickness: 20 μm
	Roughness (Ra): 20 μm
Hara et al.²²	Coating technique:
	Plasma-sprayed
	Chemical composition:
	Hydroxyapatite
	Crystallinity: 55%
	Ca/P ratio: 1.66
	Physical parameters:
	Thickness: 20 μm
	Roughness (Ra): 4.3±0.6 μm
Fini et al.³⁹	Chemical composition:
	Hydroxyapatite
Hayashi et al.²³	Coating technique:
	Plasma-sprayed
	Chemical composition:
	Hydroxyapatite
	Physical parameters:
	Roughness (Ra): 0.8±0.21 μm
Hayashi et al.²⁸	Chemical composition:
	Hydroxyapatite
Vidigal et al.⁴⁷	Coating technique:
	Plasma-sprayed
	Chemical composition:
	Hydroxyapatite
Jung et al.⁴⁸	Coating technique:
	Ion beam-assisted deposition
	Chemical composition:
	Hydroxyapatite
Rocca et al.²⁵	Coating technique:
	Plasma-sprayed
	Chemical composition:
	Hydroxyapatite
	Crystallinity: >70%
	Ca/P ratio: 1.66
	Physical parameters:
	Roughness (Ra): 7.3 μm
	Coating adhesion strength: >30 MPa
	Porosity: 4%
Borsari et al.²⁹	Coating technique:
	Vacuum plasma-sprayed
	Chemical composition:
	Hydroxyapatite
	Crystallinity: >70%
	Ca/P ratio: 1.66
	Physical parameters:
	Thickness: 100±19 μm
	Roughness (Ra): 12±2 μm
	Coating adhesion strength: >30 MPa
	Porosity: 5.0±1.2%
Aldini et al.²¹	Coating technique:
	Plasma-sprayed
	Chemical composition:
	Hydroxyapatite
	Crystallinity: >70%
	Ca/P ratio: 1.66
	Physical parameters:
	Thickness: 80±20 μm
	Roughness (Ra): 7.3±2 μm
	Coating adhesion strength: >30 MPa
	Porosity: 4%
Frenkel et al.⁴⁰	Coating technique:
	Plasma-sprayed
	Chemical composition:
	Hydroxyapatite
	Physical parameters:
	Thickness: 50 μm

Highly detailed information with regard to chemical and physical composition is generally not presented. Only one publication reports a more detailed chemical composition,³⁰ whereas the other 13 articles roughly describe the composition as a hydroxyapatite.

CaP-coated implants versus noncoated implants in osteoporotic animal models

First, the hypothesis that in osteoporotic animal models (test) CaP ceramic-coated implants (test) are associated with improved bone healing as compared to noncoated implants (control) was tested. In osteoporotic animal models, the difference of the weighted means of BIC percentage between CaP ceramic-coated implants and noncoated implants was 19.2%. The 95% CI of the difference ranged between 13.8% and 24.6%. This difference was statistically significant (p<0.05). The difference of the weighted means for mechanical push-out testing between CaP ceramic-coated implants and noncoated implants was 105.4 N. The 95% CI of the difference ranged between 53.5 and 157.4 N. Also this difference was statistically significant (p<0.05). Moreover, the difference of the weighted means for mechanical torque-out testing between CaP ceramic-coated implants and noncoated implants was 115.9 Ncm. The 95% CI of the difference ranged between 31.6 and 200.2 Ncm. However, this difference did not reach statistical significance (p>0.05; Table 4).

Table 4.

Calcium Phosphate-Coated Implants Versus Noncoated Implants in Osteoporotic Animal Models

	BIC [%](n=15 trials)		Push-out [N](n =5 trials)		Torque-out [Ncm](n=4 trials)
Osteoporotic animal models	Mean±SD	95% CI	Mean±SD	95% CI	Mean±SD	95% CI
CaP-coated implants	82.9±3.4	[76.2, 89.6]	164.0±14.3	[136.0, 192.0]	190.0±56.5	[79.5, 301.0]
Noncoated implants	63.4±3.9	[55.5, 71.2]	57.5±27.3	[4.2, 111.0]	73.4±26.3	[21.7, 125.0]
	MD	95% CI	MD	95% CI	MD	95% CI
CaP-coated implants vs. noncoated implants	19.2	[13.8, 24.6]	105.4	[53.5, 157.4]	115.9	[31.6, 200.2]
p-value	0.0001		0.0005		0.0613

SD, standard deviation; CI, confidence intervals; MD, difference of the weighted means.

Noncoated implants in osteoporotic animal models versus nonosteoporotic, healthy animal models and CaP ceramic-coated implants in osteoporotic animal models versus noncoated implants in nonosteoporotic, healthy animal models

Second, the hypothesis that CaP ceramic-coated implants (test) overcome a possible negative impact of osteoporosis (test) on bone healing was investigated. Comparing noncoated implants in nonosteoporotic, healthy animal models versus osteoporotic animal models revealed that the difference of the weighted means of BIC percentage was 8.2%. The 95% CI of the difference ranged between −4.1% and 20.5%. This difference was not statistically significant (p>0.05). The difference of the weighted means of mechanical push-out testing was 22.4 N. The 95% CI of the difference ranged between −41.2 and 86.0 N. Also, this difference was not statistically significant (p>0.05). Further, the difference of the weighted means of mechanical torque-out testing was 15.0 Ncm. The 95% CI of the difference ranged between −45.1 and 75.1 Ncm. Again, this difference failed to reach statistical significance (p>0.05; Table 5, Fig. 2A–C). Further, comparing CaP ceramic-coated implants in osteoporotic animal models versus noncoated implants in nonosteoporotic, healthy animal models uncovered that the difference of the weighted means of BIC percentage was 11.3%. The 95% CI of the difference ranged between −0.4% and 23.0%. This difference was not statistically significant (p>0.05). The difference of the weighted means of mechanical push-out testing was 84.1 N. The 95% CI of the difference ranged between 39.8 and 128.4 N. This difference was statistically significant (p<0.05). Besides, the difference of the weighted means of mechanical torque-out testing was 101.6 Ncm. The 95% CI of the difference ranged between −13.4 and 216.6 Ncm. Again, this difference failed to reach statistical significance (p>0.05; Table 6, Fig. 3A–C).

FIG. 2.

(A) Forest plot of the mean difference between noncoated implants in healthy versus noncoated implants in osteoporosis for bone-to-implant contact [%]. (B) Forest plot of the mean difference between noncoated implants in healthy versus noncoated implants in osteoporosis for push-out testing [N]. (C) Forest plot of the mean difference noncoated implants in healthy versus noncoated implants in osteoporosis for torque-out testing [Ncm].

FIG. 3.

(A) Forest plot of the mean difference between calcium phosphate (CaP)-coated implants in osteoporosis versus noncoated implants in health for bone-to-implant contact [%]. (B) Forest plot of the mean difference between CaP-coated implants in osteoporosis versus noncoated implants in health for push-out testing [N]. (C) Forest plot of the mean difference between CaP-coated implants in osteoporosis versus noncoated implants in health for torque-out testing [Ncm].

Table 5.

Noncoated Implants in Osteoporosis Versus Noncoated Implants in Healthy

	BIC [%](n =15 trials)		Push-out [N](n =5 trials)		Torque-out [Ncm](n= 4 trials)
	MD	95% CI	MD	95% CI	MD	95% CI
Noncoated implants in healthy vs. noncoated implants in osteoporosis	8.2	[−4.1, 20.5]	22.4	[−41.2, 86.0]	15.0	[−45.1, 75.1]
p-value	0.1904		0.4897		0.6249

Table 6.

Calcium Phosphate-Coated Implants In Osteoporosis Versus Noncoated Implants in Healthy

	BIC [%](n=15 trials)		Push-out [N](n=5 trials)		Torque-out [Ncm](n=4 trials)
	MD	95% CI	MD	95% CI	MD	95% CI
CaP-coated implants in osteoporosis vs. noncoated implants in healthy	11.3	[−0.4, 23.0]	84.1	[39.8, 128.4]	101.6	[−13.4, 216.6]
p-value	0.0581		0.0001		0.0833

Effects of CaP coatings on bone healing in osteoporotic animal models and in nonosteoporotic, healthy animal models

Third, the hypothesis that the effect of CaP ceramic coatings (test) is greater in osteoporotic animal models (test) as compared to nonosteoporotic, healthy animal models (control) was examined. As already mentioned above, in osteoporotic animal models the difference of the weighted means of BIC percentage between CaP ceramic-coated implants and noncoated implants was 19.2%. The 95% CI of the difference ranged between 13.8% and 24.6%. This difference was statistically significant (p<0.05). The difference of the weighted means for mechanical push-out testing between CaP ceramic-coated implants and noncoated implants was 105.4 N. The 95% CI of the difference ranged between 53.5 and 157.4 N. Also, this difference was statistically significant (p<0.05). Moreover, the difference of the weighted means for mechanical torque-out testing between CaP ceramic-coated implants and noncoated implants was 115.9 Ncm. The 95% CI of the difference ranged between 31.6 and 200.2 Ncm. However, this difference did not reach statistical significance (p>0.05). In nonosteoporotic, healthy animal models the difference of the weighted means of BIC percentage between CaP ceramic-coated implants and noncoated implants was 13.7%. The 95% CI of the difference ranged between 6.8% and 20.6%. This difference was statistically significant (p<0.05). The difference of the weighted means for mechanical push-out testing between CaP ceramic-coated implants and noncoated implants was 147.7 N. The 95% CI of the difference ranged between 91.1 and 204.3 N. Again, this difference was statistically significant (p<0.05). Moreover, the difference of the weighted means for mechanical torque-out testing between CaP ceramic-coated implants and noncoated implants was 132.6 Ncm. The 95% CI of the difference ranged between 41.7 and 223.6 Ncm. Again, this difference reached statistical significance (p<0.05). Heterogeneity testing of the differences of the weighted means of BIC percentage, push-out testing, and torque-out testing between CaP ceramic-coated implants and noncoated implants in osteoporotic animal models and nonosteoporotic, healthy animal models revealed that the included experiments presenting BIC percentage data and the experiments presenting push-out testing data were heterogeneous (Q-value: 48.5, p<0.05; Q-value: 9.7, p<0.05; respectively), whereas the experiments presenting torque-out testing data were not heterogeneous (Q-value: 4.7, p>0.05). The applied statistical effect models, the random effect model for BIC percentage, and the push-out testing, and the fixed effect model for torque-out testing did not indicate statistically significant overall differences regarding the investigated outcome.

Discussion

The aims of the present report were (1) to systematically review the literature on the subject of bone healing around CaP ceramic-coated implants in osteoporotic animal models and (2) to investigate BIC data and data retrieved from mechanical stability tests by means of a meta-analytical approach. It was hypothesized that in osteoporotic animal models (1) CaP ceramic-coated implants are associated with improved bone healing as compared to noncoated implants, that (2) CaP ceramic-coated implants overcome a possible negative impact of osteoporosis on bone healing, and (3) that the effect of CaP ceramic coatings is greater in osteoporotic animal models as compared to nonosteoporotic, healthy animal models. As in osteoporotic animal models statistically significant differences between CaP ceramic-coated- and noncoated implants with regard to BIC percentage and mechanical push-out testing were found in favor of CaP-coated implants, we tend to accept the proposed H₁-hypothesis. On the other hand, it should be kept in mind that for mechanical torque-out testing the alternative H₀-hypothesis of no difference could not be rejected. Further, the proposed H₁-hypothesis that CaP ceramic-coated implants overcome a possible negative impact of osteoporosis on bone healing could not be validated. Nonetheless, for mechanical push-out testing with regard to CaP ceramic-coated implants in osteoporotic animal models versus noncoated implants in nonosteoporotic animal models, the proposed H₁-hypothesis seems to be substantiated. Additionally, statistically significant effect size differences for BIC percentage, mechanical push-out testing, and mechanical torque-out testing were not observed for CaP ceramic-coated implants versus noncoated implants in osteoporotic animal models and nonosteoporotic animal models. Therefore, the proposed H₁-hypothesis could not be confirmed.

As already indicated above, the retrieved findings have to be interpreted with caution.³¹ In principle, sample-size and thereby statistical power of the reviewed preclinical in vivo experiments tended to be low. For example, the compared eight weeks BIC data within the group of osteoporotic animals of Aldini et al.³⁰ have, with an assumed α-error of 0.05, a statistical power as low as 0.092. For that reason, a meta-analytical approach was intended to increase the power of statistical analysis by pooling the results of all retrieved available trials. Preferably, all research should be done in a similar manner when their results are being combined in a meta-analysis. As shown in Table 2, this is clearly not the case for the presently included articles. The publications that were eligible for inclusion in the present study display experimental variability for the utilized animal model, the method applied for induction of osteoporosis, the used technique to identify osteoporosis, the number of enrolled implants and/or animals, the anatomical site of implantation, the healing time after implant placement, and the type of implant and coating used. Although surgical technique might also have a critical effect on implant fixation,^18,29 it was difficult to consider the surgical variables in this systematic review. This was because all included studies did not estimate all the possible effects of surgical technique on the implant outcomes. Not surprisingly, statistical heterogeneity of the evaluated experiments regarding BIC percentage and push-out testing was found and might be explained in part by experimental variability.^32,33 However, subgroup analysis was unable to clarify the origin of the dissimilarity in terms of treatment effects. Consequently, for the outcome variables BIC percentage and push-out testing random effects meta-analysis was selected.³⁴ The applicability and problematic aspects of random effects meta-analysis are thoroughly discussed elsewhere.^35,36 Further, as heterogeneity was not found for torque-out testing, fixed effects meta-analysis was chosen for that outcome variable.³⁴

Next to the meta-analytical method, to be confident that the combined outcome estimate points toward a meaningful picture of the effect of CaP ceramic coatings for the potential use in humans, it is important to critically appraise the utilized osteoporotic animal models.

Obviously, the used bilateral ovariectomy-induced, estrogen- deficient osteoporosis rat model,^{22–24,30,37–39} sheep model,^21,25,29 and dog model⁴⁰ are respected,^41–43 well documented, and in general, accepted osteoporotic disease models to simulate human osteoporosis for mucoskeletal research.^41,44–46 In contrast, the employed rat unilateral limb sciatic neurectomy-induced osteoporosis model²⁸ and the bilateral ovariectomy-induced, estrogen-deficient osteoporosis rabbit model^47,48 are evidently suitable osteoporosis models, but less accepted to mimic the human situation.^46,49–51

For the careful interpretation of the current findings it should be mentioned that 11 of the 14 meta-analyzed research articles are published by just two different groups [group 1:^{21,25,29,30,39}, group 2:^{22–24,28,37,38}]. Apparently, only Frenkel et al.,⁴⁰ Jung et al.,⁴⁸ and Vidigal et al.⁴⁷ seem to be unrelated.

Unfortunately, highly detailed information with regard to chemical and physical composition is generally not presented. For example, only one publication reports a more detailed chemical composition,³⁰ whereas the other 13 articles roughly describe the composition as a hydroxyapatite. However, chemical and physical surface properties as ionic composition, hydrophilicity, and roughness are parameters that play a major role in implant-tissue interaction.¹⁵ In other words, different CaP implant surface coatings possess different and unique chemical and physical surface properties, and might potentially lead to differences in the bone-to-implant reaction.⁵² That finally implies that the statistical heterogeneity of the evaluated experiments regarding BIC percentage and push-out testing might be explained in part by differences of the chemical and physical composition of the applied CaP coatings. Accordingly, more detailed information regarding chemical and physical surface properties as ionic composition, hydrophilicity, and roughness might have enabled a subgroup analysis to clarify the origin of the dissimilarity as regards treatment effects.

Conclusion

Within the limits of the current meta-analysis, we conclude that (1) in osteoporotic animal models CaP ceramic-coated implants are associated with improved bone-to-implant healing as compared to noncoated implants. Further, (2) essentially due to quality characteristics of the analyzed original research articles, a negative impact of osteoporosis on bone-to-implant healing could not be confirmed. In addition, (3) the established positive bone-to-implant healing effect of CaP ceramic coatings does not differ between osteoporotic and nonosteoporotic, healthy animal models.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Lerner

Bone remodeling in post-menopausal osteoporosis. J Dent Res, 85:584. 2006.

Cooper

, Campion

, Melton

Lr.

Hip fractures in the ederly: a world-wide projection. Osteoporsos Int, 2:285. 1992.

Cooper

Epidemiology of osteoporosis. Osteoporos Int, 9,Suppl 2:S2. 1999.

Gullberg

, Johnell

, Kanis

Worldwide projections for the hip fracture. Osteoporos Int, 5:407. 1997.

Raisz

Pathogenesis of osteoporosis: concepts, conflicts, and prospects. J Clin Invest, 115:3318. 2005.

Munakata

, Tachikawa

, Honda

, Shiota

, Kasugai

Influence of menopause on mandibular bone quantity and quality in Japanese women receiving dental implants. Arch Osteoporos, 6:51. 2011.

von Wowern

General and oral aspects of osteoporosis: a review. Clin Oral Invest, 5:71. 2001.

Tsolaki

I.N.

, Madianos

P.N.

, Vrotsos

J.A.

Outcomes of dental implants in osteoporotic patients. A literature review. J Prosthodont, 18:309. 2009.

Mellado-Valero

, Ferrer-García

, Calvo-Catalá

, Labbaig-Rueda

Implant treatment in patients with osteoporosis. Med Oral Patol Oral Cir Bucal, 15:e52. 2010.

10.

, Zhang

, Pan

, Liu

, Zheng

, Chan

et al. Impaired bone healing pattern in mice with ovariectomy-induced osteoporosis: a drill-hole defect model. Bone, 48:1388. 2011.

11.

Albrektsson

, Wennerberg

Oral implant surfaces: part I—review focusing on topographic and chemical properties of different surfaces and in vivo responses to them. Int J Prosthodont, 17:536. 2004.

12.

Albrektsson

, Wennerberg

Oral implant surfaces: part II—review focusing on clinical knowledge of different surfaces. Int J Prosthodont, 17:544. 2004.

13.

Wennerberg

, Albrektsson

Effects of titanium surface topography on bone integration: a systematic review. Clin Oral Impl Res, 20,Suppl. 4:172. 2009.

14.

Junker

, Dimakis

, Thoneick

, Jansen

Effects of implant surface coatings and composition on bone integration: a systematic review. Clin Oral Impl Res, 20,Suppl. 4:185. 2009.

15.

Le Guéhennec

, Soueidan

, Layrolle

, Amouriq

Surface treatments of titanium dental implants for rapid osseointegration. Dent Mater, 23:844. 2007.

16.

de Groot

, Wolke

J.G.

, Jansen

J.A.

Calcium phosphate coatings for medical implants. Proc Inst Mech Eng H, 212:137. 1998.

17.

Daculsi

, Laboux

, Malard

, Weiss

Current state of the art of biphasic calcium phosphate bioceramics. J Mater Sci Mater Med, 14:195. 2003.

18.

Davies

Understanding peri-implant endosseous healing. J Dent Educ, 67:932. 2003.

19.

Morris

, Ochi

, Spray

, Olson

Periodontal-type measurements associated with hydroxyapatite-coated and non-HA-coated implants: uncovering to 36 months. Ann Periodontol, 5:56. 2000.

20.

Barrere

, van der Valk

, Meijer

, Dalmeijer

, De Groot

, Layrolle

Osteointegration of biomimetic apatite coating applied onto dense and porous metal implants in femurs of goats. J Biomed Mater Res, 67:655. 2003.

21.

Aldini

N.N.

, Fini

, Giavaresi

, Giardino

, Greggi

, Parisini

Pedicular fixation in the osteoporotic spine: a pilot in vivo study on long-term ovariectomized sheep. J Orthop Res, 20:1217. 2002.

22.

Hara

, Hayashi

, Nakashima

, Kanemaru

, Iwamoto

The effect of hydroxyapatite coating on the bonding of bone to titanium implants in the femora of ovariectomised rats. J Bone Joint Surg Br, 81:705. 1999.

23.

Hayashi

, Uenoyama

, Mashima

, Sugioka

Remodelling of bone around hydroxyapatite and titanium in experimental osteoporosis. Biomaterials, 15:11. 1994.

24.

Nakamura

, Hayashi

, Abu-Ali

, Naito

, Fotovati

Effect of preoperative combined treatment with alendronate and calcitriol on fixation of hydroxyapatite-coated implants in ovariectomized rats. J Bone Joint Surg Am, 90:824. 2008.

25.

Rocca

, Fini

, Giavaresi

, Aldini

N.N.

, Giardino

Osteointegration of hydroxyapatite-coated and uncoated titanium screws in long-term ovariectomized sheep. Biomaterials, 23:1017. 2002.

26.

Team

R: A Language and Environment for Statistical Computing. Vienna, Austria: Computing RFfS, 2009.

27.

DerSimonian

, Laird

Meta-analysis in clinical trials. Control Clin Trials, 7:177. 1986.

28.

Hayashi

, Uenoyama

, Matsuguchi

, Nakagawa

, Sugioka

The affinity of bone to hydroxyapatite and alumina in experimentally induced osteoporosis. J Arthroplasty, 4:257. 1989.

29.

Borsari

, Fini

, Giavaresi

, Rimondini

, Consolo

, Chiusoli

et al. Osteointegration of titanium and hydroxyapatite rough surfaces in healthy and compromised cortical and trabecular bone: in vivo comparative study on young, aged, and estrogen-deficient sheep. J Orthop Res, 25:1250. 2007.

30.

Aldini

N.N.

, Fini

, Giavaresi

, Martini

, Dubini

, Ponzi Bossi

M.G.

et al. Osteointegration of bioactive glass-coated and uncoated zirconia in osteopenic bone: an in vivo experimental study. J Biomed Mater Res A, 68:264. 2004.

31.

Thompson

, Pocock

Can meta-analysis be trusted?

Lancet, 338:1127. 1991.

32.

Song

, Sheldon

, Sutton

, Abrams

, Jones

Methods for exploring heterogeneity in meta-analysis. Eval Health Prof, 24:126. 2001.

33.

Glasziou

, Sanders

Investigating causes of heterogeneity in systematic reviews. Stat Med, 21:1503. 2002.

34.

Hunter

, Schmidt

Fixed effects vs. random effects meta-analysis models: implications for cumulative research knowledge. Int J Sel Assess, 8:275. 2000.

35.

Higgins

, Thompson

, Spiegelhalter

A re-evaluation of random-effects meta-analysis. J R Statist Soc A, 172,Part 1:137. 2009.

36.

Petitti

Approaches to heterogeneity in meta-analysis. Stat Med, 20:3625. 2001.

37.

Hayashi

, Fotovati

, Abu Ali

, Nakamura

, Inagaki

, Naito

Effect of a prostaglandin EP4 receptor agonist on early fixation of hydroxyapatite/titanium composite- and titanium-coated rough-surfaced implants in ovariectomized rats. J Biomed Mater Res A, 92:1202. 2010.

38.

Hayashi

, Fotovati

, Ali

S.A.

, Oda

, Oida

, Naito

Prostaglandin EP4 receptor agonist augments fixation of hydroxyapatite-coated implants in a rat model of osteoporosis. J Bone Joint Surg Br, 87:1150. 2005.

39.

Fini

, Giavaresi

, Aldini

N.N.

, Torricelli

, Morrone

, Guzzardella

G.A.

et al. The effect of osteopenia on the osteointegration of different biomaterials: histomorphometric study in rats. J Mater Sci Mater Med, 11:579. 2000.

40.

Frenkel

S.R.

, Jaffe

W.L.

, Valle

C.D.

, Jazrawi

, Maurer

, Baitner

et al. The effect of alendronate (Fosamax) and implant surface on bone integration and remodeling in a canine model. J Biomed Mater Res, 58:645. 2001.

41.

Turner

The sheep as a model for osteoporosis in humans. Vet J, 163:232. 2002.

42.

Yamaura

, Nakamura

, Nagai

, Yoshihara

, Suzuki

Reduced mechanical competence of bone by ovariectomy and its preservation with 24R, 25-Dihydroxyvitamin D₃ administration in Beagles. Calcif Tissue Int, 52:49. 1993.

43.

Thompson

, Simmons

, Pirie

, Ke

FDA Guidlines and animal models for osteoporosis. Bone, 17,4, Suppl:125S. 1995.

44.

Barlet

, Coxam

, Davicco

, Gaumet

Animal models of post-menopausal osteoporosis. Reprod Nutr Dev, 34:221. 1994.

45.

Turner

Animal models of osteoporosis—necessity and limitations. Eur Cells Mater, 1:66. 2001.

46.

Newman

, Turner

, Wark

The potential of sheep for the study of osteopenia: current status and comparison with ther animal models. Bone, 16,4, Suppl:277S. 1995.

47.

Vidigal

G.M.

Jr. , Groisman

, Gregorio

L.H.

, Soares Gde

Osseointegration of titanium alloy and HA-coated implants in healthy and ovariectomized animals: a histomorphometric study. Clin Oral Implants Res, 20:1272. 2009.

48.

Jung

Y.C.

, Han

C.H.

, Lee

I.S.

, Kim

H.E.

Effects of ion beam-assisted deposition of hydroxyapatite on the osseointegration of endosseous implants in rabbit tibiae. Int J Oral Maxillofac Implants, 16:809. 2001.

49.

Sevastik

, Lindgren

Osteoporosis: experimental and clinical studies. Clin Orthop Relat Res, 191:35. 1984.

50.

Mori

, Manabe

, Kurachi

, Nagumo

Osseointegration of dental implants in rabbit bone with llow mineral density. J Oral Maxillofac Surg, 55:351. 1997.

51.

Castañeda

, Calvo

, Largo

, Gonzáles-González

, de la Pietra

, Díaz-Curiel

et al. Characterization of a new experimental model of osteoporosis in rabbits. J Bone Miner Metab, 26:53. 2008.

52.

Coelho

, Granjeiro

, Romanos

, Suzuki

, Silva

, Cardaropoli

et al. Basic research methods and current trends of dental implant surfaces. J Biomed Mater Res Part B: Appl Biomater, 88:579. 2009.

53.

Gao

, Luo

, Hu

, Xue

, Zhu

, Li

Effect of combined local treatment with zoledronic acid and basic fibroblast growth factor on implant fixation in ovariectomized rats. Bone, 44:225. 2009.

54.

Gao

, Zou

, Liu

, Bao

, Hu

The effect of surface immobilized bisphosphonates on the fixation of hydroxyapatite-coated titanium implants in ovariectomized rats. Biomaterials, 30:1790. 2009.

55.

Hasegawa

, Inufusa

, Imai

, Mikawa

, Lim

T.H.

, An

H.S.

Hydroxyapatite-coating of pedicle screws improves resistance against pull-out force in the osteoporotic canine lumbar spine model: a pilot study. Spine J, 5:239. 2005.

56.

Kurth

, Eberhardt

, Müller

, Steinacker

, Schwarz

, Bauss

The bisphosphonate ibandronate improves implant integration in osteopenic ovariectomized rats. Bone, 37:204. 2005.

57.

Keller

J.C.

, Stewart

, Roehm

, Schneider

G.B.

Osteoporosis-like bone conditions affect osseointegration of implants. Int J Oral Maxillofac Implants, 19:687. 2004.

58.

, Feng

, Gao

, Luo

, Liu

, Hu

Strontium ranelate treatment enhances hydroxyapatite-coated titanium screws fixation in osteoporotic rats. J Orthop Res, 28:578. 2010.

59.

, Li

, Zhu

, Luo

, Li

, Feng

et al. The effect of strontium-substituted hydroxyapatite coating on implant fixation in ovariectomized rats. Biomaterials, 31:9006. 2010.

60.

, Zou

, Wang

, Feng

, Bao

, Hu

The effect of hydrofluoric acid treatment on titanium implant osseointegration in ovariectomized rats. Biomaterials, 31:3266. 2010.

61.

Motohashi

, Shirota

, Tokugawa

, Ohno

, Michi

, Yamaguchi

Bone reactions around hydroxyapatite-coated implants in ovariectomized rats. Oral Surg, Oral Med, Oral Pathol, Oral Radiol, Endodontol, 87:145. 1999.

62.

Ohkawa

, Tokunaga

, Endo

Intermittent administration of human parathyroid hormone (1–34) increases new bone formation on the interface of hydroxyapatitecoated titanium rods implanted into ovariectomized rat femora. J Orthop Sci, 13:533. 2008.

63.

Ozawa

, Ogawa

, Iida

, Sukotjo

, Hasegawa

, Nishimura

et al. Ovariectomy hinders the early stage of bone-implant integration: histomorphometric, biomechanical, and molecular analyses. Bone, 30:137. 2002.

64.

Pan

, Shirota

, Ohno

, Michi

Effect of ovariectomy on bone remodeling adjacent to hydroxyapatite-coated implants in the tibia of mature rats. J Oral Maxillofac Surg, 58:877. 2000.

65.

Peter

, Gauthier

, Laïb

, Bujoli

, Guicheux

, Janvier

et al. Local delivery of bisphosphonate from coated orthopedic implants increases implants mechanical stability in osteoporotic rats. J Biomed Mater Res Part A, 76:133. 2006.

66.

Stadelmann

V.A.

, Gauthier

, Terrier

, Bouler

J.M.

, Pioletti

D.P.

Implants delivering bisphosphonate locally increase periprosthetic bone density in an osteoporotic sheep model. A pilot study. Eur Cell Mater, 16:10. 2008.

67.

Tokugawa

, Shirota

, Ohno

, Yamaguchi

Effects of bisphosphonate on bone reaction after placement of titanium implants in tibiae of ovariectomized rats. Int J Oral Maxillofac Implants, 18:66. 2003.

68.

Tresguerres

I.F.

, Clemente

, Donado

, Gómez-Pellico

, Blanco

, Alobera

M.A.

et al. Local administration of growth hormone enhances periimplant bone reaction in an osteoporotic rabbit model. Clin Oral Implants Research, 13:631. 2002.