Marcin Kozakiewicz
1, Anna Marciniak-Hoffman
1, Marek Olszycki
2Comparative Analysis of Three Bone Substitute
Materials Based on Co-Occurrence Matrix
Analiza porównawcza trzech materiałów kościozastępczych
oparta na danych z macierzy zdarzeń
1Department of Maxillofacial Surgery, Medical University of Lodz, Poland 2Department of Radiology, Medical University of Lodz, Poland
Abstract
Background. Nowadays, bone substitute materials have become more popular.
Objectives. This article covers the comparison of tree bone-substitute materials – ChronOS®, Cerasorb® and Straumann Bone Ceramic® (all pertain to β-tricalcium phosphates group), the quality of newly created bone was compared.
Material and Methods. For the purposes of this paper, the analysis of X-ray photographs of 110 patients was
con-ducted. Intraoral periapical radiographs were made on the day of a surgery and 3, 6, 9 and 12 months after the sur-gery such as extraction, resection and cyst removal. In the comparative analysis co-occurrence matrix was used.
Results. The above-mentioned materials belong to the same group of tricalcium phosphates, but have
dif-ferent shape, size, and surface structure of granule, as a result of which there are slight differences in the bone regeneration activity observed during a few months post-operativelly. Moreover, Straumann Bone Ceramic contains hydroxylapatite, which is visible in all radiographic images during 12 months.
Conclusions. During 12 months of tests, only the place where ChronOS is applied does not differ from reference
bone on radiographic images. Cerasorb and Straumann Bone Ceramic need longer period than 12 months in order to create a full-value bone (Dent. Med. Probl. 2010, 47, 1, 23–29).
Key words: bone substitute materials, β-tricalcium phosphates.
Streszczenie
Wprowadzenie. Materiały kościozastępcze stają się obecnie coraz bardziej popularne.
Cel pracy. Porównanie 3 środków kościozastępczych – ChronOS®, Cerasorb® i Straumann Bone Ceramic® (wszyst-kie z grupy β-trójfosforanów wapniowych), pod względem jakości nowo powstałej kości.
Materiał i metody. W pracy dokonano analizy zdjęć RTG wykonanych u 110 pacjentów. Zdjęcia RTG zostały
wykonane w dniu zabiegu oraz po 3, 6, 9 i 12 miesiącach po zabiegach, takich jak: ekstrakcja, resekcja lub usunięcie torbieli. W analizie porównawczej wykorzystano macierz zdarzeń.
Wyniki. Materiały te należą do jednej grupy β-trójwapniowych fosforanów, różnią się jednak między sobą
kształ-tem, wielkością i strukturą powierzchni cząstek. Wpływa to na niewielkie zaobserwowane różnice w szybkości rege-neracji kości w pierwszych miesiącach po zabiegu chirurgicznym. W składzie Straumann Bone Ceramic występuje ponadto hydroksyapatyt, który jest widoczny na wszystkich zdjęciach RTG z 12 miesięcy.
Wnioski. W ciągu 12 miesięcy badań tylko miejsce aplikacji materiału ChronOS nie różni się na zdjęciu RTG od
kości referencyjnej. Cerasorb i Straumann Bone Ceramic potrzebują więcej czasu niż 12 miesięcy, aby uformowała się na ich miejscu nowa, pełnowartościowa kość (Dent. Med. Probl. 2010, 47, 1, 23–29).
Słowa kluczowe: materiały kościozastępcze, β-fosforany trójwapniowe.
Dent. Med. Probl. 2010, 47, 1, 23–29 ISSN 1644-387X
ORIGINAL PAPeRS
© Copyright by Wroclaw Medical University and Polish Dental Society
Nowadays, bone substitute materials have be-come more popular. Many doctors are aware of their advantages, that is biocompatibility, angio-genesis stimulation, porosity, haemostatic activity,
full resorption [1]. The manufacturers are improv-ing the content of bone substitite materials.
Recently, the most commonly used are β-tricalcium phosphates (β-TPC). TCP is a
syntet-ic. It is chemically and mineralogically similar to osseous tissue and thanks to that it is bio-compat-ible and osteo-conductive. Moreover, it has good tolerance, lack of innflamation reaction from host tissues, and most importantly lack of any danger connected with HIV, icterus, prions [2].
In the given article tree substitute bone ma-terials are compared: Cerasorb®, ChronOS®,
Straumann Bone Ceramic®. Cerasorb and ChronOS
are the materials where the chief component is composition of elements is calcium phosphate [3, 4]. However Straumann Bone Ceramic is a mixture of β-tricalcium phosphates and hydroxyapatite [2]. In all these materials the element composition indicates the presence of β-tricalcium phosphate, amount of which is proportional to the bioresorp-tion rate, which influences significantly the qual-ity of newly built bone [5–8]. This feature enables to shorten the period of vital and functional bone creation in place of osseous defect.
Material and Methods
Three β-tricalcium phosphates granulated bio-materials were applied into 110 patients. Cerasorb®
(Curasan, Germany) of gradation 1–2 mm particles was utilized in 58 cases, ChronOS®
(Synthes, USA) of gradation 0.7–1.4 mm parti-cles was put into 32 cases and Straumann Bone Ceramic® (Straumann, Switzerland) of gradiation
0.4–0.7 mm particles was put into 20 cases. Materials were implanted into bone defects after apicoectomies, impacted tooth removals and cyst enucleations (i.e. periapical surgery and intra-alveolar surgery).
Immediately after surgery was completed, intra-oral radiological examination was tak-en. Digora Optime digital radiography device (Soredex, Finland) and X-ray apparatus Focus (Instrumentarium Dental, Finland) were used. Technical parameters of exposure were the same in all included cases: 7 mA, 70 mV and 0.06 s. Radiographs were taken in a standardized way [9]. A modified RINN (Dentsply Rinn, USA) system was applied. A bite index was prepared using a sil-icone material (occlusal bite duplicates the shape of the film plate holder and also occlusal surfaces of the teeth). The X-ray detector was placed in the RINN positioner and the bite index with the con-nection bar and ring was replaced in the mouth of the patient then fixed to the tube. Follow-up ex-aminations were performed in the same condition 3, 6, 9 and 12 months post-operationally.
Regions of interest (ROI) were selected in ra-diographs. The ROI outlining limit was the line running on the external border of the bone
sub-stitute materials image (and not bone defect). Another region of normal bone with the least diverse gray level was also selected in the radio-graphs as a reference.
Total registered number of gray levels was decreased to 128, to reduce random noise in the analyzed data. Then each ROI optical density was normalized according to mean (mean ± 3 standard deviation of optical density). Co-occurrence ma-trix was analysed in MaZda ver. 4.5. The second-order histogram was defined as the co-occurrence matrix hdθ(i,j) [10]. When divided by the total num-ber of neighboring pixels R(d,θ) in ROI, this ma-trix becomes the estimate of the joint probability,
pdθ(i,j) , of two pixels, a distance of 5 pixels apart along a given direction θ having particular (co-oc-curring) values i and j. Formally, given the image
f(x,y) with a set of 128 discrete levels of optical
density, the matrix hdθ(i,j) is defined such that its (i,j)th entry is equal to the number of times that
f(x1y1) = and f(x2y2) = j, where (1)
(x2y2) = (x1y1) + (d cos β, d sin β) (2)
This yields a square matrix of dimension equal to the number of intensity levels in the im-age, for selected distance of 5 pixels and orienta-tion θ;. Angles θ; = 0°, 45°, 90 and 135° were in-vestigated. Aritmethical mean of these four direc-tions was calculated due to eliminate directional dependence.
Sum of average of co-occurrence matrix was extracted to describe bone structure variability. It is defined by the equation that follow (3), where µx denotes the mean of the row sums of the co-occurrence matrix (related to the marginal distri-butions px(i) and px(j)).
∑
= + = 128 1 (), where (3) i ipx y i SumAverg i + j = k∑
∑
= = + = 128 1 128 1 ) , ( ) ( j i y x k p i j p k = 2, 3, ..., 256 [11, 12] (4)The geometric adjustment was performed with the use of ToothVis program on the ra-diographs of each patient, about 4–5 reference points were selected. On the basis of these points the program matched the radiographs with each other. Then, using MaZda ver. 4.5 program, the reports regarding parameter sum of average co-occurrence matrix for each patient’s radiographs were calculated.
Statistical evaluation was performed in Statgraphics Centurion XV.II version 15.2.06. Multiple-sample comparison based on analysis of variance was performed. Statistically
signifi-cant difference amongst the standard deviations was not observed (p = 0.5953). Next, ANOVA was applied to reveal the factor which influenced the variability of texture.
Results
The ANOVA table (Tab. 1) decomposes the variance of the data into two components: a be-tween-group component and a within-group component. The F-ratio, which in this case equals 2.62208, is a ratio of the between-group estimate to the within-group estimate. Since the P-value of the F-test is 0.0016, there is a statistically signifi-cant difference between the means of the 16 vari-ables at the 95 confidence level.
Multiple Range Tests (Tabs. 2 and 3) deter-mines which means are significantly different from which others.
There is no statistically significant differenc-es between radiological texture of implanted site by Cerasorb and Straumann Bone Ceramic after 12 months.
Discussion
The difference in surface properties may then have profound effects on the protein adhesion and the following cellular attachment [13]. The blood protein adsorption starts the process of new bone creation [14, 15]. The next phase is an attachment phase, in which physical and chemical interactions between cells and material take place [16]. After that step is adhesion phase, when osteoblast adhesion appears [16]. The most important parameters in cell interaction are surface topography – size [17], shape
[17, 18], surface texture [19] of the material and the physical and chemical features [3, 4] of surface.
One of the bone substitute material feature is the creation of scaffold (which can be observed radiologically). The products from scaffold disin-tegration may become then the substrate for new bone deposition. During bone substitute material resorption, the calcium and phosphatic ions are created in that place. These ions have the ability to mineralize the newly created connective tissue from which a new bone is created.
This study presents 3 bone-substitution mate-rials that differ from one another as regards the microscopic surface structure (Figs. 1–4).
Cerasorb® molecules are almost ball-shaped
with diameter 1–2 mm, pore with diameter 1–8 μm and the whole molecule is created from micro-molecules with diameter 4–8 μm [4]. ChronOS®
molecules have different shapes with diameter 0.7–1.4 mm, pores with diameter 20–500 μm and the whole molecule consists of 2–5 μm micro-ecules [3]. While Straumann Bone Ceramic mol-ecules has 0.4–0.7 mm and pores 100–500 μm [2]. 9 months after surgery the radio-texture of implanted site by ChronOS became similar to normal bone. It indicates a superiority of this bone substitute material over two competitors.
Shape and structure of the surface may have an influence on the new bone creation rapidity [13]. ChronOS® is characterized by the most favourable
proportion of pores and molecule diameter. The textural alteration visible in radiographic images of Straumann Bone Ceramic® is different
from the two other materials. It is caused by the fact that this material contains slowly resorbing hydroxylapatite [2], and therefore the newly cre-ated bone was incrusted with hydroxylapatite for 12 months of the test.
Table 1. ANOVA table Tabela 1. Tabela ANOVA
Source (Suma
kwadratów) Sum of squares(Źródło zmienności) Df (Stopnie swobody) Mean square (Średnia kwadratów) F (Współczyn-nik Fischera) P (Stopień istotności statystycznej) Between groups (Międzygrupowe) 34.4529 15 2.29686 2.62 0.0016 Within groups (Wewnątrzgrupowe) 126.14 144 0.875969 Total (Razem) 160.592 159 Df – degrees of freedom. F – Fisher coefficient.
p – level of significant statistics. Df – stopnie swobody. F – współczynnik Fishera. p – stopień istotności statystycznej.
Radiographic image of Cerasorb® during
12 months also differs from the reference bone, which is caused by the different structure of gran-ules between Cerasorb® and ChronOS® – there
are discrepancies in the diameters of granules and pores.
Table 2. Multiple Range Tests. Fisher’s least significant
difference (LSD) procedure
Tabela 2. Wielokrotny test uszeregowujący. Procedura
Fishera najmniejszych różnic (LSD) Groups
(Grupy) Mean(Średnia) Homogeneous groups* (Homogenność grup*) ReF 63.6652 X CHR12M 64.2627 XX CHR03M 64.4883 XXX STR09M 64.6167 XXXX CHR09M 64.6233 XXXX CHR06M 64.6271 XXXX STR03M 64.7098 XXXX CHR00M 64.7685 XXXX CeR03M 64.8028 XXXX STR12M 64.9549 XXXXX STR06M 64.9694 XXXXX CeR12M 65.034 XXXXX CeR06M 65.1093 XXXXX CeR09M 65.3664 XXXXX STR00M 65.4188 XXXXX CeR00M 65.6754 XXXXX
* Homogeneous groups - a graphical illustration of which means are significantly different from which others, based on the contrasts displayed. each column of X’s indicates a group of means within which there are no statistically significant differences.
ReF – reference bone.
CHR...M – ChronOS ... months after surgery. CeR...M – Cerasorb ... months after surgery.
STR...M – Straumann Bone Ceramic ... months after sur-gery.
*Homogenność grup – graficzna ilustracja różnic między średnimi na podstawie stwierdzonego kontrastu. Każda kolumna X wskazuje grupę średnich, w obrębie których nie ma różnic statystycznych.
ReF – kość referencyjna.
CHR...M – ChronOS ... miesięcy po zabiegu. CeR...M – Cerasorb ... miesięcy po zabiegu. STR...M – Straumann Bone Ceramic ... miesięcy po zabiegu.
Table 3. Details of observed between group differences Tabela 3. Szczegóły obserwowanych różnic
między-grupowych Contrast
(Kontrast) Significant difference (Różnica statystyczna)
ReF - CeR00M –2.01025 ReF - CeR03M –1.13767 ReF - CeR06M –1.44412 ReF - CeR09M –1.70122 ReF - CeR12M –1.36888 ReF - CHR00M –1.10333 ReF - CHR06M –0.96192 ReF - CHR09M –0.958164 ReF - STR00M –1.75362 ReF - STR03M –1.0446 ReF - STR06M –1.30423 ReF - STR09M –0.95158 ReF - STR12M –1.28976 CeR00M - CeR03M 0.872579 CeR00M - CHR00M 0.906924 CeR00M - CHR03M 1.18708 CeR00M - CHR06M 1.04833 CeR00M - CHR09M 1.05209 CeR00M - CHR12M 1.41267 CeR00M - STR03M 0.965652 CeR00M - STR09M 1.05867 CeR06M - CHR12M 0.846544 CeR09M - CHR03M 0.878054 CeR09M - CHR12M 1.10364 CHR03M - STR00M –0.930455 CHR12M - STR00M –1.15604
ReF – reference bone.
CHR...M – ChronOS ... months after surgery. CeR...M – Cerasorb ... months after surgery.
STR...M – Straumann Bone Ceramic ... months after sur-gery.
ReF – kość referencyjna.
CHR...M – ChronOS ... miesięcy po zabiegu. CeR...M – Cerasorb ... miesięcy po zabiegu. STR...M – Straumann Bone Ceramic ... miesięcy po zabiegu.
Fig. 1. Transformations of the texture (parameter on vertical axis: sum of square) of site implanted by different
tri-calcium phosphates: CeR – Cerasorb, CHR – ChronOS, STR – Straumann Bone Ceramic, ReF – reference bone, M – months
Ryc. 1. Zmiany tekstury (cecha na osi pionowej: suma kwadratów gęstości optycznej) miejsca wszczepienia różnych
fosforanów trójwapniowych: CeR – Cerasorb, CHR – ChronOS, STR – Straumann Bone Ceramic, ReF – kość refe-rencyjna, M – miesiące 63 64 65 66 00M 03M 06M 09M 12M CER CHR STR REF
Fig. 2. A) Cerasorb implantation site in patient after radicular cyst removal (red area), the radiological texture is
dif-ferent from reference bone (green area) till the RTG e, B) 3 months after the implantation, C) 6 months after the im-plantation, D) 9 months after the imim-plantation, e) 12 months after the implantation
Ryc. 2. A) Implantacja materiału Cerasorb u pacjenta po usunięciu torbieli korzeniowej (obszar czerwony), struktura
radiologiczna jest odmienna od kości referencyjnej (obszar zielony) do ostatniego badania e, B) 3 miesiące po implan-tacji materiału, C) 6 miesięcy po implanimplan-tacji materiału, D) 9 miesięcy po implanimplan-tacji, e) 12 miesięcy po implanimplan-tacji
Fig. 3. A) ChronOS implantation site in patient after radicular cyst removal (red area), the radiological texture is
dif-ferent from reference bone till (green area) the image D, B) radiological examination 3 months post-operationally, C) 6 months after the implantation, D) 9 months after the implantation, e) 12 months after the implantation
Ryc. 3. A) Implantacja materiału ChronOS u pacjenta po usunięciu torbieli korzeniowej (obszar czerwony), struktura
radiologiczna jest odmienna od referencyjnej kości (obszar zielony) do badania D, B) badanie radiologiczne 3 miesiące po zabiegu chirurgicznym, C) 6 miesięcy po implantacji materiału, D) 9 miesięcy po implantacji materiału, e) 12 mie-sięcy po implantacji
Fig. 4. A) Straumann Bone Ceramic implantation site in patient after radicular cyst removal (red area), the
radio-logical texture is different from reference bone till (green area) the RTG e, B) 3 months after the implantation, C) 6 months after the implantation, D) 9 months after the implantation, e) 12 months after the implantation
Ryc. 4. A) Implantacja materiału Straumann Bone Ceramic u pacjenta po usunięciu torbieli korzeniowej (obszar
czerwony), struktura radiologiczna jest odmienna od referencyjnej kości (obszar zielony) do ostatniego badania e, B) 3 miesiące po implantacji materiału, C) 6 miesięcy po implantacji materiału, D) 9 miesięcy po implantacji, e) 12 mie-sięcy po implantacji
Straumann Bone Ceramic has the small-est particles among invsmall-estigated series of ma-terials and main part of its mass was rapidly resorbed, but this phenomenon was balanced by hydroxyapatite remnants from Straumann Bone Ceramic®, which was much more stable
than residual mass of Cerasorb® and ChronOS®.
Finally, statistically, there is no significant dif-ference in recovery of Cerasorb® and Straumann
Bone Ceramic® after 12 months from surgery.
Furthermore, the long-term observation of bone substitute material after the implantation we published previously [20].
Concluding, during 12 months after surgery no total bone regeneration could be achieved in case of all tricalcium phosphates. The creation of a full-value bone may take much more time and longer observation of patients.
References
[1] Antoun H., Chemaly C., Missika P.: Bone substitutes. In: Bone augmentation in oral implantology. eds.: Khoury F., Antoun H., Missika P. Quintessence books Co, Ltd, London 2007, 341–372.
[2] Puchała P., Kucharski G., Jaremczuk B., Monkos-Jaremczuk e.: Przegląd biomateriałów na podstawie piśmiennictwa. TPS 2008, 10, 28–36.
[3] Kozakiewicz M., Klimek L.: Analiza powierzchni, składu chemicznego i fazowego materiału kościozastępczego ChronOs. Magazyn Stomatol. 2005, 15, 4, 30–33.
[4] Kozakiewicz M., Klimek L.: Analiza powierzchni, składu chemicznego i fazowego materiału kościozastępczego Cerasorb. Magazyn Stomatol. 2003, 9, 13, 44–47.
[5] Uchida A., Nade S.M.L., McCartney e.R., Ching W.: The use of ceramics for bone replacement. J. Bone Joint Surg. (Br.) 1984, 66, 269–275.
[6] Blitterswijk C.A., Grote J.J., Kuypers W.: Bioreactions at the tissue/hydroxyapatite apatite interface. Biomaterials 1985, 6, 243–251.
[7] Shimazaki K., Mooney V.: Comparative study of porous hydroxyapatite and tricalcium phosphate as bone sub-stitute. J. Orthop. Res. 1985, 3, 301–310.
[8] Klein C.P.A.T., Drissen A.A., DeGroot K.: Biodegradation behaviour of various calcium phosphate material in bone tissue. J. Biomed. Mater. Res. 1983, 17, 769–784.
[9] Kozakiewicz M., Bogusiak K., Hanclik M., Denkowski M., Arkuszewski P.: Noise in subtraction images made from pairs of intraoral radiographs: a comparison between four methods of geometric alignment. Dentomaxillofac. Radiol. 2008, 37, 40–47.
[10] Dash M., Liu H.: Feature selection for classification elsevier Science Inc. 1997 http://www−east.elsevier.com/ida/ browse/0103/ida00013/article.htm/
[11] Materka A., Strzelecki M.: Texture analysis methods – a review, COST B11 report (presented and distributed at MC meeting and workshop in Brussels, June 1998), Technical University of Lodz, Poland. Available from: http:// www.eletel.p.lodz.pl/programy/cost/pdf_1.pdf
[12] Materka A., Strzelecki M., Lerski R., Schad L.: Feature evaluation of texture test objects for magnetic reso-nance imaging. Workshop on Texture Analysis and Machine Vision, Oulu, Finland 1999, 13–19.
[13] Meyer U., Buchter A., Wiesmann H.P, Joos U., Jones D.B.: Basic reactions of osteoblasts on structured mate-rial surfaces. eur. Cells Mat. 2005, 9, 39–49.
[14] Boyan B.D., Hummert T.W., Dean D.D., Schwartz Z.: Role of material surfaces in regulating bone and cartilage cell response. Biomaterials 1996, 17, 137–146.
[15] Meyer U., Meyer T., Jones D.B.: No mechanical role for vinculin in strain transduction in primary bovine oste-oblasts. Biochem. Cell. Biol. 1997, 75, 81–87.
[16] Meyer U., Buchter A., Wiesmann H.P, Joos U., Jones D.B.: Basic reactions of osteoblasts on structured mate-rial surfaces. eur. Cells Mat. 2005, 9, 39–49.
[17] Matlaga B.F., Yasenchak L.P., Salthouse T.N.: Tissue response to implanted polymers: the significance of sample shape. J. Biomed. Mater. Res. 1976, 10, 391–397.
[18] Misiek D.J., Kent J.N., Carr R.F.: Soft tissue responses to hydroxylapatite particles of different shapes. J. Oral Maxillofac. Surg. 1984, 42, 150–160.
[19] Hench L.L., Wilson J.: Surface-active biomaterials. Science 1984, 226, 630–636.
[20] Kozakiewicz M., Marciniak-Hoffman A., Denkowski M.: Long term comparison of application of two beta-tricalcium phosphates in oral surgery. Dent. Med. Probl. 2009, 46, 284–388.
Address for correspondence:
Marcin Kozakiewicz Cyberskiego 2 m. 14 92-447 Łódź Poland Tel./fax: +48 42 656 65 47 e-mail: marcin.kozakiewicz@umed.pl Received: 1.02.2010 Revised: 15.03.2010 Accepted: 17.03.2010
Praca wpłynęła do Redakcji: 1.02.2010 r. Po recenzji: 15.03.2010 r.
