Professional Documents
Culture Documents
DOI 10.1007/s00784-016-2030-x
ORIGINAL ARTICLE
Received: 18 March 2016 / Accepted: 6 December 2016 / Published online: 24 December 2016
# Springer-Verlag Berlin Heidelberg 2016
washing off (i.e., medium exchange) non-adherent cells from The non-parametric Wilcoxon test was used for within
the well bottom, the biocompatibility was assessed by the use group comparisons of implant diameter reductions (R1-R3/
of a cell viability assay, which quantifies the key energy me- R6). Between group comparisons of mean EDX and CPS
tabolite adenosine triphosphate (CellTiter-Glo®, Promega, values were accomplished using analysis of variance
Mannheim, Germany) in a luminometer (Victor 2030, (ANOVA, post-hoc testing using Bonferroni’s correction)
PerkinElmer, Rodgau, Germany). The signal was assessed in and the unpaired t test, respectively. Results were considered
counts per second (CPS). After cell lysis induced by the statistically significant at p ≤ 0.05.
CellTiter-Glo® reagent, the titanium implants were removed
from the wells.
In particular, median implant diameters in the TL 3 mm/ element at all surface areas investigated. However, its quantity
6 mm groups changed from 3.05/3.00 mm by 0.10/0.20 mm to and the presence of other elements varied considerably be-
2.95/2.90 mm, respectively. Similarly, median implant diam- tween groups (Fig. 3d–f). In particular, IP and M surfaces
eters in the BL 3 mm/6 mm groups changed from 3.90/ were characterized by a comparable quantity (Wt%) of ele-
3.80 mm by 0.10/0.10 mm to 3.75/3.70 mm, respectively ments carbon (C), oxygen (O), sodium (Na), chloride (Cl),
(Table 1). Perforations of the implant walls were not observed. potassium (K), and silicon (Si), but a significantly different
quantity of elements nitrogen (N), Ti, and aluminum (Al). In
SEM and EDX analyses particular, mean Wt% values for element N was 6.3 ± 6.2 in
the M group, but amounted to 13.9 ± 5.3 at IP surfaces (p =
Representative SEM views of IP, M, and SLA surface areas 0.001, ANOVA). Conversely, mean Wt% for element Ti de-
are presented in Fig. 3a–c. creased from 84.8 ± 12.4 in the M group to 76.7 ± 10.0 at IP
IP commonly resulted in a smoothing of the sharp-edged surfaces (p = 0.023, ANOVA). At some specimens of the IP
SLA microstructure. However, the procedure was associated group, EDX analysis has also pointed to the presence of ele-
with the presence of irregular grooves and flat pits, which ment Al, thus reaching statistical significance over the M
appeared to be more uneven than the structure noted for M group (p = 0.001, ANOVA) (Table 2).
surfaces. When compared to SLA surfaces, significant differences
The quantification (Wt%) of chemical elements at IP, M, were noted for elements C, O, Na, Ti, and Al. In particular,
and SLA surfaces is summarized in Table 2. Basically, EDX mean Wt% values for element Ti was 55.0 ± 10.8 in the SLA
analysis revealed that titanium (Ti) was the predominant group, but amounted to 76.7 ± 10.0 at IP surfaces (p = 0.001).
Fig. 3 Representative SEM views (original magnification ×1000) and EDX—IP. e EDX—TL transmucosal machined. f EDX—TL
EDX quantification reports for different surface areas. a SEM—IP. b endosseous SLA
SEM—TL transmucosal machined. c SEM—TL endosseous SLA. d
Clin Oral Invest (2017) 21:2355–2361 2359
M refers to the the transmucosal machined part of TL implants and SLA to the endosseous structured part of TL
and BL implants
Comparisons to IP surfaces: *p < 0.05, ** p < 0.01, *** p < 0.001; ANOVA, respectively
Conversely, mean Wt% for elements C and O decreased from of the implants. These outcomes were neither influenced
16.8 ± 5.1 and 14.6 ± 6.1 in the SLA group to 6.8 ± 5.8 and 2.0 by the extend of the procedure (i.e., 3 or 6 mm) nor the
± 3.9 at IP surfaces (p = 0.001; p = 0.001, ANOVA), respec- macrodesign of the implant (i.e., BL or TL). In this con-
tively. Some specimens in the SLA group were associated text, it must be emphasized that a potential drawback of the
with the presence of elements Na and Al, thus reaching statis- present study was the lack of further analyses to evaluate
tical significance (p = 0.04; p = 0.01, ANOVA) over respective the mechanical strength of the smoothened implants.
IP surfaces (Table 2). However, recent in vitro data employing implants with a
tapered design have indicated that IP may weaken the
Cell viability assay strength of narrow-diameter specimens [14]. In particular,
the bending strength was significantly reduced from 613.9
Cell viability measurements after 6 days of incubation, ± 42.8 N (untreated control) to 511.4 ± 55.9 N (5 mm IP
expressed as mean CPS values, are presented in Table 3. In using 30/15-μm egg-shaped diamond burs, arkansas
both TL and BL groups, the extension of IP from 3 to 6 mm stones, and fine silicon polishers), thus resulting in a body
was associated with an increase in the luminescence signal. In fracture of all 3.75-mm diameter implants. In contrast, at
particular, mean CPS values increased from 255.25 ± 128.36 4.7-mm implants, the fractures solely occurred at the level
to 263.50 ± 84.64 in the TL 3/6 mm groups and from 155.25 of the abutment screws [14]. When interpreting the results
± 71.82 to 437.25 ± 289.21 in the BL 3/6 mm groups, respec- of the latter study, however, one has to keep in mind that
tively. In the BL group, mean CPS changes reached statistical the authors did not assess the diameter reductions, and
significance (p = 0.029, unpaired t test). therefore, it is impossible to estimate to what extend a
variation in the final thickness of the implant walls may
have influenced the measurements in respective groups.
Moreover, since the aforementioned study employed ta-
Discussion pered implants with different size parameters, it is
The aim of the present in vitro study was to further analyze the
influence of a specific IP protocol, which had been success-
Table 3 Cell viability at 6 days after seeding of gingival fibroblasts on
fully used in a series of clinical trials (i.e., combined surgical the M, IP, and SLA surface areas of BL and TL implants (n = 32 implants,
resective/regenerative therapy of peri-implantitis) [3, 8, horizontally positioned in non-binding 24-well plates)
10–13], on the diameter, chemical surface composition, and
Group Mean SD Median
biocompatibility of titanium implants.
Basically, it was observed that the reduction in implant TL 3 mm 255.25 128.36 192.0
diameter after a careful application of fine/extra-fine dia- TL 6 mm 263.50 84.64 269.0
mond burs and arkansas stones was commonly considered BL 3 mm 155.25 71.82 134.0
as minimal. As evidenced by the radiographic examina- BL 6 mm 437.25** 289.21 374.0
tions, the smoothing procedure was mainly restricted to
the threads and did obviously not affect the core diameter Compared to BL 3 mm: ** p < 0.05; unpaired t test
2360 Clin Oral Invest (2017) 21:2355–2361
impossible to estimate to what extend these results obtain- profiles noted for M and SLA areas basically corroborate
ed under different experimental conditions may be applied a recent elemental analysis of the same surfaces [20].
to either BL or TL implants. Nevertheless, the clinician When comparing these data with the specific characteris-
should be cautious when applying this procedure to tics obtained after IP, it was apparent that smoothened and
narrow-diameter implants. M surfaces revealed a comparable quantity (Wt%) of ele-
When further analyzing the results of the present study, it ments C, O, Na, Cl, K, and Si, but a significantly different
was noted that IP resulted in a smoothing of the sharp-edged quantity of elements Ti and Al. However, when compared
SLA microstructure, but still revealed the presence of irregular with SLA surfaces, IP resulted in a significantly different
grooves and flat pits. In this context, it must be emphasized quantity of elements C, O, Na, Ti, and Al. Similar EDX
that another drawback of the present analysis was a lack of profiles were also reported subsequent to IP using 30-mm
surface roughness measurements. However, the Ra value particle size egg-shaped burs [22]. In particular, the analy-
resulting from a similar sequence of diamond (106, 40, sis (refers to mass%) of elements Ti, O, Al, and C for SLA-
4 μm) and arkansas burs applied at SLA surfaced TL implants surfaced implants was 67.5, 26.75, 8.23, and 3.01. After
has recently been reported to be within the range of 0.39 ± surface treatment, these values were 92.28, 1.59, 3.3, and
0.13 μm [9]. These outcomes were slightly improved to 0.32 1.73 [22]. All these data taken together with the results of
± 0.14 μm, when the diamond burs were combined with sili- the present analysis seem to indicate that the elemental
cone polishers. In contrast, the Ra value assessed at the ma- composition of IP-modified surfaces closely resembles that
chined neck of TL implants was 0.10 ± 0.01 μm and of conventional M surfaces. In this context, however, one
amounted to 1.94 ± 0.47 μm at SLA surfaces [9]. has to keep in mind that the titanium implants/discs as used
Since silicone polishers are commonly associated with a in the present or latter study were unworn and the intraoral
pronounced pollution of the surgical site, one has to question contamination of commercially pure titanium foils was as-
their slight beneficial influence on surface roughness over sociated with an increase in C from 48 to 70% [23]. Since
arkansas stones. Even though a further Ra reduction, at least biofilm removal alone did not result in a reestablishment of
to a threshold level of 0.2 μm, appears to be desirable to limit the biocompatibility of contaminated titanium surfaces
plaque biofilm formation [15], the surface characteristics at [24], one might hypothesize that the effects of IP on the
the transmucosal aspect of titanium implants had obviously elemental composition and cell to implant interactions be-
no effect on the development of peri-implant soft tissue in- come even more relevant at diseased implant sites. These
flammation [16–18]. issues need to be carefully addressed in future studies.
Moreover, the results of a previous experimental animal In conclusion, the present analysis has indicated that IP
study employing the ligature-induced peri-implantitis de- applied to SLA implants was associated with (i) a minimal
fect model have indicated that this specific IP procedure diameter reduction, (ii) an undisturbed cell viability, and (iii)
(i.e., fine/extra-fine diamond burs and arkansas stones) re- a chemical elemental composition comparable to that noted
sulted in a smooth transmucosal area at SLA-surfaced im- for M surfaces.
plants, which in turn supported a close adhesion of the
subepithelial connective tissue. The formation of a new Acknowledgements The authors kindly appreciate the skills and com-
bone to implant contact after IP could not be identified in mitment of Dr. Daniel Martens (performing the implantoplasty) and Ms
any of the specimens investigated, thus indicating that Tina Hagena (in vitro analyses) (both Department of Oral Surgery,
Universitätsklinikum Düsseldorf, Düsseldorf, Germany).
these surfaces do obviously not attract osteoblasts [7]. A
positive effect of an extension of IP (i.e., 3 to 6 mm) on the
Compliance with ethical standards
cell viability of gingival fibroblasts has, at least in part
(refers to the BL group), also been confirmed by the pres- Conflict of interest The authors declare that they have no conflict of
ent study. In this context, it must be emphasized that the interest.
spreading and proliferation of fibroblast are commonly
pronounced at smooth when compared with rough titanium Funding The study was self-funded by the Department of Oral
surfaces [19, 20]. However, in addition to the microstruc- Surgery, Universitätsklinikum Düsseldorf, Germany. The titanium im-
ture, cellular interactions may also be influenced by the plants were kindly provided by the Institut Straumann AG, Basel,
Switzerland.
chemical elemental composition of the implant surface.
This mainly refers to the presence of potential contami- Ethical approval This article does not contain any studies with human
nants from the atmosphere, such as hydrocarbons and C, participants or animals performed by any of the authors.
which have been shown to reduce the surface free energy
and subsequently cell adhesion [21]. The present EDX Informed consent For this type of study, formal consent is not required.
Clin Oral Invest (2017) 21:2355–2361 2361