Effect of Dehiscences to the Bone Response of Implants With an Acid-Etched Surface: An Experimental Study in Miniature Pigs

In recent years, endosseous titanium implants have been proven to be a successful modality for treating completely and partially edentulous patients.1 Several implant surfaces have been used for the rehabilitation of the edentulous jaws, and an understanding of their biological behavior is of relevant importance to improve surface geometries and implant designs to achieve predictable results.

In 1991, Buser and coworkers2 evaluated the bone-to-implant contact (BIC) using 5 different titanium surfaces in miniature pigs. Histomorphometric analysis was performed after a relatively short healing period of 3 and 6 weeks, and implants with a sandblasted and acid-attacked surface (large grit; HCL/H₂SO₄) showed mean BIC values of 50% to 60%. More recent works have supported these findings, reporting a higher BIC with rougher implant surfaces when compared with smoother ones.3–6 According to De Lima Fernandez and colleagues,7 the bone formation of rough-surfaced implants follows a 3-dimensional pattern, with cellular extensions inside the surfaced irregularities, while in implants with smooth surfaces, bone deposition occurs in more uniform layers along the microgrooves produced by the machining process. Regarding the surface roughness, moderately rough surfaces (between 1.0 and 2.0 µm) seem to display stronger bone response, but the differences often were small and insignificant.8 

The characteristics of titanium implant surfaces have been modified by additive methods (eg, titanium plasma spray) to increase the surface area and provide a more complex surface macro-topography. Subtractive procedures (eg, blasting, acid etching) have also been used to raise the surface area and to alter its microtopography or texture.9 Among the subtractive methods of surface technology, acid etching is of particular interest because it creates a micro-textured surface that appears to enhance the early endosseous integration and stability of the implant.10 Several reports have been published, demonstrating that the machined and acid-etched surface was superior as compared with the machined one in both clinical11,12 and histometric studies.10,13 

During the osseointegration healing phase of submerged implants, complete mucosal coverage and isolation of the implant from the oral cavity avoid trauma and infection and establish favorable conditions for osseointegration.14,15 Whether spontaneous early exposure may be an additional harmful factor resulting in early crestal bone loss around submerged implants is still unclear. Hale and coworkers16 and De Porter and colleagues17 reported that the lack of soft-tissue coverage over the implants caused bone loss in the coronal third of the fixture. If healing bone is not allowed to remodel in an environment free of movement and bacterial invasion, then osseointegration will fail and connective tissue encapsulation will occur.18 By contrast, Lum et al19 noted in a study in Rhesus monkeys that several fixtures perforated the mucosa prior to abutment attachment but did not report about any adverse effects of premature uncovering. However, standardized readings to quantify bone loss were not performed.

The purpose of the present study was first to evaluate the BIC of an implant featuring a rough acid-etched surface after 3 months of submerged healing. By means of fluorescence microscopy, the peri-implant bone formation rate as well as the direction of ossification was assessed. Furthermore, the evaluation of the impact of observed dehiscences on the osseointegration process of implants with an acid-etched surface was carried out.

In this study, 15 titanium WI.TAL implants (Wieland Dental Implants GmbH, Wiernheim, Germany) with a length of 9 mm and a diameter of 3.5 mm were used. The surface of the WI.TAL implant was modified by means of acid etching reaching up to the platform and termed OsseoAttract surface.

Five female Göttingen mini pigs (Ellegard Göttingen Minipigs ApS, Dalmose, Denmark), 2 years of age with an average body weight of 32.2 kg, were used in this study. They were housed in pens with concrete floors and received a standard pelleted diet (ssniff Mpig-H, ssniff Spezialitäten GmbH, Soest, Germany) and water ad libitum. The miniature pigs were not fed 12 hours before anesthesia to prevent vomiting. For a period of 1 week preoperatively and 4 weeks postoperatively, the diet was softened with water, so that masticatory pressure on the healing sites would be minimized to avoid damaging the newly operated regions.

The study was approved by the regional authorities for health and social affairs under the reference number G 0403/05.

All surgeries were performed under sterile conditions. The animals were sedated with an intramuscular injection of Ketamine (WDT, Garbsen, Germany; 20 mg/kg), Atropine sulphate (Braun, Melsungen, Germany; 0.03 mL/kg), Stressnil (Janssen Cilag GmbH, Neuss, Germany; 0.05 mL/kg), and Rompun (Bayer AG, Leverkusen, Germany; 3 mg/kg). An endotracheal tube was used for intubation, and a mixture of isofluran (1.5%–2.0%) and N₂O∶O₂ (1∶1) was administered. For amplifying the analgesia, Fentanyl (Janssen-Cilag, Neuss, Germany; 0.1 mg intravenous [iv]) was applicated every 30 minutes during the surgical intervention. Volume replacement was performed using Ionosteril (Fresenius, Bad Homburg, Deutschland; 1000 mL iv) and HES 6% (Fresenius, Bad Homburg, Deutschland; 500 mL iv). An antibiotic regimen comprising the intramuscular (im) application of Baytril 10% (Bayer AG, Leverkusen, Germany; 3 mg/kg once per day) and Borgal 24% (Intervet Deutschland GmbH, Unterschleißheim, Germany; 4.5 mg/kg once per day) was performed 24 hours preoperatively, operatively, as well as after the surgical procedures for 3 more weeks. Postoperative pain control was procured by im application of Temgesic (Essex Pharma, Grünenthal, Germany; 0.03 mL/kg thrice per day) for 3 days and Rimadyl (Pfizer GmbH, Karlsruhe, Germany; 0.06 mL/kg once per day) for 7 more days.

In the first surgical procedure, bilateral extractions of the premolars and the first molar of the maxilla and mandible were performed using extended mucoperiosteal flaps. After wound closure, the sites were allowed to heal for at least 2 months to achieve a fully healed alveolar crest in the upper and lower jaw. In the second surgery, dental implants were inserted according to the surgical guidelines provided by the manufacturer. For this, buccal and lingual mucoperiosteal flaps were elevated, and a total of 15 acid-etched implants were inserted in the region of the first molar of the upper respectively lower jaw. Whereas 2 mini pigs (Nos. 12, 25) received 4 implants, 1 animal (No. 24) got 3 fixtures, and in 2 miniature pigs (Nos. 22, 26) 2 implants were placed. The exact position of the implants is presented in Table 1. Whereas the implants inserted in the maxilla were tested under shortened healing conditions, the fixtures placed in the lower jaw served as control. To avoid loading, a submerged approach was performed. The mucosal tissues were adapted above the implants with continuous sutures using absorbable suture material (Monocryl 3-0, Johnson & Johnson, St-Stevens-Woluwe, Belgium).

The animals were inspected during the first few postoperative days for signs of wound dehiscence or infection and weekly thereafter to assess general health. An additional wound control under sedation was performed at weeks 2, 5, and 8 postimplantation.

The sequential administration of fluorescent dyes was allowed to follow the direction and the topographic localization of new bone formation. During the mineralization process, the fluorescent dyes were incorporated in the matrix of the front of mineralization by a chelation.20 The amount of new bone formation during a specific time period can be depicted by evaluation of the distance between the different fluorochrome labels.21 Therefore, the miniature pigs received fluorochromes in a specific time range. Xylenol orange (Sigma-Aldrich Chemie GmbH, Munich, Germany; 90 mg/kg), calcein green (Sigma-Aldrich Chemie GmbH, Munich, Germany; 20 mg/kg), and alizarin complexone (VWR International GmbH, Darmstadt, Germany; 30 mg/kg) were administered intravenously 2, 5, and 8 weeks postimplantation. The application of the fluorochromes was performed under sedation in the same dosage mentioned above.

At the end of the experimental period (12 weeks after implant insertion), the animals were killed by an iv application of Trapanal (Byck-Gulden, Konstanz, Germany; 2 g/animal), Pancuronium (Delta-Select, Pfullingen, Germany; 4 mg/animal), and a solution of 7.45% potassium chloride (60 mval/animal). Following euthanasia, the block specimens containing the implants and the surrounding tissues were dissected from all of the animals.

The implants, embedded in the surrounding bone, were fixed in a neutral solution of formaldehyde and alcohol for 2 weeks. Following, the specimens were dehydrated in a graded series of ethanol and after all degreased by pure xylol. Thereafter, the samples were embedded in methylmetacrylate (Technovit 9100 Neu, Haraeus Kulzer GmbH & Co. KG, Wehrheim, Germany). Using the sawing and grinding technique according to Donath and Breuner,22 longitudinal sections were ground to about 150 µm. Via an exact microgrinding system (Exakt Apparatebau GmbH, Norderstedt, Germany), further grinding and polishing procedures were conducted to obtain sections of 40- to 60-µm thickness.

After fluorescence microscopic analysis, the sections were tinged by von Kossa staining.

The fluorescence and light-microscopic analysis was performed using the photo microscope Axiophot (ocular: PI 10×/25 Br foc, used objectives: plan-neofluar 10×/0, 30; plan-neofluar 20×/0, 50; Carl Zeiss, Göttingen, Germany). Via an AxioCam MRc 5 camera (5 megapixel [2584 × 1936], 36-bit RGB shade, sensor size 2/3; Carl Zeiss), the favored details were recorded and digitized. Using the AxioVision Rel.4.6 program (Carl Zeiss), the histological images were calibrated and the BIC was measured.

To analyze the rate of osseointegration at the apical, middle, and coronal level, the fixture was divided into sections (Figure 1a). The BIC ratio was defined as the length of the bone surface border in direct contact with the implant compared with the complete implant periphery (%) starting from the implant shoulder up to the apex.

To evaluate peri-implant bone regeneration, the fixture was divided into sections (Figure 1b). One longitudinal histologic mesial-distal section from each implant was evaluated using a fluorescence microscope Axiophot (Carl Zeiss) with appropriate excitation and barrier filter combinations (calcein filter No. 02, xylenol orange filter No. 09, alizarine complexone filter No. 14; Carl Zeiss). The bone mineral apposition rate (in µm) was expressed as the mean distance between the different fluorescence labels divided by the interval of days between the 3 administrations of the fluorescence dyes. At magnification up to ×20, the images were assessed digitally by Axiocam (Carl Zeiss). The AxioVision imaging software (Carl Zeiss) was used for image analysis.

Besides this quantitative data acquisition, it was also possible to distinguish the different directions of new bone formation and to make statements about the degree of implantofugal and implantopetal bone growth. To quantify this, in each of the 9 implant sections, the direction of neo-ossification was estimated (sum per implant = 9), whereas contact osteogenesis implied a direct contact of the xylenol band to the implant surface. The incidence of a xylenol band close to the surrounding bone bed as well as an alizarin band adjacent to the implant surface was seen as indicators of distance osteogenesis.

Statistical analysis was performed with SPSS V13.0 and SAS V9.1 software (SPSS Inc, Chicago, Ill). Differences between independent samples were tested with the Mann-Whitney U test. The Wilcoxon signed ranks test was used to compare 2 interdependent parameters in the same subject for significant differences. Varieties of more than 2 dependent samples were performed by the Friedman test. All tests were 2 tailed, and statistical significance was set at P < .05.

The animals recovered well after surgery, as documented by daily controls of the general condition, with 1 exception (No. 12) showing an elongated convalescence. A control of wound healing was regularly performed under sedation. A high incidence of mucosal dehiscence was observed, whereas 9 of 15 inserted implants displayed dehiscences during the first 8 weeks postoperatively, up to which they had gradually closed. The precise distribution of the occurred dehiscences is presented in Table 2. All implants were clinically stable during the entire experimental period.

The application of the fluorochrome dyes (xylenol orange, calcein green, alizarin complexone), however, raised difficulties. Whereas the administration of the former fluorochromes was unproblematic, the iv injection of alizarin complexone implicated many difficulties. All probands reacted to the application of alizarin complexone with shock symptoms, presenting a soft pulse, increased heart rate, and brady-respectively apnea.

Histology revealed an osseous anchorage of all implants. No histological signs of inflammation were present. The bony apposition revealed a lamellar and in some parts trabecular structure containing individual osteoblasts at the surface of bone trabeculae.

The BIC data ranged from 14.51% to 68.97% (Figure 2), whereas the mean value was 54.19%. A precise schedule of all BIC values is depicted in Table 2. The comparison between the acid-etched implants inserted in the upper respectively lower jaw revealed for the maxilla (test) a mean BIC of 52.72% (minimum, 14.51%; maximum, 68.77%), whereas in the mandible (control), the value constituted 55.87% (minimum, 41.34%; maximum, 68.97%). The results did not show a significant difference (P = .908).

A comparison between the mean BIC values of implants showing dehiscences (50.65%) to those without dehiscences (59.5%) was performed. This difference was not significant (P = .29).

The analysis of the particular implant sections revealed a mean BIC value of 40.49% for the coronal level of all implants. The middle section showed a mean BIC of 58.43%, whereas the value of the apical section averaged 55.74%. The difference between the coronal and the middle (P = .036) was significant, whereas the difference between the middle and the apical section (P = .165) was not significant.

The mean BIC values for the coronal region of the implants inserted in the lower respectively upper jaw (Figure 3) constituted 46.03% and accordingly 35.64%, whereas the difference was not statistically significant (P = .298). The data for the middle section averaged 58.32% for the maxilla and 58.55% for the mandible. The difference was also not significant (P = .728). Values of 51.3% and 60.81% were detected for the apical section of the upper respectively lower jaw, not offering a statistically significant difference (P = .643).

Also, the separate consideration of the BIC of the coronal part of the fixture revealed no significant difference (P = .814) between implants, with (41.17%) respectively without dehiscences (39.48%).

The investigation of new bone formation was performed using polychrome sequential labeling with 3 different dyes (xylenol orange, calcein green, alizarin complexone) administrated at a predefined time. The measurement and statistical analysis of the different fluorescent bands revealed a mean new bone formation rate of 2.33 µm/d (minimum, 1.75 µm/d; maximum, 3.16 µm/d) for the first time period (week 3 to 5 postimplantation). Between week 6 and 8 after implantation (second time period), the daily neo-ossification rate was 1.99 µm (minimum, 1.55 µm; maximum, 3.01 µm). The mean new bone formation rate for the whole time period (week 3 to 8 postimplantation) was 2.32 µm/d (minimum, 1.76 µm/d; maximum, 2.82 µm/d).

The new bone formation rate of the different implant sections for the first and second time period is shown in Table 3. It was particularly noticeable that the neo-ossification rate for all subdivided regions decreased during the observation period. Only in parts of the middle region (subdivision top and bottom) were the differences significant (P = .03; P = .035).

The difference concerning the neo-ossification rate of the upper respectively lower jaw was not significant (P > .05).

Besides the quantitative determination of new bone formation, fluorescence microscopy also allows for the assessment of the direction of ossification. For this, each of the 9 implant sections was estimated in terms of the sequence of the fluorochrome bands related to the implant surface. Consecutively, a panoramic view (Figure 4) and detail views (Figures 5–7) illustrated the different directions of new bone formation. Because of the arrangement of the fluorescent bands, both kinds of bone formation were detected in the present study. Whereas 46.97% of all investigated implant sections have shown an implantofugal bone growth, 53.03% of all analyzed segments indicated distance osteogenesis. This difference was not significant (P = .5).

The direct contact of the xylenol orange band with the implant surface suggested an incipient contact osteogenesis 2 weeks after implantation (Figure 6b). In other parts, there was a space observable between the implant surface and the xylenol orange band, composed of bone trabeculae, indicating an earlier start of bone formation within the first 2 weeks (Figure 6a).

Nowadays, osseointegration is still a topic of interest, aimed at solving the related shortcomings and guaranteeing a reliable long-term success, particularly under compromised local implantation conditions, notably low bone density and reduced bone height.23–25 Previous studies have shown that implants with surface treatment exhibit more BIC in the early stages of osseointegration,2,26–28 and on this account, several modifications of specific surface properties such as topography, structure, chemistry, surface charge, and wetability have been investigated.29,30 

In the present study, the authors specifically assessed the success of an implant modified by means of acid etching reaching up to the platform. Although other acid-etched dental implants are now available, the success and predictability of a particular implant must be based on its own merits, since even small alterations in an implant design can affect its outcome.31 From a scientific point of view, it is difficult to define particular osseoattractiveness and differentiate such surfaces from moderately roughened surfaces that are quite attractive for bone formation. All implants described as osseoattractive may, in fact, share the characteristic of being moderately roughened and thereby more attractive for new bone formation than smoother turned or rougher plasma-sprayed implants.8 

This research model provided the opportunity to directly compare the rate of osseointegration of acid-etched implants inserted in bone of different qualities, whereas the results revealed no significant difference between the BIC values of the upper respectively lower jaw. On the basis of this outcome, it could be assumed that the poorer bone quality of the maxilla did not interfere with the unloaded osseointegration within the shortened healing period of 12 weeks. This statement was in accordance with other in vivo studies in miniature pigs that have shown that surface treatment of titanium implants increases BIC ratios in short-term periods compared with machined implants, even in bone with poor quality.2,5,32 

A study by De Lima Fernandes et al7 in rabbits investigated acid-etched surfaces compared with machined-turned surfaces of implants inserted in the tibia 9 weeks postimplantation. It was particularly noticeable that despite a shorter healing time of 9 weeks, the BIC value of the acid-etched surface ranged clearly above the presently determined data, but a comparison of the data is inadequate as the rabbit shows a faster bone turnover compared with other species.33,34 According to Roberts,35 the bone turnover rate of rabbits is 3-fold higher compared with that in humans. Because of the similarity of the bone regeneration rate between humans and pigs,36 the same ratio of bone formation could be assumed for mini pigs.

Mangano and colleagues6 also evaluated the bone response to acid-etched respectively machined surfaced implants inserted in the lower jaw of nonhuman primates and humans. Because of their phylogenic similarity to human beings, nonhuman primates are an attractive animal model for studies on bone growth.37 The investigation demonstrated a higher BIC in the acid-etched implants compared with the machined implants in both primates and humans at 1 and 2 months, respectively, after implantation. The BIC value of the acid-etched surface in nonhuman primates was clearly below the determined findings of the present investigation. This discrepancy could be explained by the different healing period of the study. The BIC in humans was within the range of the results within this study.

A clinical study revealed a similar BIC of dual acid-etched surfaces after 2 months of healing in the posterior human maxilla.38 The considerable wide BIC range observed by Trisi and colleagues38 was also demonstrated in our investigation.

All implant surfaces were modified in terms of acid etching. The BIC data of the cited studies were widespread, in part due to small but maybe relevant differences in the surface topography, varying surgical techniques, and different animal models. Also, the selection of the used acids as well as the sequence of processing may be of importance.39 

A study design analogous to that of the present investigation (animal species, healing time, terms of healing) but with the appropriation of different treated implant surfaces was conducted by Zechner and coworkers.5 They examined the osseous healing characteristics of machined surface implants, HA-coated implants, and anodized titanium surface implants after a healing time of 3, 6, and 12 weeks, which were inserted into the mandibles of 12 adult mini pigs. Because of the similarity of the experimental protocol conducted by Zechner et al5 to that of the current investigation, an approximate comparison can be made with regard to the different roughened implant surfaces. The rough-surfaced implants investigated by Zechner and colleagues5 offered comparable BIC values to that determined in the current study. The BIC data described by Zechner et al5 were collected from the mandible after a recommended healing protocol of 3 months.

The high incidence of mucosal dehiscences observed in our investigation was also found by Zechner and colleagues5 and was supposedly provoked by the habits and parafunctional masticatory load of the miniature pigs. Despite a softened diet, the animals may have exerted excessive stress by heavy chewing patterns and by gnawing their cages. Whereas Hale et al16 stated that the mini pig appeared to be a suitable animal model for use in implant research, our opinion is that this aforementioned characteristic, among further anatomical features, challenge the assignment of mini pigs. The frequent occurrence of observed dehiscences could have also been associated with cross-linked resorbable membranes, placed in close proximity (1 tooth distance away) from the investigated acid-etched implants. The analysis of this special type of collagen membrane was the subject of further study and has been performed concomitantly to economize the number of experimental animals. The comparison between the mean BIC values of implants showing dehiscences to those without dehiscences offered no significant difference, suggesting that a dehiscence does not compromise osseointegration. The division of the implant in sections permitted a more detailed analysis of the BIC and was also applied by Hale and coworkers16 to evaluate the osseointegration rate of Brånemark implants. In the current study, there was a significant greater bone contact in the middle and apical region vs the coronal section of the implant. The detected low BIC values of the coronal region were independent of the occurrence of mucosal dehiscences. This observation was in accordance with the study by Hale and coworkers16 that stated that the lack of soft-tissue coverage over the implants did not significantly affect the osseointegration process in the coronal third. In contrast to the present investigation, Hale and coworkers16 examined only dental implants showing dehiscences, so that a comparison to fixtures without mucosal exposure was missing. Therefore, their conclusion has to be scrutinized, and although it framed the same statement, a direct comparison was not possible because they compared the coronal part of the implant with the middle and apical section. Unlike that approach, we matched the BIC of the coronal third of implants with and respectively without dehiscences. The crestal bone loss observed in the present investigation occurred independently from the incidence of dehiscences. A correlation between spontaneous early exposure of submerged endosseous implants and crestal bone loss, reported inter alia by Van Assche et al40 and Tal and coworkers,41 could not be verified in the present investigation. The observed crestal bone loss was attributed to the implant geometry rather than to the missing mucosal coverage.

A further factor influencing crestal bone changes was investigated by Hermann and coworkers.42 They claimed that these changes were dependent on the surface characteristics of the implant and the presence/absence as well as the location of an interface (microgap). If the microgap of the implant-abutment connection was located apically to the alveolar crest, bone loss occurred. Herrmann et al42 also stated that an epicrestal position of the microgap, as conducted in the present study, caused crestal bone loss. The surgical technique of submerging or not submerging the implant, however, has no influence on the amount of crestal bone loss that occurs.42 

Polychrome sequential labeling was applied to analyze the ossification because this method offered a dynamic picture of the time sequences in bone turnover. This procedure is well established and has been used in several studies.43,44 From the findings of the present and a prior study,45 we discourage the intravenous administration of alizarin complexone in miniature pigs and advise choosing a different fluorochrome for polychrome sequential labeling or a different application procedure. In the present trial, the new bone formation rate was higher in the first weeks. The current literature provides only few fluorescence microscopic data concerning the new bone formation rate with respect to the bone mineral apposition rate (BMAR) in miniature pigs. Hönig and Merten46 examined the new bone formation rate of the tibia in mini pigs, also showing a reduction of the neo-ossification rate during the observation period. The mean bone formation rate was lower in the tibia in mini pigs. The rationale for this was the unequal remodeling dynamics in various bones. There was preliminary evidence to suggest that bone turnover is 10-fold greater in the mandibular process of certain teeth than in the mid shaft of the tibia in a canine model.47 Also, Huja and colleagues48 denoted bone turnover in the canine alveolar process of the maxilla and mandible that was 3- and 6-fold higher, respectively, than that ascertained for the femur.

Nkenke et al43 compared the BMAR of immediately loaded implants with an unloaded control during the early healing phase in the partially edentulous mandible of miniature pigs. The mean BMAR values for an observation period of 2 until 16 weeks postimplantation did not differ from the values found within this study. A minimal decrease of the BMAR during the observation period was also identified.

Polychrome sequential labeling not only offers the chance to ascertain the amount of new bone formation but also gives detailed information about the direction of the ossification.49 The combination of recruitment and migration of osteogenic cells (osteoconduction) and bone formation by those cells on the implant surface is known as contact osteogenesis (implantofugal bone growth).50 Lowenberg and coworkers51 demonstrated that differentiating osteoblasts were capable of laying down a mineralized collagen-free matrix in direct contact with the metal oxide surface of titanium. These cells have shown an initial attachment to rough titanium surfaces, and furthermore, the titanium surface roughness has been shown to affect osteoblast proliferation and differentiation.52 By contrast, in distance osteogenesis (implantopetal bone growth), new bone is formed on the surface of old bone at the peri-implant site.51 

A study by Piatelli and coworkers53 indicated that the incidence as well as the time of a beginning contact osteogenesis seemed to be dependent on the configuration of the implant surface. Using light microscopical analysis, they examined the type of bone growth around machined and sandblasted implants inserted in the femur of rabbits. A different type of bone growth was found: in the first group, the bone growth was implantopetal (ie, from the host bed toward the implant surface), whereas the second group showed an implantofugal growth (ie, from the implant toward the host bed). Huang et al54 evaluated implants with a porous modified TiO₂ surface (TiUnite) in cynomolgus monkeys and performed a sequential labeling 2, 3, 4, and 16 weeks postimplantation. The first labeling with alizarin complexone could not be identified in any of the 24 implants. Because of this outcome, they followed that new bone formation started not until the third week after implant insertion. This finding was in contrast to that of our study, in which neo-ossification started before the second week after implant placement. In a further study, Eckelt and colleagues55 coated the implant surfaces with components of the extracellular matrix. The fluorescence microscopic analysis revealed bone growth starting in the surrounding peri-implant bone toward the implant surface. However, in single cases, extracellular matrix–coated implants showed an inversion of bone growth. Thus, bone formation started on the implant surface in an implantofugal direction toward the peri-implant tissue. Biomechanical interactions between the implant surface and endogenous bone morphogenic protein and accordingly cytokines were discussed by Eckelt et al55 as a possible explanation for the implantofugal bone formation, observed for extracellular matrix–coated surfaces.

The investigated acid-etched implants (WI.TAL) have also shown a concomitant occurrence of distance and contact osteogenesis. In consideration of the outcome of the present investigation as well as that of the cited studies, the question arises as to which local conditions must be given and which attributes the implant surface should possess to generate a contact osteogenesis as early as possible.

It could be assumed that acid-etched implants offered a prerequisite to osseointegrate under a shortened healing period. The observed dehiscences seemed not to have compromised the rate of osseointegration. The crestal bone loss was not dependent on the missing mucosal coverage. The rate of new bone formation was comparable with that obtained in other mini pig studies.

BIC

bone implant contact

BMAR

bone mineral apposition rate

The authors thank Mrs Siebert for the statistical evaluation of the established data. The study was financed by a grant from Wieland Dental Implants GmbH.