Evaluation of the Surrounding Ring of Two Different Extra-Short Implant Designs in Crestal Bone Maintanence: A Histologic Study in Dogs

The aim of this study was to compare the implant stability and bone resorption and formation of two different extra-short implant designs with different diameter rings placed in a dog´s maxilla. Thirty-six extra-short, 5 mm diameter × 4 mm length (Short DM®, Bioner Sistemas Implantológicos, Barcelona, Spain), delayed implants were placed in each hemimaxilla of six dogs at the bone crest level. Eighteen implants of each design (wide and narrow ring) were installed. After 8 and 12 weeks of healing, histomorphometric analyses of the specimens were carried out to measure the crestal bone level values and the tissue thickness around the wide and narrow ring implant designs. In the microscopic analysis, less buccal bone resorption was observed in the narrow ring implants with a statistical significance (p < 0.001). For the peri-implant tissue thickness, the distance from the implant shoulder to the external portion of the epithelium was significantly higher for the implants installed with a wide ring with statistical significance (p < 0.001). Our findings suggest that the amount of peri-implant tissues (crestal bone loss) after remodeling over a period of 12 weeks was smaller in the narrow ring extra-short implant installed in the healed maxilla, compared with the wide ring extra-short implants.

In the clinical examination, all of the animals had a good general health; their maxillas' were all intact, with minimal resorption and without major oral lesions.
The animals were given vaccines and vitamins against rabies, and were then put in quarantine. The dogs were kept in individual cages throughout the project and they also received adequate veterinary care. After each surgery (two procedures), the animals received an antibiotic of 6 mg/kg Clindamycin (Clindaseptin 75 mg, Chanelle Pharmaceuticals, 20 Ireland) twice daily, and an anti-inflammatory of 0.30 mg/kg Caprox Vet 100 mg (Vibrac, Spain) three times per day, systemically.

Surgical Procedure
The animals were pre-anaesthetized with acepromazine (0.12-0.25% mg/kg), buprenorphine (0.01 mg/kg), and medetomidine (35 µg/kg). The mixture was injected intramuscularly in the femoral quadriceps. The animals were then taken to the operating theater where, at the earliest opportunity, an intravenous catheter was inserted (diameter 22 or 20 G) into the cephalic vein, and propofol was infused at a rate of 0.4 mg/kg/min, at a slow constant infusion rate. The conventional dental infiltration anesthesia (articaine 40 mg, 1% epinephrine) was administered at the surgical sites. These procedures were carried out under the supervision of a veterinary surgeon. Maxilary premolar extractions (P2, P3, and P4) were performed bilaterally. After two months of healing, the crestal incisions were performed bilaterally in the premolar region of the maxilla. The full-thickness mucoperiosteal flaps were elevated, and the recipient sites in the premolar regions on both sides of the maxilla were prepared for the present experiment, while the other regions were used for different experimental purposes, the results of which are reported elsewhere. The healed bones were prepared in order to place extra-short implants with two different types of rings. The tested implant was a tissue level implant with a 1.9 mm smooth neck, therefore leaving space for the biological width and for reducing the marginal bone loss; this helps us measure the marginal bone reaction to the tested ring device.
Thirty-six implants, Short DM ® (Bioner, Sistemas Implantológicos, Barcelona, Spain), of 4 mm in length with a 5 mm diameter, were placed. One implant was used with a narrow cervical ring with a 4.2 mm diameter, and the other was a wide cervical ring with a 5.3 mm diameter ( Figure 1). In the clinical examination, all of the animals had a good general health; their maxillas' were all intact, with minimal resorption and without major oral lesions.
The animals were given vaccines and vitamins against rabies, and were then put in quarantine. The dogs were kept in individual cages throughout the project and they also received adequate veterinary care. After each surgery (two procedures), the animals received an antibiotic of 6 mg/kg Clindamycin (Clindaseptin 75 mg, Chanelle Pharmaceuticals, 20 Ireland) twice daily, and an antiinflammatory of 0.30 mg/kg Caprox Vet 100 mg (Vibrac, Spain) three times per day, systemically.

Surgical Procedure
The animals were pre-anaesthetized with acepromazine (0.12-0.25% mg/kg), buprenorphine (0.01 mg/kg), and medetomidine (35 μg/kg). The mixture was injected intramuscularly in the femoral quadriceps. The animals were then taken to the operating theater where, at the earliest opportunity, an intravenous catheter was inserted (diameter 22 or 20 G) into the cephalic vein, and propofol was infused at a rate of 0.4 mg/kg/min, at a slow constant infusion rate. The conventional dental infiltration anesthesia (articaine 40 mg, 1% epinephrine) was administered at the surgical sites. These procedures were carried out under the supervision of a veterinary surgeon. Maxilary premolar extractions (P2, P3, and P4) were performed bilaterally. After two months of healing, the crestal incisions were performed bilaterally in the premolar region of the maxilla. The full-thickness mucoperiosteal flaps were elevated, and the recipient sites in the premolar regions on both sides of the maxilla were prepared for the present experiment, while the other regions were used for different experimental purposes, the results of which are reported elsewhere. The healed bones were prepared in order to place extra-short implants with two different types of rings. The tested implant was a tissue level implant with a 1.9 mm smooth neck, therefore leaving space for the biological width and for reducing the marginal bone loss; this helps us measure the marginal bone reaction to the tested ring device.
Thirty-six implants, Short DM ® (Bioner, Sistemas Implantológicos, Barcelona, Spain), of 4 mm in length with a 5 mm diameter, were placed. One implant was used with a narrow cervical ring with a 4.2 mm diameter, and the other was a wide cervical ring with a 5.3 mm diameter ( Figure 1).  According to the ARRIVE, the information about the allocation/randomization must to be provided. Information about the allocation/randomization of a total of 36 implants were randomly installed. Eighteen extra-short dental implants, six per dog, with wide diameter ring (5.3 mm), and 18 with a narrow diameter ring (4.2 mm), were installed in the healed maxillas (Figures 2 and 3). According to the ARRIVE, the information about the allocation/randomization must to be provided. Information about the allocation/randomization of a total of 36 implants were randomly installed. Eighteen extra-short dental implants, six per dog, with wide diameter ring (5.3 mm), and 18 with a narrow diameter ring (4.2 mm), were installed in the healed maxillas (Figures 2 and 3).  The flaps were sutured with silk 4.0 (Lorca Marin, Lorca Murcia, Spain). After the surgical procedures, the animals received antibiotic treatment (Amoxicillin 500 mg, twice a day) and analgesics (ibuprofen 600 mg, three times a day), systemically. In addition, the dogs were fed a soft diet for seven days, and plaque control was maintained by the application of Sea4 Encías  (Blue Sea Laboratories, Alicante, Spain). The wounds were inspected daily for postoperative clinical complications. Two weeks after surgery, the sutures were removed.

Histological and Histomorphometric Analysis
Three of the animals were sacrificed after 8 weeks, and the other three animals were sacrificed after 12 weeks, after the insertion of the implant, through an overdose of Pentothal Natrium ® (Laboratorios Abbot, Madrid, Spain), and were perfused through the carotid arteries with a fixative According to the ARRIVE, the information about the allocation/randomization must to be provided. Information about the allocation/randomization of a total of 36 implants were randomly installed. Eighteen extra-short dental implants, six per dog, with wide diameter ring (5.3 mm), and 18 with a narrow diameter ring (4.2 mm), were installed in the healed maxillas (Figures 2 and 3).  The flaps were sutured with silk 4.0 (Lorca Marin, Lorca Murcia, Spain). After the surgical procedures, the animals received antibiotic treatment (Amoxicillin 500 mg, twice a day) and analgesics (ibuprofen 600 mg, three times a day), systemically. In addition, the dogs were fed a soft diet for seven days, and plaque control was maintained by the application of Sea4 Encías  (Blue Sea Laboratories, Alicante, Spain). The wounds were inspected daily for postoperative clinical complications. Two weeks after surgery, the sutures were removed.

Histological and Histomorphometric Analysis
Three of the animals were sacrificed after 8 weeks, and the other three animals were sacrificed after 12 weeks, after the insertion of the implant, through an overdose of Pentothal Natrium ® (Laboratorios Abbot, Madrid, Spain), and were perfused through the carotid arteries with a fixative The flaps were sutured with silk 4.0 (Lorca Marin, Lorca Murcia, Spain). After the surgical procedures, the animals received antibiotic treatment (Amoxicillin 500 mg, twice a day) and analgesics (ibuprofen 600 mg, three times a day), systemically. In addition, the dogs were fed a soft diet for seven days, and plaque control was maintained by the application of Sea4 Encías ® (Blue Sea Laboratories, Alicante, Spain). The wounds were inspected daily for postoperative clinical complications. Two weeks after surgery, the sutures were removed.

Histological and Histomorphometric Analysis
Three of the animals were sacrificed after 8 weeks, and the other three animals were sacrificed after 12 weeks, after the insertion of the implant, through an overdose of Pentothal Natrium ® (Laboratorios Abbot, Madrid, Spain), and were perfused through the carotid arteries with a fixative containing 5% glutaraldehyde and 5% formaldehyde. Radiographs were taken after sacrifice at 60 days for the first three dogs, and at 90 days for the remaining three ( Figure 4). containing 5% glutaraldehyde and 5% formaldehyde. Radiographs were taken after sacrifice at 60 days for the first three dogs, and at 90 days for the remaining three ( Figure 4). The specimens were washed in saline and were fixed in a 10% buffered formalin. The specimens were processed to obtain a thin section of soil with the automated system Precise 1 (Assing, Rome, Italy). The specimens were dehydrated in ascending series with alcohol, and were embedded in a glycol methacrylate resin (Technovit 7200 VLC, Kulzer, Wehrheim, Germany). After polymerization, the specimens were sectioned along their longitudinal axis with a high precision diamond disk, at about 150 to 30 μm. A total of two slides were obtained for each implant ( Figure 5). The slides were stained with toluidine blue and were observed under a normal transmitted light microscope and a polarized light microscope (Leitz, Wetzlar, Germany).
The histological preparation evaluates the distance from the top of the implant collar to the first contact of the buccal and lingual bone (BBC and LBC), as well as the heights of the buccal and lingual The specimens were washed in saline and were fixed in a 10% buffered formalin. The specimens were processed to obtain a thin section of soil with the automated system Precise 1 (Assing, Rome, Italy). The specimens were dehydrated in ascending series with alcohol, and were embedded in a glycol methacrylate resin (Technovit 7200 VLC, Kulzer, Wehrheim, Germany). After polymerization, the specimens were sectioned along their longitudinal axis with a high precision diamond disk, at about 150 to 30 µm. A total of two slides were obtained for each implant ( Figure 5). containing 5% glutaraldehyde and 5% formaldehyde. Radiographs were taken after sacrifice at 60 days for the first three dogs, and at 90 days for the remaining three ( Figure 4). The specimens were washed in saline and were fixed in a 10% buffered formalin. The specimens were processed to obtain a thin section of soil with the automated system Precise 1 (Assing, Rome, Italy). The specimens were dehydrated in ascending series with alcohol, and were embedded in a glycol methacrylate resin (Technovit 7200 VLC, Kulzer, Wehrheim, Germany). After polymerization, the specimens were sectioned along their longitudinal axis with a high precision diamond disk, at about 150 to 30 μm. A total of two slides were obtained for each implant ( Figure 5).  The slides were stained with toluidine blue and were observed under a normal transmitted light microscope and a polarized light microscope (Leitz, Wetzlar, Germany).
The histological preparation evaluates the distance from the top of the implant collar to the first contact of the buccal and lingual bone (BBC and LBC), as well as the heights of the buccal and lingual bone ridges, with respect to the neck of the implant (Figures 6 and 7). The resorption of the buccal bone wall compared to the reabsorption of the lingual bone wall was expressed as a linear measure.
Materials 2018, 11, x FOR PEER REVIEW 6 of 11 bone ridges, with respect to the neck of the implant (Figures 6 and 7). The resorption of the buccal bone wall compared to the reabsorption of the lingual bone wall was expressed as a linear measure.  The buccal and lingual bone plates were measured from the implant shoulder to the first Bone to implant Contact (BIC) and to the top of the bony crest. The percentage of BIC of the native bone was also measured along the perimeter of the implant, between the coronal end of the osseointegration in the buccal and lingual aspects. The apical portion of each implant was excluded from the measurement. The total amount of bone in contact with the implants was calculated as the sum of the native bone and the newly formed bone (BIC%). A histomorphometry of the BIC bone ridges, with respect to the neck of the implant (Figures 6 and 7). The resorption of the buccal bone wall compared to the reabsorption of the lingual bone wall was expressed as a linear measure.  The buccal and lingual bone plates were measured from the implant shoulder to the first Bone to implant Contact (BIC) and to the top of the bony crest. The percentage of BIC of the native bone was also measured along the perimeter of the implant, between the coronal end of the osseointegration in the buccal and lingual aspects. The apical portion of each implant was excluded from the measurement. The total amount of bone in contact with the implants was calculated as the sum of the native bone and the newly formed bone (BIC%). A histomorphometry of the BIC The buccal and lingual bone plates were measured from the implant shoulder to the first Bone to implant Contact (BIC) and to the top of the bony crest. The percentage of BIC of the native bone was also measured along the perimeter of the implant, between the coronal end of the osseointegration in the buccal and lingual aspects. The apical portion of each implant was excluded from the measurement. The total amount of bone in contact with the implants was calculated as the sum of the native bone and the newly formed bone (BIC%). A histomorphometry of the BIC percentages was performed using a light microscope (Laborlux S, Leitz) connected to a high-resolution video camera (3CCD, JVC KY-F55B, JVC ® , JVC, Yokohama, Japan), and interconnected to a monitor and Personal Computer (Intel Pentium III 1200 MMX, Intel ® , Intel, Santa Clara, CA, USA). This optical system was associated with a scanning pad (Matrix Vision GmbH, Oppenweiler, Germany) and a software package for histometry, with image capturing capabilities (Image-Pro Plus 4.5, Media Cybernetics Inc., Immagini and Computer Snc, Milano, Italy). The total amount of bone in contact with the implants was calculated as the sum of the native bone and the newly formed bone.

Statistic Analysis
The data were compared using the one-way ANOVA (Analysis of Variance) statistical tests (α = 5%), because we had two different periods of time for the evaluation (8 and 12 weeks) and two different types of implants.
The mean values and standard deviations were calculated using a BIC descriptive test and the bone resorption measurements. The values were recorded as the mean ± standard deviation. The Wilcoxon test was applied to the comparison of the mean averages and to quantify the relationships between the differences, with a 95% interval of confidence. The Bruner and Langer non-parametric was applied to the mean values for the crestal and subcrestal implants. All of the histomorphometric parameters were analyzed using descriptive methods (SPSS 19.0, SPSS, Chicago, IL, USA). For all of the tests performed, the significance level chosen was 5% (p < 0.05).

Results
The operative surgical sites healed without incident. All of the implants were available for histological analysis.
The mean insertion torque for the implants was 40.21 ± 0.87 N-cm in P2, 42.87 ± 0.11 N-cm in P3, and 44.68 ± 0.17 N-cm in P4. Using a paired two-sample t-test, a significant difference between the average insertion torques was found (p = 0.005) ( Table 1). The mean Implant Stability Quotient (ISQ) values were above 70 ISQ, which indicates a high primary stability, and they were increasing from day 0 to day 90. We can see, in Tables 2 and 3, the ISQ values for wide ring implants and narrow ring implants. The mean bone loss for the narrow ring implants is 0.75 ± 0.22 at 60 days and 0.89 ± 0.18 at 90 days in P2, 0.78 ± 0.19 at 60 days and 0.86 ± 0.59 at 90 days in P3, and 0.71 ± 0.11 at 60 days and 0.75 ± 0.11 at 90 days in P4, which indicates more bone loss at 90 days that at 60 days (Table 4). The mean bone loss for the wide ring implants is 0.82 ± 0.11 at 60 days and 0.97 ± 0.91 at 90 days in P2, 0.80 ± 0.56 at 60 days and 0.89 ± 0.23 at 90 days in P3, and 0.79 ± 0.25 at 60 days and 0.79 ± 0.67 at 90 days in P4, which indicates more bone loss at 90 days that at 60 days (Table 5). In the microscopic analysis of the crestal bone remodeling, the distance from the implant shoulder to the first bone-to-implant contact was higher for the implants installed with a small ring in the buccal aspect with statistical significance (p < 0.001). For the peri-implant tissues thickness, the distance from the implant shoulder to the external portion of the epithelium showed no differences and no statistical significance was found in both types of implants.

Discussion
Short (length ≤ 8 mm) implants offer a minimally invasive alternative in the rehabilitation of atrophied alveolar bone [5].
Short implants present a similar success rate compared to conventional ones [14,[29][30][31]. Those implants depend specifically on the cortical bone anchorage, because they are mainly used in highly resorbed areas, where the amount of cortical bone remains stable in comparison to the trabecular bone [32]. The main drawbacks of short implants are, on one hand, the lack of primary stability due to their small size [16], and the unfavorable crown-to-implant ratio [33,34]; therefore, adding elements to maximize the contact area and the mechanical retention in the dense cortical bone can be beneficial. In this experimental study in dogs, we tested a new short implant design in which a ring is added to the implant cervical area to improve the support and primary stability at the cortical bone level, in a similar way to the extraoral implants [35]. The addition of the ring would also prevent the implant from being inserted deeper than planned, which is very important when working next to delicate anatomical structures, such as the inferior alveolar nerve. The top of the ring is polished and the bottom has a rough surface, so it can become osseointegrated. To achieve a homogeneous seating of the ring on the bone crest, we used a round flattening reamer to achieve a flat surface, where the ring can rest homogenously.
Although a cervical ring can have some advantages from a mechanical point of view, it is important to test the biological behavior of this element, because the osseointegration of the bottom surface of the ring can increase the BIC area of the implant and improve the load transmission, but if the bone does not adhere to the rough bottom surface of the ring, the marginal bone loss will be increased and a higher incidence of peri-implantitis can be expected. No previous studies on the addition of such a ring on the osseointegration of this device have been published so far. There are very few animal studies on short implants [36,37], and they are in mandible and not in maxilla like this study. In 2016, our group published a pilot study with 60 extra-short 4 mm implants in a posterior mandible splinted with 10 mm length implants, with a 100% success rate at the one year follow up [38]. All of the implants of this study were correctly integrated, which is in line with the studies in humans that have a high success rate [39]. The perfect flattening of the bone crest is technically difficult and if the ring and the osteotomy are not perfectly aligned, the implant stops at the first bone contact. This fact explains why, when measuring the total values of the marginal bone loss, some higher values can appear. This would explain why the data have a lot of rank, and in the same implant there are areas with much more bone loss. If the measurements are made from the first bone implant, the contact the results will show different values. The latter is an important finding, because adding a circular element to the cervical area of a tissue level implant with a 2.0 mm neck is going to maintain the bone and can therefore provide a clinical benefit of a more primary and greater stability surface area of the implant in contact with the bone. More studies are needed with a smaller diameter ring that is more adapted to the animal´s jaw of the experiment, as well as the modification of the technique of insertion to be able to validate this assertion. Another issue is the long-term stability of the marginal bone in the ring area and the bone's reaction to loading. Within the limitations of this study, the crestal bone resorption was reduced in the narrow extra-short ring implant designs, compared with wide ring implants in the healed maxilla. These data could be an important factor for humans, because the use of short implants with rings in soft and resorbed bone can be used with a high predictability, but can be managed with skillful technique.
More long term studies with loading protocols and different ring sizes must be performed.

Conclusions
Our findings suggests that the amount of peri-implant tissues (crestal bone loss) after remodeling over a period of 12 weeks was smaller in the narrow ring extra short implant installed in the healed maxilla, compared with the wide ring extra-short implants.