Factors Inﬂuencing Marginal Bone Loss around Dental Implants: A Narrative Review

: Implant supported dental prostheses are increasingly used in dental practice. The aim of this narrative review is to present the inﬂuence of transmucosal surface of prosthetic abutment and implant on peri-implant tissue. The article describes causes of bone loss around the dental implant. Moreover, properties of different materials are compared and discussed. The advantages, disadvantages, and biomechanical concept of different implant-abutment connections are presented. The location of connections in relation to the bone level and the inﬂuence of microgap between the abutment and implant are described. Additionally, the implant abutments for cemented and screwed prosthetic restorations are compared. The inﬂuence of implant and abutment surface at the transmucosal level on peri-implant soft tissue is discussed. Finally, the biological aspect of abutment-implant connection is analyzed.


Introduction
Dental implants are widely used for oral prosthetic rehabilitation in case of partially (single or more missing teeth), as well as fully edentulous patients. It was demonstrated that osseointegrated implants have a high survival rate (cumulative mean of 94.6%, SD 5.97%) with a follow-up period of up to 20 years [1]. Among the absolute contraindications for dental implants are poor oral hygiene, drug abuse, psychiatric illness, and patients' unrealistic expectations. Whereas, circulatory system diseases, diabetes, xerostomia, endocrine, and metabolic disorders (with an adequate treatment) are generally considered as relative contraindications. In addition, relative contraindications include aged patients, as well as patients with a low quality and density of bone, with bruxism, periodontal diseases, oral carcinomas, and generally immunocompromised patients [2,3]. An individual approach to the patient allows considering the negative influence of systemic diseases on the dental implant treatment outcome [4]. The overall implant loss limited by implant region and bone quality varied from 0.3 to 3.3% [5]. In medically compromised patients, the implant failure was 0.0-22.5% [6][7][8]. Most of the implant failure was observed in patients with smoking history (37%), hypertension (20.8%), and diabetes (20.3%) [9].

Causes of Bone Loss around the Dental Implant
Factors that have an impact on bone loss around implants can be divided into local, systemic, and social. The local factors include the implant body, occlusal loading, size of implant, and biological aspects. Structure-related factors of bone loss involve the type of connection between the implant and abutment (internal hex, external hex, conical, and their modifications), as well as the size of a microgap between the implant and abutment. Moreover, the type of an implant (one-piece, two-piece, and multi-part implant), its shape

Types of Implants
The implant systems can be categorized according to the number of parts of mechanical design to one-piece and two-piece implants. The one-piece and two-piece implants can be used in one-stage, as well as in two-stage treatment procedure [40]. In a one-piece implant, the endosseous and the abutment part are one unit, with prosthetic restoration placed on top of the implant. Whereas, two-piece dental implants consist of a part placed in the bone (implant) and a separate supragingival part (prosthetic abutment). Abutment (in two-piece implants) can comprise several parts (e.g., multi-unit constructions) used depending on the clinical situation. Such multi-unit abutments are indicated when the angulation correction of inadequately/disadvantageously positioned implants is needed, e.g., in case of implant-borne multi-unit or full-arch prosthetic restorations. Then, several multi-unit abutments are used to maintain the aesthetics and emergence profile in compromised cases of edentulous spaces [41].
One-piece implants are placed in one-stage surgery with immediate implant loading. The over 2-mm bone loss was observed in 6% of cases of one-piece implants, while in 16% of cases of two-pieces implants in 1-year follow-up [42]. Such advantageous results can be attributed to the absence of a microgap between the abutment and implant in one-piece implants [40]. On the contrary, over 2-mm bone loss was reported in 49% of one-piece implants in contrast to 7.7% of two-piece implants [43]. However, the higher apical migration of soft tissues was observed for one-piece implants. The apical migration of soft tissues can adversely affect the contour and the aesthetics of soft tissues around the implant due to the lack of suitable shape and surface of a prosthetic abutment [44].

Location of Implant-Abutment Connection in Relation to the Bone Level
There are several implant systems categorized according to their location in relation to the bone level and the construction of the implants.

Subcrestal Implants
Implants placed below the edge of the bone are subcrestal implants. It is worth mentioning that a comparable number of inflammatory cells around the subcrestal and crestal implants was found [45]. The implant abutment connection promotes bacterial colonization, which can result in peri-implantitis [45]. Therefore, it is very important for implants at the bone or subcrestal level to transfer the prosthetic substructure as far as possible on the free gingival margin by choosing a longer prosthetic abutment. In addition, it allows for better aesthetics of the restoration. Moreover, the subcrestal position of implants diminishes the probability of exposure of implants and reduces inflammations around implants [45].

Bone-Level Implants
Another type are implants placed at the bone level. In case of bone-level implants, the transmucosal component is a separate part of the implant and is located on the buccal side of the bone. Bone-level implants are placed in such a way that the prosthetic platform of the implant is at the level of the bone. This allows for a very good emergence profile and possible soft tissue corrections. It has been proven that the smallest bone loss occurs around implants, where the microgap between the abutment and implant is placed 1-2 mm above the bone [46]. However, in this case, the rough part of the implant, which should adhere to the hard tissues, may come into contact with the soft tissues and cause inflammation. The risk arises due to the positioning of the implant above the bone. Therefore, bone-level implants without the concept of platform-switching should have approximately 1 mm of polished cervical surface, which allows for the adhesion of the soft tissue [46].

Tissue-Level Implants
In addition, there are many implant systems that are embedded at the level of soft tissues. Their growing popularity can be attributed to the fact that placing the implant at the soft tissues level causes just a slight bone loss and allows for maintaining good hygiene of the prosthetic restoration. In the tissue-level systems, the transmucosal component is the main part of the implant. It is important to position this part above the bone level since by contact with the polished surface, it causes bone loss. The biological width will thus be facing this transmucosal neck. It is supposed to be as biocompatible as the implant, which cannot become loose or be removed, and which harbors no microgap. Moreover, a lower number of bacteria is found for tissue-level or one-piece implants since there is no contaminated connection at the crestal bone level. Bacteria can have an adverse effect on the biological width or contribute to the formation of peri-implantitis and the implant failure [47]. However, there were no differences in the size of the epithelial attachment with implants at the bone and soft tissue level [48,49].

Platform-Switching Concept
Platform-switching is a concept of placing a prosthetic abutment of a diameter smaller than that of an implant [50]. It results in smaller bone loss around the implant neck and favourable stress distribution [51][52][53]. Platform-switching allows the correct adhesion of marginal epithelium to an implant surface and the proper profile of soft tissues [54]. It can be applied in either the external or internal implant-abutment connection [53]. This concept applied in bone level implants causes a lower decrease of epithelial attachment in comparison to other abutments [55]. In the supracrestal implants with platform-switching, bone loss was slightly less pronounced than in the case of subcrestal implants, yet the statistical significance was not observed. Still, it is recommended to position implants on the crestal level or 1-2 mm subcrestally [56].

Platform-Switching Concept
Platform-switching is a concept of placing a prosthetic abutment of a diameter smaller than that of an implant [50]. It results in smaller bone loss around the implant neck and favourable stress distribution [51][52][53]. Platform-switching allows the correct adhesion of marginal epithelium to an implant surface and the proper profile of soft tissues [54]. It can be applied in either the external or internal implant-abutment connection [53]. This concept applied in bone level implants causes a lower decrease of epithelial attachment in comparison to other abutments [55]. In the supracrestal implants with platform-switching, bone loss was slightly less pronounced than in the case of subcrestal implants, yet the statistical significance was not observed. Still, it is recommended to position implants on the crestal level or 1-2 mm subcrestally [56].

External Connection
External connection, where the abutment forms the external edge of an implant, was introduced by Branemark [58]. Its distinctive feature is the shape of a hexagon or octagon (Infinity Implants, Neodent, Dentoflex, Titanium Fix, SIN, Conexao). The abutmentimplant junction with an external connection is localized close to the crestal bone. Higher bone loss and increased papillary bleeding index (PBI) were observed within the external connection compared with the internal connection [59].
The external connection is most commonly made of an hex, in which the prosthetic part is located towards the outside margin of an implant construction. This causes bone loss around the implant [60]. Many modifications of an external connection with or without hex were introduced over the years. In external hex, with time, the construction loosening caused by the weakened quality of connection may occur [57]. External connections were modified to decrease the possibility of rotating the abutment, to reduce the screw loosening (during implant load), to simplify the prosthetic procedure, and to increase the abutment alignment in various vertical positions towards the implant [61].

External Connection
External connection, where the abutment forms the external edge of an implant, was introduced by Branemark [58]. Its distinctive feature is the shape of a hexagon or octagon (Infinity Implants, Neodent, Dentoflex, Titanium Fix, SIN, Conexao). The abutment-implant junction with an external connection is localized close to the crestal bone. Higher bone loss and increased papillary bleeding index (PBI) were observed within the external connection compared with the internal connection [59].
The external connection is most commonly made of an hex, in which the prosthetic part is located towards the outside margin of an implant construction. This causes bone loss around the implant [60]. Many modifications of an external connection with or without hex were introduced over the years. In external hex, with time, the construction loosening caused by the weakened quality of connection may occur [57]. External connections were modified to decrease the possibility of rotating the abutment, to reduce the screw loosening (during implant load), to simplify the prosthetic procedure, and to increase the abutment alignment in various vertical positions towards the implant [61].

Internal Connection
However, to reduce the bone loss, increase the implant stability, and enhance its durability and maintenance in one position, the internal connection was developed. The main characteristic of the internal connection is a prosthetic part placed inside the implant and surrounded with the implant surface below the level of implant platform [62]. The internal hex and internal octagon differ in the design of their geometric features. The internal connection improves the distribution and dissipation of forces, which results in better stability. The larger the platform of the implant, the higher the fatigue load limit [63]. In the internal connection, the vertical height from the implant platform to the top of the abutment is reduced [64]. The screw tightening that attaches the prosthetic abutment to an implant, occurs with elongation and blocking the abutment construction, which increases its stability. The distribution of lateral forces is transferred deep within the implants. The internal connection is more resistant to screw damage and to wear in comparison with the external connection [65].

Conical Connection
A conical connection is currently considered to be one of the best. Its main advantage is the load transfer along two conical constructions, that causes limited abutment loading [66,67]. This design provides separation of the interior of an implant from the surrounding tissues, limiting microleakage. The most frequently used type of a conical connection is Morse taper, which ensures integrity between the implant and abutment, with the smallest micromovements observed at a narrow (2 • -4 • ) inclination angle of conical prosthetic abutment [68,69]. Thanks to that, the risk of rapid loosening of the prosthetic construction significantly decreases [70].

Tube-in-Tube Connection
A tube-in-tube connection is made of two cylinders that allow good load distribution to be transmitted to the implant. If the length of the tube-in-tube connection between the abutment and implant in comparison to its diameter is bigger, it gains higher durability in the long-term of loading the construction of an implant [71,72].

Microgap between the Abutment and Implant
The microgap enables bacteria to colonize the connection between an implant and abutment [59]. The aim is to obtain the smallest possible microgap between implantabutment parts. With a larger microgap the risk of bacterial leakage and rising, so-called the pumping effect, significantly increases. The pumping effect is a phenomenon of bacterial movement inside and outside the connection under the influence of pressure changes. This leads to crestal bone loss around the implants due to bacterial contamination, which incubates in intercellular fluid [73]. The exact fit of the prosthetic abutment to the implant prevents plaque accumulation that can possibly cause inflammation and subsequent soft tissue loss around the implant [74].
Studies show that the implant-abutment microgap is measured between the free edge of the implant and the abutment in a horizontal and vertical position [75]. The presence of microgap is measured with an optical and scanning electron microscope, as well as with microleakage studies [76]. The vertical microgap may vary from less than 10 µm up to even 100 µm. In some studies, microgaps vary from 2.3 to 5.6 µm. In another study, this distance varies from 22.6 to 62.2 µm for vertical and from 27.1 to 16.0 µm for horizontal microgap [64]. It is very difficult to evaluate the average, optimal value of the microgap measurements due to the lack of standardized studies [75].
The cement-retained abutments produce a far smaller microgap in comparison to screw-retained ones. The smallest size of microgap can be observed with the internal conical connection [77,78].
The position of the microgap in relation to the level of the bone crest also plays a role. It was found that moving the microgap coronally away from the bone crest resulted in less bone loss around the implants [79].
The study showed that the microgap magnitude, tested in vitro, may vary depending on the material of the abutment and the abutment manufacturing technique [80]. The smallest microgap was achieved with CAD/CAM fabricated zirconia abutments, as well as the milled Cr-Co abutments.

Abutments for Cemented Implant-Supported Restorations
Prosthetic restorations are located on abutments in two ways: By cementing or screwing. Cemented prosthetic restorations are often cheaper than screw-retained restorations [81]. The cementing procedure is required for one-piece implants due to the inability to perform screwed restorations. With a precisely carried out cementing procedure, there is no microgap, in which bacteria can multiply. In addition, there are no complications related to chipping of the ceramics at the fixing screw hole, through which the prosthetic restoration is screwed. On the other hand, in case of cemented restorations it is very important to thoroughly remove the excess cement, which may cause inflammation around the implant. Along with the depth of the restoration up to 3 mm of subgingival, the amount of unremoved excess cement increases. Additionally, the removal of the excess may lead to damage to the connector's structure [82]. There is also a problem with the repair of the prosthetic restoration located by cementation, due to its destruction risk after removal from the prosthetic abutment [83]. All in all, cemented crowns/bridges are not commonly used in modern implant-borne prosthetics due to the inability to obtain full dryness of the operating field and maintaining a sterile implant platform during the cementation procedure [81]. As a result, bacteria that are in the microgap between the prosthetic abutment and the implant can quickly penetrate beyond the connection and consequently cause inflammation. Such bacterial colonization can occur during implantation, exposing the implant, as well as during screwing or cementing the restoration [84].

Abutments for Screwed Implant-Supported Restorations
Nonetheless, prosthetic restoration on screwed abutments are most often used due to the easier repair and periodic inspection, better hygiene of the restorations, and fewer complications related to the tissues around the implant. Reduced resistance is on the screw with which the prosthetic abutment is screwed to the implant. There is also no cement deposit that can compensate for deformations caused by mismatches in the structure. There can be too much stress on the bone and the implant if the restoration is imprecise [85]. A significantly better fit of the abutment to the implant was observed with machined abutments versus cast abutments. Additionally, the screw is loosened more often and the mobility of the structure is increased [86]. The disadvantages of screw-retained restorations include frequent chipping of the crown ceramics at the fixing screw hole. Taking into account the mechanical properties, a greater susceptibility to cracking of the zirconium structure was observed due to its hardness [87].
It is also possible to combine the cementing and screwing technique, in so-called hybrid restoration. It consists of cementing the restoration on a titanium base and then screwing it intraorally to the implant. As a result, the excess cement can be accurately removed [88]. This solution prevents peri-implantitis, which may constitute 33-100% of failures, when the excessive cement is left behind around the implant abutment and prevents corrosion and galvanism using prefabricated abutments adjacent to the soft tissues [89]. Screw-retained restorations revealed a 16.9% lower level of inflammation in comparison to conventional cemented restorations [90]. Moreover, the success rate of the combined type of restorations were up to 92.4% [91]. Hybrid restoration reduces the risk of damage to the interface between the abutment and implant due to the contact between titanium and titanium. There is no such relationship with a prosthetic abutment made of one part.

Implant and Abutment Surface at the Transmucosal Level-Their Influence on Peri-Implant Soft Tissue
The attachment between the mucosa and the dental implants abutment surface depends on the implant chemical composition, surface free-energy, and topography. The surface of implants may have variable topography, that can be modified, i.e., using machining, etching (sulphuric acid, hydrofluoric acid, hydrochloric acid), oxidizing (anodizing, which modifies the microstructure of the surface increasing its roughness), sandblasting (titanium dioxide, aluminium oxide, zirconium dioxide, silicon carbide), and laser [92]. To determine dental implant surface topography, there are several surface roughness parameters which have to be examined. The most commonly evaluated parameters in the studies are Ra (arithmetic medium value of the deviations in relation to the medium line) and Rz (peak-to-valley roughness) parameters [93][94][95]. The classification of implant neck surface roughness based on the value of Ra parameter regards surface as: Smooth when Ra is less than 0.5 µm, minimally rough with Ra = 0.5-1.0 µm, moderately rough with Ra = 1.0-2.0 µm, and highly rough with Ra more than 2.0 µm [96][97][98]. Other studies use the Sa parameter (arithmetic mean of absolute value of the height) to assess the dental implant surface roughness [99][100][101] and presented a general classification of abutmentsurface-roughness (ASR) based on the Sa value. The classification describes the surface as too smooth at Sa < 0.10 µm, optimal when Sa is within the range of 0.15-0.25 µm, and too rough when Sa > 0.35 µm. The values of Sa parameter within ranges either 0.10-0.15 µm or 0.25-0.35 µm depict the surface of tolerable roughness, describing the surface as not too smooth and not too rough, respectively [101].
In general, it was observed that fibroblasts and epithelial cell proliferation on zirconium are higher than on titanium surface. As for titanium alloy surfaces, fibroblasts adhere better to smooth (polished with silicon carbide paper discs or textile discs together with diamond suspension and silicon suspension) rather than to rough surfaces (prepared airborne-particle abrasion with 110 µm alumina particles). Whereas, for zirconium surfaces, the structured surface enhances fibroblast proliferation. For epithelial cells, smooth surfaces (polished or machined) allow for an increase in the proliferation rates [93]. The adherence of soft tissue to the implants is said to ensure a lower number of bacteria and absence of peri-implantitis, which prevents bone loss around implants [44]. Several studies have shown a very good connective tissue alignment to implant surfaces at the transmucosal level that were mechanically machined [102][103][104]. Machined titanium dental implants obtained the Ra value of 0.65 ± 0.11 µm and Rz of 6.09 ± 0.37 µm [94]. The microgrooves created during the process resulted in smaller apical soft tissues migration. On such surface, a directed growth of fibroblasts was observed together with a long connective tissue alignment and small bone loss [105]. Similar properties as on the machined one were noticed on the sandblasted surface, where collagen fibers were parallel to the surface [106]. The sandblasted titanium dental implants surface revealed the Ra value of 0.75 ± 0.05 µm and Rz = 5.55 ± 0.21 µm [94]. Comparing the neck design of the implant, it has been proven that the smaller depth on probing and bone loss occurred with machining implants with a wide rough neck compared to the rough reduced neck of the implant [107].
Some studies demonstrated the microgap between the epithelial attachment and a transgingival part of the implant with a machined and etched surface. Hydrochloric acid (HCL), sulphuric acid (H 2 SO 4 ), and hydrofluoric acid (HF) are used to etch the implant surface [108]. The acid-etched (with hydrochloric or sulfuric acid solution) titanium dental implants surface exhibit the Ra value of 0.51 ± 0.10 µm and Rz of 5.09 ± 0.46 µm. Whereas, the anodized surface exhibit Ra = 0.87 ± 0.14 µm and Rz = 5.14 ± 0.69 µm [94]. The oxidized and etched surfaces of implants, at the site of soft tissue attachment, showed a smaller epithelial growth and longer connective tissue compared with machining surfaces [109]. The chemically modified etching process allows a thin layer of hydrocarbons and carbonates to establish on the surface. Thanks to that, the received hydroxylated surface initially has a contact angle close to 0, which guarantees very good attachment [110]. In contrast, the oxidized surface allows soft tissue proliferation on an implant surface. The anodic oxidation coating can improve resistance to attrition and corrosion and stability of titanium surface colour. Additionally, pink-shaded anodized titanium enhances the epithelial tissue attachment to the prosthetic abutment surface. However, after 24 h, a lower cellular adhesion to the anodized surface was observed compared with the machining surface [44]. Early bone loss can be affected by producing the polished implant abutment surface, which was replaced in many implant systems with the machining one.
Moreover, the connective tissue has shown the enhanced attachment to laser-microtex tured surface of implants and abutments. Namely, clinically less bone loss has been found when laser-microtextured collar implants were used, in comparison with the smooth surfaces, regardless of the loading protocol [111]. The epithelial tissue exhibits better alignment ability to rough than to smooth surfaces. Regrettably, it does not guarantee perfect soft tissue adhesion to the transgingival part of implant or exclude peri-implantitis. Proper (comprehensive) preparation of the surface ensures adhesion of soft tissues and fibroblasts proliferation, but also osseointegration. Simultaneously, it prevents excessive plaque deposition [112]. On the other hand, the rough surface does not allow for the proper epithelial attachment and increases plaque accumulation. Excessive plaque accumulation results in inflammation of soft tissue surrounding the implant. That is why selecting the appropriate roughness is very important especially for the transmucosal part of the implant-abutment connection. As mentioned before, the average roughness (Sa value) should vary from 0.15-0.25 µm [101]. Below this value, the stable soft tissue attachment will not be maintained. Furthermore, the higher the value of average roughness, the more microbial contamination may occur [101]. In contrast to the soft tissue connection, a higher roughness value determines a larger bone formation. When the roughness increases, the removal torque of implants increases up to the critical value [94,98]. Consequently, low preservation of the marginal bone level with the polished neck of an implant has been proven [113,114].

Biological Width
The biological width occurs around implants and natural teeth. It consists of a marginal epithelium surrounding an implant, called epithelial attachment, and connective tissue attachment, in which collagen fibers adhering to the implant surface are situated [104]. It is thought that the epithelium lining the sulcus around the implant has structural and functional features similar to gingival tissue. Soft tissues attach with hemidesmosomes around the implant, however they create much weaker connections than with natural teeth [115]. The keratinized mucosa around the implant has a positive impact on the maintenance of hard and soft tissues, as it prevents bone loss. Initially, bone resorption begins from its cortex. Bone changes can be caused by an excess of cytokines that stimulate osteoblasts or inhibit osteoclasts that affect resorption and apposition of the bone [116]. Moreover, neutrophils, plasma cells, and macrophages can be found around dental implants with bone loss [117]. The cells have a fibroblastic phenotype and lead to the reduction in bone matrix around implants [118]. On the other hand, the increased number of osteocytes was found around immediately loaded implants and it is important in maintaining the bone matrix. In contrast to natural tooth, around implants, transverse collagen fibers are replaced with circumferential fibers surrounding the implants. Circumferential fibers can be found on the first thread of implant. Moreover, by immediately loaded implants more circumferential fibers were found, which are most resistant to compressive forces resulting in high bone-to-implant contact [119,120]. In platform-switched implants the biological width is located more coronally compared with implants with other types of connections [121]. Granulation tissue, which occurs in transgingival connection of soft tissues with the implant, prevents apical connective tissue movement [122].

Emergence Profile
The emergence profile of implants is an important issue to facilitate favourable aesthetic outcome and maintain peri-implant health. The over-contoured prosthetic abutments causes an initial pressure on interproximal papilla, which leads to ischemia and, in consequence, to bone loss and apical migration of soft tissues [115]. However, a too small abutment in width may lead to an increased plaque and bacteria accumulation [123]. It has been proven, that there is an increased bone loss with convex abutment profile in comparison with the concave profile [106]. The frequent removal of the abutment causes apical migration of soft tissues [124]. Mechanical trauma to soft tissue leads to subsequent bone loss and low soft tissue profile [125]. The platform-switching type of implant-abutment connection influences the emergence profile. Thanks to this concept, the excessive pressure of the abutment on soft tissues can be eliminated. Moreover, it allows for maintaining proper hygiene around implant-supported restoration. A lower number of bacteria has been found at the implant-abutment interface when platform-switching was used, ensuring healthy soft tissues with a proper emergence profile around implants [126,127].

Summary
On the market, there are more and more manufacturers offering advanced solutions, which allow decreasing the number of failures and providing longer lasting prosthetic implant-supported restorations. With adequate procedures and correct selection of the system and tools, bone loss can be reduced and soft tissues without pathological changes can be retained. Finding the balance between function and aesthetic in implant rehabilitation is crucial.