For Figures, Tables and References We Refer the Reader to the Original Paper

Critical horizontal dimensions of interproximal and buccal bone around implants for optimal aesthetic outcomes: a systematic review

For figures, tables and references we refer the reader to the original paper.

Since it was first introduced, implant therapy has witnessed a substantial evolution. At the time that the osseointegration concept was pioneered by Branemark and many others, the focus was primarily directed towards a functional rehabilitation (Branemark et al. 1977). A wide array of guidelines, treatment protocols and physical and chemical modifications of implants have resulted in implant therapy obtaining a high predictability from a functional point of view (Adell et al. 1990; Buser et al. 1997; Lekholm et al. 1999). Survival rates over 95% in non-compromised patient's are now frequently reported in the literature (den Hartog et al. 2008; Tomasi et al. 2008). Although implant survival, or the establishment and maintenance of a stable osseointegration over time, is a prerequisite for functional rehabilitation, it does not take into consideration collateral factors such as aesthetic outcome and patient satisfaction (Glauser et al. 2006). Consequently, implant success rates that incorporate aesthetic factors can be considerably lower than implant survival rates (den Hartog et al. 2008).

Today, in the anterior, aesthetic zone, patients consider the aesthetic outcome as an essential factor often priming over the functional factor of a dental implant. The aesthetic outcome might even influence the patients self-esteem (Burkhardt et al. 2000). Similar to teeth, soft-tissue topography and stability around implant-supported crowns are important to achieve optimal aesthetics (Belser et al. 1998). Preferably, the appearance of the peri-implant soft tissues should be in harmony with the soft tissues around the adjacent teeth, and the implant crown should be in balance with the neighbouring dentition (Meijer et al. 2005). Recently, den Hartog et al. (2008) summarized various implant treatment strategies for the accomplishment of optimal aesthetics. These include approaches to rehabilitate the underlying bone structures by augmentation procedures (Weber et al. 1997; Jensen et al. 2006; Pelo et al. 2007), techniques to manipulate and enhance the architecture of the peri-implant soft tissues (Zetu & Wang 2005; Esposito et al. 2007) and methods for alveolar ridge preservation following tooth extraction (Lekovic et al. 1997; IrinakisTabesh 2007). Furthermore, implants and abutments with specific configurations have been introduced to sustain hard and soft tissues (Wohrle 2003; Morton et al. 2004; Lazzara & Porter 2006; Maeda et al. 2007; Noelken et al. 2007), together with provisionalization techniques to restore the soft-tissue contour (Jemt 1999; Al-HarbiEdgin 2007), and the introduction of ceramic-customized abutments and ceramic implant crowns (Canullo 2007; Schneider 2008).

Nevertheless, the peri-implant bone dimensions constitute the base for the supra-crestal soft tissues (Burkhardt et al. 2008). Because the latter determines the aesthetic result and has relatively constant dimensions (Linkevicius & Apse 2008), the bone dimension is one of the key factors in determining the soft-tissue contour.

Achieving aesthetic success is suggested to be dependent on an ideal three-dimensional implant position within the available bone dimensions (Buser et al. 2004) and the maintenance of adequate buccal bone over the buccal implant surface (Grunder et al. 2005). As a general principle, endosseous implants should be installed within the alveolar envelope at implant placement (Qahash et al. 2008). Despite dimensional changes due to inflammatory processes or biomechanical factors, basically three bone remodelling processes determine the bone dimensional changes at implant recipient sites under normal physiological conditions. Briefly, these are remodelling processes that occur after tooth extraction (Johnson 1969), due to surgical trauma (Gargiulo et al. 1961) and due to saucerization (Adell et al. 1981).

Implant positioning in relation to the bucco-oral and mesio-distal dimensions of the alveolar ridge is a factor thought to influence the degree of bone remodelling following implant placement (Esposito et al. 1993). Such bone remodelling may in turn negatively influence the soft-tissue topography and the aesthetic outcome of the implant therapy (Cardaropoli et al. 2006).

The respect for these biological concepts has led to several clinical guidelines regarding the correct implant positioning in relation to bucco-oral and mesio-distal bone dimensions (Buser et al. 2004; Grunder et al. 2005). For the bucco-oral dimension, it has been suggested that the buccal bone thickness should be at least 2 mm and preferably 4 mm. In relation to mesio-distal dimensions, it has been suggested that the distance between a tooth and an implant should not be <1.5 mm, and between two implants, not <3 mm. In order to fulfil these criteria, bone augmentation procedures, orthodontics, enameloplasty or restorative materials are often recommended (Buser et al. 2004). Although these criteria might be correctly mentally deducted from the biological concepts of bone remodelling, the impact at the patient level is not to be underestimated. Therefore, clinical evidence is needed to endorse the above-mentioned impactful guidelines. This review was initiated to evaluate the critical horizontal interproximal and buccal bone dimensions around implants, to ensure an optimal aesthetic outcome.

Material and methods

Focused question

Do the horizontal interproximal and buccal bone dimensions at the time of implant placement influence the aesthetic outcome of the treatment in the partially edentulous patient?

Search strategy

A Medline (PubMed) search was performed to identify all articles investigating the addressed question. The search was restricted till 1 January 2009. A similar search was conducted on the Cochrane database.

Additional hand searches were performed and included: (1) bibliographies of previous reviews on the topic of implant aesthetics (Belser et al. 2004; Buser et al. 2004; Zetu & Wang 2005; den Hartog et al. 2008; De Rouck et al. 2008), (2) bibliographies of all publications considered in this review, and (3) cited reference searches of all publications considered in this review using the ISI Web of KnowledgeSM online database.

Search terms

The following MeSH terms and key words were used: ‘implant’ AND ‘buccal bone’, ‘implant’ AND ‘bucco lingual’, ‘implant’ AND ‘bucco oral’, ‘implant’ AND ‘facial bone wall thickness’, ‘tooth implant distance’, ‘inter-implant distance’, and ‘distance between implants’.

Inclusion criteria

A priori, this review was restricted to peer-reviewed publications dealing with endosseous, solid titanium implants in English language. Only data from human studies were evaluated. If articles reported on case series, at least five consecutive cases had to be enrolled. Regarding the study design, clinical trials including randomized trials, controlled trials, prospective and retrospective cohort studies and prospective and retrospective case series were considered. The studies had to reflect on the aesthetic outcome in relation to the buccal and interproximal bone dimensions around implants. Buccal bone dimensions were interpreted as the thickness of the buccal bone after osteotomy preparation. Interproximal bone dimensions were interpreted as tooth-to-implant distance or as interimplant distance. Vertical bone dimensions were not directly considered in this review. The aesthetics had to be judged by dental professionals and not by questionnaires or patient interviews (Chang et al. 1999a, 1999b; Gotfredsen 2004).

Exclusion criteria

Publications not meeting the inclusion criteria were excluded from the review. Studies on temporary implants (mini-implants), on implants subjected to guided tissue regeneration and on retrieved implants from histological evaluations were also excluded.

Data extraction

Article selection was decided upon screening of the titles and the abstracts by two independent reviewers (J.M. and W.T.). In case of disagreement between the reviewers, an inclusion/exclusion decision was made by discussion after screening the full-text article. Data were extracted simultaneously by the two reviewers and recorded in a data extraction sheet. The heterogeneity of the studies and outcome variables rendered a meta-analysis impossible.

Results and discussion

Buccal bone dimensions and aesthetic outcome

Several clinical guidelines have been proposed in relation to buccal bone dimensions and aesthetic outcome. Depending on the reference, it has been suggested that it is crucial to have a buccal bone plate of at least 1 mm (Belser et al. 2008) or 2 mm (Grunder et al. 2005; Buser et al. 2007). This buccal bone thickness was advocated to ensure proper soft-tissue support, avoid resorption of the facial bone wall following restoration and by this minimize the risk for peri-implant soft-tissue recessions. The latter is an important factor when it comes to aesthetics.

Using the outlined search strategy and criteria, 358 abstracts were found. After title and abstract reading, 23 articles were considered for full-text analysis. The reference list and cited reference search of these 23 articles resulted in an additional seven articles that were subjected to full-text analysis. Of these 30 papers selected for full-text analysis, eight were excluded because they were animal studies (Carmagnola et al. 1999; Polimeni et al. 2004; Araujo et al. 2005, 2006a, 2006b; Wikesjo et al. 2006; Qahash et al. 2008; Schwarz et al. 2008). Of the remaining 22 articles, 14 were excluded because they did not provide information regarding buccal bone thickness (Nir-Hadar et al. 1998; Grunder 2000; Small & Tarnow 2000; Covani et al. 2003, 2007; Kan et al. 2003; Henriksson & Jemt 2004; Cornelini et al. 2005; Tsirlis 2005; Bischof et al. 2006; CanulloRasperini 2007; KoutouzisWennstrom 2007; Sammartino et al. 2007; Evans & Chen 2008) and 1 study was excluded because the implants were placed in conjunction with guided tissue regeneration (Evans & Chen 2008). Three papers appeared to be review articles (London 2001; Chen et al. 2004; Grunder et al. 2005), and the remaining four papers did not provide direct information on the question of whether buccal bone thickness influences the final aesthetic outcome (Spray et al. 2000; Botticelli et al. 2004; Covani et al. 2004; Cardaropoli et al. 2006).

Therefore, the original question had to be rephrased in order to compensate for this lack of clinical studies and to provide at least some meaningful clinical information. Because supra-crestal soft-tissues around implants seem to have relatively constant dimensions [the biological width (Linkevicius & Apse 2008)], one could eventually hypothesize that a vertical buccal bone resorptionwill result in a marginal soft-tissue recession. In turn, this will influence the aesthetic outcome negatively. The original question was therefore rationalized to: ‘Does horizontal buccal bone thickness influence vertical buccal bone resorption?’

Using the above-mentioned abstracts and full-text articles, four papers (Spray et al. 2000; Botticelli et al. 2004; Covani et al. 2004; Cardaropoli et al. 2006) were withheld because they provided information regarding buccal bone thickness. The other 26 papers were excluded for the same reasons as mentioned above. Of the four withheld papers, three provided information regarding vertical bone resorption in relation to buccal bone thickness (Spray et al. 2000; Botticelli et al. 2004; Cardaropoli et al. 2006).

From these selected papers, two articles dealt with buccal bone thickness and vertical bone resorption in healed extraction sites (Spray et al. 2000; Cardaropoli et al. 2006), and one paper dealt with immediate implant placement (Botticelli et al. 2004). Although the biological processes involved in bone remodelling after immediate implant placement might be fairly different from those after implant placement in healed bone, the scarcity of data persuaded the authors to incorporate the latter study.

In a prospective, multi-centre study, Spray et al. (2000) measured the buccal bone thickness and height of the buccal bone at the time of implant placement and compared it with the buccal bone height at the time of abutment connection. Data could be obtained from 2667 implants with different surfaces, different designs and original implants as well as replacement implants. There was no discrimination between intra-oral regions (anterior, posterior, maxilla, and mandibula). Following preparation of the osteotomy site, the thickness of the buccal bone wall was measured to the nearest 0.5 mm approximately 0.5 mm below the crest. The distance between the buccal bone crest and the top of the implant was measured with a periodontal probe to the nearest 1 mm. The initial bone level could be superior or inferior to the top of the implants. After implant placement, a submerged healing of at least 3 months was respected. When the healing abutment was placed, the distance from the bone crest to the top of the implant was re-measured. The group of implants that showed no loss of facial bone height had an ‘average’ bone thickness after preparing the osteotomy site of≥1.8 mm, whereas for implant groups showing loss of facial bone height, the average bone thickness was <1.8 mm. The more the loss of bone height, the lower the average thickness of the buccal bone at implant placement. The authors suggested therefore that a buccal bone thickness of around 2 mm would reduce the incidence and amount of vertical bone loss. However, some remarks are necessary. This study focused on vertical bone resorption and not on aesthetics. There are no data that indicates that a resorption of the buccal bone will lead to soft-tissue retraction. Moreover, it needs to be noted that the data were grouped and analysed according to the amount of facial bone loss/gain and not according to the initial facial bone thickness. The authors observed that the initial mean facial bone thickness was smaller for implants that showed more facial bone resorption at the time of abutment placement. With respect to these interesting findings, one needs to consider the large standard deviations for mean facial bone thickness, even for implants that did not show facial bone loss. For the ‘no facial bone loss’ group, with an average buccal bone thickness of 1.8 mm, a standard deviation of 1.1 mm was reported. This means that approximately 95% of the buccal bone thicknesses were located between 0 mm [1.8−(2 × 1.1)] and 4 mm [1.8+(2 × 1.1)]. Therefore, these data do not allow the extrapolation or reverse calculation that no facial bone resorption will occur when the initial buccal bone thickness is >1.8 mm.

Cardaropoli et al. (2006) recorded the alterations in bucco-oral dimensions of 11 single-tooth replacements with implant-supported restorations in the maxillary incisor region. After the initial extraction sockets had healed for a period of at least 6 months, a two-stage implant procedure was used with a 6-month submerged healing period. All implants belonged to the same implant system (Nobel Biocare, Götenborg, Sweden). After preparing the osteotomy, the implants were placed with the top of the cover screw positioned even with the buccal bone crest. This means that the implant shoulder was approximately 1.1 mm below the buccal bony crest. Buccal bone thickness was measured with an especially designed device at the level of the implant shoulder (1.1 mm below the buccal bone crest), and 2 and 4 mm more apically. Bone crest levels were measured with a periodontal probe from the implant shoulder. The measurements were performed at implant placement and at abutment connection. The buccal bone thickness at implant placement was 1.2, 1.3 and 0.9 mm at the three different depth levels. At the two most marginal measurement points, a non-significant mean reduction of 0.4 mm of the buccal bone thickness was observed at the second-stage surgery. At that time, a loss of bone height at the buccal aspect of the implant had taken place, averaging 0.7 mm. A correlation between buccal bone thickness and loss in vertical height was not considered. If reverse calculations based on the Spray and colleagues study are used, the mean vertical facial bone resorption in the study of Cardaropoli and colleagues should have been around 3 mm. However, the latter authors only report an average 0.7 mm vertical bone resorption at the time of abutment placement. This discrepancy might be related to the different implant systems used in both studies. Additionally, in the Spray study, buccal bone thickness data from different intra-oral regions were mixed, whereas in the Cardaropoli study, only implants in the maxillary incisor region were considered. It should also be noted that Cardaropoli and colleagues positioned the implant shoulder approximately 1.1 mm below the buccal bone crest. This information is not provided in the study by Spray and colleagues, or perhaps the large standard deviations in the Spray study explain this discrepancy.

Vertical bone changes in relation to buccal bone thickness after immediate implant placement can be derived from a case series published by Botticelli and colleagues. Twenty-one extraction sockets in 18 healthy patients, located in the anterior region (maxilla and mandible), received a solid screw implant with an SLA-modified surface (Straumann AG, Waldenburg, Switzerland) immediately after extraction. The vertical distance between the implant shoulder and the SLA portion was 2.8 mm in the type of implant used. The implant was placed so that the marginal level of the SLA portion was placed apical of the marginal level of the buccal or the oral wall of the socket. Before implant installation, the thickness of the buccal wall was measured with a caliper, 1 mm below the bone crest. After implant placement, the vertical distance between the implant shoulder and the bone crest and the horizontal distance between the implant surface and the outer side of the bone crest were assessed. All implants experienced a 4-month semi-submerged healing period before healing caps were placed and bone dimensions were reassessed. Because Botticelli and colleagues provided the measurements for each patient in the case series, the following data could be derived. No statistical analysis was performed due to the limited number of patients available for analysis. Implants placed in sites with an initial buccal bone thickness of 1 mm (N=8), 1.5 mm (N=8) or 2 mm (N=5) exhibited a change in vertical bone height of, respectively, 0.1±0.8, −0.6±0.4 and −0.3±0.4 mm. For implants where the horizontal distance between the implant surface and the outer side of the bone crest was ≤2.5 mm (N=5, average 2.4±0.2 mm), 3–3.5 mm (N=7, average 3.3±0.3 mm) or 4 mm (N=8), the corresponding change in vertical bone height was, respectively, 0±0.8, −0.3±0.4 and −0.4±0.8 mm. There was no obvious relationship between buccal bone thickness or the horizontal distance between the implant surface and the outer side of the bone crest at implant placement and the amount of vertical bone resorption at abutment placement. These, albeit limited, data are in contrast to the observations of Chen et al. (2007). They recently suggested that the initial thickness of the buccal bone crest may be a factor in determining the extent of crestalresorption during the healing phase of immediate placed implants (Chen et al. 2007). This study was, however, excluded from this review because the buccal bone thickness data for immediately placed implants that healed without grafts and membranes were mixed with data from implants that healed with grafts or a membrane. With this in mind, it is still interesting to note that, in the study of Chen et al. (2007), for sites showing a buccal bone dehiscence after 6 months of semi-submerged healing, the average initial buccal bone thickness was approximately 50% of the average initial buccal bone thickness of sites that healed without a dehiscence. The vertical resorption in the former sites was up to three times greater than that in the latter sites. This made the authors conclude that the initial buccal bone thickness influences vertical bone resorption but not horizontal resorption. Despite the limited amount of available data, this appears to be in contradiction to what could be calculated from the study of Botticelli et al. (2004), where no correlation could be seen between buccal bone thickness and vertical bone resorption.