Biologic Width around Dental Implants: An Updated Review

Document Type: Review Article


1 Assistant Professor, Department of Periodontics, Dental School, Shahid Beheshti University of Medical Sciences, Evin, Tehran, Iran

2 2Associate Professor, Department of Periodontics, Dental School, Shahid Beheshti University of Medical Sciences, Evin, Tehran, Iran

3 3Postgraduate Student, Department of Periodontics, Dental School, Shahid Sadoughi University of Medical Sciences, Yazd, Iran

4 Associate Professor, Department of Periodontics, Dental School, Shahid Beheshti University of Medical Sciences, Evin, Tehran, Iran

5 Professor, Department of periodontics, Bauru School of Dentistry, University of São Paulo, Bauru, SP, Brazil

6 6Associate Professor, Laboratory for Immunoregulation and Tissue Engineering (LITE), Division of Periodontology, Diagnostic Sciences &


Soft tissue-implant interface is an important anatomical feature contributing to the long-term success of dental implants. Based on the available evidence, different factors may influence biological width around implants including the surgical technique, implant loading, implant surface properties, abutment materials, implant position, and width of the peri-implant mucosa. The purpose of the present review was critical evaluation of the available data, regarding the factors that may influence the biologic width around implants and their subsequent effect on clinical performance of implants. Available literature on this subject published primarily in English from 1921 to 2014, was found by searching several electronic databases and by hand searching relevant journals as well. Totally, 70 relevant articles were selected for this narrative review. The structure of peri-implant mucosa has many similarities, as well as differences with its periodontal counterpart. Most studies report larger values for peri-implant biologic width compared to that of natural teeth. This literature review yielded contradictory data regarding the dimensions of the biologic width when different influential factors were taken into account.


Main Subjects

Historical perspectives

The ‘‘epithelial attachment’’ around teeth was first described in 1921 by Gottlieb (1). The ‘‘gingival crevice’’ or sulcus was later defined(2), followed by description of the connective tissue as three-dimensionally oriented fibers firmly connecting tooth structures to the adjacent gingiva (3). Marfino, Orban and Wentz (4), were the first to demonstrate that the attachment of gingiva to tooth is composed of gingival connective tissue attachment and junctional epithelium. In 1959, Sicher investigated the morphology of epithelial and connective tissue attachments to the teeth, described as the dentogingival junction (5). In 1961, Gargiulo et al. (6) quantified the vertical components of this structure in human cadavers and coined the term “biologic width”. Biologic width is normally composed of 0.97mm junctional epithelium (JE) and 1.07mm connective tissue attachment (CTA). Accordingly, the biologic width is acknowledged 2.04 mm, reflecting the sum of the epithelial and connective tissue measurements. In addition, sulcus depth (SD) was normally observed to be 0.69mm. These findings were substantiated by Vacek (7). After detailed assessment of 171 cadaver tooth surfaces, the mean measurements for sulcus depth, epithelial attachment and connective tissue attachment were found to be 1.34 mm, 1.14, and 0.77 mm, respectively. Vacek also realized that the connective tissue attachment was the most stable measurement, with the least degree of variance. On the other hand, significant variations were observed in epithelial attachment ranging from 1.0 mm to 9.0 mm.

Peri-implant tissues have many similarities, as well as some anatomical differences with periodontal attachment apparatus. The differences include lack of a periodontal ligament around implants, different orientation of connective tissue fibers and vascular distribution (8). Peri-implant biologic width has been investigated and measured in histological animal studies as well as clinical human studies. The purpose of this review was to draw comparisons and contrasts between biologic width around implants and biologic width around teeth and evaluate factors that may influence the peri-implant biologic width.


Structure and biological dimensions

  Listgarten et al. (9) in a comprehensive review article stated that biologic width around implants is composed of three distinct zones: sulcular epithelium, junctional epithelium, and connective tissue. Junctional epithelium around implants is derived from the oral epithelium, while the junctional epithelium around teeth originates from the reduced enamel epithelium (10); however, the structures appear morphologically similar (11-14).

  Junctional epithelium facing the implant or abutment surfaces is thin in its apical portion (40μm mean width), consisting of only a few cell layers (stratum basale and stratum granulosum) (15). The first animal studies by Berglundh et al. (8) confirmed that the peri-implant mucosa established a cuff-like barrier adhering to the surface of the titanium abutment. The peri-implant mucosa, similar to gingiva, has a well-keratinized oral epithelium that is contiguous with the junctional epithelium that faces the titanium surface. The structure of the peri-implant junctional epithelium is similar to that of natural dentition, with the exception that it is shorter and thinner (8, 15-19).

Berglundh et al. (18) in a canine study of non-submerged mandibular implants reported that epithelial proliferation begins around 1–2 weeks post-operatively, with a mature epithelial barrier establishing after 6–8 weeks. Fibroblasts are the dominant cell type at the connective tissue/implant interface at two weeks post-operatively, but their density decreases by week 4. After 4–6 weeks of healing, the collagen fibers are well organized. Hence, it was concluded that 6-8 weeks is required for the formation of a mature soft tissue attachment following surgery.

The mode of attachment of junctional epithelium to the implant surface has been demonstrated to be similar to that of teeth, which is by means of a basal lamina and hemidesmosomes (20). These findings have been verified both in vitro (23), and in vivo in rodents (27), canines (12, 15), non-human primates (21, 22) and humans (24, 25, 26). Contradictory findings were reported in an ultra-structural study by Shioya et al, who failed to observe hemidesmosomes and basal lamina adjacent to the implant surfaces (28).

A number of studies have investigated the composition of peri-implant tissues. Berglundh et al. (8) using an experimental model, observed collagen fibers to dominate peri-implant connective tissue, with fewer fibroblasts and vascular structures than normally seen in the gingiva around teeth. Importantly, collagen fibers were arranged parallel to the titanium surface in contrast to the orientation of gingival fibers, which tended to be arranged perpendicular to the cementum surface of the tooth root. Elsewhere in the marginal gingiva other fiber groups were arranged in a variety of different patterns (8, 29). In a recent study, Shioya et al. (28) reported a zone of dense collagen fibers, surrounded by loose connective tissue, consisting of a 3-dimensional network of collagen fibers running in different directions. A number of animal and human histologic studies have indicated that peri-implant collagen fiber bundles, while arranged in varying directions, are functionally oriented (30, 31). In contrast, Schierano et al. (32) in a human histologic study of nine retrieved abutments from seven patients demonstrated primarily horizontally and vertically directed connective tissue fibers around implants.

 The nature of the collagen fiber contacted with implant surfaces, as well as the anatomic details of peri-implant tissues have been studied by a number of investigators. Buser et al. (11) noted that connective tissue fibers are in direct contact with the implant surface although without true attachments. The direct connective tissue contacted to the implant surface was approximately 50 to 100 µm wide, consisting of dense, avascular circular fibers. In the adjacent outer zone, the connective tissue appeared less dense, with horizontal and vertical collagen fibers and a large number of blood vessels. Berglundh et al. (33) indicated that the zone of connective tissue near the junctional epithelium had a number of blood vessels, but the vessels were smaller in diameter and sparser than those found around teeth.

In a dog model, Moon et al. (19) demonstrated the presence of only a few blood vessels in peri-implant tissues, while noting numerous fibroblasts orientated with their long axes parallel to the implant surface (Astra Tech Implants). This attachment tissue was composed of approximately 80% collagen, 13% fibroblasts, 3% blood vessels and 3% residual tissue, resembling scar tissue. Lateral to this region, fewer fibroblasts with more collagen fibers and more vascular structures were observed which were divided into two zones: the inner avascular zone (0-40 µm) with more fibroblasts, and the outer zone (40–200 µm) with dense collagen and substantial numbers of vascular structures. It appears from these and other similar findings that the connective tissue attachment between titanium surfaces and connective tissue is established and maintained by fibroblasts.

 The peri-implant junctional epithelium appears similar to that found in the natural dentition. However, since there is no peri-implant cementum layer, most supracrestal connective-tissue fibers are oriented in parallel alignment to the implant surface. The presence of an avascular zone (50 to 100 µm) of dense, supracrestal circular connective-tissue fibers that are in direct contact with the implant surface has been confirmed through histologic examination (34).

  Berghlund et al. (33) compared the vascular system of the periodontal and peri-implant tissues in beagle dogs. The vascular supply to the gingiva originates from two different sources, large supraperiosteal blood vessels and the vascular plexus of the periodontal ligament. In contrast, the vascular system of the peri-implant mucosa of dogs appears to be derived solely from large supra-periosteal blood vessels lateral to the alveolar ridge. Interestingly though, scanning electron microscopic study in rats by Selliseth et al. (35) revealed that capillary loops in the connective tissue under the peri-implant junctional and sulcular epithelium appear to be anatomically similar to those found in normal periodontium.


Factors influencing peri-implant biologic width

a) Surface topography

Albrektsson and Wennerberg (36) have classified surface topography of implants into three general categories according to mean roughness (SA). The lowest degree of surface roughness is minimally rough with SA values of 0.5-1 µm. Moderately rough implants have SA of 1-2µm and rough ones have Sa greater than 2 µm. Buser et al. (11) investigated the soft tissue dimensions around three different titanium surfaces, namely a rough surface, a sandblasted surface and a polished surface. There were no significant differences in terms of soft tissue responses among these three implant surfaces. The soft tissue barrier consisted of a sulcus with a non-keratinized sulcular epithelium, a junctional epithelium, and a supra crestal connective tissue with an area of dense circular fibers near the implant surface. In the inner zone of connective tissue, next to the titanium surface, circular fibers were found. In the outer layer, horizontal and vertical fibers were seen running from the periosteum and the alveolar crest towards the oral epithelium. According to the authors, the orientation of fibers was different in rough and smooth surfaces. The fibers forming on smooth surfaces were mostly parallel to the implant surface, while porous-coated surfaces promoted the formation of upright fibers.

Watzak et al. (37) investigated the influence of implant design and surface topography on peri-implant soft tissue dimensions and peri-implant bone levels. The authors compared the response to screw-shaped machined surfaces (SA=0.53 μm), sandblasted acid-etched surfaces (Ra=2.1 μm), and cylindrical titanium plasma-sprayed surfaces (Ra=1.82 μm) in baboons after functional loading without oral hygiene. A histomorphometric examination of sulcus depth (SD), the dimension of the junctional epithelium (JE) and the amount of peri-implant connective tissue contact (CTC) showed that there were no significant differences between these three implant designs, neither in the maxilla nor in the mandible. Moreover, implant design and surface modifications did not have any impact on plaque accumulation or propagation of peri-implant mucositis after 1.5 years of functional loading.

Radiographic evaluation of marginal bone levels around dental implants with different designs after one year demonstrated that cylindrical implants with shorter high polish surface displayed less bone resorption (38).


b) Implant and abutment materials

     Multiple studies have documented the relationship between implant and abutment material composition and the nature of the resulting soft tissue attachment, which are summarized in (table 1) (17, 30, 39-47).

  Rompen et al, 48 in a review article, concluded that titanium was the only material that showed consistent soft tissue biocompatibility. Zirconium and aluminum oxide demonstrated favorable histological outcomes, whereas dental porcelain and gold were less biocompatible.






Table  1:  Studies about the effect  of implant and abutment materials on peri-implant biologic width.



Implant/ abutment  surface



The connective tissue contact were not significantly affected by the type of implants; but that the junctional epithelium and biologic width dimensions were larger around the implants with the machined collars.

The amount of inflammation was not different between the two implant types. Slightly more bone formation and more mature collagen formation were detected around the implants with the roughened collars compared to the implants with machined collars.

Loaded implants with machined and  roughened (SLActive) collars


Cochran et al (2014)39

compared with machine surfaces, the presence of a 0.7 mm laser ablated micro-channeled zone was associated with increased fibroblastic activity on the abutment-grooved surface, resulting in a denser interlacing complex of connective tissue fibers oriented perpendicular to the abutment surface.


Machined, laser microchannel surface abutments


Nevins et al (2010)40

At Au/Pt-alloy abutment sites  in comparison with Ti and ZrO2 :

1-apical migration of the barrier epithelium along with marginal bone loss occurred between the second to fifth months of healing.

2-the connective tissue zone(80 µm wide) contained less collagen and fewer fibroblasts and larger fractions of leukocytes.

3-Soft tissue healing appeared to be less stable.

Ti, ZrO2, Au/Pt-alloy abutments


Welander et al (2008)41

The controls had more soft tissue down-growth, greater osteoclastic activity, and increased saucerization compared with sites near the laser micro textured experimental implants.

Laser micro textured, machined collars


Weiner et al (2008)42

Junctional epithelium and connective tissue formed direct contact with the experimental implants. In the same study, TEM evaluation of the junctional epithelial cell membrane facing the surface-treated implant demonstrated dense plaques of hemidesmosomes. Nanoporous sol–gel-derived TiO2 thin film on ITIs Straumann implants improved soft tissue attachment in vivo.

Nanoporous TiO2 thin film on CPT implants, unmodified standard implants (ITI Implants)


Rossi et al (2008)43

No peri-implant marginal soft tissue dimensional differences between any of the Ti or Au designed implants.

CPT or gold alloy implants with Four different combinations of metal in coronal, central and apical zones (Ti/Ti/Ti, Ti/Au/Au, Au/Au/Au, Au/Ti/Ti).


Abrahamsson et al (2007)44

The formation of well-organized collagen fibers and abundant blood vessels in a newly formed loose connective tissue lateral to modSLA implants. While some fibers were oriented in a parallel alignment, others were extended and attached perpendicularly to the implant surface. In contrast, SLA implants appeared to be associated with a dense connective tissue area with parallel-running collagen fibers and sparse blood vessels.

SLA, modSLA implants


Schwarz F et al (2007)30

A biologic width is approximately 4.0 to 4.5 mm, consisting of an epithelial and a supracrestal connective tissue barrier around the experimental one-piece mini-implants that was similar to that described in animal studies. The oxidized and acid-etched implants experienced less epithelial down-growth and longer connective tissue barriers than machined implants.

One-piece mini-implants made of CPT with either oxidized, acid-etched, or  machined surfaces


Glauser et al (2005)45

9 months after implant placement, no significant differences between the responses to the two abutment materials.

Zirconia and titanium abutments


Kohal et al (2004)46

After 6 months of healing, the peri-implant mucosal attachment of the two types of abutments was similar in both linear dimensions and connective tissue composition.

Dual thermal acid-etched surface,

turned surface abutments


Abrahamsson et al (2002)17

After 6 months of healing, at CPT or Al2O3 abutment sites, a mucosal attachment had been formed consisting of an epithelial and a connective tissue portion that were approximately 2 mm and 1–1.5 mm in height, respectively. Abutments made of gold alloy did not allow the formation of a proper soft tissue abutment attachment, resulting in soft tissue marginal recession and bone resorption.


highly sintered Al2O3,

gold and porcelain fused to gold abutments


Abrahamsson et al (1998)47

CPT = commercially pure titanium; SLA = sandblasted, large grit and acid-etched; mod = modified; TEM = transmission electron microscopic




c) Surgical protocol 

A number of studies have examined the potential role of surgical protocol on peri-implant soft tissue healing. The effect of one- versus two-stage protocol on soft tissue healing of three different implant systems (Astra Tech Implants, Brånemark and Bonefit-ITI) was investigated and compared15. The histologic results demonstrated similar dimension and composition of epithelial and connective tissue components of biologic width with 1- or 2- stage procedures for all three implant systems. Similar findings have been reported in canine studies (16, 49-51). The current consensus appears to suggest that surgical protocol, especially one- versus two-stage procedures, has little effect on peri-implant soft tissue healing.


d) Loading time

  A number of investigators has examined the effects of loading protocols on biologic width. Cochran et al. (52) evaluated biologic width dimensions around non-submerged loaded and non-loaded implants testing two different surfaces (SLA and TPS) in a canine model. Biologic width dimensions for the unloaded implants after three months of healing were 0.49 mm for sulcus depth, 1.16 mm for junctional epithelium, and 1.36 mm for connective tissue. The corresponding measurements in the loaded group were 0.5 mm, 1.44 mm, and 1.01 mm respectively. Results were similar after 12 months of loading, confirming that the biologic width around implants dimensionally resembles biologic width around teeth. In addition, the dimensions of its constituents appear to be independent of loading time. These results have also been confirmed by Pontes et al. (53) (Conexao System), demonstrating that loading times had no influence on soft tissue healing. Similar findings were confirmed by Hermann et al. (51) who compared non-loaded with loaded implants (ITI Implant System) and submerged with non-submerged healing at different time intervals. Histometric measurements revealed significant changes within individual tissue compartments (SD, JE, CTC) over time. However, over a 15-month healing period, the total biologic width remained constant. According to Hermann et al, no statistically significant differences were detected among groups during the study period. Similar results were also reported by Siar et al. (54) comparing immediate versus delayed implant loading at 18 sites in six monkeys after three months of follow-up. The overall mean value of the biologic width was 3.9 mm in the immediate group and 3.8 mm in the delayed group. The authors concluded that there were no statistically significant differences in the dimensions and compositions between the two groups.

  In a study by Bakaeen et al. (55) the dimensions of peri-implant soft tissues around immediately- and early-loaded one-piece implants were compared with those of conventionally loaded one-piece implants. Forty-eight titanium sandblasted/acid-etched (SLA) implants were placed in four foxhounds. The implants were placed 3 months (group A), 21 days (group B), ten days (group C), and two days (group D) before restoration. Histometric analysis included dimensional measurements of the sulcus depth, junctional epithelium, the connective tissue seal, and gingival recession. There were no statistically significant differences among the four groups.

          Finally, in a systematic review of marginal soft tissue around implants subjected to immediate loading or immediate restoration, Glauser et al. (56) reported the occurrence of soft tissue healing comparable to that in conventionally loaded implants. Therefore, available evidence suggests that the loading time has little effect on the biologic width.


e) Implant macro-design and microgap position

The effects of implant macro-design, especially one- versus two-piece, and the position of the microgap on the fate of biologic width have been widely studied. A variety of implant designs are available, including one-piece implants with contiguous endosseous, transmucosal and abutment segments, and two-piece implants with separate endosseous and abutment segments. Among two-piece implants, some are at bone-level, i.e. the endosseous portion ends at bone crest and the transmucosal portion joins the abutment. Alternatively, the transmucosal portion can be contiguous with the endosseous segment. At each of these designs, the microgap between the implant and abutment is positioned at different levels. In bone level positioned implants, the microgap is potentially near the bone crest, whereas in transmucosal fixtures the microgap is above the bone level. Findings of related studies are summarized in (table 2) (15, 49, 53, 57-69).

          Results of mentioned studies in (table 2), seem to indicate that the dimensions and composition of the biologic width are not significantly influenced by the type of implant (i.e. one- versus two-piece implants) or the surgical protocol (i.e. one- versus two-stage). Limited evidence suggests, however that more deeply placed implants lead to a longer biologic width.









Table 2. Studies about the effect of implant macro-design and microgap position on peri-implant biologic width



Implant macro-design and

microgap position



S-BIC distances both lingually and buccally between test and control groups was similar.

Only buccal aBE-BC parameter presented statistically significant differences between test and control groups.

Test group presented 0.57 mm less recession than control group, being this difference statistically significant between the two groups

Platform switching abutment

at crestal bone level

1)in the test group, all prosthetic procedures were carried out direct to multi-base abutment without disconnecting it

2) in the control group, the multi-base abutment was connected/ disconnected five times during prosthetic procedures


Alves et al (2015)57

After 4 months of healing, marginal bone loss, gaps, and fibrous tissue were not detected around two types of implants.

The biologic width dimension ranged between 2.55 ± 0.16 and 3.26 ± 0.15 to one- and two-piece implants, respectively. This difference was influenced by the connective tissue attachment, while the dimensions of sulcus depth and junctional epithelium were similar between two groups.

Unloaded one- and two-piece implants