Osseo IQ
Chapter 1 · Foundations · §1.4

Peri-Implant Tissues & the Mucosal Seal

How soft tissue attaches to an implant — and why that seal is biologically weaker than the gingival attachment at a tooth.

Compiled by
Tan Khuu, DDS
Licensed dentist (CA & SC)
Audience
Oral surgeons, prosthodontists, periodontists & residents
Edition
1.0 · June 2026
Reviewed
June 2026 · next review June 2027
Reading time
~15 minutes
Evidence basis
Consensus statements + systematic reviews + primary literature
§1.4.1 — Overview

What seals an implant from the mouth

Every implant that succeeds does so behind a soft-tissue seal — a cuff of mucosa that walls off the underlying bone and osseointegrated surface from the contaminated oral environment. That seal resembles the gingiva around a natural tooth and is often described in the same vocabulary, but it is not the same tissue, and the differences are the whole point of this chapter. The peri-implant mucosa has a keratinized outer surface and a sulcus, an epithelial barrier that attaches to titanium by hemidesmosomes, and a band of connective tissue above the bone crest. What it lacks — a periodontal ligament, cementum, and the perpendicular fiber insertion and rich dual blood supply they bring — is precisely what makes the implant seal more fragile.1

The classic histometric account comes from Berglundh and Lindhe, who showed in a controlled canine model that the mucosa re-forms a relatively constant supracrestal dimension — a biologic width of roughly 3–4 mm — regardless of the starting tissue height.1 That total comprises approximately 2 mm of junctional (barrier) epithelium and approximately 1.3–1.8 mm of supracrestal connective tissue. Cochran and colleagues confirmed an analogous, stable dimension around titanium implants and tied it to the position of the implant–abutment interface.2 The clinically decisive feature lives in that connective-tissue band: where a tooth anchors collagen fibers perpendicularly into cementum, the implant offers no cementum, so its fibers run parallel to the surface and merely adhere. The tissue is also fibroblast-poor, collagen-rich, and scar-like, and — lacking the periodontal-ligament vascular plexus — it is fed only by supraperiosteal vessels, leaving it less vascular than gingiva. The downstream consequences are deeper probing and a greater susceptibility to peri-implant disease.3

The peri-implant seal is an adhesion, not an attachment — a scar-like collar that holds against the mouth by proximity rather than by anchorage.
◆ Key concept · Adhesion, not attachment

At a tooth, the soft-tissue connection is a true attachment: connective-tissue fibers insert perpendicularly into root cementum and are anchored by a periodontal ligament. At an implant there is no cementum and no ligament, so the connective-tissue collagen runs parallel to the surface — it adheres rather than inserts. The epithelial component (hemidesmosomes to titanium) is broadly comparable at tooth and implant; the decisive weakness is below the epithelium, in the fiber orientation and the impoverished blood supply of the connective-tissue zone.

§1.4.2 — Anatomy of the seal

From sulcus to crest, zone by zone

Read from the mouth downward, the peri-implant seal is layered. The coronal-most lining is the sulcular epithelium — thin, non-keratinized, unattached — bounding the sulcus where the probe and biofilm sit. Apical to it is the junctional (barrier) epithelium, the principal epithelial seal, attaching to titanium through hemidesmosomes and a basal lamina and measuring approximately 2 mm in height.1 Below the epithelium lies the supracrestal connective tissue, a band of roughly 1.3–1.8 mm whose collagen runs parallel to the implant surface. Together the epithelial and connective-tissue components constitute the biologic width.

Beneath and around all of this is the vascular bed. Around a tooth, the marginal tissue is supplied both by supraperiosteal vessels and by the periodontal-ligament plexus; at an implant the ligament — and therefore its plexus — is absent, so the tissue depends on supraperiosteal vessels alone. The result is measurably lower vascularity, reduced immune surveillance, and slower healing, which is why the same bacterial challenge is more dangerous at an implant than at a tooth.3

The figure makes the central contrast visible: at the tooth, fibers fan perpendicularly into cementum and a probe is arrested at the attachment; at the implant, fibers parallel the surface and a probe slides closer to the crest.

Natural tooth Dental implant biologic width ~3–4 mm root + cementum bone sulcus JE ~2 mm CT ~1 mm fibers insert ⟂ (perpendicular) probe stops at attachment titanium (no cementum) bone (BIC) sulcus JE ~2 mm CT ~1.3–1.8 mm fibers run ∥ (parallel) — adhesion only probe penetrates closer to crest Dimensions are mean values (Berglundh & Lindhe 1996; Cochran 1997) and vary by site, biotype, and implant design.
Figure 1. The soft-tissue seal at a tooth versus an implant. Both share a sulcus, a junctional/barrier epithelium (~2 mm) and a supracrestal connective-tissue zone, together forming a ~3–4 mm biologic width. The decisive difference is fiber orientation: perpendicular insertion into cementum at the tooth (true attachment, probe arrested) versus parallel adhesion along titanium at the implant (no insertion, deeper probe penetration).12

Tissue-zone explorer

The seal is conventionally divided into four zones from sulcus to vascular bed. Select any zone to review its structure, dimension, and clinical significance.

Tap a zone to expand.

§1.4.3 — Comparison

Tooth versus implant soft tissue

The seven features below summarize how the mucosal seal at an implant departs from the gingival attachment at a tooth. Each row carries an evidence grade: dimensional and structural claims rest on controlled histometric studies and systematic syntheses, while a handful of inferences about defense and disease behavior are drawn more cautiously from consensus interpretation.

Table 1 · Soft-tissue seal — natural tooth versus dental implant
FeatureNatural toothDental implantEvidence
Attachment apparatusPeriodontal ligament + cementumNone — direct bone-to-implant contact below the sealSyst. review
Connective-tissue fibersInsert perpendicularly into cementum (true attachment)Run parallel to the surface — adhesion, no insertionSyst. review
Epithelial sealJunctional epithelium, hemidesmosomes (~2 mm)Junctional/barrier epithelium, hemidesmosomes (~2 mm)Syst. review
Supracrestal CT height~1.0–1.1 mm~1.3–1.8 mm (fibroblast-poor, scar-like)Syst. review
VascularityHigher — gingival and PDL plexusReduced — supraperiosteal vessels onlySyst. review
ProbingProbe arrested at the attachment apparatusProbe penetrates closer to crest; deeper readingsConsensus
Defense / repairMore robust barrier and healingWeaker seal; greater susceptibility to peri-implant diseaseConsensus
✦ Clinical pearl · Read the probe in context

A 4 mm probing depth at an implant is not the same finding as 4 mm at a tooth. Because the connective tissue adheres in parallel rather than inserting, the probe tip travels closer to the crest for any given true tissue height, so baseline readings run deeper. What matters in surveillance is the change from each implant's own baseline, together with bleeding or suppuration on probing — not a single absolute number compared against tooth norms.3

▲ Common pitfalls
  • Treating a deeper peri-implant probing depth as pathology by tooth standards, and re-entering a healthy site.
  • Placing a rough surface subcrestal in a thin (≤2 mm) mucosa and expecting no crestal change — the seal will resorb bone to re-establish its minimum biologic width.
  • Assuming any implant surface or abutment can recreate a true perpendicular fiber insertion; it cannot, and the seal remains an adhesion.
  • Repeated abutment dis/reconnection, which disrupts the immature epithelial seal and can drive additional crestal bone loss.
§1.4.4 — Clinical translation

Why the seal shapes planning

Because a minimum supracrestal dimension re-forms regardless of the tissue it starts with, soft-tissue height is not a passive bystander — it is a determinant of crestal bone behavior. Where mucosa is thin (≤2 mm) at the time of abutment connection, healing characteristically includes crestal bone resorption to create the vertical space the seal requires.1 This is the biological argument for thickening thin tissue before or at placement, for thoughtful interface positioning, and for restraint in disturbing the seal once formed. The grafting decision itself is developed in its own chapter (see Soft-Tissue Grafting Algorithm →).

The same weak-seal biology explains why peri-implant inflammation, once established, can advance toward bone more readily than gingival inflammation at a tooth: parallel fibers and reduced vascularity offer the bacterial front a less-resistant apical path and a less-competent defense.3 Surveillance, maintenance, and the staged management of established disease are therefore inseparable from the anatomy described here (see Supportive Peri-Implant Care → and Peri-Implant Disease Management →).

◆ Key concept · The seal sets the crest

The biologic width is not negotiable: if the soft tissue cannot accommodate its ~3–4 mm minimum, the bone yields to make room. Treat mucosal thickness as part of bone-level planning, not as an afterthought to it.

§1.4.5 — Glossary

Key terms

Peri-implant mucosa
The soft-tissue cuff surrounding a transmucosal implant or abutment; resembles gingiva but lacks a periodontal ligament and cementum.
Mucosal seal
The combined epithelial and connective-tissue barrier that isolates the osseointegrated implant from the oral environment.
Biologic width (supracrestal tissue attachment)
The relatively constant supracrestal soft-tissue dimension (~3–4 mm) that re-forms after surgery; ~2 mm junctional/barrier epithelium + ~1.3–1.8 mm connective tissue.
Junctional / barrier epithelium
The epithelial component of the seal, attaching to titanium by hemidesmosomes and a basal lamina; ~2 mm in height.
Supracrestal connective tissue
The collagen-rich, fibroblast-poor band between epithelium and bone crest in which fibers run parallel to the implant.
Perpendicular fiber insertion
The dentogingival/dentoperiosteal fibers anchoring into root cementum at a tooth — a true attachment absent at implants.
Parallel fiber orientation
The arrangement of peri-implant collagen running alongside the surface; adhesion without insertion.
Supraperiosteal vessels
The sole vascular source of peri-implant mucosa in the absence of a periodontal-ligament plexus; account for its reduced vascularity.
§1.4.S — Self-test

Self-Test

1. At a healthy implant a probe reads deeper than at an adjacent tooth of the same true tissue height. The best structural explanation is:
B is correct. At a tooth, fibers insert perpendicularly into cementum and arrest the probe at the attachment. At an implant the collagen merely runs parallel (adhesion, not insertion), so the probe passes closer to the crest. There is no PDL at an implant (D), and an epithelial barrier does exist (A/C wrong).
2. The peri-implant biologic width (~3–4 mm) is typically apportioned as:
B is correct. Berglundh & Lindhe described a relatively constant seal of roughly 2 mm epithelium plus ~1.3–1.8 mm connective tissue. The dimension is reproducible (not random, D); A and C misstate the split.
3. A thin-mucosa site (~2 mm) receives an implant with the rough surface placed slightly subcrestal. The most likely consequence is:
A is correct. A minimum supracrestal dimension re-forms, so thin mucosa is established partly at the expense of crestal bone. The seal still forms epithelially (B wrong); implants never gain perpendicular insertion (C wrong); thin tissue is not more vascular (D wrong).
4. Peri-implant mucosa has reduced defensive and healing capacity compared with gingiva, chiefly because of:
B is correct. Without a PDL the tissue is fed only by supraperiosteal vessels, lowering vascularity and immune surveillance; the connective tissue is collagen-rich but fibroblast-poor and scar-like. C is false (supply is reduced), and the implant has no perpendicular insertion (D).
5. Which apparatus is present at a natural tooth but absent at an implant?
C is correct. Both tooth and implant share a sulcus, a keratinized surface, and a junctional/barrier epithelium. The PDL and cementum — and the perpendicular fiber insertion they permit — are unique to the tooth.
6. The junctional/barrier epithelium attaches to the titanium surface by means of:
B is correct. As at a tooth, the epithelial seal adheres through hemidesmosomes and a basal lamina. Sharpey's fibers and perpendicular collagen describe cementum insertion (A/C), and there is no PDL at an implant (D).
7. The approximate height of the junctional/barrier epithelium in the peri-implant seal is:
B is correct. Berglundh & Lindhe measured junctional epithelium of about 2.0–2.1 mm; the connective-tissue band adds ~1.3–1.8 mm, giving a ~3–4 mm total.
8. The approximate height of the supracrestal connective-tissue zone is:
B is correct. The connective-tissue band measured ~1.3 mm (test) to ~1.8 mm (control) in the original canine model; ~3–4 mm (C) is the whole biologic width, not the CT zone alone.
9. Compared with gingiva, peri-implant connective tissue is best described as:
B is correct. The peri-implant connective tissue is collagen-rich but fibroblast-poor, resembling scar tissue — a key reason the seal is less defensible. It contains no cementum (D).
10. The sole vascular source of peri-implant mucosa, in the absence of a PDL, is:
C is correct. Without a ligament, the tissue is supplied only by supraperiosteal vessels — the structural basis for its reduced vascularity. The PDL plexus (A) is precisely what is missing.
11. The orientation of supracrestal collagen fibers at an implant is predominantly:
B is correct. Lacking cementum, the fibers run parallel (and circumferentially) along the surface — adhesion, not insertion. Perpendicular and oblique insertion (A/C) require cementum.
12. The seminal description of peri-implant mucosal dimension ("biological width revisited") is attributed to:
B is correct. Berglundh & Lindhe (J Clin Periodontol 1996) measured the relatively constant epithelial and connective-tissue components of the peri-implant seal. Cochran (1997) later corroborated the dimension around titanium implants.
13. Cochran and colleagues (1997) contributed which key observation about the peri-implant seal?
B is correct. Cochran et al. demonstrated a reproducible biologic width around titanium implants and linked its apical extent to the implant–abutment interface. The other options contradict established findings.
14. Why can peri-implant inflammation progress toward bone more readily than periodontitis at a tooth?
B is correct. Adhesion-only parallel fibers, scar-like tissue, and supraperiosteal-only blood supply combine to make the seal easier to breach and slower to defend — so lesions can extend closer to bone.
15. A 5 mm probing depth recorded at a healthy implant baseline is best interpreted as:
B is correct. Because the probe penetrates deeper at implants, an absolute depth is less informative than change over time combined with inflammatory signs. It is not equivalent to a tooth reading (D).
16. The keratinized outer surface and a sulcus are:
C is correct. Both the gingiva and the peri-implant mucosa present a keratinized surface and a sulcus; the differences lie deeper, in fiber insertion, cementum/PDL, and vascularity.
17. Which intervention most directly addresses a thin peri-implant biotype to protect the crest?
B is correct. Thickening thin mucosa supplies the vertical dimension the seal needs without sacrificing bone. Deeper rough-surface placement (A) and repeated disconnection (C) tend to worsen crestal loss.
18. Repeated dis/reconnection of the abutment is undesirable primarily because it:
B is correct. Each disconnection reopens the immature seal, shifting the epithelium apically and provoking further crestal remodeling — a rationale for one-abutment-one-time approaches.
19. "Biologic width" at an implant is most precisely termed:
B is correct. The 2018 classification renamed "biologic width" the supracrestal tissue attachment — the epithelial plus connective-tissue seal above the crest. It is distinct from keratinized width (C) and from BIC (D).
20. The reported soft-tissue dimensions in this chapter should be applied as:
B is correct. The ~2 mm / ~1.3–1.8 mm figures are means; real seals vary with anatomy, biotype, and design. Human studies broadly corroborate the canine model while showing site-to-site variation.
1. Compare the soft-tissue attachment at a natural tooth with the mucosal seal at an implant, and defend why the implant seal is considered biologically weaker.
Model answer. Both have a keratinized surface, a sulcus, and a junctional/barrier epithelium attaching by hemidesmosomes. The decisive differences are below the epithelium: a tooth has a periodontal ligament and cementum into which connective-tissue fibers insert perpendicularly, giving a true, load-resistant attachment fed by both gingival and PDL vascular plexuses. An implant has neither PDL nor cementum, so connective-tissue collagen merely runs parallel to the surface — adhesion, not insertion — within a fibroblast-poor, scar-like tissue supplied only by supraperiosteal vessels. The result is a seal that resists the probe and bacterial challenge less effectively and heals more slowly, which is why peri-implant tissues are more susceptible to disease.
Examiner follow-ups:
  • Why does this make probing readings deeper at an implant?
  • How does reduced vascularity affect peri-implantitis risk?
  • What does keratinized mucosa width add, if anything?
2. Define the peri-implant biologic width (supracrestal tissue attachment) and justify why understanding its minimum dimension changes how you plan implant depth in a thin biotype.
Model answer. The biologic width is the supracrestal soft-tissue seal that re-forms above the bone crest — about 3–4 mm total, comprising roughly 2 mm of junctional/barrier epithelium and 1.3–1.8 mm of connective tissue (Berglundh & Lindhe). It re-establishes to a relatively constant minimum after surgery or abutment connection. Clinically this means that if available mucosal height is less than this minimum, the body obtains the needed vertical dimension by resorbing crestal bone. In a thin biotype I therefore favor measures that protect the crest: appropriate (often slightly supracrestal or platform-switched) interface position, soft-tissue augmentation to increase mucosal thickness, and avoiding repeated abutment disconnection.
Examiner follow-ups:
  • How does mucosal thickness predict first-year bone loss?
  • How does platform switching interact with this concept?
  • Would you graft tissue before or at placement?
3. Explain the orientation of supracrestal connective-tissue fibers at an implant versus a tooth, and defend the clinical implication for seal integrity and disease progression.
Model answer. At a tooth, dentogingival and dentoperiosteal fibers insert perpendicularly into cementum, anchoring the connective tissue and creating a mechanically robust barrier. At an implant there is no cementum, so collagen fibers run parallel (or circularly) to the surface and merely adhere — there is no perpendicular insertion. This adhesion-only arrangement, combined with fibroblast-poor scar-like tissue and reduced vascularity, makes the seal easier to disrupt and gives bacterial challenge a less-resistant path apically. Consequently peri-implant inflammation can progress faster and further toward bone than equivalent gingival inflammation at a tooth — a key reason peri-implantitis behaves more aggressively than periodontitis.
Examiner follow-ups:
  • How does fiber orientation relate to probing depth differences?
  • Why might peri-implantitis lesions extend closer to bone than periodontitis lesions?
  • Can any implant surface induce perpendicular fiber insertion?
4. A referring colleague is alarmed by a 5 mm probing depth at a two-year-old implant. Walk the examiner through how you interpret peri-implant probing and what would actually concern you.
Model answer. First I would not equate 5 mm at an implant with 5 mm at a tooth. Because the connective tissue adheres in parallel rather than inserting, the probe penetrates closer to the crest, so peri-implant baselines run deeper and an absolute number is weakly informative. What matters is the trajectory: I compare against this implant's own documented baseline and look for an increase in depth over time. I weight bleeding on probing and especially suppuration heavily, and I correlate with radiographic crestal bone level. A stable 5 mm with no bleeding, no suppuration, and no progressive bone loss can be compatible with health; a rising depth with bleeding or bone loss signals mucositis or peri-implantitis and triggers the disease pathway. I also use gentle, standardized probing force to avoid disrupting the immature seal.
Examiner follow-ups:
  • What probing force do you use, and why does it matter at an implant?
  • Which single finding most strongly distinguishes peri-implantitis from health?
  • How does the deeper baseline change your recall radiograph strategy?
5. Defend the statement that "soft-tissue thickness is part of bone-level planning, not an afterthought," using the biology of the seal.
Model answer. The supracrestal seal re-forms to a relatively constant minimum (~3–4 mm) irrespective of the tissue it starts with. If the mucosa is thinner than that minimum at abutment connection, the only way the body can create the vertical space is to resorb crestal bone — so thin tissue is, in effect, traded for lost bone. Because the crest yields to the seal, I treat mucosal thickness as a bone-preserving variable: I assess biotype before surgery, consider soft-tissue augmentation to build thickness, choose interface position and platform geometry to keep the rough surface away from a thin crest, and minimize disruptions to the maturing seal. Planning bone level without planning the soft tissue that sits above it ignores the structure that actually dictates where the crest ends up.
Examiner follow-ups:
  • What mucosal thickness threshold prompts augmentation in your hands?
  • How does this interact with subcrestal versus crestal placement?
  • Is augmenting before placement or at placement preferable, and why?
§1.4 — References

References

  1. Berglundh T, Lindhe J. Dimension of the periimplant mucosa. Biological width revisited. J Clin Periodontol. 1996;23(10):971–973. PMID: 8915028
  2. Cochran DL, Hermann JS, Schenk RK, Higginbottom FL, Buser D. Biologic width around titanium implants. A histometric analysis of the implanto-gingival junction around unloaded and loaded nonsubmerged implants in the canine mandible. J Periodontol. 1997;68(2):186–198. PMID: 9058338
  3. Araújo MG, Lindhe J. Peri-implant health. J Clin Periodontol. 2018;45(Suppl 20):S230–S236. doi:10.1111/jcpe.12952

Soft-tissue dimensions are mean values and vary by site, biotype, and implant design. Evidence grades: Systematic review Consensus Preclinical.

About this chapter

This chapter is part of Osseo IQ — a clinical reference for implant dentistry. Content is sourced from consensus statements, systematic reviews, and primary literature; each key recommendation carries an evidence grade, and every page records its review date. Material is reviewed on a rolling annual cycle.

How to cite: Khuu T, ed. Peri-Implant Tissues & the Mucosal Seal. In: Osseo IQ, 1st ed. §1.4. June 2026. Accessed [date]. [URL]

Compiled by: Tan Khuu, DDS — Doctor of Dental Surgery and a licensed dentist in California and South Carolina. Osseo IQ summarizes published evidence and clinical guidelines and is not a substitute for individual clinical judgment. Image credits: Figure 1 original schematic illustration © Osseo IQ, 2026.

For licensed clinicians — educational use only. This chapter summarizes published evidence and is not a substitute for individual clinical judgment, examination, or the standard of care in your jurisdiction. Verify drug doses, devices, and protocols against current manufacturer instructions and local guidelines.

© 2026 Osseo IQ · Edition 1.0 · Chapter 1 Foundations · §1.4 · Last reviewed June 2026