Journal of Dentistry 158 105808

Contents lists available at ScienceDirect

Journal of Dentistry
journal homepage: www.elsevier.com/locate/jdent

Review article

Systematic review and meta-analysis on marginal bone loss of dental

implants placed in augmented or pristine bone sites: Findings from clinical
long-term studies
Sonja Memenga-Nicksch a,1 , Felix Marschner b,1,* , Nele Henrike Thomas c ,
Dennis Holzwart c , Ingmar Staufenbiel a
a
Hannover Medical School, Department of Conservative Dentistry, Periodontology and Preventive Dentistry, Hannover, Germany
b
University Medical Center G!
ottingen, Department of Preventive Dentistry, Periodontology and Cariology, G!
ottingen, Germany
c
Hannover Medical School, Institute for Biostatistics, Hannover, Germany

A R T I C L E I N F O A B S T R A C T

Keywords: Objectives: Bone augmentation can be a risk factor for complications in dental implant therapy. Aim of this
Dental implants systematic review and meta-analysis was to assess the marginal bone loss (MBL) between dental implants placed
Long-term in augmented or pristine bone sites.
Bone loss
Study selection: Four electronic databases (PubMed, Embase, Scopus, and Web of Science) were searched for
Augmented bone
Pristine bone
clinical studies with a follow-up period of at least five years. Additionally, a hand search was performed. Risk of
Systematic review bias was assessed using the Newcastle-Ottawa-Scale for non-randomized studies and Cochrane risk-of-bias-tool-2
Meta-analysis for randomized controlled trials. Random-effect meta-analysis was performed for the mean MBL at implant level
after 5 years of loading for dental implants placed in pristine and augmented bone sites. The study was registered
in PROSPERO: CRD42024615716.
Sources: A total of 11 sources reporting on 10 studies were included. Four studies examined dental implants
placed in pristine bone sites, four in augmented bone sites, and two in both conditions.
Data: Random-effect meta-analysis for pristine bone sites estimated a mean MBL of 0.79 mm (95 % CI:
0.32–1.26) and for augmented bone sites a mean MBL of 1.90 mm (95 % CI: 1.73–2.07) after five years with a
high heterogeneity of MBL reported in included studies.
Conclusions: Pre-implant augmentation appears to be one but not the only risk factor for MBL. Future studies on
implant complications should describe in detail the patient (e.g., adherence, periodontitis history) and local
factors (e.g., cleanability of superstructure, attached mucosa) in order to identify further risk factors in the long
term.
Clinical Significance: Dental implants in augmented bone sites show greater MBL after five years compared to
pristine bone sites. Augmentation may increase early bone remodeling. Clinicians should consider augmentation-
related remodeling dynamics and patient-specific risk factors when planning implant therapy to optimize long-
term outcomes and reduce complications associated with peri‑implant bone loss.

Introduction studies have indicated, that within the first year following tooth
extraction, up to half of the alveolar ridge width may be resorbed [4,5].
Tooth loss has a significant impact on the quality of life of the This can significantly impact implant planning and positioning. To
affected individuals [1]. Dental implants are a well-established treat- achieve primary implant stability during the surgical procedure and
ment option for the rehabilitation of tooth loss [2]. Tooth loss is typi- ensure successful osseointegration, simultaneous or staged lateral
cally accompanied by a reduction in alveolar ridge bone [3]. Clinical and/or vertical bone augmentation procedures are often necessary to

* Corresponding author at: Department of Preventive Dentistry, Periodontology and Cariology, University Medical Center G!
ottingen, Robert-Koch-Str. 40, 37075
G!
ottingen, Germany.
E-mail address: [email protected] (F. Marschner).
1
These authors contributed equally as first authors to this work

https://doi.org/10.1016/j.jdent.2025.105808
Received 15 April 2025; Received in revised form 2 May 2025; Accepted 5 May 2025
Available online 6 May 2025
0300-5712/© 2025 The ( . 5 Published by Elsevier Ltd. This is an open access article under the CC BY license . 4 5 )/ 2 5 0/ 2 :
S. Memenga-Nicksch et al. Journal of Dentistry 158 105808

reconstruct atrophic alveolar ridges [6,7]. Advances in bone augmen- Research question
tation procedures now allow for minimally invasive techniques through
the use of bone substitutes and barrier membranes [8]. Regarding the PECO criteria [27], the following question was
Marginal bone loss (MBL) of ω1.5 mm has been defined as a positive defined:
outcome for successful implant placement after one year of loading [9]. Are there differences in radiographically assessed MBL between
However, recent research on new implant surfaces and designs indicates implants placed in augmented versus pristine bone sites in long-term
that a bone loss of 1.5 mm is no longer considered the standard after one clinical studies with a follow-up period of at least 5 years in patients
year [10]. with osseointegrated dental implants?
Long-term studies have reported similar survival rates for dental
implants placed in both – augmented and pristine sites [11,12]. This Eligibility criteria
finding is consistent with a systematic review and meta-analysis [13].
Regarding long-term outcomes on MBL, studies have shown comparable The inclusion criteria were clinical studies conducted since 2000, of
results between implants placed in augmented and pristine sites [7,14]. all levels of evidence, with a follow-up of at least five years. These
On the other hand, some studies have reported inferior outcomes for studies had to report on titanium or titanium alloy implants with a
implants placed in augmented sites [15–17]. Progressive MBL around defined reference point for measuring bone loss, describe bone
the neck of a dental implant, assessed radiographically is considered an augmentation before or at the time of implantation, or involve im-
early indicator of peri‑implantitis [18]. Novel findings further suggests plantation in pristine bone. Additionally, only studies that used the
that elevated levels of molecular biomarkers such as active-matrix parallel technique for radiographic examinations at both baseline and
metalloproteinase-8 in peri‑implant sulcus fluid, may also serve as po- follow-up were included. The methods had to be consistent with those
tential indicator [19]. Therefore, it is crucial to prevent or reduce this established at baseline after prosthesis placement (at the earliest after 10
initial bone remodeling as much as possible, starting from the time of weeks), and the studies had to be published in English or German.
implant placement. Studies have investigated the effectiveness of Exclusion criteria comprised studies those with less than ten implants
different approaches, including platform switching, different per group, insufficient data, immediate and early loading (ω10 weeks)
apical-coronal implant positions relative to the alveolar crest, and of implants after implant placement, review articles, no standardized or
modified implant surfaces [20–24]. While these methods offer prom- no parallel technique (e.g., panoramic radiograph) for radiographic
ising strategies for reducing bone loss and enhancing implant success, examinations, and a follow-up of less than five years.
the condition of the bone at the implant site plays a critical role.
Moreover, various factors such as patient-specific characteristics, the Search strategy
extent of bone loss, the surgical technique, and the kind of soft tissue
matrices employed can influence the outcomes significantly [25]. A systematic literature search was performed in four electronic da-
Augmented bone, which requires reconstruction procedures due to tabases, including MEDLINE via PubMed, Embase via Ovid, Scopus and
insufficient volume or quality, may behave differently compared to Web of Science in November 2024. The search term was formulated as
pristine bone. Understanding how these two types of bone affect follows: (("peri‑implant bone loss" OR "bone loss" OR "biological com-
long-term implant stability and radiographically assessed MBL is plications" OR “peri‑implantitis” OR "peri‑implant disease") AND
essential for optimizing treatment outcomes. ("dental implants" OR "dental implantation") AND (“pristine” OR “pris-
The aim of this systematic review and meta-analysis was to investi- tine sites” OR “augmented sites” OR “augmentation” OR "bone
gate potential differences in radiographically assessed MBL between augmentation" OR "augmented sites" OR "autologous bone graft" OR
dental implants placed in augmented versus pristine bone sites in long- "autogenous bone grafting" OR "autologous bone" OR “autologous” OR
term clinical studies (randomized controlled trials, prospective and "xenogenous bone graft" OR "xenogenous bone grafting" OR "xen-
retrospective case series) with a follow-up period of at least five years in ogenous bone" OR xenogenous)). The complete search terms for all
patients with osseointegrated dental implants. electronic databases are provided in Supplemental Table S1. Addition-
ally, a hand search using Google and Google Scholar was performed, and
Materials and methods all reference lists of included studies were screened.

Study registration and protocol Study selection

This systematic review and meta-analysis was conducted in accor- After the initial search, duplicative records were identified using the
dance with the Preferred Reporting Items for Systematic Reviews and reference management software EndNote 21.2 (Clarivate, Philadelphia,
Meta-Analysis statement and checklist (PRISMA) [26]. The study pro- PA, USA). The screening of titles, abstracts and full text was indepen-
tocol was registered on the Prospective Register of Systematic Reviews dently conducted by two authors (SMN and FM). Any disagreements
(PROSPERO) under the identification number: CRD42024615716 prior between the authors were resolved through discussion, and, if necessary,
to its initiation. by consulting a third author (IS). Reasons for exclusions after the full-
text review are presented in Supplemental Table S2.

Population (P), exposure (E), comparison (C), and outcome (O) Data extraction

The population, exposure, comparison, and outcome criteria [27] Data were extracted independently by two authors (SMN and FM)
(PECO) were defined as follows: using a pilot-tested spreadsheet. These comprised authors, year of
publication, country, study design, follow up, number of subjects, age,
→ Population (P): edentulous and partially edentulous patients with gender, subjects’ characteristics, number of implants, implant system,
osseointegrated titanium and/or titanium alloy dental implants; type of restauration, time point of augmentation, augmentation mate-
→ Exposure (E): dental implants placed in augmented sites prior or rial, type of membrane, smoking history, diabetes mellitus, history of
simultaneous to implant placement; periodontitis, oral hygiene, and mean MBL in mm ↑ SD (standard de-
→ Comparison (C): dental implants placed in pristine bone sites; viation) from baseline at both the implant and subject level after 5 and
→ Outcome (O): radiographically assessed mean MBL after 5 years of 10 years of loading. In case of missing data, the corresponding author
loading in millimeters from baseline. was contacted via email. After two weeks, a second email was sent to

2
S. Memenga-Nicksch et al. Journal of Dentistry 158 105808

Fig. 1. Flowchart of literature search and study selection.

non-responders. Common and random-effects meta-analyses were conducted. Hetero-

geneity was assessed through Cochran’s Q test and the I² statistic [31].
Risk of bias and quality assessment Given the substantial heterogeneity observed across analyses (Cochran’s
Q test p-values ω 0.05, I² ↓ 50 %) [32], the random-effect model was
Risk of bias and quality of the included non-randomized and non- used for final analysis. A significance level of α ↔ 0.05 was applied.
interventional studies were assessed using the Newcastle-Ottawa-Scale
(NOS) [28]. All studies were rated with stars, and the topics selection Quality of evidence
of study groups, comparability of participants, and exposure/outcome
were evaluated (maximum 9 stars). To align NOS scores with the Agency Quality of evidence was assessed in this meta-analysis using the
for Healthcare Research and Quality (AHRQ) standards [29], studies are Grading of Recommendations Assessment, Development, and Evalua-
classified as: good quality (3 or 4 stars in the selection domain AND 1 or tion (GRADE) system [33]. Evaluation criteria included the risk of bias,
2 stars in the comparability domain AND 2 or 3 stars in the exposur- inconsistency, indirectness, imprecision, and publication bias.
e/outcome domain), fair quality (2 stars in the selection domain AND 1 Following these criteria the evidence of findings is rated as high, mod-
or 2 stars in the comparability domain AND 2 or 3 stars in the expo- erate, low, and very low.
sure/outcome domain), and poor quality (0 or 1 star in the selection
domain OR 0 stars in the comparability domain OR 0 or 1 stars in the Results
exposure/outcome domain). Further details are provided in Supple-
mental Table S3. Study selection
Risk of bias and quality of the included randomized controlled trials
(RCTs) were assessed using the criteria presented in the Cochrane risk- 4889 potentially relevant sources were identified through electronic
of-bias tool 2 (RoB 2) [30]. The RoB 2 tool consists of five domains: databases (MEDLINE via PubMed, Embase via Ovid, Scopus and Web of
bias arising from the randomization process, deviation from intended Science). After the removal of duplicates and the exclusion of records
intervention, missing outcome data, measurement of outcomes, and based on title and/or abstract level, 57 studies were considered eligible
selection of results [30]. The overall risk of bias was assessed as low, for full-text screening. On full-text level, additional 49 studies were
some concerns or high. The evaluations were conducted independently excluded, reasons for exclusion are outlined in Supplemental Table S2.
by two authors (SMN and FM). Disagreements between the authors were Three studies were included through hand search [34–36]. A total of 11
resolved through discussion. Further details are provided in Supple- sources reporting 10 studies that met the inclusion criteria were
mental Figure S1. included in this systematic review [34–45]. A complete description of
the literature search is shown in the PRISMA flowchart in Fig. 1.
Meta-analyses
Study characteristics
Meta-analyses were performed for the mean MBL (mm) at implant
level after 5 years of loading for implants placed in pristine bone and The study characteristics are reported in Table 1. Out of the 10 in this
augmented bone sites. Statistical analysis was performed using R systematic review included studies, four evaluated MBL around dental
(version 4.3.0) in combination with the packages “meta” (version implants placed in pristine bone sites [34,35,39,42], four in augmented
8.0–2), “metasens” (version 1.5–2), and “metafor” (version 4.8–0). bone sites [37,38,43,45], and only two in both pristine and augmented

3
 Table 1
Characteristics of included studies.
Study, country Study design Subjects Mean age ± SD Male Follow-up Subject’s characteristics Im
(Group) (n) (years) (%) (years) (n

Base- Follow- Base-line Follow- Base- Follow-

line up up line up

Wennstr! om et al. Prospective study 51 47 59.5 – 39.2 – 5 Periodontitis-susceptible subjects 13

2004 [35], Sweden ↑ 9.7 following basic periodontal therapy
Pristine bone with partial edentulism
Jemt and Johansson Prospective study 76 61↗, 41↘ 60.1 – 63.2 – 15 Maxillary edentulous subjects 34
2006 [34], Sweden ↑ 11.6
Pristine bone
Pikner et al. 2009 Retrospective 602 421↗, – – – – 20 Edentulous and partially edentulous 20
[39], Sweden study 296↘ subjects 15
Pristine bone
Slotte et al. 2015 Prospective study 32 24 64 – 34.4 – 5 Partially edentulous subjects in the 71
[42], Sweden / posterior mandible requiring 3 or 4
Norway implants
Pristine bone
Simion et al. 2001 Retro-spective ​ ​ 50.4 – – – 5 Partially edentulous subjects ​
[37], Italy / USA study (median)

Augmented bone: ​ 6 6 ​ ​ ​ ​ ​ ​ 17
autograft
Augmented bone: ​ 32 32 ​ ​ ​ ​ ​ ​ 82
autograft
4

Juodzbalys et al. Prospective study 17 17 39.6 – 58.8 58.8 5 Partially edentulous subjects with 20
2007 [38], bony defects needing GBR
Lithuania
Augmented bone:
xenograft
Meijndert et al. 2017 Randomized ​ ​ ​ ​ ​ ​ 10 Partially edentulous subjects with ​
[43], The controlled study horizontal bone deficiency in the
Netherlands maxilla
Augmented bone: ​ 31 29 33.3 – 41.9 – ​ ​ 29
autograft
Augmented bone: ​ 31 24 34.6 – 51.6 – ​ ​ 24
autograft
Augmented bone: ​ 31 19 32.2 – 48.4 – ​ ​ 19
xenograft
Choi et al. 2023 [45], Retrospective 28 28 61.1 – 46.4 46.4 67.5 Partially edentulous subjects with ​
South Korea study ↑ 14.2 ↑ 10.3 bony defects needing GBR
(months)
Augmented bone: ​ – – ​ – ​ ​ 69.2 ​ 24
mixed grafts ↑ 10.5
(months)
Augmented bone: ​ 15 15 – – – – 65.8 ​ 24
xenograft ↑ 10.0
(months)
 S. Memenga-Nicksch et al. Journal of Dentistry 158 105808

bone sites [36,41,44]. Regarding the follow-up period, all included

studies reported MBL with a follow-up of at least 5 years, while three

Fixed restauration
studies additionally reported a follow-up period of 10 years or more [34,
restauration

restauration 39,43].
removable
Fixed and
Type of

Pristine bone sites

​

​
The number of subjects reported in the included studies with a
follow-up at least five years for implants placed in pristine bone sites

Palm Beach Gardens,

Brånemark System,

ranged from 24 [42] to 47 [35], with a total of 2730 dental implants.

oteborg, Sweden
Implant system

Regarding implant systems used in pristine bone sites, half of the studies
Nobel Biocare,

utilized Brånemark implants from Nobel Biocare, Go !teborg, Sweden

Biomet 3i,
ZimVie,

[34,36,39], and the remaining studies employed various other implant

USA

systems.
G!

The smoking history [34,35,41,44] and oral hygiene [35,36,42] of

​

​ participants were reported in half of the studies, while other potential

Implants

confounders, such as periodontitis [35] and diabetes mellitus [34], were

112

112

considered in only one study – each at the time of follow-up. No study

(n)

41

58

55
​

addressed all the investigated potential confounders. Further details are

provided in Table 2.
Subjects requiring two implants: one
with bony defects needing GBR and

posterior mandible requiring 2 or 3

Partially edentulous subjects in the

The included studies reporting a wide variation in MBL values. After

5 years of loading, MBL ranged from 0.41 ↑ 0.78 mm [35] to 1.73 ↑
one fully surrounded by bone

0.70 mm [36] in pristine bone sites. Long-term data beyond 5 years was
Subject’s characteristics

SD: standard deviation; USA: United States of America; GBR: guided bone regeneration; ↗: data for 5 year follow-up; ↘: data for 10 year follow-up.

limited, with only two studies reporting MBL at 10 years (range: 0.6 ↑
0.59 mm [34] to 0.93 ↑ 0.91 mm [39]). Table 3 presents MBL values at
implant and subject level for included studies.
implants

Augmented bone sites

The total number of subjects with implants in augmented bone sites
ranged from 15 [45] to 66 [36], with a total of 434 dental implants
​

reported at follow-up. The most common augmentation material was

Follow-up

xenogeneic bone substitute [36,38,41,44,45], most studies using bovine

(years)

bone substitutes, and only one study using porcine bone substitute [45].
Two studies reported the use of autogenous bone [37,43], while only
5

5
​

one study used a mixture of xenogeneic bone substituted and autoge-

Follow-

nous bone materials [45]. Barrier membranes, such as native collagen

membrane [36,38,41,43–45] or expanded polytetrafluoroethylene
up

–
​

​
​

(e-PTFE) [36,37], were frequently employed to enhance bone regener-

ation in these sites. The majority of studies included in this systematic
Base-
Male

25.3

50.0

23.3
line
(%)

review reported an augmentation procedure performed simultaneously

​

​
​

with implant placement [36–38,45].

Regarding implant systems in augmented bone sites, various implant
Follow-

systems were used. Two studies used Brånemark implants from Nobel
up
Mean age ± SD

Biocare, Go !teborg, Sweden [36,37], while all other studies employed

​

​
​

different implant systems [38,41,43–45]. Regarding confounders such

Base-line

as smoking [41,44,45], fewer studies have investigated these potential

(years)

56.1

factors, while oral hygiene was reported in 66.7 % of the included

55

56

studies [36–38,43].
​

​
​

After five years of loading the MBL ranged from 0.39 ↑ 0.54 mm
Follow-

[45] to 2.21 ↑ 1.26 mm [36]. Further details on MBL values at the

up

66

25

25

implant and subject levels in the included studies are shown in Table 3.
​

​
​
Subjects

Base-

Meta-Analyses
line
(n)

75

30

30
​

​
​

Marginal bone loss in pristine bone sites

Prospective study

controlled study

Five studies [34–36,39,42] were included in the meta-analysis model

Study design

for the mean MBL at implant level after 5 years of loading for dental
Randomized

implants placed in pristine bone sites. The overall estimated mean MBL
for implants placed in pristine bone sites after 5 years of loading was
0.79 mm (95 % confidence interval: 0.32–1.26 mm, Fig. 2- data from
​

random-effect model). Substantial heterogeneity was found between

Zitzmann et al. 2001
Table 1 (continued )

studies (p ω 0.0001, I² ↔ 99.2 %).

[36], Switzerland

Felice et al. 2018

Augmented bone:

Augmented bone:

Augmented bone:
Felice et al. 2014
Study, country

Pristine bone

Pristine bone
[41], Italy

[44], Italy

Marginal bone loss in augmented bone sites

xenograft

xenograft

xenograft

Meta-analysis was conducted for the mean MBL at the implant level
(Group)

after 5 years of loading for dental implants placed in augmented bone

&

sites, using data from three studies [36–38]. Several studies included in

5
S. Memenga-Nicksch et al. Journal of Dentistry 158 105808

Table 2
Characteristics of augmentation procedures and confounding factors in included studies.
Study, Typ of Timepoint of Augmentation Type of membrane Confounding factors
(Group) augmentation augmentation material
Subjects Subjects Subjects with Oral
with with periodontitis hygiene
smoking diabetes history (%)
history (%) mellitus
(%)

Wennstr! om – – – – 33.3 – 100 5.3

et al. 2004 ↑ 16.5≃
[35]
Pristine bone
Jemt and – – – – 61.8 2.6 – –
Johansson
2006 [34]
Pristine bone
Pikner et al. – – – – – – – –
2009 [39]
Pristine bone
Slotte et al. – – – – – – – 13.3≃
2015 [42]
Pristine bone
Simion et al. Vertical ridge Simultaneous with ​ Titanium-reinforced e- – – – ​
2001 [37] augmentation implant placement PTFE; Gore- Tex, W.L.
or delayed Gore/Implant
Innovations, West
Palm Beach, USA
Augmented ​ ​ Blood cloth ​ ​ ​ ​ 0.23
bone: ↑ 0.42↗
autograft
Augmented ​ ​ Autogenous bone ​ ​ ​ ​ 0.317
bone: chips ↑ 0.533↗
autograft
Juodzbalys GBR Simultaneous with Bovine bone Native collagen – – – 0.75
et al. 2007 implant placement substitute; BioOss, membrane; BioGide, ↑ 0.55↗
[38] Geistlich Geistlich Biomaterials,
Augmented Biomaterials, Wolhusen, Switzerland
bone: Wolhusen,
xenograft Switzerland
Meijndert et al. Vertical ridge Before implant ​ ​ – – – ​
2017 [43] augmentation placement
Augmented ​ ​ Autogenous chin – ​ ​ ​ 0↘
bone: bone
autograft
Augmented ​ ​ Autogenous chin Native collagen ​ ​ ​ 0↘
bone: bone membrane; BioGide,
autograft Geistlich Biomaterials,
Wolhusen, Switzerland
Augmented ​ ​ Bovine bone Native collagen ​ ​ ​ 0↘
bone: substitute; BioOss, membrane; BioGide,
xenograft Geistlich Geistlich Biomaterials,
Biomaterials, Wolhusen, Switzerland
Wolhusen,
Switzerland
Choi et al. 2023 Horizontal ridge Simultaneous with ​ ​ 10.7 – – –
[45] augmentation implant placement
Augmented ​ ​ Porcine bone Native collagen ​ ​ ​ ​
bone: mixed substitute; The Graft, membrane (in 20 sites)
grafts Purgo Biologics,
Seongnam, South
Korea / Autogenous
tooth bone graft
materials;
Korea Tooth Bank,
Seoul, South Korea /
Allograft
Augmented ​ ​ Porcine bone Native collagen ​ ​ ​ ​
bone: substitute; The Graft, membrane (in 12 sites)
xenograft Purgo Biologics,
Seongnam, South
Korea
Zitzmann et al. ​ ​ ​ ​ – – – 15.53≃
2001 [36]
Augmented GBR Simultaneous with Bovine bone Native collagen ​ ​ ​ 11.31≃
bone: implant placement substitute; BioOss, membrane; BioGide,
xenograft Geistlich
(continued on next page)

6
S. Memenga-Nicksch et al. Journal of Dentistry 158 105808

Table 2 (continued )
Study, Typ of Timepoint of Augmentation Type of membrane Confounding factors
(Group) augmentation augmentation material
Subjects Subjects Subjects with Oral
with with periodontitis hygiene
smoking diabetes history (%)
history (%) mellitus
(%)

Biomaterials, Geistlich Biomaterials,

Wolhusen, Wolhusen, Switzerland
Switzerland
Augmented GBR Simultaneous with Bovine bone Titanium-reinforced e- ​ ​ ​ 17.48≃
bone: implant placement substitute; BioOss, PTFE; Gore- Tex, W.L.
xenograft Geistlich Gore/Implant
Biomaterials, Innovations, West
Wolhusen, Palm Beach, USA
Switzerland
Pristine bone – – – – ​ ​ ​ 18.17≃
Felice et al. ​ ​ ​ ​ ​ – – –
2014 [41]
&
Felice et al.
2018 [44]
Augmented Vertical ridge Before implant Bovine bone Native collagen 36.7 ​ ​ ​
bone: augmentation placement substitute; BioOss, membrane; BioGide,
xenograft Geistlich Geistlich Biomaterials,
Biomaterials, Wolhusen, Switzerland
Wolhusen,
Switzerland
Pristine bone – – – – 40.0 ​ ​ ​

GBR: guided bone regeneration; USA: United States of America; e-PTFE: expanded polytetrafluoroethylene; ↗: mean modified plaque index [62] values with standard
deviation; ≃: mean plaque values with standard deviation (%); ↘: modified plaque index [62].

this systematic review were excluded from meta-analysis due to the lack greater initial mean MBL within the first five years, followed by stabi-
of comparability between the types of bone augmentations used. The lization. Similar trends were observed at the subject level. The review
estimated mean MBL for implants placed in augmented bone sites after 5 indicated, that the type of augmentation material influenced mean MBL,
years of loading was found to be 1.90 mm (95 % confidence interval: and sites where simultaneous augmentation during implantation was
1.73–2.07 mm, Fig. 3- data from random-effect model). A considerable performed generally exhibited greater bone loss. Autogenius bone
level of heterogeneity was observed among the included studies (p ↔ grafts, considered the gold standard due to their osteogenetic potential
0.0322, I² ↔ 65.9 %). [46,47]. Our review indicate that autograft materials are associated with
reduced MBL compared to xenografts.
The findings of the meta-analyses suggest a lower mean MBL in
Quality assessment
implants placed in pristine bone sites (0.79 mm) compared to
augmented bone sites (1.90 mm) after 5 years of loading at implant
Quality assessment for included non-randomized and non-
level, which may be attributed to differences in bone remodeling dy-
interventional studies revealed a overall good quality (Supplemental
namics after the augmentation procedure [48]. These findings are in
Table S3). Regarding the quality of included RCTs all included RCTs had
accordance with findings from clinical studies showing, that augmented
some concerns. Further information are provided in Supplemental
bone exhibits greater mean MBL [15,16,49,50]. However, a statistically
Figure S1.
comparison (pristine bone sites vs. augmented bone sites) was not
possible due to the lack of available data. Results should therefore be
Certainty of evidence interpreted with caution. Nevertheless, recent studies have reported
comparable long-term outcomes between both groups, particularly
The quality of evidence regarding the mean bone loss after a 5-year when modern bone augmentation techniques and materials are used
follow-up at implant level in pristine bone and augmented bone sites, is [13,51].
presented in Table 4. The overall evidence has been downgraded to a Two-dimensional radiographic techniques are commonly used for
very low level. assessing mean MBL due to their accessibility, and lower costs [52,53].
Only studies reporting parallel technique for radiographic examinations
Discussion were included in this systematic review. However, this two-dimensional
method is limited by its inability to visualize buccal and lingual bone
This systematic review and meta-analysis analyzed the mean MBL in dehiscences [54,55]. Studies have shown, that buccal bone loss, and the
dental implants placed in pristine and augmented bone sites with a shape of bony defects, often remains undetected in two-dimensional
follow-up period of at least five years. imaging due to superimposition of structures [54,56,57]. Cone beam
According to the results of this systematic review, at the implant computed tomography (CBCT) has been proposed as an alternative for
level, the mean MBL in pristine bone sites ranged between 0.41 mm and detecting peri‑implant bone loss, as it provides three-dimensional
1.73 mm after five years, whereas in augmented bone sites, it varied visualization and allows for accurate assessment of buccal and lingual
between 0.39 mm and 2.21 mm. After ten years, mean MBL in pristine bone sites [57]. However, due to concerns regarding radiation exposure,
bone sites ranged from 0.6 mm to 0.93 mm, while augmented bone sites scattering artifacts that may significantly compromise image reliability,
exhibited values between 0.16 mm and 0.74 mm. These findings suggest and the associated costs, CBCT is not routinely employed for longitu-
that pristine bone sites undergo a more consistent yet lower level of dinal assessments of MBL in clinical practice [53,56]. Clinical
resorption over time, while augmented bone sites tend to experience

7
S. Memenga-Nicksch et al. Journal of Dentistry 158 105808

Table 3 systematic search strategy in four electronic databases: PubMed,

Marginal bone loss after 5 and 10 years of loading. Embase, Scopus, and Web of Science. In addition, gray literature was
Study, MBL from baseline in MBL from baseline in mm ± examined through manual searches. Another strength is the inclusion of
(Group) mm ± SD after 5 years SD after 10 years of loading long-term clinical studies, as only those with a minimum follow-up
of loading period of five years were considered. The focus on long-term data en-
Implant Subject Implant Subject hances the clinical relevance of the findings, as peri‑implant bone
level level level level remodeling is a progressive process that continues well beyond the
Wennstr! om et al. initial healing phase [58]. In addition, the GRADE system was appli-
2004 [35] 0.41 ↑ 0.41 ↑ – – cated to assess the quality of evidence. However, the overall certainty of
Pristine bone 1.01 0.78 evidence was downgraded to a very low level due to substantial het-
Jemt and Johansson
erogeneity of the available data. The risk of bias was evaluated using two
2006 [34] 0.5 ↑ 0.47 – 0.6 ↑ 0.59 –
Pristine bone established tools: NOS for non-randomized studies and RoB 2 for ran-
Pikner et al. 2009 domized controlled trials.
[39] 0.79 ↑ 0.78 ↑ 0.93 ↑ 0.91 0.95 ↑ 0.66 Unfortunately, meta-analyses were only possible for data relating to
Pristine bone 0.75 0.55 mean MBL after five years at the implant level. Significant heterogeneity
Slotte et al. 2015
[42] 0.53 ↑ – – –
was observed in both meta-analyses, with an I² value of 99.2 % for
Pristine bone 0.08 pristine bone sites and 65.9 % for augmented bone sites, which limits the
Simion et al. 2001 ​ ​ ​ ​ generalizability of the findings.
[37] Patient-related factors significantly influence mean MBL outcomes
Augmented bone: 1.71 ↑ – – –
and contribute to the heterogeneity observed in the meta-analyses.
autograft 0.97
Juodzbalys et al. Clinical parameters such as probing depths and mucosal inflammation,
2007 [38] 2.01 ↑ – – – which are essential for accurately assessing peri‑implantitis, were not
Augmented bone: 0.35 considered in this systematic review, limiting its clinical interpretability.
xenograft The included studies encompassed a wide range of patient populations
Meijndert et al. 2017 ​ ​ ​ ​
[43]
with varying risk factors known to affect peri‑implant bone stability.
Augmented bone: – – 0.16 ↑ 0.91 0.16 ↑ 0.91 These factors include smoking, a history of periodontitis, systemic
autograft (mesial) (mesial) conditions such as diabetes mellitus, and differences in oral hygiene
0.28 ↑ 1.23 0.28 ↑ 1.23 procedures or maintenance care [59–61]. While some studies controlled
(distal) (distal)
for these confounding variables, others did not, making it difficult to
Augmented bone: – – 0.74 ↑ 1.20 0.74 ↑ 1.20
autograft (mesial) (mesial) assess the effect of bone augmentation procedures on long-term mean
0.24↑1.05 0.24↑1.05 MBL. Unfortunately, risk factors with limited evidence for peri‑im-
(distal) (distal) plantitis and peri‑implant bone loss, such as inadequate keratinized
Augmented bone: – – 0.64 ↑ 1.46 0.64 ↑ 1.46 mucosa, occlusal overload, presence of submucosal cement, and implant
xenograft (mesial) (mesial)
0.37 ↑ 1.47 0.37 ↑ 1.47
position or a superstructure that does not allow effective plaque control,
(distal) (distal) are not assessed in a standardized way in the included studies [52,59].
Choi et al. 2023 [45] ​ ​ ​ ​ These factors could also not be taken into account in this systematic
Augmented bone: 0.39 ↑ – – – review.
mixed grafts 0.54
Overall, the findings of this systematic review indicate that dental
Augmented bone: 0.41 ↑ – – –
xenograft 0.35 implants placed in pristine bone generally exhibit lower mean MBL
Zitzmann et al. 2001 ​ ​ ​ ​ compared to those placed in augmented bone sites.
[36]
Augmented bone: 1.83 ↑ – – – Conclusions
xenograft 0.63
Augmented bone: 2.21 ↑ – – –
xenograft 1.26 Future research should focus on standardizing augmentation pro-
Pristine bone 1.73 ↑ – – – tocols, ensuring consistent radiographic assessment methods and clin-
0.70 ical examination, and conducting long-term controlled studies with
Felice et al. 2014 ​ ​ ​ ​
larger patient cohorts that account for patient-specific risk factors
[41]
& associated with peri‑implantitis. Furthermore, there is a need to estab-
Felice et al. 2018 lish the routine assessment of clinical parameters in implant studies, as
[44] marginal bone loss alone is no longer sufficient to define treatment
Augmented bone: – 2.34 ↑ – – success.
xenograft 0.75
Pristine bone – 1.49 ↑ – –
0.40 Authors
MBL: marginal bone loss; SD: standard deviation.
Dr. Sonja Memenga-Nicksch
Hannover Medical School, Department of Conservative Dentistry,
Periodontology and Preventive Dentistry, Hannover, Germany
measurements such as probing depths and bleeding on probing are Dr. Felix Marschner; ORCID iD: 0009-0009-6913-5855
essential for detecting early peri‑implantitis, including buccal/lingual University Medical Center Go !ttingen, Department of Preventive
bone dehiscences in MBL. Further studies should apply standardized Dentistry, Periodontology and Cariology, Go
!ttingen, Germany
probing protocols with fixed reference points alongside radiographs to Nele Henrike Thomas, M. Sc.; ORCID iD: 0000–0003–1133–0171
reliably assess and compare MBL across treatments. Unfortunately, the Hannover Medical School, Institute for Biostatistics, Hannover,
studies in this review lacked such comparable clinical data, limiting Germany
conclusions on peri‑implant tissue changes. Dennis Holzwart, M. Sc.; ORCID iD: 0009–0005–3912–0450
A key strength of this systematic review and meta-analysis is the Hannover Medical School, Institute for Biostatistics, Hannover,
Germany