Wound Healing Complications Following

Guided Bone Regeneration for Ridge Augmentation:

A Systematic Review and Meta-Analysis
Glendale Lim, DDS, MS1/Guo-Hao Lin, DDS, MS2/Alberto Monje, DDS, MS3/
Hsun-Liang Chan, DDS, MS4/Hom-Lay Wang, DDS, MSD, PhD5

Purpose: The rate of developing soft tissue complications that accompany guided bone regeneration
(GBR) procedures varies widely, from 0% to 45%. The present review was conducted to investigate the rate
for resorbable versus nonresorbable membranes and the timing of soft tissue complications. Materials
and Methods: Electronic and manual literature searches were conducted by two independent reviewers
using several databases, including MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials, and
Cochrane Oral Health Group Trials Register, for articles published through July 2015, with no language
restriction. Articles were included if they were clinical trials aimed at demonstrating the incidence of soft
tissue complications following GBR procedures. Results: Overall, 21 and 15 articles were included in the
qualitative and quantitative synthesis, respectively. The weighted complication rate of the overall soft tissue
complications, including membrane exposure, soft tissue dehiscence, and acute infection/abscess, into
the calculation was 16.8% (95% CI = 10.6% to 25.4%). When considering the complication rate based on
membrane type used, resorbable membrane was associated with a weighted complication rate of 18.3%
(95% CI: 10.4% to 30.4%) and nonresorbable membrane with a rate of 17.6% (95% CI: 10.0% to 29.3%).
Moreover, soft tissue lesions were reported as early as 1 week and as late as 6 months based on the
included studies. Conclusion: Soft tissue complications after GBR are common (16.8%). Membrane type
did not appear to significantly affect the complication rate, based on the limited number of data retrieved in
this study. Technique sensitivity (ie, soft tissue management) may still be regarded as the main component
to avoid soft tissue complications and, hence, to influence the success of bone regenerative therapy. Int J
Oral Maxillofac Implants 2018;33:41–50. doi: 10.11607/jomi.5581

Keywords: alveolar ridge augmentation, guided bone regeneration, soft tissue complication,
systematic review

1Graduate Student, Department of Periodontics and Oral

Medicine, University of Michigan School of Dentistry,
Ann Arbor, Michigan, USA.
2 Assistant Professor, Department of Surgical Sciences,
Marquette University School of Dentistry, Milwaukee,
Wisconsin, USA.
3Resident and Research Fellow, Graduate Periodontics,
Department of Periodontics and Oral Medicine, University of
Michigan School of Dentistry, Ann Arbor, Michigan, USA.
4 Clinical Assistant Professor, Department of Periodontics and
Oral Medicine, University of Michigan School of Dentistry,
Ann Arbor, Michigan, USA.
5Professsor, Department of Periodontics and Oral Medicine,
University of Michigan School of Dentistry, Ann Arbor,
Michigan, USA.
Correspondence to: Dr Hsun-Liang Chan, Department of
Periodontics and Oral Medicine, University of Michigan School
of Dentistry, 1011 N University Avenue, Ann Arbor, MI 48109,
USA. Email: hlchan@umich.edu
©2018 by Quintessence Publishing Co Inc.
T he functional and esthetic outcomes of implant-
supported prostheses are mostly related to the
dimensions of the edentulous ridge. Moreover, in addi-
tion to implant width, approximately 2 mm of so-called
critical buccal bone is needed in the buccal flange to
avoid major physiologic bone resorption in the buc-
colingual plane, especially in the maxillary anterior
region.1 Therefore, in many cases, bone augmentation
procedures may be required simultaneously or prior to
implant placement. Several surgical techniques, such
as guided bone regeneration (GBR), distraction osteo-
genesis, and bone block grafts have been attempted
to increase bone volume and to properly withstand
implant function.1
The GBR procedure, applying particulate autog-
enous bone graft or bone substitutes and an oc-
cluding membrane, has shown to be a predicable
way to augment bone. This principle was initially
described for periodontal tissue regeneration as a

The International Journal of Oral & Maxillofacial Implants 41

© 2018 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Lim et al
‘compartmentalization’ approach to exclude undesired
cells (ie, epithelial cells and fibroblasts) from the tar-
geted area for regeneration.2 This procedure was first
applied by Hurley in an animal spinal fusion experi-
ment and later applied in oral surgery.3 The regenera-
tion procedure can promote bone defect healing after
6 to 10 months.4
There is strong clinical and histologic evidence of the
effectiveness and predictability of GBR in bone augmen-
tation of ridge deficiencies.4 Membrane type plays an
important role, since it determines the space needed
for osteogenic cells to repopulate for bone formation.5
For minor and moderate defects, the use of resorbable
membrane barriers has been advocated; when intend-
ed to augment severe horizontal defects or those with
a vertical component, nonresorbable membrane, such
as titanium-reinforced polytetrafluoroethylene (PTFE)
membrane, may present a better choice.5
Although GBR procedures have fewer complica-
tions when compared with distraction osteogenesis
or block grafting, complications associated with soft
tissue dehiscence remain a major concern. This factor
is of paramount importance, since these complica-
tions are associated with inferior hard tissue gain. In
the presence of soft tissue dehiscence and membrane
exposure, there is six times less bone formation com-
pared to nonexposed sites.6 The increased failure rate
for soft tissue dehiscence may be more apparent in
patients exhibiting risk factors, eg, systemic diseases,
smoking history, and anatomical abnormality.7
These complications challenge clinicians’ ability to
provide proper clinical intervention in a timely manner.
To date, no systematic review has focused on explor-
ing the complication rate based on membrane type
and timing of the first sign of soft tissue complications
following bone augmentation procedures. Therefore,
the aims of this review article were to identify the fre-
quency of soft tissue complications, influence of mem-
brane types, and the time point for the complications
to develop.
MATERIALS AND METHODS
Focused Question
The focused question was: What is the soft tissue com-
plication rate after GBR for horizontal ridge augmen-
tation utilizing resorbable and nonresorbable barrier
membranes?
PICO (Problem, Intervention, Comparison,
Outcome)
P: Maxillary or mandibular complete or partial eden-
tulous healthy subjects in need of dental implants
to restore oral function. I: GBR for horizontal ridge
42 Volume 33, Number 1, 2018
© 2018 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
augmentation to augment the bone width for ad-
equate 3D implant placement. C: Resorbable vs non-
resorbable barrier membranes with particulate bone
graft (xenogeneic, allogeneic, and/or autogenous
graft). O: (primary outcome)—Soft tissue complication
rate depending on barrier membrane type used; (sec-
ondary outcome)—timing of soft tissue complications.
Information Sources
Electronic and manual literature searches were con-
ducted by two independent reviewers (GL and AM)
in several databases, including MEDLINE, EMBASE,
Cochrane Central Register of Controlled Trials, and
Cochrane Oral Health Group Trials Register, for articles
up to June 2015 without language restriction. The re-
viewers independently extracted the data from stud-
ies. Publications that did not meet the inclusion criteria
were excluded. In cases of disagreement, consensus
was reached by discussion with a third reviewer (HLC).
Screening Process and Data Extraction
For the PubMed library, combinations of controlled
terms (MeSH and EMTREE) and keywords were used
whenever possible. In addition, other terms not in-
dexed as MeSH and filters were applied. As such,
the key terms used were: ((((“dental implants”[MeSH
Terms] OR “dental implantation, endosseous”[MeSH
Terms]) OR “dental implantation, endosseous”[MeSH
Terms]) AND “bone regeneration”[MeSH Terms]) OR
“bone substitutes”[MeSH Terms]) OR “periodontal
ligament”[MeSH Terms] AND (Clinical Trial[ptyp] AND
“humans”[MeSH Terms]).
A second broader screening was conducted to iden-
tify articles not adequately indexed with the primary
screening strategy: (“alveolar ridge augmentation”[all]
OR “ridge augmentation”[all] OR “guided bone
regeneration”[all]) AND (“dental implants”[mh] OR “ab-
sorbable implants”[mh] OR implant [tiab] OR implants
[tiab]) AND (“complication”[all] OR “complications”[all]).
For the EMBASE and Cochrane Libraries, the key
terms used were (Title, Abstract, Keywords): ‘dental
implant’ OR ‘endosseous implant’ AND ‘guided bone
regeneration’ OR ‘alveolar bone regeneration’ AND
‘complications’.
The screening in such databases was limited to
“clinical trials” AND “humans.” No restrictions in regard
to language were applied in any of the screening strat-
egies. In addition, an electronic screening of the grey
literature at the New York Academy of Medicine Grey
Literature Report (http://greylit.org) was conducted
as recommended by high standards for systematic re-
views (ie, Assessment of Multiple Systematic Reviews
[AMSTAR] guidelines.
Additionally, a manual search of periodontics-re-
lated journals, including Journal of Dental Research,

Journal of Clinical Periodontology,
Journal of Periodontology, and The
International Journal of Periodontics
and Restorative Dentistry, from Janu-
ary 2014 to July 2015, was performed
to ensure a thorough screening pro-
cess. References of included articles
were also screened to check all avail-
able articles.
Eligibility Criteria
Studies were included for the sys-
tematic review if they fulfilled the
following inclusion criteria: clinical
human prospective or retrospective
study, study participants of ≥ 10,
GBR procedure for ridge augmenta-
tion on partially edentulous ridges,
simultaneous or delayed implant
placement. Studies with a smaller
sample size, nonparticulated bone
graft material (ie, block graft), no
information regarding soft tissue
complication, augmentation proce-
dures other than GBR (distraction
osteogenesis, sinus elevation, etc),
or focus on vertical ridge augmenta-
tion were excluded from this review
study. Articles were screened first
based on their titles and abstracts for
the inclusion eligibility.
Data Extraction
Data from the selected articles were
independently extracted. Demo-
graphic information from each study
recorded included: the study design,
number of participants, bone graft
material used, membrane type se-
lected, flap design, surgical method,
initial bone width and height, soft tis-
sue complication rate, complication
type, timing of complication occur-
rence, and duration of follow-up pe-
riod after the augmentation surgery.
Data Analyses
The primary outcome was the in-
cidence and types of postsurgical
complications of the GBR technique.
The contribution of each article was
weighed. The risk ratio and relatively
weighed complication rate were
calculated using a computer pro-
gram (Comprehensive Meta-analysis
© 2018 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Lim et al
Records identified through Additional records identified
Identification
database searches through other sources
(n = 1,127) (n = 27)
Records after duplicates
removed
(n = 1,127)
Screening
Records screened Records excluded
(n = 485) (n = 464)
Full-text articles assessed for Full-text articles
eligibility excluded
Eligibility
(n = 25) (n = 4)
Studies included in qualitative
synthesis
(n = 21)
Included
Studies included in quantitative
synthesis (meta-analysis)
(n = 15)
Fig 1   PRISMA flowchart demonstrates the study inclusion selection process.
Version 2, Biostat). In addition to the overall incidence rate of complica-
tions, two subgroup analyses, the use of resorbable and nonresorbable
membranes, were also performed to analyze the complication rate us-
ing different types of membranes. The random-effects model was ap-
plied when performing meta-analyses to account for methodological
differences among studies. Forest plots were produced to graphically
represent weighted complication rate and 95% confidence interval
(CI) for the primary outcomes using the number of surgical sites in
each study as the analysis unit. Heterogeneity was assessed with the
chi-square test and the I2 test, which ranges between 0% and 100%
(lower values represent less heterogeneity). In addition, funnel plots
were used to assess the presence of publication bias. To assess poten-
tial confounding factors, meta-regression was performed to analyze
the effect of defect types (horizontal vs vertical ridge deficiency) on
soft tissue complications. The reporting of these meta-analyses ad-
hered to the PRISMA (Preferred Reporting Items for Systematic Review
and Meta-Analyses) statement.8
Risk of Bias Assessment
The criteria used to assess the quality of the selected randomized
control trials (RCTs) were modified from the randomized clinical trial
checklist of the Cochrane Center, which provided guidelines for the
following parameters: sequence generation, allocation concealment,
masking of the examiner, addressing incomplete outcome data, and
free of selective outcome reporting. The degree of bias was catego-
rized as: low risk if all the criteria were met, moderate risk when only
The International Journal of Oral & Maxillofacial Implants 43
Lim et al

Table 1   Summary of Excluded Studies Table 2   Features of the Included Studies
Study Reason for exclusion Study No. of cases Graft
Study design (age range) material
Merli et al (2010)11 No data reported
Buser et al (1990)16 PCS 12 (18–54 y) AUTO
Buser et al (1996)9 Use of block graft
Lee et al (2013)10 Use of block graft
Mayfield et al PCS 11 (55–90 y) AUTO
Proussaefs and Lozada Use of block graft (1997)20
(2006)12
Parodi et al (1998)24 CR 16 (22–65 y) AUTO
Rabelo et al (2010)13 Use of block graft
Roccuzzo et al (2004)14 Use of block graft
Becker et al (1999)15 PMultiS 44 (18–83 y) AUTO
Von Arx and Kurt PCS 15 (17–66 y) AUTO
(1999)27
one criterion was missing, and high risk if two or more
criteria were missing. Two reviewers (G-HL and GL) as- Carpio et al (2000)18 RCT 48 (18+ y) COMBO
sessed all the included articles independently.

Merli et al (2006)21 RCS 19 (30–70 y) AUTO

RESULTS
Llambés et al PCS 11 (39–61 y) AUTO
Data Extraction (2007)19
A total of 1,127 records were identified through the Merli et al (2007)22 RCT 22 (29–69 y) AUTO
electronic and manual search after exclusion of du-
plicates (Fig 1). Of them, 485 were screened based
Park and Wang RCT 25 (28–71 y) ALLO
on title/abstract and 21 were selected for full-text as-
(2007)23
sessment and, thus, to be included in the qualitative
Torres et al (2010)26 RCT 30 (48–76 y) XENO
evaluation. Finally, 6 studies9–14 were excluded from
the quantitative analysis due to lack of data (Table 1), Urban et al (2011)5 CR 22 (30–63 y) COMBO
and 15 studies4,5,15–27 reporting the incidence of soft Annibali et al (2012)4 CR 17 (NA) COMBO
tissue complications after GBR were meta-analyzed.
The kappa value between the two reviewers was 0.88
for titles and abstracts evaluation and 0.87 for full-text
Cardaropoli et al PCS 20 (34–80 y) XENO
evaluation. The main features of the included studies (2013)17
are summarized in Table 2. Poli et al (2014)25 CR 13 (NA) COMBO

Features of Included Studies

Study Design and Patient Features.
Four RCTs,18,22,23,26 1 prospective multicenter study,15
5 prospective clinical studies,16,17,19,20,27 1 retrospec-
tive cohort study,21 and 4 case reports4,5,24,25 were
included. All but 2 studies reported the age range
(18–90 years) and gender of the participants.4,25 In
addition, all 15 studies reported the follow-up period
(between 4 weeks to 5 years) after the ridge augmen-
tation procedure and/or implant placement. Initial
ridge width and height were examined in all stud-
ies; however, the mean width and/or height were
reported in only 5 studies (2–7 mm4; < 5.5 mm16;
3.63–4.36 mm18; < 4 mm20; ≤ 3 mm26). The other stud-
ies stated only that the initial ridge was considered
clinically deficient. All 15 studies excluded patients
with systemic or local conditions that are consid-
ered contraindications for intraoral surgery, but only
4 studies specifically stated the conditions (patients
with mitral valve prolapse, rheumatic heart disease, or
other valvular diseases15; diabetes, hyperparathyroid-
ism, osteoporosis, severe liver or kidney condition, ac-
tive sinusitis, cancer and use of immunosuppressive
agents or corticosteroids18; history of irradiation in
the head and neck area, uncontrolled diabetes, sub-
stance abusers21,22; excessive alcohol consumptions5).
Six of the studies reported the selection criteria based
on smoking status with specific restriction (more
than 10 cigarettes per day4; 2 or more packs of ciga-
rettes per day18; more than 20 cigarettes per day21,22;
smokers25; current smokers5). Five studies reported
specific exclusion criteria based on oral hygiene and
periodontal disease status (full-mouth plaque score

44 Volume 33, Number 1, 2018

© 2018 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Lim et al

Membrane Complication Complication

type Initial bone width Initial bone height rate (%) timing Complication type
N < 5.5 mm Sufficient 35.7 3–4 mo Acute infection,
dehiscence
R < 4 mm Sufficient vertical bone height 0 NA NA
(≥ 10 mm)
R < 5 mm NA 31.3 Paresthesia, 0–4 Paresthesia, post-
(mean 3 ± 0.5 mm) mo; hemorrhages, op hemorrhage,
12 h membrane exposure
N Deficiency Deficiency 0 NA NA
Ti-mesh Dehiscence or fenestrated Mean vertical size of the peri- 5.3 3 mo Soft tissue
defects implant bone defect was 6.5 dehiscence,
mm (4–12 mm) membrane exposure
N, R Collagen group = 3.63 ± Collagen group = 4.39 ± 0.49 R: 21.7 N: 16 6 mo Soft tissue
0.28 mm; ePTFE group = mm; ePTFE group = 4.18 ± dehiscence,
4.36 ± 0.40 mm (P < .05) 0.39 mm (P < .05) membrane exposure
N, R Deficiency Deficiency R: 25 N: 7.7 2–4 wk Abscess, soft tissue
dehiscence
R Deficiency Deficiency 13.3 3 wk Soft tissue
dehiscence
N, R Deficiency 2.9 mm in the resorbable R: 36.4 1–3 wk Abscess, soft tissue
group, 2.7 mm in the N: 45.5 dehiscence
nonresorbable group
R Dehiscence defects Deficiency 34.5 2 wk Membrane exposure

Ti-mesh < 4 mm ≤ 7 mm 20 1 mo Membrane exposure

R Knife-edge ridge Sufficient 0 NA NA
N, R Group A (H-GBR), Group C (V-GBR), 4.28 ± 1.23 R: 0 Membrane Membrane exposure,
3.80 ± 2.64 mm; Group B mm (range 2–6.5 mm) N: 23.5 exposure, 1–2 mo; abscess
(ridge expansion), late exposure, 4
4.63 ± 1.30 mm mo; abscess, 6 mo
R Deficiency Mean = 4.25 ± 1.34 mm 0 NA NA

Ti-mesh Deficiency Deficiency 7.7 4 mo Membrane exposure,

soft tissue
dehiscence
RCT = randomized controlled trial; PMultiS = prospective multicenter clinical study; PCS = prospective clinical study; RCS = retrospective cohort study;
CR = case report; NA = not available/applicable; AUTO = autogenous bone graft; ALLO = allograft; XENO = xenograft; COMBO = combination graft; N =
nonresorbable membrane; R = resorbable membrane; H-GBR = horizontal guided bone regeneration; V-GBR = vertical guided bone regeneration.

and full-mouth bleeding score of greater than 25%4;
uncontrolled periodontal disease19; poor oral hygiene
and poor motivation for improvement21,22; poor oral
hygiene, uncontrolled periodontal disease5).
Types of Membrane and Bone Graft Materi-
als Used. Six studies used only resorbable mem-
brane (Bio-Gide [Geistlich] collagen membrane;
copolymer of polyglycolide and polylactide [Reso-
lut, Gore]; synthetic bioabsorbable glycolide and
trimethylene carbonate copolymer fiber [Gore
Resolut Adapt LT Regenerative membrane]; bovine
collagen membrane [BioMend Extend, Zimmer
Biomet]; bioabsorbable collagen membrane
[Paroguide]). 5,17,19,20,23,24 Four studies used only
nonresorbable membrane (titanium-reinforced ex-
panded polytetrafluoroethylene [Gore-Tex]; PTFE;
ePTFE; titanium micromesh).16,25–27 As for the other
five studies, the study participants were randomly
assigned to receive either resorbable or nonresorb-
able membrane.4,15,18,21,22 In regard to the bone
grafting material used, eight studies used intraoral
autogenous bone graft.15,16,19–22,24,27 One study
combined the autograft with a layer of demineral-
ized freeze-dried bone allograft (DFDBA).4 Three
studies mixed the autogenous graft with anorganic
bovine bone mineral.5,18,25 One study used mineral-

The International Journal of Oral & Maxillofacial Implants 45

© 2018 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Lim et al

Fig 2  Forest plot presenting es-

Event Lower Upper
timated incidence of soft tissue
N (%) limit limit
complications for all included stud-
Buser et al16 14 35.7 15.7 62.4 ies. The weighted mean incidence
Mayfield et al20 17 0 0 32.2 is 16.8% (95% CI: 10.6% to 25.4%).

Parodi et al24 16 31.3 13.7 56.7

Becker et al15 55 0 0 12.7
Von Arx and Kurt 27 19 5.3 0 29.4
Carpio et al18 48 18.8 10.1 32.3
Merli et al21 21 14.3 4.7 36.2
Llambés et al19 15 13.3 3.3 40.5
Merli et al22 22 40.9 22.8 61.8
Park and Wang 23 29 40 24.1 58.4
Torres et al26 30 20 9.3 37.9
Urban et al5 25 0 0 24.4
Annibali et al4 30 13 4.9 30.2
Cardaropoli et al17 35 0 0 18.7
Poli et al25 13 7.7 1.1 39.1
All 389 16.8 10.6 25.4
–1.0 0 1.0

ized human allograft. 23 Two studies used xenograft,
ie, anorganic bovine bone.17,26
Incidence of Developing Soft Tissue Complica-
tions. Of the included studies, three studies report-
ed 0% soft tissue complications following the GBR
procedure.5,15,20 All the other studies reported some
forms of complications occurring from 12 hours up
to 6 months following the surgery. In the 20 reported
complication cases, membrane exposure (35%) and
soft tissue dehiscence (35%) were the most com-
mon complication types following GBR. Soft tissue
dehiscence was determined clinically with signs of
membrane- or implant-perforated oral mucosa dur-
ing the follow-up period. This was considered a mi-
nor biologic complication, since the sites showed
gradual improvement without treatment or with
chlorhexidine application. Membrane exposure was
determined clinically when the overlying soft tissue
showed signs of significantly large dehiscence dur-
ing the follow-up period. This was considered a ma-
jor complication, since abscess was common, which
required immediate attention or treatment. Mem-
brane exposure was reported as early as 2 weeks
subsequent to the surgery, while soft tissue dehis-
cence was reported as early as 1 week following the
surgery.21 Poli et al documented one incidence of
late membrane exposure as the result of soft tissue
dehiscence.25 Seven studies reported that soft tissue
dehiscence occurred as early as 1 week and as late
as 6 months following surgery.16,18,19,21,22,25,27 Stud-
ies also reported signs of infections and abscesses
in addition to the membrane exposure.4,16,21,22 The
weighted complication rate of the overall soft tis-
sue complications, including membrane exposure,
soft tissue dehiscence, and acute infection/abscess
into the calculation, was 16.5% (95% CI = 10.7% to
24.5%) (Fig 2). When considering the complication
rate based on membrane material, resorbable mem-
brane had a comparable weighted complication rate
to nonresorbable membrane: 18.3% (95% CI = 10.4%
to 30.4%, Fig 3) and 17.6% (95% CI = 10.0% to 29.3%,
Fig 4), respectively. Funnel plots for evaluating pub-
lication bias are shown in Supplemental Figs S-1 to
S-3 (found at the end of this article). Meta-regression
performed to identify the potential influence of de-
fect types on soft tissue complications indicated no
statistically significant difference in complication in-
cidence in terms of defect type treated (P = .30).
Risk of Bias Assessment. The results of the risk of
bias assessment for the four included RCTs are sum-
marized in Table 3. Two studies had a low risk of bias,
one had a moderate risk of bias, and one had a high
risk of bias.
Heterogeneity Test. The heterogeneity test showed
a low value of heterogeneity among studies for over-
all analysis (I2 = 21.47 and P value for chi-square test =
.45), resorbable membrane (I2 = 21.55 and P value for
chi-square test = .24), and nonresorbable membrane
(I2 = 19.38 and P value for chi-square test = .27) groups.
However, funnel plots were unsymmetrical, indicating
the risk of potential publication bias (Figs S-1 to S-3,
representing overall analysis, resorbable membrane
sub-analysis, and nonresorbable membrane sub-
analysis, respectively).

46 Volume 33, Number 1, 2018

© 2018 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Lim et al

Fig 3  Forest plot presenting es-

Event Lower Upper
timated incidence of soft tissue
N (%) limit limit
complications for included studies
utilizing resorbable membranes. Mayfield et al20 17 0 0 32.2
The weighted mean incidence is
Parodi et al24 16 31.3 13.7 56.7
18.3% (95% CI: 10.4% to 30.4%).
Carpio et al18 23 21.7 9.3 42.8
Merli et al21 8 25.0 6.3 62.3
Llambés et al19 15 13.3 3.3 40.5
Merli et al22 11 36.4 14.4 66.2
Park and Wang 23 29 34.5 19.7 53.1
Urban et al5 25 0 0 24.4
Annibali et al4 13 0 0 38.4
Cardaropoli et al17 35 0 0 18.7
All 192 18.3 10.4 30.4
–1.0 0 1.0

Fig 4  Forest plot presenting es-

Event Lower Upper
timated incidence of soft tissue
N (%) limit limit
complications for included studies
utilizing nonresorbable membranes. Buser et al16 14 35.7 15.7 62.4
The weighted mean incidence is
Becker et al15 55 0 0 12.7
17.6% (95% CI: 10.4% to 29.3%).
Von Arx and Kurt 27 19 5.3 0.7 29.4
Carpio et al18 25 16.0 6.1 35.7
Merli et al21 13 7.7 1.1 39.1
Merli et al22 11 45.5 20.3 73.2
Torres et al26 30 20.0 9.3 37.9
Annibali et al4 17 23.5 9.1 48.5
Poli et al25 13 7.7 1.1 39.1
All 197 17.6 10.0 29.3
–1.0 0 1.0

Table 3   Risk Assessment of Publication Bias for the Included RCTs
Carpio et al Merli et al Park and Wang Torres et al
Criteriaa (2000)18 (2007)22 (2007)23 (2010)26
Sequence generation Yes Yes Yes Yes
Allocation concealment method Yes Yes ? Yes
Examiner masked No Yes ? Yes
All patients accounted for at end of study Yes Yes Yes Yes
Incomplete outcome data adequately addressed NA NA NA NA
Free of suggestion of selective outcome reporting Yes Yes Yes Yes
Estimated potential risk of bias Moderate Low High Low
a According
to Higgins JPT, Green S (eds). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The
Cochrane Collaboration, 2011. Available from http://handbook.cochrane.org.
NA = Not applicable.

DISCUSSION nonresorbable membrane for GBR horizontal ridge

Principal Findings
Data from the present literature demonstrated a
similar frequency of soft tissue complications among
different types of membrane barriers. The use of
augmentation was associated with a slightly lower
incidence of soft tissue complications (ie, tissue de-
hiscence, acute infection, abscess, membrane expo-
sure) compared to resorbable membrane; however,
the weighted mean incidence (17.6% vs 18.3%) was

The International Journal of Oral & Maxillofacial Implants 47

© 2018 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Lim et al
not different. Additionally, no statistical significance
was found when comparing the membrane exposure
rate of the earlier studies (1990 to 2000) with the later
studies (2006 to 2014) (15.7% vs 17.8%, respectively).
Soft tissue lesions such as abscess or soft tissue dehis-
cence could occur as early as 1 week, and membrane
exposure could occur months following the bone
augmentation surgery.
Clinical Implications
Out of the 15 included studies, only 3 studies per-
formed GBR prior to implant placement.5,16,23 One of
the 2 studies that used resorbable membrane report-
ed membrane exposure at the 2-week follow-up.23 The
study that used nonresorbable membrane reported
acute infection and dehiscence after 3 to 4 months
that resulted in an insufficient ridge for implant place-
ment.16 As for the 3 studies that used titanium mesh
for alveolar ridge augmentation, membrane exposure
occurred at 1, 3, and 4 months and resulted in no or
partial bone augmentation.25–27 Although the current
study found a slightly higher soft tissue complication
incidence with resorbable membrane, it is worth not-
ing that the majority of the studies reported complete
healing (unassisted, with chlorhexidine/saline irriga-
tion and/or systemic antibiotics) of the soft tissue de-
hiscence sites that had previous membrane exposure
without significant negative impact on the bone aug-
mentation procedure.
The association between soft tissue complica-
tion occurrence and membrane type may indicate
that when possible (with small defects, ≤ 3 mm, and
moderate defects, 4 to 6 mm), the use of resorbable
barriers is preferred. Nevertheless, these are not free
of complications. Passive soft tissue primary wound
closure is believed to aid in stabilizing graft material
and increase the success of bone augmentation pro-
cedures.7 This has to be accomplished by adequate
flap design. For instance, Park et al, in an interesting in
vivo study, showed that by releasing one vertical inci-
sion, 1.1 mm could be obtained; two vertical incisions
could advance 1 mm more; and if more coronal ad-
vancement is needed, the release of the periosteal in-
cision may provide 5.5 mm.28 It is crucial to note that
at least 3+ mm of coronal advancement beyond the
crestal incision are needed to attain tension-free flap
closure.29 Thereby, operator skill to attain tension-free
closure remains the most important factor to achieve
successful regenerative procedures. Other factors
that may contribute to wound healing success/
complications include soft tissue flexibility, incision
design, thickness of the flap, operator’s experience
and learning curve in performing the procedure, and
patient compliance with the pre- and postoperative
instructions.22,23
48 Volume 33, Number 1, 2018
© 2018 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Agreement/Disagreement with
Previous Studies
It has been demonstrated that events during the heal-
ing period may potentially impact upon the regen-
erative outcome. For example, one trial demonstrated
90% to 100% bone regeneration when the surgical site
remained undisturbed during the initial 6 months fol-
lowing surgery, while only 42% to 62% regeneration
was achieved when the surgical site developed infec-
tion during the healing period.30 Uneventful soft tissue
healing not only confines the graft material to the de-
sired site, but also provides vascular and nutrient sup-
ply to the surgical site as well as forming a protective
barrier against biologic and mechanical irritation. The
amount and width of keratinized mucosa, flap thick-
ness, flap tension, soft tissue biotype, type and size of
the underlying bony defect involvement, membrane
material selection, suture diameter used for wound
closure, and the clinician surgical skill are all potential
contributing factors for soft tissue complications and
early membrane exposure.31
Contrary to other studies, the present systematic
review revealed a higher weighted complication rate
of developing soft tissue complications when resorb-
able membrane was used. According to previous
animal studies, nonresorbable membrane interfered
with revascularization of the healing site while re-
sorbable membranes allowed early anastomosis of
the flap and regenerated tissue at the surgical site.32
In other human studies comparing the use of titani-
um mesh (Ti-mesh) with nonresorbable membrane,
early membrane exposure led to early membrane
removal and thus compromised bone regeneration
results.12,33,34 The biologic response may be different
for membranes, which completely occlude cells, ver-
sus Ti-mesh, which allows, eg, angiogenesis and space
maintenance. The strength and rigidity of Ti-mesh
provide excellent mechanical stabilization and com-
partmentalization of the bone grafts underneath,
while the elasticity and plasticity properties prevent
mucosal compression and allow perfect adaptation
to accommodate a wide range of bony defect mor-
phology. Porosity has an additive benefit of enhanc-
ing GBR outcome in terms of improving epithelium
stability and allowing blood supply and extracellular
nutrient free diffusion across the mesh.
Rigid, nonresorbable membranes are more prone
to early exposure due to their tendency to revert
back to their original shape after adapting to cover
the bony defect site.31 Nonetheless, the present re-
view found that nonresorbable membrane demon-
strated similar long-term vascularization response
as resorbable membrane; however, the timing of the
membrane removal became crucial for nonresorbable
membrane, since delayed membrane removal may

potentiate the needed vascular supply to the flap,
which may lead to early membrane exposure and/or
soft tissue complications. Bacteria are capable of pen-
etrating through prematurely exposed nonresorbable
membrane within 4 weeks following the surgery.35 As
demonstrated in a 2006 study, the timely membrane
removal at the first sign of membrane exposure can
still lead to a successful clinical outcome as long as
acute infection is prevented.12
Limitations and Future Directions
Several limitations of this study may need to be consid-
ered. Among the 15 articles included in the systemic
review, only 4 were RCTs; the majority were clinical
studies or case reports. Confounding factors such as
different study designs, follow-up periods, type of
membranes and grafting materials used, and clini-
cal documentation and categorization of the surgical
sites were not adjusted uniformly across the included
studies. Therefore, future clinical trials must focus on
investigating the effect of membrane type on soft tis-
sue complications following standardized approaches
to better understand the impact of the material prop-
erties on predictable GBR. Moreover, there is a need
for long-term (> 10 years) regenerative outcomes for
horizontal augmentation to know the fate of the bone
formed following the principle of GBR and the influence
of newly formed bone properties upon peri-implant
tissue stability.
CONCLUSIONS
It can be concluded that soft tissue complications
after guided bone regeneration are estimated to
occur in 16.8% of cases. Soft tissue management
remains the key factor to avoid soft tissue compli-
cations and, hence, to increase the success of bone
regenerative therapy.
ACKNOWLEDGMENTS
The study was supported by the University of Michigan Periodon-
tal Graduate Student Research Fund. The authors do not have
any financial interests, either directly or indirectly, in the prod-
ucts or information listed in the paper.
REFERENCES
1. Monje A, Fu JH, Chan HL, et al. Do implant length and width matter
for short dental implants (<10 mm)? A meta-analysis of prospective
studies. J Periodontol 2013;84:1783–1791.
2. Melcher AH. On the repair potential of periodontal tissues. J Peri-
odontol 1976;47:256–260.
© 2018 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Lim et al
3. Simion M, Trisi P, Piattelli A. Vertical ridge augmentation using a
membrane technique associated with osseointegrated implants.
Int J Periodontics Restorative Dent 1994;14:496–511.
4. Annibali S, Bignozzi I, Sammartino G, La Monaca G, Cristalli MP.
Horizontal and vertical ridge augmentation in localized al-
veolar deficient sites: A retrospective case series. Implant Dent
2012;21:175–185.
5. Urban IA, Nagursky H, Lozada JL. Horizontal ridge augmentation
with a resorbable membrane and particulated autogenous bone
with or without anorganic bovine bone-derived mineral: A pro-
spective case series in 22 patients. Int J Oral Maxillofac Implants
2011;26:404–414.
6. Machtei EE. The effect of membrane exposure on the outcome of
regenerative procedures in humans: A meta-analysis. J Periodontol
2001;72:512–516.
7. Fugazzotto PA. Maintenance of soft tissue closure following
guided bone regeneration: Technical considerations and report of
723 cases. J Periodontol 1999;70:1085–1097.
8. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for
reporting systematic reviews and meta-analyses of studies that
evaluate health care interventions: Explanation and elaboration. J
Clin Epidemiol 2009;62:e1–e34.
9. Buser D, Dula K, Hirt HP, Schenk RK. Lateral ridge augmenta-
tion using autografts and barrier membranes: A clinical study
with 40 partially edentulous patients. J Oral Maxillofac Surg
1996;54:420–432.
10. Lee JY, Kim YK, Yi YJ, Choi JH. Clinical evaluation of ridge augmen-
tation using autogenous tooth bone graft material: Case series
study. J Korean Assoc Oral Maxillofac Surg 2013;39:156–160.
11. Merli M, Lombardini F, Esposito M. Vertical ridge augmentation
with autogenous bone grafts 3 years after loading: Resorbable
barriers versus titanium-reinforced barriers. A randomized con-
trolled clinical trial. Int J Oral Maxillofac Implants 2010;25:801–807.
12. Proussaefs P, Lozada J. Use of titanium mesh for staged localized
alveolar ridge augmentation: Clinical and histologic-histomorpho-
metric evaluation. J Oral Implantol 2006;32:237–247.
13. Rabelo GD, de Paula PM, Rocha FS, Jordão Silva C, Zanetta-Barbosa
D. Retrospective study of bone grafting procedures before implant
placement. Implant Dent 2010;19:342–350.
14. Roccuzzo M, Ramieri G, Spada MC, Bianchi SD, Berrone S. Vertical
alveolar ridge augmentation by means of a titanium mesh and
autogenous bone grafts. Clin Oral Implants Res 2004;15:73–81.
15. Becker W, Dahlin C, Lekholm U, et al. Five-year evaluation of
implants placed at extraction and with dehiscences and fenestra-
tion defects augmented with ePTFE membranes: Results from
a prospective multicenter study. Clin Implant Dent Relat Res
1999;1:27–32.
16. Buser D, Brägger U, Lang NP, Nyman S. Regeneration and enlarge-
ment of jaw bone using guided tissue regeneration. Clin Oral
Implants Res 1990;1:22–32.
17. Cardaropoli D, Gaveglio L, Cardaropoli G. Vertical ridge augmenta-
tion with a collagen membrane, bovine bone mineral and fibrin
sealer: Clinical and histologic findings. Int J Periodontics Restor-
ative Dent 2013;33:583–589.
18. Carpio L, Loza J, Lynch S, Genco R. Guided bone regeneration
around endosseous implants with anorganic bovine bone mineral.
A randomized controlled trial comparing bioabsorbable versus
non-resorbable barriers. J Periodontol 2000;71:1743–1749.
19. Llambés F, Silvestre FJ, Caffesse R. Vertical guided bone regenera-
tion with bioabsorbable barriers. J Periodontol 2007;78:2036–2042.
20. Mayfield L, Nobréus N, Attström R, Linde A. Guided bone
regeneration in dental implant treatment using a bioabsorbable
membrane. Clin Oral Implants Res 1997;8:10–17.
21. Merli M, Migani M, Bernardelli F, Esposito M. Vertical bone aug-
mentation with dental implant placement: Efficacy and compli-
cations associated with 2 different techniques. A retrospective
cohort study. Int J Oral Maxillofac Implants 2006;21:600–606.
22. Merli M, Migani M, Esposito M. Vertical ridge augmentation
with autogenous bone grafts: Resorbable barriers supported by
ostheosynthesis plates versus titanium-reinforced barriers. A pre-
liminary report of a blinded, randomized controlled clinical trial.
Int J Oral Maxillofac Implants 2007;22:373–382.
The International Journal of Oral & Maxillofacial Implants 49
Lim et al

23. Park SH, Wang HL. Clinical significance of incision location on guid-
ed bone regeneration: Human study. J Periodontol 2007;78:47–51.
24. Parodi R, Carusi G, Santarelli G, Nanni F. Implant placement in
large edentulous ridges expanded by GBR using a bioresorb-
able collagen membrane. Int J Periodontics Restorative Dent
1998;18:266–275.
25. Poli PP, Beretta M, Cicciù M, Maiorana C. Alveolar ridge augmenta-
tion with titanium mesh. A retrospective clinical study. Open Dent
J 2014;8:148–158.
26. Torres J, Tamimi F, Alkhraisat MH, et al. Platelet-rich plasma may
prevent titanium-mesh exposure in alveolar ridge augmentation
with anorganic bovine bone. J Clin Periodontol 2010;37:943–951.
27. von Arx T, Kurt B. Implant placement and simultaneous ridge aug-
mentation using autogenous bone and a micro titanium mesh: A
prospective clinical study with 20 implants. Clin Oral Implants Res
1999;10:24–33.
28. Park JC, Kim CS, Choi SH, Cho S, Chai JK, Jung UW. Flap extension
attained by vertical and periosteal-releasing incisions: A prospec-
tive cohort study. Clin Oral Implants Res 2012;23:993–998.
29. Greenstein G, Greenstein B, Cavallaro J, Elian N, Tarnow D. Flap
advancement: Practical techniques to attain tension-free primary
closure. J Periodontol 2009;80:4–15.
30. Lang NP, Hammerle CH, Brägger U, Lehmann B, Nyman SR. Guided
tissue regeneration in jawbone defects prior to implant place-
ment. Clin Oral Implants Res 1994;5:92–97.
31. Chao YC, Chang PC, Fu JH, Wang HL, Chan HL. Surgical site as-
sessment for soft tissue management in ridge augmentation
procedures. Int J Periodontics Restorative Dent 2015;35:e75–e83.
32. Vergara JA, Quiñones CR, Nasjleti CE, Caffesse RG. Vascular re-
sponse to guided tissue regeneration procedures using nonre-
sorbable and bioabsorbable membranes in dogs. J Periodontol
1997;68:217–224.
33. von Arx T, Hardt N, Wallkamm B. The TIME technique: A new meth-
od for localized alveolar ridge augmentation prior to placement of
dental implants. Int J Oral Maxillofac Implants 1996;11:387–394.
34. Simion M, Jovanovic SA, Trisi P, Scarano A, Piattelli A. Vertical
ridge augmentation around dental implants using a membrane
technique and autogenous bone or allografts in humans. Int J
Periodontics Restorative Dent 1998;18:8–23.
35. Simion M, Trisi P, Maglione M, Piattelli A. Bacterial penetration in
vitro through GTAM membrane with and without topical chlorhex-
idine application. A light and scanning electron microscopic study.
J Clin Periodontol 1995;22:321–331.

Supplemental Figures

0.0 0.0

0.5 0.5
Standard Error

Standard Error

1.0 1.0

1.5 1.5

2.0 2.0
–5 –4 –3 –2 –1 0 1 2 3 –5 –4 –3 –2 –1 0 1 2 3
Logit event rate Logit event rate

Fig S-1   Funnel plot of overall analysis. Fig S-2   Funnel plot of meta-analysis of studies utilizing resorb-
able membranes.

0.0

0.5
Standard Error

1.0

1.5

2.0
–5 –4 –3 –2 –1 0 1 2 3
Logit event rate

Fig S-3   Funnel plot of meta-analysis of studies utilizing nonre-

sorbable membranes.

50 Volume 33, Number 1, 2018

© 2018 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.