J Clinic Periodontology - 2019 - Naenni - Efficacy of Lateral Bone Augmentation Prior To Implant Placement A Systematic

| |
Received: 19 July 2018 Revised: 9 November 2018 Accepted: 1 December 2018
DOI: 10.1111/jcpe.13052
SUPPLEMENT ARTICLE
Efficacy of lateral bone augmentation prior to implant

placement: A systematic review and meta-analysis
Nadja Naenni1 | Hyun‐Chang Lim1,3 | Spyridon N. Papageorgiou2 |

Christoph H. F. Hämmerle1
1
Clinic of Fixed and Removable
Prosthodontics and Dental Abstract
Material Science, Center of Dental Aim: The aim of the current systematic review was to critically appraise evidence
Medicine, University of Zurich, Zurich,
Switzerland from randomized and prospective non-randomized comparative clinical trials about
2
Clinic of Orthodontics and the efficacy of lateral bone augmentation prior to implant placement and their out-
Pediatric Dentistry, Center of Dental
come regarding bone width gain.
Medicine, University of Zurich, Zurich,
Switzerland Materials and Methods: Eight databases were searched until May 2018 for rand-
3
Department of Periodontology, omized and prospective non-randomized comparative trials on lateral bone augmen-
Periodontal-Implant Clinical Research
tation prior to implant placement. After elimination of duplicate studies, data
Institute, School of Dentistry, Kyung Hee
University, Seoul, Republic of Korea extraction and risk-of-bias assessment according to the Cochrane guidelines, random-
effects meta-analyses of mean differences (MD) or relative risks (RR) and their 95%
Correspondence
Nadja Naenni, Clinic of Fixed and Removable CIs were performed, followed by subgroup, meta-regression and sensitivity analyses.
Prosthodontics and Dental Material Science,
Results: Overall, 25 trials (16 randomized/9 non-randomized) were identified, which
University of Zurich, Zurich, Switzerland.
Email: nadja.naenni@zzm.uzh.ch included a total of 553 patients (42.2% male; mean age of 43.9 years). In these in-
Funding information cluded studies and populations, various modalities for primary lateral bone augmen-
This work was initiated by the European
tation rendered implant placement feasible. Small discrepancies were found between
Federation of Periodontology (EFP) and
the Osteology Foundation (OF) and was overall clinical and radiographic gain (pooled gains of 3.45 ± 1.18 mm versus 2.90 ±
funded by the Clinic of Fixed and Removable
0.83 mm, respectively), but were not statistically significant. Bone width gain was
Prosthodontics and Dental Material
Science, University of Zurich and the Clinic significantly inversely associated with baseline bone width (pooled effect: −0.35 mm/
of Orthodontics and Pediatric Dentistry,
mm; 95% CI: −0.63 to −0.07 mm; p = 0.01). Additionally, % graft resorption was as-
Center of Dental Medicine, University of
Zurich, Zurich, Switzerland. sociated with patient age (36%/year, 95% CI: −0.62 to −0.11 mm; p = 0.01). The pres-
ence of xenograft added to autologous graft led to less resorption compared to
autologous graft alone (MD: 1.06 mm; 95% CI: 0.21 to 1.92 mm; p = 0.01). Barrier
membrane did not yield significant difference in terms of bone width gain (MD:
−0.33 mm; 95% CI: −2.24 to 1.58 mm; p > 0.05) and graft resorption (MD: 0.84 mm;
95% CI: −1.42 to 3.09 mm; p > 0.05). However, the quality of evidence ranged from
very low to moderate due to bias and imprecision.
Conclusions: Initially smaller bone dimensions are associated with favours larger
bone width gain, which indicates that a severe lateral bone deficiency can be effec-
tively augmented applying primary lateral bone augmentation. Both Patients’ age and
recipient site (maxilla or mandible) seem to influence graft resorption. The addition of
a xenograft can be helpful in reducing graft resorption. Existing evidence from
J Clin Periodontol. 2019;46(Suppl. 21):287–306. wileyonlinelibrary.com/journal/jcpe © 2019 John Wiley & Sons A/S. | 287
Published by John Wiley & Sons Ltd
1600051x, 2019, S21, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jcpe.13052 by Cochrane Chile, Wiley Online Library on [28/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
|
288 NAENNI et al.
randomized and prospective non-randomized trials on humans indicates that lateral

bone augmentation prior to implant placement can successfully increase bone width.
There are some indications that patient-related, site-related, and technique-related
characteristics might influence the amount of gained bone width, but the quality of
evidence is for the most part hampered by the small number of existing studies and
methodological limitations that might lead to bias.
KEYWORDS
dental implant, implant placement, lateral bone augmentation, primary bone augmentation,
ridge width, systematic review
1 | I NTRO D U C TI O N
Clinical Relevance
1.1 | Rationale
Scientific rationale for the study: Whether solely primary
Implant placement at edentulous areas with severe bone deficiencies
lateral bone augmentation leads to sufficient ridge width
often represents a challenging situation for clinicians. In order to cor-
for subsequent implant placement is still scarcely reported
rect bone dimensions insufficient for implant placement, numerous
in detail. Principal findings: Most of the studies (24 out of 25
bone augmentation procedures have been described in the literature
included) reported that lateral bone augmentation allowed
(Esposito et al., 2009; Jensen & Terheyden, 2009). Bone augmentation
for subsequent implant placement. Nevertheless, only a
may be carried out prior to or simultaneous with implant placement.
small number of studies reported on the proportion of im-
The choice whether or not primary bone augmentation is needed de-
plants placed according to the planning/ the placement in a
pends on a number of factors, including among others: the amount
prosthetically ideal position, the need for a narrower im-
of missing bone; the site within the dental arch; feasibility of implant
plant diameter or the proportion of implants which needed
placement in a proper position and with primary stability; and the ex-
additional bone augmentation procedures. The ridge width
pected size of implant surface exposure at the time of implant place-
gain is significantly influenced by the ridge width prior to
ment. Available evidence indicates that predictable correction of bone
augmentation. Furthermore, bone width gain and graft re-
deficiencies using bone augmentation procedures is possible (Benic &
sorption are influenced by baseline ridge width, age and
Hammerle, 2014). Moreover, implant survival rates are similar both for
jaw. The presence of xenogeneic graft material seems fa-
implants placed in pristine bone as well as for implants placed in aug-
vourable in terms of dimensional stability. Practical implica‐
mented bone (Hammerle, Jung, & Feloutzis, 2002).
tions: Primary lateral bone augmentation represents a
The use of a variety of techniques and materials has been described
predictable procedure in order to gain sufficient ridge
(Aghaloo & Moy, 2007; Chiapasco & Casentini, 2018; Donos, Mardas, &
width for placing implants.
Chadha, 2008). The materials applied encompassing autologous bone,
allogenic bone, xenogenic bone and synthetic bone substitute mate-
rials. All of these biomaterials have been applied as blocks or in par-
ticulated form. Autologous bone grafts are predominantly harvested materials is to reduce or completely avoid the morbidity associated
from intraoral sources such as the mandibular ramus, the chin region, with bone harvesting procedures. Finally, growth factors such as re-
the maxillary tuberosity, and the nasal spine (Chappuis et al., 2018; combinant human bone morphogenetic protein and enhancers such
Cordaro, Amade, & Cordaro, 2002; Cordaro, Torsello, Morcavallo, & as platelet-rich plasma (PRP) and platelet-rich fibrin (PRF) have been
di Torresanto, 2011; Meijndert, Raghoebar, Meijer, & Vissink, 2008; used alone or in combination with the above-described materials (Badr,
Tolstunov, 2009). When larger amounts of autologous bone are needed, Coulthard, Alissa, & Oliver, 2010; Barbu et al., 2016; de Freitas et al.,
extraoral sites are chosen including the iliac crest and the external plate 2013).
of the calvarium (Chiapasco, Autelitano, Rabbiosi, & Zaniboni, 2013; A variety of techniques, such as ridge splitting, guided bone re-
Chiapasco, Di Martino, Anello, Zaniboni, & Romeo, 2015). Moreover, generation (GBR) procedures, transplantation of autologous bone,
autologous bone and allogenic bone are often combined with bone and application of bone substitute materials, have been introduced
substitute materials (Wang, Misch, & Neiva, 2004) in order to expand (Chiapasco et al., 2015; Da Costa, Pelegrine, Fagundes, Simoes
the volume of grafting material available and to reduce the resorption Mde, & Taha, 2011; Hammerle et al., 2002; Scipioni, Bruschi,
of the transplanted autologous or synthetic bone. A clinical benefit & Calesini, 1994; Urban, Nagursky, & Lozada, 2011). Although
of using allogenic bone, xenogenic bone or alloplastic bone substitute these techniques have shown positive results for the correction
NAENNI et al. |
289
of alveolar bone defects, no material and/or technique has so far resorbed (defined as bone width at re-e ntry for implant insertion
demonstrated to be superior in terms of feasibility of implant place- minus immediate post-augmentation bone width); (c) % bone graft
ment and bone gain. resorption from the added graft (defined as the amount of bone
Numerous studies have investigated horizontal bone augmen- graft resorbed divided by the post-augmentation width addition);
tation both prior to and simultaneous with implant placement (d) adverse events at recipient site (such as infection, graft failure,
(Sanz-S anchez, Ortiz-V igon, Sanz-Martin, Figuero, & Sanz, 2015). wound dehiscence) or at donor site (nerve damage, infection); (e)
Most of the systematic reviews available have concomitantly ad- patient-reported outcome measures (PROMS) (morbidity: swell-
dressed staged and simultaneous bone augmentation (Aghaloo ing, pain); (f) implant failure (defined either as need for implant
& Moy, 2007; Esposito et al., 2009; Sanz-S anchez et al., 2015). removal or as salvageable complication/peri-implant disease); and
However, only limited evidence exists regarding solely primary (g) marginal bone level changes after implant insertion/loading.
horizontal bone augmentation or the various existing protocols Positive values for the primary outcome of bone width gain in-
(including materials and/or techniques) with respect to their abil- dicate that the bone width was increased compared to baseline.
ity to successfully regenerate bony ridge defects prior to implant Negative values for the secondary outcome of bone graft resorp-
placement. tion indicate that the bone graft added during the augmentation
The primary aim of the reconstruction of deficient ridges is to procedure had been resorbed at the re-e ntry surgery where im-
improve the ridge profile in order to facilitate subsequent prosthetic plant placement was planned.
rehabilitation, which ideally would be exclusively implant-borne. The focused question this systematic review answered was as
Hence, the success of primary bone augmentation procedures may follows: “In patients presenting with insufficient alveolar ridge width
be assessed by the feasibility of implant placement following bone for implant placement, does primary lateral bone augmentation lead
healing. to sufficient gain in bone width to allow for subsequent implant
placement.”
Excluded were clearly retrospective studies, case reports or case
1.2 | Objectives
series, animal studies, pre-clinical and non-clinical studies. No lim-
The aim of the current systematic review was to critically appraise itations were set regarding publication year, publication language or
evidence from randomized and prospective non-randomized com- publication type.
parative clinical trials on humans about the efficacy of primary lat-
eral bone augmentation prior to implant placement and the primary
2.2 | Information sources and search
outcome of bone width gain.
Electronic searches were performed in the following eight electronic
general, open-access, regional and grey literature bibliographic da-
2 | M ATE R I A L S A N D M E TH O DS
tabases: MEDLINE (searched via PubMed), Cochrane Database
of Systematic Reviews, Cochrane Central Register of Controlled
2.1 | Protocol, registration and eligibility criteria
Trials (CENTRAL), Cochrane Database of Abstracts of Reviews of
The review protocol was developed a priori according to the Effects (DARE), EMBASE, Scopus, Web of Knowledge and Virtual
PRISMA (Preferred Reporting Items for Systematic Review and Health Library (including Bibliografia Brasileira de Odontologia
Meta-A nalyses) statement (Liberati et al., 2009) and registered and LILACS). Additionally, Directory of Open Access Journals,
in PROSPERO (international prospective register of systematic Digital Dissertations (searched via UMI ProQuest), metaRegister
reviews; CRD42018093073). The Participants, Intervention, of Controlled Trials, WHO trials search portal and Google Scholar
Comparison, Outcome, Study design (PICOS) framework was es- were searched manually. No search filters were applied other than
tablished as follows: participants—patients in need of primary trials on humans and dentistry, where available. Hand searching was
lateral bone augmentation for increasing bone width in order to also performed from the reference/citation lists of eligible studies
place dental implants at one or more sites in either jaw; interven- and relevant systematic reviews for additional studies (Supporting
tion—primary lateral bone augmentation using different materials/ Information Appendix S1).
techniques; comparison—any material/technique other than the
Intervention group; outcome (primary)—amount (width) of alveolar
2.3 | Study selection
ridge gained through the augmentation procedure (measured ra-
diographically or clinically), which was defined as bone width at re- Study selection was performed in duplicate by two authors (NN and
entry for implant insertion minus pre-augmentation bone width; LHC) independently. Any disagreement in the process was solved
and study design—randomized and prospective non-r andomized by discussion with the other two authors (SNP and CHFH). First the
comparative clinical trials. Additional secondary outcomes in- titles and then the abstracts of all studies derived from the search
cluded the following: (a) feasibility of implant placement after an were screened against the inclusion criteria. If title or abstract did
adequate healing time (with or without additional bone augmenta- not provide sufficient information regarding exclusion of a study, the
tion at the time of implant placement); (b) amount of bone graft decision was based on the study's full text.
|
290 NAENNI et al.
and non-randomized studies were assessed with the ROBINS-I (Risk

2.4 | Data items and data collection process
Of Bias In Non-randomised Studies—of Interventions) tool (Sterne
Data extraction was made independently by two authors (NN and et al., 2016).
LHC) using pre-specified extraction tables covering study character-
istics (design, setting, and country), patient characteristics (number,
2.6 | Summary measures and synthesis of results
sex, age, systemic health, and smoking), intervention (graft, mem-
brane, number of augmented sites/implants) and outcome measure- The mean difference (MD) for continuous variables and the rela-
ments (follow-up, measurement method, and primary/secondary tive risk (RR) for binary variables with their corresponding 95%
outcome). Any disagreements were discussed with the other two confidence intervals (CIs) were chosen a priori for data synthesis
authors (SNP and CHFH). of primary or secondary outcomes. As various patient-, site-, and
graft-s pecific factors were assumed to influence the true effects
of lateral bone augmentation, a random-effect model was a priori
2.5 | Risk of bias of individual studies
chosen in order to incorporate this variability and calculate the
Two authors (NN and LHC) independently evaluated the risk of bias mean distribution of these effects. The Paule and Mandel vari-
of included studies, and any disagreements were discussed with ance estimator was chosen a priori instead of the DerSimonian
the other two authors (SNP and CHFH). Randomized trials were as- and Laird one, due to its improved performance (Veroniki et al.,
sessed with the Cochrane risk-of-bias tool (Higgins & Green, 2011), 2016).
F I G U R E 1 Flow diagram for the identification and selection of eligible studies for this systematic review. *Descriptive data were missing
to include the studies in the analyses
NAENNI et al. |
291
TA B L E 1 Implant information by included studies
Number of implants Additional GBR at the time

Nr Study placed/planned Implant diameter of implant placement Comment
1 Antoun et al., 2001; NR/NR (100%) NR NR “Implants were successfully

placed in an optimal position
at all grafted sites.”
2 Araujo et al., 2016 23/23 3.75 mm NR All implants measured
3.75 × 10 mm
3 Badr et al., 2010; 48 (group PRP) 4 mm (mean) One patient needed GBR
37 (group control)/NR at the time of implant
placement
4 Barbu et al., 2016; 47/NR 3.7, 4.1, 4.7 mm NR Information on diameter is
given, but unknown which
implant was placed at which
site
5 Beitlitum et al., 2010; NR NR 34% overall “Proper implant placement”
but not specified which
dimension
6 Caldwell et al., 2015; 23/26 Standard sized implant NR
7 Chiapasco et al., 1999; 73/74 4.1 mm (Straumann- 24), One patient needed GBR One patient required an
3.75 mm (Brånemarkt at the time of implant additional chin block bone
-50) placement graft (regarded as a failure)
8 Cordaro et al., 2002; 20/NR 4.1 mm in 15 sites, NR
3.75 mm in 3 sites
9 Cordaro et al., 2011; 55/NR 4.0, 4.1, 4.8 mm NR
10 Da Costa et al., 2011; 40/NR NR NR
11 de Freitas et al., 2013, 62/NR 3.5 mm NR
2016;
12 Eskan et al., 2014; 28/28 Reduced/regular/wide NR Pre-augmentation planning;
one site received smaller
diameter than planned
13 Kheur et al., 2018; 49/NR NR No
14 Maiorana et al., 2005; 47/NR NR NR
15 Mazzocco et al., 2011; 23/23 NR NR
16 Meijndert et al., 2008; 15/15 4.1 mm NR
Meijndert et al., 2017
17 Meijndert et al., 2005; 93/93 4.1 mm No
18 Monje et al., 2014; 43/43 NR NR
19 Mordenfeld et al., 2014; 71/NR NR NR
2017
20 Mordini et al., 2016 10/10 NR NR “… provided sufficient bone
for implant placement as
planned prosthetically”
21 Pourabbas & Nezafati, NR NR NR “Suitable sized implants were
2007; placed”
22 Shalash et al., 2013; NR NR NR “Suitable implants was
screwed”
23 Thoma et al., 2018; NR
24 Thor et al., 2005; 76/76 3.5 mm NR
Dasmah et al., 2012;
Dasmah et al., 2013;
25 Urban et al., 2011 58/NR 3.75, 4.0 mm NR
Placed: number of implants placed; planned: number of implants planned; Implant diameter: implant diameter in mm; additional GBR at the time of
implant placement: guided bone regeneration (GBR) performed at the time of implant placement; Comment: additional information given by the
authors.
|
292 NAENNI et al.
TA B L E 2 Characteristics of included studies
Design; setting; Sites (implants); Jaw;

Nr Study countrya Pats (M/F); ageb SD; smoking Tx region
1 Antoun et al., 2001; pRCT; Univ; FRA G1: 7 (NR); NR NR; NR G1: AUG (block; chin) 13 (NR); NR; NR
G2: 5 (NR); NR G2: AUG (block; chin) +
MEMNRES
2 Aarújo, 2016; wpRCT; Univ; BRA G1/G2: 11 (1/10); No; NR G1: XEN (block) + MEMRES 23 (23); Max; ANT/
49.1 G2: AUG (block; ramus) + POS
MEMRES
3 Badr et al., 2010; pRCT; Univ; UK G1: 9 (2/7); 32.0 No; Some G1: AUG (block; ilium) 39 (85); Max; ANT/
G2: 13 (6/7); 38.0 G2: AUG (block; ilium) + POS
PRP
4 Barbu et al., 2016; uNRS; Univ; NR ROM G1: 12 (7/5); 42.6 No, No G1: AUG (block + part; 24 (47); Mnd; POS
G2: 12 (4/8); 48.9 ramus) + PRF
G2: AUG (block + part;
ramus) + XEN (part) +
MEMRES
5 Beitlitum et al., pNRS; Univ/pract; ISR G1: 12 (NR); NR No; Some (≤10) G1: ALG (block) + AUG 28 (NR); NR; NR
2010; G2: 15 (NR); NR (part) + MEMRES
G2: ALG (block) + MEMRES
6 Caldwell et al., pRCT; Univ; USA G1: 12 (2/10); No; Some (≤10) G1: ALG (part) + AUG 29 (33); Max/Mnd;
2015; 50.4 (part; ramus, symphysis) ANT/POS
G2: 14 (3/11); + MEMres
53.6c G2: ALG + MEMres
7 Chiapasco et al., pNRS; NR; ITA G1: 15 (8/7); 43.3 No; Some (not G1: AUG (part; ramus/ 30 (74); Max/Mnd;
1999; G2: 15 (4/11); heavy) chin) + MEMNRES ANT/POS
40.9 G2: AUG (block; ilium/
chin/calvaria)
8 Cordaro et al., pNRS; NR; ITA G1/G2: 15 (NR); NR; NR G1: AUG (block; ramus) 9 (20); Max/Mnd;
2002; NR G2: AUG (block; symph) ANT/POS
9 Cordaro et al., pRCT; Univ; ITA G1: 8 (NR); NR No; Some (≤10) G1: AUG (block; ramus) 22 (55); NR; NR
2011; G2: 9 (NR); NR +XEN (part) + MENRES
G2: AUG (block; ramus)
10 Da Costa et al., pRCT; hosp; BRA G1: 5 (NR); NR No; No G1: ALG (block) + bone 10 (40); Max; ANT
2011; G2: 5 (NR); NR marrow aspirate
G2: ALG (block)
11 de Freitas et al., pRCT; Univ; BRA G1: 12 (6/6); 47.7 No; No G1: AUG (part; ramus) + 24 (62); Max; ANT
2013, 2016; G2: 12 (4/8); 42.4 MEMNRES
G2: collagen sponge
(origin NR) + rhBMP2 +
MEMNRES
12 Eskan et al., 2014; pRCT; Univ; USA G1: 14 (8/6); 52.0 No; No G1: ALG (part) + MEMRES 28 (28); Max/Mnd;
G2: 14 (6/8); 49.0 G2: ALG (part) + PRP + ANT/POS
MEMRES
13 Kheur et al., 2018; pRCT; Univ; IND G1: 12 (9/3); 51.1 NR; Some (≤10) G1: ALG (small or large NR (49); Mnd; POS
G2: 11 (5/6); 50.6 part) + RS
G2: ALP (small or large
part) + RS
14 Maiorana et al., uNRS; NR; ITA G1: 11 (NR); 28.1 NR; NR G1: AUG (block+part; 26 (43); Max/Mnd;
2005; G2: 12 (NR); 35.6 chin/ramus) + XEN (part) ANT/POS
G2: AUG (block+part;
chin/ramus)
15 Mazzocco et al., wpRCT; Univ; USA G1/2: 8 (4/4); NR; No G1: XEN (part)+ ALG 23 (23); NR; NR
2011; 47.0 (part) + MEMRES
G2: RE + XEN (part) + ALG
(part) + MEMRES
(Continues)
NAENNI et al. |
293
TA B L E 2 (Continued)
Design; setting; Sites (implants); Jaw;

Nr Study countrya Pats (M/F); ageb SD; smoking Tx region
17 Meijndert et al., pRCT; Univ; NLD G1: 5 (2/3); 34.0 NR; No G1: AUG (block/part; chin) 15 (15); Max; ANT
2005; G2: 5 (2/3); 33.8 G2: AUG (block/part;
G3: 5 (3/2); 36.6 chin) + MEMRES
G3: XEN (part)+ MEMRES
16 Meijndert et al., pRCT; Univ; NLD G1: 31 (13/18); No; No G1: AUG (block/part; chin) 93 (93); Max; ANT
2008; 2017 33.3 G2: AUG (block/part;
G2: 31 (16/15); chin) + MEMRES
34.6 G3: XEN (part)+ MEMRES
G3: 31 (15/16);
32.2
18 Monje et al., 2014; uNRS; Univ; ESP G1: 4 (3/1); 45.4 NR; NR G1: AUG (block, calv) + 14 (43); Max; ANT/
G2: 5 (2/3); 50.5 XEN (part) + MEMRES POS
G2: AUG (block, ilium) +
XEN (part) + MEMRES
19 Mordenfeld 2014; wpRCT; hosp; SWE G1/2: 13 (6/7); No; No G1: AUG/XEN 90:10 35 (71); Max/Man;
2017 59.6 (part, ramus) + fibrin glue ANT/POS
+ MENRES
G2: AUG/XEN 60:40
(part, ramus) + fibrin glue
+ MENRES
20 Mordini 2016 pRCT; Univ; USA G1: 3 (1/2); 55.0 No; No G1: ALG (part) + MEMRES 10 (10); Mnd; POS
G2: 3 (0/3); 64.7 G2: ALG (part) + MEMNRES
21 Pourabbas & pRCT; Univ; IRN G1: 16 (8/8); NR No; No G1: AUG (block, chin) + 32 (NR); Max; ANT
Nezafati, 2007; G2: 16 (9/7); NR MEMRES
G2: AUG (block, ramus) +
MEMRES
22 Shalash et al., 2013; uNRS; NR; EGY G1:10 (NR); NR No; No G1: ALP (part) + MEMNRES 18 (19); NR; NR
G2:10 (NR); NR G2: ALP (part) + XEN
(part) + MEMNRES
23 Thoma et al., 2018; pRCT; Univ; MC G1: 12 (7/5); 56.2 No; some (<10) G1: XEN (block) + NR (NR); NR; NR
(CHE/AUT) G2: 12 (6/6); 47.5 rhBMP2 + MEMRES
G2: AUG (block; ramus,
chin) + XEN (part) +
MEMRES
24 Thor 2005; Dasmah wpNRS; hosp; SWE G1/G2: 19 (2/17); Some; Some (<10) G1: AUG (block; ilium) 38 (76); Max; ANT
2012; Dasmah 58.0 G2: AUG (part; ilium) +
et al., 2013; PRP
25 Urban et al., 2011 pNRS; pract; HUN G1: 22 (5/17); No; NR G1: AUG (part; ramus, 25 (58); Max/Mnd; NR
49.9 surgical site)
G2: AUG (part; ramus,
surgical site) + XEN (part)
ALG: allogenic graft; ALP: alloplastic grafts; ANT: anterior; AUG: autologous graft G1, G2: group 1, group 2; hosp: hospital; Max: maxilla; MC: multi-
centre; MEM: membrane; Mnd: mandible; NR: not reported; NRES: non-resorbable; Pats: patient; pNRS: prospective non-randomized study; POS:
posterior; pract: private practice; pRCT: parallel randomized clinical trial; PRF: platelet-rich fibrin; PRP: platelet-rich plasma; RE: ridge expansion; RES:
resorbable; RS: ridge splitting; SD: systemic disease; Univ: university clinic; uNRS: non-randomized study with unclear design; wpRCT: within-person
randomized clinical trial; XEN: xenograft.
a
Countries are given with their ISO Alpha-3 codes. bAge is given in years as either mean (one value) or if mean not reported as range (two values in
parenthesis). cExcluding two patients being in both groups from the age counting.
The extent and impact of between-s tudy heterogeneity were 2011). Heterogeneity was roughly categorized as low, moder-
assessed by inspecting the forest plots and calculating the τ 2 (ab- ate and high according to I2 values of 25%, 50% and 75%, al-
solute heterogeneity) and the I2 (relative heterogeneity), respec- though the heterogeneity's localization on the forest plot was
tively; I2 defines the proportion of total variability in the result also judged. Additionally, the 95% CIs around τ 2 and I2 were cal-
explained by heterogeneity and not chance (Higgins & Green, culated (Ioannidis, Patsopoulos, & Evangelou, 2007) to quantify
|
294 NAENNI et al.
TA B L E 3 Outcomes assessed by included studies
Nr Study FUa; Imp Outcome Unit CoI
1 Antoun et al., 2001; 6.0; Yes BW (CL, RAD); BQ; ImpIns Site GrantCom/NCom
2 Aarújo, 2016; 6.0; Yes; 3y (implant BW (RAD); Imp survival Site NR
survival)
3 Badr et al., 2010; 3.0–4.0; Yes BW (CL); ISQ; ImpIns; Site No
ImpInt
4 Barbu et al., 2016; 4.0; Yes (but 1 in G2) ImpIns; BW (RAD); Compl Site No
5 Beitlitum et al., NR BW (CL, RAD); Compl Patient NR
2010;
6 Caldwell et al., 2015; 6.0; Yes BW (CL) Site No
7 Chiapasco et al., 6.0–8.0; Yes BW (CL); Imp success Site NR
1999;
8 Cordaro et al., 2002; 6.0; Yes BW (CL) Site NR
9 Cordaro et al., 2011; 4.0; Yes BW (CL); Imp success; Site Grant/Foundation
Compl
10 Da Costa et al., 6.0; Yes BW (RAD) Site Industry material
2011; support
11 de Freitas et al., 6.0; Yes BW (CL, RAD) Site Grant
2013; 2016
12 Eskan et al., 2014; 4.0; Yes BW (CL) Site Industry
13 Kheur et al., 2018; 3.0; Yes BW (CL); Imp success Site No
14 Maiorana et al., 5.4; Yes BW (CL) Site NR
2005;
15 Mazzocco et al., 6.0; Yes BW (CL, RAD) Site NR
2011;
16 Meijndert et al., 3.0 (AUG) mBL Site Industry
2008; 2017 6.0 (XEN); Yes survival rate
10y: 2 losses in G3 BW not reported
17 Meijndert et al., 3.0 (AUG) BW (CL) Site Industry
2005; 6.0 (XEN);
Yes
18 Monje et al., 2014; 6.1; Yes ISQ Site Grant/Foundation
19 Mordenfeld 2014; 7.5 (8.1 for implant BW (RAD) Impl Industry (material
2017 placement); Yes support)
2.1 y: Implant survival
20 Mordini et al., 2016 6.0; Yes BW (CL) Site NR
21 Pourabbas & 6.0; Yes BW (CL) Site NR
Nezafati, 2007;
22 Shalash et al., 2013; 6.0; Yes BW (RAD) Site NR
23 Thoma et al., 2018; 4.0; Yes BW (CL) Site Grant
24 Thor 2005; Dasmah 6.0; Yes BW (RAD); Implant NR NR
2012; Dasmah survival; mBL
et al., 2013;
25 Urban et al., 2011 8.12; Yes BW (CL) Site No
BW: bone width; BQ: bone quality; CLIN: clinical; CoI: conflict of interest; Compl: complication; FU: follow-up; Imp: implantation; ImpIns: implantation
insertion; Imp success: implant success; Imp survival: implant survival; NR: not reported; RAD: radiography.
a
Follow-up after augmentation is given in months as either mean (one value) or if mean not reported as range.
uncertainty around these estimates. Ninety-f ive per cent pre- correct interpretation of random-effects meta-analyses (IntHout,
dictive intervals were calculated for meta-analyses of ≥3 trials to Ioannidis, Rovers, & Goeman, 2016). All analyses were conducted
incorporate existing heterogeneity and provide a range of pos- in Stata SE version 14.2 (StataCorp LP, College Station, TX, USA)
sible effects for a future clinical setting, which is crucial for the by one author (SNP) with the data made freely available in Zenodo
NAENNI et al. |
295
F I G U R E 2 a, Risk-of-bias summary
of included randomized trials with the
Cochrane risk-of-bias tool. b, Risk-of-bias
summary of included non-randomized
trials with the ROBINS-I tool
(Naenni, Hyun‐Chang, Papageorgiou, & Hämmerle, 2018). A two- The overall quality of clinical recommendations for all meta-
sided p < 0.05 was considered significant for hypothesis testing, analysed outcomes was rated using the GRADE approach, as very
except for p < 0.10 used for tests of between-s tudy or between- low, low, moderate or high (Guyatt, Oxman, Schunemann, Tugwell,
subgroup heterogeneity (Ioannidis, 2008). & Knottnerus, 2011), and a Summary of Findings table was con-
structed using the improved format proposed by Carrasco-L abra
et al. (2016) and recent guidance on incorporating non-randomized
2.7 | Risk of bias across studies and
studies (Schunemann et al., 2018). The minimal clinical important
additional analyses
(Norman, Sloan, & Wyrwich, 2003) large and very large effects were
Possible sources of heterogeneity were a priori planned to be sought defined as half, one and two standard deviations (using the average
through mixed-effects subgroup analyses about the combination of standard deviation for an outcome across included studies), respec-
different grafts and the use of a membrane. Furthermore, the fol- tively. Arbitrary cut-offs of 1.5, 2.0 and 5.0 (Schünemann, Brożek,
lowing meta-analyses (corresponding to individual patient data cu- Guyatt, & Oxman, 2013) were adopted for RR. The produced forest
mulative meta-regressions) were performed by meta-analysing the plots were augmented with contours denoting the magnitude of the
regression coefficients that originated from re-analysis of available observed effects (Papageorgiou, 2014) to visually gauge heteroge-
raw data: patient age, patient sex, jaw, mouth region, baseline bone neity, clinical relevance, and imprecision.
width, healing time and donor site for autologous grafts. Additional
analyses for subgroups, meta-regressions, and reporting biases
3 | R E S U LT S
were planned, but were not conducted, due to lack of available
studies (Supporting Information Appendix S2).
3.1 | Study selection
Robustness of the results was planned a priori to be checked
with sensitivity analyses based on (a) inclusion/exclusion of The literature search yielded a total of 8,167 hits, while seven ad-
non-r andomized trials, (b) inclusion only of studies with ditional records were identified by hand searching (Figure 1). After
low risk of bias, and (c) improvement of the Grades of eliminating 1,771 duplicates, the titles and abstracts of the remain-
Recommendations, Assessment, Development, and Evaluation ing papers were scrutinized, leading to the exclusion of another
(GRADE) classification. 6,332 records. Finally, 71 full texts were checked for eligibility with
|
296 NAENNI et al.
TA B L E 4 Complications reported by the included studies.
Nr Study Tx Type of Complication & treatment
1 Antoun 2001 G1: AUG (block; chin) G2: membrane exposure (n = 1/12) − antibiotics, chlorhexidine
G2: AUG (block; chin) + MEMNRES gel/rinse, membrane removal 8%
2 Araujo 2016 G1: XEN (block) + MEMRES G1: ecchymosis (n = 1) 9%
G2: AUG (block; ramus) + MEMRES G2: wound dehiscence (n = 1/11), temporary paraesthesia (n = 1),
ecchymosis (n = 2)
Complications were not properly divided according to donor or
recipient site
3 Badr 2010 G1: AUG (block; ilium) G1: soft tissue dehiscence (n = 1/22) − healed spontaneously 5%
G2: AUG (block; ilium) + PRP G2: antral communication (n = 1) − surgical correction under
general anesthesia
Donor site: mild hematoma (n = 1)
4 Barbu 2016 G1: AUG (block + part; ramus) + PRF G2: bone exposure and necrotic change − surgical removal
G2: AUG (block + part; ramus) + XEN Donor site: prolonged pain (resolved 3 weeks later)
(part) + MEMRES
5 Beitlitum 2010 G1: ALG (block) + AUG (part) + MEMRES G1: membrane exposure (n = 1/12) − no specific remark n = 5/27
G2: ALG (block) + MEMRES 18.5%
G2: membrane exposure (n = 4/15) − no specific remark
Recipient site: slight paresthesia of the lower lip (all disappeared)
6 Caldwell 2015 G1: ALG (part) + AUG (part; ramus, G1: acute infection & loss of the majority of graft material
symphysis) + MEMres (n = 1/12) n = 2/26 7%
G2: ALG + MEMres G2: acute infection & loss of the majority of graft material
(n = 2/14)
7 Chiapasco 1999 G1: AUG (part; ramus/chin) + MEMNRES G1: membrane exposure (n = 2/30) − early removal of the
G2: AUG (block; ilium/chin/calvaria) membrane & secondary healing 6%
Donor site: transient paraethesia of the lower lip (n = 8), transient
paraethesia of the frontal mandibular teeth (n = 12), transient
alteration of de-ambulation (all patients receiving bone
harvesting from iliac crest) ) n = 20/30 66%
8 Cordaro 2002 G1: AUG (block; ramus) G1: no wound dehiscence or infection n = 1/15 7%
G2: AUG (block; symph) G2: no wound dehiscence or infection
Donor site: temporary numbness in lower incisor area (n = 1),
ecchymosis, pain, swelling
9 Cordaro 2011 G1: AUG (block; ramus) + XEN G1: wound dehiscence (n = 3/8) − partial removal of the
(part) + MENRES membrane and bone graft material in 2 cases & secondary
G2: AUG (block; ramus) healing.
G2: wound dehiscence (n = 1/9) − chlorhexidine rinse &
secondary healing ) n = 4/17 24%
10 Da Costa 2011 G1: ALG (block) + bone marrow aspirate No information
G2: ALG (block)
11 de Freitas 2013; 2016 G1: AUG (part; ramus) + MEMNRES Facial swelling, local edema and/or erythema were more
G2: collagen sponge (origin frequent and severe in the recipient site of the G2 compare to
NR) + rhBMP2 + MEMNRES the G1. n = 1/24 4%
Donor site (G1): temporary discomfort, pain
12 Eskan 2014 G1: ALG (part) + MEMRES “Membrane exposure was considered a possible postoperative
G2: ALG (part) + PRP + MEMRES healing event and was not reported as an adverse event”
13 Kheur 2018 G1: ALG (small- or large-part) + RS possibly no complication
G2: ALP (small- or large-part) + RS
14 Maiorana 2005 G1: AUG (block+part; chin/ramus) + XEN G1: infection (n = 2/11) − Non-specified treatment
(part) G2: infection (n = 1/12) − Non-specified treatment ) n = 3/23
G2: AUG (block + part; chin/ramus) 13%
15 Mazzocco 2011 G1: XEN (part) + ALG (part) + MEMRES No information
G2: RE + XEN (part) + ALG
(part) + MEMRES
16 Meijndert 2008; 2017 G1: AUG (block/part; chin) No complication
G2: AUG (block/part; chin) + MEMRES
G3: XEN (part) + MEMRES
(Continues)
NAENNI et al. |
297
TA B L E 4 (Continued)
Nr Study Tx Type of Complication & treatment
17 Meijndert 2005 G1: AUG (block/part; chin) No complication

G2: AUG (block/part; chin) + MEMRES
G3: XEN (part)+ MEMRES
18 Monje 2014 G1: AUG (block, calv) + XEN Not reported
(part) + MEMRES
G2: AUG (block, ilium) + XEN
(part) + MEMRES
19 Mordenfeld 2014; 2017 G1: AUG/XEN 90:10 (part, ramus) + fibrin G1: wound dehiscence (n = 4/6) − non-specified treatment
glue + MENRES G2: wound dehiscence (n = 3/7) − non-specified treatment
G2: AUG/XEN 60:40 (part, ramus) + fibrin n = 7/13 54%
glue + MENRES
20 Mordini 2016 G1: ALG (part) + MEMRES G2: membrane exposure (n = 2/6) − membrane removal 33%
G2: ALG (part) + MEMNRES
21 Pourabbas 2007 G1: AUG (block, chin) + MEMRES “without any major complications at the recipient or donor sites”
G2: AUG (block, ramus) + MEMRES
22 Shalash 2013 G1: ALP (part) + MEMNRES G1: membrane exposure (n = 1/10) − membrane removal
G2: ALP (part) + XEN (part) + MEMNRES n = 2/20 10%
G2: membrane exposure (n = 1/10) − membrane removal
23 Thoma 2018 G1: XEN (block) + rhBMP2 + MEMRES G2: bone block exposure − surgical removal
G2: AUG (block; ramus, chin) + XEN
(part) + MEMRES
24 Thor 2005; Dasmah 2012; G1: AUG (block; ilium) “two patients experienced localized (posterior sites) infections
Dasmah 2013 G2: AUG (part; ilium) + PRP that resolved with local drainage and oral clindamycin” n = 2/19
10%
25 Urban 2011 G1: AUG (part; ramus, surgical site) Possibly no complication in the recipient site of G1 and G2
G2: AUG (part; ramus, surgical site) + XEN Donor site: no major complication, such as hemorrhage,
(part) postoperative infection, mandibular fracture and neurosensory
dysfunction
BW, bone width; BQ, bone quality; CLIN, clinical; CoI, Conflict of Interest; Compl, Complication; FU, follow-up after lateral augmentation; ImpIns,
implant insertion; Imp success, Implant success; Imp survival,Implant survival; NR, not reported; RAD,radiography.
respect to the inclusion/exclusion criteria. Of these, 41 articles number of patients) with a mean age of 43.9 years (data from 18
were excluded for various reasons, leaving 30 papers pertaining studies reporting on age), of whom 42.2% were male (from 18 stud-
to 25 unique studies for inclusion in the present systematic review ies reporting on sex). The recipient sites of the included studies were
(Supporting Information Appendix S3). All but one included articles as follows: nine studies solely in the maxilla (six in the anterior area
reported on feasibility of implant placement after healing times and three in either anterior or posterior areas), three studies solely in
ranging from 3.0 to 8.1 months, whereas four out of these studies the posterior mandible, and seven studies investigated in both jaws,
followed the inserted implants up to 10 years (Table 1). whilst 6 studies did not report on the region investigated. After pri-
mary lateral augmentation, sites had been left for healing between
3.0 and 8.1 months before re-entry/ implant placement (Table 3).
3.2 | Study characteristics
At least 761 implants (from 15 studies reported on the number of
The characteristics of the 25 included studies are descriptively ana- implants) were placed in at least 574 sites (reported in 25 studies).
lysed in Table 2. Sixteen studies (64%) were randomized clinical trials Twenty-three studies reported clinically and/or radiographically
(13 with parallel and three with within-patient randomization), while measured bone width changes. Implant feasibility was reported in
the remaining nine (36%) were non-randomized comparative studies all but in one study, complications in seventeen studies (Table 4),
(five explicitly prospective and four with unclear design). Most stud- implant survival rate in four studies, implant success rate in three
ies were performed in a university setting (n = 16; 64%), followed studies, PROMS and marginal bone loss in two studies. The other
by hospitals (n = 3; 12%), a private practice (n = 1; 4%) and a com- outcomes irrelevant to the present review included bone quality,
bination of university and private practice (n = 1; 4%). The included implant stability quotient (ISQ) value, peri-implant health aesthetic
studies were conducted in fifteen different countries, encompassing index, bone microstructure, bone volume, histologic outcome and
a patient population of 553 individuals (all studies reported on the soft tissue volume.
|
298 NAENNI et al.
F I G U R E 3 a, Contour-enhanced forest plot on the amount of bone width gained from lateral augmentation. AUG, autologous graft CI,
confidence interval; MD, mean difference; WO/W, with/without; XEN, xenograft. b, Contour-enhanced forest plot on the amount of graft
resorption during lateral augmentation. AUG, autologous graft; CI, confidence interval; MD, mean difference; WO/W, with/without; XEN,
xenograft
risk of bias for any of the assessed domains (Eskan et al., 2014; de
3.3 | Risk of bias within studies
Freitas et al., 2016) and were thus judged to have an overall unclear
The risk of bias for the included randomized trials was assessed risk of bias. Only three studies presented more than 50% of low
using the Cochrane Risk-of-Bias tool and is presented in Figure 2a risk in all seven domains (Badr et al., 2010; Eskan et al., 2014; de
and Supporting Information Appendix S4a. None of the included Freitas et al., 2016) and eight studies presented more than 50% of
studies had an overall low risk of bias. Only two studies had no high unclear risk (Aarújo, 2016; Antoun, Sitbon, Martinez, & Missika,
NAENNI et al. |
299
TA B L E 5 (a) Results of the pooled analyses for all available studies (b) Results of the pooled analyses only for the 7 studies having data for
all 3 assessed time-points (pre-augmentation, post-augmentation, and re-entry).
Outcome Studies Average (95% CI) I2 (95% CI) 95% Prediction
(a)
Baseline width 14 2.96 mm (2.69 to 3.23 mm) 91% (83% to 97%) 1.86 to 4.06 mm
Post-augmentation width 8 7.68 mm (7.11 to 8.24 mm) 96% (89% to 99%) 5.64 to 9.71 mm
Re-entry width 13 6.36 mm (5.73 to 6.99 mm) 98% (95% to 99%) 3.77 to 8.95 mm
Width gain 20 3.30 mm (2.81 to 3.79 mm) 97% (94% to 98%) 0.96 to 5.65 mm
Graft resorption 10 −1.33 mm (−1.78 to −0.88 mm) 93% (93% to 98%) −2.98 to 0.32 mm
(b)
Baseline width 7 3.00 mm (2.82 to 3.18 mm) 46% (0% to 88%) 2.52 to 3.47 mm
Post-augmentation width 7 7.89 mm (7.40 to 8.38 mm) 90% (67% to 98%) 5.64 to 9.71 mm
Re-entry width 7 6.45 mm (6.02 to 6.89) 85% (61% to 96%) 4.97 to 7.94 mm
Width gain 7 3.43 mm (2.83 to 4.03 mm) 92% (78% to 98%) 1.31 to 5.55 mm
Graft resorption 6 −1.54 mm (−2.17 to -0.91 mm) 92% (74% to 99%) −3.77 to 0.69 mm
2001; Caldwell, Mills, Finlayson, & Mealey, 2015; Da Costa et al., were re-analysed with generalized linear regression models and are
2011; Kheur et al., 2018; Mazzocco, Nart, Cheung, & Griffin, 2011; reported in full in Supporting Information Appendix S5-S13.
Pourabbas & Nezafati, 2007; Thoma et al., 2018). The most prob-
lematic domains with high risk of bias were missing or incomplete
3.5 | Synthesis of results
blinding, incomplete outcome data and other sources of bias.
Additionally, none of the identified studies had a pre-defined reg- Overall, bone width pre-augmentation measured 2.96mm (95% CI:
istered protocol and the risk of selective outcome reporting bias 2.69 to 3.23; I2: 97%), whilst bone width post-augmentation meas-
was unclear in all publications. ured 7.68mm (95% CI: 7.11 to 8.24; I2: 96%). Due to graft resorption
The risk of bias for the included non-randomized studies accord- (-1.33mm (95% CI: -1.78 to -0.88; I2: 93%), bone width at re-entry
ing to the ROBINS-I tool is presented in Figure 2b and Supporting resulted in a mean of 6.36 mm (95% CI: 5.73 to 6.99 mm; I2: 98%).
Information Appendix S4b. None of the studies were in a low risk of Thus, an overall mean bone width gain of 3.30mm (95% CI: 2.81
bias: one had signs of moderate bias (Beitlitum, Artzi, & Nemcovsky, to 3.79; I2: 97%) could be achieved (Table 1e). Data resulting from
2010), five of serious bias (Barbu et al., 2016; Cordaro et al., 2002; re-analysis of the seven studies that provided raw data (Aarújo,
Maiorana, Beretta, Salina, & Santoro, 2005; Shalash et al., 2013; Urban 2016; Chiapasco, et al., 1999; Cordaro, et al., 2011; Eskan, et al.,
et al., 2011) and three of critical bias (Chiapasco, Abati, Romeo, & 2014; Maiorana, et al., 2005; Mordenfeld, Johansson, Albrektsson
Vogel, 1999; Dasmah, Thor, Ekestubbe, Sennerby, & Rasmusson, 2013; & Hallman, 2014; Thoma, et al., 2018) resulted in very similar pooled
Monje et al., 2014). The domain “bias due to confounding” was a source averages (Table 1f). Only a limited number of factors could finally
for critical risk of bias for three studies with overall critical risk of bias be analysed with random-effects meta-analyses: (a) a comparison
(Chiapasco et al., 1999; de Freitas et al., 2016; Monje et al., 2014). The pertaining to the use of an autologous graft (AUG) versus a xeno-
domain “bias due to missing data” was judged as unclear in all studies. graft (XEN) mixed with/without AUG (Aarújo, 2016; Barbu et al.,
The domains “bias due to confounding” and “bias in measurement of 2016; Cordaro et al., 2011; Maiorana et al., 2005; Urban et al., 2011)
outcomes” were judged as serious in more than 50% of the studies. (Table 4, Figure 3a, b) and (b) a comparison for lateral bone augmen-
tation with or without the use of a membrane (Antoun et al., 2001;
Chiapasco et al., 1999) (Table 5; Figure 3a, b).
3.4 | Results of individual studies and
Meta-analysis on the use of XEN (with/without AUG) com-
synthesis of results
pared to the use of AUG alone (Table 4, Figure 3a) did not find a
In general, a wide variability regarding materials and techniques statistically significant difference in the amount of BW gain (Aarújo,
investigated as well as parameters reported was observed in the 2016; Barbu et al., 2016; Cordaro et al., 2011; Maiorana et al.,
included studies (Tables 2 and 3). All studies gave to some ex- 2005; Urban et al., 2011) (MD: 0.32 mm; 95% CI: −0.19 to 0.83 mm;
tent aggregate data of their results, while nine studies (four rand- p > 0.05). On the other hand, sites augmented with XEN (with/with-
omized [Aarújo, 2016; Caldwell et al., 2015; Meijndert, Raghoebar, out AUG) showed significantly less absolute graft resorption com-
Schupbach, Meijer, & Vissink, 2005; Thoma et al., 2018] and five pared to sites augmented with AUG alone (MD: 1.06 mm; 95% CI:
non-randomized [Barbu et al., 2016; Chiapasco et al., 1999; Maiorana 0.21–1.92 mm; p = 0.01) (Figure 3b), with moderate heterogeneity
et al., 2005; Monje et al., 2014; Urban et al., 2011]) also provided raw across studies (I2: 63%). This was also reflected by graft resorption
data in tabular form within the published report. Available raw data compared to the amount of bone graft added to the site, where sites
|
300 NAENNI et al.
augmented with XEN (with/without AUG) resulted in 11.6% less re-

3.6 | Additional analyses and risk of bias
sorption of the added graft compared to AUG alone (MD: 11.6%;
across studies
95% CI: 5.2%–18.1%; p < 0.001).
The use of a membrane with lateral augmentation could be A large number of subgroup and meta-regression analyses had been
assessed in a meta-analysis of two studies (Antoun et al., 2001; initially planned, but could ultimately not be conducted (Supporting
Chiapasco et al., 1999) (Table 5). No statistically significant advan- Information Appendix S2). The only subgroup analysis that could be
tage could be observed regarding neither BW gain (MD: −0.33 mm; performed pertained to the meta-analysis of BW gain with the use
95% CI: −2.24 to 1.58 mm; p > 0.05) nor absolute graft resorption of an autologous (AUG) versus a xenograft (XEN) mixed with/with-
(MD: 0.84 mm; 95% CI: −1.42 to 3.09 mm; p > 0.05) with the use of a out AUG (Table 6; 5 studies). No statistically significant differences
membrane compared to no membrane (Figure 3a, b). could be found when comparing (a) the use of XEN versus AUG and
Furthermore, meta-analysis of the regression coefficients the use of XEN-AUG versus AUG or (b) the use of a membrane or
from the re-analyzed raw data of each study listed in Supporting not.
Information Appendices S5–S13 was used to assess the effect on A sensitivity analysis was conducted by excluding non-
the outcome of lateral augmentation of the following patient-, graft- randomized studies and including only randomized studies in the
or surgery-related factors: age, sex, jaw, region, baseline BW, heal- analysis (Supporting Information Appendix S15). The comparisons
ing time and donor site for AUG (Supporting Information Appendix of lateral bone augmentation with AUG versus a xenograft XEN
S14a–c). Regarding BW gain, the single predicting factor was base- mixed with/without AUG were relatively robust to the inclusion of
line BW, where sites with greater pre-augmentation BW tended to only randomized trials, as still no statistically significant differences
have less benefit in terms of BW gain (pooled effect: −0.35 mm; were found. On the other side, sensitivity analysis on membrane
95% CI: −0.63 to −0.07 mm; p = 0.01) (Table 5 & Supporting use gave slightly different results. For one, now the BW gain when
Information Appendix S14a). As far as absolute bone graft resorp- using a membrane was greater compared to no membrane use,
tion is concerned, no significant modifying factor was identified, Nevertheless, this difference did not reach statistical significance.
although there was a trend for maxillary sites to present greater This was in contrast to the original analysis that showed the opposite
post-augmentation resorption compared to mandibular sites (pooled (MDs of −0.33 mm for the original and 0.80 mm for the sensitivity
effect: −0.21 mm; 95% CI: −0.47 to 0.05 mm; p = 0.12) (Supporting analysis). Additionally, membrane use was associated with a signifi-
Information Appendix S14b). Finally, as far as % bone graft resorption cantly lower amount of post-augmentation graft resorption com-
of the added graft is concerned, BW at baseline was again inversely pared to no use of membrane (MD: 2.0 mm; 95% CI: 1.3–2.8 mm;
associated with BW gain. Increased baseline BW of the site was as- p < 0.001), which was contrary to the original analysis from random-
sociated with increased % resorption of the added graft (pooled ef- ized and non-randomized studies.
fect: −4.7%; 95% CI: −9.9 to 0.5%; p = 0.07) (Supporting Information The quality of evidence from all meta-analyses was finally gauged
Appendix S14c). Additionally, patient age had a significant influence, with the GRADE approach, using the original analyses, except for
with older patients presenting higher % resorption of the added graft the case of bone graft resorption with/without membrane, where
(pooled effect: −0.4% per year; −0.6% to −0.1% per year; p = 0.01). the sensitivity analysis was used (Table 7). All meta-analyses were
TA B L E 6 Random-effects meta-analyses on use of xenograft for lateral augmentation (experimental) compared to use of autograft
(reference)
Outcome Studies MD (95% CI) p I2 (95% CI) τ 2 (95% CI) 95% prediction
BW gain 5 0.32% (−0.19% to 0.22 55% (0%–95%) 0.18 (0–2.59) −1.26 to 1.90
0.83%)
Graft resorption 3 1.06% 0.01 63% (0%–99%) 0.36 (0–28.27) −8.32 to 10.44
(0.21%–1.92%)
% bone graft resorption 2 11.62% <0.001 0% (NC) 0 (NC) NC
from the added graft (5.15%–18.09%)
BW: bone width; CI: confidence interval; MD: mean difference; NC: non calculable.
TA B L E 7 Random-effects meta-analyses on membrane use during lateral augmentation (experimental) compared to no membrane use
(reference)
Outcome Studies MD (95% CI) p I2 (95% CI) τ 2 (95% CI) 95% prediction
BW gain 2 −0.33 (−2.24 to 1.58) 0.74 71% (NC) 1.38 (NC) NC

Graft resorption 2 0.84 (−1.42 to 3.09) 0.47 96% (NC) 2.53 (NC) NC
BW: bone width; CI: confidence interval; MD: mean difference; NC: non calculable.
NAENNI et al. |
301
judged to provide very low quality of evidence due to the inclusion 4 | D I S CU S S I O N
of non-randomized studies, the high risk of bias of both randomized
and non-randomized included studies and imprecision due to lim- 4.1 | Summary of evidence
ited sample sizes. The only exception was the sensitivity analysis as-
The literature search yielded a total of 25 studies, of which 16 were
sessing bone graft resorption with or without the use of membrane,
randomized and nine were non-randomized comparative studies in-
where moderate quality of evidence favoured membrane use—with
cluding at least 553 patients (mean age 43.9 years/42.2% male) and
the only limitation being the high risk of bias in the included random-
at least 761 dental implants that had been placed after a healing time
ized trial.
of 3.0–8.1 months following primary horizontal augmentation.
TA B L E 8 Subgroup analysis for the

p for subgroup
meta-analysis on use of xenograft for
Factor Subset Studies MD (95% CI) interaction
lateral augmentation (experimental)
compared to use of autologous graft XEN mixed with No 1 1.09 (0.35–1.84) 0.22
(reference) and with the outcome of bone AUG Yes 4 0.14 (−0.30 to 0.57)
width gain
Membrane used No 3 0.51 (−0.26 to 1.28) 0.29
Yes 2 −0.06 (−0.60 to
0.47)
AUG: autologous graft CI: confidence interval; MD: mean difference; XEN: xenograft.
TA B L E 9 Summary of findings table according to the Grades of Recommendations, Assessment, Development, and Evaluation (GRADE)
approach
Anticipated absolute effectsa (95% CI)

Outcome Quality of the evidence
Studies (sites) CTR EXP Difference (GRADE)c What happens
AUG XEN w/o AUG

Bone width gain 4.2 mm – 0.3 mm more gain ⨁◯◯◯ very lowd,e Little to no difference in
5 studies (115 sites) (0.2 mm–0.8 due to bias; imprecision gained bone width
more)
Bone graft resorption −1.4 mm – 1.1 mm less ⨁◯◯◯ very lowd,e Probably less graft
3 studies (67 sites) resorption due to bias resorption with XEN w/o
(0.2–1.9 mm less) AUG
% bone graft resorption −26.4% – 11.6% less of the ⨁◯◯◯ very lowd,e Probably less % resorption
from the added graft added graft due to bias of the added graft with
2 studies (45 sites) resorbed XEN w/o AUG
(5.2%–18.1% less)
No Membrane
membrane
Bone width gain 3.5 mm – 0.3 mm less gain ⨁◯◯◯ very lowd Little to no difference in
2 studies (43 sites) (2.2 mm less to due to bias gained bone width
1.6 mm more)
Bone graft resorption −1.8 mm – 2.0 mm less ⨁⨁⨁◯ moderatef Less graft resorption with
1 study (12 sites) resorption due to bias the use of a membrane
(1.3–2.8 mm less)
Factors associated with outcome of lateral bone augmentation.

Population and intervention: partially/fully edentulous adult patients with resorbed alveolar bone receiving lateral bone augmentation prior to dental
implant treatment.
Settings: university clinics and private practice.
CI: confidence interval; CTR: control category; EXP: experimental category; GRADE: Grading of Recommendations Assessment, Development and
Evaluation; MD: mean difference.
a
The basis for the risk in the control group (e.g., the median control group risk across studies) is provided in footnotes. The risk in the intervention group
(and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
b
Response in the control group is based on pooled Paule–Mandel meta-analysed effect in the control group. cGRADE for both randomized and non-
randomized studies starts from “high.” dDowngraded initially to “low” due to the inclusion of non-randomized studies; further downgraded to very low
for lack of blinding serious limitations (high risk of bias). eImprecision also identified due to the limited sample size finally included; however, GRADE is
already at very low. f The single identified trial was in high risk of bias due to incomplete reporting of data, downgraded by one.
|
302 NAENNI et al.
The results of the current systematic showed that (a) feasibility implant diameter and length—should be determined preceding pri-
of implant placement was achieved in the majority of the studies, mary bone augmentation (Chiapasco & Casentini, 2018). Based on
although sometimes requiring additional GBR at the time of implant the literature search, only one study reported specifically on pre-
placement; (b) bone width (BW) gain was significantly influenced by augmentation implant planning (Eskan et al., 2014) (Table 1). In that
BW at baseline; (c) age revealed a significant association with bone study, the predictability of planned implant placement reached 100%
graft resorption favouring younger patients; (d) groups using xeno- in one group and 93% in the other group. Eleven studies reported
graft showed significantly less graft resorption compared to autolo- on implant diameter, although without giving further information
gous alone, but did not lead to a statistically significant difference in regarding the placement thereafter. The rest of the included stud-
BW gain; and (e) the use of a membrane did not result in superior BW ies did not provide information regarding implant diameter. Authors
gain compared to augmentation without a membrane. used terms such as “proper implant placement” and “suitable sized
implant.” However, these only give unclear information. The above-
described main findings were not reported in a way that allowed for
4.1.1 | Bone width gain
proper statistical analysis, but are reported in a descriptive manner
This systematic review revealed that primary horizontal ridge aug- within this investigation. Hence, the original primary aim of this sys-
mentation prior to implant placement was successful in terms of BW tematic review had to be changed to BW gain and graft resorption.
gain. Nevertheless, some studies reported on the need for additional
bone augmentation at the time of implant placement (Badr, et al.,
4.1.3 | Graft resorption
2010; Beitlitum, et al., 2010; Chiapasco, et al., 1999; Kheur, et al., 2018;
Meijndert, et al., 2005) or the need for the placement of a narrower Meta-analysis of the regression coefficients indicated a relationship
implant diameter than planned (Eskan, et al., 2014). In case of insuf- between % graft resorption and patient age (p < 0.01; Supporting
ficient ridge width after bone augmentation, the placement of a smaller Information Appendix S14c). Age has been controversially discussed
diameter implant or a simultaneous bone augmentation might be con- as a factor influencing implant success (Bartold, Ivanovski, & Darby,
sidered. Interestingly, BW at baseline (prior to augmentation) was sig- 2016; Ikebe, Wada, Kagawa, & Maeda, 2009). Presumably, older
nificantly inversely correlated with the obtained BW gain (Supporting patients may exhibit poorer local bone condition, require longer
Information Appendix S14a), which indicates that the thinner the alveo- healing periods and display more systematic diseases related to
lar process at baseline was, the more BW gain could be achieved. This healing potential in general. Based on the meta-analysis, this review
observation might be due to two potential reasons: (a) the resorptive revealed that every additional year of age at the time of primary
pattern may tend to follow the natural anatomy of the original ridge augmentation led to 0.05 mm more absolute resorption of the aug-
and/or (b) the more ridge width is present prior to augmentation, the mented bone (Supporting Information Appendix S14b). For example,
less bone graft material is applied. a 60-year-old patient would experience 1 mm more graft resorption
One study demonstrated that bone augmentation within the post-augmentation compared to a 40-year-old patient. This age-
bony envelope is more predictable compared to that outside of the dependent resorptive pattern could perhaps become more clinically
bony envelope (Tinti & Parma-Benfenati, 2003). Thus, when over- relevant due to current aged or ultra-aged society (Muramatsu &
building is performed out of the ridge anatomy, one should consider Akiyama, 2011).
the possibility of future resorption. However, there has been no Regarding the outcomes for the recipient jaw, results were po-
long-term study to investigate this issue. tentially in favour of the mandible. Thus, absolute graft resorption
Moreover, when performing a ridge augmentation procedure, it is was more pronounced in the maxilla (difference: −0.21 mm; 95%
important to know what amount of graft material is required depending CI: −0.47 to 0.05 mm), even though this was not statistically signifi-
on the prosthetic planning (Chiapasco & Casentini, 2018). Within the cant (p = 0.12), probably due to low power (Supporting Information
present review, most of the authors seemed to aim at placing regular Appendix S14b). Possible reasons behind this might include the level
diameter implants ranging from 3.5 to 4.1 mm (Table 1), which may in- of surgical difficulty especially in the posterior maxilla, extensive
dicate that 6–7 mm of BW could be considered sufficient for implant overbuilding in the anterior maxilla or uneven pressure of soft tissue
placement. If, for example, 4 mm of residual BW is present (reported as after closure of the flap. One might initially expect that extensively
baseline BW in some studies), there might be no reason to apply a great resorbed mostly cortical mandibles might yield unfavourable aug-
amount of bone graft material. As a consequence, the anticipated final mentation results due to reduced vascularization. However, radio-
BW has to be taken into account when performing ridge augmentation graphic evidence indicates that the blood vessels remain within the
procedures, without disregarding expected resorptive processes. central canals of resorbed osteons and even the densest-appearing
cortex is actually porous (Atwood, 1963).
4.1.2 | Implant feasibility
4.1.4 | Presence of xenograft
Implant feasibility reflects the most important clinically relevant
parameter in assessing the success of primary bone augmentation. The presence of a xenograft leads to a decreased graft resorption com-
Ideally, implant planning—such as prosthetically ideal positioning, pared to autologous alone. These results seem somehow contradictory,
NAENNI et al. |
303
originating from five studies for BW gain (showing no significance) might indicate greater measurement accuracy for the radiographic
compared to three studies having been analysed for graft resorption. method. Additionally, radiographic evaluation of bone width enables
Three authors (Barbu et al., 2016; Cordaro et al., 2011; Maiorana et al., a more comprehensive assessment of the alveolar width at differ-
2005) have used the autologous block with particulated xenogenic ent heights. For example, the same study (de Freitas et al., 2013)
graft material (“added at the periphery,” “in order to fill gaps and de- reported considerable differences in bone width gain according to
fects” and “over the graft”), whereas one study had used particulated height below the alveolar crest (pooled gains: 1.00 ± 0.94 mm 2 mm
autologous mixed with particulated xenograft in a 1:1 ratio (Urban below crest, 2.90 ± 0.83 mm 6 mm below crest and 1.75 ± 0.98 mm
et al., 2011) and one had used a xenograft block alone (Aarújo, 2016). 10 mm below crest). This indicates a possibly uneven pattern of
Sites augmented with xenogeneic graft materials with or without au- post-augmentation resorption and final bone width gain that might
tologous bone particles XEN (with/without AUG) revealed 11.6% less have a direct clinical relevance and is therefore an interesting aspect
resorption compared to sites augmented with autologous grafts alone for future research.
AUG (MD: 11.6%; 95% CI: 5.2%–18.1%; p < 0.001) (Table 4). These re-
sults support the assumption that the presence of a xenograft prevents
4.2 | Strengths and limitations
from bone resorption. This might be due to its slow resorption rate and
thus long standing time (Schlegel & Donath, 1998; Skoglund, Hising, & The strengths of this systematic review consist of the registra-
Young, 1997). tion of its a priori protocol in PROSPERO (Sideri, Papageorgiou,
& Eliades, 2018), its exhaustive literature search, its improved
analytical methods (Veroniki et al., 2016), the use of the GRADE
4.1.5 | Use of membrane
approach (Guyatt et al., 2011) to assess the quality of the meta-
Although the use of a membrane is supposed to prevent soft tissue evidence and the transparent provision of the study's data (Naenni
ingrowth (Dahlin, Linde, Gottlow, & Nyman, 1988; Kostopoulos & et al., 2018). Additionally, only randomized and prospective non-
Karring, 1994), augmentation without membrane coverage did not randomized comparative studies were included in this systematic
demonstrate a significant differences in terms of BW gain and graft review, which are less biased than retrospective non-r andomized
resorption within this study. The two studies investigating the use of studies.
a non-resorbable membrane (Antoun et al., 2001; Chiapasco et al., However, certain limitations do exist mainly due to the vast
1999) applied an autologous block bone graft without a membrane methodological and clinical heterogeneity of the identified stud-
and compared this treatment with either a GBR procedure (particu- ies. First, the initially planned primary outcome of the review
lated autologous bone and membrane) or an autologous block graft pertaining to implant feasibility was discarded as the primary
with membrane coverage. These results were to some extent unex- outcome, since no study but one transparently reported on de-
pected. A wide body of evidence exists on the procedure of bone viations from implant size that was originally planned and im-
augmentation and the use of membranes in order to maintain space, plant size that was eventually inserted. Results on feasibility are
prevent soft tissue ingrowth and to help stabilize the augmented reported as secondary outcomes and in a descriptive manner.
area. This may have played a minor role in the two before-mentioned Second, data on post-augmentation BW gain were measured in
studies, as bone block grafts were used. The cortical layer of the the identified studies (Beitlitum et al., 2010; de Freitas et al., 2013,
bone block might be sufficient to prevent soft tissue ingrowth and 2016; Mazzocco et al., 2011) either clinically or radiographically,
resist to resorption. The role of a barrier membrane is to prevent although hints exist that the measurement method might have a
soft tissue ingrowth and give stability to the graft. The cortical plate direct influence on the observed BW. Third, a few secondary out-
of the block bone may not allow soft tissue ingrowth. Additionally, comes could not be statistically analysed due to study character-
the fixating screws may have given the bone block graft sufficient istics and thus are only descriptively reported within this review.
stability. Evidence regarding no need to use a membrane shall be Peri-implant health would have been an outcome of most interest,
perceived while keeping this in mind. especially in implants placed after primary lateral augmentation.
Unfortunately, only two of the included studies investigated this
parameter (Cordaro et al., 2011 & Meijndert CM et al., 2017).
4.1.6 | Bone width measurements
Furthermore, methodological issues existed for all included stud-
Bone width gain was measured in the included studies either clini- ies, as has been often reported for clinical trials in implant den-
cally and/or radiographically, while only one study transparently tistry (Papageorgiou, Kloukos, Petridis, & Pandis, 2015), and these
provided results of both clinical and radiographic measurements might have influenced the review's results. This is especially the
(de Freitas et al., 2013). Although small discrepancies were found case for included non-r andomized studies that were not clearly
between overall clinical and radiographic gain (pooled gains of retrospective and not clearly prospective—even though a sensi-
3.45 ± 1.18 mm vs. 2.90 ± 0.83 mm, respectively), these were not tivity analysis of only randomized studies indicated robustness
statistically significant. However, both the bone gain magnitude and of the results. Furthermore, the identified studies reported pre-
the variation of the measurements (seen through the standard devi- dominantly on small sample sizes and this might have introduced
ation) were consistently lower radiographically than clinically, which small-s tudy effects (Cappelleri et al., 1996). Also, analysis was
|
304 NAENNI et al.
performed on augmentation site level, which ignores clustering ef- ORCID

fects and might lead to information loss, except for studies where
Nadja Naenni https://orcid.org/0000-0002-6689-2684
full data were openly available and were analysed appropriately
(Supporting Information Appendix S5–S13). The limited number Hyun‐Chang Lim https://orcid.org/0000-0001-7695-1708
of included studies and their suboptimal reporting did not enable Spyridon N. Papageorgiou https://orcid.
robust assessments of heterogeneity, as well as the conduct of org/0000-0003-1968-3326
several analyses for subgroup, and small-s tudy effects that were Christoph H. F. Hämmerle https://orcid.
planned. Finally, no formal assessment of reporting biases was org/0000-0002-8280-7347
possible due to the small number of included studies and the lack
of published trial protocols, even though our comprehensive and
unrestricted literature search might partially safeguard against re- REFERENCES
porting bias. Aarújo, P. M. (2016) Análise clínica de substituto ósseo em bloco para
aumento horizontal e influência no sucesso dos implantes osseointe-
grados- estudo em humanos Odontologia do Centro de Ciências. Brazil:
Saúde da Universidade Federal de Santa Catarina.
5 | CO N C LU S I O N S
Aghaloo, T. L., & Moy, P. K. (2007). Which hard tissue augmentation tech-
niques are the most successful in furnishing bony support for implant
Based on the results from the included randomized and prospective placement? International Journal of Oral and Maxillofacial Implants,
non-randomized trials, primary lateral bone augmentation leads to 22(Suppl), 49–70.
Antoun, H., Sitbon, J. M., Martinez, H., & Missika, P. (2001). A pro-
sufficient bone width and allows for subsequent implant placement
spective randomized study comparing two techniques of bone
in most cases. augmentation: Onlay graft alone or associated with a mem-
In general, initially smaller bone dimension is associated with larger brane. Clinical Oral Implants Research, 12, 632–639. https://doi.
bone width gain, which indicates that a severe lateral bone deficiency org/10.1034/j.1600-0501.2001.120612.x
can be effectively augmented applying primary lateral bone augmen- Badr, M., Coulthard, P., Alissa, R., & Oliver, R. (2010). The efficacy of
platelet-rich plasma in grafted maxillae. A randomised clinical trial.
tation. Patients’ age and recipient site (maxilla or mandible) seem to
European Journal of Oral Implantology, 3, 233–244.
influence graft resorption. The addition of a xenograft can be helpful Barbu, H. M., Andreescu, C. F., Lorean, A., Kolerman, R., Moraru, L.,
for reducing graft resorption. However, the quality of clinical recom- Mortellaro, C., & Mijiritsky, E. (2016). Comparison of two techniques
mendation that can be drawn at the present time ranges from moder- for lateral ridge augmentation in mandible with ramus block graft.
Journal of Craniofacial Surgery, 27, 662–667. https://doi.org/10.1097/
ate to very low, due to the limited number of existing studies and their
SCS.0000000000002561
methodological shortcomings. Bartold, P. M., Ivanovski, S., & Darby, I. (2016). Implants for the aged patient:
Biological, clinical and sociological considerations. Periodontology,
2000(72), 120–134. https://doi.org/10.1111/prd.12133
AC K N OW L E D G E M E N T S Beitlitum, I., Artzi, Z., & Nemcovsky, C. E. (2010). Clinical evaluation
of particulate allogeneic with and without autogenous bone grafts
The authors thank Daniel S. Thoma (University of Zurich) for provid- and resorbable collagen membranes for bone augmentation of atro-
ing additional data from their study. phic alveolar ridges. Clinical Oral Implants Research, 21, 1242–1250.
https://doi.org/10.1111/j.1600-0501.2010.01936.x
Benic, G. I., & Hammerle, C. H. (2014). Horizontal bone augmentation
by means of guided bone regeneration. Periodontology, 2000(66),
C O N FL I C T O F I N T E R E S T 13–40. https://doi.org/10.1111/prd.12039
Caldwell, G. R., Mills, M. P., Finlayson, R., & Mealey, B. L. (2015). Lateral
Dr.Lim had received an osteology scholarship in 2017. The rest of
alveolar ridge augmentation using tenting screws, acellular dermal
the authors have stated explicitly that there are no conflicts of inter- matrix, and freeze-dried bone allograft alone or with particulate au-
est in connection with this article. togenous bone. International Journal of Periodontics and Restorative
Dentistry, 35, 75–83. https://doi.org/10.11607/prd.2260
Cappelleri, J. C., Ioannidis, J. P., Schmid, C. H., de Ferranti, S. D., Aubert,
AU T H O R S ’ C O N T R I B U T I O N S M., Chalmers, T. C., & Lau, J. (1996). Large trials vs meta-analysis of
smaller trials: How do their results compare? JAMA, 276, 1332–1338.
Nadja Naenni was involved in protocol development, inclusion/ https://doi.org/10.1001/jama.1996.03540160054033
Carrasco-Labra, A., Brignardello-Petersen, R., Santesso, N., Neumann, I.,
exclusion of studies, drafted the article and critically revised and
Mustafa, R. A., Mbuagbaw, L., … Schunemann, H. J. (2016). Improving
approved the article. Hyun‐Chang Lim was involved in protocol de- GRADE evidence tables part 1: A randomized trial shows improved
velopment, inclusion/exclusion of studies and drafted the article. understanding of content in summary of findings tables with a
Spyridon N. Papageorgiou was involved in protocol development, new format. Journal of Clinical Epidemiology, 74, 7–18. https://doi.
collected data, analyzed and interpreted the data and critically re- org/10.1016/j.jclinepi.2015.12.007
Chappuis, V., Rahman, L., Buser, R., Janner, S. F. M., Belser, U. C., & Buser,
vised and approved the article. Christoph H. F. Hämmerle was in-
D. (2018). Effectiveness of contour augmentation with guided bone
volved in protocol development and critically revised and approved regeneration: 10-year results. Journal of Dental Research, 97, 266–
the article. 274. https://doi.org/10.1177/0022034517737755
NAENNI et al. |
305
Chiapasco, M., Abati, S., Romeo, E., & Vogel, G. (1999). Clinical out- or autogenous bone grafts: A proof-of-concept randomized clini-
come of autogenous bone blocks or guided bone regeneration with cal trial. Journal of Clinical Periodontology, 40, 968–975. https://doi.
e-PTFE membranes for the reconstruction of narrow edentulous org/10.1111/jcpe.12148
ridges. Clinical Oral Implants Research, 10, 278–288. https://doi. de Freitas, R. M., Susin, C., Tamashiro, W. M., Chaves de Souza, J. A.,
org/10.1034/j.1600-0501.1999.100404.x Marcantonio, C., Wikesjo, U. M., … Marcantonio, E. Jr. (2016).
Chiapasco, M., Autelitano, L., Rabbiosi, D., & Zaniboni, M. (2013). Histological analysis and gene expression profile following augmen-
The role of pericranium grafts in the reduction of postopera- tation of the anterior maxilla using rhBMP-2/ACS versus autogenous
tive dehiscences and bone resorption after reconstruction of se- bone graft. Journal of Clinical Periodontology, 43, 1200–1207. https://
verely deficient edentulous ridges with autogenous onlay bone doi.org/10.1111/jcpe.12601
grafts. Clinical Oral Implants Research, 24, 679–687. https://doi. Guyatt, G. H., Oxman, A. D., Schunemann, H. J., Tugwell, P., & Knottnerus,
org/10.1111/j.1600-0501.2012.02485.x A. (2011). GRADE guidelines: A new series of articles in the Journal
Chiapasco, M., & Casentini, P. (2018). Horizontal bone-augmentation of Clinical Epidemiology. Journal of Clinical Epidemiology, 64, 380–
procedures in implant dentistry: Prosthetically guided regenera- 382. https://doi.org/10.1016/j.jclinepi.2010.09.011
tion. Periodontology, 2000(77), 213–240. https://doi.org/10.1111/ Hammerle, C. H., Jung, R. E., & Feloutzis, A. (2002). A systematic review
prd.12219 of the survival of implants in bone sites augmented with barrier
Chiapasco, M., Di Martino, G., Anello, T., Zaniboni, M., & Romeo, E. membranes (guided bone regeneration) in partially edentulous pa-
(2015). Fresh frozen versus autogenous iliac bone for the rehabili- tients. Journal of Clinical Periodontology, 29 Suppl 3, 226–231; discus-
tation of the extremely atrophic maxilla with onlay grafts and en- sion 232-223. https://doi.org/10.1034/j.1600-051X.29.s3.14.x
dosseous implants: Preliminary results of a prospective comparative Higgins, J. P., & Green, S. (2011) Cochran handbook for systematic reviews
study. Clinical Implant Dentistry and Related Research, 17(Suppl 1), of interventions version 5.1.0. http://handbook-5-1.cochrane.org/.
e251–e266. https://doi.org/10.1111/cid.12191 Ikebe, K., Wada, M., Kagawa, R., & Maeda, Y. (2009). Is old age a risk fac-
Cordaro, L., Amade, D. S., & Cordaro, M. (2002). Clinical results tor for dental implants? Japanese Dental Science Review, 45(1), 59–64.
of alveolar ridge augmentation with mandibular block bone https://doi.org/10.1016/j.jdsr.2009.02.001
grafts in partially edentulous patients prior to implant place- IntHout, J., Ioannidis, J. P., Rovers, M. M., & Goeman, J. J. (2016). Plea
ment. Clinical Oral Implants Research, 13, 103–111. https://doi. for routinely presenting prediction intervals in meta-analysis.
org/10.1034/j.1600-0501.2002.130113.x British Medical Journal Open, 6, e010247. https://doi.org/10.1136/
Cordaro, L., Torsello, F., Morcavallo, S., & di Torresanto, V. M. (2011). bmjopen-2015-010247
Effect of bovine bone and collagen membranes on healing of Ioannidis, J. P. (2008). Interpretation of tests of heterogeneity and bias
mandibular bone blocks: A prospective randomized controlled in meta-analysis. Journal of Evaluation in Clinical Practice, 14, 951–957.
study. Clinical Oral Implants Research, 22, 1145–1150. https://doi. https://doi.org/10.1111/j.1365-2753.2008.00986.x
org/10.1111/j.1600-0501.2010.02093.x Ioannidis, J. P., Patsopoulos, N. A., & Evangelou, E. (2007). Uncertainty
Da Costa, C. E., Pelegrine, A. A., Fagundes, D. J., Simoes Mde, J., & Taha, in heterogeneity estimates in meta-analyses. BMJ, 335, 914–916.
M. O. (2011). Use of corticocancellous allogeneic bone blocks im- https://doi.org/10.1136/bmj.39343.408449.80
pregnated with bone marrow aspirate: A clinical, tomographic, and Jensen, S. S., & Terheyden, H. (2009). Bone augmentation procedures in
histomorphometric study. General Dentistry, 59(5), e200–e205. localized defects in the alveolar ridge: Clinical results with different
Dahlin, C., Linde, A., Gottlow, J., & Nyman, S. (1988). Healing of bone defects bone grafts and bone-substitute materials. International Journal of
by guided tissue regeneration. Plastic and Reconstructive Surgery, 81, Oral and Maxillofacial Implants, 24(Suppl), 218–236.
672–676. https://doi.org/10.1097/00006534-198805000-00004 Kheur, M. G., Kheur, S., Lakha, T., Jambhekar, S., Le, B., & Jain, V. (2018).
Dasmah, A., Thor, A., Ekestubbe, A., Sennerby, L., & Rasmusson, L. (2012). Does graft particle type and size affect ridge dimensional changes
J Craniomaxillofac Surg., 40(8), 654–9. https://doi.org/10.1016/j. after alveolar ridge split procedure? Journal of Oral and Maxillofacial
jcms.2011.10.032 Surgery, 76, 761–769. https://doi.org/10.1016/j.joms.2017.11.002
Dasmah, A., Thor, A., Ekestubbe, A., Sennerby, L., & Rasmusson, L. (2013). Kostopoulos, L., & Karring, T. (1994). Augmentation of the rat mandible
Marginal bone-level alterations at implants installed in block versus using guided tissue regeneration. Clinical Oral Implants Research, 5,
particulate onlay bone grafts mixed with platelet-rich plasma in atro- 75–82. https://doi.org/10.1034/j.1600-0501.1994.050203.x
phic maxilla. a prospective 5-year follow-up study of 15 patients. Liberati, A., Altman, D. G., Tetzlaff, J., Mulrow, C., Gotzsche, P. C.,
Clinical Implant Dentistry and Related Research, 15, 7–14. https://doi. Ioannidis, J. P., … Moher, D. (2009). The PRISMA statement for re-
org/10.1111/j.1708-8208.2011.00377.x porting systematic reviews and meta-analyses of studies that eval-
Donos, N., Mardas, N., & Chadha, V. (2008). Clinical outcomes of im- uate health care interventions: Explanation and elaboration. Journal
plants following lateral bone augmentation: Systematic assessment of Clinical Epidemiology, 62, e1–e34. https://doi.org/10.1016/j.
of available options (barrier membranes, bone grafts, split osteot- jclinepi.2009.06.006
omy). Journal of Clinical Periodontology, 35, 173–202. https://doi. Maiorana, C., Beretta, M., Salina, S., & Santoro, F. (2005). Reduction of
org/10.1111/j.1600-051X.2008.01269.x autogenous bone graft resorption by means of bio-oss coverage: A
Eskan, M. A., Greenwell, H., Hill, M., Morton, D., Vidal, R., Shumway, B., & prospective study. International Journal of Periodontics and Restorative
Girouard, M. E. (2014). Platelet-rich plasma-assisted guided bone re- Dentistry, 25, 19–25.
generation for ridge augmentation: A randomized, controlled clinical Mazzocco, F., Nart, J., Cheung, W. S., & Griffin, T. J. (2011). Prospective
trial. Journal of Periodontology, 85, 661–668. https://doi.org/10.1902/ evaluation of the use of motorized ridge expanders in guided
jop.2013.130260 bone regeneration for future implant sites. International Journal of
Esposito, M., Grusovin, M. G., Felice, P., Karatzopoulos, G., Worthington, Periodontics and Restorative Dentistry, 31, 547–554.
H. V., & Coulthard, P. (2009). Interventions for replacing missing Meijndert, L., Raghoebar, G. M., Meijer, H. J., & Vissink, A. (2008). Clinical
teeth: Horizontal and vertical bone augmentation techniques for and radiographic characteristics of single-tooth replacements pre-
dental implant treatment. Cochrane Database Systematic Review, 4, ceded by local ridge augmentation: A prospective randomized clin-
CD003607. doi:10.1002/14651858.CD003607.pub4 ical trial. Clinical Oral Implants Research, 19, 1295–1303. https://doi.
de Freitas, R. M., Susin, C., Spin-Neto, R., Marcantonio, C., Wikesjo, org/10.1111/j.1600-0501.2008.01523.x
U. M., Pereira, L. A., & Marcantonio, E. Jr. (2013). Horizontal ridge Meijndert, L., Raghoebar, G. M., Schupbach, P., Meijer, H. J., & Vissink, A.
augmentation of the atrophic anterior maxilla using rhBMP-2/ACS (2005). Bone quality at the implant site after reconstruction of a local
|
306 NAENNI et al.
defect of the maxillary anterior ridge with chin bone or deproteinised Scipioni, A., Bruschi, G. B., & Calesini, G. (1994). The edentulous ridge
cancellous bovine bone. International Journal of Oral and Maxillofacial expansion technique: A five-year study. International Journal of
Surgery, 34, 877–884. https://doi.org/10.1016/j.ijom.2005.04.017 Periodontics and Restorative Dentistry, 14, 451–459.
Meijndert, C. M., Raghoebar, G. M., Meijndert, L., Stellingsma, K., Vissink, Shalash, M. A., Rahman, H. A., Azim, A. A., Neemat, A. H., Hawary, H. E.,
A., & Meijer, H. J. (2017). Single implants in the aesthetic region & Nasry, S. A. (2013). Evaluation of horizontal ridge augmentation
preceded by local ridge augmentation; a 10-year randomized con- using beta tricalcium phosphate and demineralized bone matrix: A
trolled trial. Clinical Oral Implants Research, 28, 388–395. https://doi. comparative study. Journal of Clinical and Experimental Dentistry, 5,
org/10.1111/clr.1281 e253–e259. https://doi.org/10.4317/jced.51244
Monje, A., Monje, F., Galindo-Moreno, P., Montanero-Fernandez, J., Suarez, Sideri, S., Papageorgiou, S. N., & Eliades, T. (2018). Registration in the in-
F., & Wang, H. L. (2014). Microstructural and densiometric analysis of ternational prospective register of systematic reviews (PROSPERO)
extra oral bone block grafts for maxillary horizontal bone augmen- of systematic review protocols was associated with increased review
tation: A comparison between calvarial bone and iliac crest. Clinical quality. Journal of Clinical Epidemiology, 100, 103–110. https://doi.
Oral Implants Research, 25, 659–664. https://doi.org/10.1111/clr. org/10.1016/j.jclinepi.2018.01.003
12159 Skoglund, A., Hising, P., & Young, C. (1997). A clinical and histologic ex-
Mordenfeld, A., Johansson, C. B., Albrektsson, T., & Hallman, M. (2014). amination in humans of the osseous response to implanted natural
A randomized and controlled clinical trial of two different composi- bone mineral. International Journal of Oral and Maxillofacial Implants,
tions of deproteinized bovine bone and autogenous bone used for 12, 194–199.
lateral ridge augmentation. Clinical Oral Implants Research, 25, 310– Sterne, J. A., Hernan, M. A., Reeves, B. C., Savovic, J., Berkman, N. D.,
320. https://doi.org/10.1111/clr.12143 Viswanathan, M., … Higgins, J. P. (2016). ROBINS-I: A tool for assess-
Mordini, L. Analysis of Bone Volume Changes Following Lateral Guided ing risk of bias in non-randomised studies of interventions. BMJ, 355,
Bone Regeneration. Master Thesis, Tufts University School of Dental i4919. https://doi.org/10.1136/bmj.i4919
Medicine, 2016 Thoma, D. S., Payer, M., Jakse, N., Bienz, S. P., Husler, J., Schmidlin, P.
Muramatsu, N., & Akiyama, H. (2011). Japan: Super-aging society pre- R., … Jung, R. E. (2018). Randomized, controlled clinical two-centre
paring for the future. Gerontologist, 51, 425–432. https://doi. study using xenogeneic block grafts loaded with recombinant human
org/10.1093/geront/gnr067 bone morphogenetic protein-2 or autogenous bone blocks for lateral
Naenni, N., Hyun‐Chang, L., Papageorgiou, S. N., & Hämmerle, C. H. ridge augmentation. Journal of Clinical Periodontology, 45, 265–276.
F. (2018) Effectiveness of lateral bone augmentation prior to im- https://doi.org/10.1111/jcpe.12841
plant placement: a systematic review and meta-analysis. http://doi. Thor, A., Wannfors, K., Sennerby, L., & Rasmusson, L. (2005). Clin Implant
org/10.5281/zenodo. 1713769 Dent Relat Res., 7(4), 209–20.
Norman, G. R., Sloan, J. A., & Wyrwich, K. W. (2003). Interpretation of Tinti, C., & Parma-Benfenati, S. (2003). Clinical classification of bone
changes in health-related quality of life: The remarkable universality defects concerning the placement of dental implants. International
of half a standard deviation. Medical Care, 41, 582–592. https://doi. Journal of Periodontics and Restorative Dentistry, 23, 147–155.
org/10.1097/01.MLR.0000062554.74615.4C Tolstunov, L. (2009). Maxillary tuberosity block bone graft:
Papageorgiou, S. N. (2014). Meta-analysis for orthodontists: Part II–Is all Innovative technique and case report. Journal of Oral and
that glitters gold? Journal of Orthodontics, 41, 327–336. https://doi.or Maxillofacial Surgery, 67, 1723–1729. https://doi.org/10.1016/j.
g/10.1179/1465313314Y.0000000110 joms.2009.03.043
Papageorgiou, S. N., Kloukos, D., Petridis, H., & Pandis, N. (2015). An Urban, I. A., Nagursky, H., & Lozada, J. L. (2011). Horizontal ridge aug-
assessment of the risk of bias in randomized controlled trial re- mentation with a resorbable membrane and particulated autogenous
ports published in Prosthodontic and implant dentistry journals. bone with or without anorganic bovine bone-derived mineral: A pro-
International Journal of Prosthodontics, 28, 586–593. https://doi. spective case series in 22 patients. International Journal of Oral and
org/10.11607/ijp.4357 Maxillofacial Implants, 26, 404–414.
Pourabbas, R., & Nezafati, S. (2007). Clinical results of localized al- Veroniki, A. A., Jackson, D., Viechtbauer, W., Bender, R., Bowden, J.,
veolar ridge augmentation with bone grafts harvested from Knapp, G., … Salanti, G. (2016). Methods to estimate the between-
symphysis in comparison with ramus. Journal of Dental Research, study variance and its uncertainty in meta-analysis. Research
Dental Clinics, Dental Prospects, 1, 7–12. https://doi.org/10.5681/ Synthesis Methods, 7, 55–79. https://doi.org/10.1002/jrsm.1164
joddd.2007.002 Wang, H. L., Misch, C., & Neiva, R. F. (2004). “Sandwich” bone augmen-
Sanz-Sanchez, I., Ortiz-Vigon, A., Sanz-Martin, I., Figuero, E., & tation technique: Rationale and report of pilot cases. International
Sanz, M. (2015). Effectiveness of lateral bone augmentation on Journal of Periodontics and Restorative Dentistry, 24, 232–245.
the alveolar crest dimension: A systematic review and meta-
analysis. Journal of Dental Research, 94, 128S–142S. https://doi.
org/10.1177/0022034515594780 S U P P O R T I N G I N FO R M AT I O N
Schlegel, A. K., & Donath, K. (1998). BIO-OSS–a resorbable bone sub-
stitute? Journal of Long-Term Effects of Medical Implants, 8, 201–209. Additional supporting information may be found online in the
Schünemann, H., Brożek, J., Guyatt, G., & Oxman, A. (2013) Handbook Supporting Information section at the end of the article.
for grading the quality of evidence and the strength of recommenda-
tions using the GRADE approach. Updated October 2013: Grading
of Recommendations, Assessment, Development and Evaluation
How to cite this article: Naenni N, Lim H, Papageorgiou SN,
(GRADE) Working Group.
Schunemann, H. J., Cuello, C., Akl, E. A., Mustafa, R. A., Meerpohl, et al. Efficacy of lateral bone augmentation prior to implant
J. J., Thayer, K., … Sterne, J. (2018). GRADE guidelines: 18. How placement: A systematic review and meta-analysis. J Clin
ROBINS-I and other tools to assess risk of bias in nonrandomized Periodontol. 2019;46(Suppl. 21):287–306. https://doi.
studies should be used to rate the certainty of a body of evidence.
org/10.1111/jcpe.13052
Journal of Clinical Epidemiology, [Epub ahead of print]. doi:10.1016/j.
jclinepi.2018.01.012

J Clinic Periodontology - 2019 - Naenni - Efficacy of Lateral Bone Augmentation Prior To Implant Placement A Systematic

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

J Clinic Periodontology - 2019 - Naenni - Efficacy of Lateral Bone Augmentation Prior To Implant Placement A Systematic

Uploaded by

Copyright:

Available Formats

| |

Received: 19 July 2018 Revised: 9 November 2018 Accepted: 1 December 2018

Efficacy of lateral bone augmentation prior to implant

Nadja Naenni1 | Hyun‐Chang Lim1,3 | Spyridon N. Papageorgiou2 |

randomized and prospective non-randomized trials on humans indicates that lateral

and non-­randomized studies were assessed with the ROBINS-­I (Risk

TA B L E 1 Implant information by included studies

Number of implants Additional GBR at the time

1 Antoun et al., 2001; NR/NR (100%) NR NR “Implants were successfully

TA B L E 2 Characteristics of included studies

Design; setting; Sites (implants); Jaw;

Design; setting; Sites (implants); Jaw;

TA B L E 3 Outcomes assessed by included studies

Nr Study FUa; Imp Outcome Unit CoI

TA B L E 4 Complications reported by the included studies.

Nr Study Tx Type of Complication & treatment

Nr Study Tx Type of Complication & treatment

17 Meijndert 2005 G1: AUG (block/part; chin) No complication

Outcome Studies Average (95% CI) I2 (95% CI) 95% Prediction

augmented with XEN (with/without AUG) resulted in 11.6% less re-

BW gain 2 −0.33 (−2.24 to 1.58) 0.74 71% (NC) 1.38 (NC) NC

TA B L E 8 Subgroup analysis for the

Anticipated absolute effectsa (95% CI)

AUG XEN w/o AUG

Factors associated with outcome of lateral bone augmentation.

performed on augmentation site level, which ignores clustering ef- ORCID

You might also like

and non-randomized studies were assessed with the ROBINS-I (Risk