Mardas Et Al-2015-Clinical Oral Implants Research

Nikos Mardas Does ridge preservation following
Anna Trullenque-Eriksson
Neil MacBeth
tooth extraction improve implant
Aviva Petrie treatment outcomes: a systematic
Nikolaos Donos
review
Group 4: Therapeutic concepts & methods
Authors affiliations: Key words: alveolar ridge preservation, dental implants, tooth extraction
Nikos Mardas, Anna Trullenque-Eriksson, Neil
MacBeth, Nikolaos Donos, Unit of Periodontology,
UCL Eastman Dental Institute, London, UK Abstract
Neil MacBeth, Defense Dental Service, RAF, Objective: (1) Primary focused question (Q1): to evaluate the effect of alveolar ridge preservation
London, UK
(ARP) on implant outcomes (implant placement feasibility, need for further augmentation, survival/
Aviva Petrie, Biostatistics Unit, UCL Eastman
Dental Institute, London, UK success rates, marginal bone loss) compared with unassisted socket healing (USH) and (2) secondary
focused question (Q2): to estimate the size effects (SE) of these outcomes in three different
Corresponding author:
Nikos Mardas
interventions (GBR, socket filler, socket seal).
Unit of Periodontology Material and methods: Electronic (MEDLINE, EMBASE, Cochrane Central Register LILACS; Web of
UCL Eastman Dental Institute Science) and hand search was conducted up to July 2014. Randomised controlled trials (RCT),
256 Grays Inn Road
London WC1X 8LD, UK controlled clinical trials (CCT) and prospective cohort studies with USH as controls were eligible in
Tel.: +44 (20) 7915 2379 the analysis for Q1. RCTs, CCTs and prospective case series, with or without USH as control, were
Fax: +44 (20) 7915 1137 eligible for Q2.
e-mail: [email protected]
Results: Ten (8 RCTs, 2 CCTs) and 30 studies (21 RCTs, 7 CCTs, 2 case series) were included in the
analysis for Q1 and Q2, respectively. The risk for bias was unclear or high in most of them. Q1:
Implant placement was feasible in ARP-treated and USH sites. These implants presented similar
survival/success rates and marginal bone levels. The need for further augmentation decreased
when ARP was performed (Relative risk: 0.15, 95% CI: 0.070.3). Q2: The SE for implant placement
feasibility was 98.5% (95% CI: 96.499.6) in GBR and 96.2 (95% CI: 93.198.2) in socket filler group.
The SE for need for further augmentation was 11.9 (95% CI: 5.619.9) for GBR and 13.7% (95% CI:
5.025.6) for socket filler groups. GBR and socket filler presented similar SE for survival/success rates
and average marginal bone loss. Limited data were available for implant-related outcomes in sites
treated with socket seal.
Conclusions: There is limited evidence to support the clinical benefit of ARP over USH in
improving implant-related outcomes despite a decrease in the need for further ridge
augmentation during implant placement. Similar implant placement feasibility, survival/success
rates and marginal bone loss should be anticipated following ARP or USH. Currently, it is not clear
which type of ARP intervention has a superior impact on implant outcomes.
The success of implant-supported restorations complicating implant placement in the ideal

depends on the interaction between a number prosthetic position. When extensive ridge
of anatomical, technical, surgical and pros- resorption has occurred, ridge augmentation
thetic factors. Amongst them, restorative-dri- procedures, prior or simultaneously with
ven implant placement allows the optimal implant placement, are required to allow a
support of the surrounding soft and hard tis- prosthetically driven implant placement.
sues and a satisfactory emergence profile of Although, up to a certain extent, extraction
Date: the final prosthesis (Chen & Buser 2009). sockets heal, with new bone, the extensive
Accepted 12 May 2015
However, the alveolar process is a tooth- resorption and remodelling of socket walls
To cite this article: dependent anatomic structure which, when during the early healing stages will position
Mardas N, Trullenque-Eriksson A, MacBeth N, Petrie A,
Donos N. Does ridge preservation following tooth extraction permanent teeth are lost, undergoes signifi- the post-extraction alveolar ridge margins at a
improve implant treatment outcomes: a systematic review.
cant dimensional changes (Atwood 1971; more apical level and to a more lingual/palatal
Clin. Oral Impl. Res. 26 (Suppl. 11), 2015, 180201
doi: 10.1111/clr.12639 Cardaropoli et al. 2003; Farmer & Darby 2014) position resulting in changes to the overlying
180 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Mardas et al Implant outcomes after ridge preservation
soft tissue contours (Atwood 1971; Ara ujo & such as implant placement feasibility, need Populations of studies
Lindhe 2005; Tan et al. 2012). Systematic for further augmentation during implant Healthy individuals, without any age limit,
reviews have reported a weighted mean in placement and the relevant survival/success who underwent any type of ridge preserva-
alveolar ridge reduction in 3.87 mm in width rates and marginal bone loss of implants in tion following permanent tooth extraction
and 1.67 mm in height (Van der Weijden et al. comparison with unassisted socket healing. with the aim of facilitating future implant
2009) after 3 months of healing or a 63% and Furthermore, no biomaterial or type of inter- placement, were included.
22% dimensional loss in a horizontal and ver- vention used for ARP can be claimed to be
tical plane taking place at the first 6 months superior than others in terms of implant out- Types of interventions
after the extraction (H ammerle et al. 2012; comes (Horvath et al. 2013).
Tan et al. 2012). The horizontal bone loss/ Therefore, the aim of this systematic Test groups
Studies reporting on any of the following
resorption of the socket is generally more pro- review and meta-analysis was to investigate
ARP interventions were included: (1) socket
nounced at the buccal plate, and the vertical the additional effect of ARP on implant-
filling with various bone grafts, substitutes
resorption is also more evident on the buccal related outcomes in comparison with unas-
or biologically active materials (growth fac-
contour of the ridge (Pietrokovski & Massler sisted socket healing and to estimate the size
tors) in different type of carriers, (2) socket
1967; Ara ujo & Lindhe 2005). effects of the above-mentioned outcomes
seal with soft tissue grafts of different origins
Alveolar ridge preservation (ARP; also according to the type of intervention for
and (3) GBR with various barrier membranes
known as socket preservation) involves ARP.
and combinations of the above filler/grafting
any procedure developed to eliminate or limit
materials. The minimum number of subjects
the negative effect of post-extraction resorp-
Methods per group was 10 in controlled studies and 20
tion, maintain the soft and hard tissue con-
in case series.
tour of the ridge, promote bone formation
A detailed protocol was developed based on
within the socket and facilitate implant
Cochrane guidelines and the European Asso- Control groups
placement in a prosthetically driven position
ciation of Osseointegration requirements for The studies attempting to answer focused
(Horv ath et al. 2013). Different types of pro-
this consensus report. question 1 presented unassisted socket healing
cedures have been described in the literature,
following atraumatic tooth extraction without
including guided bone regeneration (GBR),
Focused question 1 any other intervention as a control group.
socket filler and socket sealing. The main focused question of this systematic
Various grafting materials have been uti- review was the following: Is there any addi- Outcome variables
lised individually or in combination with tional benefit of alveolar ridge preservation For both focused questions, four dichotomous
resorbable (Iasella et al. 2003; Barone et al. techniques over unassisted healing in terms (yes/no) and one continuous implant-related
2008; Mardas et al. 2010) or non-resorbable of the following: (i) implant placement feasi- outcome variables were evaluated:
GBR barriers (Lekovic et al. 1997) to reduce bility (ii) need for further augmentation (iii) 1. Feasibility of implant placement:
the post-extraction remodelling effect. The implant survival, (iv) implant success and (v) expressed as a percentage (%) of implants
described grafting/filler materials include the marginal bone loss. placed with satisfactory primary stability.
following: autografts (Becker et al. 1994),
2. Need for further augmentation: expressed
allografts (Becker et al. 1994, 1996; Froum Focused question 2 as a percentage (%) of implants that
et al. 2002), xenografts (Iasella et al. 2003; Ba- In order to examine data published in case required further ridge augmentation pro-
rone et al. 2008; Mardas et al. 2010), allop- series in controlled clinical studies, where cedures during implant placement for the
lasts (Serino et al. 2003; Mardas et al. 2010) unassisted socket healing has not been used management of residual dehiscence or
and growth factors in different type of carri- as a control group, but different ARP proce- fenestration defects.
ers (Fiorellini et al. 2005). Furthermore, dif- dures were compared or data published in 3. Implant survival: expressed as the per-
ferent soft tissue grafts such as autogenous case series, the same systematic review centage (%) of loaded and functional
free gingival grafts (Jung et al. 2004; Araujo attempted to address a second focused ques- implants present in the arch after
et al. 2014), dermal allografts or collagen tion: What are the estimated size effects of 12 months of loading.
matrix xenografts (Jung et al. 2013) have been implant placement feasibility, need for fur- 4. Implant success: expressed as the per-
used to seal the socket entrance (for ther augmentation and the survival, success centage (%) of successful implants at
review, see Wang & Lang 2012; Horvath and marginal bone loss of implants placed 12 months after loading based on speci-
et al. 2013; Avila-Ortiz et al. 2014; Vignoletti following different alveolar ridge preservation fied sets of success criteria.
et al. 2014). Previous systematic reviews and techniques? 5. Marginal bone levels: average of mesial
consensus papers have reached the conclu-
and distal proximal bone loss at
sion that, although ridge resorption was not Types of studies 12 months after loading. If mesial and
prevented, ARP procedures reduce the post- For focused question 1, only longitudinal pro-
distal measurements were presented, but
extraction dimensional changes and were spective studies, that is RCTs and CCTs
no mean figure was provided, the mesial
compatible with bone regeneration in the with unassisted socket healing as a control
measurement was used for the analysis.
extraction socket (Vignoletti et al. 2012; group, were included in the meta-analysis.
Wang & Lang 2012; De Risi et al. 2013; For focused question 2, in addition to the
Risk of bias and methodological quality
Horv ath et al. 2013; Avila-Ortiz et al. 2014; previous studies, RCTs, CCTs and large pro- assessment
Morjaria et al. 2014). There remained, how- spective case series, without an unassisted A modification of the Cochrane tool for eval-
ever, a lack of consensus whether ARP would healing control group, were included in the uating risk of bias (Higgins & Green 2011)
directly improve implant-related outcomes meta-analysis. was used to evaluate the methodological
2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd 181 | Clin. Oral Impl. Res. 26 (Suppl. 11), 2015 / 180201
quality and risk of bias of the included stud- without unassisted socket healing as a York, NY, USA), and all further steps of
ies. Non-randomised studies are by definition control group and prospective case series. screening were performed on this interface. A
at higher risk of bias than well-designed 2. Controlled studies reporting on a mini- three-stage selection of the resulted hits was
randomised clinical trials; however, we did mum of 10 patients per group or case series performed independently and in duplicate by
not evaluate randomisation in the quality reporting on a minimum of 20 patients. two reviewers (ATE and NMcB). In order to
assessment, as both randomised and non- reduce errors and bias, a calibration exercise
randomised clinical trials were included. The Search strategy was performed with the first 24 articles iden-
following six parameters: allocation conceal- The search strategy incorporated both elec- tified from the journal hand-searched. In case
ment, blinding of participants and personnel, tronic and hand searches. The following elec- of disagreement at the title selection stage,
blinding of outcome assessment, incomplete tronic databases were utilised: (1) MEDLINE the trial was included in the abstract stage.
outcome data, selective reporting and other In-Process & Other Non-Indexed Citations and At the abstract and full-text selection, any
sources of bias (related to study design or MEDLINE 1950 to present via Ovid interface; disagreements between the above reviewers
other problems), were evaluated as low risk, (2) EMBASE Classic + EMBASE 1947 to present were resolved by discussion including a third
unclear risk or high risk of bias. If all the via Ovid interface; (3) The Cochrane Central reviewer (NM). The reasons for exclusion
parameters were judged as low, the study Register of Controlled Trials (CENTRAL); (4) were recorded in a specific data extraction
was at low risk of bias. If at least one param- LILACS; (5) Web of Science. The electronic form at the full-text selection stage. The
eter was judged as unclear or as high risk of search strategy included terms related to the level of agreement was determined by a
bias, the studies were included at unclear or intervention and used the following Kappa score calculation of agreement during
high risk of bias, respectively. combination of key words and MeSH terms: the title and abstract selection process.
(tooth extraction OR tooth removal OR
General inclusion criteria socket OR alveol ORridgeORcrest Research synthesis & meta-analysis
1. Studies on healthy individuals, without OR toothsocketOR alveolarbonelossOR For all included studies answering both
any age limit, who underwent ARP fol- boneresorptionOR boneremodelling)AND focused questions, a descriptive synthesis
lowing tooth extraction in order to (preserv OR reconstructORaugment OR was undertaken. The studies were classified
receive implants. fillORseal OR graftORrepair OR according to research design and type of
2. Studies providing information regarding alveolar ridge augmentation OR bone intervention, and the outcomes were
feasibility of implant placement and/or regeneration OR bone substitutes OR recorded in evidence tables.
the need for further augmentation during transplantation). Cochrane search filters for For focused question 1, meta-analysis was
implant placement and/or reporting sur- RCTs and CCTS were implemented, with conducted utilising the available data from
vival/success and/or proximal bone loss cohort trials also included. The results were the RCTs and CCTs using a parallel design
around the implants placed in extraction limited to human studies. The full electronic with the patient as unit of analysis.
sites treated with ARP. search strategy can be found in the Appendix. For focused question 2, meta-analysis was
An extensive hand search was also per- conducted utilising the available data from
General exclusion criteria formed encompassing the bibliographies of all the studies included in the meta-analysis
1. Retrospective studies. the included papers and other narrative and of focused question 1 and data from RCTs
2. Studies on medically compromised systematic reviews. In addition, the following and CCTs without unassisted socket healing
patients or under specific medication. journals were screened from 2001 to July as control group, as well as larger prospective
3. Studies reporting on immediate implant 2014: Clinical Oral Implants Research, Clini- case series, as long as they used a parallel
placement. cal Implant Dentistry and Related Research, design with the patient as unit of analysis.
4. Studies reporting solely on third molars European Journal of Oral Implantology, The studies included for meta-analysis were
extractions. Implant Dentistry, International Journal of divided into three different groups (GBR,
5. Publications reporting data on the same Oral and Maxillofacial Implants, International socket filler and socket seal) according to the
sample and procedures as other publica- Journal of Periodontics and Restorative Den- type of intervention. When ARP was per-
tions. tistry, Journal of Clinical Periodontology, formed utilising a resorbable or non-resorb-
Journal of Dental Research, Journal of Oral able barrier with or without bone grafting,
Specific inclusion criteria for focused question 1 and Maxillofacial Surgery, Journal of Peri- the study was categorised in the GBR group.
1. Longitudinal prospective studies, that is odontology, Oral Surgery, Oral Medicine, Oral When the socket was treated just with a fil-
RCTs and CCTs where one of the above- Radiology, Oral Pathology and Endodontics. ler/graft including collagen sponges/plagues,
mentioned types of interventions, were No language restrictions were applied, and the study was categorised in the socket graft-
carried out in the test group and which translations were carried out if necessary. ing group. Finally, the study was categorised
had unassisted socket healing as a con- Unpublished trials and abstracts were not in the socket seal group when a soft tissue
trol group. included in the search process. When the graft was used to seal the entrance of the
2. Studies reporting on a minimum of 10 results of a study were presented in a number socket with or without grafting of the socket
patients per group. of publications, the most complete data set following a flapless approach.
was included in the analysis. In case of miss- MedCalc (version 14.12.0, MedCalc Soft-
Specific inclusion criteria for focused question 2 ing or incomplete data, the authors were con- ware bvba, Ostend, Belgium) software was
1. Longitudinal prospective studies, that is tacted via email allowing a period of 3 weeks used for the meta-analysis relating to binary
RCTs, CCTs, cohort studies, where one for their reply with the missing data. outcome variables in relation to both focused
or more of the above-mentioned types of The extracted data were copied into End- questions, and the continuous variable for
interventions, were carried out, with or Note X7 software (Thomson Reuters, New focused question 1. Assessment of statistical
182 | Clin. Oral Impl. Res. 26 (Suppl. 11), 2015 / 180201 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
heterogeneity was performed using Cochrans

Q-test and determination of the I2 index (Hig-
gins et al. 2003). The I2 index provides an esti-
mate of the amount of variation attributable
to heterogeneity (I2 = 25%: low; I2 = 50%:
moderate; I2 = 75%: high heterogeneity). The
different outcome variable estimates were
pooled using a fixed effects analysis if non-sig-
nificant statistical heterogeneity was found
between studies. If there was evidence of sta-
tistical heterogeneity, a random effects model
was used.
Depending on the type of variable, different
pooled estimates were provided.
For Question 1: (a) Relative risk for
dichotomous variables (e.g. need for fur-
ther augmentation). (b) Standardised mean
difference (i.e. the difference in means
divided by the standard deviation) for the
continuous variable (e.g. marginal bone
loss in mm).
For Question 2: (a) Proportion of positive
responses for dichotomous variables (inci-
dence). This figure was obtained by back-
transformation after taking the arcsine-
square-root transformation of each propor-
tion (Freeman & Tukey 1950). (b) Mean
implant proximal bone loss (mm) for the
continuous variable.
Forest Plots were created to illustrate the

effects of the different studies, shown against
the global estimate. Statistical significance
was achieved if P < 0.05.
Results
Study selection
The initial search yielded a total 14,399
records including 72 papers that were Fig. 1. Selection process and search strategy flowchart.
selected through hand search and one more
through cross-reference. After removal of
duplicates and the title and abstract screen- Thirty studies (Iasella et al. 2003; Serino intending for implant placement; and dupli-
ing, a total of 103 articles were left for full- et al. 2003; Vance et al. 2004; Neiva et al. cate reports. Several articles were excluded
text assessment (Fig. 1). The authors of 34 of 2008; Crespi et al. 2009a,b, 2011a,b; Beck & for more than one reason. The excluded
these 103 articles were contacted at this Mealey 2010; Barone et al. 2012; Gholami papers and the reasons for exclusion for both
stage in order to provide additional data on et al. 2012; Hoang & Mealey 2012; Mardinger focused questions are listed in Table 1.
implant outcomes before the final selection. et al. 2012; Perelman-Karmon et al. 2012; The Kappa score for agreement between
Ten papers (Iasella et al. 2003; Serino et al. Sisti et al. 2012; Wood & Mealey 2012; the reviewers (ATE, NMB) at the title and
2003; Crespi et al. 2009a, 2011b; Barone et al. Barone et al. 2013a,b; Cook & Mealey 2013; abstract selection level was 0.94 indicating a
2012; Sisti et al. 2012; Barone et al. 2013a; Festa et al. 2013; Leblebicioglu et al. 2013; high level of agreement.
Festa et al. 2013; Cardaropoli et al. 2014; Spi- Patel et al. 2013; Poulias et al. 2013; Wallace
nato et al. 2014) were eligible for inclusion in et al. 2013; Barone et al. 2014; Cardaropoli Study design and population
the qualitative analysis for focused question et al. 2014; Coomes et al. 2014; Eskow & The study design and study population char-
1. The most common reason for exclusion Mealey 2014; Lindhe et al. 2014; Spinato acteristics of the included studies for both
for this focused question was the lack of con- et al. 2014) were included in the qualitative focused questions are presented in Table 2.
trol group with unassisted socket healing; analysis for focused question 2. The most
insufficient number of patients; and not common reason for exclusion for this focused Controlled studies answering the focus question 1
reporting on implants outcomes or not question was insufficient number of patients; Of the 10 studies (8 RCTs, 2 CCTs) that were
intending for implant placement. not reporting on implants outcomes; or not eligible for inclusion in the qualitative analysis
Table 1. List of excluded full-text papers and reasons for exclusion following full-text screening for focused question 1, two selected RCTs
Author and year Reasons for exclusion (Crespi et al. 2011b; Festa et al. 2013) and two
Aimetti (2009) Excluded due to no relevant outcome measures being provided CCTs (Serino et al. 2003; Crespi et al. 2009a)
Alkan (2013) Insufficient number of patients were excluded from the quantitative analysis
Al-Khaldi (2011) Not ARP due to split-mouth or unclear design, which
Anitua (1999) Insufficient number of patients
made pooled meta-analysis not feasible. All
Artzi (2000) Insufficient number of patients
Babbush (2003) Insufficient number of patients remaining publications had a parallel design
Barone (2008) Duplicate report (Barone 2012) with the patient being the unit of analysis.
Brkovic (2012) Insufficient number of patients The study population ranged from 15 to 58
Brownfield (2012) Insufficient number of patients
Camargo (2000) Not reporting on implants
patients. In the six studies included in the
Canullo (2013) Insufficient number of patients meta-analysis, 221 sockets were treated in a
Canuto (2013) Not reporting on implants total of 214 patients. The distribution of the
Cardaropoli (2012) Duplicate report (Cardaropoli 2014)
extracted teeth was fairly heterogeneous and
Carmagnola (2003) Insufficient number of patients
Casado (2010) Not reporting on implants included both single- and multi-rooted teeth.
Clozza (2012) Duplicate report (Clozza 2014), insufficient number of patients Four studies included smokers (Barone et al.
Clozza (2014) Insufficient number of patients 2012; Sisti et al. 2012; Barone et al. 2013a;
Collins (2014) Insufficient number of patients
Cardaropoli et al. 2014), while in two studies,
De Coster (2011) Not all patients/sockets intended for implant placement
Engler-Hamm (2011) Insufficient number of patients, not all patients/sockets smoking habits were not reported (Iasella
intended for implant placement et al. 2003; Serino et al. 2003).
Farina (2013) Early implant placement
Fernandes (2011) Not all patients/sockets intended for implant placement
Fiorellini (2005) Not reporting on implants Studies answering the focus question 2
Fotek (2009) Insufficient number of patients Thirty studies (21 RCTs, seven CCTs, two
Geffre (2010) Animal study case series) were eligible for inclusion in the
Geurs (2014) Insufficient number of patients, not reporting on implants qualitative analysis for focused question two.
Hanser (2014) Study seems to be retrospective
Hauser (2013) Insufficient number of patients From those, twelve studies compared a GBR
Heberer 2008 Early implant placement approach to unassisted socket healing or to
Heberer (2011) Insufficient number of patients another GBR approach (Iasella et al. 2003;
Heberer (2012) Early implant placement
ndez-Alfaro (2005)
Barone et al. 2012, 2013a,b, 2014; Gholami
Herna Insufficient number of patients, reports on a mixture of clinical
situations (ARP, discrepancy implant-socket, reconstruction et al. 2012; Cook & Mealey 2013; Leblebicio-
after removal of implants, etc.) glu et al. 2013; Patel et al. 2013; Poulias et al.
Hsuan-Yu (2012) Insufficient number of patients 2013; Wallace et al. 2013; Cardaropoli et al.
Huh (2011) Not reporting on implants
Irinakis (2006) Review
2014). Fourteen studies compared a socket fil-
Jung (2004) Not reporting on implants ler approach to unassisted socket healing or to
Jung (2013) Not all patients/sockets intended for implant placement different socket filler materials (Serino et al.
Kim (2011) Not all patients/sockets intended for implant placement
2003; Neiva et al. 2008; Crespi et al. 2009a,b,
Kim (2013) Insufficient number of patients
Kim (2014) Not reporting on implants 2011a,b; Beck & Mealey 2010; Hoang & Mea-
Kotsakis (2014a) Insufficient number of patients ley 2012; Sisti et al. 2012; Wood & Mealey
Kotsakis (2014b) Not all patients/sockets intended for implant placement 2012; Festa et al. 2013; Coomes et al. 2014;
Lambert (2012) Insufficient number of patients, not all patients/sockets
Eskow & Mealey 2014; Spinato et al. 2014)
intended for implant placement
Lekovic (1998) Not reporting on implants and two studies reported implant outcomes
Luczyszyn (2005) Not reporting on implants following ridge preservation with a socket seal
Madan (2014) Insufficient number of patients technique (Mardinger et al. 2012; Lindhe et al.
Mahesh (2012) Not reporting on implants
Mardas (2010) Duplicate report (Patel 2013) 2014). Finally, two studies compared a GBR
Mardas (2011) Duplicate report (Patel 2013) approach to a socket filler approach (Vance
Mardinger (2009) Duplicate report (Mardinger 2012) et al. 2004; Perelman-Karmon et al. 2012).
Misch (2010) Insufficient number of patients
Twenty-one studies were considered for
Moghaddas (2012) Insufficient number of patients
Nam (2011) Not all patients/sockets intended for implant placement meta-analysis of at least one outcome vari-
Neiva (2011) Insufficient number of patients able following categorisation in three inter-
Nevins (2014) Insufficient number of patients vention groups: (a) GBR: 8 RCTs (Iasella
Norton (2002) Insufficient number of patients
Oghli (2010) Not reporting on implants
et al. 2003; Barone et al. 2012; Barone et al.
Pellegrini (2014) Insufficient number of patients, not reporting on implants 2013a; Cook & Mealey 2013; Patel et al.
Pinho (2006) Not reporting on implants 2013; Poulias et al. 2013; Barone et al. 2014;
Ruga (2011) Not all patients/sockets intended for implant placement
Cardaropoli et al. 2014) and one case series
Scheyer (2012) Insufficient number of patients
Schneider (2014) Not reporting on implants (Leblebicioglu et al. 2013), (b) socket filler:
Serino (2008) Insufficient number of patients seven RCTs (Neiva et al. 2008; Hoang &
Simon (2011) Excluded due to no relevant outcome measures being provided Mealey 2012; Sisti et al. 2012; Wood & Mea-
Shakibaie (2013) Insufficient number of patients
ley 2012; Coomes et al. 2014; Eskow & Mea-
Suttapreyasri (2013) Insufficient number of patients, not reporting on implants
Tal (1999) Not reporting on implants ley 2014; Spinato et al. 2014) and one CCT
Thalmair (2013) Insufficient number of patients, not reporting on implants (Beck & Mealey 2010) and (c) socket seal: 1
Toloue (2012) Not reporting on implants CCT (Lindhe et al. 2014) and one case series
Table 1. (continued) mation on implant type (Iasella et al. 2003;

Author and year Reasons for exclusion Crespi et al. 2009a; Festa et al. 2013; Card-
Vanhoutte (2014) Duplicate report (Lambert 2012), not all patients/sockets aropoli et al. 2014).
intended for implant placement
Villanueva-Alcojol (2013) Insufficient number of patients Studies answering the focus question 2
Weiss (2007) Not reporting on implants In most of the studies utilising GBR for ARP,
Wu (2014) Not reporting on implants
a muco-periosteal flap was elevated with the
exception of two studies (Barone et al. 2013a,
2014-one group). On the contrary, 9 studies
(Mardinger et al. 2012). Two RCTs where 2012; Festa et al. 2013), muco-periosteal flaps including a treatment group where socket fil-
two different types of interventions were were elevated both at the ARP-treated and ler was used (Neiva et al. 2008; Crespi et al.
compared (Vance et al. 2004; Perelman-Kar- control extraction sites and in one study only 2009a,b; Beck & Mealey 2010; Crespi et al.
mon et al. 2012; GBR vs. socket grafting) in the ARP-treated sites (Cardaropoli et al. 2011a,b; Sisti et al. 2012; Coomes et al. 2014;
were categorised in both GBR and socket fil- 2014). It was unclear whether flaps were ele- Eskow & Mealey 2014; Spinato et al. 2014)
ler groups. The data from the GBR and the vated in the ARP-treated sites in one study and the 2 studies using socket seal tech-
socket filler treatment groups in these two (Crespi et al. 2011b). In the remaining four niques for ridge preservation (Mardinger et al.
studies contributed separately in the meta- studies (Crespi et al. 2009a; Sisti et al. 2012; 2012; Lindhe et al. 2014) utilised a flapless
analysis of each type of intervention. Five Barone et al. 2013a; Spinato et al. 2014), a approach. Very few studies besides the two
studies were excluded from the quantitative flapless approach was followed in the ARP- studies using socket seal attempted and/or
analysis due to a split-mouth design (Crespi treated sites. Primary closure was attempted, achieved primary closure of the soft tissues
et al. 2009a; Crespi et al. 2011a,b; Gholami however, only in one study (Barone et al. over the augmentation materials (Barone
et al. 2012; Festa et al. 2013) and three due to 2012). In the six included studies that speci- et al. 2012, 2014; -one group, Gholami et al.
an unclear study design (Serino et al. 2003; fied number of intact walls, all four or at 2012; Perelman-Karmon et al. 2012; Wallace
Crespi et al. 2009b; Barone et al. 2013b), least 3 walls of the socket walls should have et al. 2013). Most of the included studies for
which made pooled meta-analysis not feasi- been intact after extraction allowing only all three types of interventions reported that
ble; one more study was excluded due to data partial loss of the buccal wall (Crespi et al. all four or at least three walls of the socket
not being provided with the patient as the 2009a, 2011b; Barone et al. 2012; Festa walls should have been intact after extraction
unit of analysis (Wallace et al. 2013). et al. 2013; Cardaropoli et al. 2014; Spinato in order to proceed in ARP.
In the studies included in the qualitative et al. 2014). Regarding the studies included in In most of the studies, GBR was performed
analysis, the study population ranged from 12 the meta-analysis, 2 studies required full using a collagen barrier for GBR in combina-
to 64 patients. Following categorisation into integrity of all socket walls (Barone et al. tion with either a porcine or bovine xenograft
intervention groups, 280 patients were consid- 2012; Spinato et al. 2014), one study required (Vance et al. 2004 one group, Barone et al.
ered for the meta-analysis of GBR group, 242 minimal buccal bone loss (Cardaropoli et al. 2012, 2013a,b; Cook & Mealey 2013 one
patients for the meta-analysis of the socket fil- 2014) (<80%), whereas socket integrity was group, Barone et al. 2014; Gholami et al.
ler group and 60 patients for the meta-analysis unclear in the remaining two studies. In four 2012 one group, Perelman-Karmon et al.
of the socket seal group. The distribution of of the 10 included studies, ARP was per- 2012 one group, Patel et al. 2013 one
the extracted teeth was heterogeneous includ- formed using a collagen barrier for GBR in group, Cardaropoli et al. 2014), hydroxyapa-
ing both single- and multi-rooted teeth. Eleven combination with a porcine or bovine xeno- tite (Gholami et al. 2012 one group, Cook
studies included smokers (Barone et al. 2012; graft (Barone et al. 2012, 2013a; Cardaropoli & Mealey 2013 one group), or synthetic
Mardinger et al. 2012; Sisti et al. 2012; Barone et al. 2014) or an allograft (Iasella et al. 2003). ceramic (Patel et al. 2013 one group), or
et al. 2013a,b; Cook & Mealey 2013; Patel In four studies, a collagen sheet was combined freeze-dried bone allograft (Iasella et al. 2003;
et al. 2013; Poulias et al. 2013; Barone et al. with an alloplast (Crespi et al. 2009a; Sisti Leblebicioglu et al. 2013). An acellular der-
2014; Cardaropoli et al. 2014; Coomes et al. et al. 2012), xenograft (Crespi et al. 2011b) or mal matrix barrier in combination with an
2014; Eskow & Mealey 2014; ), and 5 studies allograft (Spinato et al. 2014). In one study, a allograft with or without the rhPDGF growth
non-smokers (Neiva et al. 2008; Perelman- porcine xenograft with a porcine cortical layer factor (Wallace et al. 2013) and a resorbable
Karmon et al. 2012; Wood & Mealey 2012; was used for grafting of the sockets (Festa polylactide barrier with cancellous allograft
Leblebicioglu et al. 2013; Spinato et al. 2014; et al. 2013), and in another study, a polylac- with or without bovine xenograft were used
), while five studies did not report on smoking tidepolyglycolide acid sponge was placed in the remaining study (Poulias et al. 2013).
habits (Iasella et al. 2003; Vance et al. 2004; (Serino et al. 2003). Socket filling was performed using either
Beck & Mealey 2010; Hoang & Mealey 2012; Implant placement was attempted at 1.9 allografts (Beck & Mealey 2010), xenografts
Lindhe et al. 2014; ). (Sisti et al. 2012), 3 (Crespi et al. 2009a), 4 (Crespi et al. 2009b one group, Crespi et al.
(Crespi et al. 2011b; Barone et al. 2013a; 2011a,b; -two groups, Crespi et al. 2011b;
Intervention characteristics Cardaropoli et al. 2014; Spinato et al. 2014), Perelman-Karmon et al. 2012; one group,
The study intervention characteristics of the 6 (Serino et al. 2003; Festa et al. 2013) or at Festa et al. 2013; Neiva et al. 2008; -one
included studies for both focused questions 7 months after ARP (Barone et al. 2012). In group), synthetic materials (Serino et al.
are also presented in Table 2. one study, implants were placed at 4 or 2003), a combination of alloplast and collagen
6 months (Iasella et al. 2003). In the studies (Crespi et al. 2009a,b; -two groups, Crespi
Controlled studies answering the focus question 1 where information on implant type was pro- et al. 2011a,b; -one group, Sisti et al. 2012), a
In four of the 10 included studies (Iasella vided, implants with a rough surface were combination of allograft and collagen (Wood
et al. 2003; Serino et al. 2003; Barone et al. placed; four studies did not report any infor- & Mealey 2012; Eskow & Mealey 2014;
Table 2. Study characteristics of included papers
Setting
(country, Who carried Number of
number, out patients Mean age SD Smokers
Reference type centre) Funding Study design procedures (sockets) and/or range included
Barone et al. Italy, 1 P Unclear RCT, Parallel Unclear 40 (40) 2669 Y
(2012)*,
Barone et al. Italy, 1 H Unclear RCT, Parallel Specialists 58 (58) 40.5 (2063) Y
(2013a)*,
Barone et al. Italy, Industry RCT, Unclear Unclear 38 (62) 51 14 Y

(2013b) Germany,
Spain; 6,
U and P?
Barone et al. Italy, 1 H Unclear RCT, Parallel Specialists 64 (64) 32.7 12.4 Y
(2014) (1847)
Beck & Mealey USA, 1 U Self-funded CCT, Parallel Unclear 33 (38) 57.4 (3976) ?
(2010)
Cardaropoli et al. Italy, 1 P Unclear RCT, Parallel Unclear 41 (48) 47.2 12.9 Y
(2014)*,
Cook & Mealey USA, 1 U Industry RCT, Parallel Specialist 38 (40) 56 (2378) Y
(2013) trainees
Coomes et al. USA, 1 U Industry RCT, Parallel Unclear 34 (34) 1979 Y

(2014)
Crespi et al. Italy, 1 H Unclear CCT, Specialists 15 (45) 51.3 (2872) N

(2009a) Split-mouth
Crespi et al. Italy, 1 U Unclear CCT, Specialists 15 (45) 54.6 (3468) N

(2009b) Unclear
Crespi et al. Italy, 1 H Unclear CCT, Split-mouth Specialists 15 (45) 53.7 (3270) N
(2011a)
Crespi et al. Italy, 1 H Unclear RCT, Split-mouth Specialists 15 (30) 53.7 (3270) N
(2011b)
Eskow & Mealey USA, 1? U Unclear RCT, Parallel Unclear 35 (35) 54 (2779) Y
(2014)
Reported
Pre- or follow-up
post- Healing after
Socket location and Materials (details, number of Atraumatic Flap Primary operative period Description of implant
defect morphology patients/sockets) extraction raised closure antibiotics (months) implants placed placement
Non-molar; 4 walls GBR (porcine bone + ? Y Y Y 7 Premium, Sweden & 3 years
collagen barrier; 20/20) vs Martina
USH (20/20)
Molar or premolar GBR (porcine bone + ? N N N 4 Ossean surface, Intra- None
resorbable collagen barrier; lock
29/29) vs USH control (29/
29)
Molar or premolar; GBR (bovine bone mineral Y Y N ? 6 NanoTite Tapered None
excluded if facial (BBM) + collagen barrier; Certain, Biomet 3i
soft tissue and ?/31; T1) vs GBR (bovine
buccal plate xenograft + resorbable
markedly reduced collagen barrier; ?/31; T2)
Molar or premolar; 4 GBR (corticocancellous Y T1: N T2: Y T1: N T2: Y Y 3 Intra-lock None
walls porcine bone + resorbable
collagen barrier; 32/32; T1)
vs GBR (corticocancellous
porcine bone + resorbable
Single root; excluded Grafting (nonfreeze-dried Y N N Y Approx 2.5 or Unclear None
if >50% of any cancellous mineralised approx 5.5
socket wall absent human bone allograft +
collagen; 19/22; T1) vs
grafting (nonfreeze-dried
cancellous mineralised
human bone allograft +
collagen; 14/16; T2)
Molar or premolar; 3 GBR (bovine bone mineral Y T1: Y C: N N Y 4 Unclear None
intact walls and at blended with collagen +
least 80% of fourth resorbable collagen barrier;
wall intact 21/24) vs unassisted healing
(20/24)
Non-molar; excluded GBR (bovine bone mineral Y Y N Y 45 Unclear 1 month
if bony dehiscence blended with collagen +
>50% of total resorbable collagen barrier;
socket depth 20/21; T1) vs GBR
(hydroxyapatite +
resorbable collagen barrier;
18/19; T2)
Buccal bone Grafting (collagen + rhBMP- Y N N Y 5 SLA or SLActive, None
destruction 2; 18/18; T1) vs grafting Institute Straumann
(collagen; 16/16; T2)
Molar or premolar; 3 Grafting (MHA + collagen; ? N N Y 3 Unclear None
bone walls and loss 15/15; T1) vs grafting (CS +
of buccal plate collagen; 15/15; T2) vs
unassisted healing (15/15)
3 walls and loss of Grafting (MHA + collagen; ? N N Y 3 Seven, Sweden & 24 months
buccal wall 14/15; T1) vs grafting (CS + Martina
grafting (corticocancellous
porcine bone + collagen;
15/15; T3)
One molar or Grafting (MHA + collagen; Y N N Y 4 Unclear None
premolar on each 15/15; T1) vs grafting
side of jaw and one (corticocancellous
additional randomly xenogenic bone + collagen;
located tooth to be 15/15; T2) vs grafting
used as a control (collagen; 15/15; T3)
One molar or Grafting (corticocancellous ? T1: ? C: N T1: N C: ? Y 4 Sweden-Martina None
premolar on each xenogenic bone + collagen;
side of jaw; 3 bone 15/15) vs unassissted
walls and loss of healing (15/15)
buccal plate
Non-molar; excluded Grafting (cortical FDBA + Y N N Y Approx. 4 Unclear None
if >50% of socket collagen; 17/17; T1) vs
walls vertical grafting (cancellous FDBA +
dimension absent collagen; 18/18; T2)
Table 2. (Continued)
Setting
Festa et al. (2013) Italy, 1 U Unclear RCT, Split-mouth Unclear 15 (30) 2858 N
Gholami et al. Iran, 1? Unclear RCT, Split-mouth Unclear 12 (28) 44.6 11.4 ?
(2012) (2160)
Hoang & Mealey USA, 1 U Self-funded RCT, Parallel Unclear 30 (30) 56.1 (2976) ?
(2012)
Iasella et al. Unclear Unclear RCT, Parallel Unclear 24 (24) 51.5 (2876) ?
(2003)*,
Leblebicioglu USA, 1 U Institutional Prospective Specialist 24 (25) 2483 N

et al. (2013) case series trainees
Lindhe et al. Unclear Unclear CCT, Parallel Unclear 24 (24) 2554 ?
(2014)
Mardinger et al. Israel, ?, Unclear Prospective Unclear 36 (43) 50.75 (2475) Y

(2012) U and P case series
Neiva et al. USA, 1 U Industry RCT, Parallel Specialists 24 (24) 2576 N

(2008)
Patel et al. (2013) UK, 1 U Industry RCT, Parallel Specialists 26 (26) 37.3 11.4 Y
(2058)
Perelman-Karmon Unclear Unclear RCT, Parallel Unclear 23 (23) 2668 N

et al. (2012)
Poulias et al. USA, 1 U Self-funded RCT, Parallel Specialist 23 (23) 52 16 (2677) Y

(2013) trainees
Serino et al. (2003) Unclear Unclear CCT, Unclear Unclear 36 (39) 3564 ?
Reported
Pre- or follow-up
post- Healing after
Premolars; excluded Grafting (corticocancellous Y Y T1: N C: Y Y 6 Unclear None
if buccal or palatal/ porcine bone + soft cortical
lingual bony wall membrane; 15/15) vs
fractured/lost unassisted healing (15/15)
Non-molar; four-wall GBR (DBBM + resorbable Y Y Y Y 68 Unclear None
sockets collagen barrier; 12/14; T1)
vs GBR (nanocrystalline HA
embedded in silica gel
matrix + resorbable
Molar; excluded if Grafting (demineralised Y ? N Y 45 Unclear None
buccal bony bone matrix in a carrier of
dehiscence >50% of bovine collagen and
length of socket sodium alginate + collagen;
16/16; T1) vs grafting
(demineralised bone matrix
in a carrier of bovine
collagen and sodium
alginate + collagen; 14/14;
T2)
Non-molar GBR (FDBA + resorbable Y Y N Y 4 or 6 Unclear 2 months
collagen barrier; 12/12) vs
unassissted healing (12/12)
Molar or premolar GBR (FDBA + resorbable Y Y ? Y 3.78 Unclear None
collagen barrier; 24/25)
Excluded if buccal Sealing (DBBM + Mucograft; Y N Y ? 6 Astra Tech System None
dehiscence defect 13/13; T1) vs sealing
2 mm (Mucograft; 11/11; T2)
Site not completely Sealing (porous bovine Y N Y Y 6 Seven MIS Implants 6 months
surrounded by bony xenograft + intrasocket Technologies;
walls; excluded if reactive soft tissue; 36/43) Tapered Screw-Vent,
less than two bony Zimmer Dental;
wall defects Screwplant Implant
Direct; Osseotite 3i/
Implant Innovations
Biomet
Maxillary premolars Grafting (anorganic bovine- Y N N ? 3.7 Unclear None
with >80% bone derived HA matrix
volume in all combined with a synthetic
dimensions cell binding peptide P-15 +
grafting (collagen; 12/12;
T2)
Non-molar; excluded GBR (60% HA + 40% b- Y Y N Y 8 Straumann standard 12 months
if major part of tricalcium phosphate + plus SLActive post-loading
buccal or palatal resorbable collagen barrier; implant
wall damaged or 13/13; T1) vs GBR (DBBM +
lost resorbable collagen barrier;
13/13; T2)
Non-molar; at least GBR (bovine bone mineral + ? Y Y N 9 Unclear None
50% of sockets resorbable collagen barrier;
partially resorbed/ 11/11; T1) vs grafting
destructed at one to (bovine mineral bone; 12/
two walls, but not 12; T2)
circunferentially
Non-molar GBR (cancellous allograft + Y Y N T1: N T2: ? 4 Unclear None
resorbable polylactide
barrier; 12/12; T1) vs GBR
(cancellous allograft +
bovine xenograft +
resorbable polylactide
barrier; 11/11; T2)
Unclear Grafting (polylactide ? Y ? N 6 Astra Tech None
polyglycolide acid sponge;
24/26) vs unassisted healing
(12/13)
Table 2. (Continued)
Setting
Sisti et al. Unclear Unclear RCT, Parallel Unclear 20 (20) 50.85 (3670) Y
(2012)*,
Spinato et al. Unclear, 3 P Self-funded RCT, Parallel Unclear 31 (31) 48.5 (2774) N
(2014)*,
Vance et al. Unclear Industry RCT, Parallel Unclear 24 (24) 56 ?

(2004)
Wallace et al. Unclear Industry CCT, Parallel Unclear 30 (34) 1870 N

(2013)
Wood & USA, 1 U Industry RCT, Parallel Specialist 33 (33) 56.7 (2078) N
Mealey trainees
(2012)
All studies included for Question 2; highlighted studies included for Question 1.
*Selected for meta-analysis Question 1.
Selected for meta-analysis Question 2.
U, university; H, hospital; P, private practice; Y, yes; N, no; ?, unclear; RCT, randomised clinical trial; CCT, controlled clinical trial; GBR, guided bone
regeneration; T1, test group 1; T2, test group 2; T3, test group 3; C, control group; MHA, magnesium-enriched hydroxyapatite; CS, calcium sulfate; FDBA,
freeze-dried bone allograft; DBBM, deproteinised bovine bone mineral; HA, hydroxyapatite; DFDBA, demineralised freeze-dried bone allograft.
Spinato et al. 2014), a combination of syn- Mealey 2012; Wood & Mealey 2012; Cook & ther ridge augmentation was 0.150 (95%
thetic polymer, ceramic material and allo- Mealey 2013; Festa et al. 2013; Leblebicioglu CI 0.0740.302) indicating a decrease in
graft (Vance et al. 2004; -one group), a et al. 2013; Cardaropoli et al. 2014; Eskow & the need for further ridge augmentation
bovine-derived hydroxyapatite combined with Mealey 2014), while in all other studies, when ARP was performed.
a synthetic peptide P-15 and collagen (Neiva implants with a moderate rough surface were 3. Implant survival: Eight of 10 included
et al. 2008; -one group), a demineralised bone placed. studies reported data on implant survival.
matrix in bovine collagen and sodium algi- All implants placed in the ARP sites sur-
nate carrier (Hoang & Mealey 2012), or a col- Outcome characteristics vived at 12 months post-loading. One
lagen carrier with and without rhBMP-2 The outcomes of the included studies for study (Barone et al. 2012) reported a 95%
(Coomes et al. 2014). both focused questions are presented in survival rate, where only one implant
Socket seal was performed with a porcine Table 3. placed in the untreated control sites did
collagen matrix and bovine allograft (Lindhe not survive at the same observation per-
et al. 2014) or with granulation tissue har- Controlled studies answering the focused question 1 iod. As in all of studies but one, the
vested from the extraction site and bovine 1. Feasibility of implant placement: all the implant survival was 100% for both ARP
xenograft (Mardinger et al. 2012). included studies reported that implant and control, meta-analysis was not
Implant placement was attempted at differ- placement was feasible in all the patients reported for this outcome variable.
ent healing periods ranging among all the of both test (ARP) and control (unassisted 4. Implant success: Eight of 10 included
included studies between 1.9 months (Sisti socket healing) groups. For this reason, studies reported data on implant success
et al. 2012) to 9 months (Perelman-Karmon no meta-analysis was performed for this at 12 months post-loading. Five of these
et al. 2012). In the GBR studies, the healing outcome variable. eight studies utilised the Albrektsson cri-
period ranged between 3 and 9 months in 2. Need for further augmentation: Nine of 10 teria (Albrektsson et al. 1986). One study
comparison with the socket filler studies included studies reported data on the need evaluated success based on the absence
where the healing period was less than for further ridge augmentation ranging of peri-implantitis (Serino et al. 2003),
5 months (range: 1.99 months). between 0% and 15% for the ARP-treated and another one on the absence of proxi-
In both studies that used a socket seal sites and between 0% and 100% in the mal radiographic bone loss more than
approach, the implants were placed at unassisted socket healing sites. Six studies 2 mm (Spinato et al. 2014). Another
6 months after ARP. Eleven studies did not were included in the meta-analysis for this study considered an implant restoration
report on the type of implants (Iasella et al. outcome variable (Fig. 2a), which appeared as successful when the implant was sta-
2003; Crespi et al. 2009a, 2011a; Beck & to be highly homogeneous (I2 = 0%, ble, without a radiolucent zone around it,
Mealey 2010; Gholami et al. 2012; Hoang & P = 0.707). The pooled relative risk for fur- no mucosal suppuration and no pain
Reported
Pre- or follow-up
post- Healing after
Non-molar; buccal Grafting (Mg-e HA granules Y N N Y 1.9 Premium, Sweden & None
bone defect >5 mm + collagen; 10/10) vs Martina
unassissted healing (10/10)
Maxillary non-molar; Grafting (cancellous Y N N Y 4 Tapered Screw Vent; None
four intact bony allograft + collagen; 19/19) Zimmer Dental
walls vs unassisted healing (12/
12)
Non-molar Grafting Y Y N Y 4 Stage-1; Lifecore 4 months
(carboxymethylcellulose + Biomedical
CS + DFDBA; 12/12; T1) vs
GBR (bovine bone mineral
+ resorbable collagen
barrier; 12/12; T2)
18 intact and 16 GBR (allograft + rhPDGF-BB Y Y Y ? 4 Internal RBT None
sockets with buccal + acellular dermal matrix Laser-Lok,
wall defects barrier; ?/19; T1) vs GBR BioHorizons
(allograft + acellular dermal
matrix barrier; ?/15; T2)
Single-rooted non- Grafting (DFDBA + collagen; Y ? N Y 44.7 Unclear None
molar 17/17; T1) vs grafting (FDBA
+ collagen; 16/16; T2)
(Crespi et al. 2011b). The success rates placement was not feasible at 8 months relevant cohorts/groups from 10 studies
were high for both test and control following ridge preservation with a por- (Vance et al. 2004; one group, Neiva et al.
groups ranging between 95.2% and 100% cine collagen barrier and bovine bone 2008; Beck & Mealey 2010; Hoang & Mealey
for the ARP-treated sites and between mineral (Patel et al. 2013). Eleven studies 2012; Perelman-Karmon et al. 2012 one
90% and 100% for unassisted socket contributed with one or two treatment group, Sisti et al. 2012; Wood & Mealey
healing sites. Five of these 8 studies were groups in the meta-analysis (Iasella et al. 2012; Coomes et al. 2014; Eskow & Mealey
included in the meta-analysis (Fig. 2b). 2003; Vance et al. 2004; one group, Ba- 2014; Spinato et al. 2014) were included in
The studies were highly homogeneous rone et al. 2012; Barone et al. 2013a; the meta-analysis (Fig. 3b). The studies
(I2 = 0%, P = 0.953), and the relative risk Cook & Mealey 2013; Leblebicioglu et al. appeared homogeneous (I2 = 0%, P = 0.791),
for implant failure was 1.055 (95% CI 2013; Patel et al. 2013; Perelman-Karmon and the estimated pooled size effect was
0.9451.177). et al. 2012 one group, Poulias et al. 96.204% (95% CI: 93.08998.182).
5. Marginal bone levels: Five studies 2013; Barone et al. 2014; Cardaropoli When a socket seal type of intervention
reported data on proximal bone levels et al. 2014). The studies were homoge- was used (Mardinger et al. 2012; Lindhe et al.
at 12 months post-loading (Crespi et al. neous (I2 = 0%, P = 0.999), and the esti- 2014), implant placement was feasible in all
2009a; Barone et al. 2012; Sisti et al. mated pooled size effect was 98.54% extraction sites treated, and for this reason,
2012; Barone et al. 2013a; Spinato (95% CI: 96.4299.58) (Fig. 3a). no meta-analysis was performed.
et al. 2014). One study was excluded from 2. Need for further augmentation: All stud-
Implant placement feasibility following
the meta-analysis (Crespi et al. 2009a). ies in the GBR group except one (Gho-
ARP with socket filler ranged between 88.9%
The studies appeared homogeneous lami et al. 2012) reported on the need
and 100%. Implant placement was not feasi-
(I2 = 0%, P = 0.881). The standardised for further ridge augmentation (range
ble in 8.3% of the sockets filled with a colla-
mean difference in implant proximal bone 069.2%) during implant placement.
gen wound dressing material implant
loss (mm) between ARP- and non-treated Treatment groups from 11 studies were
placement (Neiva et al. 2008), in 10.5% and
extraction sites was 0.039 mm (95% CI: included in the meta-analysis (Iasella
7.1% of the sockets filled with cancellous
0.358 to 0.280) (Fig. 2c). et al. 2003; Vance et al. 2004; one
nonfreeze-dried bone allograft after 2.5 and
group, Barone et al. 2012; Barone et al.
5 months of healing, respectively (Beck &
2013a; Cook & Mealey 2013; Leblebicio-
Studies answering the focus question 2 Mealey 2010), in 5.9% of the sockets filled
glu et al. 2013; Patel et al. 2013; Perel-
1. Feasibility of implant placement: Implant with demineralised freeze-dried bone allograft
man-Karmon et al. 2012 one group,
placement was feasible in all GBR-treated (Wood & Mealey 2012) and in 11.1% of the
Poulias et al. 2013; Barone et al. 2014;
sites with the exception of one study sockets treated with cancellous freeze-dried
Cardaropoli et al. 2014) (Fig. 3c). Due to
reporting that in one site (7.7%), implant bone allograft (Eskow & Mealey 2014). The
high heterogeneity among the studies Spinato et al. 2014). Six studies evaluated suc- Spinato et al. 2014). All these studies were
(I2 = 72.97%, P < 0.0001), a random effect cess based on Albrektsson criteria and the considered to have an unclear risk of bias. All
model was used for the analysis. The remaining three on either presence of implant other studies were considered to be at high
estimated pooled size effect was 11.85% stability, absence of a radiolucent zone around risk of bias. Reporting and attrition were the
(95% CI: 5.64219.952). the implants, no mucosal suppuration and no most common sources of bias, while selec-
pain (Crespi et al. 2009b) or absence of peri- tion, performance and other sources of bias
Two of 16 included in the qualitative implantitis (Serino et al. 2003) or radiographic were of less concern in most of the studies
analysis studies that used socket filler for proximal bone levels (Spinato 201). All these (Fig. 4b). In several studies, however, alloca-
ARP did not report on the need for further studies reported success rates of 100%; there- tion concealment, blinding of participants and
ridge augmentation (Beck & Mealey 2010; fore, no meta-analysis was performed. personnel or outcome assessment was not
Festa et al. 2013), which ranged between 0% None of the two studies utilising a socket clearly reported resulting in unclear risk of
and 81.3%. Treatment groups from 9 studies seal approach reported on implant success at bias for these domains.
were included in the meta-analysis (Vance 12 months post-loading.
et al. 2004; one group, Neiva et al. 2008; 5. Marginal bone levels: Data on radio-
Hoang & Mealey 2012; Perelman-Karmon Discussion
graphic proximal bone levels at
et al. 2012 one group, Sisti et al. 12 months post-loading were reported in
2012; Wood & Mealey 2012; Coomes Objective and main findings
4 studies where GBR has been used for
Recent systematic reviews and meta-analyses
et al. 2014; Eskow & Mealey 2014; Spinato ARP (Barone et al. 2012, 2013a,b; Patel
demonstrated that although post-extraction
et al. 2014) (Fig. 3d). Due to the high hetero- et al. 2013). The average radiographic
resorption of the alveolar ridge cannot be
geneity among the studies (I2 = 81.22%, bone loss in implants proximal sites ran-
totally prevented, various ARP procedures
P < 0.0001), a random effect model was used ged among these studies between
will reduce vertical and horizontal hard and
for this analysis. The estimated pooled size 0.12 0.4 mm and 1.4 0.9 mm. One
soft tissue dimensional changes and will sup-
effect was 13.65% (95% CI: 5.04225.588). study was excluded from meta-analysis
port new bone formation in the extraction
Finally, no further augmentation was nec- (Barone et al. 2013b) because of unclear
socket (Horvath et al. 2013, Vignoletti et al.
essary in any extraction site for both study design and the patient not being
2012; Wang & Lang 2012; De Risi et al. 2013;
included studies when a socket seal type of the unit of analysis. As only 3 studies
Avila-Ortiz et al. 2014; Morjaria et al. 2014).
intervention was used. were finally included in the meta-analy-
However, the same systematic reviews em-
3. Implant survival: The qualitative analysis sis, the results are not presented.
phasised the fact that there is limited evidence
included seven of 14 studies that used Five of 16 studies included in the qualita-
on the influence of ARP on implant-related
GBR and nine of the 16 studies that used tive analysis that used socket filler reported
outcomes. Considering that these procedures
socket filler together with the two stud- on radiographic proximal bone levels at 12-
are performed mostly to facilitate implant
ies that used socket seal reporting data month post-loading (Neiva et al. 2008; Crespi
placement in the post-extraction sites, such
on implant survival. In all studies, the et al. 2009a,b; Sisti et al. 2012; Spinato et al.
information could be of high clinical relevance
implant survival for the ARP-treated sites 2014). Average radiographic bone loss ranged
as clinicians may argue against the extensive
was 100%; therefore, meta-analysis was between 0.11 0.08 mm and 1.28
use of ARP procedures if implant placement
not performed for this outcome variable. 0.32 mm. A meta-analysis was not performed
feasibility and need for ridge augmentation are
4. Implant success: The qualitative analysis for this type of intervention as two studies
not significantly decreased in comparison with
included six of 14 included studies that were excluded because of split-mouth design
unassisted socket healing (Horv ath et al. 2012,
used GBR reporting data on implant suc- (Crespi et al. 2009a,b) and one study because
Vignoletti et al. 2012; Wang & Lang 2012; De
cess at 12-month post-loading based on no standard deviation/standard error of the
Risi et al. 2013; Avila-Ortiz et al. 2014; Morja-
the Albrektsson criteria. The success mean was reported (Neiva et al. 2008). None
ria et al. 2014). Therefore, the present system-
rates in the different GBR treatment of the two studies utilising a socket seal
atic review evaluated the evidence deriving
groups in these studies ranged between approach reported on radiographic proximal
from existing RCTs, CCTs and large prospec-
83.3% and 100%. Five studies contrib- bone levels.
tive case series that reported on implant-
uted one or two treatment groups to the
Quality assessment & risk of bias related outcomes following different ARP sur-
meta-analysis (Barone et al. 2012, 2013a;
The quality assessment of all the included gical protocols. The outcome variables selected
Perelman-Karmon et al. 2012; Patel et al.
studies for both focused questions is pre- were implant placement feasibility, need for
2013; Cardaropoli et al. 2014) (Fig. 3e).
sented in Fig. 4a and b. There was only one further ridge augmentation during implant
The studies were not homogeneous
study (Neiva et al. 2008) complying with all placement, implant survival/success rates and
(I2 = 54.77%, P = 0.0503), and the ran-
the criteria for bias, and it was considered to proximal bone levels of the implants placed in
dom effect model showed an estimated
have a low risk of bias. Three studies pre- the preserved alveolar ridges at least
pooled size effect for implant success of
sented low risk of bias in 5 of the 6 domains 12 months after implant loading. The meta-
93.976% (95% CI 85.70898.833).
and unclear risk in one domain (Barone analyses results from both focused questions
Nine of 16 included in the qualitative analy- et al. 2012, 2014; Gholami et al. 2012), one demonstrated that dental implants could be
sis studies that used socket filler reported data study presented low risk in four domains placed in the vast majority of the patients that
on implant success at 12-month post-loading and unclear in 2 (Patel et al. 2013) and three were treated with ARP and that all implants
(Serino et al. 2003; Neiva et al. 2008; Crespi studies presented low risk of bias in three survived and presented high success rates with-
et al. 2009a,b, 2011a,b; Perelman-Karmon domains and unclear risk in the other three out significant proximal bone loss indepen-
et al. 2012; one group, Sisti et al. 2012; (Festa et al. 2013; Cardaropoli et al. 2014; dently of the intervention for ARP used. These
Table 3. Study outcomes of included papers
Implant placement Need for further Implant survival at Implant success at Marginal bone loss (mean SD or
Reference Comparison feasibility augmentation 12 months 12 months* other if provided)
Barone et al. (2012), GBR vs unassissted healing T 100%, C 100% T 15%, C 50% T 100%, C 95% T 100%, C 95% T 0.75 0.3, C 0.76 0.3
Barone et al. (2013a), GBR vs unassissted healing T 100%, C 100% T 6.9%, C 44.8% T 100%, C 100% T 100%, C 100% T 0.9 0.8, C 0.9 0.7
Barone et al. (2013b) GBR (T1) vs GBR (T2) T1 100% T2 100% T1 3.2%, T2 6.5% T1 100%, T2 96.8% T1 96.8%, T2 93.5% T1 1.2 0.8, T2 1.4 0.9
(data per socket) (data per socket) (data per socket) (data per socket)
Barone et al. (2014) GBR (T1) vs GBR (T2) T1 100%, 100% T1 6.3%, T2 9.4% T1 100%, T2 100%
Beck & Mealey (2010) Grafting (T1) vs grafting (T2) T1 89.5%, T2 92.9%
Cardaropoli et al. (2014), GBR vs unassissted healing T 100%, C 100% T 4.8%, C 60% T 100%, C 100% T 95.2%, C 90%
Cook & Mealey (2013) GBR (T1) vs GBR (T2) T1 100%, T2 100% T1 10%, T2 16.7%
Coomes et al. (2014) Grafting (T1) vs grafting (T2) T1 100%, T2 100% T1 33.3%, T2 81.3%
Crespi et al. (2009a) Grafting (T1) vs grafting (T2) T1 100%, T2 100%, T1 0%, T2 0%, C 0% T1 100%, T2 100%, T1 100%, T2 100%, T1 0.18 0.09, T2 0.16 0.12,
vs unassisted healing C 100% C 100% C 100% C 0.52 0.23
Crespi et al. (2009b) Grafting (T1) vs grafting (T2) T1 100%, T2 100%, T1 0%, T2 0%, T3 0% T1 100%, T2 100%, T1 100%, T2 100%, T1 0.19 0.09, T2 0.11 0.08,
vs grafting (T3) T3 100% T3 100% T3 100% T3 0.13 0.10
Crespi et al. (2011a) Grafting (T1) vs grafting (T2) T1 100%, T2 100%, T1 0%, T2 0%, T3 0% T1 100%, T2 100%, T1 100%, T2 100%,
vs grafting (T3) T3 100% T3 100% T3 100%
Crespi et al. (2011b) Grafting vs unassissted healing T1 100%, C 100% T 0%, C 0% T 100% C 100% T 100% C 100%
Eskow & Mealey (2014) Grafting (T1) vs grafting (T2) T1 100%, T2 88.9% T1 23.5%, T2 27.8%
Festa et al. (2013) Grafting vs unassisted healing T 100%, C 100%
Gholami et al. (2012) GBR (T1) vs GBR (T2) T1 100% T2 100%
2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Hoang & Mealey (2012) Grafting (T1) vs grafting (T2) T1 100%, T2 100% T1 6.3%, T2 14.3%
Iasella et al. (2003), GBR vs unassissted healing T 100%, C 100% T 0%, C 25%
Leblebicioglu et al. (2013) GBR 100% 8.3%
Lindhe et al. (2014) Sealing (T1) vs sealing (T2) T1 100%, T2 100% T1 0%, T2 0% T1 100%, T2 100%
Mardinger et al. (2012) Sealing 100% 0% 100%
Neiva et al. (2008) Grafting (T1) vs grafting (T2) T1 100%, T2 91.7% T1 0%, T2 33.3% T 100%, T2 100% T1 100%, T2 100% T1 0.32 (range 00.5), T2 0.79
(range 0.51.0)
Patel et al. (2013) GBR (T1) vs GBR (T2) T1 100%, T2 92.3% T1 69.2%, T2 61.5% T1 100%, T2 100% T1 84.6%, T2 83.3% T1 0.12 0.4, T2 0.2 0.58
Perelman-Karmon GBR (T1) vs grafting (T2) T1 100%, T2 100% T1 0%, T2 0% T1 100%, T2 100% T1 100%, T2 100%
et al. (2012)
Poulias et al. (2013) GBR (T1) vs GBR (T2) T1 100%, T2 100% T1 0%, T2 0%
Serino et al. (2003) Grafting vs unassisted healing T 100%, C 100% T 0%, C 0% T 100%, C 100% T 100%, C 100%
Sisti et al. (2012), Grafting vs unassisted healing T 100%, C 100% T 0%, C 100% T 100%, C 100% T 100%, C 100% T 1.28 0.32, C 1.45 0.58
Spinato et al. (2014), Grafting vs unassisted healing T 100%, C 100% T 5.3%, C 16.7% T 100%, C 100% T 100%, C 100% T 0.68 0.72, C 0.61 0.8
Vance et al. (2004) Grafting (T1) vs GBR (T2) T1 100%, T2 100% T1 0%, T2 0%
Wallace et al. (2013) GBR (T1) vs GBR (T2) T1 100%, T2 100% T1 0%, T2 0%
Wood & Mealey (2012) Grafting (T1) vs grafting (T2) T1 94.1%, T2 100% T1 0%, T2 0%
All studies included for Question 2; highlighted studies included for Question 1.
*Most studies used (Albrektsson et al. 1986) success criteria; criteria (Crespi et al. 2009b) presence of implant stability, absence of a radiolucent zone around the implants, no mucosal suppuration and
no pain; (Serino et al. 2003) no Peri-implantitis; (Spinato et al. 2014) marginal bone loss < 2 mm both mesially and distally.
Information provided by author.
193 | Clin. Oral Impl. Res. 26 (Suppl. 11), 2015 / 180201

(a) Meta-analysis superior to that of unassisted socket healing.

For the second focused question, controlled
Barone 2012 studies where the control group was other
Barone 2013a than unassisted socket healing and large pro-
Cardaropoli 2014 spective case series were also included in the
search in order to utilise as many of the
Iasella 2003
available data as possible. The studies were
Sisti 2012
categorised according to the type of interven-
Spinato 2014 tion for ARP in order to provide average inci-
dences for each of the previously described
Total (fixed effects) implant-related outcomes according to the
Total (random effects) type of intervention used for ARP.
Although a comprehensive search strategy
0.001 0.01 0.1 1 10 including five databases, extensive hand and
Relative risk cross-reference search and no language
restriction were applied, it is possible that
(b) Meta-analysis
some grey literature may not have been
Barone 2012
included as only published studies were
selected. In order to obtain as much data as
Barone 2013a
possible from published studies, the authors
Cardaropoli 2014 of 34 of 103 studies selected for full-text
screening were conducted with e-mails in
Sisti 2012
order to provide further information and clar-
Spinato 2014 ifications especially in relation with implant
outcomes. Therefore, a large part of the pro-
Total (fixed effects) vided information that was used in the analy-
Total (random effects)
ses for both focused questions was provided
directly by the authors and is not published
0.1 1 10
in the original papers. On the other hand,
Relative risk some authors failed to respond within the
requested period of time; therefore, it is pos-
(c) Meta-analysis sible that the available data were more than
what has been reported.
Barone 2012
The total number of subjects and selected
Barone 2013a studies for focused question 1 could be con-
sidered sufficient for the assessment of effect
Sisti 2012
size differences between ARP and unassisted
Spinato 2014 socket healing for some of the implant-
related outcomes like the need for further
ridge augmentation and implant placement
Total (fixed effects)
feasibility. Similarly, the total number of
Total (random effects) patients/studies included in the meta-analy-
sis for focused question 2 could be considered
1.5 1.0 0.5 0.0 0.5 1.0 sufficient for the assessment of average inci-
Standardized dences of some of the investigated outcomes,
Mean Difference for example need for further augmentation,
Fig. 2. (a) Meta-analysis results and heterogeneity test for Q1; Relative risk for need for further augmentation. (b) when GBR or socket fillers were used for
Meta-analysis results and heterogeneity test for Q1; Relative risk for implant success at 12 months post-loading. (c) ARP. However, limited data were available to
Meta-analysis results and heterogeneity test for Q1; standardised mean difference for marginal bone loss. evaluate the influence of socket seal tech-
niques on implant-related outcomes. Further-
data are in agreement with previous systematic focused questions. For the first focused ques- more, the meta-analysis results for some
reviews assessing qualitatively implant-related tion, the search and analysis were limited to treatment outcomes (e.g. success rates and
outcomes in general (Horvath et al. 2012, Vig- RCTs, CCTs and prospective cohort studies marginal bone loss) should be evaluated with
noletti et al. 2012). with a control group of unassisted socket caution, given the limited number of studies
healing, based on the fact that the clinical reporting on these outcomes that were eligi-
Strengths & weakness of the systematic review merit of applying ARP in order to facilitate ble for the quantitative analysis.
In comparison with previous systematic implant treatment could only be validated, if Finally, the sample sizes of the selected tri-
reviews, the present review has evaluated implant placement feasibility, need for fur- als were relatively small and only few trials
exclusively implant-related outcomes follow- ther ridge augmentation and implant sur- included a sample size calculation; this may
ing ARP and tried to answer two different vival, success and marginal bone loss are have reduced the power of the studies.
Meta-analysis (d) Meta-analysis

(a)
Coomes 2014 - 1
Barone 2012
Barone 2013a Coomes 2014 - 2
Barone 2014 - 1 Eskow 2014 - 1
Barone 2014 - 2 Eskow 2014 - 2
Cardaropoli 2014 Hoang 2012 - 1
Cook 2012 - 1 Hoang 2012 - 2
Cook 2012 - 2 Neiva 2008 - 1
Iasella 2003 Neiva 2008 - 2
Patel 2013 - 1
Patel 2013 - 2 Perelman-Karmon 2012 - 2
Perelman-Karmon 2012 - 1 Sisti 2012
Poulias 2013 - 1 Spinato 2014
Poulias 2013 - 2 Vance 2004 - 1
Vance 2004 - 2 Wood 2012 - 1
Leblebicioglu 2013 Wood 2012 - 2
0.6 0.7 0.8 0.9 1.0 0.0 0.2 0.4 0.6 0.8 1.0
Proportion Proportion
(b) Meta-analysis (e) Meta-analysis
Beck 2010 - 1
Beck 2010 - 2 Barone 2012
Coomes 2014 - 1
Coomes 2014 - 2 Barone 2013a
Eskow 2014 - 1
Eskow 2014 - 2 Cardaropoli 2014
Hoang 2012 - 1
Hoang 2012 - 2
Patel 2013 - 1
Neiva 2008 - 1
Neiva 2008 - 2 Patel 2013 - 2
Perelman-Karmon 2012 - 2
Sisti 2012 Perelman-Karmon 2012 - 1
Spinato 2014
Vance 2004 - 1
Wood 2012 - 1 Total (fixed effects)
Wood 2012 - 2
Total (fixed effects) Total (random effects)
0.6 0.7 0.8 0.9 1.0 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Proportion Proportion
(c) Meta-analysis
Barone 2012
Barone 2013a
Barone 2014 - 1
Barone 2014 - 2
Cardaropoli 2014
Cook 2012 - 1
Cook 2012 - 2
Iasella 2003
Patel 2013 - 1
Patel 2013 - 2
Perelman-Karmon 2012 - 1
Poulias 2013 - 1
Poulias 2013 - 2
Vance 2004 - 2
Leblebicioglu 2013
0.0 0.2 0.4 0.6 0.8 1.0

Proportion
Fig. 3. (a) Meta-analysis results and heterogeneity test for Q2; estimated size effects for implant placement feasibility in GBR-treated sites. (b) Meta-analysis results and hetero-
geneity test for Q2; estimated size effects for implant placement feasibility in sites treated with a socket filler. (c) Meta-analysis results and heterogeneity test for Q2; estimated
size effects for need further augmentation in GBR-treated sites. (d) Meta-analysis results and heterogeneity test for Q2; estimated size effects for need further augmentation in
sites treated with a socket filler. (e) Meta-analysis results and heterogeneity test for Q2; estimated size effects for implant success at 12 months post-loading in GBR-treated
sites.
(a) (b)
Fig. 4. (a) Quality assessment of the included papers: Risk of Bias Summary. (b) Quality assessment of the included papers: Risk of Bias Graph. Please note that the risk of bias
evaluation is based on the original publications only.
Focused Question 1 analysis of a previous systematic review implant placement was feasible in all the
Qualitative and quantitative analysis of the (Horvath et al. 2012) or studies that were not patients of both test (ARP) and control (unas-
included studies for focused question 1 dem- included in the specific analysis due to differ- sisted socket healing) groups. These results
onstrated that the application of ARP proce- ent reasons (Fiorellini et al. 2005; Pellegrini are in agreement with Serino et al. 2008
dures will significantly decrease the need for et al. 2014). This was actually the only out- (excluded from the current analysis) who
further ridge augmentation during implant come variable where statistically significant reported that implants could be placed in all
placement in comparison with unassisted differences were detected between ARP and patients independently of applying ARP or
socket healing (Table 3, Fig. 2a). This obser- unassisted socket healing. Interestingly not. It could be argued, however, that several
vation was in agreement with the qualitative enough, all the included studies reported that surgical factors such as the anatomic loca-
tion, the angulation of implant placement interventions, we could speculate that differ- the included studies presented with different
according to a prosthetically driven implant ences in implant size and design; the surgical anatomical and dimensional characteristics
placement protocol or the diameter of the protocol; implant angulation; and the ana- and gingival tissue biotypes and implants
used implants may have influenced both the tomic location may have contributed to the were placed at various healing periods after
implant placement feasibility and the need great variance observed. The high level of ARP. This lack of consistency and standardi-
for further ridge augmentation (Mardas et al. heterogeneity between the studies included sation may have contributed to the high het-
2010). in each intervention type may also reflect erogeneity observed especially in relation to
In the present study, there was no clear these differences. the need for further augmentation when
evidence that ARP procedures increased extraction sites were treated with GBR or
implant placement feasibility, improved the Risk of bias, quality assessment & confounding socket fillers and should be taken under con-
survival or success of the implants placed in factors sideration in the interpretation of the results
The quality of the included studies for both
post-extraction alveolar ridges or contributed of both meta-analyses.
focused questions has been assessed in this
in maintenance of marginal proximal bone Previous systematic reviews suggested a
review to estimate the source and magnitude
levels better than unassisted socket healing. possible beneficial effect of flap elevation
of potential bias that may lead to inaccurate
Histological healing of the socket should be (Vignoletti et al. 2012; Avila-Ortiz et al.
conclusions. Based on the quality assess-
also considered when implant placement is 2014), use of membrane (Horv ath et al. 2012,
ment, the results presented should be evalu-
scheduled after extraction with or without Vignoletti et al. 2012; Avila-Ortiz et al. 2014)
ated with caution as only one study included
ARP (H ammerle et al. 2012). It is possible and the use of xenograft or allograft (Avila-
in the qualitative and quantitative analysis
that besides a reduction in dimensional Ortiz et al. 2014) on preservation of pre-
for focus question 2 presented with a low
changes post-extraction, ARP does not fur- extraction ridge dimensions. In the present
risk of bias (Neiva et al. 2008), while none of
ther promote or accelerate new bone forma- study, it was not possible to apply any sub-
the trials included in the qualitative and
tion in comparison with unassisted socket group analyses exploring potential differences
quantitative analysis for focus question 1
healing, been unable to guide the histological in treatment effect between flap and flapless
have qualified as low risk of bias. Six of the
events or limited physiologic resorption of protocols, different bone grafts/fillers or dif-
10 studies included in the analysis for ques-
the bundle bone. (H ammerle et al. 2012; De ferent healing periods; therefore, we were not
tion 1, and 22 of the 30 studies for question 2
Risi et al. 2013). able to estimate the effect of these confound-
were qualified for a high risk of bias. These
ers on implant treatment outcomes. On the
studies presented with high risk of bias in at
Focused Question 2 other hand, this systematic review found
Although various surgical techniques and least 2 domains. Similar concerns about the
comparable implant-related outcomes follow-
materials have been used for ARP, no mate- quality of the currently available studies on
ing ARP with GBR and socket fillers, failing
rial or type of ARP intervention can be the effect of different ARP procedures have
to identify a beneficial effect of using a bar-
claimed to yield superior results to another been raised in other systematic reviews (Vig-
rier membrane on any of the investigated
(Horvath et al. 2012). Previous systematic noletti et al. 2012; Horvath et al. 2013; Mor-
implant-related outcomes. Furthermore, the
reviews concluded that the use of barriers for jaria et al. 2014). Inadequate or selective
present qualitative analysis showed that the
GBR appeared to be more effective in limit- reporting and incomplete data outcomes were
use of allografts for socket grafting in 3 stud-
ing post-extraction dimensional changes of the most common sources of bias in our
ies (Beck & Mealey 2010; Wood & Mealey
the alveolar ridge (Horv
ath et al. 2012, Vigno- study.
2012 Eskow & Mealey 2014) resulted in
letti et al. 2012; Avila-Ortiz et al. 2014). In The lack of universally accepted success
decreased implant placement feasibility in
the present systematic review, we evaluated and survival criteria for implant-supported
comparison with other grafting/filler materi-
implant-related outcomes following ARP restorations is a significant obstacle in com-
als. This finding is in contradiction with the
with three different types of interventions; paring the different studies and surgical pro-
systematic review by De Risi et al. (2013)
GBR, socket filler and socket seal. Although tocols in implant dentistry (Donos et al.
who reported higher values of bone formation
direct statistical comparison was not possi- 2008). The fact that different success criteria
following ARP with allografts in comparison
ble, the reported average incidences for were used in the included studies, and pro-
with xenografts. Differences in the clinical
implant placement feasibility or need for fur- gressive marginal bone loss around the
management of these materials and in
ther augmentation were not much different implants was evaluated only short term,
implant placement protocol, as well as the
between subjects treated with one interven- using in most of the cases not standardised
fact that different studies were included in
tion or another. In the vast majority of the X-rays, makes difficult the interpretation of
the two systematic reviews, should be con-
subjects, implants were successfully placed the results for these treatment outcomes.
sidered for these discrepancies. Currently,
and restored, while all implants survived at A plethora of confounding factors may
there is no evidence correlating histological
12 months after loading independently of the have influenced implant-related outcomes
healing and implant-related outcomes and it
intervention used for ARP. However, the following ARP procedures. In the included
is not known which type of grafting material
need for further augmentation varied signifi- studies, different biomaterials, surgical tech-
could serve superior to another in terms of
cantly within the different studies and type niques and protocols have been combined in
histological healing and implant-related out-
of interventions ranging from 0 to 81.3% in the qualitative and quantitative analyses for
comes.
total indicating that several factors besides both focused questions. The different ARP
The timing of implant placement following
the type of intervention may have contrib- procedures have been applied after extraction
ARP varied significantly between the
uted to the clinical decision to perform fur- of different types of teeth, due to different
included studies for both focused questions
ther grafting during implant placement. reasons in a great variety of individuals,
(Table 2). It might be argued that longer heal-
Besides the different materials and type of including smokers. The extraction sockets in
ing periods could have improved implant-
related outcomes allowing more time for the 4. No evidence was identified to inform on bone plate, flap reflection and closure,
mineralisation of bone tissue in the socket. the possible superior impact of a type of should be investigated.
However, this was not obvious in the present ARP intervention (GBR, socket filler and 4. In all future trials, special emphasis
systematic review where comparable survival socket seal) on implant outcomes. Cur- should be given in the following issues:
and success rates were achieved between rently, it is not known if a biomaterial or
a. Decrease heterogeneity and control all
studies placing implants at different healing a treatment protocol is superior to others.
the previously reported sources of bias
periods after ARP. Similar suggestions have 5. The majority of the studies evaluating
b. Need for further augmentation and
been made in another systematic review eval- implant-related outcomes after ARP pro-
implant placement feasibility should
uating histological outcomes following ARP cedures are presenting high or unclear
be evaluated based on a prosthetically
where the authors suggested that implant risk of bias; therefore, any clinical recom-
driven implant placement protocol
placement could be performed after 3 or mendation derived from these studies
and implant size selection.
4 months of healing independently of the should be applied with caution.
c. Survival & success of implant treat-
grafting materials used (De Risi et al. 2013).
ment should be evaluated on standar-
Recommendations for further research dised and universally accepted set of
Conclusions 1. There is still a need for high-quality criteria that will allow a better com-
RCTs on adequately powered sample parison between the studies.
Within the limitations of present study, the sizes to evaluate differences in implant- d. Radiographic assessment of marginal
following conclusions can be drawn: related outcomes between specific bone levels should be performed on
1. Alveolar ridge preservation procedures ARP procedures and unassisted socket standardised radiographs taken at spe-
may decrease the need for further ridge healing. cific period of times including implant
augmentation during implant placement 2. There is significant need for high-quality placement, implant loading and a
in comparison with unassisted socket RCTs on adequately powered sample minimum of one year after loading.
healing. sizes to evaluate differences in implant- Long-term studies evaluating marginal
2. There is no evidence to support the fact related outcomes between different proto- bone levels of implants placed in alve-
that implant placement feasibility is cols and materials for ARP in order to olar ridges previously treated with
increased following ARP in comparison identify the most successful treatment ARP are not currently available.
with unassisted socket healing. options. Socket seal procedures should be e. Aesthetic indexes of final implant-
3. The survival, success and marginal bone further investigated in comparison with supported prostheses, patient based
levels of implants placed in alveolar other ARP interventions. outcomes and cost-benefit indicators
ridges following ARP are comparable 3. The role of possible confounding factors, should be included in future trials
to that of implants placed in untreated like smoking, reason for extraction, tooth evaluating implant-related outcomes
sockets. type and location, integrity of buccal following ARP.
References
Albrektsson, T., Zarb, G., Worthington, P. & Eriks- versus extraction alone for ridge preservation comparison between flapless and flapped proce-
son, A.R. (1986) The long-term efficacy of cur- after tooth removal: a clinical and histomorpho- dures in a randomized clinical trial. Journal of
rently used dental implants: a review and metric study. Journal of Periodontology 79: 1370 Periodontology 85: 1423.
proposed criteria of success. International Journal 1377. Beck, T.M. & Mealey, B.L. (2010) Histologic analysis
of Oral Maxillofacial Implants 1: 1125. Barone, A., Orlando, B., Cingano, L., Marconcini, of healing after tooth extraction with ridge preser-
Araujo, M.G., da Silva, J.C., de Mendonca, A.F. & S., Derchi, G. & Covani, U. (2012) A randomized vation using mineralized human bone allograft.
Lindhe, J. (2014) Ridge alterations following graft- clinical trial to evaluate and compare implants Journal of Periodontology 81: 17651772.
ing of fresh extraction sockets in man. A random- placed in augmented versus non-augmented Becker, W., Becker, B.E. & Caffesse, R. (1994) A
ized clinical trial. Clinical Oral Implants extraction sockets: 3-year results. Journal of Peri- comparison of demineralized freeze-dried bone
Research 26: 407412 [Correction added after first odontology 83: 836846. and autologous bone to induce bone formation in
online publication: volume and page numbers Barone, A., Ricci, M., Tonelli, P., Santini, S. & Co- human extraction sockets. Journal of Periodontol-
corrected]. vani, U. (2013a) Tissue changes of extraction ogy. 65: 11281133.
Araujo, M.G. & Lindhe, J. (2005) Dimensional ridge sockets in humans: a comparison of spontaneous Becker, W., Urist, M., Becker, B.E., Jackson, W.,
alterations following tooth extraction. An experi- healing vs. Ridge preservation with secondary soft Parry, D.A., Bartold, M., Vincenzzi, G., De
mental study in the dog. Journal of Clinical Peri- tissue healing. Clinical Oral Implants Research Georges, D. & Niederwanger, M. (1996) Clinical
odontology 32: 212218. 24: 12311237. and histologic observations of sites implanted
Atwood, D.A. (1971) Reduction of residual ridges: a Barone, A., Todisco, M., Ludovichetti, M., Gualini, with intraoral autologous bone grafts or allografts.
major oral disease entity. Journal Prosthetic Den- F., Aggstaller, H., Torres-Lagares, D., Rohrer, M.D., 15 human case reports. Journal of Periodontology
tistry 26: 266279. Prasad, H.S. & Kenealy, J.N. (2013b) A prospective, 67: 10251033.
Avila-Ortiz, G., Rodriguez, J.C., Rudek, I., Bena- randomized, controlled, multicenter evaluation of Cardaropoli, G., Araujo, M. & Lindhe, J. (2003)
vides, E., Rios, H. & Wang, H.-L. (2014) Effective- extraction socket preservation comparing two Dynamics of bone tissue formation in tooth
ness of three different alveolar ridge preservation bovine xenografts: clinical and histologic out- extraction sites. An experimental study in dogs.
techniques: a pilot randomized controlled trial. comes. International Journal of Periodontics and Journal of Clinical Periodontology 30: 809818.
The International Journal of Periodontics and Restorative Dentistry 33: 795802. Cardaropoli, D., Tamagnone, L., Roffredo, A. &
Restorative Dentistry 34: 509521. Barone, A., Toti, P., Piattelli, A., Iezzi, G., Derchi, Gaveglio, L. (2014) Relationship between the buc-
Barone, A., Aldini, N.N., Fini, M., Giardino, R., G. & Covani, U. (2014) Extraction socket healing cal bone plate thickness and the healing of
Calvo Guirado, J.L. & Covani, U. (2008) Xenograft in humans after ridge preservation techniques: postextraction sockets with/without ridge preser-
vation. The International Journal of Periodontics Fiorellini, J.P., Howell, T.H., Cochran, D., Malm- maintenance following tooth extraction. Report
and Restorative Dentistry 34: 211217. quist, J., Lilly, L.C., Spagnoli, D., Toljanic, J., of 10 cases. Journal of Periodontology 68: 563
Chen, S.T. & Buser, D. (2009) Clinical and esthetic Jones, A. & Nevins, M. (2005) Randomized study 570.
outcomes of implants placed in post extraction evaluating recombinant human bone morphoge- Lindhe, J., Cecchinato, D., Donati, M., Tomasi, C.
sites. International Journal of Oral & Maxillofa- netic protein-2 for extraction socket augmenta- & Liljenberg, B. (2014) Ridge preservation with
cial Implants 24(Suppl.): 186217. tion. Journal of Periodontology 76: 605613. the use of deproteinized bovine bone mineral.
Cook, D.C. & Mealey, B.L. (2013) Histologic compari- Freeman, M.F. & Tukey, J.W. (1950) Transforma- Clinical Oral Implants Research 25: 786790.
son of healing following tooth extraction with ridge tions related to the angular and the square root. Mardas, N., Chadha, V. & Donos, N. (2010) Alveo-
preservation using two different xenograft proto- The Annals of Mathematical Statistics 21: 607 lar ridge preservation with guided bone regenera-
cols. Journal of Periodontology 84: 585894. 611. tion and a synthetic bone substitute or a bovine-
Coomes, A.M., Mealey, B.L., Huynh-Ba, G., Barboza- Froum, S., Cho, S.-C., Rosenberg, E., Rohrer, M. & derived xenograft: a randomized, controlled clini-
Arguello, C., Moore, W.S. & Cochran, D.L. (2014) Tarnow, D. (2002) Histological comparison of cal trial. Clinical Oral Implants Research 21:
Buccal bone formation after flapless extraction: a healing extraction sockets implanted with bioac- 688698.
randomized, controlled clinical trial comparing tive glass or demineralized freeze-dried bone allo- Mardinger, O., Vered, M., Chaushu, G. & Nissan, J.
recombinant human bone morphogenetic protein graft: a pilot study. Journal of Periodontology 73: (2012) Histomorphometrical analysis following
2/absorbable collagen carrier and collagen sponge 94102. augmentation of infected extraction sites exhibit-
alone. Journal of Periodontology 85: 525535. Gholami, G.A., Najafi, B., Mashhadiabbas, F., Goe- ing severe bone loss and primarily closed by in-
Crespi, R., Cappare, P. & Gherlone, E. (2009a) Dental tz, W. & Najafi, S. (2012) Clinical, histologic and trasocket reactive soft tissue. Clinical Implant
implants placed in extraction sites grafted with dif- histomorphometric evaluation of socket preserva- Dentistry and Related Research 14: 359365.
ferent bone substitutes: radiographic evaluation at tion using a synthetic nanocrystalline hydroxyap- Morjaria, K.R., Wilson, R. & Palmer, R.M. (2014)
24 months. Journal of Periodontology 80: 16161621. atite in comparison with a bovine xenograft: a Bone healing after tooth extraction with or with-
Crespi, R., Cappare, P. & Gherlone, E. (2009b) Mag- randomized clinical trial. Clinical Oral Implants out an intervention: a systematic review of ran-
nesium-enriched hydroxyapatite compared to cal- Research 23: 11981204. domized controlled trials. Clinical Implant
cium sulfate in the healing of human extraction Hammerle, C.H.F., Araujo, M.G., Simion, M. & Dentistry and Related Research 16: 120.
sockets: radiographic and histomorphometric Araujo, M.G. (2012) Evidence-based knowledge Neiva, R.F., Tsao, Y.-P., Eber, R., Shotwell, J., Billy,
evaluation at 3 months. Journal of Periodontol- on the biology and treatment of extraction sock- E. & Wang, H.-L. (2008) Effects of a putty-form
ogy 80: 210218. ets. Clinical Oral Implants Research 23: 8082. hydroxyapatite matrix combined with the syn-
Crespi, R., Cappare, P. & Gherlone, E. (2011a) Com- Higgins, J.P.T., Green, S. (2011) Cochrane handbook thetic cell-binding peptide p-15 on alveolar ridge
parison of magnesium-enriched hydroxyapatite for systematic reviews of interventions version preservation. Journal of Periodontology 79: 291
and porcine bone in human extraction socket 5.1.0 The Cochrane Collaboration. 299.
healing: a histologic and histomorphometric eval- Higgins, J.P., Thompson, S.G., Deeks, J.J. & Altman, Patel, K., Mardas, N. & Donos, N. (2013) Radio-
uation. The International Journal of Oral & Max- D.G. (2003) Measuring inconsistency in meta- graphic and clinical outcomes of implants placed
illofacial Implants 26: 10571062. analyses. British Medical Journal 327: 557560. in ridge preserved sites: a 12-month post-loading
Crespi, R., Cappare, P., Romanos, G.E., Mariani, E., Hoang, T.N. & Mealey, B.L. (2012) Histologic com- follow-up. Clinical Oral Implants Research 24:
Benasciutti, E. & Gherlone, E. (2011b) Cortico- parison of healing after ridge preservation using 599605.
cancellous porcine bone in the healing of human human demineralized bone matrix putty with Pellegrini, G., Rasperini, G., Obot, G., Farronato, D.
extraction sockets: combining histomorphometry one versus two different-sized bone particles. & Dellavia, C. (2014) Soft tissue healing in alveo-
with osteoblast gene expression profiles in vivo. Journal of Periodontology 83: 174181. lar socket preservation technique: histologic eva-
The International Journal of Oral & Maxillofa- Horvath, A., Mardas, N., Mezzomo, L.A., Needle- luations. International Journal of Periodontics
cial Implants 26: 866872. man, I.G. & Donos, N. (2013) Alveolar ridge pres- and Restorative Dentistry 34: 531539.
De Risi, V., Clementini, M., Vittorini, G., Mann- ervation. A systematic review. Clinical Oral Perelman-Karmon, M., Kozlovsky, A., Liloy, R. &
occi, A. & De Sanctis, M. (2013) Alveolar ridge Investigations 17: 341363. Artzi, Z. (2012) Socket site preservation using
preservation techniques: a systematic review and Iasella, J.M., Greenwell, H., Miller, R.L., Hill, M., bovine bone mineral with and without a biore-
meta-analysis of histological and histomorpho- Drisko, C., Bohra, A.A. & Scheetz, J.P. (2003) sorbable collagen membrane. The International
metrical data. Clinical Oral Implants Research Ridge preservation with freeze-dried bone allo- Journal of Periodontics and Restorative dentistry
26: 5068. [Correction added after first online graft and a collagen membrane compared to 32: 459465.
publication: volume and page numbers corrected] extraction alone for implant site development: a Pietrokovski, J. & Massler, M. (1967) Ridge remod-
Donos, N., Mardas, N. & Chadha, V. (2008) Clinical clinical and histologic study in humans. Journal eling after tooth extraction in rats. Journal Dental
outcomes of implants following lateral bone aug- of Periodontology 74: 990999. Reseach 46: 222231.
mentation: systematic assessment of available Jung, R.E., Fenner, N., Hammerle, C.H.F. & Zitz- Poulias, E., Greenwell, H., Hill, M., Morton, D., Vi-
options (barrier membranes, bone grafts, split os- mann, N.U. (2013) Long-term outcome of dal, R., Shumway, B. & Peterson, T.L. (2013)
teotomy). Journal Clinical Periodontology 35(8 implants placed with guided bone regeneration Ridge preservation comparing socket allograft
Suppl): 173202. (gbr) using resorbable and non-resorbable mem- alone to socket allograft plus facial overlay xeno-
Eskow, A.J. & Mealey, B.L. (2014) Evaluation of branes after 12-14 years. Clinical Oral Implants graft: a clinical and histologic study in humans.
healing following tooth extraction with ridge Research 24: 10651073. Journal of Periodontology 84: 15671575.
preservation using cortical versus cancellous Jung, R.E., Siegenthaler, D.W. & Hammerle, C.H.F. Serino, G., Biancu, S., Iezzi, G. & Piattelli, A.
freeze-dried bone allograft. Journal of Periodontol- (2004) Postextraction tissue management: a soft (2003) Ridge preservation following tooth extrac-
ogy 85: 514524. tissue punch technique. The International Jour- tion using a polylactide and polyglycolide sponge
Farmer, M. & Darby, I. (2014) Ridge dimensional nal of Periodontics and Restorative Dentistry 24: as space filler: a clinical and histological study in
changes following single-tooth extraction in the 545553. humans. Clinical Oral Implants Research 14:
aesthetic zone. Clinical Oral Implants Research Leblebicioglu, B., Salas, M., Ort, Y., Johnson, A., 651658.
25: 272277. Yildiz, V.O., Kim, D.G., Agarwal, S. & Tatakis, Serino, G., Rao, W., Iezzi, G. & Piattelli, A. (2008)
Festa, V.M., Addabbo, F., Laino, L., Femiano, F. & D.N. (2013) Determinants of alveolar ridge preser- Polylactide and polyglycolide sponge used in
Rullo, R. (2013) Porcine-derived xenograft com- vation differ by anatomic location. Journal of human extraction sockets: bone formation fol-
bined with a soft cortical membrane versus Clinical Periodontology 40: 387395. lowing 3 months after its application. Clinical
extraction alone for implant site development: a Lekovic, V., Kenney, E.B., Weinlaender, M., Han, Oral Implants Research 19: 2631.
clinical study in humans. Clinical Implant Den- T., Klokkevold, P., Nedic, M. & Orsini, M. (1997) Sisti, A., Canullo, L., Mottola, M.P., Covani, U.,
tistry and Related Research 15: 707713. A bone regenerative approach to alveolar ridge Barone, A. & Botticelli, D. (2012) Clinical eval-
uation of a ridge augmentation procedure for Mesh terms protocol supplementary concept word,
the severely resorbed alveolar socket: multicen- 7. Tooth socket rare disease supplementary concept word,
ter randomized controlled trial, preliminary 8. Alveolar bone loss unique identifier]
results. Clinical Oral Implants Research 23:
9. Bone resorption 5. Crest*.mp. [mp=title, abstract, original
526535.
10. Tooth extraction title, name of substance word, subject
Spinato, S., Galindo-Moreno, P., Zaffe, D., Bernar-
dello, F. & Soardi, C.M. (2014) Is socket healing 11. Bone remodelling heading word, keyword heading word,
conditioned by buccal plate thickness? A clinical 12. 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 or protocol supplementary concept word,
and histologic study 4 months after mineralized 10 or 11 rare disease supplementary concept word,
human bone allografting. Clinical Oral Implants 13. Preserv$ unique identifier]
Research 25: e120e126. 14. Reconstruct$ 6. ((tooth or teeth) adj3 remov*).mp. [mp=ti-
Tan, W.L., Wong, T.L.T., Wong, M.C.M. & Lang,
15. Augment$ tle, abstract, original title, name of sub-
N.P. (2012) A systematic review of post-extract-
ional alveolar hard and soft tissue dimensional
16. Fill$ stance word, subject heading word,
changes in humans. Clinical Oral Implants 17. Seal$ keyword heading word, protocol supple-
Research 23: 121. 18. Graft$ mentary concept word, rare disease sup-
Van der Weijden, F., DellAcqua, F. & Slot, D.E. 19. Repair$ plementary concept word, unique
(2009) Alveolar bone dimensional changes of identifier]
post-extraction sockets in humans: a systematic
Mesh terms 7. Exp Tooth Socket/
review. Journal of Clinical Periodontology 36:
20. Alveolar ridge augmentation 8. Exp Alveolar Bone Loss/
10481058.
Vance, G.S., Greenwell, H., Miller, R.L., Hill, M.,
21. Bone regeneration 9. Exp Bone Resorption/
Johnston, H. & Scheetz, J.P. (2004) Comparison of 22. Bone substitutes 10. Exp Tooth Extraction/
an allograft in an experimental putty carrier and a 23. Transplantation or autologous transplan- 11. Exp Bone Remodelling/
bovine-derived xenograft used in ridge preserva- tation or heterologous transplantation or 12. 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 or
tion: a clinical and histologic study in humans. homologous transplantation 10 or 11
The International Journal of Oral & Maxillofa-
24. 21 or 17 or 12 or 20 or 15 or 14 or 22 or 13. Preserv*.mp. [mp=title, abstract, original
cial implants 19: 491497.
18 or 13 or 16 or 19 title, name of substance word, subject
Vignoletti, F., Matesanz, P., Rodrigo, D., Figuero,
E., Martin, C. & Sanz, M. (2012) Surgical proto- 25. Randomized controlled trial.pt heading word, keyword heading word,
cols for ridge preservation after tooth extraction. 26. Controlled clinical trial.pt. protocol supplementary concept word,
A systematic review. Clinical Oral Implants 27. Randomized.ab. rare disease supplementary concept word,
Research 23(Suppl 5): 2238. 28. Placebo.ab. unique identifier]
Vignoletti, F., Nunez, J. & Sanz, M. (2014) Soft tis- 29. Drug therapy.fs. 14. Reconstruct*.mp. [mp=title, abstract, ori-
sue wound healing at teeth, dental implants and
30. Randomly.ab. ginal title, name of substance word, sub-
the edentulous ridge when using barrier mem-
branes, growth and differentiation factors and soft
31. Trial.ab. ject heading word, keyword heading
tissue substitutes. Journal of Clinical Periodon- 33. Groups.ab. word, protocol supplementary concept
tology 41: S23S35. 33. 27 or 25 or 28 or 26 or 24 or 30 or 29 or word, rare disease supplementary concept
Wallace, S.C., Snyder, M.B. & Prasad, H. (2013) 31 word, unique identifier]
Postextraction ridge preservation and augmenta- 34. Humans.sh. 15. Augment*.mp. [mp=title, abstract, origi-
tion with mineralized allograft with or without
nal title, name of substance word, subject
recombinant human platelet-derived growth fac-
Ovid MEDLINE heading word, keyword heading word,
tor bb (rhpdgf-bb): a consecutive case series. The
1. ((tooth or teeth) adj3 extract*).mp. protocol supplementary concept word,
International journal of Periodontics & Restor-
ative dentistry 33: 599609. [mp=title, abstract, original title, name of rare disease supplementary concept word,
Wang, R.E. & Lang, N.P. (2012) Ridge preservation substance word, subject heading word, unique identifier]
after tooth extraction. Clinical Oral Implants keyword heading word, protocol supple- 16. Fill*.mp. [mp=title, abstract, original
Research 23: 147156. mentary concept word, rare disease sup- title, name of substance word, subject
Wood, R.A. & Mealey, B.L. (2012) Histologic com- plementary concept word, unique heading word, keyword heading word,
parison of healing after tooth extraction with
identifier] protocol supplementary concept word,
ridge preservation using mineralized versus
demineralized freeze-dried bone allograft. Journal
2. Socket*.mp. [mp=title, abstract, original rare disease supplementary concept word,
of Periodontology 83: 329336. title, name of substance word, subject unique identifier]
heading word, keyword heading word, 17. Seal*.mp. [mp=title, abstract, original
protocol supplementary concept word, title, name of substance word, subject
Appendix rare disease supplementary concept word, heading word, keyword heading word,
unique identifier] protocol supplementary concept word,
Electronic search strategy 3. Ridge*.mp. [mp=title, abstract, original rare disease supplementary concept word,
title, name of substance word, subject unique identifier]
Ovid MEDLINE heading word, keyword heading word, 18. Graft*.mp. [mp=title, abstract, original
1. Tooth extraction$ protocol supplementary concept word, title, name of substance word, subject
2. Socket rare disease supplementary concept word, heading word, keyword heading word,
3. Ridge unique identifier] protocol supplementary concept word,
4. Alveol$ 4. Alveo*.mp. [mp=title, abstract, original rare disease supplementary concept word,
5. Crest title, name of substance word, subject unique identifier]
6. Tooth removal heading word, keyword heading word,
19. Repair*.mp. [mp=title, abstract, original 4. Alveo*.mp. [mp=title, abstract, subject 24. (random* adj3 (clinic* or trial*)).ab.
title, name of substance word, subject headings, heading word, drug trade name, 25. (control adj2 group).ab.
heading word, keyword heading word, original title, device manufacturer, drug 26. Placeb*.ab.
protocol supplementary concept word, manufacturer, device trade name, key- 27. (prospective and (case or clinical)).ab.
rare disease supplementary concept word, word] 28. 24 or 25 or 26 or 27
unique identifier] 5. Crest*.mp. [mp=title, abstract, subject 29. Exp controlled clinical trial/
20. Exp Alveolar Ridge Augmentation/ headings, heading word, drug trade name, 30. Exp prospective study/
21. Exp Bone Regeneration/ original title, device manufacturer, drug 31. 24 or 25 or 26 or 27 or 28 or 29 or 30
22. Exp Bone Substitutes/ manufacturer, device trade name, key- 32. Limit 23 to (clinical trial or randomized
23. Exp Transplantation, Autologous/or exp word] controlled trial or phase 2 clinical trial or
Transplantation/or exp Transplantation, 6. Exp tooth socket/ phase 3 clinical trial)
Heterologous/or exp Transplantation 7. Exp alveolar bone loss/ 33. 23 and 31
Conditioning/ 8. Osteolysis/ 34. 32 or 33
24. 13 or 14 or 15 or 16 or 17 or 18 or 19 or 9. Exp tooth extraction/ 35. Limit 34 to (amphibia or ape or bird or
20 or 21 or 22 or 23 10. Bone remodeling/ cat or cattle or chicken or dog or ducks
25. 12 and 24 11. 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 or and geese or fish or frogs and toads or
26. Randomized controlled trial.pt. 10 goat or guinea pig or hamsters and ger-
27. Controlled clinical trial.pt. 12. Preserv*.mp. bils or horse or monkey or mouse or
28. Trial.ab. 13. Reconstruct*.mp. [mp=title, abstract, pigeons and doves or rabbits and
29. Placebo.ab. subject headings, heading word, drug hares or rat or reptile or sheep or swine)
30. Groups.ab. trade name, original title, device manu- 36. 34 not 35
31. Randomly.ab. facturer, drug manufacturer, device trade
32. Drug therapy.fs. name, keyword]
Cochrane Library, LILACS and Web of Science
33. 26 or 27 or 28 or 29 or 30 or 31 or 32 14. Augment*.mp.
#1 (tooth or teeth) and (extract* or remov*):
34. 25 and 33 15. Fill*.mp. [mp=title, abstract, subject
ti,ab,kw or socket or ridge and alveolar
35. (prospective adj3 (case or cohort or clini- headings, heading word, drug trade name,
and crest (Word variations have been
cal)).mp. [mp=title, abstract, original title, original title, device manufacturer, drug
searched)
name of substance word, subject heading manufacturer, device trade name, key-
#2 mesh descriptor: [Alveolar Bone Loss]
word, keyword heading word, protocol word]
explode all trees
supplementary concept word, rare disease 16. Graft*.mp. [mp=title, abstract, subject
#3 mesh descriptor: [Tooth Socket] explode
supplementary concept word, unique headings, heading word, drug trade name,
all trees
identifier] original title, device manufacturer, drug
#4 mesh descriptor: [Bone Resorption]
36. 33 or 35 manufacturer, device trade name, key-
explode all trees
37. 25 and 36 word]
#5 mesh descriptor: [Tooth Extraction]
38. Limit 37 to animals 17. Repair*.mp. [mp=title, abstract, subject
explode all trees
39. 37 not 38 headings, heading word, drug trade name,
#6 mesh descriptor: [Bone Remodeling]
original title, device manufacturer, drug
explode all trees
manufacturer, device trade name, key-
Ovid EMBASE #7 preserve or recontruct or augment or fill
word]
1. ((tooth or teeth) adj3 (extract* or or seal or graft or repair
18. Exp bone regeneration/
remov*)).mp. [mp=title, abstract, subject #8 mesh descriptor: [Alveolar Ridge Aug-
19. (bone and (regen* or substit*)).mp.
headings, heading word, drug trade name, mentation] explode all trees
[mp=title, abstract, subject headings,
original title, device manufacturer, drug #9 mesh descriptor: [Bone Regeneration]
heading word, drug trade name, original
manufacturer, device trade name, key- explode all trees
title, device manufacturer, drug manufac-
word] #10 mesh descriptor: [Bone Substitutes]
turer, device trade name, keyword]
2. Socket*.mp. [mp=title, abstract, subject explode all trees
20. Exp bone prosthesis/
headings, heading word, drug trade name, #11 mesh descriptor: [Transplantation]
21. Exp transplantation/
original title, device manufacturer, drug explode all trees
22. 12 or 13 or 14 or 15 or 16 or 17 or 18 or
manufacturer, device trade name, key- #12 #1 or #2 or #3 or #4 or #5 or #6
19 or 20 or 21
word] #13 #7 or #8 or #9 or #10 or #11
23. 11 and 22
3. Ridge*.mp. #14 #12 and #13

Mardas Et Al-2015-Clinical Oral Implants Research

Uploaded by

Copyright:

Available Formats

Mardas Et Al-2015-Clinical Oral Implants Research

Uploaded by

Document Information

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

Mardas Et Al-2015-Clinical Oral Implants Research

Uploaded by

Copyright:

Available Formats

Nikos Mardas Does ridge preservation following

The success of implant-supported restorations complicating implant placement in the ideal

heterogeneity was performed using Cochrans

Forest Plots were created to illustrate the

Table 1. (continued) mation on implant type (Iasella et al. 2003;

Table 2. Study characteristics of included papers

Barone et al. Italy, Industry RCT, Unclear Unclear 38 (62) 51  14 Y

Coomes et al. USA, 1 U Industry RCT, Parallel Unclear 34 (34) 1979 Y

Crespi et al. Italy, 1 H Unclear CCT, Specialists 15 (45) 51.3 (2872) N

Crespi et al. Italy, 1 U Unclear CCT, Specialists 15 (45) 54.6 (3468) N

Leblebicioglu USA, 1 U Institutional Prospective Specialist 24 (25) 2483 N

Mardinger et al. Israel, ?, Unclear Prospective Unclear 36 (43) 50.75 (2475) Y

Neiva et al. USA, 1 U Industry RCT, Parallel Specialists 24 (24) 2576 N

Perelman-Karmon Unclear Unclear RCT, Parallel Unclear 23 (23) 2668 N

Poulias et al. USA, 1 U Self-funded RCT, Parallel Specialist 23 (23) 52  16 (2677) Y

Vance et al. Unclear Industry RCT, Parallel Unclear 24 (24) 56 ?

Wallace et al. Unclear Industry CCT, Parallel Unclear 30 (34) 1870 N

193 | Clin. Oral Impl. Res. 26 (Suppl. 11), 2015 / 180201

(a) Meta-analysis superior to that of unassisted socket healing.

Meta-analysis (d) Meta-analysis

(b) Meta-analysis (e) Meta-analysis

0.0 0.2 0.4 0.6 0.8 1.0

You might also like

Barone et al. Italy, Industry RCT, Unclear Unclear 38 (62) 51 14 Y

Poulias et al. USA, 1 U Self-funded RCT, Parallel Specialist 23 (23) 52 16 (2677) Y