Critical Size Defects for Bone Regeneration Experiments in the Dog Mandible: A Systematic Review

Abstract

Objectives: To perform a systematic literature review of studies involving critical size defects (CSDs) in the dog mandible and calvarium to find out the common characteristics of CSDs.
Materials and Methods: Internet search of the literature was performed on June 2016 based on specific keywords. The search process included 5 databases. The Animal Research Reporting In Vivo Experiment (ARRIVE) guidelines was used to assess the quality of the included studies.
Results: Nine studies have met the inclusion criteria and subjected to quality evaluation. All the defects (N = 156) were located in the mandible. Only 2 articles showed randomized controlled studies, whereas the remaining 7 were nonrandomized controlled studies. The geometry of the defects was either rectangular, box, cylindrical (circular), arch, or saddle shaped.
Conclusion: There is a lack of homogeneity in reporting data on CSDs in the dog mandible. Future animal studies should include a negative control group for an objective comparison and evaluation of any new biomedical materials. More awareness is needed for the Animal Research Reporting In Vivo Experiment (ARRIVE) guidelines to improve data reporting, which can facilitate comparison and reproducibility of future studies.

Full text

Critical Size Defects for Bone
Regeneration Experiments in the Dog
Mandible: A Systematic Review
Hesham F. Marei, PhD,* Khalid Mahmood, PhD,†and Khalid Almas, FDS (RCSEd)‡
The concept of critical size defect
(CSD) was developed more than
25 years ago as an attempt to
standardize research on bone regener-
ative materials. Originally, it referred
to the smallest defects that would not
heal by natural process during the life-
time of the animal.
1
Because healing
can involve fibrous tissue formation,
the CSD concept was clarified few
years later to involve any defect that
has less than 10% of bone regeneration
during the first year of defect healing.
Such period was considered to indicate
that bone union is not expected to
occur during the animal lifetime, and
such defect was considered as a true
CSD.
2
Because the animal lifetime in
most of clinical research is bounded by
the completion of the study, the CSD
concept was revisited recently to refer
to the smallest size of a defect that
does not heal spontaneously when left
untreated for a certain period.
3
Linking the defect size to the dura-
tion of the study opened the gate to
animal studies that have reported healing
of different defect sizes over different
durationswithin the same animal model,
and all the nonhealing sites were still
considered as CSDs. Delgado-Ruiz
et al
4
reviewed systematically CSDs on
rabbit calvariae and found a lack of
homogeneity between different studies.
Moreover, the authors advocated that
all animal research protocols should fol-
low the quality checklist of Animal
Research Reporting In Vivo Experiment
(ARRIVE) guidelines to increase unifor-
mity and to enable comparison and the
reproducibility of the studies.
Describing the CSD by referring
only to the size of the surgical site has
been appraised by Cooper et al,
3
as such
description undermined other factors
such as the age of the animal, blood
supply during the healing process, and
the topography of the defect, which
have a direct impact on bone healing.
Multiple studies confirmed that the
healing requirements of the vertical de-
fects are different than those of the hor-
izontal defects,
5
and 3D defects with
loss of continuity are higher than mono-
cortical defects created by trephine
burs.
6,7
Moreover, preserving the
*Associate Professor, Department of Biomedical Dental
Sciences, College of Dentistry, Imam Abdulrahman Bin Faisal
University, Kingdom of Saudi Arabia; Associate Professor, Oral
and Maxillofacial Surgery Department, Faculty of Dentistry, Suez
Canal University, Egypt.
†Professor, Department of Information Management, University
of the Punjab, Lahore, Pakistan.
‡Professor, Department of Preventive Dental Sciences, College
of Dentistry, Imam Abdulrahman Bin Faisal University, Kingdom
of Saudi Arabia.
Reprint requests and correspondence to: Hesham F.
Marei, BDS, MSc, FDS (RCS-Eng), PhD, College of
Dentistry, Imam Abdulrahman Bin Faisal University,
PO Box 1982, Dammam 31441, Kingdom of Saudi
Arabia, Phone: +966548027750, Fax: +966-3-8572624,
E-mail: hmarei@iau.edu.sa
ISSN 1056-6163/17/02701-001
Implant Dentistry
Volume 27 Number 1
Copyright © 2017 Wolters Kluwer Health, Inc. All rights
reserved.
DOI: 10.1097/ID.0000000000000713
Objectives: To perform a system-
atic literature review of studies
involving critical size defects (CSDs)
in the dog mandible and calvarium
to find out the common character-
istics of CSDs.
Materials and Methods: Inter-
net search of the literature was
performed on June 2016 based on
specific keywords. The search pro-
cess included 5 databases. The Ani-
mal Research Reporting In Vivo
Experiment (ARRIVE) guidelines
was used to assess the quality of
the included studies.
Results: Nine studies have met
the inclusion criteria and subjected
to quality evaluation. All the defects
(N ¼156) were located in the man-
dible. Only 2 articles showed ran-
domized controlled studies, whereas
the remaining 7 were nonrandom-
ized controlled studies. The geome-
try of the defects was either
rectangular, box, cylindrical (circu-
lar), arch, or saddle shaped.
Conclusion: There is a lack of
homogeneity in reporting data on
CSDs in the dog mandible. Future
animal studies should include a neg-
ative control group for an objective
comparison and evaluation of any
new biomedical materials. More
awareness is needed for the Animal
Research Reporting In Vivo Experi-
ment (ARRIVE) guidelines to
improve data reporting, which can
facilitate comparison and reproduc-
ibility of future studies. (Implant
Dent 2017;27:1–7)
Key Words: mandibular defect,
bone healing, animal model
MAREI ET AL IMPLANT DENTISTRY /VOLUME 27, NUMBER 1 2017 1
Copyright Ó2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.

periosteum influences the healing pro-
cess in different defect sizes,
8
which in-
dicates that healing of CSDs of the same
sizes could be manipulated also by the
manner in which the defect is created.
Misinterpretation of results that
may happen due to the above-
mentioned variability in reporting
CSDs, and the recent advances in
standardizing and evaluating bone de-
fects highlighted a need for evaluation of
the reporting quality of CSD experimen-
tal studies. Although Delgado-Ruiz
et al
4
have explored the most common
practice among high-quality studies in
rabbit calvariae, there is still a need to
find out the consensus among studies
that have reported CSDs in both
weight-bearing bone and weight non-
bearing bone in a bigger animal model
to cover various clinical conditions.
Rabbits differ in their bone micro-
architecture and biomechanics, skeletal
anatomy, and healing of critical-sized
defects when compared with large
animal models. Large animal models
have well-developed haversian and
Fig. 1. Flowchart for the search process. Three hundred fifty-four articles were identified, of which, there were 143 duplicate articles resulting in
211 articles ready for screening. One hundred sixty-four articles involved defects in other anatomical sites and other animal models. Full text of
47 articles was subjected to in-depth reading and evaluation. Nine studies fulfilled the inclusion criteria.
2DEFECTS FOR BONE REGENERATION EXPERIMENTS MAREI ET AL
Copyright Ó2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.

trabecular bone remodeling; large skel-
etal surface areas; and volumes that
allow testing of prosthetic devices and
similar skeletal disuse atrophy results to
human.
9
Therefore, the aim of this
study was to perform a critical, system-
atic literature review of studies involv-
ing CSDs in the dog mandible and
calvarium and to assess the quality of
the selected studies using the ARRIVE
guidelines to provide a consensus that
can improve the homogeneity and com-
parison of future studies.
MATERIALS AND METHODS
The present study is a systematic
review that has followed the Preferred
Reporting Items for Systematic Re-
views and Meta-Analyses (PRISMA)
guidelines in addressing one focused
question: What are the common char-
acteristics of CSDs among studies that
were conducted on a dog experimental
model?
Internet search that is followed by
a manual search of the literature by all
the authors was performed on June
2016. The search included 5 databases:
PubMed, Google Scholar, Web of Sci-
ence, Ebsco Dentistry, and Scopus.
Search terms involved specific key-
words, which are (dogs OR canine)
AND (“critical size defect”OR “critical
size defects”OR “critical defects”OR
“bone defects”OR “non-healing
defect”).
One author (K.M.) independently
screened the titles and abstracts of all
the identified studies to exclude dupli-
cations. Two authors (H.F.M. and
K.A.) assessed the titles, abstracts, and
full articles identified by the Internet
and manual searches. Disagreements
were resolved by discussion between
the authors, if not, the third author
(K.M.) determined the inclusion or
exclusion of a relevant article.
The inclusion criteria were as fol-
lows: (1) original experimental studies;
(2) use of dog model; (3) defects created
in the mandible and/or calvariae; (4)
inclusion of a negative control group;
(5) There was a confirmation either by
histomorphometric or radiographic
data that the negative control defects
did not heal along the entire period of
the study; and (6) articles in English
language only.
The exclusion criteria were as fol-
lows: (1) animal models other than dog;
(2) anatomical locations other than
mandible and calvariae; (3) literature
reviews; (4) circumferential defects
and/or supra-alveolar defects around
dental implants.
All the included articles were sub-
jected to extraction of sex, weight, sex
of the animals; number, location, geom-
etry, size of the defects; type of osteot-
omy, time to sacrifice, and the
techniques used to evaluate bone heal-
ing. Authors HM and KM identified the
study design and evaluated the quality
of the included studies using the
ARRIVE guidelines. The authors
followed the procedure described by
Kilkenny and Altman
10
and Kilkenny
et al.
11
To describe the quality of each
item in the checklist, the quality score/
maximum score was calculated and
compared with 3 possible quality co-
efficients: Excellent quality if the score
is .80%; average quality if the score is
between 50% and 80%; and poor qual-
ity if the score is ,50%.
Statistical Analysis
Descriptive statistics (mean and
SD) were calculated. Kappa statistics
Table 1. List of the 38 Excluded Studies and the Reasons for Exclusion
# Study Reason for Exclusion
1 Caplanis N. et al (1997) Supra-alveolar defects around dental implants
2 Cochran D. et al (1997) Circumferential defects around dental implants
3 Giannobile W. et al (1998) Supra-alveolar defects and no size are given
4 Wikesjo E. et al (1998) Supra-alveolar defects
5 Tatakis D. et al (2000) Supra-alveolar defects
6 Nagao H. et al (2002) No negative control
7 Wikesjo U et al (2002) Supra-alveolar defects
8 Artzi Z. et al (2003) No negative control
9 Wikesjo E. et al (2003)a Supra-alveolar defects
10 Wikesjo E. et al (2003)b Supra-alveolar defects
11 Wikesjo E. et al (2003)c Supra-alveolar defects
12 Wikesjo E. et al (2003)d Supra-alveolar defects around dental implants
13 Wikesjo E. et al (2003)e Supra-alveolar defects and no negative control
14 Polimeni G. et al (2005) Supra-alveolar defects
15 Tal H. et al (2005) Complete bone healing in negative control
16 Cui L. et al (2007) No negative control
17 Yuan J. et al (2007) No negative control
18 Lai H. et al (2009) Circumferential defects around implants
19 Weng D. et al (2009) Circumferential defects around implants
20 Choi S. et al (2010) No negative control
21 He D. et al (2010) No negative control group
22 Kwon H. et al (2010) Supra alveolar defects and no negative control
23 Lee J. et al (2010) Supra alveolar defects and no negative control
24 Valderrama P. et al (2010) Circumferential defects around implants
25 Yan X. et al (2010) Supra-alveolar defects
26 Yuan J. et al (2010) No negative control
27 Fernandes J. et al (2011) Supra alveolar defects and no negative control
28 Kawai T. et al (2011) No negative control
29 Kim S. et al (2011) Circumferential defects around dental implants
30 Wang L. et al (2011) Circumferential defects around dental implants
31 Koli K. et al (2012) Supra-alveolar defects
32 Machtei E. et al (2013) Complete bone healing in negative control
33 Messora M. et al (2012) Titanium mesh was placed on the control site
34 Liu X. et al (2014) No negative control
35 Messora M. et al (2014) Titanium mesh was placed on the control site
36 Zandi M. et al (2014) No negative control
37 Jung U. et al (2015) Circumferential defects around dental implants
38 Tsumanuma Y. et al (2016) Supra-alveolar defects
The main reasons for exclusion were lack negative control group, complete healing of the negative control, supra-alveolar defects, and
circumferential defects around the dental implant.
MAREI ET AL IMPLANT DENTISTRY /VOLUME 27, NUMBER 1 2017 3
Copyright Ó2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.

Table 2. List of Included Studies and the Extracted Data
Author/year Type of Study Animal Sex No. of Animals Age, mo Weight, kg No. of Defects per Animal/Geometry
Hunt D. et al (2001) Pilot study; nonrandomized controlled Male 3 18–36 25 4/Saddle-type defects
Imbronito A. et al (2002) Pilot study; nonrandomized controlled N/P 4 Adult 12 4/Rectangular
Huh J. et al (2005) Nonrandomized controlled Female 16 N/P 15–20 2/Segmental defects and continuity defects
Elsalanty M. et al (2009) Nonrandomized controlled Male 13 N/P 37 1/Segmental defects and continuity defects
Neamat A. et al (2009) Nonrandomized controlled Male 8 18–24 12–15 2/Geometry is not provided
Baba S. et al (2011) Nonrandomized controlled Male 7 20–22 N/P 2/2-wall rectangular defects
Birang R. et al (2012) Randomized controlled Male 3 N/P 25 8/bilateral/circular (cylinder)
Miura K. et al (2012) Nonrandomized controlled Male 9 17 N/P One/an arc-shaped defect
Park C. et al (2012) Randomized controlled study Male 10 15 9–13 2/rectangular 1-wall defectsdbox shape
Author/year
Total Number of
Negative Control
Defects/Study
Defect Size
Expressed, mm
Method Used to
Perform the Osteotomy
Time to
Sacrifice
in Weeks
Percentage of Healing
(Defect Closure as
a Percentage
of the Defect Area)
Measured Variables (Defect
Closure/Bone Regeneration
Obtained
From Histomorphometric or
Radiographic Data)
Hunt D. et al
(2001)
210315 Reciprocating mini-saw 12 65% Histomorphometry and radiology
Imbronito A. et al
(2002)
18312 Low-speed rotatory and
chisels
8 & 16 N/P Histomorphometry
Huh J. et al (2005) 32 5–20, 30–60 N/P 24 N/P Histomorphometry and radiology
Elsalanty M. et al
(2009)
83368.4 Reciprocating minisaw 10 N/P Histomorphometry and radiology
Neamat A. et al
(2009)
8 5 Carbide surgical bur 12, 24 N/P Histology
Baba S. et al
(2011)
733335 Dental round bur 4, 8 N/P Histomorphometry
Birang R. et al
(2012)
6636 Trephine 6 N/P Histomorphometry
Miura K. et al
(2012)
3 10 N/P 24 N/P Histomorphometry and radiology
Park C. et al (2012) 5 4 35 N/P 24 N/P Histomorphometry
Manuscripts that fulfilled the inclusion criteria: type of study, number of animals per study, total number of defects/study, total number of negative control defects/study, type of osteotomy describes the instrument used for creating the CSD, time of sacrifice was
expressed in weeks, and percentage of healing, expressed as the healing of the control/whole defect.
N/P, data not provided in results.
4DEFECTS FOR BONE REGENERATION EXPERIMENTS MAREI ET AL
Copyright Ó2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.

for the qualitative evaluation of the
studies was calculated to reflect on the
interreviewer agreement. Statistical
analysis was performed using SPSS
22 statistical software (IBM Corpora-
tion, Armonk, NY).
RESULTS
The Internet search in all the
selected databases resulted in 342 titles,
whereas 12 titles were found by the
manual search. The first round of
screening of all the titles resulted in
exclusion of 143 titles because of
duplications. The second round
involved screening of the titles and
abstracts of the 211 studies according
to the inclusion and exclusion criteria. If
any of the identified abstracts did not
provide the full information, the full
articles were retrieved to check eligi-
bility for inclusion (Fig. 1). The second
round of assessment resulted in exclu-
sion of 164 studies because they were
dealing mainly with defects in other
anatomical sites and other animal mod-
els. The full text of the remaining 47
articles was retrieved for in-depth read-
ing and evaluation. The third round of
evaluation resulted in exclusion of 38
articles because they lack negative con-
trol group, complete healing of the neg-
ative control, supra-alveolar defects,
and circumferential defects around den-
tal implants (Table 1). Extraction of the
required information and quality evalu-
ation based on the ARRIVE guidelines
were applied on the remaining 9
studies, which fit our inclusion criteria
(Table 2).
The included 9 studies, with exper-
imental sample of 73 dogs (53 males
and 16 females and 4 not provided);
showed defects in the mandible only.
The variability in the shape of the
defects among studies made statistical
comparison impossible; therefore,
descriptive statistics was performed to
show the most common characteristics
of mandibular CSDs. Only 2 articles
showed randomized controlled studies,
whereas the remaining 7 were non-
randomized controlled studies. The
earliest study was published in 2001,
whereas the latest one was in 2012. Five
studies have mentioned the age of the
animals, which ranged from 15 to 36
months. Regarding the weight of the
animals, it was reported in 7 articles and
varied from 9 to 25 Kg.
The total number of defects was
156. They were continuity defects (n ¼
45) that were fixed by 2 miniplates or
reconstruction plate, 1 wall defect (n ¼
44), and 2 wall defects (n ¼14),
whereas the type of 53 defects was not
clearly provided. The geometry of the
defects was either rectangular, box,
cylindrical (circular), arch, or saddle
shaped. For the rectangular defects,
the sizes were 3 35 mm and 8 3
12 mm; for the box type defects, the
sizes were 4 35 mm; for the saddle
defects, they were 10 315 mm; for
the arch shaped, they were 10 mm in
radius, whereas for the cylindrical de-
fects, they were 6 36 mm. All the
cylindrical defects were created using
trephine burs, whereas the segmen-
tal defects were formed using
reciprocating mini-saw. For the other
defects, dental burs and chisels were
used.
Thetimeofsacrifice of the animals
ranged from 4 to 24 weeks. The most
commonly used healing period was 24
weeksasitwasfollowedin4studies.
Regarding healing, all studies used
histomorphometric analysis for evalu-
ation of bone tissue. Only 1 study has
provided the percentage of defect clo-
sure in relation to the overall defect
volume, but all the studies have re-
flected on bone volume formation in
the studied histological sections, the
significant difference between the
study and the control sites, and if
complete healing was attained in the
control defects.
The included studies involved 72
negative control defects. All the defects
involved either less than 10% of bone
within the healed tissue or did not show
complete obliteration within the life-
time of the animal except for mandib-
ular segmental defects that are less than
40 mm in length when the periosteum is
preserved, and those that are less than
15 mm in length when the periosteum is
excised.
Quality evaluation of the included
articles showed a mean score of 27.1 6
4.8 out of 36. Items that evaluated the
housing- and welfare-related assess-
ment interventions (ie, type of cage,
bedding material, number of cage
Table 3. Quality Scores of the Included Studies Based on the ARRIVE Guidelines
Items According to the ARRIVE Checklist
References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Total/study
Hunt D. et al (2001) 1 2 1 11112000 221 102112 22
Imbronito A. et al (2002) 1 2 1 10121000 221 101220 20
Huh J. et al (2005) 1 2 1 12221010 211 202222 27
Elsalanty M. et al (2009) 1 2 2 12211011 221 212222 30
Neamat A. et al (2009) 1 2 2 12222111 221 222112 32
Baba S. et al (2011) 1 1 1 12122000 210 101222 22
Birang R. et al (2012) 1 2 2 12221101 220 212220 28
Miura K. et al (2012) 1 1 2 12221010 221 212222 29
Park C. et al (2012) 1 2 2 12222111 221 222222 34
Total/category 9 16 14 9 15 15 16 13 3 5 4 18 16 7 15 7 16 16 16 14 d
Maximum score/category 9 18 18 9 18 18 18 18 18 18 9 18 18 9 18 18 18 18 18 18 d
Items from 1 to 20 are title (1), abstract (2), introductiondbackground (3), introductiondobjectives (4), methodsdethical statement (5), study design (6), experimental procedure (7), experimental animals
(8), housing and husbandry (9), sample size (10), allocation of animals (11), experimental outcomes (12), statistics (13), resultsdbaseline data (14), number analyzed (15), outcome and estimation (16),
adverse events (17), discussiondinterpretation/scientific implications (18), general applicability/relevance (19), and funding sources and role of the funders (20). Last number on the right side represents
the total score of a maximum of 36 points, obtained by each study.
MAREI ET AL IMPLANT DENTISTRY /VOLUME 27, NUMBER 1 2017 5
Copyright Ó2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.

companions, light/dark cycle, tempera-
ture, and access to food and water) and
sample size-total number, details of cal-
culation methods were received in poor
quality (score ,50%). Items that eval-
uated experimental animal species,
strain, sex, developmental stage,
weight, source of animals, allocation
of animals to experimental group ran-
domization, baseline data characteris-
tics and health status of animals,
outcomes and estimation results for
each analysis with a measure of preci-
sion have received average quality
(score 50%–80%), whereas the remain-
ing items of the checklist have received
over 80% indicating excellent quality
(Table 3).
DISCUSSION
The aim of this study was to
explore the common characteristics of
CSDs among studies that were con-
ducted on dog experimental model to
provide a consensus that can improve
the homogeneity and comparison of
future dog studies.
Our search resulted in a total of 211
studies after removal of duplicated
titles. Out of them, only 9 articles were
included. Such number of studies is
considered as low number if compared
with the study by Delgado-Ruiz et al
4
that investigated CSDs in rabbit calvar-
ium and showed 350 articles with inclu-
sion of 25 studies. Such difference
could be due to the nature of the exper-
imental model under investigation.
Rabbit as an experimental model
has more advantages than dog. It is
easier to handle and control, cheaper,
has shorter bone sigma, and less diffi-
culties with creating large, homogenous
samples for statistical testing.
9
It is easier
to control the confounding factors in
studies investigating weight non–
bearing bone (calvarium) than in
weight-bearing bone studies. To evalu-
ate CSDs in weight-bearing bone such as
mandible, proper fixation and adequate
stability of segmental defects are crucial
for bone healing to keep the size of the
defect as the only independent variable.
Furthermore, preserving or excising
the periosteum could be another con-
founding factor. In one study, segmental
defects of 30 and 40 mm in length
showed gap obliteration, whereas gap
persisted in defects greater than 50 mm
over 24 weeks when the periosteum was
preserved. In the same study, defects in
which the periosteum was excised
showed no healing for segmental defects
longer than 15 mm in length.
8
Our study has included only studies
with a negative control group in their
design. Although the purpose of some
of these studies was not to test the CSD,
the negative control group in these
studies has still reflected on the healing
of a defect that has specific shape and
size. Our inclusion criteria were con-
sistent with those in the study con-
ducted by Delgado-Ruiz et al.
4
The included studies had 53 male
and 16 female dogs with age range from
15 to 36 months and their weights
varied from 9 to 25 Kg. The reviewed
literature showed that animal age,
weight, and sex influenced
bone mineral concentration (BMC)
and bone mineral density (BMD),
therefore affecting healing potential of
any bony defect. In general, middle-
aged dogs (3–10 years) revealed the
highest BMC and BMD levels. Mean
BMC and BMD were higher in males
compared with females.
12
Moreover, in
aging dogs, the skeletal exchange of
calcium falls to a very low level with
an increase in osteoclastic bone resorp-
tion and loss of skeletal mass.
13
On the
other side, bone defects in skeletally
immature dogs heal at a faster rate than
skeletally mature individuals, which
could result in misleadingly high poten-
tials of a tested material if skeletally
immature dogs have been used. There-
fore, Hollinger and Kleinschmidt
2
rec-
ommended that biological age that is
confirmed by radiographic investiga-
tion of epiphyseal closure be used in
conjunction with chronological age
and body weight to establish skeletal
maturity, before CSD creation.
The current study investigated the
results of 72 negative control defects of
156 mandibular defects that have dif-
ferent sizes, types, and geometries.
Although there was a great heterogene-
ity among the included studies that
made statistical comparison impossi-
ble, all the defects still fell in 2 groups,
which are segmental and nonsegmental
defects. In segmental defects, there was
a lack of agreement between 2 studies in
regard to defect size. One study consid-
ered 33 68.4 mm as the size of CSD,
7
whereas the other considered 50 mm
and more.
8
Such difference could be
related to the time of animal sacrifice.
In the first study, the animals were
killed at 10 weeks, whereas the animals
were left to 24 weeks in the second
study. Hollinger and Kleinschmidt
2
mentioned that CSD could be as great
as 45 mm. The positive correlation
between percentage of healing and time
of sacrifice was confirmed in a system-
atic review on CSD in a rabbit model.
4
In nonsegmental defects, there were
5 different defect shapes, which are
rectangular, box, arc, saddle, and cylin-
drical. Comparison between different
sizes for statistically calculating the mean
CSD is difficult because all the experi-
mented animals were killed at different
time point. These results were different
from those of the study conducted by
Delgado-Ruiz et al
4
, which revealed that
the most common CSD used in experi-
mental procedures in calvarium rabbit is
15 mm in diameter. Reciprocating saw,
round dental burs, and trephine burs cre-
ated all the bony defects. It was found that
round/fissure dental burs were the most
commonly used method for square and
rectangular defects, whereas circular tre-
phines were the most commonly used
method to create cylindrical defects.
Quality evaluation of the included
studies revealed a mean score of 27.1 6
4.8 and ranged from 20 to 34. Although
the ARRIVE guidelines have been pub-
lished in 2010, some of the earlier stud-
ies in our review have still shown high
quality score. The most common data
that were missed in all studies were rel-
evant to housing- and welfare-related
assessment interventions and sample
size calculation. We advocate that more
awareness in future about the ARRIVE
guidelines could improve the quality of
reporting animal studies.
Themainlimitationinourstudywas
the low number of the included studies
for final analysis. Inclusion of only 9
studies could be due to the strict inclu-
sion and exclusion criteria that were
followed in the study. Moreover, the
study involved high level of agreement
between authors, which ensured minimal
subjectivity in the selection and quality
6DEFECTS FOR BONE REGENERATION EXPERIMENTS MAREI ET AL
Copyright Ó2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.

evaluation process. Within the limitation
of our study, we support the previous
recommendations that were suggested
by Coopper et al.
3
The authors advocated
the discontinuing use of the term CSD
because of the various confounding fac-
tors (age, weight, sex, blood supply,
defect shape, periosteum, and time of
killing the animals) that affect the healing
process resulting in a model with a lim-
ited clinical applicability.
CONCLUSION
There is a lack of homogeneity in
reporting data on CSDs in the dog
mandible. Future animal studies should
include a negative control group for an
objective comparison and evaluation of
any new biomedical materials. More
awareness is needed for the Animal
Research Reporting In Vivo Experi-
ment (ARRIVE) guidelines to improve
data reporting, which can facilitate
comparison and reproducibility of
future studies.
DISCLOSURE
This research did not receive any
grants from any funding agencies. The
authors claim to have no financial
interest, either directly or indirectly, in
the products or information listed in the
article.
APPROVAL
The Institutional Review Board ex-
empted the study from ethical approval.
REFERENCES
1. Schmitz JP, Hollinger JO. The critical
size defect as an experimental model for
craniomandibulofacial nonunions. Clin
Orthop Relat Res. 1986;4:299–308.
2. Hollinger JO, Kleinschmidt JC. The
critical size defect as an experimental
model to test bone repair materials.
J Craniofac Surg. 1990;1:60–68.
3. Cooper GM, Mooney MP, Gosain
AK, et al. Testing the critical size in
calvarial bone Defects: Revisiting the
concept of a critical-size defect. Plast Re-
const Surg. 2010;125:1685–1692.
4. Delgado-Ruiz RA, Calvo-Guirado
JL, Romanos GE. Critical size defects for
bone regeneration experiments in rabbit
calvariae: Systematic review and quality
evaluation using ARRIVE guidelines. Clin
Oral Implants Res. 2015;26:915–930.
5. Esposito M, Grusovin MG, Felice P,
et al. The efficacy of horizontal and vertical
bone augmentation procedures for dental
implants - a Cochrane systematic review.
Eur J Oral Implantol. 2009;2:167–184.
6. Schlegel KA, Lang FJ, Donath K,
et al. The monocortical critical size bone
defect as an alternative experimental
model in testing bone substitute
materials. Oral Surg Oral Med Oral Pathol
Oral Radiol Endod. 2006;102:7–13.
7. Elsalanty ME, Zakhary I, Akeel S,
et al. Reconstruction of canine
mandibular bone defects using a bone
transport reconstruction plate. Ann Plast
Surg. 2009;63:441–448.
8. Huh JY, Choi BH, Kim BY, et al.
Critical size defect in the canine
mandible. Oral Surg Oral Med Oral Pathol
Oral Radiol Endod. 2005;100:296–301.
9. Mooney M, Siegel M. Animal models
for bone tissue engineering of critical-sized
defects (CSDs), bone pathologies, and
orthopedic disease states. In: Hollinger J,
Einhorn T, Doll B, et al, eds. Bone Tissue
Engineering. 1st ed. CRC Press; USA
2005:217–244.
10. Kilkenny C, Altman DG. Improving
bioscience research reporting: ARRIVE-ing
at a solution. Lab Anim. 2010;44:377–378.
11. Kilkenny C, Browne WJ, Cuthill IC,
et al. Improving bioscience research
reporting: The ARRIVE guidelines for
reporting animal research. PLoS Biol.
2010;8:e1000412.
12. Lorinson K, Loebcke S, Skalicky
M, et al. Signalment differences in
bone mineral content and bone mineral
density in canine appendicular bones. A
cadaveric study. Vet Comp Orthop
Traumatol. 2008;21:147–151.
13. Williams EA, Kelly PJ. Age-related
changes in bone in the dog: Calcium
homeostasis. J Orthop Res. 1984;2:8–14.
MAREI ET AL IMPLANT DENTISTRY /VOLUME 27, NUMBER 1 2017 7
Copyright Ó2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.