Root Fracture of Abutment Teeth for

Partial Removable Dental Prostheses

Yoko Mizuno, DDS, PhD1/Tomoya Gonda, DDS, PhD2/Toshihito Takahashi, DDS, PhD3/
Akiko Tomita, DDS, PhD4/Yoshinobu Maeda, DDS, PhD5

Purpose: Root fracture is a common and serious cause of abutment tooth loss. No studies to
date have comprehensively assessed the individual contributions of the factors that increase
stress on abutment teeth. The purpose of this study was to analyze the stress distribution
in abutment teeth based on a three-dimensional finite element model and to analyze the
factors that affect stress distribution. Materials and Methods: Models were generated
from the computed tomography data of a single patient, consisting of a mandibular second
premolar abutment tooth, bone, residual mucous membrane, and a partial removable dental
prosthesis (PRDP). Four models were prepared using different types of endodontic posts
and cores. Akers clasps were used for the simulated PRDPs, and a vertical load was applied
to the occlusal surface of the PRDPs. Debonding between the post and root was simulated.
The Young modulus of the residual ridge was reduced to simulate a poor fit between the
denture base and the residual ridge. Stress distribution in the abutment tooth root was
observed, and the maximum principal stress was evaluated. Results: The nonmetal post
model and the mesial rest model reduced stress concentration in the root. The stress increased
in models simulating debonding and poor fit. The results of the multiple linear regression
analysis confirmed that debonding and poor fit were significantly associated with root stress.
Conclusion: Within the limitations of this study, it is suggested that the bonding integrity
of posts and adequate fit of the denture base are important factors affecting the longevity
of abutment teeth for PRDPs. Int J Prosthodont 2016;29:461–466. doi: 10.11607/ijp.4327

R oot fracture is a common and serious cause of

abutment tooth loss. Axelsson et al1 reported that
root fracture was the main cause of tooth loss in pa-
tients who maintained a plaque control program. The
incidence of root fracture is significantly higher in
1Resident, Department of Prosthodontics, Gerodontology and Oral
Rehabilitation, Osaka University Graduate School of Dentistry,
Osaka, Japan.
2Associate Professor, Department of Prosthodontics, Gerodontology and
Oral Rehabilitation, Osaka University Graduate School of Dentistry,
Osaka, Japan.
3Assistant Professor, Department of Prosthodontics, Gerodontology and
Oral Rehabilitation, Osaka University Graduate School of Dentistry,
Osaka, Japan.
4Specially Appointed Researcher, Department of Prosthodontics,
Gerodontology and Oral Rehabilitation, Osaka University Graduate
School of Dentistry, Osaka, Japan.
5Professor, Department of Prosthodontics, Gerodontology and Oral
Rehabilitation, Osaka University Graduate School of Dentistry,
Osaka, Japan.
Correspondence to: Dr Tomoya Gonda, Department of Prosthodontics,
Gerodontology and Oral Rehabilitation, Osaka University Graduate School
of Dentistry, 1-8 Yamadaoka, Suita, Osaka 565-0871, Japan.
Fax: +81-6-6879-2957. Email: tgonda@dent.osaka-u.ac.jp
©2016 by Quintessence Publishing Co Inc.
abutment teeth for partial removable dental prosthe-
ses (PRDPs) than in nonabutment teeth,2 presumably
because of the increased loading received by abut-
ment teeth. It is believed that root fracture occurs more
frequently in endodontically treated teeth because of
the associated loss of tooth structure. Furthermore,
root fracture occurs in areas of high stress concentra-
tion. The factors that impose stress on teeth, such as
the type of post, core,3–7 and denture clasp8–10 used,
have been identified, but no studies to date have com-
prehensively evaluated the individual contributions of
these factors.
The purpose of this study was to analyze the stress
distribution in abutment teeth based on a three-
dimensional finite element model (3D-FEM) and to
analyze the factors influencing stress distribution in
abutment tooth roots.
Materials and Methods
A single patient at the Osaka University Dental
Hospital, who was missing the distal extension on
both the left and right sides of the mandible, was se-
lected to participate in this study. The right side was
selected for generation of the model. The study proto-
col was approved by the Ethical Committee of Osaka

Volume 29, Number 5, 2016 461

© 2016 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Root Fracture of Abutment Teeth

M, R FR, MR
AD AM

a b c
Fig 1   Three-dimensional finite element models (3D-FEMs). (a) Overview of 3D-FEM. (b) Four types of endodontic post: M = cast metal post and
core; R = composite resin post and core; FR = composite resin core with a fiber post; MR = composite resin core with a metal post. (c) Types of
clasp: AD = Akers clasp with a distal rest; AM = Akers clasp with a mesial rest.

Table 1   Material Properties were set according to previous studies,3,11,12 except for
Young modulus Poisson bone, which was set using Keyak formula13 (Table 1).
Component Material (GPa) ratio The models were constrained on the inferior aspect of
Dentin 14.7 0.35 the body of the mandible, and adjacent elements were
Periodontal ligament 9.8 × 10−3 0.45 connected rigidly except for the clasp and abutment
tooth.
Gutta-percha 0.2 0.45
A vertical load of 98 N was applied to the occlu-
Fiber post Glass fiber 40.0 0.26
sal surface at the first molar position on the PRDP.
Metal post Ni-Cr alloy 19.6 0.30 The models were then analyzed using the 3D finite
Cast metal core Au-Ag alloy 78.5 0.33 element analysis software to compare the maximum
Resin core Composite resin 5.9 0.30 values of the principal stress within the root.
Cast metal crown Au-Ag alloy 78.5 0.33 With long-term denture use, cement dissolution
or secondary caries can occur between the post and
Denture base Acrylic resin 4.0 0.30
root, and this was simulated in the present study by
Clasp Co-Cr alloy 2.0 × 102 0.30
debonding between the post and root. The degree
Residual ridge 1.0 ×10−2 0.40 of debonding was classified as adequate bonding,
Bone set by Keyak formula.13 debonding only in the mesial cervical area, debond-
ing only in the cervical area, or complete debonding.
Additionally, the residual ridge remodels and atrophies
University Graduate School of Dentistry (H24-E13), over time, which may result in poor fit of the denture
and informed consent was obtained from the par- base. The Young modulus of the residual ridge was
ticipant after the aims and methodology of the study gradually reduced, as follows, to simulate a poor fit
were explained. between the denture base and residual ridge: 10 MPa,
The participant planned to have implant treatment 5 MPa, 1 MPa, 0.1 MPa, and 0.01 MPa.
and had undergone a computed tomography (CT) ex- The models were created with Akers clasps with
amination. 3D-FEMs were generated using the CT data either a distal or a mesial rest. The distal rest mod-
of the patient and 3D analysis software (Mechanical els had 623,033 quadrilateral elements with 119,555
Finder version 6.1 Extended Edition, Research Center nodes, while the mesial rest models had 637,776
of Computational Mechanics). Models of the man- quadrilateral elements with 121,739 nodes. Stress dis-
dibular right second premolar, bone, residual mucous tribution in the root was observed, and the maximum
membrane, and PRDP were prepared (Fig 1a). principal stresses were compared.
Models of four types of endodontic posts and cores Multiple linear regression analysis was conducted
(cast metal post and core, composite resin post and to reveal the relationship between the maximum value
core, composite resin core with a fiber post, and com- of the maximum principal stress in the root and the
posite resin core with a metal post) for the second variables that may affect stress distribution within the
premolar were generated (Fig 1b). The diameter of the root, including type of endodontic post, type of clasp,
post was one-third the diameter of the root, and the debonding between the post and root, and poor fit
length of the post was two-thirds the length of the between the denture base and the residual ridge. A
root. The two simulated PRDPs had Akers clasps with statistical analysis program (Stata 11, Stata) was used
either a distal or a mesial rest retained on the abut- to manage and analyze the data with a threshold
ment tooth (Fig 1c). Material properties for each unit significance of P < .05 for multiple linear regression

462 The International Journal of Prosthodontics

© 2016 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Mizuno et al

M R FR MR

Maximum
principal
stress
(MPa)

a Distal side Mesial side

AD AM AB MC C CD

Maximum
principal
stress One-half
(MPa) length of
post

Maximum
principal
stress
(MPa)

b Distal side Mesial side c Distal side Mesial side

Fig 2  Maximum principal stress distribution in the Poor fit of denture base Well fit of denture base
abutment tooth root. (a) Stress distribution in the root
for each type of post and core (Akers clasp with a dis-
tal rest): M = cast metal post and core; R = composite
resin post and core; FR = composite resin core with Maximum
a fiber post; MR = composite resin core with a metal principal
stress
post. (b) Stress distribution in the root for each type of (MPa)
clasp (cast metal post and core model): AD = Akers
clasp with a distal rest; AM = Akers clasp with a me-
sial rest. (c) Changes in stress distribution in the root
with debonding between the post and root (cast met-
al post and core and Akers clasp with a distal rest):
AB = adequate bonding; MC = debonding only in the
mesial cervical area; C = debonding only in the cervi-
cal area; CD = complete debonding. Yellow line = re- 0.01 0.1 1 5 10
gion of debonding. (d) Changes in stress distribution Distal side Mesial side
in the root with a poor fit between the denture base Young modulus of the residual ridge (MPa)
and the residual ridge (cast metal post and core and d
Akers clasp with a distal rest).

analysis. In the mesial cervical debond- Results

ing model and the cervical debonding
model, the results of the stress analy-
sis with a well-fitting denture base (10
MPa) were included in the multiple
linear regression analysis. In the mod-
els simulating poor fit (5 MPa, 1 MPa,
and 0.1 MPa), the results of the stress
analysis in the adequate bonding mod-
el were included in the multiple linear
regression analysis.
Stress concentration was observed in the mesial cervical area and
the mesial coronal third of the root surface in the abutment tooth
with a cast metal post and core. In the models with other types of
post, stress concentration was observed in the mesial cervical area
(Fig 2a). There was similar stress distribution in models with both
mesial and distal rest Akers clasps (Fig 2b). The maximum principal
stresses in the mesial coronal third of the root were compared. In
the distal rest model, the maximum principal stress was lower in
the resin core with a fiber post and the resin post and core than in

Volume 29, Number 5, 2016 463

© 2016 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Root Fracture of Abutment Teeth

Maximum principal stress 3.0 3.0 3.0

Maximum principal stress

Maximum principal stress

2.0 2.0 2.0
(MPa)

(MPa)

(MPa)
1.0 1.0 1.0

0 0 0
M R FR MR M R FR MR Distal rest Mesial rest
Akers clasp with distal rest Akers clasp with mesial rest Akers clasp with distal rest
a b c

15 90 90
Maximum principal stress

Maximum principal stress

Maximum principal stress

10 60 60
(MPa)

(MPa)

(MPa)
5 30 30

0 0 0
AB MC C CD 0.01 0.1 1 5 10 0.01 0.1 1 5 10
Young modulus of residual ridge (MPa) Young modulus of residual ridge (MPa)
d e f
Fig 3   Graphs showing maximum principal stress in the abutment tooth root. (a, b) Maximum principal stress in the root in the mesial coronal
third of the root surface. M = cast metal post and core; R = composite resin post and core; FR = composite resin core with a fiber post; MR =
composite resin core with a metal post. (c) Maximum principal stress in the root in the mesial coronal third of the root surface (cast metal post and
core model). (d) Highest maximum principal stress for each debonding condition (cast metal post and core and Akers clasp with a distal rest).
AB = adequate bonding; MC = debonding only in the mesial cervical area; C = debonding only in the cervical area; CD = complete debonding.
(e) Highest maximum principal stress for Akers clasp with a distal rest (cast metal post and core model). (f) Highest maximum principal stress for
Akers clasp with a mesial rest (cast metal post and core model).

the cast metal post and core and the resin core with a the denture base and the residual ridge (Figs 2d, 3e,
metal post (Fig 3a). Results were similar in the mesial and 3f).
rest model, but the difference in stress was reduced The results of the multiple linear regression analy-
(Fig 3b). sis revealed that debonding and poor fit were signifi-
The stress in the root of the mesial rest model was cantly associated with maximum principal stress in
more evenly distributed than in the distal rest model the root (Table 2). The standardized partial regression
(Fig 2b). The maximum principal stresses in the mesial coefficient of debonding was 0.55, and that of poor
coronal third of the root were compared. Regardless denture base fit was 0.78.
of the type of post and core used, the stress in the root
of the mesial rest model was lower than in the distal Discussion
rest model (Fig 3c).
In the models with a cast metal post and core Fractures occur because of crack initiation and
and a distal rest clasp, the stress distribution varied propagation from areas of high stress concentration.
greatly according to the degree of debonding. In the Hayashi et al reported that fractures closer to the root
adequate bonding model, a high-stress area was ob- apex are more likely to result in extraction.14 The re-
served around the mesial cervical area and the me- sults of the present study show that because of the
sial coronal third of the root surface. Additionally, a high stress concentration in the mesial coronal third
high-stress area was observed in the mesial internal of the root surface of the model with a metal post and
surface of the root in the mesial cervical debonding core, an abutment tooth with a metal post and core is
model and the cervical debonding model. In the com- more likely to fracture near the apex of the root and
plete debonding model, a high-stress area was found result in extraction than one restored with a resin post
around the apex of the post (Fig 2c). The maximum and core.
stress observed in the cervical debonding and com- Previous studies reported a reduction in the stress
plete debonding models was greater than the stress intensity within the root of an abutment tooth with a
observed in the adequate bonding and mesial cervical mesial rest versus a distal rest.8,9,15 These studies did
debonding models (Fig 3d). not address stress in nonvital abutment teeth restored
In all models, the maximum stress in the root in- with a post and core; however, the findings of the pres-
creased as the Young modulus of the residual ridge ent analysis of nonvital abutment teeth are consistent
decreased—that is, where there was a poor fit between with the previous studies. This could be because the

464 The International Journal of Prosthodontics

© 2016 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
 Mizuno et al

Table 2   Multiple Linear Regression Analysis (n = 72)

Partial regression Standardized partial
coefficient regression coefficient Collinearity
B β P VIF
Endodontic post
R 47.2 0.09 .21 1.50
FR 41.5 0.08 .27 1.50
MR 30.0 0.06 .43 1.50
Clasp 10.4 0.02 .70 1.00
Debonding between post and root
(2) 145.3 0.20* .01 1.56
(3) 151.9 0.21* .01 1.56
(4) 294.7 0.55* .00 1.56
Changes in fit between denture base and residual ridge
5 MPa 143.6 0.20* .01 1.56
1 MPa 154.5 0.22* .01 1.56
0.1 MPa 207.5 0.29* .00 1.56
0.01 MPa 421.8 0.78* .00 1.56
Objective variable: highest maximum principal stress (MPa).
Explanatory variables:
Type of endodontic post (standard = cast metal post and core): R = composite resin post and core; FR = composite resin core with a fiber post;
MR = composite resin core with a metal post.
Type of clasp: Akers clasp with a distal rest = 0, Akers clasp with a mesial rest = 1
Debonding between the post and root (standard = adequate bonding): (2): debonding only in the mesial cervical area, (3): debonding only in the cervical
area, (4): complete debonding.
Fit between denture base and residual ridge (standard = Young modulus of the residual ridge): 10 MPa = satisfactory fit.
Adjusted R 2 = 0.75.
*P < .05.

distance from the point of load to the center of rota-
tion of the abutment tooth is shorter in the distal rest
model than in the mesial rest model.
Clinically, it is difficult to achieve complete bonding
between the post and the root. In this study, an area
of high stress was found in the mesial cervical area in
the case of bonding between the post and root. Crack
initiation may occur in this area, which could cause
microleakage and cement dissolution. Complete
debonding of a rigid metal post and core could pro-
duce an area of high stress concentration in the root
around the apical portion of the post and core.
There were some limitations to this study. The mod-
els were generated with CT data from only a single
patient. In future research, models should be gener-
ated and analyzed from patients with various types
of residual ridges. The residual ridge has viscoelas-
tic properties; however, the FEM software used in
this study cannot analyze viscoelastic characteristics
and therefore the results were assessed using lin-
ear analysis. Post length, diameter, and taper influ-
ence stress distribution within the root.16–18 Different
types of posts should be compared in future stud-
ies. Conventional Akers clasps were selected for this
study, and the effects of different rest positions were
compared. Additional research should be undertaken
using different types of retainers.
The results of the multiple linear regression analy-
sis showed that debonding between the post and the
root and a poor fit between the denture base and the
residual ridge were significantly associated with the
maximum principal stress in the root. These findings
suggest that examination of the bonding of the post
and prevention of secondary caries are important fac-
tors in the longevity of PRDPs. It is also important to
examine the fit of the denture and reline the denture
base if necessary. No significant differences in stress
in the root depending on the type of post or clasp
were found in the multiple linear regression analysis;
however, the results of the stress analysis suggest
that root fracture can be prevented by carefully con-
sidering the denture design and the type of endodon-
tic post used to restore abutment teeth.
Conclusions
Within the limitations of this study, it was concluded
that nonmetal posts reduce stress concentration in
the abutment teeth of PRDPs. The stress in the root of
the mesial rest model was lower than in the distal rest
model. Debonding between the post and the root and
poor fit between the denture base and the residual
ridge also increased stress in the abutment tooth root.
The results of the multiple linear regression analysis
confirm that debonding and poor fit were significantly
associated with stress in the root. Other variations
should be analyzed in future studies.

Volume 29, Number 5, 2016 465

© 2016 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.
Root Fracture of Abutment Teeth

Acknowledgments 9. Enomoto Y. Experimental studies on the influence of the selection

The authors thank the staff of the Department of Prosthodontics
and Oral Rehabilitation, Osaka University Graduate School of
Dentistry. The authors declare no potential conflicts of interest
related to this study.
References
1. Axelsson P, Nyström B, Lindhe J. The long-term effect of a
plaque control program on tooth mortality, caries and peri-
odontal disease in adults. Results after 30 years of mainte-
nance. J Clin Periodontol 2004;31:749–757.
2. Matsuda K, Ikebe K, Enoki K, Tada S, Fujiwara K, Maeda Y.
Incidence and association of root fractures after prosthetic
treatment. J Prosthodont Res 2011;55:137–140.
3. Ukon S, Moroi H, Okimoto K, et al. Influence of different elastic
moduli of dowel and core on stress distribution in root. Dent
Mater J 2000;19:50–64.
4. Pegoretti A, Fambri L, Zappini G, Bianchetti M. Finite element
analysis of a glass fibre reinforced composite endodontic post.
Biomaterials 2002;23:2667–2682.
5. Pierrisnard L, Bohin F, Renault P, Barquins M. Corono-radicular
reconstruction of pulpless teeth: A mechanical study using fi-
nite element analysis. J Prosthet Dent 2002;88:442–448.
6. Hayashi M, Sugeta A, Takahashi Y, Imazato S, Ebisu S. Static
and fatigue fracture resistances of pulpless teeth restored with
post-cores. Dent Mater 2008;24:1178–1186.
7. Eskitaşcioğlu G, Belli S, Kalkan M. Evaluation of two post core
systems using two different methods (fracture strength test and
a finite elemental stress analysis). J Endod 2002;28:629–633.
8. Nogawa A. Study on the dynamic behavior of mandibular dis-
tal-extention removable partial denture utilizing finite element
method. 2. Influence of the location of rest and displaceability
of residual ridge mucosa [in Japanese]. Nihon Hotetsu Shika
Gakkai Zasshi 1989;33:1313–1327.
of abutment teeth and clasp design on force distribution on abut-
ment teeth in free-end saddle dentures. Shikwa Gakuho 1972;
72:46–80.
10. Pellizzer EP, Ferraço R, Tonella BP, Oliveira BJ, Souza FL, Falcón-
Antenucci RM. Influence of ridge type on mandibular distal
extension removable partial denture. Acta Odontol Latinoam
2010;23:68–73.
11. Kitamura S, Tsumita M, Tsubota Y, Kokubo Y, Fukushima S.
Mechanical consideration for the post and core systems using
two-dimensional finite element. Shigaku 2007;33:21–29.
12. Huang H. Dynamic research on reinforcing structure in lower
jaw full denture. Journal of Osaka University Dental Society
1998;43:60–74.
13. Keyak JH, Rossi SA, Jones KA, Skinner HB. Prediction of femo-
ral fracture load using automated finite element modeling.
J Biomech 1998;31:125–133.
14. Hayashi M, Kinomoto Y, Takeshige F, Ebisu S. Prognosis of in-
tentional replantation of vertically fractured roots reconstruct-
ed with dentin-bonded resin. J Endod 2004;30:145–148.
15. Thompson WD, Kratochvil FJ, Caputo AA. Evaluation of photo-
elastic stress patterns produced by various designs of bilateral
distal-extension removable partial dentures. 1997. J Prosthet
Dent 2004;91:105–113.
16. Sorensen JA, Engelman MJ. Effect of post adaptation on frac-
ture resistance of endodontically treated teeth. J Prosthet Dent
1990;64:419–424.
17. Kainose K, Nakajima M, Foxton R, Wakabayashi N, Tagami J.
Stress distribution in root filled teeth restored with various post
and core techniques: effect of post length and crown height. Int
Endod J 2015;48:1023–1032.
18. Bacchi A, Dos Santos MB, Pimentel MJ, Caetano CR, Sinhoreti
MA, Consani RL. Influence of post-thickness and material on
the fracture strength of teeth with reduced coronal structure.
J Conserv Dent 2013;16:139–143.

Literature Abstract

A Detailed Analysis of Mandibular Angle Fractures: Epidemiology, Patterns, Treatments, and Outcomes

This useful retrospective study aims to determine the relationship between treatment modality patient outcomes. Mandibular angle
fractures are common and have high complication rates, and the ideal treatment modality is unclear. A total of 103 patients (mean age:
30.4 years) treated by the same operator were included. Three treatment variables (fixation type, use of IMF, and teeth in the fracture line
extracted or not) and three patient outcomes (postoperative infection, fracture healing, and patient comfort) were examined. The vast
majority of patients were men who had suffered a physical injury to the left side of the jaw in the month of June. The Champy plate and
two-plate technique resulted in 100% bony union. Use of IMF or presence of teeth in the fracture line had no statistically significant effect
on infection, healing, or comfort. The strut or ladder technique closely followed by the Champy plate was most comfortable. The authors
admit that the low numbers, retrospective design, and effect of confounding factors such as smoking habits are weaknesses in this study.
However, this is useful guidance to supplement the surgeon’s experience.

Patel N, Kim B, Zaid W. J Oral Maxillofac Surg 2016 May 13 [Epub ahead of print], doi: 10.1016/j.joms.2016.05.002. References: 35. Reprints: Dr N. Patel,
Department of Oral and Maxillofacial Surgery, Louisiana State University, 1400 Annunciation Street, Apartment 1306, New Orleans, LA 70130, USA.
Email npate9@Isuhsc.edu —Steven Soo, Singapore

466 The International Journal of Prosthodontics

© 2016 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.