Original article

Retention and stress distribution in distal extension removable partial

dentures with and without implant association

Renata Cristina Silveira Rodrigues DDS, MSc, PhD*, Adriana Claudia Lapria Faria DDS, MSc, PhD1,Ana Paula Macedo BEng, MSc2, Maria da Gloria Chiarello de Mattos DDS, MSc, PhD3,

Ricardo Faria Ribeiro DDS, MSc, PhD4

Department of Dental Materials and Prosthodontics, Dental School of Ribeirao Preto, University of Sao Paulo, Av. do Cafe, s/n, Monte Alegre, 14040-904 Ribeirao

Preto, SP, Brazil

Received 24 November 2011; received in revised form 4 May 2012; accepted 9 July 2012

Available online 27 November 2012

Abstract

Purpose: The present study aimed to evaluate the retention and stress distribution of conventional (C) RPD and compare to RPD associated to

implant for support (IS) and retention (IR).

Methods: Frameworks were cast from cp Ti (n = 18) and Co–Cr alloy (n = 18) by plasma and injected by vacuum–pressure. Conventional RPDs were

compared to implant associated RPDs using a distal implant to support (IS) or to support and retain (IR) RPD. The specimens were subjected to

insertion/removal cycles simulating 5 years of use and the retention force (N) was measured or evaluated. A mixed linear model was used to analyze the

data (a = 0.05). Photoelastic models were qualitatively examined for stress when an occlusal load of 15 kgf was applied over support teeth and RPD.

Results: Retention force of IR RPDs is greater than IS and C RPDs for both cp Ti and Co–Cr alloy specimens. Retention force of cp Ti RPDs

increased initially and was maintained throughout 5 years of simulation test while Co–Cr RPDs presented a decrease at the beginning of the test and

had their retention force maintained throughout the test. Implant placement at residual alveolar ridge decreased stress around teeth, mainly in the

first premolar. Stress concentration in the IS RPD is slightly greater than in the IR RPD.

Conclusion: The results suggest that implant placement at the distal extension improves retention and stress distribution of RPDs.

# 2012 Japan Prosthodontic Society. Published by Elsevier Ireland. All rights reserved.

Keywords: Dental implantation; Dental prosthesis retention; Denture precision attachments; Removable partial dentures

www.elsevier.com/locate/jpor

Available online at www.sciencedirect.com

Journal of Prosthodontic Research 57 (2013) 24–29

1. Introduction

Although implant-supported fixed partial dentures are the

ideal treatment option for partially edentulous patients, bone

loss in mandibular posterior regions could require bone graft

and mandibular nerve lateralization procedures for implant

placements, representing high surgical risk and cost, discoura-

* Corresponding author. Tel.: +55 16 3602 4005; fax: +55 16 3602 4780.

E-mail addresses: renata@forp.usp.br (R.C.S. Rodrigues),

adriclalf@forp.usp.br (A.C.L. Faria), anapaula@forp.usp.br (A.P. Macedo),

gloria@forp.usp.br (M.d.G.C. de Mattos), rribeiro@forp.usp.br (R.F. Ribeiro).1 Tel.: +55 16 3602 4130; fax: +55 16 3602 4780.2 Tel.: +55 16 3602 4104; fax: +55 16 3602 4780.3 Tel.: +55 16 3602 4098; fax: +55 16 3602 4780.4 Tel.: +55 16 3602 4046; fax: +55 16 3602 4780.

1883-1958/$ – see front matter # 2012 Japan Prosthodontic Society. Published b

http://dx.doi.org/10.1016/j.jpor.2012.07.001

ging patients [1,2]. Thus, removable partial dentures (RPDs)

still represent an alternative of rehabilitation for these patients.

Distal extension RPDs are complex because of the teeth and

mucous support, requiring better load distribution for both

tissues to avoid vertical, horizontal and torsional forces that

may have adverse effects [3,4]. Thus, the use of distal implants

to support and retain RPDs has been reported in the literature to

minimize dislodgement, improve esthetics and mastication,

resulting in patient satisfaction in a cost-effective manner [3–5].

Cobalt–chromium (Co–Cr) alloy is the material more used

around the world to construct RPD frameworks because of low

cost, corrosion resistance, high microhardness and modulus of

elasticity and low density [6]. Titanium alloys and commer-

cially pure titanium (cp Ti) represents alternatives to Co–Cr

alloy RPD frameworks. Some titanium advantages such as

resistance, low specific weight, high corrosion resistance, good

physical and mechanical properties, low cost and excellent

y Elsevier Ireland. All rights reserved.

R.C.S. Rodrigues et al. / Journal of Prosthodontic Research 57 (2013) 24–29 25

biocompatibility increased the interest in the application of this

metal for prosthesis frameworks after 1980 decade [6–13].

Thus, titanium and titanium alloy RPD frameworks have been

searched showing good retention throughout time [6,12,13].

Several studies have researched stress distribution of distal

extension RPDs associated with implants in different positions

of alveolar residual ridge varying implant dimensions and

attachment systems [1,2,14]. Nevertheless, the retention of

conventional and implant associated RPDs, which is important

to minimize dislodgement, resulting in improved mastication

and patient satisfaction, was not evaluated yet. The null

hypothesis was that the association of an implant in residual

alveolar ridge with RPDs improves retention and stress

distribution. Thus, the aim of the present study was to evaluate

the retention and stress distribution of conventional RPD and

compare to RPD associated to implant for support and

retention.

2. Materials and methods

2.1. Specimen preparation

A metallic matrix representing a distal extension mandibular

right hemi-arch segment, presenting first and second premolar,

was fabricated from Co–Cr alloy. Undercuts of 0.25-mm (0.01-

inch) were provided in the positions established for T-bar

clasps. In addition, a mesial occlusal rest and proximal and

lingual guide-planes were prepared on the second premolar to

standardize the path of insertion. This matrix was used to

perform the use simulation test.

The matrix was relieved for the correct waxing of the

frameworks and duplicated in silicone (Silicone Master,

Talmax, Curitiba, Brazil). Casts with Rematitan Plus (Den-

taurum, Pforzheim, Germany) were used to make the cp Ti

(Tritan1 grade I, Dentaurum) specimens, and casts with Crom-

O-Cast (Polidental Industry and Commerce Ltd., Sao Paulo,

Brazil) were used to make the Co–Cr alloy (Vera PDI, Aalba

Dent. Inc., Cordelia, CA, USA) specimens.

Casts were positioned at the dental surveyor according to

their guide planes which standardized the path of insertion.

Thus, prefabricated T-bar clasp patterns (Rewax, Renfert

GmbH, Hilzingen, Germany) were positioned on the casts

following a tray obtained from a plate vacuum formed in the

first waxing, permitting to get similar framework waxings. In

addition, plastic pins (5 mm wide and 30 mm long) were

attached to a circular retention mesh (Renfert GmbH)

positioned in the area corresponding to the distal extension

partial denture base. This pins originally served as a sprue for

the molten alloy and, secondly, it was used to position the

specimens in the simulation test machine.

The specimens were cast by plasma, in the machine

Discovery Plasma (EDG Equipments and Controls Ltd., Sao

Carlos, Brazil), where the melting was made by arc melting in a

vacuum and argon inert atmosphere, with injection of the alloy/

metal into the mold by vacuum–pressure. After casting, the

specimens were removed from the cast and airborne-particle

abraded with aluminum oxide particles (80 psi = 5.62 kgf/

cm2). When nodules and burs were noted, they were removed

using tungsten burs under magnification. Co–Cr specimens

were subject to electrolytic polishing in a VRC apparatus for

15 min (R.R. Equipments for Dental Prostheses Ltd., Sao

Paulo, Brazil), and cp Ti specimens received chemical

polishing by soaking in Kroll solution (10 mL HF, 30 mL

HNO3 and 50 mL water) for 1 min.

Before the tests, all specimens were radiographically

examined to detect possible casting defects that would

contraindicate their use in the tests using a laboratorial unit

X-Control (Dentaurum, Ispringen, Germany).

Specimens were divided into three groups according to their

support system:

– conventional RPD (C);

– implant supported RPD (IS), using a distal implant with a

healing cap to improve the support;

– implant retained RPD (IR), using a distal implant with a ball

attachment and O’ring to improve the support and the

retention.

A layer (3 mm) of polyether film [15] was applied at the

distal extension of the matrix to simulate the mucosa in the C

RPD group (Fig. 1A). For the C RPDs, framework specimens

were positioned at the matrix and the base was constructed

polymerizing the resin (Stern Tek, Sterngold, Attleboro, MA,

USA) for 4 min in a light-curing unit (EDG Lux, EDG

Equipments and Controls Ltd.).

At IS RPD group, an implant (Titamax CM 5 mm � 8 mm,

Neodent, Curitiba, Brazil) was inserted with cyanoacrylate

based glue (Super Bonder, Loctite, Sao Paulo, Brazil) at the

matrix using a dental surveyor to replace the second molar. A

healing cap (Neodent) measuring 4.5 mm in diameter and

1.5 mm in height was installed in the implant with a torque of

10 N cm. After this, the framework was positioned and the base

constructed over the healing cap at the same manner of the C

RPD group (Fig. 1B).

At IR RPD group, the healing cap was replaced by a ball

attachment measuring 1.5 mm in height (Neodent) that was

installed in the implants with the recommended insertion torque

of 20 N cm. The framework was positioned at the matrix and a

cylinder with O’ring was captured using light-curing resin

(Stern-Tek) at the moment of the base construction (Fig. 1C).

2.2. Simulation test

After specimen preparation, the simulation test was

performed in an insertion/removal testing apparatus designed

at the Department of Dental Materials and Prosthodontics of the

Dental School of the Ribeirao Preto, University of Sao Paulo,

Brazil [6,12,13]. The apparatus allowed for the insertion of the

metallic framework into its terminal position on the Co–Cr

matrix passively without any load and its subsequent removal,

thus simulating the placement and removal of an RPD. The

matrix and specimen were positioned in an acrylic box of

testing apparatus that permits deionized water immersion. Tests

were performed in a wet environment at 37 � 5 8C and

Fig. 1. Matrix for conventional (A), implant supported (B) and implant retained (C) RPD frameworks.

R.C.S. Rodrigues et al. / Journal of Prosthodontic Research 57 (2013) 24–2926

temperature was controlled by a thermostat and a resistance

[12,13]. The wet environment at 37 8C was set up based on

body temperature and previous studies [12,13] that used wet

environment to better simulate the oral condition decreasing the

attrition between the matrix and framework. Deionized water

was chosen to avoid chemical interaction of the solution with

the matrix, framework and prosthetic components.

To analyze the data obtained during the simulation test,

intervals corresponding to 0, 1/2, 1, 2, 3, 4 and 5 years were

established. A total of 7205 cycles were performed, represent-

ing the simulated insertion and removal of the RPD over 5

years, estimating that the patient would perform 4 complete

cycles per day [6,12,13]. The test was performed with

41 cycles/min at a constant speed of 35.79 mm/s. The value

established for each time interval corresponded to the

arithmetic average of 10 consecutive insertion/removal cycles.

The force required for each specimen removal was captured and

stored using data acquisition software (LabView 5.0.1, National

Instruments, Austin, TX, USA).

To evaluate the effect of time and compare the different

prostheses at the retention force, the statistical analysis was

performed using a mixed linear model, which is a general-

ization of the standard linear model (ANOVA). The compar-

isons were made by orthogonal contrasts using PROC MIXED

procedure of SAS software (SAS Institute Corp., Cary, NC,

USA). This model is used in the analysis of data in which the

responses of the same specimen are grouped, and the

assumption of independence among the observations in the

same group is not adequate [16]. For the utilization of this

model, it is necessary that the errors have a normal distribution

with mean zero and constant variance. Differences were

considered significant when p < 0.05.

2.3. Photoelastic analysis

Two rectangular (68 mm � 30 mm � 15 mm) models of

polycarbonate with orifices for insertion of teeth and implant

were used as master models. In the first model, the first and the

second inferior premolars of resin (Odontofix, Ribeirao Preto,

Brazil) were inserted in the orifices with cyanoacrylate based

glue (Super Bonder) using a dental surveyor. In the second

model, not only the teeth were inserted but also an analog of the

implant (Titamax CM 5 mm � 8 mm, Neodent) was inserted in

other orifice at the position of the right inferior second molar.

Master models were impressed using a duplication silicone

(Silicone Master, Talmax) to reproduce accurately tooth

position at the photoelastic model. Implant position was

reproduced using an impression transfer. After 24 h, transfer

was unscrewed to permit master model removal.

Tooth root received a standardized layer (300 mm) of

polyether (Impregum, 3 M ESPE, Seefeld, Germany) to simulate

periodontal tissues [17]. After teeth were fit to their silicone molds

and implant (Titamax CM 5 mm � 8 mm, Neodent) was screwed

to its transfer, photoelastic resin (Araldite CY 279 and Aradur

2963, Huntsman, Everberg, Belgium) was manipulated accord-

ing to manufacturer’s instructions and poured into the mold. The

implant was directly embedded in the photoelastic resin without

any interposed material to simulate osseointegration. After 72 h,

the photoelastic model was removed and analyzed.

The healing cap and ball attachment were installed at the

implant at the same manner of the specimen preparation for

simulation test. Similarly, the base was constructed directly

over the photoelastic model and teeth were mounted over the

base and the resin (Stern Tek) polymerized for 4 min in a light-

curing unit (EDG Lux).

Photoelastic models were qualitatively examined for stress

in the field of a circular polariscope (PS-100 SF Polarimeter,

Strainoptics Technologies, North Wales, PA, USA). A digital

camera (EOS Rebel, Canon, Tokyo, Japan) was coupled to the

polariscope to photograph each situation of interest. When any

stress was noted before load applying, models were maintained

at 50 8C in an oven trough 10 min to relieve the stress. Loads

were applied with a straining frame and monitored by means of

a calibrated 50 kgf load cell (Kratos, Sao Paulo, Brazil)

mounted on the movable head of the frame and a digital display

(IKE-01, Kratos). An occlusal load of 15 kgf was applied over

support teeth and RPD. Antagonist teeth were mounted in an

ideal occlusion against support teeth and RPD using light-

curing resin (Stern-Tek). This load was selected because they

are functional load levels and they also provide a satisfactory

optical response within the model.

The stresses around teeth and implant were monitored and

recorded photographically in the field of a circular polariscope

and the monochromatic fringe pattern findings were accom-

plished for each condition of load application [18].

3. Results

3.1. Simulation test

The retention force results of different prostheses throughout

time are presented in Fig. 2. Retention force of IR RPDs is

Table 1

Comparison among conventional (C), implant supported (IS) and implant

retained (IR) RPDs.

Comparison Metal/alloy Differencea p-Value 95% CI*

Inferior Superior

C–IS Co–Cr �1.040 0.002 �1.697 �0.384

C–IR �1.767 <0.001 �2.423 �1.111

IS–IR �0.727 0.030 �1.383 �0.070

C–IS cp Ti �0.668 0.046 �1.325 �0.012

C–IR �2.526 <0.001 �3.182 �1.869

IS–IR �1.858 <0.001 �2.514 �1.201

a Difference between means of each prosthesis type for each metal/alloy.* CI, confidence interval for the estimate of mean difference.

Fig. 2. Retention force (N) throughout simulation test.

R.C.S. Rodrigues et al. / Journal of Prosthodontic Research 57 (2013) 24–29 27

statistically greater than IS and C RPDs for both cp Ti and Co–

Cr alloy specimens (Table 1).

Retention force of Co–Cr C RPDs after 5 years is

statistically greater than at 6-month, 1 and 2-year periods.

At IS RPD, retention force after 5 years is statistically greater

than at 1-year period while IR RPD presented a significant

decrease only at first six months. However, no differences were

noted for cp Ti C and IS RPDs while IR RPDs presented

retention force statistically greater after 4 and 5 years of

simulated use than at the beginning (Table 2).

3.2. Photoelastic analysis

Fringe patterns presented in Fig. 3 show the stress

distribution around support teeth and/or implant of C, IS and

IR RPD when load is applied. When load is applied over C

RPD, stress around the second (support tooth) and first

premolar is noted and residual alveolar ridge region is free of

stress (Fig. 3A). The presence of an implant at the distal

extension contributes to stress distribution eliminating the

stress around the first premolar (Fig. 3B and C). However, stress

concentration around implant and second premolar in the IS

RPD (Fig. 3B) is slightly greater than in the IR RPD (Fig. 3C).

4. Discussion

In the present study, retention force was evaluated by

simulation test and cp Ti RPDs were less retentive than Co–Cr.

Retention force of cp Ti RPDs increased initially and was

maintained throughout 5 years of simulation test while Co–Cr

presented a decrease at the beginning of the test and had their

retention force maintained throughout the test. These results are

according to some authors [6,12] that evaluated retention force

of circumferential and T bar (cp Ti and Co–Cr) clasps relating

maintenance of retention force throughout the test. Never-

theless, the abutment tooth and its effective guide-planes used

in this study consist of a rigid system while the periodontal

ligament present in the mouth allows for small tooth movement.

Because there are usually different insertion and removal paths

in the mouth, since obtaining truly effective guide-planes is

conditioned by anatomical aspects, and patients can change the

path used to move the denture at each insertion and/or removal

cycle, greater loads are produced on the tooth, thus leading to

permanent clasp deformation in a short period of time [12].

According to Sato et al. [19], when a clasp is constructed in

clinical situations, the inner surface is sandblasted and will be

abraded and have a shiny spot, decreasing friction coefficient

and consequently the retention force. The same authors also

related that, for a metal abutment, the initial period will indicate

a large friction coefficient and after the inner surface of the

clasp is abraded, the friction coefficient is lowered. Considering

that electrolytic polishing was used for framework specimen in

the present study, the contact between inner surface of clasp and

polished metallic matrix probably decreased friction coefficient

and retention force because of abrasion at the beginning of the

test. On the other hand, cp Ti exhibit high friction values and an

increased tendency for material transfer and adhesive wear

[20]. A study that evaluated wear resistance of cp Ti and

titanium alloys suggested a deformation at the border of worn

surface [21] confirming adhesive wear of cp Ti. Thus, the

increase of retention force of cp Ti specimens could be

attributed to this adhesive wear and deformation that increases

surface contact and consequently friction coefficient.

An author argued that a RPD should present a force of

approximately 20 N to hold it in place when sticky foods are

chewed and still allow patients to remove RPD while clasps

should present a retention of 5–10 N to be appropriate [22]. As

the specimen framework of the present study is made of only

one clasp, the retention force of the Co–Cr RPDs would be

appropriate for clinical use. However, cp Ti frameworks of C

and IS RPD groups presented retention force ahead of the

values recommended in the literature [22] but these results

reinforces the clinical necessity of using cp Ti clasps in more

retentive areas than Co–Cr, once low modulus of elasticity

permits to position cp Ti clasps in more retentive areas without

oblique load transmission to support teeth at the insertion and

removal, providing better esthetics [12].

The results of the present study show that prosthesis

retention was maintained throughout 5 years of simulation test

independent of the metal/alloy or type of prosthesis. Previous

studies [6,12] that evaluated RPD framework retention without

implant association had their retention force maintained

throughout the time, as noted in the present study. Thus, the

use of implant to support or to retain RPD did not affect

retention throughout the time, but the use of implant increased

Table 2

Comparison of retention force (N) throughout the time for the different prostheses.* Data are presented as mean (standard deviation).

Initial 6 months 1 year 2 years 3 years 4 years 5 years

Co–Cr C 6.6 (2.9)ab 5.1 (2.1)a 4.9 (1.4)a 5.3 (1.2)a 5.5 (1.3)ab 6.3 (1.8)ab 7.2 (2.2)b

IS 6.9 (1.4)ab 6.4 (1.9)ab 6.1 (1.8)a 6.7 (2.2)ab 6.7 (2.4)ab 7.6 (2.1)ab 7.9 (2.1)b

IR 8.7 (1.7)a 6.9 (1.0)b 7.2 (1.7)ab 7.1 (1.6)ab 7.6 (1.8)ab 7.7 (1.9)ab 8.1 (1.6)ab

cp Ti C 2.3 (0.4)a 3.1 (1.1)a 2.9 (1.2)a 3.1 (1.2)a 3.1 (1.3)a 3.3 (1.3)a 3.3 (1.0)a

IS 2.4 (0.6)a 3.9 (1.0)a 3.9 (0.9)a 3.9 (0.6)a 3.8 (0.7)a 3.8 (0.6)a 3.9 (0.7)a

IR 4.0 (0.8)a 5.4 (1.6)ab 5.5 (1.7)ab 5.7 (1.9)ab 5.7 (1.9)ab 6.2 (1.7)b 6.1 (1.5)b

* Different letters mean statistically significant difference at the line (a = 0.05).

Fig. 3. Stress distribution in the photoelastic models using a conventional (A), implant supported (B) and implant retained (C) RPD frameworks.

R.C.S. Rodrigues et al. / Journal of Prosthodontic Research 57 (2013) 24–2928

retention when compared to conventional RPD frameworks,

mainly in the IR group.

Although implant retained overdentures using ball attach-

ment/O’ring present a decrease of retention force after clinical

or simulated use [23–26], IR RPDs did not present a decrease of

retention force probably because of RPD insertion/removal

path that avoids lateral and oblique dislodgement, protecting

O’ring rubber from wear or deformation.

In addition to the better retention of IS and IR RPDs, the

main advantage of RPD/implant association is the load

transmission. According to some authors [1,2,14], implant

movement is lower than periodontal tissues, decreasing base

movement toward residual ridge. The benefits of implant

association are to prevent bone loss, decrease stress around

support teeth and to improve retention, masticatory efficiency

and comfort [2].

Photoelastic models of the present study demonstrate the

importance of an implant at residual alveolar ridge region to

decrease stress around teeth, mainly the first premolar. These

results corroborate other study [1] that found greater load

concentration around support teeth at conventional RPD and

decrease of load transmission in the region of the residual ridge

between support teeth and implant, when an implant in the

distal extension was used even load transmission around

support teeth was not decreased. IS and IR RPDs decrease load

transmission to the first premolar and increase transmission for

implant area. Additionally, the use of a ball attachment/O’ring

at the IR RPDs improved load distribution between support

teeth and implant when compared to IS RPDs, probably

because O’ring rubber partially absorbs the load.

According to a crossover study, patients that used

conventional and implant supported RPDs, preferred implant

supported RPDs because of retention, comfort, stability and

chewing. Additionally, implant supported RPDs can be an

alternative for partially edentulous patients when an implant

supported fixed prosthesis cannot be applied because of

anatomical or economic reasons [27].

Thus, the association between RPD and implant can improve

retention and load distribution, representing an interesting

alternative for patients with clinical and financial limitations.

After analyzing images of photoelasticity and results of

simulation test, it is possible to observe that the obtained results

confirm the hypothesis that the association of implant to RPD in

the distal extension residual ridge improves the retention and

stress distribution.

Acknowledgements

The authors thank from Sao Paulo State Research

Foundation – FAPESP (2006/57559-6) for the financial support

and Mr. Luiz Sergio Soares and Mr. Marcelo Aparecido Vieira

by their technical support.

References

[1] Verri FR, Pellizzer EP, Rocha EP, Pereira JA. Influence of length and

diameter of implants associated with distal extension removable partial

dentures. Implant Dent 2007;16:270–80.

[2] Pellizzer EP, Verri FP, Falcon-Antenucci RM, Goiato MC, Gennari Filho

H. Evaluation of different retention systems on a distal extension remov-

able partial denture associated with an osseointegrated implant. J Cra-

niomaxillofac Surg 2010;21:727–34.

[3] Kuzmanovic DV, Payne AGT, Purton DG. Distal implants to modify the

Kennedy classification of a removable partial denture: a clinical report. J

Prosthet Dent 2004;92:8–11.

[4] Turkyilmaz I. Use of distal implants to support and increase retention of a

removable partial denture: a case report. J Calif Dent Assoc 2009;75:

655–8.

[5] Shahmiri RA, Atieh MA. Mandibular Kennedy Class I. implant-tooth-

borne removable partial denture: a systematic review. J Oral Rehabil

2010;37:225–34.

R.C.S. Rodrigues et al. / Journal of Prosthodontic Research 57 (2013) 24–29 29

[6] Rodrigues RCS, Ribeiro RF, Mattos MGC, Bezzon OL. Comparative

study of circumferential clasp retention force for titanium and cobalt–

chromium removable partial dentures. J Prosthet Dent 2002;88:290–6.

[7] Blackman R, Barghi N, Tran C. Dimensional changes in casting titanium

removable partial dentures frameworks. J Prosthet Dent 1991;65:309–15.

[8] Lautenschlager EP, Monaghan P. Titanium and titanium alloys as dental

materials. Int Dent J 1993;43:245–53.

[9] Kotake M, Wakabayashi N, Ai M, Yoneyama T, Hamanaka H. Fatigue

resistance of titanium–nickel alloy cast clasps. Int J Prosthodont

1997;10:547–52.

[10] Jang KS, Youn SJ, Kim YS. Comparison of castability and surface

roughness of commercially pure titanium and cobalt–chromium denture

frameworks. J Prosthet Dent 2001;86:93–8.

[11] Ohkubo C, Hanatani S, Hosoi T. Present status of titanium removable

dentures – a review of the literature. J Oral Rehabil 2008;35:706–14.

[12] Rodrigues RC, Macedo AP, Torres EM, Mattos MGC, Ribeiro RF.

Retention force of T-bar clasps for titanium and cobalt–chromium remov-

able partial dentures. Braz Dent J 2008;19:209–13.

[13] Rodrigues RCS, Faria ACL, Orsi IA, Mattos MGC, Macedo AP, Ribeiro

RF. Comparative study of two commercially pure titanium casting meth-

ods. J Appl Oral Sci 2010;18:487–92.

[14] Cunha LAP, Pellizzer EP, Verri FR, Rocha EP, Pereira JA. Evaluation of

the influence of location of osseointegrated implants associated with

mandibular removable partial dentures. Implant Dent 2008;17:278–87.

[15] Ochiai KT, Williams BH, Hojo S, Nishimura R, Caputo AA. Photoelastic

analysis of the effect of palatal support on various implant-supported

overdenture designs. J Prosthet Dent 2004;91:421–7.

[16] Schall R. Estimation in generalized linear models with random effects.

Biometrika 1991;78:719–27.

[17] Hohmann A, Wolfram U, Geiger M, Boryor A, Sander C, Faltin R, et al.

Periodontal ligament hydrostatic pressure with areas of root resorption

after application of a continuous torque moment. Angle Orthod

2007;77:653–9.

[18] Brodsky JE, Caputo AA, Furstman LL. Root typing: a photoelastic–

histopathologic correlation. Am J Orthod 1975;67:1–10.

[19] Sato Y, Abe Y, Yuasa Y, Akagawa Y. Effect of friction coefficient on Akers

clasp retention. J Prosthet Dent 1997;78:22–7.

[20] Ohkubo C, Shimura I, Aoki T, Hanatani S, Hosoi T, Hattori M, et al. Wear

resistance of experimental Ti–Cu alloys. Biomaterials 2003;24:3377–81.

[21] Faria ACL, Rodrigues RCS, Alves Claro APR, Mattos MGC, Ribeiro RF.

Wear resistance of experimental titanium alloys for dental applications. J

Mech Behav Biomed Mater 2011;4:1873–9.

[22] Shimpo H. Effect of arm design and chemical polishing on retentive force

of cast titanium alloy clasps. J Prosthodont 2008;17:300–7.

[23] Cordioli G, Majzoub Z, Castagna S. Mandibular overdentures anchored to

single implants: a five-year prospective study. J Prosthet Dent

1997;78:159–65.

[24] Walton JN, MacEntee MI. A prospective study on the maintenance of

implant prostheses in private practice. Int J Prosthodont 1997;10:453–8.

[25] Winkler S, Piermatti J, Rothman A, Siamos G. An overview of the O-ring

implant overdenture attachment: clinical reports. J Oral Implantol

2002;28:82–6.

[26] Rodrigues RCS, Faria ACL, Macedo AP, Sartori IAM, Mattos MGC,

Ribeiro RF. An in vitro study of non-axial forces upon the retention of an

O-ring attachment. Clin Oral Implants Res 2009;20:1314–9.

[27] Ohkubo C, Kobayashi M, Suzuki Y, Hosoi T. Effect of implant support on

distal-extension removable partial dentures: in vivo assessment. Int J Oral

Maxillofac Implants 2008;23:1095–101.