Cyclic loading of implant-supported prostheses: Changes in component fit

over time
Donna M. Hecker, DDS, MS,a and Steven E. Eckert, DDS, MSb
School of Dentistry, University of Minnesota, Minneapolis, Minn.; Mayo Clinic, Mayo Graduate
School of Medicine, Rochester, Minn.

Statement of problem. Dental literature suggests that an implant-supported prosthesis must exhibit a passive
fit to prevent implant fracture, component breakage, and screw loosening. From a practical standpoint, passive fit
is impossible to achieve; instead, minimal misfit may be the clinical goal. To date no specific range of misfit (below
which problems are minimal and above which catastrophic failure occurs) has been established.
Purpose. The purpose of this study was to determine whether the fit of an implant-supported prosthesis changes
through cyclic loading and to quantify the amount of change between the gold cylinder and implant abutment
over time.
Materials and methods. Fifteen implant-supported frameworks were fabricated with conventional casting
techniques and were cyclically loaded under 3 different loading conditions. Five frameworks were loaded on the
anterior portion of the framework, 5 were loaded on the left unilateral posterior cantilever, and 5 were loaded
bilaterally on the posterior cantilevers with a servohydraulic testing machine. A cyclical load of 200 N was applied
to each framework for up to 200,000 cycles. Linear measurements were made in micrometers of the gap between
the prosthetic cylinder and the implant-supported abutment at 4 predetermined reference points. These mea-
surements were recorded before the application of the cyclical load, after 50,000 cycles, and after 200,000 cycles.
A repeated measures of variance model was fit separately to the data for each load location (P⬍.05).
Results. There was a significant (P⫽.024) decrease in gap dimensions at individual reference points and a
significant (P⫽.031) decrease in the average gap when the load was applied to the anterior portion of the
framework. When the load was applied unilaterally or bilaterally on the posterior cantilever, significant gap closure
was not observed (P⫽.33 and P⫽.35, respectively).
Conclusion. Within the limitations of this study, the fit between the prosthetic superstructure and the implant-
supported abutment changed when simulated functional loading of the anterior portion of the prosthesis was
performed. Simulated functional loading applied unilaterally or bilaterally to the posterior cantilever portion of
the prosthesis did not result in changes in the measured gap sizes. (J Prosthet Dent 2003;89:346-51.)

CLINICAL IMPLICATIONS
This in vitro study suggests that force application to an implant-supported prosthesis may alter
the fit of the prosthesis to its respective supporting components. These changes may be cumulative
over time, having clinical implications toward maintenance of screw joint integrity.

T he recognition of a biocompatible union between

bone and alloplastic implant material creates a number
of new applications to the field of dentistry. Osseointe-
gration offers new treatment options for the edentulous
and partially edentulous patient. The use of dental im-
plants to support and retain dental prostheses has been
demonstrated to be clinically efficacious.1-4 Although
the bone-to-implant interface may be reliable,5-7 clinical
complications can and do occur at the prosthetic level.
Presented at the Academy of Prosthodontics Meeting, Santa Fe, NM,
May 2001.
a
Director of Maxillofacial Prosthetic Services and Assistant Clinical
Dental Specialist, Department of Restorative Sciences, University
of Minnesota.
b
Associate Professor, Mayo Graduate School of Medicine.
Technical problems continue to frustrate clinicians.
These include inability to achieve intimate fit on fabri-
cation of the prosthetic framework, inability to correct
the misfits, and difficulty in developing an occlusal
scheme that will not overload the assembly.8
The objective of achieving passive fit of the prosthesis
(gold cylinder) to the abutment is addressed by several
authors.9-18 Branemark13 suggested that components
have no more than a 10-␮m misfit, but modern dental
technology is unlikely to consistently achieve this level of
accuracy. Many manufactured components do not dem-
onstrate this level of accuracy of fit.19,20 Consequently,
misfit of prostheses is a clinical reality, but the amount of
misfit that can be tolerated without adverse mechani-
cal21-30 or biologic9,31-34 complications has yet to be
determined.

346 THE JOURNAL OF PROSTHETIC DENTISTRY VOLUME 89 NUMBER 4

HECKER AND ECKERT
Fig. 1. Model testing system included 5 implants arranged in
arc, with uniform (10 mm) inter-implant spacing and fulcrum
line-anterior abutment distance of 9 mm.
This study evaluated the changes that take place in
the gap between the implant and the prosthetic compo-
nents after cyclic loading of prostheses. The goal was to
evaluate changes qualitatively and quantitatively.
MATERIAL AND METHODS
A model system was designed to allow placement of
five 3.75- ⫻ 10-mm implants (Branemark System; No-
bel Biocare, Yorba Linda, Calif.) into a rigid base. The
system used 5 implants arranged in an arc, with uniform
(10 mm) interimplant spacing, a fulcrum line-anterior
abutment distance of 9 mm, and a distal extension can-
tilever of 18 mm (Fig. 1). The model system included
load cells (Model MB-250; Interface, Scottsdale, Ariz.)
attached to each implant for the purpose of measuring
applied loads and detecting any changes in load caused
by change in fit of the prosthesis.
Polysiloxane (Reprosil; Caulk/Dentsply, York, Pa.)
impressions of the implants were made by use of an open
custom tray to engage direct implant transfer abutments
(Nobel Biocare) that had been rigidly connected with
autopolymerizing resin (GC pattern resin; GC America,
Chicago, Ill.). After the fabrication of master casts, new
standard 4-mm abutments were placed on each of the
implant analogues. The abutments were labeled and
marked to ensure the exact position on each hex could
be duplicated on the model testing system.
Fifteen frameworks were waxed, cast, finished, and
polished with conventional lost-wax laboratory tech-
niques. Frameworks incorporated standard 4-mm gold
cylinders (Nobel Biocare). Framework design used an
L-shaped configuration in cross-section, with a lingual
vertical rise of 6 mm and horizontal dimension of
8 mm.35-37 No attempts were made to improve the fit of
the frameworks by sectioning and soldering. A manual
engraving pen was used to mark lines on the abutment at
APRIL 2003
THE JOURNAL OF PROSTHETIC DENTISTRY
Fig. 2. Cylindrical sleeve custom-machined to ensure uni-
form identification of reference points scribed on abutments
at every 60 degrees.
Fig. 3. Graphic representation of reference points labeled a,
b, c, and e. Reference points d and f eliminated because of
lack of access with microscope.
4 different positions, indicating preselected reference
points. A cylindrical sleeve with a slot every 60 degrees
was custom-machined to ensure uniformity between
placement of the engraved lines on the abutments (Fig.
2). Three engraved lines were made on the facial aspect
of the abutment and 1 on the lingual aspect. These
positions were consistent on all abutments and were
labeled a, b, c, and e, respectively (Fig. 3).
The abutments were attached to the implants in the
predetermined position and tightened to a torque of
20 Ncm with a torque driver (Machine, standard 21
mm, Nobel Biocare). The frameworks were attached to
the abutments with the gold retaining screws supplied
by the manufacturer (Nobel Biocare) and were tight-
ened manually with the slotted driver (Slot, 22/37; No-
bel Biocare). An electronic torque wrench (Torque mea-
surement device, custom; NK Biotechnical Corp,
Minneapolis, Minn.) with digital read-out capability was
used to tighten the gold screws to the recommended
preload force of 10 Ncm. These values were recorded
before load application.
347
THE JOURNAL OF PROSTHETIC DENTISTRY
Fig. 4. Frameworks were evaluated for vertical discrepancies
between components with micron microscope.
A micron microscope (Model # FS-60; Mitutoyo
Corp, Kawasaki, Japan) with a traveling stage and
digital measurement read-out system (VRZ-720-B
X-Y; Heidenhain, Schaumberg, Ill.) was used to visu-
ally aid in quantifying the vertical space between the
abutment and gold cylinder at the mating surface
(Fig. 4). The vertical discrepancies were measured and
recorded before cycling, after 50,000 cycles, and after
200,000 cycles. Measurement values were recorded in
micrometers to a resolution of 0.0005 in/0.01 mm
(Figs. 5 and 6).
The measurements were made at 4 different locations
around the abutment, to quantify the change in misfit at
specific locations. These individual points could also be
averaged to quantify change in a planar dimension. The
values were recorded in a laboratory notebook and then
entered into a spreadsheet for subsequent analysis.
The summary of the distributions of the gap values
(averaged over reference points, implant positions, and
frames by load location and cycling interval) is shown in
Table I. The summary statistics are based on 25 values (5
348
HECKER AND ECKERT
Fig. 5. View through microscope demonstrates 10-␮m verti-
cal discrepancy (gap).
Fig. 6. View through microscope demonstrates 147-␮m ver-
tical discrepancy (gap).
implants within each of the 5 frameworks). A repeated
measures analysis of variance model was fit separately to
the data for each of the 3 load locations. In this analysis,
the gap changes that occur with cycling were different
between the 4 reference points (P⬍.05).
RESULTS
There was a significant decrease in gap measurements
considering each reference point separately when the
load was applied to the anterior portion of the frame-
work (P⫽.024), but not when the load was applied uni-
laterally on the posterior cantilever (P⫽.33) or when the
load was applied bilaterally on the posterior cantilever
(P⫽.35). When the load was applied to the anterior
portion of the framework, there was a significant change
with cycling in the gap measurements at reference point
e, which corresponds to the lingual aspect of the abut-
ment (P⫽.023). The gap measurements did not change
significantly with cycling at any of the other 3 reference
points.
VOLUME 89 NUMBER 4
HECKER AND ECKERT THE JOURNAL OF PROSTHETIC DENTISTRY

Table I. Summary of the distributions of gap measurements (averaged over reference points, implant positions and frames)
by load location and cycling interval

Geometric
Load location Cycle mean Median Mean (std) Minimum Maximum 25th percentile 75th percentile

Unilateral Posterior Pre-load 0.067 0.076 0.108 (0.092) 0.009 0.334 0.038 0.142
Unilateral Posterior 50,000 0.062 0.068 0.105 (0.095) 0.007 0.334 0.033 0.150
Unilateral Posterior 200,000 0.066 0.069 0.104 (0.091) 0.010 0.325 0.038 0.131
Anterior Pre-load 0.054 0.047 0.082 (0.077) 0.014 0.346 0.039 0.115
Anterior 50,000 0.052 0.049 0.082 (0.082) 0.013 0.378 0.035 0.110
Anterior 200,000 0.049 0.048 0.075 (0.070) 0.014 0.299 0.025 0.123
Bilateral Posterior Pre-load 0.047 0.052 0.078 (0.060) 0.005 0.199 0.034 0.124
Bilateral Posterior 50,000 0.042 0.050 0.075 (0.059) 0.007 0.212 0.034 0.122
Bilateral Posterior 200,000 0.041 0.055 0.071 (0.058) 0.006 0.209 0.027 0.116

When the load was applied bilaterally to the posterior
cantilevers, there was a significant change with cycling in
the gap measurements at reference point e (P⫽.003).
The gap measurements did not change significantly with
cycling at any of the other 3 reference points.
In an attempt to summarize the gaps across the 4
reference points on the abutment, the average of these
gap measurements per abutment was used. Averaging
the 4 individual reference points allows inferences to be
made about changes occurring in the planar dimension.
A repeated measures analysis of variance model was fit
separately to the data for each of the 3 load locations.
There was evidence of a significant decrease in the aver-
age gaps due to cycling when the load was applied to the
anterior portion of the framework (P⫽.031); the aver-
age gaps were significantly less after 200,000 cycles than
the average gaps before testing.
DISCUSSION
The purpose of this study was to investigate changes
that occur in fit at the gold cylinder to implant abutment
interface over time. The investigation was initiated in
response to the recognition that there is little likelihood
of achieving a perfect fit between these components with
current dental technology. Given the inevitable nature
of imperfection in fit, it seems prudent to understand the
effect of cyclic loading on the fitting surfaces of the
implant-supported prosthesis.
In evaluating the determinants of a passively fitting
prosthesis, several factors should be considered. These
include the definition of passive fit and the accuracy of
the procedures used to achieve this degree of passivity.
Other factors include the method by which the fit is
evaluated and the role of the premachined components.
Numerous methods exist for evaluating framework
fit. Different methods apply to the clinical and labora-
tory settings. Both settings involve use of visual or tactile
inspection. Visual inspection may include the 1-screw
test,14 use of disclosing medium,15 fabrication of a cast
verification jig,20 or use of radiographic films.22 All of
these methods require visualization of the space be-
tween 2 components. Tactile evaluation of fit is per-
formed by placing the prosthesis in contact with its re-
spective abutments and feeling for even seating of the
prosthesis. Once the initial seating is confirmed, retain-
ing screws are alternately tightened and loosened to de-
termine whether movement of the prosthesis occurs
while the individual screws are secured.
The visual and tactile methods of evaluation attempt
to bring objectivity to the evaluation process, but both
retain a level of subjectivity. Conversely, laboratory
methods for measuring fit 3-dimensionally at the im-
plant-prosthodontic interface may eliminate this level of
subjectivity. There are essentially 3 methods for labora-
tory assessment: stylus contact techniques,16 laser video-
graphy,17 and photogrammetry.18 Each method pro-
vides precise measurements to the level of approximately
3.5 ␮m. Of course, any discrepancy between the oral
cavity and the master cast will continue to create clinical
misfit of the prosthetic frame.
Although there are many ways to assess 3-dimen-
sional fit of prostheses to abutments,16-18 the sophisti-
cated testing equipment required for such testing is not
readily available. The use of more common measuring
techniques still provides information on the relative fit
or misfit of a prosthesis while recognizing the fact that
this is not as precise as the other methods.
Another factor that must be considered in the fabri-
cation of prostheses is the reproducible fit of the prema-
chined components. These components are machined
by the manufacturer to ensure a specific tolerance. Ma-
chining tolerances define the limit of acceptability when
assessing the measurement of a component feature in
relation to the specified design value target. There can
be no consistently greater accuracy of the final prosthesis
fit to its abutments than the level of machining toler-
ance. Ma et al21 demonstrated a range of tolerance for
the implant components (abutment, gold cylinder, im-
pression coping, and brass abutment analog) in the hor-
izontal plane from 22 to 100 ␮m. This study also dem-

APRIL 2003 349

THE JOURNAL OF PROSTHETIC DENTISTRY
onstrated a discrepancy between the machining
tolerances of the prosthetic components and the labora-
tory analogs. The machining tolerance of brass analog
and gold cylinder was significantly larger compared with
that of the abutment and gold cylinder. This level of
discrepancy may result in a prosthesis that appears to fit
in the laboratory, whereas the same prosthesis may have
a larger misfit in the clinical setting. Rubenstein and
Ma19 found similar results when comparing components
for titanium laser-welded and standard cast implant
components. A range of 23.1 to 51.7 ␮m for machining
tolerances was reported.
This study demonstrated that under specific loading
conditions, the gap between the implant abutment and
the prosthesis can undergo significant change. In this
study, the change resulted in smaller gaps after cyclic
loading. This change is probably due to wear at the
component interface, because some point of contact
must prevent the components from achieving the man-
ufactured fit. Once this wear occurs and prosthetic re-
taining screws are retightened to ideal levels, it is possi-
ble that the net effect may be favorable toward future
component loosening or breakage. However, if compo-
nents do become dislodged, the orientation of the worn
surfaces could be altered. The consequences of such a
change are currently being investigated in a subsequent
study.
CONCLUSIONS
Within the limitations of this study, cyclic loading of
implant-supported frames caused changes in the fit of
the frame to the supporting implant abutments. The fit
of the superstructure showed significant gap reductions
when the prosthesis was loaded on the anterior segment.
Fit did not change significantly when the frameworks
were loaded on the unilateral or bilateral posterior can-
tilever portions of the prostheses.
REFERENCES
1. Adell R, Lekholm U, Rockler B, Branemark PI. A 15-year study of os-
seointegrated implants in the treatment of the edentulous jaw. Int J Oral
Surg 1981;10:387-416.
2. Adell R, Eriksson B, Lekholm U, Branemark PI, Jemt T. Long-term fol-
low-up study of osseointegrated implants in the treatment of totally eden-
tulous jaws. Int J Oral Maxillofac Implants 1990;5:347-59.
3. Cox JF, Zarb GA. The longitudinal clinical efficacy of osseointegrated
dental implants: a 3-year report. Int J Oral Maxillofac Implants 1987;2:
91-100.
4. Laney WR, Tolman DE, Keller EE, Desjardins RP, Van Roekel NB, Brane-
mark PI. Dental implants: tissue-integrated prosthesis utilizing the os-
seointegration concept. Mayo Clin Proc 1986;61:91-7.
5. Branemark PI, Adell R, Breine U, Hansson BO, Lindstrom J, Ohlsson A.
Intra-osseous anchorage of dental prostheses. I. Experimental studies.
Scand J Plast Reconstr Surg 1969;3:81-100.
6. Carr AB, Gerard DA, Larsen PE. The response of bone in primates around
unloaded dental implants supporting prostheses with different levels of fit.
J Prosthet Dent 1996;76:500-9.
7. Clelland NL, Carr AB, Gilat A. Comparison of strains transferred to a bone
simulant between as-cast and postsoldered implant frameworks for a
five-implant-supported fixed prosthesis. J Prosthodont 1996;5:193-200.
350
HECKER AND ECKERT
8. Zervas PJ, Papazoglou E, Beck FM, Carr AB. Distortion of three-unit
implant frameworks during casting, soldering, and simulated porcelain
firings. J Prosthodont 1999;8:171-9.
9. Rangert B, Jemt T, Jorneus L. Forces and moments on Branemark implants.
Int J Oral Maxillofac Implants 1989;4:241-7.
10. Rangert B, Gunne J, Sullivan DY. Mechanical aspects of a Branemark
implant connected to a natural tooth: an in vitro study. Int J Oral Maxil-
lofac Implants 1991;6:177-86.
11. Weinberg LA. The biomechanics of force distribution in implant-sup-
ported prostheses. Int J Oral Maxillofac Implants 1993;8:19-31.
12. Zarb GA, Schmitt A. The longitudinal clinical effectiveness of osseointe-
grated implants: the Toronto study. Part III: Problems and complications
encountered. J Prosthet Dent 1990;64:185-94.
13. Branemark PI, Zarb G, Albrektsson T. Tissue-integrated prostheses: os-
seointegration in clinical dentistry. Special edition for Nobelpharma. Chi-
cago: Quintessence; 1987. p. 268-71.
14. Jemt T. Failures and complications in 391 consecutively inserted fixed
prostheses supported by Branemark implants in edentulous jaws: a study
of treatment from the time of prosthesis placement to the first annual
checkup. Int J Oral Maxillofac Implants 1991;6:270-6.
15. Yanase RT, Binon PP, Jemt T, Gulbransen HJ, Parel SP. Current issues
forum: How do you test a cast framework for for a full-arch fixed implant-
supported prosthesis? Int J Oral Maxillofac Implants 1994;9:469-74
16. Tan KB, Rubenstein JE, Nicholls JI, Yuodelis RA. Three-dimensional anal-
ysis of the casting accuracy of one-piece, osseointegrated implant-retained
prostheses. Int J Prosthodont 1993;6:346-63.
17. Riedy SJ, Lang BR, Lang BE. Fit of implant frameworks fabricated by
different techniques. J Prosthet Dent 1997;78:596-604.
18. Lie A, Jemt T. Photogrammetric measurements of implant positions. De-
scription of a technique to determine the fit between implants and super-
structures. Clin Oral Implants Res 1994;5:30-6.
19. Rubenstein JE, Ma T. Comparison of interface relationships between im-
plant components for laser-welded titanium frameworks and standard cast
frameworks. Int J Oral Maxillofac Implants 1999;14:491-5.
20. Binon PP. Evaluation of machining accuracy and consistency of selected
implants, standard abutments, and laboratory analogs. Int J Prosthodont
1995;8:162-78.
21. Ma T, Nicholls JI, Rubenstein JE. Tolerance measurements of various
implant components. Int J Oral Maxillofac Implants 1997;12:371-5.
22. Cox JF, Pharoah M. An alternative holder for radiographic evaluation of
tissue-integrated prostheses. J Prosthet Dent 1986;56:338-41.
23. Jemt T, Lekholm U. Measurements of bone and frame-work deformations
induced by misfit of implant superstructures: a pilot study in rabbits. Clin
Oral Implants Res 1998;9:272-80.
24. Piattelli A, Piattelli M, Scarano A, Montesani L. Light and scanning elec-
tron microscopic report of four fractured implants. Int J Oral Maxillofac
Implants 1998;13:561-4.
25. Eckert SE, Wollan PC. Retrospective review of 1170 endosseous implants
placed in partially edentulous jaws. J Prosthet Dent 1998;79:415-21.
26. Taylor TD. Prosthodontic problems and limitations associated with os-
seointegration. J Prosthet Dent 1998;79:74-8.
27. Binon PP, McHugh MJ. The effect of eliminating implant/abutment rota-
tional misfit on screw joint stability. Int J Prosthod 1996;9:511-9.
28. Niznick GA. Bending overload and implant fracture: a retrospective clin-
ical analysis. Int J Oral Maxillofac Implants 1996;11:431-2.
29. Rangert B, Krogh PH, Langer B, Van Roekel N. Bending overload and
implant fracture: a retrospective clinical analysis. Int J Oral Maxillofac
Implants 1995;10:326-34.
30. Kallus T, Bessing C. Loose gold screws frequently occur in full-arch fixed
prostheses supported by osseointegrated implants after 5 years. Int J Oral
Maxillofac Implants 1994;9:169-78.
31. Goodacre CJ, Kan JY, Rungcharassaeng K. Clinical complications of os-
seointegrated implants. J Prosthet Dent 1999;81:537-52.
32. Eckert SE, Meraw SJ, Cal E, Ow RK. Analysis of incidence and associated
factors with fractured implants: a retrospective study. Int J Oral Maxillofac
Implants 2000;15:662-7.
33. Carr AB, Gerard DA, Larsen PE. The response of bone in primates around
unloaded dental implants supporting prostheses with different levels of fit.
J Prosthet Dent 1996;76:500-9.
34. Jemt T, Lekholm U, Johansson CB. Bone response to implant-supported
frameworks with differing degrees of misfit preload: in vivo study in
rabbits. Clin Implant Dent Relat Res 2000;2:129-37.
VOLUME 89 NUMBER 4
HECKER AND ECKERT THE JOURNAL OF PROSTHETIC DENTISTRY

35. Clelland NL, Papazoglou E, Carr AB, Gilat A. Comparison of strains Reprint requests to:
transferred to a bone simulant among implant overdenture bars with DR DONNA M HECKER
various levels of misfit. J Prosthodont 1995;4:243-50. UNIVERSITY OF MINNESOTA
36. Michaels GC, Carr AB, Larsen PE. Effect of prosthetic superstructure ROOM 6-284 MOOS TOWER
accuracy on the osteointegrated implant bone interface. Oral Surg Oral 515 DELAWARE ST SE
Med Oral Pathol Oral Radiol Endod 1997;83:198-205. MINNEAPOLIS, MN 55455
37. Stewart RB, Desjardins RP, Laney WR, Chao EY. Fatigue strength of FAX: 612-624-2660
cantilevered metal frameworks for tissue-integrated prostheses. J Prosthet E-MAIL: hecke003@tc.umn.edu
Dent 1992;68:83-92.
38. Stewart RB, Staab GH. Cross-sectional design and fatigue durability of Copyright © 2003 by The Editorial Council of The Journal of Prosthetic
cantilevered sections of fixed implant-supported prostheses. J Prosthodont Dentistry.
1995;4:188-94. 0022-3913/2003/$30.00 ⫹ 0
39. Staab GH, Stewart RB. Theoretical assessment of cross sections for canti-
levered implant-supported prostheses. J Prosthodont 1994;3:23-30. doi:10.1067/mpr.2003.71

APRIL 2003 351