Clinical Biomechanics 65 (2019) 92–99

Contents lists available at ScienceDirect

Clinical Biomechanics
journal homepage: www.elsevier.com/locate/clinbiomech

In vitro assessment of connection strength and stability of internal implant- T

abutment connections
Michelle D. Kofron , Matthew Carstens, Cong Fu, Hai Bo Wen
⁎

Biomet 3i, LLC, a Zimmer Biomet company, 4555 Riverside Drive, Palm Beach Gardens, FL, United States

ARTICLE INFO ABSTRACT

Keywords: Background: Various connections have been machined to improve the fit between the dental abutment and
Internal connection implant. In vivo, the instability created by imprecisely fitting components can cause soft tissue irritation and
Microgaps bacterial colonization of the implant system. The aim of this study was to quantify abutment stability under in
Micromotion vitro force applications.
Abutment stability
Methods: Abutment stability and fit were quantitatively measured after application of rotational, vertical, and
Axial loading
horizontal forces.
Findings: The abutment connection held by friction (Friction-Fit) was the only group to have 0° angular rotation.
A significantly greater vertical force was required to pull the abutment from the implant for the Friction-Fit
connection as compared to all other experimental groups. The abutment connection held by a mechanically
locking friction-fit with four grooves (CrossFit) and Friction-Fit demonstrated significantly lower lateral
movement as compared to all other connections. The remaining connections evaluated included two hexagon
connections that rely on screw placement for abutment fit (Conical + Hex #1 and Conical + Hex #2), one
connection with protruding slots to align with recessed channels inside the implant (Conical + 6 Indexing Slots),
and an internal connection that allows for abutment indexing every 120° (Internal Tri-Channel).
Interpretation: Internal connection geometry influenced the degree of abutment movement. Friction-Fit and
CrossFit connections exhibited the lowest rotational and horizontal motions. Significant differences were found
between Friction-Fit and CrossFit following the application of a vertical force, with the Friction-Fit requiring a
significantly greater pull force to separate the abutment from the implant.

1. Introduction Steenberghe, 1993; Steinebrunner et al., 2005; Tripodi et al., 2015;

Verdugo et al., 2014; Zipprich et al., 2016). A correlation between
Two-piece implant systems, consisting of the implant and abutment, microleakage and screw loosening has been demonstrated, as screw
are widely used in dental restorations. The advantage over a one-piece loosening enlarges the microgap (Sahin and Ayyildiz, 2014). Thus,
implant system is that it allows the implant to be unloaded during the microgaps contribute to mechanical failure.
bone healing phase and provides the benefit to adjust the prosthetic Microgap also provides space as a potential site for bacteria, fluid
angle depending on the abutment selected (angled or straight) for and macromolecules from the crevicular fluid and saliva. Abutment
placement on the implant. A disadvantage to a two-piece implant micromovements, while seated on the implant, have been reported to
system is the resultant microgaps or spaces that exist along the implant- create a pumping effect that may allow the bacteria to migrate and
abutment interface when the abutment is seated on the implant and macromolecules to flow through the implant-abutment junction to the
connected via the abutment screw. Under loading conditions, these internal aspect of the implant system (Alves et al., 2016; do Nascimento
spaces permit rotation and micromotion of the abutment and can lead et al., 2009; Quirynen and van Steenberghe, 1993; Scarano et al.,
to screw preload reduction, screw loosening, bending, or fracture (Kano 2015). The average microgap of the implant abutment junction has
et al., 2007; Yao et al., 2019). The existence of microgap has been been reported to be from 1 to 49 μm (Mishra et al., 2017; Quirynen and
demonstrated using scanning electron microscopy and microleakage van Steenberghe, 1993), while the average size of bacteria is 0.2 to
studies (Alves et al., 2016; Cavusoglu et al., 2014; do Nascimento et al., 1.5 μm in width and 1 to 10 μm in length depending on shape (El
2012; El Haddad et al., 2016; Koutouzis et al., 2014; Quirynen and van Haddad et al., 2016; Nascimento et al., 2016). The localized bacteria

⁎
Corresponding author.
E-mail address: [email protected] (M.D. Kofron).

https://doi.org/10.1016/j.clinbiomech.2019.03.007
Received 19 November 2018; Accepted 11 March 2019
0268-0033/ © 2019 Biomet 3i, LLC, a Zimmer Biomet company. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/BY-NC-ND/4.0/).
M.D. Kofron, et al. Clinical Biomechanics 65 (2019) 92–99

colonization initiates recruitment of inflammatory cells. The sub- switched implants than in platform-matched implants. As the force
sequent inflammatory reaction affects the maintenance of soft and hard distribution occurs deep within the implant wall and distributes itself
tissue formation around the implant complex (Jones and Cochran, outwards towards the bone, rather than on the abutment screw, the
2006). In extreme cases, bacterial colonization can lead to peri-im- incidence of screw loosening has been reduced to 3.7% (Levine et al.,
plantitis, which is characterized by severe soft tissue inflammation, 1997). A disadvantage to internal connections is the thinner, and
bone resorption, and loss of implant integration (El Haddad et al., 2016; thereby weaker, implant wall at the collar to allow space for the con-
Pokrowiecki et al., 2017; Raab et al., 2017; Sahin and Ayyildiz, 2014; nection. As a more recent design concept, there is also less historical
Salvi et al., 2017; Siadat et al., 2017; Tallarico et al., 2017; Tsuruta data on internal connections as compared to external connections.
et al., 2018; Zipprich et al., 2007). Because of this, connections are Conical connections are a specific type of internal connection based
designed and machined with the goal to minimize areas of microgap as on the machine taper of the two structures. A true conical connection is
these contribute to biological and mechanical failure. a Morse Taper connection that is press-fit together with significant
In an attempt to limit microgap and increase abutment stability, a friction existing between the two components that result in cold-
variety of connection designs have been manufactured. The first was welding. The Morse Taper angle is determined according to the me-
the Brånemark external connection. In this design the connection fea- chanical properties of each material (Macedo et al., 2016). Compressive
ture extends superior to the coronal portion of the implant. Because the forces may cause deeper settling of the abutment into the implant body
abutment sits on the external connection feature, it has a high center of of conical connections, minimizing the microgap and allowing the 2-
rotation relative to the implant and the screw is the only component piece system to behave as one piece (Tsuruta et al., 2018; Yao et al.,
securing the abutment. Common external connection shapes include 2019). The connection may form a hermetic seal that prevents micro-
hexagon, octagon, and spline, with external hexagon being the most bial invasion and anti-rotation through the cold-welded friction fit. It
common. Advantages to this design are that long-term follow-up data is has a high resistance to bending and rotational torque, which reduces
available and compatibility exists among multiple implant systems. the possibilities of screw loosening or screw fracture as compared to
External hexagon connections have been shown to have rotational other connections. This is because the surface area of the implant-
misfit in the range of 3 to 10°, while a rotational misfit of less than 2° is abutment interface is greater in conical connections as compared to
required to provide a stable screw joint that limits screw loosening parallel walled connections, contributing to reduced measurements of
(Binion, 1996a). Thus a disadvantage to the external hexagon connec- microleakage and screw loosening (Tsuruta et al., 2018). As compared
tion includes a higher prevalence of screw loosening, which ranges to external and other internal connections, microleakage has been
from 6% to 48% (Levine et al., 1997; Pardal-Peláez and Montero, shown to be less in Morse Taper (conical connection) geometries,
2017), a higher prevalence of rotational misfit, a less esthetic result, providing better stabilization, marginal bone loss, maintenance of soft
and often, an inadequate microbial seal. Design modifications to im- tissue anatomical dimensions, and esthetics (Macedo et al., 2016). It
prove abutment stability and fit and to limit rotational misfit have been can also support platform switching.
applied to the height and width of the external hexagon. The external In a true conical connection, the fit relates to the taper degree and
hexagon currently ranges from 0.7 to 1.2 mm in height and from 2.0 to connection area. The anti-torsional ability is dependent on the frictional
3.4 mm in width. Deepening the abutment screw engagement limits the resistance in the cone being greater than the torsion movement (Yao
tipping forces on the abutment and reduces the incidence of screw et al., 2015; Yao et al., 2019). Given high occlusal forces, this is often
loosening. Other modifications include machining a 1.5° taper to the difficult to maintain clinically, so a hybrid connection is often designed
hex flats, which creates a friction fit between the abutment and implant. to incorporate self-locking mechanisms that can mitigate abutment
Micro-stops have also been added in the corners of the abutment rotation and micromovements that can compromise restorative success.
hexagon to engage the implant hexagon and limit misfit. Modifications A disadvantage of conical connections is the potential for vertical dis-
have also been made to the abutment screw material and design that crepancy, which ranges from 22.6 to 62.2 μm, and results when the
cover the shank, number of threads, diameter, length, thread design abutment is not fully seated on the implant (Tsuge et al., 2008). This
and torque applications (Binion, 2000). Studies have shown the mean mismatch will contribute to reduced engagement of the anti-rotational
preload using a gold alloy screw to be greater than that of a titanium feature, reduced engagement of the anti-bending feature and loss of
alloy screw; a greater preload will minimize the incidence of screw abutment stability.
loosening (Binion, 2000). Masticatory forces when chewing cause micromotion between the
Internal connections have the connection feature inferior to the abutment and implant because the microgaps exist. The size range of
coronal portion of the implant, which is inside the implant body. The micromotion is from 1.52 μm to 94.00 μm (Karl and Taylor, 2014). This
connection may be a butt joint (90°, flat-on-flat), a slip-fit or passive movement causes the material to undergo fretting or material wear due
joint, where a gap exists along the implant-abutment junction, or a to small amplitude oscillatory motion, which has been shown to enlarge
friction-fit or active joint, where there is physical contact between the the microgap and subsequently increase micromotion (Yao et al.,
components (Sarfaraz et al., 2015). The internal connections exist in a 2019). Thus, the implant-abutment systems require adequate mechan-
wide variety of shapes including, but not limited to, hexagon, octagon, ical characteristics to resist functional occlusal loads. The present study
spline, cone screw, cylinder hex, tri-channel, and cam tube. Internal is the first to systematically evaluate abutment micromotion in six
connections have several advantages over external hexagon connec- commercially available internal connections. The dynamics of chewing
tions. Placing the connection internal in the implant lowers the center can cause abutment micromovements in vertical, horizontal, or rota-
of rotation and provides greater abutment stability by resisting lateral tional directions. Testing methods applied forces to the implant-as-
loads. Overall, internal connections have less screw loosening, provide sembly in vertical, horizontal, or rotational directions and quantified
better esthetics, provide an improved microbial seal, provide better the amount of misfit. The null hypothesis was that the various internal
joint strength and provide more platform switching options as com- connections would show no significant difference with regards to
pared to external connections. Porter, Lazzara, and Gardner introduced measurements in response to the application of the vertical, horizontal,
the concept of platform switching, by which a larger-diameter implant and rotational forces.
is combined with a narrower abutment, resulting in movement of the
implant-abutment gap away from the implant shoulder (Salamanca 2. Methods
et al., 2017). As a result, there is a shift in the loading forces from the
implant neck to the center of the abutment interface and blocking of 2.1. Implants and abutments
inflammatory cell infiltration (Hsu et al., 2018). The clinical benefits of
platform switching include less marginal bone resorption in platform- Table 1 lists the implant/abutment combinations evaluated in this

93
M.D. Kofron, et al. Clinical Biomechanics 65 (2019) 92–99

Table 1
Experimental groups.
Connection Key features Description Manufacturer, location

Friction-Fit • Slip-fit conical portion 4.1mmD Tapered Screw-Vent® implant + Zimmer Biomet, United
• 1°Antaper in the hex portion provides the friction-fit Straight abutment States of America
CrossFit® • Theinternal cone of 15° provides flexibility in prosthetic treatments 4.1mmD Roxolid® SLActive® implant + RC Straumann, Switzerland
• locking
15° conical-cylindrical connection incorporates a mechanically
friction fit with four internal grooves to provide stability under
Anatomic abutment

loading
Conical + Hex #1 • 1.5mm hex flats provide anti-rotation 3.8mmD Tapered Internal Laser Lok Biohorizons®, United States
• Connection of the implant and abutment relies on the screw, which
utilizes Spiralock® technology
implant + Straight Esthetic abutment of America

Conical + 6 Indexing
Slots
• 5highest
symmetrically placed slots and one additional slot located towards the
point on the implant collar allows for one-position-only
4.2mmD OsseoSpeed® EV implant +
TiDesign™ EV abutment
Astra Tech, Sweden

placement of all indexed components and index-free placement of one-

piece abutments
• Slip-fit conical portion
• Connection of the implant and abutment relies on the screw
Conical + Hex #2 • The hexagon provides anti-rotation and allows for prosthetic indexing
every 60°
4.3mmD NobelActive® Internal RP implant
+ Esthetic Abutment Conical Connection
Nobel Biocare®, Switzerland

• The hex flats provide anti-rotation RP

Internal Tri-Channel • Limited paths of insertion make it easy to restore 4.0mmD NobelReplace® Straight Groovy Nobel Biocare®, Switzerland
• Internal connection with indexing every 120° implant + Snappy abutment
• Connection of the implant and abutment relies on the screw

Note: All trademarks and product names are the property of their respective owners.

study and includes generic connection terms, key features, trade names, previously heated low melting point metal reservoir. A horizontal de-
and manufacturers. Each force application test (detailed below) in- gree scale was fabricated based on Pythagorean Theorem and the
volved the use of five independent implant and abutment replicates equipment being placed at a fixed distance from the scale. Once cured,
(n = 5). a laser pointer positioned under the construct allowed the assembly to
be centered at 0° (Fig. 2a). To measure relative angular movement, the
2.2. Internal connection images abutment was gently rotated in the clockwise direction and released.
The resultant angle was recorded. The process was repeated by cen-
The internal connection of each implant was imaged using scanning tering the assembly at 0° and gently rotating the abutment in the
electron microscopy (Model: 6440; JOEL USA, Inc., Pleasanton, counter clockwise direction. The resultant angular measurement was
California, USA) at 20 keV to view the characteristic internal features of recorded.
each connection (Fig. 1a–f). The Friction-Fit internal connection is ma-
chined to provide a slip-fit in the conical portion of the coupling and a 2.4. Abutment Pull Force (APF)
1° of taper in the hex portion to provide a friction-fit (Fig. 1a). The lead-
in bevel connection reduces horizontal stresses better than flat “butt- APF (n = 5) measured the force required to separate the abutment
joint” connections (Binion, 1996b; Karl and Taylor, 2014). The CrossFit and implant (Hsu et al., 2018). Briefly, the abutment was placed on the
connection has a 15° conical-cylindrical connection that incorporates a implant and the screw was torqued to the manufacturer's recommended
mechanically locking friction fit with four internal grooves to provide value using a digital torque gauge. The abutment screw was reverse
stability under loading (Fig. 1b). The Conical + Hex #1 has 1.5 mm hex torqued and removed. Next, the implant was inverted. Abutment-im-
flats (Fig. 1c). The connection relies on standard bolt technology, in plant connections with no retention features between the two compo-
which the implant and abutment are connected via a screw. Referred to nents had immediate separation. These abutments completely rely on
as Spiralock® technology, the abutment screw must be tightened to the the previously removed screw for abutment retention and the pull force
appropriate torque to increase the strength of the connection and re- was recorded as zero Newtons. Abutment-implant constructs that re-
duce the potential for abutment screw loosening. The Conical + 6 In- mained connected had the implant secured upside down in a collet. The
dexing Slots has five symmetrically placed slots and one additional slot retained abutment was lowered into a previously heated low melting
located towards the highest point on the implant collar (Fig. 1d). This point metal reservoir. Once cured, the force required to separate the
connection allows for one-position-only placement of all indexed implant and abutment was recorded using an Instron materials testing
components and index-free placement of one-piece abutments. The system (Model: E3000; Instron, Norwood, Massachusetts, USA)
Conical + Hex #2 is an internal conical connection with a hexagon (Fig. 2b) at a crosshead speed of 5.0 mm/min.
portion that allows for prosthetic indexing every 60° (Fig. 1e). The In-
ternal Tri-Channel is an internal connection that allows for prosthetic 2.5. Abutment MicroMotion (AMM)
indexing every 120° (Fig. 1f).
While under cyclic loading, AMM (n = 5) measured the length of
2.3. Abutment Angular Rotation (AAR) abutment micromotion, at the base of the abutment, relative to the
stationary implant (Karl and Taylor, 2014). Briefly, implants were
AAR (n = 5) was measured as the absolute difference between in- compression mounted in phenolic resin (Buehler, Lake Bluff, Illinois,
duced clockwise and counterclockwise abutment movements (Binion, USA) at bone level simulating 0 mm of clinical crestal bone resorption.
1996b; Nicholls and Basten, 2007). Briefly, the abutment was placed on Each abutment was milled to 4 mm height using a Bridgeport Milling
the implant and the screw was torqued to the manufacturer's re- machine (Model: F537A; Hardinge Inc., Elmira, New York, USA). This
commended value using a digital torque gauge (Model BGI Mark-10; was the minimal cutting height that prevented interference from the
Wagner Instruments, Copiague, New York, USA). The abutment screw abutment screw during testing. Standardization of the abutment height
was reverse torqued and removed. Next, the implant system was in- normalized the abutment bending moment under load and minimized
verted and secured in a collet. The abutment was lowered into a deflection of the post. The corresponding abutment was placed on the

94
M.D. Kofron, et al. Clinical Biomechanics 65 (2019) 92–99

Fig. 1. Internal conical connections evaluation. Scanning electron micrographis were obtained of each internal conical connection. [a] The Friction-Fit internal
connection is machined to provide a friction-fit in the hexagon portion and a slip fit in the conical portion of the coupling. The abutment hex flats are machined with
1° of taper to create the friction-fit to the walls of the implant. [b] The CrossFit connection has four internal grooves to position the prosthetic and a slip fit conical
portion that has an internal cone of 15°. [c] The Conical + Hex #1 has 1.5 mm hex flats. [d] The Conical + 6 Indexing Slots has a slip fit conical portion. [e] The
Conical + Hex #2 has an internal conical connection with a hexagon portion that allows for prosthetic indexing every 60°. [f] The Internal Tri-Channel is an internal
connection that allows for prosthetic indexing every 120°.

implant and the screw was torqued to the manufacturer's recommended demonstrated the presence of a friction-fit provided an angular rota-
value using a digital torque gauge. Implant assemblies were subjected tional freedom of 0°. Hybrid conical-hexagonal connections with a slip
to a 200 N force at a 30° angle with respect to the implant axial axis for fit (Conical + Hex #1 and Conical + Hex #2) had a gap through the
1000 cycles using the Instron E3000 materials testing system (Zipprich implant-abutment junction that allowed for a significantly greater de-
et al., 2007). These testing parameters were selected as posterior oc- gree of rotation as compared to a similar connection that incorporated a
clusal forces range up to 200 N with peak forces acting at an angle of friction-fit in the hex portion of the connection. Without the abutment
30° to the implant axis [ISO 14801] (Yao et al., 2019). The micromotion screw, the Spiralock Technology of Conical + Hex #1 was unable to
at the base of the abutment was detected using a Laser Doppler Vib- engage. Slip fit and non-indexed connections demonstrated the greatest
rometer connected to a data acquisition system (Model: OMS LaserPoint angular rotational freedom, ranging from 2° to 5°. The square CrossFit
LP01; Optical Measurement Systems, Laguna Hills, California, USA) connection demonstrated no significant difference as compared to the
(Fig. 2c and d). Friction-Fit connection.

2.6. Statistical analysis 3.2. Abutment Pull Force

Data are presented as mean (SD). Statistical analysis used a one-way The tighter the fit between the abutment and implant, the more
Analysis of Variance to determine the presence of statistical significance force is needed to pull the abutment off of the implant. APF (Fig. 4)
(α = 0.05) and a Tukey's post-hoc test to determine which groups were demonstrated the more static friction created in the connection during
significantly different. the application of abutment screw torque necessitated the strongest
forces to separate the abutment from the implant. The force required to
3. Results separate the abutment from the implant was significantly greater for
the Friction-Fit connection as compared to all other connection designs.
Quantitative testing of the various internal connections provides an When inverted, slip fit abutment connections (Conical + Hex #1, Con-
indication of the specific design features that may contribute to a suc- ical + 6 Indexing Slots, and Internal Tri-Channel) separated from the
cessful outcome following clinical implant placement. implant unaided, demonstrating the abutment screw is the only feature
that connects the abutment to the implant in these systems.
3.1. Abutment Angular Rotation
3.3. Abutment Micromotion
AAR measurements are used to show the degrees of abutment ro-
tational freedom. A lower measurement indicates a resistance to rota- AMM measured lateral abutment movement under dynamic
tion and greater abutment stability. AAR measurements (Fig. 3) loading. The Friction-Fit and CrossFit connections demonstrated

95
M.D. Kofron, et al. Clinical Biomechanics 65 (2019) 92–99

Fig. 2. Equipment descriptions. [a] Abutment Angular Rotation (AAR) equipment was used to measure the degrees of clockwise and counterclockwise rotation the
abutment can withstand when seated on the implant. An Instron E3000 materials testing system was used to measure [b] Abutment Pull Force (APF) and [c]
Abutment Micromotion (AMM). [d] For AMM measurements, the force was applied to an abutment at a 30° angle to the implant axial axis, while the laser detected
abutment motion.

significantly less horizontal micromotion as compared to the effect of loading on implant systems, internal connection implant sys-
Conical + 6 Indexing Slots and Conical + Hex #2 connections (Fig. 5). tems were exposed to rotational, vertical, and horizontal forces.
The Internal Tri-Channel connection demonstrated significantly less Internal conical connections have improved dissipation of loads and
micromotion as compared to the Conical + 6 Indexing Slots connection. reduced risk of unscrewing and/or fracture of the abutment screw as
compared to external or other internal connections (Caricasulo et al.,
4. Discussion 2018). Clinically during occlusal loading, abutment rotation creates
tension in the screw connection, which contributes to screw loosening
Prior work has demonstrated that greater dimensional variation in and screw joint failure (Binion, 1996c; Zipprich et al., 2007). Internal
the machining tolerances during manufacturing produces a more connections have less rotational freedom than external connections
asymmetric connection (Binion, 1996b; Nicoll et al., 2013). Machining (Siadat et al., 2017). The CrossFit connection demonstrated rotational
tolerances in implant dentistry have been reported in the literature to freedom ranging from 0° to 1.25°, which corroborates previously re-
range from 1 to 120 μm (do Nascimento et al., 2012; Ma et al., 1997; ported rotational measurements, acquired without the abutment screw,
Nicoll et al., 2013), representing a wide variability that influences the of mean 1.21° (SD 0.236°) (Gigandet et al., 2014). Gigandet et al. re-
fit of prosthetic components and the mechanical stability of the entire ported the greatest rotational misfit by the Internal Tri-Channel con-
implant system (Binion, 1996b; Nicoll et al., 2013). To understand the nection at mean 3.50° (SD 0.285°) (Gigandet et al., 2014). The results

96
M.D. Kofron, et al. Clinical Biomechanics 65 (2019) 92–99

manufacturer's recommended value. The abutment screw is removed

Fricon-Fit * †γ ψ prior to testing. APF results demonstrated the Friction-Fit connection
required a significantly greater force to separate the abutment from the
CrossFit® *† γ ψ
implant as compared to all other connections tested. The CrossFit and
Conical + Hex #1 *† Conical + Hex #2 measurements ranged from 0 to 66.6 N and from 0 to
Conical + 6 Indexing Slots *† 68.6N, respectively, which indicate an absence of a frictional force in
Conical + Hex #2 some, but not all, specimens. Karl et al. previously noted the occasional
Internal Tri-Channel failure for stabilizing features to engage (Karl and Taylor, 2014). The
Conical + Hex #1, Conical + 6 Indexing Slots, and Internal Tri-Channel
0.00 1.00 2.00 3.00 4.00 5.00 6.00 connections demonstrated 0 N was required to separate the implant and
abutment for each specimen tested. These particular connections, in
Degrees
mechanical terms, are considered non-self-locking (Zipprich et al.,
Fig. 3. Abutment Angular Rotation (ARR) measurements. Lower rotational 2007). The Friction-Fit connection APF ranged from 81.8 to 150.6 N.
measurements indicate a tighter fit between the abutment and implant. Each Thus, Friction-Fit was the only connection to consistently engage and
Friction-Fit connection had an angular rotational measurement of 0°. * indicates contribute to the abutment stability.
statistical significance as compared to Internal Tri-Channel. † indicates statistical AMM testing measured the horizontal abutment displacement under
significance as compared to Conical + Hex #2. γ indicates statistical sig- cyclic loading. The current results were in alignment with prior mi-
nificance as compared to Conical + 6 Indexing Slots, and ψ indicates statistical
cromotion testing that demonstrated the internal geometry of the
significance as compared to Conical + Hex #1. (α = 0.05).
connection influences the degree of micromotion (Karl and Taylor,
2014). The least amount of micromovement was measured in the
from the current study also demonstrated the greatest rotational Friction Fit and CrossFit connections. Zipprich et al. measured 4–8 μm
freedom from the Internal Tri-Channel connection at mean 5.10° (SD micromovement between the abutment and implant during dynamic
0.29°). Implant-abutment connections, fastened with the abutment loading of the Internal Tri-Channel connection (Zipprich et al., 2007).
screw, and having greater than 5° of rotational freedom, have been These results are in agreement with the micromotion measurement for
reported as extremely prone to screw loosening, which occurred be- the Internal Tri-Channel connection at mean 6.9 μm (SD 3.7 μm).
tween 1.1 and 2.5 million cycles of dynamic loading (Binion, 1996b). Measurements of micromotion indicate the presence of a microgap.
Implant-abutment connections, joined with the abutment screw, with Other studies evaluate the presence of microgaps at the implant-abut-
less than 2° of rotational freedom have the greatest resistance to screw ment junction using leakage tests. Leakage at the implant-abutment
loosening, completing an average of 6.7 million cycles of dynamic junction is dependent on the fit between the abutment and implant, the
loading (Binion, 1996b; Binion, 1996c). Connections with rotational degree of micromovement between the components, and the torque
freedom greater than 2° can result in vibration and micromovement force used to connect them. Leak teating reported in the literature in-
between the components during functional loading, which subse- dicates that 2-piece implant systems can contain a gap between the
quently decrease the clamping force until screw-joint failure occurs implant and abutment (Abdelhamed et al., 2015; Alves et al., 2016;
(Binion, 1996b; Binion, 1996c). Self-locking is the term used to define Cavusoglu et al., 2014; Dellow et al., 1997; D'Ercole et al., 2015;
resistance to rotation; the self-locking effect is caused by static friction Fernández et al., 2014; Quirynen and van Steenberghe, 1993; Scarano
created between the abutment and implant surfaces or machined ro- et al., 2016; Zanardi et al., 2012). The microgap provides a potential
tational locks (Zipprich et al., 2007). In the results presented here, the site for the accumulation of bacteria and their metabolites and allows
Friction-Fit was the only connection that provided complete resistance for the micropumping effect that allows for bacteria to colonize the
to angular rotation. This is attributed to the proprietary abutment hex internal aspect of the implant. This may lead to higher levels of in-
flat that is machined at a 1° taper to provide a tight friction-fit between flammatory cells around the soft tissue surrounding the implant-abut-
the implant and abutment. Binion previously reported the degree of ment interface, tissue redness, swelling, malodor, and bleeding (Alonso-
rotational misfit can be directly correlated to size discrepancies be- Pérez et al., 2017; Gross et al., 1999). Previous studies demonstrated
tween the flat-to-flat dimensions of the implant hexagonal and its cor- bacterial leakage can occur at the implant-abutment junction and along
responding abutment hexagonal recess (Binion, 1996b), which are re- the abutment screw channel, leading to malodor or peri-implant disease
lated to machining tolerances and quality control at the time of (Raab et al., 2017). do Nascimento et al. incubated specimens, of in-
manufacture. ternal-hex, external-hex, and Morse cone connections, for 7 days in
APF testing measured the vertical force required to overcome the human saliva (do Nascimento et al., 2012). The saliva samples con-
static frictional force established in the connection once the abutment tained a variety of bacteria and leakage was dependent on the size and
screw has been fully seated on the implant by torquing the screw to the shape of bacteria. For internal-hex and external-hex connections,

Fig. 4. Abutment Pull Force (APF) measurements. A tighter

fit between the abutment and implant are demonstrated by
use of a higher force to detach the abutment from the im-
plant. Statistical analysis demonstrated a significantly greater
abutment pull force was required to separate the abutment
from the implant for the Friction-Fit connection as compared
to all other experimental groups. (α = 0.05).

97
M.D. Kofron, et al. Clinical Biomechanics 65 (2019) 92–99

Fig. 5. Abutment Micromotion (AMM) measurements.

Displacement between the implant and abutment in vivo
creates microgaps that enable bacterial invasion.
Measurements quantified abutment micromotion under cyclic
loading. * indicates statistical significance as compared to
Friction-Fit; † indicates statistical significance as compared to
CrossFit; γ indicates significance as compared to Internal Tri-
Channel. (α = 0.05).

loaded implants showed higher bacterial counts and percentages of test results is limited and may not be indicative of clinical performance.
species occupying the internal aspect of the implant as compared to Abutment screw removal allowed for a more direct physical assessment
unloaded implants. The differences in connection geometry also influ- of the tightness and stability of the fit between the abutment and im-
enced leakage. Under loaded and unloaded conditions, the Morse cone plant. The angular rotation was indirectly applied to the abutment via a
connection demonstrated lower bacterial counts than the internal-hex manually operated lever. The lever was moved until a reasonable force
and external-hex connections (Caricasulo et al., 2018). could no longer advance the lever; however, despite all attempts to
Microgaps can cause prosthetic instability and increase the stress prevent this, the force applied to each abutment could have been
states in the surrounding bone and implant components (Coelho et al., variable. Incorporating a mechanistic method of control for component
2008). Increased stress and strain levels on the endosseous implant can rotation would be an improvement over the current system. Additional
overload the peri-implant bone and accelerate marginal bone loss. AAR testing with the abutment screw torqued and retained in position
Multiple clinical studies have demonstrated that greater levels clinical would likely reduce the amount of motion and thereby potentially
peri-implant bone loss have been measured with external connections, eliminate significant differences between the experimental groups.
while internal connections and conical connections have the lowest Lastly, limitations exist because universally accepted methods for
amounts of marginal bone loss (Caricasulo et al., 2018). testing abutment stability and fit do not exist. This makes it difficult to
Finite element analysis has supported the clinical observations of draw direct comparison between results achieved following different
crestal bone loss (Chun et al., 2006; Sarfaraz et al., 2015). Newer im- test methodologies throughout the published literature.
plant abutment connection designs have been commercialized to re-
duce stress concentrations to the marginal bone. The external connec- 5. Conclusions
tion implant-abutment interface is coronal to the implant and marginal
bone. Thus, stress generated is transferred and localized to the marginal Reliable and stable implant-abutment connections are required for
bone, inducing crestal bone loss. Internal connections have a more the clinical success of a dental implant system. Selection of an implant
apically oriented implant abutment interface, so stresses are more connection with self-locking features that will resist micromovement
broadly distributed and extend deeper into the implant and further due to applied forces is likely to provide better performance under
away from the marginal bone, thereby reducing the quantity of crestal biomechanical loading in the oral environment. Our results demon-
bone loss as compared to external connections (Chun et al., 2006). strated that connections incorporating anti-rotational features, such as
Conical connections with an active, friction-fit under loading conditions a friction-fit surface or machined micro-stops, performed the best in
had greater stress distributed on the implant and less stress distributed preventing abutment rotation, resisting separation, and resisting mi-
on the surrounding bone as compared to conical connections with a cromotion. Based on the results presented here, the Friction-Fit and
passive, slip-fit (Sarfaraz et al., 2015). Stresses on a friction-fit con- CrossFit connections performed best overall in limiting motion in the
nection are localized at the implant-abutment interface and abutment rotational, vertical, and horizontal, directions examined.
neck. The friction-fit creates a wedging effect that improves the im-
plant-abutment joint stability against lateral forces and distributes CRediT authorship contribution statement
stresses primarily on the implant. The friction-fit also provides a greater
resistance to deformation and fracture under oblique compressive Michelle D. Kofron: Conceptualization, Data curation, Formal
loading as compared to passive-fit conical connections. Ultimately, analysis, Investigation, Methodology, Project administration,
these features reduce biological and mechanical complications when Resources, Software, Validation, Visualization, Writing - original draft,
using conical friction-fit connections. As cold-welding does not occur in Writing - review & editing. Matthew Carstens: Conceptualization, Data
passive-fit connections, there are more likely gaps at the implant- curation, Investigation, Formal analysis, Methodology, Resources,
abutment interface which contribute to micromotions and interfere Software, Validation, Visualization, Writing - original draft, Writing -
with adequate stress distributions. Load transfer at the bone-implant review & editing. Cong Fu: Conceptualization, Data curation,
interface is a complex phenomenon. The implant geometry and design Investigation, Methodology, Resources, Software, Validation,
of the implant-abutment connection affect load transfer, which is also Visualization, Writing - review & editing. Hai Bo Wen:
affected by the loading protocol and the type of occlusion, the number Conceptualization, Investigation, Methodology, Project administration,
of implants and position, and the quality and quantity of the sur- Resources, Software, Writing - review & editing.
rounding bone (Sarfaraz et al., 2015).
AMM tested components were assembled according to the manu- Acknowledgements
facturer's instructions for use and subjected to cyclic loading to imitate
in vivo use. During study implementation, abutment screw removal The authors would like to thank Jim Reams for technical assistance.
prior to AAR and APF testing means the interpretation of pre-clinical The authors would also like to thank Elnaz Ajami, PhD and Prabhu

98
M.D. Kofron, et al. Clinical Biomechanics 65 (2019) 92–99

Gubbi, PhD for providing review of the manuscript. Levine, R.A., Clem III, D.S., Wilson Jr., T.G., Higginbottom, F., Saunders, S.L., 1997. A
Role of the funding source multicenter retrospective analysis of the ITI implant system used for single-tooth
replacements: preliminary results at 6 or more months of loading. Int. J. Oral
This research was supported by Biomet 3i, LLC, a Zimmer Biomet Maxillofac. Implants 12, 237–242.
company. Ma, T., Nicholls, J.I., Rubenstein, J.E., 1997. Tolerance measurements of various implant
components. Int. J. Oral Maxillofac. Implants 12, 371–375.
Macedo, J.P., Pereira, J., Vahey, B.R., Henriques, B., Benfatti, C.A., Magini, R.S., López-
Conflict of interest López, J., Souza, J.C., 2016. Morse taper dental implants and platform switching: the
new paradigm in oral implantology. Eur. J. Dent. 10 (1), 148–154.
All authors were employees of Biomet 3i, LLC, a Zimmer Biomet Mishra, S.K., Chowdhary, R., Kumari, S., 2017. Microleakage at the different implant
abutment interface: a systematic review. J. Clin. Diagn. Res. 11 (6), ZE10–ZE15.
company at the time this research was conducted and have received Nascimento, C.d., Pita, M.S., Santos, Ede S., Monesi, N., Pedrazzi, V., Albuquerque Junior,
remuneration and other compensation. R.F., Ribeiro, R.F., 2016. Microbiome of titanium and zirconia dental implants
abutments. Dent. Mater. 32 (1), 93–101.
Nicholls, J.I., Basten, C.H., 2007. A comparison of three mechanical properties of four
References
implant designs. Postgraduate Dent. 2 (1), 4–12.
Nicoll, R.J., Sun, A., Haney, S., Turkyilmaz, I., 2013. Precision of fit between implant
Abdelhamed, M.I., Galley, J.D., Bailey, M.T., Johnston, W.M., Holloway, J., McGlumphy, impression coping and implant replica pairs for three implant systems. J. Prosthet.
E., Leblebicioglu, B., 2015. A comparison of zirconia and titanium abutments for Dent. 109, 37–43.
microleakage. Clin. Implant. Dent. Relat. Res. 17 (S2), e643–e651. Pardal-Peláez, B., Montero, J., 2017. Preload loss of abutment screws after dynamic fa-
Alonso-Pérez, R., Bartolomé, J.F., Ferreiroa, A., Salido, M.P., Pradíes, G., 2017. tigue in single implant-supported restorations. A systemic review. J Clin Exp Dent 9
Evaluation of the mechanical behavior and marginal accuracy of stock and laser- (11), e1355–e1361.
sintered implant abutments. Int. J. Prosthodont. 30 (2), 136–138. Pokrowiecki, R., Mielczarek, A., Zaręba, T., Tyski, S., 2017. Oral microbiome and peri-
Alves, D.C.C., de Carvalho, P.S.P., Elias, C.N., Vedovatto, E., Martinez, E.F., 2016. In vitro implant diseases: where are we now? Ther. Clin. Risk Manag. 13, 1529–1542.
analysis of the microbiological sealing of tapered implants after mechanical cycling. Quirynen, M., van Steenberghe, D., 1993. Bacterial colonization of the internal part of
Clin. Oral Investig. 20 (9), 2437–2445. two-stage implants, an in vivo study. Clin. Oral Impl. Res. 4, 158–161.
Binion, P.P., 1996a. The spline implant: design, engineering, and evaluation. Int. J. Raab, P., Alamanos, C., Hahnel, S., Papavasileiou, D., Behr, M., Rosentritt, M., 2017.
Prosthodont. 9 (5), 419–433. Dental materials and their performance for the management of screw access channels
Binion, P.P., 1996b. The evolution and evaluation of two interference-fit implant inter- in implant-supported restorations. Dent. Mater. J. 36 (2), 123–128.
faces. Postgraduate Dent. 3 (1), 3–13. Sahin, C., Ayyildiz, S., 2014. Correlation between microleakage and screw loosening at
Binion, P.P., 1996c. The effect of implant/abutment hexagonal misfit on screw joint implant-abutment connection. J. Adv. Prosthodont. 6, 35–38.
stability. Int. J. Prosthodont. 9 (2), 149–160. Salamanca, E., Lin, J.C., Tsai, C.Y., Hsu, Y.S., Huang, H.M., Teng, N.C., Wang, P.D., Feng,
Binion, P.P., 2000. The external hexagonal interface and screw-joint stability: a primer on S.W., Chen, M.S., Chang, W.J., 2017. Dental implant surrounding marginal bone level
threaded fasteners in implant dentistry. Quintessence of Dent. Technol. 23, 91–105. evaluation: platform switching versus platform matching-one-year retrospective
Caricasulo, R., Malchiodi, L., Ghensi, P., Fantozzi, G., Cucchi, A., 2018. The influence of study. Biomed. Res. Int. 2017, 7191534.
implant-abutment connection to peri-implant bone loss: a systematic review and Salvi, G.E., Cosgarea, R., Sculean, A., 2017. Prevalence and mechanisms of peri-implant
meta-analysis. Clin. Implant. Dent. Relat. Res. 20 (4), 653–664. diseases. J. Dent. Res. 96 (1), 31–37.
Cavusoglu, Y., Akça, K., Gürbüz, R., Cehreli, M.C., 2014. A pilot study of joint stability at Sarfaraz, H., Paulose, A., Shenoy, K.K., Hussain, A., 2015. A three-dimensional finite
the zirconium or titanium abutment/titanium implant interface. Int. J. Oral element analysis of a passive and friction fit implant abutment interface and the
Maxillofac. Implants 29 (2), 338–343. influence of occlusal table dimension on the stress distribution pattern on the implant
Chun, H.J., Shin, H.S., Han, C.H., Lee, S.H., 2006. Influence of implant abutment type on and surrounding bone. J. Indian Prosthodont. Soc. 15 (3), 229–236.
stress distribution in bone under various loading conditions using finite element Scarano, A., Perrotti, V., Piattelli, A., Iaculli, F., Iezzi, G., 2015. Sealing capability of
analysis. Int. J. Oral Maxillofac. Implants 21, 195–202. implant-abutment junction under cyclic loading: a toluidine blue in vitro study. J.
Coelho, P.G., Sudack, P., Suzuki, M., Kurtz, K.S., Romanos, G.E., Silva, N.R., 2008. In vitro Appl. Biomater. Funct. Mater. 13 (3), e293–e295.
evaluation of the implant abutment connection sealing capability of different implant Scarano, A., Valbonetti, L., Degidi, M., Pecci, R., Piattelli, A., de Oliveira, P.S., Perrotti, V.,
systems. J. Oral Rehabil. 35 (12), 917–924. 2016. Implant-abutment contact surfaces and microgap measurements of different
Dellow, A.G., Driessen, C.H., Nel, H.J., 1997. Scanning electron microscopy evaluation of implant connections under 3-dimensional X-ray microtomography. Implant. Dent. 25
the interfacial fit of interchanged components of four dental implant systems. Int. J. (5), 656–662.
Prosthodont. 10 (3), 216–221. Siadat, H., Beyabanaki, E., Mousavi, N., Alikhasi, M., 2017. Comparison of fit accuracy
D'Ercole, S., Tripodi, D., Marzo, G., Bernardi, S., Continenza, M.A., Piattelli, A., Iaculli, F., and torque maintenance of zirconia and titanium abutments for internal tri-channel
Mummolo, S., 2015. Microleakage of bacteria in different implant-abutment assem- and external-hex implant connections. J. Adv. Prosthodont. 9 (4), 271–277.
blies: an in vitro study. J. Appl. Biomater. Funct. Mater. 13 (2), e174–e180. Steinebrunner, L., Wolfart, S., Bössmann, K., Kern, M., 2005. In vitro evaluation of bac-
do Nascimento, C., Barbosa, R.E.S., Issa, J.P.M., Watanabe, E., Ito, I.Y., Albuquerque Jr., terial leakage along the implant-abutment interface of different implant systems. Int.
R.F., 2009. Use of checkerboard DNA-DNA hybridization to evaluate the internal J. Oral Maxillofac. Implants 20 (6), 875–881.
contamination of dental implants and comparison of bacterial leakage with cast or Tallarico, M., Canullo, L., Caneva, M., Özcan, M., 2017. Microbial colonization at the
pre-machined abutments. Clin. Oral Impl. Res. 20, 571–577. implant-abutment interface and its possible influence on peri-implantitis: a sys-
do Nascimento, C., Miani, P.K., Pedrazzi, V., Gonçalves, R.B., Ribeiro, R.F., Faria, A.C., tematic review and meta-analysis. J. Prosthodont. Res. 61 (3), 233–241.
Macedo, A.P., de Albuquerque Jr., R.F., 2012. Leakage of saliva through the implant- Tripodi, D., D'Ercole, S., Iaculli, F., Piattelli, A., Perrotti, V., Iezzi, G., 2015. Degree of
abutment interface: in vitro evaluation of three different implant connections under bacterial microleakage at the implant-abutment junction in Cone Morse tapered
unloaded and loaded conditions. Int. J. Oral Maxillofac. Implants 27, 551–560. implants under loaded and unloaded conditions. J. Appl. Biomater. Funct. Mater. 13
El Haddad, E., Gianni, A.B., Mancini, G.E., Cura, F., Carinci, F., 2016. Implant-abutment (4), e367–e371.
leaking of replace conical connection Nobel Biocare® implant system. an in vitro Tsuge, T., Hagiwara, Y., Matsumura, H., 2008. Marginal fit and microgaps of implant-
study of the microbiological penetration from external environment to implant- abutment interface with internal anti-rotation configuration. Dent. Mater. J. 27 (1),
abutment space. Oral and Implantol. 9 (2), 76–82. 29–34.
Fernández, M., Delgado, L., Molmeneu, M., García, D., Rodriguez, D., 2014. Analysis of Tsuruta, K., Ayukawa, Y., Matsuzaki, T., Kihara, M., Koyano, K., 2018. The influence of
the misfit of dental implant-supported prostheses made with three manufacturing implant-abutment connection on the screw loosening and microleakage. Int. J.
processes. J. Prosthet. Dent. 111 (2), 116–123. Implant Dent. 4 (1), 11–16.
Gigandet, M., Bigolin, G., Faoro, F., Bürgin, W., Brägger, U., 2014. Implants with original Verdugo, C.L., Núñez, G.J., Avila, A.A., San Martín, C.L., 2014. Microleakage of the
and non-original abutment connections. Clin. Implant. Dent. Relat. Res. 16 (2), prosthetic abutment/implant interface with internal and external connection: in vitro
303–311. study. Clin. Oral Implants Res. 25 (9), 1078–1083.
Gross, M., Abramovich, I., Weiss, E.I., 1999. Microleakage at the abutment-implant in- Yao, K.T., Kao, H.C., Cheng, C.K., Fang, H.W., Huang, C.H., Hsu, M.L., 2015. The potential
terface of osseointegrated implants: a comparative study. Int. J. Oral Maxillofac. risk of conical implant-abutment connections: the antirotational ability of cowell
Implants 14 (1), 94–100. implant system. Clin. Implant. Dent. Relat. Res. 17, 1208–1216.
Hsu, P.F., Yao, K.T., Kao, H.C., Hsu, M.L., 2018. Effects of axial loading on the pull-out Yao, K.T., Kao, H.C., Cheng, C.K., Fang, H.W., Huang, C.H., Hsu, M.L., 2019. Mechanical
force of conical connection abutments in ankylose implant. Int. J. Oral Maxillofac. performance of conical implant-abutment connections under different cyclic loading
Implants 33 (4), 788–794. conditions. J. Mech. Behav. Biomed. Mater. 90, 426–432.
Jones, A.A., Cochran, D.L., 2006. Consequences of implant design. Dent. Clin. N. Amer. Zanardi, P.R., Costa, B., Stegun, R.C., Sesma, N., Mori, M., Laganá, D.C., 2012. Connecting
50, 339–360. accuracy of interchanged prosthetic abutments to different dental implants using
Kano, S.C., Binon, P.P., Curtis, D.A., 2007. A classification system to measure the implant- scanning electron microscopy. Braz. Dent. J. 23 (5), 502–507.
abutment microgap. Int. J. Oral Maxillofac. Implants 22 (6), 879–885. Zipprich, H., Weigl, P., Lange, B., Lauer, H.C., 2007. Micromovements at the implant-
Karl, M., Taylor, T.D., 2014. Parameters determining micromotion at the implant-abut- abutment interface: measurement, causes, and consequences. Implantologie. 15 (1),
ment interface. Int. J. Oral Maxillofac. Implants 29, 1338–1347. 31–46.
Koutouzis, T., Mesia, R., Calderon, N., Wong, F., Wallet, S., 2014. The effect of dynamic Zipprich, H., Miatke, S., Hmaidouch, R., Lauer, H.C., 2016. A new experimental design for
loading on bacterial colonization of the dental implant fixture-abutment interface: an bacterial microleakage investigation at the implant-abutment interface: an in vitro
in vitro study. J. Oral Implantol. 40 (4), 432–437. study. Int. J. Oral Maxillofac. Implants 31 (1), 37–44.

99