This study evaluated the effect of adding serration to the abutment-implant connection on torque maintenance before and after loading. Two implant systems with the same dimensions and connection design (internal 8° Morse taper octagon) were selected: one with nonserrated abutments (Simple line II) and the other one with serrated abutments (F & B). The removal torque value (RTV) was measured in 2 groups for each system: one group with one-piece abutments and the other group with 2-piece abutments, before and after cyclic loading (n = 10 in each group). The initial RTV of the abutment screw was measured with a digital torque meter. Each abutment received a cement-retained metal crown with 30° occlusal surface. Cyclic axial peak load of 75 ± 5 N was applied to the implants for 500 000 cycles at 1 Hz. The post-load RTV was then measured. Two-way and repeated-measures analysis of variance (ANOVA), and independent t test were applied to assess the effects of cyclic loading, connection design, abutment type, and their interaction on the percentage of torque loss (α = .05). Two-way ANOVA showed that serration of mating surfaces had a significant effect on torque maintenance before (P < .001) and after (P = .004) cyclic loading. Repeated-measures ANOVA also showed that loading had a significant effect on the torque loss percentage (P < .01). Comparison of the groups with t test showed that the torque loss of the serrated groups was lower than that of non- serrated groups. Despite the limitations of this study, the stability of the implant-abutment connection in the serrated design was higher than that of non-serrated group.
Introduction
Titanium dental implants are widely used because of their benefits such as optimal mechanical properties and great anchorage in the jawbone. The suitable choice of a particular type of a prosthetic restoration (cement-retained or screw-retained), and an appropriate implant-abutment connection system (external hexagon, internal hexagon, or internal taper) are critical for successful outcomes.1,2 However, mechanical complications such as fracture or mobility of the superstructure or abutment screw have been frequently reported.3,4 These complications may occur due to several reasons, such as the errors accumulated during the multiple steps of prosthesis fabrication, characteristics of the materials, or the surface irregularities of the mating surfaces of implant, abutment, and screw.5–9 Factors such as the connection design, screw design, screw settling, adequate preload by proper tightening torque, precision of mating implant components, and implant diameter are also important in screw joint stability.10–16 Recent studies have reported that the positional stability of the implant-abutment connection is a determinant factor for prosthesis misfit and mechanical complications.17–19
A number of studies have compared the efficacy of different connection mechanisms securing the implant-abutment connection stability, which determines the mechanical integrity, strength and stability of the assembly.9,20–23 Currently, there are over 20 different commercially available geometrical variations of implant-abutment interface, and all manufacturers claim that their designs greatly contribute to implant connection stability.24,25 Most current designs have been derived from 2 basic designs namely the “butt-joint” with an external connection, and internal “cone-in-cone” connection. According to some recent studies, the latter design shows higher resistance against the bending forces applied to the implant-abutment connection and minimizes the risk of screw loosening and fracture that typically occur in use of external butt-joint connections.1,26–28 On the other hand, some studies claim that the abutment screw characteristics, such as the screw material or surface, are the main determinant factors in this respect, and neither the internal nor the external design had any important effect on implant-abutment connection.29,30 An abutment with internal taper connection may have directly machined threads on the abutment body (1-piece abutment), or it may be fixed to the implant by a separate screw (2-piece abutment).1,2 The preference for one type of the abutment over the other (1- or 2-piece) depends on different factors such as the clinical and laboratory procedures, and the location of implant in dental arch (eg, esthetic zone).1,2 Type of abutment also affects the stress distribution, which can be related to the number of components the abutment is made of.31,32
The torque applied to a new screw causes elongation, and the subsequent elastic recovery results in generation of a compressive clamping force between the screw head and the abutment, and also between the screw threads and the internal implant threads.33,34 The contact force that clamps the implant and abutment together is called the preload.33–35 Within the elastic limits, a higher preload increases the resistance to loosening.34 Direct measurement of preload is preferable but not practical since it requires a strain gauge or load cell. However, for estimation of preload, indirect methods such as measuring the torque, angle of rotation of the screw head, or both can be used. Measuring the torque value is the most commonly used method for this purpose.36
The possibility of cold welding of the abutment inside the implant is a concern in internal taper systems, which could lead to clinical difficulties if the abutment needs to be retrieved.37,38 Sutter et al reported 10% to 15% higher removal torque values (RTVs) of abutment compared with the respective initial torque values.38 However, some other studies found that cold welding is probably counteracted by a phenomenon known as the settling effect (embedment relaxation),39–41 which works on the threads of the screw part of the abutment, and decreases the initial torque.1,5,18,42 Therefore, it has been shown that the RTVs of abutment screws immediately after screw tightening are frequently lower than the initial tightening torque values.30,38,39 Another possible cause is permanent/plastic deformation of the mating surfaces.26 During tightening of an abutment screw, the screw is damaged by friction, which may cause creeping, and decrease the tightening torque by 2%–10%.40
Dynamic loading in a chewing simulator is often used to simulate the clinical loading conditions.41,42 Realistic simulation of intraoral conditions could improve clinical relevance of the results.42,43 Generally, it is believed that screw loosening or torque loss would occur more after loading. Different studies have shown controversial results in this respect.26,40,42,44–47
Adding serration to the abutment connection is one strategy that has been claimed to enhance stability and better stress distribution at the implant-abutment connection, and subsequently decrease the rate of mechanical complications such as screw loosening (Figure 1a and b). The objective of this study was to evaluate the effects of cyclic loading on the RTV of internal taper abutments by comparing the percentage of torque loss of loaded and unloaded abutments. Two types of internal taper abutments of two implant systems, one system with serration feature on the implant-abutment connection and the other one with conventional mating surfaces were assessed. The null hypothesis of the study was that there would be no significant difference in RTVs of the 2 implant-abutment connection types.
(a) Adding the serration feature to the implant-abutment connection. (b) Conventional implant-abutment internal connection.
(a) Adding the serration feature to the implant-abutment connection. (b) Conventional implant-abutment internal connection.
Materials and Methods
Twenty implants of one implant system (4.8 × 10 mm; F&B, Busan, Korea) were divided into 2 groups. The first group (F1) included 1-piece abutments (F&B, SESA455) with serrated design (F&B 1-piece abutment, F1 group; Figure 2a) while the second group included 2-piece abutments (F&B, SEOA4305 O) with serration on the neck (F&B 2-piece abutment, F2 group; Figure 2b). Twenty implants of another implant system (Dentium, Simple line II, Seoul, Korea) with the same dimensions (4.8 × 10 mm) and connection design (internal 8° morse taper octagon) were selected and divided into 2 groups of 1-piece (D1) (SOSAB4855, Dentium 1-piece abutment; Figure 2c) and 2-piece (D2) abutments (SODAB4855O, Dentium 2-piece abutment; Figure 2d).
(a) F&B 1-piece abutment with serration on connection (F1). (b) F&B 2-piece abutment with serration on connection (F2). (c) Dentium 1-piece abutment (D1). (d) Dentium 2-piece abutment (D2).
(a) F&B 1-piece abutment with serration on connection (F1). (b) F&B 2-piece abutment with serration on connection (F2). (c) Dentium 1-piece abutment (D1). (d) Dentium 2-piece abutment (D2).
For the fabrication of standardized full-metal crowns for each abutment, the occlusal surface of one abutment of each system was waxed up with the slope of 30° to simulate the direction of oral forces applied to the cusp.18 Then, 38 duplicates were prepared by silicone material, and molten wax (blue inlay casting wax; GC Corp, Gyeonggi-do, South Korea) was used to fabricate the wax pattern of each duplicate. Each wax pattern was removed, and minor imperfections in the wax were corrected. Next, spruing, investing, burn out, and casting with a base metal alloy (Vera Bond V, Alba Dent, Fairfield, Calif) were performed.18,26 The implant-abutment assemblies were fixed in a special jig to avoid any rotation or movement. The abutments of each group were torqued to the value recommended by the manufacturer, which was 30 Ncm for D groups and 35 Ncm for F groups using an electronic torque meter (Lutron Enterprise, Coopersburg, Penn). According to the protocol suggested by Dixon et al,40 Breeding et al,48 and Deben et al,46 the abutment screw was retightened to the same torque 10 minutes later, to minimize embedment relaxation between the mating threads and help in achieving the optimal preload. Five minutes later, the reverse torque was measured using the same torque gauge and recorded.9,21 Next, the implant-abutment assemblies were mounted in acrylic resin (Acropars, Iran) vertically by a dental surveyor in the loading pot of the chewing stimulator. Then, all metal castings were cemented with temporary cement (Temp Bond, Kerr, Salerno, Italy) mixed with petroleum jelly in 50:50 ratios to ensure easy retrieval of the crowns. The full metal casting was seated on the abutment with finger pressure for 10 seconds followed by a sustained pressure of 5 kg for 10 minutes.1,9,21 The cement was allowed to set at room temperature. Upon setting and removing the excess cement, dynamic axial peak loads of 75 ± 5 N with a compressive sine wave were applied to the crowns for a duty of 500 000 cycles at 1 Hz. This loading regimen was applied by a chewing stimulator machine (S-D Mechatronic, Germany), and corresponded to 20 months of chewing in the clinical setting.1,9,21 After cyclic loading, the castings were removed and the RTV of each abutment was measured by an electronic torque meter. The reverse torque loss percentage values were measured based on the difference between the preload values of each system.
A power analysis was performed to determine the minimum sample size necessary to test the null hypothesis considering the different experimental groups. The Kolmogorov-Smirnov test was applied to verify the normality and homoscedasticity of the insertion torque and removal torque data.
Three factors were involved in this study: the abutment type (1- or 2-piece), the connection design (with or without serration), and cyclic loading. Therefore, the data were analyzed using 2-way ANOVA; paired t test was used to compare the insertion torque and the RTV in each group before and after cyclic loading. The effect of cyclic loading and its interaction effects with the implant system and abutment type were analyzed by repeated measures ANOVA. A probability level (P value) of less than .05 was considered statistically significant. All statistical analyses were performed using SPSS version 20.0. The methodology of the study was reviewed by an independent statistician.
Results
After dynamic loading, there was no screw loosening and no sign of mechanical failure in the specimens. During the experiment, loosening of cemented crowns was not noted by tactile or visual inspection, but wear of the occluding surfaces of restorations was detected. Also, following the RTV measurements, it was observed that the abutments did not have any sign of deformation or fracture. The measurements showed that the mean percentage of torque loss had a wide range of variation with a minimum value of 3.27% and maximum value of 38.3% (Table 1, Figure 3).
Torque loss percentage before and after cyclic loading in each group.
Two-way ANOVA showed that serration of implant-abutment mating surface had a significant effect on torque maintenance before (P < .001) and after (P = .004) cyclic loading (Table 2). Repeated-measures ANOVA also showed that loading had a significant effect on the percentage of torque loss (P < .01). Comparison of groups with t test showed that torque loss in F2 groups was lower than that in D2 groups, both before (P = .008) and after (P = 0.19) loading, and also was lower in F1 group compared with D1 before loading (P = .003; Table 3). The methodology and the results were reviewed by an independent statistician.
Two-way analysis of variance (ANOVA) results for the removal torque value before and after loading of experimental groups

Discussion
In this study, all abutments presented torque loss, indicating that cold welding did not occur at the abutment/implant connection. The effect of cyclic loading on implant-abutment connection, screw loosening, and the RTV has been reported in several studies.1,21,23,25,26,36,46–49 The present results demonstrated that the torque loss percentage was significantly higher after loading in comparison with the torque loss percentage before loading. This was in accordance with the findings of other studies.21,23,47 The reduction in RTV after loading conforms with the joint failure mechanisms explained by Bickford.50 The external forces progressively decrease the preload because of screw vibration, wear of mating surfaces, and settling effect. The greater the abutment micromovements, the more the torque loss would be.50 Although a high attrition force between the threads would prevent loosening, external loads may decrease the friction by compression of the screw head against the abutment base and result in screw loosening.19,23
Each implant system offers different abutment types for clinical use such as 1- and 2-piece abutments. However, the more recent provision of two-piece conical abutments for improved prosthetic flexibility raised a question regarding the stability of these joints, since the abutment screw is considered as the “weak link in the chain” and could affect the joint stability.20 Joint stability of 2-piece abutments, in comparison with 1-piece abutments, has been studied in different investigations.1,20,26 These results were in agreement with those of Norton20 who concluded that the strength of both groups was beyond the clinical loads applied to the implant-abutment connection.20 In another study, Cehreli et al1 reported that the fatigue resistance of both types of abutments was extremely high after dynamic loading and both could have convenient function with no clinical complications. However, the RTV of 1-piece abutments was significantly higher than that of 2-piece abutments, which was not in agreement with our results.
The mechanical behavior of implants is influenced by the design of implant-abutment mating and retentive properties of the screw joints. In the current study, adding serration to either 1- or 2-piece abutments caused a lower percentage of torque loss that indicates that the type of connection may affect the RTV and this result was in agreement with the findings of many other studies.2,24,27,37,51,52 Many studies have evaluated the effects of altering the mating surfaces or friction characteristics of implant-abutment connection.1,9,22–24,26,30,40,49 Application of dynamic loading without considering the thermal alterations cannot simulate the humidity of the oral environment; thus, not simulating the continuous thermal alterations of the oral environment was one limitation of this study.53,54 Another limitation of this study was the comparison of 2 systems with different recommended applied torque values. Despite these limitations, comparison of 2 different systems—serrated and non-serrated—with the same size of abutments and also evaluation of 1- and 2-piece abutments of each system together and conduction of dynamic loading for better simulation of clinical setting (compared with static loading) were the strength points of this study. Thus, this design might enhance joint stability and decrease mechanical complications such as screw loosening or fracture.
However, further studies are needed to evaluate the fatigue resistance of abutments with the serrated design. Finite element analyses are also required to evaluate stress distribution at the implant-abutment connection, compare screws made of different materials and also angled abutments or customized abutments of this design. For successful oral rehabilitation treatments, attempts should be made to improve the connection systems and increase the implant/prosthetic abutment interface stability to minimize the biological and mechanical complications.45 Thus, lower drop of torque loss percentage in serrated specimens demonstrated in this study may be helpful to increase the implant-abutment connection stability.
Conclusion
Within the limitations of this study, the following conclusions were drawn:
Torque loss occurred in all abutments before and after cyclic loading, indicating that cold welding did not occur.
Torque loss was greater for the abutment screws after dynamic loading, which highlights the role of mechanical loading in torque loss.
The type of implant system significantly influenced the torque loss before and after cyclic loading. Despite the limitations of this study, the stability of the implant-abutment connection of the serrated design was better than that of non-serrated group.
In both systems, the torque loss of 2-piece abutments was higher than that of 1-piece abutments; but this difference was not statistically significant.
The methodology and conclusion were reviewed by an independent statistician.
Acknowledgments
We would like to acknowledge Dr Mohammad Javad Kharazifard for reviewing the methodology and statistical analyses as an independent statistician. Dr Kharazifard is Research Advisor for the Dental Research Center at Tehran University of Medical Sciences, Tehran, Iran.
Note
There is no conflict of interests for any of the authors.