This study aimed to compare the primary and secondary stability, measured by resonance frequency analysis (RFA), in implants of different lengths installed in areas submitted to maxillary sinus lift. Correlation between RFA and implant insertion torque was also assessed. Twenty implants of 9 and 11 mm were inserted in areas submitted to maxillary sinus lift. The insertion torque was measured by the Bien Air motor. Osstell, through RFA, determined the implant stability quotient (ISQ) 2 times: the day of implant installation (T1) and 90 days after implant installation (T2). No differences were observed in the ISQ between T1 and T2 when the 20 implants were grouped, nor when the 9 mm implants were evaluated separately. In contrast, when the 11 mm values were evaluated separately, the ISQ was significantly higher in T2 than in T1 (P < .05). In T1, 9 mm implants had a higher ISQ than 11 mm ones (P < .05), whereas in T2, the implants of 11 mm showed a higher ISQ than did the 9 mm implants (P < .05). There was no difference in insertion torque between 9 and 11 mm implants (P > .05), nor was there a correlation between ISQ and insertion torque (P > .05). In conclusion, longer implants (11 mm) presented a significant increase in ISQ values during the healing period when installed in areas previously submitted to maxillary sinus lift. This phenomenon was not observed for shorter implants (9 mm). Finally, no correlation was observed between ISQ and insertion torque.
Introduction
Dental implants have become a predictable treatment modality that has been widely used in the rehabilitation of edentulous areas. Primary stability that is achieved shortly after implant installation is a prerequisite for the success of osseointegration, as well as one of the main factors that influence the survival rates of the implants.1,2 Primary stability consists of the frictional contact between the implant and the bone of the receptor site.3 Factors such as implant geometry, surgical technique, bone quantity, and bone quality influence primary stability.1,4 After implantation, a sequence of events in the osseointegration process occurs at the bone/implant interface, leading to a reduction in primary stability,3 which is gradually replaced by the secondary biological stability provided by deposition of newly formed bone around the implant body.5
The stability of the implant can be assessed clinically by noninvasive quantitative methods such as insertion torque6 and resonance frequency analysis (RFA).1 Described by Johansson and Strid,6 the insertion torque registers the torque required for implant installation. A high insertion torque indicates that the implant is well fixed and mechanically stable.6,7
RFA measurements are performed by Osstell and are expressed as implant stability quotient (ISQ), indicating a scale stability of 1 to 100.8,9 The technique allows the monitoring of changes in stability and stiffness at the implant/bone tissue interface, in any stages of osseointegration.10,11
Although RFA and insertion torque are associated with the assessment of implant stability, these techniques measure distinct mechanical characteristics. The insertion torque measures the frictional mechanical resistance generated at the bone/implant interface during the rotational movement of the implant on the longitudinal axis in the apical direction. The RFA, whose result is expressed in ISQ, represents the contact rigidity of the implant with the receptor bone bed and its resistance to lateral displacement.12
In this context, studies evaluating the possible correlation between insertion torque and RFA have been conducted, and these describe contradictory results as positive correlation12–14 or absence of correlation.15–18
Therefore, the present study aimed to compare the primary stability and secondary stability (measured by RFA and expressed in ISQ) of implants with the same geometrical characteristics but with two different lengths, installed in the posterior grafted maxilla that was submitted to maxillary sinus lift. The existence of a correlation between ISQ and insertion torque in the evaluation of the primary stability of the implants was also assessed.
The null hypotheses to be tested were: (1) there are no differences in the ISQ between the day of implant installation (T1) and 90 days after implant installation (T2), (2) there are no differences in the ISQ between the implants with different lengths, and (3) there are no correlation between ISQ and insertion torque.
Material and Methods
Study design
The study was approved by the Research Ethics Committee of the Pontifical Catholic University of Minas Gerais (CAAE 53955215.0.0000.5137).
This prospective clinical study was developed at the Implant Dentistry Clinic of the Graduate Program in Dentistry of the Pontifical Catholic University of Minas Gerais. The study evaluated 20 Neodent Titamax Ti cylindrical external hexagon implants, with 3.75-mm diameter and treated (acid and blasted) surface. Of the 20 evaluated implants, 12 units were 9 mm in length and 8 units were 11 mm in length. The implants were installed in 9 patients who previously underwent maxillary sinus lift grafting using Bio-Oss biomaterial alone. The choice between implants of 9 mm or 11 mm depended on the bone availability after grafting. Therefore, the implant distribution was not randomized by length. Nevertheless, this distribution was balanced (12 units of 9 mm and 8 units of 11 mm length). The patients showed residual bone height before grafting (residual alveolar ridge) ranging between 1 mm and 4 mm. The final graft height (height of newly formed bone) ranged from 10 mm to 12 mm. Therefore, the residual alveolar ridge may not be considered as a relevant variable that could influence the ISQ or the insertion torque since the majority of the implant was inserted into the newly formed bone. The grafts showed a density similar to bone type I and type II (evaluated by computed tomography).
Inclusion and exclusion criteria
Patients included in the study had previously submitted to maxillary sinus lift using Bio-Oss biomaterial after a 6-month follow-up period. The following exclusion criteria were adopted: (1) patients presenting any type of systemic disease that contraindicated surgical procedures at the time of installation of the dental implants, (2) smokers, (3) patients with active periodontal disease, and (4) patients showing a graft remnant that made impossible to install implants at least 9-mm long and 3.75-mm diameter (graft remnant after the repair period and evaluated by computed tomography).
Surgical procedure
Patients submitted to antisepsis with chlorhexidine gluconate (2%), an oral solution with chlorhexidine mouthwash (0.12%), and local anesthesia with 2% lidocaine and epinephrine (1/100 000). The surgical procedure was performed according to the protocol of the Neodent company (Curitiba, Brazil). With the use of the Bien Air motor (iChiropro, Bienne, Switzerland), we installed 20 Neodent Titamax Ti cylindrical external hexagon implants (3.75 mm diameter and treated—acid and blasted—surface and a length of 9 mm or 11 mm) in the posterior region of the maxilla previously submitted to elevation of the maxillary sinus using only Bio-Oss biomaterial. The prosthetic treatment planned for all patients involved overdentures applied after a 3-month healing period.
Measurement of insertion torque
The Bien Air electronic motor (iChiropro) was used to measure the insertion torque of the implants at the time of installation. This motor was designed to register the insertion torque (expressed in N·cm) at the same time as the implant installation. The Bien Air electronic motor was calibrated according to the manufacturer's instructions. The recording of the insertion torque was completed as soon as the implant reaches its final position and its rotation was interrupted by friction with the peri-implant bone tissue.
ISQ measurement
After the implants were installed, the evaluation of their primary stability was performed using the Osstell device. This apparatus, through the RFA, then determined the ISQ. For this measurement, a specific Smartpeg for the implant model used was adapted to the implant, and the device was brought close to it without touching it, following the manufacturer's recommendations. Three measurements were performed on each implant, separated by intervals of 5 seconds. The median of the triplicate measurements was then used as the ISQ value, avoiding possible bias due to an occasional measurement error. The measurements were made by the same operator at 2 points in time: T1 immediately after implant installation and T2 at 90 days after implant installation.
Statistical analysis
A D'Agostino-Pearson normality test showed that the data of the variable ISQ and insertion torque had normal distribution. Therefore, descriptive statistics were based on the calculation of mean and standard deviation while inferential statistics were performed by parametric tests.
The paired t-test was performed to assess the existence of differences in ISQ between T1 and T2. This analysis was performed separately: (1) for implants grouped independently of their length (9 mm or 11 mm), (2) for the 9 mm implants, and (3) for the 11 mm implants.
A Student t-test was performed to assess any differences in ISQ between the 9 mm implants and the 11mm implants. This analysis was performed separately on T1 and T2. A Student t-test was also performed to assess the existence of differences in T1 insertion torque between the 9 mm implants and the 11mm implants.
The Pearson correlation test was used to evaluate the existence of a correlation between ISQ and insertion torque in T1. This analysis was performed separately: (1) for implants grouped independently of their length (9 mm or 11 mm), (2) for the 9-mm implants, and (3) for the 11-mm implants.
The level of significance was set at 5%. Analyses were performed using GraphPad Prism 6.05 software (GraphPad Software, San Diego, Calif).
Results
In T1, when the implants were grouped (9 mm and 11 mm), the ISQ presented a mean of 62.55 and a standard deviation of 8.88. In T2, this variable had a mean of 62.60 and a standard deviation of 10.81. No statistically significant difference was observed in the ISQ between T1 and T2 (P > .05; Table, Figure a).
Mean and standard deviation of the implant stability quotient of implants grouped (9 mm and 11 mm), implants of 9 mm, and implants of 11 mm†

(a) Implant stability quotient (ISQ) in T1 and T2 in the 20 implants evaluated (9 mm and 11 mm implants). (b) The twelve 9 mm implants. (c) The eight 11 mm implants. (d) The insertion torque and its comparison between 9 mm and 11 mm implants. (e) Dispersion diagram of the correlation between ISQ and insertion torque in T1 of implants of 9 mm and 11 mm.
(a) Implant stability quotient (ISQ) in T1 and T2 in the 20 implants evaluated (9 mm and 11 mm implants). (b) The twelve 9 mm implants. (c) The eight 11 mm implants. (d) The insertion torque and its comparison between 9 mm and 11 mm implants. (e) Dispersion diagram of the correlation between ISQ and insertion torque in T1 of implants of 9 mm and 11 mm.
When the 9 mm implants were evaluated separately, the ISQ presented a mean of 65.83 in T1 and a standard deviation of 6.74. In T2, this variable had a mean of 58.67 and a standard deviation of 11.80. No statistically significant difference was observed in the ISQ between T1 and T2 (P > .05; Table, Figure b).
When the 11 mm implants were evaluated separately, the ISQ presented a mean of 57.63 in T1 and a standard deviation of 9.84. In T2, this variable had a mean of 68.50 and a standard deviation of 5.65. The ISQ was statistically higher in T2 than in T1 (P < .05; Table, Figure c).
When evaluating the existence of differences in ISQ between 9 mm and 11 mm implants in T1, the 9 mm implants had a statistically higher ISQ (65.83 ± 6.74) than the 11 mm implants did (57.63 ± 9.84; P < .05; Table). In T2, the 11 mm implants had a statistically higher ISQ (68.50 ± 5.65) than did the 9 mm implants (58.67 ± 11.80; P < .05; Table).
In the insertion torque evaluation, no statistically significant difference was observed between the 9 mm (28.93 ± 5.24) and the 11 mm (27.39 ± 4.40) implants (P > .05; Figure d).
Discussion
The purpose of the present study was to evaluate, through RFA and insertion torque, the stability of cylindrical implants installed in the posterior region of the maxilla in areas previously submitted to maxillary sinus lift using only Bio-Oss.
Mechanical tests and radiographic examinations, as well as histological and histomorphometric analyses, are used to evaluate osseointegration at the bone-implant interface. In particular, insertion torque and RFA are methods used for clinical evaluation of primary implant stability.19
In general, ISQ values are predictive of implant stability when evaluated repeatedly over a period of time.20 The measurement of ISQ at 2 points in time (T1 and T2) represents a way of comparing the primary stability and secondary stability of implants.21 In the present study, primary stability was measured at time of implantation (T1) and after 3 months (T2) to assess the possible impact of changes that occur around implants during osseointegration in a graft obtained using only Bio-Oss for sinus lift.
When all the implants were grouped and the 9 mm implants were evaluated separately, no statistically significant difference was observed in the ISQ between T1 and T2. However, when the 11 mm implants were evaluated separately, a statistically higher ISQ was found in T2.
Similar results were found by Rabel et al,22 whose study evaluated the stability of 2 implant systems and found no significant difference between the mean values of ISQ obtained in the immediate postoperative period and after 3 months of osseointegration when considering the total number of implants. In the same context, Rasmusson et al,23 in a prospective study, compared the stability of implants installed in grafted and ungrafted areas. Patients submitted to bilateral maxillary sinus lift with particulate bone showed an initial ISQ of 61.1 ± 5.5, similar to the present study, where a mean ISQ value of 62.55 ± 8.88 was obtained. In addition, as observed in our study, no statistically significant difference was found between the values of ISQ obtained in the first and second measurements (T1 and T2) when considering the total number of implants. Similar results were found by Antunes et al,24 conducting a study in dogs to evaluate the stability of implants installed in bone defects treated with Bio-Oss. These authors obtained a mean initial ISQ value of 66.22 and a final value (after 2 months) of 63.00.
The present study showed that the 9 mm implants had a statistically higher ISQ in T1 than did the 11 mm implants in the same period. This result should occur due to differences in bone density in implant installation areas, even though no differences were observed in insertion torque between 9 mm and 11 mm implants.
After a 3-month period (T2) the 9 mm implants did not reach the initial level of stability (T1). Instead, the 11 mm implants presented a statistically higher ISQ in T2 than did the 9 mm implants. In fact, only the 11 mm implants showed a statistically significant increase in the ISQ from T1 to T2. According to Olsson et al25 and Balshi et al,26 implants installed with high levels of primary stability do not reach the initial level of stability at a later second measurement (secondary stability). These authors also observed that implants with lower levels of primary stability tended to maintain or exceed the initial stability level at a later time. These data, as well as an increase in percentage of bone-implant contact (BIC) or others factors (not identified in the present study) could be the reason for the observed increase of ISQ in the 11 mm implants. Although the primary stability could also be influenced by residual bone height before grafting, in the present study, the patients showed small and similar residual alveolar bone height.
Since the ISQ was statistically higher in the 11 mm implants than in the 9 mm implants in T2 (3 months after implantation), our results suggest that implant length was a factor that influenced ISQ values during osseointegration, considering that 11 mm implants exhibited a statistically significant increase in the ISQ from T1 to T2 and the 9 mm implants did not show a statistically significant difference in ISQ when comparing T1 and T2. So, it could be inferred that 3 months after implantation, the 11 mm implants presented a higher percentage of BIC, justifying the higher ISQ. It is important to highlight that only cylindrical implants were used in the present study, avoiding influence on the stability by differences in the macrogeometry of the implants. Nevertheless, Sim and Lang27 conducted a longitudinal evaluation of the factors that influence RFA in Straumann implants of 8 and 10 mm in length showed, during osseointegration, a continuous increase in ISQ values in the 8 mm implants and no significant alteration of ISQ in the 10 mm implants—the opposite results of those observed in the present study.
In the present study, insertion torque values were recorded during implant installation, with no difference in insertion torque between 9 mm implants and 11 mm implants. These values were also used to evaluate the existence of a correlation between ISQ and insertion torque. No correlation was observed between ISQ and insertion torque at T1 for the implants evaluated, regardless of whether they were grouped (9 mm and 11 mm implants; 9 mm implants; 11 mm implants).
Similar results, showing no association between ISQ and insertion torque, were reported in previous studies.15,16,28,29 Friberg et al,15 when comparing insertion torque and ISQ measured in implants installed in the maxilla, found a significant correlation only in the cervical third of the implants. However, its final results also showed no general correlation between insertion torque and ISQ. Da Cunha et al,16 using two systems that were implanted in the maxilla and submitted to immediate loading, also did not observe a correlation between the values of insertion torque (obtained by OsseCare motor) and ISQ (measured by Osstell). Stubinger et al,28 in a study carried out on an animal model, found no correlation between the measures of torque (insertion and removal) and ISQ values. Dagher et al29 evaluated the existence of a correlation between RFA, insertion torque, and BIC in implants with 4 different surfaces (SLA, SLActive, TiUnite, and Euroteknika). A significant positive correlation between RFA and insertion torque was found only in the SLA group. Regardless of the type of implant surface, there was no correlation between RFA and BIC or between insertion torque and BIC.
Previous reports tried to demonstrate a correlation between ISQ and insertion torque.11–13 However, it must be taken into account that RFA and insertion torque, when used to assess the primary stability of implants, represent distinct characteristics: RFA is associated with resistance to bending stress, which resembles the direction of the clinical load; insertion torque represents resistance to shear forces.9
Some limitations of the present study must be indicated. They include the use of only one type of implant system, the use of one type of biomaterial for the sinus graft, the limited sample size, and the absence of randomization. Nevertheless, the results of the present study are similar to those of previous findings.
Conclusions
In conclusion, the results showed that when installed in areas previously grafted with Bio-Oss for maxillary sinus lift, longer implants (11 mm) presented a significant increase in ISQ values during the healing period. This phenomenon was not observed for shorter implants (9 mm). Finally, no correlation was observed between ISQ and insertion torque.
Abbreviations
Acknowledgments
Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (Capes), Brazil; Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Brazil; Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG), Brazil. Prof MCR Horta is a research fellow of FAPEMIG (CDS-PPM-00653-16).
Note
The authors declare no conflict of interest.