Validation of Resonance Frequency Analysis by Comparing Implant Stability Quotient Values With Histomorphometric Data

It is of interest to develop and evaluate assessment methods for implant stability, both for research purposes and for prognostic evaluation that led to the development of resonance frequency analysis (RFA) by Meredith in 1996.1,2 The use of RFA to assess implant stability has been validated by several authors.3–9 Despite widespread acceptance of RFA, as reflected by its extensive application in current research, its reliability is still debated.10–12 

The objective of this study was to validate the accuracy of RFA by measuring the implant stability quotient (ISQ) values with the Osstell instrument (Osstell AB, Gothenburg, Sweden) and correlating it with the implant-bone distance measured histomorphometrically.

Ten indigenously fabricated commercially pure titanium, single piece, root form, threaded implants with sandblasted surfaces were used (Figure 1). All implants were 4.7 mm in diameter at the level of the finish margin and 12 mm in length. An access channel for the SmartPeg (Osstell AB) of the Osstell instrument was machined in the implant. The implants are developed under the brand name “Idea.”

Implants were equally divided into 2 groups:

1. The tight fit or control group comprised implants placed in osteotomy sites prepared according to the manufacturer's recommendations to ensure primary stability.

2. The loose fit or test group comprised implants placed in oversized osteotomy sites in relation to the implant dimensions to artificially induce implant instability.

Five femurs were harvested from 3 goats immediately post mortem. Osteotomy preparations were commenced immediately following harvesting.

Ten osteotomy sites (5 test and 5 control) were identified in 5 femurs. The medial epiphyses of the femurs were selected as sites for implant placement because of the presence of adequate trabecular bone. Two sites (1 test and 1 control) were selected in each epiphysis. Radiovisiographs (RVGs) of the proposed sites were taken and analyzed to confirm the presence of adequate bone. Sites were randomly assigned to each group (Figure 2).

All osteotomies were prepared by the same operator using a drilling speed of 2500 rpm, under copious irrigation with normal saline.

Penetration of cortical bone was achieved with no. 6 round burs. A pilot drill 2.3 mm in diameter was used to prepare 12-mm deep osteotomies. The osteotomies were progressively enlarged using 2.8-mm and 3.4-mm diameter drills, until a depth of 12 mm was achieved. The osteotomies were widened to a depth of 8 mm from the crest using a 3.8-mm drill. The final size of the control osteotomies were 3.8 mm in diameter until a depth of 8 mm, and 3.3 mm in diameter from a depth of 8 mm to 12 mm.

Osteotomy preparations for the loose fit group were identical to those of the tight fit group until preparation with the 3.4-mm drill. Subsequently, 3.8 mm, 4.4 mm, and a final drill diameter of 4.8 mm were used to enlarge the osteotomies. The final size of the test osteotomies was 4.8 mm in diameter and 12 mm in depth.

Control implants were placed into designated osteotomies. The implants were rotated to the full depth of the osteotomies using a ratchet. Complete seating was visually verified by ensuring that the finish margins were at the level of the crest of the bone (Figure 3) and radiographically confirmed with a radiovisiograph (RVG) (Figure 4). Primary stability of the implants was verified by exerting lateral forces and confirming absence of motion.

Test implants were placed into osteotomies corresponding to the “loose fit” group. Implants were seated to their final positions in a manner identical to that described for the control implants. Lack of primary stability was verified by exerting lateral forces and noting motion.

The SmartPeg of the Ostell instrument was screwed into the access channels in the control implants according to the manufacturer's instructions. Three ISQ values were obtained from each of the following positions: 3, 6, 9, and 12 o'clock (Figure 5). The data were automatically stored in the device and subsequently transferred to a personal computer (Figure 6).

Measurement of ISQ values for the loose fit (test) implants was done in a manner identical to that used for the tight fit (control) implants, as described above.

The tight fit (control) and loose fit (test) implants with 5 mm of surrounding bone were removed en bloc using a diamond disc mounted in a straight handpiece under copious irrigation with tap water (Figure 7).

The specimens were fixed by immediate placement in 10% formalin solution, and dehydrated by immersion in 70% acetone, followed by 3 changes of absolute acetone, for durations of 2 hours, 90 minutes, 90 minutes, and 2 hours, respectively.

Resin embedding of the specimens in clear autopolymerizing resin was done by infiltration with methylmethacrylate monomer followed by inducing polymerization by adding polymer powder (Figure 8).

Specimens were sectioned using a diamond disc mounted in a straight handpiece at a speed of 1500 rpm, under copious irrigation with normal saline. All implant abutments were severed at the level of the finish margins. Embedded specimens of both groups were sectioned longitudinally through the centers of the implants into 2-mm thick slices (Figure 9).

Fifty micron–thick ground sections of the specimens were prepared by luting them to glass slides with cyanoacrylate followed by grinding with Emery paper of numbers 220, 280, 320, 400, and 600 sequentially under copious irrigation with normal saline. Adequate reduction of thickness was assessed by viewing the sections under a light microscope. Specimens of both groups were stained with hematoxylin and eosin (Figure 10).

Specimens of both groups were histomorphometrically analyzed using a light microscope. Implant-bone distance was measured using an ocular micrometer at ×40 and ×100 magnifications, and were expressed in microns. Measurements were made at standardized locations for all specimens, namely at the first two threads on the right and left sides of the implants. Two readings were made in each region and the data were statistically analyzed (Figures 11 through 14).

Relationship between distance and ISQ values was assessed by Pearson's correlation analysis. A P value of .05 was considered as the level of significance.

The highest and lowest mean ISQ values for the tight fit group were 68.16 and 58.91, respectively. The means of the largest and smallest implant-bone distance were 183 µm and 129 µm, respectively (Table 1). The correlation between the mean values of implant-bone distance and ISQ values for the loose fit (test) group are presented in Table 2.

Highest mean ISQ values of 68.16 corresponded to the lowest implant-bone distance with a value of 147 µm. Lowest ISQ values of 58.91 corresponded to the highest implant-bone distance with a value of 180 µm.

A negative relationship was found between the implant-bone distance and ISQ values. However, it was not statistically significant (r  =  −0.52, P  =  .19) (Table 3).

The highest and lowest mean ISQ values were 35.25 and 33.16, respectively (Table 2). The means of the largest and smallest gaps were 624 µm and 792 µm, respectively. Highest mean ISQ values of 35.25 corresponded to the least mean gap at the interface with a value of 624 µm. Least mean ISQ values of 33.16 corresponded to the largest mean gap at the interface, with a value of 792 µm.

A negative relationship was found between the implant-bone distance and ISQ values. However, it was not statistically significant (r  =  −0.65, P  =  .12) (Table 3).

While no gold standard exists to measure implant stability,10 several methods have been proposed for the same, namely clinical assessment by exerting lateral forces with 2 opposing mirror handles, percussion, Dental Fine Tester (Kyocera), Periotest device, pull-out and push-out tests, removal torque testing, and histologic and histomorphometric methods.1,4–6,13 The accuracy and practicality of these methods has been widely challenged in the literature.

One objective of this study was to quantify the stability of endosseous implants by measuring ISQ values using RFA. Undersized and oversized osteotomy preparations, in relation to implant dimensions were used to simulate high and low primary stability conditions, according to a previous study.6 The variable clinical stability of the implants was reflected by the resultant ISQ values, with higher values recorded for stable implants placed in undersized osteotomies and lower values recorded for mobile implants placed in oversized osteotomies. In a clinical situation, this would equate to higher values in case of clinically stable implants, and lower values in case of clinically mobile implants. These findings are in accordance with previous reports.6 

Mechanically, the resonance frequency (RF) of an object is strongly correlated to its boundary constraints.13 ISQ values reflect the stiffness of the implant-bone interface.13–15 The implant-bone interface was artificially designed to be stiffer in the “tight fit” or control group implants. The amount of surrounding bone in contact with the implants was greater as compared to the test group. These conditions were well reflected by the higher ISQ values for the control group, and are in accordance with findings of other studies.6–8 

Measurement of implant stability using RFA merely reflects on the implant-bone contact. It was hypothesized that ISQ values are related to the proximity of the bone to the implant and higher and lower ISQ values would correspond to lesser and greater implant-bone distances, respectively. The implant-bone distance was quantified by histomorphometric analysis of ground sections, which is a proven accurate method.7 

A negative correlation between ISQ values and the width of the interface within the tight fit and loose fit groups indicates that lower and higher ISQ values were found in cases of wider and narrower interfaces, respectively. In other words, ISQ values are inversely proportional to the implant-bone distance. These findings conform to the principles of basic vibrational theory, which states that “the RF value of a structure is strongly affected by its effective vibrational length and the tightness of the boundary contact.” The results are in accordance with other studies, where ISQ values were correlated to the density and cortical to the trabecular bone ratio at the interface.1,5 Lower ISQ values have been measured following implantation in low-density bone compared to bone of high density.6,14 This could be attributed to decreased stiffness of the interface in the former.1 Bone of low density, containing a large volume of trabecular bone, will manifest with larger gaps at the interface as compared with densely packed compact bone. Results of this experiment are in accordance with these facts.

A statistically significant correlation within and between the groups was not found. It could be speculated that this was due to the small sample size used in this study. Perhaps a significant correlation may be derived with further research with a larger sample size, which is not a realistic proposal when the constraints of animal experiments are considered.

Types of tissue reactions to implantation depend on the size of the gaps between the implant and bone cavity surface. The critical threshold of micromotion above which fibrous encapsulation results instead of osseointegration is not 0 but seen to range between 50 and 150 µm.16,17 The findings of this study indicate that high ISQ values showing clinically determined primary stability correspond to an interface of an average width of 158 µm. RFA presented higher ISQ values for deliberately stabilized implants. Results of this study indicate that RFA provides a good indication of clinical stability of an implant and the implant-bone distance.

The following conclusions were drawn from this study:

1. Stable implants present with higher ISQ values compared with mobile implants when subjected to RFA.

2. Higher ISQ values are indicative of lesser implant-bone distance and vice versa.

3. Clinical stability of implants is a function of the width of the implant-bone interface.

ISQ:

implant stability quotient

RF:

resonance frequency

RFA:

resonance frequency analysis

RVG:

radiovisiograph

We extend our gratitude to Dr B. Sivapathasundharam, Dr Saraswathi, and the entire team at the Department of Oral and Maxillofacial Pathology, Meenakshi Ammal Dental College and Hospital for their expert guidance and help.