Biomechanical and Microscopic Response of Bone to Titanium Implants in the Presence of Inorganic Grafts

After tooth extraction, an unpredictable remodeling of bone occurs, which may make ensuing prosthetic rehabilitation procedures difficult.1–5 To minimize losses in bone height and thickness, a graft material must be placed into the defect at the moment of surgery.1–5 

Autogenous bone is the most suitable and effective graft, but the scarcity and postoperative morbidity associated with its collection limits its use.2 Therefore, the search for bone substitutes that present similar levels of biocompatibility and osteointegration has been increasing.2,3 To date, inorganic xenogenic bone grafts from bovine species have shown promising results in this area due to their remarkable osteoconduction.2–4,6–8 The osteoconductive property of a biomaterial refers to its ability to serve as a scaffold on which osteoprogenitor cells can migrate, attach, proliferate, and undergo maturation into osteoblasts that produce bone matrix.9 

Initially, the grafted area becomes a combination of newly formed bone and biomaterial, which should ideally be in direct contact (ie, with no soft tissue interposed).2–4,6,7 As such, when a titanium implant is inserted in this region, the acceptable fibro-osseous integration that occurs in the first stages is not sufficient to provide implant stability in the long term.4 

The response of an implant to occlusal loads depends on the level of osteointegration achieved screw.10–14 The acceptability and thus the stability of titanium implants depend on the quality and quantity of newly formed adjacent bone and also on the extension of bone contact over the irregular surfaces of the screw.10–14 

To date, the microscopic responses of bone to xenogenic grafts2–4,6,7 and titanium implants10–12,15 have always been evaluated separately; thus, the aim of this study was to investigate the influence of a xenogenic graft on the microscopic and biomechanical responses of newly formed and mature bone to inserted titanium implants.

A total of 16 male New Zealand white rabbits weighing between 3.0 and 4.0 kg were used in this study. All experiments were carried out in accordance with the European Communities Council Directive of November 24, 1986 (86/609/EEC) regarding the care and use of animals for experimental procedures and were approved by the review board of the University of São Paulo (CEP 021/2005).

The animals were anesthetized (ketamine 50 mg/kg plus xylazine 5 mg/kg), and routine oral disinfection procedures were performed with chlorhexidine before the bilateral mandible incisors of each animal were surgically extracted. A 4-mm-diameter perforation in bone, communicating with the apical part of each alveolar socket, was created with a carbide 702 drill halfway between the incisors and lower molar teeth.

Afterward, a socket/hole defect chosen at random was filled with bovine cancellous inorganic bone graft (Gen-Ox, BAUMER SA, Mogi Mirim, SP, Brazil), whereas the contralateral defect in the same animal was left with only the blood clot, serving as a paired control. The surgical site was then sutured, and the remaining maxillary incisors were cut every 14 days.

Eight weeks after initial procedures, the bone areas of each rabbit's mandible were exposed again so that the procedures for implant placement could be performed. Under cooled sterile saline irrigation, a guide drill was first used to mark each implant location, which was then sequentially enlarged to 2 and 3 mm in diameter with twist drills prior to the insertion of a conventional titanium implant screw (3.75 × 8.5 mm; Conexão Implantes, Arujá, Brazil). The screw was hand ratcheted to bone until tight and then tapped so that the suture of gingival tissues could be made with a polyglactin 910 thread (Vycril, Ethicon, Cornelia, Ga).

Half of the animals were killed at 2 months and the other half at 6 months after implant placement (n  =  8). After sacrifice, mandibles were removed, split in half at the symphysis menti, and immersed in 10% neutral buffered formaldehyde.

Torque tests were performed with a LutronTQ8800 (Impact, Taiwan) torque gauge manometer. A counterclockwise movement was executed on the implant's external connection, and the values necessary to dislodge the implant were recorded each second during the entire process of pulling out the screw. The mean, in Newton-centimeters (N·cm), obtained for each implant was used for comparisons.

After implant retrieval, the specimens were decalcified, embedded in paraffin, reduced to 5-µm-thick longitudinal sections, and stained with hematoxylin and eosin for descriptive histological evaluation with a light microscope (binocular Carl-Zeiss, Germany). Afterward, each microscopic image was projected on a color monitor at ×10 original magnification using a CCD/RGB camera (Sony DXC151P, Wetzlar, Germany), so that histomorphometrical measurement of bone-implant contact (BIC) proportion at the final apical threads could be performed with software (KS300, Carl-Zeiss, Germany). Bone mass (BM) was evaluated at distances of 0 to 1 mm (area including the implant thread concavities) and 1 to 2 mm (area outside the implant threads).

Data were subjected to 2-factor analysis of variance (ANOVA), and the comparisons between groups and time intervals were made by means of Tukey's test adjusted to the 95% confidence interval. Statistics were calculated using SPSS software version 16.0 for Windows (SPSS, Chicago, Ill).

The ANOVA revealed no statistical differences in BIC, BM, or torque values between the experimental and control groups at each period or between the intervals studied (Table). Means, standard deviations, and the results of Tukey's test (α  =  .05) are presented in the Table.

According to the microscopic assessment of both groups, the quality of bone and the amount of osteocytes and osteoblasts adjacent to the implant surface after 2 months (Figure 1) were similar to those after 6 months (Figure 2). Farther from the implant interface, the xenogenic graft particles were in direct contact with bone, with no surrounding connective tissue (Figure 3).

After implant insertion, direct apposition of newly formed bone to the titanium surface (osteointegration) is expected; however, it is believed that this may not occur if a biomaterial impairs and limits the regular course of osteogenesis.16–18 This is the first study to apply a mechanical implant torque test to address this problem and to investigate simultaneously the microscopic response of rabbit's bone to a xenogenic graft material.

The resistance of titanium implants to removal may be positively correlated to the degree of contact between mineralized bone and irregularities at the implant surface.11 Thus, the force required to rotate an implant screw anticlockwise may be determined first by the degree of BIC and second by the biomechanical characteristics that the adjacent bone develops (quantity and quality) in the initial and later stages of osteogenesis.10–12,14 Since no difference in torque values was observed between control and experimental groups at 2 or 6 months, it can be assumed that, despite the presence of xenogenic graft particles, there was no significant interference in the quantity and quality of newly formed or mature bone contiguous to the implants.

Studying the substantial removal torque required for implants in rabbit tibias, Johansson and Albrektsson10 determined that 10, 16.8, and 68 N·cm were the mean values obtained 1, 3, and 6 months after implant insertion, respectively. In our study, the mean torque value after 2 months for the control group (19 N·cm) was slightly higher than that presented after 3 months in that work but statistically equivalent to the value for the experimental group (14.25 N·cm).

However, the torque values obtained in this study at 6 months were lower than those reported by Johansson and Albrektsson10 with conventional implants in rabbit tibias. Despite differences in the method of torque measurement in each study (peak value vs mean value), the lower values obtained here may be attributable to differences in the proportion of cortical/cancellous bone encountered in monocortical rabbit mandible in our investigation, as compared with bicortical tibia.15 In agreement with this statement, the values obtained by Ivanoff et al15 after 3 months in a monocortical model implant were comparable to those presented in this investigation.

Ideally, the torque required to rotate the implant at the moment of implant installation into the mandible of rabbits should have been noted as well, so that comparisons between the immediate postsurgical torque and that 2 months after could be made.13 However, because of the restricted aperture of a rabbit's mouth, this procedure was not feasible.

Our results for BIC and BM were comparable to decalcified and nondecalcified control thin sections reported by other authors.15,19,20 According to the work of Oltramari et al21 and Scarano et al,22 partial resorption of xenogenic bone grafts is expected after 3 and 48 months, respectively. Although the absence of difference in BIC or BM between the 2- and 6-month groups indicates that the presented data developed within 2 months do not change over time, the bone-remodeling process was in course and no difference was detected possibly because the portion of biomaterial resorpted was being replaced by newly formed bone and no distinction between formed bone and bone graft particles during BM and BIC measurements was performed. However, it should be noted that this study was limited by the small number of animals used, and for this reason, a possible reduction in precision and power of the statistical tests should be considered.

During implant retrieval for torque testing, the risk of detaching bone fragments is high, and therefore, the bone-implant contact extension can be underestimated.20 To address this problem, identical screw retrieval procedures were executed in both experimental and control groups; however, the results of this test should be interpreted cautiously.

However, it must be stressed that torque measurement is only an indirect surrogate for assessing the bone interaction with the implant. Another possibility is to measure implant mobility with accurate devices. Although not executed in this study, the procedure is believed to reflect clinically the degree of BIC,23 and for this reason, when performed in conjunction with BIC, BM and torque analysis could bring the obtained results near to clinical applications.

Moreover, the absence of statistical significance in this type of in vivo research does not necessarily imply the inexistence of clinical significance. Thus, further animal studies and clinical trials have to be performed to evaluate the effects of occlusal loads on bone remodeling in the presence of xenogenic graft particles and implants.

Based on the results of this investigation, it was concluded that the presence of an inorganic bovine graft did not interfere with the biomechanical and microscopic characteristics of newly formed and mature bone in the anterior mandible of New Zealand white rabbits.

BIC

bone-implant contact

BM

bone mass

H&E

hematoxylin and eosin

The authors wish to thank Professor José Roberto Lauris for statistical advice and FAPESP for financial support.