The results of a randomised trial with bisphosphonate-coated dental implants have been reported previously. Each patient received one coated and one uncoated implant in a double-blind split-mouth design study. After 6 months of osseointegration, resonance frequency analysis indicated better fixation of the coated implants. Reduced marginal bone resorption was also shown. However, it was not known whether the advantage of the bisphosphonate coating would persist over time. The radiographic results at 5 years after implant installation are reported herein. A blinded investigator measured marginal resorption on fresh radiographs obtained for 14 of the 16 patients (two had died) and compared these with the post-implantation images. Non-parametric statistics were used. All implants functioned well. The median marginal bone loss for control implants was found to be 0.70 mm, which is less than usually reported in the literature. The bisphosphonate-coated implants showed even less resorption (median 0.20 mm). The median difference within each pair of implants after 5 years of use was 0.34 mm (95% confidence interval 0.00–0.75 mm; P = 0.04). The present data suggest that bisphosphonate-coated implants enable prolonged preservation of the marginal bone.
The insertion of metal implants in bone is one of the most common of all surgical procedures. Brånemark et al. first defined ‘osseointegration’ in 1969 as the direct structural and functional contact (at the light microscopy level) between living bone and the implant. The clinical definition of implant osseointegration considers the level of stable marginal bone and absence of mobility in the bone. The diagnosis is therefore based on radiographic and mechanical stability criteria.
Many investigators have shown that tooth extraction stimulates osteoclastic activity with varying amounts of alveolar crest loss. Systemic alendronate was found to significantly reduce the bone loss associated with experimental periodontitis in monkeys and beagle dogs. Local delivery of alendronate can reduce the alveolar bone resorption activated by mucoperiosteal surgery. In orthodontics, the topical administration of amino-bisphosphonate was found to cause a significant reduction in tooth movement in rats when orthodontic force was applied.
Several methods have been reported to increase the bone density around experimental porous implants, but to varying degrees. In animal models, investigators have shown that surface-immobilised bisphosphonates improve the mechanical fixation of metal screws in terms of an increased pullout force and bone-to-implant contact. Systemic and local treatment with bisphosphonates has been shown to improve the fixation of orthopaedic implants. A single infusion of 4 mg zoledronate improved the initial fixation of a cementless hip prosthesis. The local application of ibandronate during total knee replacement reduced postoperative migration of the prostheses as measured by radiostereometry.
In a recent randomised trial by the present study group, it was shown that dental implants coated with a bisphosphonate-eluting nano-layer of fibrinogen became better fixed than uncoated control implants. This trial had a paired, split-mouth design, with each of the 16 patients receiving both a coated and an uncoated control implant in a randomised, double-blinded fashion. The patients were treated with a two-step procedure, i.e. the implants were submerged and left unloaded for osseointegration for 6 months. Measurements of vibration resonance frequency were performed at insertion and when the implants were exposed for abutment fixation at 6 months, which indicated better fixation when comparing each bisphosphonate-coated implant with its paired control. There was also less bone loss from the alveolar crest near the bisphosphonate implants (less marginal bone resorption). The radiographic findings after 1 and 5 years of implant loading are reported herein. It was hypothesised that the initial difference in bone loss from the alveolar crest would persist.
Materials and methods
It was demonstrated in a previously reported study that a thin bisphosphonate-eluting fibrinogen coating can improve the fixation of metal in human bone. Sixteen patients (mean age 65 years, 11 female) received one bisphosphonate-coated implant and one ordinary implant in the upper jaw at sites with similar bone quality. The patients were examined preoperatively by computed tomography (CT) scan to ensure a sufficient maxillary bone volume and shape (in practice, a buccolingual distance >6 mm). The implants used were Brånemark MK III TiUnite, 3.75 mm in diameter. The coated and control implants were both 11.5 mm long and visually indistinguishable. The coating procedure was performed as described by Tengvall et al. A cross-linked layer of fibrinogen was covalently bound to the metal, and then small amounts of pamidronate and ibandronate were bound and adsorbed to the fibrinogen matrix. After preparation and sterile packaging, the implants were sterilised with gamma radiation (25 kGy; ARTIM, Praha, Czech Republic). The effects of gamma radiation on the bisphosphonate–fibrinogen layer were analyzed in a previous experimental study. All implants were placed using a two-step process. Resonance frequency analysis (Ostell Mentor; Integration Diagnostics, Gothenburg, Sweden) was used to measure the implant stability quotient (ISQ) at implant insertion and at abutment connection (after 6 months).
Dental radiographic films were obtained for each individual implant preoperatively and after 2 and 6 months, using a long cone technique in order to obtain an X-ray direction perpendicular to the film. A blinded investigator, not previously involved in the study, measured the level of the marginal bone in relation to a defined edge on the implants. This investigator also performed the same measurement on the post-implantation radiographs, and the difference constituted ‘marginal bone resorption’. To estimate the error in measurements, the same measurements were performed independently by another investigator who had previously had access to the treatment protocols and was therefore theoretically unblinded. The results are based solely on the measurements of the blinded investigator.
In the previous randomised trial, and at 18 months of follow-up (1 year of loading), the radiographic examinations were done using traditional radiographic intraoral films (Insight Super Poly-Soft; Kodak, Rochester, NY, USA). Dental radiographic films were taken for each individual implant using a long cone technique and positioning the device in order to obtain an X-ray direction perpendicular to the film. One hundred and thirty-eight radiographs (256 implant sites) were taken and were analyzed by two examiners. Ten radiographs were excluded due to poor quality. For the 5-year follow-up, the examination of marginal bone level was performed using a digital receptor technique (Schick CDR Elite; Sirona Dental Systems GmbH, Bensheim, Germany) and diagnostic imaging (Sectra PACS; Sectra AB, Linköping, Sweden) for 14 patients (two had died). Thirty-two digital radiographs (56 implant sites) were analyzed by two investigators. Four digital radiographs were excluded due to poor quality. All radiographic measurements were adjusted according to the mean enlargement of the radiographs. The quality of the films was examined by a radiologist.
The level of marginal bone at the fixture mesially and distally was estimated from a reference point on the implant to the first implant–bone contact ( Fig. 1 ). The bony defect was defined as the most coronal point where the periodontal ligament space showed a continuous width.
The assessment of marginal bone loss was based on internal controls. The hypothesis was that bisphosphonate-coated implants would show less bone loss than controls after 5 years in function. The difference in the change in marginal bone level on digital radiographs was analyzed with the Wilcoxon test for paired data, and the confidence interval (CI) was calculated according to Hodges–Lehman, using IBM SPSS Statistics version 20.0 software (IBM Corp., Armonk, NY, USA).
At 18 months after insertion (1 year of loading)
All implants functioned well, and no complications were registered. In general, the marginal bone loss was small. The values appeared not to be normally distributed, mainly because one patient showed more resorption than the others: nearly 2 mm for both implants. The marginal bone loss for the control implants from insertion to 18 months thereafter was a median 0.66 mm, ranging from 0 to 1.75 mm. The median difference within each pair of implants was 0.50 mm, 95% CI 0.00–1.25 mm ( P = 0.04).
At 5 years of loading
The marginal bone level was assessed in 14 patients after 5 years of use (range 58–62 months). Two of the 16 patients originally included had died; next of kin were interviewed and reported no known implant problems. All other implants functioned well, and no complications were registered. The marginal bone loss for the control implants from insertion to 5 years thereafter was a median 0.70 mm (range 0–2.1 mm). The marginal bone loss for the bisphosphonate-coated implants was a median 0.20 mm, range 0–1.25 mm. The median difference within each pair of implants was 0.34 mm, 95% CI 0.00–0.75 mm ( P = 0.04). Fig. 2 shows the differences in marginal bone loss between the control implants and bisphosphonate-coated implants at baseline and 2 months, 6 months, 18 months (1 year of loading), and 5 years postoperatively.
The measurements in 0.25-mm increments were in complete agreement between the two readers for nine patients, and never differed by more than 0.25 mm ( Tables 1 and 2 ).
|Patient||Bisphosphonate-coated implant||Control implant|
|2 months||6 months||18 months a||60 months||2 months||6 months||18 months a||60 months|