The use of GBR to treat bony defects around dental implants has been extensively documented throughout the past decades (Fugazzotto et al. 1997; Zitzmann et al. 1997; Hammerle et al. 2002; Aghaloo & Moy 2007). Among all the available resorbable barrier membranes used for GBR procedures, membranes made of collagen have become the membrane of choice in many clinical situations (Zitzmann et al. 1997; Jung et al. 2003; Moses et al. 2005). A few years ago, a synthetic and resorbable membrane that can be directly custom made intra-operatively for an individual defect has been engineered (Lutolf et al. 2003; Jung et al. 2006, 2009b; Wechsler et al. 2008). This newly developed synthetic membrane is made of polyethylene glycol (PEG). PEG hydrogel has been shown to be highly biocompatible and was investigated in other medicine disciplines, for example, as a sprayable adhesion barrier (Vaage et al. 1997; Mettler et al. 2003).
In a recent randomized, controlled clinical trial, this PEG hydrogel used as a barrier membrane for GBR was compared with the standard collagen membrane in the treatment of bony dehiscence defects around dental implants after an observation time of 3 years (Ramel et al. 2012). It was concluded that the PEG membrane performed as successfully as the natural collagen membrane regarding implant survival, clinical soft tissue parameters, and marginal bone levels after a follow-up period of 3 years. Further long-term data are mainly available for collagen membranes with observation periods of at least 12 years (Jung et al. 2013a).
The criticism of these types of studies is the fact that the success of the GBR procedure is assessed through a two-dimensional measurement of the mesial and distal radiographic bone level at the implant site and further clinical parameters. However, in the majority of these GBR procedures, the bone augmentation was performed mainly on the buccal aspect of the implants. Hence, there is limited date available for long-term controlled clinical studies assessing the bone dimensions at the buccal aspect of the implants, which have been placed simultaneously with bone regeneration procedures. A very recently published prospective, cross-sectional study reported on the long-term outcome of implants placed simultaneously with GBR procedures. Stable peri-implant hard and soft tissues at the buccal aspect were reported after a follow-up time of 5–9 years (Buser et al. 2013).
The aim of this study is, therefore, to assess the buccal dimensions of peri-implant bone, the implant survival rate and the mucosal level of implants that have been placed 5 years ago with simultaneous GBR either with a polyethylene glycol or a collagen membrane in combination with deproteinized bovine bone mineral (DBBM).
- Top of page
- Material and methods
The present randomized, controlled clinical study demonstrated high implant survival rates and stable buccal regenerated bone and soft tissue 5 years after GBR around dental implants using either a synthetic PEG gel membrane or a standard porcine collagen membrane covering DBBM. CBCT and clinical measurements documented similar amounts of vertical bone height and bone thicknesses at the buccal aspect of the control and the test implants.
Both groups revealed an implant survival rate of 100% after 5 years. This is similar or higher compared with other studies evaluating implant in regenerated bone after at least 5 years. In a recent long-term clinical study, 222 implants in 58 patients could be followed for 12–14 years (Jung et al. 2013a). One of the groups used the same GBR materials as in the presents control group. They revealed a cumulative survival rate of 91.9% at the 12.5 years follow-up examination. Another 5-year clinical study with a split-mouth design demonstrated implant survival rates in sites with simultaneous implant placement and GBR of also 100% (Benic et al. 2009). A systematic review on bone augmentation procedures in localized peri-implant defects reported implant survival rates ranging from 93% to 100% after 12–60 months following implant loading (Jensen & Terheyden 2009).
One criticism of the majority of long-term studies is the fact that the success of the GBR procedure is indirectly determined by the height of the mesial and distal bone level at the implants measured on two-dimensional radiographs. It would, therefore, be of high scientific and clinical interest to assess the stability of the regenerated bone buccal to the implant site where the regeneration has taken place. This study represents so far the only RCT evaluating the efficacy of the bone regeneration procedures 5 years after implant placement with simultaneous bone augmentation. The mean distance between the implant shoulder and the first BIC was 2.24 mm for the control and 2.75 mm for the test implants after 5 years. Taking into account that the presently used soft tissue level implants have a polished neck of 1.8 mm, the effective buccal bone loss in relation to the structured SLA surface is 0.44 mm for the control and 0.95 mm for the test implants. With an overall buccal bone loss of <1 mm after 5 years, both GBR procedures can be considered very stable. Only one patient of the test group revealed a buccal bone loss to a level of 0.5 mm below the BIC before the bone augmentation procedure. In this particular case, a postoperative infection with a fistula was reported. The generally consistent buccal bone level in the present study is in contrast to a recently published clinical trial (Benic et al. 2012). In this study, single-tooth implants were immediately placed into fresh extraction sockets of anterior and premolar jaw regions with a simultaneous GBR procedure. The infrabony defects and dehiscences were grafted with DBBM and covered with a collagen membrane without over-augmenting the buccal bone plate. After 7 years, the soft and hard tissues were evaluated using CBCT scans in a very similar way as in the present study. The results revealed that in one-third of the sites, there was almost no buccal bone detectable on the CBCT, whereas in the other two-third, the buccal bone level covered the entire rough implant surface. Another clinical study with 40 patients did an assessment of anterior single tooth implants with conventional CT (Nisapakultorn et al. 2010). They revealed very similar results in terms of vertical defect height, however, with a much shorter observation period of only 12 months.
This study demonstrated not only a stable vertical level of the buccal bone but also a quite thick regenerated buccal bone with mean values ranging from 1.5 mm to 2 mm at 1 mm apically to the first bone contact (HT1 mm) and from 3.0 mm to 3.1 mm at 5 mm apically to the crest (HT5 mm). This study proved considerably thicker buccal bone compared with the previously described CBCT examination evaluating immediate implants with simultaneous GBR procedures after 7 years (Benic et al. 2012). The investigators reported on a total of 14 patients revealing an average buccal bone thickness of 0.4 mm ranging from 0 mm to 2.1 mm. As mentioned before, in one-third of the sites almost no buccal bone was radiographically detected. One possible explanation for the differences in buccal bone thickness between the two studies could be the difference in time points of implant placement after tooth extraction. The present study investigated delayed implants, whereas the other study has assessed immediate implants, which possibly reveal more hard tissue alterations after implant placement (Botticelli et al. 2004; Jung et al. 2013b). Other factors explaining the discrepancies between the results of the studies might be the differences in the GBR procedure, in the CBCT scanning protocol and in the apico-coronal location of the measurement.
Further, clinical studies have analyzed the thickness of the buccal bone around natural teeth in different areas of the jaw (Vera et al. 2012; Zekry et al. 2013). They showed mean thicknesses of 0.83 mm in the anterior maxillary area and 1.13 mm in the premolar area (Vera et al. 2012). Rarely, a width of 2 mm was yielded with a generally increased thickness toward the posterior region. Hence, this present study revealed that the regenerated buccal bone after 5 years even exceeded the amount of bone thickness generally found around natural teeth.
From an esthetic point of view the most important factor is the buccal level of the soft tissue. In all, but two patients (one control and one test) the soft tissue covered the implant shoulder by 0.5–0.8 mm revealing no unesthetic exposure of the metal crown margin after 5 years. In a recent longitudinal study, it was documented that 51% of the crown margins were visible and were located in a supramucosal position after a mean follow-up time of 12.5 years (Jung et al. 2013b). The soft tissue dimensions of the present study are in line with a previous CT study including 40 implants sites in the anterior area (Nisapakultorn et al. 2010). In contrast to the present investigation, the study by (Nisapakultorn et al. 2010) showed a correlation between the mucosa thickness and the level of the musosa.
The present RCT reveals some methodological limitations. One is that titanium implants are causing artifacts within a CBCT scan, and therefore, it might be questionable whether the peri-implant tissue can be assessed by this technique (Draenert et al. 2007; Razavi et al. 2010; Schulze et al. 2010; Benic et al. 2013). This would mean that the amount of mineralized tissue, which can be detected around dental implants in CBCTs is rather underestimated. Hence, the fact that a buccal bone plate was visible at all of the implants after 5 years seems to be strong finding, pointing out the effectiveness of GBR around dental implants. The other limitation of the study is that clinical intrasurgical measurements have been compared with CBCT measurements. However, different studies have proven a sufficient accuracy of linear measurements of dento-facial structures within CBCTs (Suomalainen et al. 2008; Lamichane et al. 2009). An in vitro experiment with fresh pic jaws showed that the clinical measurements of the peri-implant bone dimensions differed in <0.2 mm to the CBCT measurements (Mengel et al. 2006). For further improvement of the accuracy, a small voxel size and a small “FOV” focusing on the area of interest has been recommended (Molen 2010). Therefore, the present study used the highest resolution of the CBCT machine with a voxel size of 0.125 mm and a scanning time of 26.9 s. To overcome the risk of movement artifacts with such a long scanning time, every effort was performed to stabilize the patient's head. Nevertheless, one patient had to be excluded due to a movement artifact. A further limitation is the fact that only one single-blinded person (DS) performed the measurements. However, all the measurements have been further checked and discussed within a team of 2–3 persons (RJ, GB, DS).