Implant survival after graftless sinus floor augmentation in highly atrophic maxillae: a randomized controlled trial in a split mouth study

Extreme atrophy of the edentulous maxilla is a common problem that requires augmentation surgery for achieving a sufficient alveolar bone volume allowing for dental implant placement [1‐3]. Lateral sinus membrane elevation for the posterior maxilla [4, 5] is a reliable procedure for augmentation of the maxilla with autogenous, xenogenous or other bone replacement material [6‐10].

The sinus lift procedure is not straight-forward and is associated with complications that may compromise the stability of the graft and the overall success of the treatment, such as the perforation of the sinus membrane, which occurs in more than 20% of cases [11, 12]. The use of autogenous bone comes along with donor site morbidity [13‐15]. The choice of augmentation technique and augmentation material remains controversial and depends mainly on the degree of atrophy of the upper jaw, the experience of the surgeon, the medical condition of the patient and the medical–technical possibilities of the operating room [3, 5, 16, 17]. The practitioner can solve these problems with circumscribed augmentative surgery using bone graft substitutes, computer-assisted cone beam computed tomography (CBCT)-based computer-guided implant placement and short dental implants [18, 19]. Simultaneous augmentation and implantation in the posterior maxilla is another possibility and has been discussed widely in literature [20‐22]. It is acceptable only if the remaining vertical bone height and width are sufficient for primary implant stability.

Ellegaard et al. [23] introduced a sinus lift technique in the posterior maxilla with sinus membrane elevation and simultaneous implant placement without grafting material. Later, Lundgren and co-workers [24] published extensively on this topic and established it in daily practice. Over the years, this technique has been widely used [25‐27]. In 2015 we published a variant of this technique, in which we elevated the sinus membrane without augmentation material in severely atrophied edentulous patients in order to place dental implants in a second step. The aim of the study at that time was to investigate spontaneous bone regeneration [28]. The aim of the present study was to evaluate implant survival, success of prosthetic rehabilitation and stability of vertical bone gain on the non-grafted maxillary side compared to conventional sinus floor augmentation with a mixture of autogenous and xenogenous bone.

In January 2013, we started our pilot study as a prospective randomized controlled clinical trial in a split mouth study design to compare the efficacy of two different techniques for the augmentation of atrophic posterior maxillae in completely edentulous patients (Fig. 1). This human study design is a translational study that has emerged from a number of animal studies in the past. The reason we developed this particular technique is that the research group already had years of experience and evidence of new bone formation after periosteal elevation with mesh space holders in animals [29‐32]. This study [28] was approved by the medical ethics committee of the Maastricht University Medical Center: azM/UM: NL41286.068.12/METC 12-2-066. The study was conducted in full accordance with the principles outlined in the Declaration of Helsinki. All included patients were fully informed about the study design and alternatives and signed the informed consent form.

Ten completely edentulous patients from the outpatient clinic of the Department of Cranio- and Maxillofacial Surgery of the Maastricht University Medical Center (MUMC), with bilateral atrophic maxillae and insufficient total prostheses were included in the study. The inclusion criteria were: complete edentulous, age 18–75 years old, residual bone height of 1–8 mm, bone width of at least 5 mm. Patients with contraindications for general anesthesia, a history of radiotherapy in the head/neck region, treatment with bisphosphonates, poor oral hygiene, uncontrolled diabetes, pregnancy, infection and increased tendency to hemorrhages, were excluded from participation of this study.

The maxillary side being treated with the experimental technique was assigned randomly; a randomization was made in SPSS (statistical package for the social sciences, IBM Corp., Armonk, NY, USA) from patient one to ten. Five papers said “experimental side LEFT” and five said “experimental side RIGHT”, they were placed in ten sealed opaque envelopes by a medical student, without informing anyone about the content. The envelope was opened by the responsible surgeon (SL) on the day of surgery right before the start of the operation. The patients were blinded regarding which side would be the test or control side.

A total of ten patients, four males and six females, were treated according to the study protocol. The age ranged from 50 to 70 years with an average of 60.5 years (SD 7.7). A bilateral sinus membrane elevation had been performed in all patients in a split mouth design. Six months after maxillary sinus augmentation, only one patient showed too little vertical bone gain on the graftless side in the preoperatively planned implant site. On this side, only two implants could be placed. A total of 29 implants were inserted, 14 in the test group and 15 in the control group. During the sinus lift procedure two small membrane lacerations of about 1 mm occurred at the test side and two sinus membrane perforations (1 and 2 mm in length) occurred at the control side. These lacerations did not need any treatment. In one patient, the incision site for bone harvesting became infected one week postoperatively. The patient was treated with antibiotics (Augmentin 500/125 mg, GSK, Brentford, Middlesex, UK) for 7 days. The wound healed within 3 weeks. No other postoperative complications occurred. No patients dropped out of the follow-up. The follow-up period ranged from 4 years and 9 months to 7 years and 4 months.

At the end of this study, we wondered whether graftless sinus floor augmentation, as we investigated it, could be a real alternative to the many existing techniques described in the literature to create a stable bone supply in the atrophic posterior maxilla pre-implantologically. Our group recently published a systematic review [27] that concluded that both non-grafted maxillary sinus lifts and conventional sinus lifts with augmentation materials have a high implant survival rate (97.92% and 98.73%, respectively). Aghaloo et al. have addressed this question in 2007 and again in 2016 [35, 36]. The article from 2016 is a systematic review article that deals with scientific contributions from 1980 to 2014 and describes the five most important augmentation procedures. All five techniques—guided bone regeneration, interpositional grafting, maxillary sinus augmentation, inlay and onlay grafting or combinations of the above can be successfully applied in pre-prosthetic augmentation surgery of the maxilla. The technique of sinus floor grafting without graft material has not been studied, as this technique was quite new. However, there is a consensus in the relevant literature that if the residual height of the alveolar process in the posterior maxilla is less than 6 mm, augmentation measures are usually recommended [2].

Established and successful grafting procedures for augmentation on the posterior edentulous atrophic maxilla exist, which offer long-term implant survival of 85 to 100% [37‐39]. Nkenke et al. in 2004 and Kessler et al. in 2005 conducted studies on large patient series in which autogenous bone was harvested from the anterior or posterior iliac crest in pre-prosthetic indications [13, 40]. If large bone volumes are required for three-dimensional maxillary augmentation with lateral maxillary widening, these techniques can still have their justification. In 2014, Nkenke and Neukam published an article dealing only with comparative clinical trials on the harvest of autogenous bone grafts [41]. Six intraoral or distant donor sites were identified. From the patients' perspective, intraoral bone harvesting at the ramus was preferred. Interestingly, the comparison of chin bone harvesting with iliac crest bone harvesting led to a preference for pelvic bone harvesting, although this requires treatment under general anesthesia. This choice was justified by the high morbidity of chin bone harvesting. Pain, loss of superficial skin sensitivity and wound healing disorders led to a high morbidity rate after chin bone harvesting. This is astonishing, since the literature often suggests that intraoral bone harvesting causes less problems than harvesting at the iliac crest. There is sufficient literature on this issue that we cannot go into detail here. Nevertheless, the unavoidable morbidity of autogenous bone harvesting was the main motivation to submit our study to the Ethics Committee in the first place, which considered the argument of non-existent morbidity in our study protocol as decisive for approval. However, it should not be forgotten that only autogenous bone grafts provide all biological advantages desired at a defect side, such as: scaffold for osteoconduction, growth factors for osteoinduction and progenitor cells for osteogenesis, which altogether seems to be decisive with regard to bone volume stability and the structure of augmented bone sites [42]. With autogenous bone as bone source for augmentation exceptionally high implant survival of up to 100% after cumulative 3-year analysis after placement of implants in the maxilla regardless of source, chin or iliac crest bone were reported in literature [43].

The use of bone replacement materials is the classical alternative to autogenous bone. An almost unmanageable amount of alternative bone product is offered. Bone substitutes can be used just as successfully as autogenous bone. Al-Nawas and Schiegnitz published a systematic review and meta-analysis on bone substitutes as augmentation materials in maxillary sinus floor augmentation. The mean implant survival rate was 98.6% for implants placed in sinuses augmented with bone substitute materials. In the comparison groups, sinus floor augmentations with a mixture of autogenous bone and bone replacement material and autogenous bone were used exclusively. In the former group the mean implant survival was 88.6%, in the latter 97.4%. All procedures appear to allow long-term implant survival [44]. Similar results were published by many other authors [1, 45, 46].

Even the question whether sinus lifting is necessary at all is difficult to answer, if short implants of less than 8 mm can successfully be loaded in maxillary bone with a residual height of 4 to 6 mm [47]. However, their long-term prognosis is still unknown. This statement must be critically evaluated and can only be seen in the interaction of augmentation, implantation and prosthetic rehabilitation. In most patients, there is atrophy of the maxilla as well as atrophy of the edentulous mandible. If the intermaxillary space becomes too large as a result, short implants may fail if the ratio of the length of the implant and the height of the prosthetic structure is also unfavorable to the implant length. In a recently published study, Al-Nawas and Schiegnitz address the question whether narrow-diameter implants (mini implants) in the atrophic jaw are successful. They warn against mini-implants in the masticatory posterior region [48]. The same could also apply to short implants with unfavorably high prosthetic abutments. In addition, dental implant manufacturers often only guarantee the success of their products if the implant–prosthetic construction aspect ratio is 1:1 [49].

In this context, Summers' technique, which allows the insertion of longer implants, should not go unmentioned [50]. One advantage of the crestal sinus lift, a "minimally invasive" technique, is that it is less traumatic for the patient than the external sinus lift. The narrow and complex surgical access via the burr hole of the crestal sinus lift requires a high degree of skill on the part of the surgeon. An equally important factor of the crestal sinus lift according to Summers is the augmentation height of about 4 to 5 mm that is possible [51].

Ellegaard et al. [23] and later Lundgren [24] were the first to describe successful implantation after sinus membrane elevation without the use of any bone grafts or applying the crestal sinus lift. A coagulum is formed after creating a bone wound by removing the sinus membrane from the floor of the sinus, the lateral and medial sinus walls. The osteogenetic capacity of the surrounding bony walls on at least three sides of the sinus leads to the formation of bone-forming callus from the coagulum. The smaller this space is, the more stable is the coagulum formed in this space. When the space is bigger, like in completely edentulous maxillae, this may lead to a less stable callus. When no bone grafts are used in a sinus membrane elevation, this space does not contain a scaffold [33].

This study is the first study about the implant survival after graftless maxillary sinus membrane elevation that included patients that were all completely edentulous. The use of a resorbable membrane from material like PDLLA has the advantage that no material had to be removed in a second surgery and the crucial region of interest would stay untouched when the implants were placed. The insertion of the implants was challenging because the usual feeling of insertion resistance was missing. As indicated above in the preimplantation radiographic analysis, bone maturation was not yet complete at implant placement or was completed without sufficient calcification. Next to this, we do not exactly know what the resorbable material would do and what influence it has on bone growth. It is known that it is not always fully absorbed after sinus membrane elevation. This was closely studied in a publication of Cricchio et al. [52, 53]. To discover what the PDLLA membrane did in our human study, we should have retrieved histological evidence including the sinus membrane. Since this would bring too much damage to the patient, it would not be approved by the medical ethics committee.

A limitation of the technique used in this study is that the open window to the sinus cannot prevent the growth of granulation tissue into the defect. On the other hand, the periosteum directly covers the opening to the maxillary sinus. There is a possibility that the bone-forming capabilities of the periosteum play a beneficial role in stimulating bone growth.

We trusted the functional biological process that functional loading will result in bone apposition and that the PDLLA membrane would not interfere at implant level. Nevertheless, we can conclude from these clinical results that the implants are osseointegrated. Only long-term (> 10 years) follow-up might show, if the functional loading due to the placement of dental implants in conjunction with appropriate prosthetic rehabilitation will render a stable callus in a non-grafted atrophic maxillary sinus. Further studies have to prove, if earlier or maybe later implant placement will render better results with regard to implant survival. The staged procedure we chose was based on the empirically gained experience of conventional implantology.

This study has some limitations: the sample size is small, there was no option to take bone biopsies with inclusion of the sinus membrane or retrieve implants for histological prove of implant osseointegration. In addition, we know that the posterior maxilla is often more atrophied than the anterior. If we exclude the anteriorly placed implants, we see a survival rate of 100% on the conventionally treated side versus 78.9% on the side treated without bone graft (not significant).

We can also discuss whether the higher implant loss on the test side could be caused by the very small amount of residual bone shown in Table 4. In a next study, it might be interesting to exclude patients with less than 4 mm of residual bone. In this way, we could investigate whether the experimental technique leads to better implant survival.

Furthermore, a uniform definition of the term "loss of function" after prosthetic restoration does not exist and whether there will be one in the near future remains questionable. Often it remains unconsidered or undefined on the basis of which criteria the end of the functional phase of a dental restoration is reached. Mostly retrospective case series with at least 100 patients and an observation period of at least 3 years allowed a study-related estimation of the survival of prosthetically restored implants (implants as surrogate parameters), but the different endpoints as well as the study design (case series) or different forms of prosthetic rehabilitation do not allow pooling of the results [54‐57]. The 5-year implant success rates varied between 69.5% and 98.4% (the corresponding 10-year success rates between 79.4% and 94.3%), depending on the prosthesis used [58]. The long time periods make it clear that the event "loss of function" usually occurs only after a very long time, if one has a normal clinical use period in mind. Such a long time can often hardly be waited for results in randomized trials. So far, the results of implant survival in this study show acceptable results since all prosthetic restorations were still functioning.

We conducted an experiment with completely edentulous humans with atrophic maxillae by comparing conventional augmentation procedures with a sinus membrane lifting technique without using any bone grafts. The results showed new bone formation on both test sides. However, the non-grafted side showed significant less bone gain than the grafted side, with an overall mean of, respectively, 6.20 mm (SD 4.92) and 9.69 mm (SD 4.62).

Within the limitations of this study, we can conclude that the non-grafted technique might have some potential for clinical use, although it showed a higher implant failure. This technique in the complete edentulous maxilla is a challenging procedure. It can be applied to avoid the use of any bone grafts and reduce morbidity in patients where this is favorable. Further research with more statistical power will be necessary to support the hypothesis of this study.