Tissue Level Changes after Maxillary Sinus Floor Elevation with Three Types of Calcium Phosphate Ceramics: A Radiological Study with a 5-Year Follow-Up

This study evaluates the radiological changes in tissue height after maxillary sinus floor elevation (MSFE) using three types of calcium phosphate ceramics over a period of up to 5 years after dental implant placement. In 163 patients, MSFE was performed. Three groups of patients were distinguished and treated based on the type of calcium phosphate ceramic used and radiologically evaluated: 40 patients with β-tricalcium phosphate (β-TCP), 76 patients with biphasic calcium phosphate (BCP) 20% hydroxyapatite (HA)-80% β-TCP, and 47 patients with BCP 60% HA-40% β-TCP. Radiological measurements were performed on panoramic radiographs at several time points up to 5 years after dental implant placement. After MSFE, a slow decrease in tissue height measured over time was seen in all three study groups. Resorption of the grafted bone substitutes was more prominent in β-TCP than in BCP ceramics with an HA component (60/40 and 20/80). Loss of tissue height after 5 years was lowest in BCP 60/40 and highest in β-TCP. This radiological study shows a predictable and comparable behavior of the slow decrease in tissue height over time for all three types of calcium phosphate ceramics used in MSFE. The fraction of HA in calcium phosphate ceramics and dental implant loading seems to be beneficial for tissue height maintenance after MSFE.


Introduction
Maxillary sinus floor elevation (MSFE) is a common pre-implant surgical procedure to increase the vertical dimension in the posterior edentulous maxilla in order to place dental implants for oral rehabilitation [1][2][3][4]. For this internal augmentation, the surgeon has the choice between autogenous bone grafts and bone substitutes [5][6][7]. Although autogenous bone grafting, considered to be the gold standard, is a well-known and reliable grafting procedure [8,9], it has the disadvantage of the necessity of a second surgical procedure and, therefore, a higher risk of surgical morbidity and complications [10][11][12][13][14]. For this reason, in some cases a bone substitute is chosen as graft material [12,15]. Different available bone follow-up, 81 implant sites were available for investigation, and at 5 years, 43 implant sites were available for radiological measurements. The criteria for treatment selection was a minimal native alveolar bone height of minimal 4 mm in the posterior maxilla and the patient's wish for prosthetic rehabilitation with dental implants. Patients with a history of radiation of the jaws, patients with heart valve prostheses and a history of endocarditis and patients who had a MSFE with simultaneous lateral augmentation of the posterior maxilla were not included in this study. Smokers (21%) were not excluded. All patients received standard care and signed a written consent prior to both surgical procedures (MSFE and dental implant surgery) for the use of their data.

Maxillary Sinus Floor Elevation Procedure
The MSFE was performed according to Tatum's "top-hinge-trapdoor-technique" under local anesthesia [3]. Patients received an antibiotic prophylaxis, amoxicillin 500 mg, 4 times daily (or in case of allergy, clindamycin 300 mg 4 times daily) for 7 days, starting the day before MSFE. For oral hygiene, chlorhexidine digluconate 0.12% (w/w) 2 times daily (10 mL) for 2 weeks was prescribed, according to the manufacturers' instructions. A midcrestal incision was made as surgical flap design with mesial and distal buccal and vertical release incisions. In the three groups, the created area at the maxillary sinus bottom was filled with the selected graft material for that group of patients. No membrane was used to cover the lateral window [39]. The wounds were closed with Gore-Tex ® sutures (W.L. Gore & Associates, Newark, DE, USA). After 10 to 14 days the sutures were removed.

Dental Implant Placement
The 256 Straumann ® SLA Soft Tissue Level dental implants with a diameter of 3.3, 4.1, or 4.8 mm and implant length of 10 or 12 mm were placed under local anesthesia 6 months after MSFE, according to the manufacturers' instructions (Institut Straumann AG, Basel, Switzerland). The wounds were closed with Gore-Tex ® sutures. All patients received a prophylactic dose of 3 g amoxicillin 1 h prior to dental implant surgery (or in case of an allergy, clindamycin 600 mg). The dental implants were left to integrate in a non-submerged unloaded fashion.
For oral hygiene chlorhexidine digluconate 0.12% (w/w), 2 times daily 10 mL for 2 weeks was prescribed, according to the manufacturers' instructions, as part of the standard procedure for dental implant surgery. To allow postoperative radiological evaluation, a panoramic radiograph (Orthophos XG, Sirona Dental Systems GmbH, Bensheim, Germany) was taken immediately after dental implant placement. Removal of the sutures was performed 10 to 14 days after dental implant surgery. Three months after dental implant placement, prosthetic treatment was started by a restorative dentist.

Radiological Evaluation
Panoramic radiographs were taken at patient intake, approximately 2 months prior to MSFE (T0), immediately after MSFE (T1), and 4 weeks prior to dental implant surgery ridge mapping is performed, which is a measurement procedure to ensure that the diameter of a dental implant does not exceed the dimensions of available bone (T2), immediately after dental implant placement (T3), at the end of the integration period, after prosthetic loading (T4), and later during the yearly recall visits (T5-T9). On these panoramic radiographs, changes in tissue height of the grafted area were measured at the implant site (black line) and 2-3 mm distally of the dental implant site (red line) ( Figure 1A,B). All panoramic radiographs were taken with an average magnification of 1.25. The measured tissue height on the panoramic radiographs was multiplied with 0.8 to obtain the true height in mm.

Statistical Analysis
Data are expressed as mean plus or minus standard deviation. Since the heights were measured several times (T0-T9) in the same individual and 256 implants were measured in 163 patients, a mixed model analysis (T0-T9) was applied to adjust for the non-independence in the data. Correction/adjustment was performed by estimating the variance of the intercepts (random intercepts). The effect modification was checked by estimating the variance of the slope (random slope). Maximum height loss was calculated at T8 and tested using an ANOVA. p < 0.05 was considered significant. IBM SPSS version 26 was used for all statistical analyses.

Patient Data
Patient data of all 512 sites (256 implant sites and 256 distal positions) in 163 patients after MSFE with calcium phosphate ceramics were gathered and combined. In Table 2, the demographic data of the study patients and the numbers and types of calcium phosphate ceramics used as a graft material in MSFE are shown.

Maxillary Sinus Floor Elevation with Calcium Phosphate Ceramics
Tissue levels (163 patients at 256 implant sites and 256 distal positions) were measured on panoramic radiographs and calculated in millimeters ( Figure 1A,B). Table 4 shows the data from the "combined measurements" and "the individual measurements" of the three calcium phosphate ceramic groups. Table 4. Individual and combined tissue height measurements (in mm) of 512 sites (256 implant site and 256 distal positions) of three types of calcium phosphate bone substitutes in maxillary sinus floor elevation (MSFE), after 5-year follow-up. β-TCP-β-tricalcium phosphate; BCP-biphasic calcium phosphate; HA-hydroxyapatite; T0-patient intake; T1-MSFE; T2-ridge mapping; T3-dental implant placement; T4-prosthetic loading; T5-follow-up visit after 1 year; T6-follow-up visit after 2 years; T7-follow-up visit after 3 years; T8-follow-up visit after 4 years; T9-follow-up visit after 5 years.

Time Points
Months of Observation

Comparison of Three Individual Types of Calcium Phosphate Ceramics
The mean tissue heights of the three individual different types of calcium phosphates are shown in Table 4 and Figure 2.

Comparison of the Combined Calcium Phosphate Ceramics at Implant Position and Inter-Implant (Distal) Position
The pattern of the mean tissue height changes of the combined calcium phosphate ceramics at two positions: at dental implant site and distally of the implant site ( Figure 1A,B). These "distal" positions can be regarded as inter-implant or inter-implant-tooth positions or, in case of a free-ending situation, as distal positions. Table 4 shows the pattern of mean tissue height changes in the posterior (partially) edentulous atrophic maxilla after MSFE surgery with a calcium phosphate bone substitute. The maxillary sinus with limited alveolar crest height (initial height 6.5 mm = T0) is internally augmented with a bone substitute 6 months prior to dental implant placement (13.5 mm = T1). Therefore, a mean increase in tissue height is 7.0 mm immediately after MSFE. The tissue height change showed a decrease of 0.6 mm (after MSFE) at the time of dental implant placement (12.9 mm = T3). Over time, the tissue height diminishes further to 2.0 mm 5 years after MSFE (T9). The total tissue height gain after 5 years is 5.0 mm, from 6.5 mm (T0) resulting in a final tissue height of 11.5 mm (T9). The final tissue height loss is 1.6 mm at 4 years and 2.0 mm at 5 years after dental implant placement. The pattern of tissue height loss at the two positions appears to be the same but the mean tissue height loss at dental implant position is 1.3 mm at 4 years and 1.7 mm at 5 years. At distal position the tissue height loss is 1.8 mm and 2.3 mm at 4 and 5 years after dental implant placement, respectively.
On several panoramic radiographs a particular phenomenon was found, indicating that apically of the inserted dental implant some of the graft material appeared to have been lifted in the cranial direction ("Summers" effect/method) [40]. The tissue height at implant site increased after MSFE. Figure 3A-C show radiographs with clear signs of this tissue height increase. This postoperative increase in height can also be recognized in Table 4 at timepoint T3.

Discussion
In this study, the radiologically derived tissue height distribution following MSFE with three types of calcium phosphate ceramics were compared over a period of 5 years after dental implant placement. In all cases, the use of calcium phosphate ceramics in MSFE resulted in an initial tissue height gain followed by an initial decrease in tissue height which subsides after dental implant placement. This pattern is consistent with previously reported MSFE procedures using bone substitutes [5,28,30,41,42], also reported in the meta-analyses by Haugen et al. [17].
Apart from the three fully synthetic calcium phosphate bone substitutes used in the present study, xenografts, such as demineralized bovine bone mineral (DBBM), have been also widely used in MSFE procedures [43][44][45][46][47]. Clinical, radiological, and histomorphometrical studies on DBBM confirmed predictable outcomes after MSFE, meaning good quality and volume of regenerated tissue [17]. DBBM has a chemical composition similar to human bone, being osteoconductive and showing a slow resorption rate comparable to 100% HA, resulting in stabilization of the grafted material and a high dental implant survival rate [48][49][50][51][52]. Several studies compared DBBM to 100% β-TCP and/or synthetic BCPs composed of different ratios of HA and β-TCP [29,51,[53][54][55][56][57]. Overall, MSFE procedures with synthetic BCPs (HA/TCP) and DBBM seem to exhibit similar results, regarding new bone formation and the survival of dental implants. However, studies on volume stability after MSFE, comparing various calcium phosphate ceramics containing different proportions of HA and β-TCP, are scarce in the literature [29,55,58]. In this respect, no studies have been reported comparing DBBM to 100% β-TCP or 60/40 or 20/80 mixtures of HA and β-TCP.
Height loss (after MSFE) may be due to settling (or clinging) of the calcium phosphate granules in the maxillary sinus and at a later stage resorption of the grafted material. At the same time, height loss of grafted material due to air pressure from respiration in the maxillary sinus is also unavoidable [59,60].
In all three groups, the loss of tissue height at the distal positions was more prominent than at the dental implant site. This could have a mechanical explanation [61,62], since dental implants can be considered as stable pillars, which eventually may lead to some degree of scalloping. Another explanation for the height differences between implant sites and distal positions could be that the calcium phosphate granules are pushed up at implant sites during the osteotomy and placement of the dental implants at dental implant surgery or by an ossifying hematoma [63]. This can be regularly observed at postoperative radiographs after implant placement [34,38,41].
The height of vital bone within the total tissue volume is the relevant structure for dental implant stability [21,64]. This vital bone consists of the original alveolar bone and the newly formed bone between the bone substitute granules as a result from osteoconduction in cranial direction. As resorption of β-TCP seems to occur at a higher rate than HA, the question is whether the resorption results in replacement by vital bone that is supposed to connect with the titanium implant root surface, eventually resulting in osseo-integration of the dental implant. From the literature, there is little or no support for this. This is the reason why for augmentation purposes BCPs have been developed [6,29,52,65].
A recent micro-CT study by Helder et al. [66] shows BCP 20/80 might perform better, at least in the short term, as a scaffold for bone augmentation in the MSFE model than BCP 60/40 as more bone is formed, and more osteoid is deposited at the cranial side in BCP 20/80 treated patients compared to BCP 60/40 treated patients.
The term "tissue height" is used instead of "bone height" in this study, since we know that, in spite what some commercial companies tend to promote, the transition from a bone substitute to vital bone is a very slow and incomplete process [5,62].
Many publications refer to radiological bone height after sinus floor elevation, actually meaning tissue height, consisting of three layers of tissues: native bone from the original alveolar crest, a transition layer consisting of a mixture of bone substitute and new vital bone, and a cranial layer of remaining bone substitute [4,5,7]. As the process of bone generation in the grafted area is slow, time plays an important role. Six months healing time after MSFE generates approximately 3 mm of bone gain [41]. This means that a substantial native bone height is required for primary stability of the placed dental implants, even after a 6-months healing period. In fact, from our former study, it can be concluded that a 9-month healing period after MSFE using BCP 60/40 would be favorable [5].
Less than 4 mm native bone height does not seem to be the suitable indication for the use of bone substitutes. [7] A further complication could be if the cranially situated bone substitute resorbs, without turning into vital bone in time. If this bone substitute dissolves, the apical parts of the dental implants would become dehiscent. In the present study, this was not observed over a period of 5 years after dental implant placement. Although the "insolvable" hydroxyapatite component in the calcium phosphate bone substitutes appears to maintain the graft on the cranial side best, it can be regarded as a covering or protective layer of tissue to cover the apical parts of the dental implant from exposing into the elevated maxillary sinus, without giving support to the dental implant. One could regard this layer of mainly calcium phosphate and fibrous tissue as merely a shielding structure for the "new" maxillary sinus.
This study also has some limitations. One limitation is the missing values in the 5-year follow-up. As a result, the consistent trend in final tissue height loss did not show a significant difference at T9. For this reason, final tissue height loss was also calculated at T8, 4 years after the dental implantation. At this time point significant differences were found at the distal position. Nevertheless, studies with a 4 to 5-year follow-up are rare but our study provided valuable insights in MSFE, dental implant placement, and final height loss of different calcium phosphate ceramics.
At the time of surgery, 25.7% of the Dutch population was smoking. As the negative influence of smoking for dental implants was known, our patients were advised to stop smoking if they did. This resulted in a percentage of smokers of 21% in our patient population. However, the aim of this study was to evaluate the changes in tissue height in a certain patient population, not the dental implant survival rate or the influence of risk factors.

Conclusions
This radiological study shows a predictable and comparable behavior of slow decrease in tissue height over time for all three types of calcium phosphate bone substitutes used in MSFE. Mean tissue height was better maintained at implant sites than at inter-implant or inter-implant-tooth positions or free-end-positions. The use of β-TCP results in a greater tissue height loss, compared to the use of HA-containing calcium phosphate ceramics. The fraction of HA in calcium phosphate ceramics in combination with dental implant loading seems to be beneficial for tissue height maintenance after MSFE. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. All patients received standard care and signed a written consent prior to both surgical procedures (MSFE and dental implant surgery) for the use of their data.

Conflicts of Interest:
The authors declare no conflict of interest. The authors claim to have no financial interest, either directly or indirectly, in the products or information listed in the article.