Does the Presence of Pathological Change in the Schneiderian Membrane Increase the Risk of Membrane Perforation During Sinus Floor Elevation? A Systemic Review

The placement of dental implants in areas with insufficient bone height, particularly the posterior maxillary region, is a complex and challenging undertaking. To improve the vertical bone height and volume beneath the maxillary sinus, the most common surgical approaches are osteotome sinus floor elevation (OSFE) or lateral sinus floor elevation (LSFE).^1,2  The selection of the OSFE or LSFE depends on the amount of the remaining residual bone height (RBH) in the posterior maxilla. When the RBH is <5 mm, the lateral window technique is recommended.³  However, if the RBH is >5 mm, the modified transalveolar approach^4,5  or OSFE surgery may be a better choice because it is less invasive.

Generally, before sinus floor elevation (SFE) is planned, a radiographic examination is necessary to understand 2- or 3-dimensional bone volume and the status of the maxillary sinus.⁶  Panoramic radiograph and cone-beam computerized tomography (CBCT) are the most commonly used methods.^7,8  One of the important structures that should be evaluated before performing SFE is the Schneiderian membrane (SM). The SM is a mucous membrane lining the internal surface of the maxillary sinus.⁹  It consists of 3 layers: the pseudostratified columnar epithelium (respiratory type epithelium), a highly vascular lamina propria, and the periosteum.¹⁰  Different methods have been used to measure the thickness of the SM, such as cadaver examinations, CTs, and CBCTs.^7,8,11  With the advent of new technology, SM thickness has also been measured by obtaining a mucosal biopsy from the maxillary sinus of living subjects by using an endoscope.¹²  Based on the thickness and the appearance of the sinus membrane in the radiograph, the SM was regarded as normal, thickening, or antral pseudocyst (polypoid mucosal thickening).^13,14 

An SM thickness ≤2 mm is considered to be normal; however, mucosal thickening >2 mm is regarded as a pathological condition.^15–17  This condition is a common radiographic finding of the maxillary sinus, amounting to 12% and 29.2% in patients scheduled for dental treatment.^15,16,18,19  Another common pathological change in the maxillary sinus is the antral pseudocyst. It is a benign lesion, has a dome-shaped appearance on the radiograph, and is usually discovered incidentally. It consists of an inflammatory exudate surrounded by loose connective tissues,^20,21  and it has been reported to be observed in 16.4% of patients and 10% of sinuses studied.¹⁹  These pathological conditions have been considered as a potential risk for failure of bone augmentation and dental implants.²² 

Controversy regarding the effect of the presence of antral pseudocysts or thickened membranes on SM perforation and implant survival rates has been reported in the literature.²³  Lum et al¹²  concluded that patients with thin SM experienced more membrane perforation compared with those with thick SM. On the other hand, Wen et al²⁴  reported that the perforation rate of the SM was higher in sinus mucosa ≥3 mm or in membranes ≤0.5 mm.²⁴ 

Celebi et al²⁵  reported that among 4 patients with antral cysts who underwent SFE, no complications occurred during follow-up. Kara et al²⁶  performed SFE surgery on 29 patients with antral pseudocysts and obtained a 100% implant survival rate. Unfortunately, the evidence regarding SFE under the circumstance of these 2 sinus statuses (mucosal thickening and antral cyst) is still unclear.

Because of the lack of compelling evidence summarizing the risk of the presence of sinus pathology (antral pseudocyst or mucosal thickening) on SM perforation and implant survival rates during SFE surgery, this study aims to systematically review and critically analyze the available evidence and to determine whether the presence of SM thickening or pseudocysts increases the risk of membrane perforation and implant failure rates during SFE.

This systematic review and meta-analysis adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement for reporting systematic reviews.²⁷ 

The protocol of the current review has been registered at PROSPERO (CRD42020182053).

Does SFE in the presence of SM pathology increase the risk of membrane perforation and implant failure rate?

The question of the current systemic review was adopted in accordance with the PICO criteria (Table 1).

The following databases were searched without time or language restrictions in December 2019: MEDLINE (PubMed), Embase, Cochrane Library, CNKI, WANFANG, and VIP.

A manual search of the following dental journals was also performed: Clinical Oral Implants Research, Journal of Dentistry, Clinical Oral Implant Dentistry and Related Research, Journal of Clinical Periodontology, Journal of Periodontology, Journal of Oral Rehabilitation, International Journal of Oral and Maxillofacial Surgery, European Journal of Oral Implantology, Journal of Prosthodontic Dentistry, Journal of Prosthodontic Research, Journal of Oral and Maxillofacial Surgery, and British Journal of Maxillofacial Surgery.

The reference list of the identified studies and the relevant reviews on the subject were also evaluated for possible additional studies. Moreover, online databases providing information about clinical trials in progress were checked (clinicaltrials.gov; www. centerwatch.com/clinicaltrials; www.clinicalconnection.com).

Studies were included if they met the following criteria: (1) randomized or nonrandomized controlled studies that evaluated the outcomes of SFE in human patients with SM pathology and normal patients regardless of whether SFE surgery was performed with or without simultaneous implant placement; (2) clear diagnosis of sinus pathology (antral pseudocyst or mucosal thickening) on panoramic radiograph, CT, or CBCT; and (3) full-text articles reporting at least 1 of the outcomes of interest such as perforation events, and implant failure events. If more than 1 study reported findings from the same population, only the article covering the most comprehensive information was included. The eligibility was evaluated by 2 authors (M.M.A., Y.B.) independently, and disagreements were settled by consultation.

Articles were excluded for the following reasons: (1) review studies, meeting abstracts, or case reports; (2) not including the sinus pathology (antral cyst or mucosal thickening); (3) lack of information about SM perforation or implant survival/failure; and (4) fewer than 5 patients in each group.

Two investigators (M.M.A., Y.B.) independently screened and selected the studies according to the titles and abstracts and classified the studies as “include” or “exclude.” Any disagreements were resolved through discussion and negotiation. Articles that met the inclusion criteria were then reviewed by both investigators.

Two authors (Y.B., M.M.A) independently assessed the titles, abstracts, and full text of the included studies, and the following data were extracted (when available): authors, year of publication, type of study, number of patients and sinuses, age, elevation methods, time of implant placement, imaging data, grafting material, implant brand, RBH, perioperative complications (membrane perforation, membrane swelling, ostium obstruction, change of antral cyst volumes, sinusitis, etc), raised height of the sinus floor, endosinus bone gain, number of implant failures, prognosis, and follow-up. Any disagreements between the 2 investigators were resolved by a third author (Y.F.) through discussion and consensus.

The modified Newcastle-Ottawa Scale (NOS)²⁸  was independently applied by 2 authors (Y.B., Y.D.) to assess the methodological quality of the case series, case-control studies, and cohort studies. Disagreements were resolved by consulting the third author (Y.F.) to adjudicate. The NOS consists of 3 factors: object selection, comparability of the study group, and assessment of outcome. A total of 0 to 9 stars were allocated to each study. Studies achieving more than 6 stars were considered to be of high quality.

Odds ratios (OR) with 95% confidence intervals (CIs) were used to synthesize binary data such as SM perforation and implant success rate (when available). Because most of the included studies reported zero instances of implant failure, the implant survival rate was expressed as a percentage (%) by dividing the number of implants survived by the total number of implants inserted. The data regarding other outcomes were reported as a percentage (%) or mean and standard of deviation (if available). Both the Cochran chi-squared test (Q-test) and I² statistics were used to express the percentage of the total variation across studies due to heterogeneity, with 25% corresponding to low heterogeneity, 50% to moderate heterogeneity, and 75% to high heterogeneity.²⁹  The inverse variance method with the random effects model described by DerSimonian and Laird³⁰  was used when I² > 50%; otherwise, a fixed-effects model was used. A P value of <.05 was considered statistically significant. Comprehensive Meta-Analysis Software (Biostat Inc, Englewood, NJ) was used for data analysis.

The kappa value was 0.85, so the agreement between investigators was almost perfect. The process of electronic database searching is summarized in Figure 1, and the detailed search strategy to identify eligible studies for the systematic review is presented in Figure 1 and Supplementary Table S1.

The initial search yielded 803 studies, and 5 more studies^31–35  were obtained from manual searches, resulting in a total of 808 articles. A total of 691 records remained after duplicates were removed. The titles and abstracts of the remaining 691 articles were screened, and 660 were excluded because they were irrelevant. Two researchers carefully read the full text of the remaining 31 studies for potential inclusion. Based on the inclusion and exclusion criteria mentioned before, 13 studies were excluded (Supplementary Table S2), and 18 studies^{26,31–34,36–42}  were included for data extraction and quality assessment (Tables 2, 3, and 4). Among them, 11 were retrospective studies,^{9,24,26,32,34,36,39–43}  3 were prospective studies,^31,37,38  and 4 were case series.^33,35,44,45  These articles were published between 2007 and 2020.

The mean age of the included participants ranged from 17 to 94 years. Seven studies^{9,24,32,35,37,39,43}  reported a mean age of >50 years, and 9 studies^{26,31,33,36,38,40,42,44,45}  reported <50 years; the age of the participants was not reported in 2 studies.^34,41  Both male and female participants were included in the 17 studies,^{9,24,26,31–44}  and the gender of the participants was not reported in 1 study.⁴⁵  In total, 18 studies with 1542 participants (unclear gender of 14 participants, 775 male and 753 female) and 1797 SFE procedures were included.

The duration of follow-up ranged from 4 to 107 months but was not reported in 4 studies.^9,24,38,43  The mean follow-up was >6 months. The included studies excluded patients with severe systemic disease or with uncontrolled periodontal disease who were unwilling to participate in clinical research.

The basic information about each study is shown in Table 2. Intra- and postoperative complications and prognosis are listed in Table 3.

On the diagnosis of maxillary sinus pathology including antral cysts or membrane thickening, 14 studies^{9,24,31–38,42–45}  with a total of 1129 participants used CBCT to assess the status of the participant's sinuses scheduled for SFE. One study⁴¹  with a total of 104 patients used CT, whereas panoramic radiography was used in 3 included articles^26,39,40  with a total of 309 patients. Sixteen studies^{9,24,26,31,34–39,40–45}  with a total of 1449 participants reported the number of SM perforations during the sinus augmentation procedure. A total of 51 perforations occurred during 701 OSFE procedures and 96 SMs perforated during 731 LSFE procedures. Sinus augmentation methods were unclear in 3 studies.^34,38,39 

Of the 18 articles, 12 studies^{26,31,32–36,39,40,42,44,45}  with a total of 805 patients and 1204 dental implants reported implant success rates ranging from 96% to 100%.

According to the NOS, 6 articles^{9,24,35,38,44,45}  received 6 stars each, 6 studies^{32–34,37,41,43}  received a score of 7 stars, and 6 studies^{26,31,36,39,40,42}  received 8 stars each. On the whole, all studies were considered to have high risk quality (Table 4).

A total of 171 sinus membrane perforations during 1674 SFE procedures in 16 studies^{9,24,26,31,34–45}  were reported. The weighted perforation rate was 10% (ranging from 0% to 22%). Eight studies (6 retrospective studies,^{9,24,38,41–43}  2 prospective studies^31,37) with a total of 1048 SFE procedures compared the incidence of membrane perforation during surgical sinus elevation in normal and thickening sinuses. The rate of SM perforation in the normal sinus group was 14%, whereas it was 6% in antral pseudocysts and 6% in sinus mucosal thickening. In 6 included studies,^{24,31,36,40,44,45}  the OSFE method was used, and the perforation rate was 9%. The LSFE method was reported in 4 included studies,^9,35,37,42  and the perforation rate was 9%. All perforations were managed by a biological membrane, and no further complications occurred. The SFE method varied among the included studies in the meta-analysis of SM perforation: LSFE was used in 3 included studies^9,37,42  and OSFE was used in 2 studies,^24,31  whereas both techniques were used in 3 studies. The overall result of the meta-analysis comparing the incidence of SM perforation showed no significant difference between the normal sinus and sinus with mucosal thickening (fixed effects; OR, 0.896; 95% CI, 0.504–1.59, P = .707, I² = 32%). A subgroup meta-analysis showed a nonsignificant difference in the incidence of sinus perforation between normal SM and mucosal thickening in both OSFL and LSFE (Figure 2).

A total of 1204 implants were installed in the 12 included studies,^{26,31–36,39,40,42,44,45}  with a total of 805 patients, and 19 implants were lost. The total implant survival rate was 98%. The survival rate ranged from between 96% and 100% during a mean follow-up period of 4 to 56 months.

A total of 388 implants were installed in the posterior atrophic maxillae with normal sinus following SFE; of these, 7 implants failed, resulting in an implant success rate of 98%. Of 230 implants placed in patients with an antral pseudocyst, 1 implant was lost, and the survival rate was approximately 100%. In 4 included studies^31,32,34,44  with a total of 110 dental implants installed in the group of patients with mucosal thickening, no implant failure was reported, resulting in a 100% survival rate. Because of the heterogeneity among the selected studies and because most of them reported zero implant failure, a meta-analysis of implant failure was not possible for studies comparing normal vs thickened SM.

In 6 studies, the mean values of the raised height of the sinus floor was reported. The mean raised height of the sinus floor was reported to be <7 mm in 4 articles^24,31,36,45  and >10 mm in 2 articles.^9,37  The reported mean elevated height varied, ranging from 4.81 mm to 6.93 mm in the OSFE group and 6.8 mm to 11.32 mm in the group of patients who underwent LSFE.

Endosinus bone gain (ESBG) was reported in 3 studies^31,36,40  with a total of 213 patients. Gong et al³⁶  reported an average of 3.55 ± 2.46 mm (range, 0.71–7.95 mm) in ESBG in patients with normal sinus and 4.03 ± 2.33 mm (range, 0.87–8.52 mm) for antral pseudocysts after a healing time of 4–6 months. Feng et al⁴⁰  reported a mean ESBG of 3.40 ± 1.78 mm (range, 0–6.45 mm) in 21 patients with antral cysts. Qin et al³¹  reported an average ESBG of 3.25 ± 0.83 mm (range, 0.49–5.75 mm) with an average follow-up of 7.82 ± 2.02 months.

A statistically insignificant difference was observed in ESBG between the normal sinus group and the antral pseudocyst group (P = .26) in the study conducted by Gong et al.³⁶  In addition, no significant difference was observed regarding ESBG when comparing sinus pathology (antral cyst or mucosal thickening) with normal sinus (P = .15).³¹ 

Qin et al³¹  reported a mean bone height of 31.83% ± 16.89% and mean crestal bone loss of 0.50 ± 0.33.

Only 1 study⁴⁰  measured the periodontal index after an average follow-up of 27 months. The modified plaque index was 1.30 ± 0.67, the modified bleeding index was 0.60 ± 0.70, and the mean probing depth was 3.35 ± 0.74 mm.

The volumes of all 19 cysts in the included study³⁶  increased after operations. Sinus opacifications (n = 10) were observed in 2 studies.^31,37  In 1 included study,³⁷  membrane swelling (n = 49) was the most frequent complication after 53 SFE procedures. Three studies^26,38,39  reported the incidence of acute sinusitis. A total of 17 patients had postoperative sinusitis (12 sinusitis occurred among 310 patients with normal sinuses, 4 sinusitis developed among 37 patients with antral cysts, 1/17 of the patients with SM thickening). In 2 studies,^41,42  12 wound infections were reported. Other complications included ostium obstruction, wound dehiscence, and bleeding. Only 1 article⁴²  reported preprosthetic and implant-related complications and consisted of 3 failing osseointegrations, 2 implant mucositis, and 1 peri-implantitis.

Five studies^{31,35–37,45}  reported postoperative volumetric changes of the antral cysts. Of the total 76 antral pseudocysts, the volumes of 5 cysts remained unchanged (6%), 6 cysts increased (8%), 18 cysts decreased (24%), and 27 cysts disappeared (36%) after a follow-up of 4–36 months. One study⁴⁴  reported the change in thickness of the SM in the group of patients with sinus mucosal thickening after sinus augmentation. The SM thickness increased in 2 patients and decreased in 14 patients.

This systematic review included 11 retrospective studies, 3 prospective studies, and 4 case series involving 1797 sinuses in 1542 participants. To the best of our knowledge, this is the first systematic review to analyze the available evidence regarding the incidence of SM membrane perforation and implant survival rate in studies comparing SFE in patients with normal and pathologic sinus membrane (antral pseudocyst or mucosal thickening). The findings of this review concluded that there is not a statistically significant difference in the incidence of SM perforation between the normal sinus mucosa and sinus with mucosal thickening during the sinus augmentation procedure (fixed; OR, 0.896; 95% CI, 0.504–1.59, P = .707, I² = 32%). The weighted SM perforation was 14% in the normal sinus, 6% in the antral pseudocysts, and 6% in the sinus with mucosal thickening. These findings are inconsistent with studies conducted by Ritter et al⁴¹  and Rapani et al,⁴³  such that the incidence of SM perforation was significantly higher in the normal-appearing sinus mucosa group than in the mucosal thickening group (P < .05). In contrast, Lin et al⁹  found a statistically significant correlation between perforation and membrane thickness. They found that the perforation rate was lowest (7.14%) when the membrane thickness was 1–1.5 mm, and as the membrane became thicker (≥2 mm) or thinner (<1 mm), the perforation rate increased dramatically.⁹  Hence, we can infer that thickened membranes may have more sufficient mechanical capacity to resist elevated force or bone grafting than normal or thin sinus mucosa, which was consistent with the in vitro study by Pommer et al.⁴⁶ 

Sinus membrane perforation is a common intraoperative complication during SFE, and it has been reported to occur in 5%–44% of patients subjected to SFE.^47–50  The weighted perforation rate was 10% in the present study. One membrane perforation occurred in every 10 procedures of sinus lift surgery.

Several factors have been considered as potential risks for SM perforation. For instance, anatomic abnormalities such as sinus septa were reported in 13% to 59.7% of the population.^47,51–54  Other factors such as reduced membrane thickness, thin gingival phenotype (<1.5 mm), and small residual ridge height have also been considered important factors for sinus perforation.^55–57  Therefore, CT or CBCT assessment is necessary to avoid the occurrence of perforation.

With regard to the implant survival rate, because most of the included studies reported 0 events in all treatment arms, and because of the heterogeneity and inadequately comparable studies in our review, no meta-analysis could be performed. However, none of these studies reported any difference in the implant survival rate between normal sinus and sinus pathology. The weighted implant survival rate in the 18 included studies was calculated and ranged from 96% to 100% after a follow-up period of 4–107 months. A total of 388 implants were installed in the posterior atrophic maxillae with normal mucosa, and the weighted implant survival rate was 98%. However, the implant survival rate reached 100% in the groups of patients with SM thickening and antral pseudocysts. These results were in accord with the studies conducted by Gong et al³⁶  and Küçükkurt,³²  such that insignificant differences in the dental implant survival rate were shown between normal sinus and sinus pathology. This indicated that sinus pathology has little impact on implant survival. Although infection after membrane perforation may be a risk for implant failure, it did not show relevance in our study. One meta-analysis also revealed a nonsignificant difference in the implant survival rate between nonperforated and perforated sinus.⁵⁵  In contrast, Al-Moraissi et al⁵⁸  conducted a meta-regression analysis and concluded that an intraoperative SM perforation significantly increases the risk of implant failure after sinus lift surgery. To date, sinus pathology (antral pseudocyst or thickened mucosa) has not been demonstrated to result in higher failure rates.

In this review, only 2 studies^31,36  reported a raised height of the sinus floor, and 3 studies^31,36,40  reported an average ESBG of 3.25–4.03 mm. ESBG may be a key factor in implant survival and gain in long-term stability. Because the measuring sites and calculation methods on the raised height of the sinus floor and ESBG were different, an absolute comparison could not be conducted.

A sinus augmentation procedure was performed on asymptomatic patients in the presence of antral pseudocysts, and in only in 1 study³³  was an injector used to remove the cyst liquid to reduce the volumes. The sinus floor augmentation procedure had no great influence on the prognosis of cysts in our study. Although the volumes of the cysts might increase immediately after the operation, a few cysts remained unchanged or increased, and most decreased or disappeared after a healing time of several months. This is the result of the natural healing process of the lesions or the administration of antibiotics after surgery. However, the increased size of the antral pseudocyst and swelling membrane immediately after SFE may result in sinus ostium obstruction or liquid drainage difficulty, resulting in sinusitis. Based on the suggestion of some clinicians, asymptomatic and small antral pseudocysts on the sinus floor can be preserved if they do not affect the augmentation procedure. Cyst removal may increase the chance of membrane perforation or bleeding. Patients with a history of inactive sinusitis or postoperative swelling of a cyst more than one-third of the volume could obviously influence ostium patency. Therefore, simultaneous removal of the cyst with preservation of the periosteal layer of the SM membrane should be suggested during LSFE or via the use of transnasal functional endoscopic sinus surgery.⁵⁹ 

Unfortunately, there were not enough comparable studies and no randomized controlled trials available in our review. Some articles reported only 1 of the primary outcomes, and the sample size was not sufficient. Some studies did not report which SFE technique (OSFE or LSFE) was implemented. The key factors related to long-term implant survival, such as ESBG, were reported in only a few studies. Furthermore, different methods (CBCT, CT, and panoramic) were used to evaluate changes in the SM. However, the diagnostic accuracy of these modalities in the measurements of SM has not been evaluated in a comparative study. Therefore, additional studies with large sample sizes and long follow-up evaluations to determine the changes in the SM after SFE and to compare different radiographic techniques are still needed.

This review suggests that there is no significant difference in membrane perforation between normal sinus and pathologic sinus (mucosal thickening). The results of this study showed that SFE in the presence of SM thickening or pseudocysts did not increase the risk of implant failure rate. Within the limitations of the included studies, additional high-quality randomized controlled trials with large sample sizes and long follow-ups are needed.

This research was funded by the key scientific research platform and scientific research project (2018KTSCX186) of the ordinary universities of the Guangdong Provincial Department of Education and the project (201804015) of the Science, Technology, Industry and Information Bureau of Liwan District, Guangzhou.