© The authors, 2025

Defects in the posterior regions of the maxillary dental arch often result in a significant decrease in masticatory efficiency and pose a major challenge for effective prosthetic rehabilitation. Depending on the size of the defect and the condition of adjacent teeth, various prosthetic options can be considered, including removable dentures, tooth-supported fixed prostheses, or implant-supported restorations. Among these, implant-supported prostheses are regarded as the treatment of choice for rehabilitating patients with posterior maxillary defects, as supported by numerous scientific studies [1–3].

The availability of sufficient alveolar bone volume is a fundamental prerequisite for the placement of dental implants intended to serve as abutments for subsequent prosthetic rehabilitation. In the rehabilitation of posterior maxillary segments with implant-supported restorations, inadequate bone volume frequently constitutes a major limitation. This is primarily attributable to the anatomical proximity of the maxillary sinus and the progressive alveolar ridge resorption that ensues following tooth extraction. In the majority of cases, vertical bone deficiency in the posterior maxilla necessitating augmentation is managed through lateral or transcrestal sinus floor elevation techniques. These approaches are well-established within the framework of evidence-based practice, demonstrate consistently high clinical efficacy, and, in selected cases, may be performed concurrently with implant placement [4–7].

At the same time, the presence of defects in the floor of the maxillary sinus significantly compromises the conditions required for successful dental implant placement. Such defects may develop following traumatic tooth extractions or extractions preceded by periodontal pathology that resulted in perforation of the sinus floor [8,9]. Less commonly, these defects arise as a consequence of tumor resection, traumatic injuries (including gunshot wounds), or as a result of debridement procedures related to osteonecrosis of various etiologies [10,11]. These conditions often preclude or substantially complicate implant placement, thereby necessitating the use of various bone augmentation techniques or the selection of alternative prosthetic rehabilitation strategies.

Nevertheless, the literature contains only a limited number of publications analyzing the outcomes of implant-supported prosthetic rehabilitation in cases involving defects of the maxillary sinus floor [12]. These studies vary in their criteria for assessing such defects and in the protocols employed for dental implant placement. In our view, the planning of implant placement under such complex conditions should begin with a thorough evaluation of the topographic and anatomical characteristics of the sinus floor defects and their potential impact on the selection of appropriate augmentation or implantation protocols. Accordingly, the aim of the present study was to investigate the topographic and anatomical features of maxillary sinus floor defects and propose a topographic-anatomical classification of sinus floor defects based on their linear dimensions and wall preservation, and to evaluate its potential role in guiding the selection of augmentation techniques and timing of dental implantation.

Materials and methods

To address the stated objective, a retrospective cohort study was conducted to compare the baseline topographic and anatomical characteristics of maxillary sinus floor defects. The study included 91 patients who sought dental implant placement for the rehabilitation of posterior maxillary edentulous segments at the Department of Maxillofacial Surgery and Innovative Dentistry, Bogomolets National Medical University, between January 1, 2021, and December 31, 2024. The study protocol was reviewed and approved by the Bioethics and Research Ethics Committee of Bogomolets National Medical University (Protocol No. 163, dated November 7, 2022).

The inclusion criteria for the study were as follows: the presence of a maxillary sinus floor defect accompanied by a dental arch defect classified as Kennedy Class I, II, or III on the corresponding side of the maxilla, or complete edentulism with preservation of the hard palate; availability of computed tomography (CT) data; and a signed written informed consent form for participation in the study.

Exclusion criteria included patients under the age of 18; individuals with maxillary defects resulting from total or partial maxillectomy involving resection of the maxillary sinus walls; patients with maxillary sinus floor defects associated with active oroantral communications or fistulas; cases with incomplete clinical or tomographic documentation; inability to accurately determine the true geometric dimensions of the defect; insufficient data for analysis; and withdrawal of consent to participate in the study. Isolated sinus floor defects in the region of the maxillary third molars were excluded from consideration due to the lack of prosthetic rehabilitation needs in this area.

All patients underwent a standardized examination protocol that included medical history collection, assessment of general and local clinical status, utilization of additional diagnostic methods, followed by definitive diagnosis and treatment planning. For each patient included in the study, the following parameters were recorded: age, sex, localization of the sinus floor defect in relation to the missing teeth, defect dimensions (in the vestibulo-oral and mesiodistal directions), defect area, maximum residual bone height, and the etiology of the defect (post-extraction, post-debridement, post-resection, or post-traumatic).

To assess the geometric characteristics of the defects, 3D visualization and computer modeling techniques based on computed tomography data were employed (Fig. 1). For this purpose, either multislice or cone-beam CT scans were obtained and subsequently analyzed using Mimics Medical 23.0 software (Materialise, Belgium).

Tomographic data were processed as follows: series of tomographic files in DICOM format were imported into the aforementioned software environment. All slices were examined in three orthogonal planes (axial, coronal, and sagittal), with additional evaluation performed using bone window contrast settings.

Using automatic segmentation based on the radiodensity of bone tissue (within the range of 200–3000 Hounsfield units, HU), a virtual model of the maxilla was generated. Subsequently, semi-automatic segmentation of the defect was performed in a separate “mask” to enable reconstruction of the continuity of the alveolar ridge of the maxilla.

When necessary, additional virtual models were created using a mirroring algorithm of the unaffected side of the maxilla (“virtual donor” method). Based on this, a virtual model of the defect was constructed (Fig. 2), followed by analysis of its linear dimensions and cross-sectional area at the level of the maxillary sinus floor.

Statistical analysis included the calculation of means (M) and standard deviations (±) for parametric data, as well as medians (Me) and interquartile ranges (IQR) for non-parametric data. To compare differences between groups, one-way analysis of variance (ANOVA) was performed for data with a normal distribution, while the Kruskal–Wallis test was used for data not meeting normality assumptions. Linear regression analysis was applied to assess the overall influence of independent variables on the outcome variable. For variables demonstrating statistically significant effects, the strength of association was evaluated using Pearson's correlation. Spearman's rank correlation was used to identify associations among categorical variables. A p-value of less than 0.05 was considered statistically significant. All statistical analyses were performed using R software (version 4.2.2, R Core Team).

Fig. 1 Cone-beam CT scans illustrating maxillary sinus floor defects classified according to the proposed topographic-anatomical system. A. Type I: Small defects surrounded by four bony walls, with at least one linear dimension <5mm. B. Type II: Larger four-wall defects, where all linear dimensions exceed 5mm. C. Type III: Defects lacking one wall, typically the buccal wall; less frequently the palatal wall. D. Type IV: Extensive defects with only one or two remaining bony walls.

Fig. 2 Virtual model of a maxillary sinus floor defect created using Mimics Medical 23.0 software, used for analysis of defect linear dimensions and cross-sectional area.

Results

The age of the patients included in the study ranged from 25 to 83 yr, with a mean age of 49.4±12.9 yr. The majority of patients belonged to the working-age population between 20 and 50 yr. The analyzed defects originated from the following causes: tooth extraction — 67 patients (73.6%), neoplasm removal — 5 patients (5.5%), high-energy trauma (including gunshot and blast injury) — 6 patients (6.6%), and debridement surgery involving the maxillary sinus — 13 patients (14.3%) (Fig. 3).

Periodontitis and its complications — particularly odontogenic sinusitis — were the leading causes of tooth extraction in our patient cohort, ultimately contributing to the formation of sinus floor defects.

The mesiodistal extent of the defects ranged from 1.4 mm to 45.7 mm, with a median value of 5.0 mm (IQR: 2.8–7.2 mm). The median dimension of the defects in the vestibulo-oral direction was 4.9 mm (IQR: 2.6–6.2 mm), with a range from 0.9 mm to 16.0 mm. The defect area varied between 0.8 mm² and 251.3 mm², with a median of 17.1 mm² (IQR: 6.1–32.1 mm²). The residual bone height of the alveolar ridge surrounding the defect ranged from 0.25 mm to 14.2 mm, with a median value of 5.5 mm (IQR: 3.7–6.7 mm) (Tab. I).

Most defects were located in the region of the first maxillary molars (74.7% of cases), followed by the second molar region (25.3%). Defects in the area of missing second premolars were the least common, observed in 8.9% of cases. In 7.7% of cases, defects extended over areas corresponding to multiple missing teeth.

A moderate positive correlation was observed between the etiology of the defect and the number of teeth involved (R = 0.33, p = 0.001). Additionally, a statistically significant association was found between defect area and traumatic etiology (p < 0.001), while no significant associations were identified between defect area and other etiologies (p > 0.05) (Fig. 4). A moderate correlation was also found between traumatic origin and defect area (R = 0.38, p = 0.001).

Based on topographic and anatomical characteristics, all defects were classified into four types (Fig. 5). Type I included small defects surrounded by four bony walls (buccal, palatal, mesial, and distal), with at least one linear dimension measuring less than 5 mm. This threshold was selected based on clinical considerations, as it corresponds approximately to the maximum diameter of most commercially available dental implants intended to achieve reliable osseointegration. Type II defects also had four bony walls (buccal, palatal, mesial, and distal), but all linear dimensions exceeded 5 mm. Type III defects were characterized by the absence of one of the four walls, most commonly partial or complete loss of the buccal wall, less frequently the palatal wall. Type IV included the largest defects, surrounded by only one or two remaining bony walls.

Within the studied group of maxillary sinus floor defects, Type I defects accounted for 37.4% (n = 34). The median cross-sectional area for this group was 5.4 mm² (IQR: 3.1–9.0 mm²), while the residual alveolar bone height surrounding the defect ranged from 1.0 to 9.5 mm, with a mean value of 4.4±2.3 mm.

Type II defects were identified in 13.2% of cases (n = 12). The median cross-sectional area for this group was 29.1 mm² (IQR: 22.4–38.4 mm²), and the residual bone height ranged from 1.0 to 12.0 mm, with a mean of 6.2±3.2 mm.

Type III and Type IV defects were observed in 30.8% (n = 28) and 18.6% (n = 17) of patients, respectively, out of the total sample (n = 91). Detailed data on defect area and residual alveolar bone height are presented in Table I.

According to the conducted analysis, the defect types demonstrated a statistically significant progressive increase in cross-sectional area from Type I to Type IV (p < 0.001). Additionally, significant differences were observed between defect types in terms of their mesiodistal and vestibulo-oral dimensions (p < 0.001), as well as in their extension toward the buccal wall of the maxillary sinus (p < 0.001). In contrast, residual bone height at the defect site did not differ significantly across the defect types (p >0.05).

Fig. 3 Etiological distribution of 91 maxillary sinus floor defects categorized based on clinical history (e.g., extraction, trauma, or surgical debridement).

Fig. 4 Association between the cross-sectional area of maxillary sinus floor defects (mm2) and their etiology, based on regression analysis.

Fig. 5 Schematic illustration of maxillary sinus floor defects classified according to the proposed system.

Discussion

The rehabilitation of posterior edentulous segments in the maxilla remains a topic of clinical and scientific discussion, despite the well-established and evidence-based recognition of dental implants as a highly effective solution in such cases. Nevertheless, the success of implant therapy in the posterior maxilla is closely dependent on the availability of sufficient bone volume for predictable and stable implant placement. In this regard, the presence of the maxillary sinus and its topographic and anatomical characteristics may present significant limitations to implant placement in individual clinical scenarios.

In cases of alveolar ridge atrophy resulting from tooth loss, additional interventions such as bone augmentation or modifications of the implant placement protocol may be required to compensate for the limited bone availability. Importantly, the already compromised anatomy of the posterior maxilla due to ridge resorption is often further exacerbated by the presence of combined defects involving both the alveolar process and the floor of the maxillary sinus.

Numerous classifications of the residual alveolar ridge height in the posterior maxilla have been proposed to justify the need for and guide the choice of subantral augmentation techniques — including systems described by Tatum [13], Juodzbalys & Kubilius [14], and Niu et al. [15], Wang & Katranji [16]. However, none of the currently available classifications address defects of the maxillary sinus floor that arise as a result of molar extractions, debridement procedures, or traumatic injuries. These types of defects may require surgical decisions that fall outside the scope of conventional classification systems.

According to our review of the literature, only a single study — that by Wen Y et al. [12] — retrospectively analyzed the outcomes of sinus floor augmentation in patients with such defects. Despite the high clinical prevalence of these conditions, few studies have systematically investigated this category of patients.

The purpose of the proposed topographic-anatomical classification of sinus floor defects is to support its clinical application and enable a non-linear, individualized approach to selecting augmentation techniques.

Notably, in certain cases — such as Type I defects in our classification — despite limited residual bone height (M = 4.4 ± 2.3mm), the mesiodistal (Me = 2.8 mm, IQR: 2.0–4.0mm) and vestibulooral (Me=2.5mm, IQR: 2.0–3.3mm) dimensions of the defect may allow the placement of a dental implant with a diameter of 4.0mm or greater. In such situations, the vertical bone deficiency may potentially be addressed using closed sinus lift techniques, even though the majority of existing classifications recommend lateral window sinus floor elevation and even delayed dental placement in cases with comparable residual bone height [13–16]. This approach deviates from conventional protocols and is being further investigated in the present and subsequent studies.

According to our findings, slightly less than half of the patients (49.5%) presented with defects smaller in cross-sectional area than the surface of standard commercially available dental implants. However, only 15.4% of patients demonstrated conditions suitable for achieving primary implant stability, thereby rendering the site potentially viable for implant placement without prior augmentation.

The proposed classification may support clinical decision-making by stratifying patients according to the complexity of bone reconstruction required prior to implantation. For instance, Type I defects, although characterized by limited vertical height, may still allow for immediate placement using transcrestal sinus lift techniques, while Type IV defects typically necessitate staged bone augmentation. Thus, the classification may serve as a predictive tool for determining the likelihood of successful immediate implantation and the need for advanced grafting strategies.

Further clinical validation of this classification system is warranted to assess its prognostic value and to optimize treatment protocols in real-world settings. The proposed system does not replace existing and widely used classifications based on residual bone height, but rather is intended to complement them by supporting clinical decision-making in cases involving maxillary sinus floor bone defects.

The results of our study demonstrate that such combined defects most frequently arise following tooth extractions, with the majority localized in the region of the first maxillary molars. Their development is commonly facilitated by pre-existing periapical inflammatory processes extending toward the sinus floor and the use of traumatic or overly aggressive extraction techniques. The most extensive defects in terms of surface area were observed in patients with combat-related injuries, where bone loss resulted both from direct trauma caused by high-energy projectiles and from subsequent surgical debridement procedures performed during wound management.

In all other cases, the defects were associated with either loss of the crestal cortical plate of the maxillary alveolar ridge, insufficient residual bone height, or, in 31.4% of cases, involvement of one of the lateral walls (palatal or buccal). The median residual bone height along the defect periphery was 5.5 mm (IQR: 3.7–6.7 mm).

These findings suggest that, in most cases where dental implantation is considered in the presence of maxillary sinus floor defects, a combination of vertical and horizontal ridge augmentation is required, along with the reconstruction of the cortical continuity of the alveolar process. Therefore, subantral augmentation alone may not be sufficient to address the lack of available bone and should be considered as part of a broader reconstructive strategy — an aspect not addressed in most existing classifications, but explicitly included in the one we propose.

Defects extending over a span greater than two teeth were observed in 7.7% of cases and invariably involved at least one maxillary molar. The majority of such extensive defects were classified as Type IV. Among all analyzed patients, only 9.9% of cases allowed for rational prosthetic rehabilitation with implant support without requiring implant placement in the region of the maxillary sinus floor defect. In all other cases, these defects posed significant limitations to the feasibility of implant placement.

As previously noted, a subset of defects—particularly those classified as Type III—extended toward the anterior wall of the maxillary sinus, with a median vertical height of 3.8 mm (IQR: 3.1–4.7 mm). This anatomical configuration either precluded or substantially restricted the possibility of performing traditional open sinus floor elevation procedures. According to the findings of our prior study, the typical bone window area created during open sinus lift procedures measures approximately 36.04 mm² and is usually located at a height of around 4.3 mm from the sinus floor [17].

Another potential challenge to successful bone augmentation was the consistent presence of periosteal fusion between the alveolar ridge and the mucoperiosteum of the maxillary sinus in all observed cases. As a result, when vertical augmentation was indicated, It became evident that there was a necessity for alternative subantral augmentation techniques or the development of novel surgical approaches capable of overcoming the aforementioned limitations.

From a surgical standpoint, Type IV defects proved to be the most complex. Their presence effectively precluded the possibility of performing dental implantation and prosthetic rehabilitation using conventional methods. These defects were characterized by the greatest extent, area, and vertical height, invariably involving the loss of more than two teeth. They exhibited a complex three-dimensional geometry and a minimal amount of residual bone along the periphery, thereby necessitating consideration of various bone grafting strategies to enable their reconstruction.

Thus, maxillary sinus floor defects varied in terms of area, localization, extent, and their topographic and anatomical characteristics significantly complicated the planning of dental implant placement. In particular, the majority of cases were associated with the loss of crestal bone, which is critically important for achieving primary stability of dental implants in the posterior maxilla.

Conventional techniques for bone grafting and subantral augmentation have notable limitations when applied to this patient population. The clinical value of the proposed classification extends beyond identifying cases suitable for immediate implant placement. By stratifying defects according to their dimensions and the number of preserved bony walls, this system may support clinicians in selecting the most appropriate augmentation technique and estimating the risk of complications. For instance, although Type I defects are often associated with limited residual bone height, their confined geometry and intact walls may allow for transcrestal sinus elevation and immediate implantation. Notably, in 15.4% of cases in our study, bone augmentation was performed simultaneously with implant placement, significantly accelerating the rehabilitation process. Under conventional classification systems based solely on residual bone height, such clinical scenarios would have typically indicated lateral sinus floor elevation followed by delayed implantation. In contrast, the proposed classification may support more flexible and individualized treatment protocols. Extensive defects classified as Type III or IV typically lack sufficient anatomical support and require staged reconstruction with a higher risk of graft failure or reduced primary implant stability. In such cases, sinus floor elevation alone is often insufficient to provide the necessary bone volume and cortical continuity for successful implantation. Additional vertical and horizontal ridge augmentation may be required. This classification may thus contribute to improved surgical planning and more personalized treatment protocols in posterior maxillary implantology.

Conclusion

The proposed topographic-anatomical classification, based on the number of preserved bony walls and defect dimensions, may serve as a practical tool to assist clinicians in selecting augmentation techniques and timing of implantation, based not only on residual bone height but also on the morphology and wall integrity of the sinus floor defect.

Specifically, confined defects (Types I and II) may allow for transcrestal sinus lift and immediate implantation, even in cases with limited bone height. In contrast, extensive defects with wall loss (Types III and IV) typically require staged lateral window augmentation and may require additional alveolar ridge augmentation.

This classification system may thus help to individualize surgical decision-making, estimate procedural complexity, and stratify patients by surgical risk.

Further prospective studies are warranted to validate its prognostic value and define treatment algorithms based on defect type.

Funding

This work recieved no funding.

Conflicts of interest

The authors declare that there are no conflicts of interest regarding the publication of this paper.

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Author contribution statement

Artem Artemchuk conceived the study, contributed to the study design, performed clinical procedures and data acquisition/processing, carried out data analysis and interpretation, and drafted the manuscript. Bekir Osmanov contributed to the study concept and design, participated in clinical procedures and data collection/processing, assisted with analysis and interpretation, conducted the literature search, and contributed to writing and editing. Yurii Chepurnyi contributed to the study concept and design, participated in clinical procedures and data collection/processing, supervised the project, verified the analytical methods, critically revised the manuscript for important intellectual content, and approved the final version for submission. All authors read and approved the final manuscript and agree to be accountable for all aspects of the work, meeting the ICMJE authorship criteria.

Ethics approval

The research protocol was reviewed and approved by the bioethics committee of Bogomolets National Medical University, Kyiv, Ukraine (approval No. 163; November 07, 2022).

Informed consent

Written informed consent was obtained from all patients.

References

All Tables

All Figures

Fig. 1 Cone-beam CT scans illustrating maxillary sinus floor defects classified according to the proposed topographic-anatomical system. A. Type I: Small defects surrounded by four bony walls, with at least one linear dimension <5mm. B. Type II: Larger four-wall defects, where all linear dimensions exceed 5mm. C. Type III: Defects lacking one wall, typically the buccal wall; less frequently the palatal wall. D. Type IV: Extensive defects with only one or two remaining bony walls.

In the text

	Fig. 2 Virtual model of a maxillary sinus floor defect created using Mimics Medical 23.0 software, used for analysis of defect linear dimensions and cross-sectional area.
In the text

	Fig. 3 Etiological distribution of 91 maxillary sinus floor defects categorized based on clinical history (e.g., extraction, trauma, or surgical debridement).
In the text

	Fig. 4 Association between the cross-sectional area of maxillary sinus floor defects (mm2) and their etiology, based on regression analysis.
In the text

	Fig. 5 Schematic illustration of maxillary sinus floor defects classified according to the proposed system.
In the text