Authors: Ibrahim Burak Yuksel, Sumeyye Celik Ozsoy, Muhammet Emin Arslan, Fatma Altiparmak, Hatice Yilmaz, Ali Altindag, Guldane Magat
Categories: Research, Maxillary sinus morphology, Anatomic variation, Surgical planning
Source: BMC Oral Health
Authors: Ibrahim Burak Yuksel, Sumeyye Celik Ozsoy, Muhammet Emin Arslan, Fatma Altiparmak, Hatice Yilmaz, Ali Altindag, Guldane Magat
The maxillary sinus may contain cortical bony partitions, called Underwood’s septae, which protrude into the sinus cavity and can subdivide it into multiple compartments. These structures are often shaped like inverted gothic arches and may be overlooked on routine imaging. However, their presence can significantly affect surgical access and outcomes, particularly in posterior maxilla procedures. Early identification of such anatomical variations contributes to safer and more predictable treatment planning.
This study aimed to determine the prevalence and anatomical characteristics of maxillary sinus septae in a Turkish adult population by evaluating digital panoramic radiographs.
Panoramic radiographs of 1000 patients (587 females and 413 males), aged between 18 and 70 years, were retrospectively analyzed. The sample consisted of patients who visited the Department of Oral and Maxillofacial Radiology at Necmettin Erbakan University Faculty of Dentistry. Maxillary sinus septa are defined as bony extensions from the sinus walls that rise at least 2.5 mm high, as noted by Velasquez Plata et al. According to Underwood’s anatomical descriptions, their position within the sinus cavity is categorized as anterior, middle, or posterior. These established criteria support the reliable identification and localization of septa, which is crucial for evaluating surgical risks and planning interventions in the maxillary sinus. All findings were recorded based on gender, lateralization (unilateral/bilateral), crest status (dentate/edentulous), and side (right/left). Statistical analyses were performed using SPSS version 21.0, with significance at p < 0.05.
Among the 1000 radiographs, 529 revealed at least one maxillary sinus septum, yielding a prevalence rate of 52.9%. A total of 748 septa were identified, with unilateral septa in 385 patients and bilateral in 144. In 587 female patients (1174 sinuses), 428 septa were detected, while in 413 male patients (826 sinuses), 320 septa were observed. The prevalence was slightly higher in males (53.75%) compared to females (52.29%). Most septae were located in the posterior region, followed by the middle and anterior regions. The statistical analysis confirmed the findings were significant (p < 0.05).
Digital panoramic radiography is valuable for identifying anatomical variations such as maxillary sinus septae. Preoperative recognition of these structures is crucial in reducing surgical risks, particularly in sinus lift and implant procedures. Radiologists and oral surgeons should be vigilant in detecting septae to ensure safe and effective treatment planning. Ethical approval was obtained from the Ethics Committee of Necmettin Erbakan University, Faculty of Dentistry (Date: 31.10.2024; Decision No: 2024/495).
The maxillary sinus is a pyramidal-shaped air cavity located within the body of the maxilla, bounded by the alveolar process, the facial surface, and the orbital floor. Its morphology, volume, and wall thickness demonstrate substantial individual variation and may differ bilaterally within the same subject. Anteriorly, the sinus typically extends to the level of the canine and premolar teeth. At the same time, its floor often reaches the deepest point in the first molar region due to its convex anatomy [1].
One of the notable anatomical variations within the maxillary sinus is the presence of septa-cortical bony partitions that arise from the sinus walls. These structures, commonly called Underwood’s septae, may partially or entirely divide the sinus cavity and are characterized by diversity in location, height, number, and angulation [2]. Morphologically, septa are classified as either primary, originating during embryogenesis, or secondary, which may form in response to postnatal changes such as tooth loss, bone resorption, trauma, or surgical intervention [3]. Depending on their extent, septa may significantly alter the internal architecture of the sinüs. Maxillary sinus septa (MSS) represent clinically significant anatomical structures that can complicate surgical procedures involving the posterior maxilla, particularly sinus floor elevation and implant placement. Their presence has been correlated with an increased risk of Schneiderian membrane perforation, compromised primary implant stability, and potential damage to the posterior superior alveolar artery, which may lead to intraoperative hemorrhage or hematoma formation. Moreover, septa can modify the morphology of the sinus floor, resulting in alveolar crest irregularities that impede surgical access and compromise prosthetic alignment [4]. Such variations frequently necessitate modifications to established implant protocols, encompassing alterations in implant angulation or the application of patient-specific surgical guides, particularly when supported by three-dimensional imaging [5].
The accurate preoperative identification and characterization of these septa are crucial not only for minimizing intraoperative risks but also for attaining predictable, prosthetically driven treatment outcomes [6]. Although cone-beam computed tomography (CBCT) provides superior spatial resolution and enables precise assessment of sinus anatomy, its routine implementation may be restricted by factors such as cost, radiation exposure, and limited accessibility in general practice. In contrast, panoramic radiography remains a widely available and cost-effective imaging modality, particularly for large-scale epidemiological studies and initial anatomical screening [7]. Despite its inherent limitations as a two-dimensional technique, panoramic imaging can still reveal radiopaque projections consistent with sinus septa, especially when these structures are sufficiently prominent [8]. Accordingly, the present study expands upon previous research by analyzing a large Turkish adult cohort using panoramic imaging, thereby contributing clinically relevant, region-specific data for surgical planning in general practice settings [9, 10].
This study aimed to investigate the prevalence, anatomical distribution, and morphological features of maxillary sinus septa in a Turkish adult population using panoramic radiographs. Additionally, the relationship between septum presence and variables such as sex, age, crest status, and sinus laterality was evaluated to provide relevant data that may contribute to improved planning of maxillary sinus interventions.
This retrospective radiographic study was conducted at the Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Necmettin Erbakan University. Ethical approval was granted by the Institutional Review Board of the same institution (Date: 31.10.2024; Decision No: 2024/495). The study was conducted in accordance with the principles of the Declaration of Helsinki.
Panoramic radiographs of patients aged between 18 and 70 years, obtained between January 2023 and January 2024 for various diagnostic purposes, were included in the evaluation. All panoramic radiographs were acquired using the Planmeca ProOne^®^ unit (Planmeca Oy, Helsinki, Finland), operated at 68 kVp, 7 mA, with an exposure time of 10 s. Images were captured by a single trained technician following the manufacturer’s recommended protocol. Radiographic data were processed and reviewed using Romexis^®^ imaging software (Planmeca, Helsinki, Finland). Standardization was achieved during the initial image optimization phase by adjusting brightness, contrast, and filtration parameters before evaluation. Radiographs were included only if they demonstrated sufficient diagnostic quality to visualize the maxillary sinus region.
Exclusion criteria comprised radiographs with inadequate visibility of the sinus floor, poor image quality, pronounced positioning errors, the presence of motion artifacts, or incomplete patient demographic information. Patients younger than 18 or older than 70 were excluded from the study.
A single oral and maxillofacial radiologist (IBY) assessed each radiograph with seven years of clinical experience. All evaluations were performed on anonymized panoramic images, which were randomly selected and presented to the examiner blinded without access to any clinical or demographic information. To ensure consistency and assess intra-observer reliability, 100 randomly selected images were re-evaluated by the same observer after a three-week interval. The intraclass correlation coefficient (ICC) for detecting MSS was calculated as 0.88, indicating good reproducibility of the radiographic observations.
A priori power analysis was conducted to determine the minimum required sample size for detecting a medium effect size (Cramér’s V = 0.3) using a chi-square test of independence, with a significance level of 0.05 and a statistical power of 0.90. The analysis indicated that a minimum of 117 participants would be necessary to achieve sufficient power for evaluating the association between sex and the presence of MSS. Both the right and left maxillary sinuses were assessed separately. For each case, MSS’s presence, number, and localization were recorded. Septa were bony projections extending at least 2.5 mm from the sinus wall. For dentate patients, sinus localization was classified into three regions according to the protocol described by Kim et al. (2006): anterior (mesial to the distal aspect of the second premolar), middle (between the distal aspect of the second premolar and the distal aspect of the second molar), and posterior (distal to the distal aspect of the second molar) [11] (Fig. 1). In patients with missing posterior teeth where anatomical landmarks were absent, septal location was determined using the method described by González-Santana et al. and Rancitelli et al. [12] (Fig. 2). This approach involved measuring the maximum anteroposterior dimension of the sinus on the panoramic image, dividing the measurement into four equal segments. The central half was designated the middle region, while the anterior and posterior quarters were defined accordingly.
Fig. 1Method used to determine the position of the maxillary sinus septum in dentate patients (Kim et al., 2006)
Fig. 2Classification of maxillary sinus septa in edentulous patients (based on González Santana et al. and Rancitelli et al.)
All collected data were meticulously entered into Microsoft Excel and analyzed using SPSS version 21.0 (IBM Corp., Armonk, NY, USA). Descriptive statistics, including frequencies and percentages, were calculated for categorical variables such as sex, age group, crest status (dentate or edentulous), presence of MSS, laterality (unilateral or bilateral), and anatomical localization (anterior, middle or posterior).
To evaluate potential associations between the presence of MSS and the aforementioned categorical variables, chi-square (χ²) tests of independence were employed. In instances where statistically significant associations were identified, Cramér’s V was used to determine the strength of association. A Cramér’s V value approaching zero was interpreted as indicative of a negligible association. Post hoc pairwise comparisons were conducted when appropriate, and Bonferroni corrections were applied to adjust for multiple testing. All statistical tests were performed at a significance level of p < 0.05.
The present study evaluated 1000 individuals, including 587 females and 413 males, to determine the prevalence and distribution of MSS. Radiographic analysis revealed the presence of MSS in 529 participants, corresponding to a prevalence rate of 52.9%. Among those identified with MSS, 307 were female (52.29%) and 222 were male (53.75%). Although the occurrence appeared slightly higher in males, statistical analysis demonstrated no significant association between septum presence and sex (χ² = 0.242, p = 0.697; Cramer’s V = 0.012), suggesting that gender does not play a substantial role in MSS development (Table 1).
Table 1This table presents the distribution of MSS among 1,000 individuals, categorized by gender. A total of 529 participants (52.9%) were found to have MSS, with 307 females (52.29%) and 222 males (53.75%) affectedGenderTotal IndividualsIndividuals with MSSMSS Prevalence (%) Female 58730752.29 Male 41322253.75 Total 100052952.9
Regarding septum laterality, unilateral septa were significantly more frequent than bilateral ones. Out of the 529 MSS-positive individuals, 385 (72.7%) exhibited unilateral septa, while only 144 (27.3%) presented with bilateral involvement. When sex-specific distributions were assessed, similar patterns were observed across both 221 of 307 females (72%) and 164 of 222 males (73%) had unilateral septa. Notably, in these unilateral cases, a left-sided predominance was recorded for both sexes. Specifically, 107 females had left-sided septa compared to 114 on the right, and 93 males had left-sided involvement versus 71 right-sided cases. These findings indicate a slight, yet consistent, lateral preference across genders (Table 2).
Table 2Gender-based distribution of unilateral and bilateral maxillary sinus septa (MSS). Unilateral septa were predominant across both sexesGenderTotal with MSSUnilateral SeptaBilateral Septa Female 30722186 Male 22216458 Total 529385144
Age-stratified analysis revealed that the 18–44 years cohort constituted the majority of MSS-positive individuals (n = 412; 77.8%), followed by the 45–60 years group (n = 96; 18.3%) and the 61 + years group (n = 21; 3.9%). In all age brackets, unilateral septa remained more prevalent than bilateral ones. However, statistical comparison between age groups and septum laterality failed to show any meaningful correlation (χ² = 0.616, p > 0.05), indicating that age may not significantly influence the laterality of MSS (Fig. 3) (Table 3).
Fig. 3Polynomial trend illustrating septum detection across age groups. A steady decline in frequency is observed with increasing age, especially beyond 45 years. The shaded area represents the 95% confidence interval
Table 3Distribution of patients with MSS across age groups. The highest prevalence was observed in younger adults (18–44 years), with a decreasing trend in older populationsAge Group (Years)Number of Patients with MSS 18–44 412 45–60 96 61–70 21
Additionally, the presence of MSS was examined concerning the status of the crest. 290 (54.9%) of the 529 MSS-positive individuals were classified as dentate, while 239 (45.1%) were toothless. When bilateral septa were explicitly examined (n = 144), 84 individuals (58.4%) maintained their dentition, while 60 (41.6%) were edentulous. For unilateral cases, the distribution of denture status was somewhat symmetrical between the right and left sides. Furthermore, there was no statistically significant correlation detected between the crest status and septum laterality (χ² = 0.362, p > 0.05) (Fig. 4). These findings suggest that tooth loss, while anatomically relevant, may not possess a statistically significant impact on septum laterality or prevalence.
Fig. 4Distribution of maxillary sinus septa (MSS) and bilateral septa by crest status
Seven hundred forty-eight distinct septa were detected within the 529 MSS-positive patients, indicating that some individuals exhibited multiple septal formations. Anatomical mapping showed that the central region of the maxillary sinus was the most commonly affected site, accounting for 458 septa (61.2%). The posterior sinus region followed with 183 septa (24.5%), while the anterior region had the lowest occurrence, with only 107 septa (14.3%). This distribution substantiates the proposition that septa exhibit a tendency to localize preferentially within the central region of the sinus, an area of clinical significance due to its proximity to frequently employed implants and sinus lift procedures zones (Fig. 5).
Fig. 5Anatomical density map of septa across sinus regions. Color intensity indicates frequency of septa, with the middle region demonstrating the highest density
Additional comparative analysis was carried out to examine the distribution of paired configurations of septa, such as unilateral versus bilateral and right-sided versus left-sided. Statistically significant differences were noted between unilateral and bilateral arrangements and between unilateral and right-sided configurations (p < 0.05). However, other pairwise comparisons did not reach statistical significance (Fig. 6). These findings underscore unilateral septa’s predominance and distinct clinical behavior relative to different variations.
Fig. 6Pairwise comparisons of septum categories. Blue lines indicate statistically significant comparisons (p < 0.05), while maroon lines denote non-significant comparisons. Node sizes represent relative category frequency
Advanced graphical methods were employed to elucidate the interrelationships among several septal groups. A hexagonal significance map (Fig. 6) was created to graphically depict the degree and direction of statistical relationships among the septum types. This visualization employed blue connectors to denote statistically significant correlations and maroon lines representing non-significant associations. The size of the nodes corresponded to the frequency of categories, highlighting the prevalence of unilateral septa throughout the sample.
A chord diagram was developed to elucidate interaction intensities and directional tendencies among septum classifications (Fig. 7). The chord structure distinctly elucidated the significant role of unilateral septa within the dataset, as these structures exhibited stronger interrelations with other anatomical and demographic variables than their bilateral or side-specific counterparts. Such visual tools provide enhanced interpretability for complex data patterns and may assist clinicians in the processes of risk stratification and surgical decision-making.
Fig. 7Chord-style diagram illustrating the distribution and transition patterns between septum classification types. Unilateral septa exhibited the strongest interactions
The findings of the present study contribute to the growing body of literature regarding MSS morphology, specifically through the utilization of panoramic radiography within a substantial cohort from Turkey. Notably, with a prevalence of septa at 52.9%, our results significantly surpass the prevalence reported in certain studies employing CBCT. For instance, Furtado et al. (2021) documented a prevalence of 26.8% in a Brazilian population utilizing multislice CT, while Dhami et al. (2021) identified a prevalence of 49.09% using CBCT in Nepal [13, 14]. This discrepancy may be attributed to variations in ethnic origin, diagnostic thresholds, and the resolution of imaging modalities. It is crucial to note that while CBCT provides enhanced three-dimensional visualization, our findings reaffirm that panoramic imaging continues to serve as a valid screening instrument. This supports the conclusions articulated by Costa et al. (2023), who illustrated that panoramic radiographs, despite being two-dimensional, effectively delineate the spatial relationships between the sinus floor and adjacent dental structures [15].
Our finding of a significant dominance in the central (middle) region (61.2%) aligns with previous studies. Furtado et al. (2021) found the highest occurrence of septa in the first and second molar regions (44.2%), which corresponds to the middle zone. Likewise, Dhami et al. (2021) reported that 44.44% of septa are found in the middle region, reinforcing our conclusion that this area is the most frequently affected, regardless of the population studied [14].
Sex-based comparisons in our study demonstrated no significant association between the presence of MSS and gender, which aligns with the findings of Furtado et al., and Bhageshwar Dhami et al., who similarly did not identify substantial sex-related differences. The identified left-sided predominance in unilateral MSS within our cohort represents a novel finding that has not been consistently documented in prior studies, thereby meriting further investigation [13, 14].
Comparative analyses of age within our dataset revealed a greater prevalence of MSS in younger adults, specifically those aged between 18 and 44 years. This finding stands in contrast to the anticipated progressive pneumatization and osseous remodeling associated with advancing age. Nevertheless, Johnny et al. (2022) acknowledged the considerable anatomical diversity present in sinus morphology among younger adults, a factor that may elucidate these observations. Our research corroborates this anatomical variability, thereby highlighting the necessity for tailored radiographic assessments, particularly within younger demographics where surgical interventions might be contemplated [16].
The analysis of crest status revealed no significant correlation with the presence or laterality of MSS. This observation is partially consistent with the findings of Ocak et al. (2020), who established a relationship between tooth loss and increased sinus pneumatization, yet did not specifically investigate the formation of MSS. Our findings indicate that, while post-extraction alveolar resorption does influence sinus morphology, it does not seem to substantially affect the presence or distribution of MSS [17].
Our study utilized advanced visualization tools, including pairwise comparisons and network diagrams, which facilitated a detailed interpretation of MSS patterns. The prevalence of unilateral MSS and their more significant interaction profiles highlight their clinical importance. These visualization techniques align with the classification framework suggested by Johnny et al. [16], emphasizing the diversity of sinus architecture even when assessed with panoramic imaging.
In summary, our findings underscore the diagnostic utility of panoramic radiography in evaluating MSS and provide comparative insights in light of recent literature. Despite the inherent limitations associated with two-dimensional imaging, panoramic radiographs continue to serve as a practical and accessible modality for preliminary sinus assessment and preoperative planning, especially in resource-constrained environments [18]. Further studies incorporating CBCT may help validate and refine these observations, particularly regarding subtle morphological variations and their clinical implications.
Panoramic radiography remains an effective and readily available imaging modality, providing significant clinical benefits for evaluating MSS and surrounding structures [19]. Due to its cost-effectiveness, ease of use, and common presence in dental clinics, it continues to be a key imaging method for diagnostic and preoperative needs [20].
Recent investigations have further emphasized its applicability in anatomical research and clinical evaluation. For example, Johnny et al. (2022) [16] introduced a new classification system for the maxillary sinus based on digital panoramic images, highlighting the modality’s scalability in enabling large-sample studies that facilitate the analysis of sinus configurations, including the presence of septa. In a comparative study, Ferrari and Hasna (2022) [21] evaluated the reliability of panoramic radiographs versus CBCT in determining the proximity between dental root apices and critical anatomical landmarks. Their results demonstrated that, in several anatomical regions, panoramic imaging yields measurements within clinically acceptable margins, thereby reinforcing its value in settings where immediate or cost-conscious diagnostics are necessary. Furthermore, Costa et al. (2023) [15] explored the spatial relationships between maxillary molar and premolar roots and the sinus floor, emphasizing the relevance of such assessments in implant planning and surgical risk evaluation. Unlike CBCT, panoramic radiography exposes patients to relatively low radiation levels, contributing to its preference in routine clinical practice, especially when comprehensive 3D imaging is not essential.
The current body of literature highlights the clinical importance of various maxillary sinus floor contour classifications, particularly concerning their application in surgical planning for sinus floor elevation. For example, Johnny et al. (2022) [16] proposed a novel classification system derived from panoramic radiographic observations that categorizes the sinus floor into distinct contours, including scalloped, wavy, curved, straight, and V-shaped types [22]. Gao et al. (2023) [23] further demonstrated that the sinus floor contour is pivotal in predicting implant apical non-coverage in bone-added transcrestal sinus floor elevation, advocating for customized surgical approaches based on the observed anatomical features. Similarly, Stacchi et al. (2022) [24] identified that variations in the sinus floor shape correlate with peri-implant bone remodeling outcomes following osteotome sinus floor elevation, indicating that different morphological patterns may require distinct operative techniques to optimize graft stability and implant integration.
These studies collectively highlight that thorough preoperative evaluation of sinus floor contour, even with commonly available methods such as panoramic radiography, can improve surgical predictability and customize treatment planning to individual anatomical differences, thereby decreasing complication rates and enhancing clinical outcomes [25]. In line with this perspective, our study was conducted with the recognition that the morphological characteristics of MSS, their vascular associations, and anatomical variations are critical considerations in sinus surgeries. These factors should be carefully evaluated to minimize intraoperative risks and to guide appropriate surgical strategies [25].
Recent studies have focused on maxillary sinus symmetry and asymmetry, providing valuable insights into their clinical implications for treatment planning and surgical procedures [26]. Takeda (2025) [27] introduced a classification system for sinus floor contours, identifying five distinct morphological narrow tapered, tapering, ovoid, square, and irregular. This typology reflects the anatomical diversity of the sinus and its possible asymmetry concerning facial structures. Rosso et al. (2022) reviewed cases of silent sinus syndrome, a condition characterized by unilateral maxillary sinus collapse often associated with enophthalmos and hypoglobus, contributing to midfacial asymmetry [28]. Ahn et al. (2023) [29] examined patients with skeletal Class III facial asymmetry and found that the left and right maxillary sinus volumes were similar despite notable external asymmetry. This suggests that internal sinus dimensions may not always correlate with external facial contours. Supporting this, Park et al. (2021) reported that nasal floor slanting is linked to deviations in neighboring anatomical structures as an indirect indicator of nasofacial asymmetry. These findings demonstrate that although maxillary sinuses are often bilaterally symmetrical, certain pathological or developmental factors can lead to asymmetry. This is particularly relevant during preoperative assessment and planning for interventions such as sinus floor elevation, where anatomical precision is essential [29].
In addition to volumetric and morphological asymmetries of the maxillary sinus, recent imaging studies have highlighted the impact of bone-related factors on radiographic interpretation. Ketabi et al. (2024) [30] demonstrated that the visibility of maxillary sinus septa on panoramic radiographs is significantly influenced by the surrounding cortical bone thickness, particularly in cases where anatomical asymmetry exists. Their findings suggest that unilateral increases in cortical density or sinus wall angulation may obscure septal projections, leading to false-negative interpretations on two-dimensional imaging. This is especially important for patients with subtle or localized sinus asymmetry, where the internal bony structures may not align with the external facial contours. In this context, the findings of the current study highlighting regional differences in the visibility and localization of the septum underscore the necessity of considering asymmetrical anatomical structures during radiographic evaluations. Recognizing these limitations is crucial when using panoramic imaging for preoperative planning, as neglecting unilateral septa due to asymmetrical bone masking could compromise the accuracy of surgical procedures.
This study has some limitations. As a two-dimensional method, panoramic radiography is affected by magnification, distortion, and overlapping structures, which may limit accurate identification of MSS. CBCT validation was not possible, as corresponding scans were unavailable. The retrospective design and single-center setting may limit generalizability, and assessments were performed by a single observer. Future studies using CBCT in multi-center, prospective designs are recommended to enhance diagnostic accuracy.
This study substantiates the diagnostic value of panoramic radiography in evaluating sinus-related anatomical variations, particularly in contexts where CBCT is not routinely accessible. The significant prevalence of MSS at 52.9%, alongside their predominant localization within the middle sinus region at 61.2%, underscores the anatomical intricacy of the posterior maxilla and its surgical implications. While CBCT continues to be recognized as the gold standard for three-dimensional assessment, the concordance of our findings with those from CBCT-based studies reinforces the credibility of panoramic imaging. Moreover, although the interpretation of CBCT requires advanced technical expertise and specialized evaluation, panoramic radiographs enable clinicians possessing adequate anatomical understanding to efficiently identify such variations in routine practice. In light of its accessibility, cost-effectiveness, and reduced radiation exposure, panoramic radiography remains an instrumental resource for mitigating surgical risks during preoperative planning. It is advisable that future prospective investigations incorporate CBCT validation and encompass more diverse populations to enhance diagnostic precision and clinical relevance.