4.1│Sinus Septa Variations and Morphology
The present study findings found maxillary sinus septa in about two third of the patients (63.9%), and half of the sinuses exhibited septa (47%), which was noteworthy and equivalent to the literature and consistent with previous studies that showed a prevalence ranging from 9–70% in patients and 13–58% in sinuses [11–14, 20]. In the literature, the prevalence of maxillary sinus septa varies significantly between studies of various nations and ethnicities. China, Korea, Brazil, Iran, Saudi Arabia, Turkey, and South Africa reported rates of 46.9%, 22.93%, 27%, 44,4%, 44.8%, 45.9%, 47.6%, and 69%, respectively [21–28]. The sample size, criteria of septa (2–4 mm height), races and nations, and imaging modalities (CBCT, CT, panoramic, surgical observation, cadavers) could influence the different septa prevalence rates [29].
The septa were more prevalent unilaterally 33.7% (left 18% and right 15.7%) compared to bilateral (30.2%) in both sinuses. The maximum number of septa in a single sinus was higher in the left (5 septa) compared to the right (4 septa). Further, the septa were more prevalent in males than females, which is consistent with most previously published studies [4, 17]. However, Park et al. found the prevalence of septa to be higher among females [30].
In the present study, most septa were located on the floor with middle position and coronal orientation. About 43% were located in the middle region, 38.4% was located in the anterior region, and 18.6% were located in the posterior region. The most common orientations of the septa were coronal (66%), followed by sagittal (33.6%), and axial (0.4%) directions. The present study is in line with the previous studies that reported a majority of septa were oriented in the coronal/buccopalatal /transverse direction in a range of 61.8–87.6% [31].
The presence of maxillary sinus septa was significantly higher in the middle region in both the right and left sinuses, agreeing with systematic review studies that found the septa were positioned in the middle region [11–13, 20, 32], corresponding to first and second molars [22, 32] in the range between 35.7–70.94% and divided the sinus into two or more sections with a coronal orientation. However, some authors have demonstrated a higher rate of septa coincidence in the posterior molars [2, 3]. Nevertheless, our results are inconsistent with some authors who found that the highest rate of maxillary septa prevalence was located anteriorly [7, 33].
These results reveal the clinical importance for implant surgical procedures. A higher number of septa were observed in the first and second molar (middle) region, with coronal orientation, since the first and second molar regions are among those with the most bone loss after tooth extraction and often necessitating SFE treatments. According to classification system from Irinakis et al., the coronal orientation represents a lower risk of membrane perforation as Class I septa orientation increases the predictability of treatment unless it has a combination with other orientations as in Class IV [34].
In the present study, the average height of the septa of the maxillary right and left sides were 5.04 mm and 5.32 mm, respectively, with the entire mean of 5.6 mm, which was consistent with the septa average height of the previous studies in the range of 5.4–7.3 mm [24, 30, 35].
4.2│Effect of the Gender, Age, and Dental Status on the Septa
The present study indicated no significant correlation between the prevalence of septa and patient age or gender, which is consistent with prior studies [20, 25, 36–38]. Other studies showed a negative correlation only between septa and gender [22, 39–42].
The present study found no relationship between dental status (dentate, partially edentulous, edentulous) and the prevalence of septa (63.8% of dentate, 22.1% of partially edentulous, and 14.1% of edentulous). This finding is consistent with Shen et al. [22] who observed no correlation between septa prevalence and the absence of maxillary first or second molars, consistent with recent studies [2, 18, 21, 42–44]. The present study contradicts other studies [38, 45, 46] which showed edentulous ridges have more septa than dentate ridges, as concluded by Krennmair et al. and Kim et al., who proposed the hypothesis that atrophic edentulous maxilla considers secondary septa [47, 48].
Primary/congenital and secondary/acquired classification of maxillary sinus septa has limited clinical significance since septa can appear anywhere in the maxillary sinus and do not emerge as a result of sinus pneumatization following tooth loss [18]. The etiology of septa formation is unknown, but genetics may affect septa formation during middle-face region development. Hence, maxillary sinus septa should be classified based on their location, orientation, and morphology since they are clinically significant during SFE procedures. Septa cause 11–56% of sinus membrane perforations during SFE [18, 25, 49].
4.3│Effect of the Presence and Absence of Septa on the Sinus Membrane Pathology
The prevalence of mucosal thickening ranged between 35.1–66%, and thickening is considered normal between 1–3 mm [50]. The perforation rate was higher in thicker (≥ 2 mm) and thinner membranes (< 1 mm) according to Lin YH et al. [51]. The present study found no significant correlation between Sinus Membrane Pathology and the presence or absence of septa on both right and left sides (P = 0.789 and P = 0.806), respectively. Findings from N. Kocak et al. in 2019 matched our results and observed no significant difference (P = 0.863) between septa presence and maximum membrane mucosal thickness in 376 patients and 500 sinuses [43]. By contrast, Cakur et al. reported that sinus septa might cause the Schneiderian membrane's thinness, as membrane thickness is inversely related to sinus septa [52]. However, Rancitelli et al. found no positive correlation between septa height and membrane thickness. The thickness of the membrane at the septa origin is almost twice as thick as it is away from the septa origin [53].
The present study has common limitation that is a cross-sectional evaluation of CBCT images from only two cities; thus, the findings cannot be generalized to all populations and the measurements matched by single observer. Therefore, future studies are strongly recommended to confirm the present findings.