Introduction

The paranasal sinuses are air-filled cavities located within the skull bones around the nasal cavity. They consist of the maxillary, frontal, and sphenoid sinuses and the ethmoid air cells. Their development and final shape show numerous variations, and even identical twins can have different sinus configurations [1, 2].

Various imaging modalities can be used to detect paranasal sinus pathologies and anatomic variations. Cone-beam computed tomography (CBCT) is a relatively new technology in the field of oral and maxillofacial radiology, but is widely accepted as one of the pioneering tools for paranasal sinus evaluation by dentists, maxillofacial radiologists, and otolaryngologists [3]. CBCT image assessment for the nasal cavity and paranasal sinuses offers several advantages over multislice computed tomography, including easier image acquisition, greater image accuracy facilitated by high-quality bony definition, multiplanar reformation, lower radiation doses, faster scan times, and less expensive machine [4].

The maxillary sinus holds a special place in the field of dentistry because of its close proximity to the teeth and associated structures. Among the paranasal sinuses, the maxillary sinus has a crucial role in the formation of facial contours. It is the largest of the four paranasal sinuses and is the first to develop [2, 5], with onset of development in the third month of intrauterine life [6]. The average volume of the maxillary sinus at birth is 6–8 cm3. It can be detected radiographically on a standard anteroposterior view at 5 months after birth. Pneumatization of the sinus continues both laterally and inferiorly during two rapid growth periods from birth to 3 years of age and from 7 to 12 years of age [7].

The developmental pattern of the paranasal sinuses is totally unpredictable and varies greatly among individuals [8]. The maxillary sinuses can even develop in a differently from one another in the same person [5]. After the final rapid growth phase of the maxillary sinus, many chronological and pathologic events can affect its volume. The development of the maxillary sinus has a direct relationship with the alveolar bone and palate. Because of its large volume, craniofacial anatomical features can be affected if there is a considerable change in the maxillary sinus volume [9]. Moreover, maxillary sinus hypoplasia can be associated with facial asymmetry [10]. Therefore, we focused on the volumetric changes of the maxillary sinus in this study.

There have been many studies on the relationships of maxillary sinus dimensions and volume with age and sex [5, 11,12,13,14,15,16,17,18,19,20,21,22], and a few studies have investigated the relationships of maxillary sinus volume with tooth status [9, 20, 23] and adjacent anatomical variations such as nasal septal deviation (NSD) and concha bullosa (CB) [24,25,26,27,28]. However, no studies have assessed the relationships between impacted upper premolar–molar teeth and maxillary sinus volume.

The aim of this study was to determine the effects of NSD, CB, impacted teeth, and missing teeth in the maxillary posterior region on the volume of the maxillary sinus using CBCT images. We also evaluated the normal volume distribution of the maxillary sinuses according to age and sex.

Materials and methods

This retrospective study was approved by the local ethics committee of Izmir Katip Celebi University (approval no. 80). All patients in our CBCT archive were informed and provided written consent regarding the use of their data for scientific research. The study was conducted at the Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Izmir Katip Celebi University, Izmir, Turkey. CBCT images of 1368 patients referred to our clinic for various complaints between October 2012 and July 2015 were retrospectively analyzed. All images were obtained with a NewTom 5G CBCT machine (QR srl, Verona, Italy) using the following parameters: 110 kVp and 1–20 mA. CBCT images with large fields of view (15 × 12 cm and 18 × 16 cm) were selected, because they included the entire maxillary sinus. The voxel size was 0.2 mm, and the slice thickness was 1.0 mm.

The exclusion criteria were: (1) < 12 years of age; (2) poor image quality with severe artefacts; (3) previous maxillary sinus surgery or grafting; (4) maxillary sinus pathology such as tumor or cyst; (5) acute or chronic sinusitis; (6) sinus or nasal polyposis; and (7) craniofacial anomalies such as cleft lip and palate. Rak et al. [29] reported that abnormal sinusitis only occurred when mucous thickening of ≥ 3 mm was present. Therefore, patients with mucous thickening of ≥ 3 mm in the maxillary sinus were considered to have sinusitis and excluded from the study. Finally, a total of 252 patients with 504 healthy maxillary sinuses were included in the study.

The CBCT scans of the patients were exported in Digital Imaging and Communications in Medicine (DICOM) format and imported into Mimics 16.0 software (Materialise Dental, Leuven, Belgium). All CBCT images were processed and analyzed by a single researcher. For determination of maxillary sinus volumes, coronal images were selected. Thresholding for each patient was performed manually by the same researcher. The sinuses were cropped in the slice, where their largest size was apparent. The connections between the sinuses and the outer air were then cropped in a slice-by-slice manner. Cropping was also performed in axial and sagittal images. After all connections between the sinuses and the outer air were eliminated, three-dimensional models of the left and right sinuses were created separately, and their volumes were calculated (Fig. 1).

Fig. 1
figure 1

Three-dimensional reconstructed images of right and left maxillary sinuses

Full size image

Nasal septal deviation was defined as any bending of the nasal septal contour observed on coronal CBCT images, in accordance with Bhandary and Kamath [30]. The angle between a linear line drawn from the maxillary spine to the crista galli and a linear line drawn from the crista galli to the most deviated part of the nasal septum was accepted as the deviation angle. The convex side of the curvature defined the direction of the deviation. The patients were divided into three groups according to the measured angle of NSD as described by Elahi et al. [31]: mild (< 9°; Group 1), moderate (9°–15°; Group 2), and severe (≥ 15°; Group 3) (Fig. 2). The control group comprised patients without NSD. In patients with NSD, the maxillary sinus volumes on the deviation side and the contralateral side were compared. Similar comparisons were made in the NSD subgroups.

Fig. 2
figure 2

Coronal CBCT images showing the definitions of nasal septal deviation based on the measured angle. a Mild deviation. b Moderate deviation. c Severe deviation

Full size image

CB was defined as the presence of a pneumatized middle turbinate on a coronal image. CB was classified into three different patterns as described by Bolger et al. [32]: lamellar (pneumatization localized in the vertical lamella of the concha), bullous (pneumatization localized in the bulbous segment of the concha), and extensive (extensive pneumatization of the entire middle turbinate) (Fig. 3). The control group comprised patients without CB. In patients with unilateral CB, the maxillary sinus volumes on the CB side and the contralateral side were compared. Similar comparisons were made for all subtypes among the unilateral CB cases.

Fig. 3
figure 3

Coronal CBCT images showing the definitions of concha bullosa. a Lamellar concha bullosa. b Bullous concha bullosa. c Extensive concha bullosa

Full size image

Impacted maxillary premolars and molars were detected on axial, coronal, and sagittal images and recorded. A tooth was considered impacted if it was not exposed to the oral cavity and its root development was > 75% complete [33]. The maxillary sinus volumes on the impacted tooth side and the contralateral side were compared.

Missing maxillary premolars and molars were determined on axial, coronal, and sagittal images and recorded. Because of the high prevalence of congenitally missing maxillary third molars, the absence of these teeth was not recorded as missing teeth. The maxillary sinus volumes on the missing tooth side and the contralateral side were compared.

Statistical analysis

Statistical analysis was carried out using SPSS v.22 software (IBM, Chicago, IL, USA). Conformity of parameters to a normal distribution was assessed by the Kolmogorov–Smirnov test. A paired-sample t test was used to compare the mean bilateral maxillary sinus volumes. Student’s t test was used to compare continuous variables. Correlations between parameters were evaluated by Pearson’s correlation analysis. Statistical significance was determined at p < 0.05. For reliability testing, 63 CBCT images (25% of the images) were randomly selected for repeated measurements at 3 weeks after the first examination by the same researcher. The intraclass correlation coefficient was used to determine the reliability of the measurements.

Results

A total of 252 patients with 504 healthy maxillary sinuses were included in the study. The patients comprised 96 males and 156 females with an age range of 12–85 years (mean 31.21 ± 17.71 years). The coefficients of reliability for all measurements were > 0.98. The results of the paired-sample t test revealed no significant difference between the two sets of readings (p > 0.05). The mean right and left maxillary sinus volumes were 14.64 ± 5.19 and 14.66 ± 5.2 cm3, respectively, with no significant difference (p = 0.88). The overall mean maxillary sinus volume was 14.65 ± 5.19 cm3. The mean right, left, and total maxillary sinus volumes of male patients were significantly larger than those of female patients (p < 0.01) (Table 1).

Table 1 Comparisons of maxillary sinus volumes between males and females
Full size table

A negative correlation was observed between the right, left, and total maxillary sinus volumes and age (p < 0.05), meaning that maxillary sinus volume tended to decrease with age (Table 2).

Table 2 Correlations between maxillary sinus volumes and age
Full size table

NSD was found in 226 (89.68%) patients. Of these, 118 (52.21%) had convexity to the right, 98 (43.36%) had convexity to the left, and 10 (4.43%) had biconvex NSD. Twenty-six (10.32%) patients had a straight nasal septum and were assigned to the control group for NSD. Patients with biconvex deviations were not included in the NSD group. The mean right and left maxillary sinus volumes in the control group were 14.51 ± 4.07 and 14.16 ± 4.41 cm3, respectively, with no significant difference (p = 0.32). However, in the NSD group, the maxillary sinus volumes were significantly smaller on the same side as the deviation than on the contralateral side (p < 0.05) (Table 3).

Table 3  Maxillary sinus volumes according to nasal septal deviation side
Full size table

The NSD angles ranged from 4.46° to 22.13°. The patients with NSD were classified as 76 mild (< 9°; Group 1), 97 moderate (9°–15°; Group 2), and 43 severe (≥ 15°; Group 3) cases. The maxillary sinus volumes were significantly smaller on the same side as the deviation than on the contralateral side in Groups 2 and 3. There was no significant difference in Group 1 (Table 4).

Table 4  Maxillary sinus volumes in nasal septal deviation subgroups
Full size table

Among the 252 patients, 195 (77.4%) had at least one CB. Unilateral CB was present in 62 (24.6%) patients and bilateral CB in 133 (52.8%). Among the 62 unilateral CB cases, lamellar, bulbous, and extensive CB were found in 36, 8, and 18, respectively. The maxillary sinus volumes were compared in patients with unilateral CB. Similar comparisons were made for each subtype among the unilateral CB cases. No significant differences were found between the CB side and the contralateral side (p > 0.05) (Tables 5, 6).

Table 5  Maxillary sinus volumes in 62 patients with unilateral concha bullosa
Full size table
Table 6 Maxillary sinus volumes according to unilateral concha bullosa subtypes
Full size table

The investigation for impacted maxillary molars and premolars revealed that 29 (11.5%) patients had unilaterally impacted teeth and 25 (9.9%) patients had bilaterally impacted teeth. The bilateral maxillary sinus volumes were compared in patients with unilateral impacted teeth. The maxillary sinus volumes were smaller on the same side as the impacted teeth than on the contralateral side, but the difference was not significant (p > 0.05) (Table 7).

Table 7  Maxillary sinus volumes in 29 patients with unilateral impacted teeth
Full size table

The examination for missing maxillary molars and premolars (at least one molar or premolar) revealed that 39 (15.5%) patients had unilaterally missing teeth and 74 (29.4%) patients had bilaterally missing teeth. The 39 patients with unilaterally missing teeth were divided into two groups by age: group A, age ≤ 20 years; group B, age > 20 years. The bilateral maxillary sinus volumes were compared in both groups. The mean maxillary sinus volume was smaller on the same side as the missing teeth than on the contralateral side in group A, while it was larger in group B. No significant difference was observed in both groups (p > 0.05) (Table 8).

Table 8 Maxillary sinus volumes in Group A and Group B
Full size table

Discussion

In the previous studies, the volumes of paranasal sinuses were measured on cadaveric materials [34], dry skulls [35], computed tomography scans [9, 20, 24,25,26,27], magnetic resonance images [36], and CBCT images [19, 37, 38]. The techniques for volume measurement on cadaveric materials and dry skulls cannot be used in living subjects [11]. Furthermore, measurements on dry skulls may be larger than the actual sizes because of the lack of soft-tissue lining [20]. Measurement of maxillary sinus volumes using CBCT has certain advantages, such as low radiation doses compared with computed tomography and low cost compared with magnetic resonance imaging [19]. Therefore, CBCT images were used for measurement of maxillary sinus volumes in the present study.

The maxillary sinus has two rapid growth phases, one between birth and 3 years of age and another between 7 and 12 years of age. After these rapid growth phases, the maxillary sinus volume becomes approximately adult-sized between 12 and 15 years of age [7, 39]. Some authors reported that the maxillary sinus volume does not increase significantly after 12 years of age [40, 41]. Therefore, it was decided to evaluate the maxillary sinus volume in patients aged > 12 years in this study.

In the present study, a negative correlation was observed between maxillary sinus volume and age, meaning that the maxillary sinus volume tended to decrease with age. This finding was consistent with the previous reports by Cho et al. [9] and Cohen et al. [18]. In other studies, Kim et al. [14], Sahlstrand-Johnson et al. [11], and Emirzeoğlu et al. [17] found no significant correlation between maxillary sinus volume and age in patients aged > 18 years. However, these three studies all included less than 80 patients, and the lack of correlation may have been related to the low numbers of patients. Meanwhile, Karakas et al. [21] showed that the volume of paranasal sinuses increased with age in patients aged 5–55 years. However, this positive correlation can probably be explained by their inclusion of very young patients.

There have been many studies on the relationship of maxillary sinus volume with sex. Several authors reported differences in maxillary sinus volumes between males and females [5, 11,12,13,14,15,16,17,18], while other authors showed no such difference [19, 20]. We found that the mean right, left, and total maxillary sinus volumes in male patients were significantly larger than those in female patients. This difference in maxillary sinus volumes between sexes can probably be attributed to the fact that males are generally physically larger than females in most dimensions.

In the present study, there was no significant difference in mean volumes between the right and left maxillary sinuses, consistent with the previous studies [11, 20]. Because maxillary sinus volumes can vary by age and sex, as observed in our study and other studies, we compared the bilateral maxillary sinus volumes, rather than comparing different patient groups, when investigating the maxillary sinus volume changes in the presence of NSD, CB, and impacted or missing maxillary posterior teeth.

There are a few studies on the relationship of maxillary sinus volume with NSD. In our study, there was no significant difference between the right and left maxillary sinus volumes in the control group. However, in the NSD group, the maxillary sinus volumes were significantly smaller on the same side as the deviation than on the contralateral side. Orhan et al. [24] compared maxillary sinus volumes in patients with NSD and control patients. They found that maxillary sinus volumes were significantly smaller on the same side as the deviation than on the contralateral side in patients with NSD. They did not find a significant difference between the left and right maxillary volumes in the healthy control patients, as also observed in our study. Aydın et al. [42] compared the maxillary sinus volumes of 36 patients with antrochoanal polyps on the ipsilateral and contralateral sides to the deviation and found no significant difference in the mild, moderate, and severe deviation groups. Kapusuz Gencer et al. [25] reported that maxillary sinus volumes were higher on the contralateral side to severe septal deviations, while mild and moderate septal deviations had no significant effect on maxillary sinus volumes. In another study investigating the effect of septal deviations on frontal and maxillary sinus volumes, moderate septal deviation was found to significantly impact the maxillary sinus. These findings suggested that maxillary sinus volumes tended to be higher on the contralateral side to moderate septal deviations [28]. In our study, although mild septal deviations lacked any significant effect on maxillary sinus volumes, moderate and severe deviations had a significant impact on these parameters. Contrary to our study, Kucybala et al. [26] found that NSD had no impact on maxillary sinus volume, for either right or left deviations. However, they did not categorize their patients according to the angle of NSD. In our study, mild septal deviation lacked any significant effect on maxillary sinus volume, consistent with the previous studies [25, 28, 42]. Therefore, the study by Kucybala et al. [26] may have contained high numbers of patients with a small deviation angle or mild deviation.

Nasal obstruction caused by NSD can increase nasal airway resistance and cause turbulent nasal airflow, precipitating dryness and crusting of the nose, frequent nosebleeds, and recurrent sinusitis [43, 44]. The resultant impaired nasal breathing can lead to chronic mouth breathing, which in turn can result in moderate-to-severe maxillary constriction, and a vertical skeletal growth pattern characterized by long anterior lower face height, bilateral maxillary crossbite, high arched palate, low tongue posture, and insufficient lip support [45, 46]. In addition, it has been hypothesized that nasal respiration enables normal growth and development of the craniofacial structures. According to the functional matrix theory, undisturbed nasal airflow is a continuous stimulus for lowering of the palate and lateral maxillary growth, indicating a close relationship between nasal breathing and dentofacial morphology [47]. Our study showed that mild and moderate septal deviations caused insufficient growth of the maxillary sinus on the same side as the deviation. The insufficient growth probably arose from inadequate ventilation of the maxillary sinus on the deviation side. Based on these findings, we suggest that nasal respiration should be controlled in dental patients, especially adolescent patients. If any impaired nasal breathing is present, the patients should be referred to an otorhinolaryngologist. Thus, treatment of conditions that can lead to impaired nasal breathing, such as NSD, may result in avoidance of craniofacial disorders. Furthermore, based on its greater image accuracy, the authors recommend that dentists should use CBCT as a reliable tool for NSD evaluation, and further suggest that when NSD is diagnosed on CBCT examination, patients should be referred to an otorhinolaryngologist for further evaluation.

CB is defined as pneumatization of the middle turbinate and is one of the most common anatomic variations of the osteomeatal complex region [48]. We hypothesized that pneumatization of the middle turbinate may affect maxillary sinus growth and volume, because the turbinates are the first structures to develop in the lateral nasal wall. However, we did not find a relationship between unilateral CB and maxillary sinus volume. We also found no relationship between each subtype of unilateral CB and maxillary sinus volume. Demir et al. [27] also demonstrated no significant association between CB and maxillary sinus volumes. Kucybala et al. [26] compared the maxillary sinus volumes in a patient group with bilateral CB and another patient group without CB or with unilateral CB. They reported that CB did not affect the left and right maxillary sinus volumes in unilateral CB cases, similar to our study and Demir et al. [27]. However, they found larger right and left maxillary sinus volumes in bilateral CB cases compared with unilateral CB cases or patients without CB. In their study, the patients did not have a homogeneous age (range 18–97 years) or sex distribution. Maxillary sinus volume can vary according to age and sex, as seen in our study and many other studies [5, 9, 11,12,13,14,15,16,17]. The nonhomogeneous patient distribution may have affected the outcome of their comparisons performed in independent groups. We consider that a new study taking age and sex distributions into consideration may provide more reliable results when assessing the effect of bilateral CB on maxillary sinus volume.

Impacted teeth cause many complications [49]. To the best of our knowledge, no previous studies have investigated the relationship between impacted maxillary premolars or molars and maxillary sinus volume. In the present study, bilateral maxillary sinus volumes were compared in patients with unilateral impacted teeth. Although no significant difference was detected between the two sides, maxillary sinus volumes were smaller on the same side as the impacted teeth than on the contralateral side. Based on this result, we believe that impacted teeth may have a negative effect on maxillary sinus volume.

Many authors have reported that maxillary sinus pneumatization occurs after posterior maxillary tooth extraction [50, 51]. Nevertheless, there are only a few studies on the relationship of maxillary sinus volume with missing maxillary molars or premolars. Cho et al. [9] suggested that periodontitis and premolar or molar loss did not affect maxillary sinus volume. Ariji et al. [20] reported that the mean maxillary sinus volume of dentate patients was significantly greater than that of edentulous patients. However, when they took the age distribution into consideration, there was no significant difference in volume between dentate and edentulous patients. Takahashi et al. [23] reported a trend toward a reduction in maxillary sinus volume with molar loss, based on research using elderly Japanese cadavers. Velasco-Torres et al. [52] observed that total maxillary sinus volume was significantly smaller in total and partially edentulous patients than in dentate patients. These studies compared maxillary sinus volumes in a pool of patients rather than in the same patients. This can lead to reduced accuracy, because maxillary sinus volume varies with age and sex. Therefore, we compared bilateral maxillary sinus volumes in patients with unilaterally missing maxillary premolars or molars to avoid the effects of age and sex on sinus volume. While the mean maxillary sinus volume was smaller on the same side as the missing teeth than on the contralateral side in patients aged ≤ 20 years, it was larger in patients aged > 20 years. However, no significant difference was observed in both groups. According to our findings, missing maxillary premolars or molars may negatively affect sinus volume and its development in patients aged ≤ 20 years, but may have a positive effect on sinus volume in patients aged > 20 years because of sinus pneumatization.

A limitation of this study is that the numbers of patients with unilaterally impacted or missing maxillary premolars and molars were relatively small. We believe that further studies with larger sample sizes are required to clarify the association between impacted or missing maxillary posterior teeth and maxillary sinus volume in the general population.

In conclusion, the present study revealed a significant reduction in maxillary sinus volume on the deviation side for moderate and severe NSD. Because of insufficient growth of the maxilla and maxillary sinus, this reduction may result in dental malocclusion and maxillary asymmetry. Our findings suggest that CB has no association with maxillary sinus volume. Impacted or missing maxillary premolars and molars may be related to maxillary sinus volume. Further research should focus on maxillary sinus volume in patients with impacted or missing maxillary premolars and molars.